AJANTA ARGUMENTS ABOUT AJANTA
HANDBOOK OF ORIENTAL STUDIES HANDBUCH DER ORIENTALISTIK SECTION TWO
INDIA INDIEN edited by J. BRONKHORST
VOLUME 18/2 AJANTA: HISTORY AND DEVELOPMENT
ARGUMENTS ABOUT AJANTA
AJANTA: HISTORY AND DEVELOPMENT VOLUME TWO
ARGUMENTS ABOUT AJANTA BY
WALTER M. SPINK
BRILL LEIDEN • BOSTON 2006
Cover illustration: Ajanta Cave 1, front wall “Persian Embassy”, detail 477. This book is printed on acid-free paper. Library of Congress Cataloging-in-Publication Data A C.I.P. record for this book is available from the Library of Congress.
ISSN 0169-9377 ISBN-13: 978 90 04 15072 0 ISBN-10: 90 04 15072 2 © Copyright 2006 by Koninklijke Brill NV, Leiden, The Netherlands Koninklijke Brill NV incorporates the imprints Brill Academic 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 Brill 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
CONTENTS
VOLUME IIA. ARGUMENTS ABOUT AJANTA Chapter Chapter Chapter Chapter Chapter
One Two Three Four Five
A discussion of H. Bakker’s The Vakatakas Cave 26 as an inaugural monument .......... Cave 26’s complicated development ............ Cohen’s “Possible Histories” ........................ Scholarly contributions to Maharashtra Pathik ..............................................................
3 22 54 97 115
VOLUME IIB. PATTERNS OF PATRONAGE PART I: PATRONAGE: CONSISTENT VS. COLLAPSING Ajanta differs from most other Buddhist sites, generally created as community undertakings, in that it was purely “elitist”, developed by less than a dozen major patrons, who planned it with great care in essentially one great burst of enthusiasm. This enthusiasm turned to desperation immediately after Harisena’s death, when the patrons anxiously rushed their shrine images to completion, in order to secure the merit from so doing. During the fifteen years or so that the site flourished under the aegis of these proud donors, including the emperor Harisena himself, no “outsiders” ever could donate a single thing. But once the great patrons had rapidly departed from the collapsing site during the disastrous reign of Harisena’s successor, the monks still resident there, along with local devotees, briefly sponsored a helter-skelter spate of votive donations, also to make merit while they could. After about 480 this activity probably stopped completely, the craftsmen having gone away. The monks continued to live in some of the caves for perhaps as much as another decade after which the site was totally abandoned, except for the use of a few cells by Saivite sadhus and the like in later centuries.
Chapter Six Chapter Seven Chapter Eight Chapter Nine
Patterns of Patronage at Ajanta: Consistent vs. Collapsing .............................. Locating Intrusions in Time ........................ Could Any Intrusions Date Before Mid-478? ........................................................ Caves 9 and 10: Their Redecoration and their Intrusions Excavations: Dead or Alive ..................................................
149 158 161
167
contents
vi Chapter Ten Chapter Eleven
Crises and Cave 1 .......................... 174 The Breakdown of Patronage in the Period of Disruption ........................ 192
PART II: PATRONAGE: THE HINAYANA CAVES WITH EMPHASIS ON THEIR REDECORATION IN VAKATAKA TIMES Although we can assume that, being so readily available, the old Hinayana viharas and caitya halls would have been used both for residence and worship when excavation work began again at Ajanta in about 462, it appears that no one troubled to redecorate or refurbish them until the new Vakataka phase was well underway— perhaps not until the very last year of Harisena’s reign. The fact that these efforts came all too late is evidenced by the manner in which they soon had to be brought to a halt, due to the political situation. After this, “intrusive” donors took over, filling most of the still available areas with their votive offerings. All of these developments contribute to our knowledge of Ajanta’s turbulent history.
Chapter Twelve
Patronage of the Hinayana Caves: Considerations .................................. Chapter Thirteen Cave 10: Redecoration .................... Chapter Fourteen Cave 10: Intrusions: Summary ...... Chapter Fifteen Cave 10: The Aisle Paintings: Original and Intrusive .................... Chapter Sixteen Cave 10: Façade Intrusions ............ Chapter Seventeen Cave 12: .......................................... Chapter Eighteen Cave 9: ............................................ Chapter Nineteen The Anomalous Painting on Cave 9’s Rear Wall ........................ Chapter Twenty Cave 9: Triforium Paintings; Aisle Wall Paintings ........................ Chapter Twenty-One Cave 9: Palimpsests and other Transformations .............................. Chapter Twenty-Two Cave 9: Intrusions on Pillars .......... Chapter Twenty-Three Cave 9: Façade Intrusions .............. Chapter Twenty-Four Cave 9: Considerations about Usage .................................................. Appendix
Ajanta’s Inscriptions ..............................................
199 204 221 230 233 235 239 245 251 257 259 262 266 273
VOLUME IIA
ARGUMENTS ABOUT AJANTA
CHAPTER ONE
A DISCUSSION OF H. BAKKER’S THE VAKATAKAS
The Vakatakas: A Study in Hindu Iconology, by Hans Bakker (Groningen, 1997) is an impressive exposition of important Vakataka sculptures, only recently being given their due, and of their revealing religious and political context. However, some of Bakker’s conclusions are highly controversial, particularly with regard to Vakataka history in the late years of the dynasty. I shall attempt to show how a very close—in fact, approximately year by year—analysis of the remarkable developments at Ajanta (and related caves) can lead to a revision of such commonly held views. At the same time, this should greatly magnify the image of the great emperor Harisena, who (in my view) was responsible for the startling, and final, florescence of the great Vakataka empire in central India during his brief reign from about 460 through about 477 A.D. Too long regarded as poor relations of the Guptas, the Vakatakas themselves, under the powerful Harisena, were in fact the final sponsors and guardians of the so-called Golden Age.1 However, when the great emperor Harisena died, the world of central India suddenly was disrupted, and the Golden Age suddenly ended. This became evident almost immediately at Ajanta, dependent as it was upon courtly patronage. By the end of 478, Ajanta’s long established patronage had totally collapsed; and by the early 480s, the huge empire itself, which Harisena, through war, marriage, inheritance and intrigue, had put together, had been broken back into the constituent parts which he had gradually welded together as he extended his domains from the eastern to the western sea.2 And that was the end of the Vakatakas.
1 To parallel this statement, I have taken the liberty of slightly revising the labeling of the map of India in this period, to give the Vakatakas (at least during Harisena’s reign, equal status with the Guptas. 2 This spread of empire is evident from the listing of territories in his minister’s (Varahadeva’s) Cave 16 inscription, verse 18; see Volume 1; the territories which he (using Bakker’s term) “stood above” stretch from the eastern to the western sea.
volume iia, chapter one
4
Their rule did not continue into the early sixth century, as most scholars believe.3 It ended with the rapid destruction of the empire in the early 480s, during the reign of Harisena’s inept successor, Sarvasena III, who had inherited the finally unified empire from his father, but disastrously failed to hold his inheritance together.4 The sculptures with which Bakker deals are generally isolated images, often found in an architectural context. They were produced— exclusively—in the eastern Vakataka domains from the late fourth century into the last half of the fifth or (Bakker suggests) the early part of the sixth.5 Many of these works have been recently discovered, or at least only recently recognized for what they are. A number have been published previously, notably by A. P. Jamkhedkar (1991, 85–92), who has brought so much of this material to light over the course of the last twenty years.6 Earlier, we knew next to nothing about this fascinating body of work. Furthermore, Bakker focuses new attention on the Vakataka dynasty, which has been almost as disregarded as these sculptures which it produced. Clearly it is now due for a scholarly renaissance. As Bakker (1997, 2) succinctly states: “One may say that from the middle of the sixties (of the fifth century) the kingdom of the Vakatakas has come to be seen as pivotal in the history of India, being essential for our understanding of the development of its art, religion and culture; as such it is on a par with the Gupta world, of which it can no longer be considered to be merely a province.” It is gratifying to have the Vakatakas so appropriately raised in status. All too often they get little credit beyond the astute marriage of the Vakataka king Rudrasena II to the Gupta princess Prabhavati Gupta in the late fourth century—the stress being put on the Gupta rather than the Vakataka connection. In the following discussion, I shall attempt to further amplify Bakker’s assertion of the Vakatakas’
See, for instance, Khandalavala 1992, 123 ff. For a justification of such dating, and related historical considerations, see Spink 1991B, 71–99. 5 According to my chronology, the latest Vakataka sculptures in the Nagpur region would not post-date 477 or possibly 478, when the Vakataka empire was collapsing. 6 The famous Ganga from Paunar ( Jamkhedkar, 1991), plate 13 (page 92) and the related Ramayana cycle installed at Vinobha Bhave’s ashram appear to be postVakataka, probably Visnukundin—reflecting of course the previously strong political and even marital connections while Harisena controlled the empire. 3 4
a discussion of h. bakker’s the vakatakas
5
importance, in part by arguing that Ajanta, in its great later phase, is in fact a Vakataka site.7 In the past, the credit for Ajanta’s flourishing was typically diffused, with shares being given to the Guptas, the Calukyas, the Rashtrakutas, and even the Asmakas.8 This has obviously diminished the Vakataka’s reputation; whereas to see the whole site (along with Bagh, Dharasiva, Ghatotkacha, Banoti and Aurangabad Caves 1, 3, and 4A) as all having been accomplished in less than twenty years under Vakataka patronage, dramatically defines the cultural potency of the Vakataka empire which sponsored it. When such a reinterpretation of Ajanta—which reveals the high character of Vakataka culture in its various art forms—is combined with Bakker’s presentation of the previously undiscovered and/or little understood Vakataka material from Eastern Vidarbha, the reputation of the too-neglected dynasty will better come to receive its due.9 Since his concern is to elucidate the manner in which “art is embedded in a social and cultural context”—that is, to deal with “iconology” (Bakker 1997, 3)—the information that such sources provide can clearly add to one’s understanding. This is especially the case with the (Western) Vakataka site at Ajanta, which provides an “illustrated history” of Buddhist patronage, wherein so many social, political, and economic forces are at play. Admittedly, Bakker’s title—The Vakatakas—is misleading, for this implies that the book will give a relatively equal coverage to the Eastern (Nandivardana) and the Western (Vatsagulma) branches of the house. However, it is essentially about the former. He has made a “deliberate decision to leave out the specific Buddhist evidence . . . . one will not find, for instance, an art historical assessment of the
7 Even from 478 to 480, when the Asmakas had declared their independence from their Vakataka overlords, Sarvasena was still the “legal” ruler, for he had not yet been defeated (in fact, killed) during the Asmaka insurrection. 8 The Asmakas were in fact among the inaugurators of the site in around 462, and still later, from 475 on, had taken it over completely, although still (from the Vakataka point of view) nominally feudatories of the Vakatakas, even after their rebellion in 478. 9 Bakker provides a generous selection of the little known Vakataka sculptures, along with very useful maps, although one regrets the lack of views of related structures, such as the Ramagiri temples; nor does Bakker illustrate any of the important sites (such as they are), despite including a useful chapter describing the particular sites from which his illustrations are drawn. He also provides an excellent bibliography, a revised edition (unfortunately not translated) of the important KevalaNarasimha Temple Inscription, and an (arguable) “Outline of Vakataka Chronology”.
6
volume iia, chapter one
Ajanta Caves . . .” (Bakker 1997, 3). Although he breaks this rule occasionally, the focus of his book is on the Hindu materials from the Eastern Vakataka realm. These sculptures he analyzes with erudition, drawing upon an impressive body of textual sources and explicating their religious and political connections in an exemplary way, while leaving enough room for scholarly contention. Bakker’s “second strategy” is “to utilize textual and archaeological sources in combination as much as possible”. (Bakker 1997, 3) But here too he underplays his options by siding with convention, and giving short shrift to the Visrutacarita (the eighth ucchvasa of Dandin’s Dasakumaracarita) the one textual source which I am convinced can clarify Vakataka history most fully. In contrast, Bakker insists: “At best (the Visrutacarita) could be seen as a kind of Alexander Romance, valuable in itself, though no historian would use it as his primary source . . .” (Bakker 1997, 37) If I am correct—and admittedly in this I am in total disagreement with most scholars—the Visrutacarita provides us with clear evidence that the Vakataka dynasty came to an end under the Western (Vatsagulma) branch, not the Eastern (Nandivardana) branch. (Spink 1991B, 71–92) Furthermore, it ended far earlier (in the early 480s) than scholars normally suppose.10 At the same time Bakker (1997, 5), claiming the “limits set by (his) own competence”, intentionally avoids getting involved in exploring the rich evidence from Ajanta, which confirms what we speak of as the “Short Chronology”, once we see it as a “dig” and analyze its various “strata”.11 These two sources—Ajanta, and the Visrutacarita—which should neither be solely “left to art historians” (Bakker 1997, 41), nor for reading as a mere “Alexander Romance”—are the essential keys to late Vakataka history, and we exclude them at our scholarly peril. Admittedly, my friend Karl Khandalavala (now sadly deceased), speaking for many, repeatedly warned that “there is no end to Dr. Spink’s make-believe theories” (Khandalavala 1991, 123) and it is true that my “chimerical” views by no means ring of established
10 Although I owe much to Mirashi’s fundamental study, “Historical Evidence in Dandin’s Dasakumaracarita” (Mirashi 1945) my conclusions are based upon a significantly different chronology, resulting in a very different view of history. 11 See Spink 1991A (“The Archaeology of Ajanta”) for a discussion of the site’s recurring crises and revealing “strata” with which they can be linked.
a discussion of h. bakker’s the vakatakas
7
truth.12 I have nonetheless shown, utilizing my revised chronology (Spink, 1992, 23) how the Visrutacarita tallies remarkably with both textual and archaeological evidence; and I also show how an analysis of Ajanta’s development forces us to recognize the painful intrusions of the aggressive Asmakas into the dynastic history of Harisena and his successor Sarvasena III, whose great house they finally destroyed. It is too much to think that I can prove the remarkable relevance of Ajanta just by writing about it. Readers should go there and treat the caves with the same physical and mental precision that is typically involved in the excavation of any major archaeological site. However, I can at least present something of the revealing complexity of its development. It is one thing to say with Bakker (1997, 89) that “we do not subscribe to (the) opinion that all Mahayana caves (at Ajanta) were excavated in twenty years or less” or, with Williams, to “prefer to think of Vakataka patronage at Ajanta as extending roughly from 460–505 . . . (with a probable) absolute range of c. 460 to 530” (Williams 1982, 182, 186). But how could one explain to the established courtly patrons, all anxious to get their donations completed while they were still alive, why such work would have to take so long, especially since all of the major caves at the site, with the exception of two (Caves 14, 28), were started in the same single initial burst of activity in the early 460s?13 Therefore, reversing the usual objection: “How could it have been done so quickly?” one should try to answer the more relevant question, given the intense commitment of its eager patrons, most of whom were already “of a certain age”: “Why did it take so long? How could it have taken nearly twenty years”!14 Ajanta is unique in its complexity, and in the manner in which its complexity can be elucidated. Rather than believing that its “assessment is only partially relevant for the political history (of the
12 For various critiques of my theories by K. Khandalavala, see Khandalavala 1990; 1991A; 1991B. 13 A few very small and very unfinished caves, placed at a higher level where there was still room, were also started very late, probably all in 477 (See Time Chart). One other, barely undertaken in 477, was at a lower level, between Caves 21 and 23; unfortunately, it was recently mostly filled up with cement. 14 As I suggest in Volume I, fifteen (or even fourteen) years might be a closer approximation for the period of the site’s consistent development—excluding the two or three years occupied by its subsequent Period of Disruption.
8
volume iia, chapter one
Vakatakas)” (Bakker 1997, 40) we may find that Ajanta’s analysis is the most significant key of all to that assessment. Indeed, it is my conviction that a proper analysis of Ajanta’s development, together with a related reading of the Visrutacarita, must transform our whole view of Indian history in the last half of the fifth century, and will also have a great impact upon our understanding of the chaotic developments of the early sixth. Instead of seeing the great vessel of the Golden Age slowly being tarnished as the Guptas, starting with Skandagupta, gradually failed in power and authority in the last half of the fifth century, we must see the vessel polished to its greatest splendor by the Vakatakas, during the reign of the great Harisena, before being shattered irreparably by his sudden and tragic death in c. 477.15 It was left to his successors—the multiple dynasties which made up Harisena’s fold—to pick up the pieces. Indeed, they were the pieces, now become separate again, which he had earlier forged together in the constructing of his empire. In the discussion below, I will make continual references to Bakker’s opinions, even though it will be evident that I am trying to sell my own goods. Bakker’s study is excellent in so many ways that I am concerned lest the weight of its authority may damage the more delicate structure of truth, even in those areas where (in my opinion) it is clearly wrong, notably in his reconstruction of late Vakataka history as well as in his brief forays into Ajanta’s development. So I shall present my alternative view of these matters. But my interest is not so much to merely present the conclusions as to convince the reader of the possibilities of analyzing Ajanta’s development in very precise detail. Such an analysis can lead to a new synthesis which goes well beyond the study of the caves themselves. I should add a word in justification of my very explicit year-byyear dating of the developments in the caves. Ajanta’s brief development from c. 462 to c. 480 is so crowded both with forms and with transformations that, more than any other site in the world, it can be picked apart (in order to be put together), with what many
15 Obviously this exaltation of the long-neglected emperor Harisena is hardly part of current views of Indian history. Romila Thapar, the doyen of historians, does not even mention Harisena in her new edition of The Early History of India. By contrast, I have asserted that Harisena may have been the most successful ruler in the whole world in the 460s and 470s. Such a statement may perhaps be seen as a devotee’s overstatement; but so far, no one has named a rival.
a discussion of h. bakker’s the vakatakas
9
will see as a hardly believable precision.16 In fact, as I explain elsewhere, we must see my so-called “short chronology” as slightly flexible, as if all of my dates are inscribed on an elastic band which could be stretched (or perhaps better, could be contracted) by a year or two at either end as well as at certain points between.17 If I am correct, the sequential relationships on this continuum, because of the insistently revealing nature of the site’s interconnected developments, should hardly change, even if one should decide to make a slight adjustment of its outer limits or of some of its intermediate sections: the (war-produced) “Hiatus”, for example, may well have occupied slightly more or slightly less than the three years (472–474) which I have arbitrarily assigned to it on the basis of the various developments occupied within this brief period. (See Time Chart) Furthermore, to make these developments more usefully “graphic”—to make a clear distinction, for instance, between the events of the year just before his death and the year just after— Harisena has to die, rather arbitrarily, on “Dec. 31, 477”, rather than on some indeterminate day in October or November. By the same token, the Asmakas have to be expelled from the region, rather decisively, on Dec. 31, 468, while the site’s lesser patrons, already affected by the implications of this event, started efficiently rushing their shrine Buddhas to completion on New Year’s Day, in 469 (see Time Chart). The reader must forgive this obviously imprecise “precision”, for the simple reason that it allows us to divide the site’s dramatic development into more workable units. What I hope to show, in this explanation of a small portion of Ajanta’s myriad features, is that until we yield to the demands of the literally solid evidence that the site presents, we are not going to get very far along the complicated pathway which can lead us to a satisfactory understanding of its amazing history. Ideally (and actually) too, we must go to the caves themselves for our illustrations. The motifs and features which must be studied, as one moves toward
16 For a justification of such dating, and related historical considerations, see Spink 1991A, 71–99. See also Volume I: Chapters 2, 4. 17 Although my own approach has certain parallels with that of Philippe Stern, the concentration on single motifs which typify his own work and that of his associates, is diametrically opposed to my method, which involves a consideration of the interconnected myriad motifs which compose both the separate caves and the site of which they are a part. See Schastok 2000, 7–14.
10
volume iia, chapter one
conclusions, are so many, so varied, and so interconnected, that no book could properly picture or describe them sufficiently to satisfy all of the demands of a proper investigation.18 Therefore, the illustrations included in volume four can eventually only be seen more as an “aide memoire” than as sufficient evidentiary material. As noted above, Bakker’s main focus is upon the Hindu sculpture from the eastern (Nandivardana) branch of the Vakatakas instead of upon Buddhist monuments of the western (Vatsagulma) branch. However, he also notes the influence of these eastern productions upon the generally later developments—chiefly at Ajanta—sponsored by the great emperor Harisena of the dynasty’s Vatsagulma branch. Clearly the evidence at Ajanta supports his view (Bakker 1997, 44) that “a stream of artisans moving from Nandivardana to Vatsagulma was certainly only one of the creative forces that contributed to the development of the Caves”. I would only insist that this was only one of many streams, the totality of which was responsible for Ajanta’s rich complexity—even confusion—of style. But certainly, as Harisena became more and more powerful, and ultimately became the sole heir to Vakataka power, workmen would have been increasingly drawn from the major cities and sites under his control; and surely Nandivardana would have been most particularly represented. Far more concerned than most scholars about the role that cultural forces play in nourishing such monuments as Ajanta, Bakker rightly stresses the role of the merchants traveling upon the flourishing trade routes, who would have played a significant part in supporting the costly Buddhist undertakings at Ajanta. To this I would add the suggestion that the popularity in the caves of representations of Avalokitesvara as Lord of Travelers would enhance Bakker’s assumption; indeed, such images typically appear at moments of great political and/or military stress, but particularly from mid-478 to 480—for these were times when there must have been real danger upon the roads.19
18 For Khandalavala’s fervent rejection of this “fallacious methodology for dating the caves, see Khandalavala 1991, 96. 19 There are also two examples in the caves of the local king, whose patronage suddenly ended in 471 due to the Asmaka attack. One painted image is on the short front wall of the porch of Cave 17, at the left; the other on the rear of the right porch pilaster capital in Cave 20; but both appear to have been done before the situation became critical.
a discussion of h. bakker’s the vakatakas
11
One must agree too with Bakker’s perceptive suggestion that these “western” productions reflect a newly self-indulgent world, with an economy apparently based more on luxury than responsibility, still unaware and increasingly unprepared for—so I would say—the deluge which was soon to come. The fate of Ajanta’s local feudatory king, who “covered the world with stupas and viharas” (Cave 17 inscription) but was not prepared for the war which was to destroy him, was only reflected on a larger scale by the fate of the great emperor’s house hardly a decade later. Considering the obsessive energy which could produce such a vast undertaking as Ajanta in less than twenty years—even if not a single cave was ever finished—Bakker is surely right in seeing it as a “heavy drain upon the economic resources of the kingdom” (Bakker 1997, 44). But there may have been another threat too in the western realm, where the emperor Harisena himself, surely also devoted to the Buddha, “allowed his courtiers to improve their balance of merit and demerit by spending parts of their earnings or spoil on lavishly decorated Buddhist monasteries.” (Bakker 1997, 45) Indeed, Buddhism is hardly a faith that prepares one for a particularly committed defense against aggression. The fact (at least I see it as a fact) that when Harisena suddenly died, his many courtiers, including his own Prime Minister, immediately and single-mindedly turned inward, to complete the shrine images deep within their caves, only enhances Bakker’s point. At the same time he is surely wrong when he asserts that “the king himself seems to have taken only a marginal interest in these by-products of his policy” (Bakker 1997, 45). I would assert just the opposite: that Harisena was clearly committed to this great effort—even if perhaps only for reasons of politics and prestige. What cannot be denied (again this is my view) is that it was Harisena himself who was responsible for the donation of Cave 1, the most sumptuous rockcut vihara ever made in India (Spink 1981, 144–157). And the beautiful paintings with which he filled it, all focused on themes involving the character of kingship, do indeed exalt the virtues of sacrifice and of generosity, rather than the more practical values of “heroic” aggression. In fact it must have been with his approval, and perhaps with imperial support, that Ajanta burst so dramatically upon the scene in the early 460s; and his support must have been a factor in the roughly contemporaneous and equally rapid developments at Bagh, which provided a profitable sanctuary for Ajanta’ workmen during Ajanta’s troubled Recession and Hiatus from early 469 through 474.
12
volume iia, chapter one
Harisena’s religious affiliation is a matter which has by no means been settled. However, it is hard to believe that the donor of the greatest vihara in India was not in some sense a Buddhist, despite his Saivite ancestry. It is relevant to point out, with Bakker (1997, 58) that “The Vatsagulma kings call themselves Dharmamaharaja, without being more explicit about their religious orientation . . . The title . . . may therefore be interpreted as a general one indicating an ecumenical attitude towards the religions represented in their kingdom.” A similar suggestion of the perhaps fragile boundaries between the two religions is the fact that Harisena is given the title Haritiputra in the Thalner inscription, dated to his regnal year 3 (463, according to my chronology).20 This somewhat ambiguous title hardly proves that he was a Buddhist, just as the reference in the Thalner inscription (c. 463) that he gave gifts to Brahmins (Mirashi 1982, 78–85), nor the Visrutacarita’s statement that he “(maintained in order) the four castes” (Kale 1966, 349), would hardly prove that he was a Hindu. The Visrutacarita’s various references to the Saivite focus of religious activity in Mahismati, where Harisena’s “second son” was viceroy, and to which his two grandchildren were taken for safety, are more compelling. However, even Maharaja Subandhu (Prince Visruta of the account), whose adventurous conquest of Harisena’s granddaughter took place in various Saivite contexts, proudly takes credit for repairing the magnificent (Buddhist) caves at Bagh. (Mirashi 1955, 19–21) It seems evident that this impressive complex would never have been started without the support of the regional viceroy; and this in turn implies the encouragement of the emperor himself. The Bagh copper plate inscription (approximately 486 C.E.),21 of which I will include only the most relevant portion, is of great interest in showing the ruler’s active support of a Buddhist establishment; and it also revealingly describes much about the arrangement of the monastery and the activities still going on there vigorously, some years after Ajanta’s total collapse.22
20
For the Hariti Shrine in Cave 2 see Volume V, Cave 2. The date on the Bagh plate is lost, but for the dating of the closely related Barwani inscription to the Gupta era rather than (following Mirashi 1955, 17–19) to the Early Kalacuri era. 22 See Volume I for the political benefits of such generosity. See also “Bagh: A Study” (Spink 1976–77, 53–84). 21
a discussion of h. bakker’s the vakatakas
13
“(Line 4) Be it known to you that for the increase of the religious merit of my parents and myself, this village has been granted by me . . . in order that it may be used for (defraying the expenses of ) perfume, frankincense, flowers and offerings as well as for maintaining an alms-house, for repairing broken and rent portions (of the vihara) and for providing the community of Venerable Monks coming from (all) the four quarters, with clothing, food, nursing of the sick, beds, seats as well as medicine in the Monastery called Kalayana (the Abode of Art) to be constructed by Dattataka, as long as the moon, the sun, the oceans, planets, constellations and the earth would endure.” (Mirashi 1955, 19) A.K. Narain, writing at length of the ecumenical attitudes of the period, emphasizes the fact that both “the Gupta kings and most of the contemporary kings and chiefs of India personally belonged to one or the other of Brahmanical religions or sects. . . . It is also true, however, that these kings were liberal in their outlook, and not only tolerated faiths other than their own but patronized them, whether they belonged to the Brahmanical fold or to the non-Brahmanical. (Narain 1983, 48) If we assume that the attitude of Harisena and other Vakataka kings was ecumenical, one is still left with the curious fact that nearly all of the known Vakataka works from the eastern area are Hindu, whereas all of the known works (generally slightly later) from the western area are Buddhist.23 The latter sites would include Ajanta, Bagh, Aurangabad Caves 1, 3, 4A, the Ghatotkacha vihara, the cave at Banoti (except for sixth century revisions), various panels from the stupas in the Kanheri cemetery, and just possibly the cave at Lonad.24 At the very least can we not suppose that Harisena managed to subscribe to both Hinduism and Buddhism, partaking of the political advantages of such ecumenicalism? This was certainly the case with Harisena’s Prime Minister, Varahadeva, who in his Cave 16 inscription “(regarded) the sacred law as his (only companion, . . . being extremely devoted to (the Buddha), the teacher of the world”; while
23 Although not mentioned in Bakker’s study, there are a few Buddhist finds from the eastern Vakataka region, such as the bronze standing Buddhas from Hamlapuri (near Ramtek) and the Phophnar Buddhas. Jamkhedkar (1991, 87–91) dates them variously from the late 4th to early 6th, but precision is difficult with such relatively isolated figures. 24 For a brief discussion of these related sites, see Volume 1, Chapter 14.
14
volume iia, chapter one
in his Ghatotkacha inscription his forbears are praised for “performing religious duties as enjoined by the Vedas and Smritis”.25 The Vakataka feudatory Upendragupta, in his Cave 17 inscription, shows an apparently total commitment to Buddhism, while the close connections of the monk Buddhabhadra, patron of Cave 26, with the Asmaka court, suggests an ecumenical attitude at high levels there also. It is relevant to note that the (mostly) Jain caves at Dharasiva, which are located farther to the south, also appear to belong to the period of Harisena’s reign. They lie in ancient Kuntala which, according to both the Ajanta Cave 16 inscription and the Visrutacarita lay within his sphere of influence or control. Although started by the mid-60s, they appear to have been largely roughed out during Ajanta’s Recession cum Hiatus (469 through 471), when many displaced artisans from the great site would have been available. Their proximity to Asmaka was probably responsible for the decline of activity during the following conflagration. Perhaps for this reason the developed images in their long unfinished shrines reflect those of the latest phase of consistent patronage at Ajanta (i.e. 477–478).26 The influence of Buddhist (specifically Ajanta) precedents on both of these main images, and on the character of the caves in general, is noteworthy. Bakker is certainly correct in seeing the Eastern Vakataka productions as being an influential source for the sculptural work at Ajanta, even though they were at some remove in time and space. But this influence is by and large limited, as far as we now know, to material such as the much earlier finds from Mandhal, most of which were produced well before the mid-fifth century. However, a connection can be seen in the heavy corporeal forms of a number of the site’s earlier sculptures—on the base of the porch doorway of Cave Lower 6, on the façade of Cave 7, on the sculptured ceiling forms of the front aisle of Cave 16, and various other examples, while some of the “small but massive” figures on the fallen façade of Bagh Cave 3 have similar characteristics. But none of these examples can be dated prior to 465 at the earliest, for it took a few years before Ajanta’s caves were ready for any such carved decoration, 25
Ajanta Cave 16 inscription v. 21, and Ghatotkacha inscription v. 6. However, in neither inscription is Varahadeva himself (or his emperor Harisena) said to have “mastered the Vedas.” 26 See Volume I: Dharasiva Caves.
a discussion of h. bakker’s the vakatakas
15
which in the first few years was generally not applied to the monuments until they were essentially roughed out.27 Even certain images from Ramagiri, and the remarkable finds at Mansar, which are indeed linked closely with Ajanta, may show the influence of Ajanta, from around 470 or a bit later, rather than vice versa, as I shall suggest below.28 As for the sculptures from Paunar, they cannot be dated until at least 500 A.D., as Bakker himself agrees, by which time I would ascribe them to some other patronage—probably Visnukundin—as noted above. But here too there is a lingering Vakataka connection.29 Bakker is predisposed to his view of Vakataka history by his assumption—which I shall attempt to counter—that the Eastern (Nandivardana) branch of the Vakatakas ultimately absorbed the Western (Vatsagulma) branch in the fifth century. But even so, he is ready to admit (1997, 44) that the “artistic explosion” at Ajanta cannot be sufficiently explained as being nourished by the “rural small-scale economy” of Nandivardana, even though the Eastern Vakatakas must have contributed something to Ajanta’s development. It was clearly fed by a far wider world, including the inner universe of its insistently elite patronage. To give a sense of the power of this patronage, it is remarkable (if my conclusions are correct) that until the site’s controls literally fell apart in 479–480, not a single lesser devotee ever was allowed to donate a single image at the burgeoning site; and these restrictions extended even to the years between the beginning of 469 and the end of 474, when at first most and then finally all of the caves were temporarily abandoned.30 This elitist focus of Vakataka patronage is the more surprising when we realize that earlier Buddhist undertakings, including Ajanta’s own Hinayana nucleus, were generally community efforts. As the
27
See below. Cave 26 was almost entirely reamed out before any such detailing was done. This may not have been true of every planner; for instance, the carvings on Cave 7’s façade may have been done shortly after it was exposed, although the intended medallions on the ceilings of the porch projections were merely reserved, and therefore never got completed due to the subsequent interruption of work. One, at the far right, was merely outlined in red, while another, in the next “bay”, for which projecting stone had been reserved, also never got completed. 28 See Bakker plates XXXVI A and B; XXXV C. 29 For familial connections with Harisena’s family, see Volume I, In Defense of Dandin. 30 See Time Chart. Clearly all of the temporarily abandoned caves were protected in some way during the Recession and Hiatus.
16
volume iia, chapter one
first great campaign to emulate, and soon dramatically surpass, the impressive cave sites of the Satavahana period, and reflecting so clearly in its paintings “the very world of leisure and affluence that enabled (its) donors to be magnanimous!” (Bakker 1997, 44) Ajanta drew its energy as much from pride and the desire for prestige as from mere piety. Nonetheless the absolute compulsiveness with which these rich donors rushed to complete their images whenever ominous clouds hung over the ravine, and the high praise that they shared with the Buddha in their self-aggrandizing inscriptions, show how deeply their faith functioned as a motive force. At the same time the piety of Ajanta’s patrons was rivalled by the fervor of their desire to outdo both the past and the present—a goal clearly evident in the constant changes in the site’s development. These insistent transformations soon led the site from the diffident productions of the early 460s toward the demanding esthetic with which both the patrons and the artists had been for years familiar, even if they had to overcome the resistance of the unfamiliar rock before it could be properly expressed. It took only a few years of social pressure—for everyone was watching everyone else, concerned both about pay and prestige—to renounce the embarrassing simplicity of the early pillars, doorways, and decorative details, and to straighten out the crooked walls and the angled ceilings of the earlier caves.31 You can tell the earlier caves from the later merely by assessing the alignments of their walls and ceilings and cells. Compare, for instance, Cave 17, excavated in the 460s, with Cave 23, excavated (except for certain façade features) in the 470s. These continually elaborated changes in style and iconography and plans and functional features would quickly turn the more developed caves under development in the late 460s and the 470s into forms which, “clothed in the brilliance of Indra’s crown . . . cannot be even imagined by little-souled men”. (Cave 17 inscription, verse 25) Created to emulate nothing less than the “palaces of the lord of gods” (Cave 16 inscription, verse 27), these “memorials on the mountains” (Cave 26 inscription, verse 8) gradually came to look more and more like the surely luxurious structural mansions of the cities which many of the artisans must have been building all their lives.
31 See Volume I, Cave 4, for discussion of why ceilings and walls are generally out of alignment in the earlier caves. Also Volume V: Cave 4.
a discussion of h. bakker’s the vakatakas
17
Sadly, these beautiful residences are preserved now only in Ajanta’s paintings, as in the startlingly convincing palace scenes in Cave 1’s Janaka Jataka, where the elegance and authority of what once must have been can still be seen today. Indeed, in the whole book of India’s ancient architectural history, only this chapter from the world of the Vakatakas has been so remarkably preserved, in living color, as it were. This dramatic evolution from the “archaic” provincialism of the site in the early 460s to the highly developed urbanity of a decade hence—the seeming equivalent of which would take so many centuries to achieve in the world of Greece—can be explained by the fact that the desired stylistic end was already known from the start. Instead of being slowly searched out, it in fact already existed in the resplendent world to which Ajanta’s courtly patrons must have belonged. From the moment the first Vakataka excavators diffidently faced Ajanta’s recalcitrant scarp, the urgent process of bringing the site up-to-date began, urged on by the highly productive rivalries between the artists eager for the best jobs (and, during times of stress, for any jobs at all), and between their various contending sponsors, eager to best honor the Buddha and to outdo their neighbors with the supremacy of their achievement. The insistent progression toward what in fact was already known can be followed in the treatment of doorways, windows, shrines, and cells, but is most easily revealed in a comparison of the plain octagonal (“Hinayana”) pillars of the early 460s, with the elaborate and multi-facetted high-based columns being carved throughout the site and at Aurangabad a mere half-decade later. Thus little by little, having started so willingly under the influence and aid of the past, within a mere half decade they were already emulating in stone Indra’s celestial mansions—an achievement attained with what may seem to be a staggering swiftness, but in reality was only a conquest of what, as the heirs to a long tradition, they already knew. Faced with Ajanta’s soaring cliffs, and the potential palaces within, their task was ultimately one of translation. However, there is surely no reason to fault what the artists and the patrons achieved with such rapidity. When Bakker (1997, 89) suggests that “the quality of (the Ajanta) sculptures on the whole does not reach the sophisticated subtlety and the fresh originality of that of the Eastern Vakataka kingdom” it is hard to believe that he is thinking of the larger body of Ajanta’s sculptured forms, whether
18
volume iia, chapter one
images or ornamental motifs. He seems here to have been influenced by James Harle, “who (had) a very marvelous store of merit” (Cave 17 inscription verse 22) yet at one point lost a bit of it by calling the Ajanta Buddha images “generally uninspired and rather lumpy”. (Harle 1974, 24)32 The truth is that the lack of “sophisticated subtlety”—if not of “fresh originality”—far more appropriately characterizes the sculptures of the first half of the fifth century in the Eastern Vakataka domains. These are the sculptures that comprise so much of the fascinating material in Bakker’s book. With their compacted clumsiness, and with heads sometimes sprouting from their legs and shoulders and competing for position elsewhere, they do indeed embody a remarkable energy which seems eager for a more subtle expression. Just as Giotto’s ponderosities in some ways help to explain even Botticelli, the forms from Mandhal undoubtedly provide some of the inner forces that are ultimately transformed into Ajanta’s refinements. On the other hand, the characterization of the Mandhal figures, with their links to a folk tradition, can by no means apply to the unique figure of Siva (or a Saivite attendant) from Mansar, which Bakker believes can best be dated to the reign of Pravarasena II and assigns to the first half of the fifth century.33 This is much too early in my opinion, since the beautifully modelled form has such striking connections with various yaksha images of the 470s at Ajanta, as Bakker himself notes. Like Jamkhedkar (1991, 90 and 91, Fig. 10) I would date both this image, along with other superb figures from Mansar, and the stylistically related Bhararaksha figures from Nagardhan to 470 or slightly later.34 Genetically linked to its past— “it can be said to be somewhat in the tradition of the Mandhal images” ( Jamkhedkar 1991, 90)—it has, as if preparing for the future, equally telling connections with some of the fine carvings from the Uccakalpa sites (such as Nachna Kuthara and Bhumara) dating somewhat after Ajanta, as the early heirs of Ajanta’s fall.35 And still fur-
32 Harle 1974, p. 24. It is true that a number of the shrine Buddhas are overly constricted by the outgrown limitations of the central image block, but Bakker’s larger generalization regarding all of the sculptures at the site is surely unjustified. 33 Bakker 1997, Plate XXXVII. Mirashi assigns Pravarasena II’s reign to 422–57; I suggest c. 410–c. 445 (Spink, 1992, “A Revised Vakataka Chronology”. However, I date the image later, as explained herein. 34 Bakker 1997, Plate XXX, A and B. 35 For such glorious “residue” from Ajanta’s fall, see discussion of the situation of the Vakataka fall in Volume I.
a discussion of h. bakker’s the vakatakas
19
ther in the potential so solidly embedded here one can sense the profound authority of Early Kalacuri sculpture, soon to be realized first in the urgent forcefulness of the sculptures at Jogesvari and Mandapeswar, and then in the ultimate refinements of Elephanta. Beyond that, the royal line continues in the beautiful but already doomed refinements of Cave 21 at Ellora, after which and around which the promise is already wearing out, as if waiting for the harsh infusion of a new strength—the taut energy—which will after another century be part of the region’s conquest from the south. The startlingly subtle image from Mansar, one of the still too little known masterworks from that staggering site, instead of suggesting the impact of the East upon the West, as Bakker contends, suggests just the opposite, in my view. If my proposed new Vakataka chronology is correct, Harisena, now having taken over the Nandivardana branch (instead of vice versa) and having unified the empire, came into control of the Eastern Vakataka domains sometime in the mid470s. The same course of influence from West to East—rather than vice versa—can explain the stylistic connections between many Eastern Vakataka forms at Ramagiri and the far more burgeoning developments at Ajanta.36 Just as artists almost surely sought out work at Bagh, Dharasiva, and elsewhere during Ajanta’s Recession and Hiatus (469 through 474: see Time Chart), they surely went to the Nandivardana region as well. Bakker and I differ fundamentally in our interpretation of the implications of the shift of cultural energy between the two parts of the empire. Whereas he and most other scholars see the eastern branch of the dynasty as ultimately transcendent and continuing in power until the end of the fifth century, I have tried to show that by the mid 470s the holdings of Prithivisena II, the last king of the Nandivardana branch, had already been taken over (through either inheritance or conquest) by the Vatsagulma branch of the longdivided family. It was Harisena—not Prithivisena II—who finally unified the once bifurcated Vakataka dynasty, and was responsible for its greatest political, military, and artistic achievements. Both the Visrutacarita, which sees Harisena (= Punyavarman) as the single ruler over vast domains, and the Ajanta Cave 16 inscription (verse 18) which makes it clear that he controlled central India from sea to
36
Bakker Plates XXXV A and B; XXXVI A and B, for Ramagiri.
20
volume iia, chapter one
sea support this point.37 Thus it was under Harisena himself, in the 460s and 470s that the Vakataka imperium achieved its greatest heights—heights from which, only a few years after Harisena’s death, it was to so tragically and finally fall. The implications of this view are as immense as they are conventionally rejected. For if this is true, Harisena was the last great ruler of India’s so-called Golden Age. Directly or indirectly, he controlled the whole of central India from sea to sea, at a time when the vaunted Guptas were already in decline. And Ajanta, in its crescendo of activity, not only mirrors the character of that age at the highest point of its development, but finally and suddenly mirrors its destruction as well, recording, in the stone, the traumatic shock of Harisena’s great empire’s shattering during the reign of his inept successor, Sarvasena III (c. 478–c. 483).38 At this point, pace the history books, India entered an age of darkness from which it did not recover until the various lesser kings whom Harisena had once dominated were able to climb up out of the ruins of the Vakataka imperium and make their own new way into the world. To those who would read Vakataka history in the old way, this sounds offensively speculative, as if it could meet with the approval of Dandin alone. But the evidence of art history demands this new view of Vakataka history and equally explains the remarkable character and course of Ajanta’s patronage. In fact I would contend that art history turns out to be the one ultimately necessary key to our understanding of late Vakataka developments. It is not art history, with its plethora of data, but the unrooted extension of convention, which is speculative. Surprising as it may seem, one can not possibly make sense of the political and military situation during and just after Harisena’s reign, nor properly test the validity of the new Vakataka chronology, without applying the disciplines of art historical
37 Although past scholars have generally supplied the missing verb in the inscription as “conquered”, Bakker (1997, 35) suggests “stands above”; this seems preferable, since Harisena must have developed his empire through marriage and political alliances or inheritance (of Prithivisena’s domains) as well as war. However, it should be noted that Bakker uses “stands above” in the sense of “outdoing or putting to shame all the surrounding kings”, rather than, as I would suggest, transcending them and quite literally incorporating (at least to some degree) their authority by his own. For fuller discussion, see Volume I. 38 See Spink 1991, pp. 71–99; see also discussion of Sarvasena’s accession and situation in Volume I: Chapter 4 In Defense of Dandin.
a discussion of h. bakker’s the vakatakas
21
analysis. To do this we must turn, leaving our preconceptions behind, very specifically to the Vakataka donations at Ajanta, Bagh, Banoti, Ghatotkacha, Aurangabad Caves 1, 3, 4A, and Dharasiva; for it may well be deep in those caves that the answers lie.
CHAPTER TWO
CAVE 26 AS AN INAUGURAL MONUMENT
Nowhere are such investigations more relevant than for reaching an understanding of the growth of the vast Cave 26 complex, comprising the main caitya hall and its four wings. Bakker (1997, 41) rightly points out that the monk Buddhabhadra’s dedicatory inscription, with its fulsome praise of the Asmaka rulers and ministers, shows “no word of Harisena’s overlordship nor of any other ruler apart from that of Asmaka . . .”. In his view (1997, 41), this “strongly suggests that this complex was excavated after Asmakas had gained full control over the region, i.e. after Harisena’s death.” It is with this seemingly reasonable conclusion that I must so strongly disagree. It is certainly true that the Vakatakas are significantly omitted in the Cave 26 record. I would also agree that the Asmakas had gained full control over the region at the time that it was written. Furthermore, “as far as historical and epigraphical evidence goes” (Bakker 1997, 41) one can agree that it would be hard to locate the cave in relation to the Vakataka developments at the site. But art history can and does locate Cave 26 convincingly and solidly in the mainstream of the site’s Vakataka patronage. It explicates, in its stage by stage development, the crucial (and often disruptive) role that the Asmaka feudatories played from the very start in the vagaries of Ajanta’s patronage, which finally ended with the abandonment of the cave, the disruption of the site, and ultimately the destruction of the whole empire. The Asmakas, as feudatories, were as involved in the site’s development as they were implicated in the ominous course of late Vakataka history; and Cave 26, along with its inscription, far from denying this involvement, fundamentally helps to explain it. By the time that the Cave 26 dedicatory inscription was written, in mid-478, the great caitya hall complex was (not surprisingly) essentially done; and the Vakatakas were essentially done for. The inscription in effect predicts Ajanta’s collapse—a collapse that was almost inevitable as soon as the aggressive Asmakas so rapidly asserted their control over the site shortly after Harisena’s death in 477. Indeed, it is in part the evidence of this inscription and its revealing art historical
cave 26 as an inaugural monument
23
context that limits the site’s patronage activity to less than twenty years, for it defines Ajanta’s doom, by writing—between the lines— that the great emperor was dead. The evidence that clearly proves that Cave 26 was not created after Harisena’s death, as Bakker hypothesizes, but was in fact one of the site’s initial undertakings (being started in 462), is by and large obscured by its lavish overlay of late (475–478) decoration and by the many changes in plan made during the course of its excavation. But the evidence is surely there, for the “cover-up” is by no means total. We can list some of the complex’s indisputably early features here to show that it was one of Ajanta’s inaugural undertakings. In fact, with its four wings, it was the most ambitious one of all. Furthermore, as will be shown later, all of the other caves in the site’s western extremity (i.e. Caves 21–24, and 28) were also the product of Asmaka patronage; and, perhaps remarkably, their development appears to have been under the control of Buddhabhadra himself, as I shall try to show in this chapter and the next. It should be mentioned that these early features in the Cave 26 complex are often obscured, masked by changes made to keep up with the times, or to improve functional features, or to consciously cover forms that were troublingly obsolete.1 As in archaeology, one often has to “dig” down through various layers of change, to uncover the past; and whereas something of the process and of the “finds” can be detailed here, in the end one must go to the site for a number of “seasons”, in order to properly “excavate” the caves. After providing evidence here that the Cave 26 complex is, “at the core” one of the very earliest undertakings at Ajanta, I shall then provide a resume of its development (and will explain its long period of abandonment) during the troubled course of the site’s history. It might only be mentioned that the whole site follows a similar pattern, wherein the development of the caves is controlled by political realities. In nearly every case (other than the very late and often minor undertakings Caves 3, 14, 22, 23A, 24A, and 28) all of the various caves were started very early, and then were either totally
1 Most earlier caves were blocked out before any decoration started. The Cave 26 complex is a striking example; when it was first abandoned in 468, it had not a bit of imagery or decorative carving on it. In striking contrast, note the more sophisticated development of later caves, such as Cave 24.
24
volume iia, chapter two
interrupted (i.e., Caves 17–20; 29) or (in the case of all of the other caves) suffered a period of abandonment before work on them was eventually taken up again, in their patrons’ ultimately abortive attempts to complete them. (See Time Chart in Volume IV) Thus Ajanta’s Vakataka development, throughout what might, had tradition not wrought its havoc, be called its lengthy history of nearly twenty years (instead of half that time), is a history of crises. And these crises are particularly revealing; for it is at these points in time that we find breaks in development, as artists and planners and patrons respond differently to suddenly changing times. When a city is burned down, or flooded, or bombed, the pre-crisis evidence and the post-crisis evidence tell a story that hardly need be written down. And it is the same with a site such as Ajanta, or indeed the other related Vakataka sites. When the Asmakas are ejected from the site, and then when they in turn eject the local king; or when the great emperor Harisena, “a partial incarnation, as it were, of the God of Justice” (Kale 1966, 349) sent the empire into shock by his untimely death; or when, as a consequence, the region is direly threatened; and then when even the helter-skelter patronage at the dying site collapses: these events leave their marks deep in the stone, and those marks can eventually be seen to be the letters and the words with which a dramatic story can be eventually composed. And these events are written not only—or even so much—in the grandest of the caves; they are even more poignantly exposed in the tragic mini-histories of those caves which were abruptly abandoned, or had to struggle (leaving evidence of their difficult passage) over the flaws and gaps in the site’s development to finally reach their patrons’ obsessive goal.2 This goal was of course, and particularly by this time, the achievement of the merit that would be theirs, if only, in the developing storm, they could get their shrine Buddhas brought, in both the inner and the outer sense, to realization.
2 Such gaps or traumas can affect the course of a scholar’s research also, and often the positive benefit from such events is not immediately realized. A personal reflection may be of interest. Some decades ago, at the unfortunate instance of a suggestion from a high foreign official in Delhi who thought the major painted caves were in jeopardy, those important caves (1, 2, 16, 17) were completely closed for conservation, only visitors of some political significance, along with staff (and occasionally their families) being allowed in while the work, unhappily and insufficiently it turns out, proceeded. I, as a sufficiently young and interested scholar was banned
cave 26 as an inaugural monument
25
Cave 26’s Early Features: An Analysis If one agrees with Bakker (1997, 41) and many other scholars that “Cave XXVI . . . may be of a later date” than many or most of the other caves at the site, then my own view of the cave’s development must be incorrect. And if this is incorrect, the whole “Short Chronology” must also be wrong. For this reason a proper understanding of Cave 26’s development—a justification of its position as an inaugural monument, as well one whose history ultimately speaks of the fall of empire—is essential. K. Khandalavala (1991, 102) expressed the case for the opposition most decisively. “The inscription in Cave 26 is a bed-rock of any sensible chronology of the Ajanta Caves. . . . All the far-fetched attempts, with no evidence in support, which have been made by Dr. Spink to surmount what the inscription in Cave 26 states can only be regarded as exercises in sheer futility for no amount of ingenuity can ever displace the fact that Cave 26 could only have been excavated after the fall of the Vakataka dominion.” To better explain my own position, I will make a listing of some of the Cave 26 complex’s never-noticed early forms and features, and link them with the site’s overall development. As a result of this we will be better able to reconstruct the remarkable history of the whole site which the complicated development of Buddhabhadra’s cave so tantalizingly mirrors.
from the caves for that same period, despite the strenuous efforts of the then DGA M.N. Deshpande, widely known as the doyen of cave studies, to grant me access. For nearly ten years, his pleas to higher authorities, which one wishes to believe pained him, were of no avail, despite the innocence, other than scholarly, of my involvements. Thus, since I of course had no intention of giving up my work at the site, I was forced to study the many unfinished and broken caves—the very ones which P. Stern, in his faulted “Les Colonnes indiennes d’Ajanta . . .” excludes from consideration because they are broken and unfinished. But forced to undertake my work in these less studied areas, I eventually discovered that here—and not in the more fully finished caves—was where the most revealing secrets lay. This was because, due to their variously abandoned states, they far more clearly showed the telling breaks in the site’s development; and it was by an analysis of these breaks and their significance that I first constructed the stage by stage pattern of development which underlies my conclusions still today. I hardly think that thanks are necessary for the productive pain which I occasioned during this period, but it is intriguing, if ironic, that, both sadly and happily, my presumably necessary, or at least instructive, exclusion had such a positive conclusion.
26
volume iia, chapter two Astylar Plans
The astylar plan of Cave 26’s four wings is revealing as a notably early feature, showing the early dependence on Hinayana prototypes at the site.3 As in most Hinayana caves, each wing’s central hall, as planned, was to be surrounded by monks’ cells on left, rear, and right; but this notably simple arrangement, representing the plan of all four of Cave 26’s wings, undergoes major later adjustments in all cases, as conventions at the site develop. Cave 8 and Cave 15, both inaugural excavations, are also astylar, whereas both Cave Lower 6 and Cave 11 were similarly planned, but were found to require interior pillars to hold up their ceilings.4
Porch End Development All of Cave 26’s wings were conceived without porch end cells, a retardatory (Hinayana) feature found throughout the site but wisely abandoned after 465 in order to add cells at these convenient locations in all caves. Beginning in 465 and 466, simple single cells were cut wherever possible in these previously “wasted” areas, rapidly becoming conventional (and invariable) features for the next couple of years. In 467 cells with pillared vestibules come (also invariably) into fashion, replacing the single cells whenever possible, just as the single cells, whenever possible, had been cut into the still-earlier plain end walls of the porches. The surprising adjustments made to Cave 26’s left wing, under the dictates of rapidly changing taste, show this process of transformation. Originally this wing must have been planned to precisely mirror its counterpart on the right, preserving the careful and conventional symmetry of the whole complex’s façade arrangement. (See “original” plan of Cave 26 complex) But as we can see in the earlier-
3 In fact Hinayana viharas in general are astylar. M.K. Dhavalikar has suggested that Ajanta’s Cave 15 is early fifth century for this reason. Dhavalikar (1981, 136; also 1970–71). However, the wings of the Cave 26 complex are also astylar and, along with the plans of Cave 8 and Cave 20, show that this feature is commonly found early in Harisena’s reign, in the 460s. 4 For the added pillars in Caves Lower 6 and 11, see Volume V, in the discussion of those caves.
cave 26 as an inaugural monument
27
defined right wing, this involved a porch too shallow to allow the addition of porch-end cells, when (in 466, or even 465) such cells came into fashion; and in any case, the excavators had already cut a cistern at the right end of the right wing’s little porch. Such a positioning for the cistern would never have been done if a cell had been intended for this location. The cistern could easily have been cut in some area where it would not take up this now-valuable space. By the same token, there was no possibility of putting a cell at the other (left) end, an area where the needs of the abutting porch end of the main hall surely had priority. Single cells must have been planned for (and perhaps even begun in) these porch-ends of the main cave by 465 or 466, but if so they were being converted to the more complex and suddenly fashionable pillared cell-complexes by 467 or 468. At this point, like the rest of the cave, they would have been only roughly defined by the time of the Recession, being completed with their elaborate fronts only in the later phase of work on the cave, starting in 475. (See “Defining Features” chart in Volume IV.) Since the lower left wing of the caitya complex had not been started as early as the right, in 465 or 466 the caitya complex’s planners, to accord with new conventions, made the decision to shift the still not fully revealed court of the caitya hall farther to the left (west). This was done in order to make room in the set-back porch of the left wing (Cave 26LW) for a single cell at either end—specifically (even if surprisingly) to accord with the always—demanding new conventions. But in 467/468 the still merely (if at all)roughed-out right porch cell of Cave 26LW was converted—again to accord with changing conventions—into the new “cell complex”, in which a pillared vestibule fronts the residence cell behind.5 That the vestibule of the complex pillared cell at the right of the porch of Cave 26LW was previously under excavation as a cell is revealed by its uncharacteristically deep dimensions, and by the equally uncharacteristic thinness of the fronting pillars. These thin pillars, now long since broken away, were, in fact, cut out of the previously defined front wall of the now-converted cell! The present cement restorations are somewhat thicker than they should be; this was an understandable mistake on the part of the restorers, who did not understand that
5
The left porch cell was apparently cut, but has since fallen into the ravine.
28
volume iia, chapter two
they had originally been shaped from the typically thin early walls of the simple cell, and did not look to the still-intact pilasters for guidance. Such a fundamental change, from plain porch end, to the addition of single cells, and then to pillared cell complexes, all taking place during the course of a few years of excavation, provides insight into the flexibility of the planners as well as the force of the rule of taste, or of convention, at the site. There are a number of telling examples showing that such changes, if possible, were considered as conventional. The complex cell L1 in Cave Upper 6, and the cell at porch right (PR) in Cave 27 were both started as simple cells before the Recession (in fact prior to 467), and then “upgraded” to complex types after 475, when work on them finally continued. Indeed, the present porch-end complexes of Cave 26 itself were first intended as single cells; otherwise Cell PR of the Left Wing could never have has its present format. Indeed, the present vestibules (celllike in their dimensions rather than more conventionally shallow) must have been “reserved” while more crucial work went on in the great cave. Then, when they were finally cut, they were turned into pillared complexes. The very thickness of their fronting pillars would seem to prove that earlier penetration was never made, and that what we see today are pillars created after 475, and (both on the left and on the right) “extra” cells now added for extra residence space, at that same time. The door fittings, at this late date, would of course be D mode, even though throughout most of the wings the cells were penetrated earlier, in the A mode, and then converted, after 475, to the practical later type of fitting. The conversion of Cell R1 in Cave Upper 6 presents a special case since, due to lack of time, it remained in its pre-Recession state until an intrusive donor took it over in 479, and turned it into his own personal shrine. Although surely, as in its counterpart Cell L1, its front (originally intended for a simple doorway to a single cell) had been widened in preparation for the definition of the new (never completed) vestibule pillars, the “intruder” merely used this space as an unconventionally wide entrance to his new shrine, at the same time cutting his sacred image into the back wall of the previously intended residence cell; something very similar occurred in Cave 6’s right rear cell, with its intrusive image; it had been started, probably in 468, as another pillared complex, which never got completed by the original patron even after 475.
cave 26 as an inaugural monument
29
Although Caves 21 and 23 now have typically late pillared cell complexes in their porches, we can tell from the too-narrow area of scarp between them that single cells must have been originally planned here when the two caves were first laid out—the earlier (Cave 21) in c. 465 or c. 466.6 Hardly a year later, because the deeper cell complexes had just come into fashion, the space between Caves 23 and 24 had to be very significantly adjusted—a phenomenon of great interest, to be discussed later. Pillars Sitting Directly on Floor Although the standard convention which was developed at Ajanta in the Vakataka period was to support the pillars in the cave with a monolithic floor beam, over which visitors often trip today, it would appear that in the earliest excavations the pillars were set directly on the floor, just as they were in the Hinayana caitya halls, which had such a strong influence on the earliest Vakataka excavations. This “primitive” characteristic—paralleling the cutting of the earliest cell doorways without planning for door-fittings—is notably evident in the multipillared Cave Lower 6, even though at the upper level the pillars “support” the beams which in turn “support” the ceiling. Cave 11’s pillars were also at first planned without bases, even though complex bases were added when the old dormitory was converted to a shrine. Although the use of supporting floor beams came into use very early, and of course persisted, they are significantly absent in the very early Right Wing of the Cave 26 complex, which can be seen as yet another confirmation of the very early date at which it was undertaken. By contrast, the Left Wing, under excavation a few years later, shows the beams under the porch pillars. The very distinction between such features in the two wings—for they were originally planned as parallel excavations—reveals the rapid developmental changes which the whole complex was undergoing.
6 It is very unlikely that Cave 21 was underway prior to 465, prior to the time when porch end cells were in vogue; but even if this was the case, the space requirement between the two caves would have been the same because the area required by interior cells would be equivalent to that taken up by single cells at the porch ends.
30
volume iia, chapter two
Cave 25, the upper right wing of the Cave 26 complex, was probably started before 26RW (the lower right wing), for work would have progressed generally downward (as well as from front to rear) when the whole complex was being exposed. It almost certainly was planned in the simplest manner, with its porch pillars connecting directly with the floor. Indeed, that is what we see today, even though in 478 much of the long abandoned and unfinished porch was recut when, apparently, it was being expediently converted into an anomalous shrine area.7 We can see that the pillars, originally planned (and only roughed out) with heavy octagonal shafts, as would have been expected in the early 460s, were later being recut in a more modern style when all Asmaka patronage was interrupted late in 478. The floor between them, though surely still rough when these late revisions started, probably had already been enough defined, in the early phase of work on the cave, so that it was not practical to now add a more up-to-date floor beam. It seems to be true, also, from old photographs, that the front pillars of the two porticoes of Cave 7—one of the earliest excavations started—had no “supporting” monolithic base-beams; the fact that they were reconstructed recently in this way would seem to confirm this, since reconstructions generally (or at least ideally) respect the traces of old forms. However, more rearward pillars do have beams, probably because they were exposed slightly later, and thus conform to the new convention. It is not surprising for conventions to change very rapidly.8 From Shrineless to Shrines Just as in all of the caves at the site which were underway during the first four years or so of activity, the very early four wings of the Cave 26 complex were planned without any image shrines, again following Hinayana precedents. These shrines, which in all of the earliest Vakataka caves such as these were always added features, consequently were not without problems of placement; we will later 7
See Volume I, Chapter 12. The porch end complexes of Cave 7—perhaps the first created at the site and dating to 467—also have pillars with no supporting beams beneath, even though such beams are conventional later on. 8
cave 26 as an inaugural monument
31
discuss the adjustments which had to be made to accommodate such previously unplanned-for images.9 The reason that the shrine Buddhas did not choose to take up “residence” at the site until 466, by which time nearly all of the site’s Vakataka viharas were already underway, is still unexplained, and should be a challenge to Buddhologists.10 If the viharas at a site as important as Ajanta were originally planned without shrines—that is, as mere dormitories—does this reflect the contemporary situation in structural viharas elsewhere as well? Was this crucial feature in fact one of Ajanta’s innovations? Although shrines were planned for by 466, at least in the major caves, it seems likely that the first separate Buddha images were not finished and dedicated until sometime during the first half of 469. All four of the viharas in which we find the earliest Buddhas at the site—Caves 6L, 7, 11, 15—had been penetrated to varying degrees in 468. This surely reflected the revolutionary changes taking place in the important Vakataka caitya halls, Cave 19 and 26, where the idea of fronting the stupas with images had already developed, even though they would not actually be carved until slightly later than the Buddha images in the viharas Lower 6, 7, 11, and 15, all finished and dedicated in early 469. The image in Cave 19 was probably revealed by 469, whereas, its carving interrupted by the Recession, the Cave 26 image was not revealed until 477. The fact that a stupa (soon abandoned) was started before the image was carved in Cave 11, would further seem to reflect the transition seen in the caitya halls.11 The reason that we know that shrine images had to be planned (at least in the major caves) much earlier than they were actually carved is from the widening of the intercolumniation between the central pillars of the front aisle, to provide an axial focus for the shrine. However, such widening never appears in the earliest excavations which have pillared halls, namely Caves 2, Lower 6 and Upper 6. This shows that, like the various astylar halls, they were originally planned (always before 466) with no thought at all of shrines. 9
See Volume I, Chapter 12 for comments on the four wings of the Cave 26 complex. 10 Schopen (1990, 181–217) discusses the Buddha as “resident” in the caves, but does not explain why the conception did not appear at the site until about 466. 11 See Volume V: Cave 11; the abandoned stupa appears to have begun, or at least conceived, in 467.
32
volume iia, chapter two
Since many of Ajanta’s earliest excavations (8, 15, 20, 25, 26RW, 26LW, 27) are astylar, they provide no evidence in this regard, while the pillars in the small Cave 11 were created as ceiling-supports some years before the shrine was even conceived. Although much of Cave U6’s interior was excavated after 475, the front aisle colonnade was apparently defined (thus setting the pattern) very early. The nearby cells, L1 and R1, which would have been started somewhat later, were almost certainly in the A mode and thus were not defined after 467. The present windows, for instance, like those in Cave 5, are late (unfinished) conversions from very early vertical types like those of Caves 15 and 20. The idea of adding a shrine to the viharas was probably one of the many concepts transmitted to Ajanta from Bagh, where the friable nature of the relatively low sandstone scarp apparently disallowed the making of the expected caitya hall. Therefore, although the earliest of the Bagh viharas, Cave 2, (which has no axial focus) had apparently been started as a mere “dormitory”, it could be converted to a “caitya hall” by the addition of a shrine at the rear. And being the very first such converted vihara, at least in central India, it is hardly surprising that the “resident” caitya was a stupa, not a Buddha image, following the long established convention for halls of worship.12 Shortly thereafter, at Ajanta, when the idea of adding shrines— probably at first only to the largest caves—developed, it was probably also stupas, not Buddhas, that were originally intended for them. Of course, in these privileged major caves, such as 4, 16, 17, and the small (but royal!) Cave 20, the original plans (without shrines) had to be adjusted to provide for this important (indeed revolutionary) new feature. In these larger and more important caves the new conception must have developed by no later than 466, before the pillars of the front aisle had been started, or while they were still so rough that their central intercolumniations could be widened, as new conventions required.13 Only Cave 1, begun much later (c. 466) could be excavated as planned, at least in this regard. But it too may well have been first conceived with a stupa as its focus,
12 See Spink 1976–77, 53–82. Therein I considered Bagh Cave 4 as earlier than Cave 2, which I now feel is not correct. See also Volume I, Chapter 3. 13 This is discussed at length in Volume V: Caves 16, 17, etc.
cave 26 as an inaugural monument
33
a year or more before the idea of the image took precedence. Indeed, its image was not even started until 475, although its shrine had been penetrated earlier. In the larger and more important viharas, such as Caves 1, 16, 17, and 4, the plans would have been adjusted well before the excavators had reached the shrine area; and when, in the course of excavation, they were starting on the shrine itself, Buddhas instead of stupas had become the “caitya” of choice. So their past has been obscured: indeed, were it not for the “obsolete” central blocks from which all of the earliest ones were cut (or were intended to be cut) we would not know, seeing them today, that their Buddha shrines “replace” once-intended stupa shrines, and that the stupa shrines themselves would have been cut where residence cells were once intended. King Upendragupta’s beautiful Cave 20 seems to be the single exception to the “rule” that the smaller excavations were somewhat slower in developing the shrine concept. Probably because of its “priority” royal patronage, Cave 20 was the first small cave to have a shrine added to its originally simpler conception. We know that its shrine area (if not the shrine proper) was planned very early— in 466 or 467—because its shrine antechamber projects out into the astylar hall, in order to support a bad ceiling flaw; and the hall contains many very early features, the most striking being its “primitive” doorways, doorsteps, and door fittings. However, its position of privilege did not alleviate its obvious financial difficulties as the Recession progressed, for both its image and its shrine doorway had to be rushed to a most expedient completion when Upendragupta’s troubles, increasing through 469 and 470, caused him to have to abandon his connection with the site late in 471.14 In lesser caves (other than the royal Cave 20) it would appear that it took two or three years for the idea of adding a shrine to have its impact, for even though in the smaller caves it took less time to reach the rear of the cave in the course of excavation, shrines were never begun in them either until late 467 or (more probably) 468. Perhaps at first, the addition of shrines was reserved only for the more prestigious patrons and their caves; although the idea soon became universal.
14
See Volume V: Cave 20 for the anomalous projecting antechamber.
34
volume iia, chapter two
The earliest cave in which an image was actually completed may be Cave 11, where the shrine was expediently made from a converted cell; significantly, it contains a monolithic stupa fronted by a Buddha image. However, the Buddha image, which was probably not even conceived when the stupa was started, soon become the sole focus, and the stupa is paid no heed whatsoever. Not only does the Buddha block the view of the stupa—the halo, in particular, seems to have been made particularly large just for this purpose— but the Buddha alone was worked on and carefully finished in 469. It seems likely that the combined stupa/Buddha, before the “victory” of the latter, had been conceived following the lead of the influential plan for a similar combination in Buddhabhadra’s great hall of worship where, for the first time in any such caitya hall, an image was to front the stupa.15 Of course, for reasons of ritual and tradition, one could hardly “abandon” the stupas in the caitya halls, as was so abruptly done with the stupa in Cave 11. However, the developments both in the shrine of Cave 11 and in the purely Buddha oriented shrines of Caves Lower 6, 7, and 15 reveal the irrelevance of the stupa in the vihara setting. Indeed, the same could be said for the ritual of circumambulation, for in spite of what scholars generally assume, there is not the slightest reason to think that the ritual of circumambulation was ever practiced or even intended in any of the Vakatakas’ viharas.16 With the interest in stupas as a focus gone, the patrons and planners of the viharas were increasingly burdened by the outmoded convention of the “centralized” shrine format. This made sense, indeed was necessary, when the shrines were conceived for stupas, but when the stupa concept (never truly realized) was rejected in favor of Buddha images the old format imposed unacceptable restraints, particularly as the Buddha groupings expanded. This is painfully evident in the constrictions imposed by the old format on the developed image groups of Caves 1 and 4.17 The ultimate goal was to spread out the image group against the rear wall, as in the
15 For the reasons that the shrine of Cave 11 was probably the first and earliest started, but at first with a stupa, see Volume V: Cave 11. 16 For circumambulation, see Volume I, Chapter 15. 17 See Volume V: Cave 1, Cave 4.
cave 26 as an inaugural monument
35
anomalous Cave 20; but this ideal was not “legitimately” achieved until after 475, in the shrines of Caves 2, 21, Upper 6, and 26L.18 At the same time, it is logical to believe, that just as in Bagh Cave 2 and probably Bagh Cave 4, the patrons and planners first conceived of their shrines as having stupas only. Indeed, it may well be the case that when the Vakataka caitya halls Caves 26 and 19 were first planned—very much under the influence of the Hinayana Caves 10 and 9 respectively—that they too were going to have stupas alone as their focus, and that the idea of fronting Buddha images only developed, both at Ajanta and Bagh, as excitement mounted at the sites, and as new influences, including the desire for image worship, flooded in. As already noted, the first vihara shrine undertaken, in Cave 11, with its unfinished stupa fronted by the completed Buddha image, apparently appropriated the same already-conceived combination. But the stupa was abandoned, barely half finished, probably by 468, and by early 469, when it was hastily dedicated, its Buddha image, like those in Caves Lower 6, 7, and 15, had become the sole focus. The insistent switch to Buddha images surely reflects a trend toward image worship throughout the subcontinent in this general period. But in the case of these first shrines, forced by circumstance (in early 469) to be expediently rushed to the point where their Buddha images could be dedicated, time and money may have played some part in the decision to focus on the images alone. However, stupas had long been recommended by tradition and convention; and the fact that, in early 469, the Prime Minister put a relief stupa as a kind of substitute altar in his Ghatotkacha vihara, when he felt constrained to discontinue the excavation of the still incomplete cave in 468, suggests that the preferences were still somewhat in balance, as they so clearly were in the shrine of Cave 11. But by 468, the Buddha image had clearly won out, being the sole image made in the shrines of Lower 6, 7, 15 (and now in Cave 11)—even though they surely had all been originally conceived for stupas, with a characteristic central placement.19
18
For Cave 20 see Ch. 13. The seated Buddha on the Ghatotkacha stupa is a later intrusion, while the supporting yaksha below was carved along with the stupa. 19
36
volume iia, chapter two
Similarly, as we shall see, the development of the plans in the lower wings of Cave 26 proves that by 468 (after which these Asmaka caves had to be abandoned for the next six years) these viharas were being adjusted (26RW) or extended (26LW) to include shrines. Needless to say these caves—originally mere dormitories—had not been planned with shrines in mind, a fact which tested the architect’s creativity now that he had to add them. These newly introduced shrines may well have been conceived for stupas in 468, but at least in Cave 26RW and 26LW the places were still in the process of being prepared for them, when the Asmakas were ousted.20 Then, by the time these lower wings were finally completed, in 478, they were provided with typically developed Buddha images, just as we would expect. Since both of the complex’s upper wings (Cave 25 and 27) had probably been at least temporarily abandoned in 466, it is hardly surprising that early shrines were never started in them; the shrines later planned (in 478, when time was quickly running out) were, understandably, never finished. The discussion above only tells us how and when shrines, and ultimately shrine images, first made their appearance; it does not tell us why this remarkable development took place or why the eighteen mere dormitories already underway before 466, were suddenly turned into places of worship.21 Judging from the spectrum of the site’s different painting styles, as well as its rich body of sculptural and architectural motifs, Ajanta drew its forms and features, and its ideas, from many different parts of the Indian subcontinent. In this regard the established Buddhist sites in both the northwest and the southeast deserve special attention, both being linked to Ajanta by the routes of trade as well as by their strong Buddhist traditions. For instance, the bhadrasana pose of the Buddha, perhaps unknown in central India and wider regions beyond until its appearance at Ajanta, appears much earlier in Andhradesa, while the frontal projection framing the Buddha in the Ajanta caityas 19 and 26 may well reflect the “false gables” common in Gandhara.22 If the idea of putting
20
Volume I, Chapter 12. This excludes Cave 1, started only in 466. Since the courtyards of Caves 23 and 24 were already started at this same time, it is possible that they were planned as “mere dormitories”; but by the time that their elaborate porches were reached, they were certainly conceived as shrines. 22 Kurt Behrendt, who has made a detailed study of Gandhara material, writes 21
cave 26 as an inaugural monument
37
images (or even stupas) in viharas was current in the wider world at this time, would we not find shrines in the viharas at Ajanta and Bagh from the start? Is it possible that this revolutionary concept was first developed at the Vakataka caves themselves? If not, then how do we explain it?
Excavations Procedures When Cave 26 was first under excavation, the vault, as expected, was revealed first, since excavation logically proceeded downward. However, at this early date, the excavators failed to reserve matrix for the upper levels of the stupa, so that it was necessary to attach the umbrellas (now missing) separately. This was hardly very practical, and is in clear contrast to the technically more sophisticated— and later—treatment of the stupa in Cave 19, where the umbrellas were cut from matrix which had been sensibly reserved. This was also to be the case in the significantly later (469) Caitya Cave 29, where the excavation of the unfinished vault has been stopped, surely by intention, at the point where the stupa was to be carved; had work continued, the upper portions of the stupa could have been monolithic, following the later and more sophisticated mode. Another early feature seen in Buddhabhadra’s Cave 26 complex is the still diffident mode of excavation of the entire complex, which was wholly blocked out before any decoration was started. This is notably evident in various early but very incomplete caves. The very unfinished Cave 26’s upper right wing (Cave 25) shows the process very clearly, despite some obscuring late additions and adjustments made in 478. Its modest main hall was being totally shaped before any cells, even those toward the front, got started; and while the excavation was being opened up not a bit of imagery or decorative carving was added to it. In fact this was true of the whole great Cave 26 complex which, when its development was forcibly interrupted at the end of 468, had not a single such motif on any of its
( Jan. 2003) “I agree with you that the Buddha images attached to the stupas in Ajanta caves 19 and 26 related to the false gable pattern popular in Gandhara”. It seems reasonable to assume that the enframed Gandhara images anticipate rather than reflect those at Ajanta.
38
volume iia, chapter two
surfaces.23 The treatment of the Prime Minister Varahadeva’s Cave 16 appears to have been similar. Although it was among the first caves started (c. 462), and had been deeply penetrated (though hardly finished) a half-decade later, the porch doorway’s carved decoration had been but beautifully completed in 468 just before the Recession began, while work on the elaborately carved front aisle ceiling was also going on at that time. By way of contrast, the painting of Cave 16’s porch doorway, and of the area of the ceiling directly above show every sign of rush, characteristic of work at the beginning of the Recession in 469.24 By way of contrast, in Buddhabhadra’s Cave 24, which had only been barely penetrated prior to 475, the cells were being cut sequentially from front to back, as the interior was gradually opened up (see Cave 24 Plan), a very different process from that seen in the much earlier Cave 25. Although Cave 24’s great hall remains largely unfinished, the pillars of its front aisle are in various stages of completion and decoration, and its nearly completed porch is already richly detailed. Even in Upendragupta’s aborted Caitya Cave 29, started in 469, we find that certain façade and vault details were being defined even while excavation was in its initial stages—a more developed system than appears to have been used earlier, when Cave 26 was initially underway.25
Door Fittings Monolithic pivot projections—which make their first appearance in the Vakataka caves in 468—are found in the very last cell doorways cut in the Cave 26 complex before work broke off with the Recession. The projection in the doorway of the left cell (L1) of Cave 26RW gives striking proof that this little cave, like the great hall to which it is attached, was well underway when the Asmaka patrons were ousted from the site late in 468. As we might expect, cell R1, although mostly broken away, also has a B mode doorway; and like L1 it was converted to the D mode when work was renewed on the long-
23 24 25
This is true of pre-Recession work on Cave 21 and thus on all Asmaka caves. See Volume V: Cave 16. Volume V: Cave 29.
cave 26 as an inaugural monument
39
abandoned complex in 475. Early projecting fittings, paired for double-doors, were also used for the Cave 26’s main doorway, just as they were for early shrine doorways and cistern-chambers throughout the site.26 In fact, since Cave 26 had been largely roughed out by the Recession, it is reasonable to assume that its monolithic doorway projections had already been at least shaped by 466. This might well suggest that this great doorway was the ultimate source at the site of the B mode type, which appears in cells in 468 and 469.27 Most cells in the Cave 26 complex, having been roughed out prior to 468, had primitive “A mode” doorways; but this is not immediately evident, since all of them (like the B mode doorways in the Right Wing) were converted to the more efficient D mode after the renewal of work on the complex.28 However, the cells in the Cave 26 complex typically have thin walls, and this reveals the early (468 or before) date of their cutting, despite their conversion. Cells with original (non-converted) D mode doorways conventionally have very thick walls; none are found in the Cave 26 complex except in a few cases in Cave 27, where the cells had not been penetrated (or only barely so) before the Recession. Their thickness probably resulted from conventions transmitted from Bagh with the return of many workmen after 475. There the thickness was required by the friable nature of the sandstone, but it also was particularly suited for the insertion of the D mode pivot holes in a deep recess at the rear of the doorway. The problem of converting early doorways is evident in the conversion of the particularly thin-walled B mode doorway of Cave 26RW’s to the D mode after 475. The added recess is minimal, and necessarily uses the obsolete projection as a strengthening element; the same difficulties are seen in the relatively early and thin-walled doorways of Cave 1 where, apparently for esthetic reasons, the planner
26 Perhaps surprisingly, such projections are occasionally used in the doorways of shrinelets (as in Cave Upper 6) in the Period of Disruption (479–480). 27 Such pivot holders were surely not the “invention” of Ajanta; even today one can find them in various buildings; but of course not in a monolithic variant. Indeed, they were used throughout the Hinayana caves, but were apparently unnoticed(!?); in any case they did not provide an immediate source for those in the Vakataka caves. 28 As noted above, the doorways in the porch-end cells of Cave 26 were probably “original” D mode types, since the pillared cell complexes were not excavated until after 475, even though laid out prior to the Recession.
40
volume iia, chapter two
tried to preserve as much depth as possible in the doorways’ surrounds. We might also note how, considering the relative size of the caves, the B mode main doorway of Cave 26 is so much thinner than the significantly later doorway of Cave 19.29 The excavated niche in Cave 26RW’s Cell L1 would have been cut when that cell’s doorway was converted to the D mode; and it would further suggest that that conversion took place very shortly after the Asmaka restoration, since such niches, typically cut when the associated doorways were fitted out, go out of fashion after 475. There is a similar niche at the rear of the main cave’s Cell PRb, which can reasonably be assigned to the same early date in the Asmaka restoration, when there must have been an urgent need to get some cells in the complex fitted out.30 The many similar niches in Cave 1 were almost certainly cut in 475, when that cave’s doorways were by and large refitted, in a way similar to that in Cell L1 of Cave 26RW.
Converted Features The caitya hall’s complex pillars have an underlying octagonal “Hinayana” format which is still preserved in the eight undecorated pillars which can still be seen at the apsidal end of Cave 26. Being in a low priority position, they were probably, after some trimmingdown, left with their simple format to save time and money, often a factor even in such prestigious caves.31 However, just as in Cave 2’s porch, where (as opposed to those in that same cave’s interior) octagonal pillars must already have been roughed out when work was first interrupted, the rest of the Cave 26 pillars were able to be “modernized” after 475 and made into more complex polygonal
29 The doorway of Cave 19 is both efficient and anomalous. Probably finished about 470, its quasi-D mode fitting, with its inner recess, already suggests the influence of Bagh’s doorway format. Possibly it represents an early transmission from Bagh, where certain workers would have gone from the start of the Recession. 30 For another, probably later niche, in Cave 27’s Cell R1, see Volume 1, Chapter 13 (end) Cave 27. 31 Another example of economy in Cave 26 can be seen in the fact that the painted dwarfs under the capitals toward the front of the cave have four arms, and those behind only two. Similar savings were effected in both Cave 10’s aisle ceiling redecoration and in the decoration of Cave 19’s vault.
cave 26 as an inaugural monument
41
forms. However, since they had already been roughed out in an octagonal format it was now impossible to supply them with the large square bases which came into vogue during later years of activity at the site; they would have had to be cut back so much that they would be too attenuated. This convention may have started as early as 468, by the time that the interior pillars of Caitya Cave 19, followed by the porch pillars of Cave 1, were underway.32 Significantly, although Cave 2’s porch pillars show their octagonal origins, the cave’s interior pillars are of the conventional postHiatus square-based types. This notably violates the convention of having the porch pillars and those of the interior of essentially the same design. But in Cave 2, this can be explained because the interior pillars must still have been in a very roughed-out state when work was interrupted on the cave in 468.33 Thus, when work on Cave 2 was continued after 475, it was still possible to make the interior pillars into the up-to-date square-based types. As if to confirm the fact that Cave 2 is not a wholly “late” cave, as is almost universally assumed, we should note that the doorway of its cell R2 was supplied with a revealingly early (but significantly unfinished) monolithic projection cut in the characteristic B mode. This clearly proves that work on the cave, as on so many others at the site, was interrupted by the Recession at the end of 468, and that by that time the interior had been penetrated to the extent that Cell R2 (and surely a number of other generally forward cells) had already been reached by the excavators. This being the case, it stands to reason that the forward part of the interior was also underway by this time. Indeed, it must have been penetrated a few years earlier than that, because the disposition of the interior pillars (still very roughed out in the cave’s early phase) shows that the forward ones, at least, were underway before a shrine was planned; this is because the intercolumniation of the axial pairs is not stressed.34 It is hardly surprising that the porch pillars would have been roughed out in
32 The Cave 19 portico pillars also may have been originally intended as octagonal, when first roughed out. 33 See conjectural plan of Cave 2 in Volume IV. 34 It seems to be the case that in other early excavations (Caves 4, 16, 17) which had also been started before shrines were planned, it was possible to widen the intercolumniations of the still very rough axial pairs by “squeezing” them outward. It may be that there was not enough extra matrix to do both this and to create the square bases in the Cave 2 pillars.
42
volume iia, chapter two
the early octagonal format by that time, even though the work on the interior pillars had not progressed so far, and could later be supplied, after 475, with the square bases then come into fashion.35 It seems clear that Cave 26’s four porch pillars were relatively larger in format than the somewhat crowded pillars of the interior when they were first roughed out by 468.36 This helps to explain why, when finally finished after 475, they could be completed in the by now conventional square-based mode. However, these square bases are smaller than expected (compare the roughly contemporaneous pillars in the same patron’s Cave 24!), almost certainly because this was necessitated when what we can assume were the originally octagonal forms, already roughed out, were cut back—suffering some reduction—to achieve a square format. By the same token it was necessary to reduce the width of the supporting “floor” beam beneath them, when that was finally finished; its originally intended width is revealed by the now “meaningless” cutback on the court side of the related pilasters. Thus, although the evidence is typically subtle, we can see the difficulties which the planners encountered, after 475, in bringing the old cave “up-to-date”. It is only through such observations that we can understand its true “two-phase” history. Still another conversion, in the Cave 26 complex, can be mentioned. I have mentioned elsewhere how the sculptors, in 478, struggled to “update” the still rough octagonal pillars in the porch of Cave 25.37 At the same time, they made a clumsy attempt to similarly update the capital design of the left pillar, by providing it with a very late (i.e. much reduced) volute design, characteristic of developments in the post-475 (in this case 478!) work on the cave, which Buddhabhadra was trying to turn into a setting for a new shrine at the eleventh hour. Had such a ribbed capital been defined in the cave’s early phase—prior to the Recession—it would have had a correspondingly early volute design; the earliest examples, found on the earlier right side of Cave 7’s porch (like the porch pillars of Cave 11) have five and a half volutes, reduced to four and a half on the very slightly later left side and in Caves 4, 16, 17 and 35
It might be noted that the widening of the intercolumniation of the porch pillars was conventional from the start, to provide more focus for the central porch doorway. 36 The Cave 26 plan from Yazdani Ajanta Vol. 4, 16 shows the porch pillars as rectangular in plan; they should be square. 37 Volume I, Chapter 12, Cave 25.
cave 26 as an inaugural monument
43
Ghatotkacha, and finally to three and a half in Caves 5, Upper 6, and 23, all dating after 475, even though the pillars of Caves 5 and Upper 6 were roughed out by 468. Development of the Cave 26 Stupa The present bhadrasana image in Cave 26 was added to the stupa no earlier than 477. In fact, the Cave 26 stupa was originally planned to reveal a very different image on its front; it was probably to be a standing Buddha similar to that in Cave 19. This need not seem surprising, since the two sponsoring feudatories had not yet come into conflict at that time; in fact the Asmakas could only have been included among the inaugurators of the site with the approval of the local Risika king. Furthermore, the two images in question were, as far as we know, the very first ever used in caitya halls in India— a revolutionary development conceived by Ajanta’s planners in the mid-sixties—even though the images (that in Cave 26 radically changed in form) were actually carved only when the excavations were more fully developed. Therefore, it is easier to believe that a single image form—in this case a standing Buddha—would have been conceived at this inaugural moment than that two significantly different types would have been developed for these essentially contemporaneous and related undertakings. Weiner (1977, 50) speaking of the possible sources of this fusion of stupa and Buddha, states: “Thus the stupa and the Buddha are combined in a single structure, a solution which in part may have been influenced by the use of Buddha images in the northwest to adorn the stupas, and in part, by the example of the great Amaravati and Nagarjunakonda stupas to which had been added standing images at the four cardinal points.” This is indeed reasonable, and reminds us of the significance of both these sources which, linked by trade route connections, in so many ways lie in the background of Ajanta and contemporary monuments. Indeed, the bhadrasana type itself might be considered an “import” from earlier sites in the southeast, in particular Nagarjunakonda, just as it might also find a source in the “bhadrasana” posed figures of yakshas or bodhisattvas from the Gandhara area.38 38
Possibly there was a bhadrasana image (now very obliterated, so not sure) at
44
volume iia, chapter two
At this early date, shrine Buddha images were never attended by bodhisattvas at Ajanta, even though attendant figures had long been common in India’s sculptural tradition.39 This important feature makes its appearance in the site’s shrines only in about 470, quite possibly first in the now-broken attendants of the Buddha image in Cave 19. Such bodhisattva attendants become insistently conventional from this point on, appearing first in the shrines of the same royal patron’s Caves 17 and 20, and then in all subsequent shrines. Ajanta’s attendant bodhisattvas, having finally gained entrance to the shrines, yield their place of honor only when often “replaced” by attendant standing Buddhas in 479/480, a shift particularly striking in the ambulatory of Cave 26, where the earlier (479) Buddha panels (those nearer the front, in better light, on better wall surfaces) have attendant bodhisattvas, while the later (480, and often unfinished) panels have attendant Buddhas. Panel L2, which in terms of its positioning might well be seen as midway between the two other types, intriguingly shows the bodhisattva Avalokitesvara on the proper left, yielding his conventional pride of place to an attendant standing Buddha, who has occupied the position on the proper right.40 We must remember that the new bhadrasana image in Cave 26 was a very late concept developed nearly a decade after the stupa had been roughed out. Therefore it had to be placed in the tooconstricted fronting space which, it is almost certain, was originally intended to hold a standing image. Thus the now-“required” bodhisattvas had to be squeezed onto the stupa drum, where they are not even visible when one stands in front of the stupa. Similarly, kneeling devotees, also now a “required” post-Hiatus feature, had to be most expediently treated. In 477, when the image was being carved, there was still enough uncut matrix on the floor at the left to carve a group of kneeling devotees, even though the constraints of the available stone resulted in their being smaller than expected and set too far rearward, for in normal course the floor would have been more fully trimmed right up to the stupa.41 The fact that so the left of cave 4 court at Bagh; if so, it would have anticipated those at Ajanta; just as the pair of attendant Buddhas in Cave 2 Bagh would be earlier than (sculptured) counterparts at Ajanta. 39 The constriction of the conventional central block in the earliest shrines may have been responsible for this “omission”. 40 N. Morrissey has located a similar panel in Kanheri 3 (late 5th century) and also in Aurangabad Cave 2 (late 6th century). 41 This stage by stage excavation procedure helps to explain why in 468 the floor
cave 26 as an inaugural monument
45
much matrix remained adjacent to the stupa would confirm the assumption that the stupa itself, and its fronting projection, was still only roughed out when work was taken up again on the cave after 475. On the right the floor had surely been more fully exposed when work was interrupted by the ouster of the Asmakas in 468. This is probably why the remaining matrix on this side was less than on the left, and therefore insufficient for the carving of the “necessary” monolithic devotees. Instead it is evident that a separately carved group was “plugged into” a slot in the floor; it is needless to say that only the slot remains. But by its position we can tell that the group it held was more appropriately located vis-à-vis the image, being closer and more to the front. This of course destroys the perfect symmetry which is generally so dominant a feature in such compositions, but Ajanta’s planners were typically able to adjust convention in cases like this, to allow for a significant correction. It is intriguing to consider why the cutting-down of Cave 26’s floor had proceeded further on the right of the stupa than on the left, after the line defined by the front of the stupa, in normal course, had already been reached, and the workmen were continuing the excavation rearward. It is easy to note that the space between the stupa and the pillars on the right is significantly less (about 7") than that on the left, something that might not have been realized by the “authorities” when the work was ordered and the payment (surely as per contract) was authorized. If this were the case, the workman (or workmen) on the right chose both first and well, and by “Dec. 31, 468” had cut this floor area down, while that on the left, where the matrix was much deeper, had not yet been started, as the presence of the later monolithic devotees proves. Or the explanation may simply be that given a choice of where to start, the crew, guided by human nature, began the easiest task first, and so (until sometime after 475) the left side remained with much matrix still to be removed. We might note that similar considerations explain the surprising asymmetry seen on two major panels on Cave 19’s façade.42
of Cave 26, in front of the stupa, must have been at least roughly revealed, while just beyond (notably at the left) the matrix was deep enough to carve the monolithic devotees. See conjectural drawing in Volume IV. 42 See Volume V: Cave 19 for discussion of these Cave 19 panels.
46
volume iia, chapter two Penetration of Cave 24 Into Cave 26’s Right Side
Because the huge Cave 24, planned by the powerful monk Buddhabhadra in about 466, was going to boldly (and knowingly) penetrate deep into the precincts of Cave 25, work on the latter half-finished cave (and on the similarly affected lower right wing, Cave 26RW) had to be abandoned at this point. All of the porch and hall cells on Cave 24’s left side would necessarily cut deeply into the right portion of the Cave 26 complex, although since work on Cave 24 was interrupted by the Recession, such cutting of its cells did not actually take place until 475 and after. However, Cave 24’s courtyard must have been already largely exposed by 468, making the creation of a shrine in the rear wall of Cave 26’s right wing impossible, since the cutting of a shrine or even cells in this area would have occasioned a breakthrough. Consequently, Cave 26RW’s shrine, when it was finally conceived in 468, was being shifted to a totally anomalous position at the center of the little cave’s left wall when time ran out with the Recession, the effects of which were precipitously sudden in the caves of the now-expelled Asmakas. The fact that these “violated” wings of Cave 26 were already significantly underway by 466, and summarily abandoned in that same year, alone proves that these two caves (and consequently the Cave 26 complex as a whole) must have been started by 462 or very shortly thereafter. Work was not taken up again on Cave 25 until 478, in some last-minute compulsion to still make something of the abandoned cave; and Cave 26RW also has an expedient later history, for even by 468, Buddhabhadra was determined to add a shrine to it (as mentioned above) even though this too was not completed, with its characteristically late image, until 478.
Why Was Cave 26, Already Well Underway, Penetrated by Cave 24? The adjacent viharas, Caves 21, 23, and 24, were originally planned, in about 466, with single cells at their porch ends, as was conventional at that time.43 Although their excavation was undertaken sequen-
43
It is conceivable that their court areas were being exposed as early as 465, in
cave 26 as an inaugural monument
47
tially, with Cave 21 started first, the group must have been planned together, since before the expansion of their porch cells they would have fitted perfectly into the expanse of scarp available between Cave 21 and the already established Cave 26 complex. But when extended porch cell complexes with pillar-fronted vestibules came into fashion in 467, all of these caves now required more room. This could be found only by shifting them all to the left, since Cave 21 (the earliest of the triad) was already significantly underway, thus making any shift to the right impossible. Moving everything to the left, to provide space for the expanded cell complexes, with their pillared vestibules, was the only option. By viewing the western caves from across the river, it is easy to see that just enough scarp area was reserved between Caves 21 and 23 to comfortably incorporate the single cell originally planned at Cave 21’s porch left, with its counterpart at Cave 23’s porch right. It was only when such single cells were expanded that problems would occur; and in compensation, even before Cave 23’s porch was penetrated (but when its flat façade plane had already been defined, in what was the conventional fashion as excavation progressed from courtyard to cave), Cave 23 had to be expediently “displaced” some 10’ to the left, in order to make room for the newly conventional pillared cell complexes in both Caves 21 and 23. The problems which resulted are discussed elsewhere.44 When we look to the left from our vantage point across the river, we can see that even before the excavators had reached the façade of the huge Cave 24 in 467, they were planning for the expanded cells, for a much larger section of scarp has been reserved for that purpose between Caves 23 and 24. Consequently, as explained above, the upper and lower left wings of the Cave 26 complex had to be “sacrificed”, if the splendid later Cave 24 was to be properly completed in the way which evolving taste now demanded. Indeed, the displacement first of Cave 23 and then of Cave 24 now required that the pillared cell complex at the left of Cave 24’s porch would have to penetrate some twenty feet underneath the older Cave 25, making it impossible to properly cut down the floor of the latter’s
which case they would have been conceived as dormitories. But it is evident that by the time their complex porches were reached, shrines were being planned. 44 See Volume V: Caves 21–26.
48
volume iia, chapter two
still unfinished porch. By the same token, the left aisle cells of Cave 24, had they ever been able to be completed, would have had a similar effect in Cave 25’s interior; indeed, it was at this point that work on the Cave 25’s hall floor was abandoned in mid-course, probably in 466, even perhaps in 465. This “destruction” of the right wings of the great caitya complex, to accommodate the extended Cave 24 was a decision as drastic as it is surprising. However, given the fact that every good space for caves at the site was already taken by 467, it was in fact the trading off of embarrassingly obsolete older excavations—those two old wings which formed the right side of the caitya hall complex—for a new conception in which the donor could take an understandable pride. And this donor of the impressive and up-to-date Cave 24 could have been no one other than the “owner” of the caitya hall itself, namely the powerful monk Buddhabhadra, whose force of character and access to funds—already evident in his remarkable caitya complex—is revealed by his dramatic decision to essentially sacrifice the half-finished Cave 25 and Cave 26RW, in order to make his most modern and sumptuous vihara the way he wanted it to be. Indeed, since the now-expanded Caves 21 and 23 were similarly implicated in the shift leftward, they too must have been his own excavations, or under his clear control. That is to say, Buddhabhadra appears to have appropriated (or have been granted) the whole western extremity of the site as a specifically Asmaka complex, quite possibly from the time the caitya hall was begun in 462, even though Caves 21, 23, and 24 were not undertaken, in that order, until four or five years later, once the great caitya hall was well underway. It was of course long assumed that the Cave 26 caitya hall represented the latest phase of work at the site—that it “could only have been excavated after the fall of the Vakataka dominion” (Khandalavala 1991, 102); and even Mirashi took this same position, entirely omitting its crucial dedicatory inscription (along with dozens of intrusive Vakataka inscriptions) from his great publication: Inscriptions of the Vakatakas (Mirashi 1963). It was logical to conclude, therefore, that its location at the far end of the site’s western extremity was a function of the unavailability of space elsewhere at the site. But, when we realize that Cave 26 must have been one of the very first excavations to have been inaugurated at the site, such views have no force. What seems to be the case is that the Asmakas, via Buddhabhadra, located their huge caitya hall at the intended center
cave 26 as an inaugural monument
49
of an extended monastic complex, which they started developing to the right (Caves 21–24) and would eventually have continued with Cave 28 and others (never started) at the left, had time not run out in 478. It should be noted, of course, that the height of the cliff where Cave 26 is located, was an advantage, and a necessary one; but even as far to the left as Cave 21, the depth of the scarp would have been more than sufficient to contain it. Unfortunately, by the time that it was clear that Cave 23 needed to be shifted to the left, in response to Cave 21’s expansion, the excavators of Cave 23 had already exposed the façade plane in its originally planned position. Typically, the excavators had established it as a flat surface, ready to be penetrated when the cutting of the porch colonnade was begun.45 This plane can be seen, still uncut (for the porch itself had not been started) at the extreme right of Cave 23’s façade, where it now exists as a “meaningless” (vestigial) stretch of smoothed rock some ten feet in width.46 The explanation for its continued existence (unused) lies in the fact that Cave 23’s porch also now had to be expanded to allow for the newly “necessary” cell complexes at its porch ends. Consequently, the whole nowexpanded cave had to be shifted to the left to give it sufficient room. By happy chance, the porch colonnade had still not been begun at this point, so the excavators were able to “displace” the whole façade plane, not only to allow adequate room on the right, but also (by now extending the façade plane) an equivalent amount to the left, before starting the reveal the porch colonnade in its new (more leftward) position. It is evident, of course, that this displacement of Cave 23 had to have its “domino effect” on Cave 24 which, in its original (466) conception, would have had only the then-standard single cells at its porch-ends, and would have fitted closely but comfortably into the space between Cave 23 and the Cave 26 complex, as originally laid out before fashion changed.47
45 Cave 24A, although barely started, was probably being cut in this way. There is a much clearer example at Nasik (Cave 1) where the plane of the façade has been clearly revealed; its cutting was never continued. 46 It has been suggested that this blank wall was left to create a setback in that area, to provide for steps leading up to Cave 22. But there was surely no plan to make Cave 22 when the wall was cut, so such a suggestion goes against reason. 47 See reconstruction of area as of 468, in forthcoming volume of plans.
50
volume iia, chapter two
Cave 21 was well under excavation—at least the porch was essentially revealed—when the Recession caused the temporary abandonment of all of these Asmaka caves at the site’s western extremity at the end of 468. By then the excavators of Cave 21 realized, even though Cave 23 had been displaced to the left, how constricted the space was between the two caves; it now had to contain not only the two residence cells, but the pillared vestibules which fronted them. Therefore, although they cut the right residence cell and its vestibule in the normal way—for there was no problem on this side— they notably reduced the depth of both of these elements at the left, knowing that if the excavators of Cave 23 did the same thing with the abutting complex (at Cave 23’s right) it would be possible to fit everything in, since Cave 23 had already been shifted to allow more space. It seemed that the problem was solved. At the end of 468 the Recession occurred, and all work in the western extremity had to be abandoned until 475, by which time the Asmakas had won control of the site for themselves. Then work started up again. But since many of the workers, and probably the planners too, had left the site during the Recession and the following Hiatus, it seems apparent that a new crew took up the work at this time; and although they of course knew that Cave 23 was to have the now standard pillared complexes at the porch ends, they obviously were not told that they should do as the Cave 21 crew had earlier done, and to carefully squeeze these elements into a space which was in fact very constricted. They merely went ahead in the normal way, starting the problematic complex at the right in a way which was destined for disaster. The disaster indeed happened when, in cutting the inner cell, the excavator suddenly broke through into the inner porch-end cell at the left of Cave 21. At this point it was necessarily abandoned, surely to the great embarrassment of those in charge. We can be sure from such evidence that, although the porch of Cave 21 was well underway by 468, the porch of Cave 23—at least the revealing of its porch end complexes—was not even being roughed out until 475. And by that time the new workers quite understandably assumed that they could proceed in a normal fashion, not realizing that time had laid this trap for them. When the façade plane of Cave 23 was shifted leftward, just before the Recession, it is evident that the new convention for porch-ends had already been established. (This façade shift was hardly earlier
cave 26 as an inaugural monument
51
than 468, or the porch cells would have been started before work was interrupted.) Of course Cave 24, where the courtyard alone was being excavated at this time, was to have such complexes too. This explains why, in order to contain the new cell complexes the stretch of stone separating Caves 23 and 24 is nearly twice as wide as that between Caves 21 and 23. But at the same time, the result of all this accumulating leftward shifting meant that (and explains why) Cave 24 had to deeply and destructively penetrate into the outmoded right wings of the Cave 26 complex. It is of course fascinating that such a drastic course of linked action, first involving Cave 21, and then 23, and then 24, resulting inevitably in the destructive penetration of right side of the Cave 26 complex, could have been sanctioned. And since the ultimate impact is on the vast and sacred caitya hall—which alone sustained (and allowed) the “damage”, it is evident that such drastic actions could not have been done without the approval of Buddhabhadra, and at his own initiative. One should not discount, of course, his auspicious connection with the powerful Asmaka minister, “who was attached to him (the monk) in friendship through many successive births” (Cave 26 inscription, verse 9). Although by 478, when the cave was inscribed, his powerful friend had died, to be succeeded by his son (“an equally foremost personality” (ibid. verse 12) it seems safe to assume that Buddhabhadra’s good friend was alive and committed when the great caitya complex was started in 462, over fifteen years before. In any case, whether or not Buddhabhadra himself raised the funds for all of the caves in the site’s western extremity, it is evident that nothing in the area would have been done without his approval or his initiative, the expansion and shifting leftward of Caves 21, 23, and 24 being a clear case in point. The above analyses prove that Cave 21, the porch (or at least the façade) of Cave 23, and the court area of Cave 24 were all underway (to varying degrees) by 468, when the Recession forced the Asmakas’ sudden departure. The same decision that authorized the expansion and shifting of these three viharas required the abandonment of work on the consequently corrupted right wings of Cave 26, allowing us at the same time to see, at least in general terms, how far work had progressed on the caitya 26 complex as a whole when the decision was made to halt any further excavation on its right side. It is reasonable to assume that the decision to expand the porches of the three major viharas was made in 467, for the space
52
volume iia, chapter two
between Caves 21 and 23 was sufficient for only single porch-end cells, which were still in fashion in 466. Cave 21’s porch was probably started in that same year (466), with the porch cell adjustments underway in 467 and finished by 468, to “update” the cave’s conception. It is fascinating (and revealing!) to see the different manner in which the excavators handled the conversion of the single porch cells originally planned for Cave 21, to the expanded cell complexes which convention now required. At the right, where there was no problem caused by Cave 21’s abutment with Cave 23, both the pillared vestibule and the inner cell are “normal” in size—reflecting no restraints whatsoever. On the other hand, when we view the left porch cell complex, it is immediately evident that the excavators, defining this area in 468, just before the Recession, realized that they quite literally had to “squeeze in” the new cell complexes in the abutting caves, and so they significantly reduced the depth of both the vestibule and the cell beyond. Had the excavators followed suit in excavating the right cell complex in Cave 23, there would have been no problem. But this was not to be the case. As for Cave 23, it had also been started in 466, with the expectation of single cells at its porch ends, as the original positioning of the façade, located closer to Cave 21 than the present positioning of the colonnade, makes evident. However, the porch was not penetrated until at least 467 and probably 468, since its new (leftward) positioning shows that more complex porch-end units were being planned for when the colonnade was being roughed out. Admittedly, it is quite possible that there was a gap between the time that the original façade plane was defined and the time that it was moved to the left, for this required a major rethinking, and decision at a high level. It is even possible that it was put off for the time being, while work progressed vigorously on the slightly earlier Cave 21. There is no way of knowing if the façade plane was moved to the left, and the penetration of the porch begun, just before the Recession or not. All we can be sure about is that the porch-end complex at the right (and probably its untroubled counterpart at the left, where plenty of space had been allocated) was not begun until after the Asmaka restoration in 475, by which time the need to carefully “squeeze it in” had been forgotten. It is not difficult to imagine what happened when a new crew of excavators was put to work on the porch of Cave 23 at the time of
cave 26 as an inaugural monument
53
the Asmaka restoration, in 475. They were told to cut the cell complexes, and they of course had no reason to know—if they had not been specifically received word from “Delhi”—that they could not go about it in normal course; which is precisely what they did. Even if they trekked over the Cave 21, they would not have suspected the potential problems, nor would the subtle adjustments in Cave 21’s left cell have been noticed by any simple workman. Thus, having generously (too generously!) carved the pillared vestibule of the right cell complex of Cave 23, they continued their work into the inner cell. It must have been a total shock when, barely halfway through the course of the cell’s excavation, the chisel erupted into the abutting cell of Cave 21. At this point, work was of course abandoned, the revealing problem adding only to the compelling analysis of developments which such situations allow. After the Hiatus, of course, work began again with great vigor throughout all of the Asmaka caves in the site’s western extremity, for the Asmakas were now the rulers of the area. Thus all of these caves (except the very late new undertakings such as Caves 22 and 28) reveal a dramatic two-phase pattern, first continuing up until the end of 468 and then again from the beginning of 475 until after Harisena’s death, precisely reflecting the developmental pattern of the great Cave 26 complex itself, with which they were all associated. Finally, it is of interest to note that when the façade plane of Cave 23 was first revealed, most of the matrix in the court had not been fully cut away—the focus of work being to get the plane defined and then the colonnade started. Then when it became necessary to shift Cave 23 to the left, a large portion of the uncut matrix in this now “useless” area was merely left, being a low priority concern. This is the reason that it could be utilized for the steps up to Cave 22, which was probably not even begun until 477.48
48 The suggestion that the blank wall was left to provide convenient access to Cave 22 makes no sense, since Cave 22 was not even conceived until 477.
CHAPTER THREE
CAVE 26’S COMPLICATED DEVELOPMENT COMPARED WITH UPENDRAGUPTA’S CAITYA CAVE 19 AND OTHER CAVES 462–468: Ajanta’s First Flourishing under Risika: Asmaka also Vigorously Involved in Ajanta’s Patronage The vagaries of the Asmakas’ involvement in Ajanta, and in the course of the empire, form the most gripping, and painful, chapter in Ajanta’s “illustrated history”. In 462, when the site was inaugurated by a powerful and courtly consortium of patrons, the Asmakas were clearly on good (or at least workable) terms with the local feudatory of the Ajanta region, Upendragupta of Risika (See Map). In fact, the whole area must have been at peace, to provide a political environment in which such an ambitious undertaking as Ajanta could have been begun.1 The peaceful connection between Asmaka, Risika, and Anupa, early in Harisena’s reign, can probably be ascribed to the fact that they were part of the domains which the new ruler inherited from his father, Devasena. Thus it is not surprising that these central areas, all bordering on western Vidarbha, are notably absent from the list of presumably conquered and/or dominated territories listed in his Prime Minister’s Cave 16 inscription (verse 18). However, the relations between the local king and Asmaka had eroded by the end of 468. It was at this point that the development of the Cave 26 complex (indeed of the whole Asmaka series of caves at the western extremity of the site) came to an abrupt end, due to the precipitous expulsion of the Asmakas from the region. This conclusion, although the expulsion may have had little to do with relationships between the patrons at the site itself, is nonetheless supported
1 The inscription of Harisena’s third regnal year supports the assumption that this general area was at peace. Mirashi’s interpretation, which assumes that Harisena was involved in a digvijaya seems to have no basis (Mirashi 1982, 78–85). A.M. Shastri agrees that there is no basis for a digvijaya.
cave 26’s complicated development
55
by the evidence that Ajanta supplies. This shows that after a few years of strife (already smoldering in 469 but breaking into flames by 472) the tables were turned and the Asmakas, now victorious, had in fact taken over control of the site and (one can assume) the surrounding region. The emperor Harisena, although powerful, had apparently not tried to prevent this. Even though Asmaka’s strength and aggressiveness must have been a matter of concern at the central court, for the present Harisena (and/or his powerful prime minister, Varahadeva), perhaps intentionally opted for a policy of divisiveness, trusting that such a local strife might weaken a potential aggressor. Rulers—the case of Akbar springs to mind—have often improved their own position by tolerating and even encouraging the internecine rivalries of some of their vassals. Actually, if such a policy of “divide and rule” on the part of Harisena was indeed the case, it may not have had the desired effect. If we can credit the fulsome inscription of Upendragupta, the local king, it is fair to say that this exuberantly pious ruler had wasted his birthright on good and pious deeds, “expending abundant wealth” on excavated halls “which cannot even be imagined by little-souled men” (Cave 17, verse 25), as well as on other lavish indulgences. Thus he was ill-prepared for the Asmaka troops that were all too soon to come marching over the hill. The brief (c. 472–c. 474?) conflict, instead of slowly eroding the Asmaka power, must have significantly enhanced it, as they so easily took over Upendragupta’s rich domains. If we can judge Upendragupta’s Cave 17 record, and indeed the evidence of his luxurious donations, Upendragupta was far from destitute; so his great losses must have contributed directly to the future war effort which the Asmakas may already have been planning against their Vakataka overlord.2
2 It seems clear that Asmaka did not take over the whole of Risika along with the Ajanta region in the 470s, since the “King of Risika” was one of those who joined the Asmaka insurrection described in the Visrutacarita. He was probably the Thalner inscription’s Maharaja Gomika (or his successor), who was known to be ruling over northwestern Khandesh in the early 460s, and must have been a contemporary of Upendragupta, who was ruling in the southeast. Even though the Asmakas had taken over the latter area in the 470s, and apparently were increasing in strength and influence, they clearly had not taken over the domains of Maharaja Gomika, who must have been the “King of Risika” listed as one of the insurrectionists by Dandin.
56
volume iia, chapter three
This strife between Asmaka and Risika caused Ajanta’s Recession, roughly from 469 through 471, after which their conflict erupted into war, causing a total “Hiatus” in patronage activity at the site from c. 472 through c. 474.3 The manner in which the Recession is manifest in different caves at the site is significant. Only the grand Cave 1, Harisena’s own donation, continued to develop without apparent problems, at least until the war put a temporary stop to its development. Not surprisingly, the caves of the local king (17–20; 29) continued underway during the Recession also.4 However, if Harisena’s great Cave 1 shows an apparently untroubled development, Upendragupta’s involvement was much more subject to haste and expediency; for after all it was Upendragupta, not Harisena, who was anticipating attack. As for most of the other caves in the main area, only in that of Varahadeva, Harisena’s prime minister (Cave 16) was major work continued for a somewhat extended period—probably throughout most of 469. However what was now done in Cave 16 was as much involved in subtraction as in addition, in the interest of saving time. This remarkable phenomenon, a process of simplification most unexpected in India’s excavated art, will be discussed in due course.5 But Varahadeva clearly misjudged the situation which, at the beginning, he was planning to transcend. As a consequence, unlike the lesser patrons of Caves Lower 6, 7, 11, and 15, who had rushed their Buddha images to completion by the middle of 469, he could not even get his image started by the end of that difficult year. Prime Minister or not, he finally had to yield to circumstance. And his reward, when he was finally able to continue work on his cave again nearly a decade later, was that he was able to sponsor the most staggering Buddha image at the site—an image that was, in the end, nothing like the simple one that he had earlier planned.
3 I allow three years for the conflict, but it could have been over quicker. The stylistic, iconographic and technological changes that transpired at the site before the Asmaka takeover in (or by) 475 predisposed me to generosity in my estimate. 4 Upendragupta’s Cave 29 appears to have been started in 469 to “replace” (insultingly) the banned Cave 26 chaitya hall, and like the latter appears to have been oriented to the summer solstice; but it had to be abandoned as the Asmaka threat developed. 5 See Volume V, Cave 16; also Cave 21, where similar “savings” were effected. Even the cave at Banoti (Volume 1: Banoti) shows certain (later) cutbacks.
cave 26’s complicated development
57
The Prime Minister had a similar disappointment with his related Ghatotkacha vihara; because it was so hopelessly unfinished at the time of the Recession. Therefore, early in 469 he decided to hurriedly carve a projecting stupa—a kind of expedient shrinelet—out of the mass of still uncut matrix at the right end of the front aisle. The “sanctity” of this unusual shrinelet is evident from the fact that in the Period of Disruption it attracted even more intrusions (including the seated Buddha carved on its face) than did the cave’s hastily completed main image.6 Other lesser patrons, also afflicted by Asmaka’s threat, apparently had two options. In those caves where the shrine image could be rushed to (or at least toward) completion in the few weeks allowed, this course was taken. It is evident that starting, completing and then dedicating the main Buddha image was the patrons’ greatest concern in these troubled times. Such images were completed, even if with a sobering expediency, in Caves Lower 6, 7, 11, and 15, where we can assume they were successfully dedicated, particularly since this was still a time of peace. The shrine in Cave Lower 6 was the only one essentially completed as planned, as far as its basic design is concerned. However, it must have been originally planned for a stupa, rather than for the Buddha image carved there, to satisfy changing taste, in 469.7 Being of obvious high priority, the Buddha image was carefully finished, despite the fact that so many of the wall surfaces at the rear of the cave were never properly finished when, in 469, they were hurriedly surfaced with a thick layer of plaster and then hastily painted. The shrine walls and ceilings were probably finished last—the scaffolding for the image being in the way in 469—and never got fully completed, at least toward the rear. Cave Lower 6’s fine shrine doorway was probably well underway in 468, but was surely not painted until 469, since the adjacent walls were finished with a roughness characteristic only of that latter year. Therefore Cave Lower 6’s Buddha figure, like those of Caves 7, 11, and 15, belongs to the
6 See Spink, 1995, 171–180. However, herein I placed the stupa just prior to the beginning of the Recession, in 468; I now realize that it was made at the beginning of the Recession, in 469. 7 For the suggestion that a stupa was originally intended, see above, Chapter 2; also Volume IV: Cave Lower 6.
58
volume iia, chapter three
opening weeks or months of the Recession; but it is the only one whose plans were not too significantly aborted by the pressures of time. It might be noted that Cave Lower 6’s cells had still not been fitted out at this time (469); their doorways—originally A mode— were re-cut in the late D mode after 475. As for the many less fully excavated caves, where the image had not even been reached, they were abandoned completely at this point, although work on them would start up again after the Asmakas had finally taken over control of the region by 475. (See Time Chart) Significantly, the situation is very different in the Asmakas’ Cave 26 complex which, despite its importance, now suffered a harsh reversal. By 468 three of its shrine areas (the stupa in the main hall as well as the shrine areas recently added to its lower left and lower right wings) had already been blocked out, and were ready to receive their images, as soon as the “decoration phase” of the complex would be started.8 They had not been already carved because the planners of Cave 26 (and of all of the Asmaka caves underway prior to the Recession), followed an early procedure which delayed the carving of any architectural and sculptural details until the whole complex was roughed out. The shrines in the two attached viharas, neither of which had been planned when the simple astylar wings were originally conceived, reveal in their positioning the insistent demands of this new requirement—the need for now providing a “residence” for the Buddha—a shrine—in every vihara. The shrine in the Left Wing was being “attached” by 468 to what (still in 466) was to have been a “Hinayana” type residence hall, with no provision for an image whatsoever. Influenced by the recently developed (and anomalous) plan of the local king’s little Cave 20, the Left Wing’s shrine and shrine antechamber were merely “tacked on” to the back of the cave’s unfinished hall. The shrine in the Right Wing is even more expedient. It was not started until 467 or more probably 468, since the clearly associated and set-back adjacent cell has a B mode fitting, first introduced to the site in the latter year. By this time, the presence of the complex cell at the main cave’s right porch end made it impossible to cut 8 However, this treatment was not universal—the doorway of Cave Lower 6, the façade decoration of Cave 7, and many other examples show that carved “decoration” was started fairly early in certain other caves.
cave 26’s complicated development
59
the shrine back into this area. Nor could it be cut on the opposite wall where the stone was so corrupt that only one (R1) of the two expected residence cells could be excavated. Nor could the shrine be cut, as would be conventional, in the cave’s rear wall, because by, probably as early as 466) the floor of the adjacent Cave 24’s extensive court—being given priority—had cut away so much matrix that the roof of any such shrine beneath it would have immediately collapsed. This is particularly evident since the presence of a bad flaw high up along the Right Wing’s rear wall would have required the shrine ceiling to be raised up so high that a breakthrough would have been inevitable. Since work on both of Cave 26’s upper right and lower right wings was abandoned in about 466 in order to allow Cave 24, with its up-to-date porch complexes, to expand as it would, it is likely that the planners of Cave 24 were not concerned about the problem that the cutting down of the cave’s court floor would cause in 468. Indeed, it seems likely that work on Cave 26’s right wing was taken up again at that time mostly in order to provide another location for a Buddha shrine, since at the very most the little cave could now (in 468) contain only two of the standard six cells originally intended. In any case, because a shrine was now required for this Right Wing, the decision was made to cut back the thick matrix remaining along the previously unfinished left wall to form a central projection into which the intended image—in fact not realized until a decade later—would be carved.9 As mentioned above, this explains the curious “set-back” of the adjacent cell, while the fact that on the left side the wall could not be penetrated (because the area had already been utilized in 467 for Cave 26’s right porch cell) may have suggested utilizing the “wasted” space for a bed, which undoubtedly did enjoy usage. It might be noted that this small bed—the length (slightly less than 6') quite reasonable for a fifth century monk—has the same essentially “charpoy” design, with four nubbed legs, as that upon which the Buddha lies dying in Cave 26. It can be supposed, too, that somewhat similar beds were used in the cells, as their occasional adjustments of length and breadth would suggest.10
9 10
The image finally carved on this projection dates to 478. See Volume I, Chapter 3.
60
volume iia, chapter three
It is reasonable to believe that the main hall had priority over the problematic right wing; and the fact that the cells of the main hall originally had A mode doorways, while the right wing had slightly later B mode doorways, supports this view. However, the bed mentioned above, like the bhadrasana “shrine” image finally carved in the left wall’s projection in 477 or (more probably) 478, had not yet been carved—and surely not even conceived—in this early phase, because of the uniquely abrupt way that work broke off in the Asmaka caves at the end of 468. At that point neither the projection for the image nor the undefined area to the left (the bed area) were probably more than roughed out, and would have remained in that condition, because of the suddenness with which the Asmaka patrons were expelled. Surely, considering the importance of their patron and of the complex itself, and the fact that their locations were already essentially prepared, these three images would have been completed even more quickly than those in Caves Lower 6, 7, 11, and 15, had the Asmakas been able to continue work. But they were not; and so these three shrine areas are unique in not being hastily supplied with their intended Buddha images at this stressful time. As if to explicate the political situation in which the Asmakas now found themselves, work on these Asmaka shrines just suddenly stops—as suddenly as the Asmakas hurried (or were hurried) out of the region. By the time that the aggressive Asmakas returned and took over the site by 475 and, a few years later, had finally completed these “missing” images, almost a decade had passed; so understandably the new images were very different from those originally intended. Cave 26 and 26RW have notably late (477–478) bhadrasana images; the padmasana image in Cave 26LW, like the very similar image in Cave 2, probably was begun somewhat earlier in 477, before the bhadrasana type was introduced; although the addition of leonine throne legs to its base (heightened because of a flaw), the late character of its shrine doorway, and the fact that its shrine antechamber pillars were expediently finished, suggests that it was still underway in 478.11
11 The pillars had been blocked out in an “unfinished” square format in 468, and the later decoration was merely applied to the squared surfaces, rather than being more properly elaborated. For the late shrine doorway, see Volume V: Cave 26 complex. Leonine throne legs were first developed in 477.
cave 26’s complicated development
61
Drawing historical conclusions from this revealing art historical evidence, it seems clear that the Asmakas were abruptly expelled from Ajanta (and of course the surrounding region as well) at this very point. The reason for this must have been their developing aggression toward Upendragupta, the local king, who now, under an all too real threat of conquest, drastically limited the ongoing work at the site, presumably in order to prepare (too little, too late) for the likely needs of war. It is fair to assume that, with a (Buddhistic?) policy of “peace in our time”, he had used his vaunted “power of the expenditure of wealth” (17 inscription, verse 17) for precisely the wrong purpose. His numerous excavations at Ajanta were obviously only part of his lavish, even Olympian, program to “adorn the earth with stupas and viharas and to cause the joy of supplicants by conferring gifts (on them)”. (Ajanta Cave 17 inscription, verse 22) But this did not save them from the Asmakas. As Bakker (1997, 44–45) tellingly suggests, analyzing the attitudes revealed by the inscriptions, the courtly patrons such as Upendragupta were hardly disposed to deal with such harsh realities as an attack from Sparta. For they had been living in a luxurious “world of courtiers in which the ministers took over the day-to-day worries of government in order to allow the king to become ‘free from care’ so that he could engage ‘himself in the enjoyment of pleasures, acting as he liked’”, using the privileges of the gods rather than the responsibilities of the hero as his model.12 Fascinatingly, Bakker and Dandin, though still not on speaking terms, are in remarkable accord in their analysis of the ills of empire during the troubled reign of Harisena’s successor, Sarvasena III. The royal court apparently took on a fin de siecle character as compelling as it was ominous; and in fact the indulgences appear to have multiplied to such a degree that the fin de siecle—the end of this great century—arrived some twenty years before its time was due.
12
Bakker 44–5, with quotes from the Cave 16 inscription.
volume iia, chapter three
62
469–471: The Recession: Asmaka Caves Abandoned—Many Workmen Leave as Patronage Reduced and Fear of Asmaka Invasion Increases War, as the Recession developed from 469 to 471, was increasingly imminent. This is suggested by the more and more expedient fashion in which the local king’s sumptuous caves, upon which he “expended abundant wealth” (Cave 17 inscription, verse 25) were rushed toward completion and were dedicated, but were finally not properly finished at all. When, late in 471, the workers of the local king, with uncharacteristic haste, cut Cave 17’s left porch pilaster back in order to make more room for his dedicatory inscription ( just recently composed, formatted, and sent down from the capital), the boast that in some earlier time he and his brother “had attacked repeatedly (?) rich countries such as Asmaka”13 his reference to his antagonists was by now but a useless example of braggadoccio. The reference is suspiciously apotropaic—seeking to prevent what was surely going to happen in the weakening present by invoking the lost glory of the past. But it would not protect the little kingdom from the rising Asmaka tide. That the Asmakas were about to “come over the hill” at this point is evident in the manner in which all of the local king’s major caves were now nervously rushed toward completion, as if the overriding new criterion now was just to get them done. The plastering in the later (generally rearward) portions of Upendragupta’s ambitious Cave 17 covers shockingly rough wall surfaces and cell doorway openings. This must have pained the highly discriminating Upendragupta, just as he must have been pained, late in 471, to have to rush his beautiful image into dedication, when the surrounding shrine walls were still devoid of decoration. Similarly, the image group in Cave 20, along with its shrine doorway—both forms closely related to those in Cave 17—had to be roughly completed with plaster rather than fully carved. Indeed, the whole of the lavishly conceived little Cave 20, with many of its uncarved details expediently plastered and painted (or half painted), is a sobering study in how to try to finish very quickly what in better times would have been completed with care.
13
Cave 17 inscription, verse 10; trans. Bakker 1997, 25.
cave 26’s complicated development
63
The caitya Cave 19, on the other hand, was more fully finished before Upendragupta’s world began to fall apart. It was obviously his “priority” monument. But even here portions of the court decoration, including the expected large yaksha (or alternatively the yakshi Hariti) which (according to convention) should have been paired with the beautiful seated nagaraja across the way, were never finished when time ran out in 471.14 In the case of the missing yaksha (or yakshi), only the honorific roof-motif was blocked out. This confirms the hypothesis that the location had earlier been assigned to the expected subdivinity; the “vestigial” roof-motif has nothing to do with the bhadrasana Buddha image cut beneath it in 479, at which point it was merely used as a location for hooks (now missing) to hang the honorific garlands supplied to the intrusive Buddha. It is also relevant to note that, although doors were hung in the right rear court cell in 470 and in the left court cells in 471, as their fittings and related shelves reveal, they could not have been used more than briefly, if at all, in Upendragupta’s day. It is possible that the cells (like those of Cave 17) also continued in use, for practical reasons, even after Upendragupta lost his control over the region. However, it is more likely, considering how the Asmakas forbade worship in the cave and also “desecrated” the two front cells (see below), that the evident wear in the pivot holes dates from the period of the site’s breakdown in mid-478–480 (the Period of Disruption) and for a few year thereafter, when the cave was clearly in use. C mode cell doorways appear at the site only in 470 and 471, as is evident from their very late (471) positioning in Cave 17 and Cave 1. Excavated niches, as in Cave 19’s left court cells, never were cut prior to 471, as is again revealed by their positioning in Cave 17. The right cell complex, apparently dating from 470—for its vestibule as well as the connected cell were finished while the similar area at the left remained unfinished—once had pegs placed so as to hold a double shelf. This was a unique “constructed” counterpart of the left cells’ niches, which had not yet come into usage in 470, and then appears only in contexts dated from 471 through 475. Thus we can surmise that to make up for the lacking niches, the rear right court cell was fitted with this structural substitute, probably in 471.
14 After 471, such pairing became conventional. See the porch-ends of Caves 2, 21, 23, 26, 26LW (left end broken, however).
64
volume iia, chapter three
Although we must believe—from the presence of intrusions alone— that the main Buddha image in Cave 19 was dedicated, the inscription planned for placement over the doorway inside the cave apparently did not get sent down to the site in time. Perhaps there was by now no time (or literate carver) to inscribe it. Nor could the cave have been in worship more than very briefly, before (with the Asmaka takeover of the region) it received the same summary treatment that had earlier been accorded to the “rival” caitya hall Cave 26. This is why this beautiful caitya hall, which Upendragupta had planned as the ritual center of the site, shows not a trace of soot from incense or oil lamps; while even a few of its garland hooks were never inserted along the pillar tops, as would have been done in this insistently organized cave if its royal patron had had a chance to properly finish and use it. In fact, even those which were already in place were probably never used at all, since the plaster around them shows none of the damage typically caused (in many other caves) by the careless changing of such garlands with a “pole-hook”. The damage that we see today was undoubtedly caused in more recent times by people pulling the useful iron hooks out, causing some of the plaster (which was applied after the hook was in) to break away. This breakage from pulling out the hooks was quite different from that caused by the more gradual damage caused by changing the garlands. The latter is seen in one instance in the cave, however. This is at the center of the front aisle ceiling where the knockedaway (but now inappropriately restored!) plaster around the hole for the now-missing garland hook proves a degree of ritual use. Since it is clear (and understandable) that Upendragupta did indeed dedicate the hall—the presence of later intrusions alone would proved that it had been brought to life—it is possible that this damage took place during the brief period (probably just a few months) in 471 when it would have been in worship. However, this breakage is more likely to have taken place during the Period of Disruption (479–480) when many devotees were making intrusive donations there than it was to have been done during the brief period of frenzy when Upendragupta was hurrying to complete the cave—failing even to get the inscription cut into the panel prepared for it on the hall’s front wall. We can locate the near-completion of the complex court cells of Cave 19 very precisely in time. They probably were not even planned when the splendid Caitya Hall was originally conceived; and they
cave 26’s complicated development
65
certainly were not started until very late.15 This is evident because, like the very late nagaraja and (merely planned) yaksha across the court, they are not conventionally and properly aligned (See Plan). They respond instead to the surprising curvature of the façade, whereby the left front face of the otherwise precisely engineered cave is drastically (and surprisingly) twisted to achieve a solstitial alignment.16 Thus the fact that the vestibuled cell complexes are aligned not with the façade as would have been originally planned, but (like the late nagaraja relief also) with the torque of the façade, proves that they were conceived and started rather late, probably in about 470. It is clear that cave 19’s façade was not “adjusted” to connect with the (winter) solstitial axis—which it does, at its point of most extreme curvature—until the cave had already been deeply reamed out with no expectation whatsoever that the dramatic “warping” of the whole excavation was going to be required. It was only after the cave’s original positioning had been established that an order must have come from the king and his advisors in the capital to do what was in effect the impossible—to align the already roughed out cave to the winter solstice—for the original orientation of the rock-cut hall is angled well over a dozen degrees to the right of the solstitial axis.17 Had the perpetrators of this impossible demand realized the excavation problems involved, they would probably not have made it; but the planners at the site, blindly following blind orders from above, did it anyway—or at least tried to do it—wrenching the otherwise precisely planned excavation harshly away from its original axis. 15 It is likely that, like the various pillared cell complexes in many of the viharas, their ultimate source was the court cells of Cave 1. 16 Spink 1985A; although my measurements are totally incorrect, the overall observations seem to confirm the intention of the planners to achieve (impossibly) a solstitial orientation. 17 According to careful theodolite measurements made by Dr. McKim Malville, the alignment is 10.5 degrees off (to the right of ) the solstitial axis; however, when this reading was made from a position across the ravine, neither of us realized that the axis of the interior (defined by the positioning of the stupa) was still further to the right, the total to amount to at least twelve degrees and perhaps a bit more. On the morning of the solstice (Dec. 22), I was able to determine the actual solstitial axis by recording, at exact sunrise, that the sun’s rays fell at 90 degrees upon the point of the farthest point of the curvature of the warped façade at the extreme left. It should be noted that Burgess, uncharacteristically, incorrectly “corrects” the misalignment of the stupa in his frontal elevation of the cave. (See Burgess drawings in Volume IV) His groundplan, however, is essentially correct and shows the dramatic warping of the façade.
66
volume iia, chapter three
Cave 26’s excavation had progressed to nearly (but not quite) the same degree when the new concern for a solstitial orientation imposed itself on that cave too. However, the task in Cave 26 was somewhat easier, since by chance it was already nearly aligned with the summer solstice—being a mere three degrees off.18 However, even though the stupa, before its carving was finalized, was brought both forward and to the left, and even though the relationship between the front caitya arches was adjusted, it appears that an absolute solstitial alignment between the dawn rays of the sun and the stupa was not quite fully achieved. (This is discussed at greater length below, in this same chapter.) By contrast when, a few years later, the local king, having aborted the development of the Asmakas’ caitya hall, took perverse pleasure in ordering a summer solstice-oriented caitya hall of his own, it appears to have been carefully laid out from the start; for by then— in 469—the solstitial orientation could be built into the plans. In Cave 19, the harried planners did what they could. The stupa, still with some matrix around it, could be “squeezed” to the left with great difficulty, but could by no means achieve a proper solstitial alignment. If one stands in the court and looks directly through the cave’s portico, the stupa, instead of being on axis, appears to be located very much to the right; and though there are distinct even if necessarily minor (leftward) adjustments to the shape of the dome and the positioning of the standing Buddha, they hardly make a difference in bringing the stupa into alignment, for its position and shape had surely been fixed too early to allow the deserved changes. The planners, twisting the whole hall around as much as they could, even though it had already been roughed out, had better success with the pillared portico, almost surely because it was reached later in the course of cutting than the stupa, the upper parts of which (including the umbrellas) would have been revealed when the cave’s vault was being reamed out. Particularly since the portico (like most elements) would typically have been blocked out with much extra matrix, it could be somewhat shifted into alignment. But even so, its adjustment is far from sufficient.
18 The (summer) solstitial axis, as measured by Malville, is 3.2 degrees off; but just as in Cave 19 (but more successfully) the cave has been adjusted to coincide with it.
cave 26’s complicated development
67
One might have thought that the plane of the still roughed-out façade could merely have been trimmed back at the needed angle, with great depth at the left diminishing gradually to a relatively thin wall at the right. However, there were problems. At the right, the stone could not be cut back sufficiently because of the obstruction caused by the presence of the niche-like inset near the right end of the front aisle; had that not already been in place, the hard-pressed planners could have defined the new plane of the façade as a simple angle rather than with its present twisted curve. (See Plan) Since Cave 19 is so influenced by the Hinayana Cave 9, one might expect that these two niche-like insets were originally intended to be small windows, as are found in that earlier caitya hall. However, since (being blind niches) their cutting could only have been started from within the front aisle rather than from the outside, it would appear that the planner decided that windows were unnecessary in this small cave with its great caitya window, and would interrupt the elaborate program of carving on the façade. And so (not quite renouncing the connection with Cave 9) they were “replaced” with these niche-like insets. No matter what their surely soon-forgotten origin, these would have been very useful features for there is nowhere else in the cave—which was originally intended as the devotional focus of the site, where ritual implements and other necessities could be stored. We might note that this retention of the “windows” (now niches) reflects the same dependence on the cave’s Hinayana model as do the vertical beams that were attached to pillars L1 and R1: namely, that the architect was “blindly” following the precedent of the Hinayana Cave 9. It is clear from the difficulty that the right niche-like inset created relative to the efficient adjustment of the façade plane in that area, had already been started when the drastic façade revision was authorized.19 That is, the whole of Cave 19, inside and out, must have been more or less fully roughed out (but by no means finished) prior to the subsequently necessary startling angling of the façade plane to attempt a solstitial connection, and this in turn preceded the addition of the cave’s rich decorative overlay, which clearly lies on the adjusted surfaces. 19 The wall of the adjusted façade, at the left, was of course far too thick for a window; but this was certainly never intended anyway, or the cutting of the window would have been started from the exterior.
68
volume iia, chapter three
Such observations require us to believe that all of the cave’s sculptural overlay (iconic as well as decorative) was concentrated into the last few years of the king’s patronage of the cave, by which time the painful reorientation of the cave would have been completed, to the degree possible. If we allow three years for this work, which of course was still not quite finished by the time that the king’s connection with the cave suddenly ended late in 471, then the (attempted) reorientation of the cave would have taken place sometime between 466 and 468. At this time it would have been already adversely affected by the presence of the niche-like insets, which we can thus assume had been penetrated in 466 or earlier, in the pre-adjustment phase of the excavation, which had been inaugurated perhaps as early as 462 but surely by 463. By contrast, in 471, both here and in the later areas of the related Cave 17, niches now came into use, being cut (in Cave 19) into the better-lit walls of the cells at court left. (The left front cell was partly destroyed by the Asmakas, but traces of the niche are still evident.) These niches and the related shelf in the right rear cell were similarly and unconventionally placed, on the better lit side wall, to take advantage of the light reflected into the pillared vestibules from the court. If we ask why Cave 19’s planner did not order a niche for the right rear cell, when such features came into fashion in 471, we might hazard the guess that the two areas were the responsibility of different contractors and that once the job on the right had been paid for, a revision would have involved a further unexpected (and unauthorized) payment. There are many features at the site where changes would have been desirable, but such changes may have involved administrative (and pecuniary) complications. As to why it was not adjusted later, after more funding was authorized, we can assume that time had already run out for the local king and his works, due to the Asmaka invasion at just this time. The pivot holes in the doorways of the intact left and right rear cells show considerable wear. This might at first suggest that Cave 19’s cells (like those of Cave 17) continued in use, for practical reasons, even after Upendragupta lost his control over the region. It is more likely, however, considering how the Asmakas “desecrated” the two front cells (see below), and otherwise dishonored Upendragupta’s beautiful hall, that the evident wear dates from the period of the site’s breakdown in 479–480. This is when the fine intrusive panels and related paintings were added to the vestibules and when (conventionally)
cave 26’s complicated development
69
the cells were plastered, which further reveals that they were in use at that time, and probably for a few years after. Cave 19’s right front cell may already have been converted into a special functional chamber by 471, by which time a doorway (or opening) appears to have been cut on the nearest (roughly east) side of Upendragupta’s small pillared chamber, Cave 18, making a direct and convenient connection between the cistern therein—“delightful to the eyes”—and Cave 19 itself.20 Thus the “sweet, light, clear, cool, and copious water” would have been brought from the cistern (Cave 18) through this converted cell and then through the converted cell doorway into the cave’s right pillared vestibule. The front right cell was somewhat extended on the east side, perhaps to provide a low platform for holding the water pots; such low platforms are found in many cistern areas. When the Asmakas brutally cut into Cave 19’s cells to make a passage to their own complex, they would have taken advantage of this convenient opening in the (assumed) “cistern chamber’s” wall, with which they aligned their new “destructive” cuts, one through the west wall of this same converted cell, and one through the corresponding east wall of the front left cell, and finally one through the west wall of the latter; this of course further suggests that the opening in the east wall of the cistern chamber was already there when the Asmakas took over and decided to make their passage. The now-unnecessary cell door on the north side of this front right “cell” (now better termed a “chamber”) was never hung; significantly, no latch was ever cut and there is no wear in the pivot hole cut into the upper projection. Since at this time (470–471) C mode doorways were always cut, it seems likely that the expected lower projection either was eroded away (like that in the left front cell); any traces are now obscured by the thick layer of cement covering the floor of the cell.21 Since it is generally true that door fittings
20 This wall and the (assumed) doorway have broken away, but the doorway must have been there to bring in the water to Cave 19. The quotations are from Upendragupta’s description of the cistern in verse 26 of the cave 17 inscription, located close to the cistern chamber. 21 It is just possible that the fitting in question was of the B mode, which has no lower projection. If so, it would date to 468; the fact that this cell would have been reached before the other court cells in the course of excavation would support such a view. It is of course also possible that the lower projection was intentionally cut away, because it served no purpose; but if so one would think that the
70
volume iia, chapter three
(as well as pole holes and niches) were cut only at the time that the doors were about to be hung, the presence of the (unused) pivot hole would suggest that the conversion of the right front cell to a cistern chamber was made just at the time that the doors were being fitted in the other three cells. Although it would appear that water was brought from the adjacent cistern in Cave 18 it is conceivable that an old cistern, now filled in, lies under the presently cemented floor of Cave 19’s converted cell. One might well expect this particularly important cave to have had such a cistern. The connection with Cave 18 also provided a convenient pathway between Upendragupta’s Cave 17 and 19, even though we must believe that another “external” path led from the great vihara outside Cave 19’s courtyard wall (which preserves traces of a railing motif ) to join with Cave 19’s now long lost staircase up from the river.22 This undoubtedly steep ascent led one into the court past the beautiful life-size guardian naga, which can be more easily seen from the ravine below; the other naga, at the left of the now blocked entrance, has fallen away. This “external” path from Cave 17, must have continued somewhat below the “barrier” at the left front of Cave 19’s court and then over to Upendragupta’s associated Cave 20, for it is unlikely that the king or his cohort (to say nothing of the resident monks) would have been required to go down to the river to get from Cave 19 to Cave 20. And of course the brutally convenient passage that the Asmakas made when they had taken over the site by 475 would have been unthinkable when Upendragupta was excavating the caitya hall.23 It would be of interest to remove the modern cement at the left of Cave 20’s courtyard too, for this area (now much broken away) almost certainly obscures another nearby, and associated, water
upper projection would also have been removed. Actually, it is evident that the unused projections were often left (being somewhat trimmed away if necessary) to strengthen the D mode fittings which later replaced them in many caves; see various telling examples in Cave 1. 22 Recently Cave 23’s cistern has also been filled with cement, sadly destroying useful evidence. Cave 7’s cistern (which Cave 8’s excavators broke into when they enlarged that cave) was probably cemented over many years ago, along with much of Cave 7’s courtyard. 23 It is of course possible that this portion of the scarp had fallen away by 475, facilitating the Asmaka revisions.
cave 26’s complicated development
71
supply.24 As noted elsewhere, the small platform at the left end of the porch plinth of Cave 20 is a characteristic cistern feature, and even has two circular depressions which must have been made to hold the water pots steady. Furthermore, Cave 20’s left porch cell, started in 465 or 466 (see Defining Features), well after the cistern would have been excavated, is untypically elevated, surely to compensate for the presence of the earlier cistern chamber, which would have extended beneath it. The porch-end cell at the right is also elevated, but much less so, almost certainly to create an esthetic balance with the left; its positioning can be seen as a kind of compromise.25 Along with cutting through Cave 19’s monks’ cells, and apparently forbidding the use of not only the focal Caitya Cave 19 but also Upendragupta’s Cave 20, the Asmakas appear to have discouraged the carving of the local king’s “signature” image—the standing Buddha with abhaya mudra—which was the focus of Upendragupta’s caitya hall, and was reflected in a few images from the Period of Disruption (mid-478–480), when the Asmakas had given up on the Cave 17 and Cave 20 shrine doorways. It is only during their patronage of the site, because of the developing demands of war, that such images flourish once again.26
472–474: The Hiatus Begins. Asmaka Invades Risika. All Work Stops at Ajanta. Remaining Workmen Leave for Bagh and other Peaceful Sites It is clear that there was no patronage activity whatsoever at the site for a short period (roughly 472–474) after both the local king and the emperor (the patron of Cave 1), had had to abandon their ambitious undertakings. This Hiatus—to which not a single chiselcut or brush-stroke can be ascribed—must be explained as the period of conflict when the Asmakas were struggling to take over the region;
24
For further discussion see Volume V: Cave 20. A similar compromise can be seen in the slightly different heightening of the porch end cell complexes in Cave 7. See Volume V: Cave 7. 26 A similar Buddha at the upper right on the 26LW shrine doorway is actually a walking Buddha. The only apparent exceptions are in the antechamber of Cave 7; they appear to be intrusions at first glance (or thought), but actually can be ascribed to the obsessive donation activity of the patron, probably all carved early in 478. See Volume V: Cave 7. 25
72
volume iia, chapter three
and their ultimate victory is evidenced by the resumption of work, in 475, on their long-abandoned (since 468) Cave 26 complex. At the same time, work was taken up again, as workmen rapidly returned to the site, on most other caves as well. By contrast, all further development of Upendragupta’s caves was now disallowed, and nothing further would be done to them until after the Asmaka takeover of the site, when a few minor functional changes were made to the still-occupied Cave 17, almost entirely in the interest of converting a few cells (L5, L6, and possible PL) to storage areas. It is important to recognize that there was indeed this distinct break in Ajanta’s development immediately after the Recession, for this short Hiatus represents the shift of control from the local king to the Asmakas. But we cannot just say it, we have to see it. We can of course observe that work on Upendragupta’s caves came to an abrupt end at some point midway in the site’s development, and we can also observe that the Asmaka caves, which had been long abandoned, had a relatively late “renaissance” at some point after work broke off on the former caves. But how do we know that the later Asmaka work did not merely succeed the final stages of work on the excavations of the Risika king, as if the excavators of Cave 26, coming to the site, merely passed the excavators of Cave 19 on the road as the latter were leaving?27 That is, how do we know that what I call the “Hiatus”—and blame on war—did indeed involve a distinct and decisive break in the site’s development? Particularly compelling evidence of a distinct break is revealed in the study of technology at the site, most notably the changing manner in which doorways, both main doorways but especially cell doorways at the site were fitted out, for these changes were specifically dependent upon how well a particular feature worked. That is, the criteria were by and large functional, rather than being influenced, as Buddha imagery, decorative carvings, and paintings might well be, by tradition and convention, which could complicate and confuse the issue. The particularly revealing development of door fittings at the site is discussed at length elsewhere.28 However, it will be useful to briefly 27 This is a purely figurative statement since, without doubt, many of the same workmen must have worked on both caves. 28 A brief survey of the development of cell doorway fittings (and particularly
cave 26’s complicated development
73
mention of the telling changes which separate the B mode in standard use just prior to the Recession (in 468) from the very different and far more practical D mode used when, after the Hiatus, vigorous work began again in 475. When most work at the site ended at the start of the Recession, we find (significantly) that many of the projecting monolithic pivot holders for the B mode had been revealed but had never been supplied with pivot holes, since the caves were still under excavation. Then, in and after 475, these obviously unused B mode fittings were totally abandoned in favor of the very practical new D mode, a type almost certain brought back from the Bagh caves, where many workmen had gone to find work during Ajanta’s Recession and Hiatus.29 The new “D mode” is characterized by a recess, containing the necessary pivot holders (both above and below), which was cut into the rear of the doorway. Very often the old and abandoned B mode projections, including a few which had actually been put into use, have been left as strengtheners when the door fittings were “updated” in this way. If we are looking for strata in the site’s development, this and various other examples involving the complex but logical development of the site’s various doorway fittings are particularly revealing. However, as is evident from the listing in “Defining Features” (see IV), door fittings are only one of a great number of other forms and features which can be similarly analyzed. The conclusion that there was a drastic Hiatus at the site (472–474), during which work stopped completely after a brief period of exclusively royal activity in 469–471 (which continued during the Recession), perfectly accords with the assumption that the Asmakas did indeed first menace and then finally take over the Ajanta region from their once friendly (or at least originally tolerant) rival, the local king. Even the compelling evidence that many of Ajanta’s workmen had migrated to Bagh (and presumably many other peaceful areas such as eastern Vidarbha, the Konkan (Kanheri caves), and Kuntala (Dharasiva caves) during these troubled times helps us to understand the political situation. For Bagh was in Anupa, part of Harisena’s inherited central domains, and (like these other areas) was not involved
revealing re-fittings) has been provided in Volume I, Chapter 3, and is developed in detail in a later volume under preparation. 29 The A+ mode doorways (with attached projections to approximate the B mode, and the slightly later C mode doorways, were also typically converted to the D mode after 475.
74
volume iia, chapter three
in the Asmaka/Risika strife, as its untroubled development confirms. And, as is evident when work starts up again at Ajanta in 475, some of these “interim” excavations influenced Ajanta, just as the migration of workmen from Ajanta had earlier influenced them. We know from the manner in which the various excavations closely abut one another during the Vakataka (as opposed to the Hinayana) phase, obviously to avoid any unnecessary waste of space, that there must have been strict administrative controls at the site. Such strong controls may also explain the surprising fact that during the Hiatus, when the consistent patronage of the caves stopped completely, there were no intrusions added whatsoever, even though we might well expect “uninvited” devotees to have taken advantage of the break to add their own votive images, as they did so flagrantly and so disruptively in the later Period of Disruption, when the site was on the verge of collapse and the former controls no longer functioned. Whether, because of the needs of war, there was a threatening royal prohibition against such merit-making private endeavors, or whether the resident monastic community had taken it on itself to protect the “abandoned” donations, we have a situation which is of considerable interest as a reflection of the concerns of those involved in the site’s care and development.30 Of course if, as I have argued, such intrusions would be put only in caves which were already dedicated, this would significantly reduce the areas available for any such uninvited offerings which might have been put there during the Hiatus, or even the Recession. This group would comprise those dedicated in early 469 (Lower 6, 7, 11, 15) and then, a few years later, Caves 17, 19 and 20. Various carved Buddhas were added to the court area of Cave 17; mostly generic seated images, but the major panel under the inscription is clearly a developed type, dateable to 479 or 480. Similarly, the presence of various carved bhadrasana Buddhas around the court area of Cave 19 makes it clear that they and all of the images associated with them belong to the Period of Disruption. The same must be true of the intrusive bhadrasana Buddha in Cave 20, while the two Buddha panels in the porch also are late in type. Thus it is fair to say that every intrusive image at Ajanta was put there
30 The “proof ” that the caves remained sacrosanct during both the Recession and the Hiatus is discussed in Volume I.
cave 26’s complicated development
75
in the Period of Disruption, which started in 479 and ended abruptly shortly thereafter—the end of 480 seems reasonable. There were never any intrusions cut or painted at the site at any earlier date, even in times of trouble or abandonment (discussed at length in Volume III on intrusions). 475–477: Ajanta’s Renaissance under Asmaka Feudatories In 475, with Asmaka control of the site now established, work was vigorously begun again, after nearly a decade, on the Asmakas’ Cave 26 complex and on other caves at the site’s western extremity. These other caves, notably Caves 21, 23, 24, and subsequently 22, 23A, 24A, and 28 were all apparently under Buddhabhadra’s patronage. This seems evident from the fact that the powerful monk obviously had to authorize the surprising penetration of the splendid Cave 24 into the upper and lower right wings of his Cave own 26 complex— a drastic decision which totally disrupted their originally intended development. He obviously did this to give priority to the richly conceived and very up-to-date Cave 24. However, such a drastic action would have been impossible if he had not been the “owner” of both of these caves. If anyone tears away half of your old kitchen to make room for a new garage, it must be you yourself, not your neighbor, who has decided on such a drastic action. The fact that, in contrast to the situation in other caves at the site, work proceeded in a relatively normal way in Caves 21 and 23 (as in Cave 26) throughout the whole difficult course of 478, further shows that these caves were clearly under Asmaka—probably specifically Buddhabhadra’s—control. Work on the grand Caves 24 and 28 did not, however, continue in 478; they were just abandoned, since they were both hopelessly incomplete at the time of Harisena’s death. Had work on them continued in 478, we would expect some evidence of rush (as in Caves 21 and 23); and there is none. By contrast, the attempt to complete Caves 21, 22, and 23 in 478 was highly expedient, and ultimately unsuccessful, proving that even these donations by the now ruling Asmakas were also afflicted by the terminal troubles which were starting to destroy the site. In Cave 21, the walls were plastered, in preparation for painting, before they (and some of the cell doorways) had been properly defined. Furthermore the antechamber brackets that had been started so
76
volume iia, chapter three
confidently in 477 now, in 478, were being peremptorily chiseled off, in recognition of the fact that the cave would never be able to be finished in the elegant fashion originally planned. But the fact that here (and similarly in Cave 23) much lower priority work was being done while work on the high priority Buddha image was progressing only slowly, shows that the sudden ending of Asmaka patronage was not sufficiently anticipated.31 Buddhabhadra and his associates obviously recognized that the future was dark, but they did not realize how quickly the darkness would descend. By contrast, in the “Vakataka” (i.e. non-Asmaka) caves, the entire focus at this time was on finishing the Buddha images. In all of the Asmaka caves, which occupied Ajanta’s western extremity, work proceeded with particular vigor from 475 on, now that the Asmakas, though still feudatories of Harisena, were in effective control of the site. The overlay of carvings on Cave 26’s main façade—all slightly too early to include any bhadrasana images— effectively masks the warping of the façade plane, a feature reflective of the inexperience of the workers who first revealed the façade of the great hall out of the cliff many years before. Inside, the disposition of the colonnades (not evident on Yazdani’s often reproduced plan) is subtly adjusted in an attempt to align the cave to the rays of the rising sun at the summer solstice, although here too there may be some effect of early inexperience.32 Even more obviously— but this is surely to better effect a solstitial alignment—the stupa has been noticeably shifted to the right and at the same time somewhat warped for the same reason. Equally significant, this appears to be the only caitya hall in India (out of a dozen surveyed) where the space behind the stupa is not the same as that to right and left.33 The reason for such an unprecedented forward adjustment must again be to pull it better into alignment with the solstice—a requirement obviously planned when the
31 In Cave 23, the antechamber pillars must have been started in 478, at which point the expected brackets were expediently omitted. 32 The plan of the Cave 26 complex is in Yazdani 1955, 16. In terms of the shape of certain areas, the width of walls and pillars, etc., it is a poor approximation of the wonderfully precise plans made by Burgess. Even these are almost always “improved” in new publications, where everything is straightened out, making the plans worse than useless. 33 This would be true in Cave 19 too, except for the fact that the surrounding pillars have been shifted both in position and (in the case of pillar L7) in size.
cave 26’s complicated development
77
excavation was still in its early stages, and such changes were still possible.34 The inner frame of the great caitya window has also been adjusted for the same reason; its width is significantly greater on the left than on the right, thus shifting the central point through which the solstitial axis would pass. Such an obvious break in symmetry— a feature so characteristic at the site—would never have been allowed without a reason. The whole interior is splendidly decorated, and with such subtlety that one is hard-pressed to see the very early original structure beneath the now highly elaborated pillars. Like the comparable porch pillars of Cave 2, all of Cave 26’s pillars had almost certainly been blocked out as “old-fashioned” octagonal forms some ten years before.35 Similarly, the newly introduced bhadrasana image is carved upon a stupa front which was not originally planned to contain it. This new type of image, probably begun in 477, like its impressive counterpart in the prime minister’s Cave 16, has already sponsored three small “offsprings” of the same new bhadrasana type—two in the frieze above the pillars, and one directly over the main stupa image. It is reasonable to assume that all of these bhadrasana Buddhas, along with the four major panels sponsored by Buddhabhadra in the ambulatory, were started shortly after the central image, which was cut from the stupa’s too narrow fronting projection, which had been roughed out a decade earlier, almost certain to contain a standing Buddha.36 Considering the overriding popularity of this new form, as it rapidly develops over the course of the next few years, it is interesting to note that these central Buddha images in Cave 16 and in the Cave 26 caitya hall are the sole examples of such shrine images started during the main phase of the site’s patronage. These two images were completed no earlier than 478, for Varahadeva had to rush to get his image dedicated in the first few months after Harisena’s death, and Buddhabhadra clearly made work on his cave a priority during the whole course of that same difficult year. But the impact of these
34
As discussed above, the problems faced in adjusting Cave 19 were far greater, because its original orientation was so far from the solstitial angle. 35 When finished and highly decorated after 475, they could not be provided with the conventionally late high square bases, without cutting too much of the forms away, resulting in an undesirable attenuation. 36 See Volume V: Cave 26 among many other references.
78
volume iia, chapter three
great images is immediate. Already in 479, during the Period of Disruption, the bhadrasana type has become the image of choice, to be used anywhere and everywhere, and the popularity of the type continues and greatly increases in the caves of the next centuries. The Asmaka takeover of the site by 475 and the brief stability which they brought to the region was responsible for a new flourishing equivalent to that of the seven years or so of Ajanta’s inaugural activity; but now the planners and the artists and the patrons had gained confidence from both praise and experience, and approached their tasks with a growing exuberance and facility. Although some of the lesser patrons apparently had to delay or reduce their efforts until sufficient workmen had returned to the region, work on all of the caves at the site now could begin again, except on the disallowed (and still unfinished) caves of the defeated local king. Even with this proscription, it is interesting to note that Upendragupta’s Cave 17, which had been occupied by monks since 471, continued to be used as a much-needed monastic residence, probably for purely practical reasons. Remarkably, when activity at the site was renewed in 475—a mere three years before Harisena’s death— the small and very unfinished Cave 11 was the only vihara other than Cave 17 where essentially all of the cells had been fitted out for proper residence, so housing was obviously a serious problem at this time.37 Of course monks (or workmen) may have utilized some of the cells in caves still under excavation, but if so they were in no way properly fitted out, and it is reasonable to suppose that they lived in the old Hinayana viharas too, even though it seems clear that there was no attempt to refurbish (or to re-fit the doors of ) those somewhat tattered old caves (and the Hinayana caitya halls too) until perhaps as late as 477.38 Considering this great need for residence cells in the expanding site, we might conclude that one of the reasons that Buddhabhadra gave such priority to the huge Cave
37 Nearly all of the cells in Cave 11 and Cave 17 show signs of early use. However, the “added” left rear cell in Cave 11 appears to have been finished and fitted out (in D mode) only after 475. Cell R6 in Cave 17 has C mode projections, but the pivot holes had not been cut, in order to fit the door, by the time that Upendragupta lost control of the cave. In Caves 8, and 15, a few cells were also fitted out prior to the Recession, while Cell R3 in Cave 1 was fitted out shortly after 468, although was probably not used for residence, but perhaps for storage. (See Vol. V: Cave 1). 38 See Part II for the refurbishment of the Hinayana caves.
cave 26’s complicated development
79
24, instead of merely reducing it in size to fit in between Caves 23 and 26, was that it was planned to house forty four monks comfortably (two to a cell), if the space at the rear had been fully finished and utilized. By contrast, the old Cave 25 (which Cave 24’s expansion “destroyed”) was originally planned to hold only nine, while Cave 26’s troubled Right Wing, just beneath, finally was able to hold only two, largely because Cave 24’s aggressive proximity; it had originally been planned for six cells.39 Death of Harisena: “End of 477” If we move on to the next stratum in our “dig” for truth, we come to the tragic moment that presaged the site’s destruction. But it is not Harisena’s death itself that concerns us here; it is its aftermath. In fact, it is the aftermath that defines the death, the type of death, and the time of death of the great emperor. And if this is hypothetical, it surely approaches facts very closely and convincingly. Even if he were deathly ill, why would the patron of the most splendid vihara in India not finish its elaborate program of painting, when (depending on the need for urgency) it could have been completed in hardly more than a week. There were great numbers of painters at the site at this (still happy!) time, and he (or his planners) could surely have ordered any that he wanted to complete his program. And why—even if the paintings were not completed, as was so often the case in other dedicated caves—did he not dedicate the image, get the merit therefrom, and bring the cave to life—the latter being a clear requirement for later intrusions, which are tellingly lacking in this particular cave? And why did he not force the brahmins in his court to rush his inscription to completion and send it down to the site, to proclaim his piety and his power? It is hard to explain these things except as the result of (the nature of ) Harisena’s death; and, when we consider how, during the coming weeks, all of the site’s worried patrons were totally obsessed with finishing their images—and (in general) the images alone—in order to dedicate them and make the merit, it seems apparent that Harisena’s death was very sudden—a mere blunt ending to his glorious career.
39
See the plan of the Cave 26 complex with conjectural reconstruction. Volume IV.
volume iia, chapter three
80
And the cause may well have been the same cause that ultimately destroyed the empire—namely, the machinations of the Asmakas, as I have explained elsewhere.40 We need only study the development of the site in the difficult year (478) subsequent to the abandonment of Cave 1, to confirm the assumption that the situation at the site now suddenly reflects the harsh reality of Harisena’s death.41 Even if one does not wish to believe that Cave 1 is Harisena’s cave and that its abandonment signals his death, the traumatic events of 478 themselves argue that the remarkable emperor had precipitously died. His demise and the consequent problems narrated in the Visrutacarita are reflected first in the rush to dedicate the “Vakataka” caves, and then in the notably different rush involved in getting the Asmaka caves dedicated. I have explained this in other chapters, first analyzing the situation in the “Vakataka” caves (other than Harisena’s own Cave 1) along the main scarp, and then that of the Vakataka’s rebellious feudatories—the Asmakas—who sponsored all of the excavations in the site’s western extremity. 478: Year of Anxious Consolidation. Revealing Contrast Between Manner in Which “VAKATAKA” Patronage and Asmaka Patronage Ends On “December 31, 477” the sudden death of the emperor Harisena immediately threw the site into a turmoil, for all too quickly the Asmakas, taking advantage of the ineptitude of Harisena’s successor Sarvasena III, had declared their independence. This sea-change in the political situation is evident from Buddhabhadra’s Cave 26 inscription which, written in the year after Harisena’s death, tellingly omits mention of a Vakataka overlord, praising instead “the mighty king of Asmaka” and his “excellent minister”. Considering this insult to the Vakatakas—who now had only nominal, rather than actual, control of the Ajanta region, it is hardly surprising that suddenly, in the
40
See Volume I, Chapter 4: In Defense of Dandin. The evidence of developments in 478 makes it is impossible to credit Cohen’s suggestion that Harisena might have been “the great and powerful Asmaka king” See my criticism of this “turnabout” of history in the following chapter “Cohen’s “possible histories”. 41
cave 26’s complicated development
81
months preceding this ominous announcement, the normal course of work was abruptly and dramatically affected on all of the caves on the “Vakataka” side of the site, as patrons hurriedly turned their attention to the completion and dedication of their shrine images. The Asmaka caves, at the site’s western extremity, were not immune to the troubles occasioned by Harisena’s death, but their problems were more economic than political, as pointed out elsewhere, and work in their caves, continuing for some months in 478 after the Vakataka patrons had fled the site, was by no means either so anxious or so rushed. And in all this activity, Harisena’s sumptuous Cave 1—the finest vihara in India—stands alone. It appears to have been summarily abandoned from the moment of the great emperor’s death—its beautiful image neither ever dedicated nor put into worship.42 As a result of the trauma now visited upon the site by Harisena’s death, and its political consequences, Ajanta’s established patrons were obviously driven to get their shrine Buddhas dedicated, an act from which particular merit must have accrued. The evidence of their hurried efforts makes it clear that although the situation was turbulent, there were still a few weeks or months of “grace”; for in every case where it was reasonably possible to bring their shrine images to some kind of expedient completion, they managed to do so, even if they could not properly decorate (or sometimes even properly prepare) the walls around them. Remarkably, not one of these late images (unlike that in Cave 1, finished the year before) was completed in normal course.43 And if this is true of the main images, it is easy to understand why the caves themselves remained invariably unfinished. This period of “grace” can be identified as the few months between the time of Harisena’s unexpected death and the time when the unruly Asmakas, scorning the Vakataka successor, Sarvasena III, rejected the Vakataka overlordship, a decision reflected in Buddhabhadra’s revolutionary inscription, which in the press of events can be dated to about the middle of 478. Like the other major inscriptions at the site, this elaborate composition must have been written 42 For the expedient and anxious completion of the images in the “Vakataka” caves, see Volume I: Chapter 11 (Understanding the Site’s Collapse: the “Vakataka” Caves.) 43 See Volume I: Chapters 11 and 12, on “Vakataka” and “Asmaka” donations.
82
volume iia, chapter three
in the patron’s capital, as its “political” content, as well as its sophisticated mode of expression would suggest. It probably came down to the site with an order that it should be placed on the main porch wall over the right aisle doorway, for its format precisely fitted it for that position. (However, the officials who, from afar, chose the location, apparently did not know that there was both a serious horizontal flaw and a serious vertical one, which might eventually make parts of the inscription illegible. So it would appear that a kind of compromise was made; the inscription was shifted to the right enough to minimize any problem from the vertical flaw (which was perhaps the more obvious at the time), but otherwise was placed where (we assume) it was intended, with deleterious results: verse 17, which appears to have provided significant information about the donor himself and his gift, is unreadable. Needless to say, the evidence which the inscription provides about connections between the site and the capital is highly relevant to our understanding. Similarly, the very magnificence of the clearly expensive complex only validates this connection. It seems evident that the Asmakas had not actually broken their feudatory ties with the Vakatakas until at least a few months, in the ominous year of 478, had already passed. If this were not true, we could not explain how Varahadeva, as the Vakataka Prime Minister, with considerable responsibility—he publicly claims in his Cave 16 inscription (verse 20) to have himself “governed the country righteously”—would have been able to rush his own great image to completion in 478, just as so many of the other patrons working in the main (or “Vakataka”) part of the site also tried to do. However, from the worried activity in the “Vakataka” caves early in 478, we can conclude that a deep fear for the future had afflicted the site from the moment that Harisena died, a fear which could only have been amplified by the probably dark cause of his death.44 Consequently, since the Vakataka patrons thought that time was short, they immediately shifted the focus of their work to satisfy their priority concern: the completion of the main Buddhas in their caves. However, no matter how intense their involvement, we can reasonably assume that, except in the Asmaka caves themselves, all of the anxious
44 For the possible Asmaka involvement in Harisena’s death, see Volume I, Chapter 4.
cave 26’s complicated development
83
patrons’ last minute efforts to complete their main Buddhas ended abruptly with the Asmakas’ assertion of independence in about the middle of 478. That Harisena died too precipitously to get his own image dedicated is “proven” not merely by the absence of a dedicatory inscription and by the lack of any evidence of ritual use in the shrine, but also by the fact that this most significant cave is bereft of any later intrusions. There are hundreds of such intrusive images at the site. Most of them date to 479 and 480, although in the abandoned caves of the disenfranchised Vakataka patrons, this takeover process probably began in the latter part of 478. But what is relevant to our present considerations is that these intrusions are notably absent in caves where the main image had not been put into worship, no matter how ideal the location might otherwise seem to be. This alone strongly argues that the Cave 1 image was never dedicated. By contrast, in the Asmakas’ Cave 26 complex, and in other relatively complete Asmaka caves nearby (Caves 21, 22 and 23), the overall programs of excavation and decoration continued, even if in some cases very expediently. It is evident that the frantic urgency that typifies the rush to complete the main images in the rest of the site was far less a factor in this Asmaka area, for the Asmakas were now controlling both the site and the situation. However, the expediency seen in the continuation of work in 21, 22, and 23 suggests that priorities were rapidly changing from pious to military concerns, and this assumption is supported by the fact that direct Asmaka patronage of the site ended—apparently abruptly—after this final year (478) of their own continued involvement. The magnificent Dying Buddha in Cave 26, poignantly recalling the death of Buddhabhadra’s patronage in 478, together with the splendid Temptation scene, the three beautiful “Sravasti Miracle” panels in the right aisle along with panel L8 at the back, and the Eight Buddhas (seven plus Maitreya) are the last great programmed images at the site, reflecting the greater care and creativity that Buddhabhadra expended on this particularly sacred hall.45 They also sadly announce the tragedy of
45 Since the mudplaster under the decorated left aisle ceiling—clearly part of Buddhabhadra’s program—penetrates into the space cut back for the Dying Buddha, the latter cannot have been carved later than 478. As if to confirm this, the limeplaster over the Dying Buddha is by no means continuous with that over the adjacent intrusion, as Cohen (1995, 190) declares: “. . . the same white plaster Spink
84
volume iia, chapter three
Ajanta’s ending, proclaiming as they do that a new sculptural force and inventiveness was just coming into its own, and its promise now would never be realized. If, like Mozart, Ajanta had not died so young (in 478, it was still in its early maturity) one is hard pressed to imagine what delights, as the rewards of devotion and of donations, it could have provided. As it is, to see the potential of the developing tradition come to a further expression, one must seek out the work of its heirs some two generations later at sites like Elephanta, or in such “transitional” sites as Kanheri and Jogesvari. Mid-478–480: Established Patronage Replaced by Intrusive Donations The dramatic starts and stops which Ajanta underwent in 462, 468, 469, 471, 475, 477 and 478, during the period of its consistent patronage, directly reflect the developing political realities which were to lead to its own downfall, and to the fall of the great Vakataka house. And the ensuing Period of Disruption (from mid-478 to 480) may give us still further insights into the deteriorating situation. The developments in the Asmaka caves at the western extremity of the site—and particularly in Buddhabhadra’s great Cave 26 Caitya hall complex—is particularly revealing; for it shows how, even in the caves of these newly “independent” rulers of the region, patronage was finally seriously disrupted, even if not as dramatically as elsewhere at the site. When we consider the community involvement in so many Buddhist sites (including Ajanta in its Hinayana phase), it is remarkable that the work at Ajanta between 462 and 478 was both elitist and exclusive. For more than a decade and a half, while the site flourished, there was not a single painted or sculptured image donated by anyone other than the major donors themselves, whose careful programs are never disturbed by private votive offerings made by anyone not a part of this exclusive and elite group. Remarkably, this is true even during periods such as the Recession and the Hiatus, when the site’s highly controlled development first drastically suffered and then, at least for a few years, stopped altogether. But from the time that Asmaka rejected its own feudatory status late in 478, and caused the major Vakataka patrons’ involvement in the site to be aborted, the gates were finally opened, and a spate of new and lesser donors
cave 26’s complicated development
85
started to sponsor hundreds of their own intrusive images. These were overwhelmingly carvings or paintings of the Buddha, but included some relief stupas, and a number of figures of Avalokitesvara as Lord of Travelers—the latter undoubtedly offered for protection upon the once flourishing but now-dangerous routes which those leaving the dying site would have to take.46 These helterskelter donations typically usurped the most desirable areas in the caves, as long as these locations were still available; but, as already mentioned, it is important to realize that they were added only to those caves which (unlike Cave 1) were already dedicated. Not surprisingly, most of these scattered intrusions were sponsored by the monks still resident at the site who, up to 479, along with the general public, had never been allowed to make a single donation of their own.47 So it is hardly surprising that, threatened by the rapidly darkening times, they now availed themselves of the chance to make merit—surely very cheaply—at this eleventh hour, when the remaining hoards of workmen had no other work to do at the site, and little choice of anyplace else to go. We need look only at the façade returns of Cave 26 to witness graphically Buddhabhadra’s loss of authority over his own cave. This was caused of course by the Asmakas’ shift of priorities from preparations for “the attainment of the fruit of supreme awakening” (Cave 26 inscription) to preparations for war.48 Because this was an Asmaka cave, and indeed the cave of the great Buddhabhadra, work had gone on here in a relatively normal manner in 478. However, by 479, when intrusive donors took over, neither the returns of the façade nor much of the ambulatory had been decorated according to Buddhabhadra’s plan. Therefore now there was what must have been a rush among these new “patrons” to claim the cave’s unused spaces and to make some last-minute merit while they could. Understandably those first to take advantage of the situation chose the best spots available for their unwelcome additions, which had essentially no connection with the decorative and iconographic program originally planned. deems to be programmatic was also slathered over several haphazardly placed ‘intrusive’ panels to the right . . .” But in fact the plastering is distinctly different, and if anything separates the adjacent carvings in time. 46 For discussion of Cave 4 panel, Volume V: Cave 4. 47 For justification, see elsewhere in this volume. 48 Cohen translation: Cave 26 inscription Verse 15.
86
volume iia, chapter three
It is reasonable to assume that the first of these intrusive images are the two large standing Buddhas carved high up on either side, just next to the main façade, which itself had been almost fully decorated shortly after 475, following Buddhabhadra’s commendably organized plan. The excellent visibility of these two intrusive images, their impressive size, their location adjacent to the cave front proper, and the fact that no previous intrusions hindered the appropriation of such important corresponding spaces, all add up to a weighty argument for their primacy. It seems likely that at the very start of this Period of Disruption (479–480), the financial situation must surely have been better than at the end. This might explain the impressive size of these two images, to say nothing of the unusual fact that they both have incised (as opposed to painted) inscriptions.49 The inscription under the large left image refers to “the religious donation of the Sakyabhiksu reverend Gunakara” and it would appear from the identical character of the remaining portion of the right inscription, and the similarities of the images and of their placement, that he was responsible for both, and surely proud of his achievement. One can by and large determine the sequence of the other intrusive images on these façade returns, most of which are carved. Intrusive painted icons, most of which have long since been washed away in the monsoons, due in part to the breakage of the façade overhang, probably filled the few related spaces around or between them. Significantly, there is only one bhadrasana panel on the cave front, and this “late” iconography, plus the fact that it was in a “low-priority” area, not visible (prior to the collapse of the porch roof ) from the courtyard, reveals it as one of the very last images in this area. Indeed, it is even later than the two small standing Buddhas to its left, for their presence has displaced it to the right, necessitating the unusual positioning of the farther bodhisattva at a right angle on the main façade surface. And the two standing images themselves must have been among the last panels cut on the façade return, for they too are in a visually undesirable position; that is
49 Only one other image at the site from this Period of Disruption has an (extant) incised record—this is a small standing Buddha in Cave 26’s right ambulatory, dating to 480—while a great many have cheaper and more quickly painted ones, not even counting those on this and other cave fronts that must have been weathered away.
cave 26’s complicated development
87
why they were squeezed as much as possible into an area where they would be marginally visible, leaving the otherwise useful surface still uncut and thus available for the bhadrasana group. The fact that the bhadrasana Buddha found in this very lowpriority location is the single example of this newly popular form, which from the very start takes over the ambulatory in 479 and 480, suggests that the intrusions on the cave front can all be assigned to 479, and were being finished at the time when the bhadrasana type started to be utilized for such intrusions. The many Buddha images along the fronting base of the caitya arch and within the vault also are of the standard earlier types; and since they, like those on the returns, are all fully finished, it is clear that all were completed before this Period of Disruption ended. This too would support the dating of all this abundant intrusive material on the façade to 479. By contrast, the alternating seated Buddhas and floral panels painted under the vault belong to Buddhabhadra’s own program, and probably date to 477 or 478, when the whole cave was being painted. However, looking just below, it may at first be surprising to see the obviously intrusive rows of Eight Buddhas, etc., carved just below at the springing of the vault, since no matter where intrusions appear, they never cut or cover any previous Buddha images. But the explanation is that these lower vault areas, which could not be seen from the courtyard, never were painted with images, they were merely covered with plain colors, if with anything at all. Therefore, having been left “blank” to save time and money, they were perfectly satisfactory areas for intrusions, whose donors would be willing to sacrifice visibility for any relatively auspicious location, particular one on the great caitya hall. A few images slightly higher up were carefully located only where the floral motifs—not the painted Buddhas— had to be cut away. This is not immediately obvious, in part because the mudplaster used to fill in a major flaw (in order to prepare the surface for the painting) has fallen away; but the situation can be reconstructed without difficulty. The low “wall” directly beneath the (now-screened) caitya window—again not visible from the courtyard— was also used for a crowded host of intrusions. By contrast, the nowexposed rear wall of the cave’s porch which (like the porch ceiling above) was clearly part of Buddhabhadra’s original painting program, shows no intrusions. Decorated after the porch doorways, dating to c. 477, were ornamented, it probably once showed the expected paired bodhisattvas flanking the main doorway.
88
volume iia, chapter three
The new “intrusive” donors, who started in 479 to take over the façade areas, were mostly monks, judging from the preponderance of their votive inscriptions throughout the site. Some surely had been previously associated with Buddhabhadra himself, even though his control over the cave had ended after 478. And it was only after all or most of the spaces on the cave front were taken that they turned their attention to filling up the ambulatory. This is suggested by the fact that the ambulatory’s intrusive iconography, compared with the intrusions on the cave front, is notably later in character from the start. All of the first intrusive panels cut in the right ambulatory—those in the more frontal, better lit, less flawed, and still available areas—show the Buddha in the new bhadrasana pose—a form so recently introduced that it never appears among the original carvings on the façade, and indeed appears but once and in the “latest position”, as noted above, among the façade intrusions. The first such bhadrasana sculptures ever made at the site—the highly important central images here and in Cave 16—cannot date before 477, and neither of the major, and indeed revolutionary, images could have been finished and dedicated before 478. The validation of this new and assertive type of image by such eminent patrons must have sparked its remarkably immediate and rapidly increasing popularity. This influential new type, which rapidly surpassed padmasana Buddhas in popularity, is lavishly rendered in the four image panels mentioned earlier: R2, R3, R4, and finally L8, all placed by Buddhabhadra in Cave 26’s ambulatory (along with the Parinirvana, the Temptation by Mara, and the “Eight Buddhas” panel) in 478. Despite the ominous rumblings of this final year of the site’s consistent patronage, the Asmaka rulers of the area were able to continue their excavation programs with a degree of confidence not shared by the donors more linked to “Vakataka” patronage, whose excavations are all located along the main (eastern) stretch of scarp. In fact these four fine bhadrasana panels “improve” upon their prototype on the front of the stupa, for this new image did not properly fit into its projecting frame, which had been blocked out a decade earlier for a different image altogether.50 Thus in the main
50 For the development of the stupa and its image, see Volume I: Chapter 12 “Understanding the site’s collapse: the Asmaka Caves”.
cave 26’s complicated development
89
image the expected flanking bodhisattvas, required by iconographic convention at this time, had to be relegated to hardly visible and constricted locations on the stupa drum, whereas there was plenty of room for them in the slightly later ambulatory panels, which could be provided with ample room. Thus it is reasonable to assume that these new types were started sometime in 479—probably relatively late in that year, since it was the great cave façade that first attracted the new donors, and was filled up first. Since all of the many intrusions on and in Cave 26 belong to a span of no more than two years, work on many of them must have been underway at the same time. Indeed, this makes sense when we realize, from the fact that there are many dozens of “intrusive” inscriptions at the site, how many eager donors must have been impatient to sponsor them—and how many eager out-of-work sculptors and painters were only too willing to oblige. By the same token, even if we recognize a distinct time-difference between the intrusive carvings around the front of Cave 26 and those in its ambulatory, it is very unlikely that there is a significant time gap between them; nor would this be possible given the pressure of events in this final troubled period. Thus it is likely that just as space was running out on the front—at about the time the intrusive bhadrasana Buddha (so undesirably obscured by the porch roof ) was underway—work was starting up on the intrusions in the ambulatory. Thus the single bhadrasana Buddha on the front, underway rather late in 479, may have been roughly contemporary with the first intrusions started by the eager new donor in the ambulatory. Sometime in 480, by which time the more desirable locations in the ambulatories had been taken up with this type of “Sravasti Miracle” panel, a significant change occurs, which must represent a significant doctrinal change. Now such large image groups, without exception, show standing Buddhas as attendants, rather than the attendant bodhisattvas which had been conventional for a decade. This placement of “multiple Buddhas” within the same composition reflects the very late trend at the site wherein Buddhas act as attendants to the Buddha, notably in the sculptural groupings of six standing Buddhas—perhaps the six Buddhas of the past—attendant up the “historical” Sakyamuni.51 At the same time a number of similarly
51
The Buddha image flanked by two Buddha attendants appears in Cave 19
90
volume iia, chapter three
late standing Buddhas—raised up on stemmed lotuses, and often with nagas beneath—also appear, as do a number of small-sized bhadrasana Buddha panels, which never appear in this reduced format until this very last year of work.52 The fact that small-scale bhadrasana images never appear in intrusive contexts until the very last year of activity at the site—480—is one of Ajanta’s many curious but dependable idiosyncrasies. Admittedly, (as mentioned above) three small bhadrasana images were carved as part of the original (as opposed to intrusive) programs in 477/478. Furthermore, a few appear in similar positions in the closely related Aurangabad Cave 3.53 However, when the Period of Disruption started in 479, although there were many small padmasana panels donated, all of the bhadrasana images were relatively large in scale. Perhaps the sculptors found it difficult to carve such images, with their projecting knees, in a small compass. But, knowing the rule of taste at Ajanta, it seems more likely that the smaller format type merely did not become conventional until this time. In any case, it forms a useful criterion for the dating of such intrusions. The fact that there are no unfinished intrusive images in or around the front of Cave 26 makes it evident that those areas had all been filled up before time began to run out. It is reasonable to date the façade additions all to 479, the first year that intrusions appear, at least in the Asmaka caves; in the caves of the patrons allied to the now-rejected Vakataka house, the first intrusions may well date from mid-478, since by that time most of those caves in the main area of the site had been rushed to dedication and then abandoned by their threatened courtly patrons. Thus these “Vakataka” caves were surely the first to be taken over for the random votive donations of the “intruders”. The finished character of all of these external images would itself suggest that they should be dated to 479 rather than 480, a con-
wall paintings in 470 or 471, as does the bhadrasana type. Often painting precedes sculpture in this way. 52 This curious “resistance” to the making of small, as opposed to large Bhadrasana images is characteristic throughout the site. 53 The outer face of the frieze over the right pillars has surprising last-minute substitutions (appropriate for this suddenly anxious year after Harisena’s death) for the slightly earlier (477) loving couples in the same position on the left side of the cave. This breaking of the planned symmetry emphasizes the importance of the new icon, as does its appearance at the top center of the shrine doorway.
cave 26’s complicated development
91
clusion suggested by the fact that there is only a single intrusive bhadrasana image on the cave front, as noted above.54 Further confirmation that this lone bhadrasana image should be dated to 479 rather than to 480, is the fact that it has bodhisattva attendants, rather than the attendant standing Buddhas seen in the late (i.e. 480) intrusions; the latter, not surprisingly, are found toward the rear of the right ambulatory—a very low-priority area which the new intrusive patrons avoided until better spaces had been used up, because here the light was bad and the rock very faulted.55 It is just when these latest intrusions were being carved, that we begin to find a significant number of unfinished images, for time was obviously running out and in fact rapidly came to a sudden halt. Almost the last of these must be the barely started standing Buddha at the very front of the right aisle wall. One would at first assume that this frontal and well-lit location would have been appropriated very early; but the fact is that it was not a good position at all, because the image would be obscured (and perhaps damaged) when the right aisle doorway was flung open. The great Dying Buddha in the other aisle is placed much closer to the front end of the wall for the simple reason that the aisle doorway there—conventionally—opens on the same (right) side, and so would not obscure or damage or “dishonor” it. By the end of 480 (according to my relatively precise chronology) this “intrusive” phase of work itself comes to a sudden halt. In the ambulatory of Cave 26 alone, even excluding images which may never have gotten painted, there are no less than twelve unfinished (carved) Buddha figures. These relatively simple images which, like the standing Buddha just discussed, could surely have been finished in a matter of days, are to be found in the least desirable areas in terms of location (often being at a higher level, as well as at the rear), lighting, and rock quality, confirming their very late date; and all show notably late iconographic features—features which change with a remarkable rapidity in the late phase of work at the site, and
54
Refer to previous. Panel R5’s space had not been utilized at first, because of bad flaws; and when the image was finally cut there, there was no room for carved bodhisattvas. However, following the demands of convention, the attendants would have been painted on the responds of the frame, as is often the case at the site. (Cave Upper 6 front wall, Cave 22 right rear). 55
92
volume iia, chapter three
help to justify our dating not merely on a year to year basis, but in fact require us to think in terms of “early” and “late” in years such as 479 and 480, which are so crowded with change. But we cannot think past 480, for sometime during the late evening or night of December 31st, it is clear that the site’s remaining carvers and painters decided not to return. The fact that all of these abandoned images, all to be found in the latest—i.e. least desirable—positions, must have been underway at the very same time, clearly shows how rapidly and urgently and anxiously the new patrons, eager to make merit while time allowed, had been filling up the cave’s ambulatory in the last few weeks of that fateful year. Thus the evidence of the insistent, and surely anxious, intrusive work on and in Cave 26—which could be confirmed by reference to such work on many other caves as well—proves that work at the site stopped suddenly at the end of 480. The presence of so many contemporaneous unfinished images in Cave 26 compels us to recognize this break as decisive; and this conclusion is supported by a review of the intrusive Buddhas in other caves, notably Cave Upper 6, Cave 22, Cave 21, and Cave 4 (shrine), as well as in the Ghatotkacha vihara, where the latest intrusive images were similarly left unfinished. By contrast, there are no unfinished images among the abundance of intrusive forms in Caves 2, 7, 11 (porch), 16, and 19, where the available space was filled up before the crisis occurred. The reason that these intrusive donations abruptly stopped—and any hope for the merit they would provide—must be that the huge insurrection against the Vakataka house, which the Asmakas were fomenting, was now moving from the stage of political intrigue to that of serious preparation for military action. Perhaps there was even a sudden flaring up of a local conflict in the Ajanta region, initiated by the banished Vakataka patrons who, through their benefactions, had developed such a connection with the region. It is, after all, most unlikely that the site could have continued to flourish, even in this rank and undisciplined way, in the shadow of the great troubles which were already beginning. And the site clearly tells us that it indeed did not flourish; it once again explains its own history. Thus, just as it is reasonable to see 478 as the year in which the Asmakas declared their independence from their overlords, it is reasonable to see the “Period of Disruption” (479–480, but starting in mid-468 in the “Vakataka” area)—when these “eleventh hour” donors were making merit while they could—as the time when the Asmaka
cave 26’s complicated development
93
rulers were developing their plans to take over the empire itself. This poisonous process must have already begun as soon as the vulnerable Sarvasena III acceded to the throne in 478, and may have started in local skirmishes even before the total impact of the eager insurrectionists came down with full force upon the anxious empire in the early 480s.56 In fact, the remarkably precipitous ending of work at the site— the fact that on the morning of January 1, 481 not a single artist showed up to continue work on the many images already underway—may be the first hint of the more global conflict that was to come.57 Since it is clear that the Asmakas were out to make trouble anyway, the simplest and most reasonable way for them to start was to first consolidate their hold over Risika (the Ajanta region) itself. They had already declared themselves its overlords in Buddhabhadra’s Cave 26 record of mid 478, but this is not to say that the so rudely disenfranchised Vakatakas had agreed to their own debasement. So it is reasonable to assume that it was in this battle, with its flaring start early on New Year’s Day 481, that the confrontation was resolved and Ajanta’s fate at last fully and sadly sealed.
After 480: All Artistic Patronage Ends. Some Monks Stay Briefly Ajanta’s patronage includes not only the carefully controlled productions which ended so urgently shortly after Harisena’s death, but also the vast spate of intrusions which so vigorously afflicted the site in the subsequent Period of Disruption. However, after 480, even these late scattered offerings had no progeny. One could assume that, bereft of support, the artists now moved away, probably returning to their villages, or searching (with difficulty now!) elsewhere for work, or possibly being conscripted to satisfy the mounting demands of the military. The manner in which the vigorously ongoing intrusive work ended in mid-course late in 480, would seem to reflect
56 See Volume I, Chapter 4, for the passage northward of Vanavasi and the other insurrectionists to the banks of the Narmada, a “troop movement” which might well have led right through the Ajanta region. 57 In other words, work decisively ended on December 31, 480, as pointed out earlier, with apologies for such impossible but useful specificity.
94
volume iia, chapter three
the trauma which brought this helter-skelter patronage to a decisive end—sounding a signal for the artists to depart. In the stormy climate of the times, it is reasonable to assume that most, or at least some, of the monks stayed on at the site after 480, for there was really no other place for them to go. It is clear that for at least a short while, a number of them continued to reside in the caves, presumably depending upon the villagers from the surrounding areas for their reduced support. We can tell by the evidence of significant wear in the pivot holes of doorways which had not been supplied with fittings until very late (c. 477), that some of the cells at the site must have been used as residences for a decade or so, well into the 480s. But after that, the chanting surely stopped. Although the caves were essentially abandoned by the end of the 480s, we should note that occasionally, over the course of later centuries, they were used as residence by a few sadhus and the like. In Cave 11 there are a few symbols such as tridents scrawled on the walls, and in both Cave 11 and Cave 7, the ceilings of the porches are badly smoked up. This was surely the result of cooking rather than of worship, and of course while the site was flourishing this would never have happened. Also in the Left Wing of Cave 26, Cells L2 and L2a (the most “prestigious” of the cells in the cave) were plastered by (or more likely for) some such occupant, using a thin surfacing which appears to be partly composed of cow-dung, which was probably never used in the more generous plastering of the cells during its Vakataka phase. The proof that this plastering has nothing to do with the original work in Cave 26LW is to be seen in the fact that this thin plastering is slathered right up into one of the long-abandoned pivot holes of the doorway. By the same token, the thick mudplaster in the cell at the left end of the porch of Cave 11 once covered (and still partially covers) the broken-off teak wood “pins” which affixed the wooden projection which had been added in 468 to “convert” the primitive A mode doorway to the more up-to-date projecting B mode. This is clear proof that the (5th century) door had fallen away by the time the sadhu moved in, although we cannot tell whether he took up residence some years or some centuries after the site’s abandonment. A similar mix of mudplaster was also put in all of the other porch cells, suggesting that a number of sadhus might have made it their abode. However, the telling evidence provided by the cover-up of the door-fitting pegs
cave 26’s complicated development
95
is not available in these other cells, where the mudplaster (if it once covered the stumps of the pegs) has fallen away.58 It seems likely that the Asmakas, who destroyed the Vakatakas and Ajanta too, must have been hoist by their own petard. They may have been so weakened by their own insurrection that they were unable to continue their patronage even after the war was done. The fact that even their late excavations at Aurangabad, developed (but never finished) during the heyday of their patronage between 475 and 478, were located closer to their own center of power and yet were never taken up again, would seem to confirm such an assumption.59 Just possibly the Asmakas’ power was ceded to another ruler, after the great insurrection, and the new authority controlling the region was Hindu in its orientation rather than Buddhist. One suggestion that indeed the Asmakas may have lost control of the region is to be found in an elusive and now nearly illegible inscription carved on the wide wall between Cave 26 and its left wing.60 Scholars generally agree that this long and highly visible record belongs to one Nannaraja, a Rastrakuta king ruling (possibly) in the late sixth century. But it is in no sense a donative inscription, nor does it have any discernible connection with the site other than the fact that it was cut on the convenient and already smoothed wall between Cave 26 and Cave 26LW. It must be an advertisement of power by a king who was probably ruling over the Ajanta area at this time. When we ask what Nannaraja was doing here in the old Asmaka domains, perhaps a slight light emerges from a reference to an earlier Rastrakuta king, Mananka, who “frightened the countries of Vidarbha and Asmaka by his policies”.61 Admittedly Mananka is generally considered a fourth century king but (even if this is true) it could suggest a history of conflict, as well as contiguity. What is far more certain than this is that there is no way that this non-donative inscription
58 Presumably this particular matter could be solved by Carbon 14 dating. With the kind help of the Livermore Laboratory, who analyzed two seeds (part of the plaster mix) in Cave 21’s Cell PR, we were able to confirm a fifth century dating for that cave—so often considered much later. 59 See Volume I: Chapter 14: “Related Caves of the Vakatakas or their Feudatories” 60 See Yazdani, 1955, Vol. 4, 121–124. 61 For a discussion of this inscription, see Malandra 1982.
96
volume iia, chapter three
can be used to “testify against Spink’s assumption . . . (that) all work at the site should have come to a halt almost immediately after Harisena’s death.” (Bakker 1997, 41) Rather, the record is more likely a self-aggrandizing record of a visit, a bit like that of the tiger hunter who came upon the site by chance nearly two centuries ago, and scratched across the chest of a painted Buddha on pillar R13 in Cave 10: “John Smith, 28th Cavalry, 28th April, 1819”.62
62
Dhavalikar 1968, 147–153.
CHAPTER FOUR
COHEN’S “POSSIBLE HISTORIES”
In a recent article, Richard Cohen has made the not unconventional statement that “Spink has gone wrong in his historical reconstruction (of developments at Ajanta)”.1 Cohen’s arguments are arresting as an intellectual construction, partaking of the playfulness that so often recommends a significant conceptual undertaking. And since they are presented with conviction, they are likely to satisfy the many scholars who think that my own theories are “off the wall”.2 However, since Cohen’s arguments are unconvincing as an archaeological construction, it seems necessary to suggest their weaknesses here, at the same time stressing “The need for study in situ”. Cohen (1995, 51) finds the “linchpin of Spink’s historical reconstruction” in the use of “verse 10 from (the Cave 17 inscription) to document a set of stormy relations between the Vakataka, Rsika and Asmaka dynasties.” The reference is to Mirashi’s reading of verse 10 of the Cave 17 inscription. There (according to Mirashi) the royal donor of Cave 17 speaks of “having subjugated prosperous countries such as Asmaka” sometime in the past. Cohen (1995, 47) then criticizes my use of “Mirashi’s translation of one verse to substantiate a history of animosity between Cave 17’s donor and the ruler of Asmaka.” In Cohen’s very different translation (1995, 48), the “foe’s identity, name, and title remain unknown”, while Asmaka could be a reference to a battle location only. Cohen (1995, 47) continues to insist: “This revision of a single verse . . . has greatly problematized Spink’s telling of Ajanta’s history. Without Mirashi’s reading . . ., there is no indisputable proof for conflict between these powers before Ajanta’s inception; the hypothesis that work on Caves 17 and 19 stopped in 471, never to resume, due to Asmaka’s revenge becomes insupportable.” 1
Cohen, unpublished at time of writing. Khandalavala (1991, 102) speaks of “the far-fetched attempts, with no evidence in support, which have been made by Dr. Spink” in various regards. 2
98
volume iia, chapter four
But I certainly never mentioned, or even thought of, “revenge” as a significant motive behind Asmakas’ aggressive search for land and power. And I certainly never considered Mirashi’s reading a “linchpin” in my historical reconstruction. In fact, in my very specific “Reconstruction of events related to the development of the Vakataka caves” I do not even mention (Spink 1991A, 96) this crucial “linchpin” which Cohen has brought into the argument. Mirashi’s arguable reference to Asmaka’s “subjugation”, whether his reading is right or wrong, is irrelevant to my historical reconstruction. Just because we defeated Japan some years ago, does not mean that they are preparing to revenge themselves by attacking us now. History depends upon its own unpredictability. The truth is that this verse—no less disputed by scholars after Cohen’s new reading—is the “linchpin” not of my reconstruction but of Cohen’s own. This is because his novel redoing of Vakataka history is invalidated from the start if Mirashi’s, rather than his own, reading of the verse is correct. According to Mirashi’s reading, the local feudatory king, the donor of Cave 17, defeated the Asmakas, who were also feudatory to the Vakataka emperor—“that moon among princes” (Cave 17 inscription, verse 21).3 But Cohen (1995, 62) aims to upset the conventional cart with the assumption “that the Asmaka in which Cave 17’s donor fought his enemies was actually part of Harisena’s personal lands. It is, in fact, possible that the great and powerful Asmaka king to whom Cave 26’s inscription refers is Harisena himself. (Italics Cohen’s) The Asmakas referred to in the Caves 17 and 26 inscriptions may be the Vatsagulma Vakatakas themselves.” Obviously, if the Vakataka emperor Harisena was “the great and powerful Asmaka king”—Asmaka being understood as a geographical designation according to Cohen—Mirashi’s reading of this troubling verse would make no sense, for the patron of Cave 17 would certainly not be “subjugating” the very overlord whom he praises in his inscription. Although Cohen’s hypothetical identification of the great Vakataka emperor with the king of Asmaka demands a drastic revision of Mirashi’s reading of the verse under dispute—and without such a 3 This presumed Risika/Asmaka conflict occurred before the death of Upendragupta’s brother sometime before Ajanta’s. These adjacent territories, both inherited by the emperor Harisena, were clearly at peace when they cooperated on the inauguration of the site, even though by 468 this was no longer so.
cohen’s “possible histories”
99
revision collapses—the identification actually derives from his new and different interpretation of the contents of the eighth ucchvasa of Dandin’s Dasakumaracarita. Scholars in general recognize the famous tale told in this section of Dandin’s book as a historical reflex, written in the early seventh century, recalling the fall of the Vakataka dynasty a century and a half earlier. However, few credit its reliability, or approve of my insisting upon the remarkable way that it reflects and describes the calamitous disturbances in Central India precipitated by the emperor Harisena’s death and the consequent destruction of his empire. The explication of this account as a very detailed resume of this “unforgettable” event was first seriously attempted by the great V.V. Mirashi (Mirashi 1945, 20–31). Cohen (1995, 55) rightly states that “Spink’s reconstruction of the events related to the development of Ajanta’s caves is in large part a reworking of ideas presented by Mirashi”. However, Cohen fails to mention the implications of my shifting of Harisena’s regnal years from Mirashi’s (estimated) c. 475 to c. 500 to a more precise c. 460 to c. 477. This in fact makes all the difference. Only such a revision allows a series of important synchronisms to lock the events described convincingly into place, linking the account with historical realities in a way which is impossible if one follows Mirashi’s now traditional and now conventional dating. Indeed it is Mirashi’s arbitrary (but unfortunately incorrect) dating of Harisena’s accession to c. 475 (rather than c. 460) which has kept his otherwise brilliant reconstruction from being accepted, or acceptable. I have discussed elsewhere the many compelling historical synchronisms revealed by a revised dating of Harisena (Spink 1991A), but will refer to a single example here, to show the difficulty which Mirashi’s dating for Harisena’s accession necessarily causes. The Visrutacarita tells us that the emperor Harisena’s two grandchildren were taken for safety, at the time of the Vakataka fall, to the city of Mahismati, which their uncle was still controlling in the turbulent years after the great emperor Harisena’s death at the end of 477. However, we know that this flight to Mahismati had to take place sometime before the mid 480s, since in 486 or shortly before, according to the Barwani inscription (Mirashi 1955, 17–19) Maharaja Subandhu was ruling over that city. Thus this one (of many) examples makes it clear that Harisena must have died some years before, and invalidates Mirashi’s suggested regnal dates of 475 to 500, at
100
volume iia, chapter four
the same time validating our new dating to c. 460 to c. 477. In this regard it is relevant to note that Mirashi himself merely suggested assigning Harisena’s reign to “475–500”, along with his father Devasena’s reign to an equally neat 450–475. He utilized this hypothetical dating as a reasonable working hypothesis, in the absence, in his view, of reliable landmarks which might better establish it. The consequent assumptions, which have in fact so long distorted the conventional understanding of developments in the fifth century, have been caused by the unquestioning acceptance of Mirashi’s hypothesis by those who, for one reason or another, have not questioned his authority. If we are going to put the house aright, we must recognize that the old theory, no matter how generally accepted, is wrong. Mirashi’s extraction of historical data from Dandin’s famous Visrutacarita—the eighth chapter of the Dasakumaracarita—where he links the characters and actions in the story with the goings-on in the court of Harisena and his successor, seems less problematic. But here too Cohen sees a very different “reality”. Instead of identifying the great emperor Harisena with the Visrutacarita’s noble Punyavarman, and his weak son and successor Sarvasena III with the unfortunate Anantavarman, both ruling (as suggested by the Visrutacarita) over a finally unified empire, Cohen rocks the boat of convention by suggesting another “possible history”, played by different actors, in what becomes a very different drama. This intriguingly perverse but arresting reconstruction is thrown “as a sop to those (i.e., Mirashi and myself ) who swear by the (Visrutacarita’s) historicity.” (Cohen 1955, 77). If one credits Cohen’s (1995, 65) suggestion, then ”the virtuous Punyavarman becomes identified with Pravarasena II, a renowned king under whose reign this branch of the Vakataka family reached its pinnacle. Punyavarman’s weak-willed son, Anantavarman, will be equated with Narendrasena. Anantavarman’s son, Prithivisena, . . . would be the Visrutacarita’s Bhaskaravarman. The evil Asmakas will be Harisena’s own family. In fact, the scheming Candrapalita (viz.: should be Indrapalita) could be equated with Hastibhoja, Varahadeva’s father and Devasena’s prized minister. The Asmaka plant (or “mole”), Indrapalita (viz.: should be Candrapalita) who corrupted Anantavarman, could be Cave 16’s Varahadeva himself.” Although Cohen (1995, 69) takes the (very dubious!) position that “this interpretation fits the available facts with much greater fidelity than that of Mirashi”, he sees this schema of his own, like the alter-
cohen’s “possible histories”
101
nate reconstruction of Mirashi and myself, as “only one possible history” of the site—as only one possible way of filling in the blanks. Thus Cohen peoples Dandin’s drama with a set of actors totally different from those hired for the purpose by Mirashi and myself and in effect rejects them both, insisting that we do not have sufficient evidence to construct any valid history whatsoever from the evidence at hand. But the evidence at hand is in the caves themselves, when it is analyzed with a necessary precision. Indeed, Cohen (1995, 27) accepts “the general outline of (my) chronology for Ajanta’s Vakataka phase”, recognizing the various crises which define the site’s brief and turbulent development; but this outline alone provides merely an overview. We must dig beneath the surface to test Cohen’s somewhat pernicious hypothesis—his suggestion that there can be no final truth emerging from an analysis of the site. To counter such a position we must analyze specifically the circumstances attendant upon each of Ajanta’s moments of crisis or of significant change. Only then can the truth emerge. Cohen’s nihilistic conclusion—his directorial decision that the parts of Ajanta’s great drama could have been played by one troupe of actors as convincingly as the next—is hardly justified by the wealth of data awaiting any scholar willing to give the richly laden site the attention it deserves. But any such investigation must perforce go well past the gross pattern of development—the simplistic template upon which Cohen has laid these various “possible histories”—both his and my own and, by implication, other conceivable ones as well. Although Cohen does in fact follow my “Short Chronology”, he then posits different forces at work during the site’s development, to create his alternative “history”. But we cannot get at the truth—or even decide that there is no truth to be discovered—so easily. We must explore the deeper effects that each of the moments of transition—particularly the recurring crises—had on the life of the site, and then must base our judgments upon these more complex realities. We must get down to the specifics lying beneath the surface of change, for the evidence is in the details; and the details are in the caves themselves. What I shall do is specify each moment of crisis or change, as the site develops over the brief course of its history. Then I shall show in each case how the responses of the different patrons validate or—in the case of Cohen’s reconstruction, invalidate—the suggested
102
volume iia, chapter four
political situation. As for my own theory, it assumes throughout the aggressiveness of the Asmaka feudatories, whose threatening stance caused them to be expelled from the region by the local feudatory king in 468. By 471 , however, they had disrupted the latter’s rule over the region, and by 475 they had taken over control of Ajanta themselves, as part of their long-term aggressive goals. Patronage flourished again from 475, until the death of the Vakataka overlord, the emperor Harisena on “December 31, 477”.4 The Asmakas then almost immediately (in mid-478) rejected the overlordship of Harisena’s weak successor and started to prepare for the military take over of the empire. This rapidly drained the site of its resources, so that by 480 its patronage had completely ended. Cohen’s summary of Spink reconstruction of Ajanta’s History: Richard Cohen (1995, 45– 46) has succinctly outlined my reconstruction of Ajanta’s history (without necessarily agreeing with it—or I with him on every point). The reader should find his summary useful, so it is included here. * Period 1: Early 463–late 468 Situated along a major North-South trade route, the region surrounding Ajanta is an object of contention between the Rsika dynasty to its north and the Asmaka to its south. Some time before 462 the Rsikas take decisive control of the region, and with the support of the new suzerain, the Vakataka overlord Harisena, they enforced peace. Whereafter, the Rsika king (christened “Upendragupta by Spink), Harisena’s prime-minister (named Varahadeva), a monk named Buddhabhadra with ties to the Asmakas, and other patrons of unknown affiliation serially initiate the excavation of cave monasteries for Buddhist monks. * Period 2: Early 469–late 471: Asmaka attacks Rsika This conflict stops work on all caves except those sponsored by the Rsika king and Harisena himself. Many workmen go to Bagh several hundred miles to the north; the decorative motifs of Bagh’s earliest phase are the same as Ajanta’s second.
4 For a justification of my usefully arbitrary dating, see Volume I, Chapter 3 The dates are of course (close) approximations. I speak of full years in order to simplify the sequence of developments.
cohen’s “possible histories”
103
* Hiatus: Early 472–late 474” The Rsika/Asmaka conflict heightens and all work at Ajanta stops. It is important to note that during this period not only did no programmatic work continue on the caves, but no intrusive Buddha images, no “graffiti” as it were, were added either. This highlights an important facet of Ajanta’s patronage. Namely, as long as a donor maintained an interest in the cave for which He paid, that excavation was treated as his exclusive properly, and was not available for alteration of decoration by anybody else. A clear demarcation can be made between “programmatic” and “intrusive” periods and iconographies at Ajanta. * Period 3: Early 475–late 477 Asmaka wins this time around. Work resumes in earnest, this time under Asmaka supervision. This period ends with Harisena’s probably sudden, probably unexpected, death. * Period 4: Early 478–late 478: The shock of Harisena’s death, the evident weakness of his successor, and the recognized ambitions of Asmaka to destroy the entire Vakataka empire impels Ajanta’s patrons to hurriedly complete and dedicate their caves. * Period 5: Early 479–late 480: Asmakan machinations devastate the Vakataka polity; Harisena’s empire is rent in “civil war”. Original patrons give up their control, enabling the monks and artisans still living at Ajanta, as well as traveling merchants, to commission images in the caves already excavated. Eventually, the long and bloody war renders the Ajanta pass impassable. The monks leave Ajanta in search of support, for the original patrons of the caves no longer maintain their establishments and travelers no longer come near. The site is abandoned. In opposition to my own view, Cohen considers aggressors responsible for the troubles of the site and its perhaps final demise to be the so-called “main” or eastern branch under king Prithivisena II. Although Cohen (1995, 70) admits that “no direct evidence exists for this hypothesis”, he suggests that Prithivisena may have waged war against Harisena’s western branch in order to recover territories previously lost to the latter. Although the evidence at the site proves that Prithivisena did not succeed in taking over the Ajanta
104
volume iia, chapter four
region, Cohen suggests that his aggression at least affected the site’s development, whose problems culminated in the death of the great emperor Harisena. During the reign of Harisena’s successor Sarvasena III, the problems were exacerbated and eventually, for reasons which (Cohen suggests) cannot be clearly understood, Ajanta’s patronage ended completely. This summarizes Cohen’s view. As noted above, to Cohen these opposing scenarios are merely two “possible histories” of Ajanta, either of which could satisfactorily connect with the evidence of the Visrutacarita. It is of course very possible that in situations such as this the “real” truth cannot be finally determined. But I would insist that this is not the case here, and that this will be clear once we get deeply into the available evidence. This evidence must go well beyond the recognition that the various moments of crisis or change at the site are characterized by the disruption of its patronage, or alternatively by its starting up again. We must see the effects on individual donors, and recognize who they are, for this will bear directly upon the question of whether the aggression at the site was at the behest of the feudatory Asmakas or of Prithivisena, Harisena’s hypothetically contentious relative. We can start with Ajanta’s first crisis, using here as in all later cases, my hypothetical dates, which Cohen accepts. (For a justification of the specificity of my dating see Volume I, Chapter 2) I should stress again that the argument is not concerned with the pattern of development of the site, about which Cohen and I by and large agree; rather, our disagreement reflects our opinions about the very different forces responsible for this pattern of development. The Asmaka Expulsion (End of 468) and the Recession (469–471) The first crisis took place at the end of 468 when, in my view, the aggressive Asmakas were expelled from the region by the local king who, concerned about the Asmakas’ future military intentions, drastically cut down or cut off much of the ongoing course of excavation at the site, initiating what I have characterized as the “Recession”. (See Time Chart) According to Cohen’s hypothesis, this same disruption of activity at the site may have been caused by the invasion of the eastern branch of the family, attempting to restore some of their lost domains.
cohen’s “possible histories”
105
What is significant for our considerations is that the caves, all connected with a number of different patrons, were affected in different ways by the crisis responsible for the Recession, which started on “January 1, 469”,5 only a half decade or so after most of the caves at the site had been begun. There is a clear—and understandable—distinction between what happened in the case of the courtly caves and those of lesser status. Work continued vigorously on the three already developing excavations of the local king— Caves 17, 19, and 20. Indeed, despite the troubled nature of the times, Upendragupta even began a major new excavation; this is caitya Cave 29, which is placed at a higher level, since by this time every available space along the main surface of the scarp had already been taken.6 Although Cave 29’s development was soon aborted, due to the mounting troubles of the times, it was certainly intended to “replace” the nowabandoned Asmaka caitya Cave 26. At the same time work progressed vigorously on Cave 1, which I have elsewhere insisted (Spink 1981, 144–155) must be the donation of the Vakataka overlord, Harisena himself. Furthermore, a somewhat normal course of work continued on the overall excavation and decoration of the prime minister’s Cave 16 during 469, even though surely, before the end of that year, work would have to be abandoned, again due to the troubles of the times. This “elite” activity was notably different from what happened with the caves of the site’s lesser patrons, or of the caves at the western (Asmaka) end of the site. In all of these cases—some seventeen!— the progress of work was cut off abruptly, except in those four excavations (L6, 7, 11, 15) where excavation had progressed sufficiently to get the Buddha image expediently finished and rushed into dedication within the first months of the Recession.7 In these four cases, the sumptuary restrictions, which must have been imposed by edict of the local king, were briefly relaxed, so that the Buddha images
5
See previous note. Cave 29’s anomalous numbering is explained by the fact that it was still covered with debris when the caves were put in their present sequence. Both Cave 29 and Cave 26 appear to have been oriented to the summer solstice. 7 Of course the prestige of the patrons (as in Cave 16) may have been a consideration; but of the four only Cave Lower 6 was a particularly ambitious undertaking, at least as it turned out. 6
106
volume iia, chapter four
could be hastily completed.8 This compulsion to get such images done and dedicated clearly supports the conclusion that this was the primary mode of obtaining merit at the site, as the patrons’ actions in later crises will confirm. However, at the time of this first crisis, in late 468, there were at least two and possibly three other excavations in which the anticipated Buddha images could also have been hurriedly added, but were not, even though their locations had already been prepared— indeed more fully prepared than was generally the case in Caves Lower 6, 7, 11, 15. This obviously has to be explained. The locations in question are the crucial central stupa in Cave 26, whose front projection had already been roughed out in preparation for an image by 468; the anomalous shrine-projection which had been expediently (and anomalously) cut in the still-rough left wall of the right wing of the same caitya hall, and the shrine of the caitya hall’s left wing, where the shrine must have been well underway since all of the surrounding cell doorways (revised after 475 to the D mode) had been cut (in the A mode) by 467. It might be noted that these were all part of the great caitya complex which the monk Buddhabhadra made “in honor of Bhavviraja who served the mighty king of Asmaka as the latter’s minister, (who was) attached to him in friendship through many successive births” (Cave 26 inscription, verse 9). Although the proof of my theory of a mounting Asmaka aggressiveness is cumulative, becoming increasingly evident as we move from crisis to crisis, the evidence provided by these three “missing” images in the Cave 26 complex is telling from the start. This is because the three excavations under consideration are all Asmaka undertakings, as we know from the presence of the Cave 26 inscription in the same cave complex. And of course, although these three shrines all have Buddha images now, these were all added much later, when the Asmakas had turned the tables and had gained possession of the site. According to my view of the site’s development, the local feudatory king expelled the threatening Asmakas in 468. Since this brought the work on their caves to a total halt, we can understand why the
8 For the special case of the Prime Minister’s Cave 16, see Volume V, Cave 16; also Volume I, Cave 16.
cohen’s “possible histories”
107
three shrines in the Cave 26 complex, unlike Caves Lower 6, 7, 11, and 15, never were supplied with images nor hastily dedicated at this crucial point in time. Indeed this decisive and surprising action of the local king, who allowed only his own caves and those of the emperor and prime minister to continue in a somewhat normal manner, must have had its source and justification in his fear of his aggressive neighbors. Thus the failure to add the expected Buddha images to these already prepared shrines in the Asmakas’ Cave 26 complex, to say nothing of the insistent abandonment, in 468, of all of the Asmaka excavations then underway, supports the conclusion that the Asmakas, although at peace with the local king when work at the site began, all too soon started making trouble, probably because of their territorial ambitions. We shall see how their story plays out when we discuss later crises, all of which have their ultimate source in Asmaka aggressiveness. If the sudden abandonment of the Asmaka excavations late in 468 supports the view which sees them as aggressors, the same evidence clearly runs counter to Cohen’s alternative “history”. If, as he suggests, the Asmakas (those living in the region of Asmaka) were actually the western Vakatakas, ruled by the great Harisena, how can we explain the total abandonment of the Asmaka (i.e., what Cohen calls “Vakataka”) caves at the western extremity of the site in 468, when the less prestigious donors of Caves Lower 6, 7, 11, and 15 were able to get their Buddha images done and dedicated early in 469? Also, since both the major dedications of the local king (Caves 17, 19, 20, 29) and of the Vakataka prime minister, to say nothing of the great Cave 1 of the emperor himself, proceeded vigorously (at least at first) after 468, the fact that nothing at all was done in the site’s Asmaka complex again controverts Cohen’s view that the area was the domain of the Vakataka overlord. If the eastern Vakatakas were indeed the aggressors, as Cohen would have us believe, they surely never took over the site during these early years, since work continued in varying degrees upon a number of the caves (as noted above) throughout the period of the Recession. Indeed, if the eastern Vakatakas were in fact attacking the region, as Cohen proposes, one can hardly believe that the local king and the emperor Harisena himself (Cave 1) would be working consistently on their excavations in this troubled period. It is certainly more reasonable to see Asmaka as the aggressors who, having been
108
volume iia, chapter four
expelled by the local king in 468, were marshalling their forces with a view to taking over the region (ancient Risika, wholly or in part) as soon as possible. It was a goal that was soon to be accomplished. The pattern of developments during the Recession (469 through 471) clearly reveals the interplay that went on between the Asmakas— the aggressors—and the local king, Upendragupta. In 468 Upendragupta was successful in expelling these troubling rivals from his region; but it is clear from the fact that he immediately forced work to stop on most of the (non-royal) caves that he was uneasy about the future. Even so, he continued work on his own caves and— surely unwisely—began a new caitya hall (Cave 29) as if to “replace” the now-abandoned Asmaka Cave 26. There may have been a touch of bravado in his continuation of his patronage in such critical times, when a show of strength may have offered both political and psychological benefits. And at the same time, he may have been eager for the merit which he could achieve through finishing his own donations—a goal which he was of course in a good position to achieve, or attempt to achieve. However, even though Upendragupta lauds himself as “expending abundant wealth” to “adorn the earth with stupas and viharas” (Cave 17 inscription), the reality of the Asmaka aggression obviously dawned on him before very long. Thus, he apparently abandoned work on his new caitya hall in the same year that it had been started (469); and at the same time he began to cut back on, and to expediently rush, the work on his beautiful Cave 20 and on Cave 17, where the walls and doorways at the rear (i.e. the late portions) of the cave are cut with increasing and evident haste. His sumptuous caitya hall (Cave 19) alone—although he could not quite finish it—was being carved and painted with consistent care, before time ran out late in 471. The situation of the Vakataka Prime Minister’s Cave 16 is equally revealing. Although the lesser patrons of Caves Lower 6, 7, 11, and 15 were allowed hardly more than a few weeks or months to complete and dedicate their shrine Buddhas, after abandoning the overall course of excavation completely, it is clear that when the Recession began in 469, the Prime Minister expected that he could bring his whole very unfinished cave to completion, even though from the very start he realized that the work done in this difficult year would have to be highly expedient. Indeed, he authorized “short-cuts” which would have been inconceivable in normal times; he did not properly
cohen’s “possible histories”
109
finish the leveling of the aisle walls before painting them; he painted the left wall before painting the ceiling, and even sliced off the stillrough capitals of the side aisle pillars to hasten the work.9 In the end he had barely been able to reveal the pillars of the shrine antechamber (some ten years later converted to an anomalous shrine), and was forced to abandon the whole excavation before the Buddha image was even reached. It was only Cave 1—the emperor Harisena’s cave—which progressed without any problems during the Recession, probably because as overlord, Harisena could insist on his cave’s priority; nonetheless, although work was progressing in normal course, it too had not been completed when the Recession ended disastrously late in 471.10
The Hiatus 472–474 The gradual cutbacks in Upendragupta’s patronage during the Recession, the rushed completion of the images in the local king’s Caves 17 and 20, and the fact that the inscription in his splendid caitya hall was never incised in the large panel prepared for it, reflects the increasing threat from the neighboring Asmakas, a threat which apparently culminated in full scale war in about 472. It seems likely that the Asmakas actually invaded the area at this time, for all work was now cut off completely, and the site went into a kind of limbo while—we can assume—the contending parties—Asmaka and Risika—fought it out. In the end, of course, Asmaka, which had long been seeking such a victory, emerged victorious. As we would expect, the Asmakas immediately continued the patronage of their great series of caves at the western extremity of the site. At the same time they not only disallowed worship in the rival local king’s Caitya Cave 19, long planned as the ceremonial focus of the site, but actually desecrated it by cutting through two of its court cells to make a convenient passage to their own complex, which lay beyond. Such a scenario clearly shows the developing pattern of Asmaka aggression, as well as the deepening establishment of their power at the site. An analysis of their further involvements will make this clear.
9 10
Ibid. See Volume V, Cave 1; Volume I, Cave 1.
110
volume iia, chapter four
By contrast, if one assumes with Cohen that the aggressors were the eastern Vakatakas, it seems impossible to find any evidence at the site of their incursion into the region. As for the Hiatus, Cohen is obliged to conclude, using an unsatisfying argumentum ex nihilo that they waged a battle but that it did not finally disturb the status quo; that is, we have no evidence at all that it in fact took place. This is very different from our alternate theory, which sees a drastic shift of regional control from the local king to his Asmaka antagonists, whose work now is to have a dramatic rebirth, while that of their royal adversary, the local king, finds a sudden and harsh termination. Evidence from Bagh One other matter relevant to the situation during the Hiatus should be mentioned. It seems evident from the mutual borrowing of various features that many workmen from Ajanta went to Bagh during the Hiatus, and then later came back to Ajanta with new forms which had been developed at that sister site. The Visrutacarita tells us that Anupa (in which Bagh lies) was then part of the Vakataka empire, with one of Harisena’s sons acting as the imperial viceroy. The latter would, like Harisena himself, have been an observer, rather than a participant, in the local Asmaka/Risika conflict. This assumption is supported by the consistent and untroubled development of the Bagh caves, as opposed to those at Ajanta, particularly throughout the years of the Recession and the Hiatus (see Time Chart). As a consequence Bagh was a logical place for Ajanta’s artisans to go for work and safety in those difficult years. This “migration to Bagh”, which in situ evidence clearly reveals, makes sense if the Asmakas were the aggressors. By contrast, if the eastern Vakatakas were the aggressors, as Cohen suggests, it certainly does not make sense. If there was such an inter-family conflict between the two branches of the Vakatakas, Anupa, as part of the original imperial domains of the western branch, would surely have been caught up in the conflagration. In such a situation, surely the patronage there would have been stopped during the period of the interfamilial contentions, just as patronage suffered a distinct “Hiatus” at Ajanta, according to my view, because of the Risika/Asmaka war. However Bagh is quite innocent of any such effects, and appears to have been a haven of refuge, while down in the Ajanta region, Asmaka and Risika fought it out.
cohen’s “possible histories”
111
It would appear from the character of Bagh’s paintings, sculptures, and its architectural and technological features, that work had stopped at Bagh somewhat earlier than the main phase of patronage came to an end (in 478) at Ajanta.11 At Bagh one does not find certain very late and significant “Ajanta” features, such as bhadrasana images, loving couples carved on the doorways, richly conceived pilaster medallions, and highly elaborated painted ceilings. Even the dearth of narrative murals in the Bagh interiors (as opposed to the porches) suggests that the Bagh interiors were underway first; the earliest narrative murals started in the interiors of Ajanta were probably those in Caves Lower 6 and 16, started in 468 and 469 respectively. In general, it appears that features moved from Bagh to Ajanta, affecting late developments there.12 The trabeated (quasi-structural) T-shaped doorway which does not appear at Ajanta until 477 would seem to find its source in the porch doorways of Bagh 2 and 4, as would the inclusion of standing door guardians and coin-pouring dwarfs.13 The Bagh mode of fitting doors and windows also dramatically affects Ajanta from 475 on in the new and highly practical “D mode” developed at Bagh in response to the weakness of the sandstone. Also the “attendant” standing Buddha figures in the shrine antechamber of Bagh Cave 2, even though probably added some time after the shrine had been completed, surely pre-date the use of paired standing Buddhas as attendants in sculptured groups at Ajanta, where such pairs never appear until 478. This usage may actually have been suggested by the slightly earlier use of groupings of six standing Buddhas as attendants of a central seated shrine image. Here again the groups found at Ajanta (never before 477) in Caves 4, Upper 6, 7, and the 24 court cell are anticipated by those in Bagh Cave 3.14 It is hard to believe that work would have stopped at Bagh just because so many workmen went back to Ajanta starting in 475. Surely a cadre, originally from Ajanta, would have stayed, along
11 See Spink 1976–1977, pp. 53–83, ill. 14–15, 8–10; although my conclusions are slightly different in some cases today. 12 For the probable influence of Bagh shrine arrangements see Volume I, Chapter 3. A badly damaged Buddha group at the left of the courtyard of Cave 4 could conceivably (but by no means surely) have been a bhadrasana image; if so it probably anticipated similar sculptures at Ajanta, which do not appear before 477. 13 A few earlier doorways at Ajanta (i.e. Cave 17 shrine) are “T-shaped” but not trabeated. 14 See Spink 1976/1977.
112
volume iia, chapter four
with local workmen. However, it may well be that most work underway had been essentially completed by then, and that no more was possible because all of the limited space at the site had already been used up. If we can believe the Visrutacarita, Anupa was not one of the many territories that rose up against the Vakataka overlord in the early 480s. In fact, Dandin tells us that it was the last stronghold of the empire, for it was to Anupa that Harisena’s grandchildren fled for safety, and it was in Anupa that they were protected from the empiredestroying Asmakas by the Gupta prince Visruta (= Maharaja Subandhu).15 Although the monastic community at Bagh could hardly have been totally insulated from the empire-shattering troubles of the early 480s, there is no evidence revealing the kind of deaththroes that we see at Ajanta in the Period of Disruption. While at Ajanta hundreds of intrusive images declare the breakdown of authority, it appears to be the case that there is not a single identifiable apparent “private” offering at Bagh to suggest any similar breakdown. Of course Maharaja Subandhu’s decision, in about 486, to defray the expense of ritual materials, and to provide “the community of Venerable Monks . . . with clothing, food, nursing of the sick, beds, seats as well as medicine”, might suggest—not surprisingly— that the local economy had suffered from the war.16 Nonetheless, it also suggests the establishment’s continuity, under strong control, over the course of the few intervening years. In this regard, it is relevant to note that Maharaja Subandhu’s gift was for “maintenance”, “repairs” and for general support.17 He did not sponsor any new excavations, even though that might have been more tempting and more “meritorious”; again, this would suggest that space was no longer available.
15
For Visruta’s identification with Maharaja Subandhu, see Volume I, Chapter 4. Maharaja Subandhu’s Bagh inscription, trans. Mirashi 1955, 19–21. 17 See Mirashi 1955, 19–21 (Bagh plate); 17–19 (Barwani Plate). Schopen 1990, note 13, states: Spink argues in part that the presence of such a provision in the Bagh grant indicates that Subandhu actually made “repairs” to the caves and that they were, therefore, excavated earlier than had been previously thought”. In fact, in my article on Bagh (Spink 1976–1977, 53–84) I take pains to prove that they are very much later than generally thought, dating to the 470s and 480s under the Vakatakas, rather than (following Mirashi) the late fourth (!) century. The various repairs and gifts made by Subandhu would date from the mid or late 480s, as opposed to Mirashi’s circa 417, since he dates Subandhu’s inscriptions to the Kalacuri rather than Gupta era. 16
cohen’s “possible histories”
113
Thus, quite possibly because of the forcefulness of Subandhu, who took over the rule of the region, it seems likely that Anupa largely escaped the embroilment in the great conflict which tore central India apart in the mid-480s. More than most of the many other rulers who, if their goal was independence, would benefit from the Vakataka fall, Subandhu could go about the business of setting up his new kingdom quite expeditiously. Indeed (if my hypothesis is correct) he was now laying the foundations of the great Early Kalacuri dynasty, which would restore something of the domains he had gained through his wife (Harisena’s granddaughter) to their rightful owners. He was of course now one of them.18 Asmaka Control over the Site 475–8 If my reconstruction of the political situation is correct, the Asmakas had taken over the Ajanta region by 475. This clearly explains the vigorous and well-organized rush of renewed excavation activity throughout the site’s western extremity—the Asmaka area—at that time. And since the Asmakas were (still) feudatory to Harisena, we can understand how work on the emperor’s Cave 1, his prime minister’s Cave 16, and most of the other caves also took on a new life, now that the Asmaka/Risika conflict had been resolved. The only exceptions—and they are of great significance—are the splendid caves (17–20; 29) of the Risika king, which remained in the same unfinished state as prior to the Hiatus, even though work was now otherwise renewed throughout the site. It would be hard to find a stronger confirmation of the Asmaka victory over the king of Risika, and the turnabout, in 475, of the site’s local control. By contrast, Cohen’s alternative theory does not “compute”. If the “Asmakas referred to in the Cave 17 and 26 inscriptions may be the Vatsagulma Vakatakas themselves”, as Cohen (1995, 62) suggests, one could understand why work on the Asmakas’ complex would have started up vigorously with their (re)establishment of control over the site in 475, after the Hiatus. It would also provide a reason for work on many of the other caves at the site being renewed
18 See resume of evidence from the Visrutacarita in Volume 1, Chapter 6 (summarizing events).
114
volume iia, chapter four
at this time—for such activity could only consolidate the connection of the new feudatory rulers with the various donors and with the people of the region. But (unless I have missed the point) one would be at a total loss to see why the new local rulers banned any further work on the great complex of caves at the center of the site (Caves 17, 19 and—I would insist—Cave 20) especially since their donor specifically honors the emperor in his Cave 17 inscription (verse 21), written “when Harisena, that moon among princes, was ruling the earth”. Cohen’s suggestion, which Cohen himself (1995, 73) admits “may not be correct, or convincing”, is that it is possible that work on these royal caves “did not start up again because their patron died in the fight” with his hypothetical invaders of the Eastern Vakataka Branch. But this is surely an argument in extremis, with two dubious hypotheses leaning upon each other for support. It takes no account of the gratuitous desecration of Upendragupta’s caitya hall, the lack of any ritual grime in the interior, the fact that the garland hooks over the pillars were never used and the fact that the door fittings of the intact court cells were not upgraded, as was common after 475—all evidence showing that the new rulers of the site disallowed its use.19 Nor can we explain this by saying that even though the hall was essentially complete, it had not been dedicated, and was therefore not ritually “alive”. The evidence of the site is very clear: only caves in which the image had been dedicated—no matter what their state of completion—attracted votive intrusions in the Period of Disruption; and Cave 19 is covered with them, while its sister caves, 17 and 20, have a good number too.
19
Actually, the garland-hook hole on the ceiling between the front pillars showed clear signs of use, but it is likely that its usage was in the Period of Disruption, for the cave has dozens of intrusions. Unfortunately, this hole has recently been cemented in. Some of the (intended) hooks over the pillars were never even applied. In other cases, the damage to the surrounding plaster appears to have been done not by usage but when the hooks were pulled out in later centuries.
CHAPTER FIVE
SCHOLARLY CONTRIBUTIONS TO MAHARASHTRA PATHIK SOME CONFLICTING VIEWS AND A REPLY By K. Khandalavala, A. Jamkhedkar, B. Deshpande, with final reply by Walter M. Spink. All are responses to Chapter 1 (W. Spink: Ajanta in a Historical and Political Context: Pathik Sept. 1990) It seems appropriate to retain the spellings given by these three authors. For these contributions and for all of the volumes, a Sanskrit glossary will be provided in the final volume. Introduction by Walter M. Spink The view of Vakataka chronology utilized as a general introduction in Volume I, Chapter 1, and under discussion in the commentaries below, demands (if correct) a new interpretation of the development of India’s Golden Age.1 These conclusions were developed, little by little, by the evidence of Ajanta itself; for Ajanta alone, standing so exuberantly and so precariously at the very apogee of that remarkable period, provides the crucial clues that allow the political and cultural situation in India in the third quarter of the fifth century to be read aright. In the process, of course, Ajanta itself emerges as a uniquely preserved record of the final burgeoning of India’s “classic age” on the one hand, and of that age’s subsequent and sudden fragmentation on the other. As a richly illustrated historical “text”, written with an inspired intensity and consistency, at the right time and the right place by the right participants, it clearly qualifies as one of mankind’s greatest creative achievements. And the individual who more than any other person made it possible can now be brought out of the shadows of neglect. More than anyone else it was the Vakataka emperor Harisena who, through the vigor of his
1 Originally published in Maharashtra Pathik (September 1990) as a result of the interest and energy of Devaratra Mehta, then Managing Director, MTDC.
116
volume iia, chapter five
rule and the breadth of his ambitions, sustained and nourished India’s great heritage for this brilliant climactic moment, even as the Guptas were starting their inexorable decline. Although I myself started with the once-standard assumption that Ajanta’s “Mahayana” development must have gone on for at least two hundred years, I finally was forced to conclude that this whole main—properly Vakataka—phase was accomplished in somewhat less than two decades.2 All except the last few frenzied years of work, done when the administrative structure of the site—like that of the empire—was breaking down, were comprised within the brief compass of the emperor Harisena’s reign, which the evidence at Ajanta proves must be assigned to the 460s and 470s. This is of course not obvious from the inscriptions alone. Such conclusions, with their farreaching implications for Indian history, emerge only from a detailed analysis of Ajanta’s development. Indeed, when combined with the epigraphic data, the solid evidence of the caves tells us a host of surprising things relevant to an understanding of these times, for the storey of the inauguration of the site and of its turbulent but carefully monitored growth bears directly upon the contemporary political, religious, and economic situation. The evidence available at Ajanta even provides a unique insight into the moment and the manner of Harisena’s death, that continent-shattering event that doomed both site and empire equally. Similarly, the record in the rock clearly reveals the political tensions and traumas that that ultimately led Ajanta’s elitist patrons from their initial exuberance to their eventual despair, over the course of the site’s all-too-sudden history. Here too, we find (once a few preconceptions are abandoned) a most compelling correspondence with the relevant epigraphic data, as well as with the information supplied by the storey of Visruta in Dandin’s Visrutacarita. The latter’s startling relevance for a clearer understanding of the history of the last half of the fifth century still awaits the acceptance of the shorter Vakataka chronology which it so distinctly and so discretely supports.
2 In providing dates for H. Zimmer’s The Art of Indian Asia, published in 1955, I followed contemporary convention by dating Caves 1, 2, 26 and others at Ajanta to the seventh century, and Elephanta to the eighth century, instead of to the fifth and sixth respectively
scholarly contributions to maharashtra pathik
117
I.K. Khandalavala: The History and Dating of the Mahayana Caves at Ajanta (Maharashtra Pathik, (vol. 2, No. 1, September 1990) pp. 18–22) (Mr. Karl Khandalavala is a leading Indologist and art collector and is an authority on Indian art. He is Chairman of the Board of Trustees of the Prince of Wales Museum, a Member of the Lalit Kala Akademi since its inception in 1954 and Editor of the Lalit Kala Journal.) The dating and history of the Mahayana caves of Ajanta can never be resolved by any process of speculation nor the formation of theories which are better suited to the realm of fiction that of factual evidence. First and foremost to suggest as Dr. Walter Spink, a prolific writer on Ajanta, has done that all the Mahayana caves of Ajanta were completed in less than twenty years (seventeen to be precise) and that each must have been completed in five years at the most, is a proposition that discloses the unpalatable truth which more than one Indian historian has remarked upon, namely, that most western writers on history of Indian art can never understand the Indian way of life and the Indian mind. The comparison which Dr. Spink had made between Michelangelo and the Indian artisans who created the sculptured and painted Mahayana caves of Ajanta, in order to support his time factor of five years per cave, is an instance of how little western writers can comprehend the role of the India artisan, his unchanging methods of work, his unchanging habits, his unchanging lifestyle including the inevitable visits to his village which take precedence over all else, and the unhurried pace of his craftsmanship. Apart from all this there was undoubted stoppage of all work during heavy rains, while even in other seasons work could only be done in the darker interiors of the caves when the sun afforded light, which was only for a limited period of time each day depending on the position of the cave in the Ajanta gorge. Now the only way to take sunlight into the deeper areas of each cave in order to work therein was by the use of white cloth reflectors, a method by which the present writer was able to bring sufficient light to the darkest areas of Cave No. II where he carried out the experiment. Any suggestion that the architectural and sculptural work and the mural paintings at Ajanta could have been done with the use of inherently dim oil-fed lamps
118
volume iia, chapter five
and constant soot would be too far-fetched to be given even passing consideration. How slow would the pace of excavating great caves and then fashioning intricate and elaborate pillars, façades, ceilings and sculptures be can only be realized by one who has seen the Indian artisan working on the renovation and replacement of damaged temple sculptures as the present writer has witnessed in restorations at the Jain shrines of Kumbharia. But Dr. Spink and those who follow him choose to forget that there is the most cogent inscriptional evidence that the important Satavahana period Balasri cave at Nashik took twenty years to complete as we have had occasion to point out before. And be it remembered that the Balasri cave was one of which the patrons were royalty so its progress could never have lacked funds. Even assuming that the phase of work at Ajanta was little speedier than at the Nashik cave it is inconceivable each of the great Mahayana caves of Ajanta namely Cave Nos. XVI, XVII, XIX, XXVI, I and II could ever have been completed in less than fifteen years each at the very least. Only a western writer who can never comprehend the Indian way of life and work amongst the Indian artisan classes would subscribe to so untenable a proposition namely that each of the Mahayana caves was completed in five years and that all the work at Ajanta was completed in less that twenty years—seventeen according to Dr. Spink with hundreds of craftsmen when probably there were no more than twenty-five working on each cave at a time. It must not be forgotten that the excavation of the Mahayana caves of Ajanta was not like the work of a high-speed engineering drill and latest machinery of the year of grace 1990. Nor were the guilds of excavators, sculptors, carvers and painters, who worked at Ajanta. Each comprised of a very limited number of guild craftsmen as we note to this very day. Apart from fantasies relating to the time factor for the completion of each of the great Mahayana caves of Ajanta we find a disregard by Dr. Spink of well established historical events in order to close his chapter on the Mahayana caves of Ajanta by A.D. 477 despite the existence of the most cogent inscriptional evidence which when taken along with the well accepted account of Dandin’s Visrutacarita clearly leads to the conclusion that Cave No. XXVI was commenced only after the death of Harisena in circa A.D. 505 and the fall of the Vakataka dynasty in circa A.D. 508 and since it
scholarly contributions to maharashtra pathik
119
could not possibly have been completed till circa A.D. 523 (a period of fifteen years) it is all too evident from the inscription in Cave No. XXVI that it was completed when the usurper of the Vakataka kingdom namely the former feudatory ruler of Asmaka was still ruling over the usurped Vakataka kingdom in circa A.D. 523. He is described in the above inscription in Cave No. XXVI as “the mighty Asmaka king”. This inscription in Cave No. XXVI is crucial and completely falsifies all Dr. Spink’s fanciful theories. Most significantly this inscription in Cave No. XXVI where the Asmaka usurper is referred to pointedly as “the mighty Asmaka king” refrains equally pointedly from making any mention of Harisena whatsoever which omission is unthinkable if Harisena was still ruling and the downfall of the Vakataka kingdom had not taken place. Be it remembered that in Cave No. XVI of the minister of Harisena as well as in Cave No. XVII of the Risika feudatory, the overlordship of Harisena is prominently mentioned in the inscriptions in each of these two caves. The inscription in Cave No. XXVI affords striking corroboration to the account in Dandin’s Visrutacarita of the usurpation of the Vakataka kingdom, after the death of Harisena, by the Asmaka feudatory ruler. It was only sometime after the completion of Cave No. XXVI in about circa A.D. 523 that Cave Nos. I and II were excavated, sculptured and painted. Again most significantly Cave Nos. I and II have no inscriptions at all and thus contain no inscriptional evidence to show under whose patronage they were constructed. This circumstance clearly indicates that they were undertaken after the restoration of the Vakataka dynasty not by the new young ruler or his ministers or feudatories but by wealthy monastic Orders with many donations and hence there was no single patron, royal or otherwise. For the restoration of the Vakataka dynasty see the Dasakumaracarita. Dr. Spink’s flight of imagination that Cave No. I was donated by Harisena and is “The King’s Cave”, though Harisena was long dead, is an instance of how the storey of the Mahayana caves of Ajanta is transformed from indisputable inscriptional and historical data into fiction. It is beyond the bounds of all possibility that Cave No. I if donated by Harisena would not bear a glorious dedicatory inscription of this monarch even if the Cave No. I was completed after his death. Moreover, if stylistic perception is to have any meaning there is a clear generation gap between the paintings in Cave Nos. I and II on one hand and those in Cave Nos. XVI and XVII on the other, as we have pointed out in our Heras Memorial Lectures
120
volume iia, chapter five
published as The Development of Indian Painting, by the Heras Institute. Moreover there is no evidence whatever of Harisena himself having any devotion to Buddhism or any interest in Buddhist monastic Orders. The Vakatakas were staunch Vaishnavas by faith, as aptly pointed out by Dr. Meena Talim in the Heras Institute Journal while refuting the suggestion, which is a historical impossibility both timewise and patronage-wise, of Cave No. 1 being the donation of Harisena. The history and dating of the Mahayana caves of Ajanta can be ascertained with reasonable certainty not on the basis of fictional theories such as those of Dr. Spink but on: 1) Proper interpretation of the vital inscriptional evidence in Cave Nos. XVI, XVII and XXVI. 2) The Ghatotkacha cave inscription and the highly significant fact that this cave which was also the donation of Varahadeva, the minister of Harisena and donor of Cave No. XVI, remained unfinished due without any doubt to the downfall of the Vakataka kingdom shortly after the death of Harisena. That Varahadeva, in very old age, be it particularly noted, fled the Vakataka kingdom with the young heir to the Vakataka throne thus leaving his donation of the Ghatotkacha cave unfinished is a historical circumstance established by the Dasakumaracarita.
Even the supposedly later addition of Dandin’s drama were obviously based on the earlier well known but lost original parts thereof and give us an authentic account of what happened after Harisena’s death and the Asmaka usurpation. 3) The important negative evidence of there being no inscriptional evidence in Caves Nos. I and II to indicate the donor of each of these caves respectively. 4) The account given in Dandin’s Dasakumaracarita is accepted by all sober historians as a true account of (a) what happened to the Vakataka dominion after the death of Harisena and the slaying of his son by the Asmaka led feudatories of Harisena; (b) the coming into being of the Asmaka interregnum which lasted over fifteen years; (c) and the defeat and death of the Asmaka usurper and the restoration of the Vakataka dominion to its rightful heir namely the grandson of Harisena.
We may pause here to give an instance of the kind of theories Dr. Spink is wont to put forward totally at variance with historical and inscriptional evidence and the authentic data of the Dasakumaracarita.
scholarly contributions to maharashtra pathik
121
In Dandin’s drama Visruta is a noble character who ensures the defeat of the Asmaka usurper and regains the Vakataka kingdom for the true heir namely Harisena’s grandson and has him crowned and set on the throne of his ancestors. But Dr. Spink to suit his imaginative theories equates Visruta with King Subandhu of the Bagh and Barwani inscriptions, though Subandhu dies several decades prior to the commencement of the Mahayana caves of Ajanta. Thus have resurrected Subandhu from his grave or funeral pyre Dr. Spink proceeds to transform him as a person acting under the guise of being the saviour of the Vakataka kingdom but in reality being an unscrupulous villain who first marries the granddaughter of Harisena, then pretends to restore the true heir (whose sister he has married) to the Vakataka throne but thereafter slays the true heir and appropriates the Vakataka kingdom himself. Thus, Visruta, a truly noble character in Dandin’s historical drama, is then first equated by Dr. Spink, by some magical process, with the long dead Subandhu and transformed into an unscrupulous villain who appropriates the Vakataka kingdom and that is the end of the storey of the Mahayana caves of Ajanta according to Dr. Spink with some caves unfinished and other abandoned with the change in regime. All this may be high drama and excellent fiction but is not authentic history. Subandhu ruled in the very early part of the 5th century A.D. over the Bagh area, for as shown by that great epigraphist, the late Dr. Mirashi, the Bagh inscriptions are inevitably to be dated in the Abhira era being identical in every respect, including the royal sign manual, with the Khandesh inscriptions which are regarded by all sober scholars as dated in the Abhira era. But Dr. Spink transfers the Bagh and Barwani inscriptions to the Gupta era though both on sound inscriptional evidence as well as stylistic considerations the Bagh caves are several decades earlier than the Mahayana caves of Ajanta. All this is done by Dr. Spink in order to equate Visruta with the long dead King Subandhu and the extinction of the Vakataka kingdom and what he assumes was the end of the Mahayana caves of Ajanta by circa A.D. 477. Thus Dr. Spink seeks to justify a period of about seventeen years for the completion of the entire complex of the Mahayana caves of Ajanta. Let us now turn fiction to factual evidence. The date of Harisena’s accession to the throne is not known save that it must by after A.D. 458 the date of the Hisse Borala inscription of his father Devasena. It is apparent from the inscription in
122
volume iia, chapter five
Cave No. XVI that Harisena’s father had a long life and when tired of administering his kingdom he gave the reins of government into the hands of his long-trusted minister namely Varahadeva’s father Hastibhoj and then led a life of ease. Most historians think that Harisena’s accession was not till circa A.D. 475 but for the sake of meeting Dr. Spink’s theories we will assume with him that Harisena came to the throne in A.D. 460 with Varahadeva’s old father as his minister for a short while to be succeeded by Varahadeva himself as Harisena’s minister. Now it is evident that in A.D. 460 Varahadeva was at least 40 years of age himself, being the son of an aged father and as one who was mature enough to should the great responsibilities as the minister of the Vakataka realm. The inscription of Varahadeva in Cave No. XVI further tells us that a time came when Varahadeva had no interest left in life and material benefits and the only thing that mattered in his mortal existence was his great devotion to the Buddha. Though it may well be beyond the comprehension of western writers, every Indian conversant with his religious heritage knows too well that is only in old age that kings, princes, generals, ministers, officials, merchants as well as other men reach that stage in life when all that matters is only absolute devotion to the Supreme Being himself, far away from great cities in the heart of lonely forests, or in far away mountain caves or by the banks of lonely rivers. This was the stage attained in Varahadeva’s life though obviously in loyalty to his sovereign he continued to perform his ministerial duties. It is thus all too evident that when Varahadeva became possessed with the desire to donate a cave dwelling to a Buddhist brotherhood of monks, because of his great devotion to the Buddha, he could not conceivably have been less than sixty years of age, at the very least. Thus even on the assumption of Dr. Spink that Harisena’s reign commenced in A.D. 460 it could never be earlier than A.D. 480 or more probably a couple of years later (A.D. 482) that Varahadeva commenced the actual project of creating a Buddhist cave dwelling at Ajanta namely Cave No. XVI. Since only a western writer with his mind closed to the Indian way of life and thought and unable to comprehend it (Spink’s Michelangelo comparison) would think that Cave No. XVI could have been completed in five years. In fact Cave No. XVI could not have been completed in less that fifteen years even if we assume that the work was speedier than at Balasri’s Nashik cave which was twenty years in the making. Thus Cave No. XVI was completed during the period A.D. 480 to A.D. 500 most probably A.D. 497.
scholarly contributions to maharashtra pathik
123
In the meantime after work on Cave No. XVI of Varahadeva had progressed for some years, the Risika feudatory of Harisena, as evidenced by his inscription in Cave No. XVII, greatly grieved by the death of his brother and being an ardent devotee of the Buddha emulated Varahadeva’s example and created not only the Vihara Cave Nos. XVII where his inscription is engraved but also the splendid Caitya Cave No. XIX as well as the Vihara Cave Nos. XVIII and XX. Significantly he refers to Harisena as that moon amongst princes who is ruling the earth. To digress, this inscription make it unthinkable that Harisena’s name was not inscribed in Cave No. I if donated by him (as Dr. Spink would have us believe) whether he was alive or dead at the time of its completion. Thus the Risika feudatory’s Cave No. XVII together with his Caitya Cave No. XIX and Cave Nos. XVIII and XX were in all probability commenced about A.D. 490 and since this cave complex took at least fifteen years to complete it must have been completed by A.D. 505 the likely date of Harisena’s death after which the Vakataka kingdom collapsed under the Asmaka usurpation. That Cave No. XVII is even stylistically a little later and more advanced than Cave No XVI should be apparent to any one sensitive to stylistic features. In the meantime the religious-minded Varahadeva desired to make yet another cave donation to the Buddhist monastic Orders and hence commenced work on a cave dwelling at Ghatotkacha about eleven miles from the Ajanta ravine. But the Ghatotkacha cave remained unfinished due no doubt to the downfall of the Vakataka kingdom and Varahadeva having to flee for safety with the young heir to the Vakataka throne, as narrated in the Dasakumaracarita. Thereafter when the Asmaka feudatory usurped the Vakataka kingdom, a monk name Buddhabhadra got the Caitya Cave No. XXVI excavated praising the Asmaka usurper as “the mighty Asmaka king” and pointedly ignoring the name of Harisena and the Vakatakas altogether for not only were they no more, but the whole kingdom was in the hands of the mighty Asmaka usurper. If the Asmaka ruler was still a feudatory at the time when Cave No. XXVI was excavated and Harisena was still the overlord of the Vakataka domain, as Dr. Spink had contended in an effort to support his theory of all the Mahayana caves of Ajanta being completed by circa A.D. 477, then only a person devoid of all historical sense could believe that Harisena though alive and overlord of his kingdom was still not mentioned in the inscriptions of Cave No. XXVI by its donor who dared
124
volume iia, chapter five
to humiliate the great Harisena so unabashedly that he referred to a mere feudatory ruler as “the mighty Asmaka king”, pointedly but ignored the great overlord namely “that moon amongst princes who rules the earth” (Harisena) as described in this inscription of the Risika feudatory in Cave No. XVII. Only a western writer could believe that such a thing could happen in India in the 5th-6th century A.D. Cave No. XXVI was commenced by about A.D. 508 as can clearly be deduced from its stylistic affinities to Cave No. XIX of the Risika feudatory completed only in about A.D. 505. Since Cave No. XXVI could never have been completed before circa A.D. 523, the Asmaka usurpation must have lasted till at least that date. Thereafter we know from the Dasakumaracarita that the Asmaka ruler was defeated by the noble Visruta and the Vakataka kingdom restored to the grandson of Harisena. Thereafter Caves Nos. I and II were donated by rich monastic Orders with numerous donations no doubt, and hence there are no donatory inscriptions therein as very rightly suggested to us by Dr. R.N. Misra, the well known scholar of the Gwalior University. The commencement of Cave No. I can be assigned to circa A.D. 525 and its completion to circa A.D. 540 while the later Cave No. II was probably commenced in circa A.D. 530 not long after the restored Vakataka dynasty came to an end. So the period of the Mahayana caves of Ajanta is from circa A.D. 480/482 till circa A.D. 545 if inscriptional evidence, historical data in the Dasakumaracarita and marked stylistic features are to have any meaning. II. Dr. Arvind Jamkhedkar: Some Relections on Professor Spink’s Chronology at Ajanta, April 1991 Prof. Spink has been working on the Ajanta Caves right from the fifties and his contributions on Ajanta which appeared from time to time as presentations in various scholarly gatherings and in research journal run into about four hundred printed pages and deserve more serious consideration than it has received, especially in India. In the last century when the Caves were put to systematic study, the later or Mahayana phase at Ajanta was dated to roughly A.D. 500–680. This envisaged again two phases in which the Group II and Group III Caves (the Group I being that excavated in the
scholarly contributions to maharashtra pathik
125
Hinayana phase) were supposed to be successively excavated respectively in C. 500–600 A.D. and C. 600–680 A.D. The Group II included Cave Nos. 6 and 7 and 16 to 20, while the Group III included Caves 1 to 5 and 21 to 27. Caves 28 and 29, though known, were inaccessible and the Ghatotkacha Cave, though described, was not put to any special scrutiny from the point of view of chronology. Cave 11 was thought to be of the Hinayana Group. The epigraphical studies of Mahamahopadhyaya V.V. Mirashi brought a new dimension to the studies of Ajanta because of his better reading and interpretation of the inscriptions in Caves 16 and 17 at Ajanta and the Ghatotkacha Cave. This proved that Cave No. 16 at Ajanta and the Ghatotkacha Cave were creations of one and the same person, i.e. Varahadeva, Minister of King Harisena of the Vakataka family. The researches of Mm. Mirashi also dispelled the misconceptions regarding the genealogy of the Vakatakas and established that the Vakatakas who were responsible for the excavation at Ajanta (through their courtiers) were a separate though cognate branch of Vakatakas at Nandivardhan which line was better known through their copper-plates and matrimonial relations with the Gupta family. Ms. Mirashi also showed that the patron king of Cave No. 17 was a feudatory of Harisena and the latter must have ruled a vast empire including Malava in the North and north Karnataka (Kuntala) in South and the coastal regions of east (Kalinga and Kosala) and West (Lata and Trikuta) and that he flourished during 475–500 A.D. With this new understanding of the political vicissitude of the Caves at Ajanta, there was a modification of the chronology of the Caves at Ajanta. The orthodox school still maintaining the twophased development of Ajanta put the first group in A.D. 475–500 bracket and proposed A.D. 500–550 bracket for the remaining i.e. Group III Caves. Prof. Spink on the other hand proposed that the creations at Ajanta, both Caves and paintings, were creations of Vakataka period, especially that of Harisena (whom he proposed a shorter reign period of A.D. 465–490) and the intrusive figures to the period of confusion which followed his sudden death (in C. 490 A.D.) and ineffective rule of his heir (for whom Mm. Mirashi had proposed a time bracket of 500–510 A.D.). Besides the Indian scholars who have been contributing significantly to our understanding of the art historical studies of Ajanta, like C. Sivaramamurti, K. Khandalavala, M.N. Deshpande to mention some,
126
volume iia, chapter five
European and American scholars have studies the subject from different perspectives. To mention only some, the names Sheila Weiner, Wayne Begley and Phillipe Stern come to the fore. The last mentioned has studied the chronology of Ajanta in the light of his studies of the development of pillars at Ajanta and Ellora. An observation in this connection is worth mentioning. All the scholars, India and foreign, who have studied this problem come to more or less similar conclusions. And this fact in important in that they have arrived at these conclusions independently. This brings us to the most difficult question, viz. how and why these scholars suggest different time brackets; how is it that some have suggested that some of the Caves at Ajanta are of post-Vakataka period. The question is also whether the end of the Vakataka dynasty has something to do with the end of the so-called classical art of India. The above questions have something to do directly or indirectly with the ingenious reconstruction of the last days of the Vatsagalma branch of Vakataka dynasty done by Mm. Mirashi. On the basis of the Katha of Visruta in the Dasakumaracarita of Dandin, Mm. Mirashi has made this reconstruction. It is postulated on the basis of certain strands in the storey that at Mahismati, a son of last known king of the Vakataka dynasty, viz. Harisena was ruling. Probably the grandson of Harisena took refuge at the Mahismati Court after his rescue from Vatsagalma along with the old minister, as his father was defeated in battle by the feudatories who revolted against him. Mm. Mirashi proposed that Harisena must have ruled from C. 475–600 A.D. and his weak son who was abandoned to luxuries and ultimately succumbed to the machination of the Asmaka minister may have ruled from C. 500 to 510 A.D. This terminus could not be stretched further down the time stream as the Kalacuris had already started their career at Mahismati by the beginning of the second quarter of the sixth century A.D. Prof. Spink, who agreed with this reconstruction of the history by Mm. Mirashi made certain modifications: he proposed a shorter time bracket for the reign of Harisena., viz. C. 465–490 and remaining ten years, for the reign of his unfortunate son and disintegration of the Vakataka empire. He also thought that the Trikuta kingdom which paid tribute to Harisena (inscription in Cave 16) claims itself in A.D. 494 to be having ‘an augmenting kingdom’ which shows that the Vakataka empire had not only lost its hold but was on its
scholarly contributions to maharashtra pathik
127
way to disintegration (Ajanta to Ellora, Marg, Vol. XX, No. 2 March 1967). However, after 1972 Prof. Spink further truncated the time bracket for the excavation of the Caves as also the rendering of paintings at Ajanta. This he has presented in a nutshell in the paper (“Ajanta’s Historical/Political Context”—Pathik of MTDC). He furthered realised that Maharaja Subandhu of Kalacuri family was already ruling around C. 487 A.D. in the region Mahismati. This meant that the Vakataka creations at Bagh were earlier than that time. Prof. Spink had also made several art historical observations about the Mahayana phase Caves which showed clear evidence of precedence of Bagh Caves over certain elements from the Ajanta Caves. Not only this, both early and late elements showed themselves in the same Cave; thereby showing that work at certain Caves was stopped at a juncture and then was resumed. The rule at Mahismati of the Junior Vakataka (son of Harisena) could be a solution, he thought. On the basis of other observations like presence of excessive soot, and wear and tear of pivot holes in certain Caves and their total absence in the later complex Caves like 1 and 2; the passage broken through the Cave 19, the inscriptional evidence about Asmakas in Cave 26: all these can be explained on the basis of the political events which have been reconstructed very ingeniously by Prof. Spink. Scholars have expressed their own doubts about the further shortened time bracket for the Mahayana Caves at Ajanta. Arguments have been put forward by Karl Khandalavala showing total disagreement with Prof. Spink (Rock Temples etc., pp. 209–230 in Maharashtra Issue of Marg 1985 regarding the relative dates of Bagh and Ajanta and Kalacuri patronage of Elephanta Caves. He accepts Mm. Mirashi’s date for Harisena (C. 475–500 A.D.) and his successor (C. 500–510 A.D.) but does not accept total reliance on Dasakumaracarita for understanding Vakataka history. He thinks there was a post Vakataka phase at Ajanta from C. 500–550 A.D. Prof. Joanna William, who has given due consideration to the views of Prof. Spink and agrees with his reconstruction of Vakataka history, is cautious regarding the negative and vague nature of evidence from inscriptions of Traikutakas and Subandhu. It will be too much to read the type of meaning in these inscriptions which Prof. Spink prefers to read, according to her. The time bracket suggested earlier by Prof. Spink (viz. 465–500 A.D.) is a more satisfactory one,
128
volume iia, chapter five
she say (The Art of Gupta India: Empire and Province, Princeton University Press, 1983, pp. 181–187). Another difficulty in accepting the modified time bracket suggested by Prof. Spink is that he has overlooked apparently the situation in the Nandivardhan House. As per the accepted understanding Rudrasena II ruled from C. 400–405 A.D., Pravarasena II ruled from 420–450 A.D., Narendrasena ruled from C. 450–470 A.D. and Prithvisena from C. 470–490 A.D. The reign-period of Pravarasena is required to be augmented at least by two years because of the Pauni plates which mention his 32nd year of reign. The Mahurzari plates of Prithvisena II, which were issued in his 17th regnal year, show that the period assigned to his reign is more or less correct. According to the chronology of Washim Brauch as understood both by Mm. Mirashi and Prof. Spink, Harisena was a contemporary of both Narendrasena and Prithvisena. Balaghat plates of Prithvisena describe that Narendrasena was paid tribute by kings of Kosala, Mekala and Malava. The evidence moreover of the inscriptions at Ganj of Vyaghradeva show that Prithvisena was his overlord; and Vyaghradeva was ruling region in the vicinity of the origin of Narmada. This shows that the present shorter time bracket forwarded by Prof. Spink can be accepted only if the aforesaid difficulties are satisfactorily solved by him. And, therefore, till then, the earlier time bracket would be more acceptable. However, his art historical observations regarding the relative chronology of the Caves and pertaining architecture, sculpture and painting at Ajanta, his observations regarding the relationship of the Caves at Ajanta and the other Caves Centres of contemporary times as at Aurangabad, Bagh etc., must be viewed seriously and taken into consideration. III. Brahmanand Deshpande: Dating of Ajintha Caves in the Context of Vakataka Decline, Pathik July 1992 Sanskrit scholars, turned Indologists, are always prone to hunt for historical scanning Sanskrit works like Karpuramanjiri, Viddhasalabhanjika, Dasakumaracarita and many other for this purpose. The last mentioned work, according to him, has a bearing on the decline of the Vakataka empire. His inferences have been utilised by Dr. Walter Spink, without proper scrutiny, in a fervent attempt to fix the date of Ajintha Caves and hence a reappraisal of the entire
scholarly contributions to maharashtra pathik
129
episode is much needed. For a careful perusal, the main contentions of Dr. Mirashi are summed up as follows: data in literary works. Late Dr. Mirashi, for example, has laboriously worked in this direction, 1. On the left wall of the verandah of Cave No. 16 at Ajintha, a much damaged inscription has been located. This inscription alludes to Varahadeva, a minister of Harishena, the last known Vakataka emperor. The inscription contains a long eulogy of Harishena who is said to be ruling over a vast empire consisting of countries like Kuntala, Avanti, Kosala, Trikuta, Lata, Andhra and many others. According to the available epigraphical evidence, Harishena seems to have ruled from 475 to 500 A.D. Half a century later, Vidarbha seems to have passed under the sway of the Kalachuris of Mahishmati. We wonder at this fast decline and are at loss to solve the riddle of this sudden decline of this vast empire in such a short span of time. 2. Dasakumaracarita, composed by the famous prose writer Dandin, sheds valuable and welcome light on this problem. Damodara, the great-grandfather of Dandin was a resident of Achalapura, an ancient town of Vidarbha region. He migrated to Karnataka, received patronage of a king called Durvinita, a Western Ganga King and finally settled at Kanci and became the court poet of Pallava king, Simha Vishnu, who ruled from 580 to 600 A.D. Thus the date of Dandin, his great-grandson, can be fixed from 650 to 675 A.D. Thus Dandin lived a century and a half later than Harishena. We can say that he possessed detailed information on the Vakataka empire passed from generation to generation in his family. 3. Dasakumaracarita, or the episodes of ten princes, is not available in its complete form. Its Purvapithika (prologue) and Uttarapithika (epilogue) are not the handiworks of Dandin himself but were composed and appended to it by other scholars. The Purvapithika contains the account of two ‘kumaras’ and those of remaining eight ‘kumaras’ are narrated in the following eight Uchchhvasas (chapters). 4. The eighth Uchchhvasa, narrates the storey of Visrutakumara and hence is called Visrutakumaracarita, which runs as follows: a) The pious and benevolent king Punyavarman, who ruled over Vidarbha region was succeeded by his son Anantavarman, who was given to sensual pleasures and neglected his royal duties. His faithful and old minister, Vasurakshita, advised him to mend ways and study statecraft but Anantavarman was under the vicious influence of Viharabhadra, his companion in sensuous affairs. He turned a deaf ear to the advices of Vasurakshita and even insulted him. b) Now Anantavarman became utterly careless about his empire and spent most of his time in pursuit of sensual pleasures.
volume iia, chapter five
130
c)
d)
e)
f)
Vasantabhanu, the king of neighboring Asmaka country, was clever enough to take advantage of this situation. A son of his minister Candrapalita entered Vidarbha under the false pretext that as a punishment for a quarrel with his father he had been banished. Anantavarman, the generous Vidarbha king, gladly granted him asylum. Candrapalita was a wily politician. Among with him came bards, dancers, singers, musicians, hunters and a host of domestic servants who were spies in disguise. They kept the king busy with hunting, gambling, drinking bouts and sex orgies. The administration collapsed and chaos took over. Now Vasantabhanu instigated Bhanuvarman, the ruler of Vanavasi region to invade Vidarbha, which he did and conquered some bordering districts of the Vidarbha empire. Anantavarman, now realising his folly, tried to tide over the coming danger. He hastily gathered the remnants of his army and called his feudatories for help. Some feudatories responded to his entreaties and came to his succour. Vasantabhanu, the master-plotter, was first to come. A sanguinary battle was fought on the banks of the Varada river. Vasantabhanu very skillfully sowed the seeds of dissension amongst the feudatories of the Vidarbha king and turned his faithful allies like Varasena, the lord of Murala; Ekavira, the lord of Rsika; Kumaragupta, the lord of Konkana and Nagapala the lord of Nasikya against him. They betrayed their master and went over to Bhanuvarman. Anantavarman fell while fighting. Vasantabhanu grabbed the Vidarbha empire and the spoils of wars were shared by the unfaithful feudatories. The lives of Vasundhara the widowed queen, Bhaskaravarman her son, and Manjuvadini her daughter were in jeopardy. Vasurakshita, the old devoted minister, now rose to the occasion. He took these surviving members of the royal family to Mahishmati, where Mitravarman, the stepbrother of Anantavarman was ruling. But here too, Vasundhara, the unfortunate queen was not safe. Mitravarman tired to seduce her and hence the poor trio had to take resort in the dense forests of the Vindhya ranges. Here they met Visrutakumara who turned out to be a distant relative of Bhaskaravarman. This gallant youth brought about the downfall of Mitravarman and Bhaskaravarman was seated on the throne of Mahismati. As noted above, Dasakumaracarita is an incomplete work and we do not know whether Bhaskaravarman, got back his ancestral Vidarbha kingdom or not. The stories of Dasakumaracarita are generally imaginary but Visrutakumaracarita seems to contain important historical information. Mark Collins was the first to note this fact. This Visrutakumaracarita explains the ‘sudden downfall’ of the Vakataka empire. The unfortunate Vakataka ruler must have been the son and successor of Harishena.
scholarly contributions to maharashtra pathik
131
5. Thus Dr. Mirashi has identified Anantavarman and Bhaskaravarman with the son and grandson of Harishena respectively. The Vanavasi king Bhanuvarman has been identified with Ravivarman, a Kadamba ruler of Belgaum district of Karnataka state. This is the bulwark of Dr. Mirashi’s arguments. Dr. Walter Spink has accepted these inference in toto and proposed some new identifications also. All this seems plausible but a careful perusal of the available epigraphic material gives rise to many doubts and differences. 1. At the outset we maintain that Dr. Mirashi is elusive on many grounds and Dr. Walter Spink too followed suit. 2. Damodara was a resident of Vidarbha but he migrated to Karnataka and finally accepted the patronage of Simhavishnu, the Pallava ruler of Kanci. His son was Manoratha and grandson was Viradatta. Dandin was the son of this Viradatta from his spouse Gauri. Dandin lost his parents at an early age. The Pallavas had to face invasions of their rivals. Poor Dandin had to leave the Pallava court and he was a homeless wanderer for twelve long years. His intimate knowledge of Vakataka history should be taken with a pinch of salt. 3. Dasakumaracarita is an incomplete work. Purvapithika and Uttarapithika are clearly later additions. Even the account of Visrutakumara is not complete. It seems that many anonymous pen-pushers have dared to supplement this great work. 4. Dr. Mirashi accepts that the accounts of the ten kumara are imaginary. Nevertheless he is tempted to accept the storey of Visruta as historical. Following the footsteps of Mark Collins, he believes that this storey can best explain the sudden downfall of the Vakataka empire. 5. Ajintha Cave inscriptions of Varahadeva lists a number of countries which Harishena is said to have conquered during the course of his ‘digvijaya’. Dr. Mirashi is aware that this account is traditional and stereotyped. There is no proof, whatever, for the defeat of the Traikutakas at the hands of Harishena. To the contrary the Kanheri plates of Saradvatiputra, Buddhist Mendicant of far away Sindhuvishaya speaks of the augmenting sovereignty of the Traikutakas. Hence the extensive Vakataka empire and its sudden downfall should be discussed more carefully. 6. Dasakumaracarita tells us that the invader of the Vakataka empire was Vanavasya Bhanuvarman. Dr. Mirashi has safely assumed that he was the king of Vanavasi, modern Belgaon district of Karnataka state. Now this Vanavasi tract was ruled by the illustrious Kadambakula. Dr. Mirashi has identified the Vanavasya Bhanuvarman of Dasakumaracarita with the Kadamba king Ravivarman who ruled from 497 to 537 A.D. Now this Ravivarman actually had a younger brother called Bhanuvarman, who issued his Halsi copper plates in 545 A.D. Dr. Mirashi is cautious enough not to identify the Vanavasya Bhanuvarman of
132
volume iia, chapter five Dasakumaracarita with his namesake in Kadambakula. The reason is not difficult to understand. He himself has opined that he whole of Vidarbha passed under the Kalachuri rule around 550 A.D. Hence the Kadamba ruler Bhanuvarman of Halsi plates issued in 545 A.D. could not be identified with Bhanuvarman of Dasakumaracarita.
But even this trick has not solved the problem completely. Vanavasi or the modern Belgaon district of Karnataka state is far away from Vidarbha. It seems well nigh impossible for a Kadamba ruler to invade a distant country. Contemporary epigraphic records do not speak of any rivalry between the Kadambas and the Vakatakas. It has been correctly suggested by Dr. K.R. Joshi that ‘Vanavasya’ does not necessarily mean to be the king of Vanavasi region. According to him Vanavasya means a person or a king of a forest-clad region. He has drawn our attention towards Bastar region which is covered with dense forest and has a common border with Vidarbha. The Nala dynasty of Bastar region is well known to the students of Indology. Bhavadattavarman, the Nala king of Bastar, seems to have captured a portion of Vakataka empire. His copper plate grant dated in his 11th regnal year has been issued from Nandivardhana, the capital of the Vakatakas. The paleography of this grants suggests its date around 450 to 550 A.D. Coins of Varahadeva, Bhavadattavarman and Arthapati, the Nala rulers of Bastar, are found at a place called Edenga in Bastar district. The foregoing discussion clearly proves that the invader of the Vakataka empire was the Nala king Bhavadattavarman of the neighboring Bastar region and not the Kadamba rule Ravivarman of the distant Vanavasi region. Vanavasya simply means a resident of forest country and should not be taken to mean to be the far off Vanavasi kingdom. This much for the invader. Now who was the Vakataka ruler that faced this invasion? The Balaghat copper plate inscription of Prithvishena II throws some light on this problem. In this inscription Pravarasena II, the grandfather of Prithvishena II is given the epithet ‘Uchchhinnasarvadvishah’ or the vanquisher of all enemies. It is quite possible that Bhavadattavarman invaded the Vakataka kingdom during the reign of Pravarasena II. His Chammak and Mandhala copper plates were issued from Pravarapura and not from Nandivardhana, their original capital. This change in the capital was due to the occupation of Nandivardhana by Bhavadattavarman, the Nala invader. Similarly in the Balaghat plates of Prithvishena II,
scholarly contributions to maharashtra pathik
133
Narendrasena, the father of Prithvishena II, is described as ‘Purvva = dhigataguna—Visvasad = aparhitavamsasriyah’ or one who has restored the lost fortunes of his family. Even Prithvishena II the successor of Narendrasena is called ‘Dvimagnavamsoddhartuh’ or the savior of his family which suffered twice. Thus it is clear that the invasions on the Vakataka empire were repelled by Pravarasena II, Narendrasena and Prithvishena II. All these three rulers belonged to the Nagaradhana branch or the main branch of the Vakataka dynasty. While Harishena belonged to the minor or Washim (ancient Vatsalgulma) branch. Hence it is quite obvious that neither Harishena nor his son and grandson encountered any invasion whatever. Any attempt to connect the rout of the king Anantavarman as narrated in Dasakumaracarita with the downfall of the Vakataka empire is bound to be futile. Now let us turn towards the theories of Dr. Walter Spink. As noted above, he has accepted the inferences of Dr. V.V. Mirashi in toto. The new identifications which he has proposed are as follows: 1. Harishena dies without warning. Sarvasena III, his son and successor, already under pernicious Asmaka influence i.e. c. A.D. 477. 2. Asmaka, still nominally feudatory begins plotting the overthrow of Vakataka power i.e. c. A.D. 478. 3. Open war against Vakataka overlord Sarvasena III, the son and successor of Harishena, begins i.e. c. A.D. 481. 4. Vakataka dynasty falls. Sarvasena III is killed. Harishena’s grandson and grand-daughter flee taken by old minister (Varahadeva?) to Mahishmati where another son of Harishena (Mitravarman?) still remains in control of that region i.e. c. A.D. 483. 5. Anupa, last remnant of Vakataka empire with capital at Mahishmati now controlled by Maharaja Subandhu probably identical with Prince Visruta of Dasakumaracarita. Visruta after killing the Vakataka Viceroy married Harishena’s granddaughter and started ruling at Mahishmati as a regent of Harishena’s young grandson who eventually dies or was possibly killed by Subandhu. This Subandhu may have founded the Kalachuri dynasty of Mahishmati i.e. c. A.D. 485. 6. 486 A.D. provides the terminus ante quem for period of Vakataka rule. Barvani plates of Maharaja Subandhu prove that he was in control of Mahishmati. Bagh copper plates prove that he repaired the caves at Bagh. 7. Traikutakas, former Vakataka feudatories, now independent as proved by Surat (490 A.D.), Kanheri (494 A.D.), and Matvan (505 A.D.) plates. This also confirms that Vakataka fall occurred before this time.
Let us now examine these theories in the light of available epigraphic records.
134
volume iia, chapter five 1. We have already established that there is no mention, whatever, of any invasion on the kingdom of Harishena. That he died without warning is a baseless hypothesis. We do not know whether his son Sarvasena III was under Asmaka influence or not and also the cause of his death. 2. It is adventurous to identify the Vakataka minister Varahadeva of Ajintha Cave No. 16 inscription with Vasurakshita the minister of Anantavarman, alluded to in Dasakumaracarita. There is absolutely no proof for this. 3. Similarly it is not correct to identify Visruta of Dasakumaracarita with Subandhu, the ruler of Malwa. Subandhu is known to us through his Bagh and Barvani copper plate inscriptions. The Bagh copper plates record a donation for the Buddhist monastery at Pathaka in the Dasilakapalli region. The Barvani copper plates are dated on Bhadrapada Suddha Saptami of the year 167 of an unknown epoch. Dr. Halder has taken this to be a year of Gupta Samvat and placed the Barvani plates in 486 A.D. But according to Dr. Mirashi, Malwa was not under the sway of the Guptas at the close of 5th century A.D. He has suggested the Kalachuri epoch of the year 249 for this date. Thus the date of the Bagh copper plate is 416 A.D. Three more copper plates from Khandesh confirm this date. 4. The Dasakumaracarita portrays Visruta as a noble prince and the savior of the offsprings of Anantavarman. Dr. Walter Spink has painted him as a ruthless intriguer and that too without any proof. 5. If the date of the Barvani copper plate given by Dr. Mirashi is correct, Subandhu preceded Harishena by more that half a century and his identification with Visruta, who helped the grandson of Harishena, goes wrong. 6. Whatever may be the date of the Vakataka downfall, it has nothing to do with the Traikutaka rule. Traikutakas were never the feudatories of the Vakatakas. All the Traikutaka records clearly speak of their independent rule and never mention Vakatakas as their overlords. They issued independent coinage which also speaks of their sovereign position.
To sum up we stress the following points: 1. The Visruta-Katha of Dasakumaracarita has nothing to do with the successors of Harishena. 2. There is no epigraphic record of any invasion on the Vatsagulma branch of the Vakatakas. 3. Identification of Subandhu with Visruta does not hold ground. 4. Traikutakas were never the feudatories of the Vakatakas. 5. All attempts to fix the date of the Ajintha caves in context of Vakataka downfall should be done in the light of a careful scrutiny of available epigraphic and literary sources.
scholarly contributions to maharashtra pathik
135
Walter M. Spink: Repy to K. Khandalavala, A. Jamkhedkar, B. Deshpande a. Reply to Karl Khandalavala’s criticism The poet Rilke has written about the value of wrestling with a great—he may even have said divine—adversary, if one is going to grow; and in this regard it is both a challenge and a privilege to be in contention with my friend Karl Khandalavala. However, even the devas can make mistakes; and in the matter of Ajanta’s history and development, Mr. Khandalavala, and conventional wisdom in general, is surely wrong. However, he is gloriously wrong. That is, he is wrong in the service of a great cause, which is nothing less than the task of proving that Ajanta, in its Vakataka phase, may well be mankind’s most remarkable creative achievement. (What other great monument has such a rapid, and tumultuous development, and at the same time can be analyzed, literally, year by year?) By aligning all of the forces of the Indian tradition and the Indian mind (which, he declares, “most Western writers . . . can never understand”), and his own opinions in particular, against my view that Ajanta’s whole later development took place in less than twenty years, Mr. Khandalavala advances my larger purpose, which is to establish the site’s global preeminence as a creative achievement. For by arguing so insistently that what I say about the site breaks all the limits of credibility— that such an amazing accomplishment could never have been realized in such a short time—he clearly reveals that such an achievement, if it did happen, would be “unbelievable” indeed, or at least would seem to be. It is my goal, of course, to make the “incredible” credible by going to the relevant evidence and by trying not to indulge in mere opinion. And in doing this one finds that Ajanta’s incredible achievement, while surely remarkable, is not in fact “unbelievable”. The point of my Michelangelo example, which so offends Mr. Khandalavala, is that one need not allow two or three hundred years to get the caves painted. Even with the workers’ trips to the village and all of the other distractions and interruptions that Mr. Khandalavala (I expect wrongly) mentions, a couple of decades would be far more than sufficient to accomplish all of the painted work, which careful scrutiny shows was produced with both intensity and efficiency, under
136
volume iia, chapter five
a very demanding patronage rather than at the whim of the workers. Painter friends of mine, making precise copies of some of the great mural scenes for the Department of Archaeology, insisted that any given wall or ceiling must have been decorated in a matter of weeks— so free and vigorous was the style. At the same time, they themselves, working with very different goals, and being allowed much tea and talk, extended such assignments out to a matter of years. In the same way, we should know Balasri’s particular situation before criticizing her for taking twenty years to complete her very simple and conventional cave at Nasik, which a few dozen workmen, each cutting away merely a few cubic feet of rock a day, could have at least roughed out in far less than a year. Even the vast and elaborately decorated temple of Rajaraja I at Tanjore, where all the stone had to be hauled in, was accomplished in less than a decade. As in the Nasik example, the problem at Ajanta is ultimately not how it could all have been done in such a short time, but why it took so long. We can hardly believe that any given Ajanta patron intended to wait fifteen years to see his cave completed, any more than anyone today would want to wait so long to have their house or their place of business finished. And yet the great majority of the Ajanta excavations were underway almost that long; and of course most of them were still very unfinished when time ran out. Thus one of my major concerns, in reconstructing the history of the site, has been to explain when and why and in which caves work stopped, either temporarily or permanently, during the course of the site’s patronage; for only in this way can we explain what otherwise would seem to be a most dilatory endeavor. (Of course, I never did say “that each of the Mahayana caves was completed in five years”—I like to believe that blanket generalizations like the present misquote are the very antithesis of my approach.) The truth is, Ajanta’s development was not dilatory at all. Despite Mr. Khandalavala’s assumption that the artists—as if they ruled the site—let their royal patrons wait while they went off to play Holi, and apparently refused to work even inside the caves during the rainy season, the reality of Ajanta appears to be something very different, and something far more impressive. We are dealing with a unique achievement—the very culminating monument of the Golden Age (at least the greatest surviving one!).3 Even if the artists did not 3
The great brick structure at Mansar is a somewhat comparable achievement,
scholarly contributions to maharashtra pathik
137
know that this was what they were producing, the patrons and the monks and the travellers and the merchants were applying pressures upon them that are clearly revealed in the insistently unceasing evolution of the work at the site. We are not talking here of a hit-ormiss, decade-by-decade or century by century, production. This is the old way of viewing the site—still shamelessly reflecting the “colonial” assumption that the Indian workers could never get their act together and produce a complex site with such efficiency—should now be set aside for a correct and appropriately laudatory view. What the site reveals instead is a cohesive and exciting era of discovery, almost entirely compressed into the course of the brief reign (c. 460–477) of a single remarkable king, whose name and reputation deserve to be restored to the highest levels of history. Needless to say, my conclusions, which rest upon a complex construct of evidentiary data, are immediately rendered invalid if certain assumptions that I make cannot be sustained. Chief among these, in Mr. Khandalavala’s mind, is my apparently impious disagreement with “that great epigraphist the late Dr. Mirashi” regarding the matter of the date of Maharaja Subandhu of Mahismati. However, the assumption that, the evidence notwithstanding, Dr. Mirashi (or any other such great man) is automatically correct, does a disservice to the cause of truth and (as I expect Dr. Mirashi would agree) to Dr. Mirashi himself. I considered him not only my friend but my generous mentor, and could never have done my work without his fundamental studies, which I constantly use. However, he was wrong about Bagh, and about Maharaja Subandhu.4 Furthermore, his excessively late dating of Harisena denies the validity of a host of compelling synchronisms, while rendering the revealing evidence of the Visrutacarita confusing rather than clarifying. At the same time, his great Inscriptions of the Vakatakas (see Bibliographic References) omits something like eighty percent of available Vakataka inscriptions—notably, the plethora of donative records scattered throughout Ajanta during its final disruptive years (c. 479–480, according to my reconstruction).5 Indeed, Mr. Khandalavala himself innocently
which I suggest was made (or at least remarkably enlarged) as a donation to Siva by Harisena when he took over eastern Vidarbha about 475. See discussion in Chapter 2. 4 For Bagh, and Maharaja Subandhu’s inscriptions, see Mirashi, 5 I now suggest that intrusions may have been added to the “Vakataka” caves— those along the main scarp—in mid-478, when the Asmaka insurrection apparently began.
138
volume iia, chapter five
restores some of these many records to the dynasty, when he assigns Cave 2 to his previously-unheard-of “Vakataka restoration”. This is because although he says “there are no donatory inscriptions therein” the truth is that there are many, all associated with these late intrusive images; and all have been published, even if not by Dr. Mirashi.6 The Subandhu situation comes down to this. If Maharaja Subandhu’s inscriptions are assigned to the Abhira era, then as Mr. Khandalavala, following Dr. Mirashi, says: “Subandhu ruled in the very early part of the fifth century over the Bagh area”. By the same token, since his Bagh plate refers to repairs to the caves, the caves must already have been in existence by about 400 A.D. (Mirashi himself assigned them to the late fourth century). But the complex pillar forms, elaborate doorway and window designs, peristyle plans, technologically sophisticated modes of hanging the cell doors, and many other features, all link the Bagh caves to the latest excavations at Ajanta—notably Cave 2 and its counterparts, which Mr. Khandalavala assigns to c. 530–545. Thus it boggles the art-historical mind to read his opinion that on the basis of stylistic considerations “the Bagh caves are several decades earlier than the Mahayana caves of Ajanta.” And the same statement boggles the historical mind too, for “several decades earlier” than the Ajanta caves, the earliest of which he dates from c.480, would place the Bagh caves in the mid-fifth century. So how could Maharaja Subandhu repair them if he had long since been dead, as would have to have been the case if we assign his inscriptions to the Abhira era? Here again, the only solution is to recognize that Subandhu dated his records in the Gupta era, which allows the Bagh caves to fall where they belong, namely in the 460s and the 470s, as yet another one of Harisena’s achievements.7 The use of the Gupta era makes perfect sense if I am correct in my identification of Visruta with Subandhu, for according to the Visrutacarita he was a prince from Magadha. As a minor footnote to these considerations, it might be suggested that the repair of the beautiful Bagh caves by Subandhu = Visruta, shortly after he had come to power and had taken the young Vakataka princess as his bride, was but one more means of taking upon himself the role
6 7
For a useful and thorough listing of all inscriptions, see Cohen, Appendix. Spink, 1976–77.
scholarly contributions to maharashtra pathik
139
of the protector of the Vakataka heritage, and of continuing the strong support of the Buddhist faith that had characterized the reign of his wife’s grandfather, the great Harisena. It may well be that Mr. Khandalavala, like so many other scholars, is convinced that the Bagh caves belong to a period earlier than Ajanta because some of the shrines there contain stupas rather than Buddha images as the objects of worship. This tends to be the “fallback” position in campaigns to preserve Dr. Mirashi’s sacrosanct dating, since his assumptions about Bagh’s style are immediately self-defeating when one is confronted by the evidence itself rather than by assumptions about it. However, the stupa argument also cannot bear examination. There is incontrovertible evidence that some of the Bagh shrines (e.g.: those of Caves 3 and 7) did indeed have (now-missing) central images; and even more revealing, we can find a similar stupa, without any image whatsoever, in one of the much more developed caves at Kanheri (Cave 34), which dates from the period of Traikutaka patronage in the very late fifth to early sixth century. Having given our reasoning about Maharaja Subandhu and about the early decline of the Vakataka “main line”, in this task of trying to make the “incredible” credible, let us now look at the evidence surrounding Harisena’s trusted old minister, Varahadeva, whom we have just mentioned. We are very privileged, in our Ajanta studies, to be able to focus in some detail upon actual historical figures— figures for whom, in many cases, Dandin’s remarkable Visrutacarita gives a tangibility and a credibility that is all too often lacking in historical reconstructions. Mr. Khandalavala rightly recognizes that the great minister Varahadeva was one of the patrons who inaugurated the late phase at Ajanta; in fact, he dates the beginning of work on Varahadeva’s Cave 16 and the beginning of work on the site to the same year— c. 480. But what he does not respond to is my contention that when the unfinished Cave 16 (like Varahadeva’s other excavation at Ghatotkacha) was inscribed by the minister about a decade and a half later (as he and I would both agree was the case, although our dates would differ), the work then underway both on Cave 16 and on the Ghatotkacha vihara proves to be extremely late indeed, in terms of the site’s stylistic development. That is, the whole development of the site is essentially comprehended by the span of
140
volume iia, chapter five
Varahadeva’s patronage.8 If Mr. Khandalavala is going to criticize my conclusions, it would be better to suppress gratuitous thoughts about the general lack of comprehension shown by western scholars and to get down to the evidentiary nitty-gritty. The unique shrine in Cave 16, with its complex late pillars, which like so much of the cave never could be finished, because the Vakatakas’ time ran out, contains an authoritative new type of Buddha (in the pralambapadsana pose) that became the rage at other sites from this time on; but it was totally new when Varahadeva decided to use it at Ajanta when, in about 477, he decided to re-design the still incomplete Cave 16. At almost precisely the same moment, Buddhabhadra was sponsoring the same type of image in the Caitya Cave 26. It also makes its appearance in the shrine of the equally late Aurangabad Cave 3, also a Vakataka period (even if more precisely, an Asmaka) dedication; but it had never been used in any Vakataka cave before. Would Mr. Khandalavala care to assert that this Buddha, with its distinctive throne-back features and associated bodhisattvas, is not one of the very latest images at the site, as I have tried to establish in previous studies? Or that—“if stylistic perception is to have any meaning”—the associated paintings on the Cave 16’s rear wall are not equally late, and very different indeed from the early paintings toward the front of the cave that I would date from nearly a decade before? These rear-wall paintings even have the distinctive (and characteristically late) use of lapis-lazuli eye shadow, like their counterparts in Cave 2, paintings that Mr. Khandalavala quite rightly puts near the very end of Ajanta’s development. One must also note how late type doorfittings (of the type conventional for Cave 2, which in turn had been influenced by Bagh) were used to replace the neverutilized old-fashioned ones, once Cave 16 was finally nearing completion; and one must also consider how the different stages of plastering in the cave bear out the assumption that although Varahadeva did indeed start the cave very early, he finished it very late—at least to the degree that he was able to in the crisis situation of c. 478. The second example is the reverse of the first, but argues the same point. The great caitya hall, Cave 26, built by the monk Buddhabhadra in honor of the great minister of Asmaka, is generally
8
Spink, 1991A.
scholarly contributions to maharashtra pathik
141
seen as one of the latest caves—Mr. Khandalavala dates its completion to the 520s. However what Mr. Khandalavala does not recognize is that, despite its being finished so late, it was one of the very first caves started, as can be determined by an analysis of many relevant structural features.9 The astylar character of its associated upper and lower wings alone would seem sufficient proof of this, but one can further point to the cave’s thin walls, door fittings, pillar configurations, and even its constricted central image to justify the conclusion that the cave was started extremely early, being finally finished with its typically “late” stylistic overlay only when the Asmakas (feudatory to the Vakataka emperor Harisena and then briefly to his son) had gained control of the region by defeating Upendragupta, the local king. Thus, just as in the case of Varahadeva’s Cave 16, the development of Cave 26, from its very early inauguration to its very late conclusion, was all the work of one man—the powerful monk Buddhabhadra—whose patronage of the cave (and indeed of the site’s whole western complex—from its inauguration to its abandonment spans Ajanta’s whole Vakataka period development, except for the final year or two of its accumulating disruption. To this group of persons who saw the whole site develop from start to finish we could further add the monks Bhadrabandhu and Dharmadatta, “who (saw) to the excavation and completion of (Cave 26)” for Buddhabhadra (Cave 26 inscription, verse 14), as well as the rich but long suffering Mathuradasa, whose massive Cave 4, the largest at the site, suffered a disastrous collapse of its ceiling in mid-course.10 Indeed, Harisena’s reign, following an apparently normal course of succession after his father’s death, also comprehends the whole period of the site’s consistent development—a flowering which was suddenly stopped by his death. As I pointed out in the previous issue, but beg to repeat here, the implications of such “actuarial” considerations cannot be escaped; nor should they be, for they allow us to assign Ajanta, once it has been properly analyzed, to its rightful place in history. Assuming that these powerful patrons (to say nothing of their many anonymous contemporaries) were all established and mature individuals when they inaugurated the site, the
9 10
See Defining Features in Volume IV. Inscr. 17.
142
volume iia, chapter five
striking fact that (with the exception of the just-deceased Harisena) they (and surely most of the unnamed patrons too) were all still alive at the time of its decline would make it impossible to believe that the site’s total development comprised more than a few decades at the very most. And this is of course precisely what the epigraphic evidence, when read anew, demands. Since its inauguration and its flowering are so indissoluably linked to the reign of Harisena and to the tenure of his minister Varahadeva, it must have been begun sometime after 458 A.D.; for at that date, according to the Hisse Borala inscription, Devasena (Harisena’s father) was still reigning. Similarly, since the disruption of all patronage at the site is so clearly connected with the events leading to the Vakataka’s fall, work must have ended a number of years prior to 486 A.D., by which time Maharaja Subandhu (probably the impressive but scheming “Visruta” of the Visrutacarita) was already reigning over Mahismati, once a central city in the Vakataka domains. It was here, according to the Visrutacarita, that the old minister (perhaps Varahadeva) sought refuge for Harisena’s two grandchildren, after their father, Sarvasena III, had been killed in the great insurrection organized by the Asmakas; for it was here that the Vakataka viceroy, a second son of Harisena, had been able to briefly cling to a fragment of the shattered empire. There seems to be no reason to question the validity of the consequent dating—sometime in the early 480’s—for the Vakataka fall, which is well-nigh demanded by the proper dating of Subandhu’s inscription to the Gupta era. Indeed, the same conclusion (supported by the evidence in the Visrutacarita) is demanded by inscriptions of the Traikutakas, as well as one of the ruler Gauri, which confirm our assumption that western India could no longer have been under Vakataka control in the 490s.11 Similarly, although scholars will have to resist the conclusions as long as they insist that Harisena reigned in the last quarter of the fifth century, there is strong evidence that the Visnukundins, the afflicted Guptas, and various lesser powers, were similarly taking advantage of the Vakataka fall, by establishing their control over eastern Vidarbha and adjacent regions in that very period. This is what the site itself reveals to us when we look critically at its forms, and realize how the powerful patrons who started it with
11
Spink 1991B, esp. “Chart Referring to Historical Synchronisms”.
scholarly contributions to maharashtra pathik
143
such exuberance ended up so despairingly when time ran out at the moment of the death of the great Harisena in c. 477. The trouble was of course caused by the Asmakas. Having taken over the region, they now planned to take over the whole empire, seizing the opportunity afforded by the accession of Harisena’s weak and ineffectual successor (Sarvasena III), even before he had got settled on the throne. (Mr. Khandalavala somehow concludes that I think Harisena was still ruling at this time, which totally confuses the discussion.) The long inscription in Cave 26–made, significantly, in honor of the Asmaka minister—clearly signals the insurrection which, starting just at this time (c. 478) so quickly spread disaster throughout the site by disrupting its financial and administrative support. Just as happened in the case of certain newly assertive Gupta feudatories, who praised themselves and not their overlords during the hard days when that empire was going into decline, so here the Asmakas exalt themselves while (ominously!) the Vakataka overlord is not even mentioned. Indeed, within a less than a decade he was dead, and his dynasty with him. The reconstruction of late Vakataka history and of the history of the site itself which is demanded by the evidence at Ajanta makes a perfect “fit” with the rich body of epigraphic and literary evidence available from other sources. As I have tried to explain, everything works together to provide a coherent picture of one of the most important and interesting moments in India’s history. Given such a trove of mutually supportive documents, it is both amazing and sad that, by and large, neither Ajanta, nor the great emperor Harisena, have as yet been given the particular recognition they deserve; for if my analyses are correct, what might at first seem “unbelievable”, is believable after all. “All this may be high drama and excellent fiction, but (it) is not authentic”, says Mr. Khandalavala. I suggest that that is for the reader, after checking out the evidence at first hand, to decide. b. Reply to Dr. A. Jamkhedkar I have attempted to supply a convincing reply to Dr. Jamkhedkar’s challenging queries about Vakataka chronology in Chapter 3 above, entitled A Revised Vakataka Chronology. This chapter is so closely based on the reply originally written for Pathik, that it would serve no purpose to repeat it here.
144
volume iia, chapter five
c. Reply to Shri Brahmanda Deshpande Shri Deshpande gives a very succinct and useful summary of Dr. Mirashi’s presentation of historical data in the Ajanta Cave 16 inscription and the Visrutacarita, which bears upon the decline of the Vakataka empire. Whether or not I have utilized the inferences of Dr. Mirashi “without proper scrutiny” is for the reader to decide; I would have thought such a criticism might apply better to those who accept them without question. However, the later assertion that I accept the inferences “in toto” I must take exception to. I am trying to establish a very different view of this premier period in Indian history; and to do so all of the available evidence must be considered afresh, with presuppositions kept as far in the distance as possible. If we believe, with Mirashi, that Harisena ruled in the fourth quarter of the fifth century, rather than in the third, and then try to make sense of the relevant documentation, whether it be inscriptions or monuments or the Visrutacarita, we may find only confusion; in fact, I am insisting that we will. Any scholar doing research fifteen hundred years from today who tries to show that the events surrounding Indian Independence took place in the 1970s rather than the 1940s may be able to convince their colleagues, by a sufficiently insistent presentation of the fragmentary evidence, but this does not mean that they will be correct, or that others should not properly scrutinize their work. I was interested in Shri Despande’s information that Dandin’s ancestors migrated from the Vakataka heartland, for it gives another reason why Dandin apparently took such an interest in reflecting the historical situation so precisely in the eighth chapter of the Dasakumaracarita. However, Dandin’s remarkably “total recall”, evidence for which emerges when (and only when) we date Harisena’s reign to the 460s and 470s, must still ultimately depend upon the magnitude and “unforgettability” of the events recorded: the continentshattering trauma of the empire’s fall. Shri Deshpande feels “at a loss to solve the riddle of this sudden decline”; but it seems to me that Dandin’s detailed explanation is clear and convincing. If there are occasional excursions into hyperbole, we must remember that the storey is, after all, cast as a novel. Incidentally, both the Visrutacarita and the Cave 16 inscription would counter Shri Deshpande’s assertion that the “Traikutakas were never the feudatories of the Vakatakas.” It is true, as he says, that none of the Traikutaka records mention the Vakatakas as their over-
scholarly contributions to maharashtra pathik
145
lords; but again he is following Dr. Mirashi’s dating of Harisena to the period when (according to my view) the Traikutakas were indeed independent. Significantly, there are no such records in the 460s and 470s when Harisena apparently did control the region. It seems to me that Shri Deshpande unnecessarily rewrites the Visrutacarita, in assigning the account of the invasion of the Vakataka empire by the king of “Vanavasi” to a point in time when it has nothing to do with the sobering role which Dandin assigns it. It is true that “there is no epigraphic record of any invasion”; but why should we expect such a record, when the perpetrators of the insurrection left no positive achievement to make up for the destruction of the empire that they attacked. The lack of inscriptions—the “argumentum ex nihilo”—proves essentially nothing. As for the suggestion that Vanavasi should be read not as “Vanavasi” but as a “forested region” (which rather mysteriously turns out to be the Bastar region in central India), such an interpretation hardly aligns with the straightforward readings given (and apparently accepted by Shri Deshpande) for all of the other regions that Dandin so precisely lists. Admittedly, as Shri Deshpande points out, “Vanavasi or the modern Belgaon District of Karnataka state is far away from Vidarbha”. However, what he refuses to recognize is that the huge Vakataka empire, as is suggested by the evidence of the Ajanta Cave 16 inscription as well as that of the Visrutacarita, extended well into the south at this time, apparently incorporating much or all of both “Andhra” and “Kuntala”. So “Vanavasi” was by no means “a distant country” which could not invade the Vakataka domains. Indeed, there is even evidence that Harisena’s inheritance had included portions of northern Karnataka, as A.M. Shastri has pointed out in his recent “Vakatakas in Karnataka”. So once again the available evidence would advise us not to manufacture a new reading when we have a convincing one before us. I would be the first to admit that my “identification (of Maharaja Subandhu) with Visruta” of the Visrutacarita of course goes wrong if, as Shri Deshpande believes, “the date of (Subandhu’s) copper plate given by Dr. Mirashi is correct.” That is self-evident. The question requires not such an unsupported assertion, but rather a critical scrutiny of my view that the plate has to be dated in the Gupta era, a conclusion based primarily (although by no means entirely) upon a precise analysis of both Bagh and Ajanta, showing their equally intimate links to Harisena’s reign. Dr. Mirashi’s dating of
146
volume iia, chapter five
the Bagh caves, as a result of his epigraphic considerations, to a period nearly one hundred years earlier than Ajanta produces pure chaos from an art-historical point of view. At least Shri Deshpande does not go so far as to similarly place the Bagh caves in the late fourth century, even though he accepts the epigraphic assumption that demands such a dating. The validity of the assertion that Harisena died without warning in the late 470s can only be judged by analyzing the situation in his own poignantly unfinished Cave 1, and in those other caves throughout Ajanta that reflect, in their suddenly frenzied rush to complete the shrine image, this same event.12 Admittedly, one is wise not to swallow such a statement without chewing on the evidence; but on the other hand, one should not merely spit it out, without some fair attempt to digest it. Such a digestive process might include taking into account the lengthy arguments that I have published elsewhere, even if they are expected to be unpalatable. If this also should require going to Ajanta to partake of the feast so lavishly offered there, that is all to the good. Indeed, my overall conclusions, while admittedly dependent upon epigraphic and literary evidence, are at the same time so linked to a detailed analysis of the site itself that any proper critique of my chronology must take that art-historical material into very particular account; and this is nowhere more true than the evidence bearing upon Harisena’s death. Not the least of the benefits of setting the Vakataka house in order is the manner in which the history of many of the Vakataka’s contemporaries—the Guptas, the Traikutakas, the Visnukundins, the Ucchakalpas, the Nalas, the Kadambas, and so many others—must be reconsidered, and clarified in the process. Ultimately, of course, the evidence accumulating around Harisena and his activities proves that he was by far the greatest king of his age. Again, the pride of my Maharastrian critics should be responsive to my assertion: that it was under the Vakataka emperor Harisena—the ruler of ancient classical culture Vidarbha—not under the weakening Guptas, that Indian achieved its final, even perhaps its fullest, flowering.13 And Ajanta, magnificently, reflects those halcyon days, preserving and revealing the otherwise vanished picture of those times. 12
Spink 1991A. The gradual weakening of Gupta power and authority, starting during the reign of Skandagupta, in the 470s, was probably exacerbated by Vakataka inroads, particularly in territories along their common border. 13
VOLUME IIB, PART I
PATTERNS OF PATRONAGE
CHAPTER SIX
PATRONAGE: CONSISTENT VS. COLLAPSING
Technically, everything done at Ajanta between its inauguration in 462 and the abandonment of its artistic patronage in 480 may be considered Vakataka. This is because Asmaka, although it insultingly dishonors the Vakataka overlord in mid-478 (Cave 26 inscription), was still in a de jure even if not de facto feudatory status from the Vakataka point of view; the Asmakas’ war for independence was yet to come. However, from mid-478 the Asmakas, established in Ajanta’s western extremity (Caves 21–28), were clearly in control of the whole site. The old “Vakataka” patrons, shocked by Harisena’s sudden and ominous death on “December 31, 477”, had already rushed their shrine Buddhas to dedication and fled the area by the middle of 478. As far as we can tell from the frantic course of patronage ignited by the death of the great Harisena, all of the fifth century caves in the “main” or eastern section of the site were made by these suddenly threatened “Vakataka” patrons, whereas work on the Asmaka caves continued with nothing of the same type of anxiety until, surely due to the erosion of funding, they too had to be abandoned at the end of 478. To allow a distinction between the “western” caves sponsored by the Asmakas on the one hand and the caves along the “main” scarp, suddenly under a highly unstable patronage due to Harisena’s death, I speak of the latter as “Vakataka” caves, and the threatened (non-Asmaka) patrons similarly as “Vakataka” patrons. Used without quotes, the term does not suggest this distinction but refers to the two feudatories’ common overlord. See also Volume I, Introduction: A Few Conventions. Ajanta’s Vakataka phase of patronage, dating approximately 462 through 480, took place exclusively during the reigns of that dynasty’s last two emperors, the great Harisena (c. 460–c. 477) and his inept son and successor, the ill-fated Sarvasena III (c. 478–c. 483). As argued earlier, Ajanta’s startling and sudden abandonment was due in large part to the latter’s shortcomings.1 By way of contrast, the 1 See particular the arguments in Volume I, involving evidence both from the site itself and from Dandin’s Visrutacarita.
150
volume iib, part i, chapter six
site’s remarkable flowering was due to his father, the great Vakataka emperor Harisena, “a partial incarnation . . . of the God of Justice” (Kale 1966, 349). To give credit where credit is due, I have asserted that this sadly short-lived emperor may well have been the greatest ruler in the world in the 460s and 470s.2 Although Harisena’s own palaces have turned to dust, and even his name is still largely forgotten, the energy and the expansiveness of his reign are reflected for all to see in Ajanta’s mirroring of the Vakatakas’ illustrious heyday. Clearly intended to emulate but at the same time to outdo the impressive rock-cut monuments of three to five hundred years before, the Vakataka caves, nearly all started in the early 460s, were conceived by their courtly patrons as sumptuous offerings “which cannot be even imagined by little-souled men” (Ajanta 17 inscription, verse 25) “Having expending abundant wealth” (ibid.) and clothing them “in the brilliance of Indra’s crown” (Ajanta 16 inscription, verse 29), the courtly patrons did indeed intend them to “rival the palaces of the lord of gods.” (Ajanta 16 inscription, verse 27) From such comments by the various patrons about their donations and from what we know about the status of the major donors, it is evident that Ajanta, until its sudden collapse under Sarvasena III, was an insistently elitist site, with productive connections to both the imperial court and its feudatory satellites. The imperial Prime Minister, “having amassed a large store of religious merit . . . and being extremely devoted to the Buddha” (Ajanta 16 inscription, verse 21) was probably the most powerful inaugurator of the site. He put his huge vihara, “adorned with windows, doors, beautiful picturegalleries, ledges, statues of the nymphs of Indra, . . . beautiful pillars and stairs . . .” (ibid., 24) at the very center of the site, an appropriate position for a patron of his eminence. His monumental “Elephant Gate” was still remembered, even after the site’s collapse, and referred to by Zuan Xang, as the “entrance to the site” where the great stone elephants still “sometimes. . . . utter a great cry and the earth shakes throughout.”3 2 This somewhat grand statement has received much amused criticism; but the critics have not yet named a rival. I might have more difficulty defending my claim (Spink 1992, 17) that Ajanta “may well be mankind’s most remarkable creative achievement”, which Cohen (1995, 55) justifiably classifies as a “memorable hyperbole”. Even so, in terms of quality, quantity, efficiency, and historical significance, Ajanta can hardly be bettered. See also Volume I, Chapter 2: A Final Comment 3 Zuan Xang (Hsuan Tsang), trans Beal, 1881, book XI, section on Maharashtra.
consistent vs. collapsing
151
Upendragupta, the local king, profligate in his piety—in his wish to “cover the earth with stupas and viharas” (Ajanta 17 inscription, verse 22)—availed himself eagerly of “the power of the expenditure of wealth” (ibid., verse 17), but used it for pious purposes which were of no avail when the aggressive Asmakas attacked and absorbed his kingdom. Nonetheless, of the twenty four excavations (See Time Chart) started in the first decade of Ajanta’s activity, before Upendragupta’s ignominious defeat, no less than five lavishly conceived caves were the products of his misguided munificence—excesses which he must have thought of as a source of virtue. The great monk Buddhabhadra, who was personally responsible for all of the vast Asmaka developments in the site’s western extremity (Caves 21–28) was also “born of a noble family” (Ajanta 26 inscr., vs. 16); benefiting from his high-born status, he was “endowed with great learning . . . (and) perfectly mastered the course of the Buddha’s teachings” (ibid.). Although “he became a monk in his early age” (ibid.) he kept his connections with the Asmaka court; and in his inscription not only praises “the mighty king of Asmaka” (Ajanta 26 inscr., vs. 9) but eulogizes both the recently-deceased Asmaka minister, “who was attached to him in friendship through many successive births” (ibid.) and also that “equally foremost personality” his son, the new minister (Ajanta 26 inscr., vs. 12). Although devotees from the courts and from the region, as well as the merchants and mercenaries coming to Ajanta via the nearby trade route, must have all made merit by contributing to the development and support of the site in various ways, we might suspect that Buddhabhadra, in his desire “to set up a memorial on the mountains that will endure for as long as the moon and the sun continue” (Ajanta 26 inscr., vs. 8), drew most of his funding from this close connection with the Asmaka court. The patron named Mathura, who sponsored Cave 4, left the only other inscription connected with the development of the site during its heyday. This record, incised hastily at the “last minute” in the troubled context of early 478, had to “make do” in place of the long and lavishly composed one which he surely would have supplied, had time not been so short. We know him only as the Zuan Xang did not go to the abandoned site when he traveled through the region in the early 7th century, but heard reports about it, showing that it was still known and famous.
152
volume iib, part i, chapter six
“viharaswamin”—the owner of the cave (Ajanta Cave 4 inscr., line 2)—who managed to get his colossal Buddha image dedicated even though the cave itself was not even fully cut from the rock. However, he was obviously a rich and powerful man, since his vihara is by far the largest at the site; and he persisted in its excavation despite the setbacks caused first by the sumptuary restrictions imposed by the local king in anticipation of the Asmaka/Risika war, and then by the further delays occasioned by the war itself. As if the problems caused by the region’s turbulence were not enough, Mathura’s dreams must have turned to nightmare when the vast main hall ceiling, by far the largest at the site, dramatically collapsed. Just as the awed architects, excavating the hall’s vast span, might have predicted: “Pride goeth before a fall”. But still the proud and persistent Mathura did not give up; and his reward, in the end, was to have created (albeit hastily) the largest Buddha image at the site, as the focus of worship in his troubled cave.4 With what seems to be a convention among donors—who almost invariably “transfer” the merit from their donations, he declares: (Inscr. 17) “Whatever merit there is in it, may that serve [his] mother, father, and paternal grandmother—to whom belongs the principle share—, as well as all sentient beings’ attainment of unexcelled knowledge.” Finally, and perhaps most important for establishing the view that Ajanta, during its heyday, was a purely elitist undertaking, we have evidence of the direct involvement of the emperor Harisena himself in this vast project. Although he surely approved and must have helped to fund the site from the start, he did not actually get his superb Cave 1 started until almost five years had passed, at which point all of the more desirable spaces had been appropriated by patrons whose caves were already underway. From Cave 2 at the eastern end to the Cave 26 complex at the western extremity, there was no satisfactory spot, at least at the “normal” level, where excavation was not already proceeding.5 The poor rich emperor had to be satisfied with the relatively low rise of rock just to the right of
4
For the Cave 4 Buddha image and its development see Volume I, Chapter 11 Late in the site’s development, a few caves (3, 14, 22, 23A) were excavated at a higher level, above the earlier excavations. The available scarp at the extreme end of the site’s western extremity may have seemed too remote as a location, and relatively difficult of approach. 5
consistent vs. collapsing
153
Cave 2, the very last workable area available at the accessible level. But of course he made the most of it; the vihara (Cave 1) which he created turned out to be, not surprisingly, the most sumptuous in India. Perhaps Harisena’s tardiness in getting involved was due to the fact (it is almost certainly a fact!) that, unlike his Prime Minister and the other courtly donors who patronized the site, he was a Siva worshipper.6 Indeed, I have suggested that when he absorbed the “eastern branch” of the Vakatakas in the mid-470s, he may also have been responsible for the staggeringly impressive enlargement of the Hidimba Tekdi near Mansar, a Saiva monument which rivals Ajanta as an achievement, and may have been his own compensatory offering to the Hindu god of his ancestors.7 In any case we must assume that Harisena, although perhaps not a “pure Buddhist”, was clearly supportive of Buddhism; and he may surely have gained political as well as personal benefit from his remarkable benefaction, whose highclass innovations rapidly became a model for the site.8 If the site was already flourishing even before the emperor Harisena’s direct involvement, the excitement and the accomplishments further burgeoned under his increasing commitment, even surviving the various crises with which it was afflicted. His central significance as a patron—but more particularly the force of his rule over the whole of central India from sea to sea—was tragically revealed only a little more than a decade from the beginning of work on his Cave 1, when he suddenly and unexpectedly died. Within less than a year from this date, at about the end of 477, all of the old established patronage of the site had collapsed due to the political repercussions—the dire consequences—of his death. What can only be
6
Bakker 1997, 40 and 58ff. For Hidimba Tekdi, see Bakker 1997, 87–88; he connects the remains with Pravarasena II. However, the splendid sculptures found at the site appear to be so allied, stylistically, with Ajanta, that I would prefer to see them as being done (perhaps even by some of the same artists) when Harisena took over this region and (I would hypothesize) enlarged Pravarasena’s earlier monument. Compare, for instance, the famous Siva image with a seated yaksha in the frieze at the right end of the porch of Ajanta Cave 21, for which I suggest a date of 477. Bakker has similarly compared the Mansar figure with the great guardians of Cave 19 (Bakker Plate XXXVIIIA), which I would date to about 470. 8 For the arguments supporting the ascription of Cave 1 to Harisena, see Spink 1981. For the matter of religious tolerance, see Narain 1983. 7
154
volume iib, part i, chapter six
described as a “shock wave” sent its deadly tremors through the site as soon as this startling news arrived.9 During the year following Harisena’s death, first the “Vakataka” patrons, after rushing to finish and dedicate their shrine Buddha images, hurriedly left the site. Then, only a few months later, the Asmaka patrons also left, their intense connection with the site now a casualty of the war effort. It was only after the established patrons had departed, leaving their remarkable excavations now unattended, that the many monks who remained in a sense stranded at the site, along with local devotees, rushed to make their own votive offerings. They now filled the available and desirable spaces on the façades and the interiors with a spate of “private” images. Essentially, these images are all Buddhas, although in a number of cases the Buddha is in stupa form, and in a few in his imminent form as the bodhisattva Avalokitesvara.10 But wherever these intrusions appear, he is only found in caves which had been ritually brought to life by the generally hurried dedication of its central image.11 The point of the above review is to support the assertion that Ajanta, during its crescent phase—that is, during the reign of Harisena—was a purely elitist, exclusive, courtly site. In fact, there are no less than eighteen caves which are connected either directly or indirectly through inscriptional evidence with either the imperial, or the feudatory courts, while it seems reasonable to add Caves 1 and 4 (as argued above) to this high-priority list.12 By contrast, there are only ten caves whose patrons are not known, and even in those cases it is likely, given what we know of other patrons, that all or most of these patrons were also
9 For the situation of the anxious patrons in 478, see Volume IB (The Shock Wave and the Site). 10 Throughout the site, during the Period of Disruption, relief stupas can be substituted for Buddha images, their “meaning” apparently being the same. 11 The requirement regarding intrusions, which are only placed in caves which have been “brought to life” through the dedication of their shrine image, is discussed at length, particularly in Volume I. 12 I have tried to show that the monk Buddhabhadra, with his connections at the Asmaka court, either directly or indirectly controlled Caves 21, 22, 23, 23A, 24, 24A, 25, 26, 26RW, 26LW, 28. Caves 17, 18, 19, 20, and 29 were sponsored by the local king Upendragupta, Cave 16 by the Prime Minister Varahadeva. Cave 4’s patron, Mathura, was so well established, that he must have had courtly connections. Cave 1 is not inscribed, but is undeniably the donation of the emperor Harisena, as discussed in Volume I, and elsewhere.
consistent vs. collapsing
155
either courtiers or had courtly connections.13 Indeed some of these “anonymous” excavations were probably sponsored by the same donors listed above; patrons, sometimes for reasons of time, did not necessarily inscribe every monument, and of course it is reasonable to assume that some inscriptions have also been lost. In any case, it seems fair to say that it would have been impossible to get into this upscale monastic “housing society” without “a friend at court”. However, even if one had such a “friend at court”, one could not just come into the site, either when it was being started in the early 460s, or was developing over the course of the following years, and decide to sponsor an excavation on one’s own in some space (generally very constricted after 466) that was still available. The site, from the start, was clearly under strong administrative controls, which extended both to planning and protection, like some benign Mafia compound. Whereas in the Hinayana phase, the five excavations of that period were expediently located-the old caitya halls are at a relatively low level, close to the riverine approach and spaced almost at random— in the Vakataka phase it is evident that space was neither wasted nor allowed to be wasted between the caves.14 In general, they are so “efficiently” and tightly placed adjacent to each other, that there were often problems when, due to changing taste, certain enlargements (as in both Cave 7 and Cave 8) or extensions (as in many porch cells) were desired. We might also note the evidence for shared approaches and walkways, which not only connect the various caves with each other, but suggest a deeper collegial connection and attitude of cooperation. Only the Asmaka caves (from 21 to 28) are separated from the main group, and this was (at least originally) merely due to geological factors; the seasonal waterfall between Cave 20 (belonging to the local
13 Caves where the patrons are not known (but some may have been sponsored by patrons known from other caves): 2, 3, 5, Upper 6, Lower 6, 7, 8, 11, 14, 15 14 No excavation abuts Cave 5 on the right, because the rock there was so bad; it is for this reason that Cave 4 had to be set much higher up. Cave 20 has a monsoonal waterfall on its left, and is at a slight distance from Cave 19 to take advantage of a better rock face. Of course there was more room up above on the scarp face, but this was obviously low-priority; see Cave 3, Caves 14, 22, 23A— all very late “additions” at the site. For the shift from single to complex porch cells, and the consequent space problems, see especially discussions of Caves 21 through 24, with the consequent “invasion” of the Cave 26 complex see earlier in this Volume.
volume iib, part i, chapter six
156
Risika king) and Cave 21 (belonging to the Asmaka group) necessarily divided them. As noted above, the old “entrance to the site” was probably through the prestigious minister’s Elephant Gate, of Cave 16, at the very center of the site. Even today, this is an appropriate and convenient access. We have referred elsewhere to the power of the rule of fashion at the site.15 Over and over, artistic motifs, plan characteristics, or technological features that were in fashion one year were out of fashion the year after. This clearly suggests “controls”, even if they often operated through subtlety rather than statement. What is clear is that the site, during its heyday, was in no way a “free-for-all”. In the same way, the fact that the monk Buddhabhadra clearly oversaw both the location and the development of all of the many caves at the western extremity of the site, supports the view that throughout the whole site, a firm and insistent supervision was the rule. These administrative controls almost certainly extended not only to “where” but to “whom”. It is a remarkable fact that, although there are hundreds of “intrusions” at Ajanta—sculptures and paintings that have nothing to do with the patrons’ programs of decoration, and indeed typically violate them—such intrusions are never found until the years when the site was collapsing, after Harisena’s death, as well as after Sarvasena III’s very brief overlordship of the area. Even though there are many years (notably 469 through 471) when work on all but the royal caves was totally abandoned, not a single eager “uninvited” devotee was ever allowed to put a single image in them. Even during the Hiatus—the period of the Asmaka/Risika warwhen the site was so deeply affected that all work totally stopped even on the royal caves—not a single intrusive image was ever put in or on any of the caves. This is the more remarkable because, in the Period of Disruption, the intrusive donations were mostly made by monks—presumably those monks still resident at the site. Yet such monks neither made nor were allowed to make such gifts in this earlier period. Again, this emphasizes the exclusiveness—the elite patronage—of the work done while the site was under effective and assertive imperial control—a control which appears to have parallels in the admin-
15
See chapter 2.
consistent vs. collapsing
157
istrative virtues of the emperor Harisena himself, who (back in the capital) “himself looked after his treasury and conveyances, took great pains to supervise (and appreciate) the work of the heads of the various departments, and encouraged with honor and fitting rewards those who well acquitted themselves of the duties entrusted to them” (Kale 1966, 349). Even admitting a degree of overstatement in the statement, we have the evidence of Harisena’s remarkable empire as witness to its credibility.
CHAPTER SEVEN
LOCATING INTRUSIONS IN TIME
It is important for us to locate the plethora of intrusions at Ajanta to the period in which they were actually done—a period which turns out to be strictly limited to the brief years after Harisena’s death. To assume, or conclude, that some are referable to the site’s heyday would confuse the proper understanding of their brief development—a development revealed by the surprising changes not only in their iconography but in the manner in which they were selected, located, and utilized. Thus we must satisfy ourselves that all of the intrusions at Ajanta and other Vakataka sites do indeed belong solely to what I have described as the Period of Disruption. For those patrons occupied with donations along the main face of the scarp—the “Vakataka” patrons, whose fates were linked to that of the “de jure” imperial overlord—the disruption began in mid478, the crisis in their patronage being reflected in their hasty departure from the site. For the patrons busy on the caves at the site’s western extremity—the Asmaka patrons—troubling concerns about the future clearly began at this same time, but their efforts were not in fact totally disrupted until some months later, at the very end of 478. At that point, it appears that the previously well-established Asmaka support of the development of the latter caves was suddenly cut off. This was probably because of the mounting monetary needs of their military buildup, but possibly involved the need for a fighting force too; in any case, the result was that Asmaka patronage stopped completely at the end of this ominous year. Thus from mid-478 in the caves along the main scarp—the old “Vakataka” section—and from the end of 478 in the Asmaka section to the west, the site was open to “intruders” and the Period of Disruption began. It was at this point, and never before, that the new disruptive activity with which we are here primarily concerned started so vigorously. But it too had a precise conclusion; at the end of 480, the donations suddenly stopped, often in mid-course. As I shall argue, this was because the growing insurrection of a large coalition of Vakataka feudatories against the imperial house, insti-
locating intrusions in time
159
gated and led by the Asmakas, suddenly impacted directly upon the Ajanta region. The consistency of certain significant (even if sometimes seemingly minor) iconographic or stylistic or technological features is of particular help in determining that all of these votive intrusions belong to the same very late period, between mid-478 (starting in the “Vakataka” caves) and the end of 480. For instance, one would never find the ground-brick plaster characteristically used to surface walls for many late intrusions in contexts earlier than 477; nor would one find the use of blue pigment, which must have got cheaper as time went on, applied both lavishly and wastefully even to the armpits of certain Buddha images in contexts earlier than the Period of Disruption.1 Similarly, there are a great number of exclusively late iconographic or stylistic features which become almost standard in the Period of Disruption, but will never be found prior to 477. Thus a few carved bhadrasana Buddhas, to take a prime example, can be found in 477, when the type first came into favor, and they become increasingly—almost tumultuously—popular over the course of the next three years; but after 478, they are invariably intrusive. To establish the point that none of the multifarious intrusions at Ajanta can be assigned to the period when the site’s original patrons were still in control, we can briefly list a few of the more evident features which are characteristic of Buddha images (specifically carved ones) in the Period of Disruption (See Chart: Development of Iconographic Features).2 These features or forms cannot ever be found at the site prior to 477—the final year of Harisena’s reign, when the developments at the site were probably more vigorous than they had ever been. Something of the old energy, but now fed by anxiety rather than by optimism, continued in the “Vakataka” caves during the first months of 478, but the focus now was almost exclusively on completing the shrine images before time ran out. Happily, developments continued more confidently in the Asmaka excavations throughout
1
Maybe as the site collapsed the price went down drastically. But is it really lapis lazuli; could it be cobalt blue? Tests have never been made on these pigments. 2 The chart refers primarily to carved images. Curiously, a few important forms are found in painting a decade in advance of their sculptural representations. E.g.: the bhadrasana image, Buddhas and attending Buddhas.
160
volume iib, part i, chapter seven
the whole of 478, for the simple reason that the Asmakas were the source of the threat rather than the object of it. Thus, at least wherever Buddhabhadra focused his interest and his surely declining funds, most particularly in the decoration of his great Caitya Cave 26 with its sumptuously carved scenes of the Buddha’s Parinirvana, Temptation, and Sravasti Miracles, one can sense this continuing confidence. As one can understand, many of the features which are found—sometimes for the first time—in the Asmaka caves during their continued development throughout 478 have a strong impact on the intrusions of 479 and 480. Even in the brief and troubled Period of Disruption itself, as will be clear in the detailed discussions, we will find developments which will identify certain images as belonging to (or defining) an early context and others as belonging to a late one, despite the fact that we are dealing with a time-span comprising only mid-478 to 480. For instance, as a study of the intrusions on and in Cave 26 suggests, bhadrasana images, despite their oncoming popularity, probably were not carved in intrusive contexts until late in 479, and then only in large format; small bhadrasana panels appear only in 480.3 Groups of carved padmasana images in 479 typically have alternating mudras, whereas in 480 they are shown only with the dharmacakra mudra. This was perhaps due to the “authority” of the new bhadrasana image, which is invariably depicted with this particular teaching mudra, a factor that may have influenced the surprisingly rapid evolution of the type.4
3 Both in Cave 26 and in the closely related Aurangabad 3, a few bhadrasana figures, dating to 477 or more probably 478, are woven into the decorative schema; but they are neither carved as separate “panels” nor donative in type. 4 My “absolute” dating is of course a convenience, to separate the earlier carvings (“479”) from the later (“480”).
CHAPTER EIGHT
COULD ANY INTRUSIONS DATE BEFORE MID-478?
Except for the special case of Cave 4, where the cave’s anxious patron had to be satisfied with a hasty dedicatory record applied to the shrine Buddha itself at the last minute, rather than placing it in more “public” positions like those of Caves 16, 17, 20, and 26, the presence of donative inscriptions on or beneath iconic imagery in the Vakataka caves invariably identifies them as intrusions. Of course, this does not tell us when the intrusions were done; nor does it prove that they are referable to the Period of Disruption. Is it possible that intrusions—whether inscribed or not inscribed—could have been put in at least some of the caves by eager private donors during the years of the Recession (469 through 471) when many patrons had left off working on their own caves? Or could intrusions have been added, with more ease, to caves during the Asmaka/Risika war (472–474), when no work was going on whatsoever, and when the local king, who had imposed his sumptuary restrictions earlier, was fighting other battles now? The answer to such questions seems clear, and clearly negative. During the “Recession”, with most work stopped at the site by executive fiat, it is hard to imagine that some “outsider” (even an eager monk residing at the site) would have been able to “invade” these dormant caves. It is not likely that such aspiring “outsider” donors could violate the sumptuary restrictions; nor is it likely that the original patrons were totally neglecting the “security” of their unfinished and temporarily abandoned excavations, even in time of war. Indeed this war period (= the Hiatus of roughly 472 through 474) appears to have been a time when the site was totally closed down with regard to patronage activity. Even the emperor’s cave—although he was generally immune from local concerns—was not continued at this time, as we know from clear evidence of a break in its development at the time of the Hiatus. So this was hardly a time for intrusions, even if they had been allowed by the “establishment” at the site. In fact, most of the painters and sculptors, if not conscripted, must have gone away for this brief period of conflict, to find work
162
volume iib, part i, chapter eight
at Bagh and in other areas outside of the realm of war; so there probably would have been no one to make such images anyway. Of course, we might ask why eager donors did not put intrusions into the caves of the local king, after his defeat by the Asmakas. It is clear that three of these caves (17, 19, and 20) had been dedicated (and thus “brought to life”) by Upendragupta in late 471 before he lost his control of his caves and indeed of the whole site. However, although for practical reasons Cave 17 continued in use to house the monks and was in worship, Caves 19 and 20, apparently because of their connection with the defeated local king, were “off-limits”, so it seems clear that no intrusions were allowed in them, or even in Cave 17, while the Asmakas were controlling the site.1 However, the reason that they attracted no intrusions may be simpler than this. The probable explanation is that, throughout the site, all such private donations were totally disallowed while an overriding authority was present. The fact that this authority was probably imperial rather than local would explain why, during the Hiatus, when Asmaka and Risika were surely more concerned with the requirements of conflict rather than with the site’s fate, not a single intrusion was put in or on Upendragupta’s caves, or for that matter in any of the caves at the site. It was only when the site was in its great final decline, after its established patrons had departed, that Caves 17, 19, and 20 were taken over by eager new donors and were utilized as fit locations for intrusions. Only Upendragupta’s Caitya Cave 29 remained untouched at this time, for it had never reached a stage where it could be dedicated, and thus could attract no intrusions. This same reasoning could explain the surprising fact that the contemporaneous Vakataka site at Bagh apparently has not a single intrusion, even though it continued in worship long after Ajanta’s patronage had totally collapsed. But its situation was quite different. It was in Anupa, protected first by the authority of Harisena’s viceroy, and immediately thereafter by authority of the Gupta prince Maharaja Subandhu of Mahismati—Visruta of the Dasakumaracarita—who not only took over the control of the region but married Harisena’s granddaughter. And we know that he took a personal interest in the
1 For the usage (or non-usage) of Upendragupta’s caves during the Asmaka ascendancy, see Chapter 6.
could any intrusions date before mid-478?
163
established site, for he authorized extensive “repairs” at the old site, as well as support for its various needs.2 The only caves at Ajanta where we might logically seek for intrusions carved or painted earlier than the “normal” Period of Disruption (mid-478 through 480) are Caves Lower 6, 7, 11, and 15.3 These are the caves of the patrons favored in 469 by the local king, who briefly relaxed his sumptuary strictures on their behalf. As I have tried to establish, the main images in these four caves were all rushed to completion and dedication early in that gloomy year, while nine of the cells in Cave 11 and two in Cave 15 had been fitted out by then also. (It is of course possible that the cells in Cave Lower 6 and 7 had also been occupied, but their doors had not been hung). However, even in these first four caves dedicated at the site we would be hard put to locate any of their intrusions earlier than this same final Period of Disruption. Cave Lower 6, unlike the other three caves in this somewhat privileged group, had been almost fully painted by the original patron before (or at the same time as) being dedicated. This left little room for intrusions in any case, but the rear wall of the shrine, and a small wall area at the left rear, could have been utilized anytime after 469. Even though these spaces behind the image were hardly very desirable in any case, the fact that intrusions were never put there lends at least a modicum of support to the assumption that no intrusive activity was allowed (or desired?) during the period of the site’s development.4 The situation in Cave 7 is more telling, for it is evident that when the image was hastily finished (and thus presumably dedicated) at the start of the Recession (early 469) there was plenty of space both in the shrine and in the shrine antechamber where intrusions could have been carved or painted. That this was not done—that the area was not taken over during the long years of the Recession and Hiatus—is evident from the fact that after 475 it was totally filled 2 See Mirashi 1955, 19–21, for Subandhu’s donations at Bagh. However, he dates Subandhu’s Barwani inscription to the Early Kalacuri era, rather than the (correct) Gupta era. 3 For observations about Cave 8, where the (presumed) loose image is missing, see Volume I, Chapter 11, Cave 8. 4 It is perhaps of more interest to note that even in shrines where there are intrusions, they are always placed on the front wall, or at the more forward parts of the side walls.
164
volume iib, part i, chapter eight
up with a myriad images authorized by the cave’s patron as part of his compulsively meritorious program.5 Obviously he would not have been able to fill these areas so completely if any intrusions had been earlier placed there, for Buddha images are never thus cut away or covered over. In fact, there is only one extant intrusion in Cave 7—a large and damaged representation of bhadrasana Buddhas painted on the left porch wall, an area which the cave’s patron never got decorated as part of his total program. We know that it could not have been painted until the patron had all too hastily plastered the porch area early in 478 and had already painted the narrative Birth of the Buddha scene on the roughly prepared right wall. Thus this iconic scene can confidently be assigned to the Period of Disruption, sometime after mid-478. Significantly, it appears to have been inscribed. Cohen, regarding the record (Inscr. 21) states: “probably a donative record, it is oxidized beyond the point of intelligibility”. Similarly, even though Cave 15’s shrine Buddha was probably originally dedicated in 469, there is no evidence that intrusions were added to the cave until the Period of Disruption.6 Had intrusive donors wanted to (or been allowed to) utilize the cave for their own votive purposes, they could have added images almost anywhere to the still-unplastered and unpainted walls. The fact that it was later consistently plastered, probably in 477 and (in the case of the shrine and shrine antechamber) 478, makes it clear that no such early intrusions were “in the way” when this work was done. In terms of their iconography, the two intrusive panels on the left rear wall clearly belong to the Period of Disruption. Of these four caves, the case of Cave 11 is the least clear, for the simple reason that both the porch and the interior hall had already been plastered, as part of the ongoing original program, by the time the cave was temporarily abandoned in 469. The beautiful ceiling of the porch and the splendid but now ruinous bodhisattvas flanking the porch doorway must have been painted in 468, while the first phase of work at the site was still flourishing, but otherwise the plaster was still bare. A few of the more “generic” intrusions that appear throughout Cave 11 could indeed have been placed there anytime after 469. 5 6
For the Cave 7 patron’s compulsive program, see Volume I, Chapter 11, Cave 7. See Volume I, Chapter 11, Cave 15.
could any intrusions date before mid-478?
165
That they were not—that they belong to the Period of Disruption— must depend in part on the observations made in our analysis of Cave 11.7 But even without that evidence, it would certainly be unlikely that this little cave would be the only one at the site which had any intrusions which could be dated prior to the Period of Disruption. Even the fact that from about 468 (when the old A mode doorways were fitted out in the A+ mode) monks had occupied all but one of the cells, and were presumably beholden to the original patron, would seem to militate against the “takeover” of the cave by intruders.8 Furthermore many of the intrusive paintings in Cave 11’s hall are placed over pieces of fabric which almost certainly were added a decade or so after the walls were first plastered, in order to repair or stabilize these still unpainted old plastered surfaces.9 This would of course demand that the intrusions covering them (and often overlapping the edges of the applied pieces of cloth) be seen as very late additions; and the notably late iconographic character of many of them—particularly the elaborate bhadrasana Buddha paintings— also supports a late dating. It might also be noted that the inscribed Avalokitesvara Litany at the left end of the porch was also a particularly popular iconic form during the Period of Disruption, when travel away from the site must have been as dangerous as it was desired. It is clear that the anomalous votive stupa was added to Varahadeva’s Ghatotkacha vihara in early 469, when the excavation of the cave had to be abandoned for the first time. In fact, this could be only reason that Varahadeva cut it in the still only half-completed cave at this early date. There are many intrusions around this anomalous relief stupa, but it is clear that none of them pre-date the cave’s second abandonment in mid-478. This is because the area would not have been fully excavated until the time when the cell was added; and the character of the cell doorway locates that work well after 475—probably to 477 or perhaps late 476, for that seems to be the surprisingly late date when Varahadeva renewed work on his even more important Cave 16. It is unlikely that he would have resumed work on the Ghatotkacha vihara before that same time.
7 8 9
See Volume I, Chapter 11, Cave 11. For the fitting out of the doorways, see Volume I, Chapter 3. For this anomalous use of cloth, see Spink 1968A.
166
volume iib, part i, chapter eight
For the above reasons, it seems clear that the intrusions could not be earlier than mid-478; that is, they must be assigned to the Period of Disruption. The same late date must be assigned to the various intrusions in the shrine antechamber, which had not even been fully exposed before 475, while the image (which “copies” Cave 1’s) and the somewhat unfinished shrine doorway, must have both been still underway in early 478, again dating any intrusions to the subsequent Period of Disruption.10
10 The Cave 1 image was not completed before 476, while the Ghatotkacha shrine doorway is modeled upon a type not developed at Ajanta before 477.
CHAPTER NINE
CAVES 9 AND 10: THEIR REDECORATION AND THEIR INTRUSIONS
Preliminary Considerations One would not expect to find intrusive Buddha images in the old Hinayana viharas, Caves 12, 13, and 15A since, being without shrines, they were not ritually “alive” and thus not a proper or desirable location for votive offerings; and indeed, none are found in these caves. It is of interest to ask why when, in about 465, all of the Vakataka patrons at the site were urgently adding shrines to what were previously conceived as mere dormitories, someone did not add shrines to these old caves too; it would have been an easy task to do this by incorporating one or more rear center cells, as was done in Caves 11, 16, 27, and perhaps 26 Left Wing. Why this transformation was resisted—as was so long the case with regard to the redecoration of most or all of the Hinayana caves—must have something to do with the importance of making something meritorious of your own, as well as something up-to-date, rather than taking over someone else’s previous (and long out-of-date) donation. Is it possible that since the merit would already have been made by the original donor(s), there was little “left over” for further acquirement? But such questions must be left to Buddhologists and textual experts. As noted later, no one appears to have had the slightest interest in redecorating the ancient caitya halls, Caves 9 and 10, until very late in the site’s Vakataka phase—indeed, probably not until 477. Had the “intruders” in Caves 9 and 10 been allowed into the caves prior to this time, when the more carefully programmed (even if never completed) work had been accomplished, they would surely have put their images on the most visible pillar faces, rather than having to be satisfied with the less desirable faces toward or at the back of the pillars, where such intrusions now appear. Such an argument, based on the powerful criterion of the relative priorities of positioning, seems conclusive: the intrusions in Cave 10, being relegated to the lower-priority positions by the previously painted images
168
chapter iib, part i, chapter nine
belonging to the program of consistent redecoration carried out in 477 and early 478 (described below) all must post-date the middle of 478, by which time the “Vakataka” patrons had all departed. It is of interest too, to examine the two sculptured panels at the right of Cave 10’s façade. If the hypothesis is valid—that images reflecting the central Cave 19 standing Buddha type were disallowed while the Asmakas were in control of the site—this alone would confirm the post 478 date of these intrusions, as would the late character of their pilastered frames.1 As for the carved images on the façade of Cave 9, the very fact that so many never got finished, confirms their very late dating to the end of 480, and it is reasonable to suppose that even the completed carvings would belong to the same Period of Disruption, an assumption supported by the very late character of the many intrusive paintings inside the cave. If we add all of the above evidence together, we can conclude that there were absolutely no “uninvited” images—no intrusions— added anywhere at the site until the middle of 478 at the earliest, when the first few may have been carved and/or painted on or in the suddenly abandoned “Vakataka” caves along the main scarp. Then, thereafter, from 479 through 480, intrusions appear throughout the site as a whole, in every cave in which the main image had been dedicated. Such a conclusion is clearly supported by an analysis of various other caves, such as Cave 2 and Cave 16.2 In Cave 2, there are many intrusions, but they are invariably located in areas which had not yet been painted (but had already been plastered) when work broke off due to Harisena’s death. Thus, on the left wall, the Birth of the Buddha scene (part of the original program for the cave) ended abruptly at the end of 477 when the original patron lost control of the cave or could pay the workers no longer. But at this point the painters had not reached the already-plastered stretch of wall beyond, which was then (in 479) filled with intrusions. However, the intrusions were certainly not placed there before, as their position alone would suggest; they merely take advantage of the fact that this wall had been readied for painting before the original work program broke off, and was therefore a convenient and time-saving location. 1
For the “proscription” against Upendragupta’s central image type during the period of Asmaka control of the site, see Volume I, Chapter 15. 2 See Volume I, Chapter 11, Cave 2 and Cave 16.
caves 9 and 10
169
The fact that it was close to the rendition of an important event in Buddha’s life was surely not of any particular interest to the new and “uninvited” donor; the important thing was that the main Buddha image had been dedicated and that therefore the cave was “alive”. The same thing happened in the shrine antechamber, where the ceiling had already been painted, and the walls below had already been plastered, when such lesser-priority work had to be abandoned due to Harisena’s death. However, these already conveniently plastered and still undecorated-surfaces were used for intrusions sometime after mid-478. But for our present purposes the important thing to note is that the areas taken over for intrusions were those wall spaces which had been prepared but never utilized as part of the original patron’s program; and since this original painting program can clearly be assigned to 477, with “essential” compositions (the “Vajrapani” etc.) hurriedly added in early 478, it is evident that the intrusions must be still later; there is no way that they could have been added at any earlier date. For the sake of completeness, we should consider the dating of the small number of intrusions in the Vakataka caves (1, 3, 4A) at Aurangabad. There are many on or around the very unfinished Cave 1, but they are clearly icons of the sixth century, when the old prohibition against putting images in “dead” caves was no longer in force. In the Vakataka phase, Cave 1 had obviously been left untouched, once it was abandoned. As for the nearby cave 3, like Cave 1 it was not even started until after 475; and since it was fully completed and apparently fully painted too, it is hardly surprising that it remained untouched even in the Period of Disruption; there was no space left. Only the little Cave 4A, the iconographic character of which almost certainly proves that it dates to 478, has a few intrusions; they are little Buddha images carved wherever space allowed.3 5. Excavations: Dead of Alive? A particularly telling reason for concluding that the many intrusions at Ajanta all belong to the Period of Disruption, involves their choice
3
See Volume I, Chapter 14, Aurangabad Cave 4A.
170
chapter iib, part i, chapter nine
of the excavations which they inhabit or, we might better say, invade. First of all, although it may seem strange and surprising, they appear only in caves where the shrine image had been dedicated; they are never found in any others, no matter how seemingly attractive the location nor how available the space.4 Nor, as pointed out earlier, do they ever appear in the three Hinayana viharas, which had no shrine. In fact, this appears to have been the most important concern of the new donors in the Period of Disruption; only such Buddha images (and the caves where they dwelt) that were “alive” and thus efficacious in fulfilling the donor’s anxious quest for merit, were acceptable. It is for this reason that the new donors, over and over, crowded their images, if necessary, into cramped spaces in caves where the Buddha image was in worship, while they left untouched the spacious wall surfaces of excavations where, for one reason or another, the image was never brought to sufficient completion to be put into ritual use. Obviously the original patrons shared the same concern that the image (and the cave it was in) be “alive” in order to be operative. It was this intense concern that drove them, whenever the site was in crisis, to rush their shrine images to completion. This happened at the beginning of the Recession when the donors of Caves Lower 6, 7, 11, and 15 got special dispensation to finish their shrine images, even if they had to abandon all further work on the caves themselves.5 Except for Cave Lower 6, where the image was already well underway by the end of 468, all of these caves were left in an extremely incomplete state at this time, for their patrons clearly put all their funds and efforts solely into getting the images completed— or at least completed enough to be dedicated, even if the shrines themselves were still in a very rough condition.6 So we can probably assume that all of these four caves were “alive”, their images having been dedicated in 469. However, any intrusions in them cannot be dated prior to the Period of Disruption, for the reasons given above.
4
For Cave 22 as an anomalous exception, see Volume I, Chapter 12, Cave 22 Cave 16, the Prime Minister’s cave, also had this special dispensation, but for different reasons the image was not completed at this time. See Volume I, Chapter 11, Cave 16. 6 All three images were reworked (and rededicated?) a decade later. 5
caves 9 and 10
171
In striking contrast to the favored treatment of some of the “Vakataka” patrons, the Asmaka patrons, who had been developing the site’s western extremity right up to the moment of crisis at the end of 468, were ejected from the site at that point without being given any chance whatsoever to complete and dedicate their images, even though the stupa in Cave 26 and an anomalous shrine-projection in Cave 26 Right Wing were already sufficiently prepared to contain the intended images. This peremptory treatment clearly suggests the threat which the Asmakas posed to the rule of the local king and to the region over which he held sway. Even the privilege of completing the main Buddhas in a few of their caves was completely disallowed. At this point, at the end of 468, the Asmakas ongoing, but very incomplete, excavations were totally abandoned, before even a single image or decorative motif had been started. It must be the fact that from a ritual point of view these caves which filled up the western extremity of the site had never been brought to life which accounts for their being left completely alone until the Asmakas, returning in force, had taken over the region by 475. At this point the Asmakas of course reaffirmed their proprietary rights to their caves, and continued their normal course of work; intrusions never are to be found until 479 and 480, by which time the Asmakas were exercising no controls whatsoever over their abandoned caves. That potential private donors, such as monks at the site, did not put at least a few images in these (temporarily) abandoned caves when the Asmakas were not only gone but in total disfavor, between early 469 and late 474, reinforces the importance of such caves having been dedicated, if they were to be suitable loci for intrusions. It is even possible that the whole western extremity may have been considered “out of bounds” too, after the trouble-making Asmakas were ejected from the site, just because those patrons were in such disfavor. The rush to get shrine images at the site dedicated happened again at the end of 471, in the case of Caves 17, 19, and 20, when the local king was anticipating (correctly) his imminent defeat by the invading forces of the returning and avenging Asmakas.7 So the tables
7 Cave 19’s image had probably been completed a year or so earlier, but may not have been put into worship until just before Upendragupta had to flee, at which
172
chapter iib, part i, chapter nine
were turned once more, even though Upendragupta (the local king) managed better than the Asmakas had, to bring his images to life; for whereas the Asmaka patrons woke up one morning (we might say) and found that they had to leave within a few days—as the suddenly abandoned state of their unfinished main Buddhas suggests—the local king had some advance warning that his kingdom and his caves were in jeopardy. So his main shrine images were all (even if hastily) completed before his painful departure; and therefore his caves, being alive, might have attracted intrusions for, except for Cave 17, they remained unused during the site’s final heyday.8 The fact that potential private donors clearly avoided them until the Period of Disruption must have been because of the opprobrium connected with them, as the offerings of the defeated local king; for after the Asmakas were caught up by the needs of war and were no longer exercising any control over the site, during the Period of Disruption, all of Upendragupta’s donations became widely used for intrusions.9 Finally, in one final dramatic surge of anxious effort in 478, when the established patronage at the site was rapidly collapsing due to the threat and the needs of war, at least eleven and possibly twelve of the site’s original patrons rushed their images to an expedient completion. This was the patrons’ final act of devotion at the site, and it was surely self-seeking, even though they typically offer the benefits of their pious gifts to their mother and father and sometimes to “all sentient beings” or the whole world, as well. These Buddha images, all completed in 478, although generally started a year or more earlier, are: Caves 2, 4, Upper 6, 7, 8?, 11, 15, 16, 21, 26, 26 Left Wing, 26 Right Wing.10 They, along with the following, are discussed at length elsewhere. To this remarkable
point work in the courtyard was still going on, and the dedicatory inscription had not been installed. See Volume I. 8 See Volume I for Cave 29. Cave 18 does not come into the discussion, being only a cistern chamber. 9 Possibly this intrusive activity did not start in the old offerings of the local king until his enemies, the Asmakas, had given up their control over the site early in 479, even though other “Vakataka” caves may have attracted intrusions as early as mid-478. 10 For Cave 8 (loose image now missing) see Volume I, Chapter 11, Cave 8; the image in Cave 26 might have been fully carved in 477, but so much work was still going on in the cave in 478 that it is unlikely that it was yet fully painted and dedicated.
caves 9 and 10
173
grouping—the result of a sudden and anxious surge of patronage— can be added the Buddha images in the Vakataka excavations at other related sites.11 These include the main images in Aurangabad Caves 3 and 4A and in the Ghatotkacha vihara—and possibly also the Jain images in Caves 2 and 3 at Dharasiva. It is conceivable that one or two of the latest images at Bagh would have been completed at this same time, but since Anupa appears to have been quite immune to the Asmakas’ depredations, there was no rush there. As for the relatively minor and quite unfinished cave at Banoti, work on it had been given up as soon as Harisena died, just as was the case with most of the very unfinished caves at Ajanta. It might be added—I think it is fair to say—that this determination of the dating of all of these disparate images could never be done on the basis of “style”. This disparity of “styles”, which do not seem to revealingly develop during the site’s decade and a half of vigorous patronage, suggests that no one was telling the artists, who came from a great variety of places, following their local traditions— how to carve or paint; but they surely were being told what to carve or paint, for image types or features evolve, unlike style, with a remarkable consistency. Any art historian who, on such grounds, might suggest that all of these shrine Buddhas must have been completed in a single year, would have to be abnormally perceptive. Iconography, which develops almost by “clockwork”, is a surer guide, but even that could not provide a sufficient reason to assign all of these seemingly (in fact, actually) disparate Buddha images to a single year. The determination ultimately depends on context—on evidence connected with the overall development of each cave as well as evidence deriving from the varying states of incompletion. This must of course be combined with the crucial criteria provided by the historical situation of each cave, and of the whole site, as revealed through epigraphic and literary sources.
11
See Volume I, Chapter 14 (Related Caves of the Vakatakas and their Feudatories).
CHAPTER TEN
CRISES AND CAVE 11
The history of Ajanta’s Vakataka phase is defined by a number of major crises that shape its exuberant, but at the same time sporadic, development. The first crisis, which initiated the Recession early in 469, began with the expulsion of the Asmakas by the local king. The second crisis occurred late in 471 when the Asmakas, returning in force, but still as Vakataka feudatories, took over the control of the Ajanta region. Then, at the end of 477, the third and most crucial of all occurred: the great Vakataka emperor Harisena, Ajanta’s ultimate sponsor, suddenly died. The repercussions of this tragedy were immediately felt, not only at Ajanta itself, but throughout the empire. By mid-478, the feudatory Asmakas decided to avail themselves of the weakness of Harisena’s new successor, who was “totally averse to the science of politics” (Kale 1966, 360), and whom they themselves had managed to subvert through their pernicious influence at the imperial court. These aggressive feudatories therefore renounced their allegiance to the new and untried Vakataka overlord, and assumed independent control of the Ajanta region. This fourth crisis—this bold act of insurrection, this insulting theft of power—was in effect a declaration of war. But before this blow actually fell, the established Vakataka patrons, realizing that time was running out, rushed to complete their shrine Buddhas, and (most important) to get them dedicated, during the first months of 478. Driven by fear of the future, they had already fled from the troubled region by the time that the Asmaka takeover was actually effected. The Asmaka patrons, now enjoying the privilege of power, were able to keep on working at the site for some months after their Vakataka counterparts had fled.2 They continued, although not without some difficulty in these changing times, to develop their caves 1 This chapter expands upon some of the points already discussed in Volume I, Chapter 3. The peculiar development of Cave 1’s shrine is discussed in Volume I, Chapter 9. 2 The patrons along Ajanta’s main scarp were apparently all connected with the
crises and cave i
175
throughout 478. However, by the end of that year they too had to rush to get their main images dedicated, for it is evident that by the end of the year great pressures were put upon them, as both funds and perhaps workers as well were requisitioned for the needs of war. This was Ajanta’s fifth crisis, when Asmaka patronage was strangled by the rising requirements of war: at the end of 478, all established patronage at the site ended finally and forever. Thus, from mid-478 in the “Vakataka” area and from the end of 478 in the Asmaka area of the site, the Period of Disruption begins. At this point old programs and the well-laid plans were totally abandoned; now the site starts to be filled with a helter-skelter array of votive intrusions given by previously excluded donors now eager to make merit while it was still possible. But this surge of new and scattered patronage, even though responsible for hundreds, even thousands, of images, lasted only until the end of 480.3 Then it abruptly ended, leaving a host of unfinished images throughout the site, their development suddenly cut off (so it would seem) by the impact of the war against the empire upon the region. This was the site’s sixth major (and final) crisis, dating to the end of 480, after which, with the destruction of the empire and a shift in the control of the Ajanta region, Ajanta’s silent fate was sealed.4 It is remarkable that, in every Ajanta vihara, and in the contemporaneous times caves at Aurangabad and in the Ghatotkacha vihara too, the shrine Buddhas in the caves of the original patrons were all completed and dedicated only in one or another of these four of crises: early 469, late 471, mid 478, and the end of 478. Indeed, three of the shrine Buddhas (7, 11, and 15) were first dedicated in early 469 and then refurbished (and rededicated?) by mid-478. It seems likely too, that the main images in the two new caitya halls
Vakataka power, now in jeopardy from the insurrectionist Asmakas, who sponsored the caves at the site’s western extremity. By mid-478, the Asmakas, at least in their own eyes, were independent; so I make this dual (and convenient) distinction. 3 The multiple-Buddha compositions still visible in Cave 2, Lower 6, 7, 9, 10, 11 and elsewhere amount to a few thousand in themselves, while many others which once covered the front interior wall of Cave 26 and the rear porch wall of Cave Upper 6 are almost totally lost. 4 One might include still another “crisis”. In 465/466, perhaps by order but more likely via consensus, the patron of every Vakataka vihara at the site had to suddenly and drastically revise the vihara plans to include a shrine. Every vihara was indeed affected, even those where the rear of the cave had already been defined without such a shrine.
176
volume iib, part i, chapter ten
were also dedicated in these crucial years. Caitya Cave 19’s dedication can surely be assigned to late 471, since the cave was still underway until that time, and its inscription never even got put in place. Similarly, Cave 26’s hall was still vigorously underway in 478, and its famous inscription, probably contemporaneous with the cave’s dedication, clearly belongs to the midpoint of that same year, since the disruptive action which it records and represents precipitated the flight of all “Vakataka” patrons from the site; and it is evident that their departure must have taken place some months prior to that of the Asmaka patrons themselves, who were able to continue working on their caves—briefly—until the gradual erosion of necessary support, perhaps mingled with a rising apprehension, required their departure, “not with a bang but a whimper”. It should not be surprising that work continued on such major caves even after they were inscribed; in fact not a single one of the four with extant inscriptions was finished, and it is likely that none had yet been dedicated when their respective inscriptions were applied, even though the presentation of the caves is written into the record. At least some of the greatest images in Cave 26—the Dying Buddha, the Temptation scene, the beautiful Sravasti Miracle triad, the (tellingly) unfinished Sravasti Miracle panel at the rear, and quite possibly the main Buddha itself, were not completed until well after Buddhabhadra’s proud but ominous inscription was applied. The great main Buddha in Varahadeva’s Cave 16 was surely not completed and rushed into dedication until early 478, although it is evident that the dedicatory record had already been written and applied the year before, since it speaks with pride both of Harisena himself and the vast territories which he controlled. Mathura, the patron of the vast and unfinished Cave 4, must have ordered his short and hasty inscription cut on the base of the main image just prior to it being thinly plastered and painted in the rushed context of early 478; it is clear that, surprised by Harisena’s death, he did not have time to have a proper prasasti written and sent down to the site. The inscription composed for the dedication of the fine Vihara 17 of Upendragupta, the local king, was applied with uncharacteristic haste in 471—the adjacent porch pilaster being all too roughly cut away to make room for the record under the protective eave.5
5
Since the record did not quite “fit”, perhaps the original plan was to locate it
crises and cave i
177
The shrine was being worked on (and never fully completed) with a similar haste at this same time, suggesting that the image would not have been dedicated until an even later moment (but also in 471) in this desperate final year of Upendragupta’s control over the region. His problems with his beautiful Caitya Hall (Cave 19) were even greater, and in this particular case the image was indeed dedicated, again in 471, before the record was incised. In fact, it was never put in place, although an extremely large recessed panel was made for it over the doorway, just inside the cave. It was surely intended to be the longest and the most lush encomium at the site. But one can almost see the Asmaka forces “coming over the hill” while the Brahmins were still laboring over it, too late, up in the capital; or perhaps it was delayed by the frantic traffic on the roads— blocked by the developing Asmaka takeover of the region. This remarkable “coincidence” of all the Buddha images at the site being rushed into dedication at one or another of the times of crisis, can probably be ascribed to the impact of these different troubling events, which urged—indeed compelled—donors to rush their images, at each moment of crisis, to completion and dedication. Had the times been “normal”, all or at least most of these images would surely have been finished in a more leisurely and often more disciplined fashion. Our present concern, however, is with the latest group of dedications, all completed (and generally rushed to completion) in 478: Caves 2, 4, Upper 6, 7, 8?, 11, 15, 16, and Ghatotkacha vihara, all dedicated by mid-478, and Caves 21, 26, 26 Right Wing, 26 Left Wing, and Aurangabad Caves 3 and 4A, all dedicated by the end of 478. For in explaining how and why the continuation of work on these still-unfinished excavations was suddenly abandoned, as soon as their shrine Buddhas had been rushed to completion in 478, we shall be able to understand why, in the very next year, the site became a votive “free-for-all” until its expiration at the end of 480. If we can understand the forces which, within a matter of months, transformed Ajanta’s proud achievement into a time of anxiety, and Ajanta’s heyday into the Period of Disruption, we can go far in
on the rear wall of the porch, like the dedicatory inscriptions of Cave 26 and the Ghatotkacha vihara; it appears that Cave 16’s inscription was also originally planned for the recessed area near the right aisle doorway rather than for its present (more effective) position at the left of the court.
178
volume iib, part i, chapter ten
reconstructing not only Ajanta’s proper history, but the history of much of India in the mid and late fifth century. Thus Ajanta, perhaps more specifically than any other site in India, can reveal how through close analysis of its development, we can literally work out and explain—pick apart and put together—its evolution almost year by year. As noted earlier, at Ajanta, perhaps better than at any other site, we can elicit “history from art history”.6 In this reconstruction we can use, and at the same time confirm, the much-questioned reliability of the well-known account of the Vakataka debacle in Dandin’s Visrutacarita, analyzed in detail elsewhere. The “archaeological” evidence at Ajanta and related excavations, the historical evidence in the Visrutacarita, and significant evidence from various epigraphic sources all point to one conclusion: that the collapse of consistent patronage at the site in c. 478, after a mere seventeen years or so of vigorous (even if sporadic) activity from c. 462 through c. 478, was indeed due to the death of the emperor Harisena on “December 31, 477”.7 On a local level, this is the crisis that opened the floodgate for intrusions at the site. But from a larger perspective, this is the crisis, we can finally say, upon which the drama of India in the late fifth century essentially turns. And, if I am correct, the confirmation of Harisena’s unexpected death is to be found in an analysis of Cave 1, the most splendid vihara in India. Even without the evidence of Cave 1 the story could be written, but in far less compelling detail; for the plot, and the key to the tragedy, turns on Harisena’s sudden death, and it is the evidence of that which Cave 1 convincingly reveals. I shall review this here, discussing concerns about the general development of the image concept, in which Cave 1 played a crucial part, in a later chapter. I have asserted elsewhere that Cave 1 was the emperor Harisena’s donation, and that he suddenly died before it was dedicated, and before its inscription had been brought down to the site. Of course, describing the “imperial” character of Cave 1 does not prove that
6 This was the title of a paper given at the 26th International Congress of Orientalists (1963), in which I first presented my original (later adjusted) arguments for a “Short Chronology”. See Spink 1964 (= date of publication of ICO Summaries). 7 In Volume I, Chapter 2 I suggest that fourteen years might be more reasonable than seventeen, from an actuarial point of view.
crises and cave i
179
it is the emperor’s donation, even though it may urge the point. It is the evidence bearing upon how and why it was left unfinished, how and why its unfinished situation is unique, and its total lack of intrusions, that provides the ultimate “proof ” of its imperial patronage. But before turning to that I shall describe certain “internal” features that also suggest this. Cave 1 is easily the most impressive excavated vihara in India.8 Remarkably, it is the only vihara with an elaborately sculptured façade, unusual in its incorporation of a winged courtyard, and unusual too in being provided with a splendid (now lost) portico.9 Its complex and beautifully organized friezes depict the particular prerogatives of kingship—war, the hunt, and amorous dalliance— while friezes of “royal” lions’ heads border the palatial structures. Even the narratives of Prince Siddhartha’s search for spiritual insight, on the façade responds, show the ideal vocation of the ideal man, to which all serious seekers must pay heed even in the midst of the responsibilities of rule. Inside the cave, the same dignity, complexity, and order reigns, suggesting authoritative planning from on high, both in the disposition of the architectural elements, and the organization of the mural decorations.10 Indeed, this is the only vihara with a clearly focused program of narrative decoration.11 The same emphasis on a “vicarious” spiritual evolution, involving the same implied royal protagonist as evidenced on the façade is to be seen in the famous murals on the walls of the hall. Except for the expected and probably
8 The closest rivals would be the great viharas at Bagh, although the nature of the rock there disallowed much fine sculptural decoration. The Bagh caves were also done during Harisena’s time, and presumably with his support; it is surely relevant to note that one of Harisena’s sons, described in the Visrutacarita as Sarvasena’s queen’s “husband’s brother by another mother” (Kale 1966, 361) was viceroy of the region at the time. 9 The original portico, although badly damaged, can be seen in a nineteenth photograph in ASWI 4. In the early 20th century the portico, having fallen, was disposed of in the river below. For the old photograph also see ACSAA Ajanta fiche 2:19. 10 There is one minor “mistake”, probably caused by a misunderstanding between workers and planners. The absolute design symmetry of the four corner pillars is broken by the diagonal flutings of the one at the right front not corresponding to the general pattern. This minor digression from absolute order is hardly noticeable, but it still must have been a source of embarrassment to the chief planner. 11 Leela Wood has shown convincingly that Cave 17’s paintings reveal a programmatic organization which is both subtle and complex; see Wood 2004.
180
volume iib, part i, chapter ten
conventional iconic scenes in the shrine area—scenes which may have been replicated in great numbers of now-lost structural temples of the time—every mural is a tale specifically focused on the virtues of good kingship. In every one, the future Buddha is a royal figure, surely reflecting the cave’s royal patronage, as well as an order and discipline reflective of (or at least suggestive of ) imperial authority. One should also note the particular emphasis and honor accorded to nagas, who have particular places of honor throughout this cave. Not only do nagas worshipfully encircle the stupas on the sumptuous pillar capitals of the important rear center pillars, but large carved nagas, both protective and welcoming, guard the base of the shrine doorway. Significantly, in a significant percentage of the Jataka tales painted on the walls of the hall the future Buddha, or chief protagonist, is a serpent king, a nagaraja, while in a few other tales nagas play an important role.12 Such an emphasis must have reminded the visitors of that day that the Vakataka house traced its origins back to the ancient naga dynasty; and of course, as in many other caves at the site, it recognized the presence and the power of these lords of the ravine. The germinal influence of Cave 1 in being the apparent conduit for stylistic, iconographic, and technological innovations coming from the contemporary Vakataka caves at Bagh, and as the most authoritative source for the establishment of certain defining features at the site, also confirms the central importance of its patron. This is also suggested by the authoritative manner in which the cave’s contracts were carried out. This is revealed throughout the cave: in the coordinated painting of the remarkable ceiling by a number of coordinated “ateliers” using slightly different palettes; in the swift but surely controlled work on the extensive murals; in the carving of the elaborate sculptural program; even in the careful and consistent refitting of all of the cell doorways in 475 (immediately after the cessation of the local war) and the addition of a rock-cut niche at the same time, in order to fit the cells out in a newly popular and practical way, drawing upon the influence of Bagh. The priority position of Cave 1’s patron is even more evident when we realize that Cave 1 is the only cave at the site which con-
12 See Schlingloff 1999: Sankhapala, Nagakumara and Champaka Jatakas (Cave 1). In the Sudhana Jataka a naga king also plays an important role.
crises and cave i
181
tinued to flourish both when the local king stopped work on the Asmaka caves, and then when the Asmakas stopped work on the local king’s caves. (See Time Chart) That is, the patron of Cave 1 was essentially unaffected by—“stood above”, as Bakker might say— this battling of his feudatories, which at one time or another (depending on allegiance) cut off work in every other cave at the site. When one analyzes this evidence—sees the site divided along partisan lines, but always with the exception of Cave 1—it is hard to think that Cave 1’s patron could be anyone but the emperor Harisena, the Vakataka overlord. This bifurcated pattern of development at Ajanta distinctly proclaims this. Even the evidence that Cave 1 was the last cave to get started during the site’s first decade of activity, as its “last available” position in the scarp and its relatively developed character suggests, can be argued as supporting the view that it is the emperor’s donation. Why did the great emperor, who (with or through his courtiers) clearly supported Ajanta’s renaissance from the start, not actually establish an excavation at the site until four or five years had passed, and all of the best locations had already been taken? The answer is probably that as the fame of the developing site spread, it became increasingly incumbent upon him to support it in a directly personal way. As Bakker says: “By all we know of Harisena he was a Hindu; nevertheless, the material remains of his reign that we possess are almost all directly linked to the Buddhist faith”.13 The great emperor was, quite obviously, ecumenical in his attitudes. It is relevant (and perhaps surprising) to note that not only Harisena’s Prime Minister, but two of the major Vakataka feudatories, who supported the developments at Ajanta, all declare themselves to be fervent Buddhists, despite the fact that they probably retained their Hindu connections too. This dual connection seems evident, in the case of the Prime Minister who, “being extremely devoted . . . to (the Buddha) the teacher of the world” (Ajanta Cave 16 inscr. Vs. 21) details his brahmana lineage in his Ghatotkacha Cave inscription. And in a sense, the emperor Harisena, if he is indeed the patron of the sumptuous Cave 1, must have been a “fervent Buddhist” too. 13 (Bakker 1997, 40) I have suggested that the remarkable Saivite dedications at Mansar were probably due to Harisena’s patronage, after he came into control of eastern Vidarbha late in his reign. However, this suggestion is supported more by intuition than by proper research.
182
volume iib, part i, chapter ten
Yet according to the Dasakumaracarita, Harisena “(maintained in order) the four classes according to the path laid down by Manu” (Kale 1966, 349) while some of the dramatic action in that same story takes place in a Durga temple, with a courtly audience in attendance (Kale 1966, 365–366). It would seem that in the refreshingly open climate of the Gupta/Vakataka period, we must believe that Harisena-and his courtiers too—could support both of these related forms of belief.14 We can well imagine that when his courtiers inaugurated Ajanta, Harisena was supportive but not yet deeply involved; but when some half-decade later, with everyone talking about the developing wonders of the site, it is reasonable to suppose that he also wanted “to get in on the act”, if for no other reason than to encourage a site which, lying on a great trade route, was bringing so much attention and so much excitement to his reign. It was surely clear to everyone involved that they were building a remarkable lasting monument not just to the Buddha, but to the present ruler, and that it was indeed expected, as Upendragupta, in his Cave 17 inscription (vs. 29), states: “(to) cause the attainment of well-being by good people as long as the sun dispels darkness by its rays!” As is so often the case with religious donations, Harisena’s involvement was probably as much for political as for spiritual reasons. Ajanta’s richness alone best suggests the power and popularity of Buddhism in these times, so it is understandable if Harisena was swayed in the Buddha’s direction. Having mentioned the many features and facts that suggest that Cave 1 is Harisena’s cave, we must now consider the implications of the fact that it is so surprisingly unfinished. This is revealed first of all by the half-finished state of the dark front wall which here, as in most other caves at the site, was almost always the last one to be decorated, when work (as here) was going on in normal course. The famous (really infamous) “Persian Embassy” scene (see Volume I, Chapter 8) on the right wall, now identified by Schlingloff (1996, Pl. 1.2) as the Mahasudarsana Jataka, is completed only at the upper levels, while the areas further to the right have nothing at all on them except a single preliminary red-line sketch, in characteristic red pigment, of a crowned head. By the same token, the rearward area
14
Narain 1983, 34–45.
crises and cave i
183
of the right wall has been plastered, but never painted, while the scenes on the area up as far as the right front pilaster appear to have been still in process when work was suddenly interrupted. Such evidence that the cave’s carefully and elaborately organized decorative program was cut off in mid-course makes it evident that something drastic, involving the patron, must have happened. This conclusion is confirmed when we realize that the soot from the oil lamps used for worship which, to a greater or lesser degree, has begrimed so many other caves, depending upon how long and actively they were used, is absolutely not in evidence here. Then when we see that the fragile plaster around the garland hook (now missing) at the center of the antechamber’s ceiling medallion is completely undamaged, we can conclude that no monk ever hung a garland there, to say nothing of changing it every day.15 This is because the act of doing so always rapidly damaged the plaster and paint around the hook. One need only look at the center of the various medallions in Cave 2 (not even put into use until 478, when the image was dedicated) to confirm such a statement. It is clear the Cave 1’s shrine, along with the shrine image, had been fully painted by the time work broke off in the cave. There is no evidence of the anxious rush that we see in Caves 2, 4, 6U, 11, 15, and 16, to say nothing of the earlier-abandoned Caves 17 and 20, in all of which the image—violating normal procedures—was rushed to completion “at the last minute” before the surrounding shrine was completed.16 At the same time, what is curious is the fact that although the plastering and painting of Cave 1’s antechamber is in splendid condition, there are merely a few traces of the original surfacing left in Cave 1’s shrine itself—the most significant remains being in the better protected parts of the image itself, notably on the halo and the carved umbrella above it. There are also traces of an almost obliterated bodhisattva on the right front shrine wall, while a tiny trace of a painted halo and ushnisa on the right wall of the left “corridor” surely allows us to assume that this whole area was filled with small multiple Buddhas. 15
The hook, like many at the site, has either fallen out or (more probably) has been removed by local people. Unfortunately, the small hole for the hook has been cemented in. 16 Cave 2’s situation is slightly different, for the shrine was fully decorated; but the “rush” is evident from the fact that the antechamber walls, although plastered, were not painted as part of the original program.
184
volume iib, part i, chapter ten
These losses must be explained; and they were surely not due to damage caused in the course of worship, since there is not a trace of the soot from oil lamps which covers the walls and ceilings of such crucial areas in various other caves. Perhaps the best clue is to be seen in the surprising fact that the pivot holes in the shrine doorway, which must have been fitted in place when the decoration of the shrine was complete, show distinct evidence of significant wear caused by the opening and closing of the double doors. It seems reasonable to suggest that, since the cave, never having been dedicated, was ritually “dead”, but since monks continued to live in the cells (with their well-worn pivot holes), the shrine was taken over for use as a storage area, which could be well secured by the shrine’s huge doorway. A number of the caves at Ajanta (Caves 17, 6L, 11, and 25), along with the related caves at Bagh, have such facilities, which must have served a useful and necessary purpose at such a busy site.17 Even more telling as evidence that Cave 1 was ritually “dead”, is the fact that it is the only cave (of some fifteen) at the site with a completed shrine Buddha, which nevertheless has no later “intrusions”, either painted or carved.18 And this is true despite its grandeur, the impressiveness of its shrine, and the fact that there was plenty of available space on the rear and right walls. All of those other caves at Ajanta and related sites, in which the Buddha image had not only been completed but also dedicated, were “alive”; these caves, even if they had far less available space than Cave 1, were taken over by the “intruders” in the Period of Disruption. Conversely, all of the ten unfinished excavations at Ajanta and other sites, in which the shrine image was not carved at all, are totally bereft of intrusions. Thus we must conclude that Cave 1 was never in worship—never dedicated—even though we know that monks had moved into the 17 Bats could have added to the damage; a distinct and slightly depressed line at the top of the walls was probably created by bats (with their sharp claws) hanging upside down in this area. However, there is surprisingly little soiling on the walls below, as was more typically the case in other caves. 18 Caves 2, 4, 6 Upper, 6 Lower, 7, 11, 15, 16, 17, 19, 20, 21, 26, 26LW, 26RW. Cave 26LW shows no intrusions, probably because the available walls had been fully painted in 478; the walls now being so exposed, no traces of paintings are left. Cave 22 has a central image, but it is an intrusion, not in the intended shrine, but on the rear wall of the antechamber. Aurangabad Cave 1 was apparently fully painted, so has no room for intrusions. This may have been the case with Cave 8 too, considering the remaining surfacing.
crises and cave i
185
cells around 475, when the still unused B mode doorway fittings were significantly improved by conversion to the D mode, and the various shelves and clothes poles and niches were also added. But they certainly did not worship the image at this time, because it was not yet “brought to life”. This would seem to tally with the fact that no dedicatory inscription has ever been found, although the encomium to the emperor would surely have been extensive, and would have required a place of honor. This in itself is not surprising since many of the shrine Buddhas at the site were not dedicated until 478, and yet monks must have already been using them for residence. Interestingly, the B mode fitting in Cave 1’s Cell R3 shows definite pivot wear, ascribable to the period prior to its conversion to the D mode. Since B mode fittings went out of fashion after 469, the door must have been originally fitted by that time. However, since the excavation of the hall would have been very much in process at that date, it is hardly likely that this single cell was used as a residence. It is far more likely that it was provided with a door, which could be secured, for storage of the workers tools or other supplies while excavating work in the cave was underway, and that it was then converted to the D mode along with the cave’s other cells in 475.19 Incidentally, the situation here provides valuable evidence pertaining to the rapidity with which the pivot holes in such doorways were worn smooth by the turning of the doors. Since we can be sure that the original B mode pivot was cut (and presumably then fitted out) in either 468 or 469, and since it was replaced, and never used again, when the doorway was supplied with the more efficient D mode in 475, we can see that such pivot holes showed the effect of the door’s turning quite rapidly: the original B mode pivot hole is distinctly smoothed, only slightly less than is the case with the later D mode fitting, which would have been in use from 475 to sometime in the 480s, by which time the remaining monks at the site had moved away. We might also note that the rapidity with which such wear could take place—and of course this depended on the degree of the doors’ usage—is equally revealed by evidence of such wear in a number of cells at the site which would not have been ready for occupancy until 477, the last date at which cells 19 I have suggested that at some point in or after the Period of Disruption the shrine (whose image had never been brought to life) was converted to a storage chamber. See Volume V.
186
volume iib, part i, chapter ten
would have been finished in normal course. In such cases the amount of wear would seem to confirm the reasonable assumption that monks continued residing at the site for at least a few years after the original patrons had had to leave the site. Since there does not seem to have been any place reserved for the emperor’s inscription on the heavily decorated façade, and we have no precedents for placing such records on a cave’s approach, it seems likely that, like Buddhabhadra’s highly important record in Cave 26 or at the Prime Minister’s Ghatotkacha vihara, the plan would have been to locate it somewhere on the rear wall of the porch, the actual size of the panel being determined when the size of the inscription was known.20 At this point a recessed panel could have been prepared for it, as in the case of the local king’s Caitya Cave 19. But we can assume that this information, like the inscription itself, never got to the site. Unfortunately, any evidence that some area on the main porch wall had been reserved for a still-uncut panel area is now lost, because the walls are largely barren now; there are only the merest traces, all high up near the ceiling, to suggest that the conventionally large representations of the bodhisattvas Avalokitesvara and Vajrapani in a forest setting were once painted there. The losses in the porch are largely due to the buildup of debris over the centuries—the huge courtyard having served all too well to collect fallen material from the hill above.21 At the same time, since most of the upper levels (including the porch ceiling) suffered great losses as well, it seems likely that the whole porch area was affected by the alternating (seasonal) dampness and desiccation of the debris, aided by the depredations of birds, chipmunks, and the like, whose claws, over the
20 As in Caves 16, 17, and 26, the size of the panel would be determined by the length and shape of the inscription. The extremely large size of the (never inscribed) panel in Cave 19 suggests that the size needed was already known. It appears that a location to the left of the right aisle doorway of Cave 16 had been recessed earlier for the dedicatory inscription, but that its location was changed for more visual impact when one came up the tunneled stairway. This change of location may well have been suggested by the assertive location of Cave 17’s inscription, cut some years earlier; or perhaps the panel earlier planned turned out to be too small; the painting around it would already have been done by the time the inscription was incised in 477. See Volume I, Chapter 15. 21 It is said that the porch doorway was so filled up with debris that vandals could not get into the cave in early times to cause damage. Whether or not this story is true, the cave is remarkably free of graffiti.
crises and cave i
187
centuries, grain by grain, can do great damage.22 It is of interest to note that although Cave 1’s grand courtyard was appropriately “imperial”, its extensiveness was in fact necessitated by the lay of the land; because the scarp slopes back much more gradually here than (for example) in Caves 16 and 17, a very deep courtyard was required to achieve the necessary façade height. What seems to be the case, and tallies with other evidence bearing on the cave’s “decease”, is that the inscription had not yet been brought down from the capital for installation by the carver/scribes when work on the cave so suddenly ended. Perhaps its composition had not yet been completed and submitted to Harisena and his courtiers for approval. Ultimately, it may well have been left unfinished, as one more consequence of the deep shock—and resultant “shockwave”—caused by the sudden death of the emperor. We may reconstruct the situation as follows. Probably because of the status and the financial resources of its patron—whose superintendent of works could preempt workmen as he wished—Cave 1’s work program went on with a particular efficiency. The cave’s “priority” also helps to explain why Cave 1’s Buddha image is the only image being worked on after 475 that got finished prior to Harisena’s death; and everything else in the cave had also been completed by late 477, except for a few murals (see above) on the front and right walls. The speed and character of such developments remind us that in 477 times were good. Indeed, in those days it must have seemed that, under Harisena’s firm and happy rule, the good times would “go on forever”. Who would have thought that only a few months later, the site would be in a frenzy—in fact in the throes of death— with time so quickly running out. Thus, in 477, when Cave 1’s image itself was finally completed and the final stages of the cave’s decoration were progressing well, there would have been no particular urgency about arranging the dedication ceremonies, except to have the satisfaction of seeing everything brought to a meritorious completion. In 477, the “Ceremonies Department” in Harisena’s court was already hard at work, with 22 Jogesvari’s great east entrance (with its scene of Ravana Shaking Kailasa) looks like it has been extremely weathered, but is in a spot well-protected from the elements. The damage almost certainly can be ascribed to the ever-present pigeons, sparrows, and chipmunks, whose claws have gradually scratched away the already rather corruptible basalt, which may have been prepared for their ravages by the effects of the salt air from its proximity to the ocean.
188
volume iib, part i, chapter ten
neither fax nor phones to help them, making the complicated practical, political, and ritual arrangements for the visit of the emperor and his huge and hierarchical retinue to the site. To put their plan into action, they were merely waiting for the word that the painting of the cave was finally complete; that the Brahmin pundits had finished the composition of the emperor’s panegyric; that it had been sent down to be carefully inscribed on the porch wall or (less likely) at the top of the grand approach up from the river; and that the court astrologers had chosen the most auspicious day for Harisena and his court to attend the dedication functions at the site. However, before this could happen, Harisena very suddenly died; and because of that his great cave died too, never having been “brought to life” by virtue of the completion of the necessary ceremonials. I have suggested that Harisena’s death was quite possibly due to the treachery of the ambitious and aggressive Asmakas, who had almost certainly infiltrated his court even before the great emperor’s death and (if they had not already done so) would soon corrupt his weak son and successor, Sarvasena III. But how do we know that Harisena’s death was so sudden; as if it might have resulted from a stroke or heart attack or (perhaps more likely) as the result of an assassination plot at the hands of the “spies in various disguises” (Kale 1966, 356) that the Asmakas planted within the imperial court, even as they were building up their forces of war?23 The possibility that the emperor was assassinated gains some support from the curious reference to his death in the Dasakumaracarita: “He came to be numbered among the immortals, owing to the want of religious merit on the part of his subjects” (Kale 1966, 349). Ryder translates the same passage as: “yet for the unworthiness of his subjects (he) was translated into divinity” (Ryder 1927, 200).24 It seems 23 The presence of such “guests” at court was obviously conventional at this time. When Visruta was finally establishing himself in power in Mahismati, he enlisted the help of his newly appointed minister to select “truthful, incorruptible counselors” and, along with them, “spies diversely disguised” (Ryder 1927, 224). 24 I have suggested (Volume I, Chapter 4) that Dandin may have felt it inappropriate to speak of the emperor’s assassination, as representing a shortcoming on Harisena’s part, in not developing adequate security controls. Dandin may have had a similar reluctance to ascribe any fault to his presumably exemplary protagonist, Visruta, if in fact the latter might have been implicated in the suspicious death of Harisena’s grandson, who stood in the way of Visruta’s imperial ambitions). Such reluctance would be particularly understandable if, as seems likely, Dandin was describing a recent historical situation with a necessary or at least understandable discretion.
crises and cave i
189
clear from such a muted reference that his death was not from natural causes, but from some kind of subversive activity, in which the Asmakas, surely, proved their “unworthiness”. Ajanta itself would seem to confirm the conclusion that the emperor’s death was sudden: if he had a wasting disease—if he had had cancer or tuberculosis or the like—he would certainly have seen to it that his cave got dedicated, in order to acquire the merit from his donation. It is clear from the evidence at Ajanta that getting the main Buddha image finished and finally functional was the over-riding compulsion felt by every patron at the site. Had Harisena known he was going to die, he would certainly have insisted on the dedication ceremonies being performed, either in his presence, or (if necessary) in his absence.25 But he would not have “died intestate”; he would not have foregone this most essential result of his involvement in the site if he had had time to give the orders. That he apparently did not finish what he had begun clearly heightens the likelihood of his death being sudden and totally unexpected. We can assume, considering Harisena’s high position, that the officials in charge of excavating and decorating the cave would have wanted it fully completed before the dedication ceremony, and would have wanted the prasasti in place as well. The fact that the cave was left, frozen in time, from the date of his death just as it remains today, makes it clear that it was still being prepared for Harisena’s arrival when its development was so abruptly cut short. And if work on it was never continued by his son—a “duty” often neglected, in any case, in lines of royal descent—this can be ascribed to the very turmoil which followed directly upon the deadly blow which had been dealt to the Vakataka court and all associated with it. The fact that the “Vakataka” patrons renounced their overall work programs and focused compulsively now on finishing their shrine Buddhas suggests that an overriding concern about Asmaka aggression was already in the air as soon as Harisena died; and the “Vakataka” patrons’ fears were clearly confirmed only a few months later (by mid-478) with the insulting omission of even the slightest reference to the Vakataka emperor in Buddhabhadra’s Cave 26 25 I assume that the cave could have been dedicated on Harisena’s behalf, if time, interest, or the situation had allowed. I do not know the conventions for this but Bakker (1997, 41) suggests that it could have been done, even though he disagrees with my view of the circumstances of Harisena’s death.
190
volume iib, part i, chapter ten
inscription—which at the same time generously praises “the mighty king of Asmaka” and his minister. This repudiation of the authority of the new Vakataka emperor, Sarvasena III, placed for all to see on what was now the most important cave at the site was startling; but it was surely only a local reflection of the dire “declaration of independence” that must have been broadcast from the Asmaka capital. Such a manifesto must have both initiated and at the same time energized the developing insurrection which, according to the Dasakumaracarita, the feudatory rulers of Kuntala, Murala, Risika, Konkana and Nasikya soon joined, eager to divide up the empire. However, one need not think that the new “privileges” of the Asmaka patrons at Ajanta lasted very long in this troubled period. Due to the compressed burden of events in these anxious days, the Asmaka patrons too were forced to yield to military priorities by the end of that same troubled year (478), and to finish what they could in the limited time allowed. By the beginning of 479, following the earlier lead of the “Vakataka” patrons, the Asmakas too had sadly renounced their connection with the site forever. It is the character of this departure—first of the “Vakataka” patrons and finally of the Asmaka patrons—and the nature of what they did in those last days before they had to leave, that most surely confirms the assumption that it was the death of the great emperor which rapidly put an end to all of Ajanta’s established patronage. And if no one had “had the presence of mind to inscribe the prasasti on (Cave 1’s) wall” (Bakker 1997, 41) this is not difficult to understand, given the turbulence of the times. Even if anyone had any interest in doing so after Harisena’s death, Ajanta was not a safe place to be traveling to in the early months of 478, to say nothing of the months thereafter. So Harisena’s grand vihara, conceived and executed with such vitality, finally turned out to be dead. It would hardly change India’s history if a truly anonymous cave— even the most beautiful cave at Ajanta—was ritually “dead”.26 What changes history is the fact that the dead cave was the cave of the
26 This does not mean that it was not used for a monastic residence. As pointed out above, many of the caves at the site were occupied well before being finished— in fact, not a single cave at Ajanta was ever fully completed. Furthermore, Cave 1, like many others, continued to be occupied throughout the Period of Disruption, and quite possibly into the 480s.
crises and cave i
191
great emperor, Harisena, who by the time the cave was reaching completion, had extended his empire from the western to the eastern sea, and was surely the greatest king in India in the mid-fifth century. Significantly, what the dead cave tells us is that the emperor too was dead; and it was this death of the emperor that presaged the destruction of the great Vakataka dynasty early in the 480s and plunged the radiant world reflected at Ajanta into decades of darkness.
CHAPTER ELEVEN
THE BREAKDOWN OF PATRONAGE IN THE PERIOD OF DISRUPTION
The rejection of the Vakataka’s overlordship, the key to Ajanta’s sudden decline and eventual death, was publicly announced at the site in the monk Buddhabhadra’s Cave 26 inscription (verse 9, 12), where he handsomely honors the “steadfast, grateful, wise, and learned minister of the mighty king of Asmaka”, who had been his friend “through many births”, as well as the minister’s son (“an equally foremost personality”), who had succeeded him at his death. But crucially, and of course intentionally, he insultingly omits any reference whatsoever to Harisena’s weak and inexperienced successor, Sarvasena III, who, as Dandin kindly puts it (Ryder 1927, 200) “happened to be somewhat inattentive to administrative duty.” This ominous debasing of Vakataka imperial authority, caused in large part by the ineptitude of his successor, was surely anticipated from the moment of the great emperor Harisena’s mysterious death late in 477, and explains the great urgency with which the worried “Vakataka” patrons rushed their shrine images to an expedient completion and dedication early in 478. It seems likely that these established “Vakataka” patrons—surely with both deep regrets and deep forebodings—had all left the site by the time that Buddhabhadra’s sobering encomium to Asmaka power was inscribed on the porch of Cave 26 in about the middle of 478. This, or the political fiat to which it gave rise, must define the moment when the now abandoned caves along the main (“Vakataka”) stretch of Ajanta’s scarp could be invaded by the new intrusive donors eager to put their own offerings in the caves, in order to (finally!) make merit on their own. However, if these onceso-exclusive caves were opened up to the “intruders” as early as the middle of 478, this frantic donative activity only increased in intensity when the Asmaka caves were themselves finally abandoned by their powerful patrons some six months later, at the end of that same year. As it happens, the rapid departure of the Asmaka patrons—or at
the breakdown of patronage
193
least the interruption of their work on the caves—was even more unexpected than the flight of the “Vakataka” patrons some six months before. This is understandable; for the harsh order imposed upon the Asmakas came from their own court, while that responsible for the flight of the “Vakatakas” was imposed by their enemies. It is instructive to note that throughout the difficult year after Harisena’s death, work on the Asmaka caves, though showing some retrenchment from the start, was proceeding as if the patrons felt that their excavations, even if somewhat diminished in splendour, would get done and then would get dedicated in normal course.1 It is only that they were not anticipating the insistent call of war, or at least of the preparations for war which, in the end, was so drastic that it demanded an ending to all activity within a matter of weeks—or even days. The long-committed Asmaka patrons were obviously caught unprepared. The unexpectedly expedient ending of all Asmaka excavation programs (at Aurangabad too!) at the end of 478 now exposed all of the Asmaka excavations to a pious “invasion” by a horde of eager new donors—every man (or monk) for himself—starting in 479. And at the same time it left the site without any authoritative controls whatsoever. Eventually, this loss of the same royal patronage which had made the site grow with a vitality which nourished its inheritors for years to come, was responsible for its death. After the end of 480, even the frantic patronage of its pious intruders had ceased, and although a small band of monks lived there for a few more years, eventually, devoid of support, they too left this isolated (and therefore remarkably preserved!) mountain vastness to nothing more troubling than the “the chirping of birds and the chatterings of monkeys.” (Ajanta 26 inscription, verse 18) There were hundreds of separate votive images carved or painted at Ajanta during this Period of Disruption—the two or three years of Ajanta’s sudden decline following the exodus of Ajanta’s original patrons from the site. If we count each of the extant images in depictions of the Sravasti Miracle, or other multiple Buddha compositions, the image count would run to a few thousand—not counting those which have long since totally disappeared. Yet no matter how
1
For the most striking example of the Asmakas’ misjudgment, see Volume I, Chapter 11, Cave 21.
194
volume iib, part i, chapter eleven
staggering this number may seem, all of these images—standing Buddhas, seated Buddhas, Buddhas ministering or making miracles, representations of stupas, and images of Avalokitesvara as Lord of Travelers, were created in somewhat less than three years. It is interesting to note that only a single votive figure of the female bodhisattva Tara was created in this same period, even though her presence is magnified and multiplied over the course of the next century in this same general region.2 Of course, the very fact that these years were so troubled certainly increased the intensity with which such “protective” votive gifts were offered, or at least desired. For it is clear from what we know of the manner in which shrine Buddhas were anxiously rushed to completion and dedication by the earlier (original) donors at the site, that the goal of these gifts was merit, the importance of which is confirmed by the many inscriptions on the intrusive imagery of this latter period as well. Since we know that most of these new and “uninvited” donors were monks—and almost certainly monks still resident at the site— we can well believe that many of them took this opportunity to make their own pious offerings at this time.3 And since the departure of the original “Vakataka” patrons must have left many workmen with nothing or little to do, Ajanta—in a dramatic reversal—had now become a “buyers’ market”; carvers and painters, instead of being hard to get, as was the case a few years earlier, immediately after Harisena’s death, were now eager and probably desperate for employment. Their skills were now very high, after years of experience, but their charges now were surely low; perhaps they even bartered their artistic abilities against the ritual knowledge of the monks, accepting the latters’ prayers in lieu of payment. It was of course the presence of this “captive” cadre of workmen at the site, and the fact that all “gross” excavation work had now ceased, which explains how so many carved and painted images could be made, and with such efficiency, in the site’s last few years of disruptive patronage. Furthermore, it was surely not easy for the artisans to leave. In this
2 Actually, a second Tara appears as a subsidiary figure in panel L7 in Cave 26. Although this image was probably carved in late 478, the intrusive phase did not begin in Asmaka caves until 479. 3 Cohen (1995, 412) lists 36 intrusive inscriptions by monks, 6 by laymen, 1 by a laywoman; 13 are indeterminate.
the breakdown of patronage
195
time of mounting troubles, with all related sites were in the same decline, where were they to go? Although—or perhaps because—Ajanta’s intrusions, all made by uninvited (or previously excluded) donors, were all created in a remarkably short space of time, the sequence of their development, the criteria for their placement, and the reason for the precipitous cessation of all such pious efforts, can all be revealed with some degree of precision. We shall see that this is true in the case of the redecoration (and subsequent intrusions) in the Hinayana caitya halls, discussed in the next section; and the study of intrusions in the various Vakataka period caves will further clarify this complex matter. But for the present we will only mention the evidence bearing upon the startlingly sudden disruption of all intrusive patronage throughout the site at a date which we arbitrarily define as December 31, 480.4 The evidence for this very sudden cessation of work is to be found very convincingly in Cave 26, Cave Upper 6, Cave 4, and Cave 22. In all of these caves many of the intrusions which were underway last and are typically late in type are located in the least desirable— generally rearward spots—where the rock is bad, or the light is poor; and they were clearly left unfinished at the same point in time. That is, it appears that, almost certainly on the same day, the workmen laid down their chisels or their brushes, and left, or perhaps fled, the site, never to return.5 Obviously, some trauma drove them away, even before they had collected their pay. When we ask what this trauma might have been, we can perhaps get some idea from the evidence of the Visrutacarita, which relates that when the Asmakas had gathered together the forces of the lord of Vanavasi and of other insurrectionist feudatories, they started from the Asmaka homeland and marched north toward the Narmada river, where they encountered the (inadequate) forces of the ineffectual
4
I am of course using a modern calendar, but the point is that work ended at a very specific point, representing the decisive end of patronage at the site. 5 See particularly the rear areas (or low-priority rear walls) of Cave 26, Cave 4 and Upper 6 for these “terminal” works, all in various stages of incompletion. Understandably, they are usually sculptured images, but in at least one extant example time also ran out on a painted image—the hastily inscribed Litany scene on the left wall of Cave Upper 6. The various unfinished paintings in Cave 1 have no connection with this group, having been cut off by Harisena’s death some years earlier.
196
volume iib, part i, chapter eleven
Vakataka emperor, Sarvasena III. In the ensuing battle the ill-fated emperor “became mincemeat, because he despised political science” (Ryder 1927, 214).6 Then, “after bestowing a petty fraction (of the plunder) on the lord of Vanavasi, (the Asmaka leader) faced about and reduced the entire realm of Anantavarman (= Sarvasena III)” (Ryder 1927, 215). We can well believe that this march northward, which may well have been on a route through ancient Risika (whose ruler had joined the insurrection), posed a dire threat for the Ajanta region.7 Although we can hardly believe that the Asmaka armies came to the site itself, it is conceivable that they took the same present-day route which leads through the famous pass nearby. In any case they must have posed a material threat to the people of the region, and perhaps even the threat of conscription; of course conscription might have been a desirable alternative in a world where there was now all too little to eat or do. Of course the greedy imperial plans of the Asmakas did not stop with this local campaign. Within a few years it appears to be the case that they had extended their control, whether strong or fragile, over most of the great Vakataka empire. But they could not hold it together. It all too soon fell apart, back into the pieces which Harisena had fused together, or into new pieces rising out of the chaos which the ultimately unsuccessful Asmaka insurrection engendered. We must assume that at this point the Ajanta region was taken over by rulers, probably Hindu, who had no interest in continuing to develop the great cave site developed by the defeated Vakataka power; for no matter what the explanation, not a single image was ever added to that once-flourishing site again.
6 7
Kale 1966, 360 translates: “fell a prey to those princes”. See discussion of Gomika of Risika in Volume I, Chapter 15.
VOLUME IIB, PART II
PATRONAGE: THE HINAYANA CAVES WITH EMPHASIS ON THEIR REDECORATION IN VAKATAKA TIMES
CHAPTER TWELVE
PATRONAGE OF THE HINAYANA CAVES: CONSIDERATIONS
I have made the point that Ajanta, during its Vakataka phase, was an elitist site, in which no one but courtly and/or highly privileged patrons could find a place. However, one should not confuse the patronage situation at Ajanta during the Vakataka period with the situation which existed many centuries before, when the site’s Hinayana caves were created. At that time it is evident that patronage was essentially a more characteristic community effort: in Cave 10 an incised inscription on the right of the great arch states that “the façade is the gift of Vasisthiputra Katahadi” (Inscr. 40. An incised inscription on the left wall (Inscr. 41) states that “the prasada is the gift of Dharmadeva”, although there is no agreement on the meaning of prasada. Another incised inscription on the left wall, also near the front records: “The wall is the gift of Kanhaka of Bahada” (Inscr. 42). The last of the early records (Inscr. 43) yet found in Cave 10 is painted, is fragmentary and of imprecise import; Cohen suggests that it may be descriptive rather than donative. In Cave 12, just left of the right rear cell doorway, an incised inscription speaks of “the religious donation of the merchant Ghanamadada” (Inscr. 66), referring to the cell, and further confirming the “community” character of the donations of this early phase. This pattern of involvement is typical of most early Buddhist sites, and of certain later ones as well. However, it stands in significant contrast to the exclusivity of Ajanta in its Vakataka phase, where each cave was the proud precinct of a single patron who, if he managed to get it inscribed at all, put a generally lengthy donative inscription in a place where it could both visibly and verbally announce his virtues. Such virtuous “advertisements for myself ” appear in the expectedly prominent locations in Caves 16, 17, 26, and Varahadeva’s Ghatotkacha Vihara. There are also hasty donative inscriptions, truncated by the unforgiving impositions of time, in Mathura’s great Cave 4 and in Upendragupta’s sumptuous but aborted Cave 20, while the latter ruler failed to get his exceedingly long inscription,
200
volume iib, part ii, chapter twelve
surely planned in the capital but never brought down to the site, incised in the panel which had already been prepared for it on the front wall inside Cave 19. The original Vakataka patrons never troubled, or felt it appropriate, to put their name anywhere but in their dedicatory inscriptions, which alone would sufficiently (and impressively) establish the reality of their patronage. It is only the later intrusions, all added in the Period of Disruption, that are identified by separate inscriptions, placed beneath separate images or beneath image-groups. According to the count made by Cohen (1995, 412) no less than fifty six of these are still extant, all but three of them painted, often hastily, perhaps because time and funds were short. The goal, in any case, in these years of the site’s impending demise, was not esthetic or social or political achievement, but the acquisition of merit. It seems likely that certain literate monks wrote these brief records; and it seems likely that illiterate and unconcerned painters then merely copied the letter forms from the written “originals” provided by the donors, whether monastic or lay. Someone qualified could analyze the number of “hands” involved in painting the inscriptions, to suggest whether a few scribes (and roughly how many) were assigned this task, and/or whether certain errors might suggest that the painters themselves added them. One can hardly believe that the monks themselves painted the images, although this has often been proposed or assumed. Of course the fact that there are obviously far fewer “hands” than there are figures of the painted Buddhas negates any assumption of “one hand/one image”, and the same conclusions would surely emerge from a study of the painted records, many of which were obviously written by a limited number of scribes, or perhaps by painters copying the words they were given. It seems somewhat unlikely that monks would be put into service as the scribes who added the inscriptions to the painted donations. Such donative and at the same time intrusive inscriptions—many long since missing—are still to be seen in certain caves in abundance, as listed by Cohen (1995, 412) and described fully in his very useful Appendix: Ajanta’s Inscriptions, reproduced here. Cave 2 has six still extant intrusive inscriptions (plus seven descriptive labels and verses), Cave Upper 6 has two, Cave 11 has three, Cave 16 has five (plus a few graffiti), the tiny Cave 22 still has three (plus two descriptive labels), Cave 26 has three (all carved), and a number of other
patronage of the hinayana caves
201
Vakataka viharas anywhere from one to three still visible. The old Hinayana caitya halls have even more of these late Vakataka period inscriptions identifying the monks or lay persons who donated many of the Buddha images on the still available and thus generally lower priority faces of the pillars during the Period of Disruption. Cohen (1995, 412) counts no less than sixteen extant intrusive records in Cave 9 and eighteen in Cave 10. Many other such “intrusive” inscriptions are now missing, we can safely assume, since they were typically painted on a panel beneath the image. The lower portions of such donations were often destroyed by debris which built up over the centuries, while any painted inscriptions at the base of intrusions cut on the façades were rapidly obliterated by the weather, especially when, over the course of time, the protective eaves broke away. It is clear that the two Hinayana caitya halls Caves 9 and 10 had retained their sanctity from early times, for they both show attempts at redecoration late in Harisena’s reign, as well as intrusions added immediately thereafter. The vihara Cave 12 was also redecorated, so we can perhaps assume that the smaller Caves 13 and 15A (sometimes numbered 30) were repainted as well, but both are very exposed, so any confirming evidence is now lost.1 Of course, being “mere” viharas (without shrines) Caves 12, 13, and 15A did not have the “sanctity” of the old caitya halls; so it is hardly surprising that intrusions were never added to them. Significantly, in Cave 10, the extant votive inscriptions, fourteen referring to gifts by monks and four indeterminate (Cohen 1995, 412) are only to be found in the lower priority areas, notably on the less readily visible faces of the more forward octagonal pillars, or on more frontal locations on some of the more rearward pillars, where the latest of the intrusions were probably painted. Such generally less desirable locations for these intrusions had to be chosen now because the cave’s programmatic redecoration had been started in the cave late in the site’s heyday, and although it was aborted, it had already filled up many of the more desirable areas. Needless to say, it is important to distinguish the work ordered by the original (Vakataka) patrons and accomplished as part of this carefully 1 Cave 15A, located just to the right of later Elephant Cave of Cave 16, was particularly damaged by dirt and rock which fell down from the steep scarp above. Like the Vakataka Cave 29, it was still hidden from sight completely when the caves were numbered, explaining its necessarily late and expedient numbering.
202
volume iib, part ii, chapter twelve
controlled program, from the intrusive imagery added helter-skelter during the Period of Disruption. It is also of interest to ask why this great cave—Cave 10—as well as its smaller neighbors, Caves 9 and Cave 12, were not redecorated early in the site’s renaissance, so that it could better be of service during the long years in which the site’s elaborate new caitya halls were still in process of excavation. Since Buddha imagery was of such fundamental importance in the later caves, and of course never shown in the earlier ones, one might well wonder why the early caves were so long left without such significant additions. By the same token we might ask why the old Hinayana doorways—all in our so-called B mode or C mode—were not “upgraded” to the far more workable D mode after 475, as was the case with the majority of those outmoded doorways in the Vakataka period caves. There is, as we would of course expect, evidence of the usage of the cells in the Hinayana viharas—in a number of cases the upper pivot holes were actually repositioned (surely in Hinayana times) to make the doors hang better, while blocks of wood were often inserted below to make for smoother pivoting. However, it is still remarkable that there are generally no signs of wear made by the doors’ pivoting, as is always expected in those cells of the Vakataka period which saw significant usage.2 Although it seems reasonable to assume that the monastic community, burdened by the need for residence space from (indeed particularly from) the start of Vakataka activity in 462, would have used the old caves for sleeping quarters, it may well be that nobody took up—or desired the meritorious credit for— the refurbishing the old doorways in the outdated Hinayana excavations. The new patrons apparently had other and more self-interested concerns, and for this same reason they did not trouble to redecorate any of these old caves until very late in Harisena’s reign. By this time, the site was mostly filled up; and with enthusiasm mounting, this may have finally been perceived as a significant thing to do. If we cannot certainly answer the question of why the refurbishing of these old caves was not started until so late, at least we can 2 There are some curious exceptions, as in many of the cells of Cave 16, where it may be that wooden “sleeves” (to smooth the doors’ pivoting) were inserted in the “pristine” upper holes, even though the lower holes, when not obscured by the ubiquitous cementing-in (!) do generally reveal the expected wear. Many of the upper holes in the various “un-worn” Hinayana doorways, however, were too small to contain such (hypothetical) sleeves.
patronage of the hinayana caves
203
locate the programmed redecoration activity fairly closely in time. We can show what portion of it was accomplished in the happy and productive days just before Harisena’s death, and what was done by the remaining but troubled “Vakataka” establishment in early 478, before the planned program was aborted in the middle of that year, at which point the old esthetic (even if pious) controls were replaced by the more selfish desire for merit.
CHAPTER THIRTEEN
CAVE 10: REDECORATION
The “redecoration” program of Cave 10 probably started, as would be expected, at the upper level, with the repainting of the main vault, surely in bad condition after some four hundred years. This later decoration, of which the merest traces remain today, apparently consisted of nothing more than simple and (compared to their Hinayana counterparts) hastily-painted lotus medallions filling the multitude of “coffers” between the old wooden fittings; the latter must have still remained (or were repairable) at that time, judging from the “grid” pattern left by the plastering, which is still visible today. Although only one medallion is more or less preserved in the huge vault of the interior; those on the eave of Cave 10 have all been lost, along with all of the other façade painting. However, although the decorative pattern of the vault is very simple, the process of achieving it could not have been. The huge interior space must have been filled with a ton or two of scaffolding before the vault surface was reached, and the whole hall must have been a jungle, filled with the bamboo supports and braces that we can assume were used.1 And of course this was only the start of the total program planned. When the scaffolding was finally taken down in the hall as well as from the façade everything else—which would also generally require scaffolding—had still to be done. If it is understandable that this program had been put off for years, when there was so much to occupy the already pressured crews, it is also evident that this vast redecoration scheme was not the incidental inspiration of a few pious devotees. Getting the work done, finding the money to pay for it, and somehow co-opting the necessary crews from the artists busy at work throughout the site, must have required decisions and commitments at the highest levels. Nor is it surprising that this work did not even begin until about
1 Possibly the hall was treated section by section; but the manner typical of such scaffolding today, when large buildings are constructed, suggests no such division.
cave 10: redecoration
205
477, when the site’s cresting energies seemed to make all things possible, and when everyone was surely—with a blind exuberance— looking forward to years of further fulfillment. That this was not to be was soon to be realized in the ultimate abandonment of the ambitious program, not long after it was beginning to be realized at the lower level. It seems reasonable to assume that the huge stupa, which must have been in particularly bad condition as far as its surfacing is concerned, would have been one of the first things worked on in this program of redecoration. However, only a small fragment of painting has survived on it, somewhat to the left of center on the dome. This painting, in a style rather like that of the Nanda scene on the left wall in Cave 16, is clearly from the period of Vakataka patronage, rather than a Hinayana survival. Portions of the faces and upper bodies of a few vigorously drawn figures in a simple architectural setting can be seen. The attitude of what appears to be a flying dwarf at the upper right suggests that the whole group, probably originally quite numerous, was focused upon a seated Buddha located (we may surmise) precisely at the front center of the stupa’s dome. When we realize the excitement that the addition of Buddha images to the stupas in Caitya Caves 19 and 26 must have caused, the assumption that a Buddha image—necessarily painted—was added to the Cave 10 stupa seems eminently reasonable, while the presence of the fragmentary “attendant” group lends support to the reconstruction. Considering the manner in which the whole program in Cave 10 developed, even though this (hypothetical) image was probably one of the first things realized in the redecoration process, it was not likely to have been done earlier than 477, or possibly late 476, as we shall see. Thus it would most probably have been of the newest bhadrasana type, in fact emulating that being carved on Cave 26’s stupa at this very time.2 The semi-vaulted ceilings of the side aisles were perhaps the next areas to be repainted, showing facilely rendered but attractive seated (and attended) Buddha images in the ceiling areas on the curving surface, along with an array of similarly painted standing Buddha 2 In the unlikely event that this hypothetical bhadrasana Buddha was painted as early as 476, it would pre-date the carving (if not the conception) of the first bhadrasana Buddha images in Caves 26 and 16; but this would not be surprising, since painted bhadrasana images predate carved ones at the site by about a decade.
206
volume iib, part ii, chapter thirteen
images on the projecting faces of the stone “rafters” of these same aisle ceilings. However, if we start with the very exposed examples now just outside the frame of the cave’s protective screen, there are thirteen “bays” filled with such imagery in the left aisle, but only nine in the right, stopping between pillars L6 and L7 and pillars R4 and R5 respectively. Interestingly, at these points and proceeding rearward, the seated Buddhas were “replaced” by simple lotus medallions which quite possibly filled the ceiling of the whole (already plastered) ambulatory, although time, aided by bats and dampness, has obliterated any that might have been in the apse area. The money-saving (and time-saving) shift to simple lotuses which could hardly have taken an hour apiece to paint—the whole ambulatory having been plastered in advance—would in itself be suggestive, but the fact that the shift from Buddha figures to mere lotus medallions occurs at such different points at the left and the right would seem to describe a sudden change of plans—a reduction of expectations—in mid-course.3 The decoration still proceeded, but no longer with the exuberance of 477; considered together with the course of the redecoration of the pillars, it is reasonable to ascribe this shift to the impact of “reality” caused by Harisena’s death, and to the troubled context of the early months of 478, when in the adjacent viharas the “Vakataka” patrons were giving up their grand plans in an anxious effort to get their Buddha images finished and dedicated. There was no attempt to redecorate the wall surfaces below these aisle ceilings at this time. This may have been due in part to the fact that they still retained their long stretches of Hinayana jataka paintings. However, these were by now obscured by perhaps as much as five centuries of sooty grime, which makes them almost illegible today; thus it seems unlikely that there was any strong wish to preserve them, particularly since they contained no figures of the Buddha—which would of course have been preserved by convention.4 It is likely that these walls were merely accorded a low pri3 Similar economies, but planned in advance, are revealed in Cave 19, where the vault starts off filled with painted Buddha images, but they alternate with (cheaper!) stupas starting above pillars R4 and L4. The first four splendidly roiling zoomorphic motifs on the left and right sides of the same cave’s triforium yield to simpler floral forms farther down. In Cave 26, the four-armed dwarfs “supporting” the pillars, become two armed at a point well towards the rear. 4 It might be noted that these Hinayana narratives appear on a carefully lime-
cave 10: redecoration
207
ority since they were dark and mostly not readily visible; it is reasonable to assume that had the whole redecoration program been continued, they would eventually have been included in it.5 Instead, during the Period of Disruption, intrusive donors started to fill these wall areas, at the more convenient lower level, sometimes impinging upon the old narrative cycles, deeply begrimed, slightly higher up. Since the left aisle ceiling was obviously getting decorated somewhat faster than that of the right when this change occurred, it is not surprising to find that the continuation of the program, involving the repainting of the pillars, also was started on this more advanced left side. However, it too suffered an interruption just when it was proceeding well. Although some of the pillars are now “blank”, being damaged or replaced, we can see that the intended program was to start (logically) at the tops of the pillars and that it continued down as far as Pillar L10 before breaking off.6 Probably in the interests of efficiency, once they were up on a scaffold, two ranks of images were painted at the same time, with a single painter (using a distinctive palette) responsible for the work on a particular pillar or on adjacent ones. (It appears that the same painter did this highlevel work on both Pillars L4 and L5, another being responsible for L9 and L10). Emphasizing the fact that the highest row (with bhadrasana Buddhas) and the next (with standing Buddhas) were planned and executed together, it is clear that they shared their enclosing “bead” borders, as well as the flowers in the spaces between them. It seems evident that now (in 477), when there was no expectation of serious cutbacks, the plan was to “surround” the pillar tops completely with these images. Yet surprisingly, only five of the eight
plastered surface which, before removal, surprisingly covered over a carefully incised inscription stating that “The wall is the gift of Kanhaka of Bahada (Inscr. 42); the same seems to have been the case with an early painted inscription on the left wall under the fourth rib (Inscr. 43). This “cover-up” seems to prove that the plastering and painting, quite possibly of the whole cave, was done sometime (even if not much) after its original excavation; otherwise it would be strange to have covered over these early records. 5 In the more rearward areas all traces of paint in the aisles are gone, so we cannot be certain that the well-organized program was continued there. However, traces of plaster suggest that refurbishment of these areas was intended and therefore possibly accomplished. 6 Pillars L1, L2, L3, L6 and L8, and most of L4 and L7 are reconstructed in cement.
208
volume iib, part ii, chapter thirteen
faces were completed, the three faces at the rear being left merely plastered, on all of the pillars in question (from L4 through L10, with a few missing). We could of course assume that the painters (or more likely the planners) were eager to see the work that “counted most” completed first. But a better explanation might be, since the decoration of the aisle vault was started at the same time, that the latter’s scaffolding was in the way; there was no easy way to work on everything at once, and so the painting of the relatively “unimportant” rear faces of the pillars was temporarily put off—and then of course never got finished, due to the disruptive death of the emperor. Once again, we see how Ajanta’s life was intertwined with his. Had all of this consistent decoration been finished, as was of course originally intended, the great hall, with Buddha imagery covering all of its thirty nine pillars would have presented a veritable litany of significant forms, with a staggering esthetic as well as pious impact. It is relevant to note that the painted bhadrasana images at the tops of the first left pillars have a number of late features which, even if minor, are helpful in establishing the date of these carefully painted figures to the very last years of Harisena’s reign. The coiling fronds arcing out of the makara heads on the thrones, and the leonine throne legs (as opposed to lions merely placed beneath the throne) never occur in sculptured Buddha images prior to 477, and the same appears to be true in painting. The manner in which the lions are raised up high on pedestals would also never be found prior to 477. The “radiant” white lines (stylized flames) on the green halos are commonly found on images in the Period of Disruption, but never appear prior to 477. Finally, the bhadrasana type of Buddha never appears in sculpture prior to 477, nor do capping “nubs” appear on the corners of the thrones of carved Buddhas until that time, although (surprisingly?) both features appear in painting nearly ten years earlier, and for this reason are less useful than the motifs above for establishing the absolute date of these pillar images.7 In terms of the sequence of this redecoration work, we can tell that the aisle ceiling decoration was already underway when the decoration on the pillar tops began. This would be expected anyway, as the redecoration program logically proceeded downward, but it
7
See Cave 19, left wall, for precedents.
cave 10: redecoration
209
is “proved” by the fact that one of the latter Buddhas (on pillar L5, above face H) slightly overlaps—and therefore must have been painted subsequent to—the lotus medallion on the adjacent beam, which itself would have been done as part of the aisle ceiling’s new painting program.8 However, this does not necessarily separate them significantly in time, except to tell us that the aisle vault was finished somewhat in advance of the pillar tops—a conclusion which is confirmed by the unfinished rear faces of the left pillars. It is significant to note, too, that the painting of the right aisle vault appears to have lagged behind that of the left aisle vault, since the vault Buddhas, belonging to 477, end behind Pillar R6 and R7, while those at the left continued down to Pillar L4 and L5, before yielding to cheaper and quicker lotus medallions. This delay of work on the right equally explains why the consistent decoration of the pillar tops, seen on the left colonnade, was never started on the right. Instead, clear traces of the earlier Hinayana decoration remain visible at the upper levels of some of these pillars (note R5–7) and in a few other areas which, being at an inconvenient height for intrusions, were never covered over.9 These early painted motifs have often been compared to the floral and geometric decorations on the gates at Sanchi. Many other Hinayana designs are either long since lost or are covered with the many Vakataka period intrusions, or by the plain slip applied in preparation for the latter; but there are many other pillar faces, especially in less visible areas, which apparently were not decorated at all, other than with their simple lime plaster surfacing or plain colors either in Hinayana or Vakataka times. What seems to have happened is this: the refurbishment of the cave’s great vault was started first, and when it was done (or possibly shortly before) and the vast array of scaffolding removed, the new mud-plastering and redecoration of the aisle ceiling and the painting of the upper levels of the left pillars was started. Work in the left aisle was clearly started slightly in advance of work in the right aisle, where fewer Buddhas were painted in the aisle ceiling, and where no images at all were added to the upper levels of the
8
This overlapping was clearer a few decades ago! A small area of Hinayana painting also remains on Pillar L5, Face C, confirming, as we might expect, that the left colonnade had also been painted in Hinayana times. 9
210
volume iib, part ii, chapter thirteen
pillars before work ended.10 We do not know why the workmen (or their bosses) chose to “favor” the left side—nor does it seem very consequential!—but it could be because the light was better on that side, or that the old pillars were in better condition; often at the site such decisions depend upon the quality of the rock, or the prior preparation of the surface, or the quality of the illumination. Shortly after the carefully organized work on the forward part of the left aisle ceiling and on the upper levels of the more forward left pillars had been accomplished—with a smaller amount of painting on the right aisle ceiling started just slightly later—the shock of Harisena’s death impacted upon the site. At this point the painters were ordered down from their scaffolds and told that their subsequent decoration work would have to be speeded up, not only by using lotuses instead of Buddha images in the redecoration of the aisle ceilings, but by drastically limiting the redecoration of the cave’s pillars. The happy days of 477 were past; it was now early in 478. The site’s long-held controls had not been forsaken yet, in early 478, but those in charge of the redecoration now were required to face the changed situation very realistically. They did not give up completely, for the old “Vakataka” patrons clearly still exercised some authority over developments at the site during the early months of 478, prior to the time (in the middle of that year) when the Asmakas boldly rejected Sarvasena III’s overlordship. However, just as the main Buddha images in the surrounding Vakataka viharas now had to yield to the demands of expediency, the redecoration of the great hall had to suffer similar money and time-saving cutbacks at this time. That it was continued at all in the anxious milieu of early 478 suggests that significant merit must have been expected from the endeavor, even though in this phase of work there appear to have been no inscriptions assigning credit. There is not a single donative inscription on the Buddha-groups at the top of the left pillars or on the paintings of the Buddhas on the ceiling of the aisles, nor will we find any on the “processions” of Buddhas described below. Thus, at the start of 478, with concerns mounting, these “Vakataka” patrons ordered the expedient shift to “cheaper” lotuses on the aisle ceilings, and gave up their plans to totally cover all of the hall’s pillars “from top to bottom” with a well-disposed myriad of painted 10 There are Buddhas high up on pillar R2, but they are clearly intrusive, being painted only on the most visible faces. One is inscribed (Inscr. 54).
cave 10: redecoration
211
Buddhas on all eight faces, in the manner started on the front left pillars. Instead, they decided, expediently, to concentrate solely on painting Buddha images on the more accessible lower half of the pillars, with concentration on the two most visible faces, arbitrarily designated as Face A (= the most visible face as one enters the cave) and Face B (= the axial face, also immediately visible).11 We can assume that the painters started with a row of Buddha images (all now totally missing due to the buildup of debris in ancient times) at the base level of the pillars. They then added another row slightly higher up; this is by now largely destroyed, except for the upper levels of the orange-haloed figures. Then they added a final (still higher) row of white-robed and white-haloed Buddhas, which occupy a position more or less midway on the pillars. Only this relatively high row remains relatively intact, even though in a precarious state of preservation. Had time allowed, they would presumably have continued this same mode of work on the pillar levels above, of course reversing (in the interests now of convenience and speed) the more consistent and (from a decorator’s point of view) logical program which was developing on the left pillar tops just before the shock of Harisena’s death threw the intended scheme into disarray.12 What results is (or would once have been) a virtual procession of painted Buddhas moving, as it were, toward the shrine. It is possible that the now-obliterated lower two “processions” continued right down to the stupa (at about pillars L13 and L14) but the higher row of white-robed Buddhas stops at the tenth pillar on both left and right.13 Apparently this work was done as a contractual assignment, as part of the overall redecoration program, for (like the orangehaloed fragmentary figures just below) it was surely the work of a single group of painters, working in much the same style and with
11 Thus, starting with Face A, the lettering proceeds clockwise for the pillars of the right colonnade, and counterclockwise for those of the left colonnade. For a diagram, see Cohen 1995, 326, reproduced here in the Appendix on Inscriptions. 12 The painted surface on the most rearward pillars has been almost totally lost, but was probably not painted at this time; the so-called Buddha with the One-eyed Monk, on Pillar 17 Face A is an intrusion, yet is on a major face, along with a many other intrusions on the rear pillars, which were accorded a lower priority. 13 Pillar L11 is reconstructed in cement. The “procession” at this third or upper level might have been continued as far as L12, but if so only on the most visible face (Face A) where a ruinous figure might be seen as part of the sequence. But L12 Face B breaks the pattern, as do all of the images on the further pillars, some of which (L15A, L17B and L18A) are indisputably intrusive, since they are inscribed.
212
volume iib, part ii, chapter thirteen
much the same palette. However, despite its processional impact, the painter of the white-robed Buddhas surely did not start his assignment at the very front of the cave, for there are very differently painted images on the highly damaged pillars L1 and L2; the situation on the right side was apparently the same.14 Thus it seems safe to say that the “contract” for the painting of the “white-robed procession” started at pillars L5 and R5, and ended at pillars L10 and R10. The reason for ending at this point, before getting close to the stupa, was probably because the planners of the second row (the row just beneath) became excessively enthusiastic as their work progressed into this same area. The standing Buddhas on Faces A and B of these more rearward pillars (R12, R13, R14 remain intact), along with the charming celestials above, take up so much space on the pillar faces that there was no possibility of the white robed procession continuing past pillar R10, and the same situation appears to have affected the procession at the left too. It seems clear that work on the three ranks (the lowest totally missing), as it proceeded upward quite expediently, did not take any particular account of what was to come next in the rank above. It can be assumed that now, early in 478, painters were expected to do their work quickly, with less concern for quality than in those happy, even if final, days of Harisena’s reign, when the fine decoration of the highest levels of the left colonnade and the confidently facile painting of the Buddhas on the aisle ceiling was underway. And as if to speed up the process, it is evident that a fairly large number of separate painters were involved, even though in every case they were constrained by the intended overall plan. Furthermore, work was logically divided. We can tell by studying the style and palette of the few and fragmentary orange-haloed Buddhas on the second (middle) row, and then of the white-robed procession above, that each painter was responsible for painting the Buddhas on both Face A and Face B of his particular pillar. And this is equally true of the “expanded” images on Pillars R12–R14. Furthermore, every one of the pillars on which these processional images are still visible was surely painted by a different artist, with the likely exception of the paired white-robed Buddhas on pillars R5
14 Pillars R3 and R4, as well as L1, L2 and L3 are now totally cement; but the images in this area on L4 are clearly the work of a different painter.
cave 10: redecoration
213
and R6. However, although the “hands” are so various, the general pattern of the compositions and the coloring of robes, the haloes and of the background are carefully controlled as part of a master plan. Thus every white-robed Buddha on the right pillars stands against a neutral/ochre-ish plain background at shoulder level and a plain reddish background at the level of the feet. Perhaps in the interest of speed, enhanced by the fact that the painters would get paid by the job and not by the hour, these backgrounds have been left very plain. In this they are quite unlike the various intrusions in the cave, which have many added (even if often sloppy) flowers or other decorative details; quite possibly this is because each of the latter intrusions was the offering (we assume) of a particular devotee/donor, whereas the “procession” paintings were merely part of an imposed decorative scheme. And if the intrusions often have donative inscriptions, the “decorative” (even if spiritually significant) processions never do. A painting on your own wall, with meaning for you, might well be signed, but that is not true of the wallpaper with which the room was decorated. However, although various distinguishable groups of images can be seen, these were in no sense private offerings, as was the case with the many Buddhas added, where space was still available and time allowed, to these same pillars in the Period of Disruption. As we might expect, these “intrusive” Buddhas often have painted dedicatory inscriptions, whereas the “programmed” processions, like the rest of the apparently authorized redecoration of the cave, never have any. Just as in the caves sponsored by the various major Vakataka patrons, if credit was to be given, and merit made from the (attempted) redecoration of the hall, this would typically be “concentrated” in a single donative record; but if such a record pertaining to this ambitious attempt at total redecoration ever existed in Cave 10, it has long since disappeared.15 The “processions” in Cave 10 were surely truncated because of time/cost considerations, but at the same time the manner in which their location was chosen helps to confirm the assumption that fifth century donors and devotees had little interest in circumambulation. 15 It is interesting to note that the merit accruing from any donation, if assigned at all, was generally to the donor’s mother and father and/or to the whole world, not to the donor himself. Such generosity perhaps only added to the merit of the grantee.
214
volume iib, part ii, chapter thirteen
The very impact of Buddha figures being added to the stupas in Caves 19 and 26—for the first time in this region!—and (consequently) of Buddha images being cut in the vihara shrines, was surely great. At Ajanta, the focus was so much on a face to face confrontation with the image that circumambulation as a ritual was much suppressed. There was no main image in any of the Vakataka viharas that was planned for circumambulation—in fact the organization of the shrines increasingly denies the very possibility of circumambulation, even though the earliest ones were all intended to have the image at shrine center.16 As for the Vakataka caitya halls, they too very much underplay the significance of the space behind the stupa in favor of the image which fronts it. In Cave 26, the eight rearmost pillars have no carved decoration while, although to a lesser degree, the decorative forms at or toward the rear of Cave 19 are also generally simplified. Thus it is hardly surprising that the Vakataka phase “processions” under discussion in Cave 10 are not continued past the stupa, which itself would already have been redecorated by this point with what was almost certainly a Buddha image at front center, even though only traces of the attendants remain, as noted above. Thus, with the painted processions not only leading down to the stupa, but being disposed on the two pillar faces most easily visible as one enters and then proceeds down the hall, it would seem clear that the interest and action was not in pradaksina, but in abhigamana (approach to the image). Certainly this is true with regard to the Buddhas in the viharas, where one either never could or never would circumambulate within the shrine, and in the Caitya halls too (Caves 19 and 26) the focus is clearly on an axial approach. Although a few different types of Buddhas were included in the “processions” on the pillars, there is only one remaining instance where three instead of two pillar faces were painted at this time. The “extra” image here is painted on the next-most-visible face (face H of pillar R5) in the same style and coloring as the other two. Possibly the missing lower row also had three instead of two images on each pillar, but it would appear that just as the middle row got started, concerns about time and money or the painter’s contractual
16 For discussion see Volume I, Chapter 11: Cave 16 (Circumambulation Reconsidered).
cave 10: redecoration
215
agreement intervened; perhaps for this reason, the number of figures was reduced to two immediately after the middle “procession” was started. Certain scholars have suggested that the white-robed Buddhas that take part in these somewhat hurried but carefully disposed “processions” show the influence of Gandhara Buddha imagery.17 As Dr. Anand Krishna (1981) has pointed out, “the samghati as a whole gives the impression of a thick cloth covering both the shoulders in (this) tradition.” He reasonably suggests that these “white” Buddhas show a fusion of the Gandhara idiom with elements of the GuptaVakataka style. At the same time we should note that the disposition of the robe has certain parallels in Mathura and Gandhara imagery as well as at Ajanta itself. In any case, such an influence from the northwest would not be surprising, for Ajanta’s development was surely much affected by the fact that it lay close to a significant trade-route—upon which trucks and cars move so busily today. But it may be relevant to mention (as pointed out below) that nearly every pair of these “Gandhara” images was done by a different Ajanta painter, making quite different types of (albeit white) robes, as if there was no insistent consensus. If we wish to support the “Gandhara” theory, perhaps we are seeing here the effect of an authoritative painting master, who may indeed have come from that region, and here imposed his will on the workers with varying degrees of success. This would seem to be the only way that we can explain why the dozen or more painters assigned to work on these particular pillar faces managed to conform, although each with certain idiosyncracies, to a rather novel formula.18 Furthermore, if the painters involved were truly adepts in a “Gandhara” tradition, why are there practically no comparable figures painted anywhere at the site, even in Cave 10 itself ? Are we to assume that these various painters, who (to speak generously) spent hardly more than a day on these particular assignments, worked nowhere else at the site? There is one unexpected variation seen in the sequence of whiterobed Buddhas on the left colonnade (Pillar L9A); this is a anomalous 17
Yazdani 1964, 38–39; Krishna 1981, 96–99; 1991, 254–257. Krishna 1991, 256 suggests that “the patron showed a preference to this special type which is projected in the present instances.” This recognizes the same imposition of authority that I suggest, but there is no reason to think that a single patron (as opposed to a master painter, contractor, or sub-contractor) determined the style of this group of pillar faces. 18
216
volume iib, part ii, chapter thirteen
seated Buddha image. Although it is white-robed, its presence breaks the insistent sequence of white-robed walking (or standing) Buddhas which catch the attention as one enters the cave. However, the reason for it being different is clear. The image just below (barely the head remains), like some of the images on the right pillars (R12–14) takes up much more space than most; especially with its fine flying devotees, there was no room for a standing white-robed image above it, when the subsequent “white” procession was painted at a higher level. Should we require confirmation that the seated white-robed Buddha was painted subsequent to the figure beneath, we can note that its lower portions overlap details of the flying devotees belonging to the figure beneath. This and other similar evidence makes it clear that the painting of the three “processions” on either side started with the most convenient lowest group and then worked upward, until this expedient final stage of redecorating the great colonnade broke off suddenly at about the midpoint of the colonnade. This explains the obvious gap between the processions (assignable to the first half of 478) which filled the lower half of faces A and B on the left pillars, and the paintings filling the upper part of those pillars, which had been started so confidently and consistently in 477. This “gap” on the more desirable faces, A and B, of the more forward pillars at the left was soon filled up by intrusions (sometimes inscribed) in the Period of Disruption. The same was true of the roughly corresponding area above the processions on the right pillars, where of course there was no gap, since the tops of those pillars had never been filled with the (unfinished) “eight Buddha” sequences. Needless to say, these still relatively desirable areas were not the only ones taken over for intrusions in the Period of Disruption. Intrusions gradually filled up the lower priority spaces, with an understandable preference given to the frontal Face H, and to the relatively visible Face C; however, intrusions also appear in a rather random and expedient fashion on the darker rear faces and, as any particular pillar filled up, at higher levels too. As we might expect, these private donations become sparser on the harder-to-see pillars at the back of the cave. Indeed, since the earlier “processions” never were continued down to this dark part of the cave, a number of intrusions are actually placed on the lower level of faces A and B in this area. The intrusive character of these individual offerings of the Period of Disruption, suggested by their non-programmatic char-
cave 10: redecoration
217
acter, is clearly proved by that fact that, despite occupying Faces A and B, some have individual donative inscriptions (L17B, L18A). In contrast, where the processions were painted, all other extant inscribed Buddhas are either in the areas just above the processions (L9A, L9B, R5B) or on the lesser faces: L9G, L10G, R6H, R7H, R10F, R10G, R12F.19 One intrusive inscribed image requires special mention, since it appear on Face B of Pillar R2, along with a hastily painted disorganized array of other Buddhas on this pillar and the adjacent pillar R1 (R3 and R4 are totally reconstructed). As mentioned earlier, it appears that the three processions (the lowest lost) did not start until Pillars L5 and R5. This was possibly because they were more “readable” if started at this point rather than at the very front of the cave. Or perhaps the first four pillars were in an area obscured by paraphernalia at that time—possibly scaffolding put up for the repainting of the now much-fallen façade. Or possibly, during the programmed phase in early 478, they were reserved for another contractor or contractors who never took up the work. In any case, it seems clear from the evidence of pillars R1 and R2 that, at least on the right, these most forward areas were still unpainted when the Period of Disruption began and thus were later filled with intrusions, as were the still unpainted areas in the rear of the cave.20 It is very important that we properly separate the “programmed” paintings from the slightly later intrusions, if we are going to establish a meaningful pattern of development in the redecoration of the cave. Cohen has failed to do this, by concluding that the procession of white-robed Buddhas was painted subsequent to those just above, which we have distinguished as intrusive and therefore later rather than earlier than the programmed works. Discussing the inscribed standing Buddha, “wearing yellow/orange robes in a green mandorla” relatively high up on Face B of Pillar R5, he states that beneath it is “a standing Buddha wearing white robes, his head surrounded by a white halo (this halo intrudes over the background of the higher figure)”. If this observation were correct, it would mean
19 Two inscriptions, originally on Pillar L8 and one on Pillar R5 are no longer extant; although there is no record regarding their positions, we can assume they followed the same pattern as these just discussed. (Inscr. 46, 47, 55). 20 The processional decorative program of early 478 may have ended at Pillar L10, all images beyond being intrusive.
218
volume iib, part ii, chapter thirteen
that the lower figure—the white-robed Buddha—would have to have been painted after the inscribed intrusion above. It is indeed likely that the ruinous second (orange-mandorlaed) procession was painted after the (now-missing) lowest row and would have overlapped it; and we can clearly see that in turn the third row (of white-robed Buddhas) overlaps the second. However, the row of white-robed Buddhas by no means overlap the figures (sometimes inscribed) still farther up, as Cohen suggests. It does indeed appear that the white-robed Buddha’s halo, not only in Cohen’s Pillar R5, but in virtually every other instance in this long procession as well, does “intrude over the background of the higher figure”. But when we study the juncture of the figures, we can see that the painter of the higher figure (probably subject to a “horror vacui”) has filled in the background space around the halo of the Buddha image below with the flower covered ground upon which his own image stands. Even the disposition of the hastily painted flowers conforms to the prior presence of the halo. Such arguments are fully convincing in themselves, but they are “clinched” by the fact that, for some reason, the white-robed Buddha on Pillar R5A retains its total background area, showing us what all of the others were like too, before they were covered up by flowers falling down from above. Furthermore, if we study the smaller intrusions (limited in size by the presence of the early Buddhas above) on the major faces (A and B) of Pillar L5, we discover that in both of them, the lower margin “jogs” distinctly downward at the right, to make room for a painted devotee, and that this adjustment of the margin reflects the prior presence of the halo of the white-robed Buddha below. Furthermore, we might note that although there is a seeming order in the intrusions painted at this (slightly) later level, with their rather standardized green mandorlas and orange robes, the more impressive figures appear only on the single axial face (Face B). This face was apparently given priority by the intrusive donors, who had little concern about the effect of the “processional view” as one entered the hall in the Period of Disruption; nor were they concerned to continue the paired patterning of the carefully programmed Buddhas below. We might also note that the coloring of these intrusions is quite characteristic of the many other scattered intrusions in the cave, possibly reflecting the better availability, or perhaps the lower cost, of such pigments during the Period of Disruption. Green mandor-
cave 10: redecoration
219
las, edged with “radiant” white lines, now seem to be the design of choice, not only in Cave 10 (and Cave 9) but throughout the intrusive work at the site. Significantly, the treatment of the slightly earlier programmatic Buddhas in the processions is quite different. As we have noted, even on the pillars which have survived (many are restored fully or in part with cement) the painting on the lowest of the three rows and on the lower level of the second row is obliterated today. This is because destructive debris built up over the centuries to some six feet in depth, at least toward the front of the cave. Today, in what was originally the second row from the floor of the cave, little more than the heads and shoulders of most of the images survive, for wherever the seasonally dampened debris, its powers exacerbated by a good percentage of bat and bird excrement, came into contact with the paint and its mud-plaster base, it literally dissolved them away. The action of the debris, as it dried and then became wet again as the seasons changed, also weakened the natural (geological) faults running through the pillars. This caused the lower sections, despite their size and weight, to be loosened and displaced, millimeter by millimeter, over the centuries, until these lower sections sometimes literally fell over onto the floor, leaving the upper parts of the pillars still connected with the ceiling. Old photographs and prints show the damage, with chunks of the pillar bases lying on the cave floor. It is the very depth of the debris—now long since removed—that built up in Cave 10 after its abandonment from the fifth century on, that explains why the famous inscription on Pillar L13: “John Smith 28(th) Cavalry, 28 April 1819” is located about eight feet from the floor. Even though the record is quite high above one’s head today, the sponsor of this first of many modern graffiti at Ajanta obviously had no trouble writing it—by scratching it with some sharp instrument—quite easily and with a certain innocence across the chest of the painted Buddha which adorns the axial face of the pillar.21 The fact that he could do this so easily was of course due to the fact that when he (probably along with his fellow soldiers) came into the cave on this first recorded visit by any foreigner, even the
21 Smith probably did not know that he was damaging a Buddha image. Early writers often considered the images Jain or Hindu, those religions being more familiar to them.
220
volume iib, part ii, chapter thirteen
more rearward portion of the old hall was filled with at least four or five feet of debris. The sequence of the (mostly) standing Buddhas, to which John Smith’s rather innocently “violated” image belongs, are all related in terms of style, types of costumes, and priority positioning, as well as by their total lack of votive inscriptions. Thus, they are distinctly different from the many intrusions which, being started only after the best locations had already been utilized, were necessarily relegated to the less well illuminated and/or less visible shaft faces or to the less easily accessible locations higher up on the shafts. However, compared with the insistent completeness of the work on the ceilings, which in normal course would have been started first, this “authorized” program of pillar decoration, which occupies only the most visible and easiest-to-paint areas, falls far short of what must have been the original plan conceived in the happy days when Harisena was still alive.
CHAPTER FOURTEEN
CAVE 10: INTRUSIONS: SUMMARY
If we except the scattered and often inscribed “intrusions” described above, all of the decoration that we have so far discussed in Cave 10 is “anonymous”, part of a carefully developed (and then reduced) program which, to sum up again, was originally aimed at redoing the whole great hall in a thorough and up-to-date manner, but which, after mid-478, had to greatly limit its expectations. This left the great cave, during the latter part of that year and continuing up through 480, open to the pious depredations of a very different group of donors. Sometime after mid-478, and throughout the Period of Disruption, the floodgates suddenly opened for the plethora of “intrusions” which start to fill up the places thought to be most desirable in terms of location and light. Just as the shift from the Buddha images to cheaper lotus medallions on the aisle ceilings came as a result of Harisena’s death, the decision to give up the time-consuming covering up of the whole pillars and to concentrate only on painting the two most visible faces, as far down as the stupa, can be explained in the same way. Beyond, behind, and above the better organized triple processions, assignable to the first half of 478, the painting on the pillars breaks into a helter-skelter arrangement typical of the intrusive phase, with many images separately inscribed by anxious donors, as was common in the Period of Disruption, but never before. By Cohen’s count (1995, 412) fourteen such inscriptions in Cave 10 alone are still extant, referring to the pious gifts of monks, most of whom were undoubtedly still resident at the site; the four other inscriptions are “indeterminate”.1 However, even if these intrusions are all “selfish”
1 For the import of these donative inscriptions, and the fact that the great majority refer to donations by monks, often honoring parents, see Schopen 1997, esp. 61–63, 31. The count is much increased by Cohen 1995, 412, who has investigated such images at the site with great care.
222
volume iib, part ii, chapter fourteen
personal donations, their arrangement is to some degree visually disciplined by the very constraints of their necessary placement on the narrow faces of the pillar shafts. The apparent break in the consistency of Cave 10’s redecoration program suggests, as we have noted, that this whole abortive program of redecoration was started very late in Harisena’s reign and then was hurriedly continued (but never finished) early in 478. This would explain the interruption of the originally grand conception, and its reduction to the point where only the most “essential” or readily visible work was accomplished. And of course this left the way for the intrusions which, starting almost immediately (perhaps as early as mid-478), filled up the many spaces which would never have been available had the original program been brought to completion. The difference between the two “campaigns” in Cave 10—the programmed, started shortly before Harisena’s death and continuing for a few months thereafter, and the personal, belonging to the Period of Disruption—is tellingly revealed on the left pillars, near the front. There, between the carefully disposed bhadrasana images seen at the top, dating to late 477, and the processionally programmed whiterobed Buddhas dating to early 478, located just above eye-level, a helter-skelter group of hastily painted intrusions fills up the space between.2 These disorganized panels, at least two of which (L9A and L9B) are inscribed, are typical of the intrusive Buddhas which were painted on the still-available and often less than desirable faces of the right pillars too; and although it is clear that the more visible locations were the more desirable, the pressure to donate caused much of the available space at the rear of the pillars to fill up too, in a somewhat random, expedient, fashion. Happily for the intruders, it seems likely that the cave’s pillars had already been plastered when the total program was first conceived while Harisena was still alive. A number of these in the undesirable dark and isolated space in the apse, behind the stupa, still retain the mud-plastering, but are bereft of Buddha images. Many other pillars toward the rear have
2 As on the right pillars, there must once have been a programmed series of standing Buddhas painted at the lowest level on these left pillars; but they have been lost.
cave 10: intrusions
223
few if any images on the rear pillar faces; for instance the low priority aisle faces (D, E, F) of pillars L15 to L19 are all blank, while the better-lit faces (A, B, C, G, H) all have been completely or partially filled up with this rather randomly placed imagery. As we might expect, the painters worked on the most convenient areas first, so generally one can find informative overlappings of the lower panels by those higher up. On the right pillars, where the tops never got started in late 477, as they did on the left, the new offerings, now usually proceeding upwards rather than downwards, seldom reached the top of the shafts. Thus the fine floral and geometric decorations applied in Hinayana times can still be seen at the upper levels of a number of these pillars. Cohen (1995, passim), in his discussion of the pillar figures in Caves 9 and 10 and elsewhere, continually refers to the small figure or figures kneeling, generally at the proper right, at the Buddha’s feet, as “donors”, sometimes properly qualifying the identification with quotation marks. If there are two figures side by side, they are assumed to be “parents”. For instance, on the inscribed (no. 60) representation of a white-clad standing Buddha on Pillar R10 Face G in Cave 10, he sees “three donor figures, all dressed in white. To the Buddha’s proper right is a monk, to his left, two “parents”. Yet if we credit what this and indeed all other such inscriptions at the site say, the gift is always that of a single donor. By Cohen’s count (1995, 412), thirty-six records on these intrusive offerings name a (single) monk as donor, six records name a layman, and thirteen are indeterminate. Only one image (flanked by standing attendants rather than by a kneeling figure) was given by a lay woman (Inscr 13). Quite apart from the question as to whether one can count three “donors” in a painting of the Buddha, when the inscription specifically refers to only one, we must note that throughout the site, particular after 475, when large sculptural representations of the Buddha become more and more popular, we find such kneeling attendants, sometimes singly, sometimes in groups, generally male but often with females also, carved in contexts where they could not possibly be representations of “donors”. That is, these multiple figures are carved at the feet of Buddha figures which are part and parcel of the single(!) donor’s plan. Although the first shrine images completed at the site—up until 471 (Caves 17 and 20)—never are flanked by such attendants, the kneeling figures at the throne base soon become con-
224
volume iib, part ii, chapter fourteen
ventional, and this convention impacts strongly upon intrusive images also, from late 478 onward.3 The kneeling devotees—they cannot be called donors—attending the great Buddhas in Harisena’s Cave 1, Mathura’s Cave 4, Aurangabad 3 and throughout Cave 26, are among the many examples of what is obviously a convention. By the same token, we must note that the “processional” Buddhas painted on the pillars of Cave 10, as part of a consistent (even if aborted) scheme of “Vakataka” redecoration, often have one or more kneeling attendants; yet they could hardly be “donors”, since this whole complex of imagery was part of a single scheme of re-decoration. If any individual or “corporation” was planning to get the credit—the merit—from this obviously authorized project, it would probably have been stated in a major inscription when the whole was done, or perhaps when work was so peremptorily interrupted early in 478. However, no such record exists; and even if it once did, it is hardly surprising that it did not survive the great losses to the cave. It is significant to note that the slightly later “intrusive” representations on the pillars use the same general formula, with one or more attendants sometimes shown, sometimes not, surely dependent upon space or funds available and the desire to show the Buddha’s worshipful significance, rather than upon any need for the many different donors to magnify their own “image”. This would seem to have been particularly the case, when they were so careful to assure the world that their meritorious efforts were on behalf of “their mother and father, and all sentient beings”; not themselves.4 On the other hand, we should not rule out the possibility that a few donors are indeed represented, starting with the anomalous and surprising representation of the kneeling figure who crowds himself into the consequently not quite symmetrical motifs on the base of
3 Those in Cave Upper 6 and 21 they were cut away to save time. In Cave 26 they were added when the main image was redesigned in 477. 4 Williams 2000, 268 (note 18) provides a useful observation about the transference of merit, although he does not explain what “meritorious benefit” might be gained by the donor himself from such a gift. “All Buddhist traditions held that it is an appropriate religious act to pray that the merit that might otherwise accrue from a particular virtuous deed should be transferred to another party. In Mahayana it is held to be the appropriate response of a bodhisattva to gaining merit that the merit should be given away for the benefit of all sentient beings (i.e. all beings with consciousness, all beings that are alive and can therefore feel pleasure and pain).”
cave 10: intrusions
225
the shrine image of Cave 11. This is the only instance where a single figure kneels before the main image, as opposed to the later representations of pairs or groups of figures. Its uniqueness might credit the notion that the figure is in fact a donor, but on the other hand it was obviously created before the convention of balancing such figures was established, so it might just as well be an early and anomalous devotee. No other such single figures appear during the heyday of the site. Single kneeling figures are often found beneath large standing Buddhas in the Period of Disruption, the most famous being the charming painting of the incense-bearing worshipper painted next to the carved figure of a monk, both appearing below the first to the left of the intended Six Buddhas in the shrine antechamber of Cave Upper 6. The situation here is quite unique, for the carved kneeling monk and the standing Buddha which he attends were part of the original donor’s aborted program for the antechamber’s decoration, and must be dated to 477 or just possibly early 478. But work on the intended Six Buddha grouping had to be interrupted because of Harisena’s death, and it was not until the Period of Disruption that the antechamber, still not completed, was in any case plastered and painted.5 Therefore the painted worshipper (although so much “a part of the picture”) is surely intrusive, dating to late 478 at the earliest. Zin has perceptively described this figure, the “fame” of which (despite its original unimportance) would seem to justify repeating what she says: “The worshipper kneeling at the Buddha’s feet holds an incense vessel in his right hand and in his left three white lotus flowers. This man is not a monk, as often stated in literature; he is not wearing a monk’s clothing but a swathe of light cloth that exposes his stomach. He has earrings and rather than being shaved, his hair falls in curls down his back. A flower is fastened in his hair. It is no coincidence that this picture has become one of the most-published paintings in Ajanta: The worshipper embodies the piety of the Buddha’s followers.” (Zin 2003, 19 fig. a) It is hard to be sure, but considering the special character of this worshipper, the fact that it was “unnecessarily” added to a composition
5
See Volume I, Chapter 11, Cave Upper 6.
226
volume iib, part ii, chapter fourteen
that already had a (carved) devotee, and the fact that the new “devotee” is so clearly distinguished from its monkish counterpart and indeed from all of the other intrusions in the antechamber, it might be argued that this is indeed a “portrait” of the donor himself, who (alone or with others) was responsible for plastering and then painting the antechamber early in the Period of Disruption. At the same time, one could argue that the ample space it occupies was planned for it originally, and that the panel was merely being completed by someone who knew the original patron. In fact, the new “intrusive” donor, who helped to complete the decoration of the antechamber (even if to make merit for themselves), might well have been living in the cave already, and possibly even connected in some significant way with the evicted patron. We should remember that either single or multiple kneeling devotees typically accompany standing Buddhas carved during the site’s heyday; yet there is no reason to think that they have any specific connection with the cave’s patron or his family. Such figures appear, datable to 477 or 478, singly or in pairs, male and female, under the five large standing Buddhas in the shrine antechamber of Cave Upper 6 just discussed; and since this grouping (minus the sixth) must have been part of the original patron’s intended iconographic program, these figures can hardly be “donors”. They must be “generic” devotees; and this argues strongly that the kneeling worshipper must have been planned as the same. Among many other examples of “generic” devotees which might be chosen, those attending the monk Gunakara’s huge intrusive standing Buddha at the left of the Cave 26 façade return are striking. Like the Cave 6 image just discussed, the devotees here are also rendered in “mixed media”, with one carved and one painted devotee, the latter a kneeling male. Below the closely related image on the right return, probably also Gunakara’s donation, we find a similar grouping, but the painted figure is a female, and surely not a donor, since Gunakara’s inscription credits the gift to himself alone. Just as elsewhere throughout the site, it seems unlikely that the donor would have ordered “extra” devotees to be carved or sculpted in such panels, if he was actually interested in being depicted himself, as the donor. Zin is properly cautious in not calling the figure a donor, probably because of its being one of a pair of kneeling figures that attend the large image; but this may well be because she does not realize that the two were by no means done at the same time. Her
cave 10: intrusions
227
circumspection extends to a number of other intrusive images, but these are all clearly single kneeling figures, whose “uniqueness” (i.e. singleness) separates them, perhaps meaningfully, from paired or grouped attendants. In all of the examples cited (II 10a, IX 21c, X22c, XVI 25a, XVII 33a) she speaks of these separate figures (in all cases kneeling monks) as “possibly or probably” the “donor” or “donors”. I would agree with her “indeterminate” conclusion, but would tend toward the opinion that (even in the case of the Upper 6 and Cave 11 worshippers) that these figures are probably devotees rather than donors, for the simple reason that if the idea of “putting yourself into the picture” was current at the site, then everyone would have been tempted to get into the act. This is how things tended to develop, and to develop with intensity, at the site. Only a single image of Avalokitesvara—arguably the only bodhisattva treated as an intrusive votive image in this and other contexts at the site—breaks the pattern of repetitive Buddha imagery on the pillars of Cave 10; it appears on pillar R10D.6 Although we might expect stupa-representations also, since stupas are sometimes “substituted” at the site for the more common Buddha images (having equivalent meaning), there are none shown here, although only a single example appears on Pillar L8A in the adjacent Hinayana Caitya Cave 9.7 Interestingly, a few do appear along with other intrusions, on the pillars of the ancient Pitalkhora caitya hall, which was repainted at about this same time, with a similar involvement of many different (and clearly “intrusive” donors) rather than as the result of an authoritative overview.8 The painted faces of the pillars in Cave 10 are so constrained by their narrow format, and the painting so quickly rendered and so unconcerned with the absolute integrity of margins, that one can often determine which panel was done first and which last by carefully scrutinizing the points where lines and/or the colors overlap. 6 Other intrusive representations of Avalokitesvara are found in Cave 4 (left wall of porch); Upper 6 (left end of façade, left rear wall of porch, hall’s left wall, and Cell PL); Cave 9 (Pillars L4G, and two on projecting element on L1); Cave 10 (Pillar R10D); Cave 10A (right wall); Cave 11 (left rear wall of porch); Cave 20 (rear of right porch pilaster); Cave 26 (two on façade right; one on panel under main arch). 7 Perhaps this was because the stupa format is poorly suited to pillar faces. “Intrusive” stupas, either painted or carved, are often found among intrusive Buddha images during the Period of Disruption (see Cave 11, Cave Upper 6). 8 For Pitalkhora, see Volume I, Chapter 14.
228
volume iib, part ii, chapter fourteen
Such observations reveal how typically (and understandably) the direction of the intrusive panels moved from the more ideal to the less ideal positions, whether the latter were to be found toward the back of the pillars, or at the more inconvenient higher reaches of the shafts. Even more significantly, we can establish the priority of the “programmed” (and earlier) images on the most readily visible pillar faces by finding “overlaps” from the “intrusive” panels just above them, or on the less desirable faces toward the rear. As noted above, there are sufficient instances to clearly confirm the somewhat earlier dating—to early 478—of the “processional” Buddhas vis-à-vis the scattering of adjacent intrusions, all done subsequently. It is perhaps not surprising, considering the size of Cave 10, and the brevity of the Period of Disruption, that the old hall never got completely filled up with the new intrusive paintings. As far as we can tell from the inscribed images, most of the donors in this troubled period were monks, as noted above. Although their interest in making such donations must have been very great, especially after all the years when they could not participate in such merit-making activity, it is reasonable to assume that not only time but money was running out as the Period of Disruption continued. Furthermore, we must assume that the artists, now that their employment by major patrons had ended, were rapidly leaving the expiring site. But the work, despite its gradual erosion, ended more with a bang than a whimper. I have shown earlier that it ended precipitously, probably because the region came under a sudden military threat by the Asmaka/insurrectionist forces. That is, work on the intrusions, throughout the site, was proceeding in a normal fashion on one day, and had ended peremptorily by the next. This sad day was, if the reader will allow an oversimplification: “December 31, 480”, after which not a single image was ever painted or carved at the site again. Because so many of these intrusive Buddhas have idiosyncratic features—in their dress, their haloes, their coloring, the treatment of background—it should be possible to (roughly) divide them into a number of different hands, giving some idea of the number of painters at work in the Period of Disruption.9 Furthermore, one might also 9 Cohen 1995, 3 discussing inscription 56 also recognizes “some administrative control over the decorative program” of Cave 10, but the analysis is faulted by his assumption that the intrusions were done at the same time as the “consistent” (even if reduced) redecoration program of which the “processions” form a part.
cave 10: intrusions
229
be able to determine if some of the same painters responsible for the “processions” of early 478 turned their hands to the intrusions being made a few months later, and if (as one would of course expect) at least some of them also worked on the intrusions in the adjacent Cave 9. Furthermore one could analyze the inscriptions on so many of the intrusions in both Caves 9 and 10 and indeed, elsewhere at the site; and in doing this many questions emerge. Were they the work of various scribes or could the (presumably illiterate) painters have been given a written guide to follow? Could the monks (or other devotees) have written their own inscriptions, or those of some of their associates? Can the paleography, or grammatical idiosyncrasies, provide clues in this regard?
CHAPTER FIFTEEN
CAVE 10: THE AISLE PAINTINGS: ORIGINAL AND INTRUSIVE
Long stretches of Hinayana paintings are found on the upper level of Cave 10’s aisle walls, slightly above eye level. Although much has been lost by today, they probably continued without a break throughout the very rear. It is also likely that other similarly early scenes were represented on the wall space below them, even though any such paintings (some or all obscured by intrusions, now mostly lost) were obliterated by the action of the debris which built up in the cave over the years of the site’s abandonment from the late fifth century onward. These remaining Hinayana narratives, which in format are like a long scroll unrolled, show scenes from the life of the Buddha on the left, along with the story of Udayana, and the Syama and Saddhanta Jatakas on the right.1 Although heavily begrimed, they have been in large part preserved, even though this was not due to any particular reverence for them or desire to save them on the part of donors during the Period of Disruption. Rather, in the case of Cave 10, it appears that these late donors were primarily concerned that their intrusions be placed at a convenient lower level both for making and for seeing. For this reason they sometimes imposed their later Buddha images on such earlier narratives which (it seems likely) may once have been painted below the extant jataka compositions. It is hard to say whether the original planners of the redecoration, when it was begun during Harisena’s reign, would have had the same attitude, or might have wished to preserve these stories, for it seems clear that their planned program of work was aborted before reaching this point. There must have been many later intrusive images painted on these long wall areas, but nearly all have long since disappeared due 1 Those which probably existed at the rear have been lost. For a most useful visual and verbal description of the remaining Hinayana Jataka paintings, see Schlingloff 1999, pages (with drawings) 24–27.
the aisle paintings
231
to the action of the debris which over the centuries gradually built up against or around them at this unfortunately (as it turns out) “convenient” level. Today, only the very top portions of few images or image groups remain, all intruding upon the lower margin of the old jataka scenes without any apparent guilt, since there were no actual representations of the Buddha in the Hinayana narratives. Assuming that other Hinayana narratives were indeed painted below the still-remaining ones, those now-lost stories would probably have been quite fully obscured by these intrusions so conveniently painted over them. The remaining (extreme) upper part of the late iconic group on the left aisle wall (near Pillar L10) hides the lower portion of nothing less than the Wheel of the Law—representing the Buddha’s First Teaching in the Deer Park. This famous subject was the focus of the whole extensive Hinayana composition, and even four or five centuries later, it is hard to believe that it would not have been recognized by the “intruders”, even though they obviously did not consider it sacrosanct, or they would not have obscured it.2 That the “Vakataka period” Buddha images in question are intrusions is revealed by the fact that their placement is anomalous rather than programmed, but also because it seems clear that the original course of redecoration, continued in early 478, ended before work on these somewhat out-of-the-way and dark areas could ever get started. All of these fragmentary intrusions on the aisle walls (the one on the left behind Pillar R10 and three on the right at Pillars R4, R5 and R13) were probably representations of seated Buddhas with attendants on either side. Those on the right seem to be linked by a long stretch of what appears to be a white roof structure (much obscured), as if they might once have comprised a long series under a pavilion; but this can hardly be said with any conviction. These late intrusions, fragmentary as they are, are clearly distinguishable as late paintings because they have hardly any grimy deposits on them, in great contrast to the deeply blackened earlier paintings upon which their remaining upper levels impinge. Needless to say, these
2 See Schlingloff 1996, X,12, where the remaining top portion of the wheel is shown; the remaining top of the “intruding” painting is not shown; it has recently been better revealed (showing the upper part of the Buddha’s umbrella) by Mr. M. Singh and his associates carrying on conservation at Ajanta.
232
volume iib, part ii, chapter fifteen
intrusions, like those on the pillars, did not suffer from similar deposits because the cave, and the site, must have been essentially abandoned so shortly after they were completed. By the same token, the deep blackening of the Hinayana paintings strongly suggests that the cave may have been in continual use right up to the time of its repainting late in the site’s Vakataka renaissance. This is because these paintings were located at quite a distance from the stupa where, we would assume, worship with begriming oil lamps would have been carried on; that they have been so heavily blackened surely suggests that worship in the cave had gone on for many years.3 This matter of the usage of the Hinayana complex is an important one, but one about which little evidence is available. We only know that although Buddhist patronage surely suffered a general decline in Western India after the second century A.D., the old site was still functioning to some degree when Fa hsien traveled through India very early in the fifth century. However, Fa hsien did not visit the site, probably because, as he reported, “the roads . . . are dangerous and hard to travel”, while those “who wish to go there have to present money or goods to the king of the country”; indeed, he suggests that the quickest and safest way to visit was by air: “The people of the country often see men flying to the monastery here.”4 If such observations show that the new donors in Cave 10, during the Period of Disruption, had no compunction about intruding upon, or indeed covering up, the ancient Hinayana painting, whether narrative or decorative, this is even more obviously true in Cave 9, as we shall see. The Hinayana Jataka tales and the like which remain in Cave 10, were apparently preserved only by chance or by convenience; had the intrusive phase gone on longer, they would surely have been covered over. The new donors both here and in Cave 9 found it easy, quick, and cheap, to locate their iconic intrusions expediently, without concern for anything they might be covering up, no matter how sacred such representations might have seemed to the ancient donors, or to us today. 3
It is possible (but seems unlikely) that oil lamps were put in the aisle areas to make the old paintings more legible; if this were indeed the case, they would have had to have been put on the floor, since there are no hook-holes either in the adjacent pillars or in the walls themselves. 4 Fa Hsien: A Record of Buddhist Countries, trans Li Ying hsi, Peking 1957, 74–75.
CHAPTER SIXTEEN
CAVE 10: FAÇADE INTRUSIONS
There was relatively little space available for intrusions on the façade of Cave 10; but those that were done (or which still are extant) are clearly late in type, just as we would expect. Two standing Buddhas, under arches bereft of carved decoration, appear on the lower right respond; both show a series of repeated smaller Buddha images on the left—the right sides must have been similarly treated but are now broken away. The presence of such “redundant” frame images— as if various donors were sharing the same panel—suggests a late date of 480 for these panels—as does the manner in which the small seated Buddhas at the side of the lower panel are raised up on projecting lotus pedestals. One should also note the unexpected use of the abhaya mudra utilized for two of the small standing Buddhas in the upper panel. As mentioned elsewhere, this gesture appears to have been only rarely used while the Asmakas were controlling the site, from the mid-470s until the end of 478, but gained favor again in the Period of Disruption, after the Asmaka departure. In the case of the higher of the two, it is evident that the expected varada mudra was given up because of a fault in the rock at the point where the lowered hand would have been; either the hand broke away in the course of carving, or the sculptors feared that it would. We have seen elsewhere how the same change was made in other areas where faults in the rock, or breakage in the course of carving, militated against the carving of the proper right hand in the much more standard varada mudra.1 It is harder to be sure that the lower figure shows abhaya for the same reason, but this does indeed seem to be the case. Note the problem which has been caused by the thin vertical fault which runs close to the left edge of the panel. Indeed, a small piece of rock has split away because of the fault, just where the lowered hand would have been. It would seem that the sculptor decided
1
See Spink 1986.
234
volume iib, part ii, chapter sixteen
to make the abhaya mudra at this point; probably the chipped-off thigh area was built up with plaster, prior to being painted. The lower Buddha panel has a wide area beneath that was being smoothed when work broke off. Like the large Buddhas on Cave 9’s façade, it was probably to have an inscription; indeed, were the inscription a painted one—as were all but three (extant) in the Period of Disruption—it may have once been present, but has now long since washed away. The higher panel also shows a roughly defined area at its base; it too must have been planned for an inscription, but perhaps had to be finished with plaster in order to get the inscription written. The fact that the carving of the panel suddenly broke off, suggests that the panel was probably underway when all such work at the site came to a sudden end in the late 480s.2 It is hard to date the intrusions of the fronts of Caves 9 and 10 more specifically than to assign them to the time between mid-478 (when the “Vakataka” patrons had left) and 480 when the Period of Disruption suddenly ended. The “abhaya” images suggest that the margins, at least, date from after the Asmakas gave up their authority over the site; that is they would not date until after the beginning of 479. Of course, the fact that, on both Cave 10 and Cave 9, the panels were never quite finished, a few of the images being merely sketched in, makes it fairly clear that work was continuing right up to the end of the Period of Disruption.
2
For the sudden termination of work, see Volume I, Chapter 3.
CHAPTER SEVENTEEN
CAVE 12
A further suggestion that the refurbishment of the old Cave 10 did not start until shortly before Harisena’s death—perhaps as late as 477—is the fact that the associated Hinayana vihara, Cave 12, was also redecorated at an equally late date. We know this because the ceiling, which would normally be refurbished before the lower areas, was repaired with the red ground-brick plaster which was never used at the site until the very last years of its consistent patronage—probably not until 477. The red plaster was of course widely used thereafter, during the Period of Disruption; but such repair work was of no interest to donors during that late period, so it seems evident that, whether completed or not, this work was started during the site’s late heyday. Just as was intended (although never completed) in Cave 10, this refurbishing of Cave 12 involved the whole cave, walls as well as ceiling. This seems clear, for the once lime-plastered and then painted arch motifs carved along the upper level of the walls were now totally covered over with the same characteristic late red-plaster. This red plaster, both on the walls and on the ceiling, has fallen away, partly because it has poor adherence, but also because the broken cave was long exposed to the elements. However, it has the virtue, from the investigator’s point of view, of leaving a distinctive red stain on the stone surfaces to which it was applied. We can clearly see that the painted designs under and around the old arch-motifs were in extremely bad condition when the red plaster was applied many centuries later, for the red stain distinctly identifies the early losses. This is very evident in the “tattered” caitya window designs inside the stone arches; in fact, it appears that the major decorative forms were intentionally built up using the thick lime-plaster, with a paste-like consistency; this may have been done to emulate the carved stone designs in some of the rock-cut arches on the left wall. The whole design would then (presumably) have been painted, although all one can now see are traces of the early
236
volume iib, part ii, chapter seventeen
white limeplaster with a few traces or stains of the late red plaster remaining on its surface. Happily, we can determine with certainty that the gray mud-plastered ceiling dates from Hinayana times. The evidence can be seen on both sides of the hall, at the juncture of the ceiling and the wall, where the old ceiling (which in normal course would be painted first) is slightly (but convincingly) covered by the old Hinayana lime plaster which had been applied to the walls before the early painting was done. Once the ceiling had been repaired with the red plaster, it is reasonable to suppose that the whole cave was then repainted, but the only evidence suggesting this is the decoration which can still be seen on the remaining areas of earlier Hinayana mud-plaster on the ceiling. The painting in these areas is in bad condition and “hard to read” but it is reasonable to conclude that it remains from the fifth century refurbishment, and that the whole ceiling was replastered after the application of the red plaster repairs, which are mostly in areas where the stone is cracked and evidently allowed moisture to seep in. Unfortunately the decoration on the ceiling has disappeared wherever there were red plaster repairs, because all of that plaster has fallen, leaving only its typical (and informative) red stain; thus we do not have proof that the remaining designs on the formerly-plastered parts of the ceiling are definitely due to a later repainting, even though it would be surprising if the ceiling had been extensively patched and then not properly and fully decorated. It seems likely that this refurbishing was done very late in Harisena’s reign, along with the beginning stages of the work in Caves 12 and 10. However, it might have been done, like much of the redecoration in Cave 10, early in 478, when for meritorious reasons hasty lastminute work was going on throughout this section of the site. It is unlikely that it was done during the Period of Disruption, when the total focus of donors and devotees was on more “significant” concerns.1 Hopefully, the date of the remaining ceiling design can be determined through chemical analysis, or by discovering if there might
1 On the other hand, many of the monks’ cells at the site were never plastered until the Period of Disruption, so it is conceivable that some donor might have made merit by refurbishing the cave at that time.
cave 12
237
be an earlier layer of Hinayana painting beneath. In any case we can confidently conclude that the replastering of the whole cave (except for the still intact old mud-plastered ceiling areas) was accomplished in 477, because one would not find red plaster used earlier than that, nor would one find such a consistent program of repair done later than that. The very fact that the plastering was done so close to the time of Harisena’s death of course might allow one to surmise that the painting of both walls and ceiling hardly got started before having to be abandoned. I have mentioned water damage to the old plastered ceiling, which, along with various “normal” losses, necessitated later repairs, but it is of interest to note that there is one further area where red plaster repair was needed. This was in the area around the monolithic eye-hook, clearly created when the cave was being excavated, probably for hanging a garland, at the exact center of the hall.2 It seems apparent that this damage—although not very extensive—was caused by the changing of the garlands which presumably were hung there, probably starting in early times. We have seen similar plaster breakage around hooks (or remaining hook holes) in Cave 2 and in many other excavations which enjoyed some usage. The fact that the damage (necessitating the red replastering around this carved device) was less extensive in Cave 12 than in the Vakataka caves where this evidence is found, may be because the ceiling is lower and so the hook was more accessible; but it is perhaps more likely that a secondary, more easy to utilize, hook or hanger was tied to the monolithic one above, and the garlands were hung from the latter. Or perhaps the damage—clearly early, because of the loss of the old plaster—was minor because garlands would not be used much in such a mere “dormitory”. It is just possible that some kind of an icon was hung from the hook, and such an object would not require much (damaging) attention. There is no evidence that it would be for an oil lamp, because there is no evidence that soot collected here more than elsewhere on the ceiling over the many centuries when the vihara may have been in use. In fact, it appears that this cave did not interest the “intruders” in any case during the Period of Disruption for, unlike Vakataka
2 Cave 15A has a hole, similarly positioned at the center of its tiny hall; the metal garland hook appears to have broken off, the stub being up in the hole.
238
volume iib, part ii, chapter seventeen
viharas, it had no image. Although it had provided the shrine-less model for the first viharas to have been started in the Vakataka period, by 466 they of course were all being “upgraded” from mere dormitories to sacred halls by the addition of image shrines. Why image shrines were not cut into the rear walls of the old Hinayana viharas (Caves 12, 13, and 15A), which would have required only the sacrifice of a residence cell or two, as in Cave 11 or cave 16 or Cave 27, is hard to explain. It is hard to believe that such old “dormitories” were particularly sacrosanct; it may have more to do with donors believing that merit would not accrue from taking over someone’s previous donation; or perhaps the change was resisted because monks were already living in the cave. In any case, it is clear that no one seems to have been particularly eager to redecorate these old caves; it took almost a decade and a half before such a process was started, even in the old caitya halls. This may seem less surprising when we realize that for nearly a half millennium no one had added any images whatsoever, as far as we can tell, to any of these ancient caitya halls or viharas, even though Fa hsien’s report (see above) makes it clear that they were in use prior to the Vakataka phase.
CHAPTER EIGHTEEN
CAVE 9
Although it took a long time for all of these Hinayana caves to attract the interest of patrons during the site’s heyday, by the last year or years of Harisena’s reign, reflecting its crescent enthusiasms, they had finally begun to gain attention and support. This attention continued unabated during the Period of Disruption, at least in the old caitya halls. We can imagine that this was in part due to the very sanctity which attached to any caitya hall, and possibly to such very ancient ones in particular. But there was another more practical reason: unlike the façades of the later caves 19 and 26, which were totally covered with Buddha images, on those of Cave 9 and (less conveniently) Cave 10 there were well prepared and “empty” spots available for later intrusions. By the same token, the interior colonnades of the “new” (Vakataka period) caitya halls were covered with up-to-date carved decoration, which disallowed the addition of Buddha images, whereas the plain old pillars of Cave 9 and 10 were “readymade” for the application of votive standing Buddhas. The assumption that the Hinayana caves, which were of course the ancient nucleus of the site, had been left in bad repair throughout most of the period of consistent Vakataka patronage—that no attempt at redecoration had been started until very shortly before Harisena’s death—ains further support from the evidence of the ancient Caitya Cave 9. Here, just as in the larger and more impressive Caitya Cave 10, and also in the Hinayana vihara Cave 12, the process of redecoration was started while Harisena was still ruling, but as in Cave 10, it was started so late that it could not be finished. The fact that the situation is so similar in these three major Hinayana caves supports the conclusion that the idea of redecorating them all may have been sponsored by the great enthusiasm which the site witnessed in the very final years of its heyday, when the future seemed vastly more promising than it in fact turned out to be. In any case, it seems clear that this work was not inaugurated until very late in the reign of the great emperor, probably not until 477, the very year of his death. The very fact that they had apparently
240
volume iib, part ii, chapter eighteen
remained “sacrosanct”—had attracted no private intrusions before this time—would seem to suggest that, even if not earlier refurbished, they were at least protected right up until 477.1 There was relatively little programmed redecoration work done on the walls and pillars of Cave 9, as compared with the more extensive work in Cave 10. However, it is likely that the whole vault— far smaller than Cave 10’s—may have been refurbished when such a program was started in 477. However, since no traces of paint now remain there, to reconstruct the missing program, we must depend upon the evidence of the redecoration of the underside of Cave 9’s eave. This shows simple and hastily rendered lotus medallions—white or ocher on a red background—quite similar to the medallions (only one remains!) which would have been painted on the vault of the adjacent Cave 10 at approximately this same time. The “beams” dividing these “coffers” show typically fifth century “Greek meander” and other designs similar to those in a number of the Vakataka caves proper. Cave 9’s overall façade surface, having been long exposed, may well have been redecorated at this same early moment in the planned program; but if so, no traces remain, due to exposure. The thick layer of lime-plaster found in a few protected areas (notably between the parallel stone “beams” under the projecting vault) would have been applied in Hinayana times, but whether or not it once had one layer of painted decoration or two (i.e., Hinayana and 477), its surface has retained no coloring at all. All of the votive sculptures on the cave front are obviously intrusions, dating from the Period of Disruption. Possibly there were a few intrusive paintings on the façade too, but if so, they have long since been washed away, along with the votive inscriptions painted on the smoothed panels beneath the large intrusive Buddhas on either side of the façade. It is reasonable to suppose that if Cave 9’s eave was redecorated at this time—and such “non-essential” (decorative) work would hardly have been undertaken after Harisena’s death—then the inner reaches of the vault would have been redecorated in 477 also, although here too all traces of such decoration are now gone. The wooden rafters
1 The (apparent) fact that even in the centuries before, neither the sangha nor private donors added votive images to them is of course of interest, since Fa Hsien’s report suggests that they were still in use.
cave 9
241
originally “completing” the vault of Cave 9 must (like those of Cave 10) still have been in place in Vakataka times, although all had already fallen or been taken down by the villagers for their fires or building purposes sometime after Ajanta’s abandonment in the fifth century. Examining the grid-pattern of bare rock left by the loss of the wooden attachments in the cave’s vault, we can see that these “structural” elements were put up before the vault was originally plastered.2 Needless to say, in any redecoration scheme, the monolithic stupa would surely have been given very high priority; and after so many centuries as the focus of worship the stupa would have been badly in need of redecoration. Of course, there is no way to tell whether it was the first thing repainted, and whether its redecoration was delayed until the scaffolding needed to repaint the vault was taken down. Indeed, because of its exposed position, the new painting on it (which surely would have been centered on a new and up-to-date Buddha image) has totally disappeared.3 Only traces of its original (Hinayana) lime plastering remains, in its more protected areas. It might be noted that in the interior of the hall, lime plaster alone was used in areas which were more smoothly finished by the excavators: the stupa, the pillars, and the aisle walls. The vault and triforium areas were left with more “tooth”, for they would be covered with a relatively thick layer of mud-plaster; this roughness was appropriate for mud plaster, but would have required too much of the expensive lime plaster which (as on the old ceiling) appears to have been sometimes used as a thin final surface. It is logical to conclude that the “triforium” walls above the pillars were treated in the same way as the vault, with the vertical members affixed first and the plastering of the spaces between them done subsequently. However, here it would appear that the wooden members had fallen and/or were taken away, probably well before
2 The excavated area of the vault above the apse appears to have been recessed deeply, perhaps to remove a flawed section, which might otherwise have become detached and fallen. 3 Surely an image painted on the old stupa would have had priority, if the lower levels were reached, as the work on the rear aisle suggests. As if the very form was sacrosanct, no Hinayana stupas have been re-cut to include a carved image, and of course the stupas in Caves 19 and 26 were planned with fronts which would contain the image. The base (but not the dome) of the relief stupa in the Ghatotkacha vihara was cut for a Buddha image in the Period of Disruption.
242
volume iib, part ii, chapter eighteen
the site’s renaissance in the fifth century. We can assume that this attractive area was due to be redecorated, along with the rest of the cave, starting late in Harisena’s reign, but the many images on it are clearly intrusive, one even having a donative inscription (Inscr 27). The plastering too has the variety and expediency of application that suggests that it was applied until the Period of Disruption, as discussed below. The fact that the flat aisle ceilings retain much of their fine ancient (Hinayana) decoration of lotus medallions, and that this was not covered over later, might lead one to believe that time had run out, with Harisena’s death, before any further ceiling redecoration could be taken up. Even though the medallions are well preserved on the left and right aisle ceilings and might have been left as is for this reason, the very rear areas probably had suffered significant losses by Vakataka times; but the little surfacing that remains there today all belongs to the earlier period, so it seems clear that the old aisle ceiling remained untouched even at the rear in the later period. In Hinayana times, the simple but careful manner in which the aisle ceiling was painted apparently pertained to the modest decoration of the pillars too. Although much obscured by later paintings and by the actions of time, one can still reconstruct the situation. The very tops of the pillars were encircled by two simple and varied bands of plain color, red and ochre and green. These appear to be the very same pigments used in the medallions of the old aisle ceiling, which of course is directly adjacent to these pillar tops. It further appears that all of the pillars, after first being thinly limeplastered, were decorated with these same colors and that there was no attempt to paint them with simple early designs, as was done in Cave 10 in Hinayana times. Although on the more visible faces this coloring has been mostly hidden by the addition of later paintings, in “lower-priority” areas, where even the thin surfacing of later Vakataka mud-plaster (to prepare for paintings) was not applied, the plain colored surfaces can still be seen. It is clear that redecoration work had progressed less far in Cave 9 than in Cave 10, when Harisena’s death occurred. Although a beautiful narrative composition (described below) may have been started in 477 on the cave’s conveniently flat and already lime-plastered rear wall, it was not finished until later; nor had a consistent course of redecoration been started on the aisle ceilings and the more forward
cave 9
243
pillars, as was the case in Cave 10. However, it does seem to be clear that in the early months of 478, before the “Vakataka” patrons fled, there was an attempt to start a program of pillar decoration, albeit with the now appropriate limitations. In fact, what was now done in Cave 9 in early 478, seems to follow the lead, a bit tardily, of the larger Cave 10. There, as we have seen, the redecoration plans were reduced expediently as soon as Harisena died, to the degree that work was continued only on the most visible pillar faces, for the cheapest, quickest, and most effective impact as one entered the hall. The situation in Cave 9 seems to directly parallel Cave 10’s program, clearly supporting our reconstruction of the cave’s (and the site’s) history at this crucial moment. Although work did not continue nearly as long (or as far) as in Cave 10, it seems clear that in Cave 9 as well, a “procession” was planned, but was going to involve only the most visible (A and B) faces of the hall pillars, and that (as in Cave 10) it would not be initiated at the extreme front of the cave (where the clustered entrance pillars were not so readily visible) but at a more effective point somewhat farther down.4 Thus here in Cave 9 we see that it was begun on Pillar L4 and Pillar R4, continued decisively on Pillar L5 and Pillar R5, but then apparently had to be abandoned. This is hardly surprising considering the tensions and the demands in these early months of 478, when the “Vakataka” patrons were so anxious to hastily finish their main Buddhas and then to hastily leave. Just as in Cave 10, we can see that on all four of the pillars being decorated, the two relevant faces (A and B) on each pillar were painted by the same artist; but here, perhaps because the whole conception was much more modest than in Cave 10, it would appear that a single artist did all of the work, as the character of the features, haloes, umbrellas, and coloring would suggest.5 It is of course relevant to note that here, as in the less harshly truncated processions in Cave 10, there are no donative inscriptions; nor would any be expected. 4 If there was a lower “procession”, as in Cave 10, the evidence is not sufficient to reconstruct it. 5 Mere traces of halo, etc, remain on R5A; but it “matches” the work on the relevant faces of the other three pillars.
244
volume iib, part ii, chapter eighteen
The only painting on Cave 9’s pillars which postdates the “programmatic work” on the fourth and fifth pillars consists of “intrusive” private donations of the Period of Disruption (to be discussed below). At that time various intrusions filled up the desirable space of the “triforium” and soon thereafter work began on the less desirable walls of the dark aisles.
CHAPTER NINETEEN
THE ANOMALOUS PAINTING ON CAVE 9’S REAR WALL
It is clear that, except for the images on the major faces of the fourth and fifth pillars, most of the many Buddha images in the interior—like the sculptured images added to the façade—are clearly helter-skelter offerings of the Period of Disruption, which finally filled up the desirable space on the triforium, as well as the more forward reaches of the aisle walls. However, there is one anomalous exception, which probably belongs neither to the Period of Disruption, nor to the final year of Harisena’s rule. This is a series of particularly beautiful paintings, which Schlingloff has suggested represents the visit of the Buddha to the sage Kasyapa, as told in the Catusparisatsutra.1 The identification is controversial, since the two central panels have nothing to distinguish them from the standard iconic presentation of bhadrasana Buddhas.2 However, the specificity of action and the non-iconic character of the composition in the first and last panels would seem to support Schlingloff ’s view. From our present point of view the matter of interest is that two of the four (or really six) scenes in question are clearly narrative in content, which would seem to locate them in the consistent (rather than intrusive) phase of work at the site; narrative compositions do not appear in the Period of Disruption. But at the same time at least some (and probably all) of the separate sections of the total composition had donative inscriptions (a single one survives), which never appear in
1
Catus.paris.atsu-tra, ed. E. Waldschmidt, vols. 1–3, London, 18990–1911 (24 a-j). “Indra. . . . has made a spring , represented by two Nagas . . . so that the Buddha can carry out the necessary ablutions”. (Schlingloff, 1999, p. 23); but the placement and treatment of the two nagas beneath the Buddha’s feet is conventional, being repeated in many instances at the site. Nor is the presence of two fly whisk bearers, standard for such compositions, so specifically identifiable as “the four kings of heaven”. 2
246
volume iib, part ii, chapter nineteen
association with separate images during the consistent phase of activity.3 Thus the anomalous series fits into neither period, the parameters of which are clearly defined; and to solve this problem we can make little more than a suggestion. Since Cave 9’s vault (and probably the whole façade and the stupa) had been repainted while Harisena was still alive, it is reasonable to assume that work had also started on this long stretch of well lit and accessible rear wall during these happy times. But since nothing else in the cave was started in 477, we can well imagine that the panel in question had just got barely underway when it was abruptly stopped by the death of the emperor. We know that redecoration work then continued, very briefly indeed, in early 478, on the fourth and fifth pillars, but that nothing else was done in the cave at this time; and there is little reason to think that an old story like the abandoned narrative on the rear wall now be accorded any priority by planners trying to get the cave decorated as expediently as possible, and now interested only in purely iconic rather than narrative formulations. It was only in the Period of Disruption, perhaps starting as early as mid-478, that the cave came to a new and anxious life. Suddenly, everyone was eager to fill these old caves, which for years had received no attention at all, with their votive intrusions; and the fine rear wall clearly attracted their pious interest. My conclusion (offered without absolute conviction) is that the long narrative composition planned for the rear wall had had to be abandoned when no more than its two outer scenes had been completed; and the reason that only these lateral areas, painted well above eye level, had been completed might be found in the requirements of the scaffolding. In a study of carving procedures, we have found a number of instances where, in an area to be carved, the scaffolds must have been first erected at either end, so that work could proceed on the two sides of the composition at once, the center to be taken care of last.4 Thus I am suggesting that the beautiful narrative scenes at either end had already been painted when work on the composition was
3
Cohen 1995, 338–9 Inscriptions 23–26. See both sides of antechamber in Cave Upper 6, panels R1–R3 in Cave 26. Even the Parinirvana and Mara scenes in Cave 26 were surely separated so that their scaffolding would not cause a problem. 4
the anomalous painting
247
suddenly abandoned at the time of Harisena’s death, when, throughout the site, work on a number of narrative cycles and on so many sculptural and architectural details peremptorily ended.5 Then, probably soon after the middle of 478, some of the eager intruders who were just starting to fill up many of the caves with their own votive offerings decided to take over this fine area for their own spiritual benefit. Since they could hardly obliterate (or want to obliterate) the scenes already made, which contained sacrosanct Buddha images, two of the expectant donors appear to have taken them over for their own private donative purposes.6 Possibly they themselves paid for the work which had been done but abandoned, and thus could legitimately get the merit, which would surely have accrued from the Buddha represented rather than the narrative context in which he appears.7 It is even possible that, if these scenes had been abandoned too abruptly some months before, they hired the original painter to finish the whole, since the character of the work is consistent as well as high in quality.8 Indeed, the stylistic uniformity of the completed composition, and even the care with which the separate sections are organized and “shared”, suggests that this was to some degree a group effort, just possibly by the same (hypothetical!) “planning committee” which had, only a year or so before, started the now-transformed narrative composition as part of an ambitious scheme for handsomely decorating the whole cave. My assumption is that when the outer “narrative” scenes were appropriated by two “intrusive” devotees, other devotees then “filled in” the central area with a conventional bhadrasana Buddha toward either side, a hieratic figure of Avalokitesvara at the left, the latter’s now-missing counterpart (surely another bodhisattva or a standing Buddha) at the right, and an inscribed stupa in between. Surely by intention, this central stupa is placed exactly on the axis of the cave, echoing the old monolithic focus of worship in its form. 5 E.g.: front wall and left wall scenes in Cave 1, left wall scene in Cave 2, to say nothing of much abandoned excavation work. 6 We can understand why new donors would rather take over than obliterate such paintings, even if they were technically “dead”, since (presumably) they had not been dedicated. Indeed, if they had been expediently dedicated, it seems unlikely that later donors could appropriate (and inscribe) them. 7 This is a matter for Buddhologists to settle. 8 Is it possible that certain narrative details (the flames, the snake, mentioned below) got omitted when work on the narrative sections was abandoned; and that there was no interest in adding them later?
248
volume iib, part ii, chapter nineteen
So this incomplete narrative composition, happily already figuring Buddhas at either end would have been transformed into a series of separately dedicated and inscribed votive offerings. Although much of the plaster and the covering paint has fallen away from the very damaged lower margins of the composition, it is clear from old reports that a number of separate donative records had once been painted there. Cohen 1995, 338–339 lists three previously recorded donative inscriptions (Inscr 23, 25, 26) under the bhadrasana Buddhas at the near left, near right, and far right of the carefully centered painted stupa. The latter’s donative inscription, well preserved, is the only one extant. If we reasonably assume that the panel at the extreme left, with its very damaged lower section, also once had a donative inscription beneath, then we can say that there were a total of five donative inscriptions (all seriously damaged now) ranged beneath those major sections of the composition.9 In fact, it would be most reasonable to assume that originally all seven sections, counting the Avalokitesvara and the missing standing image, were once inscribed. Even in the Period of Disruption, when inscribed donations were extremely common, the series here would not be typical; for donative inscriptions seldom appear in the organized fashion that one sees here, on a composition conceived of as a unified (or at least balanced), even if unusual, sequence. Such a precise organization of narrative episodes, separated with what Schlingloff calls “scene dividers” is also not at all characteristic of the “free-flowing” mural compositions made during Ajanta’s heyday, even though the care taken in the organization of the composition and the quality of the imagery has parallels with the well-planned redecoration, as it was started in 477, in the adjacent Cave 10. It is as if a compromise has been made here between the narrative demands of the story, which is much reduced in its detail, (and surely of little interest now in any case), and the more compelling iconic demands of the two “standard” bhadrasana Buddha groups. The latter icons have surely been too hopefully “identified” by Schlingloff as the “night visit” of the gods to the Buddha on the left and the preparation for the Buddha’s ablutions on the right.10 In the many different contexts in other caves
9 10
Cohen 1995, 331–332; Inscriptions 23, 24 (still extant), 25, 26. Schlingloff 1999, 23.
the anomalous painting
249
where extremely similar compositions appear, these two panels would be seen as normal Sravasti Miracle figurations with the conventional two nagas supporting the lotus below. It is hard to see, with Schlingloff, the supporting nagas as representing a spring made for the Buddha’s ablutions, especially when he is busy preaching, while the night visit scene shows him in the same conventional and “standard” Sravasti Miracle attitude.11 In any case, Schlingloff (1999, 23) and the ancient planners (or revisers) of this composition have this in common—that they are adjusting the narrative’s representation, or its interpretation, to serve their different goals. However, this by no means invalidates the point that they wish to make. Even in the first scene, involving a poisonous snake (all too discreetly hidden from view in the Buddha’s bowl) or showing Kasyapa’s disciples trying to extinguish a fire in his hut (but with no flames shown) one might argue that such narrative details have yielded to the iconic importance of the Buddha represented. Here again, it seems reasonable to assume a kind of “conversion” from a traditional narrative formulation to one in which the Buddha image itself has become the focus, surely making a more merit-producing figuration than would derive from the narrative context in which they appear. This would be understandable and indeed expected in the Period of Disruption, when a sense of urgency and anxiety, and the desire for merit, was paramount. We must remember that no “uninvited” private (including monastic) donors ever were allowed to put a single votive donation in any of the Vakataka undertakings at the site until the beginning of the Period of Disruption. So, despite the very high quality of this total span of rear wall decoration, it would be hard to assign this whole composition, with its revealing inscriptions, to the years of consistent Vakataka patronage just prior to Harisena’s death, when narratives are never combined with icons in this way, and when single narrative scenes (like the end panels here) never were given donative inscriptions.12 On the other hand, to assign it purely to the Period of Disruption is hardly possible, since narratives (like the first two panels) are never
11
Schlingloff (1996), Cave 9, 6, 67; 1999, 23. A few narrative scenes have short “label information”, but this had nothing to do with private meritorious gifts. 12
250
volume iib, part ii, chapter nineteen
painted in that period, having suddenly become irrelevant, in a world now so anxiously focused on making merit before time ran out. This would seem to be why the two outer narrative panels, almost certainly painted, or at least planned, in better days, appear to have been unapologetically “appropriated”, and transformed into “selfish” votive offerings when the whole series of panels was finished and then inscribed at the behest of the (presumably) seven (or possibly five) monks eagerly involved in this ultimately intrusive donation. If the beginning of work on the end panels started in 477, the whole was not finished and its parts inscribed until at least mid-478, during the Period of Disruption. As an alternative “solution”, it seems possible that, in the confused period immediately after Harisena’s death, when everyone was rushing to get their Buddha images finished, the administrative control over this old Hinayana cave may have been relaxed or forsaken, and that monks were able to continue (and transform) this composition for their own purposes. In this case, we might date the finishing and inscribing of the seven panels to early 478. In support of this suggestion, we can note that this is the very time when Mathura, another “Vakataka” patron, drastically changed the decoration of his shrine doorway to a purely iconic formulation; and that the patron of Cave 2, giving up work on the story of Buddha’s birth, concentrated his efforts on the more “essential” iconic images in his cave’s shrine and rear wall; and that the patron of Cave Upper 6 turned his attention totally to finishing (or trying to finish) the great images in his shrine and shrine antechamber; and that the Prime Minister Varahabhadra himself renounced all lesser work to concentrate on his great shrine Buddha and its immediate iconic context.
CHAPTER TWENTY
CAVE 9: TRIFORIUM PAINTINGS; AISLE WALL PAINTINGS At the triforium level, the story is quite different, for it was filled up, in the Period of Disruption, with fine but conventional intrusions. Here too, as was common in many Hinayana caitya halls, there were wooden attachments—a series of parallel vertical members aligned with the pillars and the mid-point of the spaces between. Apparently “locked” at the top on the ridge under the vault, there must have been long crossbeams stretching across the nave to hold their lower ends in place, since no socket-holes were cut. All too tentatively attached, it appears that these had all fallen or (being quite accessible) had been taken for village use sometime before the Vakataka phase. If any remained at all, they would have been removed before the refurbishment of the whole cave was started, even if it was not to be fully realized, late in Harisena’s reign. In any case, it is clear that all of the wooden fittings were gone by the Period of Disruption, when the whole surface of the previously mud plastered triforium could now be replastered in preparation for the intrusions in that area. Perhaps the scaffolding, put up for the discontinued program of redecoration just a few months before, was still in place. This replastering was apparently accomplished in separate sections for different donors, who then filled the areas which they had appropriated with intrusions. Although the plastering was apparently done, quite typically, section by section, with a few different mixes, this surfacing bears no connection with the earlier positioning of the wooden fittings, its continuity confirming that the old fittings were no longer in place at this time. If they had been, the replastering would have only filled the spaces between them. On the left triforium the old mud-plaster, clearly much damaged, was not removed before the Vakataka layer was added, so that in many areas there are two distinct layers. Where the rock surface beneath the vertical beams was bare, the new mud plaster adhered differently from the old plaster, which probably accounts for the
252
volume iib, part ii, chapter twenty
remaining beam “pattern” which is seen in many areas. This is admittedly confusing, since sometimes the new mud has adhered well to these originally bare rock surfaces, while at other times it has fallen away in a similar pattern, adhering better to the surrounding old mud plaster. On the right triforium, the mud plaster surfacing is much more consistent than on the left. It would appear that here (except for a meter or two close to the front) all of the old plaster was efficiently removed before the new was applied. Even so (near the front) there are subtle traces on the still painted surface reflecting the bare areas once under the old beams, which had to be filled in with the mud plaster. Since the adjacent areas are better preserved, this would suggest that it may have been the different character of the varied painted surfacing on the left and right triforiums which accounted for the losses in certain areas but not in others. We know, of course, that a painted surface often protects the plaster beneath; so whether a painted surface was weak or strong might account for the degree and character of the losses (or preservation) of the plaster in such filled-in areas. The splendid intrusive Buddha groups on the right wall are distinguished to some degree by their different design, and by the fact that the third major group (near the stupa) are also differentiated by their preparatory plastering, reflecting the “each man for himself ” approach to donations in the Period of Disruption. The splendid (but ruinous) enthroned Buddha at the center was painted on the characteristic red ground-brick plaster which never came into use at the site until late 476 and possibly even 477, and was used increasingly throughout the Period of Disruption. Since it happily leaves a red stain when (as is very often the case) it falls away, we can clearly see here, in damaged areas, that the stained rock surface had been fully cleaned of its earlier mud plaster before this new plaster, for the intrusion, was applied. These triforium compositions are among the most complex and splendid at the site and the fact that they are “uninvited” intrusions in no way reduces their quality; in fact, they take full advantage of the fact that the painters by now had years of demanding experience behind them. The sadly damaged central group (painted on the red plaster) is remarkable not only in the quality of its painting, but in the authority with which the painter has shown the enthroned
cave 9: triforium & aisle wall paintings
253
Buddha in a three-quarter view, with the throne base (note the legs) in assured perspective. Conceivably the artist who painted this splendid figure, with its strong but subtle volumetric character, and its surprising (for the Period of Disruption) positioning on an angled throne, was the product of the same sophisticated workshop responsible for the “seminarrative” scenes on this same cave’s rear wall. One finds Buddhas seated in this angled “informal” way in both, as well among the figures dating to about 470 on Cave 19’s left wall, but the pose is hardly that of choice in the Period of Disruption.1 It might also be noted that, among other comparable details, the same distinctive red and white striped garment is found both here and in the fine painting on the rear wall on a figure in each of the compositions, as one might expect could be the case if the same artist or family group did them both. It would not be surprising, of course, if some or all of the scenes on both the rear and upper (triforium) walls of this cave were painted by the same artist or same workshop, since they are both so close in space and time. It is probably right to assume that all of these triforium compositions, along with the presumed transformation of the panel on the rear wall, were among the first paintings made in the Period of Disruption, at a time when time, funds and workmen were not at such a premium as a year or two later. The paintings on the left triforium were probably of the same fine type as those on the right, but they are sadly damaged, probably because the plaster surfacing was a veritable patchwork of old and new, one layer and two. Although the goal of such expedient refurbishing may have been to save time and money, the many variations of surfacing, which may have reacted to the environment in different ways, perhaps is what caused it to fall away more rapidly than on the right. In a number of areas we can see that the “late” red mud plaster was applied in patches, confirming (not surprisingly) the late character of the work. Possibly, needing a bit of extra plaster, the workers found that it was convenient to borrow some of the red ground-brick mix from the workers on the other side of the hall;
1 An intrusive Buddha image in a similar pose does appear on Pillar L17 in Cave 10.
254
volume iib, part ii, chapter twenty
the variation in color would have been of no importance, because it would be covered with paintings in any case.2 Fortunately, the fragmentary old Hinayana hunting scene—actually a herding scene including mythical animals—painted on the border just above the pillars, was not “cleaned away” by the later intruders. This certainly was not because of any particular interest in it, but because the narrow strip of stone that it occupies was too small to be of use to the later donors; any votive Buddhas painted there would have been all too tiny, especially when there was still so much space on the already plastered aisle walls which could be used instead. As if to confirm the “uselessness” of this strip of stone, we find, near the front right, that to make the intrusive composition above fit better on the triforium space allotted, the feet of one of the standing Buddhas are painted in this area where the ancient herding scene must once have existed. We might also note that the herding scene is painted on a mud plaster base, applied when the whole triforium was originally mud plastered in Hinayana times. This old plaster was left in place by the later “intruders” not out of any interest or respect, but only because the narrow recessed space had little connection with their concerns. Although the new donors in the Period of Disruption did not choose the location for their intrusions in a particularly consistent manner, they generally opted for available locations which were well lit, and in areas which had already been prepared by careful earlier chiseling and (ideally) plastering as well. In the Vakataka caves they were generally quite respectful of already-existing scenes (never of course cutting into or covering an actual Buddha image) but in the old Hinayana caves they appear to have had little compunction about covering over not only decorative features, but even narrative scenes, many of which had of course been quite obscured by grime by the fifth century; of course none of these early scenes, being aniconic, ever included Buddha images. Ready visibility, good illumination and ample space all support the assumption that the intrusions on the wide and well-lit stretch of wall above the hall pillars (the “triforium”) were probably the first, at least in the interior, to have been begun in Cave 9 during the
2 For a similar “loan” by one artisan to another, see Cave 17 doorway (Volume I, çç).
cave 9: triforium & aisle wall paintings
255
Period of Disruption, even though they would have required a good deal of scaffolding. The fact that this area appears to have been completely filled up strengthens this assumption. By contrast, the intrusions on the dark walls of the left and right aisles had to be abandoned before they filled those areas.3 The intrusive scenes on the front wall, at the left and right of the doorway, are painted directly over the original paintings.4 These ancient depictions of the Pandara and Mahagovinda Jatakas can still be made out with difficulty where portions of the later compositions have fallen away, but the later “intruders” were more interested in appropriating these good positions than in preserving the old narratives, particularly since the latter were aniconic. Thus, in the eyes of the new donors, these damaged old stories, which by these later times may not even have been understood, would have been accorded a relatively low priority compared with their own iconic intrusions, from which merit would be gained. In any case, certainly they no longer cared about them, for the Period of Disruption was not a time to be entertained, even if instructively. Their overweening interest was in depicting Buddha imagery, and multiple depictions of the Buddha seemed to be particularly favored. M. Zin has identified the present almost illegible composition inside, at the right of the doorway, as a Sravasti Miracle composition, with multiple emanations of the Buddha connected with branching lotuses.5 It is inscribed: “This is [the religious donation] of the Sakyabhiksu Sanghapriya in honor of [his] mother and father.”6 Other groups of intrusive Buddhas, both standing and seated, were painted over other early paintings on both the left and right aisle walls. Understandably, they are positioned toward the front where they were more accessible and the light was stronger. However, farther down the aisles these intrusions stop, probably not so much because the light was poorer at these more distant points in the aisles, but because time ran out. Work on these aisle wall intrusions, perhaps started in early 480, would have been abandoned (along with all other intrusive donations at the site) at the end of 480.
3 The aisles would have been better illuminated originally. The two small windows, like the cave’s “sun window” have been closed in modern times. 4 See Schlingloff and Zin. 5 Zin 2003, 21 b. 6 Cohen 1995, 338, Inscription 22.
256
volume iib, part ii, chapter twenty
The composition near the front of the left wall shows six standing Buddhas, possibly representing the six Buddhas of the past, a grouping common in very late contexts at the site.7 The composition near the front of the right wall shows, possibly by chance but presumably by convention, the familiar seven seated Buddhas, three above and four below. Both groups are painted directly over the early Hinayana narratives; where the surfaces are broken, the old lime plaster shows through. Happily, the old Hinayana paintings in the more rearward stretches of Cave 9’s aisles escaped being covered over in the fifth century phase, if for no other reason than that time ran out. Thus one can still see episodes from the Sasa, Kunala, and Udaya Jatakas on the left, and the stories of the nagas Elapatra and Pandaraka on the right, in both case somewhat rearward of the later intrusions. Although still able to be identified, they are now practically illegible due to the passage of time.8
7 See Cave 4 shrine antechamber, Cave Upper 6 shrine antechamber, Cave 7 shrine. Here in Cave 9 the focal image appears to be the fourth from the left; but this is so subtle that it may not have any particular meaning. 8 Useful drawings and commentaries on the old paintings are to be found in Schlingloff 1999, 22; and Zin 2003, 21. I find their publications indispensable.
CHAPTER TWENTY-ONE
CAVE 9: PALIMPSESTS AND OTHER TRANSFORMATIONS It may seem surprising that the paintings of the fifth century often quite heedlessly encroach upon the Hinayana cycles, as if those old compositions—perhaps now not even fully understood—had become irrelevant remnants of the past. Of course the Hinayana narratives contain no Buddha images, which were by definition sacrosanct, but it should also be realized that by the fifth century they would generally be very hard—sometimes almost impossible—to see, for those in the ambulatory are almost totally obscured by the sooty grime which collected in the years prior to the fifth century. By contrast, the Vakataka period intrusions are hardly begrimed at all; and in fact this is no more than what we would expect, since they were painted in the Period of Disruption, so close to the time of the site’s abandonment. It is intriguing that the new donors of the Period of Disruption, who so willingly covered over the ancient Hinayana scenes, never did this with the narrative scenes painted in the caves during the site’s heyday. The few rare instances where they trespassed at all on the donations of the established patrons during the site’s main phase never involve the murals, which were always left intact, even though the plastered wall surfaces planned for the painting programs which never got done were very often sensibly taken over. There are remarkably few actions by intrusive donors which might be classified as “destructive” with regard to work done by the established patrons of the Vakataka phase. The most surprising, perhaps, involved the cutting away of a few of the leaves of the Dying Buddha’s sal tree in Cave 26 to make more room for the crowded intrusive panels cut in that area a year or two later. In a somewhat similar instance, some of the stems and leaves of the lotus supports in the Sravasti Miracle scene (itself intrusive) on the rear left wall of Cave 2 have been covered over by another intrusive donor, apparently desirous of getting his offering as close to the shrine as possible. The base of the front left pilaster in Cave 17, which must earlier have
258
volume iib, part ii, chapter twenty-one
had a simple decorative design, appears to have been covered by an intrusive bhadrasana Buddha.1 In the same way, the bases of the demi-pillars fronting Cave 19, which would have been painted with simple designs in about 470, also were grounds for intrusive carved Buddha images sometime after mid-479; but of course, considering their very exposed position, their original paintings had probably already been weathered away by that time. A similarly “harmless” imposition probably characterized the placement of the carved intrusive Avalokitesvara on the rear wall of the Cave Upper 6 porch, where it would seem that this slightly later image may have cut away some of the low-priority portions of the complex painted Buddha groups in the same general area. Smaller painted and/or sculptured Buddhas were also added around the large Six Buddhas in Cave 7 (and in one instance on the right wall of the antechamber in Cave Upper 6) but they in no way destroy the original figures, although they may cover or cut into the edge of the robe. The only “exceptions” to this rule respecting the integrity of previous Buddha (or Avalokitesvara images) involve main shrine images which were never properly completed, in the eyes of the patron or his associates, and so were refurbished at a later date. But, like the improvements to monumental stupas sometimes effected even today in various parts of Asia, this was in no sense a destruction of the original image; instead it probably was thought of as a way of increasing its power, its sanctity, and its contemporaneity. The main image in Cave 7 is the most startling example of such a transformation, but Cave 15 also presents a striking example. Then, of course, there are other images which were planned in one way, but finished in quite another, as when (as in Cave 7 and probably in Cave Lower 6 also) plans for shrine stupas yielded to the new compulsion for Buddha images. Remarkably, in Cave 11, an abandoned stupa remains. The truth is, the majority of shrine images at Ajanta usually did not turn out at all as originally intended. We have noted elsewhere that prior to 468 the new shrines being planned were probably intended for stupas, and that the consequent idea of cutting the shrine out with a central block, appropriate for a stupa, caused significant problems when Buddha images had to be fitted in to the same outmoded plan. 1 Leela Wood, who has concentrated upon a study of Cave 17, agrees that this anomalous pilaster painting is a later intrusion.
CHAPTER TWENTY-TWO
CAVE 9: INTRUSIONS ON PILLARS
Time obviously ran out for the donors making the many intrusions in Cave 9, eighteen of which have extant or previously recorded inscriptions. We can assume this because by and large these intrusive iconic paintings—by and large Buddha images—appear only on the relatively visible faces; a few pillars at the very rear of the cave, hidden by the stupa, have no intrusions at all on their faces. These later Buddha figures, being intrusions, are by no means “programmed” like those added as part of the redecoration program seen on Pillars 4 and 5 here, and on a greater number in Cave 10. Now, in the Period of Disruption, it appears to be “every donor for himself ”, resulting in a disorganized array. One can sense the presence of a number of different hands at work in the cave, some quite accomplished. Occasionally, as on the major faces (A and B) of Pillar L6, it is clear that the same artist did adjacent images. But these do not fit the program started on the fourth and fifth pairs of pillars. In fact, their large green mandorlas, with spiky white rays, are found most particular in images painted in the Period of Disruption, not only in Cave 9, but in Cave 10, and in many other intrusive contexts throughout the site. Although most of the intrusive images are single Buddha figures, in a few instances a central Buddha figure is flanked by two attendant Buddhas on the adjacent pillar faces, a formulation which, at least in shrinelet images, never occurred prior to 477.1 Furthermore, there is a single representation of a stupa on Pillar L8, and three figures of (presumably) Avalokitesvara, two on the beam-like extension
1
See the right front and right rear shrinelets of Cave Upper 6. These are anticipated, and may have been suggested, by the shrine arrangement in Cave 7, where two of the six “Buddhas of the past” are so close to the main image that they could be thought of as part of the main image group, although the conjunction probably was “accidental”. Curiously, Buddhas flanking Buddhas are found in painting (Cave 19 right wall) a decade earlier.
260
volume iib, part ii, chapter twenty-two
of Pillar R1, and one on Pillar L4 Face G; the latter has a fragmentary donative inscription by a monk (Inscr 32).2 Cohen (1995, 342) considers these three bodhisattva images, along with the one in Cave 10 (Pillar R10 Face D), to be “possibly Maitreya”. However, the attributes he describes—water bottle, varada mudra in right hand, antelope skin over left shoulder, and hair in jatamukuta—are all equally (and even more?) characteristic of Avalokitesvara.3 Admittedly, there is no representation of a seated Amitabha (the Buddha from which Avalokitesvara emanates) in his diadem, but this detail, although seemingly important, is often missing from Avalokitesvara figures at the site, just as the stupa (generally considered characteristic of Maitreya) is equally missing here. Bodhisattva iconography is always in a fascinating flux at Ajanta, even to the degree that the normal positions of Avalokitesvara (representing karuna or priestly virtues on the proper right), and Vajrapani (representing prajna or kingly virtues on the proper left) are not infrequently reversed.4 What better argues the case for these intrusions being Avalokitesvara is the fact that he is very commonly depicted as a separate votive icon in the Period of Disruption and often earlier as well, and that the preponderance of these images are clearly references to his role as the Protector of Travelers.5 Indeed, even those few sculptural representations where some or all of the dangers of the road are not carved at the sides, they may have been once painted in; or per-
2
For stupas at Pitalkhora, see Volume I, Chapter 15. Cohen 1995, 199, refers (riskily) to Kushan sculptures for precedents, following Mitterwallner 1986, 118, who declares that Kushana sculptors gave Avalokitesvara “the hair-do of an ascetic in the form of straight hair-strands and to Maitreya long curly locks”. “Based upon this morphology” Cohen declares, “we can say that icons of both bodhisattvas were likely donated at the site.” He then suggests the identification of the bodhisattva painted on Cave 9 Pillar L4G as Maitreya, on this basis. 4 I take the karuna/prajna concept from John Huntington. For discussion of problems of bodhisattva identification, see Cohen 1995, 199a. 5 Intrusive images of Avalokitesvara appear in Cave 4 porch (carved), Cave Upper 6 porch (carved) and left porch cell and interior left wall (painted), Cave 9 rear wall and on pillar L4, face G and two on extension of R1, Cave 10 Pillar R10 face D (all painted), Cave 26 right façade (two panels) and at center of panel under vault (all three carved). Cave 26 hall (carved in frieze at left, near front) There are earlier (non-intrusive) separate figures of Avalokitesvara in Cave 2 shrine (ptd, 478), Cave 17 porch (ptd, c. 470), Cave 20 porch (carved, c. 469), Cave 26 int. left frieze (carved c. 477). 3
cave 9: intrusions on pillars
261
haps were merely omitted to save time and money.6 We can well understand the enhanced importance of Avalokitesvara, in his role as the protector from the astabhaya (eight dreads), during this period when, willingly or not, many devotees (surely including some of the numerous body of monks at the site) had to brave the dangers of the unknown as their stable world was in the process of collapsing.
6 In the small panels in Cave 20 (behind right porch pilaster capital), and on the left “triforium” of Cave 26, the number of side figures is much reduced. In the porch (left rear wall) of Cave Upper 6, they are omitted entirely, but may have been painted in. Generally they are included in the many large painted representations, which did not suffer from the restraints imposed by the pillar shaft format.
CHAPTER TWENTY-THREE
CAVE 9: FAÇADE INTRUSIONS
Cave 9’s façade was surely plastered and painted in Hinayana times but, even if not long since washed or worn away, the paintings would have involved decorative designs which, being non-iconic, would have been seen as readily expendable by donors of the Period of Disruption. It is significant to note that of a total of nineteen extant intrusions (both Buddhas and stupa-reliefs) on Cave 9’s façade, no less than five were either unfinished, or abandoned, when still in the sketch stage, and then re-cut. The unfinished images were almost certainly abandoned, like so many other suddenly abandoned intrusions elsewhere at the site, when work stopped at the end of 480. The images which were re-cut were of course abandoned somewhat earlier, but the very fact that they were left in mid-course (before their space was appropriated by another donor) hints at the anxieties which surely characterized these last difficult years at the site. The apparent fervor with which these already “used” spots were appropriated by subsequent donors suggests the desirability, and the sanctity, of this ancient hall, which like the other caitya halls at the site (10, 19, 26) was eagerly taken over by a wide variety of donors in the Period of Disruption, even though it had been given relatively little heed in the years before. The splendid large (9’) Buddhas across from each other on either side of Cave 9’s façade can be counted, like Gunakara’s paired standing images on Cave 26, among the first intrusions. Although, at this time (in the Period of Disruption) it is unlikely that symmetry in itself was of much or any concern, the size and visibility of the two available spaces made them so desirable that they would not have been “up for grabs” for very long. The fact that not only one, but two, such spaces were appropriated for these seemingly “paired” Buddha images even further confirms the priority of these two figures; they may well have been given by the same donor—surely using the same master sculptor—as was probably also the case with the huge intrusive Buddhas flanking the façade of Cave 26. Furthermore, the fact that these splendid images were carefully
cave 9: façade intrusions
263
finished also recommends their relatively early date. Each figure, with the right hand lowered in the expected gesture of offering, towers hugely over a small devotee kneeling below, while similarly dwarfed flying celestials converge over the colossus from above. Like Gunakara’s figures, there are smoothed panels beneath them which almost certainly once held painted inscriptions identifying the donor or donors. These presumed inscriptions, like any painted intrusive images which might have been put on the very exposed surface, have long weathered away; but there are a number of other carved intrusions, both Buddha images and stupas, fitted into available areas. The intrusive images which fill up most of the remaining usable space on Cave 9’s façade are more haphazardly placed, with no manifest interest in design symmetry. Two fine standing Buddhas appear at the left of the arch, while two padmasana images appear at the right. The latter are framed by the arched panels first introduced for padmasana images in about 477, on Cave 26’s triforium; however, during the Period of Disruption, the decorative carving of such arches was generally given up, in favor of more expedient painting.1 No paint is now visible on these two images, whose positioning suggests they were among the last intrusions. However, the fact that a pair of holes for garland-hooks were cut above them suggests that they were probably completed and put into worship; at the same time we must remember that such hooks were normally put in place before images were plastered and painted, so the presence of such holes is no clear proof of usage. Another such image was barely begun above the two just mentioned, while another was merely sketched out high up on the right façade frame; it was probably not finished because time ran out at the end of 480. A very small standing Buddha just below the colossal standing image on the right has also been left unfinished. The large triple-umbrellaed stupa relief on the right frame and a rather similar one on the left frame are impressive versions of a motif that was very common in the Period of Disruption, as we have seen. Both show, on their drums, a characteristic fifth century “Vakataka” ornamental band, with alternating oval and rectangular
1 One of the “Six Buddhas” in the Cave 24 court shrinelet (first at right; unfinished) shows a red sketch of an arched framework above; but this may only be to define its outlines, not to decorate it with carvings.
264
volume iib, part ii, chapter twenty-three
“gems”. The large size and relatively “high-priority” positioning of the stupas suggests that they were among the first of the intrusions on the façade—we would date them to 479 or possibly even late 478, by which time the old “Vakataka” patrons would have already fled from the site. Another stupa was barely sketched out at the lowest level of the right façade frame. However, it was abandoned when it was far too incomplete to have been dedicated, and thus was ritually “dead”. Consequently, taking advantage of this very accessible and very visible location, another donor ordered a standing Buddha under a makara arch to be superimposed upon it, probably in 480; once it was plastered and painted, the abandoned stupa would not be seen. We have seen two other still-visible examples of such revisions— one on Cave U6’s left rear wall, where a series of bhadrasana images covers sketches of an abandoned group of padmasana Buddhas, and one in Cave 26’s right ambulatory between panel R4 and R5, where a standing inscribed Buddha is cut over the sketch of a seated Buddha. One can understand why this could easily happen in the troubled Period of Disruption, when planning was rather helter-skelter and when so many different individual (and individualistic) donors were involved. We can easily understand how sickness or death or lack of funds or the departure of workmen—all of the many problems that were clearly afflicting the site in this Period of Disruption must have taken their toll on many other intended donations too, probably including some that can no longer be seen, having been more fully obliterated by later reworking than the three examples we have noted. What appears to be a related situation is to be seen on the right façade respond where two other still-rough stupas (perhaps started by the same ill-fated donor?) seem to have been abandoned, perhaps at the same time as the adjacent cut-away stupa on the right frame. The conveniently located lower one, which was the more unfinished of the two, was then quite heedlessly recut in order that a seated Buddha (above) and a standing Buddha (below) could be carved over the now functionless (ritually dead) stupa. It seems reasonable to date the new images—along with the abandonment of the stupa—to that same troubled year. The effect is very disruptive indeed, particularly since the new images have no essential connection with the abandoned stupa; they in no sense emerge from it, as if in emulation of the great stupas in Caves 19 and 26, but instead
cave 9: façade intrusions
265
are imposed upon it. The upper Buddha sits anomalously on the harmika, the unfinished state of which allowed its re-cutting, while the lower (standing) Buddha is expediently positioned slightly off-center on the rough unfinished stupa drum instead of being placed symmetrically on it. The very presence of this added off-center image would be enough to prove that the re-used stupa was ritually “dead”— that it had been abandoned—making it possible for the subsequent devotee (or devotees) to use the space for their own purposes. Still higher up, a relatively large (and damaged) standing Buddha, attended by bodhisattvas, is of particular interest because of the plummeting pose of the flying dwarf at the left; it would seem to be a very late (late 480) and rather novel motif, a dating which its very inconveniently high placement would support. Like the smaller padmasana image nearby, it has the lotus pedestal so common in very late images, as well as characteristically late arched (but plain) frames. Other images must also have been cut on the now-fallen parts of the façade responds; indeed, a few traces can be seen along the broken outer edge, at the right. It should be noted that the standing Buddha on the lower left façade frame and the partly revealed stupa that it fronts, are both cement additions, planned as if to balance the group on the right, and perhaps to fool (often successfully) visiting scholars as well. Such a concern for symmetry, so pervasive in the site’s heyday, was not, however, a feature of the intrusive phase, even though sometimes the very nature of the spaces available often resulted in intrusions mirroring each other, as with the large pair of standing Buddha’s flanking this same cave’s façade. This recent inappropriate cement creation should be removed.
CHAPTER TWENTY-FOUR
CAVE 9: CONSIDERATIONS ABOUT USAGE
One might have thought that, with many years ahead of them before they could expect their new caitya halls to be ready for use, the inaugurators of the site’s Vakataka phase would have wished to redecorate the tattered Hinayana caitya halls almost as soon as the site’s renaissance began in the 460s, to prepare and then utilize them more appropriately for worship. However, the patrons at the burgeoning site were apparently more concerned, in those early days, with making something new than they were in refurbishing something old and probably far less meritorious. Thus, as it turned out, the refurbishing of Caves 9, 10 and 12 was not undertaken until roughly the very year in which Harisena died. It seems reasonable to assume that, even in their tattered state, the old caitya halls 9 and 10 would have been used as centers for worship during those early years when the monks at the site would have had no other place nearly as appropriate to conduct their rituals. The heavy begriming of those old caves stands as witness to this. By the same token, it seems reasonable to assume that the many available cells in the three old Hinayana viharas were used for residence in the early years of the new Vakataka patronage when the excavations at the site were still barely underway. Even so, it would seem that the authorities at the site were relatively unconcerned about these old caves and their usage, to say nothing of their obviously desirable redecoration. One rather surprising (or at least suggestive) confirmation of the lack of concern for the old caves is to be seen in the relationship— or lack of it—between the door fittings in the Hinayana viharas and those in the caves made during the site’s heyday. In the site’s Vakataka caves, the first cells to have been excavated have doorways which have no provision for fittings whatsoever. These initial doorways— which I have called A mode—were nothing more than “mere holes in the wall”, cut before the planners had taken cognizance of the problems which might confront them, or later workers or planners, when some kind of closure had to be effected.
cave 9: considerations about usage
267
Presumably the early excavators assumed that when the doors were fitted they would either ”hang themselves” or would be held by wooden frames set into notches cut into the doorframes for the purpose of fitting them. This de post facto approach had been taken at various Hinayana sites elsewhere in western India, and can still be seen in some of the doorways of the ruinous Cave 8, the only cave at the site where some of the doorways appear to in fact have been fitted out before 468. But by 468 the excavators had “discovered” a far more practical way to hang the cell doors, a device which immediately made the “useless” A-mode doorways go completely and unapologetically out of style. The “revolutionary” new form that replaced it is what I call the B mode, wherein monolithic projections, made from stone carefully reserved by the excavators, secure the upper door pivots. This new mode seems to have struck the planners of the site as a compelling new idea, for they accepted it without exception for their work in progress and immediately started converting the old A mode doorways to this more functional new type, even if these “conversions” (distinguished as the A+ mode) were necessarily created by somewhat shaky wooden attachments. Then, by 470, an improved variant of the B-mode, called the C-mode, came into fashion. It merely added a related projection at the base of the doorway, probably to avoid the possibility of the cell door’s scraping on the floor when it was opened or closed. However, remarkably, one or both of these “revolutionary” modes (B and C) had been used five centuries before in the Hinayana viharas! Of course when the Vakataka were started, the doors in the old Hinayana cells were probably no longer either present or functioning, but nonetheless the fittings were there for all to see, if anyone had taken the time and the trouble to check the matter out. Although the plans of Caves 9 and 10 had a strong influence on Caves 19 and 26, just as the plan of simple Cave 12 provides the model for most of the earliest Vakataka viharas, the fact that the later architects did not appropriate the Hinayana viharas’ fitting modes for their cell doorways until 468, when about six years had passed, suggests exactly what the schedule of redecoration in the Hinayana caves also seems to reveal: that little attention was being paid to the past by the later donors and planners once the vigorous Vakataka patronage was well underway. It seems reasonable to suppose that, from the start, monks took
268
volume iib, part ii, chapter twenty-four
up their abode in these old caves, but there is no evidence that the cell doors (long since lost) were re-hung for their convenience; in fact this was not even done when the caves’ redecoration was started in 477, so many years later. We know that this was the case, because if doorways were refurbished in 477, their ancient and out-of-date B-mode and C-mode fittings would have been converted to the more convenient and at the same time very simple D-mode. Such conversions were not only common but conventional throughout the site after the Hiatus; but they are not found here. We might assume that once such redecoration was completed in the viharas (assuming that Caves 13 and 15A, where all traces of decoration are lost, would have followed the lead of Cave 12) planners and/or patrons would have next turned to improving the doorfittings. However, since time ran out for such consistent redecoration programs all too quickly, with Harisena’s death, this may explain why the changes were not made. The old doors had probably been lost centuries before the beginning of the Vakataka renaissance, and if the viharas were used for residence at all during this later phase, they were certain door-less. In the same way, certain later (Hindu) residents, living at the site after the Vakataka downfall, did not trouble to affix doors to the cells which they took over. In Cave 26LW’s Cell R2A, which some sadhu must have taken over for use in later times, not only is the (later) plaster—apparently largely composed of cowdung and smeared on very thinly—not of any known Vakataka type, but it also has been carelessly spread up into the upper pivot hole, showing that the hole was no longer in use. Similarly, in Cave 11’s PLA a similarly anomalous type of plaster is smeared over one of the peg-holes which once held the applied projection characteristic of A+ mode doorways. Obviously the applied (wooden?) pivot holder had broken away by this time, and had probably been taken to one of the nearby villagers for burning or to salvage the wood for other uses. Only the two teak pegs (often broken off but still in place), or the holes into which they were set, here, as in all of the site’s many converted A+ mode doorways, remain to show that the applied upper projections once were there. As noted above, all of the Hinayana caves have either B mode (with upper projection alone) or C mode (with dual projections) doorways. Judging from the evidence in Caves 13 and 15A (Cave 12’s floor level being totally reconstructed in cement) the C mode was the fitting of choice. Some cells—in the “proper” C mode—have
cave 9: considerations about usage
269
monolithic projections above and below, but sometimes (perhaps to correct a poor fit) there is a large and deep square hole, which once would have held a wooden insert in which a hole was cut to receive the lower pivot.1 Such wooden fittings had the advantage of not only being easily replaceable but would also facilitate a smoother pivoting. In such cases it is impossible to tell whether the lower pivot holder, if monolithic, would have been of the B or C type; in the Vakataka caves the distinction bears directly upon the dating of the doorway, but the point is hardly of concern here. Similar wooden fixtures were often set into Vakataka period cell doorways, but they were invariably smaller than these old Hinayana ones.2 In a few cases the upper pivot holes have been repositioned, the newly cut one being a bit smaller (there being only limited room for such a revision), and set nearer to the edge of the door. This was surely because repositioning was required, the original holes being incorrectly aligned; if there is any sign of use, it always appears in the second hole cut. In one instance (Cave 13 Cell L2) a second hole was cut in the block below, as well as in the upper projection, and both of the holes nearest the doorway show the only signs of wear; perhaps the whole door (having already been constructed) had to be moved over when the time for door hanging came.3 The need to set wooden pivot blocks at the bottom of the doorways may equally suggest inexperience; for they may well have been set in to correct a misplaced hole below. In this same regard, we should note that in a significant number of cases, the upper monolith was originally cut too close to the top of the door; when necessary, it is generally very roughly cut back, as if done by less committed later workers, when the time came to actually hang the doors. The evidence of wear in the cell doorways of all these Hinayana viharas is surprisingly sporadic. For instance, of the twelve upper 1 Two such holes for now missing wooden inserts, instead of stone projections at the base, are found among the seven cell doorways in Cave 13; in Cave 15A all three doorways had such holes for inserts, although two have been filled with cement. 2 A few similarly large ones are sometimes found in the larger doorways of shrines; notably Cave 11, where one still remains! All of those in the Hinayana cells are gone. 3 Here the holes in the lower projection are cut square, almost surely to hold a wooden “sleeve” for the pivot, as in many Vakataka examples as well, probably to facilitate turning and reduce the greater wear which could be caused by direct contact with the stone.
270
volume iib, part ii, chapter twenty-four
projections in Cave 12, only three (R2, 3, and Ghanamadada’s cell in the rear wall at the right) show significant wear. The others, unless one is somewhat imaginative in fingering the holes to test their wear, reveal no use at all. Cave 13 presents a similar situation, with three of the seven upper pivot holes revealing usage. Only in the small Cave 15A, with its three cells, do all of its doorways show usage. It seems likely that many of these doorways, clumsily cut in any case, had never worked very well, or in some cases might not have even been fitted out, in Hinayana times. Certainly they were not in use when the Vakataka caves were started; and it is evident that there was never any attempt to put them into use even late in the site’s history, when an interest in refurbishing the Hinayana caves finally emerged about 477. This is certain, because by that time, as was done throughout the site, the old projections would have been disregarded, and they would have all been re-cut in the far more practical D mode. But this was never done. The neat round holes that the Hinayana workers cut into the opposite walls to hold clothes-poles are distinctly different from the generally larger and less precise fifth century holes. Furthermore, single holes—presumably for hooks or pegs—are very rare in the Hinayana viharas, and there is no evidence that any new ones were added in Vakataka times. Nor were paired (and opposed) holes for shelves, so often found in Vakataka caves, ever added to these Hinayana residences, as one would have expected them to be, if the caves were being later refurbished. Thus we must assume that if monks took up residence here in the Vakataka period, “improvements” (such as doors!) were never added, either by the monks themselves, or by interested donors. Furthermore, it is both surprising, and suggestive, that the cells in the Hinayana viharas were never plastered during (or even before) about 479, when this was commonly done in many caves at the site. It is not clear just why the plastering of the cells took place in the frenzied Period of Disruption—was it some kind of offering to the resident monks?—but the fact that the activity did not extend to these refurbished Hinayana caves emphasizes the low priority accorded them.4 Indeed, with no clear evidence that the doors in the Hinayana
4 Admittedly, mud plaster would not adhere well to the very smooth Hinayana cell walls; but lime plaster could have been used.
cave 9: considerations about usage
271
cells were ever re-hung, or other useful fittings added, or the cells ever plastered, one might well wonder if monks actually did live in these old caves in Vakataka times. Admittedly, it is hard to believe that, being so available and even so necessary when the Vakataka viharas were merely in the first stages of excavation, the old viharas were not put into use for residence purposes. However, appearances can be deceptive; even if monks had indeed taken up residence in them, starting in the early 460s, this does not mean that any patron necessarily desired to put money into improving such old and obsolete quarters—at least until the decision was made to redecorate some or all of them late in Harisena’s reign. It seems possible that the initial resistance to refurbishing the ancient viharas and caityas had something to do with the merit that might or might not be made from such work, as opposed to that which could be acquired from fresh new undertakings. Of course there is also the matter of money and workmen, both at a premium in the Vakataka’s inaugural phase, when nearly every cave at the site was started within the first five years. This understandable focus on the exciting modern developments at the site, combined with the low-priority status of the Hinayana excavations, surely left these old caves unattended (even if used) for something like fifteen years. And of course we must also remember, speaking of lack of attention by patrons, how for well over three hundred years, before Ajanta was started, nearly all of the hundreds of ancient caves in Maharashtra survived without any evident additions or changes whatsoever.
APPENDIX
AJANTA’S INSCRIPTIONS Richard S. Cohen The University of California, San Diego
[email protected]
Abbreviations AJ2:
John Allen. “A Note on the Inscriptions of Cave II.” Appendix to G. Yazdani. Ajanta. vol. 2. London: Oxford University Press, 1933. AJ3: N. P. Chakravarti. “A Note on the Painted Inscriptions in Caves VI–XVII.” Appendix to G. Yazdani. Ajanta. vol. 3. London: Oxford University Press, 1946. AJ4: N. P. Chakravarti & B. Ch. Chhabra. “Notes on the Painted and Incised Inscriptions of Caves XX–XXVI.” Appendix to G. Yazdani. Ajanta. vol. 4. London: Oxford University Press, 1955. BCTTI: James Burgess. Report on the Buddhist Cave Temples and their Inscriptions. London: Trübner & Co., 1883. CII: Vasudev Vishnu Mirashi. The Inscriptions of the Vàkà†akas. Corpus Inscriptionum Indicarum, vol. V. Oocatamund: Government Epigraphist for India, 1963. COHEN: Richard S. Cohen. “Appendix A: Ajanta’s Inscriptions.” Setting the Three Jewels: The Complex Culture of Buddhism at the Ajanta Caves. Ph.D. Dissertation, University of Michigan, Ann Arbor, 1995. DAJI: Bhau Daji. “Ajunta Inscriptions,” Journal of the Bombay Branch of the Royal Asiatic Society. 7 (1863): 53–74. EI33: Dinesh Chandra Sircar. “Inscription in Cave IV at Aja»†à,” Epigraphia Indica. 33 (1959–60): 259–62. EI37: A. Ghosh. “Two Early Brahmi Records from Aja»†à,” Epigraphia Indica. 37 (1967): 241–44. FAI: James Prinsep. “Facsimiles of Various Ancient Inscriptions,” Journal of the Asiatic Society of Bengal. 5 (1836): 348–9, 556–561, plate ix, #4, plate xxviii, #9, #10, #11. GCI: Vasudev Vishnu Mirashi and P. Sreenivasachar. The Ghatotkaca Cave Inscription. Hyderabad: The Archaeological Department, Government of Hyderabad, 1952. ICTWI: James Burgess and Bhagwanlal Indraji. Inscriptions from the Cave Temples of Western India with Descriptive Notes, &c. Bombay: Government Central Press, 1881. KERN: Hendrik Kern. The Jàtakamàlà: Stories of Buddha’s Former Incarnations, Otherwise Entitled Bodhisattva-avadàna-màlà, by Àrya-≤ùra. Harvard Oriental Series, vol. 1. Cambridge: Harvard University Press, 1891. NIA: M. K. Dhavalikar. “New Inscriptions from Aja»†à,” Ars Orientalis. 7 (1968): 147–53.
274
appendix
NOTES:
James Burgess. Notes on the Bauddha Rock-Temples of Ajanta. Bombay: Government Central Press, 1879. OBM: Peter Khoroche (translator). Once the Buddha was a Monkey: Àrya •ùra’s Jàtakamàlà. Chicago: University of Chicago Press, 1989. STUDIES: Dieter Schlingloff. Studies in the Ajanta Paintings: Identifications and Interpretations. Delhi: Ajanta Publications, 1988. Conventions [ ]
( ) { } ˙ ‘
In the Sanskrit texts these brackets mark akßaras that are damaged or unclear but can be determined with a reasonable degree of certainty. In the translations bracketed words are either based upon reconstructed text or supplied by the editor for proper meaning. Akßaras for which the text physically has adequate space but are now lost due to time and wear, reconstructed by present editor or at a previous editor’s suggestion. Supplied by editor. For the most part, these items are whole akßaras, medial vowels, or case endings omitted due to scribal error or dialectal variation, or are conventional punctuation not utilized in the epigraphs. upadhmànìya or jihvàmùlìya as appropriate. avagraha. Never present in the inscriptions, this is always supplied by the editor. Pillar Faces for Caves Nine and Ten
ajanta’s inscriptions
275
CAVE ONE
Inscription 1 Cave: One Location: Vihàra’s right wall; on a green patch directly to the right of the third cell Medium: Painted Type: Second or post-second period graffiti Editions: Not previously noticed Copies: None published Notes: Written in yellow pigment, this record uses a script with which I am not familiar. The handwriting and medium are reminiscent of graffiti in Caves 16 and 21 (inscriptions #75, #76, and #87). Inscription 2 Cave: One Location: Vihàra’s right wall, between cells 4 and 5; beneath the stool and feet of a seated Buddha in the Nàgakumàra avadàna (as identified in STUDIES: 60–2) Medium: Painted Type: Second or post-second period graffiti Editions: Not previously noticed Copies: None published Notes: The paleography and nature of this record resemble those of #1. Inscription 3 Cave: One Location: Shrine antechamber’s left wall; beneath two of Màra’s daughters, below and to the right of the seated Buddha Medium: Painted Type: Second or post-second period graffiti Editions: Not previously noticed Copies: None published Notes: It is possible that other scholars have noticed these black strokes, but judged them to be part of the narrative. The fact that they are not colored in – as leaves or insects would have been – but do present an intentional pattern suggests their epigraphic nature. Like the previous two records, this seems to be a graffito, neither descriptive nor donative in nature.
appendix
276
CAVE TWO
Inscription 4 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Left porch-end cell’s rear wall; upon a painted cane stool Painted Second period, programmatic, descriptive label ICTWI: 81, #3; AJ2: 58, #1 ICTWI: Cave II, #5; AJ2: 58 This record identifies the figure as the bodhisattva in his incarnation as Kßàntivàdin, the Preacher of Forbearance. Here one sees Kßàntivàdin preaching to a king’s wives while they listen rapt with interest. A fuller description is found in AJ2: 58–9, n. 1.
Text L1: kßàntivàdi˙ Inscription 5 Cave: Two Location: Left porch-end cell’s rear wall; on a long green band beneath the scene containing the preceding record Medium: Painted Type: Second period, programmatic, didactic verse Editions: ICTWI: 81, #4; AJ2: 59–60, #2 Copies: ICTWI: Cave II, #3; AJ2: 59 Notes: This inscription—verses 4, 15, and 19 from the Kßàntivàda Jàtaka in Àrya •ùra’s Jàtakamàlà—is now virtually unreadable. Accordingly, I have reproduced the text published in AJ2 without emendation. The translation is from OBM: 194, 196. The Text Notes reproduce Àrya •ùra’s original verses, from KERN: 182, 184. Text L1: . . . . . yatrava rànta sadgu . bhußa»à {|} tan ma . . . . nàjña na . . . . . . . . {|| 4 ||}1 agahi . nìtim avàpya mànußa . . . . . pà†u . . . stachendriyai {|} avakyam‰tyu nnaá karoti ya˙ ≤ubhaá . . . . . chà . khyaham ema daányate {||15||}2 L2: . . . yanta kusumair mahitahà . . dàguse . . . ntin(à)ghanà {|} saràsi mantabhramatessaroru . . . rvvicàva . . . kßà hi»a {||19||}3 Text Notes 1 nivasanti hi yatraiva santa˙ sadgu»abhùßa»a˙ {|} tan ma«galyaá manojñaá ca tat tìrthaá tat tapovanam {|| 4 ||} 2 agarhitàá jàtim avàpya mànußìm anùnabhàvaá pa†ubhis tathendriyai˙ {|} ava≤yam‰tyur na karoti ya˙ ≤ubhaá pramàdabhàk pratyaham eßa vañcyate {|| 15 ||}
ajanta’s inscriptions
277
3 alaákriyante kusumair mahìruhàs ta∂idgu»ais toyavilambo ghanà˙ {|} saràási mattabhramarai˙ saroruhair gu»air vi≤eßàdhigatais tu dehina˙ {|| 19 ||}
Translation 4. For, wherever a holy man of outstanding character chooses to settle, the place becomes auspicious and delightful—it becomes an object of pilgrimage, a hermitage. 15. Anyone born in the human condition—which is not to be despised—who is of sound constitution and has acute senses, who though death inevitably awaits him, is yet so feckless that he fails to do a good deed every day, must be under some misconception. 19. Trees are adorned with blossoms, low-hanging rain clouds with streaks of lightning, lakes with lotuses and their drunken bees, and human beings with virtues that have been brought to perfection. Inscription 6 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two No longer extant; formerly on the left porch-end cell’s rear wall Painted Second period, programmatic, didactic verse DAJI; ICTWI: 81, #5; AJ2: 60, #3 DAJI; ICTWI: Cave II, #4; AJ2: 60 This inscription—verse 56 from the Kßàntivàda Jàtaka in Àrya •ùra’s Jàtakamàlà—is no longer extant. Accordingly, I have reproduced the text published in AJ2 without emendation. The translation is from OBM: 202. The Text Notes reproduce Àrya •ùra’s original verse, from KERN: 190.
Text L1: nàtracchedity akßatakßànticiraá citta tasya prakßa . . . . . . {|} . . . . prìtisaányàna mahatàá nasa . . nàtra nàdevàdhà . . {||}1 Text Note 1 gàtracchede ‘py akßatakßàntidhìraá cittaá tasya prekßamà»asya sàdho˙ {|} nàsid du˙khaá prìtiyogàn n‰paá tu bhra߆aá dharmàd vìkßya saátàpam àpa {|| 56 ||}
Translation 56. Even as he silently looked on while his body was hacked to pieces, his spirit remained unbroken in its constant forbearance. And, because of his kindly disposition, he felt no sorrow. But to see the king fallen from the path of virtue caused him anguish. Inscription 7 Cave: Two Location: Left porch-end cell’s right wall Medium: Painted Type: Second period, programmatic, descriptive label
appendix
278
ICTWI: 82, #7; AJ2: 62, #6 ICTWI: Cave II, #7; AJ2: 62 This record allows for an easy identification of the story painted on this wall, now almost completely lost: the bodhisattva in his jàtaka as King Maitrìbala. According to AJ2: 62–3, the scribe “clearly writes Chai for Mai.” However, a careful examination of the record in situ attests to the scribe’s correct use of Mai.
Editions: Copies: Notes:
Text L1: maitrìbalorkaràjà˙1 Text Note 1
ICTWI: caitrivalorkìràjà; AJ2: caitrìbalorkaràja
Translation King Maitrìbala Inscription 8 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two No longer extant; formerly on the left porch-end cell’s right wall Painted Second period, programmatic, didactic verse ICTWI: 82, #8; AJ2: 62, #5 ICTWI: Cave II, #8; AJ2: 62 This inscription—verse 44 from the Maitrìbala Jàtaka in Àrya •ùra’s Jàtakamàlà—is no longer extant. Accordingly, I have reproduced the text published in AJ2 without emendation. The translation is from OBM: 54. The Text Notes reproduce Àrya •ùra’s original verse, from KERN: 50.
Text L1: . . . . . . . . . . . . nà {|} na prasehe manasvasyàtraiva du˙khàdi . hi . {||}1 Text Note 1 hriyamà»àvakà≤aá tu dànaprìtyà puna˙ puna˙ {|} na prasehe manas tasya cchedadu˙khaá vighàhitum {|| 44 ||}
Translation 44. The joy of giving constantly distracted his mind from dwelling on the pain of the sword cuts.
ajanta’s inscriptions
279
Inscription 9 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Left porch-end cell’s right wall; on the back of a cow, which is part of the Maitrìbala Jàtaka Painted Second period, programmatic, descriptive label ICTWI: 82, #6; AJ2: 61, #4. ICTWI: Cave II, #6; AJ2: 61 Schlingloff proposes that this inscription “was probably added at a later date. [It] has no connection with the subject of the picture; it is probably the work of a scribe who wanted to show off his learning by demonstrating his knowledge of the fact that the holy name of Sarasvatì can also be applied to a cow” (STUDIES: 140). Scribal motivations aside, the hand that wrote this word on the back of a cow also wrote all of the still-extant words in this cell, including the Jàtakamàlà verses. So, unless these verses are also intrusive (unlikely, given their integration with the painted narratives), “sarasvati” was written on the cow when the wall was first painted.
Text L1: sarasvati1 Text Note 1 ICTWI and AJ both have “sarasuti” and deem this a Prakrit inscription. However, the subjoined v is clearly a closed triangle, not the open-topped hook of a medial u.
Inscription 10 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Left porch-end cell’s right wall; in the center near several disembodied heads Painted Second or post-second period Not previously noticed None published This seems to have been the entire record. It may have been initials or an abbreviation since it has no sense standing on its own.
Text L1: ja ña Inscription 11 Cave: Location: Medium:
Two Vihara’s left wall, between cells L3 and L4 Painted
280 Type: Editions: Copies: Notes:
appendix Second period, intrusive, donative Not previously edited, but noticed in: AJ2: 63, #8 AJ2: 63 This record runs across the petals of the lotus-thrones for a group of ten vajraparya«kàsana Buddhas, all displaying the dharmacakra mudrà (albeit rather indifferently), and all having both shoulders covered by their outer robes. These ten Buddhas were painted by a different artist than the one responsible for a group of 100 Buddhas that fills most of the same wall. In AJ2, Allen despaired of finding any coherent sense from this record. A close examination of the inscription in situ reveals a great deal more information than is available from a photograph. For when the inscription was written, the brush left traces in painting beneath it. The contours of these impressions can often be determined through skillful manipulation of a flashlight, even when the painted letters themselves are mostly lost.
Text L1: deya(dha)rmmo ≤àkya(bhi)[kßo] . . . . . [gu]ptasya yad atra (pu»yaá tad bhava)tu (màtàpit‰m udisa) sarvasatvànàá ca ànuttarajñà(nàvàptaye) Translation This is the religious donation of the •àkyabhikßu . . . gupta. [Let the merit] therein [be in honor of his parents] and [for the attaining of] supreme knowledge by all living beings. Inscription 12 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Vihàra’s rear wall, to left of ante-chamber Painted Second period, intrusive, donative NOTES: 34; ICTWI: 80, #1; AJ2: 64, #9 ICTWI: Cave II, #1; AJ2: 64 This record runs across the petals of the lotus-throne for a Buddha in vajraparya«kàsana, his hands in dharmacakra mudrà with the pinky of the left hand being touched by the joined fingers of the right. The Buddha is flanked by two attendants. Both stand on lotuses, are bedecked with jewels, and have their heads encircled by halos.
Text L1: deya(dharmmo ‘yaá ≤àkyabhikßo)[r bhadanta budha]guptasya yad atra pu[»yaá] L2: [ta] . . . . . . . . . . . . . . . . . . . . . (sa)rrvasatvà(nàá) . . . Translation This is [the religious donation of the •àkyabhikßu] reverend Budhagupta. . . . Let the merit therein . . . all living beings.
ajanta’s inscriptions
281
Inscription 13 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two The pedestal of the left ante-chamber pillar Painted Second period, intrusive, donative Not previously edited, but noticed in: AJ2: 64, #10 AJ2: 64 This record is painted on the petals of a lotus-throne supported by a flowering stalk upon which a Buddha sits in vajraparya«kàsana, his hands in dharmacakra mudrà. The Buddha was flanked by attendants, but the one on his left is now effaced. The right attendant shoulders a chowry at his right; his left hand holds either a reliquary, book, or offering plate aloft (the object is unclear). It is worth noting that this inscription and the other records of this Cave 2 group that are sufficiently intact use udisa instead of uddi≤ya (usually found udi≤ya at Aja»†a). Likewise, this record, #14, and #63 in Cave 11 have sàkya instead of ≤àkya. Damsteegt records this variation as part of a “tendency to the development ≤ > s” (Epigraphical Hybrid Sanskrit. [Leiden: E.J. Brill, 1978]: 45). But Damsteegt’s examples show further evidence of Pràk‰tization such as assimilation of the conjunct consonants and concern for morae, whereas these Ajanta inscriptions only vary from Sanskrit (in terms of phonology) in this change of sibilant.
Text L1: (deyadha)rmmo ya(á) sàkyo-upàsi[ka] . . . [saá]1pa∂i . . . L2: màtàpit‰(m u)[disa sarvva]sa(tvànàá ca) Text Note 1
Alternatively, may be se, si, sì, or tsa
Translation This is [the religious donation] of the •àkyopàsika . . . saápa∂i . . . in honor of her parents [and] . . . all [living beings]. Inscription 14 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Antechamber, rear wall, right of shrine door Painted Second period, intrusive, donative ICTWI: 80–1, #2; AJ2: 64, #11 and AJ3: 85, n. 2 FAI: plate ix, no. 4; ICTWI Cave II, #2; AJ2: 64 This inscription, painted with bold thick strokes upon a white field, records the donation of “1000 Buddhas” adorning the four walls of the shrine antechamber. I have counted only 979 Buddhas in the group. The heads and halos of one row of Buddhas intrude upon the ample space set aside for this record, pushing it over the top edge. Perhaps this was done because the artist, realizing 1000 figures would not fit in the available space, wanted to make use of every inch. Be that as it may, because of the damage at
appendix
282
the beginning of the second line, we cannot be certain whether these are intended to represent the 1000 Buddhas of the present Bhadrakalpa, of which our •àkyamuni is the fourth and Maitreya will be the fifth. The Buddhas are all seated vajraparya«ka upon lotuses, but there is an unsystematic diversity with respect to the mudràs displayed, whether a Buddha’s right shoulder is bared or covered, and the color of the individual Buddhas’ robes (four are used: yellow, red, white, and a dark hue, probably blue). An interesting account of the significance and use of different color robes is found in the frame-story introducing Pàli Jàtaka no. 172, in which a foolish monk, who wishes to prove his scholarship, takes up the challenge: “The ‘yellow robe’ which he put on was blue as a bluebell; his outer robe was pure white. Thus clad, he entered the meeting, greeted the Elders, stepped up to the Preaching Seat . . . and sat down, ready to begin his recitation” (E.B. Cowell, [ed]. The Jàtaka, or Stories of the Buddha’s Former Births. [London: Pàli Text Society, 1981]: 2:46). Again, in the Cìvaravastu of the Mùlasarvàstivàda vinaya we find a discussion of robe colors: “aparo ‘pi bhikßur bhagavantam idam avocat | icchàmy ahaá bhadanta sarvanìlaá cìvaraá dhàrayitum | bhagavàn àha| àgàrikà hy enaá dhàrayanti | tasmàn na bhikßù»à sarvanìlaá cìvaraá dhàrayitavyam | pùrvavadyàvat sàtisàro bhavati | evaá sarvapìtaá sarvalohitamavadàtaá na kalpayaty eva |” (N. Dutt. Gilgit Manuscripts. [Delhi: Sri Satguru, 1984]: vol. 3.2, 95). To paraphrase: A monk wants to wear entirely blue robes. The Buddha objects that such is householders’ clothing, and stipulates that monks may not wear blue robes exclusively. A similar restriction holds for green, red, and white robes as well. At a later date, the color blue/black for a Buddhist monk’s robe seems to have become associated with Tantric practitioners (See W. B. Bollée. “Buddhists and Buddhism in the Earlier Literature of the •vetàmbara Jains.” In Buddhist Studies in Honour of I. B. Horner. Ed. by L Cousines, et al. [Dordrecht: D. Reidel, 1974]: 33). One more citation of note on this topic comes from the Padmapurà»a, verses 6.236.6–7: In order to destroy demons, Viß»u, in the form of Buddha, taught the false Buddhist religion, [as well as the doctrines of] the naked [ Jains] and the bluerobed [ Tantrikas?] (daityànàá nà≤anàrthàya viß»unà buddharùpi»à | bauddha≤àstram asat proktaá nagnanìlapa†àdikam || [The Padmamahàpurà»am. [Delhi: Nag Publishers, 1984]: 3: 237]). The “main” image of this group is found set in the center of the antechamber’s right wall. This Buddha, larger than all the others, is flanked by a pair of generic, regally bedecked attendants holding chowries. Unfortunately this group is too damaged to yield many details. Text L1: (de)yadharmm[o] ‘ya(á) sàkyo-ußakasya ra[m] . . (yad atra pu»yaá tad) [bhavatu] (mà)t[à]pit‰m1 u(disa sarvvasatvànàá) ca ànu{tta}ra[jñà](nàvaptaye) L2: . . . budhà sahasaá | Text Note 1
ICTWI: [mà]tàpit‰mm a . . . ca
Translation This is the religious donation of the •àkyopàsaka Ram . . . [Let the merit therein] be [in honor of his] parents and [for the attaining of] supreme knowledge [by all living beings]. . . . One Thousand Buddhas.
ajanta’s inscriptions
283
Inscription 15 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Antechamber’s rear wall, to the right of the shrine entrance; on lotuses of the first two rows of Buddhas beneath the preceding inscription Painted Second period, intrusive, donative Not edited previously, but noticed in AJ2: 64, #11 AJ2: 65 One cannot be certain whether this record also commemorates the 1000 Buddhas mentioned in the preceding inscription, or whether a zealous devotee decided to have his name painted over somebody else’s donation. The same hand wrote this inscription and #14, suggesting an alliance between, or the mutual knowledge of, these two donors. In point of fact, because the donor’s full name is lost in #14, #15, and #16, it is even possible that the same person was responsible for all three records. Multiple dedicatory records written by a single donor on a single donation are found in Cave 16, for instance.
Text L1: . . . . . . . yad atra pu»ya ta[d] bhavatu màtàpit[‰]m udisa sa[rvva]sa(tvànàá ca) L2: anutara[jñà](nà)vapta(ye) Translation . . . Let the merit therein be in honor of [his] parents and for the attaining of supreme knowledge by all [living beings]. Inscription 16 Cave: Location: Medium: Type: Editions: Copies: Notes:
Two Antechamber’s rear wall, to the right of the shrine entrance; on lotuses of the third, fourth, and fifth rows of Buddhas beneath inscription #14 Painted Second period, intrusive, donative Not edited previously, but noted in AJ2: 64, #11 AJ2: 65 As with #15, there is a question as to what this “donor” actually donated.
Text L1: . . . . . . . [bhimasya ya]d atra pu»ya tad bha[va]tu L2: (màtàpit‰m u)disa [sa]rvva(sa)[t]vànàm anuttarajñànàva L3: (ptaye) Translation . . of ?bhima. Let the merit therein be in honor [of his parents and for the attaining] of supreme knowledge by all living beings.
284
appendix
CAVE FOUR
Inscription 17 Cave: Location: Medium: Type: Editions: Copies:
Four Shrine, on the front face of the central image’s throne base towards the top Incised. The inscription was coated with a layer of lime plaster after it was cut. Second period, programmatic, donative EI33: 262 EI33
Text L1: [siddham]1 deyadharmmo ‘yaá vihàrasvàmino ‘bhayanandiskandavasuputrasya màthurasya kàrvva†eyasagotrasya2 yad atra pu»yaá L2: tad bhavatu màtàpitros tàtàmbàyà≤ càgràn≤atàyà STU3 s4 sarvvasatvànàñ cà[nu]ttarajñà»àvàptaye5 Text Notes 1 Shown by a symbol 2 EI33: kàrvva†iy[à]sagotrasya 3 There is an orthographic symbol of uncertain meaning here, see the next note for a discussion thereof. 4 EI33 reads this as càgrànva[và]yasu . . . s, and suggests restoring “suh‰dàá≤ càtmanas” in the ellipsis. EI33 was apparently thrown off by two peculiarities of this inscription. First, °tàyà is separated from s sarvva° by a little over six inches, the span of 5 or 6 akßaras. In part, this break occurs because the throne’s dharmacakra cuts across the line of text. Although the text on the left of the cakra abuts it, on the cakra’s right 5.75 inches were left unincised before the inscription started up again ssarva°. This record’s second peculiarity that might have thrown EI33 is the presence of character that looks like stu, su, ssu, ssa, or sbhu—read by EI33 as the su of su[h‰daá]—between °tàyà and the cakra’s edge. In point of fact, a similar character is found in two other inscriptions (Cave 11, #65 & Cave 16, #70). This character does not seem to have either a grammatical or syntactical function in any of these three caves. Rather, in all three instances it indicates that the inscription continues after a physcial break. In Cave 4, the break is caused by the dharmacakra; the Cave 11 record is unsure because whatever occupied the space between this continuation sign and the inscription’s recommencement is effaced; the Cave 16 use could either allude to the fact that each of the four Buddha’s dedicated by Dharmadatta is individually inscribed as well, or to the fact that Dharmadatta inscribed a second set of Buddhas adjacent to the first. 5 EI33: °àptaye t[i], which he suggests restoring to iti. The akßara read by EI33 as “ti” does not correspond paleographically to other t-màtras or medial i-s in this inscription. Perhaps this is an auspicious symbol or abbreviation. It is also to be found in Cave 11, inscription #65.
Translation This is the religious donation of the vihàrasvàmin Màthura, son of Abhayanandin and Skandhavasu, [a member of the] Kàrva†eya gotra. Let the merit therein be for the attaining of supreme knowledge by [his] mother, father, and paternal grandmother – to whom belongs the principle share – as well as by all living beings.
ajanta’s inscriptions
285
CAVE SIX, UPPER
Inscription 18 Cave: Location: Medium: Type: Editions: Copies: Notes:
Six, upper Vihara, left wall to right of the second cell door Painted Second period, intrusive, donative AJ3: 88 AJ3: Plate Ia This records the donation of an a߆amahàbhaya Avalokite≤vara. The figure that AJ3 described as a donor is actually somebody fleeing from an approaching lion or other such horror, now lost.
Text L1: deyadharmmo ‘yaá ≤àkyabhi[kß]o gu»akìrttya1 L2: sya yad atra pu»[y]aá (tad bhava)[t]u sar[v]vasatvànàá L3: m a[nuttarajñànà](vaptaye) Text Note 1
AJ3: tara»akìrtt.
Translation This is the religious donation of the •àkyabhikßu Gu»akìrtti. Let the merit therein be for the attaining of supreme knowledge by all living beings. Inscription 19 Cave: Location: Medium: Type: Editions: Copies: Notes:
Six, upper Vihara, on the front face of the front right pilaster about eight feet from the floor, beneath what seems to have been a Buddha seated pralaábapadàsana upon a lion throne. Painted Second period, intrusive, donative Not previously noticed None published Spink has speculated that this record commemorates the conversion of cell R1 into a shrine, but the placement of this record beneath a typical “intrusive” image as well as the fact that six nuns are sculpted as donor figures beneath R1’s main Buddha militate against Spink’s view.
Text L1: (deya)dharmo ya[á] ≤àkyabhik[ß]o[r ggo]vin[d]asya yad a(tra pu»yaá) . . . Translation This is the religious donation of the •àkyabhikßu Govinda. Let the [merit therein] . . .
286
appendix
CAVE SEVEN
Inscription 20 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seven Rear wall of the porch, to the right of the shrine doorway, about midway up the wall painted upon a white background Painted Second period, or later Not previously noticed Too illegible to reproduce photographically If, as is suggested in AJ3: 14–15, the scene here depicts the birth of Buddha, this record conceivably either labeled the figures or was a verse pertaining to the adjacent action. Inscription 21
Cave: Location: Medium: Type: Editions: Copies: Notes:
Seven Rear wall of the porch, to the left of the shrine doorway, beneath the preaching Buddha described in AJ3: 13–14 Painted Second period Not previously noticed Too illegible to reproduce photographically Probably a donative record, it is oxidized beyond the point of intelligibility.
ajanta’s inscriptions
287
CAVE NINE
Inscription 22 Cave: Location: Medium: Type: Editions: Copies:
Nine Front interior wall, over the right corner of the door, close to the head of a monk Painted Second period, donative, intrusive DAJI; ICTWI: 82, #9; BCTTI: 136, #1; AJ3: 88, #1 DAJI; ICTWI: Cave IX, #9; BCTTI: Plate LIX, #1; AJ3: Plate Ib
Text L1: L2: L3: L4: L5:
(deyadharmmo) ‘yaá ≤àk(yabh)i[kß]o [s sa]«ghappr(i)[yasya] màtàpi [ta](ra)[m] u[dd]i [≤ya]
Translation This is [the religious donation] of the •àkyabhikßu Sa«ghapriya in honor of [his] mother and father. Inscription 23 Cave: Location:
Medium: Type: Editions: Copies:
Nine No longer extant. Rear wall, to the left of the central painted stùpa, beneath a Buddha seated pralaábapadàsana, flanked by two richly dressed attendants in the background, and being revered by two figures in ja†a head-dress and red-and-white striped robes in the foreground. See AJ3: 20–21 for a fuller description of the scene. Painted Second period, intrusive, donative NOTES: 48 Never copied
Text L1: . . . yadharma . . . Inscription 24 Cave: Location: Medium:
Nine Rear wall, in the center of the wall, under a painted stùpa Painted
appendix
288 Type: Editions: Copies:
Second period, intrusive, donative DAJI; ICTWI: 83, #12; BCTTI: 137, #4; AJ3: 89, #8 DAJI; ICTWI: Cave IX, #12; BCTTI: Plate LIX, #4; AJ3: Plate IIe
Text L1: de(ya)ddharm[m]o1 [‘yam] bha . . . sya m . . . Text Note 1
ICTWI: deyaddharma; BCTTI: deyadahama
Translation This is the religious donation of . . . Inscription 25 Cave: Location: Medium: Type: Editions: Copies:
Nine No longer extant. On the rear wall, to the right of the central painted stùpa, beneath the image of a Buddha seated in pralaábapadàsana. See AJ3: 21 for a full description of the accompanying scene. Painted Second period, intrusive, donative NOTES: 49 Never copied
Text L1: . . . (para)mopàsaka . . . Translation . . . supreme upàsaka . . . Inscription 26 Cave: Location: Medium: Type: Editions: Copies:
Nine No longer extant. On the rear wall, to the far right side, beneath the image of a Buddha seated in pralaábapadàsana. See NOTES: 49 for a description of the accompanying scene. Painted Second period, intrusive, donative NOTES: 49 Never copied
Text L1: . . . ddha . . . saka . . .
ajanta’s inscriptions
289
Inscription 27 Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Triforium, above right aisle’s ninth pillar Painted Second period, intrusive, donative NIA: 151 NIA: figure 5 The image corresponding to this record is rather damaged. To the right, however, there is a similar scene (uninscribed) in much better repair. In this latter painting, we see a Buddha clad in red robes, right shoulder bared; seated in pralamapadàsana upon a grand lion throne, his feet rest on a large lotus, his hands make the dharmacakra mudrà. Immediately to the Buddha’s right and left are two attendants, both dressed regally and holding flywhisks in their right hands, their left hands rest on the corresponding thigh. Further out from the sitting Buddha are two more Buddhas, both stand upon lotuses, both have their right hands in varada mudrà, both are turned so as to face towards the sitting figure, both have regally dressed attendants with chowries, albeit not as fine as the main figure’s attendants. None of the attendants stands upon a lotus, is circumscribed by a halo, or bears any distinct attributes. Finally, directly above each of the three Buddhas was painted a second smaller Buddha seated vajraparya«ka upon a lotus and also flanked by attendants. In the inscribed image, several donor figures can be seen surrounding the central Buddha’s feet. To the proper right is a figure dressed in a fancy red robe, trimmed with gold, apparently a sybaritic instead of cenobitic monk (Raviprabha?). There also seems to be a male figure behind him, although this second is less clear. At the main Buddha’s proper left kneels at least one woman, and perhaps a second figure.
Text L1: L2: L3: L4:
[siddham]1 deyadharmmo ‘yam . . . ravi2 prabhasya3 [ya]d atra (pu»yaá) tad [bha] vatu màtà(p)it(r)os sarvvasattvà(nàá) ca
Text Notes 1
NIA records a siddham symbol here
2
NIA; ‘yaá [bhadanta go]pì
3
NIA: putrasya
Translation Success! This is the religious donation of . . . Raviprabha. Let the [merit] therein be for [his] mother and father and all living beings. Inscription 28 Cave: Location: Medium: Type: Editions:
Nine Axial face of the right entrance pillar Incised Indeterminate time, intrusive?, graffiti? Not previously noted
appendix
290 Copies: Notes:
None published This “shell-character” inscription is undecipherable given the current state of epigraphical knowledge. The mirror-image inscription on the left pillar is a concrete copy courtesy of the Archaeological Survey of India. Inscription 29
Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Pillar L1, face A Painted Second period, intrusive, donative DAJI; ICTWI: 83, #10; BCTTI: 136, #2; AJ3: 89, #2 DAJI; ICTWI cave IX, #10; BCTTI: Plate LIX, #2; AJ3: Plate Ic This inscription is found under a Buddha standing upon a lotus, right hand apparently in abhaya mudrà, left at waist level. The Buddha’s body visibly blazes through his robes, which are revealed only by a white lacy outline. A donor figure, dressed in white robes at the Buddha’s lower proper right, holds garlands.
Text L1: deyadharmmo (u)pàsakajasade L2: vasya1 Text Note 1
DAJI: deyadharma upàsaka casakisya
Translation This is the religious donation of the upàsaka Jasadeva. Inscription 30 Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Pillar L1, on the back of a jamb attached to face B, over a Buddha’s parasol Painted Second period, intrusive, donative AJ3: 89, #3 AJ3: Plate IIa The accompanying figure is lost.
Text L1: [de]yadharm[m]o ‘yaá ≤àkyabhikßor bhadaáta . . . . sya Translation This is the religious donation of the •àkyabhikßu reverend . . .
ajanta’s inscriptions
291
Inscription 31 Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Pillar L1, face D, on the petals of a lotus beneath a standing Buddha Painted Second period, intrusive, donative AJ3: 89, #4 None published The Buddha’s right hand is in what Herbert Härtel has called vyàv‰tta mudrà (similar to the abhaya mudrà, but here the hand is turned 3/4 outwards, not palm-to-front), symbolic of the Buddha’s addressing an audience according to Härtel. Although few of the inscribed Buddha images adopt this gesture, many of the Buddhas on the Caves 9 and 10 pillars are depicted thus. A single donor figure, a monk wearing a yellow robe, kneels at the Buddha’s proper right. Other detail are obscured or lost.
Text L1: deyadhar(mm)o ‘yaá ≤àkyabhikßo . . . Translation This is the religious donation of the •àkyabhikßu . . . Inscription 32 Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Pillar L4, face G, above a standing bodhisatva figure Painted Second period, intrusive, donative AJ3: 89, #5 AJ3: Plate IIb Possibly Maitreya, the bodhisatva’s left hand, raised to his chest, holds a water bottle, cradling it in the crook of his arm; his right hand is in varada mudrà; an antelope skin is draped over his left shoulder; his hair is a ja†àmuku†a, over which is set a tiara with a large diadem (there is no Buddha-figure in the diadem). The bodhisattva’s head is circled by a radiant green halo. The image is lost from the belt down, so we cannot see if there were donor figures. Two similar images, both lacking inscriptions, are found on the front of the pilaster attached to the first pillar on the right aisle in this cave as well as in Cave 10 on pillar R10, face D.
Text L1: deyaddharmo ‘yaá ≤àkyabhikßo bhadaátà . . .1 L2: [se]nasya Text Note 1
AJ3: bhadaáta [àcàrya?] . . .
292
appendix
Translation This is the religious donation of the •àkyabhikßu reverend . . . ?sena. Inscription 33 Cave: Location: Medium: Type: Editions: Copies:
Nine Pillar L5, face H, at ceiling level, above a standing Buddha circumscribed by a mandorla Painted Second period, intrusive, donative Not previously noticed None published
Text L1: (de)yaddharm[m]o ‘yaá ≤àkyabhikß[o] . . d[dha]rmma[sya] Translation This is the religious donation of the •àkyabhikßu . . dharma. Inscription 34 Cave: Location: Medium: Type: Editions: Copies:
Nine Pillar L8, face D, on an umbrella, probably over a standing Buddha Painted Second period, intrusive, donative AJ3: 89, #6 AJ3: Plate IIc
Text L1: deyadharmm[o] ‘yam ≤àkya(bhikßor) [à]càrya bhadaáta . . . . [ya]d atra pu»yaá tta L2: (d bha)vatu m(àtà)pitro . . . . . . . . . . . . Translation This is the religious donation of the •àkyabhikßu teacher reverend . . . . Let the merit therein be for . . . [his] mother and father . . . Inscription 35 Cave: Nine Location: Pillar L9, face H, below a Buddha standing on a lotus Medium: Painted Type: Second period, intrusive, donative Editions: DAJI; ICTWI: 83, #13; BCTTI: 137, #5; AJ3: 89, #7 Copies: DAJI; ICTWI: Cave IX, #13; BCTTI: Plate LIX, #5; AJ3: Plate IId Notes: The Buddha figure is turned 3/4 so that his right hand—in varada mudrà—
ajanta’s inscriptions
293
thrusts towards the cave’s stùpa. A monk donor image is found at the Buddha’s proper left. Although proper right is the customary spot for such figures, the Buddha’s twisting places this donor directly beneath his boon-bestowing hand. The Buddha and monk wear robes of the same yellow color. At the Buddha’s proper right (behind his back as it were) is a second kneeling figure, an unadorned layman, whose shoulders are both covered by a white robe, with his hands folded in supplication. Text L1: deyadharmm[o] ‘yaá ≤àkyabhikßo bhadaáta bhadrase(nasya) Translation This is the religious donation of the •àkyabhikßu reverend Bhadrasena. Inscription 36 Cave: Location: Medium: Type: Editions: Copies:
Nine No longer extant. Pillar L9, face H, below a Buddha standing on a lotus (also lost) Painted Second period, intrusive, donative ICTWI: 83, #11; BCTTI: 137, #3; AJ3: 90, #12 ICTWI: Cave IX, #11; BCTTI: Plate LIX, #3
Text L1: deyadharmo1 ‘yaá bhadata [dha]rmasena[sya] Text Note 1
BCTTI: °dharmmo
Translation This is the religious donation of reverend Dharmasena. Inscription 37 Cave: Location: Medium: Type: Editions: Copies:
Nine Pillar R3, face D Painted Second period, intrusive, donative AJ3: 90, #11 None published
Text L1: de(ya)dharmmo ‘yaá ≤àkyabh[i](kßo) . . .
294
appendix
Translation This is the religious donation of the •àkyabhikßu . . . Inscription 38 Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Pillar R10, face A, on a white field beneath a red-robed Buddha standing upon a lotus Painted Second period, intrusive, donative Not previously noticed None published The Buddha’s right hand in the vyàv‰tta mudrà, at whose proper right kneels an indistinct donor figure. The record is too fragmentary to hazard a reading. Inscription 39
Cave: Location: Medium: Type: Editions: Copies: Notes:
Nine Pillar R10, face H Painted Second period, intrusive, donative Not previously noticed None published The image resembles the preceding but here the kneeling donor is clearly white clad and mustachioed. Again the record is too fragmentary.
ajanta’s inscriptions
295
CAVE TEN
Inscription 40 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Left arch of the facade’s kìrtimukha Incised First period, donative NOTES: 50; DAJI; ICTWI: 67, #1; BCTTI: 116, #1 DAJI; BCTTI: Plate LVI, #1 There is nothing to link the donor of this cave directly to the royal •àtavàhana family, whose Vàsi߆hìputra Pu∂umàvi (c. 130–159 C.E.) was so prominent at Amaràvatì and Nàsik. ICTWI proposes that the paleography of this record predates that of this •àtavàhana king by two centuries. The Cave Temples of India observes that this record’s paleography is akin to those found mentioning Vàsi߆hìputra Pu∂umàvi at Nàsik. However, this monograph’s reading is so fraught with mistakes that I would hesitate to accept its dating. In Cave Temples, Burgess tells of a wall of enormous, regular bricks built underneath the inscribed arch. He suggests this brick-work may have been Vàsi߆hìputra Ka†ahàdi’s donation, and that the cave may have been of an earlier date. The donated object, a gharamukha, (literally “house-door”) lends an air of credibility to Burgess’s suggestion. But gharamukha could also be a variant of kìrtimukha, the architectural feature on which the inscription was incised.
Text L1: vàsi†hiputasa ka†a L2: hàdino gharamukha L3: dànaá1 Text Note 1 Vàsiputasa ka†ahà dito gharmukha danam ( James Fergusson and James Burgess. The Cave Temples of Western India. [London: W.H. Allen, 1880]: 293).
Translation The facade is the gift of Vàsi߆hìputra Ka†ahàdi. Inscription 41 Cave: Location: Medium: Type: Editions:
Ten Left wall, on the first horizontal cross-beam in the vault, before the first rib Incised First period, donative NIA: 149; EI37: 244, B
appendix
296 Copies: Notes:
NIA: figure 2; EI37: 245, B One troublesome aspect of this record is the word pasàdà. This may be read as the Sanskrit pràsàda, meaning “palace” or “temple,” suggesting that this donor was responsible for the majority of the work done in the cave. Or alternately, pasàdà might be the Sanskrit prasàda, meaning faith or favor. NIA understands it in the latter meaning. EI37 compares this record with one at Bhàjà, that has the words Dhamabhàgasa pasàdo incised on a wooden rib in the caitya’s vault. Because it is unlikely that either this or the Bhàjà record means to designate the entire cave as the donation (due to the presence of other contemporaneous donative records) and because both inscriptions are located on ribs, EI37 suggests that “pasàda has the restricted meaning of roof-component of a structure,” though he admits that there is no support for this definition. The fact that pasàdà may be a plural form supports EI37’s hypothesis. Further, given that the other first period donative inscriptions (#40, #42, #66) all stipulate the object donated, NIA’s translation of pasàda as “faith” is improbable.
Text L1: dhamadevasa . . . nasa1 L2: pasàdà d[à]naá pavajitasa2 Text Notes 1
NIA: [pa]-[tha]nasa; EI37: [ma] . . . nasa
2
NIA: pasàdo [dà]na pavajita
Translation The prasàda is the gift of Dharmadeva . . . for the renunciates. Inscription 42 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Left wall, below and between the second and third ribs Incised First period, donative NIA: 148; EI37: 244, A NIA: figure 1; EI37: 245, A M.N. Deshpande equated this record’s Bàhada with the the present village of Bahàl, near in the district of Jalgaon, about 50 miles west of Ajanta (“The Rock-Cut Caves of Pitalkhora in the Deccan,” Ancient India. 15 [1959]: 69). Bahàl was a node at which two trade-routes divided. Heading south from Ujjain, after crossing over the Narmadà River, caravans would go west from Bahàl to Nàsik or south to Prati߆hàna. South-going caravans would have climbed the Indyadri range near Ajanta. The importance of the Bahàl-to-Ajanta road during the Vàkàtaka period is shown by the presence of the contemporaneous Gha†otkaca and Banaoti Caves marking its progress.
Text L1: kanhakasa bàha∂asa dànam bhiti
ajanta’s inscriptions
297
Translation The wall is the gift of Kanhaka of Bàha∂a. Inscription 43 Cave: Location: Medium: Type: Editions: Copies:
Ten Left wall, under the fourth rib Painted First phase, descriptive label NOTES: 51; ICTWI: 84, #14; BCTTI: 137, #6; AJ3: 90–91, #1; STUDIES: 5 ICTWI: Cave X, #14; BCTTI: Plate LIX, #6; AJ3: Plate III; STUDIES: 344, fig. 2
Text L1: bhagavasa yat(i) puvad(e)va [h/l/p](a)n(a)[na]t[i] patisa yasa . . . 1 Text Note 1 ICTWI & BCTTI: bhagavasa yatipuvade[va] . . . tayatipatisa yasa . . .; AJ3: restored to, bhagavasa yat. puvadevasa—t.n. ti patisa yasa—; STUDIES: bhagav(aá) s. y. t. puva(á) d(e)v(e)[h](i) t. [n]. t[à] pa ti s. y. s.
Translations ICTWI: Of Bhagava (Buddha) first deva of Yatis . . . master of Yatis. STUDIES: The Exalted One . . . first [received] by the gods . . . Inscription 44 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Left wall, below and between the ninth and tenth ribs, on a white ground Painted Second period, intrusive, donative DAJI; ICTWI: 84, #15; BCTTI: 137, #7; AJ3: 91, #2 DAJI; ICTWI: Cave X, #15; BCTTI: Plate LIX, #7; AJ3: Plate IVa I have found no specific donation accompanying this record. Indeed, the remaining fragments of painted plaster both beneath and around it hold traces from Ajanta’s first phase. If, as Burgess says (ICTWI: 84, #15 & 86, #24), there were other such inscriptions on this and the opposite wall (all now lost), perhaps they severally allocated responsibility for the vault’s redecoration during the Vàkà†aka phase.
Text L1: [à]càryya . . . . . sacivasya1 L2: d[eyadha]rmmo yad atra pu»ya[á] tad bhavatu sarvvasa L3: tvànà[á] du˙khamocà[y]à
appendix
298 Text Notes 1
DAJI: sadevasya
Translation The religious donation of teacher . . . saciva. Let the merit therein be for liberating all living beings from suffering. Inscription 45 Cave: Ten Location: Right wall, under seventeenth rib upon a white background Medium: Painted Type: Second period, intrusive, donative Editions: DAJI; AJ3: 93, #13 Copies: DAJI; AJ3: VIa Notes: Several heads are visible, but not enough to make any sense of the inscription’s purport or the donation’s subject. A similarly placed record, now wholly unreadable, may be found under the tenth rib on this side. Text L1: deyadharmmo ‘yaá càkya1bhikßor buddhi[nà]gasya2 . [su] kara»∂[à]tra ba≤a3 . . . Text Notes 1
DAJI: ≤àkya; AJ3: ≤vàkya
2
DAJI: v‰ddhi (ka) sya
3
DAJI: (à) karu»atrapa˙
Translation This is the religious donation of the •àkyabhikßu Buddhinàga . . . . Inscription 46 Cave: Location: Medium: Type: Editions: Copies:
Ten No longer extant; Pillar L8 is now a concrete reconstruction Painted Second period, intrusive, donative ICTWI: 85, #19; BCTTI: 137, #10; AJ3, 94, #18 ICTWI Cave X, #19; BCTTI: Plate LIX, #10
Text L1: bhadantasudattasya Translation Reverend Sudatta’s.
ajanta’s inscriptions
299
Inscription 47 Cave: Location: Medium: Type: Editions: Copies:
Ten No longer extant; Pillar L8 is now a concrete reconstruction Painted Second period, intrusive, donative ICTWI: 85, #20; BCTTI: 137, #11; AJ3: 94, #19 ICTWI: Cave X, #19; BCTTI: Plate LIX, #11
Text L1: [bhada]ntasudattasya Translation Reverend Sudatta’s. Inscription 48 Cave: Location: Medium: Type: Editions: Copies:
Ten Pillar L9, face G, on an umbrella over a standing Buddha Painted Second period, intrusive, donative AJ3: 91, #3 None published
Text L1: deya[dharmmo ‘yaá ≤àkya]bhikßo(r) bhadaáta [àcàryya ≤à] . . . Translation This is the religious donation of the •àkyabhikßu reverend [teacher] •à . . . Inscription 49 Cave: Location: Medium: Type: Copies: Notes:
Ten Pillar L9, face A, beneath the image of a sitting Buddha Painted Second period, intrusive, donative None published This inscription was noted in ICTWI and AJ3, but is too faded to read. Inscription 50
Cave: Location: Medium: Type: Editions:
Ten Pillar L9, face B, on a white field beneath a throne upon which sits a Buddha in vajraparya«kàsana Painted Second period, intrusive, donative ICTWI: 85, #18; AJ: 91, #5
appendix
300 Copies: Notes:
ICTWI, Cave X, #18; Aj 3: Plate IVb Apparently, this inscription records a donation to be paired with that described in #49.
Text L1: tasyaiva Translation His too Inscription 51 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar L10, face G, on an umbrella over a Buddha’s head Painted Second period, intrusive, donative ICTWI: 85–6, #21; BCTTI: 137, #12; AJ3, 92. #7 ICTWI: Cave X, #21; BCTTI: Plate LIX, #12; AJ3: Plate IVc The image is mostly lost.
Text L1: deyadharm[m]o ‘yaá ≤àkyabhikßor bhadaáta saághaguptasya Translation This is the religious donation of the •àkyabhikßu reverend Sa«ghagupta. Inscription 52 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar L17, face B, beneath a Buddha standing upon a lotus styled to look like a rug Painted Second period, intrusive, donative ICTWI: 86, #24; AJ3: 92, #8 ICTWI: Cave X, #24 The Buddha’s right hand is in varada mudrà, the left upraised. It is possible that this Buddha’s antarvàsa (under robe) or nivàsa (skirt) is white, while his sa«ghà†ì (outer cloak) or uttaràsa«ga (upper garment) is red. This would parallel the image of Vipa≤yin Buddha painted on pillar R7 (#58), who wears a distinctly two-tone outfit. If the white beneath the red is not an undergarment, it might be a donor kneeling at the Buddha’s proper right.
Text L1: . . . . . . gu»o . . . . . . yà1 bhàsurad[ì]ptayas te L2: . . . . . . yanàbh . . . . . ye kàrayaátìha jinasya bimbaá L3: [deyadha]rm[m]o ‘yaá [≤]àkyabhikßo(r àcà)r[ya] bhada(á)ta (bu)ddhasenasya
ajanta’s inscriptions
301
Text Note 1
ICTWI: . . . ndriyà
Translation This is the religious donation of the •àkyabhikßu teacher reverend Buddhasena. For a translation of the verse on lines 1 and 2, see inscription #90. Inscription 53 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar L18, face A, below the figure of a yellow-robed Buddha standing upon a lotus Painted Second period, intrusive, donative AJ3: 92, #9 AJ3: Plate Va The Buddha is wholly circumscribed by a mandorla, his right hand in varada mudrà. A monk, also yellow-clad, kneels at the Buddha’s proper right.
Text L1: deyadharmmo [‘yaá] bha(daáta) L2: bu[ddha]somasya Translation This is the religious donation of reverend Buddhasoma. Inscription 54 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R2, face B, beneath a standing Buddha Painted Second period, intrusive, donative AJ3: 94, #17 None published The Buddha’s right hand is in the vyàv‰tta mudrà. A monk holding an incense-burner in his right hand kneels at the Buddha’s feet.
Text L1: [deya]ddharm[o] ‘ya[á] L2: . . . . . . màtà . . . . . L3: . . . . . . . . . . . . Translation This is the religious donation . . . mother . . .
302
appendix Inscription 55
Cave: Location: Medium: Type: Editions: Copies:
Ten No longer extant. Pillar R5 Painted Second period, intrusive, donative ICTWI: 86, #22; BCTTI: 138, #13; AJ3: 94, #21 ICTWI: Cave X, #22; BCTTI: Plate LIX, #13
Text L1: deyaddharmmo ‘yaá bhadanta ≤ìlabhadrasya màtà L2: pitaram udi(≤ya) Translation This is the religious donation of reverend •ìlabhadra in honor of [his] mother and father. Inscription 56 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R5, face B Painted Second period, intrusive, donative AJ3: 93–4, #16 AJ3: Plate VI, D The image belonging to this inscription is not unusual: a Buddha standing upon a lotus, surrounded by a green mandorla, his right hand in vyàv‰tta/varada mudrà, with the figure of a monk wearing the same yellowcolored robes as the Buddha kneeling at the master’s proper right. It is interesting to note, however, that during the Vàkà†aka period there seems to have been some administrative control over the decorative program of this cave, of which this image is a part. The general iconographic features of the axial Buddhas (faces A, B, C) were set such that there is a line of standing Buddhas wearing yellow/orange robes in a green mandorla (one per pillar); beneath each of these there is a standing Buddha wearing white robes, his head surrounded by a white halo (this halo intrudes over the background of the higher figure); and beneath these white Buddhas is a line of standing orange-mandorlaed Buddhas. The individualism of the various Buddhas across each band, the variety of donor figures, and the dedicatory inscriptions suggest that this decorative scheme was regulated within very broad restrictions, and did not mandate strict uniformity. One may contrast this patronage pattern with that indicated by the Buddha images to be found atop many of the pillars on the left side of the cave: these are quite uniform and have no inscriptions or donor figures to even hint at individuation in the source of funding. The programme of which this inscription’s image is a part is found on pillars from both the right and left arcades of this stùpa’s chamber. Though a pattern is clearly evident, it seems to have only been followed in the middle of these pillar groups; the pillars at the far ends of both lines do not conform to this pattern.
ajanta’s inscriptions
303
Text L1: deyadharmmo ‘yaá ≤à L2: kyabhikß[o] bha(da)nta dro L3: »a[va]rmmasya Translation This is the religious donation of the •àkyabhikßu Dro»avarman. Inscription 57 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R6, face H, at the feet of a Buddha standing on a lotus Painted Second period, intrusive, donative DAJI; ICTWI: 84, #16; BCTTI: 137, #8; AJ3: 93, #15 DAJI; ICTWI: Cave X, #16; BCTTI: Plate LIX, #8; AJ3: Plate VIc His right hand in varada mudrà, the Buddha wears an orange robe, as does the aged monk with a freshly shaved head kneeling in front of him. Behind the Buddha we see a second “donor” figure, clad in white, having both shoulders cloaked.
Text L1: deyadharmmo ‘yaá ≤akya L2: bhikßor bhadaáta drà∂ha1 L3: dharm[m]asya Text Note 1
DAJI: prau∂ha
Translation This is the religious donation of the •àkyabhiku reverend Drà∂hadharma. Inscription 58 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R7, face H, at the feet of a standing Buddha Painted Second period, intrusive, donative DAJI; ICTWI: 85, #17; BCTTI: 137, #9; AJ3: 93, #14 DAJI; ICTWI, Cave X, #17; BCTTI: Plate LIX, # 9; AJ3: Plate VIb Vipa≤vin’s right hand is in abhaya mudrà; at chest level, the left holds his robe, which does not cover the right shoulder. Like many other Buddhas in this cave, Vipa≤vin is circumscribed by a green mandorla from the top of which rises a three-tiered parasol, with two garland-bearers hovering nearby. Atypically, Vipasvin does not stand on a lotus. Another interesting peculiarity of this image is that Vipasvin’s undergarment—colored in bands of light and dark tan like rough-spun khadi—peeks out from beneath the
appendix
304
hem of his outer-robe. The inclusion of the word cetika, in this record has been widely read as an indication that the present donor was a member of the Cetika nikàya, a sub-sect of the Mahàsà«ghika. According to Vasumitra, this sect’s name derives from the fact that its founder lived on Caitya-hill near Amaràvatì, and does not indicate anything about the sect’s doctrinal stand ( Jiryo Masudo. “Origin and Doctrines of Early Buddhist Schools: A Translation of the Hsüan-Chwang Version of Vasumitra’s Treatise,” Asia Major. 2 [1925]: 15). In fact, Vasumitra stipulates that one of the Cetika nikàya’s characteristic tenets is, “Even if one makes offerings to a stùpa one cannot acquire great fruits” (Masuda: 38). Assuming this inscription refers to the Cetika nikàya, it is worth noting that the only other nikàya mentioned at Ajanta, the Apara≤aila of inscription #90, was also a sub-sect of the Mahàsà«ghika, also originally from the Amaràvatì region. Text L1: vi(pa)≤v[ì]1 samya[k]saábu[ddha˙] cetika.rikasya2 Text Notes 1 DAJI: vipa≤yì; ICTWI & BCTTI: vipa≤≤i 2 DAJI: cetikadarikasu; ICTWI & BCTTI: cetika[ya]rikasya; AJ: cetika[pa]rikasya; °vàrikasya is also possible.
Translation Vipa≤vin, the Complete and Perfect Buddha. Belonging to Cetika ?rika. Inscription 59 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R10, face F, near the feet of seat Buddha Painted Second period, intrusive, donative ICTWI: 86, #23; AJ3: 92, #11 ICTWI: Cave X, #23; AJ3: Plate Vd The Buddha’s hands are held in dharmacakrapravartana mudrà, with a wheel flanked by deer before him, suggesting his first sermon at Sàrnàth. There are three figures sitting in front of the Buddha; all are clad in white. One, sitting at the Buddha’s proper right, holds an incense burner; to the Buddha’s left are seated a man and woman, possibly the donor’s parent’s, their hands folded in reverence.
Text L1: L2: L3: L4:
màtàpi taram udi≤ya [sarva](satvànàá ca) [deyadha](rmmo ‘yaá ≤àkyabhikßor) bhadanta ke≤avasya
Translation In honor of [his] mother and father [and] all [living beings], this is the religious donation of [the •àkyabhikßu] reverend Ke≤ava.
ajanta’s inscriptions
305
Inscription 60 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R10, face G, near the feet of a white-clad Buddha Painted Second period, intrusive, donative AJ3: 93, #12 AJ3: Plate Vc The Buddha is seated in vajraparya«kàsana upon a lotus, his hands in dharmacakrapravartanamudrà. There are three donor figures, all dressed in white. To the Buddha’s proper right is a monk, to his left, two “parents.”
Text L1: d[e](ya)dharm[m]o ‘yaá ≤àkyabhi . . . . . . (mà)tàpit‰m udi≤ya Translation This is the religious donation [of] the •àkya[bhikßu] . . . in honor of [his] mother and father. Inscription 61 Cave: Location: Medium: Type: Editions: Copies: Notes:
Ten Pillar R12, face F, over an umbrella Painted Second period, intrusive, donative AJ3: 92, #10 AJ3: Plate Vb The remainder of the image is lost, but the paleography and background red color suggest that this image was made by the same artisan as that responsible for the inscribed image (#59) on pillar R10, face F. If so, this is interesting, for this image is placed in the same location on the pillar as R10’s, suggesting that artisans may have been given or bought blocks of space on which to work, perhaps with an eye towards maintaining visual harmony within the cave.
Text L1: deyadharmmo ‘yaá ≤àkkyabhikß[or] buddha.ma(sya)1 L2: màtàpita[ram uddi≤ya] sa[rvva]satvà L3: nà(á ca) Text Note 1
AJ suggests Buddhasomasya
Translation This is the religious donation of the •àkyabhikßu Buddha . . ., in honor of [his] mother and father [and] all living beings.
appendix
306
Inscription 62 Cave: Location: Medium: Type: Editions: Copies:
Ten Pillar R13, face B Scratched into the paint Intrusive graffiti NIA: 152 NIA: figure 7
Text L1: John Smith 28 Cavalry L2: 28 April 1819
ajanta’s inscriptions
307
CAVE ELEVEN
Inscription 63 Cave: Location: Medium: Type: Editions: Copies:
Eleven Porch, left wall, back corner, accompanying a very faded a߆amahàbhaya Avalokite≤vara scene Painted Second period, intrusive, donative Not previously noticed None published.
Text L1: deya[dha]rmmo upà(saka)sya yad atra p[u»ya]n [tad bhavatu] sarvvasa[tvas]ya . . . Translation This is the religious gift of an upàsaka. Let the merit therein be for . . . all living beings . . . Inscription 64 Cave: Location: Medium: Type: Editions: Copies:
Eleven No longer extant Painted Second period, intrusive, donative NIA: 149 NIA: figure 3
Text L1: 1deyadharmmo ‘yam upà L2: saka2 mitradharmmasya L3: yad atra pu»yam tad bhava[tu] L4: màtàpitro3 sarvvasatvànàn ca Text Notes 1
NIA sees the siddham symbol here
2
NIA: °saka[sya]
3
NIA: °pitro[˙]
Translation Success! This is the religious donation of the upàsaka Mitradharma. Let the merit therein belong to [his] mother and father and all living beings.
308
appendix Inscription 65
Cave: Location: Medium: Type: Editions: Copies:
Eleven Rear wall, between the cells, at the feet of a completely obliterated Buddha Painted Second period, intrusive, donative Note previously noticed None published
Text L1: (deya)ddhar[mmo] (yaá) . . . (màtà)[p]it‰[pù]rvva«ga(ma)[k‰]tvà (yad atra pu»yaá tad bha)va(tu) [sa]rvvasatvànà anutta[ra]jñànàvàpta[y]e STU1 . . . [tà]2py asau tya . . . [va]tàlo . . . Text Notes 1 This appears to be an auspicious symbol of the sort found in Cave 4’s inscription, #17; and Cave 16’s #70. 2 I am uncertain whether this is an akßara or a text-marker of some sort. It is isomorphic with the final character of inscription #17, which EI33 read as t[i]. However, it does not look like ti. To know its significance one will have to make better sense of the akßaras that follow.
Translation This is the religious donation . . . . [Let the merit therein be for] the attaining of supreme knowledge by all living beings, having set his mother and father at the fore. . . .
ajanta’s inscriptions
309
CAVE TWELVE
Inscription 66 Cave: Location: Medium: Type: Editions: Copies:
Twelve Rear wall, to the left of the rightmost cell Incised First period, donative DAJI; ICTWI: 68, #2; BCTTI: 116, #2 DAJI; BCTTI: Plate LVI, #2
Text L1: thànako deyadhamaá L2: ghanàmada∂asa1 va»ija2 L3: sa uvavarako3 sa-upà4 Text Notes 1
DAJI: ghanàvha° 2 DAJI & BCTTI: vanija[sa] saupà[sayo]; BCTTI: saupà[satho]
3
DAJI: sapavanvareka
4
ICTWI:
Translation A dwelling, the religious donation of the merchant Ghanàmada∂a, a cell (uvavarako = apavaraka) along with an upà . . .
310
appendix
CAVE FIFTEEN-A
I am informed by Dr. S. Vasant that a shell-script inscription was found on this cave’s front interior wall when it was first discovered. The record was effaced before an estampage could be made.
ajanta’s inscriptions
311
CAVE SIXTEEN
Inscription 67 Cave: Location: Medium: Type: Editions: Copies: Notes:
Sixteen Left wall, in front of porch, above eye level Incised Second period, programmatic, donative DAJI; ICTWI: 69–73, #3; BCTTI: 124–128, #3; CII: 103–111, #25. DAJI; FAI: plate xxviii, nos. 9 & 10 (partial); BCTTI: plate LVII; CII: plate XXV I have reedited the long, verse inscriptions in Caves Sixteen, Seventeen, and Twenty Six (#67, #77, #93) based upon a review of estampages kept at the Archaeological Survey of India’s archive in Mysore, as well as in situ examination. I am unable to provide translations here, but may in a later volume. Tentative but serviceable translations of these inscriptions may be found in COHEN. Meters: Upajàti: 1, 2, 3, 4, 10, 11, 12, 13, 14, 15, 16, 20, 22, 24, 25, 30; Upendravajra 5; Màtràsamaka (?): 6, 7, 8, 9; Aupacchandasika 17, 19, 21, 23, 27, 28; Vaá≤astha 18; Praharßi»ì 26; Vasantatilakà 29; •àrdùlavikrì∂itam 31; Màlinì 32
Text L1: udìr»»alokatrayadoßavahninirvvàpa»a1 ¯ ‘ ‘ ¯ ‘ ¯ ¯ {|} ’ ¯ ‘ ¯ ¯ ‘ (a)bhipra»amya2 pùrvvàá pravakßye kßitipà[nu]pùrvvì(m) {|| 1 ||} L2: mahàvimarddeßv abhiv‰ddha≤akti˙ kruddhas surair apy anivàryya ¯ ’ 3 {|} ’ ¯ ‘ ¯ ¯ [ra]»adàna≤akti˙ dvija˙ prakà≤o bhuvi vindhya≤a(kti˙) {|| 2 ||} L3: purandaropendrasamaprabhàva˙ svabàhuvìryyà[rjji]ta[sa](rvvaloka˙) {|} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ànàá4 babhùva vàkà†akava«≤aketu˙5 {|| 3 ||} L4: ra»e[ßu] haryyuddha6tare»[u]jàlasañcchà7ditàrkkas sa ca ¯ ‘ ¯ ’ 8 {|} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ naràtìn k‰tvàbhivàdaprava»àá≤9 cakàra {|| 4 ||} L5: [vini]rjjitàris suraràjakàryye10 cakàra pu»yeßu paraá prayatnaá {|} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ {|| 5 ||} narendra11maulivinyastama»ikira»a12lì∂hakramàmbuja˙ {|} L6: pravarasenas tasya putro ‘bhùd vikasannavendìvarekßa»a˙13 {|| 6 ||} ravimayùkha ¯ ‘ ¯ ’ ’ ’ ’ ’ ’ ¯ ‘ ¯ ¯ ‘ ¯ ‘ ’ {|} (sarvva)sena˙14 pravarasenasya jitasarvvasenas suto ‘bhavat {|| 7 ||} L7: sa(t)p[u]tra˙ pàrtthivendrasya pra≤a≤àsa dharmme»a medinìm {|} kuntalendraá15 vi[jitya] ¯ ’ ’ ’ ≤rìviádhyasena˙ 16 ‘ ¯ ‘ ’ 17 {|| 8 ||} pravarasenas tasya18 putro ‘bhùt pravaro ‘rjjitodàra≤àsana(˙)19 pravara˙20 {|} L8: ’ ’ ‘ ¯ ‘ ¯ ‘ ¯ ’ ’ ’ ’ ’ ¯ ‘ ¯ ¯ ‘ ¯ ‘ ’ {|| 9 ||} [ta]syàtmaja[˙ kàma]21 ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ {|} ’ ¯ ‘ ¯ ¯ ‘ [m a]vàpya ràjyam a߆àbdako ya˙22 pra≤a≤àsa samya[k] {|| 10 ||} L9: tasy[àtmajo] ‘bhùn naradeva ¯ ’ ’ ¯ ‘ ¯ ¯ [bhu]vi devasena˙ {|} yasyopabhogair llalitair vvih[à]rann ’ devaràjasya23 ‘ ¯ ‘ ¯ bhù[t]24 {|| 11 ||} pu»yànubhàvàt kßitipasya25 L10: [samya](k) ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ {|} ¯ ‘ ¯ yagu»àdhivàso26 ¯ ‘ ko≤o27 bhuvi hastibhoja˙ {|| 12 ||} pra28 ¯ ‘ ¯ ¯ ˙ p‰thupìnavakßàs saroruhàkßa˙ kßapi-
312
appendix
L11: (tàripakßa˙) {|} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ bàhur ddiggandhahastipratimo babhùva {|| 13 ||} hito vinìta˙ pra»ayapra[dàno] mano ‘nukùlo ‘nuvidhànavarttì29 {|} niratyayaá L12: ¯ ‘ sa»à ‘ ¯ ¯ bhu ¯ ‘ nghe»a ‘ ¯ ‘ ka≤ ca30 {|| 14 ||} ta(thai)va lokasya hità≤ayatvàt31 sukhena samya[kpari]pàlanena32 {|} piteva màteva sakheva nityaá priyo ‘bhigamya≤ ca babhùva L13: ‘ ’ 33 {|| 15 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ {|} svasthas samàve≤ya sa tatra ràjà sasañja34 bhogeßu yathe߆ace߆a˙35 {|| 16 ||} atha tasya suto babhùL14: (va) ràjà sajalàmboru[ha] ¯ ‘ ¯ ‘ ¯ ’ {|} hariràmamàrasmarendrakàntir36 hariße»(o) harivikkramapratàpa˙ {|| 17 ||} sa kuntalàvantikali«gakosalatrikù†alà†àndhraL15: ‘ ¯ ‘ jàn imàn {|} ‘ ¯ ‘ ¯ ¯ ‘ ‘ [≤aurrya]vi≤rutàn api svanirdde≤a(gu»àti) ¯ ‘ ¯ {|| 18 ||} prathito bhuvi hastibhojasùnus sacivas tasya mahìpater bbabhùva {|} sakalakßiti L16: ¯ ‘ ¯ ‘ ryyapra ‘ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ’ {|| 19 ||} ràjàpraje߆a˙ sthiradhìracetàs tyàgakßamaudàryyagu»air upeta(˙) {|} dharmme»a dharmmaprava»a≤ ≤a≤àsa de≤aá ya≤a˙puá»yagu»àá≤u37 L17: ¯ ¯ 38 {|| 20 ||} vi≤eßata˙ ‘ ¯ ‘ ¯ ‘ ¯ ¯ prati pu»yopacayaá paraá cakàra {|} yata ùrddhvam adas39sahàyadharmmà[nuca]r[i]to40 lokagurau cakàra kàràm41 {|| 21 ||} àyurvvayovittas[a]kh[à]»i42 L18: ¯ ’ ’ ¯ ‘ ¯ ¯ ‘ ‘ [cañca]làni {|} u(ddi≤ya) màtàpitaràv udàraá nyavìvi≤ad43 ve≤ma yatìndra(sevyam) {|| 22 ||} sajalàmbuda[v‰]ndalambitàgre44 bhujagendràdhyußite45 mahìdharendre {|} L19: ‘ ‘ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ¯ ‘ [v]ìra≤rìpatinà dharànikuñje46 {|| 23 ||} gavàkßaniryyùhasuvìthivedikà[su]rendrakanyàpratimàdyala«k‰tam47 {|} manoharastambha48vibha«ga L20: ¯ ‘ 49 [ni]ve≤itàbhyantaracaityamandiram50 {|| 24 ||} ’ ¯ ‘ ¯ ¯ talasannivi߆aá vi[tàna] ¯ ¯ ‘ 51 manobhiràmaá52 {|} ’ ¯ [prakà]màmbu53mahànidhànaá nàgendrave≤màdibhir a ¯ ‘ L21: ¯ ’ 54 {|| 25 ||} ¯ ¯ ¯ ‘ hita55samìra»[aá]56 [samantàt] ¯ ¯ ¯ vivi[dha]vi[làsa] ¯ ‘ ¯ ¯ 57 {|} grìßmàrkkasya ca kira»opatàpataptaá sarvartuprathitasukho58pabhogayoga[m]59 {|| 26 ||} L22: ˙ ˙ ˙ ˙ [su]rendramandirà»àá ruciman mandarakanda[rànurùpaá]60 {|}61 ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ janair62 yathepsita(m)63 {|| 27 ||} asamasya virocane girer vvika64 L23: ˙ ˙ ˙ ˙ ˙ ˙ ≤ramà[nta]ka65tayà66 (nivahena) ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ ˙ {|| 28 ||} (≤rì)[vaijayantam i]ti67 yasya janena nàma prìtiprasàdavikacapra»a[yena] cakre {|}68 etaL24: ¯69 ‘ ¯ ‘ ‘ ‘ ¯ 70 layanaá surendramauliprabhopaci71ta[ma«galasaágrahàya]72 {|| 29 ||} nivedya73 sa«ghàya ‘ ¯ 74 pabhogyaá75 sabandhuvarggas sa varàhadeva˙76 {|} n‰devasaukhyàny anubhùya ¯ ’ 77 L25: ’ ¯ [»a]78 ≤às[t]à sugatapra≤asta˙ {|| 30 ||} sàndràmbhodabhuja«ga79bhoga(nikarai)r yyàvat ‘ ¯ ¯ ‘ ¯ 80 nùtnamana≤81 ≤ilàlakapilair yyàvat karair bhàs[k]ara˙82 {|} tavàc che83 L26: ‘ ‘ ¯ ‘ ¯ ‘ ‘ ‘ ¯ ¯ ¯ ‘ ¯ sevyatàm antarmma»∂aparatnam etad amalaá rat-
ajanta’s inscriptions
313
natrayodbhàvita[á] {|| 31 ||} vividhalayanasànus sevyamàno mahadbhir ggirir aya L27: ‘ ‘ ¯ ¯ ¯ ‘ ¯ ¯ ‘ ¯ ddhya˙84 {|} jagad85 api ca samastaá vyastadoßaprahà»à[d86 vi]≤atu padam a≤okaá nirjvaraá ≤àntam àryyaá {|| 32 ||} Text Notes 1 ICTWI: nirvàpano; BCTTI: nirvvàpa»à. 2 BCTTI: ti pra°; CII: (ya)tiá pra°. CII’s suggestion violates the meter. 3 ICTWI, BCTTI, & CII: propose reading “anivàrrya(vìryya˙)” 4 ICTWI: . . . [ya≤oá]≤ukànàá; CII suggests restoring uddh‰takaá†akànàm 5 ICTWI, BCTTI: °vaá≤a° 6 ICTWI: haryyutthi°; BCTTI: ra»e [sva]haryyutthi°; CII: ra»e[ßu] haryyuddha 7 ICTWI, CII: °sañchà° 8 CII: sa ca [karmma?] . . . 9 ICTWI: °prava»àñ 10 ICTWI, BCTTI: kàryya≤ 11 BCTTI: °»à° 12 ICTWI: nareádra; BCTTI, CII: (ari)narendra 13 ICTWI: °kßa»a, and takes the beginning of the next verse as part of this compound 14 ICTWI: [rudra?sena˙]; BCTTI: . . .[dra]senah 15 ICTWI: kuntaleádra 16 BCTTI: [p‰]thivì[ße»a˙]; CII: ≤rìviádhyasena˙ 17 CII proposes reading “n‰po ‘bhavat” here 18 ICTWI: pravarasenasya 19 BCTTI ends half verse here 20 ICTWI: °≤àsanapravara[˙] 21 ICTWI: kàmata; BCTTI: . . . yàtmaja.ma . . .; CII: tasyàtmaja˙ . . . 22 CII: ya˙ 23 ICTWI: vvipàvanair n‰devaràjasya 24 ICTWI, BCCTI, CII: . . .bhù˙ 25 BCTTI suggests inserting “tasya” here 26 CII: °vàsa˙ 27 CII reading “≤rìhastiko≤o” 28 ICTWI: h‰≤ya . . . 29 ICTWI: °kùlànu° 30 ICTWI: [≤à]sana . . . saka≤ ca 31 BCTTI: loka suhità° 32 ICTWI: °paripàlanàya 33 ICTWI: [nityaá]; CII: (satyam) 34 ICTWI: sasarjja 35 ICTWI: °ce[߆àm?] 36 ICTWI: harir eßa hara[˙] smareva kàntir; BCTTI: hari[ràma]harasmare[va]kàntir; CII: hariràmaharasmarendukàntir 37 ICTWI, CII: ya≤a˙pu»yagu»àá≤u; BCTTI: °gu»à≤u 38 CII suggests restoring “≤ubhra˙” here. 39 CII: imàn 40 ICTWI & BCTTI: °dharmmàparito; CII: sahàyadharmà [nirato] 41 BCTTI: kàrà. 42 CII: °sukhà»i 43 CII: nyavì≤ad 44 ICTWI: . . . rudhitàgre; CII: [sajalà]mbuda[v‰]ndalambitàgre 45 BCTTI: °gendràdv yußite 46 ICTWI: ≤arànikuñje 47 BCTTI: °laák‰tam 48 ICTWI: °stamma° 49 BCTTI: [meduraá] . . .; CII supplies “bhùßitam”. Even if these suggested readings are not correct, clearly whatever was here must have been their rough synonym. 50 BCTTI: . . . [uddhu]racaityamandira[á] 51 ICTWI: ma . . . talasannivi߆aá visa . . . na; BCTTI: ma . . . talasannivi߆avi . . . 52 ICTWI suggests ending verse here. 53 ICTWI: va . . . ñcàmbu°; BCTTI: . . . ñcàmvu° 54 ICTWI: °àdibhir apy ala«k‰tam, and for ICTWI this completes a half verse; CII supplies “apy ala«k‰tam” 55 CII: . . . [rmma]ha[ti] 56 ICTWI: . . . samìra»a . . .; BCTTI: . . . samara»à . . .; CII: samìra»e 57 ICTWI suggests ending verse here. 58 ICTWI & CII: °gu»o° 59 ICTWI: °yogyam; BCTTI: °yoga[á]; CII: °yogya[m] 60 CII: [°rùpam] 61 ICTWI suggests ending verse here. 62 ICTWI: . . . sukho; BCTTI: . . . ha[á] 63 ICTWI: yathepsita˙; BCTTI: yathepsita. Mirashi’s estampage shows another akßara here, it could be a final m-màtra but is too unclear to make out. I see neither a visàrga nor an anusvàra however. 64 ICTWI: asamapratirocane girau vika[ca]; BCTTI: asama[prati]rocane girer vaka 65 ICTWI: . . . ≤ramàntaka˙, and suggests ending verse here; BCTTI: . . . sramà . kà ., and suggests ending first half verse here 66 BCTTI: laya . . . 67 CII: . . vi≤àlam iti 68 ICTWI suggests ending verse here 69 ICTWI: ceta˙; CII: etasya 70 CII suggests “. . . girer” as a possibility 71BCTTI: °cì° 72 . . . hàyaá to end his first half-verse 73 ICTWI: nivevedya 74 CII suggest reading “suve≤ma” in this lacuna 75 Read upabhogyaá; ICTWI: . . . thyaá; CII: bhaktyà 76 ICTWI suggests ending verse here 77 CII suggest reading “samyak” in this lacuna 78 CII: suggests reading “dharmme»a” in this lacuna 79 janga° 80 ICTWI: . . . tamàràma . . . for this lacuna; BCTTI: . . . ma . . . 81 ICTWI & BCTTI: . . . nnamana≤; CII: °mana˙ 82 BCTTI: bhàsvara˙, but suggests reading as above. 83 ICTWI: chi 84 ICTWI: . . . bhya˙; BCTTI: . . . ddhya 85 BCTTI: yad 86 ICTWI: °prahànam; BCTTI: samastavyasta°
appendix
314
Inscription 68 Cave: Location: Medium: Type: Editions: Copies: Notes:
Sixteen Vihàra, front wall, to the left of the door Painted Second period, programmatic, descriptive labels AJ3: 96, #6 AJ3: Plate VIIIc–e AJ3 points out that although several narratives at Ajanta are based specifically on Àrya •ùra’s Jàtakamàlà, that text was probably not the source for this painting. It adduces two reasons. First, Àrya •ùra does not name the evil Bràhman “Yujaka.” Second, Àrya •ùra’s hero is named “Vi≤vantara,” not “Vai≤vantara,” the latter corresponding to the Pàli “Vessantara.” STUDIES, by contrast, understands this painted narrative to be part of a larger “cycle” on this wall, depicting Àrya •ùra’s Jàtakamàlà in its entirety. Nevertheless, STUDIES finds only incidents common to all known literary versions of the Vi≤vantara story here, and nothing particular to the Jàtakamàlà’s account.
Text L1: Vai≤vantara˙ Indra˙ Yu[jaka] Inscription 69 Cave: Location: Medium: Type: Editions: Copies: Notes:
Sixteen Vihara, right wall between cells R3 and R4 Scratched into paint Second period, intrusive, graffiti NIA: 152 NIA: figure 6 “Sùstradhàra” is either wrong for, or a dialectal variation of, “sùtradhàra,” literally “thread-holder.” This title ranges in meaning from architect to painter, broadly including any artist whose work includes the formal delineation of line and proportion. (See Shridhar Andhare. “Sùtradhàra as a Painter.” In The Art of Ajanta: New Perspectives. vol 2. Ed. by Ratan Parimoo et al. [New Delhi: Books and Books, 1991]: 351–55.)
Text L1: ≤ri1 yugadhara sùstradhàra Text Note 1
NIA: sri
Translation •rì Yugadhara, architect (or painter).
ajanta’s inscriptions
315
Inscription 70 Cave: Location: Medium: Type: Editions: Copies: Notes:
Sixteen Vihàra, left wall, over cell L4, below and between the first two painted Buddhas of a series of four Painted Second period, intrusive, donative ICTWI: 87, #27; AJ3: 95, #1 ICTWI: Cave XVI, #27; AJ3: Plate VIIa Iconographically, these four Buddha images are not particularly interesting. All wear yellow robes, and sit on lotuses under three-tiered parasols. They vary in that the first (from the left) has his right shoulder bare and holds his hands in dharmacakrapravarta»a mudrà; the second has both shoulders covered and his hands are in dhyàna mudrà; the third is like the first, and the fourth like the second. Such pairings of robe-style and mudrà are not unusual, nor is this alternation. All four Buddhas are set within a single painted frame, suggesting that they are to be viewed as a group. Groups of four Buddhas include: the Buddhas of the four directions, the four Buddhas of the La«kàvatàra Sùtra, and the four Buddhas of the Bhadrakalpa that have heretofore awakened: Krakucchanda, Kanakamuni, Kà≤yapa, and •àkyamuni (Paul Demieville. Hobogirin. Fasicule 3. [Paris: A. Maisonneuve, 1974]: 195, s.v. butsu). I would guess that the latter group is intended here. For other inscribed image groups that name or have a definite concern with previous Buddhas see Cave 10, #58 and Cave 22, # 90 and #91. There are two more points to be made about these images and this inscription. First, these four Buddhas are centered upon the cell L4 door, even though space was available on either side. One might guess that L4 was Dharmadatta’s own cell. The cell seems to have been occupied: it was plastered and the plaster is thickly coated with soot. However, the cell’s walls were not finished; nor was a door or locking mechanism ever hung, though matrix was reserved. The second point of interest is that Dharmadatta was so very territorial about his Buddhas. This group of Buddhas is marked by two dedications (#70 and #71), and each one of the four Buddhas also bears a record stating that it is Dharmadatta’s property (#72). Caves 2 and 10 show other instances wherein there may be multiple inscriptions on a single dedication or group of dedications by a single patron. To put Dharmadatta’s hypothesized anxieties into perspective, let me repeat a tale from Xuanzang. It takes place in Gandhàra: “Some old people said that there was formerly a poor man who sustained himself by working as a laborer. Once he earned one gold coin and wished to make a portrait of the Buddha. He came to the stupa and said to a painter, ‘I wish to make a portrait of the Tathàgata’s excellent features, but I have only one gold coin, which is really insufficient for remuneration. This has been my long-cherished desire, but I am poor and lacking in money.’ In consideration of the poor man’s sincerity, the painter did not argue about the payment and promised to accomplish the job. Another man under the same circumstances came with one gold coin to request the painter to draw a portrait of the Buddha. Thus the painter accepted the money from the two men, and he asked another skillful painter to work together with him in drawing one portrait. When the two men came on the same day to worship the Buddha, the two painters showed them the portrait, pointing at it and saying, ‘This is the portrait
316
appendix you ordered.’ The two men looked at each other bewildered, and the painters realized the two men were doubtful about the matter and said to them, ‘Why are you pondering over the matter for so long? Whatever object we undertake to produce is done without the slightest fault. If our words are not false, the portrait will show miracles.’ As soon as they had uttered these words, the portrait manifested a wonder: the body split into two busts, while the shadows intermingled into one, with features shining brilliantly. The two men were happily convinced, and delightedly fostered faith” (Xuanzang. The Great Tang Dynasty Record of the Western Regions. trans. Li Rongxi. [Berkeley: Numata Center for Buddhist Translation and Research, 1996]: 72–73). A final point. The donor Dharmadatta is figured to the far left of the group, over the left corner of cell L4’s door. Should this monk be identified with the Dharmadatta mentioned in Cave 26’s verse inscription (#93, v. 14)? Another unanswered question.
Text L1: deyadharmmo ‘yaá ≤àkyabhikßor bhbhadantadharmadattasya yad atra pu»yaá L2: tad bhavatu màtàpitros [sa]rvvasatvànàñ cànuttarajñànàvàpta L3: ye STU1 Text Note 1
AJ3: [ssu˙]. See the Text Notes to inscription #17 for a discussion of this akßara. Whether continuation is indicated because each of the individual Buddha’s is inscribed with Dharmadatta name, or because he wants one to read on to the other dedicatory record is uncertain. Translation This is the religious donation of the •àkyabhikßu reverend Dharmadatta. Let the merit therein be for the attaining of supreme knowledge by [his] mother and father and all living beings. Inscription 71 Cave: Location: Medium: Type: Editions: Copies:
Sixteen Vihàra, left wall, over cell L4, below and between the third and fourth Buddhas Painted Second period, intrusive, donative DAJI; ICTWI: 87, #26; BCTTI: 138, #15; AJ3: 95, #2 DAJI; ICTWI: Cave XVI, # 26; BCTTI: Plate LIX, #15; AJ3: Plate VIIb
Text L1: deyadharm[m]o ‘yaá ≤àkyabhikßor bhbhadantadharmmadatasya yad atra (pu»yaá) L2: tad bhavatu [m]àtàpitros sarvvasatvànàñ cànuttarajñànàvàptaye
ajanta’s inscriptions
317
Translation This is the religious donation of the •àkyabhikßu reverend Dharmadatta. Let the [merit] therein be for the attaining of supreme knowledge by [his] mother and father and all living beings. Inscription 72 Cave: Location: Medium: Type: Editions: Copies:
Sixteen Vihàra, left wall, over cell L4, on the lotus throne of each of a series of four Buddhas Painted Second period, intrusive, donative AJ3: 95 None published
Text bhadantadharmadattasya Translation Reverend Dharmadatta’s Inscription 73 Cave: Location: Medium: Type: Editions: Copies: Notes:
Sixteen Vihàra, left wall, L5 and L6, below the third Buddha a series of Buddhas Painted Second period, intrusive, donative DAJI; ICTWI: 87, #25; BCTTI: 138, #14; AJ3: 95, #3 DAJI; ICTWI, Cave XVI, #25; BCTTI: Plate LIX, #14; AJ3: VIIc Most likely, this donation was a series of eight (seven Buddhas plus Maitreya). At present only six Buddhas are visible. Bàpuka’s Buddhas are located directly above the four Buddhas painted for Dharmadatta. And like Dharmadatta, Bàpuka is concerned to make sure all of his Buddhas are duly marked as his property.
Text L1: deyadharmmo ‘yaá ≤àkyabhikß[o]r bhbhadanta1 bàpukasya2 L2: yad atra [pu]»yaá tad bhavatu màtàpitro sarvvasatvànàñ cà L3: m anu[tta]rajñànàvàptaye Text Notes 1
BCTTI: bbhadanta
2
DAJI: buddhakasya; ICTWI & BCTTI: dàpukasya
Translation This is the religious donation of the •àkyabhikßu reverend Bàpuka. Let the merit therein be for the attaining of supreme knowledge by [his] mother and father and all living beings.
appendix
318
Inscription 74 Cave: Location: Medium: Type: Editions:
Sixteen Vihàra, left wall, between cells L5 and L6, on the lotus throne of each of a Buddhas donated by Bàpuka Painted Second period, intrusive, donative AJ3: 95, #3
Text bàpukasya / bhadantabàpukasya1 Text Note 1 “bàpuka” is painted on the lotus of the 4th and 6th Buddhas in this group; “bhadantabàpuka” on the 5th.
Translation Bàpuka’s / Reverend Bàpuka’s Inscription 75 Cave: Location: Medium: Type: Editions: Copies: Notes:
Sixteen Vihàra, left wall, between cells L4 and L5 Yellow pigment (more like chalk than paint) over painted plaster Second or post-second period graffiti AJ3: 95, #4 AJ3: Plate VIIIa AJ3 writes that both this and the subsequent record belong paleographically to the “nail-headed variety,” intimating their contemporaneity. with the site. This is possible. In Cave 21 we also find words in a similar script scribbled with what seems to be chalk on unfinished walls (cf. #86, #87). If these inscriptions do date from the Vàkà†aka period, it seems possible that, given their location and medium, they informed the artists which scenes to paint on the marked walls; or they may be the names of the artists themselves, who claimed these spaces for their own use. Militating against this interpretation are the graffiti in Cave 1 (#1 & #2)—these use a similar script and medium but are scribbled over already-painted scenes— as well as some writing on the ceiling of the right end porch cell in Cave 21—this was scribbled over unpainted plaster that had been blackened by soot. However, this interpretation finds possible support from a similar record written in red paint or crayon on a pillar in the main stùpa hall of Pi†halkhora, on a plastered but unfinished area at the top of a pillar. One wonders why a pious vandal would have chosen such an out-of-theway spot at Pi†halkhora, given that more accessible places were available. Inscription 76
Cave: Location:
Sixteen Vihàra, left wall, between cells L3 and L4
ajanta’s inscriptions Medium: Type: Editions: Copies: Notes:
319
Yellow pigment (more like chalk than paint) over painted plaster Second or post-second period graffiti AJ3: 95, #5 AJ3: Plate VIIIb See the note for #75. The wall from which inscriptions #75 and #76 were taken is, in fact, covered by graffiti, some from the fifth century. This is an excellent project for somebody with broad palaeographic knowledge, time and patience.
appendix
320
CAVE SEVENTEEN
Inscription 77 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seventeen Porch, left wall, above eye level Incised Second period, programmatic, donative DAJI; ICTWI: 73–76, #4; BCTTI: 128–132, #4; CII: 120–129, #27 DAJI; BCTTI: Plate LVI, #4; CII: Plate XXVII Regarding the lack of a translation, see the note to #67. Meters: Vaá≤astha 1, 10, 11, 12, 13, 18, 28, 29; Upajàti: 2, 8, 9, 14, 15, 16, 17, 19, 20, 22, 23, 24, 25, 26, 27; Aupacchandasika 3, 4, 5, 6, 7, 21
Text L1: L2: L3: L4: L5: L6: L7: L8: L9: L10: L11: L12: L13: L14: L15:
‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ drumà≤ani[á]1 pra»amya vidyàtrayapàraga[á] munim2 {|} vihàradàtur vvyavadàtakarmma»o gu»àbhidhànopanaya˙ karißyate || {1 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ [pà]lanena3 labdhàtmabhàvasya naràdhipasya4 {|} dh‰tàtapatrasya babhùva putras sitàtapatro5 dh‰tarà߆rasaájña˙ || {2 ||} ‘ ‘ ¯ ‘ ‘ ¯ ‘ ¯ [sya] ràjño harisàmbo ‘mburuhendukàntavaktra˙ {|} n‰pates tanayo babhùva tasy(à)py amala≤rì[˙] kßitipàla≤aurisàmba˙ {|| 3 ||} ‘ ‘ ¯ ‘ ‘ ‘ ‘ ¯ ‘ ¯ ¯ 6 p‰thukìrttir dyutimàn7 upendragupta˙ {|} samabhùd a[varas s]uto8 ‘tha tasya kßitipa˙ kàca iti prakà≤anàmà {|| 4 ||} ‘ ‘ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ¯ [dyu]ti9kìrttinyasanàya bhikßudàsa˙ {|} prathito bhuvi nìladàsanàmà n‰patis tasya suto naràdhipasya || {5 ||} ‘ ‘ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ¯ 10 prathita˙ kàca iti pradìptakìrtti˙ {|} n‰pater atha tasya k‰ß»adàsa˙ kulavaá≤adyutivarddhano babhùva || {6 ||} ‘ ‘ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ¯ s tanayà11 candrakaràvadàtave≤à {|} abhavat paripùr»»acandravaktrà vinayàcàravibhù12ßa»à[ti]candrà13 || {7 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ (artthisthalo)dyo14takarìm avàpa {|} tasyàñ ca15 tasyàmburuhàyatàkßàv uttaptacàmìkarakàntar[ù]pau || {8 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ (pradyu)mnasàmbapratimau kumàrau {|} ekàdhipatyaá prathamo ‘vatàra(˙)16 daddhre dvit(ì)yo ravisàmbasaájñàm || {9 ||} ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ¯ ‘ ‘ m a≤makàdi[ßu]17 {|} [k‰]tàrddhasatvàm abhibhùya18 bhùyasà raràjatu≤ candradivàkaràv iva || {10 ||} ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ¯ ha nibaddhamànayo˙ {|} viv‰ddhasauhàrddaya≤a˙pratànayos19 sadànukùlyena sukhaá vijahrußo˙20 {|| 11 ||} ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ (amàna)vair apy anivàryya≤àsana˙ {|} puràk‰todbhàvitabhìmavikrama˙ kanìyasi pràkhyad anityatà≤ani[m]21 || {12 ||} ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ¯ ‘ ¯ vadhairyyàd iva [kà]yadhìruja˙22 {|} anityasaájñàsacivas23 tata˙ paraá vyavìv‰dhat24 pu»yamahàmahìruham || {13 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ tatoßam25 {|} bhùya≤≤ruta26tyàgadayàpramodamaitrìkßamàvìryyadhiyas sißeve || {14 ||} ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ¯ ‘ ¯ ¯ ‘ ‘ ¯ narendràn {|} pra≤astav‰ttàn suvi≤uddhav‰tto v‰ttena samya[kßubhi]to27 ‘nucakre || {15 ||}
ajanta’s inscriptions L16: L17: L18: L19: L20: L21: L22: L23: L24: L25: L26: L27: L28: L29:
321
¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ¯ ‘ ¯ ¯ ‘ ‘ ¯ cak[à]ra {|} a[n]yàrtthika28syàrtthijanas tathaiva kìrttiá k‰tàrttha˙29 prathayàm babhùva || {16 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ [yà]n bha[ya]viplutàkßàn30 {|} amùmucad vittavisargga≤aktyà putràn ive߆àn31 karu»àbhim‰ß†a˙ || {17 ||} ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ¯ (pra»a)yena putravat {|} anùcivàá so ‘pi hi yasya h‰dgatàá vidan‰vadhyà≤a[ya]≤uddhisampadam32 || {18 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ ¯ ‘ karà»i sadya˙ {|} sarvvajñatà ca pra»idhànasiddhi[˙]33 satyàbhidhànà[bh]ibhavàd34 apeyu˙ || {19 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ (sa)mbhàracayàdhiyoga˙35 {|} ya≤oá≤ubhi≤36 candramarìci≤ubhrair37 jjagat samagraá samala«cakàra38 || {20} || ‘ ‘ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ¯ [vada]nàravindacandre {|} paripàlayati kßitìndracandre hariße»e hitakàri»i prajànàm || {21 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ (n a)tyadbhutapu»yarà≤i˙ {|} [cakre] bhuva39 stùpavihàrabhùßàá40 dànodayai≤ càrtthijanapramodam {|| 22 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ nàkulanàdavadbhi˙41 {|} ni[tyaá] vitànà[rthadhiyà va]hadbhir42 ambhodharai≤ [≤]rìmati43 sahyapàde || {23 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ (gàmbh)ìryyagu»air upetam {|} nive≤itàntarmuniràjacaityam ekà≤makaá ma»∂aparatnam etat || {24} || ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ¯ ‘ ¯ ¯ vipulàá44 vis‰jya {|} acìkarad d[i]tyam a[mà]nakalpam45 alpàtmabhi˙ kalpanayàpy a≤akyam || {25 ||} ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ ¯ nayanàbhiràmam46 {|} nyavìvi≤at svàdu47laghuprasanna≤ìta48prakàmàmbumahànidhànam || {26} || ’ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ’ ’ ¯ ‘ ¯ n netra[ma]nobhiràmàm49 {|} anyà[á]gade≤e ‘sya di≤i pratìcyàm acìkarad gandhaku†ìm udàràm || {27} || ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ (jaga)ddhitàyodyatasarvvakarmma»a[˙]50 {|} munìndra[bhàva]pra»idhànasiddhaye51 bhavantv abhì߆à bhuvi sarvvasampada˙ || {28 ||} ‘ ¯ ‘ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ ‘ ¯ sapra»ayena ma»∂apa˙ {|} karotu tàvat ku≤alodayaá satàá nihanti52 yàvad ravir aá≤ubhis tama˙53 || {29 ||} siddhaá54
Text Notes 1 ICTWI, BCTTI: . . . màvaniá 2 ICTWI: munìm 3 ICTWI: . . . ne≤a; CII suggests reading “namnà janapàlanena” 4 CII: °patya 5 BCTTI: sitata° 6 ICTWI: . . . ne≤a; BCTTI: . . . ≤a; CII: . . . [te]na 7 ICTWI: °kìrtidyutimàn 8 ICTWI: a[gaja]s; BCTTI: ava.s. BCTTI suggests reading “avanas suto” or “agajas suto”; CII suggests “apara” as still another possibility. As far as I can tell, CII’s facsimile shows, “atejas,” meaning dull or feeble; but such candor about one’s progentitors is unkown in these elegies. 9 ICTWI: . . . dbhubi°; BCTTI: [d bhu]vi 10 CII: . . . [rai˙] 11 ICTWI: ptanayà 12 ICTWI: °bhu° 13 BCTTI: °[su]candrà 14 Read °loddyo° 15 CII: tasyàá ca 16 ICTWI: ekàdhipatyaprathamàvatàraá; BCTTI: ekàdhipatyaá prathamovatàraá, but suggests reading “prathamobabhàra”; CII: dharàdhipàrakhyàá prathamo babhàra. Despite BCTTI and CII’s urgings, it is difficult to see babhàra for “vatàra.” Insofar as both are willing to ignore a mark that looks like an anusvàra over the
322
appendix
“ra” (one can almost never be sure that a seeming anusvàra or visarga is not just a pit in the rock), I too ignore it and read “avatàra” syntactically as a nominative singular. Indeed, there should be nothing shocking in this verse’s calling these brothers “avatàras” after having said that they are the image of Pradyumna and Sàmba; although these gods are, properly speaking, vyùhas of Nàràya»a, not avatàras, we cannot expect such theological exactitude in a text of this sort. The fact that their father’s name is K‰ß»adàsa of course completes the pun. Finally, it would have been very inelegant indeed for the verse to rely upon two verbs (babhàra & dadhre) where one is sufficient. 17 ICTWI: . . . m a≤makàdi; BCTTI: . . . sam a≤makàdi[bhi˙]; CII: [nìyocchrita] a≤makàdi[kam]. 18 ICTWI: . . . nutàbhyàm atibhùya; BCTTI: .nu tàbhyàm abhibhùya; CII: [k‰]tàrtthasatvà[v]. 19 ICTWI: °ya≤apra° 20 ICTWI: vijahrato˙ 21 ICTWI: °≤ani˙; BCTTI: pra[syata] . . tà≤ani˙ 22 ICTWI: . . . yàdhiràja˙ 23 ICTWI: aciátyasaájña˙ sacivas; BCTTI: acityasaájña˙ sacivas 24 BCTTI: vyavìvigdhat 25 ICTWI: . . . nu[yàtatoßa]m; CII: . . . nujàtatoßà[n] 26 ICTWI bhùya≤ruta° 27 Read samyak kßubhito. ICTWI: samyak kßubhito; CII: sabhyànucito 28 ICTWI: a. . . rtthika°; tavaiva; BCTTI: artthika° 29 ICTWI: kìrttik‰° 30 BCTTI: . . . sa . . bhiplutàkßàn 31 ICTWI: putràbhice߆àn; BCTTI: putràdhice߆àn 32 ICTWI: vindan n‰vadhyà≤a-ya≤utdhisampadam; BCTTI: vidanravadhyà°; CII: vida[n] n‰vadhyà°, suggests reading n‰vaddhyà° 33 ICTWI & BCTTI: sarvvajñabhàvapra»idhànasiddhi˙; CII: sarvvajñabhàvapra»idhànasiddhiá 34 CII: satyàbhidhànàá vibhavàd 35 ICTWI: . . . mbàraca°; CII: °yogà(t) 36 ICTWI: ya≤ombubhi≤ 37 BCTTI: °bhrai˙ 38 ICTWI: samalaácakàra; BCTTI: [vi]malañ cakàra 39 BCTTI rightly suggests reading “bhuvaá” 40 ICTWI & BCTTI: °bhùßà 41 ICTWI:. . . [nànilanàdavadbhi˙] 42 ICTWI: . . . ≤arayàbhahadbhi˙ 43 ICTWI: . . . mavi 44 ICTWI: . . . ߆iá vipulàá; BCTTI: vipulaá 45 ICTWI: acìkarac caityam ihànakalpam; BCTTI: acìkarac caityam [ahì]nakalpam 46 ICTWI: . . . ≤e naya° 47 CII: sàdhu 48 ICTWI: °≤ìtaá 49 ICTWI & CII: °ràmam 50 BCTTI: °karmma»a, but suggests reading °karmma»àm 51 CII: munìndranàthapra»idhànasiddhaye 52 CII: vihanti 53 CII: aá≤ubhastima˙ 54 Expressed by a symbol. Inscription 78 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seventeen No longer extant. Porch, left wall, inside the Wheel of Existence Painted Second period, programmatic, didactic verse Never recorded Never copied In FIA: 560, Mr. Ralph, one of the cave’s first Western visitors, remarks that an inscription can be seen within the Cave 17’s Wheel of Existence (which he calls the “zodiac”). The Divyàvadàna advises that a Wheel of Existence should include the following two didactic verses: àrabhdavaá nißkràmata yujyadhvaá buddha≤àsane | dhunita m‰tyuna˙ sainyaá na∂àgàram iva kuñjara˙ || yo hy asmin dharmavinaye apramatta≤ carißyati | prahàya jàtisaásàraá du˙khasyàntaá karißyati || (E.B. Cowell and R.A. Neil [eds]. The Divyàvadàna. [Cambridge: The University Press, 1886]: 300). These verses come from the Udànavarga, IV, vv. 37–8, and may be translated: Begin, renounce, and join the Buddha’s Order! Death’s army clamors, like an elephant in a hut of reeds. A man who progresses in this doctrine and discipline, unshaken, Will escape the cycle of birth and make an end of misery.
ajanta’s inscriptions
323
Inscription 79 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seventeen Porch, left wall, in the lower left-hand corner, under the figure of a richly bejeweled green yakßa Painted Second period, intrusive (?), descriptive label ICTWI: 87, #28; AJ3: 96, #1 ICTWI: Cave XVII, #28; AJ3: Plate IXa Aside from the obvious interest this label possesses vis-à-vis Ajanta’s architectural, decorative, and apotropaic programmes, this is an important record for its paleography. This is the only inscription at Ajanta to use a “northern” form of the m found in some Gupta inscriptions from Mathurà (including the Gupta year 135 [= 454 C.E.] image inscription) as well as sixth century Buddhist manuscripts from Gilgit (see Lore Sander. Paläographisches zu den Sanskrithandschriften der Bernliner Turfansammlung. [Wiesbaden: Franz Steiner Verlag GMBH, 1968]: Tafel IV–B).
Text L1: mà»ibhadra˙ Inscription 80 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seventeen Porch, rear wall, left corner Painted Late-sixth to eighth-century graffiti AJ3: 96, #4 AJ3: Plate IXd Along with the next two records, inscription #85 from Cave 20, #86 from Cave 21, and the long inscription #97 between Cave 26 and 26 lower left, this inscription clearly dates from a time much later than that proposed by Walter Spink as the end of Ajanta’s efflorescence. The fact that none of these later records possess discernable Buddhist content supports Spink’s contention that there was no Buddhist community at Ajanta in Xuanzang’s time. Though we cannot be sure whether Xuanzang himself visited Ajanta, the paleography of this record tells us that his contemporaries did. Still, it should also be noted that inscription #99 from Gha†otkaca derives from a similarly late date, but is definitively Buddhist, showing that at least one Buddhist did come to this area in the seventh or eighth century.
Text L1: ≤rì . . . pauka . . . (?)
appendix
324
Inscription 81 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seventeen Porch, rear wall, to the right of #80, just to the left of the door way Painted Late-sixth to eighth century, graffiti Not previously noticed None published This inscription may be classed with the preceding as late, intrusive, and probably non-Buddhist. It is located directly to the left of the porch’s left side doorway. Spink has proposed that Cave 17’s two side doors were partially filled, in order to provide additional space on which painters might work. The location of this inscription lends credence to Spink’s supposition. For if this inscription was any longer than “≤rì,” the only place the additional akßaras could have gone was over what is presently the doorway’s empty space.
Text L1: ≤rì . . . Inscription 82 Cave: Location: Medium: Type: Editions: Copies: Notes:
Seventeen Vihàra, left wll, to the left of cell L1 Painted Late-sixth to eighth century, graffiti DAJI; AJ3: 96, #3 DAJI; AJ3: Plate IXe See inscription #80. Although most of this record is preserved, I have been unable to make sense of it. It appears to be a name and perhaps titles. A mantra?
Text L1: ≤rì bha1[mà]»amà»àvapaukagai (gau?) radeva L2: vidußà ca»∂a-2 Text Notes 1
AJ3: Ru (Bha?);
2
DAJI: ≤rì bhamà»amàtavaloka ≤ivadevavidu≤ara»am Inscription 83
Cave: Location: Medium: Type: Editions: Copies:
Seventeen Vihàra, right wall, on both sides of cell R1 Painted Second period, programmatic, descriptive labels DAJI; ICTWI: 87, #29; AJ3: 96, #2 DAJI; ICTWI, Cave XVII, #29; AJ3: Plate IXb–c
ajanta’s inscriptions Notes:
325
Painted on front aisle, right wall. Of these, I have been able to find only two “≤ibiràjà”s.
Text L1: ≤ibiràjà ≤ibiràjà indra ≤ibiràjà indra˙
326
appendix
CAVE NINETEEN
The area over the entrance on the interior front wall appears to have been prepared for a verse record like those in Caves 16, 17, and 26. No inscriptions have been found in this cave to date.
ajanta’s inscriptions
327
CAVE TWENTY
Inscription 84 Cave: Location: Medium: Type: Editions: Copies:
Twenty Porch, left pilaster Incised Second period, programmatic, donative DAJI (called “Cave XIX”); ICTWI: 76, #5; BCTTI: 132, #5; AJ4: 113 DAJI (called “Cave XIX”); BCTTI: Plate LVIII, #5; AJ4: 113
Text L1: L2: L3: L4: L5: L6: L7: L8: L9: L10: L11:
. . . . . . . . . . . yaá1 ma»∂apa . . . . . . . . . . . . . . . . . . . . pautrasva2 kß[a]3. . . . . . . . . . . putrasya upendra4. . . . . . . . . . . . . . . . . . . . . sya dharmma[haga]5. . . . . . . . . . . . . . trasya jayata6. . . . . . . . . . . . . . . . . . . . .[sya] kulapri7. . . . . . . . . . . . . . . . . . . . .mocàsaka[sya]8. . . . . . . . . . . . . . . . . . . . tvàmika[syau]9. . . . . . . . . . . . . . . . . . . . . pu»yan tad10. . . . . . . . . . . . . . . . . . . . . . . . gamàmànà[á] sa11. . . . . . . . . . . . . . . . . . . . nàvàp12. . . . . . . . . . . . . . . . . . . . . . . .
Text Notes 1
Read deyadharmmo ‘yaá 2 DAJI: . . . trasya ka . . .; ICTWI, BCTTI, AJ4: °trasya, read with this 3 ICTWI: . . . prapautrasya k‰. BCTTI & AJ4: . . . pautrasya [K‰]. . . . AJ4 allows this might be read as kßa, but kßa (or kya) seem the only possibilities. 4 ICTWI: . . . pautrasyàcàryendra; AJ: upendra[sya] 5 DAJI: °dharmmapa . . . ; ICTWI: dharmmahaga . . . [pu]. The reading “haga” is dubious. 6 DAJI: jayasa; ICTWI: jayatà; BCTTI & AJ4: jayatàá 7 DAJI: . . . sya kalapri . . .; ICTWI: kulapri . . . [para]; BCTTI: lyakalapri . . . 8 DAJI: mohasakasya ICTWI: mopàsakasya; AJ4: mo[pà]saka[sya]. The reading suggested in ICTWI and AJ4, “paramopàsakasya,” is to be preferred for its sense, but the “cà” is clear and cannot be a “pà,” albeit “cà” could be deemed a scribal error. 9 DAJI: nvàmikasyau; ICTWI: nvamikasyai . . . [yada]. This might also read °tvàmika° 10 ICTWI: [tra] pu»yaá tad bhavatu màtàpit‰; BCTTI: pu»ya(á) tad bha[vatu màtàpit‰] 11 ICTWI: [pùrvaá]gamànàá sa[kalasatvànà]; BCTTI: [pùrvvaá]gamànàá sa[kalasatvànàm utta] 12 ICTWI: m anuttarajñànàvàpta[ye]; BCTTI: [rajñà]nàv[à]p[taye]. Read this latter part yad atra pu»yan tad bhavatu màtàpit‰pùrvvaágamànàá sarvvasatvànàm anuttarajñànàvàptaye Translation This pavillion is [the religious donation] of [X . . . the . . . of] Upendra . . . the son of Kßa . . . , grandson of X . . . . . . [Let the] merit [therein, be for the attaining of supreme knowledge by all living beings,] beginning with [his mother and father].
328
appendix Inscription 85
Cave: Location: Medium: Type: Editions: Copies: Notes:
Twenty Vihàra, left wall Painted in large red letters Late-sixth to eighth century, graffiti Not previously noted None published Apparently somebody’s name and titles like the graffiti in Cave 17. Not enough of this record remains to make transcription worthwhile.
ajanta’s inscriptions
329
CAVE TWENTY ONE
Inscription 86 Cave: Location: Medium: Type: Editions: Copies: Notes:
Twenty One Porch, rear wall, right of door Paint Second or post-second period, intrusive, graffiti AJ4: 111 None published Aside from the painting in large red letters, this wall, that to the left of the door, and many of the interior walls are covered by graffiti written with white chalk. See my note to #75 for further discussion of Ajanta’s graffiti. The record has deteriorated since the days of AJ4. Chakravarti was apparently able to read the fourth line of this record as “[Dasa]ratha [ma*?]»∂a»a,” and he suggests that this and other records on the cave’s walls (most of which are too faded or fragmentary to bother reproducing, though they could yield secrets to a very tenacious epigrapher having plenty of time and the ASI’s permission to use a bright white-light source) provided the titles of scenes to be painted on the marked portion of the wall. Inscription 87
Cave: Location: Medium: Type: Editions: Copies: Notes:
Twenty One Vihàra, over cell L1 White chalk Second or post-second period, intrusive, graffiti Not previously noted None published Again see my discussion at inscription #75. Paleographically, this record is akin to #75 and #76.
330
appendix
CAVE TWENTY TWO
Inscription 88 Cave: Location: Medium: Type: Editions: Copies:
Twenty Two Vihàra, right wall, lower back corner Painted Second period, intrusive, donative Not previously noticed None published
Text L1: deyadharmmo ‘yaá ≤àkyoposa(ka) .vasya màtàpit‰m uddi≤ya sarvvasatvànàá (ca) Translation This is the religious donation of the •àkya-upàsaka ?va, in honor of [his] mother and father, [and] for all living beings. Inscription 89 Cave: Location: Medium: Type: Editions: Copies:
Twenty Two Viharàra, rear wall, left of door, on the throne of a Buddha image Painted Second period, intrusive, donative NIA: 151 NIA: figure 4
Text L1: L1: L2: L2:
left: [siddham]1 deyadharmmo ‘yaá ≤àkya right: bhi[kßo]r bhadanta bha . . . [sya] màtàpitro2 left: m udi≤ya sa[rvva]sa3 right: tvànàñ ca bhavatu4
Text Notes 1 Shown by a symbol 2 NIA: màtàpitro[˙] [»yam tad bha]vatu cà(nuttaràjñànà)vàptaye
3
NIA: -dasya ya[d atr] pu
4
NIA:
Translation [Success!] This is the religious donation of the •àkyabhikßu reverend Bha?. Let it be in honor of his parents and for all living beings.
ajanta’s inscriptions
331
Inscription 90 Cave: Location: Medium: Type: Editions: Copies:
Twenty Two Shrine, right wall Painted Second period, intrusive, donative DAJI (called Cave XXI); ICTWI: 88, #30; AJ4: 112 DAJI (called Cave XXI); ICTWI: Cave XXII, #30; John Griffiths. The Paintings in the Buddhist Cave-Temples of Ajanta. vol. 1. (London: 1896): plate 91. The first edition of Griffiths’ book source provides the best evidence for the first half of the inscription; the modern reprint is too blurry. AJ4 merely reproduces Griffiths’ plate as its Plate I.
Text L1: siddhaá1 deyadharmmo2 ‘yaá ≤àkyabhikßo m apara≤aila3.i . . . nìyasya4 màtàpit‰5 . . . [u]tranya . [o] [sa]rvvasatvànàm anuttara[jñà]nàvàptaye saurupya6saubhàgyagu»opapaánnà7 gu»endriye8 bhàsvaradìptayas te9 bhavaánti10 te11 nayanàbhiràmà L2: ye kàrayaátì[ha] ji[nasya] biábaá Text Notes 1 Represented by a symbol, not in DAJI’s or ICTWI’s translations although it is on their plates. 2 ICTWI: deyadhammo 3 Read ≤àkyabhikßor apara≤ailasya; ICTWI’s plate however shows that the next akßara had a medial i, perhaps it read ≤àkyabhikßo m apara≤ailanikayasya. ICTWI: ≤àkyabhikßo maßara≤aila; AJ: ≤àkyabhikßo(r) ma[hà]yàna 4 ICTWI: nìsusya 5 DAJI: °pitro; AJ suggests “. . . nìyasya màtàpit‰”, but this is based upon the ICTWI plate which is clear 6 DAJI: rùpya; AJ suggests reading °rù° 7 DAJI: °panna; ICTWI: °pannà 8 DAJI: °yo; ICTWI: °yair 9 DAJI: dìptayeßu; ICTWI: °diptaya߆e; ICTWI’s eye-copy clearly deviates from the actual inscription here. 10 DAJI & ICTWI: bhavanti 11 AJ suggests reading ch = aite here, which is fine since it adds the long syllable necessary for a proper Upajàti pàda, but does not alter the sense.
Translation Success! This is the religious donation of the •àkyabhikßu Apara≤aila . . . for the attaining of supreme knowledge by [my] mother and father . . . [and] all living beings. . . . Those who commission an image of the Conqueror here Will be endowed with beauty, good fortune, and good qualities. The power of their merits blazing like the sun, they will delight the eye. Inscription 91 Cave: Location: Medium: Type: Editions: Copies:
Twenty Two Shrine, right wall, beneath the series of Buddhas whose donation occasioned the previous record Painted Second period, intrusive, descriptive labels DAJI (called Cave XXI); ICTWI: 88, #30; AJ4: 111–112 DAJI (called Cave XXI); ICTWI, Cave XXII, #30; Griffiths (see #90)
332
appendix
Text L1: vipa≤vì1 ≤ikhì vi≤vabhù (krakucchanda˙)2 ka[naka]muni˙ kà≤yapa˙ ≤àkyamuni maitre[ya˙] Text Notes 1 DAJI: vipa≤yì 2 This name is lost, but supplied based upon conventional lists of the seven Buddhas.
Inscription 92 Cave: Location: Medium: Type: Editions: Copies:
Twenty Two Shrine, right wall, above the trees of the seven Buddhas Painted Second period, intrusive, descriptive labels AJ4: 112 Griffiths (see #90)
Text L1: . . . pu»∂arìka . . . ≤irìßa˙ udum(b)a(ra˙) nyagro(dha˙). . . . . .1 Text Notes 1 “pu»∂arìka” is found above •ikhì, “≤irìßa˙” above so-called Krakucchandah, “udumbara˙” above Kanakamuni, and “nyagrodha” over Kà≤yapa.
ajanta’s inscriptions
333
CAVE TWENTY SIX
Inscription 93 Cave: Location: Medium: Type: Editions: Copies: Notes:
Twenty Six Porch, rear wall, over the right window Incised. It appears that this inscription was coated with a layer of lime plaster after being incised. The inscribed verses were probably then copied in paint onto this plaster. Second period, programmatic, donative DAJI (called “Cave XXV”); ICTWI: 77–79, #6; BCTTI: 132–136, #6; AJ4: 114–118, #1 DAJI (called “Cave XXV”); BCTTI: Plate LVIII, #6; AJ4: Plate II Regarding the absence of a translation, see #67 above. Meters: Drutavilambita 1; Aupacchandasika 2, 4; Upajàti 3, 10, 16; Vasantatilakà 5; Àrya 6, 7, 8; Vaá≤astha 9; Anu߆ubh 11, 12, 13, 14, 15, 19; •ikhari»ì 17; •àrdùlavikrì∂itam 18
Text siddham1 {|} jayati lokahitàvahitodyato ‘ (a)sukhà[nta]kara˙2 paramàrtthavit {|} trividhanirmmalasarvvagu»odayo muni[r abhi]3karu»àmalacandrika˙ {|| 1 ||} L2: punar4 api5 mara»àdi yena samya[gj]itam6 ajaràmaradharmmatà ca labdhà {|} ≤ivam abhayam anàlayaá gato ‘pi pra≤amapuraá agatàá karoti càrtthaá {|| 2 ||} L3: tato namaskàragu»àbhidhànaá bhavaty avandhyaá vipulaá mahàrtthaá {|} pradattam ekaá kusumaá ca yatra svarggàpavarggàkhyaphalasya hetu˙ {|| 3 ||} L4: ata iha vidußà tathàgateßu prathitagu»àdhikalokavatsaleßu {|} k‰tam anusaratà janena kàryyà dravakaru»àh‰dayeßv atìL5: va bhakti(˙) || {4 ||} devà nirastavijayàs savipattikatvàc chàpena ≤aábhur api kàcaralocano ‘bhù(t) || k‰ß»o ‘va≤o ‘pi va≤am àpatito ‘ntakasya tasmàj jayaáti L6: sugatà bhayavipramuktà˙ || {5 ||} sthaviràcalena muninà ≤àsanam udbhàvayaá k‰tajñena7 {|} k‰ta8k‰tyenàpi satà ≤ailag‰haá kàritaá ≤àstu˙ || {6 ||} L7: pràg eva bodhisatvair bhavasukhakàmai≤ ca mokßakàmai≤ ca {|} saávidyamànavibhavai˙ kathaá9 na kàryyà bhavet10 kìrtti˙ || {7 ||} yàvat kìrttir lloke tàvat varggeL8: ßu modati ca dehì {|} candràrkkakàlakalpà kàryyà kìrttir mahìdhreßu || {8 ||} anekajanmàáttara11baddhasauh‰daá sthiraá k‰tajñaá sudhiyaá vipa≤citam {|} L9: suràsuràcàryyamateßu12 kovidaá mahànubhàvà≤makaràjamantri»am13 || {9 ||} lokajñam ekàntasamantabhadraá sarvvàrtthi14nàm artthakaraá suvàcaá {|} gu»onnataá pra≤rayaL10: namramùrttiá15 khyàtiá gataá saccaritai˙ prithivyàá16 || {10 ||} da»∂asàddhyàni17 kàryyà»i vyàyàmaikarasàny api {|} yas sàdhayati sàmnaiva nripater18 mant‰puá«gava(˙)19 || {11 ||} itthaáL1:
334
appendix
L11: bhùto ‘sya putro ‘pi devaràjo dhuraáddhara˙20 {|} pitaryy uparate yena padam unnàmitaá gu»ai˙ || {12 ||} taá bhavviràjam uddi≤ya màtàpitaram eva ca21 {|} bhikßu»à buddhabhadre»a L12: kàrita˙ sugatàlaya(˙)22 || {13 ||} àgamya dharmmadattaá[c] ca23 bhikßuá24 sacchißyam eva ca {|} bhadrabandhum25 idaá ve≤ma tàbhyàá nißpàditaá ca me || {14 ||} yad atra pu»yaá tat teßàá L13: jagatàá ca bhavatv idaá {|} sarvvàmalagu»avyàta26mahàbodhiphalàptaye || {15 ||} yo buddha≤àsanagatiá samabuddhya jàto bhikßur vyayasy27 abhinave ‘bhijanopapanna(˙) {|} L14: bahu≤ruta˙28 ≤ìlavi≤uddhacetà lokasya mokßàya krità[dhikà]ra˙ || {16 ||} na sa(á)sàràpaánnaá ≤ubham api [tu k]iñcic29 chubhakaraá vipàko divyo L15: ¯ ‘ ‘ ‘ ‘ ‘ ¯ ¯ ca niyamà˙30 {|} ‘ ¯ lokàrtthàya pras‰tamana(sàá) pu»yamahatà(á)31 vipàko (dhì)rà»àá bhavati su(kha) ¯ ¯ 32 L16: najagatàm || {17 ||} ¯ ¯ ¯ ‘ ‘ ¯ ‘ ¯ ‘ va[labhi]r nn[à]nà»∂ajavyàhrite33 go]là«gùlani34nàdapùritadare pràgbhàravi35 ¯ ‘ ¯ {|} ¯¯¯‘ ‘ ¯‘ ¯‘ ‘ ‘ ¯ L17: yogì≤varàdhyàsite ve≤medaá jana ¯ ‘ ¯ ‘ janakaá36 bhùtyai prati߆hàpitaá || {18 ||} pùrvvàpi ceya[m] tenaiva dribddhàcàryyena saugati37 {|} loke ci(ntà)[m upà]dàya ’ ’ ’ ’ ‘ ¯ ‘ ’ 38 {|| 19 ||} Text Notes 1 Expressed by a symbol 2 ICTWI: . . . hi sukhànta°; AJ4: sukhà.kara˙. An examination of the rock suggests this reconstructed “[nta],” but it is unclear 3 ICTWI: mu[ßitabhì˙ ?]; BCTTI drops ICTWI’s “?”, but keeps the brackets; AJ4 reads as I have above, though he suggests reading, munir=a-bhì˙ 4 BCTTI: pu»ar 5 ICTWI suggests reading “vi” 6 ICTWI: samyakßivam; BCTTI: samya . . . ivam, but suggests reading samyagjitam instead of samyagjivam 7 ICTWI: adbhàvayak‰°. ICTWI, BCTTI, & AJ4 suggest reading udbhàvayat k‰° 8 ICTWI: krita°, but suggests k‰ta° as the correct reading 9 ICTWI: kathan 10 BCTTI: bhave 11 Read °janmàntara°; ICTWI: °janmàátara°; BCTTI: °janmàntara°; AJ4: °janmàttara°, but suggests °ànta° 12 Read °ryya°. ICTWI & BCTTI: °ryya° 13 ICTWI: °maátri° 14 ICTWI: °àrtdhi° 15 BCTTI: nàmra°, but suggests reading namra° 16 ICTWI & BCTTI: p‰th° 17 ICTWI & BCTTI: °sàdhyàni 18 ICTWI & BCTTI: n‰pater 19 ICTWI & AJ4 suggest reading, mantripu«gava˙ 20 ICTWI: dhurandhara˙; BCTTI: dhuraádhara˙ 21 AJ4 suggests reading, màtaraá pitaraá tathà 22AJ4: °à[layaá], but suggests reading °àlaya˙ 23 ICTWI & BCTTI: °dattañ ca 24 ICTWI: bhikßaá 25 BCTTI & AJ4: bhadrabundum 26 Read °vràta°; ICTWI: °dhyàta°; BCTTI suggests reading °vyàpta° or °vyàtta°; AJ4 suggests reading °vràta° 27 Read vvayasy; ICTWI, BCTTI: vvayasy 28 ICTWI & BCTTI: bahuvrata˙ 29 ICTWI: api tu kiñcic; BCTTI: api [tu k]i«cic. AJ4: api . . ncic 30 ICTWI: . . . tya . . . ca niyamà; BCTTI: . . . ca niyamà 31 BCTTI: °mahato, but suggests reading, °mahatàá 32 ICTWI: sukhabhogàya. BCTTI disputes ICTWI’s reading, as the “na” is distinct before “jagatàm” on line 16 33 This akßara not in ICTWI; BCTTI & AJ4: na 34 ICTWI: °vyàh‰te 35 ICTWI: °lìna°; BCTTI: golàngù° 36 ICTWI: pràgbhàvi . . . 37 ICTWI: saugatim; AJ4: saugatì 38 ICTWI & BCTTI: lokacintàm upàdàya . . .; AJ4: loke ciraá . . .
ajanta’s inscriptions
335
Inscription 94 Cave: Location: Medium: Type: Editions: Copies:
Twenty Six Facade, left side, under a colossal standing Buddha Incised Second period, intrusive, donative DAJI (called “Cave XXV”); ICTWI: 79, #7; BCTTI: 136, #7; AJ4: 118–9, #2 FAI: plate xxviii, no. 11; DAJI (called “Cave XXV”); AJ4: 119, #2
Text L1: siddham1 deyadharmmo ‘yaá ≤àkyabhikßor bhadanta gu»àkarasya yad atra pu»yaá tad bhavatu màtàpitaraá pùrvva«gamaá k‰itvà L2: sarvvasatvebhya anutarajñàna àptaye2,3 Text Notes 1 Expressed by a symbol. Not recorded in DAJI, ICTWI or BCTTI. 2 ICTWI: anuttarajñàna àptaye; DAJI & BCTTI: anuttarajñànàvaptaye 3 This line is followed by several auspicious symbols.
Translation Success! This is the religious donation of the •àkyabhikßu reverend Gu»àkara. Let the merit therein be for the attaining of supreme knowledge by all living beings, beginning with [his] mother and father. Inscription 95 Cave: Location: Medium: Type: Editions: Copies:
Twenty Six Facade, right side, under a colossal standing Buddha Incised Second period, intrusive, donative DAJI (called “Cave XXV”); ICTWI: 80, #8; BCTTI: 136, #8; AJ4: 119–120, #3 DAJI (called “Cave XXV”); BCTTI: Plate LVIII, #8; AJ: 119, #3
Text L1: siddham1 deyadharmmo [‘yaá]. . . . . . . . . . . L2: yad atra pu[»yaá]. . . . . . . . . . . . . . . . . . . . . . . L3: sarvvasatvànà[m a]. . . . . . . . . . . . . . . . . . . . . . Text Note 1
Expressed by a symbol. Read by ICTWI and BCTTI as oá
Translation Success! This is the religious donation . . . Let the merit therein . . . of all beings . . .
appendix
336
Inscription 96 Cave: Location: Medium: Type: Editions: Copies: Notes:
Twenty Six Ambulatory, right wall, to the left of the third “niche” Incised Second period, intrusive, donative ICTWI: 80, #9; BCTTI: 136, #9; AJ4: 120, #4 BCTTI: Plate LVIII, #9; AJ4: 120, #4 This inscription is not as unique as the accompanying image. A seated Buddha had been planned for this spot by one donor, but that first conception was replaced by a standing Buddha. Apparently, before the seated Buddha could be fully executed, its donor either 1) changed his mind about the iconography he desired, or 2) was unable to pay the artisan, and so lost his image, or 3) left the site, assured by an unscrupulous artisan, who naturally demanded up-front payment, that his image would be made to specification, or 4) you decide. Whatever the reason, we see here that while no fully realized Buddha at Ajanta was ever desecrated so, a roughed out Buddha-form had no sacral priority.
Text L1: deyadharmo ‘yaá ≤àkya L2: bhikßu saághamitrasya Translation This is the religious donation of the •àkyabhikßu Sa«ghamitra. Inscription 97 Cave: Location: Medium: Type: Editions: Copies: Notes:
Twenty Six On the front wall between Caves 26 and 26 left wing Incised, very lightly Post second period AJ4: 121–124 AJ4: plates III, IV, V The text given here is a reproduction of that found in AJ4. Refer to AJ4 for more on this odd record. The only other discussion I am aware of is Geri Hockfield Malandra. “The Date of the Aja»†à Cave 27 Inscription,” Wiener Zeitschrift für die Kunde Sudasiens. 26 (1982): 37–46, wherein Malandra criticizes AJ4’s treatement of the inscription, and proposes a date of the late seventh or early eighth century for its composition.
Text Part A L1: da hi ni .à-ràya-ra†h†e pu»∂ra sa ya la pa [ri] iva saá - ti Part B L2: ≤rì[˙] Dakßi»àpatthe Rà߆rakù††a-kule [sùto] Nanaràjasya bhràttà [du]hi[ka] -- llacandrasya [ka]----L3: [n]ì asya Nàmnà Vajra††adev-eti vikhyàto bhuvana-ttraye {|} tasyeyaá aákapa - - [rà] || Llalità-
ajanta’s inscriptions L4: L5: L6: L7: L8: L9: L10: L11: L12: L13:
337
valoka-≤atru-pràkàra-pralaya-divàkara-ni[ddhà]da .à[nu]ràga-praca»∂a-vidyàdhara-ca»∂a-Nàràya»a-rip[ù].ematta-gaja-kesari-mahimà»aá Mallaye≤a-kàrttà[nta] - - [sra] - - - [ràjà]dhi ara[yara] - samara-staábha- .u. i- - [ya]da[m‰]a [≤aurya]-saágràma-gaja-paridrava[»a]-davagi àhava-ha»ua | samara-Caturbhuja || ta»∂a -[i] ta»u atula- - - gaja-la . i n.ù [sa]cchahe»a ca»∂-à»ila-vea-calia-chàru-la- yara caácale»a Cholla i»»iaràya-ra††u-jìem àyàsa ma »a ja i - bha i ja sa»»ibhà»u - »»a vihartuá kiá»»are»a || ko ajar-àma[ra] e thu jage i mù jà »aáta vidì»a - »∂abhaá | caá ti - [saá ti] -[»»a]ya kàpara[r»»a] - ra»»ihi -[ja]hi pu»∂ra-jìvia kuha»∂a - tì eßa sa»»i-bhà»[ù]lla bha i sura-jua - - »»ava[bhuja] riva[kha]-bhaáge ma»∂avo i»»i aràya-[ra]††ur»»i[tu]- vihimillà[bhu]-ja»a sa ya la sa làhe »i -ja[ta] - ma - [sa]para-gràha || lla - kha»∂e[≤va]ra vi»àu càu[ra vartta]»a e càcupatthe durjaha - »e ----- - bhima - »∂a-pràya - bhaá - [ga]sa ma the mahaá gu»a e - dà si e .e ku[lla]»a - [sa]á ti -
Part C L14:
Kapa††atù«gena li(khi)tam etat
appendix
338
GHAˇOTKACA CAVE
Inscription 98 Cave: Location: Medium: Type: Editions: Copies: Notes:
Gha†otkaca Porch, rear wall, at far left Incised Second period, programmatic, donative ICTW: 88, #12; BCTTI: 138, #13; GCI: 15–18; CII: 112–119, #26 BCTTI: Plate LX; GCI: Plate I; CII: Plate XXVI I have reproduced the CII edition, without significant change. Meters: Upajàti 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12; Àryà 13; Pußpitàgrà 14, 15, 17; Vasantatilakà 16, 18
Text L1: L2: L3: L4: L5: L6: L7: L8: L9: L10: L11: L12: L13: L14: L15: L16: L17:
munir munìnàm amaro ‘marà»àá1 gurur gurù»àá pravaro varà»àá {|} jayaty anàbhogavibuddhabuddhir buddhàbhidhàno2 nidhir adbhutànàá {|| 1 ||} dharmmas tato dharmmavidà pra»ìtas tathà (ga)[»a]≤ càgryatamo ga»ànàá {|} bhavanti ya[smi]n nihi(tà˙)3 su[pà]tre kàràpakà[rà]s tanayà hy udàrà˙4 {|| 2 ||} asti prakà≤o di≤i dakßi»asyàá vallùranàmnàá dvijasattamànàá {|} à brahma»as saábh‰tapu»yakìrttir5 vva[«]≤o6 mahìyàn ma[hi]to mahadbhi˙ {|| 3 ||} ta[sminn a]bhùd àhitalakßa»ànàá7 dvijanmanàá [prà]thamakalpikànàá8 {|} bh‰gvatrigarggà«gira[sàá] samàno dvijarßabho [ya]jña[pati]˙ prakà≤a˙ {|| 4 ||} tadà[tma]jo deva [ivàsa] deva˙ k‰tì g‰hì[tì]9 nayavàn kriyàvàn {|} saràjakaá rà߆ram upetya yasmin dharmyàn akriyà˙10 [pà]rtha iva11 pracakkre {|| 5 ||} somas tata˙ soma [i]vàparo ‘bhùt sa bràhma»a12kßatriyava«≤ajàsu13 {|} ≤rutism‰tibhyàá vihità[rttha]kàrì dvayìßu14 bhàryyàsu mano dadhàra {|| 6 ||} sa kßatriyàyàá kula≤ìlavatyàm utpàdayàm àsa narendracihnaá15 {|} sutaá surùpaá ravinàmadheyaá k‰tàdhipatyaá vi(ßa)ye16 samagre {|| 7 ||} dvijàsu cànyàsu sutàn udàràn sa (pràpa) vedeßu17 samàptakàmàn {|} vallùranàmnà18 di≤i dakßi»asyàm adyàpi yeßà[m va]sati[r] dvi]jànàá {|| 8 ||} raves19 suto ‘bhùt pravaràbhidhàna˙20 ≤rìràmanàmàtha babhùva tasmàt {|} tasyàtmaja˙21 kìrttir22 abhùt sukìrttir23 bbabhùva tasmàd atha hasti[bhoja˙] {|| 9 ||} vàkà†ake ràjati devasene gu»ai˙ pràkà≤o24 bhuvi hastibhoja˙ {|} adyàpi yasyàbhijanapra[dà]na25 . . . . . . {|| 10 ||} dhìreßu dhìmatsu mahotthiteßu26 k‰tya[pra]mà[»eßu] gu»ànviteßu {|} . . . . . n‰pater27 yad e . . . . . . . . . {|| 11 ||} yasyàsanàrddhaá [pu]rußà28 . . . . . . màtsaásadi . . . . . {|} . . . . . . {|| 12 ||} atha devaràjasùnur29 hariße»o . . . hastibhoja . . . . . . . . . {|| 13 ||} ≤a≤ikaradhavalàá30 ni[rì]kßya kì[rttiá] . . . . . . . . {|| 14 ||} atha gu»a≤atasaábh‰tàtmabhùta˙31 priyatana(yo) . . . . . . . . . {|| 15 ||} tasyàtmajeßu bahußu . . . . . . . . . {|| 16 ||} phalam akhila32 . . . . . . . . . {|| 17 ||}
ajanta’s inscriptions L18: L19: L20: L21: L22:
339
samyagvibhàvita33 ihàsti varàha(deva˙) . . . . . . . . . {|| 18 ||} . . . . . . {|} . . . . . . {|| 19 ||} . . . . . . {|} . . . . . . {|| 20 ||} . . . . . . {|} . . . . . . {|| 21 ||} . . . . . . {|} . . . . . . {|| 22 ||}
Text Notes 1
CII: aroma°, a printer’s error. 2 BCTTI: bbuddha° 3 ICTWI: nihità 4 ICTWI & BCTTI: tanayà py[u]dàrà˙, but BCTTI suggests reading either °nayà˙ pyudàrà˙ or °nayà hyudàrà˙ 5 BCTTI: °pu»ykìrttirttir, but suggests reading °pu»yakìrttir 6 ICTWI: vvaá≤o; BCTTI: vaá≤o 7 ICTWI & BCTTI: àhata° 8 ICTWI: °kalpakà° 9 ICTWI: g‰hasto; BCTTI: g‰hì 10 ICTWI: dharmyà˙ kriyà˙ 11 BCTTI: dharmyà˙ kriyà [nà]tha iva 12 ICTWI: sabràhmana˙ 13 BCTTI & ICTWI: °vaá≤a° 14 ICTWI: dvayìsu 15 ICTWI: nareádra° 16 ICTWI & BCTTI: malaye 17 ICTWI: [soma?] vedeßu; BCTTI: sa[ma]stavedeßu 18 ICTWI: valùranàmnà; BCTTI: vall[ù]ranàmà, but suggests reading vallùranàmnàá 19 ICTWI: rave˙ 20 CII: °ràmidhà°, a printer’s error. 21 ICTWI: tadàtmaja˙; BCTTI: tadàtmaja˙ 22 ICTWI: kirttir 23 ICTWI: sakìrtir 24 ICTWI & BCTTI: gu»ai[ßiko≤o] 25 BCTTI: adyàpi . tasyàbhimana . . . 26 BCTTI: sahot° 27 ICTWI: . . . kàryyaá n‰p°; BCTTI: . . . [yaá] n‰p° 28 ICTWI: purußàkßa . . .; BCTTI: surasà 29 BCTTI: devaràjasya . . . 30 ICTWI: °dhavalàni; BCTTI: ≤a≤ìkaradhavala [?] 31 ICTWI: °bh‰to . . . 32 ICTWI: phalamàkhilamila . . .; BCTTI: phale [makhilamila ?] 33 BCTTI: °vibhàvì Inscription 99 Cave: Location: Medium: Type: Editions: Copies:
Gha†otkaca Vihàra, front left pilaster Incised c. 8th century, intrusive, religious creed GCI: i GCI: Plate 12 (very sketchy, not particularly useful)
Text L1: 1ye dharmmà hetupra(bhavà he-) L2: [tuá teßàá] tathà(gato) hy a(vadat) L3: . . . L4: . . . Text Note 1 This likely read: ye dharmmà hetuprabhavà hetuá teßàá tathàgato hy avadat {|} teßàá ca yo nirodha evaávàdì mahà≤rama»a˙ {||}
Translation The Tathàgata teaches the cause of those things that arise from a cause, and also their cessation. Such is the Great •rama»a’s doctrine.
HANDBBOOK OF ORIENTAL STUDIES Sect. II: INDIA ISSN 0169-9377 Band 1. Die indischen Sprachen 1. Gonda, J. Old Indian. 1971. ISBN 90 04 02642 8 Band 2. Literatur und Bühne 1. Zvelebil, K.V. Tamil Literature. 1975. ISBN 90 04 04190 7 Band 3. Geschichte 3. Scharfe, H. The State in Indian Tradition. 1989. ISBN 90 04 09060 6 Band 4. Religionen 1. Gonda, J. Vedic Ritual. The Non-Solemn Rites. 1980. ISBN 90 04 06210 6 3. Schimmel, A. Islam in the Indian Subcontinent. 1980. ISBN 90 04 06117 7 Band 7 Fürer-Haimendorf, C. von. Tribal Population and Cultures of the Indian Subcontinent. 1985. ISBN 90 04 07120 2 Band 8/1 Verhagen, P.C. A History of Sanskrit Grammatical Literature in Tibet. 1. Transmission of the Canonical Literature. 1994. ISBN 90 04 09839 9 Band 8/2 Verhagen, P.C. A History of Sanskrit Grammatical Literature in Tibet. 2. Assimilation into Indigenous Scholarship. 2001. ISBN 90 04 11882 9 Band 9 Zvelebil, K.V. Lexicon of Tamil Literature. 1995. ISBN 90 04 10072 5 Band 10 Driem, G.L. van. The Languages of the Himalayas. An Ethnolinguistic Handbook. 2001. ISBN 90 04 10390 2 Band 11 Willemen, C., Dessein, B. and Cox, C. Sarva-stiva-da Buddhist Scholasticism. 1998. ISBN 90 04 10231 0 Band 12 Brockington, J. The Sanskrit Epics. 1998. ISBN 90 04 10260 4 Band 14 Cahill, T.C. An annotated Bibliography of the AlaÒk§raá§stra. 2001. ISBN 90 04 12133 1 Ergänzungsband 1 Gaeffke, P. Hindiromane in der ersten Hälfte des zwanzigsten Jahrhunderts. 1966. ISBN 90 04 00871 3 Ergänzungsband 2 Gaeffke, P. Grundbegriffe moderner indischer Erzählkunst aufgezeigt am Werke Jaya- a«kara Prasa-das (1889-1937). 1970. ISBN 90 04 00872 1 Ergänzungsband 3 Mishra, V.B. Religious Beliefs and Practices of North India During the Early Mediaeval Period. With a Foreword by A.L. Basham. 1973. ISBN 90 04 03610 5 Ergänzungsband 4 Bhatia, T.K. A History of the Hindi Grammatical Tradition. Hindi-Hindustani Grammar, Grammarians, History and Problems. 1987. ISBN 90 04 07924 6 Ergänzungsband 5 Zvelebil, K.V. Companion Studies to the History of Tamil Literature. 1995. ISBN 90 04 09365 6
Band 16 Scharfe, H. Education in ancient India. 2002. ISBN 90 04 12556 6 Band 17 Behrendt, K.A. The Buddhist Architecture of Gandh§ra. 2004. ISBN 90 04 13595 2 Band 18/1 Spink, W.M. Ajanta: History and Development. Vol. 1: The End of the Golden Age. 2005. ISBN 90 04 14832 9/978 90 04 14832 1 Band 18/2 Spink, W.M. Ajanta: History and Development. Vol. 2: Arguments about Ajanta. 2006. ISBN 90 04 15072 2/978 90 04 15072 0 Band 18/3 Spink, W.M. Ajanta: History and Development. Vol. 3: The Arrival of the Uninvited. 2005. ISBN 90 04 14833 7/978 90 04 14833 8 Band 18/4 Spink, W.M. Ajanta: History and Development. Vol. 4: Painting, Sculpture, Architecture. Forthcoming. ISBN 90 04 14983 X/978 90 04 14983 0
