Slippage in Buddhist Geography, by Todd Gibson

Swat River, Pakistan - Wikicommons

“Slippage” in Buddhist Geography: 

Orissa, Bengal, Kashmir, and Nepal as Sacred Proxies

Todd Gibson

Note: Today’s blog is an original essay by Todd Gibson. It is extracted from a chapter of his forthcoming book Inner Asia and the Nyingmapa Tradition of Tibet: The Case of Shri Singha.

 

Abstract: The reconstruction of the history of Buddhism in India and Tibet is complicated by the often-haphazard approach to geography found in the Indic sources, and a changing understanding of Indian names in the Chinese material. It is becoming apparent, however, that the shifts in perception of what geographical terms referred to was not always due to a mere lack of adequate information; the article discusses four cases which demonstrate deliberate, ideologically-motivated relocation of sacred sites away from the Indian border areas in the wake of historical changes.

It has long been noticed that geographical accuracy was not a great concern for many Indian scholars of the past. In his masterly survey of esoteric Buddhism in India, Davidson notes (2002, p. 33) that “geographical terms are used in a hazy and imprecise manner” and “alternate names are encountered with alarming suddenness in epigraphs and literature”. As a result, “The problem of the relationship between designation and locale can be acute, especially in medieval Buddhist literature.” This relationship is even more tenuous in Indian accounts of the countries of the mleccha peoples, those which lay beyond the civilized area of Āryavārta.

Chinese sources dealing with Indian Buddhist history must also be used with caution. When Buddhism first began to appear in China, the Chinese had only the haziest geographical understanding of the Indian subcontinent. As a result, the names of Indian locales in Chinese sources altered over time, sometimes (but not always) reflecting an improved understanding of the facts on the ground. Another factor that must be taken into account in weighing geographical references in Chinese Buddhist accounts is the veneration of central India as the holy land of Buddhism, and the increasing desire to distinguish the Buddhism of India from that of the Inner Asian peoples who were China’s neighbors, and who had an often-troubled relationship with the Chinese.[1] These factors, however, do not tell the whole story; creative geography in the service of religious legitimation that is found in some sources (both Indian and Tibetan), is a factor that has not yet been well examined. This article treats four cases in which this factor contributed to several persistent but inaccurate identifications of sacred lands.

The first of these cases is the country of Odiyana.[2] The current scholarly consensus in the West is that the name (in Sanskrit: Uḍḍyāna, Uḍḍiyāna, Oḍḍiyāṇa, Uḍyāna,[3] etc.; in Tibetan: Urgyan, Orgyan), narrowly defined, is conterminous with the region of Swat in the north of present-day Pakistan, a valley west of the Indus and separated from the Indus valley by a single range of mountains. This identification, based first on the testimony of Chinese pilgrims such as Xuanzang,[4] was accepted by early scholars such as Aurel Stein and Giuseppe Tucci,[5] and has been supported by the vast number of Buddhist archaeological data that are still coming to light.[6] Their opinion been followed by the majority of Buddhologists since (dissenting opinions will be discussed below), and can be accepted with some important qualifications.

As is the case with so many toponyms in India and elsewhere, Uḍḍiyāna has not been an entity with fixed boundaries throughout its history. Upasak (1990, pp. 20- 24) has asserted that in the early Indic sources, it designated modern eastern Afghanistan while Kapiśa meant that country’s central region, and Bālhīka its north. He cites Ptolemy as indicating that the name included the whole region to the north and west of the realm of Gandhāra, centered on what is now Jalalabad, and including the Swat valley, but adds that by the fourth or fifth century, Uḍḍiyāna sometimes meant only the eastern part of this region, up to the Indus, while the land from the Kabul River to the Khyber Pass was called Nagarahāra. In considering references to Odiyana, particularly in Chinese sources, it is often necessary to distinguish between these meanings. 

While among Buddhologists, the view now seems to be that, when discussing Buddhist esoterism, Odiyana is indeed to be located in Swat, there have been attempts to claim that the Oḍḍiyāna of the Sanskrit Buddhist sources refers not to the Eastern Afghanistan or Swat regions, but rather the area of Orissa/Odishya (Oḍra) in eastern India. Among modern Indian scholars, this may even be the majority[7] view, but some Western commentators have also thought among similar lines. Huntington (1975, p. 8 n. 12), for example, claimed that “there is substantial indication in Tāranātha’s History of Buddhism in India . . . that the Uḍḍiyana of the 8th and 9th centuries would have been in the Orissan or the Orissan-Bengal border region.” He bases this evaluation on the idea that most of the mahāsiddhas mentioned in the History were supposedly active in this area.[8] On the other hand, Tāranātha closes his book (Chimpa and Chattopadhyaya, 1970, p. 351) by saying that he was not able to include an account of how Buddhism spread to Kashmir, Odiyana (Urgyan), Tokharistan, or South India, for lack of detailed source material. This seems to argue against a location in Orissa. In any case, Tāranātha composed his work in the 17th century, and some of his Indian sources may well have placed Uḍḍiyāna in east-central India.

In his translation of the “official” biography of the master Shan wu wei, the first of the recognized patriarchs of the main East Asian Buddhist tradition, Chou (1945, p. 252 n. 4), notes that the same Chinese logograph can be used to represent Odiyana and Orissa/ Odishya, and clearly this was one of the reasons for a confusion between the two.[9] A possible example of this confusion is found in the history of a major sutra, the Avataṃsaka, in China. The first complete translation of this sutra into Chinese was carried out in 420 by Buddhabhadra, a monk from Afghanistan,[10] who had obtained his copy from the king of Karghalik, just west of Khotan (Hamar, 2013, p. 85). Similarly, another Afghan monk, Prajñā, who hailed from Kapiśa,[11] brought and translated a late version of the Gaṇḍavyūha chapter of the Avataṃsaka to China. The mainstream tradition accepts that Prajñā received the scripture from the king of Orissa, but given the history of the earlier version and Prajñā’s own provenance, it seems more likely that he obtained the scripture from a ruler of Odiyana in the larger sense (i.e. the eastern Afghanistan or Swat area). As for Shan wu wei himself, Chou assumes — for no clear reason — that he was from Orissa, but given the documented association of Odiyana with esoteric currents in Buddhism from early times[12] the subject deserves a thorough revisiting.[13]

In the present case, it is likely that the perceived connection with Orissa is not only a matter of an honest misunderstanding taken up and repeated, but also reflects certain later Buddhists’ attempts to locate all milestones of Buddhist history, especially its esoteric aspects, well within the bounds of the Indian subcontinent. This assertion is borne out by other cases in the later Tibetan (and Chinese) Buddhist literature in which sites in Inner Asia and the northwestern border areas of the Indian subcontinent have evidently been “moved” to India proper. Particularly noticeable in this regard is the case of a country called Zahor. A certain king of Zahor, Indrabhūti (also sometimes known as Tsa, or Dza in the Tibetan tradition), is regarded by the Nyingmapas, the oldest Buddhist school in Tibet, as having transmitted the tantric literature known to them as Mahāyoga. This tradition usually locates Indrabhūti’s kingdom in Odiyana, but sometimes in Zahor.[14] An Indrabhūti also appears in the historical accounts of the newer Tibetan schools as the source of their so-called Yogatantra literature, including the fundamental tantra called the Compendium of Buddhas (Sarvatathāgatatattvasaṃgraha).[15] The confusion brought about by the Tibetan attempts to make a coherent whole out of a variety of conflicting accounts of this literature’s history has resulted in various versions, in which one or two Indrabhūtis are posited, and Odiyana is considered to be either the same as or different from Zahor.[16] A few scholars (Davidson, 2002, pp. 242-245; Karmay, 2009, pp. 76-93; Van der Kuijp, 2010) have attempted to deal with this tangle of material, but it is clear that the legend is so fraught with textual corruption and so tied up with issues of religious legitimation that a definitive description of the evolution of these traditions will probably never be had. Nevertheless, from the point of view of the present work, it is significant that the Nyingmapa almost invariably locate Zahor (whether identified with Odiyana or not) in the northwest,[17] while the newer Tibetan schools in general support the idea that Zahor was in eastern India, probably Bengal (Karmay, 2009, pp. 80- 81; van der Kuijp, 2010, p. 148, citing Drigung Paldzin). Karmay has concluded from this (ibid., p. 89) that “Just as the later Tibetan tradition made the first king of Tibet an Indian, in the same way it also desired to connect king Tsa /Dza with the land which gave birth to Buddhism.”

A third case of geographical confusion or dissimulation can be seen in the region referred to in the Chinese sources as Jibin. Many or most Buddhist scholars (and some secular historians) have traditionally regarded the word as referring to Kashmir (cf. Zurcher, 2012), because the Chinese word represents an early attempt at transliteration of that name. Kuwayama, however, has noted (2006, p. 110; see also Kuwayama, 2002, pp. 142-146) that “In ancient Chinese accounts and maps Jibin’s location shifted from place to place as the Chinese geographical knowledge changed over time.”[18] Kuwayama uses the travel accounts of the Chinese Buddhist pilgrims, as well as monastic biographies of fourth- and fifth-century figures to demonstrate that for them, Jibin referred to the general area of Gandhāra, the pilgrimage region where the Buddha’s begging bowl was kept; he shows that on the routes these pilgrims took to and from China, any visit to Kashmir would have represented an extensive detour. More recently, Falk (2022-23, esp. n. 36) has also taken up the question of the Chinese pilgrimage routes, and has asserted that for Faxian, Jibin referred very specifically to the area around Hadda in present-day Afghanistan.

Kuwayama (2006, p. 108) also says that “In the Gaoseng zhuan” — a sixth-century Chinese compilation of Buddhist biographies — “Jibin surpasses any other region of India as the goal of Buddhist pilgrimage, and most of the foreign monk-translators in fourth-to-fifth century China were closely associated with Jibin, whether or not they were natives of that area.” Many celebrated teachers and translators from Inner Asia were trained in the Jibin of this era. The first of these was Fotucheng, who afterwards went to China, arriving in Luoyang in 310. Although Kuwayama believes he was a native of Gandhāra, an earlier, more detailed treatment of his career concluded he was probably originally from Kucha, and only studied in the former region.[19] His impact on Buddhism in China was considerable; he was able to increase the acceptance of Buddhism among both the upper classes and the common people mostly through practical means like rainmaking, war magic, and medicine, at a time of “chaos and misery”, as Wright puts it. This acceptance meant Fotucheng was able to lay the foundation for a future state-supported Buddhism, besides passing on his knowledge to disciples, who came to China from as far away as India and Sogdiana to study with him (Wright, 1948, p. 367).

The next illustrious alumnus of Jibin’s Buddhist schools was the great translator Kumārajīva (344-413). Like Fotucheng, he was from Kucha, and was taken as a child to be educated in Gandhāra.[20] After a few years, he returned to Kucha, but was kidnapped by an army from China and taken to Liangchou, where he remained for almost twenty years. Upon the capture of that city by the Later Qin dynasty, however, Kumārajīva was taken to Changan. Since by that time he knew not only his native Kuchean language, but also Northwestern Prakrit, Sanskrit, Chinese, and possibly Agnean and Sogdian (Hansen, 2012, p. 56), he was put to work at the head of the bureau that was rendering Buddhist scriptures into Chinese, and became the most prolific translator before Xuanzang. His renderings of many sutras remain the preferred versions up to the present, and he trained many Chinese disciples who became famous in their own right. Two other translators from Gandhāra, Buddhayaśas and Bhīmarakṣa, went to China to aid Kumārajīva in his work.

In sum, while there are some Chinese materials in which Jibin does refer to Kashmir, the identity can no longer be taken for granted, particularly in a Buddhist context.

The fourth example of this phenomenon is so well known that there can be no controversy about it. The kingdom of Khotan, stronghold of Mahāyāna Buddhism for over five hundred years, was well known to Tibet from at least the beginnings of the imperial period under the name of “the Li Country” (Li Yul), which name is found among some of the oldest surviving writings in Tibetan. Tibet conquered Khotan between 665 and 670, and solidified their rule in 675. Khotan maintained close contact with the Dunhuang area before and during the Tibetan imperial period, and many later documents in Khotan Saka were found at Dunhuang. Until the coming of Islam to Khotan in 1006, the three areas had strong cultural connections, and the religious chronicles of Khotan were translated into Tibetan. Nevertheless, by the twelfth or thirteenth century, some Tibetans had begun writing as if Li Yul referred to Nepal (in effect, to the Kathmandu Valley) (Snellgrove, 1987, pp. 366, 417; first remarked on by Brough, 1948).[21] It is possible that the relationship with Khotan had gradually been forgotten in Tibet during the institutional chaos that followed the fall of the Tibetan empire, but it is also likely that when the Tibetans began to reassemble their memories of that time, some of them had a vested interest in associating the country with India rather than the north. 

It is well known that sacred sites can “migrate” as part of either the spread or the domestication of a religion; the reduplication of sacred Indic place names throughout Southeast Asia was at least meant to establish a connection with India, and probably also served to supply surrogate pilgrimage sites to those who could not make the journey to the subcontinent. Sacred places may be also moved in the popular imagination, even to unapproachable areas, when the situation on the ground precludes their existence in their former location; the mythic land of Shambhala is probably the best-known example. However, the converse seems to be true as well. As Grenet has shown in relation to Zoroastrianism,[22] a religion may move its sacred sites geographically “inward” in order to be in harmony with changes in the prevailing politico-religious situation. In point of fact, it can hardly be coincidence that three of the four areas discussed above were “relocated” to the single area in India (Bihar-Bengal-Orissa) where Buddhism was still prospering and perhaps even predominant in the ninth century and after, following the Brahmanic Hindu resurgence and Muslim conquests elsewhere (Sanderson, 2009, p. 80). The fourth area, Kashmir, was also able to maintain the Buddhistic aspects of its culture much longer than was Gandhāra. There was probably a felt imperative to situate the holy places of Buddhism, particularly Vajrayāna Buddhism, within the small remaining Buddhist heartland on the subcontinent, and this resulted in the imaginative shift seen in later Indian accounts, in which all important developments were traced back to this area. The Tibetans, who for a variety of reasons became more and more inclined to equate scriptural authenticity with Indian origin, were for the most part happy to fall in with this reading of history.

•

Notes

[1] Chinese Buddhists increasingly differentiated the people of India (fan) from those living to the west of China (hu), and only the former land was held by them to be the domain of true sages. For a treatment of this trend, see Yang, 1988.

[2] I use a phonetic representation of the name rather than the more customary Sanskrit because the earliest named people of the area, the Odis, were neither Sanskrit speakers nor culturally “Sanskritized.” Settlements in the Swat valley date back to the Chalcolithic period, obviously predating the Indo-European incursions to the Indian subcontinent. The Odis themselves first appear in the historical record as satraps of a Saka ruler. Pāṇini, who was himself from the area of modern Peshawar, knew the name as Urḍi or Aurḍayāni (Upasak, 1990, p. 20). As late as the composition of the Purāṇas and the great epic Mahābhārata, Odiyana and its northern and western neighbors were not considered part of Āryavārta (Bronkhorst, 2016, pp. 17-34, pp. 124-25). In sum, it is misleading to insist on the Sanskritized spelling.

[3] But now see Falk (2022, pp. 17-19), who rejects the reconstruction of the name as Uḍyāna.

[4] Xuanzang’s name for the region was previously reconstructed as Wu chang, but this too seems to have been a scholarly error; see Li (1996, p. 82).

[5] Tucci’s 1977 contribution was the first to point out the importance of this area in detail, and pioneered the ongoing work by the Italian, Pakistani, and Japanese archaeological expeditions that have contributed decisively to its study.

[6] Callieri (2006); Neelis (2011). Neelis (ibid., p. 245) speaks of the valley’s “especially rich archaeological, artistic, epigraphic, and literary heritage with seemingly innumerable remains of Buddhist stupas, monasteries, and rock carvings.” The earliest Buddhist site, Butkara I, dates back to the third century BCE.

[7] See Mohanti and Panigrahi (2016) for a typical presentation. As with most such claims, their assertions rely largely on very late Indian and Tibetan literary sources. Chandra (1980) even proposed Kanchipuram, in southern India, as an alternative. 

[8] While Orissa was without doubt a major Buddhist stronghold in the ninth century, Tāranātha’s evaluation of the Buddhism prevailing there at that time has been called into question (Kinnard, 1996, pp. 284-287). 

[9] The confusion is not found only in Chinese sources. Sircar (1973, pp. 12- 13) has noted that in the Kālikā Purāṇa, Oḍra is inconsistently substituted for Oḍḍiyāna as one of the four major tantric seats, but he concludes that in relation to the Hindu tantric literature, the notion that the seat in question was actually located in Orissa is “unworthy of serious consideration.” Davidson (2002, pp. 206-211) also discusses the confusion and inconsistency surrounding the location of these seats.

[10] The usual reckoning has Buddhabhadra as Kashmiri, or just “Indian”, but Kuwayama (2002, p. 146; 2006, p. 109) shows that he was from Nagarahāra, specifically the area between Jalalabad and the Khyber Pass.

[11] Copp (2011, p. 360).

[12] As has been often noted, Xuanzang recorded the emphasis that the people of Swat laid on spell literature and meditation; see Li (1996, p. 83-84). 

[13] It should be taken into account that the attempts to connect this master with a later royal family of Orissa — the so-called Bhauma-kara dynasty — are based on a reconstruction of his Sanskrit name (Śubhakarasiṃha) that can in no way be considered equivalent to Shan wu wei (Chou, ibid., p. 251 n. 3). Furthermore, Shan wu wei’s official biography was written by at least two different people (Chou, ibid., pp. 250-251), and while the first half, dealing with his purported birth and career in India, contains the usual tropes of royal descent and miraculous but otherwise unknown teachers, the second begins with his travels towards China and is supported throughout by much historically verifiable information, including the rule of a Türkic khan in Odiyana (in the expanded sense) when he passed through. 

[14] The connection between Indrabhūti and Odiyana seems to date back at least to around 800 (van der Kuijp, 2010, p. 130).

[15] Karmay (2009, p. 80-82). There is a close relationship between a collection of these tantras, which were brought to China by the Sogdian master Amoghavajra, and the tantras of the Mahāyoga, though the two collections are not identical. This was first noticed by Eastman (1981), and the similarities have more recently been taken up and analyzed by Giebel (1995) and Almogi (2014). The East Asian tradition claims that Amoghavajra’s collection was originally taken from a legendary Iron Stupa that was in South India. That the tradition does not claim a provenance in the Orissa-Bengal area might be because the legend of the Iron Stupa was already established in China before the shift in Zahor’s supposed location came about. 

[16] According to Davidson (2002, p. 244), however, the Sakyapas, oldest of the “new” schools, have a tradition of three Indrabhūtis – all from Odiyana.

[17] The Fifth Dalai Lama, who came from a Nyingmapa background, nevertheless felt compelled to invent a convoluted history of his own Zahor ancestry so that he could trace it back to Bengal (van der Kuijp, 2010, pp. 147 ff.). An interesting sidelight to this question is the fact that the traditional Tibetan account of the creation of their alphabet claims that they took the letter za from Zahor (van der Kuijp, 2010, p. 137). While this is usually understood to mean that they took the form of the letter from the alphabet used in that region (which was not the case), it might mean instead that they had to find a way to indicate the voiced sibilant /z/ which is found in Tibetan words, but not in Indic; the /z/ was a native phoneme in Northwestern Prakrit, but only used in “Iranian names and loanwords” in Sanskrit (Salomon, 1990, p. 269). This would also obviously indicate a northwestern location for Zahor.

[18] See Kuwayama (2006, p. 110 n, 14) for a discussion of various reckonings on the question. Kuwayama (2002, pp. 193-199) had earlier used the testimony of later pilgrims, as well as Chinese historical materials, to demonstrate that during the Tang, Jibin came to mean the kingdom of Kapiśa in what is now central Afghanistan. 

[19] See Wright (1948, pp. 332-335) for a discussion. Wright errs, however, in following the then-current consensus that Jibin always referred to Kashmir. Some sources claim that Fotucheng was a native of Central India and a descendant of royalty, but the trope of the royal (or high-caste) descent of prominent Buddhist figures found in Chinese biographical literature is far too commonplace to be accepted at face value, and can be seen as part of the hu vs. fan dynamic mentioned above (note 2).

[20] While McRae (2004, p. 442) continues to place Kumārajīva’s studies in Kashmir, Hansen (2012, p. 66) correctly locates them in Gandhāra. As usual, there also exist accounts that make Kumārajīva the descendent of Indian royalty (Hansen, op. cit.).

[21] Tāranātha was one Tibetan who placed Li Yul south of the Himalayas (Chimpa and Chattopadhyaya, 1970, p. 60, n. 41; see also Brough, 1947-48, p. 338).

[22] Grenet (2015, esp. p. 26) demonstrated that while the sites described in the oldest Avestan literature as being created by Ahura Mazda are all located in southern Inner Asia, Afghanistan, or western Pakistan, the later Pahlavi literature moved them to the Iranian plateau.

SOURCES

Almogi, Orna (2014) “The Eighteen Mahāyoga Tantric Cycles: A Real Canon or the Mere Notion of One?” Revue d’Etudes Tibétaines 30, pp. 47-110.

Bronkhorst, Johannes (2016) How the Brahmins Won: From Alexander to the Guptas, Brill, Leiden.

Brough, John (1948) “Legends of Khotan and Nepal,” Bulletin of the School of Oriental and African Studies 12, pp. 333-339.

Callieri, Pierfrancisco (2006) “Buddhist Presence in the Urban Settlements of Swat, Second Century BCE to Fourth Century CE,” in Gandhāran Buddhism: Art Archaeology, Texts, ed. P. Brancaccio and K. Behrendt, UBC Press, Vancouver, pp. 47-82.

Chandra, Lokesh (1980) “Oḍḍīyāna: a New Interpretation,” in Tibetan Studies in Honour of Hugh Richardson, ed. Michael Aris and Aung San Suu Kyi, Aris & Phillips, Warminster, pp. 73-78.

Chimpa, Lama and Alaka Chattopadhyaya (1970) History of Buddhism in India (by Tāranātha), Indian Institute of Advanced Study, Simla.

Chou, Yi-liang (1945) “Tantrism in China,” Harvard Journal of Asiatic Studies 8, pp. 241-331; partially reprinted in Payne, 2006.

Copp, Paul (2011) “Prajñā” in Esoteric Buddhism and the Tantras in East Asia, ed. Charles Orzech, Handbuch der Orientalistik series no. 24, Brill, Leiden, pp. 360-362.

Davidson, Ronald (2002) Indian Esoteric Buddhism, Columbia University Press, New York. 

Falk, Harry (2022-2023) “Faxian and Early Successors and Their Route from Dunhuang to Peshawar: In Search of the ‘Suspended Crossing’,” Bulletin of the Asia Institute n.s. vol. 31, pp.1 -48.

Giebel, Rolf W. (1995) “The Chin-kang-ting ching yü-ch’ieh shih-pa-hui chih kuei: an Annotated Translation,” Journal of Naritasan Institute for Buddhist Studies 18, pp. 106-199.

Grenet, Frantz (2015) “Zarathustra’s Time and Homeland: Geographical Perspectives” in The Wiley Blackwell Companion to Zoroastrianism, ed. Michael Stausberg and Yuhan Sohrab-Dinshaw Vevaina, Wiley Blackwell, Chichester, pp. 21-30.

Hansen, Valerie (2012) Silk Road: A New History, Oxford University Press, Oxford.

Hamar, Imre (2007) “The History of the Buddhāvataṃsaka-sūtra: Shorter and Larger Texts” in Reflecting Mirrors: Perspectives on Huayen Buddhism, ed. Imre Hamar, Harrassowitz Verlag, Weisbaden, pp. 151-178.

Huntington, John C. (1975) The Phur-pa, Tibetan Ritual Daggers, Artibus Asiae Publishers, Ascona (Switzerland).

Karmay, Samten (2009 [first published 1997]) The Arrow and the Spindle: Studies in History, Myths Rituals and Beliefs in Tibet, vol. 1, Mandala Book Point, Kathmandu.

Kinnard Jacob N. (1996) “Re-evaluating the Eighth-Ninth Century Pala Milieu: Icono-Conservatism and the Persistence of Śākyamuni,” Journal of the International Association for Buddhist Studies 19.2, pp. 281-300.

Kuijp, Leonard W. J. van der (2010) “On the Edge of Myth and History: Za hor, its Place in the History of Early Indian Buddhist Tantra, and Dalai Lama V and the Genealogy of its Royal Family,” in Studies on Buddhist Myths: Texts, Pictures, Traditions, and History, ed. Bangwei Wang, Jinhua Chen, and Ming Chen, pp. 114-164.

Kuwayama, Shoshin (2002) Across the Hindukush of the First Millennium, Kyoto University Institute for Research in Humanities, Kyoto. 

—— (2006) “Pilgrimage Route Changes and the Decline of Gandhāra” in Gandhāran Buddhism: Art, Archaeology, Texts, ed. P. Brancaccio and K. Behrendt, UBD Press, Vancouver, pp. 107-134.

Li, Rongxin (1996)(trans. Xuanzang) The Great T’ang Dynasty Record of the Western Regions, Numata Center for Buddhist Translation and Research.

McRae, John R. (2004) “Kumārajīva” in Encyclopedia of Buddhism, ed. Robert Buswell, Macmillan Reference, New York pp. 442-43.

Mohanty, Bimalendu and Varish Panigrahi (2016) “Guru Padmasambhava of Odiyana (Odisha): The Founder of Lamaism in Tibet,” Journal of Bhutan Studies 34, pp. 80-86.

Neelis, Jason (2011) Early Buddhist Transmission and Trade Networks, Brill, Leiden.

Payne, Richard C. (2006)(ed.) Tantric Buddhism in East Asia, Wisdom Publications, Somerville MA.

Sanderson, Alexis (2009) “The Śaiva Age: the Rise and Dominance of Saivism in the Early Medieval Period” in Genesis and Development of Tantrism, ed. Shingo Einoo, Institute of Oriental Culture, University of Tokyo, Tokyo, pp. 41-349.

Sircar, D.C. (1973) The Śākta Pīṭhas, Motilal Banarsidass, New Delhi.

Snellgrove, David (1987) Indo-Tibetan Buddhism: Indian Buddhists and Their Tibetan Successors, 2 vols., Shambhala Publications, Boston.

Tucci, Giuseppe (1977) “On Swat. The Dards and Connected Problems,” East and West n.s. 27, pp. 9-85.

Wright, Arthur Frederick (1948) “Fo-tu-teng: A Biography,” Harvard Journal of Asiatic Studies 11, pp. 321- 371.

Yang, Jidong (1998) “Replacing hu with fan: A Change in the Chinese Perception of Buddhism during the Medieval Period,” Journal of the International Association of Buddhist Studies 21.1, pp. 157-170.

Zurcher, Eric (2012) “Buddhism across Boundaries: the Foreign Input,” in Buddhism Across Boundaries: The Interplay of Indian, Chinese, and Central Asian Source Materials, ed. John MacCrae and Jan Nattier, Sino-Platonic Papers, pp. 1-25.

South India in Tibetan Geography

Shree Ayyappa

It was only a few years ago that I first visited South India. I wonder what took me so long. Today I am wishing I had taken a camera with me. All my earlier Indian travels had been in the north, mostly in the Gangetic Plains, Delhi area, and Himachal Pradesh. But I’ve felt an increasing interest in South India that can in some part be explained by my personal liking for some special persons who study the area or who originated there. And all this knits together in an inexplicable way with my interest in Padampa Sanggyé, and the ever-increasing conviction that I may need to know more about South India in order to understand him better. 

I believe Padampa lived his young life somewhere in the coastal regions of Andhra Pradesh, probably in the mouth of the Krishna River. My reasons for thinking so are several. The most obvious indications in early sources are that he must have lived quite close to Shri Parvata (Dpal-gyi Ri), by which I understand the Shri Parvata that is in  the same area as Dhanyakataka Stūpa in Amaravati (there is this wiki with a map location if you are curious). We also have (less specific) testimony in the frequency of sea turtles in his metaphors, along with the story that his father was a sea captain. A number of early sources say that he was from Be-ta’i Yul. Some have interpreted this to mean the kingdom of Vidarbha, but I don’t believe this is necessarily correct. I think the name Be-ta Country would just mean Coconut Country, as a cover-all term for the entire South Indian region, where coconuts are one of the most important items in the cuisine.  Be-ta is the Tibetan word for the coconut palm.* 

(*Berthold Laufer’s remarkable work on Loanwords, known to readers of earlier blogs, discusses this on pp. 457-8, but it can be found in quite a few modern English-Tibetan dictionaries under ‘coconut.’ I say this for the benefit of those who might pride themselves on their unwillingness to trust me.) 

I’m eager to read Karl Brunnhölzl’s new book Straight from the Heart, supposedly already in the mail. Among many other things, it has what looks like a very substantial chapter on Padampa entitled Padampa Sangyé’s Meetings with Milarepa and the Nun Düdsi Gyi. It can be seen in part at Googlebooks. There (p. 203) he gives the country’s name as Veta and says that it was in a particular part of that land in a place called Carasimha that Padampa was born. With no footnote or other identification, both place names are left drifting across the page as blank signifiers, with not much of a clue as to where we might be, although it does say, without clueing us in on how anybody would know this, that he was a South Indian.

I’m not ready to rule out the Vidarbha idea quite yet. It was more-or-less synonymous with ‘the south’ in the Mahābhārata, but south in those days didn’t seem to go very far on the map. (And nowadays the name Vidarbha is limited to eastern Maharashtra with its population of Marathi speakers — try this wiki.) Anyway, ‘coconut country’ could be equally descriptive of the Kingdom of Vidarbha and of South India as a whole. Still more mysterious is the identity of Padampa’s hometown Tsa-ra-sing-ha. I’ve found some new sources that might shed light on that, but it’s not too relevant at the moment, and would provoke some lengthy technical arguments. Another time, perhaps.

Today I will limit myself to a fairly simple question. What general knowledge did educated Tibetans of the past have about South Indian geography?

No sooner is this simple question framed than the doubts and complications raise their tiny heads to assert their importance. If names are preserved in books, it doesn’t mean that people necessarily knew about them or knew how those names ought to be found on the ground. Let’s keep those tiny hydra heads at bay and try heading for the big picture based on what textual sources we have, in an attempt to avoid being ruled by our overactive imagination.

Butön (Bu-ston Rin-chen-grub, 1290‑1364 CE) was one of the most intelligent among the leading Tibetan historical personages, no doubt about it.  Tibetan iconography normally depicts him with an abnormally shaped head, extremely wide at the temples with a huge cranial bump, but narrowing sharply downward to his jutting v-shaped chin.  (As you can see, I’m trying to get a job writing art show catalogues.)

 Butön

I find myself entirely unable to make up my mind whether [1] the cranial bump is an iconographic convention to denote superior intelligence or [2] this way of depicting him his based on his actual physical appearance, which is not to neglect the possibility the answer might be [3] both (see the Heller book).

In volume 24 of the old Lokesh Chandra edition of his collected works (that’s right, he wrote more than 24 volumes of collected works!), at pages 866-7, is one of the most interesting of classical Tibetan lists of countries. This is part of his famous history composed in 1322 CE. I’ll give this list in a moment, with some of the countries identified in parentheses. I believe all of these country names are names of real countries, and No, that really does not go without saying. If a few of them might not be immediately identifiable with real countries, my assumption is that we need to go back to the earlier sources for parts of the list and think harder until we come up with the answers.  (That’s right.  I’ve got faith.  But I feel I have good reasons to have faith.) You will soon see that I give links to Wikipedia articles, which are, you know, sometimes not exactly on the cutting edge... But I believe many will benefit from having the quick reference, on the understanding that no seal of approval has been applied from my side.

First he gives a list of a few places in which gigantic versions of tantras are said to exist: 

1. lha’i gnas (divine abodes).

2. Shambha-la (Shambhala).

3. U-rgyan (Oḍḍiyāna), etc.

Then he gives a much longer list of countries where the Buddha-Dharma has spread at some time or another in one form or another:

4. Rgya-gar (India).

5. Kha-che (Kashmir).

6. Bal-yul (Nepal Valley).

7. Li (Khotan).

8. Rgya-nag (China).

9. Rgya-nag Chen-po (Greater China, or China together with Tang Dynasty conquests).

10. Par-sig (Persia).

11. Tsam-pa-ka (Campaka).

12. Spre’u (this being Tibetan for ‘monkey,’ but most likely intending Sanskrit Vānara).

13. Gser-mig (This being Tibetan for ‘gold eye,’ but probably a misreading of Skt. Suvarṇākhya as Suvarṇākṣa; to simplify, this just means the ‘Gold Country’ or the country known as ‘Gold’.)

14. Rug-ma, with an added note reading Shambha-la, implying that it is part of that country.

15. Ram-ma.

16. Zangs-gling (Tibetan meaning ‘Copper Isle’), with an added note reading Rgya-gar, ‘India,’ implying that it is part of India.  It may be Sri Lanka, a part of Bengal, or a coastal area of Burma. It’s the place Sinhala was when his lamp laughed at him.

17.  Sing-ga-la’i Gling (Isle of Siṅgala), with an added note reading Rgya-gar, ‘India,’ implying that it is part of the Indian sphere.

18. Pri-yang-ku’i Gling (Isle of Priyaṅku).

19. Ya-mu-na’i Gling (Isle of Yamunā [River?]).

20. Gser-gling (‘Gold Isle’), with an added note reading Rgya-gar, ‘India,’ implying that it is part of India.  Likely to be the Śrīvijaya kingdom in Southeast Asia?

21. Zla-ba’i Gling (‘Moon Isle’).  Candra Dvīpa, sometimes identified as an island in the Ganges delta area, or a larger part of coastal Bengal.  The Das dictionary says something about it.

22. Ma-kha (Mecca).

23. Kha-sha (Khasa?).

24. Gyi-ljang (a known placename in Tibet, but not well identified).

25. Zhang-zhung (Zhangzhung kingdom of western Tibet).

26. Bru-sha (Burusho).  

27. ’A-zha (T’u-yü-hun).

28. Sum-pa (Supi).

29. Za-hor (as the birthplace of Atiśa, although disputed, this must be in area of Bengal).

30. Me-nyag (Tangut).

31. ’Jang-yul (Nan-chao).

32. Yo-gur (Uighur), with a note that says it is in the direction of China.

33. Tho-gar (Tokhar).  This ethnonym also occurs in the Hebrew Bible (Bereshith 10:3) according to Jäschke.

34. U-rgyan (Oḍḍiyāna).  This repeats no. 3 above.

35. ’Gro-lding-ba’i Yul.

36. Long-ba’i Yul.

37. Tso-la.

38. Ka-lingka.

... etcetera.

You can find a quite old, but not for that reason negligible, English translation of this.  If you have on hand a copy of E. Obermiller’s The History of Buddhism in India and Tibet by Bu-ston, look at part 2, p. 171. If you have the recently published book Thuken Losang Chökyi Nyima’s The Crystal Mirror of Philosophical Systems, The Library of Tibetan Classics series vol. 25, Wisdom (Boston 2009), p. 371, you will find a translation of a later borrowing of Butön’s list with some interesting footnotes, although the translators could have saved themselves a bit of trouble if they had known about the earlier version of the list in Butön’s history along with Obermiller’s translation of it. (I know of another copy of Butön’s list in a much-neglected mid-16th-century history.)

One significant part of our list is copied from a passage in the Vimalaprabhā commentary on the Kālacakra Tantra (chapter 1, part 4).  Following the Peking edition of the Tibetan text, these countries are: 

[1] Bod (Tibet),

[2] Rgya-nag (China),

[3] Rgya-nag-chen-po,

[4] Pār-si-ka,

[5] Tsam-pa-ka,

[6] Spre’u,

[7] Gser,

[8] Rug-ma, and

[9] Su-ra-ma. 

In each of these countries, says the Vimalaprabhā, the teachings of the Buddha have been set down in writing in their own languages (the author saw no reason to mention India here, since that would go without saying... and did; likewise for Butön’s list Tibet went without saying).  So I hope it’s clear that nos. 8-15 in Butön’s list correspond exactly in order and very closely in spellings to the earlier Vimalaprabhā list nos. 2-9. Although I have some different ideas than those he arrived at some years ago,* I would like to just send you to John Newman’s unpublished 1987 dissertation on the Kālacakra Tantra’s chapter 1 for further discussion of these particular names (on p. 362 of my UMI reprint). There is nothing better to the best of my knowledge.

(*I’m now of the somewhat supportable opinion that nos. 14-15, Rug-ma and Ram-ma both (or perhaps the latter alone) are somehow concealing the names of the two halves of early Burma Rāmañña (EoB [Encyclopaedia of Buddhism] VII 496) and Maramma ("upper Burma"), the two countries that combined to form that once-blessed country of Burma/Myanmar.)

We could go into a lot of other interesting problems, but I am eager to get to the end of the list. So let me just say, perhaps stating the obvious, that no. 17, the Isle of Siṅgala, could only mean Ceylon / Sri Lanka.  But then what is no. 18, Isle of Priyaṅku?  I believe I am the first person to identify it as the Piyaṅgu Island known to Sri Lankan sources (EoB VII 421 has an entry). I found in a Singhalese dictionary that piyaṅgu and priyaṅgu are considered variant spellings.  It is variously interpreted to mean panic seed or millet (both regarded as poor-people’s foods, known as ‘famine grains’), black mustard-seed or long pepper. Or, a tree with fragrant flowers (?). Pri-yang-ku is even known to Tibetan medicine, which would further perplex us if we were to pursue it now, which we won’t. The conclusion of the EoB entry is just that it was a small island near the northern coast of Sri Lanka.  Since anyway the monks mostly had to commute there by miraculous means, especially by flying, I imagine a better explanation may be that it is Pegu. In early times Pegu was a well-known port-of-call (true, not an island, although this doesn’t bother me much) in eastern coastal Burma (preserved in the modern city name Bago, I guess).  I would like to find good reasons to verify or reject this idea.

Okay, enough of that.  Let’s look at the last four in the list.

35. ’Gro-lding-ba’i Yul.

Translating the Tibetan, this means ‘country of hovering/soaring travel.’  Obermiller gave it in an Indic form as Dramila (a form that actually appears in a few Sanskrit texts). This means the land of Tamil speakers.  I haven’t thought of a reason why the Tibetan would be translating the name, have you?  But I’m sure that it does translate it, since it appears in this Tibetan Kanjur title:

Dramiḍa Vidyārāja (’Phags pa ’gro lding ba’i rig sngags kyi rgyal po).  Tôhoku no. 927 (also, no. 610), Dergé Kanjur, vol. E (101), folios 273‑276.  It was translated in late imperial times, or circa 800 CE.

The Tanjur contains a quite lengthy text that would appear from its title to be a Dravidian vase ritual:

’Phags pa drā bi da’i bum pa’i cho ga.  Tôhoku no. 3130.  Dergé Tanjur, vol. PU (74), folios 129‑244.

36. Long-ba’i Yul.

Translating the Tibetan, this means ‘country of the blind.’  The reason for this is not very difficult to find (and Obermiller found it long ago).  Someone at some point in the past dropped the ‘r’ from Andhra, resulting in Andha, which means ‘blind.’

37. Tso-la.

This can only mean the Chola (Cola) kingdom of south India.  Tamil in its origins, when the Chola was at the height of its power in around 1050, it controlled most of the eastern India coastal regions as well as the western parts of insular Southeast Asia.

38. Ka-lingka.

Kaliṅga was a very well-known kingdom in the history of southern India. It was centered north of Andhra in what today is Orissa.

The first two of these four would seem to correspond very nicely with the two modern states of Andhra Pradesh, with its majority of Telugu speakers, and Tamil Nadu with its majority of Tamil speakers. The two other modern states of South India, Kerala with its Malayalam speakers, and Karnataka with its Kannada speakers, may not have yet achieved identity as ethnic entities when this list was originally made. In the case of Malayalam, it could in early times have been considered a dialect of Tamil. Kannada, according to linguists, would have split off from Tamil at an even earlier date. There are about 20 more Dravidian languages, but the four I mention are today the major ones.

So at least we know that an early 14th-century Tibetan source appears to know something about the lands of South India. Like other lists incorporated into that list, the list of lands of South India is probably older than the 14th century, reflecting as it does earlier political divisions (for example, the Chola Dynasty ended in 1279). This doesn’t directly bring us closer to finding out where Padampa was coming from.  But it doesn’t hurt, either. At least we know that he, just like Nāgārjuna, was from Coconut Country, a place the Tibetan teacher Manlung Guru visited in the 13th century. That's right.  A Tibetan actually went there back in those days. But now I’m getting ahead of myself on a story that hasn’t even started yet.

And do I even need to mention that a lot of this requires further work? Please be brave and kind and write to me in the comments box if you want to criticize or discuss any of this.

Ready to read more?

Bu-ston Rin-chen-grub’s history of Buddhism is available on the world-wide web in the Wylie transcription form of Tibetan. I believe there are two versions out there.  These digital texts are extremely useful for Tibetan readers, given the ease with which you may locate Tibetan words and names in it. Limited to English? You will probably have to settle for mail ordering an Indian reprint of the 1932 Obermiller translation. The indexing is totally inadequate.

Alain Daniélou, translator, Manimekhalai (The Dancer with the Magic Bowl) by Merchant‑Prince Shattan, New Directions (New York 1989). This is an English version of the only significant Buddhist literary text that survives in Tamil, dating from around the 6th century CE or so. The translation gets mixed reviews. There may be a better one out there.

D. Martin, “Tibet at the Center: A Historical Study of Some Tibetan Geographical Conceptions Based on Two Types of Country-Lists found in Bon Histories,” contained in: Per Kvaerne, ed., Tibetan Studies, Institute for Comparative Research in Human Culture (Oslo 1994), vol. 1, pp. 517-532.

John R. Newman, The Outer Wheel of Time: Vajrayāna Buddhist Cosmology in the Kālacakra Tantra, doctoral dissertation, University of Wisconsin (Madison 1987), in 681 pages.  UMI order no. 8723348.

D.C. [Dinesh Chandra] Sircar, Studies in the Geography of Ancient and Medieval India, Motilal Banarsidass (Delhi 1971), 2nd edition.  This (plus other works by the same author) is my Bible for the historical geography of India.  You can get a good look at most of it at Googlebooks. 

Japanese readers are invited to download this PDF to find out more about a Tamil mantra in the Lotus Sutra. 

On Butön's iconography, be well advised to go and have a look at plates 63 and 64 in Amy Heller's book Tibetan Art: Tracing the Development of Spiritual Ideals and Art in Tibet, 600-2000 A.D., Jaca Book (Milan 1999), with discussion on pages 85-86. The island of Buton, unbeknownst to Butön, is located here.

And here is a marvelous passage from the  Śrīmad Bhāgavatam 2.4.18.  In the commentary, notice how the Khasa (the Kha-sha of  our list?) are explained as people with stunted hair on their upper lips, and therefore Mongolians and other equally moustache-challenged peoples. Really, I somehow doubt this. The general point of this country listing within its context is to assert that even the most dreadfully sinful foreign cultures can find their full salvation in Vishnu. Are you listening to this out there?

For a discussion on Shri Parvata and surrounding area, since around 1960 submerged beneath a huge reservoir created by the Nagarjunasagar Dam, see this blog at hridayartha.blogspot.com.

§ § §   § § §   § § §

The frontispiece is a popular print of the famous Ayyappa deity image, object of the largest pilgrimages in Kerala. Some believe this deity and its shrine were originally Buddhist. I have nothing illuminating to say on the subject, just to say that it is a fairly commonly expressed idea, that may or may not have a strong basis. You may be the judge. I might be the only one to notice it, but the deity's (iconographically speaking) rather unusual sitting position resembles the Zhijé (not the Chö) form of Padampa. I guess it is only a coincidence. But then their right hands are also displaying the same mudrā...  Hmmm...  I would also invite you to pay attention to the small detail of the coconuts in the offering tray. A gentle reminder that you are in coconut country.

§ § §   § § §   § § §

The hardcore Tibetanists among you are welcome to go and download “50 Geo Texts” HERE at New Tibetological website. I uploaded it a long time ago, but then I forgot to tell anyone it was there. It is supposed to be useful as a search file for answering some kinds of questions about Tibetan geography and geographical knowledge. Its only proper use is as a reference tool along the lines of an index. That means the user should feel obliged to look up the original publications in order to verify that the readings are actually there.

.

I imagine this to be a mother-of-pearl representation of Lao-tzu’s journey to the west on a water buffalo’s back.  I can’t explain the ears, but I think that's probably a flute he’s holding. It was found in a hidden drawer of a credenza brought home from a furniture auction in the U.S. in the 1980’s.

Success is as dangerous as failure,

hope is as hollow as fear.

Whether you go up the ladder or down it,

your position is shaky.

When you stand with two feet on the ground,

you will always keep your balance.

Hope and fear are both phantoms

that arise from thinking of the self.

When we don’t see the self as self,

what do we have to fear?

Tao Teh Ching

I also put up an informal listing of titles of Tibetan texts of the typical geographical genres. It includes guidebooks to holy places (gnas), hidden countries (sbas yul), itineraries (lam yig) and the like. It's findable by clicking here.