Hidden Realms and Pure Abodes: Central Asian Buddhism as Frontier Religion in the Literature of India, Nepal, and

Tibet
Ronald M. Davidson
Fairfield University

The notable romantic interest in Silk Route studies in the last hundred years has spread far beyond the walls of academe, and is especially observed in the excessive world of journalism. In Japan, NHK (Japan Broadcasting Corporation) has produced a series of films whose images are extraordinary while their content remains superficial. The American National Geographical Society has followed suit in their own way, with some curious articles written by journalists and photographers. With the 2001 conflict in Afghanistan, American undergraduates have also begun to perceive Central Asia as a place of interest and excitement, an assessment that will not necessarily pay dividends in the support of serious scholarship. While Indian and Arab academic commentators on popular Western cultural movements want to read the lurid hand of Orientalism into such responses, I believe something more interesting is actually happening. Over the course of the past decade, I have often been struck by statements in medieval Buddhist literature from India, Nepal, and Tibet, statements that depict areas of Central Asia and the Silk Route in similarly exotic tones. Whether it is a land of secret knowledge or mystery, of danger and romance, or a land of opportunity and spirituality, the willingness of Indians, Nepalese, and Tibetans to entertain and accept fabulous descriptions of the domains wherein silk commerce and Buddhism existed for approximately a millennium is an interesting fact. More to the point, for the Buddhist traditions found in classical and medieval India and Tibet, there has been no area comparable to Central Asia for its combination of intellectual, ritual, mythic, and social impact. Perhaps most remarkably, references to many areas of Central Asia have often been taken by scholars as signa of Indian Buddhism, based on the presumption that the use of Indic languages (Sanskrit, Buddhist Hybrid Sanskrit, or Gndhr) is indicative of Indian presence, even though we see undeniably local Sanskrit traditions that emerge.1 This Indian presumption is done with a concomitant disregard of the clear cultural disparity between Central Asia (including the Peshawar / Swat Valley /
153

154

Pacific World

Kabul zone) and that of India proper, a difference that has only seldom been noted. For the most part, scholars have been seduced into accepting Central Asias promotion of itself as an indelible part of Indian cultural life, even though we can see that it may not be true. Partly the Indophilic acceptance is a result of the geography of Aokas India, and partly it is a consequence of colonial hegemonic maps or post-colonial nationalist sentiment. In the former case, the presence of Aokan epigraphs as far afield as Afghanistan has provided both nineteenth century imperialists and South Asian nationalists with an artificial horizon of state identity. This was reinforced by the British rhetoric of colonialism, that somehow the zone from Burma to Iran was but one common area for the British Indian Empire. However, as John Brough had already noted in 1965, we would be remiss if we found that our nomenclature of Indias subcontinental geography has lead us to underestimate the distinctive nature of the NorthWest, and to think of it merely as an extension in space, stretching away to the upper Indus and beyond, but otherwise a more or less homogenous continuation of the country of the Gangetic plain and Madhyadea. The North-West is different, in terrain and climate, and in numerous other ways, of which two are directly relevant to the present discussion: geographically, the trite fact of its location and relative accessibility from both west and east; and culturally, its development of a characteristic language.2 This paper will seek to explore aspects of the mythology of Gandhra, Swat, Kashmir, Khotan, and even more ethereal lands in literature that has been accepted within South Asia and Tibet. In this, the paper will be informed by the critical movement in anthropology and literature known as reception theory.3 In our own pursuit, we will not be concerned with the reception of Buddhism in Central Asia, but rather the reverse influence on the distinctive narratives and ideas that have been developed and propagated in these areas and reimported back into the sphere of South Asian and Tibetan societies. Because Buddhist missionaries into these regions were the transporters both of Indic systems and of the Central Asian responses, it is within Buddhist literature that we find the primary South Asian response to the Central Asian cultural horizon. In distinction, medieval non-Buddhist literature tends to be concerned with the ferocious Ha peoples that threatened the classical and medieval cultures of the plains. Thus, throughout the history of Sanskrit literature, Central Asians (akas, Kunas, Yavanas, Bhoas, Cas, and others) tended to be classed as Mleccha barbarians, although there are noted exceptions to this estimation and some Mlecchas have been influential in the minor sciences.

Davidson: Hidden Realms and Pure Abodes

155

Perhaps stimulated by the simultaneous specter of barbarian attacks and the wealth of the trade routes, Buddhist descriptions are noted for their emphasis either on the fertile origin or on the immanent demise of Buddhist cultural and religious systems, and the authority eventually granted Central Asia spilled over into the remarkable acceptance of Silk Route doctrinal or ritual innovations in an unprecedented manner. We will find that these Silk Route areas will be depicted quite differently from other important Buddhist locales, such as r Lka, which has been so often represented as the abode of demons. The precise reasons for this disparity are as yet obscure, but it is my thesis that the acceptance of Indic languages as the media of official discourse, combined with the cultural geography of isolated cultureswhether through the isolation of desert travel (Khotan) or the seclusion enforced by mountain passes (Kashmir, Swat)were strong contributing factors in their perceptions. We may also wish to reflect on Central Asia as a frontier zone and consider if Central Asian cultures represented such a romantic reality to Indian Buddhists for some of the same reasons as the American West was considered romantic by those in Europe and in the Eurocentric culture of the Atlantic seaboard. In all of these estimations, we may begin to take into account that no other area has had so strong an impact on South Asian and Tibetan Buddhist perceptions of Buddhism in the way that Silk Route sites have had. While art historians and archaeologists have dominated Silk Route studies, the evidence from the literary archive is only now beginning to come into its own, and it is to that archive that we may turn.

GREEK, AKA, AND KUNA INTRUSIONS In the larger cultural sphere, fully engaged by the Indian Buddhist subculture, the Indo-Greeks and their successors, especially the akas and Kunas, have been influential beyond that of any other foreign cultures until the time of the Muslim intrusion. In some ways, we can certainly understand an Indian engagement with the Eastern Greeks, which is paralleled in our own use of the classical world as our primary reference point for art, philosophy, and politics. Yet the Indo-Greeks did not have influence in the same way in India that they have had on Europe and America. Indians, for example, remained virtually immune to the canons and functions of Hellenistic art per se, and there was certainly nothing like the Renaissance or the Neoclassical periods in the Northwest as revival periods for their artistic heritage. While the origin of the Buddha figure continues to be disputed, and is likely to remain so for some time, there can be no doubt that the simple fact of its dispute indicates the ambivalence observable in Indian art about the Hellenistic forms they encountered and ultimately rebuffed. One could even argue that, because of the Silk Route,

156

Pacific World

Gandhran art has had a greater influence in China than within the borders of India proper. Indeed, outside of explicitly Buddhist ideology, we find the primary Indo-Greek and aka influences in the areas of cultural products and, to a lesser degree, political representations. The contributions of Northwestern cultures are most explicitly the pervasive influence seen in the borrowing of Greek-style coinage (not the punch-mark coins of the Mauryans) and the appropriation of astrology. Coins, though, are the subject of archaeology and other disciplines studying material culture, and here we will restrict our emphasis to literary traces. In this respect, it is curious, certainly, that neither of the two writing systems widely employed for virtually all early inscriptions in the NorthwestGreek and Kharohshould have found much application in India per se, even though minor inscriptions in Kharoh have been found as far afield as West Bengal and China.4 Unlike the now ubiquitous use of Indian numerals (incorrectly called Arabic numbers, as the Arabs obtained them from the Indians) in the West, neither Greek nor Kharoh scripts took India by storm. For whatever reason, indigenous Indian developments from Mauryan Brhm script entirely eclipsed the employment of Mediterranean-based scripts on the subcontinent and wherever Indian influence became dominant. Astrology, though, is a discipline apart and Indian astrology did embrace Greek-based systems. Two seminal documents point to the importance that this science had for Indians over time: the Grgya-jyotia and the Yavana-jtaka. The former is in some ways the more interesting, for it not only treats the overall topic of astrology and astrological calculation according to the mythic Gargaa treatment that spawned a whole host of imitative Grgya-jyotia-sbut also incorporated into its sixty-four chapters one that remains the only Sanskrit mythological statement devoted to groups from the Northwest: the Yuga-pura. A mere 115 verses in the received text, the Yuga-pura discusses the religious circumstances and social structure at the end of the cosmic cycle (yuga), thus integrating the appearance of Indo-Greeks (Yavana) and akas into a mythology of cosmic decline. Because of the thoroughly Indian nature of this appropriation of Northwest astrology, the text became very influential outside of astrological calculations. The Yuga-pura contributed verses and other influences to such standard Sanskrit works as the Bhatsamhit, the Matsyapura, the Mahbhrata, and so forth.5 Varhamihira, the putative author of the Bhatsamhit, knew well that the source was foreign, but it mattered little in the question of astrology. Yes, these Greeks are barbarians, but this correct science [of astrology] has somehow been found among them.6 Just as Indianized, in its own way, was the Yavana-jtaka, a treatise on Greek horoscopy. This is not, as Buddhist readers might initially presume, some story of the Buddhas previous births among the Greeks. Rather, it is a 269/270 CE versified edition by Sphujidhvaja of Yavaneas 149/50 CE

Davidson: Hidden Realms and Pure Abodes

157

prose translation from the lost Greek text, which Pingree maintains was originally a composition from the illustrious city of Alexandria, Egypt. That does not mean, however, that it is a simple translation in the manner we know today, for the text discourses on topics from reincarnation to Brahmans, from Buddhists to is, and so forth. Thus the text is in reality an extremely interesting exercise in domestication of foreign materials, with subtle shifts in reference throughout the surviving Sanskrit so that an Indian reader (for whom else would such a text be written?) would gain immediate access to the Greek science, even while the transposition to the Indian cultural landscape was in process. As a measure of its success, the Yavana-jtaka, like the Grgya-jyotia, became a seminal text that engendered a whole other series of Yavana works, some of them by actual Indo-Greeks or akas in India.7 We cannot leave the issue of the Yavana influence without mentioning that famous literary figurethe Indo-Greek king Menander. His ethnic affiliation is unquestionably declared; even in the Pli translation of what must have been originally a Gndhr text, he is addressed as a Yonakarja, that is, an Ionian king. Yet the text makes this to be an Indian curiosity, for he is granted the pedigree that all Indian kings must enjoy: he is learned in all the Vedas and Vedgas, all the philosophical positions and secular sciences. The text even provides Menander with the most Buddhist of backgrounds, for his previous birth story wherein he acquired merit (prvayoga) is related as part of his domestication into the Buddhist literary horizon. The redactors strategy is clearly to refurbish Menander as an exemplary minority figure within the Indian cultural zone, and this he does by summing up Menanders attributes that there was no other king in Jambudvpa so strong, quick, courageous, and wise.8 As Fussman has recently pointed out, it would be difficult for an Indian reading the current Pli text of the Milindapaha even to identify Menander as not Indian in either name or attributes.9 Nonetheless, the Ionian kings interest in questioning ascetics, coupled with the involvement of Indo-Greeks with Buddhism in Gandhra and in the Western Satrapies, sets the background somewhat apart, and appears to be an Indianized version of the Greek cultural icon of the philosopher-prince. Like the appropriation of Greek astrology, though, this is an increasingly occluded part of the representation, so that the Greekness of such an icon has been lost by the time the Pli text was composed, even though the Northwest setting remains firm in the texts eye. Other texts take the Kuna developments into account, and we may observe that the physical/cultural landscape of Gandhra played an important part in its acceptance as authoritatively Buddhist. One of the earliest descriptions is found in the narrative of the Buddhas travels to Kashmir and Gandhra, a story that survives in places in the Mlasarvstivda-Vinaya, such as the Bhaiajyavastu, and in other analo-

158

Pacific World

gous Mlasarvstivda collections.10 Just before the Tathgata is to pass into his mahparinirva, a Yaka named Kutika brings fruits, especially grapes, from Kashmir and Central Asia. As a result, the Buddha decides to make a visit to Kashmir, leaving nanda behind and taking the Yaka general Vajrapi as his attendant instead. In the course of this trip, he converts the nga Apalla, along with other Yakas and nga-s, and brings a master potter into the Buddhist fold. In the course of his sojourn, he makes some prophecies, including that of the arhat Madhyntikas conversion of Kashmir. The other prophecy he makes, though, is that of Kanika. Seeing a young boy playing in the dust, making stpa-like figures, he prophesizes that, four hundred years after his parinirva, this boy will be the king Kanika, and will establish a stpa in the area that will do the duty of the Buddha after his parinirva.11 This is an extraordinary statement about the great Kanika stpa, on which was painted the double-bodied Buddha that is well recognized in Central Asian artistic renditions, and which was in close proximity to the shrine containing the ostensible begging bowl relic of the Buddha.12 A similar statement about doing the duty of the Buddha is applied to Upagupta, who is said to become the arhat that overcame Mra and did the work of the Buddha following his passing.13 We may not need to follow Schopens suggestion that the relics in such stpa-s somehow have been regarded the living presences of the Buddha to understand that the Kanika stpa was certainly qualitatively different from other stpa-s in the area, many of which were ascribed to Aoka.14 Indeed, the presumed relic placed in this stpa is not mentioned at all in the Vinaya episode, and the stpas celebrated artistic program is nowhere recognized in the surviving Sanskrit text. Instead, the literary statement is an acknowledgment that the Kanika stpa in Peshawar served as a focal point for Buddhists and merchants in the Gandhra/Karakorum/Indus river corridor, and was probably the best known (and certainly largest) example of the type that informs the stpa plaques, casts, and petroglyphs from Harwan and Chilas to Hunza, the Tarim Basin, and beyond.15 In fact, the Mlasarvstivda citation comes just before one important pronouncement of the Mahsammata story, in which the first kingship is located in Mathur, the southern extent of the Kuna realm and one of the two places where images of Kanika were set up in Kuna ancestral temples. Mathur, we are assured, is the place where the law was founded and where katriya-s began their dominion over the Indian political world. All told, this story from such sources as the medical division of the Mlasarvstivda-Vinaya constitutes a statement by Gangetic Valley Indians that Central Asian Buddhism had come of age, that its overall importance for and benefit to the dispensation of the Buddha had more than exceeded the threshold required to overwhelm the indigenous xenophobia of Indians and had earned it a position in the literary archive. Yet we need

Davidson: Hidden Realms and Pure Abodes

159

to recognize that there was exhibited a process of selectivity in this report, for not all such stories or sites were accepted to such a degree. We seem, for example, to have no surviving South Asian text that recognizes the importance of the Cave of the Buddhas Shadow, a site that so impressed the Chinese travelers like Hsan Tsang.16 Thus, the aura of Kanikas stpa is presented in a specifically Indian manner, with a previous-births (prvayoga) narrative sanctioning its sanctity, analogous to the prvayoga material sacralizing Menander in the Milindapaha.

YOGCRA IDEOLOGY AND RITUAL Actually, the Gandhran self-presentation effort was sufficiently successful that one of its more important traditions was fully accepted in the Indian intellectual agenda. I am speaking of the Yogcras, those monkish meditators whose philosophical, meditative, and liturgical developments became part of the received canon. As the human founder of the Yogcra tradition and the well known author of the voluminous Yogcrabhmi, Asaga became so influential that his origin is regarded as a quaint curiosity today, but virtually all of our data indicate that he and his brother Vasubandhu were the only Gandhran authors to become so esteemed. Asagas residence in Gandhra (Puruapura) is attested in the earliest hagiography of him, that which is appended to Paramrthas story of his illustrious brother, Vasubandhu.17 The placement of the brothers is independently verified by data found in Yaomitras commentary on the Abhidharmakoa in conjunction with the Abhidharma-Mahvibha-stra. In these works, an obscure point of doctrine is ascribed to both the Gandhran Vaibhikas and the Yogcras, these latter described as consisting of Asaga, etc.18 Some of the later Chinese and Korean pilgrims tended to make all famous authors associated with Gandhra into residents of the Kanika Vihra, the most famous monastery in the environs of Peshawar, but we have no confirmation of this rather late identification, and Dobbins has culled such travelers tales from the translated literature.19 Nonetheless, Asagas residence in Gandhra seems secure, with only the single instance of Hsan Tsangwho also provides a Gandhran origin for the Yogcra masterrelating a story about Asaga living at some point outside of Gandhra, in Ayodhy, where he was said to have had a teaching hall.20 We may wonder, though, if this rather late attribution of an alternative residence is simply the result of an early medieval Indian appropriation of Asagas personality, predicated on his fame. If we recall that Asagas contribution extends into the Mahynist liturgical venue, we can better estimate the importance of this area. One of the two standard lineages of the bodhisattva vow not only traces itself to

160

Pacific World

Asaga, but the form employed is found in the section on conduct found in the Bodhisattvabhmi.21 There, the aspiring bodhisattva is to find a monk or lay mentor, to set up a statue of the Buddha, and to take on the vows of the bodhisattvas discipline while visualizing all the Buddhas and bodhisattvas of the various world systems in the ten directions. His Mahynist preceptor provides him the vows, and the Buddhas and bodhisattvas themselves perceive a simulacrum of the aspirant bodhisattva receiving the vows and accepting the discipline from the preceptor. They accordingly assist the newly minted bodhisattva by helping him increase his roots of virtue and, we may suspect, provide him a degree of protection in the process. The distribution and influence of this ritual cannot be overestimated. We find that the Bodhisattvabhmi itself enjoyed a distinctive circulation, being translated three times into Chinese, twice as an independent work aside from the Yogcrabhmi, by Dharmakema in 414421 CE and Guavarman in 431 CE. We may presume, therefore, that the text was composed in the final quarter of the fourth centuryprobably not much before and doubtfully any later. The chapter containing the ritual became separately influential as well, with two Indic commentaries dedicated to its exegesis, those by Guaprabha and Jinaputra, surviving in Tibetan translation (To. 4045, 4046). The material included in the liturgical section quoted earlier was ritually so important that the virtue chapter of the Bodhisattvabhmi shares textual material with the Bodhisattvaprtimoka, also translated by Dharmakema in 414421 CE.22 In both Central Asia and China, this liturgical text and its ritual format were emulated, spawning such apocryphal scriptures as the Fan-wang Ching, widely used in Chinese monasteries and employed in Japan.23 The ritual was popular in South Asia as well, for a text calling itself the Bodhisattvaprtimok-astra, apparently a later Indian work, appropriates the above ritual almost exactly verbatim.24 We may suspect that, because of the way this latter work is organized, the Bodhisattvabhmi circulated in conjunction with another Mahynist ritual work, the Upliparipcch.25 Sections from both of them were further included in the Bodhisattvaprtimoka-stra.

THE ESOTERIC CONNECTION The other discussion widely recognized in India and elsewhere featuring a Silk Route site as the source of the Dharma concerns the identity of Oiyna as the progenitor for much of the esoteric canon. Kuwayamas 1991 rereading of previously incorrectly deciphered epigraphs has finally secured the place of the Swat Valley as Oiyna, after many claims by Indian nationalists that it was to be located in Orissa, Bengal, or South India.26 The aura Oiyna obtained, as the esoteric canon itself, really

Davidson: Hidden Realms and Pure Abodes

161

passed through three stages: the early collection of spells evident from the sixth century forward, the development of the Indrabhti myth in the eighth century, and the extensive mythologization of Oiyna in the yogin tantras beginning in the ninth century. The first esoteric canons were apparently little more than collections of spells (mantra- or dhra-piaka), and both the Mahsghikas and the Dharmaguptakas among the early schools have had such collections attributed to them.27 It is further clear that an early canon of spells was strongly associated with the areas between Kashmir and the land of Uiyna or Oiyna. Already in the early sixth century, Sung Yn heard a story in Tashkurghan that the kings ancestor had studied spells in Oiyna, using the knowledge to save his kingdom from a dragon who was drowning merchants and interrupting trade.28 Hsan Tsang also, in his 646 CE recounting of his travels, dismissed the ignorant users of spells that he encountered in Oiyna.29 Even the language of the surviving early literature, such as the early seventh century manuscript of the Kraavyha, affirms its association with places employing birch bark for manuscripts, notably employed from Kashmir to Bamiyan.30 The Abhidharmakoabhya further contributes to this sense of localization in the environs, mentioning two kinds of spells, one from the area of Gandhra (gndhr vidy) and one bringing visions of the future (kaik vidy).31 We have no indication that any other place in the subcontinent managed to represent their ritual phrases so effectively at such an early period. Similarly, the Indrabhti myth of preaching the tantras is found in an extensive form, as early as Jnamitras late eighth or early ninth century commentary on the 700-Verse Perfection of Insight scripture. There, we are assured, eighteen classes of esoteric tantras of the eighth centurythe Sarvabuddhasamyoga, the Guhyasamja, etc.have miraculously appeared in Zahor, to its king, Indrabhti. The good king, though, is befuddled: he could not penetrate the understanding of the new scriptures. However, because of his supernormal insight obtained through countless lives of virtuous activity, he understood that an outcaste personality held the key. This individual was Kukurja, who lived with a thousand dogs in Mlava, probably to be located in one of its great cities, such as Ujjain or Mahimati. Indrabhti sent a representative to invite this dog-guru to Zahor, but Kukurja had not seen the texts, which were then dispatched to him to peruse in advance. Kukurja, though, was equally clueless and eventually obtained the inspiration of Vajrasattva to secure their comprehension. While we may not know the precise location of Zahor, Indrabhtis association with Oiyna is affirmed in virtually all other forms of the myth.32 We even see a variant in the narrative that Indrabhti asked the Buddha to preach to him a doctrine that would allow those addicted to the senses a vehicle for liberation, and in response the Buddha preached the tantras.

162

Pacific World

Moreover, Oiyna is twice mentioned in the late eighth or early ninth century autobiographical narratives on seeking the esoteric scriptures whether the Sarvatathgatatattvasamgraha or the Guhyasamjaby kyamitra and Buddhajnapda. The retelling of the latters trip to Oiyna actually indicates that the land is called guodaya, the rising of good qualities. Vitapda indicates that Oiyna is granted this designation because it is the source of so many benefits. Buddhajnapda reports that his early studies were with Haribhadra in Magadha and with Vilsavajra in Oiyna. Indeed, the importance of Vilsavajra for the hermeneutics of the early esoteric system cannot be doubted. It is probably Vilsavajra, residing in his Swat Valley monastery of Ratnadvpa-vihra, who provided the fundamental interpretation to the forty opening syllables of the Guhyasamjatantra, an interpretation that became embedded in virtually all commentaries following him. Thus, according to the surviving archive, each of the syllables of the opening of the texte-va-ma-y-ru-ta, etc.is given a specific hermeneutic and represents an experience in the esoteric system. This interpretation has been widely accepted in Indian esoteric hermeneutics and is represented in such definitive texts as the Pradpoddyotana of the Tantric Candrakrti.33 Vilsavajras surviving works also include the earliest citations of such seminal texts as the Laghusavara-tantra, providing an effective chronology to the early yogin-tantra-s. Finally, a later Indrabhti reports in the opening section of his Sahajasiddhi that this yogin-tantra inspired system derived from the area of Oiyna. According to the short lineage list and the lengthy commentarial hagiography, this Indrabhti was the receptor of a teaching on sahaja, or natural reality, that began in Oiyna with a princess r-Mah-Lldev, who based the system on her experience precipitated by an encounter with an unnamed black-headed i at the forest monastery of Ratnlakra.34 Upon being blessed by him, she realized that she was an emanation of the bodhisattva Vajrapiwho is identified here as the patron divinity of Oiynaand she and her five hundred ladies in waiting all received awakening into the nature of sahaja. The teaching on sahaja was then passed down in a lineage until the time of Indrabhti, who wrote it down. Indeed, Oiynas gravity precipitated its almost universal inclusion in lists of the four great sites (caturmahpha) found in such diverse tantras as the Hevajra and the Abhidhnottara. One of the lists of these four sites eventually served to associate the great sites with the internal maala, and Oiyna-pha was given the preeminent position of becoming the seat or cakra found in the head of the esoteric yogin. Pilgrimage to the physical sites, in this reading, was replaced by pilgrimage to the internal maala, and we find it being identified as the wheel of great bliss and an internal site by Guipda in his verses in the Carygtikoa.35 The importance of Oiyna is certainly not lost on the only Indian master to

Davidson: Hidden Realms and Pure Abodes

163

write a comprehensive treatment of the various pilgrimage sites. kyarakita, in his Phdiniraya, divides them up into nine grades from lesser-lesser to greater-greater, and takes as a given the importance of Oiyna.36 While he also lists Suvaradvpa and Singhaladvpa, and though he acknowledges that there are important sites in Tibet and China as well, pride of place is provided the Swat Valley, the only site outside the normal boundaries of Indian culture to be so recognized. We need only glance at Tibet to see the great importance that Oiyna is accorded, primarily through its mythic association with Padmasambhava and his putative father, Indrabhti, especially among the rNying-ma sect of Tibetan Buddhism. As a focus of Tibetan interest, Oiyna spawned among the rNying-ma an entire mythic literature around Padmasambhava, his relationship with Indian kings and ministers, and Padmasambhavas various wives. Certainly this material is well known and its content is increasingly emphasized, but it might be just as important to note that the bKa-brgyud-pa were sufficiently stimulated that two of their more illustrious members felt themselves motivated to make pilgrimages to Swat, and their thirteenth and early seventeenth century itineraries were translated by Tucci some time ago.37 Both Grub-thob O-rgyan-pa (12301293) and sTag-tsang ras-pa (15741651) visited the Swat Valley at quite different stages in the history of the area, yet their respective observations continue to emphasize the miraculous and the importance of its spirituality to their respective communities. In fact, the legend of Padmasambhava continues to motivate Tibetans in their search for sacred places, and we see the development of a previously obscure lake in Himachal Pradesh as a current focus of interest. Rewalsar, as it is known in Hindi, is about twenty-four kilometers southwest of the regional center of Mai, and has been esteemed by Tibetans as mTsho Pad-ma, the lake of the lotus-born teacher Padmasambhava. The lakes history as a Tibetan pilgrimage site is obscure to me, but until the nineteenth century it was primarily known as a lake dedicated to the Sikh guru, Gorbind Singh, and is well over five hundred kilometers by air from Saidu Sharif in the Swat Valley. Emersons 1920 description of Mandi State indicates that the sole Tibetan monastery that had been built to that time was recent and that no ancient Buddhist remains could be found.38 Nonetheless, the Tibetan hermits of the rNying-ma order I interviewed in the hills around Rewalsar in 1996one of whom had been there for twentyfour yearsall attested that it was the real Oiyna. This is yet another example of the appropriation of sacred site designations by local groups when the original site is either lost to memory or dangerous to secure. We frequently see this movement of site names out of India to Nepal, Tibet, China, and Southeast Asia, but the reversea non-Indian site becoming located within Indiais seldom encountered, and this is the only one to my knowledge.

164

Pacific World

END OF THE DHARMA At the other end of the theoretical spectrum, it is interesting to observe that sites essential to Silk Route merchants should play such central positions in the elaboration of the mythology of the Buddhist Dharmas extinction in Jambudvpa. Given the importance of the doctrine on the Dharmas degradation and demise, we might expect that the scenario of its destruction would occur in the geographical areas of India, and in fact the majority of Buddhist narratives include exactly this proposition. However, important variants emphasize the position of Silk Route areas, especially Kashmir and Khotan. The Kashmir connection is found in the Mahynist Mahparinirvastra, a text that was to receive currency throughout East Asia and Tibet for its support of the doctrines of essential Buddha nature, the tathgatagarbha. Less well understood is its position in specifying the development and demise of the Saddharma. According to the received text in Chinese and Tibetan, verified by its surviving Sanskrit fragments, there will be seven signs of the final end of the Tathgatas truth. At the conclusion of these signs, the Saddharma will disappear into Kashmir, like mosquito urine disappears into the earth, so that Kashmir will be the final resting place of all the Mahyna scriptures.39 There are formal similarities with this narrative and the story related by Hsan Tsang that the life of the Saddharma is tied to the great stpa of Kanika. This stpa, we are told, will be destroyed by fire and rebuilt seven times after the final conflagration, and the true Dharma will end on earth.40 Beyond the mention in the Mahyna Mahparinirvastra, a site other than Kashmir was to provide an overwhelming level of importance to Buddhists in general and Tibetans in particular. This is the state of Khotan, well known for its archaeological remains but significant also for its literature, especially evident in the construction of an alternative narrative. Among the several important scriptures composed there, the Saghavardhana-vykaraa narrative is a variant on the normative Kaumb prophecy, a prophetic story that forms the core of so many of the end of the true Dharma scenarios. Nattier, who has studied these collectively in extenso, has pointed out that the normative story proposes that the demise of the true Dharma occurs when a Buddhist king becomes victorious over two royal enemies of the Dharma. He holds a convocation, and the two foremost monksiyaka and Sratadispute the nature of the Prtimoka, so that eventually each is killed in the melee that follows. The gods declare the end of the true Dharma and the earth quakes, similar to the announcements and quaking of earth just before the moment of kyamunis awakening. Even the most important iteration of this core story, the Candragarbha-stra, shows signs of Central Asian influence,

Davidson: Hidden Realms and Pure Abodes

165

and Nattier has proposed that it received its final editing in Gandhra or a proximate area.41 In the strongly Khotanese Saghavardhana-vykaraa version, those without faith (in the doctrine) will be born in every country. Kings, as well, will compete in stealing each others territory.42 Bhiku-s and bhiku-s will accumulate karma of the ten unvirtuous paths. They will habitually tell false stories to each other, and accumulate riches by engaging in commerce. They will raise cattle and keep male and female servants. Because of these behaviors, people will lose their faith. Consequently, they will steal the wealth donated to the Buddha and the Sagha. Then the few upright bhiku-s will become depressed. The monks of Khotan (who are among those few) will speak among themselves and eventually decide to leave Khotan, traveling to Bru-sha, that is, to the Burushaski speaking areas (Hunza, Yasin valley) in what is now north-eastern Pakistan. There the monks will elect to move on to Tibet, which will initially welcome them, but also eventually cast them out. The dispossessed clerics will then go to the lake of the nga king Elpatra, who served as the guardian of Gandhra, and he will usher the monks into the court of the Gandhran king. Unfortunately, the Buddhist king will pass away in two years, after which an unbelieving king will arise, and the monks will go to Kaumb, where they will pick up the normative narrative of the conflict, as related above. Nattier has observed that the parent narrative in the Candragarbhastra had some influence in Tibet, and about eighty per cent of the text is cited in the History of Buddhism by Bu-ston.43 However, the Candragarbhastra is probably Indian in origin, even if it was edited in Gandhra and was received by Tibetans from the Tarim Basin. More interesting, for our purposes, is that the Khotanese Saghavardhana-vykaraa was clearly the basis for an end of the Dharma scenario in the thirteenth century by the rNying-ma author, mKhas-pa lDeu, in his recently published religious history.44 While his text shows some similarity with the other two Khotanese works on religious history first examined by Thomas and reread by Emmerick, there can be little doubt that the Saghavardhana-vykaraa provided the basis for his summary. Yet the scenario articulated by mKhaspa lDeu is not a simple reiteration of the Khotanese work, for much of the specifically Khotanese material has simply been left out of the text, so that we are left with a bowdlerized version that elides many Khotanese place and personal names in favor of a more generalized account of the process. It is unclear whether mKhas-pa lDeu himself, or some other editor before him, contracted the material. It is obvious, though, that the Khotanese narrative was received in Tibet with sufficient authority that an important thirteenth century author blithely included this version as the normative one in his work. Lest we conceive of this estimation as an aberration, we might further reflect on another thirteenth century chronology of the demise of the

166

Pacific World

Dharma, that proposed sometime after 1216 CE by Sa-skya Paita, who was surely no champion of Tibetan innovations or of non-Indic materials. Yet Sa-skya Paita, who is consistently critical of others about their lack of Indic sources, neglects to mention that there was at least one time when he argued against Indic doctrine. In his chronological appendix to his hagiography of his uncle, Grags-pa rgyal-mtshan, Sa-skya Paita argues that the Vinaya record as calculated by kyar in 1210 CE represented a faulty chronology, since it placed the Buddhas parinirva in 543 BCE.45 For Sa-Pa, the Khotanese chronology accepted by his uncle was correct, even though this calculation placed the Buddhas parinirva in 2133 BCE.46 The irony is that, while discounting the Indian oral tradition on this matter in favor of received scriptures, the Sa-skya sources for their chronology privileged pieces of Central Asian apocryphal literature.47 From these, and other examples, we can see that Tarim literature had clearly become influential in twelfth to thirteenth century Tibet. This is true whether we are speaking of the rNying-ma, the least orthodox group and the one most given to revealed literature, or the most orthodox group, the Sa-skya, the most avid defenders of Indian precedents and Indian models. Perhaps this privileging of Tarim Basin lineages is a reflection of the authority they enjoyed during the Royal Dynastic period, where meditative traditions either from or passing through Central Asia were accorded authenticity and patronage by the Tibetan emperors.48 Perhaps it was a result of the heightened influence of the Tangut monks studying in Central Tibet, at a time when the Tangut Empire was at its height and prestige. In Central Tibet, both Tangut monks and Tibetan teachers were encountering Indian monks fleeing from the ravages of Islamic incursions, and the growing warfare between Hindu kings and their Turkic antagonists. From this vantage point, Tangut emperors like Jen-tsung (r. 11391199), whose strong Buddhist support is well authenticated, appeared the only Buddhist political hope in the twelfth century.49 Between Tibetans becoming National Preceptors of Tangut princes and the imperial largess towards Buddhist teachers, Jen-tsungs charisma may have blended in Tibetans minds into a general esteem of Central Asia.50 For whatever reasons, though, the situation is clear: while orthodox exegetes like Sa-Pa are systematically censorious of Chinese influence in Tibetan meditative systems, Tarim mythologies and chronologies informed some of their own efforts.

THE CULMINATION: AMBHALA Curiously, almost all of these trends come together in the story of ambhala, a mythic land that was to seize the imagination of virtually all who considered this sacred realm, whether we are speaking of its composer in the late tenth or early eleventh century, when the myth was formulated,

Davidson: Hidden Realms and Pure Abodes

167

or the consumers of the Shangri-la of American fiction. The texts representing ambhala synthesize astrology, geography, esoteric meditation, gnosticism, righteous kingship, and a host of other ideas that we have already seen became important to Indian consumers of Central Asian lore. Yet the paradox of the ambhala myth is that the fictive geography of this hidden realm and pure abode occurs exactly when the real Central Asia becomes lost to Islam. This dynamic is somewhat similar to the phenomenon observable with Oiyna: just when the great pilgrimage site receives its most visible affirmation in the ninth to eleventh centuries in India, it was being submerged in Islamic aggression. Consequently, we might wonder if the aura of such sites is to a degree dependent on their being actually unavailable, and thus all the more desirable and mystical. Another curiosity that has not received much attention about ambhala is that its earliest textual sources do not display unanimity on its location. In the form initially encountered, that found in the first chapter of the Laghu Klacakra-tantra, ambhala is placed to the south of Mt. Kailsa, perhaps in the upper Brahmaputra reaches or in the Purhang Valley, where one of the two centers of the Western Tibetan Empire were developed in the tenth century.51 If this latter is the case, the author of the Klacakra-tantra would have had the new principalities founded by the scions of the Tibetan royal clan in mind when he wrote of the city Kalpa, the capital of a hidden kingdom where the Kalki kings kept the Klacakra until the time for its release was correct.52 Be that as it may, by the time Puarka wrote his Vimalaprabh commentary on the Klacakra, proposing that he himself was the second of the Kalki kings of ambhala, he placed it to the north of the t River, which is one of the mythic rivers reported in both Buddhist cosmologies and puric literature.53 Authors like Sylvain Lvi have maintained that the t is to be identified with the Tarim River, but I believe we have lost the point here. The point is that our earliest Klacakra authorities are unable to place its location accurately. Yet it appears to me probable that Puarka was himself from the northwest, perhaps from Kashmir, which maintained its traditional culture until the Muslim domination following the suicide of its queen, Dev Koa, in 1339 CE. Puarka reports languages of the far north in the text of the Vimalaprabh and was certainly in a position to understand that ambhala could not have been located to the south of Kailsa, which was accessible from Himalayan kingdoms like Kashmir.54 The problem is made more complex by disagreements on the location of the first preaching of the received Klacakra. Ravirjna, the author of the Klacakra-based Amtakaik commentary to the Majurnmasagti, maintained that the great Dhyakaaka stpa was the site of the original preaching of the tantras.55 Yet Puarka maintained that the tantra was preached by Majur in Kalpa, the capital city at the center of ambhala.56

168

Pacific World

Curiously, Tibetans like Bu-ston or Gos-lo gZhon-nu-dpal have largely rejected Puarkas position in favor of that stated by Ravirjna.57 Be that as it may, there can be little question that the ambhala myth is closely tied to an ideology of a Buddhist kingdom surviving the onslaughts of the Muslim armies, depicted both in the text of the received tantra and in the commentarial literature. Thus, the interest in the mythic kingdom of ambhala combines themes already seen about the origin of the Dharma, its demise, and (in this instance) its resurrection at the hands of the victorious armies of the final future king of ambhala, Raudrakalkin.58 Orfino has pointed out that, whoever the author of the Klacakra may have been, he must have had a thorough exposure to both the ideology of Islam and the violence of the Muslim armies of the day. We might observe that resurgent Islamic threats by agressive Nakbandi Sufis in the Tarim Basin probably precipitated the resurgence of the ambhala myth for the Mongols, resulting in the 1775 CE guidebook to the imaginary land by bLo-bzang dPla-ldan ye-shes, the Third Pa-chen Rin-po-che.59 Even with the ambhala myth, the eclipse of Central Asia as a Buddhist domain was ultimately to cause its sites to become reidentified and relocated into South Asia, specifically Nepal. John Brough has already called attention to the identification of Swayambh with specific Khotanese legends, and we certainly can see that the Bodhantha stpa has appropriated Tibetan legends in its search for authenticity.60 The importance of Central Asia in all of this is, well, central, since Bodhanth itself was older than Buddhism in Tibet, as our medieval Nepalese inscriptions affirm. The movement of pilgrimage sites, involving the identification of one pilgrimage site with another by changing the names or equating their designations, is a common phenomenon in South Asia. It certainly occurred before Puarkas time and occupied much attention in kyarakitas review of such sites, where many of the pilgrimage centers are equated to each other. The same phenomenon is visible in Southeast Asia and Tibet, which have identified areas with the names of important places in Buddhist India. Yet the aura of a sanctified zone which pervades the South Asian image of Central Asia became important in the establishment of Nepalese authenticity, even with Nepals greater antiquity.

THE WILD NORTH ON THE FRONTIERS OF REALITY The idea that a frontier area became influential on the parent civilization has certainly been part of the American persona since the time of Frederick Jackson Turners frontier hypothesis. In what must be the single most influential conference paper in the history of American academia, Turners 1893 paper, The Significance of the Frontier in American History, proposed that American valuesthe speedy integration of

Davidson: Hidden Realms and Pure Abodes

169

various nationalities into a national identity, the emergence of the democratic ideal, and so forthwere not the result of European civilization but the consequence of our encounter with an uncivilized open land. Turners thesis has fallen out of fashion, for many good reasons, but the idea of Central Asia as Chinas frontier has been employed by Owen Lattimore, as well as by others.61 More recently Todd Lewis has seen the Himalayan borderlands as a frontier between Tibet and India, with Nepalese Buddhist communities effecting a network of relationships between the two cultures.62 In many of these discussions, though, frontier has been taken as if a line in the sand, a specific margin of demarcation between the civilized world and the uncivilized. This is certainly the case with His Majestys Government in Indias designation of Peshawar, Swat, Citral, and Swat Kohistan as the Northwest Frontier Agency, for over the border were the dreaded Afghan tribes. However, it may be more fruitful for our purposes if we understand the frontier as a zone rather than a line. Even the American frontier, in the words of an anonymous 1825 reporter, was literally alive with a floating population.63 Faragher, building on such observations, would like to understand the frontier as a zone of cultural confluence. He writes that, The frontier was the region of encounter in between, an area of contest but also of consort between cultures.64 From the view of East Asia, Lattimore emphasized that models of lineal demarcation are poorly formulated. While those in political power tend to draw such lines in the sand, sociologically the frontier can never be the same as a border. Instead he proposed a nomenclature of a trans-Frontier analogous to the model of a frontier zone. The linear Frontier never existed except as a concept. The depth of the trans-Frontier, beyond the recognized linear Frontier, made possible a historical structure of zones, which varied from time to time. These were occupied by a graduated series of social groups, from partly sinicized nomads and semibarbarized Chinese, in the zone adjacent to China, to steppe peoples in Mongolia, forest peoples in North Manchuria and Urianghai, and the peoples of the plateau in Tibet, of whom the more distant were virtually unmodified by such attenuated contacts as they had with China. The oasis peoples of Chinese Turkistan formed another group, with special historical functions. Within this graduated series those groups that adjoined the Great Wall held the (inner) reservoir of political control over the Frontier.65 In many respects, this analysis of the constituents of a frontier develops cross-cultural themes observable elsewhere, including the comparison between the U.S. and Southern Africa.66 However, for our purposes, if we

170

Pacific World

were to replace China in Lattimores statement with India, and spoke of the Kunas, the Sogdians, and others in the Western Tarim and Western Turkestan as he did for Mongols, Manchus, and others, much the same analysis would hold true. Lattimores observation that the oases played a unique part may also be read into Indian literary and political history. While Puruapura cannot be considered an oasis on the same scale as Kucha or Turfan (for these latter are separated from other oasis cultures by weeks of travel), in the Gandhran capitals proximity to Jalalabad, Kapa, Begram, Oiyna, and Bamiyan, it does share characteristics with the upper Oxus cities of Bactra, Surkh Kotal, Termez, Konduz, and Ay Khanum. The difference between the verdant landscape stretching from Lahore south is clearly evident to anyone venturing in the Northwest. These Northwest areas were the first locales to receive foreign incursions of akas, Kunas, Turks, or Mongols from the north; these were the cities and subcultures that most materially profited from the silk trade, and the ones that experienced the economic consequences of its demise. To some degree this exercise demonstrates that the issue of Central Asia and the cities of the silk bazaars is exactly their respective positions between Persia, India, and China. The ethnic groups and migrant nations operating in these areas were not simply barbaric nomads or corpulent oasis burghers, but highly insightful actors and self-consciously opportunistic in their appropriation of selected aspects of the civilizations that brought them all kinds of cultural products and opportunities. Conversely, I would like to suggest that this specifically frontier character of Central Asiathe in-betweenness and fluidity of populations that have been so significant to its history, its position as the meeting ground of languages and culturesis not simply exciting to us, but has been to Buddhist South Asians and Tibetans as well. As missionary monks encountered these zones of multicultural influences, either in situ or in the diasporas of the great cities of North India or China, many of them became enamored of the paradoxical presence of metropolitan sophistication and rural isolation that Central Asia afforded. Whether by chance or by design, the literary personae of such frontier zones have a decided resemblance, and they have been cast in the romantic mode by authors from Mathur to Chang-an. Inhabitants of such frontiers also found they could capitalize on their charismatic aura, representing themselves as a breed apart and the carriers of a secret knowledge developed in a distant land. CONCLUSION The acceptance of Central Asian narratives in this manner was certainly assisted by the use of Indic languages as carriers of Buddhist, astrological, economic, legal, and political information. The employment

Davidson: Hidden Realms and Pure Abodes

171

of Gndhr in Chorasmia, the use of Indic writing systems throughout the Tarim Basin, the development of regional varieties of Sanskrit in Kucha, Khotan, and Gandhra, all certainly contributed to the perception of these areas as somehow Indic. Yet, much the same could be said for r Lka, Burma, Cambodia, or Thailand. In the former case, not only was Singhalese to become the official language, but Pli was and is the authoritative religious language, even though both are Indo-European and doubtless from North India originally. r Lka, though, never enjoyed the same aura in South Asian or Tibetan literature, and the cultural memory of r Lka continues to be of an island of demonesses, even if it was sometimes proposed as the eventual residence of Padmasambhava, where he was to live with his consorts in his palace on Copper-colored Mountain (Tmraparvata) surrounded by tribes of demons. Other locales, such as Pagan in Burma, certainly demonstrated the energy and wealth to be considered at least on par with even the greatest artistic and cultural areas of Bamiyan or the Kuna kingdoms. The Cambodian employment of Sanskrit as an inscriptional language has rightly been considered one of the more interesting linguistic developments in Southeast Asian history. Even more, though, for five centuries (8th12th) Khmr monarchs worked in concert with both Brahmans and ascetics of the aiva Pupata tradition, formulating the Angkor kingdom, with its magnificent artistic and literary heritage. However, back in India, none of these areas were to receive the credit that was their due. Only the areas of Gandhra, Oiyna, Kashmir, Khotan, and the mythic region of ambhala have captured the ancient and medieval Indian imagination and developed a following among the geographically challenged Indian intellectuals of the Gangetic Valley or their later followers in the Kathmandu Valley or on the Tibetan Plateau. Each instance of Central Asian authorityof secret knowledge, hidden realms, quasi-pure lands, or as the source and conclusion of the Dharmais not in and of itself so very important. In aggregate, though, there is no other area of the world that has maintained a visibility so great in the literature and landscape of South Asia. It may be because of the wealth established in these domains while silk was the central currency in the trade between China and Western Asia. It may be because Buddhism caught on in these zones in a way that was exceptional, but then we should see a similar phenomenon in r Lka, Burma, or in other Southeast Asian countries. Instead, I believe that the romantic aura of the desert oasis or isolated culture, encountered after a period of hard travel by merchants (who were secured by their own wealth and influence in India), has been an important factor. All the influential areas are difficult to visit, easily the objects of romantic fantasies, and exceptional primarily in the disparities between their verdant cities and the surrounding desert. The lack of understanding of these areas contributed to their legends, and, even now, some Theoso-

172

Pacific World

phists and a few Indian Brahmans continue to maintain that Tibet is the residence of a superior class of religious personages, avatars of the new age, as I have occasionally been told. Those of us who know Tibet have few such illusions, but to those who have never been to the roof of the world, the esoteric sanctity of the hidden masters in the secret abode survives.

Davidson: Hidden Realms and Pure Abodes

173

NOTES 1. For the Northwest contribution, see Richard Salomon, Gndhr Hybrid Sanskrit: New Sources for the Study of the Sanskritization of Buddhist Literature, Indo-Iranian Journal 44 (2001): pp. 241252; for examples of Kuchean Sanskrit, see articles by Heinrich Lders, Zur Geschichte und Geographie Ostturestans, Sitzungsberichte der Preussischen Akademie der Wissenschaften (1922): pp. 243261; and Weitere Beitrge zur Geschichte und Geographie von Ostturkestan, Sitzungsberichte der Preussischen Akademie der Wissenschaften (1930): pp. 764. On the larger problem of the use of languages in Central Asia, see Jan Nattier, Church Language and Vernacular Language in Central Asian Buddhism, Numen 37 (1990): pp. 195219, and Richard N. Frye, The Heritage of Central Asia: From Antiquity to the Turkish Expansion (Princeton: Markus Wiener Publishers, 1996), pp. 248250. I find it likely that the Yogalehrbuch examined by Dieter Schlingloff (Ein Buddhistisches Yogalehrbuch: Sanskrittexte aus den Turfanfunden, 2 vols. [Berlin: Akademie Verlag, 1964]) was a Tarim production, possibly in Kucha. 2. John Brough, Comments on Third-Century Shan-Shan and the History of Buddhism, Bulletin of the School of Oriental and African Studies 28 (1965): p. 584. 3. On this movement, see Robert C. Holub, Reception Theory: A Critical Introduction. (London & New York: Methuen, 1984). 4. Good bibliographies on Kharoh and Gndhr are found in Grard Fussman, Gndhr crite, Gndhr parle, in Colette Caillat, ed., Dialectes dans les Littratures Indo-Aryennes, Publications de lInstitut de Civilixation Indienne, Fascicule 55 (Paris: ditions e. de Boccard, 1989), pp. 433501; and Meicun Lin, Kharoh Bibliography: The Collections from China (18971993), Central Asiatic Journal 40 (1996): pp. 188220; for the West Bengal finds, see B. N. Mukherjee, Kharoh and Kharoh-Brhm Inscriptions in West Bengal (India), Indian Museum Bulletin 25 (1990). 5. References in John E. Mitchiners Introduction to the Yuga-pura, critically edited, with an English translation and a detailed introduction by John E. Mitchiner, Bibliotheca Indica Series 312 (Calcutta: The Asiatic Society, 1986), pp. 3745. 6. Translation mine; quoted in in Mitchiners Introduction to the Yugapura, p. 11: mlecch hi yavans teu samyakstram ida sthitam |. 7. Introduction, in The Yavanajtaka of Sphujidhvaja, David Pingree, ed. and trans., Harvard Oriental Series vol. 48 (Cambridge, MA: Harvard University Press, 1978), pp. 2341.

174

Pacific World

8. Milindapaha Pli, Swami Dwarikadas Shastri, ed., Bauddha Bharati Grantamala 13 (Varanasi: Bauddha Bharati, 1979), pp. 23. 9. Grard Fussman, LIndo-Grec Mnandre ou Paul Demiville revisit, Journal Asiatique 281 (1993): p. 68. 10. M. J. Przyluski, Le Vinaya des Mula-Sarvastivadin et les Textes Apparents. Journal Asiatique, series 11, volume 4 (1914): pp. 493568; tienne Lamotte, History of Indian Buddhism: From the Origins to the aka Era, Sara Webb-Boin trans. (Louvain-La-Neuve: Institut Orientaliste, 1988), pp. 208210; and John S. Strong, The Legend and Cult of Upagupta (Princeton: Princeton University Press, 1992), pp. 2643, surveys most of the surviving sources. 11. Bhaiajyavastu, Nalinaksha Dutt, ed., Gilgit Manuscripts, vol. 3, 1, The Kashmir Series of Texts & Studies, no. 71 (E) (Srinagar: Jammu & Kashmir Research Department and Calcutta Oriental Press, 1947), p. 2: ea caturvaraataparinirvtasya mama vajrape [kuanavaya] kaniko nma rj bhaviyati | so asmin pradee stpa pratihpayati | tasya kanikastpa iti saj bhaviyati | mayi ca parinirvte buddhakrya kariyati | 12. See Kuwayama Shoshin, The Buddhas Bowl in Gandhara and the Relevant Problems, In South Asian Archaeology 1987, vol. 2, Maurizio Taddei, ed. (Rome: Istituto Italiano per Il Medio ed Estremo Oriente, 1990), pp. 945978. The double-bodied Buddha is described Ta tang hsi y chi, in Taish, vol. 51, no. 2087, pp. 880a15b3; Samuel Beal, trans., Si-yu-ki: Buddhist Records of the Western World, vol. 1 (London: Kegan Paul, Trench, Trbner & Co., 1869), pp. 1023; K. Walton Dobbins, The Stupa and Vihara of Kanishka I, The Asiatic Society Monograph Series, vol. 18 (Calcutta: The Asiatic Society, 1971), p. 13. An eighth century painting from Khocho is represented in Herbert Hrtel and Marianne Yaldiz, Along the Ancient Silk Routes: Central Asian Art from the West Berlin State Museums (New York: The Metropolitan Museum of Art, 1982), pp. 1834. Perhaps the most interesting example is the thirteenth century statue from Khara Khoto; see Mikhail Piotrovsky, ed., Lost Empire of the Silk road: Buddhist Art from Khara Khoto (X-XIIIth Century) (Milan: Electapp, 1993), pp. 1045. 13. In the Pupradnvadna of the Divyvadna, P. L. Vaidya, ed., Buddhist Sanskrit Texts, no. 20 (Darbhanga: The Mithila Institute, 1959), p. 216; for Upaguptas position, see Strong, The Legend and Cult of Upagupta, pp. 389. 14. Ta tang hsi y chi, in Taish, vol. 51, pp. 879c15880b15; Beal Si-yu-ki: Buddhist Records of the Western World, vol. 1, pp. 99103. Gregory Schopen has forwarded this position in his articles, Burial Ad Sanctos and the Physical Presence of the Buddha in Early Indian Buddhism: A

Davidson: Hidden Realms and Pure Abodes

175

Study in the Archaeology of Religions, Religion 17 (1987): pp. 193225; On the Buddha and His Bones: The Concept of Relic in the Inscriptions of Ngrjunikoa, Journal of the American Oriental Society 108 (1988): pp. 527537; and The Buddha as an Owner of Property and Permanent Resident in Medieval Indian Monasteries, Journal of Indian Philosophy 18 (1990): pp. 181217. One could certainly argue that, because the statement is used to refer to both the stpa and to an arhat, this indicates the living quality of the stpa. Be that as it may, its similarity to Upagupta is predicated on, first, its real difference from the Buddha (as Upagupta was), and on the ideology of the survival of the Buddhas message conveyed through different personalities, as Upaguptas position in the lineage of Kashmiri-Gandhran patriarchs clearly represented. 15. Martha L. Carter (Petroglyphs at Chilas II: Evidence for a Pre-Iconic Phase of Buddhist Art in Gandhara, in South Asian Archaeology 1991, Adalbert J. Gail and Gerd J. R. Mevissen, eds. [Stuttgart: Franz Steiner Verlag, 1993], pp. 349366) illustrates pre-Kuna stpa-s, very different from the plaques illustrated in Ram Chandra Kak (Ancient Monuments of Kashmir [London: The India Society, 1933], pl. XVIII), which show startling similarities to the small stpa-s of Turfan and Gansu discussed by Hubert Durt, et al. (A propos de stpa minatures votifs du Ve sicle dcouverts Tourfan et au Gansu, Arts Asiatiques 40 [1985]: pp. 92106). 16. Strong, The Legend and Cult of Upagupta, pp. 2832. 17. Pou sou pan tou fa shih chuan, in Taish, vol. 50, no. 2049, pp.190c15 16; Takakusu Junjir, The Life of Vasu-Bandhu by Paramartha (A.D. 499 569), Toung Pao, ser 2, vol. 5 (1904): p. 290. 18. This point is on the issue of the nature of the roots of goodness (three or four), and the nature of the fourth of them. AbhidharmakoabhyaVykhy, ascribed to Yaomitra, Swami Dwarikadas Shastri, ed., in Abhidharmakoa and Bhya of Acharya Vasubandhu with Sphutrth Commentary of chrya Yaomitra, Bauddha Bharati Series, no. 5 (Varanasi: Bauddha Bharati, 1970), pp. 429.23, and 795.25; Abhidharma-Mahvibhastra, in Taish, vol. 27, no. 1545, pp. 795c921. 19. Dobbins, The Stupa and Vihara of Kanishka I, p. 54. 20. Ta tang hsi y chi, in Taish, .vol. 51, pp. 896b20896c19; Beal, Si-yu-ki: Buddhist Records of the Western World, vol. 1, pp. 226228. 21. For a full translation and discussion of the value of this section, see Ronald M. Davidson, Mahyna in Gandhra: The Problem of Artistic Representations, Textual Sources and the Public Intellectual, in Proceeding of the International Conference on Gandharan Buddhism: An Interdisciplinary Approach, Kurt Behrendt and Pia Brancaccio, eds. (Toronto: University of Toronto Press [forthcoming]). The section is found in Bodhisattvabhmi, Woghihara Unrai, ed. (193036, Tokyo: Sankib

176

Pacific World

Busshorin, 1971), pp. 152.18155.21; Nalinaksha Dutt, ed., Tibetan Sanskrit Works Series, vol. 7, (Patna: K.P. Jayaswal Research Institute, 1966), pp. 105.4107.5; Taish,.vol. 30, no. 1579, pp. 514b11515a9; Taish, vol. 30, no. 1581, pp. 911b18913a5; Ui Hakuji, et. al., eds., A Complete Catalogue of the Tibetan Buddhist Canons (hereafter, To.) (Sendai: Tohoku Imperial University, 1934), 4037, 82b584a5. Cf. Mark Tatzs translation from the Tibetan, with its interpolations, Asangas Chapter on Ethics With the Commentary of Tsong-Kha-pa: The Basic Path to Awakening, the Complete Bodhisattva, Studies in Asian Thought and Religion, vol. 4. (Lewiston, NY: The Edwin Mellen Press, 1986), pp. 6062. 22. Bodhisattvaprtimoka, in Taish, vol. 24, no. 1500, pp. 1107a1110a. The translation makes difficult to ascertain clues as to whether it was based on the Bodhisattvabhmi or vice versa. 23. Paul Groner, The Fan-wang ching and Monastic Discipline in Japanese Tendai: A Study of Annens Futsu jubosatsukai koshaku, in Chinese Buddhist Apocrypha, Robert Buswell, ed. (Honolulu: University of Hawaii Press, 1990), p. 254. 24. Nalinaksha Dutt, Bodhisattva Pratimoksa Sutra, Indian Historical Quarterly 7 (1931): pp. 259286. Yamabe Nobuyoshi informs me that Fujita Kkan (Bya chub bza po cho Bosatsu ritsugi giki ni tsuite, Mikky Bunka 141 (1983): pp. 87100) has identified this work in part with the Bodhisavaravidhi (To. 3967) of Bodhibhadra (active ca. 1000 CE). 25. I wish to thank Yamabe Nobuyoshi for attracting my attention to this suggestion by Okimoto Katsumi, Bodhisattva Prtimoka, Indogaku Bukkygaku Kenky 21 (1972): pp. 130131. 26. See Kuwayama Shshin, Linscription du Gaea de Gardez et la chronologie des Turki-hi, Journal Asiatique 279 (1991): pp. 267287. 27. Ta tang hsi y chi, in Taish, vol. 51, p. 923a8; Beal, Si-yu-ki: Buddhist Records of the Western World, vol. 2, p. 165; Paul Demiville, Lorigine des sectes bouddhiques daprs Paramartha, Mlanges chinois et bouddhique 1 (1932): pp. 6061. 28. Lo yang Chia lan chi, Taish, vol. 51, no. 2092, p. 1020b18; Yang Hsanchih, trans., A Record of Buddhist Monasteries in Lo-Yang (Princeton: Princeton University Press, 1984), p. 223. 29. Ta tang hsi y chi, Taish, vol. 51, p. 882b1314; Beal (Si-yu-ki: Buddhist Records of the Western World, vol. 1, p. 120) does not represent this section entirely accurately. 30. See Kraavyha , in Adelheid Mette, ed. and trans., Die Gilgitfragmente des Kraavyha, Indica et Tibetica Monographien zu end Sprachen und Literaturen des indo-tibetischen Kulturraumes, band 29 (Swisttal-Odendorf: Indica et Tibetica Verlag, 1997), p. 97. We notice the

Davidson: Hidden Realms and Pure Abodes

177

maintenance of this language even in the Newari ms. transcribed by Mette (Die Gilgitfragmente des Kraavyha, p. 143). R. O. Meisezahl was also of this opinion in his work on the Amoghapahdayadhra; see R. O. Meisezahl, The Amoghapahdayadhra: the Early Sanskrit Manuscript of the Ruiunji Critically Edited and Translated, Monumenta Nipponica 17 (1962): p. 270. On the topic of birch bark as a manuscript material, see Richard Salomon, Ancient Buddhist Scrolls from Gandhra: British Library Kharroh Fragments (Seattle: University of Washington Press, 1999), pp. 1522, 5771, and 81109; Michael Witzel, Kashmiri Manuscripts and Pronunciation, in A Study of the Nlamata: Aspects of Hinduism in Ancient Kashmir, Ikari, Yasuke, ed. (Kyoto: Institute for Research in Humanities, Kyoto University, 1994), pp. 614. The location of Uiyna/Oiyna has been finally affirmed by inscriptions; see Kuwayama Linscription du Gaea de Gardez et la chronologie des Turki-hi, pp. 267287. 31. Abhidharmakoabhya VII. 47, Pralhad Pradhan, ed., Abhidharmakoabhyam of Vasubandhu, Tibetan Sanskrit Works Series, vol. 8 (Patna: K.P. Jayaswal Research Institute; 2nd., rev. ed., 1975), pp. 4245. 32. I have discussed the Indrabhti preaching myth in my Indian Esoteric Buddhism: A Social History of the Tantric Movement (New York: Columbia University Press, 2002); please see it for the myths sources. 33. Alex Wayman, Yoga of the Guhyasamajatantra: The Arcane Lore of Forty Verses (Delhi: Motilal Banarsidass, 1977), pp. 322; unfortunately, the translation of the verses is occasionally flawed. 34. Sahajasiddhi-paddhati, fols. 6a67b2 (Asc. Lha-lcam rje-btsun-ma dpalmo [?=Devbharikr], To. 2211, bsTan-gyur, rgyud, zhi, fols. 4a3 25a1; Pe. 3108, bsTan-gyur, rgyud-grel, tsi, fols. 4b829a7. ); this material is given greater depth in Ronald M. Davidson, Reframing Sahaja: Genre, Representation, Ritual and Lineage, Journal of Indian Philosophy 30 (2002): pp. 4583. 35. Carygtikoa (#3, in Per Kvrne, An Anthology of Buddhist Tantric Songs: A Study of the Carygti, Det Norske Videnskaps-Akademi II Hist.-Filos, Klasse Skrifter Ny Serie, no. 14 [Oslo: Universitetsforlaget, 1977], pp. 48, and 86). 36. E.g., Phdiniraya, attributed to kyarakita, To. 1606. bsTan-gyur, rgyud, a, fols. 131b7133b6. 37. Giuseppe Tucci, Travels of Tibetan Pilgrims in the Swat Valley, Opera Minora, Studi Orientali Pubblicati a cura della Scoula Orientale, vol. 4., part 2 (Rome: Rome University, 1971), pp. 369418. 38. H. W. Emerson, ed., Punjab Gazetteers: Mandi State (Lahore: Superintendent, Government Printing, Punjab, 1920), pp. 17 and 203.

178

Pacific World

39. A Sanskrit fragment of this section from the Mahyna Mahparinirvastra is found in A. F. R. Hoernle, Manuscript Remains of Buddhist Literature Found in Eastern Turkestan (Oxford: Clarendon Press, 1916), pp. 9397. 40. Ta tang hsi y chi, Taish, vol. 51, p. 880b12; Beal, Si-yu-ki: Buddhist Records of the Western World, vol. 1, p. 103. 41. Jan Nattier, Once Upon a Future Time: Studies in a Buddhist Prophecy of Decline (Berkeley: Asian Humanities Press, 1991), pp. 178 and 226. 42. Two Saghavardhana-vykaraa versions survive in Tibetan, one canonical and one from Tun Huang, and are edited by Patrizia Cannata (La Profezia dellArhat della Terra di li Riguardante il Declino della Fede nella Vera Legge, in Paolo Daffin, ed., Indo-Sino-Tibetica: Studi in Onore di Luciano Petech, Universit di Roma, La Sapienza, Studi Orientali, vol. 9 [(Rome: Bardi Editore, 1990], pp. 4379) who also provides references to the manuscript of a Chinese translation found in Tun Huang. F. W. Thomas (Tibetan Literary Texts and Documents Concerning Chinese Turkestan, Oriental Translation Fund, N.S. vols. XXXII, XXXVII, XL, [London: Luzac & Company, Ltd, 193555], vol. 1, pp. 4169 and 307) discusses and translates the canonical version, and believed that it, as well as other scriptures describing Khotan, were not written there but in Tun Huang, Sha-chou or Tibet. Jan Nattier (Once Upon a Future Time, pp. 189 n. 98, and 194 n. 121) disputes this, maintaining their composition in Khotan, and the presence of two Tibetan and one Chinese version of the Saghavardhanavykaraa would seem to support her position on at least this text. For other Khotanese scriptures in Tibetan translation, see R. E. Emmerick, Tibetan Texts Concerning Khotan, London Oriental Series, vol. 19 (London: Oxford University Press, 1967). 43. Nattier, Once Upon a Future Time, p. 228. 44. mKhas pa ldeus mdzad pai rgya bod kyi chos byung rgyas pa, Chabspel tshe-brtan phun-tshogs and Nor-brang o-rgyan, eds. (Lhasa: Bod ljongs mi rigs dpe skrun khang, 1987), pp. 398407. The chronology for this text has been proposed by Leonard van der Kuijp (On Dating the two lDeu Chronicles, Asiatische Studien 46 [1992]: pp. 468491) to be in the mid to late thirteenth century. 45. Dan Martin (Tibetan Histories: A Bibliography of Tibetan-Language Historical Works (London: Serindia Publications, 1997), p. 39) indicates that kyar is credited with three chronological calculations, 1204, 1207, and 1210. The common era equivalents for these calculations have been considered by Ariane MacDonald (Prambule la Lecture dun Rgya-bod Yig-chan, Journal Asiatique 251 [1963]: p. 97); Yamaguchi Zuih (Methods of Chronological Calculation in Tibetan Historical Sources, in Louis Ligeti, ed., Tibetan and Buddhist Studies Commemorating the 200th Anni-

Davidson: Hidden Realms and Pure Abodes

179

versary of the Birth of Alexander Csoma de Krs, Bibliotheca Orientalis Hungarica, vol. 29 (Budapest: Akadmiai Kiad, vol. 2 [1984], pp. 420421) corrects this chronology and employs the 1207 dating of kyar as well. I have (Gsar ma Apocrypha: The Creation of Orthodoxy, Gray Texts, and the New Revelation, in Helmut Eimer and David Germano, eds., The Many Canons of Tibetan Buddhism [Leiden: Brill. 2002], p. 209) erred in giving this latter date as 544 CE.
46. Bla ma rje btsun chen poi rnam thar, in Sa-skya bKa-bum (Bod Nams Rgya Mtsho, 1969) , V148.1.32.5. 47. His sources include (148.1.62.4) the acceptance of the Khotanese reckoning by Grags pa rgyal mtshan, the Khotanese Saghavardhanavykaraa (see Thomas, Tibetan Literary Texts and Documents, vol. 1, pp. 3969) and the Candragarbha-stra (see Nattier, Once Upon a Future Time, pp. 22877), the *Vimaladevvykaraa, which is the source normatively cited for the famous line that 2500 years after my nirvana, the Saddharma will spread to the country of the red faces, (Thomas, Tibetan Literary Texts and Documents, vol. 1, p. 139) and references to the Majurmlakalpa. For a further consideration of this issue, see Davidson,Gsar ma Apocrypha, pp. 203224. 48. Much has been made of the Chan lineage of Ho-shang Mo-ho-yan, but other systems came in as well; see Marcelle Lalou, Document Tibtain sur lexpansion du Dhyna Chinois, Journal Asiatique 231 (1939): pp. 505 523. He (I believe incorrectly) indicates these systems as primarily or exclusively Chinese. See also Jeffrey Broughton, Early Chan Schools in Tibet, in Robert M. Gimello and Peter N. Gregory, eds., Studies in Chan and Hua-yen (Honolulu: University of Hawaii Press, 1983), pp. 168. Broughton sees Chan systems as an extension of Chinese and Korean developments, even while acknowledging the Tun Huang connection. 49. Ruth W. Dunnell, The Great State of White and High: Buddhism and State Formation in Eleventh-Century Xia (Honolulu: University of Hawaii Press, 1996), p. 158. 50. Ruth W. Dunnell, The Hsia Origins of the Yan Institution of Imperial Preceptor, Asia Major, ser. 3, vol. 51 (1992): pp. 85111; and Elliot Sperling, Lama to the King of Hsia, Journal of the Tibet Society 7 (1987): pp. 3150. 51. Klacakra-tantra, Biswanath Banerjee, ed., A Critical Edition of r Klacakratantra-Rja (Calcutta: The Asiatic Society, 1985), I.150: bhmau kailsakhaa himigisahita tat tribhga samantt | bhye caikaikapatra dinakaraviayair bita dvpadeai | savyrdhe ambhalkhya munivaranilaya grmakoydhivsa | koigrmair nibaddho bhavati hi viayo maala grmalakai || 52. Roberto Vitali, The Kingdoms of Gu.ge Pu.hrang (Dharamsala: Tho.ling gtsug.lag.khang lo.gcig.stong khor.bai rjes.dran.mdzad sgoi go.sgrig

180

Pacific World

tshogs.chung, 1996) pp. 153299. Giacomella Orofino (Apropos of Some Foreign Elements in the Klacakra-tantra, in Helmut Krasser, et. al., eds., Tibetan Studies: Proceedings of the 7th Seminary of the International Association for Tibetan Studies, Graz 1995, vol. 2 (Vienna: Verlag der sterreichischen Academie der Wissenschaften, 1997), pp. 717724) has discussed foreign elements in the Klackara, and has concluded (p. 724) that this literature (or at least a nucleus of it) was formulated in the regions of the North-West India, a part of India that, as is well known, played a fundamental role in the formation of the Buddhist Tantras. 53. Vimalaprabh, Upadhyaya, Jagannatha, ed., Bibliotheca Indo-Tibetica Series, no. 11 (1986), Vrajavallabh Dwivedi and S. S. Bahulkar, eds., Rare Buddhist Texts Series 12, 13 (Sarnath: Central Institute for Higher Tibetan Studies, 1994), vol. 1, pp. 79: tnadyuttare sucandravaineyn aavatikoigrmanivsin paramdibuddhatantrarjena samyaksambuddhamrgalbhya sucandro rj bht bodhisattvo vajrapir iti | ; cf. Abhidharmakoabhya III.57; Pradhan, ed., p. 162.22 23: [himavata] yata catsro nadya sravanti gag sindhu t vaku ca | For puric sources, see Dinesh Chandra Sircar, Studies in the Geography of Ancient and Medieval India (Delhi: Motilal Banarsidass, 1971), pp. 678. 54. Vimalaprabha, vol. 1, pp. 312. 55. Amtakaik, p. 1, provides a quotation from the rbhaddibuddha, but this is not found in the Laghu-Klacakra: gdhraku=e yath str prajpramitnaye | tathmantranaye prokt rdhnye dharmadean || Also suggestive is the knowledge of Pli exhibited in one place in the Vimalaprabh, p. 31.2931. 56. Vimalaprabh, vol. 1, p. 3: yad vykta daabalena purlpatantra guhydhipasya gaditt paramdibuddht | tat klacakralaghutantram ida kalpe majuriy nigadita sakala munnm || Analogous statements are found Vimalaprabh, pp. 2231. 57. The reference to Bu-stons Dus skor chos byung is in George N. Roerich, trans, The Blue Annals, The Royal Asiatic Society of Bengal Monograph Series, vol. 7.(Calcutta: The Royal Asiatic Society of Bengal, 1949), vol. 2, p. 754, n (I do not have access to Bu-stons complete works at this time). 58. Vimalaprabh, vol. 1, p. 26. 59. For a good consideration of the later Tibetan and Western fascination with ambhala, see Edwin Bernbaum, The Way to Shambhala: A Search for the Mythical Kingdom beyond the Himalayas (New York: Doubleday Anchor, 1980). 60. John Brough, Legends of Khotan and Nepal, Bulletin of the School of Oriental and African Studies 12 (1948): pp. 333339; for a discussion on the

Davidson: Hidden Realms and Pure Abodes

181

literary history of the Bodhanth myth, see Anne-Marie Blondeau, Byarung Kha-shor, Lgende foundatrice du Bouddhism Tibtain, in Per Kvrne, ed., Tibetan Studies: Proceedings of the 6th Seminar of the International Association for Tibetan Studies, vol. 1 (Oslo: Institute for Comparative Research in Human Culture, 1994), pp. 3148. 61. Owen Lattimore, Studies in Frontier History: Collected Papers 1928 1958 (London: Oxford University Press, 1962). 62. Todd T. Lewis, Himalayan Frontier Trade: Newar Diaspora Merchants and Buddhism, in Charles Ramble and Martin Brauen, eds., Proceedings of the International Seminar on the Anthropology of Tibet and the Himalaya (Zurich: Ethnological Museum of the University of Zurich, 1993), pp. 165178. 63. Quoted in John Mack Faragher, Rereading Frederick Jackson Turner: The Significance of the Frontier in American History and Other Essays, with Commentary by John Mack Faragher (New York: Henry Holt and Co., 1994), p. 239. 64. Faragher, Rereading Frederick Jackson Turner, p. 239. 65. Lattimore, Studies in Frontier History, p. 115. 66. See the comparative essays in Howard Lamar and Leonard Thompson, eds., The Frontier In History: North America and Southern Africa Compared (New Haven: Yale University Press, 1981).