Printer Friendly

The early development of the Padmasambhava legend in Tibet: a study of IOL Tib J 644 and Pelliot tibetain 307.

This article offers some new evidence on Padmasambhava, the Indian master who, according to legend, was instrumental in establishing Buddhism in Tibet. In the course of my research on tantra in the Tibetan manuscripts discovered near Dunhuang, I have found two passages relating to the early development of the legends surrounding this famous Buddhist master, neither of which have been studied to date. (1) The two passages are presented below in translation, and discussed in light of the other available early evidence.

The results of this study reveal a mutability in the early biographies of Padmasambhava. The master's role in the Tibetan imagination grew and evolved in dramatic ways during the ninth to eleventh centuries, so that by the time of his first complete biography, the twelfth-century Zangs gling ma by Nyang ral nyi ma'i 'od zer (1124-1192), Padmasambhava had become the single most important figure in Tibetan narratives of their early conversion to Buddhism. The new evidence presented here contributes to our understanding of how these Tibetan conversion narratives grew over the early years. The present inquiry is therefore less concerned with Padmasambhava as a historical person than with his legend and the thematic lines along which it developed. (2) An evaluation of the early evidence helps to clarify both how Tibetans perceived themselves and how they understood their first encounters with the Buddhist religion.

This new evidence indicates that the Padmasambhava legend initially flourished during the so-called "dark period" of Tibetan history. This period stretched from 842 C.E. when the Tibetan empire collapsed, to roughly 978 C.E. when a royal court and Buddhist monastic institutions began to reappear, bringing with them a new orthodoxy. According to traditional Tibetan historical sources, this period of one and a half centuries witnessed a horrific degradation of Buddhism, as monasteries were persecuted and the teachings corrupted. Recent scholarship has begun to question this traditional version of events. (3) Certainly, the Tibetans who emerged from the dark period were far more Buddhist, however such affiliation is measured, than the Tibetans who had entered it. It seems that despite the closing of the monasteries Buddhism continued to flourish at the local level. The forms Buddhism took during these years may well have been "corrupt" in the view of later Tibetans, but these same corruptions were fundamental to the formation of the Tibetan Buddhist identity. Freed from the watchful eye of the imperial court and the monastic orthodoxy, Tibetans of the dark period were able to make Buddhism their own. The themes, the imagery, and the strategies Tibetans developed during the inchoate years of the dark age formed the cultural foundations upon which Tibetan Buddhism was built. Only by excavating these foundations and shedding some light on the dark period can we gain a clearer appreciation of the Tibetan assimilation of Buddhism.

The manuscripts discovered at Dunhuang provide a glimpse of the events of this era. It is increasingly clear that most of the Dunhuang materials date from the dark period, well after the collapse of the Tibetan empire, (4) and what the Dunhuang collections reveal about the Tibetan assimilation of Buddhism is the central role that tantric Buddhism played in this process. (5) Earlier, during the empire, the exoteric traditions enjoyed far greater support, thanks particularly to the patronage of the royal court, while the translation of tantric texts was carefully controlled, if not prohibited. With the collapse of the empire, these controls were lifted, and Tibetans plunged eagerly into the world of Buddhist tantra.

One of the constant motifs of Tibetan religion over the centuries has been the animated, and often malevolent, landscape, and the need to mollify, pacify, or subjugate it. The materials I will examine here suggest that, rather than being something projected back into Tibetan history by later histories and chronicles, this motif is a key element in some of the earliest Tibetan Buddhist legends. The Tibetans seem to have been attracted to tantra in part for its effectiveness in controlling spirits and demons. The Tibetan universe is infused with spirits--spirits that live in the rocks, the trees, and the mountains, spirits that live in one's body, that wander the landscape, that live underground and in the sky, spirits that cause illness or natural disasters. The spirit world of Tibet is an unruly domain. Spirits demand recognition and respect, yet they are forever changing names, can be associated with multiple locations, appear in different groups, escape classification, and manifest themselves in accordance with shifting iconographies. Conversely, tantric ritual is often guided by metaphors of power and control, with the practitioner seated as a virtual sovereign at the center of the mandala palace, ruling over the realm by threat of violence. Buddhism provided both ritual methods of control and overarching narrative schemes for explaining the spirits' roles in Tibetan life. Through tantra the spirits could be mapped onto the Tibetan landscape and correlated with the more orderly Buddhist system of deities.

The evidence presented below suggests that the legends surrounding the Indian tantric master Padmasambhava should be understood as part of this tantric conversion of Tibet. The theme of demon subjugation is crucial to Tibetan culture, and Padmasambhava is the demon tamer par excellence. He is also often depicted as the principal figure responsible for bringing Buddhism to Tibet. Today the geography of Tibet is covered with countless sacred sites where the tantric saint is said to have subjugated local Tibetan pre-Buddhist spirits and converted them to Buddhism. The new evidence offered here reveals much about how these two themes, of Padmasambhava and Tibetan tantra, developed in concert.


The first piece of evidence appears in a short manuscript held in the Stein collection. Though catalogued by Louis de la Vallee Poussin in 1918, ITJ644 has been overlooked by scholars, apparently because of the unremarkable entry it received: "A treatise on the Phalas (compare Abhidharma-kosa, VI, Madhyamaka-vrtti, XXIV)." (7) The remainder of the entry is limited to the manuscript's opening and closing lines, neither of which provide any indication of the work's actual significance. There are in fact two items in the manuscript, which are closely related. The first item is a presentation of a nine vehicles (theg pa dgu) doxographical system, the second a discussion of the different vidyadhara levels, which are also grouped by vehicle. (8)

The relevant passage appears in the second item, in the context of the three vidyadhara levels associated with the vehicle of Kriya tantra. Here vidyadhara 'knowledge bearer' refers to one who has mastered the teachings of the Buddhist tantras. A vidyadhara's "knowledge" is specifically one of magic spells, and throughout Indian literature these beings are endowed with abilities to fly, to travel to other realms, and to perform spells. (9) According to our text, there are three vidyadhara levels that can be attained through the practice of the Kriya tantras: the vidyadhara of accomplishments (grub pa'i rigs 'dzin), the vidyadhara who dwells on the levels (sa la gnas pa'i rigs 'dzin), and the spontaneously accomplishing vidyadhara (lhun kyis grub pa'i rigs 'dzin). The description of the second of these, the vidyadhara who dwells on the levels, reads as follows:
 Then Vajrapani arrived and granted the siddhis. Then he went to the
 Asura Cave, and upon beholding the visage of an emanation of Vajrapani
 present there, he struck the rock with his foot. It seemed as if he
 had stuck it into dough. From that footprint the sacrament (samaya)
 descended, and from within that there came a spring with eight
 streams. One flowed to the south face of Mt. Meru, so that the spring
 was called Asvakarna. Seven of them fell inside of the Asura Cave. In
 this [spring] he cleansed himself and gained accomplishment. Thus he
 became one who is called a vidyadhara who dwells on the levels. (10)

Unfortunately the protagonist of this short story remains unnamed, but several links between this story and the later depictions of Padmasambhava. First, Padmasambhava is commonly associated with an Asura Cave at Yang le shod, located near modern-day Pharping on the edge of Kathmandu Valley in Nepal. (11) Today this is one of the most important Padmasambhava religious sites and is recognized by most Newar and Tibetan Buddhists as the place where Padmasambhava perfected the ritual systems for the deity Vajrakilaya.

This modern belief is generally consistent with the principal Dunhuang source on Padmasambhava that has already been studied. PT44 is a work devoted to the ritual traditions of Vajrakilaya. (12) It relates how Acarya Sambhava, as he is called, practiced meditation, battled demons, and performed miracles at the Asura Cave at Yang la shod:
 After arriving at Yang[-la]-shod in Nepal, he performed the practices
 belonging [to all the classes of yoga] from the general Kriya up
 through Atiyoga. He proclaimed each and every transmission of the
 Kila, for the purposes of all the vehicles, from the Hundred Thousand
 [Verse] Tantra of Vajrakila, as [is affirmed] in all the secret
 tantras. In that way, having definitively established the
 transmissions concerning attainment, and having again escorted the
 Hundred Thousand [back to Nepal], Acarya Sambhava then performed the
 rites of attainment in the Asura cave with the Newari Ser-po, Indra-
 shu-gu-ta, Pra-be-se, and others. And thus he performed the rites,
 impelling the four Bse goddesses, whose embodied forms had not passed
 away. He named them Great Sorceress of Outer Splendor, Miraculous
 Nourisher, Great Witch Bestowing Glory, and Life-Granting Conjuress.
 Having performed the great attainment for seven days, he manifestly
 be-held the visage of Vajrakumara [the Adamant Youth, an epithet of
 Having acquired the accomplishment of the Kila, concerning [his
 attainment of] the signs, Padmasambhava, having set a limitless forest
 ablaze, thrust [the Kila] at the blaze. Srigupta, having struck it at
 the rock in the region of the frontier forest of India, broke the rock
 into four fragments and thus "thrust it at stone." The Newari Ser-po
 thrust it at water and so reversed the water's course, thereby
 establishing Nepal itself as a mercantile center. Such were the
 miraculous abilities and powers that emerged. (13)

A number of parallels between this account and ITJ644 immediately present themselves. Until now PT44 had been our earliest evidence of the Padmasambhava legend. In light of the parallels in ITJ644, PT44 now appears to be just one rendition of a popular story.

Both accounts have their respective protagonists travel to the Asura cave, practice Kriya yoga, behold a vision of a wrathful deity's "visage" (zhal mthong), and then perform miracles involving the alteration of natural formations in the landscape. Taking these broad parallels as our focus, we can compare the more specific differences to reveal thematic lines along which the story may have developed. In ITJ644 the master practices only Kriya, while in PT44 he practices all the tantric vehicles from Kriya to Atiyoga. Similarly, in the former manuscript he venerates the deity Vajrapani, while in the latter Vajrakilaya. Both of these differences--the additional vehicles and the altered deity--suggest that the Pelliot manuscript represents a later version of the Asura cave narrative than our shorter Stein passage. Recent scholarship has noted a marked increase in the popularity of Atiyoga and Mahayoga (with which Vajrakilaya is usually associated) among Tibetans during the ninth and tenth centuries, and we might expect these shifts to be reflected in the Dunhuang documents that date from this same period. (14)

The impression that PT44 represents a later version of events may be supported by Tsuguhito Takeuchi's recent dating of the manuscript to the late tenth century, near the time of the closing of the Dunhuang cave. (15) As Bischoff and Hartman noted in their 1971 article, a fragment of a Tibetan date is found on the cover page of PT44 that reads "the second year, the tiger year" (lo gnyis stag gi lo). They were unable to identify this date, but Takeuchi succeeds where they did not. The paper used to make PT44 was apparently recycled; in its previous incarnation it had been a letter from the Khotanese king to the Chinese ruler of Shazhou. On the basis of the Tibetan date noted by Bischoff, a Chinese seal, and a still legible fragment of the original letter written in Khotanese, Takeuchi was able to date the Khotanese letter to the year 978. This means, concludes Takeuchi, "that the Buddhist text on Padmasambhava was written even later, namely after the 980s." (16)

Thus PT44 may be one of the latest of the Dunhuang documents (given that the cave was sealed in the early eleventh century), a fact that at least does not contradict the idea that ITJ644 represents an earlier tradition of the legends surrounding Asura caves. In short, certain details of an earlier Asura cave story appear to have been reworked, as in PT44, into a more elaborate form that supported the Padmasambhava and the Vajrakilaya ritual traditions. (17)

A similar reworking of the legend may be seen in the Zangs gling ma, the early Padmasambhava biography by Nyang ral nyi ma'i 'od zer. Where ITJ644 says that our unnamed hero attained the vidyadhara of "dwelling on the levels" while meditating at the Asura Cave, Nyang ral has him attaining the vidyadhara of mahamudra. The vidyadhara of mahamudra does appear in ITJ644, but later, as the highest level of vidyadhara, attained through the practice of Mahayoga. (18) As in PT44, Padmasambhava's level of practice during his visit to Nepal was made higher in the Zangs gling ma, possibly updated to appeal to later Tibetan interests.

There is a further element in ITJ644 that deserves attention: the account of the spring that was created out of the miraculous footprint in the rock. The modern-day Padmasambhava site in Nepal is located near a natural spring. Moreover, according to the later Buddhist tradition, when subjugating the local naga spirit, Padmasambhava left an imprint of his ritual dagger (kila) in the rocks directly above the spring. From this imprint "water emerges at certain auspicious moments." (19)

Apart from these modern traces of Padmasambhava's legendary activities, I have been unable to locate any accounts of eight streams in the later biographies of the master. However, a clue does appear in the recently discovered Dba' bzhed, a narrative history of the Tibetan imperial period that may date from as early as the eleventh century. (20) The relevant passage takes us to Padmasambhava's time in central Tibet, when he is said to have assisted the Tibetan king (btsan po) Khri srong lde'u brtsan in the subjugation of demons and the irrigation of the lands around Bsam yas monastery:
 Later, as the bTsan po was washing his hair and the mKhan po
 [Padmasambhava] heard about this, he asked where the water used for
 washing the head of the bTsan po had been taken from. bZhes zla [a
 close servant of the king] answered that it had been taken from the
 rTsang chab river of 'Om bu tshal. The mKhan po said: "This is of no
 use. There is a spring called rTa rna on top of the Ri rab [i.e., Mt.
 Meru]. If the water for washing the bTsan po's hair is taken from
 there, this will bring him long life and high political
 authority." (21)

The translators of this passage, Wangdu and Diemberger, remark on the unusual name of the spring: "rTa rna, i.e., 'Horse Ears', is known as the Tibetan form for Asvakarna, one of the Golden Mountains around Mt. Meru/Ri rab. In this legendary story it has turned into a spring on top of Meru." The Dba' bzhed's transformation of Asvakarna from a mountain around Mt. Meru to a spring atop Mt. Meru is easily explained by our short passage from ITJ644. There the miraculous spring created by our unnamed hero during his time at the Asura Cave flowed from the southern face of Mt. Meru, thereby giving the spring its name, Asvakarna. Furthermore, in both passages washing in the waters of that spring grants great blessings, blessings of a religious nature for our hero and of a secular one for the Tibetan king. Once again the events described in ITJ644 appear to have been reworked towards other purposes, this time in the Dba' bzhed towards linking the master's famous successes in Nepal to the Tibetan king. Whether the protagonist in ITJ44 is Padmasambhava or not, it is at least clear that this manuscript includes a number of precursors to the later Padmasambhava legend.


The second piece of new evidence is more extensive. It appears in a short work on the female protectors in the wrathful mandala of Sri Maha Heruka. The other items on the same scroll include a discussion of the tantric vows and a relatively standard Mahayoga ritual text. Unlike ITJ644, this text refers to Padmasambhava by name (Mkhan po pad ma sam ba ba) and recounts his activities in Tibet:
 Rdo rje kun grags ma has a black-colored body. To many she appears
 pleasing, [in which case,] if their samaya is kept, she wears dharma
 robes and is adorned with various additional ornaments, displaying a
 beautiful and lovely form. Or she may be displeasing, [in which case,]
 if the samaya is kept, she is black with [her hair in] a top-knot and
 riding a mule. She is also called Rkong la de mo ('obliged to Kong
 po'). She is the leader of these seven.
 Rdo rje kun tu bzang has a white-colored body. For her seat, she
 sits upon a white all-knowing horse, and she relishes the saliva from
 vows. She is also called Sha myed gangs dkar ('fleshless white snow
 Rdo rje kun gsal ma has a pink-colored body. She usually wears black
 robes and rides a blue horse. She is also called Lha ri g.ya' ma skyol
 ('the slate bearer of Lha ri').
 Rdo rje ye shes mchog has a pink-colored body. She wears an
 assortment of clothes. She rides an emanation of a female mdzo. She is
 also called Bda' la btsan mo ('queen of the herders').
 Rdo rje sgron ma usually wears robes. For her seat, she rides an
 emanated myan. She is also called Kha rag khyung btsun ('revered
 garuda of Kha rag').
 Rdo rje 'od chags ma has a red-colored body and usually wears
 clothes. For her seat, she rides an emanation of a wild (kham yu?)
 female yak. She is also called Byang gi gser tang yi ge 'khor 'dul ma
 ('tamer of the wheel of letters, the golden tone of the north').
 Rdo rje g.yu sgron ma has a blue-colored body. As her throne she
 rides an emanated blue horse. She is also known as Lho'i ting ting
 ('chime of the south').
 These seven are also known as dakinis, the powerful women, the seven
 great mothers, or the seven great raksasis. Originally the two types
 of mandalas were arranged and arose, and then in the chapter on the
 taming of Ma tang Rudra, their vows were bestowed and their activities
 appointed. The great charnel ground was opened, and the seven
 protective guardians were addressed by the Heruka. Furthermore, they
 are an assembly which is large but seems small. Thereafter they were
 also sacred consorts to the noble one.
 After that, both the Indian master Padmasambhava and Rlang dpal gyis
 seng ge subjugated and suppressed them. Bestowing upon these ladies of
 Tibet vajras to hold, they gave them names for being in the company of
 [the buddha] Vajradhara. Since then, they have aided and supported
 those who accomplish the secret mantra in accordance with the
 scriptural systems, and they have been entrusted as the eternally
 unfailing guardians of Tibet.
 They are also sisters. These women who are greater than the great
 have promised; they are avowed. They rejoice in the saliva from vows.
 They are pleased by the remainder offerings. They strive to act
 unremittingly for meditators and for the pure substances. They wield a
 variety of terribly fearsome weapons. At other times they wield lovely
 and beautiful implements.
 Great leader Rdo rje kun grags ma, fleshless white snow mountain Rdo
 rje kun tu bzang, glorious one of Lha ri, Kun gsal ma, queen of
 herders Rdo rje ye shes mchog, revered garuda of Kha rag, Rdo rje
 sgron ma, golden tone of the north Rdo rje 'od chags ma, chime of the
 south Rdo rje g.yu sgron ma: For the welfare of sentient beings,
 perform according to the vows you took in the presence of the noble
 ones. If the previous jewels are deceived or if the vows are violated,
 then the devastations will be unbearable--one's mind will burn and one
 will shudder with horror at the hell fires. Keep me in mind! Come
 here! Perform completely the activities of pacifying, expanding,
 overwhelming, and violence. Accomplish without fault. (22)

As noted above, the Tibetan landscape is dotted with innumerable sacred sites where Padmasambhava is said to have subjugated and converted the local Tibetan pre-Buddhist spirits. It is probably no exaggeration to say that these conversion narratives represent the dominant way in which the Tibetan assimilation of Buddhism has been understood by Tibetans themselves. PT307 may be our earliest reference to such narratives.

The relationship between Padmasambhava and the seven female deities described here should be considered in light of the saptamatrka ('seven mothers'). Carvings of these seven mothers are prominently displayed on buildings throughout Northern India and the Kathmandu Valley of Nepal, and, even more significantly, systems of holy sites (pitha) and temples associated with the seven mothers are common. (23) The saptamatrka sites played an important role within eighth-century Indian tantric circles. The Guhyasamaja Tantra, for example, recommends that wrathful subjugation rites be performed "in the places of the matrka" (Skt. matrgrhe, Tib. ma mo gnas). (24) The importance of the saptamatrka was clearly recognized by the dark-period Buddhists of Tibet; references to the ma bdun appear throughout the Dunhuang manuscripts. There is little to indicate that these were references to anything other than the normative Indian set of seven. In PT307 however, we see the seven goddesses transformed and transplanted into the Tibetan landscape.

Whatever their Indian precedents, in Tibet the seven mothers came to be seen as pre-Buddhist spirits tied to specifically Tibetan sites, usually sacred mountains or lakes. Innumerable lists of such local "pre-Buddhist" spirits are found throughout Tibetan literature, and many make specific reference to the seven goddesses. The autobiographical writings of Klong chen pa (1308-1363) refer to a set of "seven sisters," though now headed by Rdo rje g.yu sgron ma (the last of our own seven), while the other sisters remain unnamed. (25) Even more significantly, Franz-Karl Ehrhard has observed the existence of a "Cult of the Seven Mothers" (Ma bdun bka' brgyud pa) continuing well into the nineteenth century. (26) It is important to note, however, that none of these later references to the seven mothers is prior to the influence of Buddhism. For this reason we may never know whether a parallel set of seven goddesses existed in Tibet before Buddhism's arrival, or if the Tibetan set appeared in response to the Indian saptamatrka.

The list of the seven goddesses in our PT307 shares names with many other groups of native Tibetan deities. (27) Any attempt to use these other lists to locate the homes of our seven sisters is frustrated by their fluidity. In a study by the eighteenth-century scholar Klong rdol bla ma, for example, the chief sister in our own text appears twice, once under her Buddhist name Rdo rje kun grags ma as the protector of the lake Gnam mtsho phyug mo, and again under her pre-Buddhist name Rkong la de mo as the protector of Bres na ri gdong in Kong. (28)

Countless lists of local spirits appear throughout Tibetan literature; Nebesky-Wojkowitz's voluminous, yet partial, survey is so overwhelming that one might wish to dismiss these lists as meaningless chaos. Yet we are left with the Tibetans' own consistent interest in these lists, and this makes them significant. Tibetans are clearly eager to make sense out of the disorder of names and places, the shifting iconographies, and the various groupings. The spirit world of Tibet is an unruly one: the Tibetan universe is filled with powerful beings who demand recognition, yet are forever evading classification. Tantric Buddhism offered early Tibetans a way to order these chaotic beings.

Thus it is not surprising to see some of our seven sisters appear again in Nyang ral's twelfth-century narrative of Padmasambhava's demon taming activities. The Zangs gling ma follows the master through a series of eight locations in central Tibet. (29) At each location he subjugates the local spirits, and finally at G.ya' ri ('Slate Mountain') he binds them all under oath as guardians of the new Buddhist religion. Nyang ral's version of events is far more elaborate than PT307's, but the fundamental theme of Padmasambhava forcing the Tibetan deities to accept vows to protect Buddhism is the same.

The theme of violent subjugation was of course not a new one in Buddhism. Of particular interest in PT307 is the explicit connection made between the popular tantric myth of the Buddha's violent subjugation of Rudra and Padmasambhava's own activities in Tibet. Here we can see that from an early date Tibetans looked to the tantric subjugation myth to understand their own conversion to Buddhism. In India, Rudra was another name for the Hindu god Siva, a fact that has caused many scholars to speculate that strong Saiva-Buddhist competition may have contributed to the myth's popularity there. PT307 reveals how the Indian narrative of the Buddhist conversion of Siva was adopted by the Tibetans and applied to their own non-Buddhist deities.

Of the many versions of the Rudra subjugation narrative, our PT307 passage almost certainly refers to that found in chapter fifteen of the Guhyagarbha Tantra. The "two types of mandalas" mentioned are probably the peaceful and the wrathful Mayajala mandalas described in this influential work. (30) And the reference to the bestowal of vajras for the sisters to hold likely refers to the closing lines of the Guhyagarbha chapter, in which all the women in Rudra's demonic host are addressed: "Then the Great Joyous Bhagavan, having bestowed the vajra into their hands, conferred the name initiation, then he arrayed them around the outer edges of the mandala." (31) Similarly, in PT307 the Tibetan goddesses are granted vajras to carry as symbols of their new roles as Buddhist guardians and given new Buddhist names bearing the prefix "Vajra." Thus Rkong la de mo is renamed Rdo rje kun grags ma ('Vajra Renown') and so on. The goddesses' position around the outside of the mandala is also implied by the line in PT307 stating that "they are pleased by the remainder offerings." (The leftover offerings are traditionally offered to the mundane protector deities around the mandala's edges.)

The connection between Padmasambhava and the tantric subjugation rites is reinforced by other early evidence. One of the few works that was attributed to the master from an early date is the Man ngag lta ba'i 'phreng ba, a commentary to the thirteenth chapter of the Guhyagarbha Tantra. (32) In addition to the Rudra myth in its fifteenth chapter, this tantra was known in Tibet for its connections to the rites of liberation (sgrol ba), or ritual killing of demonic beings. Furthermore, there is a text in the Dunhuang collections that is attributed to Padmasambhava, a commentary on another Mahayoga tantra, the Thabs kyi zhags pa padmo'i 'phreng ba, (33) and in this work the author makes direct reference to liberating beings who are "attached to wrong views." (34) Clearly, Padmasambhava's involvement in demon taming was established from an early date.

Appended to this same manuscript (ITJ321) is a short verse praising the merits of Padmasambhava. An interlinear note (mchan 'grel) attributes the verse to Santigarbha (Slobs dpon shan ti gar ba), an Indian contemporary of Padmasambhava who was also active at the court of the Tibetan king Khri srong lde'u brtsan. "Santigarbha examined this work, and having found it to be without errors, he praised Padmasambhava ..." (35) Little is known about Santigarbha. Various sources say he specialized in medicine and the rituals associated with the Buddhist deity Yamantaka. He is also said to have presided at the consecration of Bsam yas, the first Buddhist monastery in Tibet. In any case, from the perspective of the later Tibetan tradition, it is remarkable that the opinion of a relatively insignificant figure like Santigarbha would have any relevance for one with the stature of Padmasambhava.

In a similar way, our PT307 is unusual for its reference to a second player in the narrative of Tibet's exorcism and conversion; the Tibetan Rlang dpal gyis seng ge is also a relatively obscure figure. He is said to have been one of the first Tibetans to receive ordination at Bsam yas monastery in the late eighth century. The few references to him agree that he was an expert at controlling the spirits of the Tibetan landscape. Dudjom Rinpoche, for example, writes that "by propitiating Mundane Praise, Pelgi Senge was served by the eight-fold groups of spirits." (36) The deity Mundane Praise ('Jig rten mchod bstod) is usually found in the list of eight deities in the Rnying ma school's Mahayoga Sadhana Class (sgrub sde), where he is considered one of three mundane deities who were converted by Padmasambhava. Whatever the historical accuracy of such claims, for our present study it is significant that even today Dpal gyi seng ge is associated with the mundane pre-Buddhist protectors of Tibet.

The presence of such an obscure figure alongside Padmasambhava is unusual. In later traditions Padmasambhava stands in a class by himself, as the lone conqueror of Tibet's local spirits during the imperial period. PT307 suggests that Padmasambhava's role in the establishment of Buddhism in Tibet may have expanded over time, so as to eclipse others (notably a native Tibetan) acting around him. In the Tibetan imagination, Tibet's pre-Buddhist landscape required the expertise of a foreigner to tame it. The important role played by a native Tibetan was inconsistent with the later narratives and so was forgotten.


The inflation of Padmasambhava's activities has also been observed by the translators of the Dba' bzhed, who write, "This peculiar narration, consistent in several details with the scanty mention of Padmasambhava in dynastic sources (PT44), is very different from what we read in later works; it seems to indicate that this text must have preceded the great mythographical tradition." (37) The new evidence presented in this article makes it clear that in fact PT44 and the Dba' bzhed were already part of the "mythographical tradition." The Asura Cave account, the Asvakarna spring, and other evidence all indicate that by the time PT44 and the Dba' bzhed were composed in the late tenth or eleventh century, the tradition was well established, with strong roots in the dark period of Tibetan history.

Wangdu and Diemberger's conclusion leads them to speculate on the character of a historical Padmasambhava. In the Dba' bzhed Padmasambhava's visit to Tibet is far shorter than in later sources, as he is sent back to India by Tibetans opposing his activities. This prompts the following theory:
 The portrait of Padmasambhava as mainly concerned with water magic and
 sheer water technology (for example, when he suggests training rivers
 and lakes with gabions so that these can be crossed), may even hint at
 a possible attempt to import into Tibet the sophisticated irrigation
 systems used in his land of origin. In fact both in northern Pakistan
 [Padmasambhava's birth-place] and in further western regions there had
 been a long tradition of extremely advanced irrigation technology
 which allowed a very efficient use of springs and even made it
 possible to cross great expanses of desert with covered channels.
 Given the political importance of control over water resources, it is
 not surprising that the Tibetan political leadership felt more
 threatened than pleased. (38)

There is some evidence of the importance of irrigation and water rights in other regions of central Asia among the documents dating from the late eighth century. (39) One might also be tempted to use Wangdu and Diemberger's theory to explain ITJ644's account of our (proto-?) Padmasambhava creating the spring and the eight streams at the Asura Cave in Nepal. Moreover it is certainly the case that anyone working in eighth-century India or Tibet to manipulate bodies of water would inevitably have been faced with local beliefs in the naga spirits protecting those sites. (40) The tantric subjugation rites would have complemented such work, and Padmasambhava's demon-taming activities described in PT307 may well reflect this dual role. Padmasambhava's transformation of the Tibetan landscape may have been both physical and spiritual in nature.

However we should be wary of such speculations, particularly given the extensive mutability seen in all of our evidence. Whatever the case may be regarding a possibly historical Padmasambhava and his involvement in the physical landscape, it was the spiritual aspect of Padmasambhava that captured the Tibetan imagination. The theme of violent subjugation is crucial to Tibetan culture, and Padmasambhava is the demon tamer par excellence. Others have noted a tendency among Tibetans to refer to themselves as "Red-Faced Flesh-Eating Demons" in need of taming. (41) Vajrakilaya, the main deity used for pinning demons to the ground, appears throughout all kinds of Tibetan rituals. Reenactments of the Buddha's taming of Rudra, or of Padmasambhava's taming of the native demons, are performed at almost every Tibetan festival. Padmasambhava stands at the center of all these aspects of Tibetan culture. Whatever the historical realities of Padmasambhava, his legend developed along the themes of subjugation.

Ultimately, to search for an original Padmasambhava may in fact be less illuminating than to trace his continually changing manifestations through Tibetan history. Tibetans themselves hold that Padmasambhava has always acted like a mirror, perfectly reflecting whatever aspects of the Buddhist teachings are required by his faithful followers. Perhaps we too learn more by seeing Padmasambhava in this way, less as a historically locatable person than as a shifting matrix of meanings constantly calling for interpretation.


Bischoff, F. A. 1978. Padmasambhava est-il un personnage historique? In Proceedings of the Csoma de Koros Symposium, ed. Louis Ligeti. Pp. 27-33. Budapest: Akademiai Kiado.

Bischoff, F. A., and Charles Hartman. 1971. Padmasambhava's Invention of the Phur-bu: Ms. Pelliot Tibetain 44. In Etudes tibetaines dediees a la memoire de Marcelle Lalou. Pp. 11-27. Paris: Adrien Maisonneuve.

Dalton, Jacob P. Forthcoming. A Crisis of Doxography: How Tibetans Organized Tantra during the 8th-12th Centuries. Journal of the International Association of Buddhist Studies.

Davidson, Ronald M. 2002. Indian Esoteric Buddhism. New York: Columbia Univ. Press.

Dharmasri, Lochen. 1975. 'Dus pa'i mdo dbang spyi don rgyud lung man ngag gi gnad gsal byed sgron me. In Collected Works of Smin-gling Lo-chen Dharmasri, vol. 12. Dehra Dun: D. G. Khocchen Trulku.

Dowman, Keith. 1995. Power Places of Kathmandu Valley. London: Thames and Hudson.

Dudjom Rinpoche. 1991. The Nyingma School of Tibetan Buddhism, tr. Gyurme Dorje. Boston: Wisdom Publications.

Germano, David. 1994. Architecture and Absence in the Secret Tantric History of the Great Perfection. Journal of the International Association of Buddhist Studies 17: 203-335.

Germano, David, and Janet Gyatso. 2000. Longchenpa and the Dakinis. In Tantra in Practice, ed. David G. White. Pp. 239-65. Princeton: Princeton Univ. Press.

Granoff, Phyllis. 2000. Other People's Rituals: Ritual Eclecticism in Early Medieval Indian Religions. Journal of Indian Philosophy 28: 399-424.

Gronbold, Gunter. 2001. 'Saptavara'. A Dharani Collection from Nepal. In Le Parole e i Marmi, ed. Raffaele Torella. Vol. 1, pp. 369-75. Rome: Istituto italiano per l'Africa e l'Oriente.

Guhyagarbha Tantra. The mTshams-Brag Manuscript of the rNying-ma rgyud 'bum. Vol. wa, ff. 152.6-218.7. Thimphu, Bhutan: National Library.

Guhyasamaja Tantra, ed. S. Bagchi. Darbhanga: Mithila Institute.

Gunawardana, R. A. L. H. 1979. Robe and Plough: Monasticism and Economic Interest in Early Medieval Sri Lanka. Tucson: Univ. of Arizona Press.

Gyatso, Janet. 1987. Down with the Demoness: Reflections on a Feminine Ground in Tibet. Tibet Journal 12: 38-53.

Harper, Katherine Anne. 1989. Seven Hindu Goddesses of Spiritual. Transformation. Lewiston, N.Y.: Edwin Mellen Press.

Kapstein, Matthew. 2000. The Tibetan Assimilation of Buddhism. Oxford: Oxford Univ. Press.

Karmay, Samten. 1988. The Great Perfection. Leiden: E. J. Brill.

Macdonald, A. W., and Anne V. Stahl. 1979. Newar Art. Warminster: Aris & Phillips Ltd.

Mayer, Robert. 1991. Observations on the Tibetan Phur-pa and the Indian kila. In Buddhist Forum II, ed. Tadeusz Skorupski. Pp. 163-92. London: School of Oriental and African Studies.

Nebesky-Wojkowitz, Rene. 1996. Oracles and Demons of Tibet. Delhi: Book Faith India.

Panikkar, Shivaji K. 1989. Saptamatrka Worship and Sculptures. New Delhi: D. K. Printworld Ltd.

Przyluski, Jean. 1923. Les Vidyaraja: Contribution a l'histoire de la magie dans les sectes Mahayanistes. Bulletin de l'Ecole francaise d'Extreme-Orient 23: 301-18.

Rong Xinjiang. 1999-2000. The Nature of the Dunhuang Library Cave and the Reasons for Its Sealing. Cahiers d'Extreme-Asie 11: 247-75.

__________. 2004. Official Life at Dunhuang in the Tenth Century: The Case of Cao Yuanzhong. In The Silk Road: Trade, Travel, War and Faith, ed. Susan Whitfield. Pp. 57-62. Chicago: Serindia Publications.

Shaw, Julia. Forthcoming. Naga Sculptures in Sanchi's Archaeological Landscape: Buddhism, Vaisnavism, and Local Agricultural Cults in Central India, First Century B.C.E. to Fifth Century C.E. Artibus Asiae.

Skjaervo, Prods Oktor. Forthcoming. Legal Documents Concerning Ownership and Sale from Eighth-century Khotan. In From Nisa to Niya. New Discoveries and Studies in Central and Inner Asian Art and Archaeology, ed. M. Ghose and L. Russell-Smith. London: Saffron Press.

Spiro, Melford E. 1978. Burmese Supernaturalism. Philadelphia: Institute for the Study of Human Issues.

Takeuchi, Tsuguhito. 2004. Sociolinguistic Implications of the Use of Tibetan in East Turkestan from the End of Tibetan Domination through the Tangut Period (9th-12th c.). In Proceedings of "Turfan Revisited--The First Century of Research into the Arts and Culture of the Silk Road," ed. Desmond Durkin-Meisterernst et al. Berlin: Museum fur Indische Kunst.

Tsogyal, Yeshe. 1993. The Lotus Born: The Life Story of Padmasambhava, tr. Erik Pema Kunsang. Boston: Shambala Publications.

Tucci, Giuseppe. 1949. Painted Scrolls. Rome: Libreria dello Stato.

Vallee Poussin, Louis de la. 1962. Catalogue of the Tibetan Manuscripts from Tun-huang in the India Office Library. London: Oxford Univ. Press.

van Shaik, Sam. 2002. The Tibetan Dunhuang Manuscripts in China. Bulletin of the School of Oriental and African Studies 65: 129-39.

Wangdu, Pasang, and Hildegard Diemberger. 2000. dBa' bzhed: The Royal Narrative Concerning the Bringing of the Buddha's Doctrine to Tibet. Vienna: Verlag der Osterreichischen Akademie der Wissenschaften.

Wessing, Robert, and Roy E. Jordaan. 1997. Death at the Building Site: Construction Sacrifice in Southeast Asia. History of Religions 37: 101-21.

Wylie, Turrell. 1970. A Tibetan Religious Geography of Nepal. Rome: Istituto Italiano per il Medio ed Estremo Oriente.

Yamaguchi, Zuiho. 1996. The Fiction of King Dar-ma's Persecution of Buddhism. In Du Dunhuang au Japon: Etudes chinoises et bouddhiques offertes a Michel Soymie, ed. Jean-Pierre Drege. Pp. 231-58. Geneva: Droz.



@/:/rdo rje kun grags ma/ sku mdog nag mo mang dgyes sam thugs dam skongs na na bza' dar rma gsol/ rgyan gzhan la yang sna tshogs kyis brgyan pa/ gzugs mdzes shing sdug par ston / myi dgyes la thugs dam skongs na nag mo ral pa can tre'u la bcibs/ rkong la de mo zhes kyang bgyi/ 'di bdun gyi gtso mo lags//

rdo rje kun tu bzang sku mdo dkar mo cang shes dkar po'i gdan la bzhugs shing thugs dam kha chu la dgyes/ sha myed gangs dkar zhes kyang bgyi/

rdo rje kun gsal ma/ sku mdog snar mo na bza' gtsos gsol/ rta sngon po la bcibs/ lha ri g.ya' ma skyol zhes kyang bgyi//

rdo rje ye shes mchog/ sku mdog snar mo na bza' sna tshogs gsol/ /mdzo mo sprul te bcibs / brda' la btsan mo zhes bgyi/

rdo rje sgron ma/ na bza' gtsos gsol gdan/ smyan sprul te bcibs/ kha rag khyung btsun zhes bgyi/

rdo rje 'od chags ma/ sku bdog dmar mo na bza' gtsos gsol/ /gdan khri 'bri kham yu sprul te bcigs/ byang gi gser tang yi ge 'khor 'dul ma zhes bgyi/

rdo rje g.yu sgron ma/ sku mdog sngon mo gdan khri rta sngon po sprul te bcibs/ lho'i ting ting zhes bgyi'o/

@/:/ 'di bdun yang mkha' 'gro ma zhes kyang bgyi/ shug 'gro ma zhes kyang bgyi/ ma chen mo bdun zhes kyang bgyi/ srin mo

chen mo bdun zhes kyang bgyi ste/ /thog ma ni dkyil 'khor rnam gnyis bshams shing byung ba dang/ ma tang ru ta bthul

ba'i le'u dang las/ /dam stsol zhing las su bskos te/ /dur khrod chen po phye dang bdun srung ba'i srungs mar/ /he ru kas bka'

stsol to/ /de yang chen chung 'dra' ba'i tshogs so/ /de slan chad ni dpal gyi yang gzungs dam pa'o/ /de'i 'og du ni rgya gar

gyi mkhan po pad ma sam ba ba dang/ rlang dpal gyi seng ge gnyis kyis btul cing dam brnan te/ /bod khams gyi

bdag mo 'di rnams lag du rdo rje byin nas rdo rje 'chang gi gral du mying btags so/ /de nas gzhung bzhin gsang sngags

sgrub pa rnams kyi mthu dang stong grogs bgyid pa dang/ bod khams mthar myi 'jig pa'i srungs mar bcol zhing/ sring mo

yang yin/ che je chen mo rnams kyis zhal gyis bzhes shing/ dam bcas so/ /dam tsig kha chu la dga'/ mchod pa lhag

la dgyes so/ bsgom ba dang rdzas dag nan tan cher bgyi 'tshal/ /phyag mtshan yang 'jigs tshul rna tshogs pa thogs so/

(b)ar 'ga' ni phyag mtshan sdug cing mdzes pa thogs so// /bdag nyid chen mo rdo rje kun grags ma/ gangs dkar

sha myed rdo rje kun tu bzang/ dpal ldan lha ri rdo rje kun gsal ma/ brda' la btsan mo rdo rje ye shes mchog/ kha rag khyung

btsun rdo rje sgron ma zhe'o/ byang gi gser tang rdo rje 'od chags ma/ lho'i ting ting rdo rje g.yu sgron ma/ /sems can

don phyir 'phags pa'i spyan sngar khyed/ /dam bcas khas 'ches gang yin de ltar gyis/ /dkon mchog bslus

sam/ yang na dam nyams na/ rang rgyud sreg cing dmyal mye skyi g.ya' ba'i/ /'jigs chen myi bzad de soms

'dir gshegs la/ /zhi/ rgyas/ dbang dang/ mngon spyod las kyi rnams/ /rdzogs par mdzod la

skyon myed grub par gyis/ /.



1. My thanks to the International Dunhuang Project based at the British Library for making this research possible.

2. For some speculations (which remain just that) on the historicity of Padmasambhava, see Bischoff 1978.

3. See Karmay 1988: 8-10; Yamaguchi 1996; Kapstein 2000: 11-12.

4. Here I am setting aside the masses of professionally copied Mahayana sutras, many of which may have been sponsored by the Tibetan king Ral pa can during the first half of the ninth century (see van Schaik 2002: 136-37). Some preliminary work on post-Tibetan-occupation Dunhuang manuscripts has been done by Tsuguhito Takeuchi (2004). In my opinion, Takeuchi's conclusions can be taken even further: the vast majority of the Tibetan tantric materials (and possibly much of the sutric materials too) date from the tenth century. I am currently completing a book on these issues, in collaboration with Sam van Schaik.

5. By distinguishing the tantric elements from the other forces at work in the conversion of Tibet, I do not mean to underestimate the influences of the sutras. However, the tantras in particular appear to have captured the Tibetan imagination, and many of the more creative Tibetan adaptations of Buddhism seem to have involved tantric themes, from new developments in Rdzogs chen to new mythic narratives. In any case, the legends relating to Padmasambhava grew out of the tantric sphere, and these are our present focus.

6. Scholars have used a confusing variety of acronyms for referring to the Stein collection held at the British Library and the Pelliot collection held at the Bibliotheque nationale. I have opted for ITJ, an abbreviation of "IOL[India Office Library] Tib[etan] J," since this is how the Tibetan Dunhuang materials are identified at the British Library. Similarly, PT will be used to refer to the "Pelliot tibetain" materials in Paris.

7. de la Vallee Poussin 1962: 205.

8. A discussion of the nine vehicles system offered in this manuscript can be found in my forthcoming article on the tantric doxographical systems of early Tibetan Buddhism.

9. On vidyadharas, see Przyluski 1923; Granoff 2000.

10. ITJ644, 2a.6-2b.1. de nas phyag na rdo rje gshegs nas/ dngos grub sbyin ba dang/ a su ra'i brag phug du phyin pa dang/ de na phyag na rdo rje'i sprul pa gcig bzhugs pa'i zhal mthong nas brag la rkang pa gcig brgyab pa dang/zan la brgyab bzhin snang ngo/ rjes de nas dam babs nas/ nang de na chu myig yan lag brgyad dang ldan ba brgyad yod pa la/ gcig ni ri rab kyi lho ngos su rdol te chu myig rta rna zhes bya'o/ bdun a su ra'i nang na 'bab pa la khrus byed cing bsgrub pa de/ sa la gnas pa'i rigs 'dzin ces bya'o.

11. We should be cautious, however, as multiple Asura caves are mentioned in Tibetan tantric literature. See, for example, Dharmasri 1975: 45.3 and 51.4, where other Asura caves are discussed. Of particular interest are the mentions of two Chinese translations of early Kriya Vajrakumara texts (see Mayer 1991: 186; my thanks to Mayer for bringing these to my attention).

12. The entire item was introduced, transliterated, and translated in Bischoff and Hartman 1971: 11-27. The section on Padmasambhava's activities has been retranslated more recently in Kapstein 2000: 158-59.

13. PT44, 1a.3-9a.5. As translated in Kapstein 2000: 158.

14. On the "gradual articulation of a self-conscious Great Perfection movement in Tibet" during the dark period, see Germano 1994: 219.

15. Most scholars now agree that the cave containing the Dunhuang manuscripts was sealed in the early eleventh century, c. 1006. On the closing of the cave, see Rong Xinjiang 1999-2000: 247-75.

16. Takeuchi 2004: 3.

17. I remain hesitant about this argument, however, given several caveats that should be added here: First, as noted in the introduction to the present article, it is increasingly clear that the overwhelming majority of the Tibetan tantric materials from Dunhuang date from the tenth century, and perhaps even from the second half of the tenth century. This means that PT44's late date might not be so unusual, and that our ITJ644 manuscript may well date from the same period. However, even if ITJ644 represents a contemporary alternative to PT44's Yang le shod narrative, it remains significant that such alternatives existed, and the points on which the two accounts differ likely still reflect the kinds of Tibetan concerns that motivated the changing narrative. Second, in the interlinear notes to the Dunhuang version of the Thabs kyi zhags pa pad mo'i 'phreng ba commentary (ITJ321), the commentary is attributed to Padmasambhava. If we accept this attribution (and the existence of a historical Padmasambhava), we must also accept that the master was at least aware of the early Vajrakilaya ritual system, for in chapter twenty of that work the famous Phur pa bcu gnyis is cited (ITJ321, 64v.6, or P.4717, 119b.8: Ki la ya bcu gnyis kyi ta ntra), as well as its "supplementary tantra" (phur pa bcu gnyis kyi rgyiud phyi ma; ITJ321, 70b.4, or P.4717, 122b.4). The tantra is quoted in a section on violent rites, and it is only one among a series of titles cited, but it implies Padmasambhava had at least read this seminal Vajrakilaya tantra. Finally, regarding the difference between Vajrapani and Vajrakilaya, we should consider the possibility that these were considered two forms of the same deity. If so, the idea that ITJ644 represents an earlier narrative, in this regard at least, would be less significant.

18. Note that this is different from the later normative formulations of the four vidyadhara levels of Mahayoga, in which the vidyadhara of mahamudra is only the second highest level, placed below that of spontaneous accomplishment (lhun grub).

19. Dowman 1995: 96. Here Dowman is following an explanation by the fourth Khams sprul, Chos kyi nyi ma (1730-79/80), of the view of the local Newar Vajracaryas. Khams sprul's account should not be immediately accepted, however, for it is possible that the Newars were themselves influenced by earlier Tibetan pilgrims. Today, unless they have had long contact with Tibetan teachers, Newars have little to say about Padmasambhava (my thanks to Hubert Decleer for his clarification of these points). For two other late Tibetan references to the master's demon-taming activities at Yang le shod, see Wylie 1970: 31.

20. On dating, see Wangdu and Diemberger 2000: xiv and 8.

21. Wangdu and Diemberger 2000: 56 (my italics).

22. PT307, lines 10-32. For a complete transcription, see the Appendix.

23. On the saptamatrka, see Macdonald and Stahl 1979: 83-105; Harper 1989; Panikkar 1989; Gronbold 2001: vol. 1, 369-75; Davidson 2002: 300-303.

24. Guhyasamaja Tantra, XIV, 55. The interlinear notes to this verse in the Dunhuang edition (ITJ438, 53v.5) confirm that the ma mo were understood by early Tibetans to be the same as the seven mothers (ma bdun).

25. For a translation of the relevant passage, see Germano and Gyatso 2000: 258. The story is also told in Dudjom Rinpoche 1991: vol. 1, 581-85.

26. Ehrhard's explorations have been based on a twenty-two-folio text dealing with the cult by Brag dkar rta so sprul sku (b. 1775), titled Dpal ldan gur rigs mdo chen brgyud pa'i lo rgyus nyung ngu'i ngag gi brjod pa padma ra ga'i phreng ba. According to Gene Smith (oral communication), the cult was a sub-school of the Stod 'brug tradition and seems to have been popular in the Skyid grong region on the Tibet-Nepal border. The lineages of the cult were affiliated with the Zur family, and several students of Tshe dbang nor bu wrote on the ma bdun rituals. Ehrhard's findings will certainly add much to our understanding of the development of the seven mothers in Tibet.

27. The list of the twelve Brtan ma (or Bstan ma) goddesses in particular can shed some light on the homes of our seven sisters. See Nebesky-Wojkowitz 1996: 181-88.

28. See Tucci 1949: 728.

29. For an English translation, see chapter nine of Tsogyal 1993: 62-64.

30. Chapter fifteen of the Guhyagarbha Tantra opens with the wrathful mandala emanating out of the peaceful mandala that is discussed earlier in the tantra: de nas de bzhin gshegs pa thams cad kyi bdag po sangs rgyas thams cad kyi ngo bo nyid kyi khro bo'i dkyil 'khor mngon par 'du mdzod de (Guhyagarbha Tantra, 195.6).

31. Guhyagarbha Tantra, 206.1-2. de nas bcom ldan 'das dgyes pa chen pos/ lag tu rdo rje byin nas ming gi dbag bskur te dkyil 'khor gyi phyi rim du bkod do.

32. For a discussion and translation of this text, see Karmay 1988: 137-74. The early date of its attribution is based primarily on (1) its being cited in the Bsam gtan mig sgron, a work composed by Gnubs chen sangs rgyas ye shes in the early tenth century, and on (2) the existence of a sub-commentary to the Man ngag lta ba'i 'phreng ba written in the eleventh century by Rong zom chos kyis bzang po (c. 1012-88).

33. For the tantra, see Peking 458. The commentary is found at Peking 4717 but is missing from the Rnying ma collections. The attribution of the work to Padmasambhava was forgotten by the later Tibetan tradition.

34. ITJ321, 2a.2.

35. ITJ321, 84a.5. slobs dpon shan ti gar bas brtags nas ma nor nas /sam ba bha la stod pa'o.

36. Dudjom Rinpoche 1991: 535.

37. Wangdu and Diemberger 2000: 13-14.

38. Wangdu and Diemberger 2000: 14.

39. On the late eighth-century document (British Library, Or. 9268A) concerning the strictly controlled sale of water rights around Khotan, see Skjaervo forthcoming. And according to Rong Xinjiang, irrigation was important around Dunhuang (see Rong Xinjiang 2004: 60).

40. On the links between Buddhism, nagas, and water reservoirs, see especially Shaw forthcoming; also Gunawardana 1979: 215-16. On the use of violent ritual in the construction of irrigation systems, see Spiro 1978: 104-7; Wessing and Jordaan 1997: 117.

41. See Gyatso 1987.
COPYRIGHT 2004 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2004 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dalton, Jacob
Publication:The Journal of the American Oriental Society
Geographic Code:9CHIN
Date:Oct 1, 2004
Previous Article:On a five-colored cloud: the Songs of Mount Wutai.
Next Article:An Iranian name in duplicate.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters