Tibetan and Himalayan Library - THL

THL Title Text
On the Very Idea of a Tantric Canon
David B. Gray, Santa Clara University
JIATS, no. 5 (December 2009), THL #T5690, 37 pp
Section 4 of 4 (pp. 17-22)

Mythic Treasuries and the Tibetan KangyurBka’ ’gyur

At the conclusion of an essay on the political implications of Vedic conceptions of canonicity, David Carpenter argued that “[w]hile canons everywhere may function as instruments for the creation of identity and the establishment of authority, there remain real differences in the ways in which these things are accomplished in each case, and even in the degree to which they are accomplished.”53 I believe that attempts at creating and closing canons in Tibet were affected by two countervailing forces, but to different degrees, yielding different results.54 These are the conservative equation of textual authority with Indic origination, and the sometimes compatible, sometimes conflicting notion that the continuing revelation of lost texts, preserved by the Buddhas and ḍākinīs, is possible. These, in turn, may overlap with other forces at work in the Tibetan traditions, such as scholarship (khepanyimkhas pa nyid) and spiritual accomplishment (druppagrub pa), which appear to form a creative tension, sometimes converging and sometimes diverging.55

I believe that the recurrent accounts in Buddhist literature of large ur-texts of one hundred thousand stanzas, collections of tantras of this length, and even massive collections of one hundred thousand tantras, which in turn are only incomplete expressions of the celestial collections preached by the Buddhas and preserved by the ḍākinīs, had a significant impact on the development of the Tibetan canons of tantras. While I would never suggest that these myths represent a primary cause of these collections, they were a factor that shaped their development, and also problematized attempts at their closure. With respect to the tantra collection in the KangyurBka’ ’gyur, it seems that generally speaking the conservative view prevailed, resulting in the de facto closure of this canonical collection. Nonetheless, the collection was, to a limited but noticeable extent, marked by these myths.

The history of the formation of Tibetan canonical collections, such as the KangyurBka’ ’gyur, is a complex subject beyond the scope of this essay. Research thus far has demonstrated their gradual development over several centuries from relatively unsystematized textual collections.56 While the myths discussed here probably had [page 18] little direct impact on this process, it did shape it in numerous, subtle ways. We can see this influence in the language used to describe the collection, and the ways in which texts in the collection were labeled and organized.

We can see this in a pivotal work by Butön Rinchen DrupBu ston rin chen grub, one of the major figures in the early attempts to organize the canonical collections, and limit them to works that were believed to be of genuine Indic provenance. ButönBu ston was famous (or infamous) for his composition of a controversial catalogue of tantras. This work was notorious primarily for the tantras that he chose to leave out, namely, many of the tantras of the NyingmaRnying ma school, including dynastic translations that ButönBu ston considered to be apocryphal.57 ButönBu ston entitled this work “Catalog of the One Hundred Thousand Tantras” (Gyübumgyi KarchakRgyud ’bum gyi dkar chag), despite the fact that the catalogue actually listed four hundred and twenty-two works.58 It seems likely that ButönBu ston chose this title under the influence of the myths that have been examined here; in other words, in composing a catalogue of tantras he had in mind not only the relatively modest collection of tantras to which he had access, and which he considered to be valid. But he also had in mind the far larger collections imagined and described by the Indian and Tibetan Buddhist traditions.59

This title was assumed by several editions of the KangyurBka’ ’gyur. In the DegéSde dge canon, for example, the collection of tantras is likewise labeled One Hundred Thousand Tantras (GyübumRgyud ’bum).60 Needless to say, this collection has only slightly more texts than ButönBu ston’s catalogue, nor does it contain any texts that approach [page 19] that number of stanzas. In fact, if you subtract one hundred thousand from the number of tantras supposedly studied by Atiśa, 100,455, you get a number that is very close to the number of tantras actually preserved in this edition of the KangyurBka’ ’gyur.61

The myths of the legendary root tantras and treasuries of tantras appear to have affected ButönBu ston in other ways as well. For example, ButönBu ston provides the following rather unusual title for the Cakrasamvara Tantra: “The Tantra Abbreviated from the Great One Hundred Thousand [Text]: The Appendix of the Root Tantra’s Appendix” (Bumpa Chenpolé Düpé Gyü/ Tsawé Gyükyi Chimé Chima’Bum pa chen po las bsdus pa’i rgyud/ rtsa ba’i rgyud kyi phyi ma’i phyi ma).62 This is not the title provided in the standard Prajñākīrti-MardoMar do revised translation of this text, to which ButönBu ston refers, or any other translation for that matter.63 The extant Sanskrit text and the Tibetan translations give the title “Discourse of Śrī Heruka” (Śrīherukābhidhāna; Pel Heruké Ngepar JöpaDpal he ru ka’i nges par brjod pa) at the end of each chapter, and the title “Great Yoginī King of Tantras Called the Cakrasamvara” (Śrīcakrasamvaraṃ Nāma Mahāyoginītantrarāja) in the text’s colophon.64 The source of this ButönBu ston’s “title” is actually the following descriptive passage in the colophon: “It is the king of all teachings, the appendix of the appendix (uttarottaraṃ), included within the Discourse of Śrī Heruka, the one hundred thousand [stanza] great king of tantras.”65

It is certainly not the case that ButönBu ston made a mistake here, given the fact that he composed a massive commentary on this text, one that discusses the text’s title at length.66 While I am not certain how and why ButönBu ston came up with this title, [page 20] it does seem certain that in making this choice he was influenced by the myth of the tantra’s origin, which so strongly asserts its derivation from a much larger root tantra.

There are other small signs of the impact of this myth on the organization and designation of the collection of tantras and texts within it. ButönBu ston likewise referred to the Wheel of Time Tantra in his catalogue as the “Abbreviated Kālacakra Tantra” (Dükyi Khorlo Düpé GyüDus kyi ’khor lo bsdus pa’i rgyud), which is itself an abbreviation of its full name, and one that highlights the idea that it derives from a much longer text.67 The title given to the Cakrasamvara Tantra in the KangyurBka’ ’gyur is likewise “The King of Tantras, Śrī Samvara Light” (Tantrarāja-​śrīlaghusamvara-​nāma; Gyügyi Gyelpo Pel Demchok Nyungngu ZhejawaRgyud gyi rgyal po dpal bde mchog nyung ngu zhes bya ba).68 Here the term “light” (laghu; nyungngunyung ngu) points to the myth of the tantra’s origin. While the name Laghusamvara does not occur in the text itself, this became a popular shorthand name for it, and by selecting it as the text’s title the translators placed emphasis on the idea that it was derived from a much larger scripture.

This is the case with respect to other tantras as well. For example, in the possibly older One Hundred Thousand Ancient Tantra (Nyingma GyübumRnying ma rgyud ’bum) collection,69 the early dynastic translation of the Sarvabuddhasamāyoga-​ḍākinījālasamvara Tantra, one of the eighteen Mahāyoga/Magical Net tantras, is given this simple title.70 However, in the eleventh-century New Tradition translation it becomes an appendix to an absent mythical text, the Primal Buddha Tantra through its new title, the Śrī-​sarvabuddhasamāyoga-​ḍākinījālasamvara-​nāma-​uttaratantra.71

[page 21]

The collection of tantras that came to be the KangyurBka’ ’gyur is thus replete with references to absent root tantras, with various texts that refer to a greater canon that exists not in this world, but was imagined by Buddhists as existing in the more glorious past, or in more glorious realms of reality. The myths of larger tantric canons bolstered the truly conservative efforts of scholars such as ButönBu ston to preserve what they surely believed were fragments of a much larger, but largely lost (in this world and time period) tantric canon.72 This conservative vision also served to displace authority to regions difficult to access (India, Khecarīpada, and so forth). The project of creating the Tibetan canons, supported as they were by this ideology, was, on the one hand, a project that bolstered the authority of those figures – translators, lamabla mas, and so forth – who mediated access to this treasury of knowledge.73

But this ideology was double-edged, and threatened to cut the hands of those who sought to wield it. The efforts of scholars such as ButönBu ston to regulate the canon, to exclude texts that he felt were not genuine, was undermined by narratives such as the story of the breaking of Atiśa’s pride. This story could be read as a very clever assault on the scholastic arrogance that might lead a learned person, such as ButönBu ston, to attempt to close the tantric canon. If even Atiśa, with his vast knowledge of tantra, knew only a fragment of those texts preserved by the ḍākinīs, how could a lesser scholar even imagine that he has anything but a pitifully limited understanding?

While there were, of course, later attempts to catalogue and close and the tantric canon, the widespread belief in the myth that underlies this account may have undermined these attempts, and prevented the absolute closure of the canon. For these attempts were equivocal, and arguably never completely successful, insofar as these canons have never been formally closed. Even the adherents to the new tantric traditions, the sarmapagsar ma pas, by virtue of accepting the myths of the revelation of the tantras from their supramundane treasuries, could not absolutely reject the possibility of further revelations, despite their conservative privileging of all things Indic. As Ron Davidson has shown, there were several apocryphal texts included into the New Tradition canons, including a number of works that claimed to be withdrawn from the “ḍākinīs’ secret treasury.”74

[page 22]

It thus seems that the idea of a “Canon of Tantras,” a Tantrakośa, became for esoteric Buddhist traditions an empty signifier, pointing toward an absent corpus that served as a displaced locus of authority. The actual collections of tantras preserved by these traditions, such as the GyümbumRgyud ’bum or Tantra Collection preserved in the KangyurBka’ ’gyur, would then be the relics of this absent body.75 Venerated remnants, they are carefully wrapped and enshrined, but rarely actually read. While the practice traditions connected with some of them thrived, many of these vital traditions maintain only a tenuous connection to the root texts preserved in the KangyurBka’ ’gyur. They are significant primarily as ciphers of authority, vital ideological links to the Buddhas, ḍākinīs, and siddhas who are thought to have manifested them. Their gnosis is thus enveloped as the secret meaning of the obscure texts that they revealed to the world. The idea that the authority figures of tantric traditions mediate their controlled revelation, via the ritual and meditative practices that they moderate and conduct, is one of main bases of the authority of these traditions.


[53] David Carpenter, “The Mastery of Speech: Canonicity and Control in the Vedas,” in Authority, Anxiety, and Canon, ed. Laurie Patton (Albany, NY: State University of New York Press, 1994), 31-32.
[54] For an excellent collection of essays addressing the various Tibetan canonical collections, see Helmut Eimer and David Germano, eds., The Many Canons of Tibetan Buddhism (Leiden: Brill, 2002).
[55] That is, I would suggest that these myths on the one hand undoubtedly bolstered the conservative and scholastic impulse to preserve the extant canon, to prevent its further deterioration on earth. Yet, at the same time, belief in a larger canon preserved in other realms likewise kept open (in theory, if not in practice), the possibility of further revelation. Tibetan Buddhist traditions appear to be united with respect to the need for canons, but differ in their openness to their further expansion.
[56] Regarding research on the history of the KangyurBka’ ’gyur, see especially Helmut Eimer, “Some Results of Recent Kanjur Research,” in Archiv für zentralasiatische Geschichtsforschung, ed. Dieter Schuh and Michael Weiers, Heft 1 (Sankt Augustin: VGH Wissenschaftsverlag, 1983), 5-25; and “A Note on the History of the Tibetan Kanjur,” Central Asiatic Journal 32, nos. 1-2 (1988): 64-72. See also Paul Harrison, “Meritorious Activity or Waste of Time? Some Remarks on the Editing of Texts in the Tibetan Kanjur,” in Tibetan Studies: Proceedings of the 5th Seminar of the International Association for Tibetan Studies, Narita 1989, ed. Shōren Ihara and Zuihō Yamaguchi (Narita: Naritasan Shinshoji, 1992), 1:77-93; and also “In Search of the Source of the Tibetan bKa’ ’gyur: A Reconnaissance Report,” in Tibetan Studies: Proceedings of the 6th Seminar of the International Association for Tibetan Studies, Fagernes 1992, ed. Per Kvaerne (Oslo: The Institute for Comparative Research in Human Culture, 1994), 295-317.
[57] These judgments, based as they were on the criterion of whether the text was authentically Indic, often on the basis of whether the text was known to contemporary Indian scholars, were naturally problematic and contentious. For an interesting discussion of this issue see Ronald Davidson, Tibetan Renaissance: Tantric Buddhism in the Rebirth of Tibetan Culture (New York: Columbia University Press, 2005), 251-54.
[58] See Helmut Eimer, Der Tantra-Katalog des Bu ston im Vergleich mit der Abteilung Tantra des tibetischen Kanjur: Studie, Textausgabe, Konkordanzen und Indices (Bonn: Indica et Tibetica Verlag, 1989).
[59] I recognize that the Tibetan term bum’bum, much like the Sanskrit term lakṣa, can designate not only the number one hundred thousand, but also a vast quantity in general. It can function much like the English word “myriad,” or the Chinese term wan (), both of which can mean ten thousand or “a lot.” However, given the ubiquity of the number one hundred thousand in Indian and Tibetan discourse about the tantras and their canons, I do not think that the general sense is the sole denotation of the term here, although the connotation of “a lot” is clearly implied.
[60] Various editions of the KangyurBka’ ’gyur designate the tantra collection in various ways. Among the editions that I have consulted, the DegéSde dge and Urga editions term it the GyübumRgyud ’bum. The PudrakPhug brag edition terms it, less colorfully, the “Tantra Collection” (GyüdéRgyud sde); see Jampa Samten, Phug brag bka’ ’gyur bris ma’i dkar chag, A Catalogue of the Phug-Brag Manuscript Kanjur (Dharamsala: Library of Tibetan Works and Archives, 1992), ix. The NartangSnar thang, LhasaLha sa, Peking, and TokStog Palace editions simply designate it as GyüRgyud (tantra). The designation GyübumRgyud ’bum thus appears to be a relatively late designation, appearing with the eighteenth-century DegéSde dge print of the canon. But the idea of a GyübumRgyud ’bum is clearly much older, going back to ButönBu ston at least, and traceable ultimately to the persistent association of this number with the tantras.
[61] The GyübumRgyud ’bum section of the DegéSde dge canon contains 468 texts in twenty volumes. If we also count the Ancient Tantra (NyinggyüRnying rgyud) collection that immediately follows it in a separate section, then the number rises to 484 texts in twenty-three volumes. See Ui Hakuju, et al., A Complete Catalogue of Tibetan Buddhist Canons (Bkaḥ-ḥgyur and Bstan- ḥgyur) (Sendai: Tōhoku Imperial University, 1934), 67-141. The exact number of tantras varies somewhat among the collections, with certain early collections, such as the PudrakPhug brag and TokStog Palace manuscript KangyurBka’ ’gyurs containing additional texts not found in later collections. Regarding this see Samten, Phug brag bka’ ’gyur bris ma’i dkar chag; and Tadeusz Skorupski, A Catalogue of the Stog Palace Kanjur, Bibliographia Philologica Buddhica Series Major 4 (Tokyo: International Institute for Buddhist Studies, 1985).
[62] See Eimer, Der Tantra-Katalog des Bu ston, 61.
[63] ButönBu ston lists the revisers as Lotsawa Drakjor SherapLo tstsha ba grags ’byor shes rab and Marpa Chökyi WangchukMar pa chos kyi dbang phyug (1043-1138) (Eimer, Der Tantra-Katalog des Bu ston, 61). The latter figure was commonly known as Marpa DopaMar pa do pa, or MardoMar do. The name Drakjor SherapGrags ’byor shes rab (ca. 1100) is, presumably, a translation of Prajñākīrti, whose name is transliterated rather than translated in the standard canonical translation. See Toh. 368, DegéSde dgeRgyud ’bum vol. ka, 246b.
[64] The Sanskrit here reads Śrīcakrasamvaraṃ Nāma Mahāyoginītantrarāja, and the Tibetan translations differ here. The Sumarikīrti-​mal revised translation, which is preserved in the PudrakPhug brag ms. KangyurBka’ ’gyur, matches this reading: dpal ’khor lo sdom pa zhes bya ba’i rnal ’byor chen po’i rgyud kyi rgyal po (P GyüdéRgyud sde vol. nga, 144b), while the Prajñākīrti-MardoMar do translation reads dpal he ru ka’i nges par brjod pa zhes bya ba rnal ’byor ma chen mo’i rgyud kyi rgyal po, attesting Discourse of Śrī Heruka rather than Cakrasamvara (D Rgyud ’bum vol. ka, 246b). For more information on these texts and their variant readings, see my forthcoming The Cakrasamvara Tantra: Editions of the Sanskrit and Tibetan Texts (New York: American Institute of Buddhist Studies, 2009).
[65] Gray, The Cakrasamvara Tantra, 382-83.
[66] See Bu ston, Bde mchog rtsa rgyud kyi rnam bshad gsang ba’i de kho na nyid gsal bar byed pa [The Elucidation of the Secret Reality, A Detailed Exegesis of the Samvara Root Tantra], in The Collected Works of Bu ston, ed. Lokesh Chandra (New Delhi: International Academy of Indian Culture, 1966), cha (6):141-718.
[67] The longer title, preserved in the canonical translations, is “The Śrī Kālacakra King of Tantras, Drawn from the Paramādibuddha” (Paramādibuddhoddhṛta-​śrīkālacakra-​nāma-​tantrarāja, Chokgi Dangpö Sanggyelé Chungwa Gyükyi Gyelpo Pel Dükyi Khorlo ZhejawaMchog gi dang po’i sangs rgyas las phyung ba rgyud kyi rgyal po dpal dus kyi ’khor lo zhes bya ba) which likewise highlights the text’s origin myth. See Eimer, Der Tantra-Katalog des Bu ston, 60.
[68] See DegéSde dgeRgyud ’bum vol. ka, 213a.
[69] The history of the Nyingma GyübumRnying ma rgyud ’bum collection is a complex issue outside of the scope of this essay. While the final organization of this collection occurred during the latter dissemination (chidarphyi dar) period when the KangyurBka’ ’gyur was being organized (indeed, its compilation may very well have been inspired by the exclusion of its texts from the KangyurBka’ ’gyur), it clearly does contain materials that are genuine early translations from the former dissemination (ngadarsnga dar) period, that is, the eighth and ninth centuries. Regarding this see Dorji Wangchuk, “An Eleventh-Century Defense of the Authenticity of the Guhyagarbha Tantra,” in The Many Canons of Tibetan Buddhism, ed. Helmut Eimer and David Germano, 265-91 (Leiden: Brill, 2002); as well as Cathy Cantwell and Robert Mayer, The Kīlaya Nirvāṇa Tantra and the Vajra Wrath Tantra: Two Texts from the Ancient Tantra Collection (Wien: Verlag der Österreichischen Akademie der Wissenschaften, 2007), 1-4.
[70] I refer to the title of the translation preserved in the Nyingma GyübumRnying ma rgyud ’bum, Sarvabuddhasamayoga Tantrarāja-​nāma (Sanggyé Tamché Nyampar Jorwa Zhejawa Gyügyi GyelpoSangs rgyas thams cad mnyam par sbyor ba zhes bya ba rgyud gyi rgyal po). It is no. 207 in Kaneko’s catalogue (see Kaneko Eiichi, Ko-Tantora zenshū kaidai mokuroku [Tokyo: Kukusho Kankōkai, 1982], 254-255). It occurs in the TsamdrakMtshams brag manuscript of the Nyingma GyübumRnying ma rgyud ’bum at vol. tsha, 1b-26a, http://www.thlib.org/encyclopedias/literary/canons/ngb/ngbcat.php#cat=tb/0402.
[71] The latter dissemination translation (Toh. 366) is attributed to Lha RinpochéLha rin po che in the canonical text, although it is attributed to Smṛtijñānakīrti by ButönBu ston (Eimer, Der Tantra-Katalog des Bu ston, 61). It probably dates to the eleventh century.
[72] Many accounts of the origin of the tantras emphasize this notion of loss through time, much like Yi Jing’s account of the Vidyādhara Collection discussed in section two above. For example, Tāranātha (1575-1634) reported that there was a tradition that the Tārā Tantra underwent a gradual diminishment over the course of the four yugas, starting out as a ten-million-stanza text in the kṛtayuga, condensed into a six-hundred-thousand-stanza text during the tretāyuga, further condensed to twelve thousand stanzas during the dvāparayuga, and finally resulting in the extant thousand-stanza verse text during the kaliyuga. See David Templeman, trans., The Origin of the Tārā Tantra, Jo Nang Tāranātha (Revised edition, Dharamsala: Library of Tibetan Works and Archives, 1995), 3.
[73] On this see Davidson, Tibetan Renaissance, chap. 4.
[74] See Davidson, Tibetan Renaissance, 150-51; and also his essay “Gsar ma Apocrypha: The Creation of Orthodoxy, Gray Texts, and the New Revelation,” in The Many Canons of Tibetan Buddhism, ed. Helmut Eimer and David Germano (Leiden: Brill, 2002), 203-24.
[75] For an extended comparison of the tantras to relics, see my essay “Disclosing the Empty Secret: Textuality and Embodiment in the Cakrasamvara Tantra,” Numen 52, no. 4 (2005): 417-44.

Note Citation for Page

David B. Gray, “On the Very Idea of a Tantric Canon: Myth, Politics, and the Formation of the Bka’ ’gyur,” Journal of the International Association of Tibetan Studies, no. 5 (December 2009): , http://www.thlib.org?tid=T5690 (accessed ).

Note Citation for Whole Article

David B. Gray, “On the Very Idea of a Tantric Canon: Myth, Politics, and the Formation of the Bka’ ’gyur,” Journal of the International Association of Tibetan Studies, no. 5 (December 2009): 1-37, http://www.thlib.org?tid=T5690 (accessed ).

Bibliography Citation

Gray, David B. “On the Very Idea of a Tantric Canon: Myth, Politics, and the Formation of the Bka’ ’gyur.” Journal of the International Association of Tibetan Studies, no. 5 (December 2009): 1-37. http://www.thlib.org?tid=T5690 (accessed ).