Articles by alphabetic order
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
 Ā Ī Ñ Ś Ū Ö Ō
1 2 3 4 5 6 7 8 9 0


Rethinking Treasure (part one)

From Tibetan Buddhist Encyclopedia
Jump to navigation Jump to search






Robert Mayer

Dedicated with gratitude to Matthew Kapstein, exemplary scholar and perennial inspiration.

Introduction



In October 2017, Cathy Cantwell and I, with the very welcome contributions of Lopon P. Ogyan Tanzin, and the assistance of Dr Dylan Esler, began a project on Myang ral Nyi ma’i ‘od zer and early gter ma in Tibet, funded by the German DFG, and based, through the kindness of Carmen Meinert and Volkhard Krech, at the Ruhr University in Bochum (RUB). I am currently contributing to this project part-time, mainly from the UK. Much of the research I did when working on it full time, from October 2017 through June 2018, still remains unpublished, although several of the insights developed in that period are also reflected in this paper.


Earlier, in 2016, I gave a lecture that explored related themes to the Buddhist Studies Graduate Students Seminar in Oxford, and from the resultant discussion, Anna Sehnalova, Yegor Grebnev, and myself, began an ongoing interdisciplinary seminar series at Merton College, Oxford, on treasure discovery across different cultures. From Hilary Term 2019, the seminar continued at Wolfson College.1

We are exploring the rich and varied Treasure cultures mainly of India, China, Tibet, and the Islamic world, considering them individually and comparatively, as well as contemplating their often complex mutual entanglements. Participants and speakers have included Barend te Haar, Yegor Grebnev, Cathy Cantwell, Rob Mayer, Ronit Yoeli- Tlalim, Ulrike Roesler, Anna Sehnalova, Charles Stewart, Catherine Hardie, Ronit Yoeli-Tlalim, Piers Kelly, Reinier Langelaar, and others. Thus the basic outline of this paper is the outcome of work I did in Oxford between 2016 and September 2017, much enriched by subsequent work for RUB, so that earlier versions of it have already been


1 We are grateful to the Tibetan & Himalayan Studies Centre at Wolfson for supplying us with facilities, and a small grant. Scholars working on Treasure traditions in any culture, and interested in participating, are welcome to contact us.

Robert Mayer, “Rethinking Treasure (part one)”, Revue d’Etudes Tibétaines, no. 52, Octobre 2019, pp. 119-184.


presented as lectures in the course of 2017 and 2018, at the universities of Oxford, Harvard, and Vienna.2 My preliminary task has been to re-examine and reassess the sources used by previous scholarship for understanding the historical origins of gter ma in Tibet, and, if necessary, to seek out further sources too.


In reviewing the existing literature, what I have found most strik- ing is that this topic has so far been approached quite often through what we might call internal evidence, that is, through the testimonies of the gter ma literature itself, and the closely related non-gter ma writings of famous gter stons. Most widely relied upon are the bKa’ chems ka khol ma, the Maṇi bka' ‘bum, and Myang ral’s works, all from the 12th century; Guru Chos dbang’s works from the 13th century; O rgyan gling pa’s gter mas from the 14th century; and [[‘Jigs med gling pa’s gTam gyi tshogs thegs pa’i rgya mtsho] from the 18th century. A further Tibetan work by a more recent gter ma apologist, the gTer gyi rnam bshad of the third rDo grub chen incarnation, 'Jigs med bstan pa'i nyi ma (1865-1926), has also been influential upon some Western scholars, not least through Tulku Thondup’s 1986 translation of it.


I have attempted to reexamine some of these sources, to see if they are fit for purpose: Can we best understand the origins of the gter ma tradition mainly from its own internal evidence? Or might we get a more rounded view by seeking external sources too, that is, sources not directly authored or redacted by active participants in the gter ma traditions?


In that spirit, in the first part of this paper, I will reexamine the single most influential example of a gter ma text widely believed to offer us crucial insights into the early formation of the gter ma traditions, O rgyan gling pa’s rGyal po bka'i thang yig, and raise a number of questions about its suitability for such an exercise.

After that, in the second part, I will outline twelve alternative historical approaches that we hope might eventually yield broader, sounder, and deeper understandings than are available from a narrower reliance on texts produced by gter stons themselves and their

2 Many people helped me in various stages of preparing this paper. I would like to thank in particular Daniel Berounsky, Henk Blezer, Cathy Cantwell, Ronald Da- vidson, Brandon Dotson, Dylan Esler, Rolf Giebel, Yegor Grebnev, Janet Gyatso, Barend te Haar, Cat Hardie, Guntram Hazod, Dan Hirshberg, Matthew Kapstein, Samten Karmay, Piers Kelly, Yury Khokhlov, Per Kvaerne, Reinier Langelaar, Jue Liang, Stefan Mang, Dan Martin, Wiesiek Mical, Charles Ramble, Anna Sehnalova, Sam van Schaik, Henrik Sørensen, Per Sørensen, Charles Stewart, Jeffrey Sundberg, Peter Szanto, Lopon P. Ogyan Tanzin, Ben Williams, Ronit Yoeli- Tlalim. All errors in this sometimes hastily compiled paper are of course my own.


disciples. It is intended that in a series of future publications, separate papers will be produced expanding on these twelve alternative historical approaches in turn. Prominent is a more detailed look at the role in the historical formation of the Tibetan gter ma tradition of the widespread indigenous Tibetan cosmologies involving environment and wealth controlling ancestral local deities, and the so far little examined yet closely interrelated indigenous traditions of offering gter to such gzhi bdag or yul lha, in return for worldly prosperity and general environmental enrichment.


Equally important are a variety of treasure practices from the cultures surrounding Tibet, that have not yet been properly considered. Fundamental to everything is an ap- preciation of the growing anthropological literature on treasure discovery, which is now recognised by anthropologists as a widespread phenomenon, occurring in many different cultures across the world. For us, this opens the gate to understanding the emergence of gter ma, driven by particular historical conditions in Tibet, through the hybridisation of originally quite separate indigenously Tibetan and im- ported Buddhist treasure conceptions.

The second half of the present paper is thus intended largely as an advanced notice of a series of further papers, currently in various stages of preparation, while the first part is largely taken up with the critical reexamination of the rGyal po bka'i thang yig referred to above.


O rgyan gling pa and his Treasure Chronicles


O rgyan gling pa was born in 13233 at Ya rje in lHo brag, Southern Tibet, a region from earliest times strongly devoted to Padmasambhava,4 the master from Uḍḍiyāna around whose person so much of the later gter ma tradition was narrated.

O rgyan gling pa was renowned in his day for ritual and material treasures, but is nowadays best remembered for his historical treas- ures, the bka’ thang sde lnga, or Five Chronicles. These are [1] The Chronicles of Gods and Demons (lha 'dre bka'i thang yig), [2] Chronicles of the Kings (rgyal po bka'i thang yig), [3] Chronicles of the Queens (btsun mo bka'i thang yig), [4] Chronicles of the Translators and Scholars (lo pan bka'i


3 The dates of his death are unclear.


4 Padmasambhava’s particular association with a number of locations to the south of Tibet, including lHo brag, Bumthang in the modern Bhutan, and Yanglesho in Nepal, is described in the Dunhuang text PT 44. Matthew Kapstein has proposed an interesting hypothesis for a historical Padmasambhava, drawing on all extant early sources, that also links him particularly to these regions (personal communication, 29th June 2015); see also Kapstein 2000, page 159.


thang yig), and [5] Chronicles of the Ministers (blon po bka'i thang yig). In addition, he produced a major hagiography of Padmasambhava called the [6] Padma bka’i thang, or Chronicles of Padma. Taken together, these provide a comprehensive mythology of the advent of Buddhism in Tibet, narrated from various different perspectives.


The later reception of one of these Chronicles was unusual. gTer ma was principally the preserve of the culturally more indigenised but often politically more marginalised rNying ma school. gTer mas were on the whole less valued by the more powerful dGe lugs and Sa skya schools, who tended (rhetorically at least) towards privileging tradi- tions of proven Indian origins above others, yet who usually carried most weight amongst the Mongol and Chinese forces that were so influential in Tibet for several centuries. But it was precisely within the nexus of dGe lugs canonical orthodoxy and Mongol and Chinese power, that O rgyan gling pa’s Padma bka’i thang, or Chronicles of Padma, found particular favour. Several editions were printed in Beijing and Mongolia for political reasons and promoted there by the very agencies one might otherwise expect to ignore them.5


There is some possibility that the reason for this Mongol and Chinese political interest might prove germane to my discussion, so I will return to it below (for although it is not yet clear how strong the possibility might turn out to be, the currently known evidence does seem to warrant further investigation). For now, I will just mention that the prominent state promotion of O rgyan gling pa’s chronicles helped bring them to the attention of early western Tibetologists (Laufer 1911, Francke 1927, Tucci 1949), whose interest was further piqued when they found that amongst their highly mythologised narratives, these Chronicles also preserved some demonstrably ancient materials of real value to historians.6


What early materials do the Treasure Chronicles have?

Cathy Cantwell has already shown how, in the case of the ritual or tantric texts that make up such a major part of gter ma literature, although a high status lama, whether gter ston or editor, can sometimes


5 See Kapstein 2015, “Gter-ma as Imperial Treasure: The 1755 Beijing Edition of the Padma bka’ thang”, in RET Number 31, Février 2015, Papers for Elliot Sperling.

6 B. Laufer, 1911. Der Roman einer tibetische Koenigin (bTsun-mo bka’-thang), Leipzig. Francke, A. H., 1927. ‘gZer-Myig. A Book of the Tibetan Bompos’. Asia Major, Vol. iv, Fasc. 2-3, 1927. Tucci, G., 1949. Tibetan Painted Scrolls, 3 vols., Libreria dello Stato, Roma (see pages 110-111).


change the practices considerably, nevertheless it is commonplace to find older materials recycled within newer gter ma. Quite frequently, the later ritual gter mas barely change the earlier ritual texts they incorporate.7 In a paper inspired by Cantwell’s textual findings and also by the methods of literary analysis developed by the Hebrew scholar Peter Schäfer, I have analysed some of the repeated structural patterns that can frequently be seen to govern the construction of new works out of parts of older works.8

Cathy Cantwell’s forthcoming book, which looks at rNying ma phur pa gter ma over the longue durée (13th century to 20th century), will present considerably more evidence confirming these processes, and in considerably greater detail. While not in any way denying that gter ma literature can sometimes innovate, from the sample she has studied, there seems to be little evidence that gter ma is either appreciably more or less innovative than non-gter ma literature.9

But O rgyan gling pa’s Chronicles are not ritual texts. On the contrary, what makes them so fascinating is their reproduction of lengthy narratives, deployed for the purposes of creating a historiography of the advent of Buddhism in Tibet. No systematic comparison has yet been made between textual reuse in ritual gter ma and in historiographical narratives such as these: might such an analysis find, as in Cantwell’s study of ritual texts, scant evidence that historical narrative

gter ma is either appreciably more or less innovative than non-gter ma historical narrative? Tantric rituals are conceived of as unchanging and timeless, which tends to put a brake on the degree to which they should change as they cascade down the centuries from one gter ma to another; but it is not yet clear if, or to what degree, or in what ways, traditional historiographical or mythic narratives might (or might not) have been considered more malleable. Nevertheless, five significant examples of O rgyan gling pa’s tex- tual reuse have already been analysed by Tucci, Blondeau, Karmay,


7 Cathy Cantwell, 2014. “Development of a Tantric Meditation Practice: Three Generations of Tibetan Lamas” In: Dharma Drum Buddhist College Special Series 2, ed. Chuang Kuo-pin (莊國彬),Taipei: Shin Wen Feng Print Corporation (新文), pp. 374-403; see also Cathy Cantwell, 2015. “Different Kinds of Composi- tion/Compilation Within the Dudjom Revelatory Tradition”, In Journal of the International Association of Buddhist Studies, Volume 36/37 2013/2014 (2015), pages 243-280. 8 Robert Mayer, 2015. “(What's so new about Terma?) Terton and Tradent: Innovation and Conservation in Tibetan Treasure Literature”, In Journal of the Interna- tional Association of Buddhist Studies, Volume 36/37 2013/2014 (2015), pages 227-242.

9 Cantwell, C. in press. Dudjom Rinpoche's Vajrakīlaya Works: A Study in Authoring, Compiling and Editing Texts in the Tibetan Revelatory Tradition. Equinox, Sheffield.


Kvaerne, Pritzker, and now Esler. From these five examples, we can see that with only one exception, O rgyan gling pa’s Chronicles do not merely reproduce their ancient narrative source texts with little change, or even verbatim, as is not an unknown practice among ritual gter mas. The examples so far studied indicate that the newer versions resurfacing in O rgyan gling pa’s Chronicles have in all cases except one been redacted to achieve meanings significantly at variance with their textual predecessors. To what extent this was done by O rgyan gling pa, and to what extent by possible unknown intermediary sources, has not yet been ascertained, but what is clear, and important for my present argument, is that older narratives do reap- pear within O rgyan gling pa’s Chronicles in a significantly transformed manner.

In 1971, Anne-Marie Blondeau discovered that in his Chronicles of Gods and Demons (lha 'dre bka'i thang), O rgyan gling pa had adopted an older Bon po text, the gZer mig, not merely copying it, but adapting it, to give it a new, changed, Buddhist meaning.10


A few years later, Blondeau11 with Per Kvaerne12 then made a further discovery about O rgyan gling pa’s use of the gZer mig. Lau- fer had translated the Chronicles of the Queens (btsun mo bka'i thang) as early as 1911,13 and in 1927, Francke had already observed how it had parallels with the gZer mig.14 Blondeau and Kvaerne could now prove that the Bon text was older, and that once again O rgyan gling pa had redacted it to change the names and the specific episode to suit his Buddhist version of the story.

In his work of 1988, Samten Karmay describes part of the Chronicles of the Ministers (blon po bka'i thang) as ‘a pell-mell summary of chapter four of the bSam gtan mig sgron',15 a much older work attributed to the possibly 9th or 10th century gNubs chen sangs rgyas ye shes. Tucci’s earlier edition and translation had missed this de-


10 A-M Blondeau, “Le Lha-’dre bka’-thaṅ”, in A. Macdonald (ed.), Études tibétaines à la mémoire de Marcelle Lalou, Paris, 1971, pp. 29-126.

11 A.-M. Blondeau, Annuaire de l’E.P.H.E., 5ème section, 84 (1975-1976), pp. 109-119, 85 (1976-1977), pp. 89-96.

12 Per Kvaerne, 1980. ‘A preliminary study of the gzer mig’. In Michael Aris and Aung San Suu Kyi (eds.), Tibetan Studies in Honour of Hugh Richardson, Warmin- ster (Aris & Phillips), 1980, pp.185-191.

13 B. Laufer, Der Roman einer tibetische Koenigin (bTsun-mo bka’-thang), Leipzig, 1911.

14 Francke, A. H., 1927. ‘gZer-Myig. A Book of the Tibetan Bompos’. Asia Major, Vol. iv, Fasc. 2-3, 1927.

15 S. Karmay, The Great Perfection (rDzogs chen) A Philosophical and Meditative Teaching of Tibetan Buddhism. Leiden, 1988. See pp. 90ff.


pendency.16 In several pages of dense analysis, Karmay demonstrated how O rgyan gling pa redacted numerous passages in gNubs chen’s original text, to produce meanings often at variance with the original. Here, the material is as much doctrinal as historical.

A few years later, Tanaka and Robertson (1992) returned to this point in more detail, reaffirming Karmay’s findings.17 They describe Chapters 12, 13 and 14 of the Blon po bka’ thang as a disordered repetition of passages from the bSam gtan mig sgron with substantially rearranged meanings.

The same material has now been revisited again by Dylan Esler, this time in much greater detail still, and Esler comes to similar con- clusions.18 In his Oxford DPhil of 2017, David Pritzker writes that the 12th century Tholing ms. has an account of the pre-Imperial Twelve Minor Kingdoms of Tibet closer to the Chronicle of the Kings (rgyal po


16 G. Tucci. Minor Buddhist Texts, Serie Orientale Roma, IX, Part II. Rome 1958.


17 Tanaka, Kenneth K., and Raymond E. Robertson. 1992. 'A Ch'an Text from Tun- huang: Implications for Ch'an Influence on Tibetan Buddhism'. In Steven D. Goodman and Ronald M. Davidson, eds., Tibetan Buddhism: Reason and Revelation, pp. 57–78. Albany: SUNY Press.

18 Dylan Esler, The Lamp for the Eye of Contemplation. The bSam-gtan mig-sgron by gNubs-chen Sangs-rgyas ye-shes: Hermeneutical Study with English Translation and Critical Edition of a Tibetan Buddhist Text on Contemplation. PhD thesis. Louvain-la- Neuve, 2018. page 15: “However, this is not to say that the bSam-gtan mig-sgron was completely unknown in Tibet: it is mentioned in an official decree (bka’-shog) of Pho-’brang zhi-ba-’od (11th century). Furthermore, as pointed out by Karmay, extracts from the bSam-gtan mig-sgron appear in O- rgyan gling-pa’s (1323- ca. 1379) treasure the Blon-po bka’-thang. Tanaka and Robertson [Tanaka and


Robertson 1992, ‘A Ch’an Text from Tun-huang’, pp.57-78.] have demonstrated in some detail that Chapters 12, 13 and 14 of the Blon-po bka’-thang are in fact a somewhat patchwork rearrangement of verbatim sections of the bSam-gtan mig-sgron. Matter which in the bSam-gtan mig-sgron occurs in a natural order is rearranged quite artificially in the Blon-po bka’-thang. Moreover, O-rgyan gling-pa does not distinguish between main text and quotations as differentiated in the bSam-gtan mig- sgron. Finally, the text of the Blon-po bka’-thang presents the material in such a way that it appears to be a debate between only the gradual and


simultaneous approaches, whereas the bSam-gtan mig-sgron has a vaster scope, since it also covers Mahāyoga and rDzogs-chen. O-rgyan gling-pa’s motivation would appear to be to defend the rNying-ma school against the polemicists who claim that rDzogs-chen and Chan are identical. However, rather than simply disavowing a doctrinal identity between both traditions (which would have been perhaps more straightforward since this is very much one of the intentions behind the composition of the bSam-gtan mig-sgron), the author of the Blon-po bka’-thang sets about to prove the superiority of Chan to the gradual approach. In this he can base himself on the bSam-gtan mig-sgron, but he seems to bypass the distinction which this text then goes on to draw between Chan on the one hand and rDzogs-chen on the other.”

bka'i thang yig) than to any of the other sources that carry this narrative; although it is not by any means identical.19 Conversely, back in 1950, Tucci had famously demonstrated that O rgyan gling pa’s Chronicle of the Kings (rgyal po bka'i thang yig) included a faithful reproduction of an 8th century pillar inscription that still stood beside the tomb of Emperor Khri lde srong btsan (798/800-815).20 Tucci inspected the pillar when he visited Tibet in 1948, and found the version in the Chronicle of the Kings so accurate that he used it to reconstruct words that time had effaced from the pillar. Since O rgyan gling pa had written about the tombs, and knew the region, Tucci concluded he had more likely read directly from the pillar inscription, than from another text. Similarly, Orgyan gling pa’s physical descriptions of the tombs might well be accurate.


It has recently been pointed out to me by Stefan Mang, a graduate student from the Rangjung Yeshe Institute in Boudha, Nepal, that in his Chronicles of Gods and Demons (lha 'dre bka'i thang), O rgyan gling pa preserves a particular narrative detail in some respects closer to the Dunhuang text PT44 than many other extant versions of this popular narrative.21


To what degree can we rely on the Treasure Chronicles as historical sources?

In 1969, Tucci’s Danish disciple, Erik Haarh, produced a magisterial PhD on the Yar-lung dynasty and its burial traditions, which, despite being notionally a doctoral dissertation, nevertheless deservedly remains a famous classic of Tibetological writing. Perhaps inspired by Tucci’s optimistic discoveries of the 1950’s, but writing too early to be cautioned by the later caveats of Karmay, Blondeau, Kvaerne, Tanaka and Robertson, Pritzker, and Esler, Haarh repeatedly approached the 6th to 8th century Tibetan Dynasty through the medium of much

19 D. T. Pritzker. Canopy of Everlasting Joy: An Early Source in Tibetan Historiography and the History of West Tibet, Oxford DPhil dissertation, 2017. See p. 83. Other sources that carry versions of this narrative include mKhas pa’i dga’ ston, Old Tibetan Chronicle (PT 1286, ll. 22-24), PT 1060, PT 1038, and early chos ’byung such as mKhas pa lDe’u.

20 G. Tucci. The Tombs of the Tibetan Kings. Serie Orientale Roma I Rome 1950. P. 39 ff. 21 This is the story of Padmasambhava confining the troublesome goddesses he encountered at Pharping in Nepal within his hat. Both PT44 and the lHa 'dre bka'i thang preserve this detail of the hat, while a number of other sources have substituted various other kinds of containers for the hat. See O rgyan gling pa, Bka’ thang sde lnga (Pe cin: mi rigs dpe skrun khang), 23. Available at BDRC here: https://www.tbrc.org/#!rid=W17319.


later gter ma texts, not least O rgyan gling pa’s 14th century Chronicles. In fact, he understood O rgyan gling pa’s 14th century Chronicles quite unquestioningly as later editions of works originally written in the 6th to 8th centuries. Specifying his historical source materials at the beginning of his dissertation, he wrote “By far the largest group of [my] sources is constituted by …. later editions of documents dating from the time of the Dynasty, the so-called gTer-ma. Among the most important ones for our purposes, are the texts classified as bKa’i-thaṅ-yig.”22 And a few pages later, he reiterates that he has ac- cepted ten gter ma texts, including O rgyan gling pa’s bKa’i thang yig, as ‘texts from the time of the Dynasty in later editions: later Tibetan editions of documents or other texts dating back to the time of the Yar-lung Dynasty’.23


With hindsight, it now seems reasonable to enquire if Haarh’s naively optimistic acceptance, when still a graduate student, of O rgyan gling pa’s texts as substantially unmodified sources from the 6th to 8th centuries, might have established a dangerous precedent by which some later authors might have been unwittingly influenced. Thankfully, this has not been the case for Haarh’s direct student, the distinguished historian Per Sørensen, nor his equally esteemed col- laborator Guntram Hazod, both of whom have been realistic and suitably cautious in their usage of gter ma texts as historical sources. But I think there is reason

for concern that some others, outside of Haarh’s direct intellectual lineage and perhaps less aware of his subsequent intellectual development, might still be in danger of reproducing the mistake of his famous doctoral thesis, in taking much later gter ma historical narrative texts too much for granted as reliable or even unmodified historical sources for earlier centuries, even where little or no independent corroborating evidence has yet been found. For it would seem to me that gter ma sources are sometimes historically reliable, and sometimes not; so that before accepting their testimony, additional corroborating evidence should wherever possible be sought.


Nowhere is this more apparent than in the historical study of ths Tibetan gter ma system itself. In recent decades, perhaps not least through the persisting influence of Haarh’s doctoral thesis, O rgyan gling pa’s Chronicles have widely become accepted as significant sources for proposing a putative indigenous Tibetan precursor of the

22 Erik Haarh. The Yar-luṅg Dynasty : a study with particular regard to the contribution by myths and legends to the history of ancient Tibet and the origin and nature of its kings. København, Gad Forlag, 1969. Page 15. 23 Ibid p. 19.


gter ma tradition that was connected with royal burials, despite the fact that we have virtually no other independent evidence for this, apart from other still later gter ma.

Particularly influential in our understanding of the origins of gter ma have been three passages from the single 1889 Potala edition of the Chronicle of the Kings that Haarh consulted for his PhD (1969: 21). These particular three passages have been cited by subsequent scholars, but so far at least, without very much systematic attention to questions of transmissional and interpretive difficulties. Thus we don’t yet know to what extent the transmission of the Chronicle of the

Kings might have been subjected to textual variations across its different editions, nor how this might impact on Haarh’s exploratory readings and translations of the 1889 Potala edition, some of which I suspect might be open to rather different translations or interpretations, with rather different implications.24 But let’s leave aside all such possibilities of textual variation and alternative translation and interpretation for the moment, since we cannot sort them all out here, and take Haarh’s pioneering readings at face value:


First is a passage describing the tomb of the late 6th or early 7th century kingBro gnyan lde’u (rGyal po bka’ thang 1889 Potala ed., 37.54r-56r). In Haarh’s translation, we read how a gold image of the king’s corpse, along with other precious things, were buried and sealed up with earth, stones and wood. Thus valuables were hidden for future royal generations (see Haarh 1969: 349).

Second is a passage from Chapter 13 (rGyal po bka’ thang 1889 Potala ed., 37.39v-40v), describing offerings at the tomb of Srong btsan sgam po. Haarh’s translation describes how some of his surviving ministers continued to serve their dead king by living at his tomb whilst segregated from the wider population, and how they should


24 On the one hand, the transmission of the Chronicle of the Kings might have been subjected to textual variations across its different editions, yet the extant witness- es have neither been compared, nor subjected to sufficient scholarly editing, so we don’t yet know. On the other hand, some scholars might consider some of Haarh’s readings and translations of the 1889 Potala edition open to alternative interpretations: for example, to what extent and in which places do they actually describe gter ma to be recovered and removed at a later date, as Haarh is seen to be implying?

Or might some of these passages simply describe grave goods to remain buried forever, yet these nevertheless for the benefit of later generations of royalty, thereby mirroring some later Tibetan funerary practices as reported independently by two contemporary ethnographers of East Tibet, Sehnalova and Langelaar? It might therefore be wise to look at more Tibetan editions of the Chronicle of the Kings, and review Haarh’s translations and interpretations, before deciding on the meanings and significance of these passages.


enjoy the ritual offerings left for their dead monarch (see Haarh 1969: 350-2).

The third passage also comes from Chapter 13 (rGyal po bka’ thang 1889 Potala ed., 37.43v-45r), and describes treasures buried in Srong btsan sgam po’s25 ‘wealth tomb’ (nor gyi bang mdzod) for the benefit of future generations, including grave goods like copper vessels and horse-headed jars, heirlooms from previous kings, and the so-called Thirteen Royal Treasures (Haarh 1969: 352-355). 26 Haarh translates the title of this section as ‘The hiding of the inheritance for the future royal generations, directions as to the royal family’s offerings, and the sealing (of the tomb)’.27

Here O rgyan gling pa introduces the words kha byang and lde mig, words which by his time were well established as technical terms of the gter ma tradition, indicating ‘address books’ or prophetic lists or guides to a Treasure cache (we do not yet know much about any ritual meanings these terms might have had prior to their adoption by the gter stons). O rgyan gling pa claims this kha byang was first passed down within generations of the royal family, and finally hidden as a gter ma by Mu tig btsan po. The royal name Mu tig btsan po is used inconsistently in Tibetan historical literature, but here, as Dotson observes, O rgyan gling pa might indicate the Buddhist monarch Khri lde srong btsan or Sad na legs,28 son and successor to Khri srong lde’u btsan, the famous emperor who promoted Buddhism and was the first to make it a state religion (see Haarh 1969: 352-354).

Emperors, notably Srong btsan sgam po and Khri srong lde’u btsan, along with their families, were central to early and later Treasure myths. In our view, there were a number of reasons for this. More generally, it reflected pre-Buddhist beliefs of reverencing the Emperors, as well as the bodhisattva statuses increasingly ascribed to them


25 Reigned c. 617 - 649/650, and by O rgyan gling pa’s time, revered as the founding father of the Tibetan Empire, and introducer of literacy. 26 In relation to O rgyan gling pa’s pronounced interests in Imperial-period material royal treasures, it is noteworthy that Rig ‘dzin rgod ldem, who was likewise much interested in Imperial-period material royal treasures, including those found in the hidden lands (sbas yul), was a near-contemporary, a mere fourteen years younger than O rgyan gling pa (1337-1409).

27 Ma-‘ongs rgyal-brgyud nor-skal sbas-po daṅ : rgyal-brgyud bla-mchod man-ngag rgyas- btab ni: (Haarh 1969: 352). This section is cited in the slightly later rGyal rabs gsal ba’i me long, Chapter 4, on burial of treasure. See Sørensen 1994: 307-312, especially note 949. rGyal rabs gsal ba’i me long mentions treasures buried in temples, but not in tombs.

28 Dotson, Brandon. “‘EmperorMu rug btsan and the ’Phang thang ma Catalogue.”

Journal of the International Association of Tibetan Studies, no. 3 (December 2007): 1-

25. http://www.thlib.org?tid=T3105 (accessed May 7, 2017).


over time. But perhaps more importantly for the purposes of this paper, it also owed much to the usefulness of gter ma narratives for the invention of traditions of grand Imperial patronage of Buddhist lineages. A particular preoccupation was to displace earlier unwelcome historiographies highlighting the rejection of Padmasambhava by the Empire,29 and the restrictions it placed on esoteric Vajrayāna,30 with an alternative gter ma mythology, emphasising the opposites. So while the founder of the Empire Srong btsan sgam po was re- cast through gter ma narratives as an advanced Avalokiteśvara adept, his descendant Khri srong lde’u btsan was now represented as Padmasambhava’s personal patron and closest disciple.31 From the 12th century onwards, Treasure Revealers claiming (as was claimed for


O rgyan gling pa) to be Khri srong lde’u btsan’s rebirths were to reappear repeatedly down the centuries, to rediscover the Treasures given them by Padmasambhava in their previous birth as the Emperor. True to form, O rgyan ling pa was himself deemed the reincarnation of the son of Mu tig btsan po, who was in turn the immediate reincarnation of his own grandfather, Khri srong lde’u btsan. So in revealing this section of the Chronicle of the Kings on hidden royal treasures, O rgyan gling pa is in fact presenting us with the gter ma guide buried by someone who was both his son and his father in past lives.32 Complex relationships of this type, that seek repeatedly to


29 This is described, for example, in the dBa’ bzhed. See Wangdu, Pasang and Hildegard Diemberger, 2000. dBa’ bzhed: The Royal Narrative Concerning the Bringing of the Buddha’s Doctrine to Tibet. Wien: Verlag der Österreichischen Akademie der Wissenschaften, pages 52-59.


30 While the Empire officially endorsed exoteric tantras, such as Kriyā tantras and Yoga tantras, a number of which we can see listed in the lHan dkar ma and ‘Phang thang ma Catalogues, it did not endorse the open translation of more esoteric tantras. See for example the sGra sbyor bam po gnyis pa, as presented in Ishikawa, Mie, 1990, A Critical Edition of the Sgra Sbyor Bam Po Gnyis Pa. An Old and Basic Commentary on the Mahavyutpatti. Tokyo, The Toyo Bunko, page 4. Perhaps this is a suitable moment to mention that the historically and politically very significant distinction between different genres of tantra and the related empowerments is sometimes not made sufficiently explicit in some recent writings on the Empire’s relation to tantrism.


31 While, in the light of recently read rDzong ‘phrang texts attributable to gNubs and his students, it appears not impossible that Khri srong lde’u btsan did receive some esoteric tantric initiations, alongside his more public engagement with the exoteric tantras, any association he might have had with the former was not made known, and does not appear in extant early sources.

32 The precedent for Treasure Revealers to identify themselves as reincarnations of Emperor Khri srong le’u btsan and his family was established as early as the 12th and 13th centuries by Myang ral and Guru Chos dbang. I have not yet ascertained exactly how O rgyan gling pa understood himself in his own words, but in later rNying ma tradition, which might well derive in at least some key respects interweave reincarnation and heredity over many centuries, remain typical of rNying ma Treasure culture to this day.


These three passages of the Chronicle of the Kings continue to influence major contemporary academic theories of the origins of the gter ma tradition. In them, some scholars have seen evidence for an originally non-Buddhist Treasure cult focused on the Emperors and their tombs, which later flowed into the Buddhist gter ma tradition.

One should note however that O rgyan gling pa did not himself intend them to describe the origins of the gter ma tradition. To the contrary, he held an altogether Buddhist view of its origins, presented in chapters 91 to 94 of his Padma bka’ thang, where he describes Padmasambhava as the ultimate source of the vast quantities of gter ma now in Tibet, along with their profuse accompanying prophecies and the like.


I will later elaborate on why an over-emphasis on Imperial persons, living or dead, as the foundation of the later gter ma tradition, could be problematic. On the one hand, such an emphasis can fail to distinguish Imperial regalia such as the Can dgu that should not go into occultation but were openly passed down through generations of royal succession and perhaps ritually transferred at coronation rites,33 from a separate category of occulted secret items, hidden and unknown for centuries. On the other hand, it can fail adequately to disambiguate the very many items in Tibetan culture that are classified as gter and buried forever for the

benefit of posterity, from a confrom O rgyan gling pa himself, he is reported as the seventh reincarnation of rGyal sras lha rje mchog grub rgyal po. rGyal sras lha rje is the son of Mu tig btsan po and the immediate reincarnation of Khri srong le’u btsan. In this slightly convoluted way, O rgyan gling pa too is therefore also deemed a reincarnation of Khri srong le’u btsan, and the guide to the royal family treasures described in this passage of the Chronicles of the Kings was thus deemed to have been buried after his death by his own son and heir. See p.775 of Dudjom Rinpoche’s The Nyingma School of Tibetan Buddhism. Its History and Fundamentals, trans. Dorje and Kapstein, Boston, 1991.


33 See the three-part entry on these regalia by Dan Martin on his blog Tibeto-Logic with the titles ‘Regalia Untranslatable Parts 1-3’ (October 9th, 12th, 14th, 2014). Here Dan Martin describes the three inherited from the father, the three inherited from the mother, the one weapon from the brothers, and the two ornaments from the sisters http://tibeto-logic.blogspot.co.uk/2014/10/regalia-untranslatable- part-one.html and http://tibeto-logic.blogspot.co.uk/2014/10/regalia-

untranslatable-part-two.html andhttp://tibeto-logic.blogspot.co.uk/2014/10/ regalia-untranslatable-part-three.html . See also the essay by Guntram Hazod, ‘The Plundering of the Royal Tombs, an Analysis of the Event in the Context of the Uprisings in Central Tibet of the 9th/ 10th Century’, in Christoph Cüppers, Robert Mayer and Michael Walter (Editors) Tibet after Empire: Culture, Society and Religion between 850-1000, Proceedings of the Seminar held in Lumbini, Nepal, March 2011, Zentralasiatische Studien ZAS ZAS 45, 2016, 704 S, pages 113-146.


ceptually different category of gter ma that are buried specifically for future recovery. The former includes grave goods, corpses of kings and important persons, and, above all, the commonplace treasure vases and innumerable suchlike gter that might win the favour of a gzhi bdag and enrich the environment but should not be dug up again. The latter are rarer, and include things like religious texts, religious objects, wealth, medical texts or medicines, that are concealed with a definite view to future recovery.

I will also argue, inspired by clues in Diemberger’s work on Khenbalung,34 Mathes’s work on Yolmo,35 Terrone and Jacoby’s work on modern Golog,36 Martin's work on gShen chen klu dga’,37 Hirshberg’s work on Myang ral,38 my colleague and collaborator Anna Sehnalova’s forthcoming work on contemporary Golog, and many other secondary and primary sources too (such as the autobiographical passages in Guru Chos dbang’s gTer ‘byung chen mo), that it might be misleading to put excessive

interpretive emphasis on cults of the Imperial person, whether living or dead, especially if at the expense of paying attention to local deity cults. For local and ancestral deity cults can be seen as presenting the broader cosmological fields within which the Emperor cult was a particular subset; and on existing evi- dence, local deity cults per se seem to offer sounder, richer and deeper perspectives for understanding the many and powerful influences of indigenous culture in the


history of gter ma. In particular it seems that we might so far have underestimated the relevance to the emergence of the developed gter ma tradition of indigenous cults relating to the local gzhi bdag; for they have their own rich parallel ritual worlds of gter burial, but also play a key role in the Buddhist and Bon gter ma traditions. In short, while I am in total agreement with Ronald Davidson and Janet Gyatso that the Tibetan gter ma tradition must in-


34 Hildegard Diemberger, 1997. “The Hidden Valley of the Artemisia: On Himala- yan Communities and their Sacred Landscape” in A.W. Macdonald (ed.): Manda- la and Landscape. New-Delhi: D.K. Printworld

35 K-D Mathes, 2013. ‘Clouds of Offerings to Lady g.Yang ri. A Protector Practice by the First Yol mo sprul sku Shākya bzang po (15th/16th Cent.)”. In Franz-Karl Ehrhard and Petra Maurer (eds), Nepalica-Tibetica: Festgabe fur Christoph Cüppers, Band 2.

36 Terrone, A, 2010. Bya rog prog zhu, The Raven Crest. The Life and Teaching of bDe chen ‘od gsal rdo rje, Treasure Revealer of Contemporary Tibet. PhD dissertation, Lei- den. Jacoby, S, 2014. Love and Liberation. Autobiographical Writings of the Tibetan Buddhist Visionary Sera Khandro. New York, Columbia University Press.

37 Martin, D., 2001. Unearthing Bon Treasures. Life and Contested Legacy of a Tibetan Scripture Revealer. Leiden: Brill (Brill’s Tibetan Studies Library, 1). 38 Hirshberg, Daniel. 2016, Remembering the Lotus-Born: Padmasambhava in the History of Tibet's Golden Age. Wisdom Publications, Somerville.


clude a great deal of indigenous Tibetan tradition, I am not entirely confident that Davidson in his important publications of 2005 and 2006 (see note 37 below), and Gyatso too in her more recent one, have been focusing their search for these Tibetan contributions in the most promising place.39

It seems significant that despite the huge advances made over the last fifteen years, little or no evidence has yet been found suggesting an indigenous non-Buddhist (and non g.Yung drung Bon) Tibetan ritual culture of treasure extraction (although we have as yet no knowledge of the anthropologically related phenomenon of mining for minerals and metals). So far, it seems that the rich and varied in- digenous Tibetan ritual treasure traditions were primarily donative, concerned with putting treasures into or upon the ground as offerings: they were less concerned with extraction, with taking treasure out of the ground. More specifically,

as far as current research is aware, the indigenous non-Buddhist (and non g.Yung drung Bon) Tibetan ritual traditions had no practices of intentionally hiding treasures for future recovery. By contrast, Indian and Chinese Bud- dhist treasure cults are the reverse: they are more interested in extracting treasures than in inserting them; and in important cases, they are primarily concerned with recovering particular treasures that had been hidden in the past with the intention of later rediscovery. This would suggest that the very idea of recovering treasures (rather than merely burying them) likely owes at least something to Buddhism, as well as to the historical circumstances described by Davidson.


The reception of the Chronicles of the Kings within modern scholarship on Treasure


But first let’s return to our reassessment of the scholarly mining of the Chronicles of the Kings as a historical source for the origins of gter ma. Prominent here is Ronald Davidson’s book of 2005, Tibetan Renaissance, Tantric Buddhism in the Rebirth of Tibetan Culture, and its follow-up article of 2006.40 As far as I am aware, Davidson’s writings

39 Janet Gyatso’s earlier work does not emphasise the tombs, yet in her 2015 contribution on gTer ma to the Brill Encyclopaedia of Buddhism, she does adopt a focus on the tombs, citing Davidson.

40 Davidson R., 2005. Tibetan Renaissance, Tantric Buddhism in the Rebirth of Tibetan Culture. New York, Columbia University Press. See pages 211-213, and 217-224, a section entitled Guarded by Spirits: The Hidden Imperial Person. A further work by Davidson (2006) elaborates on the same themes: ‘Imperial agency in the Gsar ma treasure texts during the Tibetan Renaissance: The Rgyal po bla gter and related


remain the most comprehensive and sustained attempts so far to account for the origins of gter ma. A scholar of exceptional erudition and insight, Davidson’s basic understanding is of gter ma excavation appearing in Tibet through a conflation of indigenous and Indian sources, and in this respect, I find myself in full agreement with him. His care in balancing both Indian and indigenous Tibetan factors was a marked improvement on Michael Aris’s earlier proposal (1989) that gter ma excavation owed little to Buddhism but much to Himalayan shamanism, and was equally an improvement on my response to Aris (Mayer 1992), in which I sought only

to document the Buddhist sources. His position still compares quite favourably with more recent statements from some other authors, who appear to imply, not altogether dissimilarly to Aris, that Buddhist sources in the construction of the gter ma excavation traditions are so slight as not to warrant any further investigation. However, while Davidson acknowledges the important contributions of Indian sources several times (2005 pp. 212, 215, 216, 217, 219), he did not have

the opportunity to study them in detail.41 Likewise, while he very briefly signals important and ground-breaking insights into the importance of ancestral yul lha deities in the early historical formation of gter ma (2005 p. 218), he was not able to follow through with any sustained discussion of that either. Instead, perhaps in part because his resources of time and research material were limited, Davidson put his focus elsewhere. He proposed we should consider the emergence of gter ma excavation primarily as an attempt by Tibetans of the phyi dar period to reconnect with relics of their past emperors, as preserved in surviving Imperial period temples and tombs, and devoted the greater part of his analysis to this theme.


After citing Haarh’s presentations of passages on the royal tombs from the Chronicles of the Kings several times, Davidson sums up thus:


literature’, in Davidson R and C Wedemeyer, eds., 2006. Tibetan Buddhist literature and praxis: studies in its formative period, 900-1400. PIATS 2003. Tibetan studies: proceedings of the tenth seminar of the International Association for Tibetan Studies, Oxford, 2003. Leiden, Brill. pp.125-147

41 Another reason Davidson did not pursue these themes might have been scholarly courtesy. He cited my short article in PIATS 1992, written when I was a doctoral student, and evidently left the field open for me to complete. I regret I never did so at the time. See Robert Mayer, 1994. “Scriptural Revelation in India and Tibet. Indian Precursors of the gTer ma Tradition”. In: Tibetan Studies, PIATS 6, Vol. 2. Institute for Comparative Research in Human Culture, Oslo, pp. 533-544. This paper was intended as a response to Michael Aris's earlier proposal that gter ma derived predominantly from Himalayan shamanism with little Buddhist content (1989, p. 59ff), and so was limited to listing some Indian sources that had influenced the gter ma traditions.


‘When we turn from such descriptions to those of the early Terma, the similarities are palpable. In fact, the ideology of royal sites appears to be the primary source for the early Terma descriptions. These sites, temples and tombs, were the repositories and subsequent sources for treasure, whether precious stones and metals or written texts” (Davidson 2005: 224).


To elaborate further (pages 217-224), he says that the burial tombs were repositories of the Emperor’s life force or bla. The Imperial bla had the power to enrich and bless, hence he proposes that there was a pre-Buddhist ritual tradition for blessing the realm, by distributing the Emperor’s bla across the Empire via material treasures in which the Imperial bla resided. Since Imperial tombs remained repositories of a deceased Emperor’s bla long after his death, Davidson proposes that treasures taken out of the tombs and processed across the country could also distribute his bla posthumously, and that this practice was an important predecessor of gter ma excavation. Davidson thus sees the Tibetan gter ma excavation tradition as at root a Buddhist re- expression of putative rituals relating to the Emperor’s bla stored in material treasures intended to be excavated from the royal tombs.


Yet an important aspect of Davidson’s hypothesis still lacks any unambiguous support from contemporaneous or early sources. While his understanding of the broader social historical dynamics giving rise to a treasure cult in Rennaissance Tibet in many ways impressively anticipated the perspectives of later anthropologists such as Valtchinova (2009), Bernstein (2011), Stewart (2012), and Gazizova (2019), some of his more speculative ideas on cultural specifics are less certain. As far as I am aware, in the intervening years since he proposed them, little new evidence has been discovered to support them. For only some (by no means all) early

gter ma was discovered in old Buddhist temples, whether Imperial or otherwise. Likewise, we still have scant reliable evidence for early gter ma being excavated from Imperial period tombs, whether Royal or not. I hasten to add, that is not to deny that the presence of vast quantities of grave goods in often abandoned tombs in many parts of Tibet, ripe for looting, might have influenced the circumstances of the early phyi dar, but that is a different consideration from Davidson’s idea of items buried with the specific intention of their future recovery.


All of the sources Davidson cites for his theory of bla come from gter ma texts centuries later than the events they describe, but in some instances some scholars question Davidson’s interpretations of these later sources. For example, in relation to the crucial word bla, some of his citations occur in gter ma passages where the meanings might be to indicate the emperor’s own personal treasures, and other times to indicate an original text rather than its copies.42


Moreover, as I have already intimated above, recent historical and ethnographic research challenges one of Davidson’s fundamental premises, that power objects should be ritually taken out of the tombs after being buried there. Contemporary evidence from East Tibet (Golog and surrounding areas) does indeed support the idea that the buried bodily remains of powerful persons, such as chieftains, can be classified as gter. Likewise, Anna Sehnalova believes that the objects buried in Songtsan Gampo’s tomb for the benefit of future generations of his lineage, as described above by O rgyan gling pa, would indeed be understood as a gter burial,

at least in parts of contemporary East Tibet. Similarly, it is quite possible that the ancient Royal Tombs might resemble practices in contemporary Golog, in being situated at the bla gnas of the royal family, and thus by extension at the bla gnas of their subjects. Yet it is altogether unclear from contemporary ethnography and ancient sources alike, how bodily remains or other items such as grave goods could later be ritually removed from such tombs as Davidson suggests, unless as a hostile act.43


I should make it absolutely clear that I am not suggesting that Davidson would be on the wrong track if he were simply raising the general issue of the burials and bla of the imperial forefathers, since these ideas were in all likelihood very significant parts of the general cultural background. It is therefore also credible that early gter ma excavation included some items narratively associated with the bla of


42 I consulted the following scholars: Daniel Berounsky, Henk Blezer, Brandon Dotson, Guntram Hazod, Samten Karmay, Per Kvaerne, Dan Martin, and Charles Ramble. The consensus was still to consider Davidson’s hypothesis regarding the Imperial bla as a predecessor for gter ma excavation, as still open to question, ra- ther than accept it as now proven.


43 My colleague Dr Anna Sehnalova offered me the following additional observa- tion (personal communication, 21st October 2018): “I also have a few examples from Golog where the repositories of bones of ancestors become bla ri of related settlements/valleys/lineages/regions (in one case the hill even has the word gter in its name, and has turned into a still venerated gzhi bdag with a la btse on top, who has got his own unwritten bsang recital, etc.; the place is one of the most im- portant in the history and identity of all Golog). But these repositories can never be excavated.”

Dr Sehnalova has been working on these issues for some years, and I will reference her work at greater length below. It might be worth noting that Sehnalova's work was done in the immediate neighbourhood of Dar thang monastery, whose abbot in exile, Tarthang Tulku, resided in Berkeley at exactly the time that Davidson studied there. Might some of Davidson's unattributed re- flections on Tibetan burials, which seem so closely to reflect the local practices described by Sehnalova, have been influenced by unreported conversations with Tarthang Tulku?


past royals, and perhaps such references can be found in Guru Chos dbang's gTer 'byung chen mo.44 For reasons explained below, they might also have taken on an increased importance in treasure discov- eries subsequent to the Mongol influence in Tibet, e.g. in O rgyan gling pa and his younger contemporary Rig 'dzin rgod ldem. But I suggest they are best seen as only part of a much wider panoply of excavated gter ma items. Until and unless further supporting evidence comes to light, I believe they cannot yet (as Davidson has suggested) be accepted as an inner core or major historical inspiration of the whole gter ma excavation phenomenon, nor as significant surviv- als of a pre-Buddhist proto-gter ma cult.


Finally, while it is true, as Davidson points out, that the seminal early gter ston Myang ral was much concerned with the Emperors, quite other explanations for Myang Ral’s interest in the royalty can be found. On the one hand, some of the Emperors were revered figures, increasingly identified as the incarnations of celestial bodhisattvas, and thus very valuable sources of authority for new gter ma revelations. On the other hand, some of the earliest literary sources had described Padmasambhava and his tradition as restricted and marginalised by the Imperial state. In response, Myang ral was much concerned to counter such


suggestions by generating alternative interpretations of history that portrayed the Padma tradition as promoted and fêted by the Emperors. Myang ral seems almost as much concerned with using gter ma to consolidate this revised perspective on history, as with the discovery of new teachings per se. Without the 12th century need to reframe Padmasambhava’s relationship with Khri Srong lde btsan, I wonder if the genre we now know as gter ma would ever have developed along quite the same trajectory. This is an important reason why early gter ston are so preoccupied with the Empire: to help inspire their 12th century program for spiritual re- newal, they are promoting a popular historical narrative to show that back in the 8th century, their tradition had never been marginal, underground, or restricted, but had always been glorified by the Imperial family, at the very centre of national culture and power.


But for now I will put the various interpretive and philological issues aside, to focus again on the academic reception of O rgyan gling pa. Turning to the Chronicles of the Kings as a possible source for understanding the origins of gter ma is quite usual in contemporary

44 See for example page 98, in Guru Chos dbang's gTer ‘byung chen mo, within Gu ru chos dbang gi rnam dang zhal gdams, Rin chen gter mdzod chen po’i rgyab chos Vols 8-9, Ugyen Tempa’i Gyaltsen, Paro, 1979. TBRC Work Number 23802.

scholarship, and in particular, Davidson’s interpretations are referenced. For example, Janet Gyatso writes in her 2015 entry gTer ma for the Brill Encyclopaedia of Buddhism:

Some Treasure histories, like the Bka’ thang sde lnga, describe an old practice of concealing royal treasuries in order to preserve them for future kings and generations, and provide minute details of their contents (O rgyan, 1986, 153–208). R. Davidson proposes that this practice is closely connected to the interment of the Yarlung dynasty kings in tombs inside of large tumuli, the remains of which still stand today [Gyatso 2015:399].


Is there any independent evidence for the Treasure practices described in the Chronicles of the Kings?


But what happens if we ask if any further supporting evidence can be found in sources closer to the 8th century than O rgyan gling pa’s distance of five hundred years? If such evidence exists, I have never encountered it. Nor, it seems, have any of my colleagues. For example, I consulted Brandon Dotson, an outstanding younger scholar who has dedicated his entire career to the institutions and ritual cultures of the Tibetan emperors. He responded:

“There is no evidence that I know of for concealing relics, regalia, or texts associated with the emperor with the expressed purpose of doing so for future generations to discover, let alone doing so with a table of contents or an instruction manual [[[kha byang]]]. Other argu- ments that can be made for various "gter ma like" practices also fall short…. I think you can be confident that in the imperial period there was nothing like [and here he quotes Gyatso’s summary of Davidson] 'an old practice of concealing royal treasuries in order to preserve them for future kings and generations, and provide minute details of their contents'. This, was no doubt read back into the imperial period by Tibetans from the late 10th century onward.” (Brandon Dotson, personal communication, 8th November 2016)


Guntram Hazod, whose fieldwork in Tibet has so brilliantly trans- formed our knowledge of the ancient tombs, also knows of no evi- dence for such practices outside of O rgyan gling pa’s Chronicles or other still later gter ma sources from the 18th century. If we have no evidence from Old Tibetan sources, can we find evi- dence in gter ma literature preceding O rgyan gling pa? The Bka’ chems ka khol ma mentions temples, but not tombs. Turning to O rgyan gling pa’s famous predecessors in his home region of lHo brag, Myang ral (1124-92), and Guru Chos dbang (1212-1270), I found noth-

ing in their works either. So I consulted Daniel Hirshberg, another outstanding younger scholar who knows Myang Ral’s biographies better than anyone.45 Hirshberg confirms that despite Myang ral’s famous claims to be the reincarnation of Khri srong lde’u btsan, his biographies specify gter ma recoveries at local mountains, and at old temples, but never at tombs.


If O rgyan gling pa is therefore the earliest evidence that scholars have so far found for linking gter ma discovery with the royal tombs, and it looks possible he might be, my instinct is to exercise caution, and await further evidence. Until such further evidence has been pre- sented, it might not be wise to read so much about the 7th and 8th centuries, into so few lines from a 14th century gter ma, with no known precedents. Conversely, I would welcome any significant precedents that might eventually come to light, in which case, I would be delighted to review my assessment.


Possible entanglements with the Mongols and China


If O rgyan gling pa’s Chronicles of the Kings is the earliest Treasure text we currently know of to link gter ma discovery with the royal tombs, then why? Is it simply because, for whatever reasons, it reproduced certain ancient themes which many earlier extant Treasure texts did not? Or could there be other reasons too? Earlier I mentioned that O rgyan gling pa’s Chronicles of Padma achieved particular favour with Chinese and Mongol state power. Matthew Kapstein has made a study of this,46 informed not least by Anna Seidel’s study of the evolving notion of “royal treasure” from early China onwards, and its great political significance. 47 Seidel wrote: (ibid: 299):


“In the case of the royal family, [[[Treasure]]] constituted the sacra or regalia of the dynasty. Their presence testified to the possession of the mandate and to Heaven’s continuing support. If the royal treas- ure-houses originally contained objects like stones, jade pieces,

45 See his book on Myang ral: Hirshberg, Daniel. 2106. Remembering the Lotus-Born: Padmasambhava in the History of Tibet’s Golden Age. Wisdom Publications, Somer- ville MA.

46 See Kapstein 2015, ibid.

47 Seidel, Anna. 1983. “Imperial Treasures and Taoist Sacraments: Taoist Roots in the Apocrypha,” in Michel Strickmann, ed., Tantric and Taoist Studies in Honour of R. A. Stein, vol.2, pp.291-371. Mélanges Chinois et Bouddhiques XXI. Brussels: Institut Belge des Hautes Études Chinois.


bronzes and weapons, they in time came to include talismans, magic diagrams, charts, prophetic adages, secret recipes for personal longevity and for the prosperity of the state and, finally, dissertations on moral and political doctrines. These texts soon were valued as more efficacious than the traditional object of the family treasure… In the politico-religious propaganda that brought the first emperor of the Latter Han, Kuang-wu, to power, the written word of Heaven in ch’an texts came to be valued and utilized far more than auspi- cious objects or natural phenomena. [...] The real content of the apocrypha was a collection of ancient legends and omen-lore recorded and elaborated for the legitimation of the Han.”


Kapstein observes that O rgyan gling pa’s Chronicles of Padma seem to achieve the same goal of legitimation, through their numerous pro- phetic announcements. He writes (ibid. p.175),


“As the warrant for imperial prophecies in general, can we imagine a more suitable treasure than that of O-rgyan-gling-pa? What is remarkable in this case is the evident congruence between the Tibetan revelation and Chinese conceptions of imperial treasure. That the prophetic books thought to be in the background of Tibetan and Mongol power were published in their Tibetan and Mongolian versions under the Manchu emperors in Beijing may perhaps be seen, therefore, as part of the ongoing response to the perpetual challenge of renewing the mandate of Heaven.”


I posed two further questions to the Sinologists at our Treasure semi- nar in Oxford: [1] Might the depictions of material treasures in the Chronicles of the Kings, match the legends in the Chronicles of Padma in also being congruent with Yuan conceptions of Imperial treasure? [2] And if so, were such congruences knowing or accidental?

The answer to the first question was supplied by Barend te Haar and is a clear affirmative. Herbert Franke has already shown that in accord with their own existing Mongolian cultural traditions, it was miraculously rediscovered material palladia,48 deemed to have been concealed since the early Chinese emperors, that became of sin- gular importance for the Yuan.49 This was despite the increasing tendency over time away from material treasures towards textual treas-


48 Palladium: A safeguard or source of protection. via Latin from Greek palladion, denoting an image of the goddess Pallas (Athene), on which the safety of Troy was believed to depend.

49 Franke, Herbert, 1978. From tribal chieftain to universal emperor and god: the legitimation of the Yüan dynasty. München, Verlag der Bayerischen Akademie der Wissenschaften.

ures, that Seidel describes in the citation above. Barend te Haar has further clarified Seidel’s findings for me by explaining that material royal treasures, such as ancient bronzes, remained highly valued through the Ming era and onwards.50


The second question I can frame myself, but not yet answer to my own satisfaction. Almost three centuries separated O rgyan gling pa from the adoption of his Chronicles of Padma by the Fifth Dalai Lama’s government, while the Imperial Manchu editions were a hundred years later still. At first glance, this might suggest that any simi- larities between O rgyan gling pa’s and Chinese Imperial treasures were accidental.


Yet on closer analysis, it need not be so. We know, for example, that his contemporary Dol po pa wrote the language of Mongol gov- ernmental edicts into religious texts.51 These authors lived in the troubled first half of the 14th century, when Tibet was reorienting its gaze from the South to the East. Earlier cultural, trade, and pilgrimage links with India were dwindling as Buddhism declined there, while political, economic, and cultural links with China had in- creased, following Yuan involvement in Tibet.


And O rgyan gling pa’s lifetime and some previous decades had seen a significant development in Tibetan religious life towards a new Chinese-influenced model that entailed difficult consequences for his rNying ma school. According to Tibetan historians, the orthodox Tibetan canon, the Kangyur, was first initiated at Narthang (Snar thang) by a follower of the New Tantras, Jamyang (’Byams dbyangs), who was at the time serving in China at the court of the Yuan Emperor Renzong (仁宗) a.k.a. Buyantu Khan (r. 1311–1320). Jamyang was evidently impressed by the Chinese model of a state-sponsored closed canon within which, in theory at least, only texts of proven Indian provenance were included, and from which those produced in China, such as Chinese Buddhist Treasure texts, were excluded as apocrypha.52 Jamyang sent funds back to Narthang together with


50 Haar, B. J. ter, 1998. Ritual and mythology of the Chinese triads: creating an identity. Leiden, Brill. See especially, but not only, the chapter on messianic demonology

51 Leonard van der Kuijp, ”Reconsidering the Dates of Dol po pa Shes rab rgyal mtshan's (1291-1361) Ri chos nges don rgya mtsho and the Bka' bsdu bzhi pa'i don," 藏族俄学刊 བོད་རིག་པའི་ད+ས་དེབ་ Journal of Tibetology 14 (2015/2016), 115-159. See page 118-20

52 Strickmann 1990 gives an interesting description of the 5th century Chinese Buddhist treasure text, the Consecration Sūtra (T1331), which, as Strickmann and oth- ers have noted (Strickmann p.88, and note 38, p. 115), was revealed as a treasure text in a manner resembling that of Tibetan gter ma. Although stringent efforts were made to rid the Chinese canon of such indigenously revealed apocrypha,


requests urging such a canon be made for Tibet.53 O rgyan gling pa’s 33 years older contemporary, Bu ston (1290-1364), then embarked on a huge bibliographic exercise similar to the state-sponsored one in China, to distinguish translations of Indic texts from the indigenously redacted scriptures favoured by the rNying ma pa. Those established as unredacted pure translations from Indic languages were included in the prestigious new Kangyur, while those redacted in Tibet and favoured by the rNying ma pa, including their Treasures, were excluded as apocrypha.


This challenging context for the rNying ma pa raises the question, might O rgyan gling pa have responded by knowingly portraying his beleaguered gter ma tradition as congruent with Yuan Imperial Treasures, and thus endow it with a powerful political symbolism of his era? Right now, I have absolutely no idea, but it seems prudent to pose the question. For if the Chronicles of the Kings was, like the Kangyur, contemporaneously influenced by Yuan or Chinese models, it might not simply reflect ancient Tibetan customs, in the uncomplicat- ed manner that some other scholars seem to have assumed.


Summary


To sum up, several questions still remain to be answered about O rgyan gling pa’s descriptions of treasures being buried in the royal tombs. Firstly, is he even describing gter ma for later recovery, as many scholars have assumed? Or only grave goods to remain buried for posterity? Or some combination of both? And is he creatively adapting older materials to suit his later agenda, or is he faithfully repeating ancient sources verbatim? We have seen how heavily he redacted earlier sources in his Chronicles of Gods and Demons, Chronicles of the Queens, Chronicles of the Ministers, and in other parts of his Chronicles of the Kings. Was he

doing the same here, tweaking them to assimilate his sources on the 7th and 8th century royal burial cults into his 14th century rNying ma gter ma Tradition, perhaps even with an eye to congruence with Chinese expectations? Or was he faithfully many hundreds nevertheless remained undetected. The Consecration Sūtra is one such, a Buddhist treasure text revealed in 5th century China, but passing as a canonical translation from Indian sources. See Strickmann, M., “The Consecration Sutra: A Buddhist Book of Spells,” in: R. Buswell, Jr. ed., Chinese Buddhist Apocry- pha. Honolulu, 1990, pp. 75-118.

53 Harrison, P., “A Brief History of the Tibetan Bka’ ’gyur,” in: J. Cabezon & R. Jackson, eds., Tibetan Literature: Studies in Genre, Ithaca NY, 1996, 70–94. See also Gzhon nu dpal 1984–1985: 410–412.

reproducing an entire ancient text with no changes, as with the pillar inscription in the Chronicles of the Kings? Or at least reproducing accu- rately some particulars of an ancient source, as he does with his narrative in the Chronicles of Gods and Demons, of Padmasambhava con- fining the Pharping goddesses within his hat? These questions need further research before we can be sure about anything, so I find it premature to accept the Chronicles of the Kings as a substantially un- modified description of ancient proto-gterma practices.


What other sources are available for the origins of the Tibetan Treasure tradition?


This, the second half of the present paper, is intended as an advanced notice of a series of further papers, currently in various stages of preparation.54 I mentioned earlier how probably excessive reliance on the internal evidence of the gter ma literature itself, as well as on the closely related works of its apologists, might in recent decades have distracted research efforts away from other so far largely unexplored avenues of enquiry, some of which might hopefully prove equally or even more fruitful.

As I pointed out above, a few of these have already been briefly signaled by Davidson, but given the major outpouring of work required, he was unable to follow their analysis through in most cases, and instead, concentrated much of his attention on his hypotheses related to the Imperial tombs. But what might we find if all the additional still unexplored avenues are further investigated? Over the next two years, I plan to investigate as many of them as I can, in an attempt to arrive, bit by bit, at a more rounded and complete understanding of the emergence of the complex and heterogeneous practices nowadays known as the gter ma traditions of Tibet. Thus it is hoped that in a series of future publications, written by myself and by others, separate papers will be produced expanding on each of the alternative historical approaches outlined below.


Our approach will take account of the entanglements of Tibetan, Indian, Chinese, Mongolian, and Western and Central Asian cultures central to our Oxford seminar, for like Davidson, we see gter ma as a complex set of cultural phenomena that reflects both Tibetan and non-Tibetan sources (we can no longer take seriously the proposition that gter ma can be wholly and sufficiently accounted for from indig-

54 I thought it prudent to commit an outline of my ideas to print as soon as possible, in case I run out of time to complete and publish more developed versions.

enous Tibetan origins alone). This work is also intended to provide the necessary context for understanding the workings of such im- portant early gter ston as Myang ral. At the time of writing, I am in all likelihood not yet aware of every avenue of enquiry that needs to be followed. But I can already briefly outline twelve, with the hope that most will prove useful, and some perhaps indispensable, for understanding the origins of gter ma.


Alternative source 1: Anthropology of treasure


Over the last decade, an increasing amount of anthropological litera- ture has been devoted to treasure recovery, which is now generally recognised as a phenomenon widely attested across numerous cultures and historical periods. Perhaps best known has been Charles Stewart's well-received book of 2012, Dreaming and Historical Consciousness in Island Greece, based on prolonged fieldwork on Naxos. In his preface to that work, Stewart presents his key themes, whose res- onance with Tibetan gter ma is self-evident:

divinely inspired dreams and visions of sacred objects buried by ancient Egyptian Christians their excavation as a religious practice the establishment of sacred texts and charismatic religious movements based on these treasures the central role of prophecy in all of this a cosmology of treasure discovery in which dangerous landscape spirits (‘moors’, ‘arabs’, and serpent spirits) guard hidden treasures.55

secrecy associated with treasure retrieval struggles to authenticate discovered treasures subsequent tensions between believers and sceptics an environment in which findable treasures (vresimata) do actually exist56 but are thought to need the help of divine beings to locate, most often the Panagía (Mary, mother of Jesus).


55 See especially Stewart 2012 Chapter 5, ‘A Cosmology of Discovery’.

56 The landscape of Naxos has indeed produced many valuable archaeological finds, including Classical Greek and even Cycladic artefacts of great value, as well as more recent treasures from the Byzantine period and later. In addition, and perhaps more fundamentally, the island’s economy has been for many cen- turies been based on the mining of emery (corundite).


Stewart is no Tibetologist. His field is contemporary Christianity on the island of Naxos, although he also refers to treasure beliefs in Bulgaria, Ancient Greece, Bolivia, Mexico, Papua, the USA, Armenia, and Turkey. Other scholars too have documented various treasure recovery practices (which can sometimes be described as indigenous pursuits of archaeology), in medieval Egypt and the Middle East (e.g. Cooperson 2010), medieval Jewish medicine ( e.g. Yoeli-Tlalim 2017, 2019),57 India (e.g. Hardy 1994), China (e.g Seidel 1983), SE Asia (e.g Leider 2009, Oppitz 2006), the Philipines (e.g Kelly 2016), pre- Buddhist Mongolia (Franke 1978), Zanzibar (e.g Walsh 2018), Mexico (Foster 1964), and more. Stewart has thus formulated an anthropology of treasure recovery to address a historically and geographically widespread phenomenon.

And in many respects, it works excellently with Tibet. His central theme is that, typically in response to a time of crisis, treasure discovery emerges in relation to the production of an affective popular historical consciousness. Thus in economically ravaged Naxos, narratives appeared about ancient Egyptian Christians who had visited to conceal sacred objects for later recovery, that would regenerate Naxos in its future time of need. Treasure discovery thus acted as a temporal rivet, binding together the present (when the treasure is discovered), with an affective popular historical consciousness of the past (visits to the island by ancient Egyptians), to aspirations for the future.

And contemporary Greece’s largest pilgrimage cult has indeed now developed around the recent Naxos treasure discoveries, allowing devotees to claim their prophecies and its purposes have been fulfilled. Similarly, Galia Valtchinova (2009) describes how interwar Bulgarian national renewal was supported by the widespread excavation of long buried Christian treasures lost in the period of Ottoman domination, now recovered by divinely inspired 20th century visionaries who received prophecies of where to dig. Similar themes of treasure and national renewal can be seen in Bernstein’s fieldwork from Buryatia (2011), and in Gazizova’s field- work in contemporary Kalmykia (2019).

In such examples, one is much reminded of the gter ma upsurge in renaissance Tibet, where treasure discoveries nailed aspirations for a happier future to a popu- lar historical consciousness of an idealised Imperial past, set within a cosmology of wealth- and treasure-guarding gzhi bdag, klu, and other landscape deities, and all this against a backdrop of actual Imperial ruins containing lost treasures and texts. I have found that the anthropological literature in general, and Stewart’s work in particular,


57 Lectures given to Oxford Treasure Seminar, Merton College, 21st November 2017, and Wolfson College, 6th March 2019.

offer valuable insights into the unconscious dimensions of Tibetan treasure discovery, which help illuminate the more conscious dimen- sions we Tibetologists mainly describe.

From the recent anthropological accounts of diverse treasure beliefs and practices, we can thus see that Tibetan treasure discovery is not the unique peculiarity of a remote and singular civilisation: ra- ther, treasure discovery is something that connects Tibet with the rest of humanity, it’s an experience Tibet shares with much of the world, including its immediate neighbours, India, Central Asia, Mongolia, and China. It is rather crucial to bear in mind that treasure cults are not unique to Tibet, because, as Davidson understood so well, the historical construction of Buddhist gter ma systems in renaissance Tibet can only properly be understood in terms of the mutual interactions between previously quite separate imported Buddhist, and indigenous Tibetan, treasure cultures.


Alternative source 2: Historical circumstances

Matthew Kapstein (2000) and Ronald Davidson (2005) have already introduced some key ideas to understanding the historical circumstances of the emergence of gter ma in the phyi dar, but a lot more remains to be done. Anthropologists of treasure recovery in different cultures and historical periods have described historical conditions which typically encourage the emergence of treasure recovery cults. These include periods of crisis, that give birth to affective narratives of popular historical consciousness lending ideological shape to the trajectories of future social renewal; a cosmology that supports the idea of treasure retrieval; and the actual existence of discoverable objects in the landscape. In relation to Tibet, we now know a little more about all of these than we did when Davidson produced his work.


While popular historical consciousness in the form of nostalgia for the lost Empire, as Davidson has mentioned, was undoubtedly a major theme connected with the emergence of gter ma discovery during the Tibetan renaissance, it also took additional forms beyond the dominant Buddhist one he describes. The emergent g.Yung drung Bon movement too was faced with the task of creating an entirely new scriptural canon that could enable traditions of non-Buddhist religion to adopt new organisationally lamaistic literary forms, and much of the work of 11th or 12th century Bon gter ston such as Khu tsha zla ‘od can be seen in this light. In the Tibetan phyi dar, we can thus see at least two contrasting sets of explicitly treasure-linked popular historical consciousnesses emerging, one typified by Myang ral’s seminal historical narratives, the other by similarly seminal Bon historical narratives such as the Gling grags text.


Many scholars have remarked that a key component in the proliferation of gter ma discoveries through the 11th and 12th century was the simple fact that old texts and suchlike were lying around in abandoned temples and other Imperial period structures, waiting to be found. As Dan Hirshberg (2016) reported, Myang ral described several of his treasure finds in precisely such terms. Cathy Cantwell has confirmed Myang ral’s claim in at least one instance, by finding that a

substantial text from the Phur pa section of his bDe gshegs ’dus pa is attested verbatim, with no variation, in the centuries older Dunhuang text IOL Tib J 331.III. Cantwell (forthcoming) has more recently found that further closely connected sections of the bDe gshegs ’dus pa are also quite likely verbatim reproductions of much older texts. So here we have convincing proof that some of the early gter ma were indeed quite simply older abandoned texts, that Myang ral had rediscovered.


But Buddhist texts and other valuables were not the only discoverables within the landscape of the Tibetan renaissance. In the background were also two further types of discoverable treasures, both potentially of considerable anthropological significance in the shaping of the Tibetan gter ma traditions, but neither of which have yet been considered. As gter ma discovery itself was later to become, both of these were in all probability from the start fundamentally integrat- ed into the cosmology of Tibetan ancestral regional deities, and it is also possible (we do not yet know many details) both contributed certain items of terminology and technical vocabulary to gter ma discovery. These two additional types of discoverables were minerals, such as gold, usually accessible through mining, and grave goods in old burial sites, accessible through plundering.

Virtually no published research is so far available on mining in ancient and pre-modern Tibet, although we do know from archaeo- logical sources of extensive gold mining in ancient Ladakh and West Tibet, and no doubt various other Tibetan regions too (Samten Karmay, for example, confirms a gold mine was located in his home region of Sharkhog).58 The existence of mining as a part of the local economy has been understood as a significant factor in a number of ethnographies of treasure recovery across the world, not least in Stewart’s descriptions of Naxos. In the case of Tibetan cultures, it is highly probable that mines and mineral deposits of all kinds were believed to come under the jurisdiction and protection of ancestral


58 Personal communication, 10th July 2019.


regional deities such as gzhi bdag, the same entities who were later said to be tamed by Padmasambhava to function as his gter srung. It is also of interest that terms such as gter kha can apply to mining and to gter ma recovery alike. Further research into the cultural construction and vocabulary of mining in Tibet might prove valuable towards an anthropological understanding of gter ma recovery.


Davidson (2005) proposed a hypothetical proto-gter ma tradition of burying items within royal tombs with the deliberate intention of later recovering them. This is not a foolish idea, since burials of important bodies and their grave goods can both be referred to as gter. But as we have seen above, scant further evidence has since been found to confirm Davidson’s idea. On the contrary, as we know from the more recent research of scholars such as Hazod, and also Sehnalova, and Langelaar, the cosmology of Tibetan ancestral regional deity beliefs would deem the removal of valuables from burial sites as an illicit or even hostile act. Yet

over and above the royal tombs, we now know that thousands more burial sites covered the Tibetan landscape, and virtually all of them have been looted for their treasures. Hazod has already produced an analysis of the plundering of the royal tombs based on traditional sources (Hazod 2016), which describe how widespread grave-robbing broke out during the civil unrest (kheng log), following the collapse of the Empire. But Tibetan historians tended only to provide a snapshot of the

plundering of the few nationally significant royal tombs, even though we know that grave-robbing was a much more widespread phenomenon, probably extending over much of the Tibetan cultural region over a period of some centuries at least. What is of particular significance to the emergence of the gter ma traditions is that this widespread tomb plundering must have formed a particularly prominent part of the recent or current historical background at precisely the period in which the

gter ma traditions first became prominent.59 Motives for plundering tombs might have been complex. While tumulus burial was not favoured by Buddhist authorities, Langelaar and Sehnalova’s recent fieldwork from Golog and other regions of East Tibet would suggest that ancestral burial sites have remained important to the cosmology of ancestral regional deity beliefs, and even with the de- cline in tumulus burial, important aspects of these traditions did not simply disappear across Tibet when Buddhism became dominant,

59 Hazod 2016 describes the plundering of the royal tombs as "the end of the tumulus burial tradition in Tibet". With the breakdown in political authority and their active discrediting by Buddhist authorities, the plundering of tombs probably become more generally widespread following the kheng log, in other words, leading up to the period when gter ma began to appear.


but continued to flourish in parallel. Tomb plundering could thus have arisen from a complex mix of religious fanaticism, socially sanc- tioned economic motives, illegitimate criminal motives, and also as an expression of clan or tribal hostilities seeking to destroy the ancestral bla gnas of rival groups (see Alternative source 5 below). Which- ever way, we know little about tomb plundering. Anthropologically speaking, as Stewart has emphasised, treasure discovery is often as- sociated with covert, semi-legal, illicit, or even criminal behaviour.60 What is of particular interest for an understanding of the emergence of the gter ma traditions is that as far as we can currently deduce, in terms of traditional cosmology, the wealth plundered from tombs seems to have been under the protection and ownership of precisely the same categories of ancestral local deities that were also appointed by Padmasambhava as his gter srung. Hence from an anthropological perspective, it might prove fruitful to understand more about how these contrasting forms of treasure discovery—tomb plundering and gter ma recovery—were differentiated and conceptualised.


Alternative source 3: Tibetan local and mountain deity traditions

This topic overlaps with the next one, and my colleague and collaborator, Dr Anna Sehnalova, is focusing on both. The Tibetan local deity traditions might have supplied the crucial context where, to quote Dan Hirshberg, ‘the Buddhist hand entered the Tibetan glove’. Because of its centrality in pre-Buddhist religion, and its ongoing importance in so many Tibetan communities, I currently feel it will be- come our single most important topic for research. The works of many scholars, including Diemberger,61 Pommaret,62 Mathes,63 Terrone,64 and Jacoby,65 offer powerful indications of how


60 A recent ethnography focusing on this aspect is Alice von Bierberstein’s (2017) account of illegal treasure hunting for hidden Armenian gold in contemporary Turkish Kurdistan. Stewart (2012) mentions several other examples.

61 Hildegard Diemberger, 1997. “The Hidden Valley of the Artemisia: On Himala- yan Communities and their Sacred Landscape” in A.W. Macdonald (ed.): Manda- la and Landscape. New-Delhi: D.K. Printworld

62 Francoise Pommaret, 1998. “ “Maîtres des Trésors” (gTer bdag): divinités locales et médiums au Bhoutan” in PIATS 7, Graz, Tibetan Mountain Deities, Their Cults and Representations. ed Anne-Marie Blondeau. Vienna 1998. Pommaret shows how gter ma protectors can also function as oracle deities who possess local mediums.

63 K-D Mathes, 2013. ‘Clouds of Offerings to Lady g.Yang ri. A Protector Practice by the First Yol mo sprul sku Shākya bzang po (15th/16th Cent.)”. In Franz-Karl


indigenous Tibetan local deity traditions of gnas bdag, gzhi bdag, yul lha, etc., with all their complex regional, social, political, ancestral, and identity implications, intersected historically with the below mentioned imported Buddhist ideas of text and material treasure revelation (see especially Alternative sources 6, 7, 8 and 10 below).

Following suggestions in such sources as gShen chen klu dga’s and Myang ral’s and Chos dbang’s biographies (respectively the earliest Bon and Buddhist gter stons for whom we have good documentation), one of our hypotheses to test is if and how indigenous local and mountain deities became identified as treasure guardian spirits corresponding to those mentioned in Buddhist sources such as the Pratyutpannasūtra and the Kriyātantras (see below Alternative sources 6 and 7 on Mahāyāna and Early Tantric Buddhism), but without losing their previous ritual, social, and political, functions. Such an identification would allow much of the indigenous

Tibetan local deity system to continue unabated, but now enriched with the additional role of guarding and dispensing gter ma to predestined gter stons. Linked to all of this is the role of such gzhi bdag as gter srung within tantric and monastic protector rituals, a subject not yet adequately explored. The many intersections of gter ma with the gzhi bdag cults throw up numerous interesting questions. For example, duly prophesied gterstons can, through the process of treasure withdrawal, enjoy a very direct relation with a regional deity from a region quite other than their own. For ordinary people, this is not so usual, and the symbolic meanings of this exception needs to be considered. For rNying ma gter stons, their status as representatives of Padmasambhava might play a key role in enabling it. Padmasambhava is often (and was already in a Dunhuang text) known as Padma rGyal po and is, almost uniquely among Tibetan guru figures, depicted in art with prominent aspects of kingly iconography. He has likewise acquired a prominent mythological dominance over all of Tibet's gzhi bdag that replicates significant features of the ancient Emperor's unique power relation with all of Tibet's gzhi bdag.66


Ehrhard and Petra Maurer (eds), Nepalica-Tibetica: Festgabe for Christoph Cüppers, Band 2, Andiast, IITBS.

64 Terrone, A, 2010. Bya rog prog zhu, The Raven Crest. The Life and Teaching of bDe chen ‘od gsal rdo rje, Treasure Revealer of Contemporary Tibet. PhD dissertation, Leiden.

65 Jacoby, S, 2014. Love and Liberation. Autobiographical Writings of the Tibetan Buddhist Visionary Sera Khandro. New York, Columbia University Press. 66 Political power in Tibet entailed a particular ritual relation with the gzhi bdag of the territory controlled. Hence, as Samten Karmay points out (1989, page 438), if

At other times, gter ma discovery might remain more local, and more susceptible to being subsumed within the local social and political system connected with a gzhi bdag within whose territory the gter ma has been buried.

Anna Sehnalova’s fieldwork in Golog describes how indigenous notions of local deities can contain an interesting circularity that resonates with Buddhist reincarnation beliefs: certain important local deities are ancestors, whom their human descendants can ascend to re-join post-mortem, but the deities themselves also continue to be reborn amongst their human descendants. Similar beliefs are reported in other regions. Sehnalova points out that these notions may have exerted some sideways influence on the development of the reincarnate lama system in Tibet. To this one must add, they also resonate nicely with the reincarnation model inherent to the gter ma tradition described by Myang ral, which, as Dan Hirshberg has pointed out, might have been the first to apply the idea of concatenated reincarnations.67


The intersections and interactions of these two traditions—on the one hand, the indigenous local deities with their complex role in local and national politics, ancestry, and clan lineage, and the many types of offerings buried for them as gter so that they will grant worldly wishes and enrich the environment—on the other hand, the Buddhist ideas of text and material gter ma burial for recovery—offer a rich field for historical analysis.

As Ratna Lingpa is quoted, ‘each great land will have a treasure,

…each minor land will have a treasure…’ (Dudjom 1991, page 935). We really need to fathom the historical implications of the regionality of gter mas and their guardians, something which has not yet been done for the early period of gter ma in Tibet.


a chieftain conquered a neighbouring territory, he would also have to begin politically appropriate ritual propitiation of its gzhi bdag, to consolidate his control. In this way, the Tibetan Emperor required a politically significant ritual relation with all of Tibet's great gzhi bdag, across his entire empire, while by contrast, regional chieftains required such a relation only with the gzhi bdag of the territories they directly controlled.


67 Hirshberg, Daniel. 2016, Remembering the Lotus-Born: Padmasambhava in the History of Tibet's Golden Age. Wisdom Publications, Somerville. As Dan Martin and I have suggested previously, the complex concatenated reincarnation model introduced by Myang ral for his seminal conceptualisation of the gter ston appears highly derivative of the Mahāyāna Pratyutpannasūtra (see below). As Kapstein (1989) points out, subsequent gter ma commentators made explicit and elaborated on this rNying ma debt to the Pratyutpannasūtra.


Alternative source 4: gTer burials for worldly needs

Anna Sehnalova's research shows that Tibet has abundant indigenous categories of gter, roughly speaking, items buried within the environment to ensure health, prosperity, and the like, usually as offerings to local deities or gzhi bdag, or more rarely simply to enrich the environment in some other way. Functionally, this overlaps with the enhancement of g.yang, or good fortune, by preserving suitable objects. A widely found basic notion is that whatever inheres to the landscape and is of some value, is the natural property of the gzhi bdag; and that making further offerings of valuables (gter) into the landscape will please the presiding gzhi bdag, who will in exchange ensure economic prosperity and general wellbeing. In this aspect of owning and receiving such offerings, in contemporary usage in Golog the gzhi bdag are often referred to as gter bdag (see Alternative Source 3 above).


It is important to emphasise that since the required research has not been done so far, there is not yet any evidence indicating when the word gter first began to be applied to such buried offerings and items. However, the contemporary dispersal of such ideas across wide areas of the Tibetan cultural regions, seems to indicate that the idea of the gzhi bdag owning all the treasures within its territory is an old one.

Across a wide range of the Tibetan cultural zones, gter offerings to the gzhi bdag can range from the ubiquitous treasure vases (gter bum) nowadays produced on an industrial scale, through grains, wool, semi-precious stones and gold, to the bodily remains of important people, and many more things besides. The related term gter kha also applies to naturally occurring minerals and the like, which the local deities similarly see as their own property (see Alternative Source 2 above).


gTer practices can acquire a Buddhist aspect and are done by monks and laity alike. For example, gter bum can take Buddhist- inspired physical shapes and be associated with Buddhist deities like Jambhala or White Mahākāla, and they can be buried by monks and lamas for the well-being of the world; but much of their underlying rationale is indigenous, linked to the notions of gzhi bdag, and the enhancement of personal and environmental wealth. Such gter is very rarely (if ever) buried with the idea it should later be dug up, and unlike the Buddhist nidhi/gter ma of Harrison’s Indian sources (see Alternative source 6 below), there are no prophecies of its future recovery or recoverers.


Anna Sehnalova has made a study of the numerous types of gter burials: wool, animal manes, animal body parts, grains, weapons, jewels, gold, silver, pearls, vases, minerals, medicines, many other worldly offerings, and also religious books and religious objects.68


Perhaps then the tradition of Padmasambhava burying caskets containing religious books and objects within the domain of a gzhi bdag, and entrusting them to its protection, was developed after the introduction of Buddhism, yet in some way building upon the already existing rich conceptual framework of the indigenous tradi- tions of burying this-worldly gter as offerings to the gzhi bdag in ex- change for worldly benefits.

But the Buddhist burials are conceived as removable (although only by a prophesied gter ston), to symbolise the transcendent power and superiority of Padmasambhava and his dharmic gter ma, over the mundane gzhi bdag and the mundane gter buried by ordinary persons. This could also invoke at a symbolic but also performative level the prominent Buddhist narratives of Padmasambhava ‘taming’ all of Tibet’s gzhi bdag. Yet this system still accords a very high degree of respect and authority to the Padmasambhava-tamed gzhi bdag, because only a prophesied gter ston, who has the express approval and support of the gzhi bdag, and no one else, can remove the treasure that Padmasambhava put there.


Indigenous ideas have thus interacted with Buddhist ideas in many interesting ways. For it is in the nature of local deities, to whom the riches of the earth belong by default, to develop a powerful proprietorial interest in anything buried in their territory. This includes the holy gter ma entrusted to them by Padmasambhava, since so long as it remains concealed in their territory within its caskets, it acts as an environmental enrichment par excellence. That is why gter stons, when recovering the gter ma prophesied for them by Guru Rinpoche, must for reasons of both courtesy and prudence, insert a replacement gter (gter tshab) of suitable value to compensate for the one they have just taken out.69


68 According to a small minority of her informants, only, religious books and religious objects are the only gter that can ever be removed again after burial, and only in specific and rather rare instances, so that here, as Sehnalova points out, insofar as such occurrences do exist, there might be rare cases where the concepts of gter and gter ma might blur. But most of her informants had no awareness that any gter could ever be removed after offering through burial. However, some gter bum are not buried, but remain on domestic or monastic altars, to increase the wellbeing and prosperity of the household or monastery: these, of course, can be respectfully moved, like many other objects on a shrine.


69 Some of the earliest work on these topics comes from Antonio Terrone, 2014, “The Earth as a Treasure: Visionary Revelation in Tibetan Buddhism and its In- teraction with the Environment”, in Journal for the Study of Religion, Nature, and Culture Volume 8, Number 4 Special Issue: Religion and Nature in Asia and the Himalayas. Guest Edited by Georgina Drew and Ashok Gurung.

Our challenge now is to track the development and distribution of such ideas in older sources.


===Alternative source 5: Pre-Buddhist / non-Buddhist burial traditions

In addition to the historical context I have already referred to (see Alternative source 2 above), in some regions of contemporary East Tibet, burials in the earth still continue to be used for some important deceased people and are often described as a kind of gter. Such burials have particular significance for local ancestor cults. Sehnalova in Golog and also Langelaar in Upper Khams have documented several such cases in their fieldwork, and I have heard possibly similar ac- counts involving multi-stage burial practices in Pemako.70 Langelaar has also studied a group of texts devoted to this topic by the 17th century Khams pa author Karma Chags med, which indicate that in his day, ancestor cult practices of this sort were quite widespread in his region.


Langelaar reports from contemporary Upper Khams that ancestor cults still flourish there. They are centred around vases containing the bones of deceased important persons, together with a srog shing and a bla rdo to act as the seat for their bla, which are interred along with valuable grave goods. The sites of such burials become dwelling places of the spirits of the ancestors (mtshun, pha mtshun). They are also key residences of the community’s life force or bla. They are of- ten

referred to as treasure (Langelaar notes that terms used include gter, sa gter, gter bum, dkar rtsi bum gter, rin chen gter, rin chen gter mdzod, yid gzhin gter bum, sa yi gter du sbas, etc.). Like the burials of gter vases etc for worldly needs described by Sehnalova (see Alternative source 4 above), they act to enhance life, harvests, fertility, health, and also grant success in warfare. Conversely, as Langelaar points out, their violation by enemies will weaken the tribe.71 Regular offerings must be made to gratify the pha mtshun, to ensure wellbeing and avoid the calamity of their displeasure.


70 Reinier Langelaar, lecture, Wolfson College, Oxford, 21st May 2019; A. Sehnalova, lectures, 4th May 2017, Oriental Institute, Oxford, and 21st May 2019, Wolfson College, Oxford; Lopon P Ogyan Tanzin, personal communication, Sarnath, December 2017

71 Langelaar mentioned that the Ge sar epic and also the rLangs poti bse ru likewise describe violating the enemy's burial grounds as a way of weakening their com- munal bla to facilitate defeating them in battle


In Golog, Sehnalova too reports that corpses of important persons are understood as gter. Their mode and place of burial plays an im- portant ritual role in the gzhi bdag cult (which is in this region also an ancestral cult), and also determines the location of the tribal bla gnas. Again, they enhance life, fertility, prosperity, good fortune, etc.


It is not yet clear how such contemporary practices might compare with earlier non-Buddhist tumulus burials from the Imperial period, but Langelaar and Hazod are currently working together on this question. If it transpires that they do resemble ancient practices, and given their explicit association with gter, learning more about them might add to our anthropological understanding of the historical context within which gter ma evolved.


However, there is no indication from Sehnalova and Langelaar’s fieldwork that anything is normally intended to be removed from such burials: on the contrary, both emphasise it is crucial everything should as far as possible remain hidden, undisturbed, and intact. In this sense, they resemble the offerings of gter to the gzhi bdag as described by Sehnalova (Alternative source 4 above). They do not normally resemble the extraction of religious treasures, characteristic of the Buddhist and gYung drung Bon gter ma systems. Davidson’s hypothesis of 2005 proposed that precious items imbued with royal bla were intended for deliberate post-mortem ritual removal, and that such a feature of deliberate removal might have been a predecessor of Buddhist gter ma, but this is not so far supported by contemporary ethnographies—quite the opposite.


Alternative source 6: Mahāyāna Buddhism

Thematic parallels between the systems of revelation described in various Mahāyāna texts and the Tibetan gter ma systems72 have not infrequently been remarked by Tibetan authors, for example, they are cited in the debate between Thu'u bkwan and Sum pa mkhan po as described by Kapstein (1989), and in other sources too. But, as many modern scholars might have wondered, if subjected to detailed ex-

72 I already wrote a short paper on this topic when a PhD student. Robert Mayer, 1994. “Scriptural Revelation in India and Tibet. Indian Precursors of the gTer ma Tradition”. In: Tibetan Studies, PIATS 6, Vol. 2. Institute for Comparative Research in Human Culture, Oslo, pp. 533-544. The paper was intended specifically as a response to Michael Aris's proposal that gter ma derived predominantly from Himalayan shamanism with no appreciable Buddhist content (1989, p. 59ff), and so was limited in scope to listing some of the more obvious Buddhist components.

amination, how close are these parallels in actuality, and what might they imply? And a further question: is there any evidence to suggest that the revelatory systems described in these Mahāyāna texts still existed in India by the 10th century, or had they by that time been displaced by other more tantric methods? Is it perhaps more likely that by the 10th century, such Mahāyāna ideas functioned predomi- nantly as literary resources useful for Tibetans as frame narratives in the construction of their own Buddhist revelatory systems? One of my tasks is to assess the evidence systematically and carefully and try to ascertain what we can deduce from it.


The Buddhologist Paul Harrison researched Indian Mahāyāna scriptural revelation for many years. He concluded that its various methods of revelation became summarised from a number of older sources in Śāntideva’s late 7th or 8th century Śikṣāsamuccaya. This work survives in Sanskrit and was translated into Tibetan in the 9th century. Śāntideva cited a text identified by Jens Uwe Hartmann and Paul Harrison as the Sarvapuṇya-samuccaya-samādhi-sūtra which outlines three broad and inclusive revelatory rubrics.73 The ideas inform-

73 For the three revelatory rubrics outlined in Śāntideva’s Śikṣāsamuccaya, citing the Sarvapuṇya-samuccaya-samādhi-sūtra, see Harrison, P. 2005. “Mediums and Mes- sages: Reflections on the Production of Mahāyāna Sūtras”, The Eastern Buddhist XXXV 1 & 2, pp.124-5; Harrison’s translation, my parentheses: “For, Vimalatejas, the Buddhas and Lords resident in other world systems show their faces to reverent and respectful bodhisattvas and mahāsattvas wanting the dharma, and they cause them to hear the dharma [similar to dag snang]. Vimalatejas, treasures of the dharma are deposited in the interiors of mountains, caves and trees for bodhisattvas and

mahāsattvas wanting the dharma, and endless dharma-teachings in book form come into their hands [dharmanidhāna, chos kyi gter, similar to sa gter]. Vimalatejas, deities who have seen former Buddhas provide bodhisattvas and mahāsattvas wanting the dharma with the inspired eloquence of Buddhas [[[pratibhāna]], spobs pa, similar to some dgongs gter, eg. the late bDudjoms Rinpoche’s Spu gri reg phung, his dgongs gter inspired within his mind by Ye shes mtsho rgyal].” Sanskrit text of Śikṣāsamuccaya, in Bendall (1897-1902, p.189): dharmakāmānāṃ hi vimalatejaḥ bodhisatvānāṃ mahāsatvānāṃ sagauravāṇāṃ sa- pratīśānāṃ (sic) anyalokadhātusthitā api buddhā bhagavanto mukham upadarśayanti dharmaṃ cānuśrāvayanti | dharmakāmānāṃ vimalatejaḥ bodhisatvānāṃ mahāsatvānāṃ parvatakandaravṛkṣamadhyeṣu dharmanidhānāni nikṣiptāni | dharmamukhāny anantāni pustakagatāni karatalagatāni bhavanti | dharmakāmānāṃ vimalatejaḥ bodhisatvānāṃ purvabuddhadarśinyo devatā buddhapratibhānam upasaṃharanti ||. Tibetan text of the Sarvapuṇya-samuccaya-samādhi-sūtra / 'Phags pa bsod nams thams cad bsdus pa'i ting nge 'dzin ces bya ba theg pa chen po'i mdo/ from Lithang Kanjur, mdo sde, Na 77a1-131a6 (vol. 51, folio 105b): dri ma med pa'i gzi brjid chos 'dod pa'i byang chub sems dpa' sems dpa' chen po bsam pa phun sum tshogs pa gus pa dang bcas pa rnams ni 'jig rten gyi khams gzhan na 'dug kyang sangs rgyas bcom ldan 'das rnams zhal ston par mdzad cing chos kyang thos par mdzad do/ /dri ma med pa'i gzi brjid chos 'dod pa'i byang chub sems dpa' sems dpa' chen po rnams kyi chos kyi gter ri dang / ri sul dang / shing dag gi nang du bcug pa dag yod de/ gzungs dang / chos kyi sgo mtha' yas pa glegs

ing these three broad rubrics are quite old in Mahāyāna Buddhism and are also discernible in the works of early text revealers such as Myang ral and Chos dbang, albeit under varying terminologies. It seems to have been only at a much later stage that an exact form of words for these three broad rubrics became terminologically norma- tive for the rNying ma (dag snang, sa gter, and dgongs gter). Śāntide- va’s three rubrics of revelatory methods are:

Firstly, meeting the Buddha face to face in a vision and receiv- ing teachings (reminiscent of the later Tibetan Pure Vision or dag snang). Secondly, dharmanidhāna, or dharma treasures concealed with- in the material world (reminiscent of the later Tibetan Earth Treasure or sa gter). Thirdly, pratibhāna or direct inspiration of mind (spobs pa, rem- iniscent of the later Tibetan Mind Treasure or dgongs gter).74

The first two are also discussed in detail in the Pratyutpanna-buddha- saṃmukhāvasthita-samādhi-sūtra (henceforth PraS), a text which Harri- son has studied in very great detail, and which was translated into Tibetan from Sanskrit within the same historical period as the Śikṣāsamuccaya. The title of this sūtra can be translated as ‘The Samādhi of Direct Encounter with the Buddhas of the Present’. Later Tibetan Treasure conventions show a particularly close and detailed mapping to it at many significant points.

As its name suggests, much of this lengthy text deals with pure vi- sion, but I don’t have time or space in this introductory article to compare in detail the slightly amorphous Tibetan category of dag snang with the descriptions of scriptural revelation through direct visionary encounter with Amitābha in the PraS.

bam du byas pa dag kyang lag tu 'ong bar 'gyur ro/ /dri ma med pa'i gzi brjid chos 'dod pa'i byang chub sems dpa' sems dpa' chen po rnams la ni sngon sangs rgyas mthong ba'i lha rnams spobs pa sgrub par byed do/

74 The Sanskrit term pratibhāna is translated into Tibetan as spobs pa. The most com- mon application of the term spobs pa in later rNying ma thinking was the spobs pa gter chen brgyad, a Buddhist doctrinal category with which ‘Jigs med gling pa be- came associated, so that he is sometimes styled in inscriptions and so forth as spobs pa’i gter chen brgyad grol ‘jigs med gling. ‘Jigs med gling pa is also thought to be among the first major exponents of the term dgongs gter, although the concept in a broader sense is very much older, e.g. in Chos dbang’s usage of thugs gter, and, indeed, in Śāntideva’s citation from the Sarvapuṇya-samuccaya-samādhi-sūtra. ‘Jigs med gling pa’s most famous treasure, the Klong chen snying thig cycle, is of- ten said to have been revealed through the process of dgongs gter. It would be in- teresting to research these themes further.

However, I should briefly list the following characteristics which are shared polythetically by this probably 1st or 2nd century Indian Mahāyāna text, and the very much later Tibetan category of Earth Treasure (sa gter).75 The Tibetan Earth Treasure tradition has a very distinctive and highly complex structure, a structure which it shares on most points of detail with the narrative presented in the PraS. We do not yet have any clear understanding of how or why Tibetan sa gter seems so closely to follow the precedents presented in the PraS Chapter 13, but what does seem apparent is that their relation is somehow a cognate one (in what follows, all text locations follow Harrison, P. The Samādhi of Direct Encounter with the Buddhas of the Present: An Annotated English Translation of the Tibetan Version of the Pratyutpanna-Buddha-Saṃmukhāvasthita-Samādhi-Sūtra, Tokyo, 1990; and The Tibetan text of the Pratyutpanna-Buddha-Saṃmukhāvasthita- Samādhi-Sūtra, Tokyo, 1978).

The rationale for burying teachings, as a hedge against future times of religious decline (13 B-D) Precise prophecies of the future discoverers and of their dis- coveries (13 K, esp. vv. 3-11, 14-15) The eight future discoverers will be reincarnated direct, named, disciples of the Buddha (13; 13 K vv. 3-11) The Buddha first directly teaches and then entrusts (gtad) these named disciples with the teachings they will rediscover again and again in repeated future lives (13, 13 H)


White-robed lay status of the eight, future, treasure recoverers (13 C, 13 E; Harrison 1990: xvii) Five hundred prophesied recipients, who also heard the orig- inal teaching in their past lives, will repeatedly be reborn to accompany the eight prophesied treasure discoverers, to re- ceive, copy and propagate the newly revealed teachings (13 G-H, 13 K v. 3, 14-15) (cf. chos bdag in Tibetan gter ma) Treasure texts are stored in caskets (sgrom bu) (13K v. 8) Treasure caskets are hidden in such places as caves, stūpas, rocks and mountains (13 K v.9; 13 B) The hidden Treasure is protected during its long concealment by being put in the care of local guardian deities or nāgas (13 K v.9)


75 The PraS was translated into Chinese by 178 CE, and seems also to have had an influence upon modes of Buddhist scriptural revelation in China, but that is not my immediate concern here.

It is predicted that after revelation, Treasure texts and their discoverers might be seen as controversial, and not readily ac- cepted by the wider Buddhist public (13 F, 13 K vv. 12-13) The Tibetan translation of PraS shares some key technical terms with the rNying ma Treasure tradition: gtad pa (13 H), sgrom bu (13 K v. 8) There are also verses amenable to interpretation as prophecies of Tibet and its gter ma (13 K v.14-15), and of Padmasambhava (13 K v.20, 22)

The above resemblances shared between the Mahāyāna sources re- searched by Harrison, and the later Tibetan gter ma traditions, need to be researched further, not least because several later Tibetan apolo- gists cite these Mahāyāna texts in relation to their own gter ma traditions.76


Yet rDo grub chen 'Jigs med bstan pa'i nyi ma in his influential gTer gyi rnam bshad somewhat downplayed the similarities between sūtra texts in general and the later Tibetan gter ma tradition.77 Perhaps this was quite simply because, unlike gter ma, they were not predom- inantly tantric. But another likely factor is that the PraS, as far as we currently know, is the only sūtra text to quite explicitly describe in any great detail an equivalent to the ‘Mind mandate transmission’ (gtad rgya), a component 'Jigs med bstan pa'i nyi ma felt was indis- pensable to Tibetan gter ma; hence, as far as he was concerned, with only one solitary exposition of the important component of gtad rgya, sūtra texts as a generality could represent only partial antecedents for Tibetan gter ma practices. Janet Gyatso has possibly over-interpreted 'Jigs med bstan pa'i nyi ma on this point, perhaps thinking he is even denying that the PraS describes a ‘Mind mandate transmission’, and this in turn might have coloured her overall approach to the history of gter ma.78 My own

76 Dan Hirshberg has argued that Myang ral was the first major author in Tibetan Buddhism to describe concatenated reincarnations, and that he did so specifically in relation to gter ma recovery. Although I am not aware that Myang ral any- where refers to the PraS by name, his understanding of the roles of prophecy and reincarnation in gter ma recovery closely follows the template outlined in the PraS. We must therefore enquire if Mahāyāna texts such as the PraS might have exerted some indirect influence on the development of the reincarnate lama systems of Tibet.


77 'Jigs med bstan pa'i nyi ma, gTer brgyud kyi rnam bshad. Pages 797-858 of Vol Kha of the Shechen edition of the Rin chen gter mdzod chen mo. New Delhi: Shechen Publications, 2007-2008. TBRC W1KG14. 2: 813 - 874. 78 Gyatso, Janet, 1998. Apparitions of the Self. Princeton University Press, Princeton. See page 150, note 24, on page 294, where she writes: ‘but note Do Drubchen III’s

reading of the Tibetan agrees with Tulku Thondup and Harold Tal- bott’s translation of 1986:79 I think 'Jigs med bstan pa'i nyi ma is in- deed saying that not ALL sūtra sources on gter ma have the gtad rgya, but he is not specifically denying that the PraS does so.80 Either way, in this particular case, modern philology has earned the last word. Paul Harrison’s decade-long minutely detailed philo- logical analysis of this text, encompassing all extant versions and fragments in Sanskrit, Chinese, and Tibetan, shows that the long and complex narrative within chapter 13 of the Tibetan version of the PraS, does indeed describe an extremely close equivalent to gtad rgya. Not only is a similar meaning conveyed, as the Buddha first entrusts the sūtra to certain named disciples before it is buried for hundreds of years for them to recover and propagate, as he prophecies, in their future reincarnations; but even the word gtad is used appropriately.81 Likewise the PraS does the same for most other basic structural fea- tures of the complex Tibetan sa gter tradition, including the prophetic ones.

Davidson seems not to have been very familiar with Harrison’s work, because he makes no direct mention of any of it in his account of the origins of gter ma. Admitting only the vaguest or most general similarities between Indian and Tibetan revelation, but nothing that is detailed or cognate, Davidson argues there was never anything in India resembling Tibetan gter ma, and that the three rubrics of Pure Vision, Earth Treasure and Mind Treasure were late Tibetan inven- tions with no Indian precedents (Davidson 2005: 213). While I am happy to concur without hesitation that the Tibetan scholastic adop- tion of the exact terminology dag snang, sa gter, and dgongs gter prob- ably came comparatively late, Harrison and Hartmann’s work on the Śikṣāsamuccaya and the Sarvapuṇya-samuccaya-samādhi-sūtra demon- strates that their underlying ideas reflect a much older Buddhist pat- tern, transmitted to Tibet through these Mahāyāna texts, and which seems to have influenced those who constructed the Tibetan sa gter

point that they often lack the distinctive mark of Treasure, namely, the intention of the concealer, who commissions a particular individual to disclose the Treasure at a particu- lar time (Thondup 1986, pp. 109-10).’ 79 Tulku Thondup Rinpoche, 1986. Hidden Teachings of Tibet. Wisdom Publications, London. 80 'Jigs med bstan pa'i nyi ma, ibid, page 806: …..mtha’ dag gtad rgya sngon song can du ‘chad pa’i shes byed dka’ ‘am snyam mod /; Thondup and Talbott page 110: “But I think it is difficult to say for all of them [the previously listed sūtras that mention gter ma] that there is any proof that those teachings came through a Mind-mandate Transmission”.

81 bcom ldan ‘das/ ting nge ‘dzin ‘di skyes bu dam pa ‘di dag la gtad par gsol/ See Harrison 1978 pp 102-115, and Harrison 1990 pp 96-113.

system, a consideration which Davidson did not discuss. Not only that, but I think one can cogently argue that the PraS does indeed describe something closely resembling the complex and highly dis- tinctive structure of Tibetan sa gter. Whether any such system ever existed as an actual practice in India, or only as a literary trope, is currently unknown, and remains to be determined. Perhaps influenced by Davidson as well as by 'Jigs med bstan pa'i nyi ma, it seems to me that Gyatso too has sometimes given Harri- son’s work insufficient weighting. For while she does seem aware of its existence, 82 she has not apparently absorbed its implications. Hence she sometimes seems simply unaware of the evidence turned up by Harrison (and Hartmann), the implications of which are that Tibetan conceptualisations of revelation do indeed seem to contain important elements that are directly cognate with older Indian Bud- dhist literary forms at the least.


Alternative source 7: Early Tantric Buddhism

If a few Mahāyāna sources describe hidden textual treasures, a great many early tantric texts dwell on more material hidden treasures, including riches, magical objects, spells and rituals, and medicines that confer siddhis, immortality, longevity, or health.83 Such treasure cults are very typical of the Purāṇas and early Buddhist tantras, and are often connected with pātālā, the sensual underground heaven of asuras and nāgas, accessible to brave yogis via special Asura caves on the surface.84 Consorting with asura females can be a key for human yogis to obtain treasures. Chinese Buddhism knows these themes, in


82 Gyatso, Janet, 1998. Apparitions of the Self. Princeton University Press, Princeton. See pages 147 onwards, especially pages 150 and 152, and note 24 on page 150. 83 There are innumarable rites for finding nidhi or nidhana within early Buddhist tantric literature. Locations include: Padmacintāmaṇi-dhāraṇī sūtra translated into Chinese by Bodhiruci in 709 (Taishō 1080.20, p. 190a.); Vajrakumāratantra in Chinese translation (Taishō 1222) (translated excerpts by S. Hodge in Mayer 2007); Vajraśekhara / Sarvatathāgatatattvasaṃgraha, Amoghavajra’s Chinese translation (Taishō 865); Mañjuśrīyamūlakalpa, especially in its final chapter, the hemasādhana; Amoghapāśakalparāja Tibetan: sDe dge 686, Vol. 92-1-138a, Sanskrit: Taisho University Mikkyō Seiten Kenkyūkai, codex unicus from China, 73a 5-6; Vidyottama- tantra sDe dge D746, Volume 95 1a-237b; at least one of the two Vajrapātālatantras in the Tibetan canon, sDe dge 744; etc etc. 84 See Robert Mayer, 2007. ‘The Importance of the Underworlds: AsurasCaves in Buddhism, and Some Other Themes in Early Buddhist Tantras Reminiscent of the Later Padmasambhava Legends’, in Journal of the International Association of Tibet- an Studies, no. 3 (December 2007): 1-31.

tantric texts such as the *Kaṇikrodha-vajrakumāra-bodhisattva-sādhana- vidhi (Sheng jia ni fen nu jin gang tong zi pu sa cheng jiu yi gui jing, 聖迦抳忿怒金剛童子菩薩成就儀軌經) (T1222a).85 Likewise Xuanzang men- tions Bhāviveka gaining entrance to the Asura kingdoms using similar procedures: Vajrapān is invoked, mantras and mustard seeds are thrown at a rock face to make it open, a crowd watches the whole operation, and some of them accompany Bhāviveka into the cave.86 Comparable themes are also found in Tibetan gter ma narratives, such as Myang ral’s biographies,87 and in some crowd gter ma (khrom gter) narratives.

I have for some years suspected that a potentially important re- source for understanding the origins of gter ma in Tibet lies in the dozens of treasure rites within the Kriyātantras listed in the lHan dkar ma catalogue, which were translated into Tibetan in the late eighth century.88 Behind these Kriyātantra texts lies a rich hinterland of Purāṇic and other Indian treasure lore, but it is primarily through the translated Kriyātantra texts that such ideas entered Tibet. Thus I have recently begun to browse the Kriyātantra section of the lHan dkar ma, starting from the beginning. So far, I have only looked at the first two. The first Kriyātantra text in the lHan dkar ma is the Amoghapāśa- kalparāja (lHan dkar ma 316, D686, sDe dge 92 1b-316a, 93 1b-57b). Fortunately, a Sanskrit version also survives. The Amoghapāśakalparāja has eleven rites for finding nidhi / gter, including for example, one dedicated to taming the gter srung or nidhipāla that I present here in this footnote.89 All of them, however, are quite brief, and tell us little.

85 For some discussion and translations of passages on treasure recovery in this text, see Mayer 2007, ibid. A Chinese etext is available from CBETA here: http://tripitaka.cbeta.org/T21n1222a

86 Mayer 2007 ibid, page 11, citing Rolf Stein, Grottes-Matrices et Lieux Saints, page 26. See also Malcolm David Eckel, 1994. To See the Buddha: A Philosopher’s Quest for the Meaning of Emptiness, Princeton, Princeton University Press. Pages 11-13. 87 Hirshberg 2016 ibid, page 115

88 Since there is some possibility that some of the tantric texts listed in the extant version of the 'Phang thang ma might have been added to that catalogue at a later date, initially at least, I am restricting my investigation to texts listed in the lHan dkar ma.


89 Amoghapāśakalparāja (lHan dkar ma 316), Tibetan: sDe dge 686, Vol. 92-1-138a, Sanskrit: Taisho University Mikkyō Seiten Kenkyūkai, codex unicus from China, 73a 5-6. “Then, if you wish to dig up some treasure, at the place where you sus- pect the treasure is, make a maṇḍala of cow dung; strew it with flowers, cense it with guggul incense, offer the three kinds of tormas, that is, offer pure tormas, meat and blood tormas, and dough, fruit and lotus tormas. Offer them to Ārya Avalokiteśvara. Do ten thousand recitations. When the ten thousand recitations

The second Kriyātantra text in the lHan dkar ma is the Vidyottamatantra (lHan dkar ma 317, D 746, sDe dge 95 1a-237b)90. I am not aware of any Sanskrit survivals, and its provenance has not yet been subjected to systematic analysis. The text is very long, and I have only browsed a few sections of it, but those I have looked at all appear to be Indian. They do not appear, like the rNying ma'i rgyud 'bum texts I have previously studied, to be Tibetan-composed or compiled in any way. Nor, according to Rolf Giebel, is the existence of this text known from Chinese sources at all.91


The Vidyottamatantra has a greater number of gter related rituals than the Amoghapāśakalparāja. Fortunately, these include a long and highly informative section in over twenty pages (sDe dge Kangyur vol 95, folio 70b ff), on how to locate the treasure doors (gter sgo) be- hind which gter is hidden, and how to open them to extract the treas- ure, which I will discuss here briefly.

According to this passage, treasure seekers cannot find the gter sgo unless deities reveal its location in a dream or vision: one is reminded here of Charles Stewart's anthropological observations (Stewart 2012), and this is also a standard feature in later Tibetan gter ma dis- covery. In the later Tibetan gter ma system, the location of the gter sgo are completed, there at the place which has the treasure, the treasure guardian will actually become present, and the treasure will be made to appear. Offer a torma and drinking water to the treasure protector, and as long as you live, the treasure protector will do work for you. Wherever you send it, whatever work you command it, all will be done.”

/de nas gter brko bar 'dod pas gnas gang na gter yod par dogs pa'i gnas der ba'i lci bas maṇḍal byas la/ me tog gtor zhing gu gul gyi bdug pas bdug cing / gtor ma gtsang ma dang / sha dang / khrag gi gtor ma dang / phye zan dang 'bras bu dang / pad+ma'i gtor ma dang / gtor ma rnam pa gsum sbyin par bya'o/ /'phags pa spyan ras gzigs dbang phyug la mchod pa bya'o/ /stong phrag bcu bzlas brjod bya'o/ /stong phrag bcu tshang ba dang gnas gang na gter yod pa'i gnas der gter srung ba mngon du 'byung zhing gter 'byung bar 'gyur ro/ /gter srung ba la gtor ma dang / mchod yon sbyin par byas na gter srung ba de ji srid 'tsho'i bar du las byas pa por 'gyur ro/ /gang du mngags pa dang / gang bsgo ba'i las de thams cad byed par 'gyur ro/ /. atha nidhānam utpāṭayitukāmena yatra sthāne nidhisaṃkā bhavati / tatra sthāne gomaya- maṇḍalakaṃ kṛtvā puṣpāvakīrṇṇaṃ kṛtvā gugguladhūpan datvāt tṛvidhibaliṃ dadyāt śuklabali māmsarudhirabaliḥ saktuphalapadmabalin dātavyaṃ / āryāvalokiteśvarapūjāṃ karttavyaḥ / daśasahasrāṇi japatāḥ samāpte daśasahasrāṇi yatra sthāne nidhir bhaviṣyati


/ tatra nidhipālam uttiṣṭhati / nidhānam utpatati / nidhipālā bali.arghaṃ nivedayitavyaṃ

/ sa ca nidhipāla yāvajjīva karmakārakā bhavati yatra preṣayasi yam ājñāpayasi tat sarvvakarmāṇi kariṣyati /

90 So far, I have largely consulted the sDe dge version, but Hermann-Pfandt (2008) notes that the text also occurs at the following locations: Cone C 407, Phug brag F 531, lHa sa H 691, 'Jang sa tham J 713, London L 576, Mustang M 113, Narthang N 653, Peking Q 402, Stog S 696, Tokyo T 693, Ulan Bator U 766. It is also listed in the ‘Phang thang ma (923).

91 Rolf Giebel, Personal communication, 1st August, 2019.

and much practical information about it is disclosed through divine- ly-dispensed kha byang, or prophetic guides. It is noteworthy that this passage from the Vidyottamatantra addresses many of the practical concerns typical of kha byang: where to find, how to make visible, and how to recognise hidden gter sgo, when (or when not) to open them, what rituals to do before, while, or after opening them, how to man- age their protectors or gter srung (the same term is used in both traditions), at what depth behind the gter sgo the treasure will be found, how much will be found, how the gter sgo should be reclosed after the gter is extracted, and so on.

A complex typology of gter sgo is described, with different shapes and colours. Equally important of course are the methods to find them and open them. These include the natural arising of specific sounds that will indicate the precise location of the gter sgo, the performance of various homas in preparation for opening them and for pacifying obstacles, magical methods to make the gter sgo door out- line and door handles appear, ways of removing the kīla or phur pa that is keeping a treasure door locked shut, and so forth. Several pages are dedicated to classifying and describing the fearsome protector deities who have been assigned by Brahmā to prevent the treasure doors being improperly opened, and the specific methods by which the treasure seeker can placate them, before they attack him.


As far as Anna Sehnalova is aware, neither the term gter sgo nor its underlying concept occur in the indigenous Tibetan traditions of burying gter for environmental enhancement or as offerings to the gzhi bdag; nor has Reinier Langelaar reported them from the pha mtshun burial cults. According to our current knowledge, it actually seems quite difficult to try to account for the origin of the idea of a gter sgo from indigenous Tibetan cosmologies alone. Yet several ele- ments concerning treasure doors from this long and complex section of the Vidyottamatantra do resonate closely with later Tibetan Bud- dhist and Bon gter ma practices, and the term gter sgo and its underling concept plays an equally fundamental role in them all. Hence a more detailed study of the emergence of the idea of the gter sgo with- in the Tibetan gter ma traditions seems desirable, and the Vidyottama- tantra seems a good place to start.


The gter sgo in both traditions can often resemble the portal to an- other dimension. In Kriyātantras, this frequently pertains to Indian mythology of biladvāra, Asura caves, pātālas, etc. Thus in the Vidyottama, in a description typical of several other Kriyātantra treas- ure narratives too, once the gter sgo is open, gandharvīs and kinnarīs will appear from the magical realm of Meru, which is just the other side of the gter sgo. Similarly, when gTer bdag gLing pa enters a gter sgo, he finds himself inside a tent-like cave with crystal walls and

bright frescoes, inhabited by supernatural young men and women (Tulku Thondup 1986: 78). When Pema Lingpa enters a gter sgo, he finds himself within a large space with thrones and supernatural in- habitants (Aris 1989: 38).92

In the Vidyottamatantra, gter sgo sometimes open spontaneously, but other times iron bars are needed to smash them open. Similarly, Tibetan gter sgo sometimes open spontaneously, for example when Pema Lingpa withdrew his famous lake treasure, but other times, hammers and chisels are needed, for example, when Pema Lingpa withdrew his cliff treasure at Gedo (Aris 1989: 49). Likewise, a gter sgo opened spontaneously for Dudjom Lingpa at Ba ter, but he had to smash one open with a chisel in Nga la Tag tse (Tulku Thondup 1986: 78, 79).


I have not yet had time to browse any of the further Kriyātantras listed in the lHan dkar ma. But we can see that some Tibetans soon appropriated the various magical rites contained in such Kriyātantra texts, since many are reproduced in a probably tenth century Dunhuang Tibetan compendium of useful magical rituals (IOL Tib J 401), along the lines of the later beʼu ʼbum genre. This text includes a rite to the Indian goddess Bhṛkuṭī for locating a treasure door (gter sgo) on the top of a mountain, making the treasure door open up, and inducing its treasure protectors, here called gter bdag, to give one the treasure.93 Two aspects are of particular interest here (see Alternative Sources 3 and 4 above): mountain tops are, in indigenous Tibetan thinking, a typical abode of the ubiquitous gzhi bdag deities; and as Anna Sehnalova explains,

The term gter bdag is commonly given to gzhi bdag deities who are be- lieved to have treasures in their territories. In my understanding, these can be both [local deity this-worldly offering] gter and/or [[[Buddhist]] and


92 One is also reminded here, of the slightly later central myth of Buddhist Yogatan- tra, its key origin narrative of the Iron Stūpa, preserved in the Far East but proba- bly from Indian sources. For after entering the Iron Stūpa, the yogin, who is sometimes nameless and sometimes identified as Nāgārjuna, finds himself within an alternative dimension, a magical space, inhabited by divine beings, who re- veal to him the Yogatantra scriptures. 93 See IOL Tib J 401, Section 5, 7r.7. For an additional rite focused on Yamāntaka, which promises the finding of treasure as one of its siddhis, see IOL Tib J 401, 22v-23r. Thanks to Sam van Schaik for this information, described in his Aris Lec- ture at Wolfson College, Oxford, on 15th November 2018. The goddess Bhṛkuṭī is often associated with Avalokiteśvara, notably in his form of Amoghapāśa, which is often classified as Kriyātantra. Bhṛkuṭī is also identified with Tārā.

Bon religious] gter ma94 treasures, in both instances the gzhi bdag is called gter bdag (personal communication, 31st October 2018).

What is of interest here is the Tibetan selection and reception of such Indian treasure rites, and the ways in which Tibetans might have understood them, to appropriate them and their terminology, apparently putting them in some kind of relation to their own indigenous gter categories. But assessing how Tibetan gter ma practices might have been influenced by Kriyātantra sources is quite complex. Con- sider, for example, the idea that gter-owning indigenous gzhi bdag were tamed by Padmasambhava, made into his servants, and ap- pointed as protectors of his Buddhist gter ma. The notion of taming a gzhi bdag seems alien to pre-Buddhist tradition, where gzhi bdag pre- side as powerful patrons and humans are their repectful clients. It is so alien to indigenous thinking that even today, no lama (however great) would normally treat the still widely revered gzhi bdag as mere servants, and their taming to the Buddhist cause is commonly at- tributed only to such unique figures as Padmasambhava, or Mi- larepa, in the distant past. But behind the myth of Padmasambhava taming Tibet’s gzhi bdag, one can perhaps detect the typical Kriyātan- tra trope of the adept taming the treasure protectors and making them his obedient servants in perpetuity. Binding the treasure protec- tors into humble servitude is indeed very prominent in Kriyātantras, one of the main benefits of their nidhi rites.


Much of the Vidyottamatantra does NOT enter the Tibetan gter ma tradition: huge bird demons guarding the treasure, a prominent reli- ance on compounded ointments and power substances to open treasure doors, an exclusive focus on material wealth. And Tibetan gter ma has many features NOT found in the Vidyottamatantra: sym- bolic languages, reincarnated gter stons, consorts, elaborate treasure caskets, emphasis on religious texts as the found treasure, etc. But as Davidson pointed out, gter ma emerged from heterogeneous sources, both indigenous and foreign, fused together in the unique social- historical conditions of renaissance Tibet. Our question is not if Ti- betan gter ma merely continues Kriyātantra: rather, we are interested in which elements from Kriyātantra might have entered the mix of Tibetan gter ma.

94 Anna Sehnalova’s usage here reflects the colloquial usage of her Golog inform- ants, where gter is used as a shorthand to indicate the widespread offerings gen- erally made by local people to their local deities for this-worldly purposes, or other such valuables belonging to the gzhi bdag, and gter ma is used as a short- hand to indicate the rarer prophesied hidden treasures (sacred texts etc.) pertain- ing to the wider Buddhist or Bon religions.

Several additional Kriyātantra texts containing treasure finding rites were also translated in the phyi dar, for example, the above- mentioned Vajrapātālatantra (D 744), or the Mañjuśrīyamūlakalpa which has fourteen different treasure rites in its final chapter alone,95 and some in other chapters too (although Wiesiek Mical advises me that the final chapter, the hemasādhana, was not translated into Tibet- an), but they too remain to be studied.

The Imperial period Kriyātantra translations, with their numerous passages on nidhi or gter, seem to have remained until now unex- plored by Western scholars of gter ma. Yet even the most cursory reading shows that they contain evidence for a highly complex and well-developed Indian cult of treasure discovery, some of which ap- pears familiar to those who study the Tibetan gter ma traditions. Quite how these Kriyātantra texts impacted the (probably) 10th cen- tury construction of the Tibetan Buddhist and g.Yung drung Bon gter ma cultures, remains to be examined. But if we seek a better under- standing, we probably don’t have the option to simply ignore the testimony of these Kriyātantras, as has largely been done up to now.

Alternative source 8: the Mahāvairocanatantra, the Yogatantras, and Chinese traditions

Also listed in the lHan dkar ma are a number of tantric scriptures that are thought to represent the phases in Indian tantric develop- ment following after the Kriyātantras. Among these are the Mahāvai- rocanābhisaṃbodhi (lHan dkar ma 321) and its piṇḍārtha (lHan dkar ma 322); and the Sarvadurgatipariśodhana (lHan dkar ma 323) and its vārtik- ka (lHan dkar ma 324). Weinberger (2003: 292) believes the Sar- vatathāgatatattvasaṃgraha was also translated in the Imperial period, although it is not listed in the lHan dkar ma.

We know several of these texts played a prominent role in the offi- cial Buddhist ceremony and doctrine of the Tibetan Empire. What is perhaps less well known is that through the same particular period, these very same originally South Indian traditions were (temporarily) playing an equally or even greater official role within the Chinese



95 See Einoo, Shingo, ‘From kāmas to siddhis — Tendencies in the Development of Ritual towards Tantrism’, in Einoo, Shingo, ed., Genesis and Development of Tan- trism, Tokyo, Institute of Oriental Culture Special Series, 23, University of Tokyo. (see especially pages p. 34-35).

Empire.96 Moreover, the Dunhuang region was at the time arguably the major centre for the teaching and practice of these texts within China, since Amoghavajra (705–774), their great Indian-Sogdian translator and master, had been particularly active in the Hexi corri- dor, at the request of the local military governor, who valued his rit- uals for their efficacy in battle (Yang 2018: 41 ff.).

But even during Amoghavajra’s lifetime, this region began to be progressively overrun by the Tibetans, who remained in control there until 851. It is therefore entirely possible that the early transmission of these traditions into Tibet were influenced by the recently- established and (at the time) hugely prestigious Chinese lineages they found at Dunhuang. In an important forthcoming article, Yury Khokhlov presents new art historical evidence for what he believes is a Chinese background to some of these early teachings in Tibet (Khokhlov forthcoming).


It might take some time before these issues are fully clarified, but if there were indeed such Chinese influences, they might have im- pacted Tibetan attitudes to scriptural revelation. For central to Amo- ghavajra’s understanding of tantra was an important revelation nar- rative, describing the first ever transmission of the Sarvatathāgata- tattvasaṃgraha, from inside a miraculous Iron Stūpa in South India, to an unnamed sage. In some respects, this Iron Stūpa resembles the biladvāra of the Kriyātantras described above. The sage had to open it by reciting mantras, circumambulating, and throwing mustard seeds, but once inside, he found himself within a huge celestial space inhab- ited by numberless divine beings, who could transmit to him the pre- cious teachings. This myth of scriptural revelation within the Iron Stūpa assumed great significance at that time in some interpretations of Amoghavajra’s tradition. The founder of the Shingon school, Kūkai (774-835), for example, made it the central spiritual metaphor of his entire interpretation of Amoghavajra’s teachings, and also identified the unnamed sage as Nāgārjuna.97

Other Chinese Buddhist texts mention systems of revelation that seem to draw upon such Indian sources as the PraS. Michel Strick- mann made a study of an apocryphal 6th century Chinese Buddhist

96 Yang, Zeng. (2018). A Biographical Study on Bukong 不空 (aka. Amoghavajra, 705- 774). Networks, Institutions, and Identities. Unpublished PhD thesis. The University of British Columbia.

97 Rambelli, F., "The Myth of the South Indian Iron Stupa (Nanten tettō) in Japanese Esoteric Buddhism", in International Journal of Buddhist Thought & Culture Vol. 27. No. 2 (December 2017): 71–88.

text, the Consecration Sūtra (Kuan-ting ching, T1331).98 It has an origin myth involving the Buddha’s prophecy of the rebirth of its future discoverer, who was indeed reborn to find it within a jewelled casket within a cave in China (Strickmann 1990, p. 86). Hence Strickmann long ago suggested it shared some cognate relation with Tibetan gter ma. Likewise, as mentioned above, there are several Chinese tantric Buddhist texts believed to be translated from Indian originals, that have sometimes long and detailed rites for finding wealth and magi- cal treasure objects, for example, the *Kanị krodha-vajrakumāra- bodhisattva-sādhana-vidhi (Sheng jia ni fen nu jin gang tong zi pu sa cheng

jiu yi gui jing, 聖迦抳忿怒金剛童子菩薩成就儀軌經) (T1222a), and the Padmacintāmaṇi-dhāraṇī sūtra, translated by Bodhiruci in 709 CE.

But quite independently of any Buddhist influences, China al- ready had its own ancient and much more profuse indigenous tradi- tions of treasure recovery. Seidel focused on some aspects of the writ- ten word, starting with inscriptions on rediscovered ancient bronzes, progressing to independent texts on paper, while Franke also dis- cusses non-textual palladia. Barend te Haar has mentioned aspects of both. Several other authors (e.g. Bokenkamp, Campany, Robinet) have discussed further aspects. The bulk of the Taoist tradition, for example, is based on the revelation of sacred texts within caves or grottos, sometimes encoded within magical writing systems that or- dinary mortals find difficult or cannot understand (e.g ‘seal’ or ‘perfected’ scripts, zhuanwen 篆文 and zhenshu 真書).99 As I understand from my colleague Yegor Grebnev, very much more indeed about Chinese treasure discovery still awaits to be described. It will take a long time before we can ascertain which, if any, Chinese treasure traditions might help shed light on their Tibetan counterparts.


Alternative source 9: pre-phyi dar rNying ma tantric scriptures

Conventionally, it is said that gter ma began in the phyi dar; yet we have reason to believe that the revelation of indigenous Tibetan tan- tras had already begun in earlier times. Dunhuang manuscripts, which probably predate the phyi dar, attest a vigorous tantric culture prefiguring what is nowadays known as rNying ma. Likewise, we

98 Strickmann, M., “The Consecration Sutra: A Buddhist Book of Spells,” in: R. Buswell, Jr. ed., Chinese Buddhist Apocrypha. Honolulu, 1990, pp. 75-118. 99 Steavu, D., 2009. ‘The Many Lives of Lord Wang of the Western Citadel: A Note on the Transmission of the Sanhuang wen 三皇文 (Writ of the Three Sovereigns)’, in Journal of the International College for Postgraduate Studies Vol. XIII, 2009, 109-162.


find reference among the Dunhuang mss to the titles of several root tantras, conceived as scriptural—the utterances of the Buddhas—that to this day remain preserved in the rNying ma tantric canon (NGB). While some were probably of Indic provenance, there are indications that others were redacted or revealed in Tibet. Included, for example, are some Phur pa tantras, with the possible understanding that they were among the corpus redacted by Padmasambhava for his Tibetan and Himalayan disciples at the Asura cave in Nepal, as narrated in the Dunhuang text PT 44. Close philological examination of their extant versions indeed exposes clear signs of being redacted or revealed in a Tibetan and Himalayan cultural context from the outset, rather than translated verbatim from Sanskrit. Another such

text, which survived in full at Dunhuang, is the rNying ma Thabs zhags tantra with its commentary and marginalia (Dunhuang text IOL Tib J 321). As far as we can tell (see Alternative source 10 below), this seems to describe its own revelation by Padmasambhava, and con- tains an important iconographical feature typical of indigenous Ti- betan culture, but not so far reported from Indian counterparts. By the end of the 11th century, conservative polemicists like Pho brang Shi ba'i 'od felt the need to compile lists of indigenous Tibetan tantric scriptures (some of them still popular among rNying ma pas) that should be shunned, because they were revealed in Tibet, not translat- ed from Sanskrit. But the available evidence indicates that such revelation began before the conventional beginning of the gter ma move- ment in the phyi dar. We are not yet sure how much earlier: perhaps somewhere between one or two centuries?


This raises several questions, crucial to the understanding of the historical origins of gter ma: (i) What differentiates gter ma from the earlier revelation? (ii) What protocols and conventions might the ear- lier methods of revelation have observed? (iii) Might such methods have flowed into and influenced the development of gter ma? (iv) Might aspects of them still continue within gter ma? Without address- ing such questions, it is unlikely we will be able adequately to under- stand the historical origins of the gter ma traditions; but so far as I am aware, modern scholarship has barely begun to consider them.


My current hypothesis is that the first question will turn out to have two main avenues of enquiry, one reflecting largely internal social-historical developments within Tibet, the other more reflective of a transnational tantric zeitgeist of that period. The first has already been alluded to above (Alternative sources 1 and 2). We have seen how anthropologists propose that under certain circumstances, socie- ties can produce treasure cults formed around popular historical nar- ratives of their past; and how Tibet presented precisely such condi- tions in the phyi dar, when gter ma first began to appear. From this

perspective, what differentiated gter ma from the previous revelation was the transition from mere religious revelation, to a fully devel- oped treasure cult. The earlier revelations merely produced new scriptures through some kind of visionary process. But gter ma was socially and culturally much more complex. It was embedded within historical narratives that melded together idealised remembrances of Buddhism (or Bon) in the ancient Empire with the ubiquitous popu- lar cosmology of treasure-controlling gzhi bdag, to provide a literature for spiritual and cultural renewal in the present.

And because of its historical consciousness, gter ma could also extend into an indigenous archaeology, in which Tibetans of the phyi dar searched out and re- covered actual physical relics of their lost Empire. However, another less local factor might also have been at work. As I will discuss in Alternative source 10 below, a new tantric zeitgeist seems to have been in evidence nearby, in 10th century Kashmir, in which the act of revealing new non-dual Śaiva tantric scriptures came out from the shadows of mystification and anonymity, into the glare of public view and personal attribution. There are at least some indications, as I will suggest below, that key elements of this development might also be reflected in contemporaneous Bon and Buddhist revelation. Here, the difference between the earlier revelations and gter ma is mainly that the latter were no longer anonymous.


The next question remains difficult to answer. Apart from the Dunhuang Thabs zhags manuscript (see Alternative source 10 below), we have little direct evidence for what conventions might have guid- ed the earlier methods of revelation. Our best guess is that they might have followed whatever revelatory conventions were imported into Tibet along with tantric Buddhism, since there can be no doubt that tantric Buddhism in India was a religion of continuous ongoing revelation.


Even if we don’t know what they were, might such methods have flowed into and influenced the development of gter ma? So far, this remains a difficult question to answer, but there are some perspectives we can at least start to think about. Tibetologists have not yet explored possible continuities between the earlier genre of indig- enously revealed rNying ma tantras now designated as bka’ ma, with the slightly later genre of gter ma. One early compilation amongst the voluminous rDzong ‘phrang srog texts now preserved in the rNying ma bka’ma,100 contains several short texts which were written by gNubs chen under his secret name

100 rDzong 'phrang srog gsum gyi chings kyi man ngag in bKa' ma shin tu rgyas pa, TBRC W25983, Volume Ha 29: 15-425.

of Yang dbang gter, or by his student. These raise interesting questions about the existence back in the ninth and tenth centuries, of the later phyi dar distinction between bka’ ma and gter ma. For here, among some portions of these texts that Cantwell believes to be genuinely old, we find a number of colophons describing the re- sealing and burying of these bka’ ma teachings as gter for future recovery.101 In this context, mention should also be made of the tradi- tions surrounding the seventeen Man ngag sde tantras of rDzogs chen, which likewise were said to have been bka’ ma texts buried for later recovery as gter ma.


Whether one gives any credence to such traditions or not, we need to ask, if and why the 11th century revelation of these Man ngag sde tantras will have used a method entirely different to those by which rNying ma tantras were revealed in Tibet over the preceding possibly two hundred years. Similarly, Myang ral’s great 12th century treas- ure discovery, the bDe gshegs ‘dus pa, included at its core some entirely typical rNying ma tantric scriptures that have ever since remained integral to all collections of canonical rNying ma tantras (NGB). We need to ponder the question, would Myang ral, in the 12th century, opt to produce his own

scriptural root tantras by an entirely new method, rather than use at least some of the time-honoured methods already employed for similar rNying ma tantra productions over the preceding two or three hundred years? Of course, the present times still see the production of gter ma root tantras, eligible in principle for inclusion into future redactions of the ostensibly bka’ ma NGB.


Conventionally, scholars say that gter ma first appeared in the phyi dar. But gter ma is not one thing, it is a complex range of phenomena. A new hypothesis to test, is that what happened in the phyi dar might not actually have been altogether new, for not even gShen chen klu dga’ gives any indication that by finding gter ma, he is doing some- thing in his day considered comparatively innovative.102 Perhaps equally likely is that the phyi dar saw a stronger conceptual and nar- rative separation between the categories of bka’ ma and gter ma, ac-

101 My thanks to Changling Tulku, who drew our attention to this extremely valua- ble source, and to Lopon Ogyan Tanzin and Cathy Cantwell, for sharing with me their readings of it.

102 A related but conceptually different issue is that there are no textual survivals from the earliest gter stons who are recognised and celebrated by the rNying ma pa, so that modern academic scholars have been inclined to dismiss them, per- haps too hastily, as mythological. But even these earliest named gter stons, whose texts no longer survive, are generally situated by the traditional literature in the earlier part of the phyi dar. It is not clear to me that we have much evidence for the widespread public identification of a gter ston prior to the phyi dar.

companied by a more public profile for those who discovered gter ma; but that at least some of the varied methods of the treasure dis- coverer were not necessarily entirely discontinuous from those by which rNying ma tantras had already been produced in Tibet for over a century.


Might aspects of these pre-gter ma methods of revelation, some of which might in turn go back to tantric Buddhism in India, still per- sist within the varied revelatory repertoire of the contemporary gter ston? Although the scant surviving historical evidence can at the moment neither definitively prove nor disprove the possibility, on balance, I think the likelihood is that some aspects of them probably do. A stronger case can be made for the alternative revelatory system of Pure Vision (dag snang), which is still practiced in ways demonstrably similar to those described in ancient Indian texts. Like some other ritual and religious practices, revelation can be somewhat connervative and slow to change.


Alternative source 10: Contemporaneous Kashmirian Non-Dual Śaivism

There are interesting similarities between the scriptural revelation practices of 9th to 11th century non-dual Kashmiri Śaivism, and gter stons in nearby Tibet at a similar or very slightly later period, which I believe have not been previously discussed. Understanding these parallels might prove fruitful to researching the historical roots of gter ma, and I plan to research them more fully with Ben Williams.

A recent Phd from Ben Williams has been devoted to the topic of revelation in the traditions of Abhinavagupta.103 The revelations of earlier Śaiva traditions were typically attributed to the fabled interactions at mythical locations of intangible supernatural beings such as ṛṣis and devas. But a defining feature of non-dual Kaśmiri Śaivism became its innovative projection of scriptural revelation out of the fantastical domains of myth, into the plain view of recordable history and tangible geography. As Williams has described, this process can already be seen in the Pratyabhijñāśāstra, to a small degree with Somānanda (c. 900-950), and much more clearly with Utpaladeva’s corpus (c. 925-975). But although already in evidence earlier and elsewhere, notably in Kaula traditions, the description of revelation by named enlightened siddhas, sometimes at specified places and


103 Benjamin Luke Williams. PhD dissertation, Harvard University, August 2017. Abhinavagupta’s Portrait of a Guru: Revelation and Religious Authority in Kashmir.


even at specified times, achieves a kind of crescendo in 10th and 11th century non-dual Śaiva texts from Kashmir, not least with Ab- hinavagupta (fl. c. 975-1025).104 According to Williams, in 10th and 11th century Kashmir, the power to reveal tantric scriptures came to be seen as an integral aspect or demonstration of the guru’s spiritual status or realisation. It is interesting that much the same soon began to become apparent among the Tibetan Bon and rNying ma pa, not very far away from Kashmir.


To give one earlier Śaiva example (see Williams p. 147), the Krama scriptural source, the Yonigahvaratantra, claims to have been revealed by the Yoginīs to an actual historical person, the siddha Jñānanetra, alias Śivānanda (circa 850-900, perhaps only one generation after Padmasambhava?).105 Jñānanetra received his revelation at a tangible geographical location, the Karavīra cremation-ground in Uḍḍiyāna, nowadays Pakistan’s Swat Valley, one of the favourite sites for Kra- ma revelations (and rNying ma narratives of Padmasambhava alike). Similar narratives apply to Niṣkriyānanda, Matsyendranātha, and Vasugupta. Revelations of this kind,

situated within what we might call recordable history and the geographical landscape, rather than veiled behind myth, was a hallmark innovation of non-dual Kashmiri Śaivism, and, as Williams describes in his PhD, central to its theology of the historically existent enlightened siddha as source of revelation. If Williams' analysis proves accurate, developments in Tibet only a few decades later bear interesting comparison: the early 11th century Bon gter ston and

contemporary of Abhinavagupta, gShen chen klu dga’ (996-1035), was surely not the first to reveal scriptures in Tibet, since, as mentioned above, we know of several Tibetan-redacted strictly anonymous rNying ma scriptures that preceded him. But he was surely among the first to bring the process of scriptural revela- tion out into the open field of recordable history, at a real geograph- ical place, which is why he is rightly described as one of Tibet’s first gter ston. Equally striking are parallels in the mode of revelation. Although some of gShen chen's revelations resembled sa gter, 106 another


104 Ben Williams, personal communication 3rd December 2018.

105 For a chronology of Kashmiri Śaivism, see pages 411 ff in Alexis Sanderson, 2007, “The Śaiva Exegesis of Kashmir.” In Mélanges tantrique à la memoire d’Heĺ ène Brunner. Edited by Dominic Goodall & André Padoux, pp. 231–442. Pondicherry, India: Institut Français d’Indologie/Ecole Française d’Extreme-Orient. 106 Three other revelations are more like sa gter, extracted from a gter sgo. Here gShen chen describes the days on which he opened the treasury doors (gter sgo phye ba lags so), and the scribal work of his students in comparing his discoveries with other old texts, and writing them out correctly


seemed to bear closer comparison with the Kashmiri Śaiva model. gShen chen’s 10th century colophons describe how his Gab pa dgu skor revelation descended on his mind as a result of his realisation or siddhi (dngos grub) (Martin 2001: 50-2). This is reminiscent of contem- poraneous Kashmiri revelation, where, as Williams has documented, the reception of new scripture was an integral outcome of realisation, or siddhi. Thus the speech of the realised Śaiva siddha could be con- strued as the utterance of new scripture. The 10th century commenta- tor Rājānaka Rāma (c. 950-1000) praises as follows the speech of Va- sugupta, who revealed the Śivasūtra:107

"I praise the speech of the guru ..Vasugupta to whom the flow of nectar in the form of the essence of vibration, the secret doctrine of all esoteric [[[knowledge]]], was directly transmitted…"(Williams p. 183)

Compare a praise to Padmasambhava from the 10th century Dunhuang text IOLTib J 321, describing him uttering scriptural tantra as an outcome of achieving siddhi:

"(When) .. pure awareness (is produced) by any noble being what- ever, whatever sound is articulated by (his) speech, all without ex- ception is called, “tantra". In the supreme incomparable place of Akaniṣṭha, the Protector Great Being, turning the vajra wheel, speaks through disseminating the tongue's sense faculty108…. I pros- trate to he who has attained the supreme siddhi, of great wonder, Padma rGyal po [The Lotus King] (who) is not worldly; (he who) unravels from the expanse the tathāgata's great secret pith instruc- tions."109

A marginal note is added:



107 These were defined as scriptural by Kṣemarāja (c. 1000-1050), but Sanderson points to earlier sources that already defined the Śivasūtras as scriptural. See Williams p. 187.

108 Cantwell, C., and R. Mayer, 2012. A Noble Noose of Methods: The Lotus Garland Synopsis: A Mahāyoga Tantra and Its Commentary. OAW, Vienna. See page 96: / /skyes bu gang gis rig pa de / /ngag gis ci skad brjod pa'i sgra / /thams cad ma lus tan tra zhes / 'og myin bla myed gnas mchog du / /mgon po bdag nyid chen po yis / /rdo rje 'khor lo bskor pa na / /ljags kyi dbang po bkram las gsungs/ /

109 /dngos grub mchog brnyes ya mtshan chen po ‘i/ / 'jig rten ngam gyur pad ma rgyal po yis/ /de bzhin gshegs pa'i man ngag gsang chen rnams / /klung nas bkrol mdzad de la phyag 'tshal lo //

"this demonstrates [that it, ie this text] is not created by Padmasam- bhava idiosyncratically".110

These similarities merit further investigation, not least because of other doctrinal parallels between the two traditions, their sometimes shared veneration of Uḍḍiyāna as a tantric holy site and source of scripture, the linkage of Padmasambhava with both Uḍḍiyāna and the Tibetan gter ma tradition, and the contiguous and overlapping borders between the Tibetan and Kashmiri cultural zones. However, it seems to me that the institution of gter ston as revealer of scripture in Tibet eventually became even more pronounced, developed, and pervasive, than its Śaiva counterpart.


Alternative source 11: early Bon sources on gter ma

The only scholar so far to have published a book on early Bon gter ma is Dan Martin. Citing Paul Harrison’s work, he suggests gter ma was Indian and Buddhist in origins.111 Yet it is also possible, as he believes, that the Bon were the first to adopt the system. He sums up his position as follows: “The Bon gter-ma phenomenon appears to have been historically initiated prior to that of the Rnying-ma school. Nevertheless it is difficult to imagine where to locate the background for this if not in wider Buddhist history”.112 Nevertheless, as I point out above, I think the pre-phyi dar rNying ma tantric scriptures are an equally possible starting point for some modes of revelation in Tibet. More work needs to be done, to re-examine the evidence. As I have mentioned above, the popular historical consciousness

informing Bon gter ma recovery differs from its Buddhist counter- parts. This has already been well documented from a historical per- spective by scholars such as Dan Martin and Per Kvaerne, but perhaps yet more insights could possibly be gleaned by revisiting the Bon narratives in an anthropological light.

110 pad ma sam ba bhas rang gz[or?] byas pa + + ma yin bar ston

111 Martin, D, 2001. Unearthing Bon treasures: life and contested legacy of a Tibetan scripture revealer, with a general bibliography of Bon. Brill, Leiden. See pages 21-27.

112 Ibid, page 209.


Alternative source 12: Tantric Buddhism from the 8th century onwards

It is frustrating that potentially one of the most valuable source for understanding the historical origins of gter ma in Tibet is also the least known, and the least supported in extant sources, so that we cannot be sure of ever knowing very much about it. Nevertheless, I am hopeful that enough sources will eventually be found to advance our knowledge a bit further.


There seems to be no sign that the Indian tantras of this period were revealed by named and publically identifiable revealers, as we find in both Tibetan gter ma and non-dual Kashmiri Śaivism especially from the 10th century onwards. On the contrary, the 8th to early 10th cen- tury pattern in India seems closer to the obscure anonymous revela- tion of the contemporaneous pre-phyi dar pre-gter ma rNying rgyud tantras in Tibet. Scholars are thus unable to say by whom, where, or when, these influential texts first appeared in India. We can only guess or estimate that the Mañjuśrīnāmasaṃgīti might have appeared in the early 8th century; the

Guhyasamāja perhaps also in the 8th century; the earliest Cakrasaṃvara scriptures perhaps some time in the 9th; the Catuṣpīṭha perhaps in the late 9th; the Hevajra maybe in the early 10th; and the Abhidhānottara possibly also in the 10th; etc. It is only with the Kālacakra, probably in the early 11th century, that we find an explicit account of by whom, where, and when it was revealed: yet unlike contemporaneous Śaiva revelation in Kashmir, or gter ma discovery in Tibet, the Kālacakra revelation narrative is clearly still mythological.


From the fact that the production of new esoteric Tantric Buddhist texts was a significant activity among their contemporaries and men- tors in India, we can suggest that the earliest disciples of such tradi- tions in Tibet imbibed at least some notions of how such tantras were or should be produced. This suggestion is further supported by the fact that Tibetan tantric masters began to produce their own Tibetan tantric texts at quite an early date, and it would seem reasonable to suppose that Tibetans produced their own tantras according to methods influenced at least to some degree by their Indian mentors.


We get hints of how some new Tibetan tantras could be revealed in the Dunhuang Tibetan version of the Thabs zhags (IOLTibJ321) as described above. But we have little direct evidence from Indian sources of the actual mechanics of revelation. David Gray has compiled a range of sources illustrating how tantras were understood to exist in the heavens of Buddhas or ḍākīnis, and descend from there to

earth,113 but he has little or no information describing how, on the ground, a tantric master or siddha would actually concretely reveal a new tantra. It is of course this latter, still largely unknown information that is of more interest for those seeking to understand the historical origins of gter ma.


Conclusion

From the discussion above, it is clear that I do not yet believe we have fully come to understand the historical emergence of the Tibetan Buddhist and Bon gter ma traditions. Much of what I have presented consists of research questions that require further investigation, rather than established conclusions. Davidson (2005) rightly emphasised multiple influences in the formation of gter ma, and our task is to discover what they were, and how they combined and interacted in the unique circumstances of renaissance Tibet. On the one hand, we still need far better to understand indigenous Tibetan fac- tors, not only Tibetan cultural traditions, but also the social dynamics of phyi dar Tibet. On the other hand, we should not vainly imagine 10th century Tibet with its emergent gter ma systems as an exception- al and isolated civilisation, divorced from the wider international zeitgeists of its time.


Bibliography

Amoghapāśakalparāja, 2004-, published in instalments in Annual of The Institute for Comprehensive Studies of Buddhism, Taisho University. Tokyo, Mikkyō Seiten kenkyūkai.

Aris, M., 1989. Hidden Treasures & Secret Lives: Study of Pemalingpa (1450-1521) and the Sixth Dalai Lama (1683-1706). London, Kegan Paul International.
Bernstein, Anya, 2011. “The Post-Soviet Treasure Hunt: Time, Space, and Necropolitics in Siberian Buddhism“, in Comparative Studies in Society and History 53(3), pages 623–653.

von Bierberstein, Alice, 2017. “Treasure/Fetish/Gift: hunting for ‘Armenian gold’ in post-genocide Turkish Kurdistan”. Subjecti- vity, Volume 10, issue 2, pages 170-189.

113 Gray, David B. 2009. “On the Very Idea of a Tantric Canon: Myth, Politics, and the Formation of the Bka’ ’gyur," in Journal of the International Association of Tibetan Studies, no. 5 (December 2009): 1-37.

Blondeau, A-M., “Le Lha-’dre bka’-thaṅ”, in A. Macdonald (ed.), Études tibétaines à la mémoire de Marcelle Lalou, Paris, 1971, pages 29-126.
Blondeau, A-M., Annuaire de l’E.P.H.E., 5ème section, 84 (1975-1976), pp. 109-119, 85 (1976-1977), pages 89-96.
Cantwell, Cathy, 2014. “Development of a Tantric Meditation Practice: Three Generations of Tibetan Lamas” In: Dharma Drum Buddhist College Special Series 2, ed. Chuang Kuo-pin (莊國彬), Taipei, Shin Wen Feng Print Corporation (新文豐), pages 374- 403.

Cantwell, Cathy, 2015. “Different Kinds of Composi- tion/Compilation Within the Dudjom Revelatory Tradition”, In Journal of the International Association of Buddhist Studies, Volume 36/37 2013/2014 (2015), pages 243-280.

Cantwell, C., in press. Dudjom Rinpoche's Vajrakīlaya Works: A Study in Authoring, Compiling and Editing Texts in the Tibetan Revelatory Tradition. Sheffield, Equinox Publications.

Cantwell, C., and R. Mayer, 2012. A Noble Noose of Methods: The Lotus Garland Synopsis: A Mahāyoga Tantra and Its Commentary. Vien- na, OAW.
Cooperson, Michael, 2010. “Al-Ma'mūn, the Pyramids and the Hieroglyphs.“, in 'Abbasid Studies II, ed. John Nawas. Orientalia Lovanensia Analecta 177. Leuven, Paris and Walpole, Peeters and Departement Oosterse Studies Leuven.

Davidson R., 2005. Tibetan Renaissance, Tantric Buddhism in the Rebirth of Tibetan Culture. New York, Columbia University Press.

Davidson R., 2006. ‘Imperial agency in the Gsar ma treasure texts during the Tibetan Renaissance : The Rgyal po bla gter and related literature’, in Davidson R and C Wedemeyer, eds., 2006. Tibetan Buddhist literature and praxis: studies in its formative peri- od, 900-1400. PIATS 2003. Tibetan studies: proceedings of the tenth seminar of the International Association for Tibetan Studies, Oxford, 2003. Leiden, Brill, pages 125-147

Diemberger, Hildegard, 1997. “The Hidden Valley of the Artemisia: On Himalayan Communities and their Sacred Landscape” in
A.W. Macdonald (ed.): Mandala and Landscape. New-Delhi, D.K. Printworld.

Dotson, Brandon, 2007. “‘EmperorMu rug btsan and the ’Phang thang ma Catalogue.” Journal of the International Association of Tibetan Studies, no. 3 (December 2007): 1-25.

http://www.thlib.org?tid=T3105 (accessed May 7, 2017).
Dudjom Rinpoche, 1991. The Nyingma School of Tibetan Buddhism. Its History and Fundamentals, trans. Dorje and Kapstein. Boston, Wisdom Publications.

rDzong 'phrang srog gsum gyi chings kyi man ngag in bKa' ma shin tu rgyas pa, TBRC W25983, Volume Ha 29: 15-425

Eckel, Malcolm David, 1994. To See the Buddha: A Philosopher’s Quest for the Meaning of Emptiness. Princeton, Princeton University Press.

Einoo, Shingo, 2009. ‘From kāmas to siddhis — Tendencies in the Development of Ritual towards Tantrism’, in Einoo, Shingo, ed., Genesis and Development of Tantrism, Tokyo, Institute of Ori- ental Culture Special Series, 23, University of Tokyo.

Esler, Dylan, 2018. The Lamp for the Eye of Contemplation. The bSam- gtan mig-sgron by gNubs-chen Sangs-rgyas ye-shes: Hermeneutical Study with English Translation and Critical Edition of a Tibetan Buddhist Text on Contemplation. PhD thesis, Louvain-la-Neuve.

Foster, George, 1964. ‘Treasure Tales and the Image of the Static Economy in a Mexican Peasant Village’, Journal of American Folklore 77: 39–44.
Francke, A. H., 1927. ‘gZer-Myig. A Book of the Tibetan Bompos’.
Asia Major, Vol. iv, Fasc. 2-3, 1927.

Franke, Herbert, 1978. From tribal chieftain to universal emperor and god: the legitimation of the Yüan dynasty. München, Verlag der Bayer- ischen Akademie der Wissenschaften.

Gazizova, Valeriya, 2019. “New Buddhists, ‘Treasure’ Discoveries and (Re)constructed Protective Deities of Kalmykia“, in Inner Asia (2019), pages 61-82.
Gray, David B., 2009. “On the Very Idea of a Tantric Canon: Myth, Politics, and the Formation of the Bka’ ’gyur," in Journal of the International Association of Tibetan Studies, no. 5 (December 2009): 1-37.

Guru Chos dbang, n.d. gTer ‘byung chen mo, pages 75-193, within Gu ru chos dbang gi rnam dang zhal gdams, Rin chen gter mdzod chen po’i rgyab chos Vols 8-9, Ugyen Tempa’i Gyaltsen, Paro, 1979. TBRC Work Number 23802.

Gyatso, Janet, 1998. Apparitions of the Self. Princeton, Princeton University Press.

Gyatso, Janet, 2015. Contribution on gTer ma to the Brill Encyclopaedia of Buddhism, Leiden, Brill, pages 398-404.

Gzhon-nu-dpal, ʼGos Lo-tsā-ba, 1984-5, 2 vols. Deb ther sngon po.

Chengdu, Si-khron Mi-rigs Dpe-skrun-khang.

Haar, B. J. ter, 1998. Ritual and mythology of the Chinese triads: creating an identity. Leiden, Brill.
Haarh, Erik, 1969. The Yar-luṅ Dynasty : a study with particular regard to the contribution by myths and legends to the history of ancient Ti- bet and the origin and nature of its kings. København, Gad Forlag.

Hardy, Friedhelm, 1994. The Religious Culture of India: Power, Love and Wisdom. Cambridge, Cambridge University Press.
Harrison, P., 1978. The Tibetan text of the Pratyutpanna-Buddha- Saṃmukhāvasthita-Samādhi-Sūtra, Tokyo, IIBS.
Harrison, P., 1990. The Samādhi of Direct Encounter with the Buddhas of the Present: An Annotated English Translation of the Tibetan Ver- sion of the Pratyutpanna-Buddha-Saṃmukhāvasthita-Samādhi-Sūtra, Tokyo, IIBS.

Harrison, P., 1996. “A Brief History of the Tibetan Bka’ ’gyur,” in: J. Cabezon & R. Jackson, eds., Tibetan Literature: Studies in Genre. Ithaca, Snow Lion, pages 70–94.

Harrison, P., 2005. “Mediums and Messages: Reflections on the Pro- duction of Mahāyāna Sūtras”, The Eastern Buddhist XXXV 1 & 2, pages 124-5.
Hazod, Guntram, 2016. ‘The Plundering of the Royal Tombs, an Analysis of the Event in the Context of the Uprisings in Central Tibet of the 9th/ 10th Century’, in Christoph Cüppers, Robert Mayer and Michael Walter (Editors) Tibet after Empire: Culture, Society and Religion between 850-1000, Proceedings of the Seminar held in Lumbini, Nepal, March 2011, Zentralasiatische Studien ZAS ZAS 45, 2016, 704 S, pages 113-146.

Hermann-Pfandt, A., 2008. Die Lhan KarMa. Ein früher Katalog der ins Tibetische übersetzten buddhistischen Texte. Kritische Neuausgabe mit Einleitung und Materialien. Vienna, OAW.

Hirshberg, Daniel, 2016, Remembering the Lotus-Born: Padmasambhava in the History of Tibet's Golden Age. Somerville, Wisdom Publica- tions.
Ishikawa, Mie, 1990. A Critical Edition of the Sgra Sbyor Bam Po Gnyis Pa. An Old and Basic Commentary on the Mahavyutpatti. Tokyo, The Toyo Bunko.
Jacoby, Sara, 2014. Love and Liberation. Autobiographical Writings of the Tibetan Buddhist Visionary Sera Khandro. New York, Columbia University Press.
'Jigs med bstan pa'i nyi ma, 2007-2008. gTer brgyud kyi rnam bshad. Pages 797-858 of Vol Kha of the Shechen edition of the Rin chen gter mdzod chen mo. New Delhi, Shechen Publications. TBRC W1KG14. 2: 813 - 874.

Kapstein, Matthew, 1989. “The Purificatory Gem and Its Cleansing: A Late Tibetan Polemical Discussion of Apocryphal Texts,” HR 28/3, 1989, 217–244.
Kapstein, Matthew, 2000. The Tibetan Assimilation of Buddhism. Con- version, Contestation and Memory. New York, Oxford Univer- sity Press.

Kapstein, Matthew, 2015. “Gter-ma as Imperial Treasure: The 1755 Beijing Edition of the Padma bka’ thang”, in RET Number 31, Février 2015, Papers for Elliot Sperling

Karmay, Samten, 1988. The Great Perfection (rDzogs chen) A Philosophi- cal and Meditative Teaching of Tibetan Buddhism. Leiden, Brill.
Karmay, Samten, 1989. The Arrow and the Spindle. Studies in History, Myths, Rituals and Beliefs in Tibet. Kathmandu, Mandala Book Point.
Khokhlov, Yury, forthcoming. “Uncovering Amoghavajra’s legacy in the Hexi Corridor and Tibet.”In Journal of Tibetology.
Kelly, Piers, 2016. "Excavating a hidden bell story from the Philip- pines: a revised narrative of cultural-linguistic loss and recu- peration." Journal of Folklore Research 53 (2), pages 86-113.

Kuijp, Leonard van der, 2016. ”Reconsidering the Dates of Dol po pa Shes rab rgyal mtshan's (1291-1361) Ri chos nges don rgya mtsho and the Bka' bsdu bzhi pa'i don," 藏族俄学刊 བོད་རིག་པའི་དས་ དེབ་ Journal of Tibetology 14 (2015/2016), pages 115-159.
Kvaerne, Per, 1980. ‘A preliminary study of the gzer mig’. In Michael Aris and Aung San Suu Kyi (eds.), Tibetan Studies in Honour of Hugh Richardson, Warminster, Aris & Phillips, pages 185-191.
Laufer, B., 1911. Der Roman einer tibetische Koenigin (bTsun-mo bka’- thang), Leipzig.

Leider, Jacques, 2009. ”Relics, Statues & Predictions: Interpreting an Apocryphal Sermon of Lord Buddha in Arakan”, in Asian Eth- nology 68 no. 2, pages 333-364.
Martin, Daniel, 2001. Unearthing Bon Treasures. Life and Contested Lega- cy of a Tibetan Scripture Revealer. Brill’s Tibetan Studies Library,
1. Leiden, Brill.

Martin, Daniel, October 9th, 12th, 14th, 2014. blog Tibeto-Logic
https://tibeto-logic.blogspot.com

Mathes, K-D., 2013. ”Clouds of Offerings to Lady g.Yang ri. A Protector Practice by the First Yol mo sprul sku Shākya bzang po (15th/16th Cent.)”. In Franz-Karl Ehrhard and Petra Maurer (eds), Nepalica-Tibetica: Festgabe for Christoph Cüppers, Band 2. Andiast, IITBS.

Mayer, Robert, 1994. “Scriptural Revelation in India and Tibet. Indian Precursors of the gTer ma Tradition”. In: Tibetan Studies, PIATS 6, Vol. 2. Oslo, Institute for Comparative Research in Human Culture, pages 533-544.

Mayer, Robert, 2007. ‘The Importance of the Underworlds: AsurasCaves in Buddhism, and Some Other Themes in Early Buddhist Tantras Reminiscent of the Later Padmasambhava Legends’, in

Journal of the International Association of Tibetan Studies, no. 3 December 2007, pages 1-31.
Mayer, Robert, 2015. “(What's so new about Terma?) Terton and Tradent: Innovation and Conservation in Tibetan Treasure Lit- erature”, In Journal of the International Association of Buddhist Studies, Volume 36/37 2013/2014 (2015), pages 227-242.

Oppitz, Michael, 2006. “Die Geschichte der Verloren Schrift“, Frobenius-Vorlesung 2005, in Paideuma 52, pages 27-50.
Pommaret, Françoise, 1998. “Maîtres des Trésors” (gTer bdag): divinités locales et médiums au Bhoutan” in PIATS 7, Graz, Tibetan Mountain Deities, Their Cults and Representations. ed Anne-Marie Blondeau. Vienna, OAW.

Pritzker, David T., 2017. Canopy of Everlasting Joy: An Early Source in Tibetan Historiography and the History of West Tibet. Oxford DPhil dissertation.
Rambelli, F., "The Myth of the South Indian Iron Stupa (Nanten tettō) in Japanese Esoteric Buddhism", in International Journal of Bud- dhist Thought & Culture Vol. 27. No. 2 (December 2017): 71–88.

Sanderson, Alexis, 2007, “The Śaiva Exegesis of Kashmir.” In Meĺ anges tantrique à la memoire d’Heĺ ene Brunner. Edited by Dominic Goodall & André Padoux, pp. 231–442. Pondicherry, Institut Français d’Indologie/École Française d’Extreme-Orient. Seidel, Anna, 1983. “Imperial Treasures and Taoist Sacraments: Tao- ist Roots in the Apocrypha,” in Michel Strickmann, ed., Tantric

and Taoist Studies in Honour of R. A. Stein, vol.2, pp.291-371. Mélanges Chinois et Bouddhiques XXI. Brussels, Institut Belge des Hautes Études Chinois.
Sørensen, Per K., 1994. Tibetan Buddhist Historiography. The Mirror Illuminating the Royal Genealogies. An Annotated Translation of the XIVth Century Tibetan Chronicle: rGyal-rabs gsal-ba'i me-long, Wiesbaden, Harrassowitz Verlag.

Steavu, D., 2009. ‘The Many Lives of Lord Wang of the Western Citadel: A Note on the Transmission of the Sanhuang wen 三皇文 (Writ of the Three Sovereigns)’, in Journal of the International Col- lege for Postgraduate Studies Vol. XIII, 2009, 109-162.

Stein, Rolf A., 1988. Grottes-Matrices et Lieux Saints. Paris, EFEO (PEFEO 151).
Stewart, Charles, 2012. Dreaming and Historical Consciousness in Island Greece. Cambridge, Harvard University Press.
Strickmann, M., “The Consecration Sutra: A Buddhist Book of Spells,” in: R. Buswell, Jr. ed., Chinese Buddhist Apocrypha. Hon- olulu, 1990, pages 75-118.
Tanaka, Kenneth K., and Raymond E. Robertson. 1992. 'A Ch'an Text from Tun-huang: Implications for Ch'an Influence on Tibetan Buddhism'. In Steven D. Goodman and Ronald M. Davidson, eds., Tibetan Buddhism: Reason and Revelation. Albany, SUNY Press, pages 57–78.
Terrone, Antonio, 2010. Bya rog prog zhu, The Raven Crest. The Life and Teaching of bDe chen ‘od gsal rdo rje, Treasure Revealer of Contem- porary Tibet. PhD dissertation, Leiden.

Terrone, Antonio, 2014. “The Earth as a Treasure: Visionary Revela- tion in Tibetan Buddhism and its Interaction with the Environ- ment”, in Journal for the Study of Religion, Nature, and Culture Volume 8, Number 4 Special Issue: Religion and Nature in Asia and the Himalayas. Guest Edited by Georgina Drew and Ashok Gurung.

Tucci, G., 1949. Tibetan Painted Scrolls, 3 vols., Rome, Libreria dello Stato.
Tucci, G., 1950. The Tombs of the Tibetan Kings. Rome, Serie Orientale Roma I.
Tucci, G., 1958. Minor Buddhist Texts, Rome, Serie Orientale Roma, IX, Part II.

Tulku Thondup Rinpoche, 1986. Hidden Teachings of Tibet. London, Wisdom Publications.
Valtchinova, Galia, 2009. “Visionaries and the National Idea in Interwar Bulgaria: The Circle of the Orthodox Association of the Good Samaritan, “ in Acta Ethnographica Hungarica, 54 (2), pp. 265–285 (2009) DOI: 10.1556/AEthn.54.2009.2.2

Walsh, Martin, 2018. “Treasure island. Buried gold and the spiritual economy of Pemba“, in Kenya Past and Present, issue 45, pages 23-32.
Wangdu, Pasang and Hildegard Diemberger, 2000. dBa’ bzhed: The Royal Narrative Concerning the Bringing of the Buddha’s Doctrine to Tibet. Vienna, OAW.
Weinberger, Steven, 2003. The Significance of Yoga Tantra and the Com- pendium of Principles (Tattvasaṃgraha Tantra) within Tantric Bud- dhism in India and Tibet. PhD Dissertation, University of Virginia.

Williams, Benjamin Luke, 2017. Abhinavagupta’s Portrait of a Guru: Revelation and Religious Authority in Kashmir, Unpublished PhD dissertation, Harvard.
Yang, Zeng, 2018. A Biographical Study on Bukong 不空 (aka. Amo- ghavajra, 705-774). Networks, Institutions, and Identities. PhD the- sis, the University of British Columbia.


Source