Tibetan Manuscript and Xylograph Traditions The Written Word and Its Media within the Tibetan Culture Sphere Edited by Orna Almogi Contents Preface 3 ORNA ALMOGI & DORJI WANGCHUK Prologue: Tibetan Textual Culture between Tradition and Modernity 5 ORNA ALMOGI The rNying ma rgyud ’bum Set at the National Archives Kathmandu: The History of Its Production and Transmission 31 MICHELA CLEMENTE Different Facets of Mang yul Gung thang Xylographs 67 HILDEGARD DIEMBERGER Early Tibetan Printing in Southern La stod: Remarks on a 1407 Print Produced at Shel dkar 105 FRANZ-KARL EHRHARD Buddhist Hagiographies from the Borderlands: Further Prints from Mang yul Gung thang 127 AGNIESZKA HELMAN-WAżNY Overview of Tibetan Paper and Papermaking: History, Raw Materials, Techniques and Fibre Analysis 171 MATTHEW KAPSTEIN A Collection of Miscellaneous Kanjur Folios including Four Illustrated Pages from a rNying ma Tantra in statu nascendi 197 SAM VAN SCHAIK The Uses of Implements are Different: Reflections on the Functions of Tibetan Manuscripts 221 PETER SCHWIEGER Some Palaeographic Observations on Tibetan Legal Documents 243 MARTA SERNESI Reprinting the Buddhist Classics: On the Production and Circulation of Blockprints 267 TSUGUHITO TAKEUCHI & MAHO IUCHI Varieties of Tibetan Texts from Khara-khoto and Etsin-gol: An Introductory Remark 321 VESNA WALLACE Remarks on the Tibetan Language Manuscripts and Xylographs in Mongolia and on the Technology of Their Production 347 DORJI WANGCHUK Sacred Words, Precious Materials: On Tibetan Deluxe Editions of Buddhist Scriptures and Treatises 371 Preface Textual scholarship, including text and book cultures, has a long and rich history throughout the Tibetan cultural sphere. Since the development of the Tibetan script—according to traditional sources sometime in the 7th century—tens (or perhaps hundreds) of thousands of texts, be they of Indic origin or autochthonous Tibetan, have been written down on Tibetan soil. Consequently, a much greater number of books, be they in the form of manuscripts or xylographs, were produced, transmitted, and further reproduced throughout the centuries. Tibetan textual scholarship thus becomes highly interesting and relevant for all of us who strive to gain a nuanced and well founded knowledge of Tibetan intellectual culture, intellectual history, religion, philosophy, textual criticism, literature, or language. In recent years we have been witnessing a growing interest in Tibetan textual scholarship—including Tibetan text and book cultures—that goes beyond the mere textual and contentual matters. Issues concerning material and visual aspects of Tibetan book culture—including writing materials, economical and logistical aspects of production, patronage, codicology, palaeography, technology, craftsmanship, artistry, and art—and such concerning Tibetan text culture—including traditional textual scholarship in general and compilatory processes and editorial policies in particular—have come to the forefront of Tibetan Studies. Religious and sociological aspects of Tibetan book culture have likewise been increasingly addressed—particularly those focusing on the book as being a ritual or reverential object, an artefact possessing magical powers, a prestigious item to be owned, a merit-accruing object, or a piece of art. With the conviction that a better understanding of these aspects will advance and enhance Tibetan textual studies as a whole, a conference on “Manuscript and Xylograph Traditions within the Tibetan Cultural Sphere: Regional and Periodical Characteristics” was held at the Universität Hamburg in May 15–18, 2013. As the title suggests, the conference aimed at discussing and identifying regional and periodical characteristics of various manuscript and xylograph traditions within the Tibetan cultural sphere. The present volume contains twelve of the papers presented at the conference along with an introductory essay, which all together cover many of the above mentioned issues regarding Tibetan manuscripts, xylographs, and legal handwritten documents, stemming from different periods of Tibetan history and from various regions within the Tibetan cultural sphere, including such that had been under its influence in the past. Although the volume is far from addressing neither all traditions of text and book cultures within the Tibetan cultural sphere nor all issues concerning them, it is hoped that it nonetheless will be a modest contribution to the advancement of research in this field along with several other recent publications with a similar or related focus. I would like to particularly thank Dorji Wangchuk for his cooperation and assistance in organising the conference and in making it possible through the financial support of the Khyentse Center for Tibetan Buddhist Textual Scholarship (KC-TBTS), and likewise for his support in various ways during the editing of the present volume. Special thanks are also due to the Khyentse Foundation whose financial support of the KC-TBTS enabled both the conference and the publication of the present volume. And last but not least thanks are also due to Eric Werner for his help in solving some last-minute technical problems during the preparation of the final version of the volume. Orna Almogi Hamburg, July 30, 2016 The rNying ma rgyud ’bum Set at the National Archives Kathmandu: The History of Its Production and Transmission Orna Almogi (Hamburg)[1] 1. Introductory Remarks When I started my study on the history of the formation, production, and transmission of the rNying ma rgyud ’bum (the Collection of the Ancient Tantras), nine editions—eight manuscript and one xylograph—were accessible, for five of which there already exist catalogues of varying scope: (1) The mTshams brag illuminated manuscript edition (Tb), with an online catalogue compiled by the Tibetan & Himalayan Library (THL), (2) The Rig ’dzin Tshe dbang nor bu illuminated manuscript edition (Tn), also known as the Waddell edition, with a detailed online catalogue compiled by Cathy Cantwell, Robert Mayer, and Michael Fisher, (3) The gTing skyes plain manuscript edition (Tk), with a catalogue compiled by Eiichi Kaneko (1982) and an online catalogue compiled by the THL, (4) The Nubri illuminated manuscript edition (Nu), uncatalogued, (5) The illuminated manuscript edition (Na) stored at the National Archives Kathmandu (NAK), uncatalogued, (6) The sGang steng plain manuscript edition (Gt-p), uncatalogued, (7) The sGang steng illuminated manuscript edition (G-i), with a catalogue compiled by Cathy Cantwell, Robert Mayer, Michael Kowalewky, and Jean-Luc Achard, (8) dGra med rtse plain manuscript edition (Gm), uncatalogued,3 (9) sDe dge xylograph edition (Dg), with a catalogue complied by Jean-Luc Achard (2003) and an online catalogue compiled by the THL. Right from the outset of my study of the history of the transmission of the collection, it was clear that a thorough examination of the content of the Nubri and the NAK sets—which were microfilmed by the Nepal-German Manuscript Preservation Project (NGMPP)4—would be necessary, not only for gaining an accurate picture of the history of their formation, production, and transmission, but also for better understanding the history of the rNying ma rgyud ’bum in general, and the place these two sets occupy within the collection’s various groups of transmission in particular.5 For this purpose, the Nubri and NAK sets stood at the centre of my initial investigation of the collection. In the following, I shall present some of the main findings regarding these two editions, primarily focusing, however, on the NAK set and only by extension on the Nubri one. 2. The Tibetan-Nepalese Borderlands rNying ma rgyud ’bum Editions: General Remarks My study of these two sets—which, for reasons that will be made clear below, I collectively designate the “Tibetan-Nepalese Borderlands group”—has initially focused on their content and its arrangement, primarily following historical-philological and bibliographical methods. Some of the questions that arose in the course of the investigation were later on further addressed by employing various scientific methods, including material analysis (mainly of the inks and pigments, but to a lesser extent also of the paper) and multispectral imaging of the NAK set. [6] Both the NAK and Nubri sets have been previously studied by Franz-Karl Ehrhard. As early as 1979, Ehrhard recorded seven volumes of a rNying ma rgyud ’bum set in his catalogue of the NAK collection of Tibetan texts that had been microfilmed thus far by the NGMPP (i.e. on the basis of the microfilms). [7] But it was not until 1989, after gaining access to the NAK Tibetan collection itself, that the entire Tibetan holdings, including the rNying ma rgyud ’bum set stored there, could be microfilmed. During the microfilming, it became clear that the set is incomplete. The team determined, most probably correctly, that the original number of volumes of the set was 37, but at the time it was believed that only 32 of them had survived and were in the safekeeping of the Archives. [8] In 1992, this time during an expedition to Nubri (Samagaon), the NGMPP microfilmed a second rNying ma rgyud ’bum set. The Nubri set was complete, consisting of 37 volumes, and it soon became clear that the Nubri and the NAK sets are closely related to each other in terms of contents, organisation, and the history of their transmission. Ehrhard studied the two collections, paying particular attention to the second of the two dkar chag-s that were microfilmed together with the Nubri set, and published his findings in a 1997 article.[9] In regard to the Nubri set, Ehrhard showed that several editions were produced at the behest of Brag dkar rta so sprul sku Chos kyi dbang phyug (1775–1837) at the end of the 18th and beginning of the 19th centuries, one of them during the years 1813–1814. This latter set was reportedly prepared on the basis of a set from Glang ’phrang, which could be borrowed thank to one sKal bzang of the rDo dmar family. Ehrhard notes that the set was later kept in Brag dkar rta so and is today in the hands of Slob dpon ’Gyur med, who enabled the NGMPP to microfilm it. In regard to the edition stored at the NAK, Ehrhard suggested—on account of its similarity to the Nubri set and the fact that some of the volumes bore the seal of the Śrī Tīn Mahārāja Bhim Shumser Jang Bahadur Rana, who was the prime minister of Nepal from 1929 to 1932—that it is possibly the set reported to have been produced shortly after the Nubri set by a disciple of Brag dkar rta so sprul sku, one O rgyan ’phrin las bstan ’dzin, who belonged to the Nyang clan of gZhung in Rong shar (i.e. the family responsible for the founding and upkeep of the main temple of Junbesi in Solu Khumbu), and to have later been given to Prime Minister Bhim Shumser. He further suggests that Sangs rgyas Bla ma (1856–1939) of the Nyang clan, who was under the protection of the Raṇa family and who was the person behind the renovation of the gZhung temple in Junbesi in 1914, was involved in the gifting of the set to Bhim Shumsher. In regard to the content of these two editions, one of the interesting discoveries by Ehrhard at that time was the fact that the last two volumes of both sets contain rDzogs chen tantras that are not included in the gTing skyes edition, but are found in the Bai ro rgyud ’bum. My first step in the investigation of the two sets in question was to edit the dkar chag containing the bibliographical list that was microfilmed by the NGMPP together with the Nubri set (i.e. dKar chag 1). In addition, I located the titles mentioned there in the Nubri set itself and compared the Nubri and the NAK sets in terms of their contents and organisation. Moreover, I also attempted to identify the individual texts contained in these two sets with those found in the other known editions. Very soon, several things became clear: [10] (a) Despite the great similarity between the Nubri and NAK sets, they differ in enough ways to presuppose a slightly different history of transmission. (b) While the Nubri edition is more or less uniform in terms of its organisation, format, and layout, the NAK set is rather chaotic in these points, betraying rather poor editorial skills, and perhaps also a lack of familiarity with the literary material, on the part of its producers and editors. (c) The texts found in the last two volumes noted by Ehrhard as missing from the gTing skyes edition are all found in the editions belonging to the Central Bhutanese group.[11] (d) Both sets contain texts that could not be located thus far in any of the other accessible rNying ma rgyud ’bum editions that are catalogued. (e) The NAK set apparently suffered from the moment it was conceived from a very ill thought-out organisational concept, which is particularly evident in the assignment of the volume numbers and the foliation. The fact that the set was in Nepalese possession seems to have contributed to the disorder, for the librarians in charge could seemingly not read Tibetan (though there was an attempt to insert folio numbers in Indian numerals on the verso of folios, apparently in order to facilitate the handling of the set by staff with no knowledge of Tibetan). These circumstances are perhaps the reason, too, why during the microfilming by the NGMPP only 32 volumes could be identified, although 35 were present and were in fact microfilmed. (f) And most importantly, the two sets clearly differ from the gTing skyes and the Rig ’dzin Tshe dbang nor bu editions in various ways, and therefore the four editions cannot be classified into one and the same group called “south-central,” as previously suggested by Robert Mayer and Cathy Cantwell.[12] 3. The Six Groups of Transmission of the rNying ma rgyud ’bum In connection with this last point, I would like to briefly present now an overview of the currently accessible rNying ma rgyud ’bum sets by way of grouping them in terms of the history of their transmission. During the past years, four further sets have become accessible. Two sets were digitised in Bhutan in 2012 within the framework of the above-mentioned project “Doxographical Organisational Schemes in Manuscripts and Xylographs of the Collection of the Ancient Tantras”:13 (1) a unique set from gDong dkar la monastery (Dk), Paro (sPa gro) district, and (2) an illuminated set from sPa sgar monastery (Pg), Thimphu (Thim phu/phug) district.14 Two further sets were digitised in 2013: (3) another unique set located in Khams (Kh), the digital images of which were obtained by me in the fall of 2013 from mKhan po ’Jam blo from the ’Jam dbyangs shes rig dar spel khang in Chengdu, and (4) an illuminated set from Sangs rgyas gling monastery (Sg), Tawang (rTa dbang), which was digitised by an Oxford University project under Cathy Cantwell, Robert Mayer, and Ngawang Tsepag. While the discoveries of the sPa sgar and Sangs rgyas gling editions have little significance in terms of advancing our understanding of the history of the transmission of the rNying ma rgyud ’bum collection, that of the gDong dkar la edition has great significance in this regard, since it represents a unique, hitherto unknown line of transmission. As for the Khams edition, although its content and organisation appears to be different from all other accessible editions, it is still unclear how significant it is as far as the collection’s history of transmission is concerned, and the matter is yet to be thoroughly investigated. The structure and content of these two editions cannot, however, be discussed here and will be treated separately elsewhere. On the basis of my extensive studies of the Nubri and NAK sets and my preliminary investigations of the gDong dkar la and Khams editions (catalogues of all four sets are currently under preparation),15 I suggest the following grouping of the thirteen16 editions accessible to date:17 (a) The Central Bhutanese group comprises six sets, all of which consist of 46 volumes and obviously ultimately go back to the same origin: (i) sGang steng plain manuscript edition (Gt-p), (ii) sGang steng illustrated manuscript edition (Gt-i), (iii) mTshams brag illustrated manuscript edition (Tb), (iv) dGra med rtse plain manuscript edition (Gm), (v) sPa sgar illustrated manuscript edition (Pg), and (vi) Sangs rgyas gling illustrated manuscript edition (Sg), which is unfortunately incomplete (vols. 11, 12 and 41 being missing). The relation of these six editions among themselves has not been thus far satisfactorily clarified. It is, however, quite certain that the sGang steng sets are among the earliest ones and that they (or one of them) have played an important role in the history of the transmission of the other editions in this group. From the information obtained thus far it seems that the earliest among them is the plain sGang steng set, which, according to historical sources, was produced in 1642 by Pad gling gsung sprul III Phan pa bzang po alias Kun mkhyen Tshul khrims rdo rje (1598–1669; TBRC: P1692) as a commemoration set (dgongs rdzogs) for his teacher sGang steng sprul sku I rGyal sras Padma ’phrin las (1565–1642; TBRC: P2659) right after his death.18 According to his autobiography, Pad gling gsung sprul III commissioned various objects to commemorate the death of sGang steng sprul sku I, the most important of which was a complete 46-volume rNying ma rgyud ’bum set. The scribes and other craftsmen are stated there to have been brought from distant places and the work to have been brought to completion with much care.19 The set that served as the master copy for this edition is stated in Karma bde legs’s rGyud ’bum dkar chag as having been borrowed from lHa lung, and the production as having taken about four months and as having required about 40 scribes (at least some of whom appear to have been from sNye mo).20 No information regarding the circumstances of the production of the master copy (reportedly from lHa lung)—which is clearly older than all accessible sets of the Central Bhutanese group, and possibly the one to which all ultimately goes back—could be obtained thus far. The sGang steng illuminated set is said to have been produced in 1726/27 as a commemoration set for sGang steng sprul sku II bsTan ’dzin legs pa’i don grub (1645–172621; TBRC: P514). It has been further stated that the set was prepared under the auspices of the rGyal tshab—to be identified as Mi pham dbang po (1709–1738; TBRC: P531), the reincarnation of rGyal sras bsTan ’dzin rab rgyas (1638– 1698; TBRC: P512)—who also consecrated the set, that its front pages were written in gold on black paper, and that it consisted of “about 45!” volumes.22 The mTshams brag set is reported to have been commissioned by mTshams brag sprul sku I Ngag dbang ’brug pa (1682– 1748; TBRC: P526), the founder of mTshams brag monastery and a disciple of sGang steng sprul sku II bsTan ’dzin legs pa’i don grub (1645–1726; TBRC: P514). His biography, composed by rJe mkhan po IX Shākya rin chen (1710–1759; TBRC: P530), mentions his commissioning of the edition only briefly.23 There the origin of the master copy is stated to be Punakha. This, however, is something that needs further verification, since sGang steng monastery, which has thus far been widely believed to be the source of the master copy, is located in Wangdue Phodrang (dBang ’dus pho brang) district and not in Punakha. The date of production is not provided either but it has most probably begun in 172624 and been completed in 1728.25 These dates as the years of production of this set make one wonder what the exact connection between it and the one commissioned by rGyal tshab Mi pham dbang po and others at around the same time to commemorate sGang steng sprul sku II. One askes oneself, for example, whether mTshams brag sprul sku, too, conceived the set as commemoration to sGang steng sprul sku II (who was his own teacher), whether the same set served as the master copy, and whether there was any logistical coordination between the two projects (e.g. sharing the same scribes and editors). The production date of the Sangs rgyas gling edition26 is unclear but, provided it is a copy of the mTshams brag set, it would be later than the latter. The production dates of the dGra med rtse and sPa sgar sets have not yet been determined. This leaves us with the sGang steng plain edition produced in 1642, the sGang steng illuminated edition of 1726/27, and the mTshams brag edition probably dating from 1726–1728, that is, from around the same time as the sGang steng illustrated set, while the relationship among the Gt-p, Gt-i, and Tb sets is not completely clear. (b) The South-Western Tibetan group comprises two sets, both apparently of 33 volumes originally: (i) Rig ’dzin Tshe dbang nor bu illuminated manuscript edition (Tn) (incomplete), and (ii) gTing skyes dGon pa byang plain manuscript edition (Tk). The two sets represent the fruit of activity surrounding the production and transmission of the rNying ma rgyud ’bum in South-Western Tibet during the late 18th century and perhaps also at the beginning of the 19th century. As for the Rig ’dzin Tshe dbang nor bu set, as shown by Cantwell and Mayer, it was commissioned in honour of Kaḥ thog rig ’dzin Tshe dbang nor bu (1698–1755), perhaps by actual students of his or later followers of his tradition, and thus can likely be dated to the late 18th century.27 As already pointed out by Dan Martin, the gTing skyes set is probably the set reported in the biographies of the gTing skyes dGong pa byang throne-holders composed by mTha’ grol rdo rje. There it is listed among the numerous books that the founder of dGon pa byang (TBRC: G1KR1628), Padma chos ’phel (alias Bya btang mKhas grub lha rje alias Tshe ring don ’grub, 1772/1773–1836; cf. TBRC: P2DB20793), deposited in the monastery’s temple.28 The year of the monastery’s founding, and thus a terminus post quem of the set’s production, is unknown, but considering the dates of Padma chos ’phel, it would be reasonable to assume that he carried out activities of such monumental scale during the second half of his life and thus place the set’s production in the early decades of the 19th century. This would position it slightly later than the Rig ’dzin Tshe dbang nor bu set. The relation between the two sets is, however, yet to be determined. (c) The Tibetan-Nepalese Borderlands group also consists of two sets, apparently both originally in 37 volumes: (i) the Nubri set (Nu), an illuminated manuscript edition that was produced in Brag dkar rta so and is currently stored in Nubri, and (ii) the NAK set (Na), an illuminated manuscript edition probably produced in the Solu Khumbu area and is currently stored at the National Archives in Kathmandu (incomplete). As pointed out earlier, both were produced at the beginning of the 19th century: the Nubri set was produced in Brag dkar rta so at the behest of Brag dkar rta so sprul sku Chos kyi dbang phyug during the years 1813–1814, and the NAK set probably in the Solu Khumbu area shortly after the Nubri set at the behest of one O rgyan ’phrin las bstan ’dzin, a disciple of Brag dkar rta so sprul sku. As I shall show below, although the NAK set is probably later than the Nubri one, it reflects an earlier (i.e. older) stage in the history of the transmission than the Nubri set, as is clearly evident from its organisation. (d) The Western Bhutanese group, consisting thus far of one set, namely, the gDong dkar la plain manuscript edition (Dk). As I have shown elsewhere, the set was probably produced in 1647 in East Bhutan, at the behest of sPa gro Chos dbang lhun grub (b. 17th cent.; TBRC: P2718) and under the sponsorship of King Dewa of Kha ling, his queen and other residents of the area.29 This set comprises 28 volumes, and from a preliminary study of it, it has become clear that it differs greatly from all other sets accessible thus far and indeed represents an independent compilation. There is no doubt that it is independent of the Central Bhutanese group, which came into being around the centre of the Pad gling tradition in lHa lung. To be noted is that the earliest available set amongst the Central Bhutanese group (i.e. the plain sGang steng) was produced only a few years prior to the gDong dkar la edition. There is therefore no doubt that a thorough study of it will shed more light on the history of the rNying ma rgyud ’bum collection in general and its transmission in Bhutan in particular. An interesting feature of this set is that it is rich in editorial glosses, which give numerous hints regarding the process of its compilation and the policies followed by its editors (see fig. 2). These glosses are valuable not only for our better understanding of the collection’s history of transmission but also for our better understanding of the text and book cultures in Tibet in general, for they provide us with unique glimpses into the compilatory and editorial processes shaping large corpora of Buddhist literary collections (be they canonical, para-canonical, or extra-canonical)—processes that have been practised within the Tibetan cultural sphere for centuries.30 (e) The fifth group, which will be tentatively called (Central Bhutanese)–Eastern Tibetan,31 likewise consists thus far of only one set of 34 volumes, which are written in dBu med in black ink with rubrication in red (see fig. 3). For lack of sufficient details regarding its exact origin, it is simply referred to here as the Khams edition (Kh). As stated above, I obtained digital images of this set from mKhan po ’Jam blo (of the ’Jam dbyangs shes rig dar spel khang) during a visit to Chengdu in the fall of 2013. This set of images consists of 34 folders, numbered 1–34. To be noted, however, that the 34th folder contains images of an as yet unidentified modern reproduction of a dBu med manuscript (bearing page numbers in printed Arabic numerals), which obviously differs from the first 33 volumes in both codicological and palaeographical terms. The set—which is currently stored in gZhi chen monastery,32 a subsidiary of Kaḥ thog monastery in the Serta area—is, like the gDong dkar la set, unique in terms of content and organisation. My preliminary study of this Khams edition shows that, while it too bears witness to independent compilatory work on the part of its producers, its compilers had access to a set related to the Central-Bhutanese group. It seems that, just as in the case of the Tibetan-Nepalese Borderlands group, this edition represents a transitional or intermediate stage between the South-Western Tibetan and the Central Bhutanese groups. However, since it includes not only texts found in the editions of both the Tibetan-Nepalese Borderlands and Central Bhutanese groups, but also texts that are only found in the Central Bhutanese group, it clearly represents a stage in the transmission that is later than the one reflected by the Tibetan- Nepalese Borderlands group. Moreover, its organisation does not resemble any of the other groups, so that its exact relation to them is yet to be determined. It appears, however, to resemble more the Central Bhutanese group in terms of content (if not in terms of organisation). Thus the suggestion to tentatively name the group it represents (Central Bhutanese)– Eastern Tibetan. According to personal communication with Karma bde legs (previously of dPal brtseg, Lhasa), this edition was apparently compiled at the behest of Khyab gdal lhun grub (b. 17th cent.) from Brag dmar monastery33 (the current digitisation efforts in gZhi chen were carried out by mKhan po ’Chi med rig ’dzin from Bla rung sgar, likewise in the Serta area). The exact time and place of compilation and production, the master copies used, the persons involved, and the exact role of Brag dmar Khyab gdal lhun grub are, however, as yet to be clarified. If this edition indeed dates from the late 17th or early 18th century (provided it was indeed commissioned by Brag dmar Khyab gdal lhun grub), the circumstances of its production, if they come to light, are surely bound to shed light on the collection’s history of transmission in East Tibet, of which little is known thus far.34 (f) The (Central)–Eastern Tibetan group consists of the only xylographic edition of the collection that has ever been produced. Despite the fact that this group is represented by only a single set from East Tibet, it seems more accurate to designate it “(Central)–Eastern Tibetan” and not simply “Eastern Tibetan,” since, as is well known, it was prepared at the behest of the queen of sDe dge Tshe dbang lha mo—the wife of the sDe dge king Sa dbang bzang po, who died at the early age of 25 years—between the years 1794 and 1798. It is said to be based on several sets, but the master copy was obviously the Central Tibetan set (or a copy of it) commissioned by ’Jigs med gling pa (1729/30–1798; TBRC: P3) in Padma ’od gling some years earlier.35 A comparison of its contents with ’Jigs med gling pa’s title list of the Padma ’od gling set contained in his history-cum-catalogue of the rNying ma rgyud ’bum shows that the two are indeed very similar, though not identical.36 4. The NAK Set, Its Production, Organisation, and History of Transmission In the following I wish to return to the two Tibetan-Nepalese Borderlands editions, with an emphasis on the NAK set. On the basis of my comparison of the content of these two sets, the following conclusions could be drawn thus far: (a) Due to the fact that, unlike the Nubri set, the NAK set suffers from great deficiencies in regard to the organisation of its individual texts, volumes numbers, and foliation it clearly represents an earlier stage than the Nubri set in terms of the overall history of the transmission of the rNying ma rgyud ’bum in the Tibetan-Nepalese Borderlands. (This of course does not mean that the former was produced earlier than the latter!) Thus, the Nubri set could not have possibly served as the master copy for the NAK set—as one is tempted to believe based on the assumption that the latter was very probably produced after the former and the reported master–disciple relationship between Brag dkar rta so sprul sku, who was behind the production of the Nubri set, and O rgyan ’phrin las bstan ’dzin from Rong shar, who, it has been suggested, was the figure behind the production of the NAK set. (b) The Tibetan-Nepalese Borderlands group—of presumably originally 37 volumes in each of its sets—clearly represents an intermediate stage between the South-Western Tibetan group—consisting of the gTing skyes and Tshe dbang nor bu editions, presumably originally 33 volumes each—and the Central Bhutanese group—consisting of six sets of 46 volumes each. As Ehrhard already pointed out, dkar chag 2, which was transmitted together with the Nubri set, informs us that when Myang ston mnga’ bdag Rig ’dzin rgya mtsho (b. 17th cent.), a disciple of gTer bdag gling pa (1646–1714; TBRC: P7), was acting as the chief editor of the sNar thang xylograph edition of the bKa’ ’gyur prepared at Shel dkar at the behest of the ruler Pho lha ba/nas bSod nams stobs rgyas (1689–1747; TBRC: P346)—that is, sometime between 1730 and 1732—he asked the ruler to invite a master named Ngag dbang lhun grub grags pa from lHo brag lHa lung to participate in the endeavour. Consequently, Ngag dbang lhun grub grags pa came to dPal mo chos sding in La stod (TBRC: G1KR1648), a monastery founded by Bo dong ’Jigs med grags pa alias Phyogs las rnam rgyal (1376–1451; TBRC: P2627), and gave there teachings— including a text transmission of the rNying ma rgyud ’bum!—to various personalities of the area. It is further reported that after the teaching and reading transmission, Ngag dbang lhun grub grags pa called upon the disciples to make a copy of the entire collection of the rNying ma rgyud ’bum in order to increase the reading transmission and to likewise carry out a reading transmission of the collection at least once. As a result, several sets were produced. Considering this report, it is very plausible that the text transmission of the rNying ma rgyud ’bum given by Ngag dbang lhun grub grags pa was based on a copy of a collection that he brought with him from lHo brag lHa lung, which, as we know, was the centre of the Pad gling reincarnation lineages and the location where the prototype for the Central Bhutanese editions very probably originated.37 In this case, it is also very likely that this lHa lung set served as the master copy for the Tibetan-Nepalese borderlands editions. This would suggest that the texts contained in the last two volumes of these two editions and other texts that are not found in the editions of the South-Western Tibetan group but in those of the Central-Bhutanese one were already included in the lHa lung edition, and maybe were even added to the rNying ma rgyud ’bum there for the first time. However, we have no exact information as to its predecessor, or in other words, to the prototype I assume to have served as the master copy for the Tibetan-Nepalese Borderlands editions, and which I suggest represents an intermediate, or transitional, stage in the history of the transmission of the collection, reflecting its state at a particular point on the line of transmission between the South-Western Tibetan group, on the one hand, and the Central Bhutanese group, on the other. To be kept in mind is that if the master copy of the Tibetan-Nepalese Borderlands editions was indeed a copy brought by Ngag dbang lhun grub grags pa from lHo brag lHa lung in 1730, it would appear that Ngag dbang lhun grub grags pa brought along with him a copy representing an earlier stage of the 46-volume Central Bhutanese group despite the fact that several copies of the 46-volume edition have already existed (i.e. at least the sGang steng plain edition of 1642, sGang steng illuminated edition of 1726/27, mTshams brag edition of probably 1726–1728, and the one reported to have been prepared by Pad gling gsung sprul IV Ngag dbang kun bzang rdo rje in the 1710s or early 1720s). The reasons for Ngag dbang lhun grub grags pa’s choice of the master copy could have been very practical ones, such as availability (we still do not have any explicit information regarding the existence of the 46-volume edition in lHa lung itself), but also other matters could have played a role, such as the tradition he belonged to and the lineage of reading transmission he received. As I pointed out earlier, the NAK set is incomplete and suffers from major organisational deficiencies, a fact that initially led to difficulties in identifying and determining the number of available volumes. During my cataloguing of the NAK set—which also included a comparison with the Nubri set—it became clear that 35 of what were probably originally 37 (Ka−Ji) volumes are currently stored in the National Archives, with only volumes Za and Sha being missing. Most of the organisational deficiencies seem to have had their origin already at the time the set was being produced. They mainly take the form of the wrong assignment of volume numbers— which in turn led to some confusion in the way the individual volumes were stored in the National Archives. While the originally 37 volumes of the NAK set should have been, exactly as in the case of the Nubri set, assigned the numbers Ka–Ji, what we find is that in several cases certain volume numbers have been assigned twice, and in two other cases volume numbers have probably not been assigned at all: (a) Two volumes are assigned the number Ja. During the cataloguing it became clear that one of them, designated by me Ja2, should be in fact volume Zha, which at first seems to be missing. This volume is, however, unfortunately incomplete, the first 81 folios being missing. As I have shown elsewhere, there seem to be no particular features common to volumes Ja and Ja2 (= Zha) which could possibly explain this confusion: the hands are different and no shared pattern of the employment of the ink varieties could be observed. It could, however, be demonstrated that the ink varieties used in Ja2 (= Zha) are rather similar to those used in some of the last volumes of the collection, which are generally of poorer quality—that is, in terms of the performance of the scribes and artist, the materials used, and the editorial scrutiny—and which appear to have been possibly produced towards the end of the project, perhaps by a different group of scribes in the same or a different location, and very probably also under some financial strain. This fact may not fully explain the reason for the confusion in the assignment of the volume number, but perhaps it partly does.38 (b) In addition to one volume numbered Tha, which corresponds to its counterpart in the Nubri set and thus can be assumed to be bearing the correct volume number, we find another small volume that is likewise assigned the number Tha. This volume, which I have designated Tha2, merely consists of 133 folios and only contains one text. This text corresponds to the third text in volume Da in the Nubri set and is indeed missing from volume Da in the NAK set.39 Also to be noted is the fact that unlike in the case of the other volumes, the first two folios of Tha2 are not illuminated. Unfortunately, no evidence could be found thus far—either through palaeographical examination or by means of material analysis of the types of ink used—that could hint at the origin of Tha2 and explain its existence or coming into being.40 (c) We also find two volumes with the number Ma. The volume I designate Ma2 is incomplete, missing the first 191 folios. As I have pointed out elsewhere, it had been initially assigned the volume number Pha, but the number was later on altered to Ma. The reasons for this change have thus far not been clarified, neither through historical-philological inquiry nor by way of material analysis or multispectral imaging. Nonetheless, it could be established, through both palaeographical examination and material analysis of the ink— that Ma2 was highly likely to have been an integral part of the collection from the time of its production and so is not of external origin. It should be, however, perhaps pointed out that most of the texts contained in Ma2 are found in volume Ma of the Rig ’dzin Tshe dbang nor bu edition, which could be a useful piece of evidence regarding the history of the transmission of this cluster of texts, provided more pieces of the puzzle surface in the future. Furthermore, all five texts contained therein (i.e. fols. 192–390) match the first five texts in volume Pha of the Nubri set.41 Interestingly, the counterparts of the remaining 15 texts in volume Pha of the Nubri edition (Nu.237–244) are found in volume Da in the NAK set (fols. 1– 164),42 while the counterparts of the texts contained in volume Pha of the NAK set are all found in volume Ba of the Nubri set.43 These differences between the Nubri and the NAK sets in the distribution of texts within the individual volumes should be weighed against the fact that in general there is otherwise a rather good correspondence between the two sets. (d) At least two volumes seem not to have been initially assigned a number at all, namely, Wa and Ji. Moreover, as I have already pointed out elsewhere, both of these volumes seem to have been at some point erroneously assigned the number Tsa (while the latter has also been occasionally confused with the number Ca due to the graphical similarity of these two Tibetan letters), a mistake that was obviously soon recognised by the persons involved in the production. The reason that these volumes initially lacked a number and that they were then falsely assigned the same number (in both cases there may have been some graphical confusion) has unfortunately not been satisfactorily clarified thus far.44 Besides these obvious irregularities and confusion regarding the assignment of volume numbers, one often also observes confusion regarding the assignment of folio numbers in the NAK set. Several of these cases have been examined by way of multispectral imaging. As has already pointed out, the processed images seem to support the historical-philological and palaeographical-codicological analyses which suggest that the work load appeared to have been distributed among the scribes in bundles of ten folios and that at least in some cases the confusion in the foliation is probably a result of this division of labour (possibly when not properly supervised and coordinated).45 Despite the evidently poor editorial quality, the production of the NAK set must have been regarded as a prestigious project, as is commonly the case with the production of such huge corpora. The fact that the first two written pages (i.e. fols. 1b & 2a) of most of the volumes in the set are illuminated—the text being written with golden ink on black paper and decorated with illustrations—further confirm the prestigious nature of this undertaking. Nonetheless, the quality of the illuminated pages varies greatly, including (i) the quality of the gold (which turned out to be in most cases a mixture of gold and orpiment in varying ratios, while only in some cases was the ink found to be made of pure gold, and in some others it consisted of orpiment alone with no gold component at all), (ii) the usage of red ink instead of golden ink (particularly in the last volumes),46 (iii) the quality of the illustrations (some of which were, for example, left unpainted or only partly painted), (iv) the quality of other decorative elements, such as the curtained windows of the verso of the first folios (some windows, for example, were left uncut), and (v) inconsistency in the layout. Some examples of the varying quality of the illuminated front pages of the NAK set—of, in particular, their declining quality in terms of both material and artistry—as well as variation in their layout are provided below (see figs. 4–8).47 As I have suggested elsewhere, the inconsistency in the quality of the gold and the illustrations—which seem to be declining particularly in the last volumes—appears to hint at financial difficulties towards the end of the production project. To sum up, as we have seen, the organisational incoherency and fluctuating quality of the craftsmen’s skill and style and of the writing material used could be partly explained on the basis of the observations and findings made thus far, and several plausible assumptions could be accordingly articulated: (i) At least some of the volumes of the NAK set were produced by a different team of scribes, possibly at a different location. (ii) Different artists, likewise possibly in different locations, were responsible for the illustrations. (iii) The sponsors of this huge undertaking very probably faced financial difficulties, particularly towards its end phase. (iv) The division of labour among the scribes possibly involved the distribution of bundles consisting of ten folios each. (v) The editorial proofreading was in all likelihood carried out in a centralised manner. 5. Concluding Remarks In this paper, I have touched upon and discussed several issues regarding the history of the transmission of the rNying ma rgyud ’bum in general and that of the NAK set in particular. I first pointed out the existence of a hitherto unrecognised distinct group among the currently accessible rNying ma rgyud ’bum sets, one that reflects a separate line of transmission, namely, the Tibetan-Nepalese Borderlands group, consisting of the Nubri and NAK sets, which thus far have been grouped together with the Rig ’dzin Tshe dbang nor bu and the gTing skyes sets (under the designation “south-central”). Secondly, I have presented a scheme of six groups, presenting six distinct lines of the transmission of the rNying ma rgyud ’bum collection, with an attempt to briefly discuss the dates and the circumstances of their production, so as to enable us to better understand the relation between the sets within one and the same group on the one hand and between the groups on the other. This was followed by a brief discussion of the historical background of the Nubri and NAK sets and of their content in comparison to other sets, on the basis of which I suggested that the Tibetan-Nepalese Borderlands group represents an intermediate or transitional stage between the South-Western Tibetan and the Central Bhutanese groups. Lastly, I discussed some distinct features of the NAK set, particularly in comparison to the Nubri set, focusing on selected irregularities in the NAK set. On the basis of this examination I proposed that the NAK set represents a stage in the transmission line of the collection in the Tibetan-Nepalese Borderlands that is earlier than that of the Nubri set, even though the NAK set was probably produced later than the Nubri one. However, all the circumstances surrounding the production of the NAK set have yet to be uncovered. Considering all the irregularities that slipped through the editorial oversight—as reflected in organisational matters, such as text order, assignment of volume numbers, and foliation—and the fluctuating quality of the material and its visual impact—as reflected in writing materials, the skills and style of the craftsmen (scribes and artists), and layout—several questions arise regarding the production of the NAK set. These include (i) whether the organisational deficiencies reflect the state of the master copy; (ii) whether there was no capable scholar available; (iii) whether the set was produced in haste; (iv) whether its production was partially outsourced or staggered—that is, it was produced in different places or at different points in time; (v) what led to the logistical or financial difficulties that the production process seems to have faced; and (vi) whether all the volumes currently stored at the National Archives have been an integral part of the collection from the very beginning or whether some are later additions. While some of these questions will have to remain unanswered as long as no further evidence comes to light, attempts have been made to answer some of them, at least partially, by combining historical-philological studies with codicological-cum- palaeographical investigations and scientific examinations (including analyses of the ink and paper and multispectral imaging). The present article augments my two recent publications concerning the NAK set, which have mainly presented the research results of codicological- cum-palaeographical studies combined with scientific methods— including material analysis of the ink and paper and multispectral imaging of select folios from the NAK set—conducted in an attempt to answer some of the questions that have ensued as a result of observations made during my historical-philological (i.e. particularly bibliographical) studies of the Nubri and NAK sets. It particularly discusses findings resulting from historical-philological studies—of mainly bibliographical, biographical, and historical material of various sorts—in an attempt to shed more light on the history of the production and transmission of the rNying ma rgyud ’bum collection in the Tibetan-Nepalese borderlands in general and of the NAK set in particular. Moreover, the findings gained through the study of the circumstances of the production of the NAK set go beyond their significance to the study of the history of the formation, production and transmission of the rNying ma rgyud ’bum collection. As I have shown in my recent discussion of the gDong dkar la edition,48 the study of the circumstances leading to the formation of a rNying ma rgyud ’bum set—and in fact of any other large corpus of Buddhist literature for that matter—could shed more light on the book and text cultures within the Tibetan cultural sphere in general, and on the compilatory process, editorial policies, scribal practices, and logistical matters in particular. This is in addition to the increasing knowledge obtained in recent years regarding the material aspects of the culture of the book within the Tibetan cultural domain, particularly concerning paper but also ink and pigments. Bibliography 1. Primary Sources gTing skyes ’khrungs rabs rnam thar = mTha’ grol rdo rje. mTshungs med dpal mgon bla ma dam pa gting skyes dgon byang mchog sprul ’khrungs rabs bcu’i rnam par thar pa mdo tsam brjod pa. Biographies of successive embodiments of the Tingkye Gonpa Jang masters. Reproduced from a manuscript made in 1965 in Sikkim. Thimphu: Kunzang Tobgyel & Mani Dorji, 1979. [Scans: TBRC: W2CZ6602]. Gu bkra’i chos ’byung = Gu ru bkra shis alias sTag sgang mkhas mchog Ngag dbang blo gros, bsTan pa’i snying po gsang chen snga ’gyur nges don zab mo’i chos kyi byung ba gsal bar byed pa’i legs bshad mkhas pa dga’i byed ngo mtshar gtam gyi rol mtsho. Beijing: Krung go bod kyi shes rig dpe skrun khang, 1990. mTshams brag rnying rgyud dkar chag = Mi nyag Thub bstan chos dar. mTshams brag rnying ma rgyud ’bum gyi dkar chag gsal ba’i me long. Beijing: Mi rigs dpe skrun khang, 2009. Ngag dbang ’brug pa’i rnam thar = rJe mkhan po IX Shākya rin chen. rGyal kun brtse ba’i spyi gzugs sems dpa' chen po gsung dbang sprin dbyangs kyi rtogs pa brjod pa rig ’dzin kun tu dga’ ba’i zlos gar. In Thub dbang gnyis pa rgyal ba shākya rin chen gyi gsung ’bum. Reproduced from the unique MS. Prepared after the author’s death and preserved at his monastery of Pha-jo-ldin Bya-rgod-phun-po. Thimphu: Kunzang Topgey, 1976, vol. 2, pp. 487–585. [Scans: TBRC: W8684-4366]. rGyud ’bum dkar chag = Karma bde legs. sNga ’gyur rgyud ’bum rin po che phyogs sgrig gi dkar chag. Lhasa: dPal brtsegs bod yig dpe rnying zhib ’jug khang, 2009. rNying ma rgyud ’bum Dg = sDe dge Xylograph Edition. 25+1 vols. Scans: TBRC: W21939. Catalogue: (i) Jean-Luc Achard, “Rig ’dzin Tshe dbang mchog grub (1761–1829) et la constitution du rNying ma rgyud ’bum de sDe dge.” Revue d’Etudes Tibétaines 3: 43–89, 2003; (ii) The Tibetan & Himalayan Library (THL), www.thlib.org. rNying ma rgyud ’bum Dk = gDong dkar la Plain Ms Edition. 28 vols. Stored at gDong dkar la monastery, Paro district, Bhutan. Tibetan Manuscript and Xylograph Traditions 62 Digitisation: CSMC, Universität Hamburg & Preservation of Bhutan’s Written Heritage, Bhutan, 2012. rNying ma rgyud ’bum Gm = dGra med rtse Plain Ms Edition. 46 vols. Stored at dGra med rtse monastery, Mongar district, Bhutan. Digitisation: Karma Phuntsho in cooperation with The British Library, 2006. rNying ma rgyud ’bum Gt-i = sGang steng Illuminated Ms Edition. 46 vols. Stored at sGang steng monastery, Wangdue district, Bhutan. Digitisation: Karma Phuntsho in cooperation with The British Library, 2003. Catalogue: Cathy Cantwell, Rob Mayer, Michael Kowalewski & Jean-Luc Achard, The sGang steng-b rNying ma’i rGyud ’bum Manuscript from Bhutan. “Introduction” by Cantwell & Mayer, 4–15, “Catalogue” by Cantwell, Mayer, Kowalewski & Achard, 16–141. Revue d’Etudes Tibétaines 11, 2006. rNying ma rgyud ’bum Gt-p = sGang steng Plain Ms Edition. 46 vols. Stored at sGang steng monastery, Wangdue district, Bhutan. Digitisation: Karma Phuntsho in cooperation with The British Library, 2004. rNying ma rgyud ’bum Kh = Khams Plain Ms Edition. 34 vols. Digitisation: mKhan po ’Chi med rig ’dzin, Bla rung sgar, 2011?. [Partly = TBRC: W2PD17382, 39(!) vols.; scans currently restricted.] rNying ma rgyud ’bum Na = National Archives Kathmandu Illuminated Ms Edition. 37? vols. (incomplete, 35 vol. available). Stored at the National Archives Kathmandu, Nepal (Ms no. 5). [The set is probably originally from Junbesi, Solu Khumbu, Nepal.] Microfilmed by the Nepal-German Manuscript Preservation Project (NGMPP), 1989: Reel Nos. AT 1/1–AT 25. rNying ma rgyud ’bum Nu = Nubri Illuminated Ms Edition. 37 vols. In the possession of Slob dpon ’Gyur med, Nubri (Samagaon), Nepal. [The set is originally from Brag dkar rta so.] Microfilmed by the Nepal-German Manuscript Preservation Project (NGMPP), 1993: Reel Nos. L 426/4–L 448/1. rNying ma rgyud ’bum Pg = sPa sgar Illuminated Ms Edition. 46 vols. Was stored at sPa sgar monastery, Thimphu district, Bhutan. Digitisation: CSMC, Universität Hamburg & Orna Almogi: The rNying ma rgyud ’bum Set at the National Archives Kathmandu 63 Preservation of Bhutan’s Written Heritage, Bhutan, 2012. [Ms destroyed by fire on 29.02.2012.] rNying ma rgyud ’bum Sg = Sangs rgyas gling Illuminated Ms Edition. 46 vols. (incomplete, 43 vols. available). Digitisation: Cathy Cantwell, Rob Mayer, Ngawang Tsepag, 2013. [Scans can be viewed in person at the Weston Library of the Bodleian, Oxford; announced as soon to be made available through the TBRC: W1KG16449]. rNying ma rgyud ’bum Tb = mTshams brag Illuminated Ms Edition. 46 vols. Stored at mTshams brag monastery, Thimphu district, Bhutan. Facsimile Edition: Thimphu: The National Library of Bhutan, 1982. Scans: TBRC: W21521. Catalogue: The Tibetan & Himalayan Library (THL), www. thlib.org. rNying ma rgyud ’bum Tk = gTing skyes Plain Ms Edition. 33 vols. Facsimile Edition: Thimphu: Dingo Khyentse Rimpoche, 1973–1975. 33+3 vols. Scans: TBRC: W21598. Catalogues: (i) Eiichi Kaneko 1982. Ko Tanotora zenshū kaidai mokuroku [A catalogue of the Khyentse edition of the gTing-skyes dgon-pa-byang rnying-ma-rgyud-’bum]. Tokyo: Kokusho Kankūkai; (ii) The Tibetan & Himalayan Library (THL), www. thlib.org. rNying ma rgyud ’bum Tn = Rig ’dzin Tshe dbang nor bu (=Waddell) Illuminated Ms Edition. 33? vols. (incomplete). Stored at various places, including The British Library, London; Bodleian Library, Oxford University; Victoria and Albert Museum, London. Catalogue: Cathy Cantwell, Robert Mayer & Michael Fisher, http://ngb.csac.anthropology.ac.uk/. sDe dge rnying rgyud dkar chag = Mi nyag Thub bstan chos dar. rNying ma rgyud ’bum gyi dkar chag gsal ba’i me long. Beijing: Mi rigs dpe skrun khang, 2000. sGang steng sprul sku gnyis pa’i rnam thar = rJe mkhan po X bsTan ’dzin chos rgyal. sGang steng sprul sku bstan ’dzin legs pa’i don grub kyi rnam thar. Phuentsholing, Bhutan: KMT Publishers, [2009]. Tshul rdor rnam thar = Pad gling gsung sprul III Phan pa bzang po alias Kun mkhyen Tshul khrims rdo rje, sPrul sku tshul rdor gyi rnam par thar pa’i tshul sgrib med ston pa’i me long. Ms. [Scans: TBRC: W8LS15799]. Tibetan Manuscript and Xylograph Traditions 64 2. Secondary Sources Achard, Jean-Luc. 2003 = See rNying ma rgyud ’bum Dg. Almogi, Orna. 2015. “The Spa sgar and Gdong dkar la Rnying ma rgyud ’bum Editions: Two Newly Discovered Sets from Bhutan.” In Olaf Czaja & Guntram Hazod (eds.), The Illuminating Mirror. Tibetan Studies in Honour of Per K. Sørensen on the Occasion of his 65th Birthday. Wiesbaden: Dr. Ludwig Reichert Verlag, 1–14. ——. forthcoming-a. The Collection of the Ancient Tantras (rNying ma rgyud ’bum): The History of Its Formation, Production, and Transmission. Indian and Tibetan Studies Series. Hamburg: Department for Indian and Tibetan Studies, Universität Hamburg. ——. forthcoming-b. “Spiritual Masters and Master Copies on the Move: On Some Manuscript Editions of the rNying ma rgyud ’bum from East Tibet and Their Origin.” Almogi, Orna, Emanuel Kindzorra, Oliver Hahn & Ira Rabin. 2015. “Inks, Pigments, Paper: In Quest of Unveiling the History of the Production of a Tibetan Buddhist Manuscript Collection from the Tibetan-Nepalese Borderlands.” JIABS 36/37, 2013/2014 (2015): 93‒117. Almogi, Orna, Martin Delhey, Claire MacDonald & Boryana Pouvkova. 2015. “Recovering Lost Text and Beyond: Multispectral Imaging for Text-related and Codicological Studies of Tibetan Paper and Sanskrit Palm-Leaf Manuscripts.” Cantwell, Cathy. 2002. “Distinctive Features of the Edition.” http://ngb.csac.anthropology.ac.uk. [A revised version of “Distinctive Features of the Rig ’dzin tshe dbang nor bu (Waddell) Edition of the rNying ma’i rgyud ’bum.” In Helmut Eimer & David Germano (eds.), The Many Canons of Tibetan Buddhism. PIATS, Leiden 2000. Leiden: Brill, 2002: 359– 376.] Cantwell, Kathy & Robert Mayer. 2007. The Kīlaya Nirvāṇa Tantra and the Vajra Wrath Tantra: Two Texts from the Ancient Tantra Collection. Philosophisch-historische Klasse, Denkschriften 349. Beiträge zur Kultur- und Geistesgeschichte Orna Almogi: The rNying ma rgyud ’bum Set at the National Archives Kathmandu 65 Asiens 52. Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Cantwell et al. 2006 = See rNying ma rgyud ’bum Gt-i. Ehrhard, Franz-Karl. 1980. “Tibetan Texts in the National Archives, Kathmandu.” JNRC 4: 233–250. ——. 1997. “Recently Discovered Manuscripts of the rNying ma rgyud ’bum from Nepal.” In H. Krasser et al. (eds.), Tibetan Studies: Proceedings of the 7th Seminar of the International Association for Tibetan Studies, Graz 1995, vol. 1, 253–267. Kaneko, Eiichi. 1982 = See rNying ma rgyud ’bum Tk. Martin, Dan. unpublished. “NGB Historical Notes: A chronologically arranged list of prints and manuscripts (both available and unavailable) of the Rnying-ma Rgyud-’bum, based mainly on narrative historical sources.” Unpublished version from 1995 ================ 1 The findings presented in this article are some of the results of two research projects generously funded by the Deutsche Forschungsgemeinschaft (DFG): (a) “The Manuscript Collections of the Ancient Tantras (rNying ma rgyud ’bum): An Examination of Variance,” conducted within the framework of the Researcher Group “Manuscript Cultures in Asia and Africa” (FOR 963, 2008–2011), and (b) “Doxographical Organisational Schemes in Manuscripts and Xylographs of the Collection of the Ancient Tantras,” conducted within the framework of the Centre for the Study of Manuscript Cultures (CSMC/SFB 950, 2011– 2015), both at Universität Hamburg. I would like to take this opportunity to thank the Nepal Research Centre in Kathmandu for their assistance in various ways over this period, which greatly facilitated my research, and likewise thank Philip Pierce (Kathmandu) for proofreading my English. Thanks are also due to Kelsang Lhamo and Jeff Wallman, TBRC, for their help in obtaining access to some scanned material. 2 Cantwell and Mayer name the illuminated set catalogued by them, together with Michael Kowalewky and Jean-Luc Achard, “sGang steng-b” and the other, plain set from sGang steng, which was digitised later, “sGang steng-a,” justifying their decision as follows: “We adopt this nomenclature because, as we will explain below, we believe the finer and more expensive manuscript we photographed first and which is presented here was made later than the simpler more cheaply produced one we one (sic) discovered later.” (Cantwell et al. 2006: 5). They further support their assumption with findings gained through philological studies, but nonetheless state in their conclusion that they “do not yet know for certain” which set is the earliest, but they do justify their decision by stating that “it is more likely that the finer one presented here is later, since it is the more expensively made of the two” (ibid. 10). As I shall show below, historical sources indeed seem to support the assumption that the plain set (provided this is indeed the reported set) was produced earlier than the illuminated one. However, since the connection between the two sets is yet to be established, I suggest naming the two sets sGang steng-p and sGang steng-i (“p” standing for “plain” and “i” standing for “illuminated”). It is hoped that future studies— philological, historical, or bibliographical—will help shed more light on the relation between the two editions and on the exact role of each of them in the history of the transmission of the Central Bhutanese group. To be noted here is that the oral tradition in sGang steng monastery is not confident in this regard either, though it appears that there is a tendency to believe that the illuminated set served as the master copy for the sets from mTshams brag and dGra me rtse (personal communication with sGang steng sprul sku at sGang steng monastery on September 9, 2009, and further verifications via Khenpo Seng nge rdo rje in February 2016). Cantwell and Mayer for their part have suggested that the mTshams brag edition is not a copy of the illuminated sGang steng set, but rather that both are copies of a third, as yet unidentified, exemplar (ibid. 9–10). 3 Although I consider the dGra med rtse edition, written in black ink on white paper, to be plain, it should be nonetheless noted that some efforts have been made to lend the edition a somewhat fairer look: the title page (i.e. folio 1b of each volume) is written on thicker-layered paper, held together by what seem to be black and red leather strips, while the resulting knots serve as decorative elements; textile curtains are mounted on these same title pages; and the written area of the first two written pages (i.e. folios 1b and 2a of each volume) is smeared with reddish colour (see fig. 1). 4 The microfilming of both sets was carried out under the supervision of Franz-Karl Ehrhard, who was at the time the director of the Nepal Research Centre (NRC), the NGMPP branch in Kathmandu. 5 The cataloguing of the remaining two sets, sGang steng-p and dGra med rtse, has clearly been less urgent for this purpose due to their great similarity to the two other 46-volume Bhutanese sets—mTshams brag and sGang steng-i—for which catalogues already exist. 5 The cataloguing of the remaining two sets, sGang steng-p and dGra med rtse, has clearly been less urgent for this purpose due to their great similarity to the two other 46-volume Bhutanese sets—mTshams brag and sGang steng-i—for which catalogues already exist. 6 The findings of these scientific investigations, which took place in March 2013 in the National Archives Kathmandu, have been published in Almogi, Kindzorra, Hahn & Rabin 2015 and Almogi, Delhey, MacDonald & Pouvkova 2015. 7 Ehrhard 1980: 245–246. The volumes recorded by Ehrhard are Cha, Nya, Tha, Pha, Ba, Tsa, and Chi. Note, however, that in Ehrhard 1997: 254, the number of volumes recorded at the time is erroneously noted as being eight. 8 Ehrhard 1997: 254. 9 The first dkar chag―titled rNying ma rgyud ’bum gyi glegs bam nang gi chos tshan bzhugs byang dkar chag dpe rdzi bsam ’phel nor bu’i ’phreng ba (NGMPP Reel No. L 426/4, 26 fols. = dKar chag 1)―contains the list of titles included in the edition. The second―titled sNga ’gyur gsang chen rnying ma rgyud ’bum gyi glegs bam yongs rdzogs gzheng tshul dkar chag tu bkod pa rdzogs ldan snang ba gsar pa’i dga’ ston (NGMPP Reel No. L 426/5, 14 fols. = dKar chag 2)―contains a descriptive account of the production of the edition. Ehrhard has translated and edited two excerpts from the latter. His findings, summarised in the following paragraphs, are based on his 1997 article. 10 The information provided in the present article regarding the overall organisation of the Nubri and NAK sets—including volume and text numbers, the identification and location of the individual texts within the sets, and the comparison between the two sets—is based on the current draft of the catalogue in Almogi (forthcoming-a). 11 On the groups of the rNying ma rgyud ’bum editions in terms of the history of their transmission, see below. 12 See, for example, Cantwell & Mayer 2007: 70–78; 2012: 26–30 13 The digitisation of this set was made possible thanks to the financial support of the DFG and fruitful cooperation with the Preservation of Bhutan’s Written Heritage, directed by Karma Phuntsho. 14 For a discussion of these two sets in general, and for preliminary findings regarding the history of the transmission of the gDong dkar la set in particular, see Almogi 2015. 15 A detailed catalogue of the Nubri and NAK sets is nearly completed. Preliminary title lists of both the gDong dkar la and the Khams editions have also been compiled, and detailed catalogues of both are currently under preparation. 16 I may add that we currently know of several other sets, which are unfortunately not accessible thus far. For some examples, see Almogi 2015: 10, n. 25. 17 For more information regarding the sets listed here—including place of storage, publication details when applicable, details regarding their digitisation, existing catalogues, and accessibility—see the bibliography. 18 The identification of the set reported in historical sources as having been produced at the behest of Pad gling gsung sprul III with the sGang steng plain set is primarily based on my communication with the current sGang steng sprul sku. Moreover, as will be shown shortly, since the illuminated sGang steng set can very likely be identified with another sGang steng set reported in historical sources, which clearly describe it as being illuminated, the identification of the plain set with the one reported to have been produced at the behest of Pad gling gsung sprul III suggests itself rather forcefully. 19 Tshul rdor rnam thar (fol. 18b1–5): rje bla ma’i dgongs rdzogs su … gtso bor rnying ma rgyud ’bum pu sti bzhi bcu zhe drug tshang ba yig mkhan sogs rgyang ring mo nas bkug cing do gal bskyed de| rten bzhengs rnams ’khos* [= ’khur?] khyer gyis gang pher bsgrubs|. *This syllable seems to have undergone a correction in which the original final letter (unclear which letter it was) has been deleted and instead the final letter -s has been inserted, however, not exactly in its place but below the syllable (obviously due to lack of space). Note that the author of the Tshul rdor rnam thar is indicated by the TBRC (W8LS15799) as anonymous. However, it is made clear in the introductory part that this is an autobiography. This is also evident from the fact that the biography is written in the first person. 20 See Karma bde legs’s rGyud ’bum dkar chag (35.10–36.1), which cites an unspecified biography of Pad gling gsung sprul III Kun mkhyen Tshul khrims rdo rje as its source for this information (apparently merely relying on information provided by Karma Phuntsho). I have not been able to identify this source. Note that according to Karma bde legs (ibid. 35.17–18), the number of volumes are there said to be 45 (and not 46). One wonders, however, whether there has been a confusion here with the report on the sGang steng illuminated set (for the Tibetan text, see note 22). 21 The TBRC notes that 1726 as sGang steng sprul sku II’s year of death is questionable. However, 1726 is supported by his biography, which states that he died when reaching the age of 82. According to the Tibetan custom this is to be interpreted as his 82th year (i.e. when he was 81 years old), which, having been born in 1645, yields 1726. That he died in 1726 and not in 1727 (i.e. when he actually was 82) is further supported by the fact that his reincarnation, sGang steng sprul sku III Kun bzang ’phrin las rnam rgyal, was born in 1727 (TBRC: P3AG16). The exact time of his death is further specified in his biography as the midnight of the 15th of the 9th month (i.e. “the month in which the Buddha’s Decent Festival takes place”). See grangs brgyad cu gya gnyis bzhes pa’i skabs lha babs zla ba’i dus chen bcos lnga’i nam gung la mya ngan las ’das so||. The fact that sGang steng sprul sku II died towards the end of the Gregorian year (i.e. Nov. 9th) would mean that, even if the set was commissioned immediately after his death, its production must have stretched into 1727. 22 See the biography of sGang steng sprul sku II composed by rJe mkhan po X bsTan ’dzin chos rgyal (1701–1766/7; TBRC: P541), the sGang steng sprul sku gnyis pa’i rnam thar (208.6–12), which states: gsung rten kun bzang rdor sems kyi thugs bcud rgyud ’bum rin po che pusta ka grangs tshad bzhi bcu zhe lnga tsam yod pa| shog deb dang po’i rigs la mthing shog gser yig zhun ma las bgyis pa dag kyang dgongs rdzogs su dmigs te ’phral rang du grub bo|| de rnams kyang rgyal tshab sprul pa’i sku dang tshul ldan gyi skyes chen mang pos rab tu gnas par mdzad cing| slar yang sprul pa’i sku myur ’byon gyi gsol ’debs thugs dam yang bskul bar byas so||. See also Karma bde legs’s rGyud ’bum dkar chag (36.1–8), which refers to the same source (relying, however, on Karma Phuntsho’s report). My identification of the rGyal tshab with Mi pham dbang po (as silently done by Karma bde legs) is based on various passages in the sGang steng sprul sku gnyis pa’i rnam thar, including (196.12–13): … mi ring bar rje rgyal ba’i sras mi pham dbang po nyid rgyal tshab tu mnga’ gsol ba’i dgongs pas|…, and (206.8–9): … rgyal tshab sprul pa’i sku rje btsun mi pham dbang po…. Note that Karma bde legs states that it appears that the sGang steng illuminated version was made on the basis of the plain one. Also note that Cantwell and Mayer point out an oral tradition prevalent in sGang steng (based on oral communication with Karma Phuntsho) and conclude that the second illuminated set was produced in Me ri dkar po, a temple now lying in ruins above Nor bu lding across the pass from sGang steng. Cantwell & Mayer 1997: 68–69. 23 Ngag dbang ’brug pa’i rnam thar (561.3): gzhan yang rnying ma’i rgyud ’bum ’di dkon par gzigs te| sku gzhogs spungs thang nas ma dpe g.yar po zhus| shog bu’i rtsol ba mdzad de phral du sgrub pa gnang|. Dan Martin, in his unpublished notes, refers to another version of this same biography. 24 The passage provided by the TBRC entry of the Sangs rgyas gling edition (for which see note 25) reads rab byung bcu gcig pa (i.e. the 11th sexagenary cycle), which would place the year zil gnon me rta noted there in 1666. I tentatively suggest emending the text to rab byung bcu gnyis pa (i.e. the 12th sexagenary cycle), which would yield the year in question as 1726, and which, considering the fact that the birth year of mTshams brag sprul sku Ngag dbang ’brug pa, the commissioner of the set, is 1682, would make much better sense. This would mean that the production of the set had taken about two years (from 1726 to 1728), which could be considered long compared to the production of other sets, which reportedly took only several months. To be noted, however, is the remark found in the same passage that the production took many years (mi lo du ma’i ring bzhengs par grags), which may suggest that the author of the passage indeed took 1666 as the year in which the production started (so that the production stretched over 62 years, which is indeed a long period!). Mi nyag Thub bstan chos dar, who has obviously paraphrased and slightly shortened this same passage, notes the year in which the production started as 1728. This discrepancy may well be an attempt by Thub bstan chos dar to make up for what he probably saw as a problem in the dates provided in the passage (he indeed completely omits the sentence that identifies 1666 as the year in which production started). See the mTshams brag rnying rgyud dkar chag (16.20– 21: …rab byung bcu gnyis pa sa sprel spyi lo 1728 lor bzhengs pa’i ’go brtsams|…). The uncertainty about the dates of production are also reflected in Karma bde legs’s catalogue, which identifies the year in which the project began as 1725 and the year in which it ended as 1748, just before mTshams brag sprul sku’s death. See the rGyud ’bum dkar chag (36.12–13): …lo 1725 la dbu btsugs te 1748 lo zhing du ma gshegs gong tsam du grub pa dang|…. Note that Dan Martin, in his unpublished notes, suggested that the set was produced around 1730. 25 TBRC: W1KG16449, “Authorship Statement”: mtshams brag rnying ma rgyud ’bum ni| dpal sgang steng dgon pa’i sprul sku [b]stan ’dzin legs pa’i don grub kyi zhal slob mtshams brag sprul sku ngag dbang grub pas rab byung bcu gcig [= gnyis?] pa’i zin [= zil] gnon me rta’i hor zla dang po’i rgyal ba gsum par dbu btsugs| rab byung bcu gnyis pa sa sprel spyil [= spyi] lo 1728 lor rgyud rgyal legs par grub pa ste | mi lo du ma’i ring bzhengs par grags| pusti 46 dang| chos tshan 904| ldeb 40399 bzhugs| ldeb ’bring bris ma| dbu ldeb gser bris ma yin| bris gzugs sogs spus ka shin tu legs| dkar chag mi bzhugs| sde dge’i rnying rgyud la mtshon na| dang po gnyis med kyi rgyud sde a ti yo ga’i skor| de la’ang nang gses kyi yang ti’i skor dang| spyi ti’i skor| man ngag gi sde’i skor| klong sde’i skor| sems sde’i skor| gnyis pa ma rgyud a nu yo ga’i skor| de la’ang nang gses kyi rtsa ba’i mdo bzhi’i skor dang | mtha’ drug gi rgyud kyi skor| gsum pa pha rgyud ma hā yo ga’i skor| de la’ang nang gses kyi sgyu ’phrul gyi skor dang| tantra sde bco brgyad kyi skor| sgrub pa bka’ brgyad kyi skor bcas kyi rab dbye gnang yod||. Compare Mi nyag Thub bstan chos dar’s modern catalogue to the mTshams brag edition from 2009 which provides the same passage in a slightly paraphrased and shortened form. See the mTshams brag rnying rgyud dkar chag (16.18–17.9). 26 As I have already noted in a recent publication (Almogi 2015: 10, n. 26), the Sangs rgyas gling edition is referred to by the TBRC (W1KG16449) as mTshams brag dgon pa’i bris ma (under “Bibliographical Title”) and it is only under “Other Title” that it is recorded as the “Sangs rgyas gling manuscript of the rNying ma’i rgyud ’bum.” There I have suggested that the (somewhat misleading) title mTshams brag dgon pa’i bris ma was apparently given on the basis of a reference to mTshams brag in what seems to be the colophon to the entire collection provided on the TBRC entry under “Authorship Statement” (for the text see note 25). The source of the passage is not indicated in the TBRC entry. Kelsang Lhamo from the TBRC was not able to name its source either (email communication from 29.02.2016). It should be noted, however, that judging by some apparent modern influences on the style of writing (e.g. providing the equivalent of the Tibetan year in the Gregorian calendar), the passage seems to have come from an external source rather than being the original colophon of the collection. Moreover, the fact that it states that there is no dkar chag also supports the assumption of an external, later source. Note also the fact, likewise supportive, that it is included in Mi nyag Thub bstan chos dar’s modern catalogue to the mTshams brag edition from 2009 in an almost verbatim form (see note 25), that is, several years before the digitisation of the Sangs rgyas gling set. In any case, the passage obviously reports on the production of the mTshams brag set and not of the Sangs rgyas gling one. Whether the Sangs rgyas gling set is a mTshams brag set that later on was transported to Sangs rgyas gling, or whether it is a copy of the mTshams brag set, with the passage in question reporting the production of its master copy, is yet to be clarified. 27 Cantwell 2002. 28 See the gTing skyes ’khrungs rabs rnam thar (49.6–50.1). See also Martin (unpublished). 29 Almogi 2015: 3–7.30 Some of these glosses have been discussed in Almogi 2015: 3–9. 31 The designation of this group as “(Central Bhutanese)–Eastern Tibetan” is tentative and is based on preliminary study of the edition. 32 The TBRC also reports on a rNying ma rgyud ’bum set from gZhi chen monastery (W2PD17382). However, this set is said to consist of 39 volumes. Since the TBRC has not released the scans yet I have not been able to compare the two sets and to verify whether they are one and the same set or different ones. Nonetheless, shortly before the publication of the present paper I have been able to obtain the first three volumes of the gZhi chen dgon set with the kind help of Kelsang Lhamo from the TBRC. A brief look indeed confirmed my initial intuition that the two are in a way the same set, as these first three volumes of the “gZhi chen dgon set” turned out to be identical with folders 8, 10, and 23 of the “Khams set,” respectively. Since according to Karma bde legs (personal communication on June, 5, 2016) the original set indeed comprises only 34 volumes, it is possible that the 39- volume set reported in the TBRC consists of the 34-volume “Khams set” and additional supplementary volumes resulting from, most probably recent, compilatory work by Tibetan scholars in and around gZhi chen dgon (possibly merely scans of already published related material, as it is the case with folder 34). This matter will be looked into in the coming months.