Conclusion - Further Avenues of Research

Every age reinvents its National History to suit its audience and the changing sensibilities of its culture. In this thesis I demonstrate some of the indigenous and foreign influences on early Tibetan Buddhist historiography. I situated the earliest historical sources in chronological order to show how they transform the image of Khri Srong lde brtsan in Tibet. From the eighth to twelfth century, Khri Srong lde brtsan was apotheosised as a divine Buddhist king, then humanised as a deluded disciple of Padmasambhava.

Most subsequent Tibetan histories, even up to the present, share this more human depiction of Khri Srong lde brtsan. However, after the twelfth century the sense of national identity expressed in these histories changed enormously. Tibet was no longer a post-imperial culture looking to India for its Buddhism. The Dharma died out in India, and the Mongol empire offered Tibetan Buddhists a new means of both self-legitimisation and narrating the past.

In 1193 CE, the great library of Nalanda Monastery burned down. From the thirteenth century onwards, Buddhism in India waned and other religions gained the patronage of the kings of the Gangetic plain. Many of the Sanskrit versions of important Buddhist texts were lost, and the extant Tibetan versions now offer us the only way to reconstruct their contents. Tibetan histories provide two conflicting accounts of the Tibetan translation project. The dBa’ bzhed states:

All the Indian doctrinal scriptures without exception were fetched and translated, but the work could not be completed because in Sri Na len tra the Indian doctrinal texts were destroyed by fire.

In contrast, the twelfth-century ZL gives the impression that all the major Buddhist textual lineages were transmitted from India to Tibet in the eighth century:

The Sanskrit copies were kept in Sri Na len tra Monastery, and Master [[[Padmasambhava]]?] miraculously fetched them. Without leaving any in India, they were kept [as] the king’s copies in the bSam yas treasury.

ZL lists many higher tantras supposedly translated at this time, some of whose antiquity and Indic provenance are open to doubt. Though the above accounts represent the translation project as either partial or complete, both assert that early Tibetan Buddhists collected texts from Nalanda and created a repository of translated copies in Tibet.

After the Great Fire of Nalanda, Tibetan monasteries became the main repositories of tantric Buddhism. Having received much of its Buddhism from the south, Tibet then began to communicate its learning to the north. Tibetan Buddhist schools, beginning with the Sa skya in the thirteenth century, transmitted Buddhist lineages to the Mongols. Like the earlier btsan pos, the Mongols had both the wealth and imperial power necessary to patronise and further spread Buddhism outside of India. I would argue that this marked an important shift in the status of Tibetan Buddhism in Central Asia, akin to a lineage-holder’s transition from disciple to guru. This change influenced Tibetans’ self-representations in their own literature. For instance, the letter from ’Phags pa to Khubilai Khan (c. 1255-1259) begins (to quote Davidson):

‘We are happy to have heard that the Prince-bodhisattva's noble figure is well and that his august activity extends everywhere. We, the righteous recipients of your generosity, are also well. You have looked on all with your great gracious love and have extensively acted with the intent to benefit generally both the kingdom and the Buddha's doctrine. But especially you have included even lowly persons like us into your inner circle (lit., heart's mandala). Therefore, your speech has been like a stream of nectar. Moreover, as we have found the finer things, complete in all requisites, come into our possession by the power of your intention to invest us with them, our happiness has naturally increased.'

Compare this to the opening lines of Buddhaguhya's letter to Khri Srong lde brtsan, which tradition ascribes to the eighth-century:

You have sent [the religious envoys] Era, Aro, Manjusri, and retinue, with the best of wealth—silver and gold—to seek the Holy Dharma of India, that they might open a window to illuminate the deep darkness of Tibet.

As the veritable Buddhaguhya (one who's secret is the Buddha), it gladdens my heart that the Meridian of royal Authority in the world, the one who has straightened the crooked ways of power within his administration, the Supreme Lord in an unbroken stream of divine manifestations, the lord Trisong Detsen should order thus...

'Phags pa is writing within a tradition of letters. As Davidson notes, this dialogue between religious figures and royal patrons is ‘reflective of other discussions stretching back through the dialogues between kings and counsellors, the Buddha and Bimbisara, the monk Nagasena and the Indo-Greek ruler Menander.' The depiction of the btsan pos in Tibetan literature undoubtedly affects the Mongolian representations of their khans. For instance, Mongolian marriage rites portray Chinggis Khan as an archetypal husband on a par with Srong btsan sgam po.

Reciprocally, Mongolian literature offered new, foreign paradigms of national identity, power and patronage to Tibetan religious leaders and later historians. Mongolian forms of writing influenced histories, such as the Deb ther dmar po. Also, some schools jostling for Mongolian patronage rewrote their histories to legitimise their own Indian lineages. Others sought to free themselves of foreign influence by emphasising Tibet's indigenous traditions (for instance the rGyal rabs gsal ba'i me long). These interesting metamorphoses are beyond the scope of this thesis, but worthy of further study.

This thesis limits itself to eighth to twelfth-century historical sources. This restriction was necessary in order to do justice to these texts in the space allowed. Yet the depictions of Khri Srong lde brtsan in later histories would reward further investigation. Most reproduce one or more of his portrayals from earlier works, while some reject or rework them, leading to interesting new representations. sBa bzhed S, for instance, constitutes a condensed narrative on the reign of Khri Srong lde brtsan, to which its redactors probably added the zhabs btags end section. There is clearly a need to identify the reasons for its omissions, and unpack the details of its lengthy final addition. Willis and Gonkatsang are completing a detailed analysis on the dBa' bzhed. I hope that this will improve understanding of its relation to the later sBa bzhed tradition.

Similar in-depth text-critical research is possible in the field of Padmasambhava hagiographies too. As I demonstrate in Chapter Two, later hagiographers incorporated ZL into their bKa' thangs in different ways. Their adaptations of the ZL narrative affect their representations of Khri Srong lde brtsan. For instance, O rgyan gling pa rewrites parts of ZL and includes extra episodes from other sources in his fourteenth-century Padma bka' thang. Sangs rgyas gling pa, in the slightly later gSer phreng, follows ZL more closely than the Padma bka’ thang does. Yet he synthesises the two texts and modernises O rgyan gling pa's old orthography. Perhaps he did this to avoid the disparity between the Padma bka' thang and ZL. These changes thus give us an insight into the changing cultural presuppositions of Tibetan authors down the centuries. However, creation in bKa’ thang literature is generally a process of continual accretion, rather than deletion. These hagiographies grow in size and number after the twelfth century, generating enough data to fill lifetimes of Tibetological research.

This thesis identifies some of the literary genres that shape eighth to twelfth-century histories. Yet there is much more work to be done, for instance, to secure the dates and provenance of many of the important tales told about Khri Srong lde brtsan.

Chapter Four discusses the beginnings of the Buddhist depiction of Khri Srong lde brtsan. The literary form of the Annals owes a debt to Chinese historiography. The Annals’ entries also maintain the closest relation to everyday events in the btsan po’s inner circle of any of our sources, except perhaps Khri Srong lde brtsan’s bKa’ mchid. The bKa’ mchid attests partial authorship by Indian Buddhist spiritual friends, perhaps even Santaraksita. These external influences lead us to question the extent of Khri Srong lde brtsan’s involvement in the conversion of Tibet to Buddhism. It is perhaps necessary to interrogate the texts attributed to him and those in the lHan kar ma catalogue in order to assess what kind of Buddhism Khri Srong lde brtsan endorsed.

Both indigenous and Indian narratives also influence later histories’ accounts of the imperial period. The texts found in Dunhuang offer much insight into the mix of national literatures affecting their portrayals of the imperium. The Old Tibetan Chronicle, for example, eulogises both the religious and imperial achievements of Khri Srong lde brtsan. These appear to be the btsan po’s abiding legacy in the late or post-imperial period. The Chronicle fashions a creative precis of Khri Srong lde brtsan’s life, reflecting the aspiration of some Tibetan historians in this period to make the reign of Khri Srong lde brtsan part of an epic narrative of the imperial period.

What do these Dunhuang texts, walled up at the edge of a crumbling empire, tell us about the central Tibetan view of its military power and assimilation of Buddhism? The collection contains Annals from at least 650 CE onwards, emanating from the centralised court of the btsan pos. But they were deposited together with late tenth-century biographical sketches such as Pelliot tibetain 849, which includes an idealised representation of the bodhisattva-kings of Tibet. Future research on the palaeography of the texts may increase our understanding of who was writing what and why; which texts interested certain specific scribes, and to what extent they adapted their sources as they transmitted them. The evidence in this thesis suggests that they were written at a crossroads, at a time of great creativity, rather than in a backwater, in a dark age.

The second diffusion (phyi dar) of Buddhism, from the tenth century onwards, brought with it the rise of religious power in Tibet. This period also witnessed the growing influence of the decline narrative and cult of Padmasambahava on Tibetan historiography. As a result, the eleventh-century dBa’ bzhed goes beyond the ninth-century laudatory descriptions of Khri Srong lde brtsan. It depicts him occasionally as inferior in status to the Buddhist masters that he invites to Tibet. The sBa bzhed's redactors interpolated additional episodes into a pre-existing narrative that resembles the dBa’ bzhed, rather than create a completely new history on the imperial period. The sBa bzhed thus retains many imperial-era metaphors, but its depiction of Khri Srong lde brtsan is dominated by later religious concerns. These different themes remain to be disentangled, and their dates and provenences assessed. It remains a highly influential history, which has just been republished in China for a 21st-century readership. This suggests that these concerns continue to resonate with Tibetan audiences today.

For his hagiography of Padmasambhava, Nyang ral collected together earlier written and oral traditions concerning Khri Srong lde brtsan that were popular in his day. It appears that he used these narratives not only to glorify Padmasambhava, but also to legitimise his own religious tradition. In ZL, Padmasambhava transmits Indian Buddhist lineages to the king. When Padmasambhava prophesies that Khri Srong lde btsan will be reborn as Nyang ral, Nyang ral is then uniquely positioned to become heir to those teachings. Chapter Two highlights what may be later interpolations into ZL.

However, more work remains to be done on the earlier strata of narrative on Padmasambhava contained in ZL, dating between the ninth and eleventh centuries. Nyang ral undoubtedly based ZL on these earlier traditions, while adding his own elements. His source may have been a written exemplar, such as an eleventh-century biography by Rong zom, or oral narratives arising from the growing cult of Padmasambhava. Until these precursors come to light, it is not possible to establish the extent to which ZL represents a synthesis of both written and oral traditions, a rejection of one tradition in favour of the other, or a largely unprecedented creation by Nyang ral alone.

MTN contains a far longer history of the Dharma than ZL. As a chos ’byung rather than a rnam thar, it casts its net wider to include more of the historiographical tradition. The narrative includes earlier cosmogonic and national myths, as well as reference to many Tibetan Buddhist schools and even Bon. It is thus possible to position Nyang ral’s hagiography and history respectively at different poles of the local / universal spectrum.

Furthermore, Nyang ral did not merely copy the words of ZL, this “second Buddha’s autobiography” verbatim into MTN. He evidently followed some other textual criteria, even sometimes contradicting ZL’s account of Khri Srong lde brtsan. Future research might seek to assess the underlying historiographical tradition that surfaces in Nyang ral’s texts. This may also explain how later redactors approached his histories.

In those of Nyang ral’s historical works discussed above, Padmasambhava is always the key to Tibet’s salvation. Khri Srong lde brtsan’s intention to build bSam yas does not detract from his comparatively deluded character and inferior status. It is tempting to speculate that this portrayal stems from Nyang ral’s need to stand out in the crowded marketplace of competing lineages in Tibet. He perhaps found in Padmasambhava a “culture-hero” far better suited than Khri Srong lde brtsan to attract followers from both old and new tantric traditions. Padmasambhava is a siddha, like Nyang ral. ZL’s image of him, as a sorcerer emanated from Amitabha, could appeal to people who value magical displays and/or Buddhist devotion. This depiction was so successful that it influenced the later portrayal of Padmasambhava, and his royal disciple Khri Srong lde brtsan.

From the twelfth century onwards, Khri Srong lde brtsan is generally depicted as subservient to his contemporary religious masters. This is due to the growing influence of Buddhism in Tibet. The imperial documents contain many epithets for Khri Srong lde brtsan, most of which were gradually forgotten after the fall of the empire. Later histories had no reason to perpetuate due deference to the btsan pos. The cult of religious figures was growing and devotion to the bla ma, especially Padmasambhava, took priority. It was inconceivable to dPa' bo gTsug lag phreng ba, for instance, that Padmasambhava would have bowed to a symbol of mundane power (see Chapter One, above), even if the king was also nominally an incarnation of Manjusri. As this thesis demonstrates, charting the changing depiction of Khri Srong lde brtsan in histories requires us to also tell the story of Santaraksita, dBa' gSal snang or Padmasambhava. The analysis that I carried out on Khri Srong lde brtsan is perhaps a desideratum for all major figures that influenced the path of Buddhism in Tibet. Identifying the conflicting trajectories of their portrayals over time would enable us to delve deeper into the culture of the text in Tibet, as well as understand historiography's increasingly religious view of the past.

Appendix


The Eleven Exemplars of the Zangs gling ma
For each version analysed, I list the following information.
1) Title of the text
2) Publication or location details
3) Physical description of the text:
a) Number of folios and lines per side
b) Description of the paper and script
c) Illustrations
d) Use of red ink
e) Use of shad marks
f) Chapter breaks
g) Interlinear notes
4) Remarks on indicative errors, etc.
ZLa
1) Slob dpon padma 'byung gnas kyi skyes rabs chos 'byung nor bu'i phreng ba zhes bya ba bzhugs so / rnam thar zangs gling ma'o.
2) Ngodrup and Sherab Drimay (eds) 1976-80. Rin chen gter mdzod chen mo: a reproduction of the Stod-lun Mtshur-phu redaction of 'Jam-mgon Kong-sprul's great work on the unity of the gter-ma traditions of Tibet, with supplemental texts from the Dpal-spuns redaction and other manuscripts. Paro.
3)
a. 102 folios; 6 lines per side, except for folios 1b-2a (5 lines each).
b. A relatively clear and well presented dbu can block print. The inking or cutting is too light at times, so it is difficult to distinguish tshegs, pa from ba, kyi from gyi, da from nga etc.
c. The text contains no illustrations.
d. The printer uses black ink throughout the text.
e. The block cutter cuts gter ma marks in place of shads throughout the text. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the text's alternative title and the chapter colophon, for instance: 'The first chapter from the Stainless Life Story of Master Padma[[[sambhava]], which tells the tale] of King Indrabhuti acting as [his] father and [his] ascension to the throne.' Then follows a shad, a space, and then another shad to mark the beginning of the next chapter.

g. There are almost no corrections, except for the very occasional addition of a ba prefix before a word that is mistakenly missing one. This suggests that the blockprints were checked carefully for major errors, the blocks on which mistakes occurred being completely re-cut. The only interlinear note appears to be a gloss giving a synonym of the text's name to which it is appended. ZLc 34b5 does not include this gloss, either in the main text or as an interlinear note, so it may be unique to this version of ZL.
4) There are insufficient errors in ZLa to aid analysis or help to distinguish it from other versions. This is probably due to a long tradition of editing this tradition of ZL in order to maintain a meaningful narrative through multiple stages of transmission. However, some trivial errors remain. For example, 20b4 reads re shig instead of the correct re zhig. It shares this mistake with the very similar ZLc, and has failed to edit it out using common sense. These tiny errors of ZLa, generally made with far less frequency than one per folio, attest to 'Jam mgon Kong sprul's scholarly abilities in finding a reliable version of this redaction and seeing it into print with minimal additional errors.
ZLb

1) Slob dpon padma 'byung gnas kyi skyes rabs chos 'byung nor bu'i phreng ba zhes bya ba bzhugs so (rnam thar zangs gling ma).
2) Thub stan Nyi ma (ed.) 1989. Slob dpon padma'i rnam thar zangs gling ma Si khron mi rigs dpe skrun khang: Chengdu. 1-193. This book also contains a brief biography of Nyang ral Nyi ma 'od zer (pp. 1-4), three gsol ’debs-s (pp. 194-238) and the life-story of Padmasambhava according to Indian sources, by Jo nang rJe btsun Taranatha (pp. 239-87, probably also based on the Rin chen gter mdzod version).

3)
a. 193 pages, 17 lines per page.
b. dbu can computer type in book format, which therefore retains a uniformly high degree of clarity throughout.
c. The book contains a colour picture of a seated Padmasambhava after its title page.
d. The printer uses black type throughout the text.
e. The editor uses gter ma marks in place of bshads throughout the text. He adds
a tsheg before a shad only after nga.
f. Each chapter ends with the text's alternative title and the chapter colohon (see ZLa, above). Then follows a shad, a paragraph break, and an indent on the next line.

g. There are no footnotes or endnotes to the text.
3) ZLb shares many of its readings with ZLa against ZLc. For instance, ZLb 61.14 contains ZLa's gloss rig pa'i mdo in parentheses after the main text from ZL1. Such details suggest that ZLb is based on ZLa, at least as one of its sources. Thub stan Nyi ma appears to have edited the text, so as to be grammatically correct or more intelligible to the general Tibetan reader. In one example of the latter practice, he has changed ZLa's bod kyi lha sras btsad pos, shared with ZLc, to the more common ...btsan pos (39.7). There are also unique mistakes in ZLb, for instance yid ma chas for yid ma ches (41.6) or lang du byon for lung du byon (43.6), but mistakes like these appear to be type-setting errors, causing the omission of the vowels in each of the above cases.

Where ZLa and ZLc read re shig (se above), ZLb reads re zhig. I assume this is because Thub stan Nyi ma corrected ZLa to create ZLb.
ZLc
1) Slob dpon padma ’byung gnas kyi skyes rabs chos ’byung nor bu’i phreng ba = rnam thar zangs gling ma.
2) Theo Sorensen procured this version from sDe dge in the early twentieth century, and donated it to the Oslo University Library, Norway, where it is now held under Per Kvaerne's catalogue no. 152.
3)
a. 115 folios; 6 lines per side, except for folios 1b-2a (5 lines).
b. dbu can blockprint on quality, thin, fibrous, yellowed paper. The same block cutter appears to have cut all the texts in this dpe cha; suggesting that this blockprint is a special collection rather than one utilising other blocks (from, say, the Rin chen gter mdzod chen mo). Ink blots from the printing obscure some graphs (e.g. 76a3-4). Occasionally the faint writing makes distinguishing divergences from ZLa difficult.
c. The text contains no illustrations.
d. The printer stained the ragged edges of the dpe cha pages purple for decoration; but uses only black ink throughout the text.
e. The block cutter cuts gter ma marks in place of shads throughout the text. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the text's alternative title and the chapter colophon (see ZLa, above). Then follows a shad, then a diamond symbols with circles at each corner, then another shad to mark the beginning of the next chapter.
g. ZLc contains no interlinear notes.
3) Comparing ZLa with ZLc rules out the possibility that the former is based upon the latter. However, there are too few indicative errors in ZLa to establish whether it is the exemplar for ZLc or not. ZLc may be a variant holder or possibly even contain readings that predate those in ZLa. However, it agrees overwhelmingly with ZLa against ZLe, below, and given the likelihood that Kong sprul edited ZLa for inclusion in the Rin chen gter mdzod (in such matters as the spelling of names etc.), it appears likely that ZLc is based on the more famous ZLa.
ZLd
1) Slob dpon padma 'byung gnas kyi skyes rabs chos 'byung nor bu'i phreng ba bzhugs so / rnam thar zangs gling ma'o.
2) Khamtrul Rinpoche (ed.) 1972. The Life of Lady Ye shes mtsho rgyal Rediscovered by Stag sham Nus ldan rdo rje: with two hagiographies of Padmasambhava from the terma finds of Nyang ral Nyi ma 'od zer and A 'dzom 'Brug pa 'Gro 'dul dpa' bo rdo rje. Tashijong: Sungrab Nyamso Gunphel Parkhang. 361-595. This book also contains a seventeenth¬century biography of Ye shes mtsho rgyal (pp. 1-359), two gsol ’debs-s (pp. 596-613) and the life-story of Padmasambhava according to A ’dzom ’Brug pa ’gro ’dul dpa’ bo rdo rje (pp. 615-31).
3)
a. 235 pages; 6 lines per side, except for pp. 362-63 (folios 1b-2a) with 4 lines.
b. A generally clear photo reprint of an dbu can blockprint that differs from both ZLa and c, with four verso pages of the blockprint on each recto leaf of the book, and four recto pages on each verso leaf.
c. The text contains no illustrations.
d. The text is printed in only black ink.
e. The block cutter cuts gter ma marks in place of shads throughout the text. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the text's alternative title and the chapter colophon (see ZLa, above). Then follows a shad, a space, and then another shad to mark the beginning of the next chapter.
g. ZLd contains no interlinear notes.
3) E. Gene Smith believed that ZLd is dependent on ZLa, though he did not give any evidence in support of this statement. Analysis of the indicative errors in the text appears to confirm this. ZLd only agrees with ZLc against ZLa coincidentally.
ZLe
1) Slob dpon padma'i rnam thar dri ma med pa bzhugs so.
2) Text no. 686 of Tucci Tibetan Fund II, housed in the Library of the Istituto Italiano per l'Africa e l'Oriente (IsIAO), Rome.436 437
3)
a. 151 folios, 4 to 6 lines per page.
b. This manuscript contains two hand-written styles of clear dbu can on three different types of paper. The scribe of 144 folios writes large and square letters, filling between four and five lines per page; but occasionally the words on last line are extremely spread out—for example 13 syllables per line rather than 32 (ZLe 26a5). Some folios contain red guidelines and borders (3a-17b, 21a-27b and 43a-146b) and others are without either (28a- 42b). The folios are old, stained and burnt at the edges, but the script is still clear.
Some other folios (1a-2b, 18a-20b and 147a-151a) display signs of a different scribal hand. I suspect that this scribe wrote these folios to replace folios missing from the original through damage. Folios 3-17 and 21-135 show signs of fire damage on the right-hand side of each page; folios 110¬146 literally show signs of wear and tear on the left-hand side of each page. The other scribe attached pieces of paper to each side of the page, exhaustively but without finesse, in order to make the dpe cha square again. She/he then wrote in the text that was destroyed or covered by the pieces of paper, in tight, vertical handwriting. She/he also rewrites the content from the missing folios over strong blue guidelines and “copy book” margins; with many more syllables per line and 5 to 6 lines per page. At the beginning she/he copies from the remains of the folios that are now missing (replaced by the second scribes' folios), at the end from a later Padmasambhava hagiography entirely. The earlier added folios are on rough paper, the final folios on shiny paper, but both appear to be by the same hand.
c. Folio 1a, the title page that the second scribe created, contains bright multi¬coloured borders and two wish-fulfilling jewels (yid bzhin nor bu) in blue, red and yellow. The mantra @@@ // na ma rat na ghu ru / fills folio 1b, in red script surrounded by a blue border, and with another wish-fulfilling jewel beneath the @@@. The scribe has stitched a protective blue cloth over the mantra using red thread.
d. The second scribe writes folio 2a lines 1 and 3 (of 3) in red, but nothing else. The first scribe's extant folios contain no red script. Every chaper colophon is underlined in red, but I suspect that Tucci or the IsIAO librarian did this.
e. The first scribe uses both single and doble shads; the second scribe uses only single shads, except at the end of a chapter. Neither uses gter ma marks instead of shads.
f. Neither the first nor second scribe's chapter colophons contain a short or secondary form of the main title. Both mark the end of chapters with a double shad, a space and another double shad to mark the beginning of the next chapter. The first scibe adds a complicated, vertically stacked symbol in the space between chapters (see e.g. 9a3). I believe that a later person, perhaps Tucci or an IsIAO cataloguer underlines the colphon in red, then adds the number of the next chapter (ibid: ‘ (2 ' ) in pencil.
g. The first scribe adds numerous interlinear notes, marked with a swastika, where she/he has noticed omissions. They range from single words (e.g. 104b: srin mo) to two lines of smaller script (to be added in place of a scribbled out phrase at 13b3). The second scribe adds no interliner notes.
4) The transmission of this text is replete with changes to the text. The whole text, while generally following ZLa-d against ZLf-k, diverges from these three examples on every line from title onwards. The first scribe follows these mistakes accurately and without hypercorrection.
The second scribe also makes or copies many uncorrected mistakes, mostly of omission, and especially towards the end of his text. For example, at 144a2 she/he misses out several lines of text contained in e.g. ZLa 98b5-99a6. Then at 144a 5 she/he misses another line that once marked the end of chapter 40 (compare with ZLa 99b1-2). This omission means that Tucci reports that ZLe contains 40 chapters, as the final chapter colophon affirms, when it actually has 41 chapters (like ZLa-d) but has been mismarked.
The second scribe does not correct this latter mistake, but instead compounds it by seemingy basing the final folios on a (later?) Padmasambhava hagiography that conatins far more information than any ZL version. It is likely that the final folios of the original manuscript were not available to the second scribe, and nor was another ZL exemplar. I have not yet discovered the source of these final folios.
Where the second scribe bases his added folios on the missing original ZLe folios he then replaces, he also omits large parts of the narrative. So, between 17b4 and 18a1 the entire episode in which Padmasambhava converts Uddiyana and his father Indrabhuti (ZLa 12a2- 13a5) is missing. Perhaps these folios of the original manuscript were also unavailable to the second scribe, who makes his own segue into the story of Padmasambhava battling 500 heretics instead.
All these errors preclude the possibility that ZLe was the exemplar for any of the other ZL versions. The large degree of divergence between it and them also strongly suggests that it was not based on any of them. I believe that the time gap between Kong sprul's creation of the Rin chen gter mdzod in the nineteenth century and Tucci's mention of his version in 1949 is too short to allow for this many transmissional errors to enter the text.
ZLf
1) U rgyan gu ru padma ‘byung gnas kyi rnam thar ‘bring po zangs gling mar grags pa.
2) The Nepal German Manuscript Cataloguing Project (NGMCP) microfilmed this manuscript from the Kathmandu National Archive collection in 1989 (reel AT 28/2). The original is still held at the Kathmandu National Archive.
3)
a. 148 folios; 5 lines per side, except for folios 1b-3a (3 lines) and 3b-4a (4 lines). The scribe occasionally leaves lines blank (e.g. 100b).
b. The dpe cha was in good condition at the time of filming, except for a few stains and worm holes. The scribe writes in a readable dbu can, following his marked lines but often transgressing the right-hand-side margin. The predominance of cig for zhig/shig/gcig in ZLf (and ZLg) suggests that this recension was at one point transmitted through an dbu med manuscript exemplar, wherein "one" in all its forms tends to become the numeral "1".
c. There is a pair of named colour portraits on each of the first two folios, of the Buddha and Avalokitesvara (1a), and of Padmasambhava and Ye shes mtsho rgyal (2a) respectively.
d. The scribe stained the edges of each page with ink. She/he also writes om mani pad me hum and some other parts of the first page's prayer in another ink. She/he does not use another ink again until folio 71a1, where she/he writes the first few words of chapter 17 (de nas yang rgyal po'i) in another ink.
e. The scribe writes each shad with a clear gter ma mark throughout, except once (83a2), where the shad resembles: 1. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the chapter colophon only (unlike ZL1). Then follows a shad, a space, then another shad to mark the beginning of the next chapter.
g. The scribe often writes interlinear notes, occasionally quite long (e.g. 23b- 24a), to replace omissions in the text. These additions are probably from the exemplar, which the scribe mistakenly omitted, since the scribe is quite prone to misspellings and repetitions as well (e.g. 26b3-4). It is possible, though, that the scribe adds these interlinear notes from another exemplar, but unlikely since she/he never adds alternatives, only omitted text.
4) Unnoticed omissions include the line rgyal po'i nam bza' bzhes phyag 'tshal, missing from Padmasambhava's speech to Khri Srong lde brtsan (45b2). This line is present in ZLg 40b2, and in all other versions, which strongly suggests that ZLf was not the exemplar of ZLg (with which it shares most similarities).
ZLg
1) 'O rgyan gu ru padma ‘byung gnas kyi rnam thar ‘bring po zangs gling mar grags pa.
2) The NGMCP microfilmed this manuscript from the collection of a certain "Lobsang" in 1980 (reel E 1125/5) and holds it in copy at the Kathmandu National Archive.
3)
a. 187 folios (folio 8 missing, the folio numbers "17" and "85" are each given to two consecutive folios by mistake); 6 lines per side, except folios 1b, 2a and 3a (5 lines).
b. This handwritten dbu can manuscript is in bad condition. Its pages are cracked, obscured and soiled, especially in the first few folios. A hole or tear has been covered with rectangular pieces of paper on both sides of folio 62, obscuring some graphs.
There is no fault in the legibility of scribe's writing; the graphs are large and tidy. However, she/he writes some graphs over a line (e.g. m-dzad 7a4-5) and tends to place the vowel marker too far back on the graph, so that many times, for instance, bde resembles bed. She/he also appears to write du for tu (but not vice versa) and nga for da (again, not vice versa, but this could be partially due to the quality of the microfilm obscuring the tail of da).
c. The text contains no illustrations.
d. The scribe uses another colour of ink to write more than half of folio 1b1-3, and then a lot more regularly than ZLf—showing up on the microfilm as extremely faded and often illegible. Different ink marks important parts of the narrative, including the beginnings and ends of some chapters, names of important texts or people—but inconsistently. Sometimes the scribe even writes only half a graph in another ink (see, e.g., mdun on 57a3).
e. The scribe generally uses gter ma marks for shads, but occasionally, when a shad is used to break up items in a list, it resembles three vertically stacked dots or a circle above two vertically stacked dots (e.g. 72b). She/he is not very consistent about only adding a tsheg before a shad after nga, and so adds them after some other final letters too.
f. Each chapter ends with the chapter colophon only (like ZLf). Then follows a shad, then occasionally a complicated, vertically stacked symbol (see e.g. 11a2), then another shad to mark the beginning of the next chapter.
g. ZLg contains far fewer interlinear notes correcting omissions than ZLf. On folio 74a the scribe notices the omission of the line: yang rab tu gyur pas / bsam pa thams cad ’grub / dam pa’i chos nyi ma shar ba lta bur bsgyur nas bshad pas bka’ drin che / da yang bdag la tshe, through a skip of the eye, and so adds it several lines down in the main text rather than in the notes.
4) At some point ZLg was transmitted without being checked for major errors, and so it contains many more mistakes than ZLf. Such indicative errors rule out the possibility that ZLg is the exemplar of ZLf. However the two versions are closely related, as other shared errors and additions suggest (see below).
ZLh
1) Padma bka’ chems brgyas pa bzhugs sho.
2) The NGMCP microfilmed this manuscript at 'Khyen Tes Monastery' in 1992 (reel E 2703/10) and holds it in copy at the Kathmandu National Archive.
3)
a. 123 folios; 5 lines per side, except folios 1b-2a (3 lines) and 3a (4 lines).
b. Sometimes scruffy handwritten dbu med on very narrow dpe cha paper that has some stains but is in a generally good condition.
c. The text contains no illustrations.
d. The scribe writes some of the text in different ink, for example words like rgyal po, "king".
e. The scribe does not use gter ma marks in place of any shad. Generally she/he does not add a tsheg before any shad, even after nga.
f. Each chapter ends with the chapter colophon only (like ZLf and ZLg). Then follows a shad, a complicated, vertically stacked symbol (except, e.g., 19b3), then another shad to mark the beginning of the next chapter.
g. The text does not contain any interlinear notes.
4) There are no major corrections, additions or changes; in general this is one of the more faithfully transmitted ZL versions. However, between folio 87 and 88 several whole folios are missing, corresponding to the whole of chapter 23 and some of chapter 24 (wherein Padmasambhava leaves Tibet). Since the folio numbers do not alert us to this fact, it seems most likely that the folios were lost after they were written and before the manuscript was paginated.
ZLi
1) bKa’ thang zangs gling ma bya ba bzhugs so.
2) The NGMCP microfilmed this manuscript at ‘Khyen Tes Monastery’ in 1992 (reel E 1217/02) and holds it in copy at the Kathmandu National Archive.
3)
a. 102 folios (the title ‘folio 86’ is given to two different folios); 5 lines per side, except 1b-2a (3 lines) 2b-3a (4 lines)
b. Handwritten dbu can on a dpe cha that is in good condition and stained at the edges. The handwriting is readable and well-presented, often along very strong guidelines.
c. The text contains no illustrations.
d. The scribe stained the edges of the dpe cha pages another colour for decoration; but uses only black ink throughout the text.
e. The scribe does not use gter ma marks for any shad, but occasionally a shad resembles 1. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the chapter colophon only (like ZLf-g). Then follows a shad, a long space, then another shad to mark the beginning of the next chapter. At some point in the transmission a scribe (perhaps this one) mistakenly numbered chapter 41 as 'chapter 42' (100b3).
g. ZLi contains only minor interlinear notes; for example the addition of a missing du after the phrul [= brul (ZLh 32b4)] gyi 'og (27a5), ‘underneath rotten dough’.
3) While describing bSam yas Monastery, ZLi neglects to mention the queens and the temples which they built, and proceeds straight to a description of the chapels (27b4). Then, after describing the rdo rings with bronze dogs on top of them, it returns to the queens (28a3-5)— perhaps because a scribe noticed the mistake. ZLi shares this mistake with no other version of ZL, and so was probably not the sole exemplar for any of them.
An editor or learned scribe undoubtedly edited ZLi at some stage of its transmission. For example, ZLi differs from ZLh and ZLj-k in not following their shared mistake of locating Uddiyana in the south. An editor or scribe rectifies the line to correctly read ‘west’ (2b1). She/he also changes slob dpon to ghu ru (sic) almost throughout the text (not only in reference to Padmasambhava); and often rgyal po to rje (e.g. 52a2). These differences suggest that someone passed a rather idiosyncratic but critical eye over the text at some point. Perhaps they made these changes before the mistakes, mentioned in the previous paragraph, appeared—since no one corrected these errors.
ZLj
1) U rgyan padma 'byung gnas kyi rnam thar mnga' bdag nyang ral pas lho brag gi brag srin po'i sbar rjes 'dra ba nas gter nas ston pa bzhugso.
2) Dr. Karma Phuntsho recently photographed this manuscript in the Gangtey (Ganteng) Monastery in Bhutan, and catalogued it as Thorbu no. 196: “Pad 'byung rnam thar”.
3)
a. 157 folios; 5 lines per side, except folios 1b-3a (3 lines).
b. This Bhutanese manuscript is in very good condition. The handwriting is in large, neat and readable dbu can.
c. An artist painted eight untitled colour portraits at either side of the first three folios: Padmasambhava and Avalokitesvara (1b); a six-armed Vajrakila and Aparmita-Amitayus (2a); Yellow Manjusri and Blue Vajrapani (2b); Yellow Jambhala and Blue Jambhala (3a). She/he covered the first portraits, on folio 1b, with decoratively cut paper, and covered the writing on folio 1b and the portraits on folios 2a-3a with rough red cloth. There are also leather flowers on both sides of the first folio.
d. The borders of the portrait folios (1b-3a) are yellow. The scribe uses red ink extensively on folios 1b-2b, then occasionally throughout the text (to mark the beginning of chapters, etc.).
e. The scribe does not use gter ma marks in place of any shad. Some shads resemble 1. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the chapter colophon only (like ZL2 and 3). Then follows a shad, two symbols resembling ft, then another shad to mark the beginning of the next chapter. She/he also writes the first few graphs of the new chapter in red.
g. Folios 5a-6a contain an interlinear note, seemingly unrelated to the main text and in looser handwriting—though possibly by the same hand). The opening line suggests that it may contain one of Yama's mantras. There are no other interlinear notes in the text.
4) ZLj contains a number of indicative errors. For instance it omits rgya nag gi yul nas spyan drangs slob dpon ka ma la shi la spyan drangs / (81b3), which is present in, e.g., ZLh 65a4 or ZLk 42a3. The scribe makes less important mistakes than in some other versions (e.g. ZLg); but it appears someone has failed to pay close attention to the transmission of the text at some stage. The presence of these errors rules out ZLj as the lone exemplar of any other ZL version.
There is some slight evidence that ZLj may have maintained connections with ZL1 at some point in its history, because in very small ways it agrees with that recension rather than with versions ZLf-k. For instance, folio 81a2 reads ...snyams pa dang / de slob dpon gyis mkhyen nas / rgyal po'i thu bar..., retaining slob dpon gyis mkhyen nas like ZL1 against ZLf-i and ZLk that all omit this phrase. Such similarities could be coincidences, but if this is a case of influence, there are two options: ZLj may have been checked against one or more versions of ZL1, or one of the ancestors of ZLj may have been the basis of ZL1 before the latter added recensional changes.
ZLk
1. U rgyan gyi slob dpon chen po padma 'byung gnas kyi rnam par thar pa nyang gter zangs gling mar grags pa bzhugso.
2. Karma Phuntsho also photographed this manuscript in Gangtey (Gangteng) Monatery in Bhutan, and catalogued it as Thorbu no. 288: “Gu ru rnam thar zangs gling ma”. The same dpe cha also contains the Slob dpon gyis pa padma ’bymg gnas kyi nam (sic) par thar pa yid kyi mun sel by Sog bzlog pa bLo gros rgyal mtshan (1552-1624) in 124 folios (80aff.). dbu med is not often written in Bhutan, so the manuscript may come from over the border in Lho brag, Tibet.
3.
a. 79 folios; 6 lines per side, except for folios 1b (4 lines) and 2a (5 lines).
b. The dpe cha is very narrow, and the dbu med is compact and full of contractions (sdus yig). These factors combine to make a not easily readable text. The dpe cha is in good condition.
c. The text contains no illustrations.
d. The scribe uses red ink on folio 1b and to mark important names and the beginning of chapters elsewhere.
e. The scribe uses gter ma marks, without lines between the circles, to mark shads. Generally she/he adds a tsheg before a shad only after nga.
f. Each chapter ends with the chapter colophon only (like ZL1 and 2). Then follows a shad, a symbol resembling 1 beneath an arch, another shad, a space and then a third shad to mark the beginning of the next chapter (see 12a 3-4). She/he also writes the first few graphs of the new chapter in red.
g. The scribe adds occasional interlinear notes to correct mistaken omissions in the text. The longest of these is on folio 30a6, where someone adds, in dbu can, the text: bzhes pa sum cha gcig sol tsa na / ’bangs nang pa cig snyun gsos bas / bdag la ni bud med dkar mo cig byung nas thalcag cig rgyab pa snyam ’gyido ed pa cig bde’o roughly in line with other versions (e.g. ZLi 38b4-5).
3. Because the scribe corrected the finished text using interlinear notes, ZLk contains fewer indicative errors than ZLj, for instance. Like the other dbu med manuscript, ZLh, ZLk has been transmitted fairly faithfully.
Bibliography
Tibetan Language Works
dBa’ bzhed: 2000. dba' bzhed: sangs rgyas kyi chos su ji ltar 'byung ba'i bka' mchid kyi yi ge. In: Pasang, Wangdu and Hildegard Diemberger, Dba' bzhed: The Royal Narrative Concerning the Bringing of the Buddha’s Doctrine to Tibet. Wien: Verlag der Osterreichischen Akademie der Wissenschaften, 123-56.
rBa bzhed: Diji Zhougar (ed.). 2009. rba bzhed phyogs bsgrig. Beijing: Mi rigs dpe skrun khang.
sBa bzhed G: mGon po rgyal mtshan (ed.). 1980/1982. sba bzhed: sba bzhed ces bya ba las sba gsal snang gi bzhed pa bzhugs. Beijing: Mi rigs dpe skrun khang.
sBa bzhed P: verbatim quotes contained in: dPa' bo gTsug lag phreng ba. 2002. chos ’byung mkhas pa’i dga’ ston. Varanasi: Vajra Vidya.
sBa bzhed S: Stein, R. A. (ed.). 1961. Une Chronique ancienne de bSam-yas: sBa-bzed. Paris: Adrien-Maisonneuve.
Chos ’byung chen mo: Lde'u Jo sras. 1987. chos ’byung chen mo bstan pa’i rgyal mtshan lde’u jo sras kyis mdzad pa. Lhasa: Bod ljongs mi dmangs dpe skrun khang.
Dri med: Mi 'gyur rdo rje. 1976. sprul sku mnga' bdag chen po'i skyes rabs rnam thar dri ma med pa'i bka' rgya can la ldeb. In: Lama Sonam Topgay Kazi (ed.), bka’ brgyad bder gshegs ’dus pa: Rediscovered by mNga’ bdag Myang Nyi ma ’od zer, Reproduced from Ven. Dudjom Rinpoche’s Xylograph Collection. Gangtok: Ngagur Nyingmay Sungrab, vol. 1, 1-163.
sGra sbyor bam po gnyis pa: Ishikawa, Mie (ed.). 1990. A Critical Edition of the sgra sbyor bam po gnyis pa, An Old and Basic Commentary on the Mahavyutpatti. Tokyo: Toyo Bunko.
rGyal bu byang chub sems dpa’ la gnang ba’i bka’ yig: 'Phags pa bLo gros rgyal mtshan. 1969. rGyal bu byang chub sems dpa' la gnang ba'i bka' yig. In: bSod nams rGya mtsho (ed.), Sa ska bka’ ’bum: The Complete Works of the Great Masters of the Sa skya Sect of the Tibetan Buddhism. Tokyo: Toyo Bunko, vol. 7, 238-2-3-4.4.
bKa’ chems ka khol ma: sMon lam rgya mtsho (ed.). 1989. bka’ chems ka khol ma. Lanzhou: Kan su'i mi rigs dpe skrun khang.
bKa’ brgyad bde gshegs ’dus pa: Lama Sonam Topgay Kazi (ed.). 1976. bka’ brgyad bder gshegs ‘dus pa: Rediscovered by mNga’ bdag Myang Nyi ma ’od zer, Reproduced from Ven. Dudjom Rinpoche’s Xylograph Collection. Gangtok: Ngagur Nyingmay Sungrab.
KGT: dPa' bo gTsug lag phreng ba. 2002. chos 'byung mkhas pa'i dga' ston. Varanasi: Vajra Vidya. bLon po bka' thang: Chandra, Lokesh (ed.). 1982. blon po'i bka'i thang yig. In: bka' thang sde lnga: dga’ ldanphun tshogs gling edition. New Delhi: Satapitaka.
Mani bka' 'bum: Trayang and Jamyang Samten (eds.). 1975. Mani bka' 'bum: A collection of rediscovered teachings focussing on the tutelary deity Avalokitesvara (Mahakaurinaka). New Delhi: Trayang and Jamyang Samten.
MBNT: 1980. mi rje lhas mdzad byang chub sems dpa' sems dpa' chen po chos rgyal mes dbon rnam gsum gyi rnam par thar pa rin po che'i phreng ba. In: Dingo Chhentse Rinpoche (collect.), rin chen gter mdzod chen po’i rgyab chos. Paro: Ugyan Tempai Gyaltsen, vol. 7.
MTNa: Lopon Choedak (collect.). 1979. chos ’byung me tog sning po sbrang rtsi’i bcud: Manuscript A. Paro: Ugyen Tampai Gyaltsen.
MTNb: Lopon Sonam Sangpo (collect.). 1979. chos ’byung me tog sning po sbrang rtsi’i bcud: Manuscript B. Paro: Ugyen Tempai Gyaltsen.
MTNc: 1985. chos 'byung me tog sning po sbrang rtsi'i bcud. In Meisezahl, R. O. (ed.), Die grofie Geschichte des tibetischen Buddhismus nach alter Tradition. Sankt Augustin: VGH Wissenschaftsverlag, plates 1-366.
MTNd: Nyang ral Nyi ma 'od zer. 1988. chos ’byung me tog snying po sbrang rtsi’i bcud. Lhasa: Bod ljongs mi dmangs dpe skrun khang.
Padma bka’ thang: O rgyan gling pa. 1987. padma bka' thang: gter chen o rgyan gling pas gter nas bton pa. Cheng du: Si khron mi rigs dpe khang.
gSal ba’i me long: Myang ston Rig 'dzin lhun grub 'od zer. 1979. bka' brgyad bde gshegs 'dus pa'i gter ston myang sprul sku nyi ma 'od zer gyi rnam thar gsal ba'i me long. In: Lama Ngodrup (ed.), bka’ brgyad bde gshegs ’dus pa’i chos skor. Paro: Ugyan Tempai Gyaltsen, vol.2 199-381.
gSer phreng: Sangs rgyas gling pa. 2007. u rgyan gu ru padma 'byung gnas kyi rnam thar gser gyi phreng ba thar lam gsal byed. Lhasa: Bod ljongs mi dmangs dpe skrun khang.
bTsan lha Dictionary: Ngag dbang tshul khrims. 1997. brda dkrol gser gyi me long. Pe cin: Mi rigs dpe skrun khang.
ZLa-k: [see Appendix].