The Process of Forgetting in Tibet

John Strong charts a similar process for the famous Asokan inscriptions. Here too an indigenous “forgetting,” is followed by a foreign “discovery” of ancient inscriptional self-representations. Strong details the different interpretations that locals gave to the Asokan inscriptions, after the Brahmi script in which they were written became indecipherable. James Princep deciphered the script in the nineteenth century, ushering in a new epoch in the study of ancient India. Subsequent historians tended to either dismiss the later Asokan legends as ‘“downright absurd mythological accounts,”' or tried to synthesise the inscriptional and literary records as “history.” Strong sees behind both tendencies ‘a failure to take seriously into account the literary form and religious intent of the legends qua legends.' He goes on:

It is good to remember that the authors of the Asoka legends were as much in the dark about the content of Asoka's inscriptions as we were until James Princep deciphered them. In this book then, we shall deliberately try to return to a time when the edicts were unknown, so as to focus on the traditional significance of the legends of Asoka which, independent of epigraphy, reflected the religious preoccupation of particular communities, and inspired Buddhists throughout Asia for many centuries.

My intentions accord with Strong's aim here. However, I cannot hope to cover the entire period between the eighth and twentieth century. My more modest aim is to chart the changing depiction of Khri Srong lde brtsan in the first four centuries after his death. I confine myself to imperial and Buddhist sources. Bon po histories, told from the perspective of the “other” religion of Tibet, contain many interesting and divergent details on the life and character of Khri Srong lde brtsan. However, these Bon po portrayals have not yet been securely dated to the twelfth century or earlier. Therefore I have excluded them from this thesis. The same applies to artistic depictions of Khri Srong lde brtsan. I instead focus on the construction of Tibetan Buddhist historiography and Khri Srong lde brtsan the Buddhist king.

Chapter Outline

This thesis is in two parts. Chapters One to Three investigate the important histories that Tibetan tradition attributes to the eighth to twelfth centuries. I perform textual analyses on these sources in order to ascertain their antiquity. I rely on some post-twelfth-century works in these chapters, but only to assess the date of the older histories. Chapters Four to Six then utilises the truly old sources to chart the changing depiction of Khri Srong lde brtsan between the eighth and twelfth centuries. The early chapters' preparatory work is necessary in order to restrict my analysis in the later part of the thesis to genuinely ancient sources.

Chapter One describes the process of recension of an important history on Khri Srong lde brtsan, the sBa bzhed. The oldest available version of the sBa bzhed carries the title ‘dBa' bzhed.' The main text dates probably to the eleventh or twelfth century, but contains earlier narratives as well as later additional elements and interlinear notes. A longer, redacted version of the same narrative (sBa bzhed G) most likely dates to the twelfth century, but a condensed version of the same narrative (sBa bzhed S) represents a thirteenth or fourteenth-century redaction. I therefore conclude that I shall omit sBa bzhed S and the notes to the dBa' bzhed from my analysis in Chapters Four to Six, since they reflect a post-twelfth-century stage of Tibetan historiography.

Chapter Two and Three then investigate the antiquity of those representations of Khri Srong lde brtsan attributed to Nyang ral Nyi ma 'od zer (1124-1196). I devote a large part of this thesis to his works because they represent the furthest point in the trajectory of portraying Khri Srong lde brtsan, as I shall explain in the second half of the thesis.

Chapter Two assesses the antiquity of various extant versions of his bKa' thang Zangs gling ma (ZL). I know of 11 different versions currently available to scholars, which I categorise into three distinct recensions: ZL1 (versions a-e), ZL2 (f-g) and ZL3 (h-k). I first catalogue the different additional material present in two of the three recensions, ZL1 and ZL2, but not contained in ZL3. Second, I identify borrowings from ZL in later Tibetan Buddhist literature: Nyang ral's own Chos 'byung Me tog snying po sbrang rtsi'i bcud (MTN) and two fourteenth-century biographies, the Padma bka' thang and gSer phreng. My analysis suggests that ZL underwent several stages of redaction and that ZL3 represents the earliest recension. MTN follows this recension, therefore ZL3 holds the greatest claim to a twelfth-century origin. The Padma bka' thang also follows ZL3 in part, further strengthening this claim. The later gSer phreng perhaps synthesises the Padma bka' thang with ZL1; but this witness is of too late a date to ensure a twelfth-century origin for ZL1. ZL3 was not perhaps only the oldest, but also the most widespread and influential ZL recension, before 'Jam mgon Kong sprul legitimised ZL1 by including a version of this recenson as the first text of his popular Rin chen gter mdzod.

Chapter Three discusses MTN, which offers a somewhat different depiction of Khri Srong lde btsan. MTN incorporates ZL3's description of Khri Srong lde brtsan in toto in its narrative, but also adds elements from the sBa bzhed tradition and unique episodes that add extra detail to its depiction of the Buddhist king. I analyse some of these interpolated episodes and elements, which I found in MTN. I suspect that these are the interpolations of a later redactor into Nyang ral's work. As a result, in the rest of this thesis I confine myself to ZL3 as representative of Nyang ral's depiction of Khri Srong lde btsan.

The same chapter also investigates the Mes dbon gyi rnam thar (MBNT) attributed to Nyang ral. I conclude that MBNT is not consistent with his other historical works. MBNT almost exclusively follows sBa bzhed S (the zhabs btags ma edition) in its description of Khri Srong lde btsan. Nyang ral, in contrast, follows his ZL in MTN. I finally decide not to interpret MBNT's representation of Khri Srong lde btsan in this thesis, since it is a post-twelfth-century history.

Chapters Four, Five and Six chart the changing portrayal of Khri Srong lde brtsan between the eighth and twelfth centuries: as a Tibetan Emperor, a Buddhist King, and a tantric disciple respectively.

Chapter Four describes Khri Srong lde brtsan through the eyes of the imperial sources. Even during the course of his life (742-c.800 CE), Khri Srong lde brtsan is represented as a btsan po, a religious king, and a bodhisattva. Scholars generally agree that these imperial depictions are the oldest and most reliable representations of Khri Srong lde brtsan. These include: the Old Tibetan Annals, whose content dates to the seventh and eighth

centuries; the pillar inscriptions that were written on behalf of Khri Srong lde brtsan throughout his reign; and the later imperial-period records that remember him after his death (including pillar inscriptions and the Old Tibetan Chronicle). I do not dispute their antiquity, but rather problematise their reliability by highlighting the “self-presentational” aspect of their proclamations. These historical sources portray Khri Srong lde brtsan as a beneficent imperial ruler, at the centre of a prosperous and powerful realm. However, the place that these proclamatory texts inhabit in the public space of the empire suggests that this is a literary construct designed to legitimise the btsan po and the empire. The same chapter then describes the more religiously oriented inscriptions from his reign. The bSam yas bell inscription records a prayer that Khri srong lde brtsan may ‘attain supreme enlightenment' as a result of his accumulated merit. Later inscriptions turn this enlightenment into an actual

event in History. The 'Phyongs rgyas inscription transforms the btsan po into a bodhisattva who led the Tibetan population to enlightenment. It also calls him both an “emperor” (btsan po) and a religious king (chos rgyal). Later texts, for instance IOL Tib J 466/3 written towards the end of the imperial period, remember Khri Srong lde brtsan as a Buddhist king (chos rgyal) who has attained nirvana. These imperial documents demonstrate the

gradual apotheosis of Khri Srong lde brtsan between his birth and the fall of empire. Post-imperial histories build on these depictions to further transform his Buddhist image, leaving aside his imperial “self-presentation” to be forgotten. Chapter Five examines the later portrayals of Khri Srong lde brtsan as a Buddhist king. IOL Tib J 370/6 presents the divine image of Khri Srong lde brtsan and his ancestor Srong btsan sgam po (c.605-649). The dBa' bzhed compares these two btsan pos as legislators, borrowing an old narrative from Indian Buddhist literature. Nyang ral’s Zangs gling ma narrates that Avalokitesvara’s emanation as Srong btsan sgam po inspired Bodhisattva Manjusri to incarnate as Khri Srong lde brtsan. Whereas imperial documents cite previous royal patronage of Buddhism in Tibet, these narratives transform this notion into divine precedent.

Buddhist metaphysics also allows for the concept of predestination to enter into historiography. The dBa' bzhed portrays the king and his spiritual friends (kalyanamitras) as predestined to spread the Dharma in Tibet, due to past karmic links forged in India. The sBa bzhed emphasises these bonds and the sense of predestination by incorporating into its narrative the well-known tale of three brothers making aspirational prayers after construction a stupa for the Buddha in Magadha. Finally, the Zangs gling ma replaces one spiritual friend in this triad, the Tibetan dBa' gSal snang, with the Indian tantric master, Padmasambhava. This substitution leads Nyang ral to privilege India, the land of the Buddha, over its barbaric

borderland Tibet. Gradually, Indian religious figures eclipse the indigenous king as the primary agents who establish Buddhism in the land of snows. Finally, Chapter Six describes Khri Srong lde brtsan as a tantric disciple, especially in the Zangs gling ma. Histories only explicitly link Khri Srong lde brtsan to Mahdyoga tantras from the post-imperial period onwards. A Dunhuang text on the tantric King Tsa first creates a Tibetan synthesis

between two Indic elements: the royal transmission of tantric Buddhism and the inevitable end of the Dharma. It describes Khri Srong lde brtsan's reign as a “golden age” of tantric practice, from which Tibetan tantra has since declined. In contrast, the dBa' bzhed places the period of decline in the eighth century. Tantric masters like Padmasambhava attempt to prevent its destruction. The btsan po then hastens its demise by banishing Padmasambhava and causing a division in the samgha. The Zangs gling ma pushes the period of decline forward again into the twelfth century. It uses

the notion of reincarnation to depict the btsan po as the Dharma's saviour in a future life—reborn as Nyang ral himself. However, this is only possible because Khri Srong lde brtsan is Padmasambhava's disciple, rather than an independent and powerful btsan po. This is the process by which Khri Srong lde brtsan becomes subservient to a powerful Mahdyogin.

The word history is an unhelpfully vague and fluid term. I use it here in only three distinct forms ‘n history/ies,' ‘History' and ‘“history.”' I always qualify the first as ‘a/the (nth-century) history,' denoting the content of a text, or the sum or shared content of several texts. I define a text broadly, to include, for example, an inscription, a fragmentary document or a lengthy set of quotations in another text. The content of a text that makes up a history consists of the words that impart meaning to a reader or listener, usually consisting of a narrative or (laudatory or

defamatory) remarks that create a comprehensible depiction of the past or present for a contemporary or future Tibetan-speaking audience. I always mark the second type, History, with a capital H. This type of history constitutes the created world of the past that a narrative implies with its use of words, metaphors, and overall structure. The audience may imaginatively inhabit that History, partially suspending their present perspectives and taking on the world-view of the text and its portrayal of the past.

The third, “history” in apostrophes, is my term for a modern Western version of the second type, History, that presents itself as scientific. Such “historical” portrayals are grounded in a historical- critical method, for example the search for the “historical” Jesus. Their resulting depictions are still imaginings of the past, but claim superiority to other, say religious, representations. This thesis seeks to analyse and interpret certain histories to uncover each one's sense of History, rather than to interrogate or attack them in search of “history.”

It may be possible in the future to describe the “historical” Khri Srong lde brtsan, based more or less on the Tibetan and/or Chinese sources. Yet I do not feel that Tibetology has reached that level of certainty at present. Rather, I seek to examine the histories themselves, as literature, in order to assess their different descriptions of Khri Srong lde brtsan. Some scholars still quote from Tibetan histories as representative of the time of their creation, or worse still of the time period that forms the subject of that history. I wish instead to situate properly each history's perspective within the intertextual web of its creation. Therefore I try first to ascertain, as far as possible, the date, authorship and cultural

milieu of each history. In other words, I analyse the texts before attempting to interpret their representations of Khri Srong lde brtsan. My interpretation of these historical depictions follows Paul Ricoeur's concept of mimesis, with several important caveats. European thought on the creation of histories cannot lead to a grand overview of the workings of the “Tibetan mind.” Yet such thought may improve our understanding of

Tibetan historiography, since it elucidates many approaches to the past that are common to all complex societies. I was ddrawn to Ricoeur because he gives us a particularly perspicuous representation of the movement of history through culture in his Time and Narrative. For Ricoeur, even a king feels the disquieting aporia of life's inscrutability, and seeks to narrativise his life. This narrative gives the present a complementary past and a

future, and makes sense of the narrator's actions. Death gives the life-story an end to match its beginning. However, only another narrator can tell this completed life-story, and they naturally adapt the other person's story to their own concerns. Later narrators further adapt the anarrative to their own times. Ricoeur calls this continual process mimesis. Sheldon Lu provides a succinct precis of his three-level theory of mimesis (Ricoeur is exhaustive but seldom succinct!):

[F]or him an individual or a collectivity aspires to the coherence of a story with a plot. Historical agents “prefigure” their lives as stories with plots; historians retrospectively “configure” these historical actions into sequences of events with a beginning, a middle and an end; and finally readers mentally “refigure” lives, events, and actions and endow them with meaning.

The mimetic text expresses not only the author's version of History, mediated by a culture, but also his attempt to represent himself to his culture. Ricoeur's representation is itself a culturally situated representation of reality, with its own beguiling narrative and specifically European concerns. I therefore espouse Ricoeur's analysis in my thesis with three caveats, corresponding to each of the three levels of his scheme:

• Khri Srong lde brtsan's explanatory edict, for example (Chapter Four, below), undoubtedly “prefigures” his life and Buddhist actions within the tradition of previous btsan pos, hence as narrative. Nonetheless, I argue that the edicts' persuasive and legitimatory aspects make them more than mere “self¬representations” of an historical agent searching to narrate his life and so give it meaning. They also represent the centralized imperial administration that took a part in creating these public texts.

• I doubt that Tibetan historians “configure” their works with a necessary beginning, middle and end, as scholars such as Ricoeur and Frank Kermode (in his book The Sense of an Ending) suggest of Western history. There is perhaps more of a “sense of a beginning” (to quote A. D. Nuttall) in these histories; but they are more episodic than monolithic. While each vignette may be worked to a conclusion or not, the depiction of the main characters (in comparison or contrast with each other in each “scene”) is more important than the position that the main protagonist finds him/herself in by the final “scene.”

• I agree that readers “refigure” these narratives. Moreover, readers can also be writers, creating new works that incorporate, adapt, or reject previous narratives that they have read in the past. Well-read historians either consciously or unconsciously alter Tibetan historiography as they practice the art of creating their own histories or redact an earlier one. Specifically, Khri Srong lde brtsan is transformed as he is “passed from hand to hand” over generations of authors and redactors.

The different ways in which each historian sees the world of the past necessitates certain implicit views of History. Some cultures formulate traditions of historiography, ways of conceptualizing the past that historians within those traditions aspire to in their data collection, textual analysis and written presentation. Early Tibetan histories contain no espoused “philosophy of history,” or apology for a certain method of approaching sources or narrating the past. My task here is to look for trends in historiography, even perhaps some important trajectories in the ways that early Tibetan Buddhist communities viewed the world. These perspectives are expressed in their histories about the world that had disappeared—the imagined “golden age” of the Tibetan Buddhist Empire.