Printer Friendly


TOM J. F. TILLEMANS, one of the foremost authorities on the Indo-Tibetan Buddhist logico-epistemological tradition (pramana, tshad ma) founded by Dignaga (fifth-sixth century) and Dharmakirti (sixth-seventh century), has been intrigued by the question of the status of the Buddha and scripture in that tradition for a long time. His first public entry in this area dates from 1986, followed by another in 1990.(1) The present work is a study and annonated translation of the Ston pa tshad ma'i skyes bur sgrub pa'i gtam, "An Account which Establishes the Teacher [the Buddha] as a Person of Authority," written by the Alashan-Olot Mongol (sog po) scholar Ngag dbang bstan dar (1759-after 1839).(2) Conveniently, Tillemans chose to reproduce the English translation and the Tibetan text on facing pages (pp. 28-67), so that both are relatively easy to follow. In addition, his book contains a lucid introduction to some of the issues addressed by Ngag dbang bstan dar (pp. 1-24) and concludes with notes to his translation (pp. 69-77) and the indexes (pp. 79-91).

Tillemans (p. 2) tells us very little about this man, who is also known as Bstan dar Lha rams pa, Smon lam Rab 'byams pa, and Lha Idan Grwa skor pa, because he took the [Smon lam] Lha rams pa degree during the Great Prayer (smon lam chen mo) festivities in Lhasa in an as yet unknown year,3 and who was a disciple inter alia of the famous Klong rdol Bla ma Ngag dbang blo bzang (1719-95). Born in Bas mtha', the exiguous details of the life of this Dge lugs pa scholar and polyglot were briefly sketched by Th. Stcherbatsky, who reproduced his commentary on Dharmakirti's Santanantarasiddhi, and by W. Heissig in his well-known study of Mongol xylographs of Buddhist texts.(4) More recently, M. Taube has provided us with a relatively complete bibliographical notice of his linguistic oeuvre while studying a Tibetan lexicographic genre.(5) Ngag dbang bstam dar's work on Mongol grammar and orthotactics, the Kelen-u cimeg, was first studied by M. Taube,(6) after which other studies of this little text appeared in Ulan Ude and Seoul, of which the former, published in 1962, is the most comprehensive to date. Another small tract of his, on ontological questions in the context of tshad ma, was studied by A. Klein, and a section of one of his studies on Tibetan grammar was dealt with by Tillemans himself.(7) Ngag dbang bstan dar was also active in Beijing, where he was associated with Yonghegong monastery. There he wrote an exegesis of a little ritual text on the Tibetan "tea ceremony" (ja mchod) composed some three hundred years earlier by the second Dalai Lama, Dge 'dun rgya mtsho (1476-1542),8 and he also appears to have known Chinese rather well, since his writings contain a good number of references to this language and to several Chinese texts.(9) While he stayed for extended periods in Central Tibet and Beijing, his main see was Dge sgrub gling (or Mi pham chos gling) monastery, the "left cloister (g. yon dgon) of A lag sha[n]."

The history of the term "a person of authority" (tshad ma'i skyes bu, *pramanapurusa) that appears in the title of Ngag dbang bstan dar's text is discussed by Tillemans on pp. 5-9, and is now less of a mystery. He refers also (p. vi) to a letter from E. Steinkellner of Vienna, the author of an earlier essay on this term,(10) informing him of its occurrence in Yamari's subcommentary to Prajnakaragupta (ca. 800). Yamari (or Jamari) flourished in Kashmir sometime during the eleventh century, and this reference would be so far the term's first attestation in an Indian Buddhist pramanavada text; a juxtaposition of both elements skyes bu and tshad ma already occurs in the introductory portion.(11) If we accept that tshad ma'i skyes bu is a contraction of the expression tshad mar gyur pa'i skyes bu (*pramanabhutapurusa), and there seems little reason not to do so, then its appearance in Indian Buddhist writings can be pushed back even further. It seems that E. Franco was the first to indicate that its antecedent pramanabhuta is already found in Patanjali's Mahabhasya(12) and recently D. S. Ruegg has collected a valuable dossier on the occurrences of the expression and its cognates in a number of Buddhist and non-Buddhist texts, and analyzed its various usages.(13) To this corpus of sources for the term we may add its use in Dharmamitra's study of the Abhisamayalamkara, the Abhisamayalamkarakarikaprajnaparamitopadesasastratikaprasphutapada.(14) Dharmamitra is usually regarded as a disciple of Haribhadra, and he cites not only Dharmakirti's Pramanavarttika, but also inter alia Kamalasila (late eighth century) and Vinitadeva (eighth century). This means that he probably flourished during the first half of the ninth century, in any event sometime between the beginning of the ninth and the second half of the eleventh century, when the text was translated into Tibetan by Abhiyuktaka Tarasrimitra and Chos kyi shes rab. Not only is tshad ma'i skyes bu then ultimately of Indian origin, but also the concept embodied by the term tshad ma'i lam (*pramanamarga),(15) "the [?spiritually liberating] path of tshad ma" (albeit probably of a relatively late date), which played a rather important role in the Dga' ldan pa / Dge lugs pa "soteriology" of tshad ma.

To all appearances, it was only during the first decade of the fifteenth century that tshad ma'i skyes bu became a pivotal concept in a Tibetan understanding of the Buddha and "buddhalogical" gnoseology in connection with the Pramanavarttika's Pramanasiddhi chapter and, most importantly, with the tshad ma enterprise as a whole. In spite of affirmations of later Dge lugs pa historiography, so frequently revisionist and tendentiously forgetful, Tsong kha pa Bio bzang grags pa (1357-1419), recognized as the founder of that school, may very well have been the first to have conceptualized it in such a fashion.(16) For until now the available textual evidence strongly suggested that the central place it came to occupy in Dga' ldan pa / Dge lugs pa tshad ma soteriology had no explicit Tibetan, let alone, real Indian antecedents. Part of Tillemans' aim in his study is to " ... seek to understand something of the history of the complicated indigenous debates" (p. 1), being all the while " ... conscious that much more remains to be said" (p. 1). Of course, he nowhere states that he intends his essay to be as comprehensive a study of the issues as the published literature allows, but it is nonetheless hard to excuse him completely from not really having followed through on his stated intention, for he spends very little, arguably too little, time on the Tibetan antecedents of Ngag dbang bstan dar's little work, let alone on the expositions of the subject by Tsong kha pa and his disciples--the earlier paper of S. Kimura is thus passed over in silence. This imbalance was in part redressed by D. P. Jackson's recent discussion of the historical issues involved in Tsong kha pa's approach to Buddhist pramanavada and with but one minor reservation I believe him to be absolutely correct when he writes that: "... Tsong kha pa and his immediate circle apparently did play the most active role in actually trying to revive Tshad-ma as a living spiritual practice in the late 14th and early 15th century."(17) But I am not altogether sure whether it is apposite to speak here of a "revival" of sorts. Something had undoubtedly been "in the air" before him. For already his senior contemporary and erstwhile teacher Nya dbon Kun dga' dpal (1285-1379) had characterized the Pramanasiddhi chapter's subject matter as having to do with "soteriological" issues qua "striving for liberation" (thar pa don gnyer).(18) And, as indicated by Stag tshang Lo tsa ba Shes rab fin chen's (1405-after 1477) study of the five domains of knowledge (rig gnas lnga), Dpang Lo tsa ba Blo gros brtan pa (1276-1342) and his main disciple and maternal nephew Lo tsa ba Byang chub rtse mo (1303-80), together with Bla ma dam pa Bsod nams rgyal mtshan (1312-75), had subsumed the texts of Dignaga and Dharmakirti under the rubric of abhidharma, unambiguously suggesting therewith their inclusion in the corpus of treatises that have to do with Buddhist soteriology.(19) Most of Tsong kha pa's biographies also stipulate that Nya dbon taught him his Abhisama-yamlamkara commentary of 1371. And 'Jam dpal rgya mtsho's (1356-1428) biography of Tsong kha pa states that he had also worked on tshad ma with him,(20) so that we can on no account exclude the probability that at least these three men had exerted some influence on him in this direction as well, whether directly or otherwise.

Having said this, we are, however, at the same time confronted by the puzzling absence of an organized explication of the role and function of tshad ma'i skyes bu in Tsong kha pa's own oeuvre on tshad ma, in which he might have employed and explored the meaning of the term, or elsewhere in his writings where, in but a few instances, he alludes to its equivalence with the Buddha, the experience of buddhahood, or one (allegedly) with this experience other than the Buddha. One of these allusions occurs in a grammatically fairly tricky passage of his celebrated Rang girtogs brjod rin po che 'dun legs ma, "A Poetic Autobiographical Narrative: A [Treatise Giving] Good Council," one that is cited in part by Stag tshang Lo tsa ba as well,(21) which Ngag dbang bstan dar quotes as follows (Tillemans, p. 43):

rnam grol don du gnyer ba bcom ldan 'das //

tshad mar bsgrubs shing de las 'di yi ni //

bstan pa kho na thar 'dod [jug ngogs su //

nges pa gting nas rnyed pas ... //

Tillemans (p. 42) translates these lines:

Those who strive for deliverance establish that the

Illustrious One is authoritative and from that, they find

a profound certainty that it is only his teaching which

is the entrance-gate ([jug ngogs) for those desiring


Thus he more or less takes the phrase rnam grol don du gnyer ba as the subject / agent of the two "perfective" stems bsgrubs and rnyed of the transitive verb sgrub and transitive verbal noun rnyed [pa].(22) These lines and the larger context in which they occur have been rendered severally and quite differently in recent years. In the introduction to his translations in Sylvae (1685), J. Dryden proleptically and apologetically remarks that " ... I have both added and omitted, and even sometimes very boldly made such expositions of my Authors, as no Dutch Commentator will forgive me," suggesting that this latter is to be included among "Pedants" and "false Critticks."(23) Be this as it may and risking yet another "Amboynic" epithet I should like to offer an alternative interpretation of these lines and the passage in which they occur, but will not discuss the earlier translations of this passage by E. Obermiller, A. Wayman, and R. A. F. Thurman. These miss the boat in just too many respects? Much better interpretations, but also not altogether convincing to me, are those proposed by E. Steinkellner and, most recently, by D. P. Jackson; I have not seen the one offered by S. Matsumoto.(25)

The entire passage from the beginning to the finite verb at the end reads as follows in the Lhasa Zhol print of Tsong kha pa's collected works:(26)

de ni mi rigs smra ba'i phul byung du // [[1]

mthong nas lhag par tshul der dpyad pa na // [2]

tshad ma kun las btus pa'i mchod brjod don // [3]

tshad ma grub par lugs 'byung lugs ldog gis // [4]

rnam grol don du gnyer la bcorn Idan 'das // [5]

tshad mar bsgrubs shing de la 'di yi ni // [6]

bstan pa kho na thar 'dod 'jug ngogs su // [7]

nges pa gting nas rnyed pas theg gnyis kyi // [8]

lam gyi gnad kun 'dril bar rigs lam nas // [9]

legs par thon pas lhag par dga' ba rnyed // [10]

The passage, as is often the case with "scientific Tibetan," has essentially one fundamental problem area, namely, the identification of the subjects/agents of the transitive verbal expressions mthong, dpyad [pa] (line 2), bsgrubs (line 6), thon [pas] (line 10), and rnyed (lines 7, 10). On the basis of what precedes line 2, there is no doubt that the covert agent of mthong and dpyad is Tsong kha pa himself; the two lines thus read:

Having seen [the view that the texts of Dignaga and

Dharmakirti have no soteriological significance] to be

the height of unreasonable talk, upon a surpassing

reflection on that [textual corpus], ...

The verbal noun thon [pas] of line 10 has gnad as its "subject." The phrase theg gnyis ... thon pas is adverbial to the finite verb rnyed of the same line, which is covertly autobiographical; we therefore obtain:

surpassing joy was obtained because all the essential

points of the path [to liberation] of both vehicles

emerged well in unison through the path of logic [of


(or: [I] obtained surpassing joy because ...)

Lines 3 to 8 present us with a greater problem, but let us be clear about one thing: the pas in rnyed pas indicates that the passage is both the result of the reflection (dpyad pa [na]) and the cause of Tsong kha pa having obtained surpassing joy (lhag par dga' ba rnyed). The phrase tshad rna grub par of line 5 could also, or perhaps even should, in my view, be understood not only "as the establishment of pramana," but also as "in the Pramanasiddhi [chapter]." Why? Because, beginning with Tsong kha pa's biography by his disciple Mkhas grub Dge legs dpal bzang po (1385-1438), later studies of his life echo over and over again that he had undergone a profoundly moving religious experience when, at the age of twenty-one, he studied the early thirteenth-century Pramanavarttika exegesis by 'U yug pa Rigs pa'i seng ge (ca. 1190-after 1267)--specifically, its commentary on that particular chapter. Ordinarily, in order to accommodate the transitivity of bsgrubs, the perfective stem of the verb sgrub, "to establish, prove, bring into being, realize," lines 3 to 5 would require an agent, regardless of its overt or covert occurrence in the sentence, and a patient. Looking at just these three and a half lines, it is grammatically not possible that rnam grol don du gnyer ba functions as the "subject" (or, perhaps better, as the agent of bsgrubs) leaving aside that the author should have rendered bsgrubs by "established" rather than "establish." If so, then gnyer ba ought to have been written gnyer bas with the agentive/ergative affix -s. The function of the coordinating particle shing here is subordinating, that is to say, it makes bsgrubs function as a causal antecedent of rnyed [pa]; x first needs to be realized (bsgrubs) after which it can be something that was acquired and taken possession of (rnyed [pa],(27) Whereas [b]sgrubs is unproblematic in its transitivity and ergativity, the triangulation of the nature and function of rnyed is just a bit more difficult, for, according to some modern non-Tibetan as well as Tibetan linguists, it is a verb that not only belongs to the class of so-called "benefactive or possessive verbs," which do not require a typical ergative construction (the agent of a sentence takes the agentive/ergative marker and the patient stands in the absolutive zero-case), but it also falls among verbs that are tha mi dad pa (= Ch. bujiwu dongci), a term that is usually rendered as "intransitive."(28) Now it is not all that hard to put together a dossier of the verb's obvious transitivity, with its agent affixed with an agentive/ergative marker and the patient with zero-affixation. Examples would be the Tibetan translation of the Kavyadarsa I: 89c-d: ... 'di dag gis // go skabs rnyed par 'gyur ma yin and the second half of Sa skya Pandita's Legs bshad rin po che'i gter, verse 412, where we read: chog shes rnams kyis nor rnyed kyang.... (29) Of course, because of the various usages of the particle shing, bsgrubs and rnyed [pa] need not be necessarily linked to one and the same agent,(30) regardless of whether they be overtly present or covertly implied in the sentence, and this is also to some extant vouchsafed by the presence of the expression de las, "from that/on that [basis]," which creates what amounts to a conceptual break in the sentence. Given all this, rnam grol don du gnyer ba cannot be the agent/subject of bsgrubs and rnyed [pa], and the widespread variant reading rnam grol don du gnyer la, which tums rnam grol don du gnyer into a "dative expression" seems to be exactly the one that we would need to make sense of these lines of verse. This then leads me to propose that "*by Dharmakirti" (*chos kyi grags pas) functions as the covert agent of bsgrubs and that "(*)by me" (ngas) does the same for rnyed [pa]. Thus lines 3-8 would read, without any pretense at versification:

because [Dharmakirti] had established in the

Pramanasiddhi [chapter], by means of a forward and

backward procedure, the sense of the invocation verse

of the Pramanasamuccaya [to be that] the Illustrious

One is an authority (tshad ma) for (read: la) the seeker

of liberation, and from that, from the depth [of my

being], the conviction was gained [or: I gained the

conviction] that his doctrine alone is the embarkation

point for one desirous of liberation ...

If my argument is acceptable, then the nine lines from Tsong kha pa's versified autobiography would read unpoetically and unprettily:

Having seen that [the view that the texts of Dignaga

and Dharmakirti have no soteriological significance] is

the height of unreasonable talk, upon a surpassing

reflection on that [textual corpus]--because

[Dharmakirti] established, in the Pramanasiddhi

[chapter], by means of a forward and backward

procedure, the sense of the invocation verse of the

Pramanasamuccaya [to be that] the Illustrious One is

an authority (tshad ma) for (read: la) the seeker of

liberation, and [because] on the basis of that from the

depth [of my being] the conviction was gained [or: I

gained the conviction] that his [the Bhagavan's]

doctrine alone is the embarkation point for one

desirous of liberation--surpassing joy was obtained

(or: I obtained surpassing joy) because all the essential

points of the path [to liberation] of both vehicles

emerged well in unison through the path of logic [of


Ngag dbang bstan dar (cf. Tillemans, pp. 43, nn. 3-4, 44) draws our attention to two text-critical problems with this verse, of which the first, the only one that is of grammatical relevance, suggests that we read gnyer ba rather than gnyer la, as is found in three separate editions of the text to which he had access. Another reason that he gives for this preference (p. 45) is that Ngag dbang bstan dar himself observed that:

'on kyang dpyod ldan la las rnam grol don du gnyer ba

ma dag ste rje phur bu lcog pa'i tikar rnam grol don du

gnyer ba rnams la zhes la sgra'i don du bkral 'dug pa'i

phyir zer ro //

Nonetheless, some perspicacious ones have alleged that

rnam grol don du gnyer ba is corrupt, because in Rje

Phur bu lcog pa's [Ngag dbang byams pa (1682-1762)]

exegesis [of the Autobiographical Narrative], rnam grol

don du gnyer ba rnams la was commented upon in the

sense of the term la.

The sheer preponderance of the evidence, that is, the readings of the text in the Beijing, Bkra shis lhun po, Sku 'bum, and Lhasa blockprints, together with its citation in other sources,(31) suggests that we indeed have to read the lectio facilior gnyer la or gnyer [ba] la rather than just gnyer ba. What is significant, however, especially in view of the fact that Ngag dbang bstan dar was profoundly interested in philological and grammatical questions, is that he refrains from taking a firm stand on this issue and that he declines to argue for the adoption of one reading over the other, although he does opt for the reading gnyer ba. I would hazard the guess that the reason for his silence was that he considered both readings to be grammatically justifiable and probably also semantically equivalent; indeed, gnyer ba can easily accommodate the interpretation of gnyer ba la.

If Tsong kha pa were indeed the first to have conceptualized in a formal way tshad ma's soteriology in terms of tshad ma'i skyes bu and its cognate thams can mkhyen pa, "omniscience" or "one who is omniscient" (we find this only in the putative lecture notes taken down by his disciple Rgyal tshab in, perhaps, 1404, and known as the Tshad ma'i brjed byang chen mo, "The Great Memorandum on Tshad ma"(32) and not in his own and probably later Sde bdun la 'jug pa'i sgo don gnyer yid kyi mun sel, "A Port of Entry into [Dharmakirti's] Seven Treatises: Removing the Intellectual Darkness of Those Striving for Meaningfulness"), then it would certainly be worth our while to ask what incidents in his life might have suggested to him the nexus of this idea, and whether his own writings on tshad ma or the notes taken down by his students when he lectured on this subject (his "indirect" writings) shed any light on its formation.(33) It is of course well nigh impossible to determine to what extent the Tshad ma'i brjed byang chen mo represents Tsong kha pa's own take on Dharmakirti, especially, and how much of it is Rgyal tshab's interpretation, so that I think we need to be a great deal more careful in attributing its contents to Tsong kha pa alone than has been the norm heretofore. An interesting case, one which indirectly supports my skeptical stance on the question of the Tshad ma'i brjed byang chen mo's authorship, is presented by the substantial differences that exist between the two sets of notes taken down by Rgyal tshab and Mkhas grub when Tsong kha pa lectured on the Pramanavarttika's Pratyaksa chapter (on most likely two different occasions), and the obvious interpolations (and, perhaps editorially more serious, revisions) carried out by the takers of these notes.

The aforementioned articles by E. Steinkellner, S. Kimura, and D. P. Jackson contain important discussions of the connection between Tsong kha pa's oeuvre and the intellectual climate of his time, and I intend to supplement these elsewhere in some detail.(34) But it does strike me as a shame that, even though Tillemans is obviously quite familiar with the Tshad ma'i brjed byang chen mo, he was for some reason not inclined to inform his present study fully with these lecture notes (sole and not very useful references to it are found on pp. 2, 9, and duplicated on p. 16), especially since it contains a number of important passages that give us additional insights into the architecture and genealogy of Ngag dbang bstan dar's work. I may just mention here, for example, the fact that his quotation of the verse from Samkarapati's Devatisayastotra (no. 16) on pp. 28-29 is also already contextually and fully prefigured in this text.(35) To be sure, of considerable interest is further that the beginning portion of the Tshad ma'i brjed byang chen mo, which deals with religious issues anent tshad ma, is cognate with, but by no means identical to, a Tshad ma'i lam [gyi rim pa'i] sgrigs, "An Arrangement of the Stages of the Spiritual Path of Tshad ma," which already 'jam dbyangs bzhad pa'i rdo rje I Ngag dbang brtson 'grus (1648-1721) unequivocally attributes to Tsong kha pa himself in the first part of his large study of the Abhisamayalamkara, which he completed in 1708 while he was still in the Lhasa area.(36) But it now appears that the situation is even more complicated, for it seems that by the early nineteenth century, at the latest, two different texts with the same title of Tshad ma'i lam [gyi rim pa'i] sgrigs and both attributed to Tsong kha pa must have circulated among the more intellectually curious Tibetan scholars. This we can deduce from the fact that the wording of the text cited by Ngag dbang brtson 'grus is not found in a work with the very same title (this holds for the Tshad ma'i brjed byang chen mo, as well) included in the second volume of the two-volume printed edition of the collected writings of Ke'u [or: Ke/kel] tshang [nub (= western)](37) Blo bzang 'jam dbyangs smon lam of Se ra monastery, which, too, is explicitly ascribed to Tsong kha pa.(38) The first volume, edited by Tang Chi'an and Mi nyag Mgon po, of a three-volume catalogue of manuscripts housed in the Cultural Palace of Nationalities states that Ke'u tshang was born in 1689 and affiliated with Brag mgo monastery in northwestern Khams (or northwestern Sichuan province).(39) But this is not possible inasmuch as his oeuvre contains inter alia a dedicatory inscription for a Maitreya statue commemorating the passing of 'Chi med rgyal mo, "Dga' bzhi pa's wife,"(40) a dedicatory prayer for the statuary of the "Bka' blon Zhabs pad [= Mi 'gyur bsod nams dpal 'byor] of the Rdo ring Dga' bhzi pa [family]," the text of which is dated 1827, and prayers for the "sons of Dga' bzhi pa [= Mi 'gyur bsod nams dpal 'byor], Sde pa Nyi ma lhun grub, and Tha'i ji Rdo rje tshe ring, and their mother Srid zhi'i nyi ma and sister Rin chen sgron ma."(41) In other words, Ke'u tshang must have flourished in the latter part of the eighteenth and the first half of the nineteenth century and he was at least on fairly close terms with the Dga' bzhi family. The publishers of his writings in Dharamsala note, on the other hand, that he was born in the famous Rdo ring (or: Dga' bzhi) family "in the late eighteenth century" and that he was the son of Bsod nams bstan 'dzin dpal 'byor, one of the family's most distinguished scions. In fact, he is mentioned more than once in the autobiography of the latter. The family had also enjoyed some sporadic personal contacts with his immediate precursor Ke'u tshang Byams pa smon lam (?-1790 or 1791), and Bsod nams bstan 'dzin dpal 'byor's autobiography mentions him first in an entry for the year 1773, and then again in one for 1782.(42) In fact, Blo bzang 'jam dbyangs smon lam was installed on the Ke'u tshang throne in the first week of January of 1801(43) and, while Bsod nams bstan 'dzin dpal 'byor for some unknown reason neglected to record the year of his birth, we do learn from the autobiography that he was fourteen years old in 1805,(44) so that he must have been born in 1791.

What the presence of a Tshad ma'i lam [gyi rim pa'i] sgrigs in Blo bzang 'jam dbyangs smon lam's oeuvre plainly means is that, like the one used by Ngag dbang brtson 'grus, this text too had some currency in Central Tibet and that, at least in Blo bzang 'jam dbyangs smon lam's immediate intellectual milieu, its ascription to Tsong kha pa was unchallenged. Whether it or the other one cited by Ngag dbang brtson 'grus was ultimately integrated in the Sku 'bum edition of Tsong kha pa's collected works remains an open question in the absence of an inspection of this edition, but I have my doubts. The last argument of both the texts of the xylograph and handwritten dbu med manuscript comes to a rather abrupt halt, after which the first colophon relates, rather inelegantly, the following admonition:(45)

'di ni chos kyi rgyal po tsong kha pa chen pos mdzad

pa'i tshad ma'i lam rim yod do zhes zer ba de nyid 'khrul

med yin mod kyi / rje nyid kyis mdzad pa'i mdzad byang

mi snang zhing / 'on kyang rje'i gsung yin par gor ma

chag pas yid ma ches par mi bya'o //

As for this [text], while the very allegation, namely, that

"there is a Stages on the Path of Tshad ma that was

authorized by the great king of religion Tsong kha pa" is

indeed without error, an author's colophon authored by

the Lord [Tsong kha pa] himself is not present [in it].

Nonetheless, [its authenticity] should not be disbelieved

since it is certain that it is a statement of the Lord.

Future research may also be able to answer questions that bear on the nature of the relationship that so obviously exists between the Tshad ma'i brjed byang chen mo and the Tshad ma'i lam [gyi rim pa'i ] sgrigs of Blo bzang 'jam dbyangs smon lam's writings, for, in spite of the latter's title, the two treatises are very similar in structure and contents, the bulk of both dealing with issues that have no explicit bearing on tshad ma's "soteriology," but rather focus on problems related to perception, inference, and concept-formation. Intriguing is that when Mkhas grub's biography of Tsong kha pa informs us that his master first lectured on tshad ma in terms of a viable soteriology in the year 1404, 'Brug Rgyal dbang Chos rje asserts at this juncture that the notes taken down at that time by the Rgyal tshab became a work known either as Tshad ma'i brjed byang or the Tshad ma'i lam bsgrigs.(46) As I intend to return to this in my study mentioned above, let us for the moment rest content with the following indication of the parallels that exist between their contents. The Tshad ma'i lam gyi rim pa'i sgrigs--it does not quote Samkarapati--begins with the following statement:(47)

om swa sti / rje btsun bla ma rnams la phyag 'tshal lo /

/ de la 'dir rigs pa'i bstan bcos kyi de kho na nyid gtan

la 'bebs par byed pa la / tshad ma'i mtshan nyid(a) ston

pa'i bstan bcos brtsams pa'i dgos pa(b) / dgos pa de bstan

bcos la rag las tshul / dgos pa de ldan gyi bstan bcos la

gces par bzung ba / bstan bcos de'i brjod bya'i lus gtan

la dbab pa'o //

(a)*TSONG adds /.

(b)*TSONG adds dang.

Om Happiness! Homage to the holy preceptors! In that

[connection(48)], when analytically clarifying the essential

nature of the treatise of logic in this [text, there are

four rubrics]:

[1] The purpose(49) for having composed a treatise demonstrating

the defining feature[s] of tshad ma.

[2] The manner in which that purpose depends on [such

a] treatise.

[3] Holding precious a treatise with that purpose.

[4] An analysis of that treatise's corpus of stated topics.

All of these fall into further topics, but it suffices to show here that there is, of course, much that is identical to the Tshad ma'i brjed byang chen mo, for it begins:(50)

rje btsun bla ma dam pa rnams la phyag 'tshal lo // grol

ba don gnyer gyi skyes bu rnams yang dag pa'i lam

dang lam ma yin pa legs par phye nas / that par'jug pa'i

mig gcig pu 'gran zla med pa'i(a) bdag nyid chen po

phyogs glang yab sras kyis mdzad pa'i rigs pa'i bstan

bcos gtan la dbab pa la don bzhi ste / tshad ma'i bstan

bcos brtsams pa'i dgos pa dang / dgos pa de bstan bcos

de la rag las pa'i tshul dang / dgos pa de ldan gyi bstan

bcos la gces par bzung ba dang / bstan bcos de'i brjod

bya'i lus(b) gtan la phab pa'o //

(a) TSONG reads pa.

(b) TSONG reads yul; TSONG, 588 reads lus.

Homage to the venerable holy preceptors! Having distinguished

well between what is and what is not the true

path [for] individuals who strive for liberation, the analytical

clarification of the treatise[s] of logic written by

the father Dignaga, the great being who is the peerless,

unique eye for entering into spiritual freedom, [and his

intellectual] son [Dharmakirti], [has] four [sections]:

[1] The purpose for having composed a treatise of

tshad ma and

[2] the manner in which that purpose depends on that

treatise and

[3] holding precious a treatise with that purpose and

[4] analysis of that treatise's object[s] of discourse.

We should also note here that while identical textual divisions are found in the introductory material of Mkhas grub's Tshad ma sde bdun gyi rgyan yid kyi mun sel, "An Ornament of the Seven Tshad ma Treatises Removing Intellectual Darkness," his topical study of Dharmakirti's oeuvre as a whole, and that while something quite similar can be found in a good portion of the prefatory remarks to his later exegesis of the Pramanavarttika,(51) they are entirely absent from Rgyal tshab's tshad ma writings. Further, Mkhas grub's analyses are by and large neither an echo of, nor echoed in, the Tshad ma'i brjed byang chen mo. Rather, to some extent he shares a similar phraseology with the Tshad ma'i lam gyi rim pa'i sgrigs and he too does not quote Samkarapati.

Ngag dbang bstan dar's work is divided into two parts, the second of which (pp. 44-63) essentially presents a summary of the arguments of the Pramanavarttika's Pramanasiddhi chapter, and makes but a few references to the exegeses of this chapter by Rgyal tshab and Mkhas grub, whom the Dge lugs pa tradition considers to have been Tsong kha pa's two main disciples and especially authoritative.(52) As is well known, the Pramanasiddhi chapter is a lengthy commentary on the opening verse of Dignaga's Pramanasamuccaya, and it was recently given a penetrating analysis by E. Franco, who convincingly argued that, in this chapter, Dharmakirti not only attempted to provide a philosophical framework for a "person of authority" acceptable to both Buddhists and non-Buddhists alike, but also that his attempt was perhaps influenced by such non-Buddhist philosophers as Vatsyayana and others.(53) Tillemans states (p. 4), a trifle anachronistically, that the Dga' ldan pa / Dge lugs pa's study of the Indian Pramanasiddhi commentaries of Devendrabuddhi and Sakyabuddhi led them to situate "a break between anuloma and pratiloma at k. [karika] 146 in the Pramanasiddhi ... " chapter [= PV II: 146]. Lest one be misled by the absence of a qualifying historical statement, it ought to be pointed out that this was by no means confined to these late authors, since precisely this break is also met with in the earlier commentaries by members of the Sa skya pa school, several of which exerted crucial influence on Dga' ldan pa exegeses. Of the published pre-Rgyal tshab Tibetan Pramanavarttika commentaries, the ones by 'U yug pa and Btsun pa Ston gzhon of the year 1297 (the latter was printed only fairly recently and was of course not available to Tillemans) make this very same division at this juncture of the text.(54)

Needless to say, both 'U yug pa and Ston gzhon,(55) let alone the Pramanasiddhi exegeses of Rgyal tshab and Mkhas grub, employ the term tshad ma'i skyes bu. In addition, another early attestation of its having been linked explicitly with the invocation of the Pramanasamuccaya and, thence, the Pramanavarttika, is met with at the outset of Dol po pa Shes rab rgyal mtshan's (1292-1361) longish reply to a letter that a dpon Byang pa had sent to him by way of a certain Dkon mchog bzang po.(56) The latter would either have to have been a governor of Byang principality, or a high official who was a member of the ruling family of Byang, or perhaps even an official who came from Byang, but was not necessarily a member of its ruling family. I would like to propose that he should be identified as Nam mkha' bstan pa'i rgyal mtshan (?1315-after 1373), a member of Byang's ruling family, who rose to the rank of governor-general of Central Tibet as a whole during the last years of Mongol rule in Tibet, and, most crucially, whom several sources, albeit not the extant biographies of Dol po pa, expressly record as one of his disciples.(57) The two studies of Dol po pa's life by his disciples Kun spangs pa Chos grags dpal bzang po (1283-?1363) and Gha rung pa Lha'i rgyal mtshan (1319-1401)(58) have it that he first received his training in tshad ma from Skyi ston 'Jam dbyangs Grags pa rgyal mtshan at Skyi stengs monastery (in Dol po) in 1310, who used an unidentified Tshad rna bsdus pa, "Tshad ma Summary," as his teaching manual. The fact that he used such a Tshad ma bsdus pa might indicate that Skyi ston's expertise in tshad ma was owed to one of the transmissions of tshad ma that had their origin in Rngog Lo tsa ba Blo ldan shes rab (1059-1109)--that is, that he was an exponent of the so-called Rngog system (rngog lugs). Lha'i rgyal mtshan states that, in addition to the Pramanaviniscaya, Skyi ston also taught him the Tshad ma rigs pa'i rgyan, "Tshad ma, An Ornament of Logical Reasoning." I am unable to identify this work at the present time, but one might conjecture that it was a Pramanaviniscaya commentary belonging to the Rngog system; it occurs in the following enumeration: tshad ma rnam nges / tshad ma rig pa'i brgyan (sic) / phar phyin / mngon pa / spyod 'jug / rdo rje phreng ba la swo (= sogs) pa. When Skyi ston was urged to come to Sa skya monastery, the young Dol po pa continued his tshad ma studies with Shes rab bzang po and Gzhon nu bzang po, and, obviously citing an autobiographical fragment, whether it be textual or verbal, Lha'i rgyal mtshan writes: " ... under them [I] listened to tshad ma's textual traditions and Summaries, and trained in discussing [them]. [I] wrote both the basic text for the Rigs pa'i rgyan, 'An Ornament of Logical Reasoning,' [and its] commentary" (khong rnams la tshad ma gzhung lugs dang bsdus pa nyan cing brgo gleng sbyangs rigs pa'i [b]rgyan rtsa 'grel cha 2 bris). As far as I am aware, neither youthful work survives, but it seems transparent that these should clearly be distinguished from the earlier Tshad ma rigs pa'i rgyan. Dol po pa himself was then summoned to come to Sa skya, where he finally arrived in 1312 at the age of twenty. He studied there the so-called Sa[skya Pandita] system (sa lugs) of tshad ma--it comprises Sa skya Pandita's Tshad ma rigs pa'i gter, "Tshad ma: A Treasury of Logical Reasoning," and the Pramanavarttika--under Kun dga' blo gros (1299-1327) and Mkhas btsun chen po Nam mkha' legs pa'i rgyal mtshan dpal bzang po (1305-43). Kun spangs pa also imparted to him Dharmakirti's major works; so did Seng ge dpal. Thus, the linkage of the notion of tshad ma'i skyes bu most probably had its origin in his studies of the sa lugs transmission of Indo-Tibetan tshad ma studies.

Ngag dbang bstan dar (and he is basically followed in this by Tillemans) is not only disinclined to employ overtly the Tshad ma'i brjed byang chen mo, but also the perhaps more fundamental textual precedents for his work, namely the two Tshad ma'i lam 'khrid, "Instruction in the Spiritual Path of Tshad ma," by Rgyal tshab and Mkhas grub.(59) Tillemans does not discuss either of these or their pertinence for the text he is studying, but does mention the former on two occasions, once in his introduction (p. 8, note 11) and once in an annotation to his translation on p. 60 (p. 76, note 36). As is to be expected, the colophons of both texts suggest that these Dga' ldan pa scholars were inspired by their master Tsong kha pa. Rgyal tshab's work, but not that of Mkhas grub, squarely places the notion of tshad ma'i skyes bu in the anthropology of the stages-on-the-path (lam rim) concept of the skyes bu chen po, "superior person,"(60) a concept which, for the Tibetans, is closely associated with Atisa's literary activity in Tibet (ca. 1042-54). Again, this seems to have been in the air, so to speak, for we also find a reference to the lam rim literature's division of persons into three different categories in the discussion of "the result of knowing objects through the valid means of cognition," by the Bka' gdams pa scholar Bcom ldan rig[s] pa'i ral gri (ca. 1240-ca. 1310),(61) whose real name in religion, as Rong ston informs us, was Chos kyi rgyal mtshan.(62) Aside from Rgyal tshab and Mkhas grub, Dwags po Blo gros brtan pa (1402-78), another fifteenth century Dga' ldan pa writer, the seventh abbot of Dga' ldan monastery, a disciple of, especially, Rgyal tshab and a specialist in lam rim, also dealt with this issue in his Rnam 'grel gyi lam bsgrigs, as noted by Pan chen Bsod nams grags pa (1478-1554) in his study of Bka' gdams pa ecclesiastic history of 1529.(63) But this work has yet to surface. True, Spyan snga Blo gros rgyal mtshan (?1402-72?), a disciple of both Tsong kha pa and Mkhas grub, wrote an interesting little treatise entitled Rnam 'grel la brten pa'i lta khrid, "An Introductory Instruction Based on the [Pramana] varttika],"(64) but it has no immediate bearing on the present subject, as it discusses in a comparative fashion the understanding of ontological "essencelessness" (bdag med) in both the Pramanavarttika and Madhyamakavatara.

The first portion of Ngag dbang bstan dar's work (pp. 30-44) is aimed at demonstrating that Buddhism as a religious enterprise and, especially, as a "text" is "faultless" (skyon reed), the argumentation for which functions as a necessary precondition for establishing that the Buddha himself is a tshad ma'i skyes bu, which is then discussed in the remainder of his text. The analysis of this relationship between the two notions may very well have begun with Tsong kha pa. Tillemans does not inform his reader of this in his otherwise detailed survey of these two points (pp. 18-24), but the Tshad ma brjed byang chen mo and the Tshad ma'i lam gyi rim pa'i sgrigs, if proved to have historical veracity, are so far our earliest textual sources for this. Indeed, Ngag dbang bstan dar's text (pp. 30-31), is arguably indebted ultimately to the former or perhaps even to both.

In his overview (pp. 9-15) of the so-called "three-fold analysis" (dpyad pa gsum), by means of which the validity of scripture can be examined,(65) and of other issues, Tillemans observes on (p. 11) that the "... Dge lugs pa authors would come to see k. 215 [of the first Svarthanumana chapter of the Pramanavarttika] as showing three criteria which a scripture must satisfy to be considered valid." But he never explicitly deals with these three criteria in a systematic and comprehensive fashion, but refers the reader to a note 18 for further information, where, however, as we shall see below, the interpretation of the text in question, which is supposed to throw light on these criteria, is at first glance (and according to Tillemans) not entirely transparent. In other words, the reader is left in the dark. In any case, he judges this analysis to be a "rather revolutionary account of scriptural inferences," but it may very well turn out that this, too, had already been seen earlier in Tibet. In fact, that it may not have been a typically early Dga' ldan pa / Dge lugs pa contribution to the field can almost be inferred from Rgyal tshab's own Pramanaviniscaya commentary, where he argues against a conceptualization of the use of precisely this dpyad pa gsum proffered by "others" (gzhan dag).(66) Of course, there is nothing in the book to suggest that "others" simply refers to his contemporaries, that is, other scholars active at Dga' ldan monastery, and that there would therefore be no need to push back the use of dpyad pa gsum to the pre-Tsong kha pa era. True enough, and one might even be in good company with this scenario, for, to be sure, the expression is not found in Rngog Lo tsa ba's study of a select number of issues in the Pramanaviniscaya or Gtsang nag pa Brtson 'grus seng ge's (twelfth century) examination of the same, or in such published thirteenth- and fourteenth-century tshad ma works as those by Sa skya Pandita, 'U yug pa, Chos kyi rgyal mtshan, Ston gzhon and, as far as I can tell, Bu ston Rin chen grub (1290-1364). Verily, the absence of the expression from their oeuvre suggests at a minimum that, if it did form part of the scholarly discourse of the time, its use was confined to as yet unknown critics. On the other hand, it is perhaps no accident that we witness its occurrence in the writings of Bo dong Pan chen 'Jigs med grags pa (1375-1451), Tsong kha pa's and Rgyal tshab's junior contemporary(67)--one that we can almost designate as nonchalant, as if it belonged to a common and generally agreed upon vocabulary of scholarly discourse made legitimate by a history of its application. Apart from the obvious literature, PV, I: 215 is also alluded to in the Atindriyarthadarsipurusa chapter of Santaraksita's Tattvasatigraha and in its Panjika commentary by Kamalasila,(68) which deal with these issues as well as the cognate ones of the Pramanavarttika's Pramanasiddhi chapter. Very briefly, this important verse states that for scripture (or verbal testimony) to be infallible, it cannot be invalidated by perception and two kinds of inference, one based and one not based on scripture. Scripture is valid only when these three criteria are satisfied.

Now the notion of dpyad pa gsum occurs in the context of Dharmakirti's discussion of the epistemology of reliable oral (and written) testimony, which, according to the "orthodox" interpretation, begins at PV, I: 214 and closes with PV, I: 217.(69) At the outset, Dharmakirti cites in his autocommentary Pramanasamuccaya, II: 5a-b, in which Dignaga, albeit a trifle obliquely, argues for the inclusion of knowledge gained from a "statement of a credible person" (aptavakya/aptavada) in inferential knowledge, enabling Dignaga thereby to remain fully committed to the integrity of the parallel framework of the two valid means of knowing, perception and inference, and their corresponding objects, with which he began his work, while at the same time incorporating a form of authoritative linguistic testimony into his scheme, based as it is on inference via an argument based on causality (karyahetu).(70) Pv, I: 215-16 is the locus classicus of the central "discussion" of the criteria by means of which the validity of testimony can be determined, including the dpyad pa gsum and the subsumption of credible speech under valid inference (but see also PV, II: 1-2). In his discussion of these two verses of the Pramanavarttika, Tillemans (pp. 11-12, n. 18) signals two conundrums with respect to Rgyal tshab's explanation of PV, I: 215 in his Pramanavarttika exegesis, and he comments: "But his strange reference to invalidation by 'the two types of inference functioning on the basis of [real] entities (dngos stobs zhugs kyi rjes dpag gnyis kyis),' or his classification of paroksa [lkog gyur, 'the invisible'] as empirical (drsta) [mthong ba], is most likely an indefensible interpretation of P[ramana]-V[arttika] ...." Leaving quite unarticulated the scope and ontological presuppositions of the so-called "inference which functions on the basis of [real] entities" (dngos stobs zhugs kyi rjes dpag, vastubalapravrttanumana),(71) Tillemans does remark (pp. 19, 22) that [1] for Ngag dbang bstan dar (and most, if not all Dge lugs pa philosophers) the vastubalapravrttanumana is used to establish proofs for "major Buddhist doctrinal and metaphysical views, such as the four truths, selflessness ... in other words, proofs like those found in the Pramanasiddhi ... [chapter]," that [2] according to PV, IV: 48-59,(72) "the vastubalapravrttanunana is the method for assessing propositions in treatises," and that [3] such an inference "has to function independently of all appeals to personal or textual authority." Perhaps echoing Tsong kha pa, but certainly in the Tshad rna brjed byang chen mo and in his own bona fide writings, Rgyal tshab explicitly and repeatedly states that the most emblematic of the "major Buddhist doctrinal and metaphysical views," namely the four truths and (the Buddha's) omniscience, are not "completely invisible" (atyantaparoksa, shin tu lkog gyur) at all, and thus not the probanda of "the inference based on credibility" (aptanumana, yid ches kyi / pa'i rjes dpag). Rather, they are simply "invisible" and therefore bona fide probanda for the vastubalapravrttanumana(73) (we have to leave aside the philosophical and philological problems inherent in distinguishing between "completely invisible" and "invisible"; a few notes on this question follow below). But Rgyal tshab does suggest that this was not something on which Tibetan philosophers were in universal agreement, arguing in his Tshad ma rigs pa'i gter commentary, which together with his Pramanasamuccaya and Nyayabindu exegeses,(74) belongs to his later period, that

rnam 'grel le'u dang po dang gnyis par bshad pa ltar na

/ 'phags pa'i bden bzhi dang / bden bzhi mkhyen pa'i kun

mkhyen dngos po stobs zhugs kyi rigs pas sgrub pa'i

tshul ma shes par de dag shin tu lkog gyur du bzung na

sde bdun gyi rigs pa'i rkang stor 'gro ba yin no //

If without understanding the way in which the four noble

truths and the enlightened omniscience cognizing the

four [noble] truths are proven/established by a logic that

functions on the basis of a real entity (dngos po stobs

zhugs kyi rigs pa, (*)vastubalapravrttayukti), as explained

in the first and second chapters of the [Pramana]varttika,

one were to take [as some do] those [probanda] as

completely invisible, one would be a creature who lost

the legs of the logic of [Dharmakirti's] seven treatises.

This view was also maintained at one time by Mkhas grub, and already had its express adherents in, for example, 'U yug pa and Chos kyi rgyal mtshan.(75) The application of the vastubalapravrttanumana and the ontological shifts inherent in the atyantaparoksa-paroksa substitution in the Pramanasiddhi chapter, in which Dharmakirti articulates these "major Buddhist doctrinal and metaphysical views," became major points of contention in the fifteenth century, and were rigorously argued against by a host of philosophers, including inter alia Gser mdog Pan chen, whose primary target was of course Rgyal tshab,(76) as well as, it appears, another contemporary of Tsong kha pa himself.

It would of course have been of some help had Tillemans used his excellent understanding of the issues at stake to unravel what he refers to as Rgyal tshab's "strange reference to the two types of inference" or "his classification of paroksa as empirical." As it is, he neither discusses nor resolves them, and therefore has his readers dangle in suspense and uncertainty. And it would certainly have been a propos had Tillemans done a little more textual excavation by, for example, looking into Rgyal tshab's other tshad ma writings to determine whether he dealt with these in a different and, perhaps, more convincing manner, especially because, in terms of their relative chronology, his Pramanavarttika study figures as the first of these and does not, as the tradition was well aware, always tally with the later hermeneutic results of his investigations in the corpus of Indian Buddhist pramana literature.(77) In any case, the examples Rgyal tshab gives for "the two types of inference" are:

mthong ba lkog gyur gyi don stan[supa] pa la / dngos po

stobs zhugs kyi rjes dpag gnyis kyis gnod pa med pa /

dper na / 'phags pa'i bden pa bzhi dngos stobs rjes dpag

gi gzhal byar bstan pa la gnod pa med cing des grub pa

dang / gzhan dag lkog gyur du 'dod pa'i bdag dang gtso

bo la sogs pa lkog gyur ma yin par bstan pa la gnod pa

med cing lkog gyur ma yin par grub pa lta bu'o //

(a) Read: ston.

When [scripture] indicates an invisible, empirical object, there is no invalidation by the two types of inference functioning on the basis of [real] entities, as in, for example, [1] when [scripture] has indicated the four noble truths [of the Buddhist schools] as a [valid] object of an inference that functions on the basis of [real] entities, there is no invalidation and [they] are established by it or [2] when [scripture] has indicated that an [entity with an eternalist] ontology (bdag, *atman), the principal (gtso bo, *pradhana), etc. [ontological categories], which others [= non-Buddhist schools] claim to be invisible, are [in fact] not invisible, there is no invalidation and [they] are established as not being invisible.(78)

To be sure, these two can to some extent be brought in line with Dharmakirti who, in his own comment on PV, I: 215, speaks of an "inference independent of scripture" (agamanapeksanumana, lung la bltos pa med pa'i rjes su dpag pa) as the second of the three criteria by means of which the infallibility (avisamvada, mi slu ba) of scripture or verbal testimony can be determined, and this second criterion can proceed either positively by proving Buddhist religious concepts, or negatively, by disproving non-Buddhist ones.(79) We should also add here that Sakyabuddhi (eighth century), followed by Karnakagomin (ca. 800) and Manorathanandin (eleventh century), may have been the first to interpret Dharmakirti's agamanapeksanumana as a vastubalapravrttanumana.(80) Of course, a full discussion of the complex of questions that are at stake here must be reserved for another occasion, but suffice it to say that, in my opinion, Rgyal tshab's view is then not really all that strange and should not necessarily lead to the assumption that he was either off base, or that this passage of his Pramanavarttika commentary was a victim of sloppy textual transmission.

In this connection, I should like to make two additional observations, the first of which has to do with the passage's potential contamination. The commentary's earliest xylograph from blocks that were carved in 1449 and housed in Dga' ldan monastery(81) has the identical reading of this passage, as have the other received xylographs of the same. Thus it has a precedent dating back to the earliest printed recension of the text. But Se ra Rje btsun has pointed out that the readings of this xylograph may not always have been acceptable,(82) so that the mere fact of having a precedent for this reading arguably ends up proving very little. Secondly, inasmuch as we meet with virtually the same scenario in his much later Pramanaviniscaya commentary,(83) we can still not present an airtight case, but at least we do have a stronger argument against contamination. In short, then, Rgyal tshab's precedents go back at least to Sakyabuddhi in asserting that the vastubalapravrttanumana's use in connection with scripture's infallibility has two aspects and we may add that while Mkhas grub has "by simply the inference which functions on the basis of [real] entities" (dngos stobs zhugs kyi rjes dpag nyid kyis) in the singular, he nonetheless writes:(84)

mthong ba lkog gyur gyi don ston pa la dngos po stobs

zhugs kyi rjes dpag nyid kyis gnod pa med pa / dper na

'phags pa'i bden pa bzhi gtan la phab pa la dngos po

stobs zhugs rjes dpag gis gnod pa med cing grub la /

gzhan dag lkog gyur du 'dod pa'i bdag dang gtso bo

sogs bkag pa la rjes dpag gis gnod pa med pa lta bu /

When [scripture] indicates an invisible, empirical object,

there is no invalidation by the very inference that functions

due to [real] entities, as in, for example, [1] when

[scripture] lays down the four noble truths, there is no

invalidation through an inference that functions on the

basis of [real] entities and [they] are established [by it],

and [2] when [scripture] rejects an [entity with an eternalist]

ontology, the principal, etc. [ontological categories],

which others [= non-Buddhist schools] claim to be

invisible, [it] is not invalidated by inference.

Further, Rgyal tshab's alignment of the invisible (paroksa, lkog gyur) with the "empirical" (drsta, mthong ba), noted by Tillemans as being most likely "indefensible," is not nearly as arcane as it appears at first blush and, in fact, has an important Indian precedent. Namely, as Rgyal tshab himself states in the part of his commentary to PV, I: 215 that is not cited by Tillemans, he owes it to none other than Dharmottara, Prajnakaragupta's senior contemporary. Rgyal tshab writes:

de ltar rjes dpag rnam pa gnyis kyis kyang gnod pa reed

pa'o // 'di dag ni lung 'di yi mi slu bar log pa'i[supa] rgyu

mtshan yin no // don bsdus na / shin tu lkog gyur ston pa'i

bcom ldan 'das kyi gsung chos can / rang gi bstan bya la

mi slu ba yin te / dpyad pa gsum gyis dag pa'i lung yin

pa'i phyir / 'dir mthong ba mngon gyur rkyang pa la byed

pa don ma yin te / slob dpon chos mchog gis rten mig

mthong bas kyang mthong bar brjod do zhes bshad pa

ltar / lkog gyur la yang mthong ba zhes bya'o //

(a)Read: 'jog pa'i.

Accordingly, there is no invalidation [of testimony] by

the two types of inference as well. These are the criteria

for asserting the infallibility of this [particular] scriptural

testimony [of the Buddha]. In sum, given a discourse of

the Bhagavan [Buddha] which teaches the extremely invisible,

it is infallible in relation to the subject-matter of

what is taught, because it is a scripture that has been

purified(85) (dag pa) by the three-fold analysis (dpyad pa

gsum). To have mthong ba function in this [quatrain of

PV, I: 215] only [in the sense of] the visible is not the purport

[of the passage]; as was stated by Master Dharmottara:

"Also because the eye, the physical basis [of

perception], is [?co-]perceived [?in a perceptual situation,

though not directly visible, we] speak of perception,"

so also the invisible ought to be called mthong ba.(86)

The relevant comment of Dharmottara, which has philosophical consequences that cannot be the concern of the present paper, is found in his Pramanaviniscayatika.(87) It should be mentioned here that Dharmottara's interpretation of Dharmakirti at this juncture seems unprecedented; the latter himself does not gloss drsta in this fashion, and neither do Sakyabuddhi, Karnakagomin (or, later, Manorathanandin) in any similar or systematic manner.(88) Let us for the present be content with but noting that Mkhas grub, too, has adopted this notion of Dharmottara, as did 'U yug pa.(89) The same holds for Ston gzhon, albeit not explicitly in connection with PV, I: 215, but rather with PV, IV: 108a.(90)

PV, I: 216 states, in a partial quote of and an obvious allusion to Dignaga, that sharing the defining features of the valid means of cognition as such, authoritative or believable words of a "person of authority," that is [Buddhist] scripture, are also infallible when not empirically contradicted, even if or when their content or object of reference is invisible, because we understand them through inference and there is no other way to understand them. And in PV, I: 217, Dharmakirti suggests alternatively that when the main point of Buddhism, that is, the four noble truths, is shown to be accurate and infallibly in tune with our experience, then one can infer that the same holds for other doctrinal points, as well. It is perhaps curious that neither 'U yug pa nor Ston gzhon considered PV, I: 217 as presenting an alternative scenario to the foregoing(91)--the important va, "or," of this verse was left untranslated in the Tibetan of the received translation even though the phrase of Dharmakirti's autocommentary which introduces this verse, athavanyatha "or, alternatively, ...." / athava prakarantarena, "or, by means of an alternative method ..." does not really allow for a large margin of doubt on this point.(92) Be this as it may, whereas Ston gzhon simply held this verse to be a continuation of Dharmakirti's main argument designed to provide proof for his contention that scripture qua credible words can be a valid means of knowing, 'U yug pa topicalized it more accurately as establishing that scripture is infallible as far as its essential and defining point is concerned, but he, too, did not consider it to be a different kind of argument.

Further, Tillemans (pp. 15-16) writes that PV, I: 217 "is cited by Tsong kha pa, Rgyal tshab and Ngag dbang bstan dar as being similar to Aryadeva's position in Catuhsataka XII k. 280, a karika where Aryadeva argues that because the Buddha was right on sunyata we can infer that he was right on (atyanta)paroksa too."(93) To be noted is the absence of the correlation of these two verses in Mkhas grub's oeuvre! Aryadeva's verse reads in Sanskrit and Tibetan:

buddhoktesu paroksesu jayate yasya samsayah /

ihaiva pratyayas tena kartavyah sunyatam prati //

sangs rgyas kyis gsungs lkog gyur la //

gang zhig the tshom skye 'gyur ba //

de yi stong pa nyid bsten te //

'di nyid kho nas yid ches bya

[from the Tibetan] He in whom doubt arises about the

invisible spoken of by the Buddha should, relying on

its emptiness, have confidence [in his Teachings in

general] on account of just this very [thing].

Thus Tillemans' bracketed insertion--(atyanta)paroksa in his interpretation of Catuhsataka XII k. 280 requires his reader to make several assumptions, the first of which is that when Aryadeva wrote paroksa (lkog gyur) he really meant atyantaparoksa (shin tu lkog gyur) or that for him "imperceptible" and "completely imperceptible" or "what is inaccessible to perception and ordinary inference" (Tillemans, pp. 10, 86, 88) are semantically equivalent and interchangeable in this verse. Important precedents for this equation are found in the earliest extant Indian Catuhsataka commentary by Dharmapala, but not the later study of Candrakirti(94) and the available Tibetan commentaries. And secondly, if so, then the argument stands or falls on the assumption that emptiness (sanyata) is a paradigmatically significant instantiation of what is paroksa/ atyantaparoksa. To be sure, much more could have been said about these issues than the little we find in this book, and I do think that the author ought to have at least questioned the Tibetan enterprise (I am unaware of Indian precedents) of interpreting Dharmakirti along these lines; that is, I believe Tillemans should have raised the point whether the equation of paroksa with atyantaparoksa is not only commensurate with Aryadeva's but, more importantly, also with Dharmakirti's thought, not to mention that a little more ink might have been used in explaining their respective ontologies.(95)

As far as the Pramanavarttika and the autocommentary on its first chapter are concerned, atyantaparoksa is used in the former in PV, I: 314, 316; III: 94, and IV: 210, and, discounting the repetition in the verses, on some three occasions in the latter in connection with explanations of PV, I: 289, 291, 334. Though he does not say this in so many words, Dharmakirti himself suggests their possible interchangeability in his own comment on PV, I: 316,(96) but this appears to be an isolated instance from which it would be reckless to draw more general conclusions. Indeed, the interchange might be part of a rhetorical strategy in his argumentation against the view of a member of the Mimamsa school, and for this reason might not have sufficient philosophical force for maintaining their equivalence. Looking at paroksa in PV, I: 216, one could argue that metre dictated its use rather than atyantaparoksa. It is to be noted, however, that Dharmakirti has nothing concrete to say about it in his autocommentary on this verse, but that beginning with Sakyabuddhi, and through the earliest Tibetan exegetes, 'U yug pa, Ston gzhon,(97) and beyond, paroksa is glossed by atyantaparoksa or shin tu lkog gyur. And some Tibetan writers, Chos kyi rgyal mtshan being an important representative,(98) had no problem in expressly stating that what is atyantaparoksa for Dharmakirti can be subsumed under paroksa, but the jury is still out on this question. Mention ought also be made of the subtle distinctions found in the writings of some Tibetan philosophers between the slightly invisible (cung zad lkog gyur), the paroksa, and the atyantaparoksa.

In the absence of the relevant source, Tillemans is, of course, not culpable for not having mentioned that the juxtaposition of these verses of the Pramanavarttika and Catuhsataka preceded Tsong kha pa by (at least) a hundred years, so that we can no longer claim that it had its inception with him. Indeed, exactly the same is met with in Chos kyi rgyal mtshan's exegesis of this verse in his Catuhsataka commentary,(99) and he may not have been the first to have correlated them either. Elsewhere in his work, he uses the expression tik byed gzhan dag, which refers to commentaries other than the ones by Candrakirti (and Dharmapala), and these must be of Tibetan origin.(100) Tibetan studies of the Catuhsataka that certainly predate it by a wide margin include the as yet to be recovered one by Zhang Thang sag pa Ye shes 'byung gnas (eleventh-twelfth centuries) and a brief study by Gtsang nag pa, which I noted among the holdings of the Cultural Palace of Nationalities, Beijing.(101) So far, we do not know when or where Chos kyi rgyal mtshan wrote his text, for it has no colophon. All I can say at present is that he must have composed it after he had completed his own work on philosophical systems, Grub mtha' rgyan gyi me tog, "The Philosophical Systems: A Flower of a Bouquet," which too has yet to be discovered, for he refers to it.(102) The latter, in turn, may have been written in the middle of his career as a commentator on a large number of canonical treatises, for he also mentions it in his Mulamadhyamakakarika exegesis,(103) but not in two of his writings on tshad ma that were possibly written in the late 1270s.(104) This absence is probably not entirely insignificant, inasmuch as both afford numerous occasions on which he might have drawn the reader's attention to further discussions in Grub mtha' rgyan gyi me tog, had there been already a text in existence with this title.

It is perhaps merely of passing interest that he begins his study of the Catuhsataka's twelfth chapter with a quotation from the Rnam 'grel gyi rgyan, which of course refers to Prajnakaragupta's Pramanavarttikalamkara. But when we compare his reading of the cited verse in Tibetan with the ones of the translation found in the Tibetan canon and the Sanskrit text, the citation becomes all the more intriguing because of the remarkable differences that exist among them.(105) In all, Chos kyi rgyal mtshan's Tshad ma sde bdun rgyan gyi me tog, his topical study of Buddhist tshad ma, contains three citations, not paraphrases, of Prajnakaragupta. The first occurs in the text's first chapter, which was studied by P. Schwabland,(106) and there we read the following:

gzhal bya'i rang bzhin grub pa nyid //

der ni bya ba yin par 'dod //

ri mo la yang mthong tsam gyis //

'bras bu yongs su rdzogs pa yin //

The canonical [Sde dge] recension reads the second line as de ni byed pa yin par 'dod //, and the Sanskrit text of the verse is:(107)

jneyasvarupasamsiddhir eva tatra kriya mata /

citre 'pi drstimatrena phalam parisamaptimat //

No really significant conclusions can be drawn from the fact that Sanskrit jneya and kriya are rendered by gzhal bya and, respectively, bya ba and byed pa. For the former one would really expect Tibetan shes bya, and this relatively minor anomaly is yet another curious feature of the vocabulary of Tibetan tshad ma texts.(108) Sanskrit tatra, however, needs here to be translated by der, so that de is simply wrong, and is probably owed to the chirographic fact that the final -r of der may easily have been omitted in the course of the manuscript transmission of the text.

The second quotation is taken from the discussion of immediate perception that is mind-based in the text's second chapter; it reads:(109)

goms phyir mdun na gnas pa la //

'di 'o zhes sogs shes pa gang //

mngon du byed phyir de la ni //

yid kyi mngon sum yin par 'dod //

The difference between the canonical reading of goms las and goms phyir of Chos kyi rgyal mtshan's text is of negligible significance in view of Sanskrit abhyasat,(110) Lastly, the third quotation reads:(111)

ji ltar tshig las mthong ba yi //

'brel pa dran par gyur pa na //

rjes dpag skye ba de bzhin du //

'gar yang mngon sum skye ma yin //

And this reading is exactly the same as the one found in the canon.(112) In other words, Chos kyi rgyal mtshan's citations of three of Prajfnakaragupta's verses do not shed any light on the origin of the major differences in the Tibetan versions of the verse he quotes in his Catuhsataka commentary. It was not quoted from the readings gained from the subcommentaries by Jayanta (eleventh century) and Yamari (after Jayanta), but it is possible that this quotation may have been taken from an as yet unidentified canonical (or extra- or para-canonical) source. We know that at least two different translations of Prajnakaragupta's study, one by Zangs dkar Lo tsa ba 'Phags pa shes rab and one by Rngog Lo tsa ba and Bhavyaraja,(113) existed at one time, that only the latter of these was included in the Tibetan canon and that already Chos kyi rgyal mtshan lists it (and not the former), in his catalogue of translated Buddhist scripture.(114) This renders it unlikely that he had access to Zangs dkar Lo tsa ba's translation, if this reading of the verse were to be found therein. Another possible source for the verse in the Catuhsataka exegesis is that it was lifted from another Tibetan work in which it was cited in a somewhat corrupt form. And lastly, we cannot entirely exclude the fairly remote possibility that Chos kyi rgyal mtshan, himself a Sanskritist of some repute (he is sometimes referred to as Snar thang Lo tsa ba) and a student of Chag Lo tsa ba Chos rje dpal (1197-1264), one of the finest Sanskritists of his day, was responsible for the rendition of this verse, meaning that he had access to a Sanskrit manuscript of the text, if not one or both of the two manuscripts R. Sankrtyayana and Dge 'dun chos 'phel (1903-52) discovered in the 1930s among the immense holdings of Sa skya monastery's Lha khang chen mo. One of these was a complete paper manuscript of the text that had belonged to Vibhnuticandra,(115) while the other was an incomplete palm-leaf manuscript of the second and third chapters that had been the property of his cotraveler and associate Danagila (presumably, Danagila's original manuscript had been complete).(116) We may note here parenthetically that Chos kyi rgyal mtshan traces the origins of his understanding of Indian tshad ma back to Danasila, who in effect was his "great grand-teacher." Now, if these men had left their manuscripts at Sa skya when they stayed there for some time in the first decade of the thirteenth century, then this scenario is nonetheless rather unlikely, presupposing as it does that Chos kyi rgyal mtshan, who was not a Sa skya pa but a Bka' gdams pa monk with no recorded upper echelon connections in Sa skya, had been free to roam and read at leisure in Sa skya monastery's libraries. This would be extremely unusual and, in fact, contrary to what we know so far about the kinds of (very limited) access the average monk and the scholar-monk enjoyed in those libraries that were not of his own monastery. Of course, Chos kyi rgyal mtshan had been a major scholar active in a number of Bka' gdams pa institutions, especially in the famous monastery of Snar thang where, as Dge 'dun chos 'phel remarked with a sense of surprise, contrary to his expectations, he was unable to locate any Sanskrit manuscripts.(117) Further, Schwabland has pointed out that Chos kyi rgyal mtshan was quite at odds (for how long we do not know) with 'Phags pa Bio gros rgyal mtshan (1235-80), Sa skya monastery's fifth patriarch, and that he also frequently took critical issue with a number of points raised by Sa skya Pandita's Tshad ma rigs pa'i gter and autocommentary in his own topical and polemic study of Indo-Tibetan tshad ma.(118) To be sure, there is plenty of other evidence to indicate not only that Chos kyi rgyal mtshan was rather well acquainted with what Sa skya Pandita had written, but also that he felt the need to disagree with it so often "in print" as to make him easily a persona non grata among at least the rank and file of Sa skya. For one, his Lung dgongs pa rgyan gyi me tog, "The Intention of Scripture: A Flower for the Ornament [of the Buddha's Teaching],"(119) is quite critical of Sa skya Pandita's Thubpa'i dgongs gsal, "A Clarification of the Intention of the Sage [the Buddha]," as it is of the considerations of a good number of other scholars. In addition, his Sgra'i bstan bcos smra ba rgyan gyi me tog ngag gi dbang phyug grub pa, "A Treatise of Linguistics: A Flower of a Bouquet [on] Language; Realizing [Oneself as] the Lord of Speech," a study of Indo-Tibetan linguistics,(120) refers critically to the latter's notion of the so-called uninterrupted dandaka (rgyun chags ma bcad pa), a literary genre in which a text consists of one (very) long metrical foot, for the legitimacy of which Sa skya Pandita had argued in several of his writings.(121) Ironically, perhaps, Chos kyi rgyal mtshan's oeuvre never seems to have had an impact in Tibetan scholastic circles and appears to have enjoyed such extremely limited circulation that the vast majority of later generations of scholars were plainly unaware of it or were able to ignore it without serious repercussions to their reputation. This would explain why they usually ended up referring to the very same critique Dpang Lo tsa ba levelled against the literary genre of the uninterrupted dandaka one generation later, for which, even though Dpang Lo tsa ba does not say so, Chos kyi rgyal mtshan may very well have been the catalyst. But some pockets of Tibetan scholarship continued to be interested in his writings. Indeed, we come across what appears to be a fairly isolated reference to his critique in a long interlineary gloss in a manuscript on costly glossy-white paper of Ta'i si tu V(?) Chos kyi rgyal mtshan dge legs dpal bzang po's (1586-1632)(122)exegesis of the Sanskrit grammar Kalapasutra (or Katantra). According to its colophon,(123) the text was first begun under the direct patronage of the house of the Gtsang pa Sde srid, specifically either Karma Phun tshogs rnam rgyal or his son Karma Bstan skyong dbang po (given the other manuscripts that I have seen in which this same kind of paper was used, I think it most probable that they also provided the writing materials(124) and was then completed in Rgyal mkhar rtse, which at this time was also controlled by the Gtsang pa ruling house. The Sanskritist Lha mdong Lo tsa ba Se na gu ru karma sha sa na dha ra (= Tib.: Tshogs / Sde bla ma Karma bstan 'dzin)(125) apparently put the finishing touches on Ta'i si tu's study.

To return briefly to where we began the discussion, Chos kyi rgyal mtshan's comment on verse 280 of the Catuhsataka reads as follows:

stong pa la nyon mongs pa mi skye bar mngon sum gyis

grub pa de bzhin du sangs rgyas kyis gsungs pa'i lkog

gyur bden bzhi dang bzhi po de'i revu 'bras kyi 'brel pa

mtha' med pa so so skye ba'i mngon sum gyis mi shes

par the tshom skye ba de yi stong pa nyid kyi dpe la

bsten te gsung rab las gsungs pa 'dj nyid kho na yin par

yid ches bya ste shin tu lkog gyur gyi don ston par byed

pa'i gsung rab chos can don ji lta ba bzhin yin te sangs

rgyas kyi gsung yin pa'i phyir stong nyid ston pa'i mdo

bzhin no zhes bya'o // rnam 'grel las /

blang dang dor bya'i de nyid ni //

thabs bcas rab tu nges pa yis //

gtso bo'i don la mi bslu'i(a) phyir //

gzhan la rjes su dpag pa yin//

zhes bden 4 la mi bslu bas gzhan la'ang dpog par bshad

do //

(a)Read slut

It is obvious that conflicting emotions do not occur in

what is empty, so likely the four truths and the endless

linking of their cause and effect, invisible [objects] spoken

of by the Buddha, give rise to doubt, not being evidentially

understood by an ordinary person; on the basis of an

illustration of their emptiness, one ought to have the

confidence that what is stated by scripture is quite the

case; given scripture that indicates the completely invisible,

it is exactly in accordance with the facts, because

it is a statement of the Buddha; like a sutra teaching

emptiness. [Pramana]varttika [I: 217], states:

Because, by means of having ascertained the nature of

what is to be rejected and accepted together with [their]

cause[s, scripture] is [understood to be] infallible with

respect to the principal object, [its infallibility with respect

to] other [subsidiary topics of scripture] is inferential[ly

ascertained as well].(126)

This suggests that, inasmuch as [the Buddha's word] is

infallible with respect to the four truths, [the same] is

inferable with respect to the other [topics of his discourses/

scripture], as well.

As we can expect from Tillemans, his translation of the technical Tibetan of Ngag dbang bstan dar's text reads very well--no easy task--and eloquently evinces his philological and philosophical control over the subject matter of the text. The first verse of its invocation might be understood a trifle differently. The Tibetan and his rendition read:

rten 'byung 'chad pa'i ngag la dbang bsgyur cing //

de nyid mkhyen pa'i blo stobs rab rgyas pas //

skal bzang thar pa'i lam la 'dgud mkhas pa'i //

bshes gnyen mchog des kun tu srung gyur cig //

May we be protected always by the excellent [spiritual]

friend, who is skilled in leading the fortunate onto the

path to liberation because of his having mastered the

explanation of dependent arising and because of the expansive

development of the force of his mind which

knows the truth.

Giving the phrase 'chad pa'i ngag its legitimate due and underscoring the "spatial" rather than the "temporal" quality of kun tu, we could read this slightly differently:

May [we] be protected in all [respects] by the supreme

[spiritual] friend who, due to [his] mastery over the

speech that explains dependent arising and [due to his]

expansive, powerful intellect which intuits reality, is

skilled in leading the fortunate onto the path of liberation.

Another passage where I should wish to propose a different interpretation occurs on page 53:

de dang rgyud gcig tu gyur pa'i mthong lam rnam grol

lam pa ba chos can / rang rgyud la bde gshegs zhes pa'i

skabs nas bstan pa'i khyad par gsum ldan gyi spangs pa

bde gshegs 'byung rung yin te / bdag med mngon sum du

rtogs pa'i byang sems mthong lam pa yin pa'i phyir /

which Tillemans translates (p. 52) as:

Take as a topic the [bodhisattva] who is of the same

continuum as [Sakyamuni] and who is on the path of

deliverance (rnam grol lam = vimuktimarga) of the path

of seeing (darsanamarga); the sugata of abandonments

endowed [with] three qualities spoken about in the context

of the [epithet] sugata [127] can arise in his continuum;

because he is a bodhisattva on the path of seeing

who directly realizes selflessness.

There are several places in this passage where I should like to differ from this interpretation. In the first place, I do not think we should understand mthong lam rnam grol lam pa ba as a compound with a covert objective-genitive particle gyi, as in (*)mthong lam gyi rnam grol lam pa ba, "the [bodhisattva who] is on the path of deliverance of the path of seeing," if only because rnam grol lam is the larger Oberbegriff. Rather, I would argue that mthong lam is to be construed as standing in a restrictive appositional relation to rnam grol lam pa ba. Moreover, neither copulative-equative yin construction in the second and third main phrases is, in my view, adequately represented in Tillemans' version, and the construction "a bodhisattva on the path of seeing" in the last phrase is not grammatically possible. Similarly, I should understand the phrase khyad par gsum ldan gyi spangs pa bde gshegs not as "the sugata of abandonments endowed *with* three qualities," which is a bit wooden and counterintuitive and based on a not altogether warranted insertion of a genitive *spangs pa'i, but rather as a simple apposition, by means of which we obtain "the sugata, the one whose renunciation consists of three qualities." Hence, I would propose to read the argument as follows:

Take as a topic one who is on the path of deliverance,

the path of seeing, who has the same continuum as he

[Sakyamuni]; [he] is one in whose continuum sugata

[the one whose renunciation consists of three qualities

that were mentioned in the section on sugata] can occur;

because a bodhisattva who has directly realized selflessness

is one who is on the path of seeing.

Another minor point is that "Blo mthun bsam gtan" (p. viii) is ambiguous and should preferably be read "Bio mthun Bsam gtan"; blo mthun, "comrade," is (already) an anachronism, and bsam gtan is part of the actual name of the controversial author of Dmu dge monastery in Khams, who passed away not long ago in 1993 at the age of seventy-nine.(128) The name he often used in his writings on language and linguistics, that is, his sgra rig pa name, was "Bsam gtan rgya mtsho mi 'jigs dbyangs can dga' ba'i blo gros." Lastly, in the copy of the book Tillemans kindly sent me, there is one marginal note (p. 5, line 9): "The usual sense of pramana is ..."; this should now read "The usual technical sense of pramana is...."

In sum, then, with the above text-historical and ideengeschichtliche restrictions and qualifications in mind, this booklet, though certainly not the last word on the subject, is valuable both for the questions it raises and for the answers it seeks to provide. In Tibet, in spite of canonical precedents, the term tshad ma'i skyes bu, "the person of authority," and some of its cognates, as epithets of the Buddha or one who is regarded as enlightened, whether used hyperbolically or not, did not quickly gain widespread currency and was used very infrequently before the fifteenth century. More research will be needed before one can suggest that the frequency of its use began to change when the politics of enlightenment (Tsong kha pa as a tshad ma'i skyes bu) first began to play a predominant role in the newly configured and reconfigured school of the Dga' ldan pa (fifteenth century) and the Dge lugs pa (sixteenth century), and then in the Tibetan polity headed by the fifth Dalai Lama and his institution of the Dga' ldan pho brang (Dalai Lama as a tshad ma'i skyes bu) during the middle of the seventeenth century. At the same time, it is important to bear in mind that a good number of the issues addressed in the booklet under review, which its author implied or explicitly stated to be specifically Dga' ldan pa / Dge lugs pa insights, need to be redirected to the chronological space behind the thinkers belonging to this school of exegesis, much of which is still quite empty. There is no question that Tillemans would rightly disagree with the naive but by no means uncommon assumption, held by the pious and newly converted, that such Dga' ldan pa philosophers as Tsong kha pa, Rgyal tshab, or Mkhas grub operated in an intellectual vacuum, and that they owed very little to their Tibetan forebears and contemporaries. But, given that they did, it then becomes important to make a concerted attempt at ascertaining what it was they had or might have inherited, conscious or otherwise, and why and at what point they became responsible for the unprecedented innovations--sngon med as the Tibetans would say--in "the established scholarly tradition," for which especially Tsong kha pa was accused and criticized by some of his more "traditional" Sa skya pa contemporaries. Tsong kha pa was one of many Tibetan scholar-saints who was privy to a series of visions and other spiritual experiences, which he credited with his several departures from what had virtually been a hermeneutic norm (if there ever were such a thing).(129) And some of his contemporaries severely contested the authenticity of these visions (of, allegedly, Mafijusri) and what he had allegedly learned from the instructions of this bodhisattva, visions and teachings that, and here, I recklessly interpret, took place in response to, or in competition with, Dol po pa's truly unprecedented interpretations of Mahayana Buddhism about half a century before. There is no doubt that Tsong kha pa was an important (and later, as his stature grew with the increasingly influential Dge lugs pa church, the most important) figure in the drive to include tshad ma in the Buddhist spiritual experience, a notion that is also emphasized by Mkhas grub in his record of teachings received, where he writes that Tsong kha pa had taught him the way in which the text of the Pramanavarttika could be made into a progressive spiritual practice (rnam 'grel gyi gzhung nyams len gyi rim par bsgrigs pa'i tshul).(130) But he was not the only one of his day to have pointed tshad ma or the Pramanavarttika in this direction. Both Rong ston and Bo dong Pan chen(131) are known to have made similar efforts, although it is far from transparent whence the similar approaches enunciated by them had their origin(s). Whereas Tsong kha pa may indeed have been among the first actively to legitimize the notion of dpyad pa gsum (he may even have coined it, although I have my doubts on this score), the correlation of PV, I: 217 and Catuhsataka 280, of which much has been made in the secondary literature and by Tillemans, did not begin with him. As it now turns out, this juxtaposition had already enjoyed a rather long history in Tibet before it was adopted and repeated by Tsong kha pa and Rgyal tshab. Lastly, but not insignificantly, the title of Ston gzhon's Pramanavarttika exegesis might to some extent betray a possible soteriological presupposition, for it quite self-consciously seeks to clarify what "the three factors" (gnas gsum, *tristhanani--this is an allusion to the notion of trtiyasthana, "the third factor," the "ontology" of scriptural authority, of PV, IV: 51c) are all about, that is, the objects of perception (= aparoksa/pratyaksa), inference (= paroksa) and scriptural/verbal authority (= atyantaparoksa), and not merely those of the first two of these. To be sure, this has everything to do with what he perceived the motive of the Pramanavarttika to have been. On the other hand, his earlier work on tshad ma whose title he quotes in this (later) work of his--Tshad ma rigs pa'i de kho na nyid snang ba,(132) "Tshad ma: Illumination of the Nature of Logic"--betrays nothing of the sort. But then it was not a study of the Pramanavarttika per se.


BO Bo dong Pan chen 'Jigs med grags pa. Tshad ma rigs pa'i snang ba, Collected Works, vol. 8. New Delhi: The Tibet House, 1970.

BOD Bod rgya tshig mdzod chen mo, vol. 1. Ed. Zhang Yisun. Beijing: Mi rigs dpe skrun khang, 1985). 'BRUG 'Brug Rgyal dbang Chos rje. 'Jam mgon chos kyi rgyal po tsong kha pa chen po'i rnam thar. Ed. Tshe tan Zhabs drung. Xining: Mtsho sngon mi rigs dpe skrun khang, 1981.

CHOS Chos kyi rgyal mtshan. Tshad ma sde bdun rgyan gyi me tog. Ed. Rdo rje rgyal po. Xining: Krung go'i bod kyi shes rig dpe skrun khang, 1991. Pp. 3-138.

CHOS(m) Idem. Handwritten dbu med manuscript, 95 folios. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. Catalogue, no. 002468(2).

CHOS(1) Idem. Bstan pa sangs rgyas pa rgyan gyi me tog. Handwritten dbu med manuscript, 38 folios. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. catalogue no. 005968.

CHOS(2) Idem. Rnal 'byor spyod pa'i bzhi brgya pa rgyan gyi me tog. Handwritten dbu med manuscript, 71 folios. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. catalogue no. 007316(8).

DGE Dge 'dun chos 'phel. Thog mar lha sa nas phebs thon mdzad pa'i tshuh Dge 'dun chos 'phel gyi gsung rtsom, vol. 1. Gangs can rig mdzod, 10. Lhasa: Bod ljongs bod yig dpe rnying dpe skrun khang, 1990. Pp. 3-40.

MKHAS Mkhas grub Dge legs dpal bzang po. Tshad ma sde bdun gyi rgyan yid kyi mun sel, Collected Works [Lhasa Zhol print], vol. Tha. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 3-449.

MKHAS(1) Idem. Rgyas pa'i bstan bcos tshad ma rnam 'grel gyi rgya cher bshad pa rigs pa'i rgya mtsho las rang don le'u'i rnam bshad pa, Collected Works [Lhasa Zhol print], vol. Tha. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 611-1001.

MKHAS(2) Idem. Rgyas pa'i bstan bcos tshad ma rnam 'grel gyi rgya cher bshad pa rigs pa'i rgya mtsho las le'u gnyis pa'i rgya cher bshad pa, Collected Works [Lhasa Zhol print], vol. Dha. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 3-218.

PV Dharmakirti. Pramanavarttikakarika Ed. Y. Miyasaka. Acta Indologica, vol. 2 (1971-72); with the usual provisions about the sequence of the chapters.

PVBH(t) Prajnakaragupta. Pramanavarttikalamkara [Tibetan translation]. SDE, vol. 47, no. 4226 [4221]. Pp. 176.2-344.3 [Te la-De 282a].

PVBH(s) Idem. Pramanavarttikalamkara [or Pramanavarttikabhasya]. Ed. R. Sankrtyayana. Tibetan Sanskrit Works, vol. 1. Patna: Kashi Prasad Jayaswal Research Institute, 1953.

PVP Devendrabuddhi. Pramanavarttikapanjika [Tibetan translation]. SDE, vol. 46, no. 4222 [4217]. Pp. 487.2-580.2 [Che la-327a].

PVSV The Pramanavarttikam of Dharmakirti. Ed. R. Gnoli. Series Orientale Roma, vol. 23. Rome: Istituto Italiano per il Medio ed Estremo Oriente, 1960.

PVSV(t) Dharmakirti. Pramanavarttikasvavrtti [Tibetan translation]. SOE, vol. 46, no. 4221 [4216]. Pp. 456.4-486.1 [Ce 261b-365a].

PVSVT Karnakagomin. Pramanavarttikasvavrttitika Ed. R. Sarikrtyayana. Allahabad: Kitab Mahal, 1943.

PVT Sakyabuddhi. Pramanavarttikafika [Tibetan translation]. SDE, vol. 47, no. 4225 [4220]. Pp. 1.1-175.3 [Je la-Nye 282a].

PVV Manorathanandin. Pramanavarttikavrtti. Ed. S. D. Shastri. Bauddha Bharati Series, vol. 3. Varanasi: Bauddha Bharati, 1968.

RDO(1h) Rdo ring Bstan 'dzin dpal 'byor. Dga' bzhi ba'i mi rabs kyi byung ba brjod pa zol med gtam gyi rol mtsho. Ed. Tshe ring phun tshogs. Lhasa: Bod ljongs mi dmangs dpe skrun khang, 1988.

RDO(ch) Idem. Rdo ring pandita'i rnam thar. 2 vols. Ed. Luo Runcang. Chengdu: Si khron mi rigs dpe skrun khang, 1987.

RGYAL Rgyal tshab Dar ma rin chen. Tshad ma rnam 'grel gyi tshig le'ur byas pa'i rnam bshad thar lam phyin ci ma log par gsal bar byed pa. Collected Works [Lhasa Zhol print], vol. Cha. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 3-874.

RGYAL(1) Idem. Bstan bcos tshad ma rnam nges kyi fik chen dgongs pa rab gsal (smad cha). Collected Works [Lhasa Zhol print], vol. Nya. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 1-521.

RONG Rong ston Shakya rgyal mtshan. Gsung 'bum thor bu['i dkar chag dpag bsam ljon pa]. Handwritten dbu med manuscript, 203 folios. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. Catalogue no. 003891.

SDE The Tibetan Tripitaka. Taipei Edition [Sde dge print]. 53 vols. Ed. A. W. Barber. Taipei: SMC Publishing Inc., 1991.

SE Se ra Rje btsun Chos kyi rgyal mtshan. Rgyas pa'i bstan bcos tshad ma rnam 'grel gyi don 'grel rgyal tshab dgongs pa tab gsal. Se ra Byes xylograph, 122 folios. Ed. Bio bzang 'gyur med. No date.

STAG Stag tshang Lo tsa ba Shes rab rin chen. Rig gnas kun shes nas bdag med grub pa. Dga' ldan phun tshogs gling xylograph of 1666, 11 folios. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. catalogue no. 004330(9).

STON Btsun pa Ston gzhon. Tshad ma rnam 'grel gyi rnam par bshad pa gnas gsum gsal ba gangs can gyi rgyan. Ed. Sun Wenjing. Xining: Krung go'i bod kyi shes rig dpe skrun khang, 1993.

STON(1) Idem. Handwritten dbu med manuscript, 191 folios. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. catalogue no. 005148(4).

TSONG Rgyal tshab Dar ma rin chen. Tshad ma'i brjed byang chen mo. Collected Works [Lhasa Zhol print], vol. Da. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 581-663.

TSONG(1) Idem. Collected Works of Tsong kha pa [Lhasa Zhol print], vol. Pha. Dharamsala: Library of Tibetan Works and Archives, 1981. Pp. 810-95.

TSONG(2) Idem. Collected Works of Tsong kha pa [Sku 'bum print], vol. Pha. Xining: Mtsho sngon mi rigs dpe skrun khang, 1987. Pp. 677-749.

(*)TSONG Ke'u tshang sprul sku Bio bzang 'jam dbyangs smon lam. Tshad ma'i lam bsgrigs chos kyi rgyal po tsong kha pa chen pos mdzad pa. Collected Works, vol. Kha. Dharamsala: Library of Tibetan Works and Archives, 1984. Pp. 767-810.

(*) [TSONG(1) ?Tsong kha pa Blo bzang grags pa. Tshad ma'i lam gyi rim pa'i sgrigs / chos kyi rgyal po btsong kha pa chen pos mdzad par grags pa. Handwritten dbu med manuscript, folios 34. China Nationalities Library of the Cultural Palace of Nationalities, Beijing. C.P.N. catalogue no. 004784(1).

'u 'U yug pa Rigs pa'i seng ge. Tshad ma rnam 'grel gyi 'grel pa rigs pa'i mdzod. 2 vols. New Delhi, 1982.


van Bijlert, V. A. 1989. Epistemology and Spiritual Authority. Wiener Studien zur tibetologie und buddhismuskunde, vol. 20. Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat. Wien.

Franco, Eli. 1994. Yet Another Look at the Framework of the Pramanasiddhi Chapter of the Pramanavarttika. Indo-lranian Journal 37: 233-52.

-- 1997. Dharmakirti on Compassion and Rebirth. Wiener Studien zur Tibetologie und Buddhismuskunde, vol. 38. Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat Wien.

Gdong drug snyems blo, Dor zhi. 1987. Bod kyi brda sprod rig pa'i don 'grelphyogs bsgrigs. Lanzhou: Kan su'u mi rigs dpe skrun khang.

Jackson, D. P. 1994. The Status of Pramana Doctrine According to Sa skya Pandita and Other Tibetan Masters: Theoretical Discipline or Doctrine of Liberation? The Buddhist Forum, vol. 3 [D. S. Ruegg Festschrift]. Eds. T. Skorupski and U. Pagel. London: School of Oriental and African Studies, University of London. Pp. 85-129.

van der Kuijp, L. W. J. 1994. On Some Early Tibetan Pramanavada Texts of the China Nationalities Library of the Cultural Palace of Nationalities in Beijing. Journal of Buddhist and Tibetan Studies 1: 1-30.

--. Forthcoming. Rgyal tshab Dar ma tin chen (1364-1432), His Exegesis of Sa skya Pandita's Tshad ma rigs pa'i gter and Related Matters.

Schwabland, P. A. 1994. "A General Exposition of Valid Cognition: The Initial chapter of Bcom ldan Ral gri's Introduction to Indian and Tibetan Buddhist Epistemology" (M.A. thesis, Univ. of Washington).

Skal bzang 'gyur med. 1981. Bod kyi brda sprod rig pa'i khrid rgyun rab gsal me long. Chengdun: Si khron mi rigs dpe skrun khang.

Tillemans, T. J. E 1990. Materials for the Study of Aryadeva, Dharmapala and Candrakirti. 2 vols. Wiener Studien zur Tibetologie und Buddhismuskunde, vol. 24. Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat Wien.

Tillemans, T. J. E, and D. D. Herforth. 1989. Agents andActions in Classical Tibetan: The Indigenous Grammarians on bdag and gzhan and bya byed las gsum. Wiener Studien zur Tibetologie und Buddhismuskunde, vol. 21. Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat Wien.

Tournadre, N. 1995. Tibetan Ergativity and the Trajectory Model. In New Horizons in Tibeto-Burman Morphosyntax.

Ed. Y. Nishi et al. Senti Ethnological Studies, no. 41. Osaka: National Museum of Ethnology: 261-75.

Yaita, H. 1987. Dharmakirti on the Authory of Buddhist Scriptures (agama): An Annotated Translation of the Pramanavarttikasvavrtti. Nanto Bukkyo [Journal of the Nanto Society for Buddhist Studies] 58: 1-17.

This is a review article of: Persons of Authority: The sTon pa tshad ma'i skyes bur sgrub pa'i gram ora lag sha Ngag dbang bstan dar, A Tibetan Work on the Central Religious Questions in Buddhist Epistemology. By TOM J. F. TILLEMANS. Tibetan and Indo-Tibetan Studies, vol. 5. Stuttgart: FRANZ STEINER VERLAG, 1993. Pp. iv + 91. DM 46.

This paper incorporates some of the results obtained during my stay in Beijing from October to December of 1992 and from July to September of 1993 that was made possible by a generous grant from the Committee on Scholarly Communication with the People's Republic of China (now Committee on Scholarly Communication with China), Washington, D.C. I am also indebted to my friend E. Franco for having carefully read through a much earlier version of this paper and for having pointed out some errors. Of course, I am entirely responsible for whatever misrepresentations or outright mistakes remain in the present essay.

(1) See, respectively, his "Dharmakirti, Aryadeva and Dharmapala," Tetsugaku 38: 31-47; and Materials for the Study of Aryadeva (1990), I: 23-35.

(2) The year of his passing is often given as 1840, but in the absence of relevant sources we can only safely say that he must have died after he completed his last known work, a study on spiritual purification (bio sbyong), which he finished on July 14, 1839; see his Collected Works, vol. Kha (New Delhi, 1971), 114.

(3) For this, see his autobiographical note in his Collected Works, vol. Kha, 747.

(4) See Bibliotheca Buddhica, vol. XIX (Petrograd, 1916), iii ff., and the Die pekinger lamaistischen Blockdrucke in mongolischer Sprache (Wiesbaden: Otto Harrassowitz, 1954), 164-65.

(5) We find this in his "Zu einigen Texten der tibetischen brdagsar-rnying Literatur," Asienwissenschafiliche Beitrage, ed. E. Richter and M. Taube (Berlin: Akademie-Verlag, 1978), 189-90, n. 132.

(6) See his "Das 'Kelen-u cimeg' des Ngag dbang bstan dar: Ein Beitrag zur einheimischen mongolischen Grammatik," Wissen-schafiliche Zeitschrift der Karl-Marx-Universitat Leipzig 10 (1961): 147-55.

(7) A. Klein, Knowing, Naming and Negation: A Sourcebook on Tibetan Sautrantika (Ithaca: Snow Lion Publications, 1991), 41-87; Tillemans and Herforth, Agents and Actions in Classical Tibetan (1989), Index, 107.

(8) Collected Works, vol. Kha (New Delhi, 1971), 571-89.

(9) For a few of these, see my "Studies in the Life and Thought of Mkhas grub rje, IV: Mkhas grub rje on Regionalisms and Dialects," Berliner indologische Studien 2 (1986): 49, n. 15.

(10) See his "Tshad ma'i skyes bu: Meaning and Historical Significance of the Term," in Contributions on Tibetan and Buddhist Religion and Culture, vol. 2, ed. E. Steinkellner and H. Tauscher, Wiener Studien zur Tibetologie und Buddhismuskunde, vol. 22 (Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat Wien, 1983), 275-84.

(11) Pramanavarttikalamkaratikasuparisuddha, SDE, vol. 48, no. 4231 [4226] 166.1 (Be, 109b).

(12) For this, see his "Was the Buddha a Buddha?" Journal of Indian Philosophy 17 (1989): 84, n. 8.

(13) See his "Pramanabhuta, *Pramana(bhuta)-purusa, Pratyaksadharman and Saksatkrtadharman as Epithets of the rsi, acarya and tathagata in Grammatical, Epistemological and Madhyamaka Texts," Bulletin of the School of Oriental and African Studies 57 (1994): 303-20; "La notion du voyant et du <<connaisseur supreme>> et la question de l'autorite epistemique," Wiener Zeitschrift fiir die Kunde Sudasiens 38 (1994): 403-19; and "Validity and Authority or Cognitive Rightness and Pragmatic Efficacy? On the Concepts of Pramana, Pramanabhuta, and Pramana(bhuta)purusa," Asiatische Studien / Etudes asiatiques 49 (1995): 817-27. But see now also the remarks in Franco, Dharmakirti on Compassion (1997), 16-17, n. 3. For pramanabhuta versus pramanibhuta as an epithet of the Buddha, see, inter alia, the note in H. Krasser's review of Tillemans in the Bulletin of the School for Oriental and African Studies 61 (1996): 180-81.

(14) Further details are found in my "Fourteenth Century Tibetan Cultural History, IV: The Tshad ma'i byung tshul 'chad nyan gyi rgyan, A Tibetan History of Indian Buddhist Pramanavada," in Festschrift Klaus Bruhn zur Vollendung des 65. Lebensjahres dargebracht von Schalern, Freunden und Kollegen, ed. N. Balbir and J. Bautze (Reinbek: Dr. Inge Wezler Verlag, 1994), 376, n. 2.

(15) This expression is found in Skal ldan grags pa (Bhavyakirti), Sgron ma gsal bar byed pa'i dgongs pa rab gsai zhes bya ba bshad pa'i tika (*Pradipoddyotananusandhiprakasika namavyakhyatika), SDE, vol. 25, no. 1795 (1793), 411.6 [Ki, 63a]; Bhavyakirti appears to have flourished in the eighth or ninth centuries. The more pejorative "path of speculative logic" (*tarkamarga) occurs in the first line of the second of the three concluding verses of the sixth and last chapter of Dignaga's Pramanasamuccaya.

(16) For a detailed discussion of some of the issues involved, see S. Kimura, "The Position of Logic in Tibetan Buddhism [in Japanese]," in Buddhism and Society in Tibet, ed. Z. Yamaguchi (Tokyo: Shunju-sha, 1986), 365-401. This paper was commented on in detail in Jackson, "The Status of Pramana Doctrine" (1994), 125-28.

(17) Jackson (1994), 107-8.

(18) See D. E Jackson, "Sources for the Study of Tibetan Pramana Traditions Preserved at the Bihar Research Society, Patna," in Studies in the Buddhist Epistemological Tradition, ed. E. Steinkellner (Wien: Verlag der Osterreichischen Akademie der Wissenschaften, 1991), 102; and Jackson (1994), 107.

(19) STAG, 3b; see also the discussion in Jackson (1994), 119-21.

(20) See his Rje btsun tsong kha pa'i rnam thar chen mo'i zur 'debs rnam thar legs bshad kun 'dus, Collected Works [of Tsong kha pa, Lhasa Zhol print], vol. Ka (Dharamsala: Library of Tibetan Works and Archives, 1981), 154.

(21) STAG, 3b-4a.

(22) The form brnyed rather than rnyed (see also the discussion below), is used in Tournadre, "Tibetan Ergativity and the Trajectory Model" (1995), 265, as a root of a verb. While the former is registered in such well-known dictionaries as H. Jaschke and S. C. Das, both, as well the Dictionnaire Thibetain-Latin-Francais (Hong Kong: Imprimerie de la Soci6t6 des missions etrangeres, 1899), 390, 399, have brnyed and brnyes as perfective stems and brnyed alone as the (?alternate) future stem of rnyed; recent dictionaries take rnyed and brnyes as stems of two different but largely synonymous verbs, neither of which inflects; see, for example, BOP, 993-94, 1016. The verb tables that form part of the concluding remarks of Gser tog pa V Blo bzang tshul khrims rgya mtsho's (1845-1915) study of indigenous Tibetan grammar (1891) omit rnyed, but do have an entry for the uninflecting brnyes; see Bod kyi brda['] sprod pa sum cu pa dang rtags kyi [jug pa'i mchan 'grel mdor bsdus te brjod pa ngo mtshar 'phrul gyi lde mig (Beijing: Mi rigs dpe skrun khang, 1957), 203. And the tables in Dmu dge Bsam gtan's fairly recent Brda sprod blo gsal dga' ston (Lanzhou: Kan su'u mi rigs dpe skrun khang, 1981), 131-32, have separate entries for both. I suspect, and this is a mere hunch on my part, that *brnyed is a hyper-correction some Tibetan writers felt necessary in order to accommodate rnyed's transitivity and its [im/]perfective ergative construction, and may either be part of the grammatical complex that led Si tu Pan chen Chos kyi 'byung gnas (1699-1774) unilaterally to change (without saying so) the reading of an important line of the Rtags kyi 'jug pa, or a reflex of this change; for a discussion, see the Karma si tu'i sum rtags 'grel chen [Sde dge xylograph], ed. 'Phrin las (Xining: Mtsho sngon mi rigs dpe skrun khang, 1982), 153ff., and the pioneering studies of T J. E Tillemans, "On bdag, gzhan and Related Notions of Tibetan Grammar," in Tibetan Studies, ed. H. Uebach and J. L. Panglung (Munchen: Kommission fur zentralasiatische Studien Bayerische Akademie der Wissen-schaften, 1988), 491-502; and Tillemans and Herforth (1989).

(23) As quoted in R. Apter, Digging for the Treasure: Translation after Pound (New York: Paragon Publishers, 1987), 7.

(24) See, respectively, "Tsong kha pa le Pandit," Melanges chinois et bouddhiques 3 (1934-35): 334; "Observations on Translation from the Classical Tibetan Language into European Languages," Indo-Iranian Journal 14 (1972): 181; and Life and Teachings of Tsong Kha pa (Dharamsala: Library of Tibetan Works and Archives, 1982), 44--and a slightly different but equally faulty rendition in Thurman's Tsong Khapa's Speech of Gold in the Essence of True Eloquence (Princeton: Princeton University Press, 1984), 73-74, which remained uncorrected (as much else) in the second edition of this book.

(25) Jackson (1994), 102-3, n. 35 and references.

(26) The passage is taken from the Rje thams cad mkhyen pa tsong kha pa che n po'i bka' 'bum thor bu ba, Collected Works [Lhasa Zhol print], vol. Kha (Dharamsala: Library of Tibetan Works and Archives, 1981), 305.

(27) See M. Hahn, Lehrbuch der klassischen tibetischen Schriftsprache, Indica et Tibetica, vol. 10 (Swisttal-Odendorf: Indica et Tibetica Verlag, 1994), 153-54, on the subordinating function of the coordinating particle shing; and also Skal bzang 'gyur med (1981), 173ff., and Dor zhi Gdong drug snyems blo (1987), 95-96 [see bibliography below].

(28) See the brief observations in Tournadre (1995), 264-65; I have not seen his large study, L'Ergativite en tibetain: Approche morphosyntaxique de la langue parlee, Bibliotheque de l'information grammaticale, vol. 33 (Louvain, 1996). See also BOP 993, 3163 and the Chinese translation of a reworked version of Skal bzang 'gyur med's book (above, n. 27) by the author himself: Shiyong zangwen wenfa (Chengdu: Sichuan minzu chubanshe, 1987), 509ff. Tibetan linguists by no means accept the classification of transitive (tha dad pa) and tha mi dad pa verbs across the board, however, and it is notably absent in the discussions of the Tibetan verb in, for example, the lengthy deliberations on its functions and morphology in Dot zhi Gdong drug snyems blo (1987), 103ff., as well as in Dpa' ris Sangs rgyas, Brda sprod gsal byed ngag sgron (Lanzhou: Kan su'u mi rigs dpe skrun khang, 1987), 204ff.

(29) Ed. A. C. Banerjee (Calcutta: University of Calcutta Press, 1939), 41; see J. E. Bosson, A Treasury of Aphoristic Jewels: The Subhasitaratnanidhi of Sa skya Pandita in Tibetan and Mongolian, Uralic and Altaic Series, vol. 92 (Bloomington: Indiana Univ. Press, 1969), 119, 287. For comments on this verse, see, for example, Sangs rgyas bstan 'dzin, Legs bshad 'dod dgu 'byung ba'i gter mdzod (Kalimpong, 1974), 338-39, and Dwangs grung, Blo gsar la'jug pa'i sa skya legs bshad kyi'grel gsar (Lhasa: Bod ljongs mi dmangs dpe skrun khang, 1992), 236. They do not remark on its grammar, however.

(30) Examples for this are found in Skal bzang 'gyur med (1981: 173-74) with an example from the Legs bshad rin po che'i gter, V: 168a-b: dregs pas yon tan nyams 'gyur zhing // 'dod pas ngo tsha nyams par byed //.

(31) See Pan. chen bla ma I Bio bzang chos kyi rgyal mtshan (1570-1662), Sgra pa shes rab rin chen pa'i rtsod lan lung rigs seng ge'i nga ro, Collected Works, vol. IV (New Delhi, 1973), 632, where, however, in his interpretation of this passage on p. 638, he curiously fails to take cognizance of this particular line; see also 'BRUG 146.

(32) TSONG, 582-609 [TSONG(1), 810-37; TSONG(2), 679-701]. For a study of Rgyal tshab, his writings on tshad ma and their relative chronology, see van der Kuijp (forthcoming).

(33) If I understand him correctly, then Sde srid Sangs rgyas rgya mtsho (1653-1705) is so far unique in stating, in his Bai durya g.ya sel, vol. I (Dehra Dun, 1976), 577, that Tsong kha pa had also been responsible for a series of notes (zin bris) to Prajnakaragupta's Pramanvarttikalamkara (rgyan) and a study of the Pramanaviniscaya. It is strange that he makes no mention of the Sde bdun la 'jug pa'i sgo don gnyer yid kyi mun sel in his otherwise fairly complete listing of the tshad ma writings of Tsong kha pa and his school.

(34) I plan to return to these and other issues in a separate study of Rong ston Shakya rgyal mtshan's (1367-1449) Tshad ma'i sgom rim[s] phul du byung ba'i snying po, in three folios, and the cognate Tshad ma rnam 'grel gyi lam rim ngo mtshar snang ba in three folios, as well. I briefly signaled the former in "On Some Early Tibetan Pramanavada Texts" (1994), 19. The latter is found in RONG 28b-31a.

(35) TSONG, 586 [TSONG(1), 814; TSONG(2), 682].

(36) Bstan bcos mngon par rtogs pa'i rgyan gyi mtha' dpyod shes rab kyi pha rol tu phyin pa'i don kun gsal ba'i rin chen sgron me, Collected Works, vol. 7 (New Delhi, 1973), 39-40: rje'i tshad ma'i lam sgrig las / sdug bsngal du rtogs pa'i lag rjes la 'khor 'bral 'dod skyes kyang mi rtag pa ma rtogs na phung po dang mi 'bral ba'i sred pa 'gog mi nus shing / de rtogs pa las de dang mi 'brai ba'i sred pa 'gog nus pa'i phyir ro // and rje'i tshad ma'i lam sgrig las / des na phung po mi rtag par rtogs pa'i lag rjes la de dang 'bral des dang sdug bsngal du rtogs pas 'bral 'dod skyes kyang thar pa don gnyer gyi blo mtshan nyid tshang ba la bdag reed rtogs dgos so //.

(37) Two generations of a Ke'u tshang shar (= eastern) line of a Central Tibetan incarnation series are noted in the Bod dang / bar khams / rgya sog bcas kyi bla sprul rnams kyi skye phreng deb gzhung, Bod kyi gal che'i lo rgyus yig cha bdams bsgrigs, ed. Ma grong Mi 'gyur rdo rje, Gang can rig mdzod, vol. 16 (Lhasa: Bod ljongs bod yig dpe rnying dpe skrun khang, 1991), 352. No mention is made of a Ke'u tshang nub series, presumably because it was not yet "on the books" when this list of incarnation series and VIPs was compiled (in ca. 1825?), or perhaps because it was too minor to be worthy of inclusion.

(38) This is *TSONG; I owe the reference to its publication to Mr. Sh. Nomura of Waseda University, Tokyo. *TSONG(1) is a handwritten dbu meed manuscript in thirty-four folios of the same, and its text bears no precise or obvious indication that it derives directly from *TSONG. The former is registered in the published catalogue of the holdings of the libraries of Ngag dbang brtson 'grus' monastery of Bkra shis 'khyil, the Bod kyi bstan bcos khag cig gi mtshan byang dri med shel dkar phreng be, ed. Grags pa (Xining: Mtsho sngon mi rigs dpe skrun khang, 1985), 280, as the Tshad ma'i lam bsgrigs rje rin po ches mdzad pa, where its author "Rje Rin po che" obviously suggests Tsong kha pa. However, it is absent from the listings of Tsong kha pa's oeuvre per se (pp. 170-77) and various Indo-Tibetan works on tshad ma (pp. 617-24). The Sde srid (see n. 33) does not list it either.

(39) See the Mi rigs dpe mdzod khang gi dpe tho las gsung 'bum skor gyi dkar chag shes bya'i gter mdzod, Stod cha (Chengdu: Si khron mi rigs dpe skrun khang, 1984), 48.

(40) The "Dga' bzhi pa" in question must be the statesman and minister (bka' blon) Bsod nams bstan 'dzin dpal 'byor (1761-after 1809), who writes in his autobiography, in RDO(lh) 1084 [RDO(ch), 2: 1284], that the lady in question was his lcam chung, "junior wife," as opposed to Rnam rgyal sgrol ma, his lcam proper. According to this work, so hugely important for late eighteenth century Central Tibetan history, they had one daughter, Tshe ring sgrol dkar (1782-?), who took her initial vows and whose name in religion was Rje btsun ma Blo bzang chos 'dzoms, and four sons. Two of these were incarnate lamas, namely, Blo bzang bshes gnyen grags pa rgya mtsho (1788-?), who was initially named Don 'grub tshe ring, and was affiliated with 'Bras spungs monastery and Ke'u tshang nub Blo bzang 'jam dbyangs smon lam (1792-?), who was affiliated with Se ra monastery. The other two were the laymen and highranking officials Dngos grub rdo rje, who is better known as Mi 'gyur bsod hams dpal 'byor (1784-?1834) and who was his father's "successor" as bka' blon, and Don 'grub tshe dbang (1796-?). For a discussion of the Dga' bzhi pa family as a whole, see the notes in L. Petech, Aristocracy and Government in Tibet 1728-1959, Serie Orientale Roma, vol. 45 (Rome: Istituto Italiano per il Medio ed Estremo Oriente, 1973), 50-64, where the occasional oversights and surmises are solely due to fact that the author had no access to Bstan 'dzin dpal 'byor's work.

(41) Ke'u tshang sprul sku Blo bzang 'jam dbyangs smon lam, sku gsung thugs rten dang mchod rdzas sogs gsar bzhengs kyi 'dod gsol smon tshig, Collected Works, vol. Kha (Dharamsala: Library of Tibetan Works and Archives, 1984), 754-55.

(42) RDO(lh), 190, 367 [RDO(ch), 228, 439].

(43) RDO(lh), 977 [RDO(ch), 1155].

(44) RDO(lh), 1061 [RDO(ch), 1257]. Contrary to the publisher's preface of his Collected Works, he was therefore not the older brother of Mi 'gyur bsod nams dpal 'byor and he was neither the brother of Lho brag Nyi lde Gar dbang bio bzang mchog grub. Rather, the latter was his grand-uncle on his father's side, who passed away in 1772; see RDO(lh), 152 [RDO(ch), 183].

(45) *TSONG, 809 [*TSONG(1), 33b-34a].

(46) 'BRUG, 284.

(47) What follows is taken from *TSONG, 768 [*TSONG(1), 1a]; the fourth topic begins in *TSONG, 772 [*TSONG(1), 5b].

(48) To begin the body of a text with de la is peculiar, if only because of the deictic opacity of the referent of the demonstrative pronoun de, and might in fact point to its original orality and/or editorial incompleteness. A similar state of affairs is met with in the famous Dag yig li shi'i gur khang lexicon of 1536, attributed to Skyogs ston Lo tsa ba Rin chen bkra shis (ca. 1495-after 1577), which begins (in all the available xylograph and other editions) with de yang, "moreover"; see, for example, the edition in Tibeto-Sanskrit Lexicographical Materials, ed. Sonam Angdu (Leh: Basgo Tongspon Publication, 1973), 1. To my knowledge, none of Tsong kha pa's authenticated works begins with de la (or de yang).

(49) For a discussion of the notion of the purpose of a text in Buddhist hermeneutics, see now T. Funayama, "Arcata, Santaraksita, Jinendrabuddhi, and Kamalasila on the Aim of a Treatise (prayojana)," Wiener Zeitschrift fur die Kunde Sudasiens 39 (1995): 181-201.

(50) TSONG, 582 [TSONG(1), 810; TSONG(2), 679]; the fourth topic begins in TSONG, 588 [TSONG(1), 816; TSONG(2), 683].

(51) See MKHAS, 6 and MKHAS(1), 647ff.

(52) Rgyal tshab's comments are studied in the annotated translation of his comments on this chapter in R. Jackson, Is Enlightenment Possible? Dharmakirti and rGyal tshab rje on Knowledge, Rebirth, No-Self and Liberation (Ithaca: Snow Lion Publications, 1993). Of further pertinence is his earlier "The Buddha as Pramanabhuta: Epithets and Arguments in the Buddhist 'Logical' Tradition," Journal of Indian Philosophy 16 (1988): 335-65, which was not used by Tillemans. T. Kimura published an annotated Japanese translation of the Pramanasiddhi chapter, Dharmakirti shukyo tetsugaku no kenkyu [A Study of Dharmakirti's Philosophy of Religion] (Tokyo: Mokujisha, 1987), which I have not seen.

(53) Franco, "Yet Another Look" (1994), and now also Franco, Dharmakirti on Compassion and Rebirth (1997), 15-43.

(54) See, respectively, 'U, 1: 259 and STON, 165 [STON(1), 65a]. The existence of the manuscript of Ston gzhon's text, that is STON(1), was first signaled in Sun Wenjing and Huang Xinming, "Zangwen yinmingshu mu," Yinming xintan, ed. Liu Peiyu et al. (Lanzhou: Gansu renmin chubanshe, 1989), 358.

(55) 'U, 1: 209--an earlier assertion of mine that it was absent from his work now stands corrected--and STON, 120 [STON(1), 46b].

(56) See the Dpon byang ba'i phyag tu phul ba'i chos kyi shan 'byed dbu phyogs, The 'Dzam thang Edition of the Collected Works of Kun mkhyen Dol po pa Shes rab rgyal mtshan, vol. 5, ed. M. Kapstein (New Delhi: Shedrup Books, 1992), 474-76.

(57) See L. Petech, Central Tibet and the Mongols: The Yuan--Sa-skya Period of Tibetan History, Serie Orientale Roma, vol. 65 (Rome: Istituto Italiano per il Medio ed Estremo Oriente, 1990), 132.

(58) For references, see, respectively, Chos rje kun mkhyen chen po'i rnam thar gsal sgron gyi rnam grangs dge legs chen po nor bu'i phreng ba, The 'Dzam thang Edition of the Collected Works of Kun mkhyen Dol po pa Shes rab rgyal mtshan, vol. 1, ed. M. Kapstein (New Delhi: Shedrup Books, 1992), 298-300; and Chos rje jo nang pa kun mkhyen chen po'i rnam thar, handwritten dbu med manuscript in fifty-seven folios, C. P. N. catalogue no. 002815(1), 5a, 6a-b.

(59) See, respectively, Collected Works [Lhasa Zhol print], vol. Ca (Dharamsala: Library of Tibetan Works and Archives, 1981), 513-53; and Collected Works [Lhasa Zhol print], vol. Ta (Dharamsala: Library of Tibetan Works and Archives, 1981), 339-70.

(60) The anthropology of the three kinds of human beings is also explicitly mentioned in the lengthy prologue in MKHAS(2), 23, so that it is probably not altogether legitimate to problematize its absence in his Tshad ma'i lam 'khrid.

(61) See CHOS, 121 [CHOS[supm], 83b]. In his own catalogue of translated Buddhist scripture, in CHOS(1), 20a, he classifies the tshad ma literature as comprising "treatise[s] of logical speculation of [that is, belonging to] the non-Buddhist [domains] of knowledge" (phyi rig pa'i rtog ge'i bstan bcos). On the other hand, his final words in CHOS, 137 [CHOS[supm], 94b] evoke a much more positive assessment of tshad ma, ending as it does with the statement: "Now, by understanding [tshad ma], the [correct] view is not lost; by having become involved [in it], a person is not deceived; by teaching [it], the disciple [becomes] confident: [these] three constitute the purpose of tshad ma. Therefore, the learned should hold a treatise on valid speculative thought in extreme reverence." (de'ang rtogs pas lta ba mi 'phrogs pa / zhugs pas skyes bu mi bslu ba / bstan pas gdul bya yid ches pa gsum ni tshad ma'i dgos pa yin no // de'i phyir mkhas pa rnarms kyis rtog ge yang dag pa'i bstan bcos ni shin tu gus pas blang bar bya'o //).

(62) We come across this real name (mtshan dngos) of his in the Bcom ldan rigs pa'i ral gri mtshan dngos chos kyi rgyal mtshan la bstod pa smra ba'i rgyan in RONG, 26a-27a.

(63) See the Bka' gdams pa gsar rnying gi chos 'byung yid kyi mdzes rgyan (Gangtok, 1977), 87.

(64) This text is found in his Collected Works, vol. 4 (New Delhi, 1983), 481-88.

(65) Tillemans (p. 13) alludes to a discussion of the epistemology of scripture as an agent able to invalidate (gnod byed) or obstruct (gegs byed) propositions in what he calls a forthcoming study of G. B. J. Dreyfus. While it was not included in the latter's recent Recognizing Reality: Dharmakirti's Philosophy and Its Tibetan Interpretations (Albany: State University of New York Press, 1997), it did form part of his earlier dissertation, however, for which see his unpublished "Ontology, Philosophy of Language and Epistemology in Buddhist Tradition," vol. 2 (Ph.D. diss., University of Virginia, 1991), 746-814.

(66) RGYAL(1), 163.

(67) BO, 259, 650ff.

(68) See here Tattvasatigraha of Santaraksita with the Commentary of Kamalasila, vol. II, ed. E. Krishnamacharya, Gaekwad's Oriental Series, no. 31 (Baroda: Central Library, 1926), 868 [SDE, vol. 49, no. 4272 (4267), 391.1-2; 'E 302a-b]. This connection was already hinted at by D. S. Ruegg in his La Theorie du Tathagatagarbha et du Gotra: Etudes sur la soteriologie et la gnoseologie du bouddhisme (Paris: Ecole francaise d'Extreme-Orient, 1969), 229, n. 2. See also RGYAL(1), 111-12, 115.

(69) PVSV, 108-9 [PVSV[supt], 473.5-474.1; Ce 322a-323a]. Most of the available Tibetan Pramanavarttika commentaries, including STON, 80-82 [STON(l), 30b-31a], concur that PV, I: 214-17 form one thematic cluster. An exception is Mkhas grub, who includes PV, I: 218a-b in the discussion, for which see MKHAS(l), 920-28. A handy overview of this matter is provided in Y. Fukuda and Y. Ishihama, A Comparative Table of Sa-bcad of the Pramanavarttika Found in Tibetan Commentaries on the Pramanavarttika, Studia Tibetica, no. 12 (Tokyo: The Tokyo Bunko, 1986), 13. An earlier, questionable appraisal of some of Dharmakirti's arguments may be found in J. I. Cabezon, "The Concept of Truth and Meaning in the Buddhist Scriptures," Journal of the International Association of Buddhist Studies 4 (1981): 8-9, 21, nn. 3-7, which was in part reproduced in his Buddhism and Language: A Study of Indo-Tibetan Scholasticism (Albany: State Univ. of New York Press, 1994), 99ff. Earlier studies of this passage of the Pramanavarttika (and Svavrtti autocommentary) are found in, for example, Yaita, "Dharmakirti on the Authority of Buddhist Scriptures" (1987) and in van Bijlert, Epistemology and Spiritual Authority (1989), 117-25, 143-44. And T. Tani has analyzed the corresponding passage in the Pramanaviniscaya; see "The Problem of Interpretation on [sic] Pramanaviniscaya III ad vv. 7-21, with the Text and a Translation [in Japanese]," Kochi Kogyo Koto Semmon Gakko Gakujutsu Kiyo, Kochi [The Bulletin of the Kochi National College of Technology] 25 (1986): 1-16 (this paper was not available to me). The latter is cited by Yaita (1987), 12, n. 11, and by E. Steinkellner and M. T. Much, Texte der erkenntnistheoretischen Schule des Buddhismus, Abhandlungen der Akademie der Wissenschaften in Gottingen, phil.-hist. KI., dritte folge, Nr. 214 (Gottingen: Vandenhoeck & Ruprecht, 1995), 34; Tillemans does not mention the contributions of Yaita and Tani.

(70) Yaita (1987), 2 has noted that Dharmakirti speaks at times of three valid means of cognition, with authenticated scripture as the third, in his autocommentary on the svarthanumana chapter of the Pramanavarttika, rather than the standard two found elsewhere in this work.

(71) For this and two other kinds of inference, see the lucid discussion in D. P. Jackson, The Entrance Gate for the Wise (Section III): Sa skya Pandita on Indian and Tibetan Traditions of Pramana and Philosophical Debate, vol. II, Wiener Studien zur Tibetologie und Buddhismuskunde, vol. 17.2 (Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat Wien, 1987), 428-30, n. 146.

(72) This passage is fully analyzed and discussed in T. J. F. Tillemans, "Pramanavarttika IV (4)," Wiener Zeitschrift fur die Kunde Sudasiens 37 (1993): 136-45.

(73) TSONG, 607-9 [TSONG(1), 835-37, TSONG(2), 699--701] and also Rgyal tshab's Pramanavarttika commentary in RGYAL, 201--2; see further, for example, the Bstan bcos tshad ma rnam nges kyitik chen dgongs pa rab gsal (stod cha), Collected Works [Lhasa Zhol print], vol. Ja (Dharamsala: Library of Tibetan Works and Archives, 1981), 222, which belongs to his middle period, and his Tshad ma rigs thigs kyi'grel pa legs bshad snying po, Collected Works [Lhasa Zhol print], vol. Nya (Dharamsala: Library of Tibetan Works and Archives, 1981), 552.

(74) See A Recent Rediscovery: Rgyal tshab's Rigs gter rnam bshad, ed. G. B. J. Dreyfus and Sh. Onoda, Biblia Tibetica, vol. 3 (Kyoto: Nagata Bunshodo, 1994), 99a. See also his Tshad ma kun las btus pa'i rnam bshad mthar'dzin gyi tsha gdung ba'joms byed rigs pa'i rgya mtsho, Collected Works [Lhasa Zhol print], vol. Nga (Dharamsala: Library of Tibetan Works and Archives, 1981), 407. In theory, the latter work dates from the years 1424 to 1432, but he must have composed it prior to the former as it is referred to on fol. 13b.

(75) MKHAS(1), 922-23, 'U, 1:138-40 and CHOS, 7 [CHOS[supM], 4a]. Ston gzhon quite expressly considers the four noble truths to be atyantaparoksa; see STON, 81 [STON(1), 31a].

(76) Scattered arguments are found in his Tshad ma rigs pa'i gter gyi dgongs rgyan lung dang rigs pa'i 'khor los lugs ngan pham byed [Rtog ge'i 'khrul 'joms chen mo], Collected Works, vol. 9 (Thimpbu, 1975), 239-40; Rgyas pa'i bstan bcos tshad ma rnam 'grel [gyi] rnam bshad kun bzang chos kyi rol mtsho, Collected Works, vol. 18 (Thimphu, 1975), 391-94; and Dka' 'grel rigs pa'i snang ba, Collected Works, vol. 19 (Thimphu, 1975), 223-24. These are further amplified and systematized in his Tshad ma'i mdo dang bstan bcos kyi shing rta'i srol rnams ji ltar 'byung ba'i tshul gtam du bya ba nyin byed snang ba, Collected Works, vol. 19 (Thimphu, 1975), 73-82. For other interesting remarks anent these questions, see also the Sa skya pa scholar Go rams pa Bsod nams seng ge's (1429-89) Tshad ma rigs pa'i gter gyi dka' ba'i gnas rnam par bshad pa sde bdun rab gsal, Sa skya pa'i bka' 'bum, vol. 12, ed. Bsod nams rgya mtsho (Tokyo: The Toyo Bunko, 1969), 130.3-133.4 [Ga 260-66b].

(77) This is evident from the remarks made in Se ra Rje btsun Chos kyi rgyal mtshan's (1469-1544 or 1546) incomplete sub-commentary or Rgyal tshab's exegesis of the Pramanavarttika's first chapter (it extends but to Rgyal tshab's explanation of PV, I: 169). He notes on several occasions that Rgyal tshab had changed his mind on a number of issues after the completion of this work; see SE, 42bff., 76aff., 78a, 79a-b.

(78) RGYAL, 200. My translation differs a little from the one in Tillemans, p. 11, n. 18.

(79) PVSV, 108-9 [PVSV[supt], 473.7, Ce 322b]: tathanagamapeksanumanavisayabhimatanam tathabhavah yatha caturnam aryasatyanam / ananumeyanam tathabhavo yathatmadinam / (de bzhin du lung la ltos med pa'i rjes su dpag pa'i yul du 'dod pa dag la de bzhin du 'gyur te / dper na 'phags pa'i bden pa bzhi dag lta bu'o // rjes su dpag par bya ba ma yin pa de dag ni de ltar 'gyur te / dper na bdag la sogs pa Ita bu'o //).

(80) PVT, 70.4 [Je 244a]; PVSVT, 392, PVV, 323.

(81) See the text under C. P. N. catalogue no. 004732, 57b. This manuscript in 240 folios is described in the appendix of van der Kuijp (forthcoming).

(82) SE, 66b.

(83) RGYAL(1), 187: rang gis Ikog gyur tu 'dod pa bdag med pa la sogs pa lkog gyur du bstan pa dang / gzhan dag gis lkog gyur du 'dod pa gtso bo la sogs pa gzhal bya lkog gyur yin pa bkag pa la dngos po stobs zhugs kyi rjes dpag gis gnod pa med pa dang/ ...

(84) MKHAS(1), 922-23. And the same is met with in a late fifteenth century Pramanavarttika study of Dga' ldan khri pa VIII Smon lam dpal (1414-91), for which see his Rnam 'grel le'u dang po'i tikka, xylograph (n.p., n.d.), 86b.

(85) The use of dag pa here is most likely owed to the passage of PVSV, 109: seyam sakyaparicchedasesavisayavisuddhir avisamvadah; PVSV[supt], 473.7 [Ce 322b]: yongs su gcod par nus pa'i yul ma lus pa rnam par dag pa de ni mi slu ba yin no //.

(86) RGYAL, 200--201.

(87) Pramanaviniscayatika, SDE, vol. 48, no. 4232 [4227], 450.7-451.1 [Tshe 46a-b]: mthong ba mngon sum dang rjes su dpag pa zhes bya ba'i yul lo // rjes su dpag dpag par bya ba yang med bzhin du res 'ga' mthong ba'am / mig bzhin du rten mthong ba'i phyir mthong bar brjod de / mig rjes su dpag pa la ni chos can mthong ba yin no // (this passage comments on Pramanaviniscaya; SDE, vol. 46, no. 4216 [4211], 438.1 [Ce 197a]). Not wholly surprisingly, the same is also found in RGYAL(l), 187!

(88) See, respectively, PVT, 70.3-4 [Je 243b-244a], PVSVT, 392 [Tillemans p. 11, n. 17 "PVSV-tika 363.30" should read " ... 393.30"], and PVV, 323, which glosses it by pratyaksa, "evident, visible." For the pratyaksa/paroksa contrast, see also PV, III: 63a.

(89) MKHAS(1), 922-23, Rgyas pa'i bstan bcos tshad ma rnam 'grel gyi7 rgya cher bshad pa rigs pa'i rgya mtsho las gzhan don le'u rnam bshad, Collected Works [Lhasa Zhol print], vol. Da (Dharamsala: Library of Tibetan Works and Archives, 1981), 757-58; 'U, 1: 139; and 'u, 2: 303. RGYAL, 772 is not explicit on this point. Earlier, however, in MKHAS, 344, Mkhas grub aligns only mthong ba with mngon gyur and ma mthong ba with both Ikog gyur and shin tu lkog gyur!

(90) STON, 425-26 [STON(1), 164b]: mthong ba ni tshad mas mthong ba ste / mngon gyur dang lkog gyur gyi don dag la mngon sum dang rjes dpag pa'i tshad mas gnod pa med cing grub pa dang / ma mthong ba ni tshad mas ma mthong ba ste/ ... The comments on PV, IV: 108a by Devendrabuddhi, Sakyabuddhi, Prajnakaragupta, and Manorathanandin do not expressly correlate drsta with (*)pratyaksa (mngon gyur) and paroksa; see, respectively, PVP, 569.7 [Che 291a]; PVT, 170.6f.; PVV 317; and PVBH[sups], 528 [PVBH[supt], 313.7, The 175b]. Significantly, Mkhas grub cites PV, IV: 108 in support of the dpyad pa gsum in MKHAS, 13.

(91) 'U, 1: 139; STON, 81 [STON(1), 31a].

(92) PVSV, 109 [PVSV[supt], 474.1; Ce 323a]. Note that the Tibetan translation (by Sa skya Pandita?) is based on a text different from the one in PVSV, for there is no way that aptavada can be rendered as nyes pa zad pa'i tshig, "wording that puts an end to evil!" Either the manuscript available to the translator read something like athava prakarantarena klesaprahana ... (yang na rnam pa gzhan gyis nyes pa zad pa ...), for which see PVSVT, 394 and PVT, 71.1 [Je 246a], whereby the additional -'i tshig ((*)-vada) owes its existence to this very translator, or, more likely, he rendered his text into Tibetan with the aid of either Sakyabuddhi or Karnakagomin. The expression nyes pa zad pa'i tshig is also met with in PVSV[supt], 473.7 [Ce 322b] ad PV, 1: 216, and its Sanskrit counterpart in PVSV, 108 is once again aptavada. However, the text of PVSVT, 394 has here dosaksayasya, roughly synonymous with klesaprahana, which has its parallel nyes pa zad pa'i tshig in PVT, 70.6 [Je 245a].

(93) See also Tillemans (1990), I: 29.

(94) For Dharmapala, see Tillemans (1990), I: 31, 91-92, 207, n. 28). For Candrakirti's commentary, see Tillemans (1990), I: 120; II: 17-19. The latter equates what is beyond the senses (atindriya, dbang po las 'das pa) with paroksa, and it is perhaps significant that he does not use the term atyantaparoksa at all in this context. In fact, the term atyantaparoksa does not occur in the fragmentary Sanskrit manuscript of Candrakirti's exegesis (about one third of his work is preserved in Sanskrit), and only once in his Prasannapada, for which see S. Yamaguchi, Index to the Prasannapada Madhyamakavrtti, pt. 1: Sanskrit-Tibetan (Kyoto: Heirakuji-Shoten, 1974), 7. This indicates, among other things, that it played no role in his ontology, but it of course is still possible that he would be inclined to equate the two and disinclined to allow for their ontological distinction.

(95) While Red mda' ba does exchange lkog gyur with shin tu lkog gyur in his study of the Catuhsataka (it is undated, but I think a case can be made for it having been written before 1400, so that it predates the documentation available on his students Tsong kha pa and Rgyal tshab), he makes no connection between this verse and Dharmakirti. Indeed, he goes so far as to cite Candrakirti's Prasannapada (ed. de La Vallee Poussin, Bibliotheca Buddhica, vol. 4 [St. Petersbourg, 1903-13], 75) in his query on the kind of cognition that allows one to understand what the Buddha said; that is, inference in general by means of which one can understand or cognize the sense of the world (lokarthadhigama, 'jig rten gyi don rtogs pa); see his Dbu ma bzhi brgya pa'i 'grel pa (Sarnath: Sakya Student's Union, 1974), 145-46. For a detailed study of an interpretation of Candrakirti's epistemology, see now Ch. Yoshimizu, Die Erkenntnislehre des Prasangika-Madhyamaka, Wiener Studien zur Tibetologie und Buddhismuskunde, vol. 37 (Wien: Arbeitskreis fur tibetische und buddhistische Studien Universitat Wien, 1996).

(96) PVSV, 166 [PVSV[supt], 481.4; Ce 359b].

(97) 'U, 1: 140; STON, 81 [STON(1), 31a].

(98) CHOS, 7 [CHOS[supm], 4a]: "[Dharmakirti] claimed that inasmuch as what is extremely invisible is the probandum of an argument [based on] scripture, [it] is included in what is invisible" (shin tu lkog gyur ni lung gi rtags kyi dpag bya yin pas lkog gyur tu 'dus par bzhed do //). Of course, this remark is not wholly unproblematic.

(99) What follows is largely taken from CHOS(2), 46b-48b. In CHOS(2), 71 a, a post-colophonic editorial note reads: 'di'i rtsom pa mthar phyin pa'i mjug sdud dang mdzad byang ma theb kyang / bcom ldan rig pa'i ral gris mdzad pa'o //, and I see no reason not to accept this ascription, especially in view of the text's subtitle of rgyan gyi me tog, which is a trademark of almost every work in his oeuvre.

(100) CHOS(2), 63b.

(101) A handwritten dbu reed manuscript of this study is found under C.P.N. catalogue no. 005141; another Tibetan commentary by way of a handwritten dbu med manuscript is found under C.P.N. catalogue no. 006668.

(102) CHOS(2), 40a.

(103) See the Dbu ma rtsa she rgyan gyi me tog, handwritten dbu med manuscript in seventy-seven folios, C.P.N. catalogue no. 007316(7), 7b; on the same page he also mentions his Dbu ma rgyan gyi me tog, evidently a study of Madhyamaka thought in general, which has not yet surfaced. Later, on fol. 11a, he refers the reader to his own Pramanasamuccaya commentary which was known so far from a single referrence by Mkhas grub, for which see J. I. Cabezon, tr. A Dose of Emptiness, 86.

(104) See the two treatises in CHOS, which includes his Pramanaviniscaya study, and my review of the volume in this journal (JAOS 114 [1994]: 304-6).

(105) CHOS(2), 46b: kwa yi sgra rnams rtag tu phyogs lhung bas // sdang zhing rang yid phyogs hur mun pa ni // de nyid rtog pa sgrub min mdzad pa'i tshul // dbu mar gnas pa yin pas der blo sgrub //. These lines read substantially differently in PVBH[supt], 344.3

[The 281a]: kye yi rtsod rnams rtag tu phyogs lhung bas // sdang yid rang bzhin phyogs byas nus pa ni // de nyid rtogs pa sgrub min yid kyi tshul // dbu mar gnas pas yid phyir de blo bsgrubs//. The corresponding Sanskrit text in PVBH(s), 648, verse 617, has: he vadino na khalu samtatapaksapatadvesam manah svaparapaksakrtandhakaram / tattvaprabodhanavidhayi manasvivrttam madhyasthabhava iti tatra matir vidheya //.

(106) Schwabland, "A General Exposition of Valid Cognition"

(1994), 127, 205, anent CHOS, 5 [CHOS(m), 2b].

(107) PVBH(t), 176.6--7 [Te 3b-4a] and PVBH(s), 5, verse 9.

(108) Another one of some interest to Tibetan philology proper is an aside in Rma bya Byang chub brtson 'grus' (?-?1185) exegesis of Candrakirti's Prasannapada, where we read that the sum in mngon sum (for Sanskrit pratyaksa), "perception," is actually of Zhang zhung origin and thus not at all a "real" Tibetan word; see his Dbu ma tshig gsal zhes bya ba'i ti ka (sic), handwritten dbu med manuscript in ninety-nine folios, C.P.N. catalogue no. 005715(7), 35a. In his view sum has the sense of Tibetan 'dzin, "hold, grasp, seize," so that mngon sum would mean, etymologically, "grasping what is evident, visible." Except for its meaning of "three" in Tibetan, sum [is less than] gsum, a Zhang zhung word sum is not glossed in the earlier studies of this language by E. Haarh, R. A. Stein, or S. Hummel.

(109) CHOS, 20 [CHOS(m)], 12a].

(110) PVBH(t), 256.4 [Te 282b] and PVBH(s), 305, verse 443.

(111) CHOS, 125 [CHOS(m)], 86a].

(112) PVBH(t), 299.1 [The 123b] and PVBH(s), 477, verse 2.

(113) On this work and its relation to that of Rngog Lo tsa ba, see M. Mejor, "On the Date of the Tibetan Translations of the Pramanasamuccaya and Pramanavarttika," in Studies in the Buddhist Epistemological Tradition, ed. E. Steinkellner (Wien: Verlag der Osterreichischen Akademie der Wissenschaften, 1991), 181-85.

(114) CHOS(1), 30a-b.

(115) His life has been recently studied in C. R. Stearns, "The Life and Tibetan Legacy of the Indian Mahapandita Vibhuticandra," Journal of the International Association of Buddhist Studies 19 (1996): 127-71.

(116) Dge 'dun chos 'phel relates information on both in DGE, 31-32, but does not specify that the latter was a palm-leaf manuscript (ta la'i lo ma!), which he does for a good number of other manuscripts they were able to discover.

(117) DGE, 34.

(118) Schwabland (1994), 11-17, 37.

(119) See the slightly damaged, handwritten dbu can manuscript in twenty-two folios, C.P.N. catalogue no. 004971(5), with numerous interlineary notes. The upper Center of the title page gives what appear to be indigenous catalogue "numbers": phyi la 523 in red ink; on the right of this "number" we read ne in black ink. The title page reads (warts and all): / skyon spangs yon tan kun rdzogs pa'i // smra mchog 'gon po la btud nas // byams pa'i rgyud 'dzin rgyal ba'i sras // 'jam dbyangs sprul pa bla ma yis // bdag gzhan 'ga' zhig don ched du // dgongs pa rgyan gyi me tog brtsams // sde snod rin chen rgya mtsho las // sa bon mkhas slabs gzhung las btus // thos bsam rig pa'i 'dum rar btab // dam chos pad mo'i skyed de // shes rab nyi gzer phog nas don // kha 'byed legs snang ngo mtshar che //. Its incipit has: [lb] sa bcu'i dbang phyug rgyal sras kun gyi rtso // sum bcu tsa gnyis gsung dbyangs byams thugs ldan // skye gcig chos kyi rgyal por gshegs pa'i // rgyal ba byams pa'i mgon la spyi bos 'dud //. And the colophon reads: [22a] ... lung dgongs pa rgyan gyi me tog zhes bya ba bla ma'i gsung sna tshogs kyi don phyogs gcig tu sbyar ba la rang dga'i skyon myed do // legs so legs so // //. Its marginal notation is ka, so that it could have been the first volume of an edition of Chos kyi rgyal mtshan's collected oeuvre. On the other hand, given the colophon's ambiguous characterization of the text as a collection of "the teacher's various pronouncements" (bla ma'i gsung sna tshogs), it is also possible that the text actually constitutes a collage of opinions of one of Chos kyi rgyal mtshan's own teachers, if he had written it, or a miscellany of views entertained by him but written down by one of his students. If the latter were the case, then it could have figured as the first volume of the writings of that student.

(120) See the handwritten dbu reed manuscript in twenty-two folios, C.P.N. catalogue no. 002357(28), 15a-b. Full bibliographic details of this manuscript as well as another one of the same text are given in my forthcoming "Chos kyi rgyal mtshan (ca. 1240-ca. 1310) on the Periodization of Written Classical Tibetan."

(121) For preliminary details, 'see my "Sa skya Pandita on the Typology of Literary Genres," Studien zur Indologie und Iranistik 11-12 (1986): 41-52.

(122) In the manuscript, for which see note 125, his name in religion is prefixed by the phrase 7 karma ta yi si dhu, where 7 is a graph used as an indication of respect and is sometimes called mgo chen, and ta yi si dhu reflects Chinese tasitu. While the mgo chen's origins are elusive, we do encounter this graph already in some Tibetan edicts issued by the Mongol imperial court of Yuan China; see, for example, the imperial edict issued by Imperial Preceptor dishi Rin chen rgyal mtshan (1257-1305) in the Bod kyi lo rgyus yig tshags gces btus / A Collection of Historical Archives of Tibet, eds. Sgrol dkar et al. (Beijing: Cultural Relics Publishing House, 1995), no. 8, 11.7-8.

(123) See his Lung du ston pa ka la pa'i mdo las mtshams sbyor gyi 'grel pa legs bshad gsar du sgrogs pa'i ngo mtshar, handwritten dbu med manuscript in ninety-two folios, C.P.N. catalogue no. 002351(3), 71a. The text is part of a bundle entitled Po sti tha pa'i dkar chag, that is, "Index to Book Tha," whereby the upper center of the title page bears the vertically arranged indigenous catalogue number of phyi za 31. Part of the colophon states on fol. 92b: zhes pa sa skyong pa yi dbang phyug gtsang chos kyi rgyal po chen po'i rgyal khab spra nam gyi klu sdings su 'dug pa'i tshe zin bris bgyis pa las / slad kyis legs par dpyad pa'i cha 'ga' zhig dang mtshams sbyor drug pa'i 'grel pa dag rgyal mkhar rtse'i skyid mo tshal du dge bar grub po // slar lha mdongs lo tsa ba se na gu ru karma sha sa na dha ra zhes pas dpyad pa'i yi ge byung bar brten mi bcos su mi rung ba rnams bcos shing / 'di khungs su drangs pa'i rgya bod kyi 'grel pa dang yig cha rnams kyang rnyed pas zhus dag dang legs par gtan la phab pa'o //.

(124) van der Kuijp (1994), 2-3, 24, n. 4.

(125) He must be distinguished from another Tibetan Sanskritist also known as Lha mdong[s] Lo tsa ba, but whose name in religion was "Bshes gnyen rnam rgyal." He flourished one generation earlier in the sixteenth century. I suspect that their relationship was one of teacher and disciple. See Si tu Pan chen Chos kyi 'byung gnas and 'Be Lo tsa ba Tshe dbang kun khyab, Sgrub brgyud karma kam tshang brgyud pa rin po che'i rnam par thar pa rab 'byams nor bu zla ba chu shel gyi phreng ba [History of the Karma Bka' brgyud pa Sect], vol. 2 (New Delhi, 1972), 103, 137. Sometimes, "Lha mndongs" is spelled "Lha mdong" and even "Lha mthong."

(126) The translation is from the Tibetan; for the Sanskrit and its translation, see Tillemans, p. 15. For a somewhat different interpretation, see Yaita (1987), 9 and van Bijlert (1989), 143ff.

(127) For these, see Tillemans, pp. 44, 72-73, n. 23.

(128) An edition of his collected works in six volumes was published in Xining at the Mtsho sngon mi rigs dpe skrun khang (Qinghai minzu chubanshe) in 1997. For his autobiography and final days see the Rje dmu dge bstam gtan rgya mtsho'i gsung 'bum pod dang po, ed. Lcags byams (Xining: Mtsho sngon mi rigs dpe skrun khang, 1997), 525-694, 694-713.

(129) I am thinking in particular of his Rten 'brel la bstod pa, "Ode to Interdependent Arising," for which an adequate translation is still needed, and of course his autobiographical summain-verse, the Rang gi rtogs brjod rin po che 'dun legs ma.

(130) Mkhas grub trams cad mkhyen pa dge legs dpal bzang po'i gsan yig, Collected Works, vol. Kha (Dharamsala: Library of Tibetan Works and Archives, 1981), 41.

(131) BO, 742-76 deals quite explicitly with tshad ma as a soteriological enterprise. We only have a fragment of Bo dong Pan chen's tshad ma writings. For example, in BO, 703, he refers to his own extensive explanation of what constitutes valid and non-valid scripture, which he had apparently detailed "elsewhere" (gzhan du).

(132) STON, 388 [STON(1), 150b].
COPYRIGHT 1999 American Oriental Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:The Journal of the American Oriental Society
Geographic Code:90ASI
Date:Oct 1, 1999
Next Article:A Note on the Authenticity and Ideology of Shih-chi 24, "The Book on Music".

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters