Further Notes on Bhāviveka’s Principal Oeuvre

Chinese Academy of Social Sciences, Beijing

Leonard W.J. van der Kuijp

Harvard University

Abstract

The oeuvre with which Bhaviveka or Bhavya is associated is beset with a number of problems, ranging from exactly what works can be reasonably attributed to him, to their titles, and the reach and nature of their inter-textualities. Building on earlier contributions to these issues,we seekin this essayto gain a furtherand,itis hoped, a deeper understanding ofthese questions and tentatively propose a relative chronology ofhis main writings.

Keywords

Bhaviveka - Bhavya - Prajnapradlpa - Madhyamakahrdaya - Tarkajvala - Zhangzhen [[[lun]]] - - *Hastaratna - Tattvâmrtâvatâra - Madhyamakaratnapradîpa

In order to eliminate all conceptual thinking,

you taught the nectar-like emptiness, ... .

sarvasamkalpanasaya sünyatamrtadesana /....

nÂgÂRjuna, Lokatltastava, 23a-b

Preliminaries

The question of which writings, celebrated for their philosophical and often polemical content, should be attributed to the Middle Way-Madhyamaka intel-lectual Bhaviveka (ca. 490/500 - ca. 570)—or Bhavyai—and therefore should play a role in this scholar's intellectual biography, has been a long-standing problem. This also has, needless to say, a direct bearing on the way his oeuvre should be located in the history of Indic Madhyamaka philosophy and, more broadly, in the history of Indian philosophy tout court. It is quite true that a fair number ofin-depth discussions ofissues relating to these topics do exist, and one builds on the results obtained by one's precursors; the present essay is certainly no exception. We have especially relied on the earlier contribu¬tions of L. de La Vallée Poussin, V.V. Gokhale, Ejima Yasunori, Chr. Lindtner, and Saito Akira, as well as on H. Krasser's most recent publications, which have placed several ofthe extant texts ofBhaviveka's oeuvre in an intriguingly new light.2

The principal works that we discuss in this essay are: the Dasheng zhangzhen lun AA'AAm [ Mahâyânahastaratnasâstra] [hereafter hr], the Prajnapra- dlpa [hereafter pp] commentary on Nagarjuna's (2nd c.) Mülamadhyamaka- karika, what have been called the Madhyamakahrdaya[[[karika]]] [hereafter mh] and the Tarkajvala [hereafter tj], and the

• Madhyamakaratnapradlpa [here¬after mrp]. In so doing, we intend to re-examine some ofthe data on which basis certain conclusions were reached and to add some additional pieces of the puzzle that have so far not played an adequate role in the various proposals. Thus, in this essay, we have gathered in one place much that is by now fairlywell known and uncontroversial, but also added some data that, to our knowledge, have not yet played a role in these various discussions. Before proceeding, we may as well show our hand in the expectation that it is non-controversial: the presupposition with which we are writing this essay is that the sixth century Bhaviveka was undoubtedly the author of hr and pp. This is the main crite¬rion on which we will base several of our arguments. Save but one of these major works, namely, the two Sanskrit recensions of mh, none of the other four appear to be extant in Sanskrit.3

hr and pp are available in Chinese translations and, in fact, form the oldest witnesses of these two treatises and, indeed, of Bhaviveka's oeuvre as a whole. Xuanzang (600/2-64) translated hr, which, unlike pp, was never rendered into Tibetan. There is a French version of this work by La Vallée Poussin and a quasi-Japanese one, a so-called kundoku “translation,” by Hatani Ryo- tai Prabhakaramitra (574/5-633)5 was responsible for the Chinese

translation ofpp, which he completed in 630-632. In their translations, Xuan- zang rendered the author's name by Qingbian and Prabhakaramitra had Fenbieming ^^W. Both unambiguously reflect “Bhaviveka.” In addition, Xuanzang has given his name as Popifeijia in his Datangxiyuji MBE, Record of the Western Regions? of 646 and this, too, can only reflect *Bha- viveka. Thus his name in the Chinese sources can be unambiguously recon-structed as Bhaviveka. We will begin our discussion with an examination of the Chinese dossier and this will naturally lead us to the other works, and the Tibetan and Sanskrit dossiers, where things are not as clearcut.

As for the Chinese title of hr, zhang in zhangzhen 'Y’BE can render Sanskrit hasta or karatala, so that we should bear in mind that this work may have been called *[Mahdydna]Karatalaratna, as was indeed suggested by La Vallée Poussin, or even *[Mahdydna]Talaratna, as we find in B. Nanjio's dated but still useful catalog.7 Given that his precursor Dignaga (ca. 480 - ca. 540) had written a work with the possible title *Hastavalaprakarana, Balled Hand Treatise,8 it is likely that, especially because, to judge from his argumentation, Bhaviveka was intimately familiar with Dignaga's oeuvre, its titling may not be unrelated and thatitwas in fact called *Hastaratna, Jewelin the Hand, andnot *Karatalaratna. But whereas Dignaga espouses a Yogacara-mentalist position in the *Hastavalaprakarana, hr critiques this position in no uncertain terms, as do pp, and mh and tj. Thus, to continue the metaphor: whereas Dignaga balls his hand, that is, makes a fist, contra realist philosophical positions, Bhaviveka opens the clenched hand to discover the jewel that is the quasi-phenomenalist Madhyamaka point ofview!

In his Tibetan translation of the large and very rewarding Samdhinirmo- canasütra commentary of 664 by the Korean monk Wónch'úk [[[Wikipedia:Chinese|Chinese]] Yuance] UOÍ (613-696), Wu Facheng (9th c.), alias 'Gos Lo tsa ba Chos

grub, rendered the Chinese title of hr, Dasheng zhangzhen lun, and the Chi¬nese name ofthe author, “Qingbian,” by, respectively, [[[Bstan bcos]]] lagnarin po che and “Bha bya” ((Skt. Bhavya).9 To be sure, “Qingbian” actually ought to elicit “Bhaviveka,” so that its translation by “Bha bya” indicates a conscious decision and an explicit choice oftranslation on Wu Facheng's part. His motivation for this can only be guessed at, but it is not entirely surprising since, as we shall see below, it was one of his names that had some currency in contemporary Tibetan intellectual circles. Bhavya/Qingbian is also mentioned in connection withmh,albeitonlyonce, inthe extanttextofWónch'úk'scommentarywhere we read the following:10

... therefore it is said in the *Tattvamrtavatara Chapter of the *Madhyama- kahrdayasastra that was written by Bodhisattva Qingbian: There is no store-house perceptual awareness (*alayavi/nana) apart from the six types of perceptual awareness (*sadvijnana), because it is not included (she by the six types of perceptual awareness such as that ofthe eye, etc., just like a'flowerin sky.'

Therefore, we know that his tradition only establishes the six types of perceptual awareness. And the ninth century Tibetan translation of this passage reads somewhat differently:11

... de’iphyir slob dpon bha byas mdzadpai bstan bcos dbu ma’i snyingpoi nangnas de kho na nyid kyi bdud rtsi la 'jug pa'i le'u las / rnam par shes pa drug las gud na gzhan kun gzhi rnam par shes pa medde /migla sogs pa'irnam parshespadruggismabsus [?bsdus] pa'iphyir/nam mkha'i me tog dang mtshungs so zhes bshad pas / de'i phyir de dag gi[s] gzhung gis rnam par shes pa drug khona rnam pargzhag par rig par bya'o //

. thereforeitissaidinthe*TattvamrtavataraChapterofthe*Madhyama- kahrdayasastra that was written by Bodhisattva Qingbian [[[Bhaviveka]]]: There is no store-house perceptual awareness (kun gzhi rnam par shespa, *alayavijnana) apart from the six types ofperceptual aware-ness (*sadvijnana), because it is not encountered by [?bsdus: in¬cluded in] the sixtypes ofperceptual awareness such as that ofthe eye, etc., just like a 'flower in sky.' Therefore, one should know that their (de dag gi!) tradition gives an exposition ofonly six types ofperceptualawareness.

The passage surely deserves several comments. In the first place, aside from the fact that it is the earliest known passage where Bhaviveka is said to bemh's author, none of the other, early sources at our disposal have anything to say about a specific chapter of mh that is titled *Tattvômrtôvatâm. Wônch'ùk is really the only one to do so; Chinese pin S (Tib. le’u) leaves no doubt that he wanted to indicate a “chapter” rather than an independent tract. Further, his quotation (yun A) suggests that this chapter was if not entirely, then at leastinpart, writteninprose. And this would mean that mh was at least partly written in prose, a notion that stands in flat contradiction to the little we know aboutthe Indo-Tibetan Tradierung ofthis text, thatis, mh[s] and mh[t], both of which are extant in verse-texts. Secondly, hr contains two self-reflexive mentions of a work [or a chapter of a work] by *Bhaviveka himself with the title *Tattv0mrt0vat0.m.i2 The first occurs towards the end of his fairly detailed exposition of the Yogacara notion of “the dependent nature” (yitaqixing ®te, paratantrasvabhava). There he writes that he will not dwell on this issue any further, stating that:

This issue has already been clarified at length, as in the Ru zhen ganlu A BAA [* Tattvamrtavatara]. The second takes place in the contextofthe import ofthe notion ofhaecceity (zhenru tathata) in this system, where the quoted texts states:

Let us halt this long controversy for the time being, because it is difficult forpeoplewho despise the meaning oflong articles to believe andaccept them. This subject has already been dealt with in the Ruzhenganlu AB AA [*Tattvamrtavatara].

Though he did not develop this idea any further in print, La Vallée Poussin conjecturedinanotethatthe *Tattvamrtavatara was nothing other than tj, an idea that was evidently based on a much earlier remark made in de Benfey's French translation of V.P. Vasil'iev's Russian translation of a portion of the Tibetan scholar 'Jam dbyangs bzhad pa'i rdo rje I Ngag dbang brtson 'grus' (1648-1722) famous Grub mtha’ chen mo. This presents us with yet another problem,andwewilldiscussitbelow.Butletusfirstattendtosomeoftheissues that bear on mh and tj. Having done that, we will focus our attention on the alleged * Tattvdmrtavatdra chapter, which will then lead us to an examination of the extent to which Bhaviveka deals with the store-house perceptual awareness in a manner that replicates, or perhaps but echoes, Wonch'uk's citation from this alleged chapter and the references to it in hr.

The Madhyamakahrdaya and the Tarkajvala

The authorship as well as the transmission ofmh and tj in toto has been among the long-standing problems in the history ofMadhyamaka studies. Even their titles are somewhat problematic, and Lindtner and Saito dealt with these in some detail.14 Furthermore, Wonch'uk appears to be the very first Buddhist intellectual to mention the title *Madhyamakahrdaya[sdstra], and his contem¬poraries Huizhao (648-714)—a disciple of [Kui Ji (632-682)—and

Taixian (late 7th - early 8th c.) essentially repeated, albeit in controversy, what he had written in their respective exegeses of the Cheng weishi lun ^«.15 We do not need to repeat what they wrote, but we may mention for the record that Huaiyuan (11th c.) mentions both mh and hr in his 1061 study of the Suramgamasutra36 The Zhangzhen lun [hr] is what Bodhisattva Bhaviveka wrote one thou¬sand one hundred years after the Buddha's nirvana; in South India, he wrote the *Madhyamakahrdayasdstra and the Zhangzhen lun, etc.

The Sanskrit text of mh is available in two different and differently filliated manuscripts, concerning which Ye Shaoyong included a notice in his valuable survey ofSanskrit manuscripts ofMadhyamaka treatises ofthe Tibet Autonomous Region.17 One is an almost complete palmleaf manuscript in twenty-four leaves that was housed in Zhwa lu monastery; the other is a com-plete, but slightly damaged twelve-page paper manuscript that was evidently taken from Rwa sgreng monastery and ended up inthe Potala.The latteris not [yet] available and only the former, that is, mh[s], was published in a facsim¬ile edition.18 A first attempt at a translation of mh and tj into Tibetan was made around the year 800.

The early ninth century Lhan dkar ma and the Phang thang ma catalogs ofTibetan translations of Buddhist scripture, and a few things besides, register incomplete Tibetan versions ofwhat the first calls “the two, Dbu ma i snying po rtog ge barba [*Madhyamakahrdayatarkajvala], the basictext(rtsaba), together with a/the commentary (dang grelparbcaspa gnyis).” The second has two separate entries, namely, Dbu ma i snying po rtog pa la barba [*Madhyamakahrdayakalpanajvala] and Dbu ma i snying po rtog pala barba i

grelpa[*Madhyamakahrdayakalpanajvalavrtti], where rtogpa la mayofcourse verywellbe a mistake for rtogge la.19 Thus, itisnotunimportant to note that Tarkajvala or *Kalpanajvala is not said to be the title ofthe/a com¬mentary on the Madhyamakahrdaya. Rather, theyare acknowledged as integral parts ofthe title ofwhat appears to be the verse-text. Neither catalog lists the name ofthe translator[s]. The same holds forthe translation ofppofcirca800, which both register as well. But the attribution of this rendition to the team of the Madhyamika (dbu ma pa) Jnanagarbha and Cog ro Klu'i rgyal mtshan appears to have been entirely uncontroversial.

The only Tibetan translations ofmh and tj, that is, mh[t] and tj[t], that we currently do have in toto apparently came from the pens of Atisa (982-1054) and his disciple Nag tsho Lo tsa ba Tshul khrims rgyal ba (1011-ca. 1070). Ascribed to Bha bya [= Bhavya] intheir colophons, they were translated atthe behest of Rngog Legs pa'i shes rab in the Gtsug lag khang of Ra sa [= Lhasa], meaning that both date to the late 1040s or early 1050s. This is what we are told in the colophons of all available versions of the Tanjur in which these two texts are included.20 mh[t] has been variously shown not to have been

17 See Ye (2009: 316-318).

18 Jiang (1991: 111-118).

19 See, respectively, Lalou (1953: 337, no. 732) and Dkar chag phang thang ma (2003: 49).

20 The translators' colophon of tj[t] includes a verse that begins: dbu ma i snying po[ i] rtog ge bar ba di // mkhas btsun shes rab od gsal gnyis kyis bsgyur //. The two men who translated this work are characterized as scholars (mkhas), monks (btsun), and having clear insight (shes rab od gsal) into matters. The two are presumably Atisa and Nag tsho Lo tsa ba.

prepared on the basis of mh[s], which is the published palmleaf manuscript of the Sanskrit text. Rather, the text of m h that was embedded in the original Sanskrit manuscript of tj was evidently filliated differently from m h[s]. tj was composed in a mixed style of verse and prose in which the author reproduced all the verses of his version of mh. At some time, maybe in the course of the translation, the verses were extracted from tj[t] and the mh[t] was born. This means that the pronoun “they” may be misleading, and that only tj was translated, which was then almost artifically split into two “different” works, m h[t] and tj[t ]. Thus, to quote verses from tj[t ] and attribute them to this text rather than to m h[t] is not an unusual practice for Tibetan scholars, even if it is sometimes done inconsistently. Examples of this would be 'Gos Lo tsa ba Gzhon nu dpal (1392-1481) and Gser mdog Pan chen Shakya mchog ldan (1428-1507), and we are sure many more instances are readily available when searched for.21

But there is a slight problem with Atisa's and Nag tsho Lo tsa ba's purport¬edly joint translation ofmh/tj. Namely, the much later biographical literature onAtisaas well as the earliest catalog oftranslated scripture ofthe later period that was compiled by Dar ma rgyal mtshan (1237-1305), alias Bcom ldan [rig{s} pa'i]ralgri, appearto agree that he and Nag tsho Lo tsa ba only translated tj.22 The available manuscripts ofhis undated catalog ofpossibly the 1270s ascribe the authorship of tj and mrp, which follows it, to 'Ba' bya [= Bhavya]. The manuscripts ofhis brief1264 history of Buddhism in the Indian subcontinent and Tibet only register Bhavya's ppand tj.23 Be this as itmay, his disciple Dbus pa Blo gsal Rtsod pa'i seng ge (ca. 1255-?) writes that Nag tsho Lo tsa ba had translated inter alia mh, tj, and mrp, and that Legs ldan 'byed [*Bhaviveka]24 was theirauthor—note here the difference in the name ofthe author as com¬pared to what we find in the canonical colophons, namely, Bhavya. As a rule, Dbus pa Blo gsal does not mention the Indian or Kashmirian, etc. counterpart

21 For the first, see for example 'Gos Lo tsa ba (2003: 6-7), where he first quotes “from the Rtog ge 'bar ba,” that is, mh[t], III, 26a-26c, and then what is stated “in the commentary of what was cited” (... gsungs pa’i grelpar); see here, Heitmann (2004:120-137). For the second, see Gser mdog Pan chen (1975: 500), where he cites a verse “from the Rtog ge 'bar ba,” which in fact is mh[t], V: 20.

22 Schaeffer and van derKuijp (2009: 212). And he a dds thereto only mrp in the next entry.

23 Dar ma rgyal mtshan (2007: 79) has ’ba bca[s], which seems to be an obvious misreading ofthemanuscriptonwhichthistextisbased; seealso Ibid. (manuscript, 15b), which has 'Bab phya.

24 Dbus pa Blo gsal (manuscript, 43b). This very same attribution also holds for pp that is registered on the same folio. of the Tibetan translator, so that we can assume that he was well aware that Atisa was Nag tsho Lo tsa ba's “informant.” In sublinear glosses that are owed to his Sanskritist colleague Lo tsa ba Bsod nams 'od zer, we find the rather funny and knee-jerk “translation” of* bhedabhagavcin. for legs Idan ’byed'. Dbus pa Blo gsal was not well served with this very strange equivalent and we can only won- derwhyitwas leftinplace inthe so far unique manuscript ofthis catalog that we have at our disposal.25 Dar ma rgyal mtshan knew the author of tj[t] as 'Ba' bya or Bha bya [= Bhavya] as we witness, for example, in his catalog of translated scripture (see below), as well as in his treatise on non-Buddhist, quasi-Buddhist and Buddhist philosophical systems (grub mtha', siddhdnta) that has recently come to light. For example, inthe twentieth chapter ofthe latter, in which he discusses Mad- hyamaka thought, he quotes the Rtog ge 'bar ba, tj[t]—it is actually mh[t], V: 28cd—and Dbus pa Blo gsal does the same in his cognate work.26 In con¬nection with the latter, both cite the Tibetan text ofmh[t], V: 28c as: mdo las sems tsam gsungs pa ni //. All canonical versions of mh[t] support this read¬ing. In his comment, Mimaki Katsumi long ago drew attention to the

fact that mdo las of mh[t], V: 28c, could not be a translation of the Sanskrit text of mh[s], V: 28c, because, according to him, it read: sdstr’eva cittamdtro- ktih, which in fact was also Gokhale's reading. At first glance, this particular reading is also reflected in Tsong kha pa's justly celebrated hermeneutics of Mahayana Buddhism, the Drang ba dang nges pa'i don rnam par phye ba'i bstan bcos legs bshad snying po of 1408, which cites this verse as bstan bcos sems tsam zhes gsungs pa //, presupposing the reading of *sdstra instead of *sutra!28 Quoting Tsong kha pa, Mimaki argued that this proved “the existence ... of a Tibetan translation other than the one of the Tibetan canon.” But this may have been a little too hasty, for the passage in question occurs in Tsong kha pa's quotation of Santaraksita's (8th c.) Madhyamakdlamkdra[[[sva]]]vrtti where, however, reading

bstan bcos las sems tsam mo zhes gsungs pa ni ., the slight prose-paraphrase of this half-verse is not attributed to anyone or anyone's work. Indeed, only his student Kamalasila (8th c.) quite explicitly stated in his Madhyamakdlamkdrapanjikd that it was taken from the Madhyamakahrdaya by Skal ldan [= Bhavya].29 On the other hand, the readings ofthe Sanskrit text of this hemistich that are variously offered (by Lindtner: sutresu cittamdtro- ktih, and P. Hoornaert and Saito: sutre ca cittamdtroktih) are not exactly con¬firmed by the only extant Sanskrit manuscript which, at this juncture, is admit¬tedly rather hard to read.30 Nonetheless, it now turns out that Mimaki was right, albeit for the wrong

reason. About thirteen years after the publication of Mimaki's book, Lindtner cited some four instances in which he provided unas¬sailable evidence for the existence of a “para-canonical version” of a Tibetan translation of mh on which, he argued, mh[t] “was at least in part based ...”31 Truth be told, that later Tibetan scholars were not averse to tinkering with translated texts in the course oftheir transmission in Tibet, either as intellec¬tual exercises or with a more noble goal in mind, is made abundantly clear in the voluminous biographical literature. And a or the translation of tj was no exception. According to his biography by Zhwa dmar IV Chos grags ye shes (1453-1524), one of his disciples, the great 'Gos Lo tsa ba himself is reputed to have redacted an as yet unidentified Tibetan version of tj, albeit without recourse to a Sanskrit manuscript, in around 1470.32

At one point in the twentieth chapter of his tract on the philosophical systems, Dar ma rgyal mtshan mentions such Svatantrika philosophers as Legs ldan 'byed [*Bhaviveka] who:33 rang rgyud la grub pa'i dngos po stobs zhugs kyi rtags dang tshad ma khas len te bden 2 [the Tibetan text uses the numeral “2” rather than “two”] las rigs pa'i stobs kyis don la nges pa ni slu bar mi 'gyur te .

. accept an argument that is based on a real state of affairs (dngos po stobs zhugs, *vastubalapravrtta), which is autonomously established, and an authoritative-valid cognition, that is, an objective certainty by virtue ofa logic based on the 2 realities that is not fallible .

This is ofcourse hardly uncontroversial.

29 Ichigo (1985: 290-291). 30 See, respectively, Lindtner (2001: 61), Hoornaert (2001: 150), and Saito (2007: 224). mh[s], V: 28c, occurs on fol. 14a, line 4, of the reproduction in Jiang (1991: fol. 14a, line 4). 31 Lindtner (1995b: 97). 32 Zhwa dmar IV (manuscript, 45b) [= 2009: 523]. 33 Dar ma rgyal mtshan (2007a: 142) [= Ibid. 2007b: 160; Ibid., manuscript, 134b].

Let us continue with our exposition of the transmission of mh[t] and t j [t]. The Lhasa prints and the handwritten versions of Bu ston Rin chen grub's (1290-1364) 1322-1326 catalog that he appended to his well-known ecclesiastic chronicle and his 1335 catalog of the Zhwa lu Tanjur manuscript have slightly different entries.34 In the first we read that Legs ldan 'byed wrote pp, the Dbu ma’i snyingpo bsduspa [*Madhyamakahrdayasamgraha] in eighty sloka-s and itstj-commentary,andthemrp.Thelast,thesecond,andthe third works were translated by Nag tsho Lo tsa ba, and the fourth was done by Rgya ston Brtson 'grus seng ge (?-1040/1).While the authorship ofpp is ascribed to Legs ldan 'byed in the second, we read that Bha bya [= Bhavya] was the author of the other three treatises, and mh's title is written Dbu ma'i snying po tshig le'ur byas pa \‘*Madhyamakahrdayakarika\. The

translations of mh[t] and tj[t] are now attributed to Atisa and Nag tsho Lo tsa ba, and that of mrp[t] to these two men plus Rgya ston. Of the two catalogs of unidentified Tanjur manuscripts that were prepared by Karma pa III Rang byung rdo rje (1284-1339), thefirsthas Legs ldan 'byed (Legs ldan byed is an obvious and not uncommon misspelling), whereas the second one has *Bhavya for pp[t], mrp[t], mh[t], and for the latter dbu ma snying po 'gyur ba.35 On the other hand, the 1362 catalog of the Sne'u gdong Tanjur by Byang chub rgyal mtshan and others (wrongly attributed to Bu ston's disciple Sgra tshad pa Rin chen rnam rgyal [1318-1388]), and Mnga' ris Chos rje Phyogs las rnam rgyal's (1306-1386) undated catalog ofthe Byang Ngam ring manuscript ofthe Tanjur, only give the author's name anddispense with the name[s] ofthe translator[s].36 Finally, Ngorchen Kun dga' bzang po's (1382-1456) 1447 catalog ofthe manuscript Tanjur at Brag dkar theg chen gling monastery in Glo bo Smonthangliststhe same forpp, onlyhas Nagtsho Lo tsa ba as the translator oftj[t] and mrp—it does not register mh[t].37 Thus the entries in the early catalogs.

mh[s,t] and tj have had an as yet undetermined impact on the ways in which later Indian and Tibetan scholarship have viewed non-Buddhist Indian intellectual history, and itis equally undeniable thatthese works exerted afair amount of influence on how the Indian Madhyamaka traditions came to be

34 Bu ston (1971: 947-948) [= 2008: 1271-1272] and (1971a: 577-578) [= 2008a: 808-809]. 35 Karma pa III (2006: 555, 558) and (2006a: 696, 698). 36 See, respectively, Byang chub rgyal mtshan (1971: 511-512) [Ibid. 2008: 686-687] and Mnga' ris Chos rje (2010: 130-131).

37 Ngor chen (1999?: 583) [= 2010: 281]. Unlike the others, this catalog explicitly has a three¬fold rubric for Madhyamaka treatises: 1. primary texts (gzhung phyi mo'i bstan bcos), 2. thal 'gyur ba—prasangika corpus, and 3. rang rgyudpa—svatantrika corpus. understood in Tibet, even if, for example, in the texts' third chapter, there is a great deal that the pp has said better and in much more detail, even if less sys¬tematically. To be sure, as important Buddhist sources for Indian philosophy in general their sole rivals

within the continuum of Indian Buddhist philosophical literature are Santaraksita's Tattvasamgraha and Kamalasila's Panjikd thereon, treatises that Tibetan scholars cite far more frequently than mh[t] and tj[t] as sources of reliable information aboutnon-Buddhistphilosophicalpositions. There is also something curious about their intertextuality or, better, the lack thereof. These large-scale treatises do not even once appearto allude to mhor tj. All four must of course be considered to be the Indian Buddhist treatises on the philosophical systems (siddhdnta) that prevailed in the subcontinent. Andwe do notquite knowwhatto make ofthis absence. It now transpires that mrp is the only Indian Buddhist work that extensively uses—though it does not quote from it per se!—a verse-text of a treatise that is now known as the mh. Of the available early Tibetan expositions of philosophical systems, non¬Buddhist and Buddhist, Bya 'Chad khaba Ye shes rdo rje (1101-1175) and Phya pa Choskyi sengge (1109-1160) obviously used mh[t] and TJ[T]. It is of perhaps minor interest that 'Chad khabacites tj[t] as Rtog ge la 'barba.

Some three centuries later, when he was compiling and writing the extensive version of his encyclopedic and massive De kho na nyid kyi 'dus pa, Bo dong Pan chen 'Jigs med grags pa (1375/6-1451), alias Chos kyi rgyal mtshan and Phyogs las rnam rgyal, used tj[t] as a fundamental authority for expositions andrefutationsofthetenetsofthenon-BuddhistIndianphilosophicalsystems of the Samkhya, Vaisesika, Vedanta, Mimamsa, and the Digambara Jain. Quite mindful that the text at his disposal was often flawed and defective—we learn this in the colophon—he essentially reproduced large portions of tj[t], in which at times better readings are preserved than those ofthe texts contained in the later Tanjurs.39 The question of “contaminated texts” with reference to mh[t] and tj[t] was raised directly and indirectly by a number of other authors as well, including the Mongol scholar Urga Chos rje Ngag dbang dpal ldan (1797-?), alias Dpal ldan Chos rje.40 A final detail: we knowfromSkyo ston

Smon lam tshul khrims's (1219-1299) biography of Mchims Nam mkha' grags (1210-1285) that the latter studied Tj[t] under 'Phags pa Blo gros rgyal mtshan (1235-1280).41 This was certainly not an anomaly. As we comb through the vast Tibetan biographical and autobiographical literature we will surely come up with manymore instances ofindividuals studying mh[T] and/orTj[T].

We nowbrieflyturnto the only available Sanskrit manuscript ofmh, mh[s]. The fact that it lacks a title page should not be a cause for alarm; the manuscript ends with the statement: tarkajvala nama sutram samaptam iti, that is, “the sutra called Tarkajvala is finished.” This suggests, of course, that the verses contained in this manuscriptwere considered to be those ofasutra-work called Tarkajvala. On the other hand, mh[s], XI: 3a-b, and mh[T] have: iti madhyamakasyedam samksepad dhrdayam krtam /

de ltar 'di ni dbu ma yi // snying po mdor bsdus byas pa yin // Tj[T] explains that when the ka-affix is added to madhyama, the resulting madhyamaka suggests either a Madhyamaka-treatise or Madhyamaka-philo- sophical system. mh[s] neither has a titular identification nor the mention of an author, it but ends in:

stutilaksananirdeso nama ekadasah paricchedah tarkajvala nama sutram samaptam This means that this work, this sutra, was called Tarkajvala!On the otherhand, mh[T] and Tj[T] end in:

dbu ma'i snyingpo 'grelpa rtog ge 'barba las/bstodpa dang mtshan bstan pa'i le'u ste bcu gcig pa'o // // dbu ma'i snyingpo'i tshig le'ur byas pa'i 'grel pa rtogge'barba/slobdpon chen pobhabyasmdzadpa rdzogsso//

This phrase may, but only may, indicate that this work could have been called Madhyamakahrdaya; mh[T] has a final, dedicatory verse not found in mh[s], in which we read dbu ma'i snying po'i bstan bcos, that is, madhyamakahrdaya-

he was able to write, on p. 156, that the majority of the negating-arguments (dgag rtags) leveled in mh[t] and tj[t] are based on arguments that are founded on the perception of an incompatibility ('gal zla dmigs pa, virodhopalabdhi).

41 See Skyo ston (manuscript: 15b).

sastra. But this is not necessarily a title, for it can also mean “treatise on the essence ofMadhyamaka.” The term madhyamaka/dbu ma occurs onlyonce in the bodyofmh[s,t] initsverylaststanza. Onthe other hand, dbumarsmraba and dbu ma pa, that is to say *madhyamakavadin and madhyamika, do occur, albeit onlyvery sporadically in tj.42 This has a parallel in pp, which also explic¬itly speaks of a school of thought that is called dbu ma and whose adherents are called dbu ma smra ba [[[rnams]]] or dbu mar smra ba, and dbu ma pa, that is, madhyamakavada/in and madhyamika.43 Although two late eighth century Madhyamaka intellectuals wrote the Tattvasamgraha plus the panjika, these titles do not suggest that these texts deal with Madhyamaka thought per se and, indeed, they do not. Structurally, they move hierarchically through a cri¬tique ofaplethoraofnon-

Buddhistphilosophicalsystemsandsundrynotions, until they reach the final apotheosis, that is, a discussion of the omniscience (sarvajnata) of buddha/the Buddha. Given that mh gives Madhyamaka anal¬ysis its due in especially chapters three, four, and five, the subsequent anal¬yses of the non-Buddhist systems of thought suggest either that its architec¬ture was quite different from the Tattvasamgraha, or that these were in fact later additions. We say this because it strikes us that, just as the Tattvasam- graha, so mh does practically end in a chapter which involves an assessment of the omniscience of buddha/the Buddha. The tenth chapter of mh/tj dis¬cusses questions relating to Mimamsa and

Buddhist notions of omniscience^ and the so-called “eleventh chapter” of mh[s] consists of a two-verse obei¬sance to the lokavid, the Buddha, and one dedicatory verse in which the title is recapitulated. Thus, these three verses—and anextraone in mh[t]—hardly merit being called a “chapter,” even ifthe commentary they receive in tj is of definite interest!45 pp, hr and mrp all lack a dedicatory phrase or stanza, so that that in mh[t] may very well have been an addition by a Tibetan editor-reader or it may, once again, point to a different filiation of mh[s] and the Sanskrit manuscript on which mh[t] is based. Finally, the topical structure of mh and tj, as they stand now, do not in the least betray the more familiar hierarchical organization, the doxographical hierarchy, of those other works in which there is a discernable, progressive movement, with transitions from themostbasephilosophicalideas—variouskindsofphilosophicalrealismand theologies—to the most sublime—the Madhyamaka school. Atthe same time,

42 bg, vol. 58, no. 3083, 471, 474 and 185, 255. 43 bg, vol. 57, no. 3080, 1139, 1244. 44 See now also the recent in-depth discussion in Krasser (2012:545-550,559-569). 45 b g, vol. 58, no. 3083,793-803.

they do remind us of the structure of Dignaga's Pramanasamuccaya. For there, Dignaga first exposes his own [[[Buddhist]]] analyses of the topics under disc- ssion, after which he systematically and critically engages another Buddhist author, Vasubandhu, and then goes on to submit a variety of non-Buddhist philosophers and their positions to scathing critiques. In other words, we do not encounter the same hierarchical structure of treatises that, in the eyes of their authors, begin with what they considered to be the most naive or base point of view, and then slowly transition step by step to what they con¬sidered to be the most sophisticated view of what there is [or is not], the worldview most conducive to soteriological success and breakthrough. The Madhyamakalamkara of Santaraksita, Aryadeva [Il’s] Jnanasarasamuccaya, Candrahari pa's *Madhyamaka-Ratnamala (translated circa 1000) and Saha- javajra's (11th c.) *Sthitisamuccaya46 are classic examples oftracts that are struc¬tured according to these lines. Long ago, in his edition of a lengthy excursus or digression in the twenty¬fifth chapter of Bhaviveka's pp[t] giving a critique of the Yogacara school of thought, Lindtner pointed out that it mentions mh's Yogacaratattvavatara chapter

(skabs). The sentence in which this occurs is rather awkward if not outright ungrammatical; it reads:47 dpyadpa rgyas par [read: pa] ni dbu ma'i snying po'i dekho na la 'jug pa'i skabs nas rnal'byorspyodpa pa dag dang lhan cig byedpar'gyurro // zhar labshadpaschogste'dinimngonparbrtsonparnamsranggisdepadang lhan cig rigs pa dang rigs pa mayin pa rtogs pa'i phyogs tsam zhig bstan pa yin no // A detailed investigation is included in the Yogacaratattvatara chapter of the Madhyamakahrdaya4 enough of the digression, that is, the digres¬sion in which is shown but an aspect for understanding what is and what is not reasonable in connection with our own industrious cohort.

Neither Lindtner nor Eckel, who translated this passage into English, pointed out [or seem to have been aware] that it occurs there in a kind ofan excur-

46 For the latter two, see bg, vol. 63, no. 3130,1035-1049, and vol. 26, no. 1132,253-272.

47 See Lindtner (1984: 95-96); see also bg, vol. 57, no. 3080,1425. Our translation of this somewhat problematic passage differs from the one in Eckel (1985: 74).

48 With bg, vol. 57, no. 3069, 443, we correct de kho nala 'jug pa'i skabs nas rnal 'byor spyod pa pa to rnal 'byor spyod pa pa'i dekho na la 'jug pa zhes bya ba'i skabs nas.

sus (zhar la bshad pas chog ste), for which pp[t] offers a number of other instances. Krasser, who tabulated agood number ofthese, remarked that they appear to be “summaries of discussions” that are found in mh and tj “(and elsewhere?)”.49 They are of course of great interest for the textual history of pp. The Tibetan tradition seems to have had no special problems with their presence in pp[t] and has simply understood them to have been written by Bhaviveka. What is important to note in the present instance is that the ref-erences in this passage to the Yogacaratattvatara chapter and the Madhya- makahrdaya are absent from Prabhakaramitra's Chinese translation of the excursus.50

Nothing is known about Avalokitavrata (ca. 700), who apparently composed the [only extant] commentary on pp, which work was translated by the same team to whom we owe pp[t]—the earlier exegesis of pp by Devasarman is not extant and is known only through a few quotations in Avalokitavrata's large study. From many aspects a very rewarding treatise, the latter's opening lines contain a line in which homage is paid to Avalokitavrata himself; we find something similar in the translation of pp[t] which also includes such a line honoring Bhaviveka. This is a rather anomalous state of affairs. In all likelihood, these lines are owed to the translators, for we meet with the same in their translation of Buddhapalita's Mulamadhyamakakarika commentary^ Inconnectionwiththe above passage frompp[t], the commentaryintroduces the title Tarkajvala as a commentary of mh, as well as what appears to have been the full title ofone ofmh's chapters, namely, Rnal byorspyodpa pa i de kho nala jugpa [= *Yogacaratattvavatara]; it states:52 thegpachenpodbumapa igzhungbstanpadang/rnamparshespatsam du smra ba daggigzhung mi thadpardpyadpargyaspar[read:pa] nislob dpon legs ldan byednyidkyisdbuma irtsaba i tshig le urbyaspa dbuma i snying po i grel pa rtog ge bar ba zhes bya ba i nang nas rnal byor spyod pa pa i de kho na la jug pa zhes bya ba i skabs ston par gyur ro zhes bya baryang sbyarro // zhar la bshadpas chog ste dinimngon parbrtson pa

49 Krasser (2011:50-52). For a recent study of one of these, see Akahane (2012:8-25); see also Akahane (2013:1182-1188). 50 t.1566[XXX],131b,andvanderKuijp(2006:195,n.78). 51 bg, vol. 57, no. 3069, 443. 52 See van der Kuijp (2006: 195, n, 79). The Tibetan of this passage is not altogether clear and, oddly, in this passage the title Dbu ma rtsa ba’i tshig le’ur byas pa, that is, Nägärjuna's Mulamadhyamakakarika, is placed in an appositional relationship with Dbu ma’i snying po.

rnams rang gi sde pa dang lhan cig rigs pa dang rigs pa ma yin pa rtogs pa'iphyogs tsam zhig bstan payin no zhes bya ba ni zhar la bshadpa'i mjug bsdu ba yin no // A demonstration of the textual tradition ofMahayana Madhyamika and an extensive examination into the fact that the textual tradition of the Yogacara adherents is incorrect were also tied to the statement: “will be shown [in] the section called *Yogdcdratattvdvatdra” from the text called *Tarkajvdld, the commentary on the so-called *Madhyamakahrdaya, the *Mulamadhyamakakdrikd, by Master *Bhaviveka. The digression, that is, “This has shown but an aspect for understanding what is and what is not reasonable in connection with our own industrious cohort,” is the conclusion ofthe digression.

mh[s] titles its chapter five on Yogacara Yogdcdratattvaviniscaya. On the other hand, the expression tattvdvatdra occurs only in chapter six, which is devoted to an exposition and critique of Samkhya philosophy, whereas chapter four on srdvaka-Buddhism has tattvaviniscaydvatdra and chapter nine on Mimamsa has tattvanirnaydvatdra—to be sure, viniscaya and nirnaya are synonyms. In other words, then, a chapter with the title *Yogdcdratattvdvatdra is neither attested in the currently available Sanskrit manuscript, nor is it reflected in mh[t] ortj[t].Wedonotknowhoworwhenthemanuscriptsofppandits commentary entered the Tibetan cultural area, let alone their age. Given that both were translated by the same team

with Jnanagarbha being styled “the Madhyamika” (dbu mapa), one can imagine without much difficulty a scenario in which manuscripts of mh together with tj, and those of pp together with Avalokitavrata's commentary, were brought to Tibet at about the same time in two bundles, one containing the former and the other the latter. An argument based on this assumption might continue with the likely identical provenance and filiation of the manuscripts of pp and its commentary that Jnanagarbha had carried with him when he went to Central Tibet. This would explain why both contain the reference to mh. On the other hand, the pp text ofthe Sanskrit manuscript that was available to Prabhakaramitra did not contain this reference,andthisabsenceisbutoneadditionalindicationthathismanuscript was no doubt differently filliated from the manuscript on which pp[t] was based. It is furthermore striking that Avalokitavrata himselfnot once explicitly cites from mh, tj, or hr in his large commentary. One could have expected that he would have done so in his excurses on the philosophical positions of the Samkhya and other non-Buddhist schools. But he did not and that he did not shouldraise—ifithasnotalreadydoneso—

someinterestingquestions.More eyebrows will be raised when one begins to compare the surveys, in tj[t] ad M H [ s, T], III: 130, of the main features of the Vaisesika, Samkhya, and Digambara Jain schools of thought53—neither the Vedanta nor the Mimamsa schools are discussed in this chapter—with the respective chapters ofthese three schools inmh[s,t] and tj[t].

Even though the Tibetan text of pp does not have a possessive pronoun attached to the phrase beginning with dbu masnying po, the presence ofthe title inbothpp[t] and its commentary with its express attribution to Legs dan 'byed/byed (sic) forms a crucial link in Lindtner's argument that the author of the pp was also the author ofmh and tj. Be this as it may, it is difficult to say howthismightplayitselfoutinlightofthesupposition,towhichhehimselfhas repeatedly subscribed and for which there is some evidence, that the verse-text of what is now called the mh was extracted from a work that had originally consisted ofa mixture ofverse and prose, and that presumably at one pointin time had begun to bear the title tj. And these suspicionsgaininintensitywhen we realize that often, and with equal or better justification, many cases could have been made for making such references elsewhere in the pp[t].

On the Tattvdmrtdvatdra, Entry into the Nectar of Reality, and tattvdmrta (Nectar of Reality)

Notwithstanding the fact that mh[s,t] is written in verse, the passage Won- ch'uk cited from its so-called Tattvdmrtdvatdra-chapter was in prose. Can- drakirti cites verses from mh but does not give the title ofthe source ofthese verses, so that Wonch'uk's citation is so far the very first indication of a work titled mh, even ifthis citation produces some dissonance. It is ofcourse pos¬sible that in the course ofits transfer to a Chinese medium, the verse text was changed into prose, even ifwe do not know whence or from whom Wonch'uk may have obtained this citation. Gokhale54 first held that the core ofthe textual corpus comprising what we now call mh may have originally consisted ofits first three chapters thatare as itwere announced inmh[s,t], I: 4-5: [1] Bodhi- cittdparitydga, [2] Munivratasamdsraya, and [3] Tattvajndnaisand—and that this treatise was in actual fact called Tattvajndndmrtdvatdra,55 a phrase which,

53 bg, vol. 58, no. 3083, 221-222, 222-223, and 223-224. 54 For what follows, see Gokhale (1958: 165-166, n. 1), (1972: 41-42), and his paper written jointlywithBahulkar(1985:78).SeealsoSeyfortRuegg(1981:63). 55 The expression tattvajndndmrta occurs once in these three chapters, namely, inMH[s,T], III: 136; seeEjima(1980a: 302-303).Butwhenwe lookatthecommentitreceivesintj— he

held, occurs in the colophon of the third chapter of mh[s,T]. This was of course an oversight, since there is no such remark there. Later he maintained, a sliver more convincingly, that this core work was titled Tattvamrtavatara, albeit without really explaining why he had changed his mind on something this significant. But there is much merit in this point of view. The expres¬sion tattvamrtavatara appears thrice in mh: once in mh[s,T], I: 4, once in the “colophon” ofmh[s], III,56 and once in mh[s,T], V: 1c. The first was orig¬inally misread by Gokhale as tattvamrtavadharaya—perhaps because ofread- ing Tibetan gzung instead of gzud—but he appears to have later corrected this reading to

tattvamrtavataraya on the basis of mh[s], V: 1c, which indeed has tattvamrtavatarayo hi.57 The comment ofTj[T] ad mh[T], I: 4, is illumi¬nating and focuses on the conduct (carya) for attaining all that is meaning¬ful (sarvartha). This conduct has two dimensions: a moral-existential and an epistemological-philosophical one. The first is in the main, but by no means exclusively, reflected in the first two chapters that have to do with, respectively, retaining the commitment ofhaving resolved to become awakened (bodhicitta) andofkeepingthe vowofthe ascetic. The secondaspectismooredinthe third chapter, whose title, Tattvajnanaisana, can be rendered Quest for the Knowl¬edge ofReality [or: Quest for Salvific Realization]. Prefiguring the titles of the first three chapters, mh[s], I: 4, has: mahabodhau krtadhiyam pararthodayadlksaya / tattvamrtavataraya saktitah kimcid ucyate // mh[T] , i: 4 reads:

gzhan don bsgrub phyir brtulzhugskyis// byang chub chen por blos byas rnams // de nyid bdud rtsir gzudpa'i phyir // nuspascungzhigbrjodparbya// Without giving elegance of diction a single thought, Gokhale and Bahulkar rendered the Sanskrit verse as:58 see bg, vol. 58, no. 3083, 225—it becomes immediately clear that the text does not indicate that it is a title.

56 Ejima (1980a: 360). To be sure, the title of the third chapter, Tattva/nanaisana, is prefigured in mh[s,T], III: 1, for which see Ejima (1980a: 268-269). 57 See Bahulkar (1994: 1, 41) and Lindtner (2001: 1, 90). 58 See also Heitmann (1995: 391-392).

With penance (dlksd) for the sake of the emergence (udaya) of others' welfare (pardrtha), [the aspirants] have dedicated [their] minds (krtadh!) to the [[[attainment]] of] Supreme Wisdom (mahdbodhi);—in orderto lead [them] to the Nectar ofTruth (tattvdmrtdvatdra), something within the reach of [my] abilities (saktitah) is being spoken here. The Tibetan verse translates as:

In order to accomplish what is meaningful for others, I will explain something by whatever abilities I may have,

In order that those who focus on the great awakening with hardship, Willenter the nectar ofreality [or: salfivic realization]. To be sure, of undoubted importance for the supposition that the first three chaptersformthecoreofthisworkisthatmh[s,t],1:5a-c,isreiteratedtowards the end of tj[t] admh[t], III: 360.59 And this brings us back, if only for a moment, to the final sentence atthe end ofthe third chapter ofmh[s] where we read: tattvdmrtdvatdre tattvajndnaisandparicchedo ndma trttyah samdptah //, that is, “The third chapter titled Quest for the Knowledge of Reality in the Tattvdmrtdvatdra is completed.” The corresponding passage in mh[t] and tj[t] have

very differently: dbu ma snying po las de kho na nyidkyi shes pa tshol ba'i le'u ste gsum pa'o //, “Chapter ofthe Questforthe Knowledge of Real¬ity from the Madhyamakahrdaya, the third.” The term tattvdmrta/de nyidbdud rtsi also occurs in mh[s,t], III: 300, and mh[s], VIII: 95 {= mh[t], VIII: 88}.60 However, the comments on these in tj[t] do not in the least suggest that the uses there of tattvdmrta suggest that these were part of a larger title.61 As is known, the Tibetans translated the word amrta in two different ways, namely, as bdudrtsi, “nectar” or literally “demonic essence” and as 'chi med, “deathless.” The term “nectar,” as Heitmann underscored, originally meant “overcoming death”62 and then came to be used in the sense of “divine drink” or “essence.”

59 bg, vol. 58, no. 3083, 350-351. 60 For the first, see Ejima (1980a: 342-345), and Eckel (1992: 174, 176). For the second, Gokhale (1972:42), that it was the ninety-fourth verse, but Qvarnstrom (1989:46,97,152) counts it as the ninety-seventh, and He (2013: vol. 2, 783-784), reckons it as the ninety-fifth of mh[s], viii,andtheeighty-eighthofmh[t],VIII.Thereissomesimilaritywithmh[s,t],III:354, which contains the term prajndmrta/shes rab bdud rtsi, for which see Ejima (1980a: 358). 61 bg, vol. 58, no. 3083, 322, 324-325 and we find the second on pp. 656-657. 62 See Heitmann (1995: 391-392).

Indeed, tattva/de kho na is defined in mh[s,t], I; 2c, as “peace” (santi, zhi ba). The salvific idea behind the term tattvamrta, then, is that it is, among other aspects, ofthe essence ofpeace.

It would be difficult to argue that the first three chapters were not originally a self-enclosedtextualcorpusandthatthisworkinitsoriginalformwasnottitled Tattvamrtavatara. The recapitulation of mh[s,t], I: 5a-c, at the end of TJ[T], III, speaks for this, as do the last ninety-four verses of this chapter and their prose commentary, which present us witha Mahayana view of enlightenment, the end result ofproper conduct, spiritual practice, and the understanding of reality, the understanding of what there is, that is, enlightenment, or simply peace. To be sure, the term *tattvamrta-de nyid bdud rtsi also occurs in pp[t], but in a sense that is but in small measure relevant to this paper, for it states that the Buddha:63

... shes rab kyis shes byairgyamtsho mngonparbsrubs te/chos thams cad kyide kho na i bdud rtsi spros pa idra ba mtha dag dang legs parbralba / gzhan gyidringmi jog pa dang /rnam par mi rtog pa brnyes .

. having churned the ocean of the knowable with his discriminative awareness, acquired the nectar ofthe reality ofall phenomena, which is appropriately free ofallofconceptualproliferations, whichis notdepen- dent on something else, and which is non-conceptual .

Now Lindtner maintained, in connection with the two references Bhaviveka made to the Tattvamrtavatara in hr, that these actually refer to what is now tj[t]'s fifth chapter, in which the author reflects on the philosophical cate¬gories ofthe Yogacara school, though he did notpointto anyspecific passages ofthischapter.64mh[s] titles this chapterYogacaratattvaviniscaya, the Tibetan translation in mh[t] has Rnal byor spyodpa ide kho nanyidgtan ladbab pala jug pa [*Yogacaratattvaviniscayavatara], and the title in the Tibetan transla¬tion oftj[t] has the same. Ofthe two references hr made to the Tattvamrta- vatara, the one bearing on tathata is perhaps the most clearcut. It turns out that we do not have to go to the fifth chapter for a discussion ofthe ontolog¬ical status of tathata and its difference from the atman and any of the other

63 bg, vol. 57, no. 3080, 906, mentions that Nagarjuna experienced an intense joy (dgyes pa i shugs drag po) after he unerringly understood just what there is (tathata, de bzhin nyid). Avalokitavrata links this up with the first-pramudita Bodhisattva stage, the joyful one; see bg, vol. 58, no. 3086, 911-916, and also Ames (1993: 214, 236, n. 24). 64 Lindtner (1984a: 167, n. 20).

postulated, eternal entities ofthe non-Buddhisttraditions. Inactualfact, these are all dealt with at some length in mh[s, t], III, 129ff.65 In fact, the notion of tathata is barely mentioned in the fifth chapter. This is further evidence that the Tattvamrtavatara existed at one time as an independent treatise. On the other hand, the term paratantra/gzhan [gyi] dbang makes its first appearance inmh[s,t], V: 5,66 and tattvamr

tavatara occurs in mh[s,t], V: 1? This verse and the comments in the tj[t], the grammer of which is not altogether unprob¬lematic, state the following:67 anye pracaksate dhlrah svamtav abhimaninah / tattvamrtavataro piyogacaraih sudesitah //

rang gi lugs kyi nga rgyal gyis // mkhas parrlom gzhan 'di skadsmra // de nyid bdud rtsir 'jug payang // rnal 'byor spyod pas legs bstan zer // theg pa chen po nyid kyi slob dpon thogs med dang / dbyig gnyen la sogs pa gzhan dag nide bzhin gshegs pas lung bstan cing /sa rab tu brnyes pa'i 'phags pa klu sgrub kyis yang dag par rtogs pa'i theg pa chen po'i don gyi lugs gzhan du 'dren par byed cing ngo tsha dang khrel med pa don rnam par mi shes pa de bzhin du rnam par shes shing mkhas pa rang rgyal byed pa dag 'di skad smra ste /

dekho na nyidkyibdudrtsir'jugpa ste/rab tu'jugpadenibdagcag kho na legs par bstan pa yin gyi / dbu mar smra ba rnams kyis ni ma yin zhes zer ro //

With pride in their own scholarly tradition, Others, boasting to be learned, allege this: ‘Entering into the nectar of reality, too, Is well-stated by the Yogacara school.' 65 Ejima (1980a: 300-301ff.). 66 Lindtner (2001:58); see also Hoornaert (1999:135-136,150-154) and Saitö (2007:207-208). 67 Hoornaert (2007:131,139-140) and Saitö (2007:203); see also Eckel (2008:213-215).

Other masters of the same Mahayana such as Asanga, Vasubandhu, etc., teach differently the objective position (don gyi lugs) of the Mahayana that was correctly realized by Nagarjuna who had been prophesied by the Tathagata and who had attained a Bodhisattva stage,68 and are shameless andunabashedanddo not understand the intent of the Mahayana and, likewise, are smug in their understanding and learning, and say this:

Entering into the nectar of reality, that is to say, really entering is what only we teach well, but it is not the case on the part ofthose who speak ofthe Madhyamaka. Thus the answer is simply: “Nothing is added by this reference!” And we do not need to do anything, for, even when this echo can clearly not be entirely unintentional, tattvdmrtdvatdra is here not used as a book title. Still, it does indicate one important fact, namely, the existence of some kind of a textual continuity, some kind of a relationship, of tj [and presumably mh] with a work that had the title Tattvdmrtdvatdra. Indeed, we would be inclined to form the hypothesis that portions ofthe original mh, III, may have originally dealt with the Yogacara school, which then later grew into a separate chap¬ter.

Much earlier, Ejima had argued that a distinction needs to be made between an original tj [= *tj] and the text that we now have of it, that is, tj[t].69 The authoroftj,then,mustbedistinguishedfromtheonewhowrote*tj,pp[and hr]. Ejima's hypothesis is that this Bhaviveka was also the author of mrp, the Madhyamakdrthasamgraha, and the Nikdyabhedavibhangavydkhydna— essentially a reproduction oftj[t] ad mh[s,t], IV: 8—and for him this indi¬vidual flourished in the eighth century [or perhaps even later]. Krasser most recently reexamined the issues in connection with the Tarkajvdld and the Madhyamakdrthasamgraha.70 The latter contains several terms and concepts 68 The phrasing Bhaviveka used at the outset of his pp[t], in bg, vol. 57, no. 3080, 906, is suggestive of this characterization: slob dpon gyis . de bzhin nyid thugs su chud nas dgyes pa'i shugs drag po yang dag par skyes pas .. This passage from tj[t] minus the last verse is also quoted by 'Jam dbyangs bzhad pa'i rdo rje I (vol. Pha, 693) with one important difference: instead of the canonical / sa rab tu brnyes pa'i 'phags pa klu sgrub kyis yang dag par rtogs pa'i theg pa chen po'i don gyi lugs, it has / sa rab tu brnyes pa'i 'phags pa rnams kyis yang dag par rtogs pa'i don gyi lugs, that is, “. the objective position that was correctly realized by the holy ones who had attained a Bodhisattva stage.”

69 See Ejima (1980a: 1-38); see also Lindtner (1982: 183). 70 See Krasser (2011: 60-71) and (2012: 554-556, 569 ff.).

that are closely linked to Dharmakirti, whose floruit is generally accepted as the seventh century. The main issues with the text ofTj as it is constituted in Tj[T] is that it contains a number ofvery pedestrian remarks concerning the logical structure ofan argument that is presented in the mh verses, and that it on occasion clashes with mh[T].71 Weighing the evidence, Krasser quite inge¬niously suggested that these could be best explained by the fact that the text ofwhich Tj[T] was a translation was put together by a beginner, and that this beginner was astudentofthe authorofmh. Thus Krasser writes: I do not believe that Bhaviveka would have impressed his opponents or anyone else by showing his ability to identify the property possessor or the reason in a verse he had formulated for himself.

This is an elegant solution to the problem, and best explains the occurrence ofthe phrase slob dpon (acarya aha), “the Masterspoke,” that often introduces a verse ofmh[T] in Tj[T]. But we believe that this highly original solution is somewhat, but only somewhat, weakened bythe fact that Bhaviveka and Ava- lokitavrata have occasionally taken a very similar approach of breaking up an argument into its component parts and identifying these using the techni¬cal terminology of“property” (chos, dharma), “property-possessor” (choscan, dharmin), etc. in their respective analyses of the arguments of the Mulama- dhyamakakarika and pp. And we find Bhaviveka doing the very same thing in hr. Once More on the *Madhyamakamtnapmdlpa

Long ago, Lindtner argued against the prevailing wind that Bhaviveka was the author of mrp.72 One, admittedly minor, part of his argument was that its colophon clearly states that this was the case. Butwe all know that colophons are by no means reliable testimonies or necessarily convincing arbiters of questions relating to authorship. They can be notoriously wrong. As a matter of fact, a number of the colophons of writings that colophons attribute to

71 Krasser (2011: 69). 72 See first of all the note in Lindtner (1979: 87-91), and then in greater detail (1982: 172¬

184). His view has been adopted in Qvarnstrom (1999:177), and a few others. However, Yamaguchi Susumu Ejima, Seyfort Ruegg and others have argued against this attribution and, we believe, with good reason; see, for example, Ejima (1980a: 4-18, 263), Lindtner (1982: 186-188), and Seyfort Ruegg (1990: 62-65).

Atisa now turn out to be in profound and egregious error. Whether these can be called forgeries in the sense of a conscious act of falsification is a difficult question to answer. It goes to motive, and motives are often complicated and obscure creatures ofthe mind, not onlywhen we are concerned with those of others, but even our own motives often elude us. The fact remains, however, that colophons of texts can be highly misleading. Of course, they can tilt the conclusioninacertaindirectionwhenmostoftheevidencepointstothatsame

direction. In the present case, however, the evidence that mrp was written by the same authoraswas ppis so flimsyand so equivocal that it really cannot do much workforus. This is notto saythatthe Tibetan scholarly community was unequivocal as to mrp's putative author. On the contrary, aside from the fact that this work barely finds mention in the available literature, thereisevidence that certain intellectual circles were very much in doubt ofwhat its colophon alleged aboutits author. Ofthe early catalogs that we have used for this paper, two cast definite aspersions on the integrity and reliability ofmrp's colophon and authorship. The text ofMnga' ris Chos rje's catalog ofthe manuscriptTanjur of Byang Ngam ring contains the following gloss after “Bhavya” in its entry of mrp: “He is not *Bhaviveka, because he quotes passages of Candrakïrti and Dignaga.” ( di slob dpon legs ldan byed

[read: byed] ni ma yin te zla grags dang phyogsglanggilung drenpa iphyirro//).73The mentionofDignaga(6thc.) isa trifle unfortunate, forthere is no doubt that Bhaviveka as the authorofppand hr [and also at least ofsome version ofmh and tj] was very well acquainted with Dignaga and his logic.74 What this of course meansis that this mentionof Dignaga cannot be used as a criterion for anything. At the same time, the clear indicationthatmrpusedandreferredto Candrakïrti'sDbumaphungpolngapa [*Madhyamakapancaskandhaka =

Pancaskandhaprakarana] for further details on the question of Abhidharma ontology and its refutation is right on the mark, since Candrakïrti had taken pp to task on several occasions, and is thus at least posterior to the “publication” of Bhaviveka's major work; mrp also cites Candrakïrti's Madhyamakdvatdra. Ngor chen's use of the term grags, “it is known”—grag, “it is alleged,” would ofcourse be the better reading—in the entry for mrp in his catalog ofthe Brag dkar theg chen gling Tanjur manuscript 73 Mnga' ris Chos rje (2010: 130-131).

74 See Krasser (2012: 535-537) and also Lindtner (1986: 58-84), and Shi Ruyuan (available on-line). Indeed, Dignäga's Älambanapanksa is quoted in, for example, TJ[T] adMH[S,T], V: 31-32; see Hoornaert (1999:131,139-140), and Saitö (2007: 226-227). He Huanhuan is currentlyworkingonanessaythatistentativelytitled“Dignäga,Bhäviveka,andXuanzang on the 'Restriction of the Thesis' (*prati/navisesana)” also suggests that there was a problem with its authorship, just as his use ofgrag in his entry for the Akutobhaya-commentary on the Mulamadhyamakakarika that had been wrongly attributed to Nagarjuna.75

Perhaps, the glossator had made an error, and instead of Dignaga intended Dharmakirti. But this, too, may not be entirely unproblematic, for Krasser has takenanotherlookatanumberofpassagesfrommh[s] andtj[t] that indicate that there definitely exists a close intellectual relationship between Bhaviveka and Dharmakirti, even if the direction of the influence may not be immedi¬ately transparent.76 To this end, Krasser pointed that tj[t] ad mh[s,t], IX: 45a-b, involves an inference techically known as a sattvanumana. As formu¬lated by tj[t], it runs in part: . gang dang gang yod pa de ni mi rtag pa ste ., that is, “. whatever exists is impermanent . .” In his view, then, the use ofthis inference shows that the author ofmh[s,t] and tj[t] was either well aware ofDharmakirti, or that Dharmakirti borrowed this argument from him. However, in her essay on the development of this argument, which Krasser cites in the passage of his just mentioned articl, Yoshimizu Chizuko

pointed out that a version of the sattvanumana was already quite hard at work in Vasubandhu's Abhidharmakosabhasya and Karmasiddhiprakarana,77 so that we are not altogether convinced that, in this instance, Dharmakirti must have been indebted to Bhaviveka, even if Sakai Masamichi ’/§#0, whom Krasser cites, quoted a passage from Arcata (8th c.) to the effect that Dhar- makirti had given a radically new spin to the sattvanumana.78 Summarizing his findings,Krasserproffersaseriesofotherpointsthathemadeinhislongarticle that might possibly suggest that the directionofinfluence was such that Dhar- makirti was slighty anterior to, or better a junior contemporary of, Bhaviveka. Iftrue, this would ofcourse have serious consequences for Dharmakirti's dates, which would need to be pushed back to the sixth instead ofthe seventh cen¬tury. We have in principle nothing against a little chaos, but this would mean that we have to reevaluate much of what we thought we knew about Indian

75 See Ngor chen (1999?: 583) [= 2010: 281]. It would appear that the suggestion that the Akutobhaya was Nägärjuna’s autocommentary was based on a statement to this effect in Avalokitavrata's pp commentary; see bg, vol. 58, no. 3086, 868. It is not known on what he may have based this assertion or whence he took it. This was then repeated in the Bodhimargadipapanjika, for which see Bg, vol. 64, no. 3178,1781. For the Akutobhaya, see, for example, Huntington (1995).

76 See Krasser (2012:556-558). 77 See Yoshimizu (1999:231-254). 78 Krasser (2012: 557, n. 56); on this theme, see now also more recently Sakai (2011/12: 187-202).

intellectual history. As an example, the assertionofDharmottara(late 8thc.), a disciple of Arcata, that Uddyotakara was a contemporary of Dharmakirti would then also lead to a reassessmentofthe developmentofearlyNyaya. Theauthorofmrpfrequentlycitesmh,albeitwithoutsayinganythingab out its authorship,79 and Ejima, Lindtner, and Seyfort Ruegg all have understood the unique phrase bdag gis bkod pa rtog ge 'bar ba to mean “the Rtog ge 'bar bathatwas written by me,” so that the authorhad explicitly stated that he was infactthe same Bhaviveka who authored tj.80This notwithstanding, were we to maintain that the authorofthe tj[and mh] is identical to the Bhaviveka of pp, then the more important linchpin ofLindtner's argument for holding that mrp is an authentic work ofBhaviveka is ofcourse based on the fact that its author expressly states that he was also the author oftj. But did he? Miyazaki has recently re-examined this veryphrase and, noting an important parallel in the Bodhimdrgadlpapanjikd, suggested that bkod could have been used in the sense of “to cite.” As an example of such usage, he refers to a single passage in the latter where a

citation from Bodhibhadra is glossed by . ces gsungs pa de bdag gi rtsa bar bkod pa yin no //.81 And, indeed, the verse in question is found in Bodhipathapradlpa, 45. Be this as it may, in our admittedly limited experience we have never seen bkod used elsewhere in this sense and neither, for that matter, in any of its attested Sanskrit equivalents.82 But we believe that Miyazaki's point is well taken and requires serious consideration, even if it is weakened by the fact that several dictionaries register “to write” ('bri) as a secondary use of the verb 'god, of which bkod is the perfective form. It is furthermore somewhat disturbing that, though he cites so many unattributed verses that are actually

taken from mh, the author not once says that he himself had written the work from which these are quoted. mrp's reference to tj and the pp's reference to mh led Lindtner to the obvious inference that its author was therefore the same as the author of pp. But we need to be much more circumspect. All that we can say with certitude is that the author knew mh and tj, and that he flourished after the tenth century, because he cites the Alokamdd.83 of Prajnamitra, alias Kambala, and the Pancakrama, the latter of 79 See Ejima (1980: 950-947), and also the list ofquotations, from mh, III, V, and IX [: 13 = Vdkyapadiya 1:42], in Lindtner (1982:178-182). We do not agree with Seyfort Ruegg (1990: 63) that the passage he cites from mrp can be interpreted in the sense that its author says that he also wrote mh.

80 See,respectively,Ejima(1980:951-950),Lindtner(1992:177),andSeyfortRuegg(1990:63). 81 Miyazaki (2006: 453-448). 82 See, for example, Negi (1993: vol. 1, 122-124). 83 See, for example, the remarks in Moriyama (1992: 436-433). which was authored by a later, possibly ninth century, person who was called, or who signed himself, Nagarjuna.

Absent from mh/tj and pp, an even more problematic passage in mrp is its putative citation from the Manjusrimulakalpa of a prophecy of Nagarjuna in which a clear attempt is made to collapse the authors of the Mulama- dhyamakakdrikd and the Pancakrama. There the alleged prophecy runs as fol- lows:84 dge slong klu sgrub ces bya ba// rang bzhin meddon de nyidrig // sa nidrugpa sgrubpaste// lo ni drug brgyar 'tsho bar 'gyur //

A monk called Nagarjuna, Will know the nature ofthe essenceless object, Realizing the sixth [[[Bodhisattva]]] stage, He will live for six hundred years.

The sixth Bodhisattva stage is called abhimukhd (mngon du gyur pa). The currently available Sanskrit text of the Manjusrimulakalpa devotes one and a halfverses [449c-450] to the prophecy ofNagarjuna. Itreads:85

(caturthevarsasateprdpte nirvrte mayi tathdgate /) ndgdhvayo ndma 'sau bhiksuh (sdsane 'smim hite ratah //) (muditdm bhumilabdhas tu) jived varsasatdni sat // 449 mdyuri ndmatovidyd siddhd (tasya mahdtmanah//) (ndndsdstrdrthadhdtvartham) nihsvabhdvdrthatattvavit//450 Translated several times into Tibetan, the only version that has so far come down to us in its entirety is that from the mid-eleventh century that was prepared by Kumarakalasa and Lo tsa ba Shakya blo gros.86 There the full prophecy reads as follows:87

84 bg,vol.57,no.3081,1555. 85 See Jayaswal (1988: 18), and the Sanskrit text on *35-36. 86 Forthis,seevanderKuijp'sdiscussioninhisForthcomingb,andtheremarkofGsermdog Pan chen (1975a: 176-177). 87 See Bka' 'gyur [[[dpe]] sdur ma], ed. Krung go'i bod rig pa zhib 'jug lte gnas kyi bka' bstan de bzhin gshegs pa nga 'das nas //

lo ni bzhi brgya lon pa na // dge slong klu zhes de 'bod 'byung // bstan pa la ni dad cing phan // rab tudga ba'isathob pa // lo ni drug brgya dag tu 'tsho // rig pa rma bya chen mo ni // bdag nyid chen po des kyang grub // bstan bcos sna tshogs don dang ni // dngos po med pa'i don de rigsa // a Read rig. gang tshe lus de bor ba na // bdeba can du'ang skye bar'gyur// de nas mthar gyis sangs rgyas nyid// nges pa kho nar yang dag thob //

We immediately notice that there are startling differences between the San¬skrit text and the two Tibetan versions, the summary in mrp and this canonical translation. For one, the Sanskrit text of450d: nihsvabhavarthatattvavit corre-sponds nicely to line two ofthe quote in mrp, dngos po medpa'i don de rig //, while the canonical translation appears to be a bit off. For the sixth line, the Sanskrit text [449e] has: muditam bhumilabdhas tu. Here the canonical transla¬tion is accurate, for, itshouldgo without saying, [pra]mudita is the first and not the sixth Bodhisattva stage, as we find in mrp's citation! The prophecy ofNagar- juna in the Lankavatarasutra also has him having attained the first stage, and this is as a matter of course echoed a number of times by Jnanasribhadra who, in his study ofthe Lankavatarasutra, repeatedly refers to himas “the sovereign ofthe first stage” (sa dangpo'i dbangpo) or“the sovereign ofthe first stage who will become a Buddha” (sang rgyas su 'gyur ba'i sa dang po'i dbang po). Not unlike the author ofmrp, Atisa too seems to summarize the verses from the Manjusrimulakalpa when he quotes it in his Ratnakarandodghata:88

dpe sdur khang, vol. 88 (Beijing: Krung go'i bod rig pa dpe skrun khang, 2006), no. 0571, 879-880; see also Obermiller (1932: 111). 88 Miyazaki (2007: 64 and n. 127). dge slongklusgrub cespa 'byung// rang bzhin meddon de nyidrigs// rma bya zhesbya'igzungsbsgrubsnas// lo ni drug brgya dag tu 'tsho //

The Tibetan translation suggests a few differences from the Sanskrit or, perhaps more likely, a reshuffling oftwo verses. The received Sanskrit text [449c, 450a] has nagahvayo namasau bhiksuh for the first line, and mayurl namato vidya siddha for the third. Though fully aware of the quotations of various tantric texts in mrp, Lindtner did not pay full attention to this reference. The quo¬tation of the Manjusrlmulakalpa and other tantric texts may in fact tell us something about the age ofmrp. In this connection, SeyfortRuegg quite rightly pointed out that the passage in which this prophecy occurs in the text men¬tions King Gopala (8th c.), the founder of the Pala dynasty in circa 750, even if he leaves open the very slight posibility that the version of the tantra to which the author ofthe mrp may have had access did not have in its entirety the parade ofprophecies that we find uniquely in this passage in the Tibetan translation ofthe tantra's fifty-third chapter.89 In mrp, the prophecy from the

ManjusrlmulakalpaisprecededbyonefromtheSuvarna[pra]bhasottamasutra and an admonishment to the reader to look into the Lankavatarasutra—this sutra states ofcourse that Nagarjuna would reach the first Bodhisattva stage— and other sutras in which the Buddha had foretold him, and it is followed by a series of quotations from the Mahameghasutra. It is important to note in this connection that, when theyreferto several prophecies ofNagarjuna's coming, neither Candrakirti nor Avalokitavrata cite the Manjusrlmulakalpa, and prob- ablyforgood reason. They either had no access to this portion ofthe tantra or, rather more likely, it had yet to be composed! Bhaviveka on the Store-House Perceptual Awareness (ddyavijndna)

Bhaviveka's criticism of a number of important Yogacara concepts did not sit altogether well withthe exponents ofthis school ofthought. Thus, in his com-mentary on the sixteenth chapter of Aryadeva's (3rd c.) Catuhsataka, his junior

89 Seyfort Ruegg (1990: 61-62). This may be true, because the late tenth century Chinese translation of this work does not have the chapter [53] that contains these prophecies. Foritsdate,seethenowsomewhatdatedessayinMatsunaga(1985:882-894).

contemporaryDharmapala (ca. 530 - ca. 561) directly took him to task on many an occasion, even if he neither mentions him by name nor explicitly refers to any book titles. But this is fairly standard Indian scholarly practice. In this connection, Kajiyama Yuichi — signaled the existence of many parallel

passages in pp and the text ofDharmapala's critiques in his aforesaid commen-tary. This would mean that Dharmapala reacted to pp and, indeed, this finds additional support in Avalokitavrata's pp commentary, in which the sthavira- elder Chos skyong [= Dharmapala] ismentioned,evenifonlyoncebyname, in connection with his criticism of Bhaviveka's interpretation in his pp of the verse of homage that is found at the outset of the Mulamadhyamakakdrikd.90 Onthe otherhand, J.P.Keenanissomewhatmore ambivalent, ifnot muddled, with regard to the object of Dharmapala's reaction in his study and transla¬tion ofthe tenth and last chapter ofDharmapala's work, which comments on Catuhsataka, XVI: 23. By and large ignoring Kajiyama's remarks, he writes at one point that Bhaviveka replied to Dharmapala's criticism in the fifth chapter,

theYogdcdratattvdvatdra,ofmhandtj,whereasheisalsowillingtoconsider at the same time that Dharmapala “. could also have been directly replying toBhavaviveka'sdismissalofYogacarain.”thischapter!91Ofcourse,onecan- not have it both ways. In his introductory remarks, Keenan cites a potentially telling verse that occurs at the end ofDharmapala's work in which, he alleges, Dharmapala used the terms tarka and jvdld, which he then interprets to the effect that this “is perhaps the source for the title” oftj!92 The Chinese text in question reads:

90 See Kajiyama (2005: 200-203). 91 See Keenan (1997: 3, 43). 92 Keenan (1997:46) [= Dharmapala, Guangbailun shilun T. 1571 [XXX] 25oa23- 24]. This stands in contrast towhathewritesinanote[92]tohis translation of this verse on p. 126: “. Thus Dharmapala is responding to a previous critique of Bhavaviveka.”

93 His translation of this verse in Keenan (1997: 126) is also different and for the first two lines he now has: “In order to burn back the fire of unorthodox view points, [[[Aryadeva]]] has covered himself with the balm (turuska) of the Tathagata's true doctrine[.]” Two points: wo Q, “I,” with which the sentence begins, is not accounted for and Hirakawa (1997:1175, no. 3839), lists some five Sanskrit equivalents of su ¡^. As is to be expected, none correspond to turuska! Earlier, Keenan (1997: 46) has: “In order to consume in fire unorthodox viewpoints, I wash in the ghee of the Tathagata's true doctrine . .” Indeed, theChinesetextofthesetwolinescanonlyberenderedaswehaveabove. I, in order to stoke the fire that burns evil doctrines, splash [read po for wo the ghee of the Tathagata's true doctrine [on the fire] and fan the fire withthe stormoflogic;who dares to dart like amoth into the big flame?

We notice that Keenan bases his supposition on his own translations of yin- ming by tarka and mengyan by jvala. Xuanzang, the translator of the Guangbailun shilun consistently uses yinming for hetuvidya and sunsi for tarka, and this throws some doubt on the veracity of Keenan's

translation. But there is an even more serious problem with this interpreta¬tion. Namely, this very same verse also figures as the last verse in the Chinese version of Aryadeva's Sataka, the Guang bai lun, a work that was also translated by Xuanzang.94 This means, ofcourse, that Dharmapala was not the authorof this verse and that, therefore, it cannot be employed in the way Keenan had proposed. In a yet more recent study ofthe relevant portion of Dharmapala's commentary, Hoornaert has not only pointed out that Keenan's translation of the admittedly difficult text of the Guangbailun shilun is often quite mislead¬ing, but also that another, important portion of the Dharmapala-Bhaviveka debate is extant in Wonch'uk's Haesimmilgyong so Let it be said

that Hoornaert's studyis a model of scholarship. Unless we are missing some¬thing (which is quite possible), the Dharmapala-Bhaviveka polemics may tell us something about the internal chronology of Bhaviveka's writings and pp's terminus ad quem, since Dharmapala was aware ofthe fifth chapter of m h/tj as wellaspp, including its twenty-fifth chapter. A final remark: Bhaviveka cites and interprets Catuhsataka, XVI: 23, in pp[t] anent his comment on Mulama- dhyamakakarika, XXII: 11, without indicating any issues with his interpretation of this verse, while Avalokitavrata does indeed signal a polemic against Bha- viveka, albeit witout naming any names.96 Of course, Dharmapala was notthe onlyone to take aim at Bhaviveka. Xuanzang, his intellectual heir, did the same inhis Chengweishilun

94 1.1570 [XXX] 186C28-29. This verse was not translated in G. Tucci's translation and study of this work in Tucci (1923-1925:567). It is also not found in the Tibetan translation of the Catuhsataka.

95 See Hoornaert (2004: 121, 134ff.). Itshouldbe saidthathis treatment ofthe sources is far superior to that ofKeenan. 96 See, respectively, bg, vol. 57, no. 3080, 1345 {= pp[c] 120a24-25}, and bg, vol. 59,no. 3086, 1422 ff. 97 See t. 1585 [XXXI] 16a6-8 [=La Vallée Poussin(1928-1929: Tome 1, 188-189), Moro (2013: 167-175), and now also He Huanhuan Forthcoming.

As we have seen near the beginning of this essay, Wonch'uk explicitly cited a passagefrommh's*Tattvdmrtdvatdrachapterinwhichtheterms*dlayavijndna and *sadvijndna allegedly occurred. He did so to show that Bhaviveka did not accept the Yogacara idea of the store-house perceptual awareness. Be this as it may, the terms dlaya, dlayavijndna or sadvijndna do not occur anywhere in mh[s]. This does not really mean much, since we know that this text shows obvious contaminations in numerous instances. However, neither do we find their Tibetan counterparts, kun gzhi, kun gzhi['i] rnam [par] shes [pa], or rnam [par] shes [pa] drug, in mh[t]. Further, the term kun gzhi (*dlaya) occurs only once in pp[t], namely just after the aforementioned “digression” of its twenty-fifth chapter, and then in a very off-hand, throw-away manner, in a quotation from the Lankdvatdrasutra, III: 48 [= X: 91], where the very existence of the dlaya is [ontically] denied (na ca dlayah).98 We encounter neither *dlayavijndna nor *sadvijndna in hr. In fact, Bhaviveka does not deal with these in any way in anyofthese works.

But it is slightly different with the prose text of tj[t]. The term kun gzhi rnam parshes pa first occurs in tj[t] admh[s,t], I: 1a: andlayam/gzhi mayin, where he quite expresslydeniesthe dlaya/gzhi, thatis, the store-house percep¬tual awareness which, to add more emphasis, he puts on par with the Samkhya idea of the “person” (skyes bu, purusa/ pums).99 Indeed, he took the notion of the purusa quite vehemently to task in the sixth chapter of mh[s,t] and tj[t],100 but did not mention there its functional equivalence to the dlaya- vijndna, and the same holds for the Yogacara chapter where he also did not underscore its functional equivalence to the purusa/pums.Buthedoes suggest there that the idea of the dlayavijndna is on par with non-Buddhist notions of a permanent self (bdag, dtman).101 What is peculiar and hence a tad dis¬concerting is that the former passage is not referred to, even superficially, by any of the interpreters of his oeuvre when they state their positions on how he dealt with the idea ofthe kun gzhi [[[rnam par shes pa]]]. tj[t] admh[s,t], V: 4, 46, 50, 98, contains several passages in which kun gzhi rnam par shes pa is mentioned.102 The term rnam par shes pa drug also occurs rather innocu-

98 See bg, vol. 57, no.3080, 1426, and Eckel (1985: 75).pp[c]does not have the corresponding passage in t. 1566 [XXX], 131b! 99 See bg, vol. 58, no. 3083, 105, and Gokhale and Bahulkar (1985: 83). 100 See He (2013: vol. 2, 424ff.). 101 bg, vol. 58, no. 3083,519; see also Hoornaert (2001:163 [187]), and Saito (2007:236). 102 bg, vol. 58, no. 3083, 486-489, 516, 519, 546-547; see, respectively, Hoornaert (1999: 150 [135-136]), (2001: 161 [184], 163-164 [187-188]), and (2003: 145-146 [163-164]); and Saito (2007: 205-206, 234, 236-237, 263-264).

ously in tj[t] ad mh[s,t], III: 71, IV: 8, VIII: 83[76],103 and none ofthese are particularly eye-opening or relevant for this essay. The first reference to the alayavijnana is purely expository, and Hoornaert has shown that its survey is by and large based on Vasubandhu's Trimsika and Sthiramati's (6th c.) bhasya- commentary.104The other three instances occur in those sections ofthis chap¬ter where the author subjects basic Yogacara doctrines, including the alayavi- jnana, to a critique.

Giventhatnone ofthe interpreters ofhis oeuvre used the outright rejection ofin tj[t] admh[s,t], I:1, did Bhaviveka really reject the ideaofastore-house perceptual awareness, ofan alayavijnana? Although no grounds for their con¬tention is given, Dar ma rgyal mtshan cites some unidentified individual[s]— Tibetan[s] or Indian[s]—whose views implied that Bhaviveka had in fact ac¬cepted the notion ofa store-house perceptual awareness, forhe writes:105

kha1[[[Tibetan]] has the numeral “1” instead ofcig] slob dpon 'ba' bya drima na [= med] pa'iyid dang rnam shes tshogs dgur 'dod zhes zer ro // Some allege: “Master Bhavya accepted, together with the consciousness that is without taint, nine types ofperceptual awareness.” What all this means ofcourse is that these unnamed individuals must have held that Bhavya did in fact accept the alayavijnana, the eighth, because that is a pre- conditionforacceptingtheso-called*amalavijnana,theninthvijnana. Strictly speaking, the termdrima medpa'iyid translates something along the orderof *amalamanas, and its occurrence is attested in a few later tantric tracts. As far aswecantell,thetermdrimamedpa'irnamparshespa(*amalavijnana) isonly found in the Tibetan translation ofWonch'uk's Haesimmilgyong so

[= 'Phagspa dgongspa zab mo ngespar'grelpa'i mdo rgya cher'grelpa]106 In an exhaustive studyofthe notionofamalavijnana and following an earliernotice by Hakamaya Noriaki M. Radich made the observation that the term amalavijnana already occurs in Vasubandhu's Abhidharmakosa, V: 29, and his bhasya-commentary thereon.107 He also argued that metrical considerations had led Vasubandhu to use amalavijnana instead of anasravavijnana in the

103 bg, vol. 58, no. 3083, 181, 366, 652. 104 Hoornaert (1999: 142, n. 1). 105 See, respectively, Dar ma rgyal mtshan (2007a: 133) [= Ibid. 2007b: 148; manuscript: 128a]. 106 bg, vol. 68, no. 3248, 544, and vol. 69, no. 3248, 433. 107 See Radich (2008:48-52).

verse-text, after which he made the terminological substitution in his bhdsya- commentary. Indeed, it was the Indian monk Paramartha (499-569), himself the first biographer of Vasubandhu and well-versed in the Abhidharmakosa and bhdsya, and their first translator, who introduced the term amalavijndna and the idea behind it into the discussion of Buddhist doctrine while he was in China, after which it soon spread far and wide in East Asia. Wonch'uk dis¬cusses this idea in his Haesimmilgyong so and it is his work from which the Tibetans no doubt drew their knowledge ofthe notionofnine types ofvijndna. Of course Wonch'uk nowhere asserts that this was Bhaviveka's idea. Truth be told, we have

noideaofthe origin ofthis unattributed assertion that Bhaviveka wasaware ofthe conceptofamalavijndnaoronwhatbasis itwas made, unless this idea had gained some currency in the Indian subcontinent and someone had passed it on to a Tibetan disciple, which then ultimately entered into Dar ma rgyal mtshan's doxographical treatise. Tsong kha pa and his disciples were quiteadamantthatBhavivekaacceptedarealisticontology(phyidon) and thus did not accept the notion of the store-house perceptual awareness, whereas Gser mdog Pan chen and also Go rams pa Bsod nams seng ge (1429-1489) had somewhatdifferentviewsonthesubject.Theirargumentswillformthesubject ofa separate essay.

Finally, before we conclude, we wish to point to two terminological curiosi¬ties in connection with the writings thatare attributed to Bhaviveka orBhavya or Bhavyakirti, as he also seems to have been known in the subcontinent. In fact, we find the name Bhavyakirti as the author ofpp used in a work on Can- drakirti's Prasannapadd that is associated with circa 1100 Hasumati and Pa tshab Nyi ma grags.108 The first is the occurrence ofthe phrase ltaba'i [']phreng ba(*darsanamdld), “garland ofphilosophical views,” and the numerical deter-mination of these views to the amount of three hundred and sixty-three in tj[t] and the mrp[t]. tj[t] first speaks of these ad mh[s], III: 135, which then prompts the author to give a superficial survey ofthe philosophical sys- tem[s] ofthe other[s] (parasiddhdnta, gzhangyigrub mtha'), as they are called in mh[s,t], III: 136.109 The number three hundred and sixty-three recurs in the second and third chapters ofmrp[t].110 Secondly, both tj[t] admh[s,t], III: 140, and Bhavya[kirti]'s (10th c.) Pradlpoddyotanavydkhydtlkd use the fairly uncommon

108 Hasumati and Pa tshab (2006: 34).
109 See, respectively, bg, vol. 58, no. 3083, 221-225, and Ejima (1980: 300-303).
110 For the first of these sources and this number in particular, see bg, vol. 57, no. 3081, 1495, and 1506; for the latter, see Lindtner (1986: 187, 191, n. 21), and also Eckel (2008: 28ff.).
expression (tshad ma’i lam, *pramanamarga),m whereby of minor interest is the fact that Kamalasila also appears to have used it once in his Madhya- makaloka.112


Conclusions


What can be concluded, ifanything, from the foregoing? We began this essay with a quotation from the seventh century Korean scholar Wonch'uk who cited a passage that ostensibly derived from the *Tattvamrtavatara chapter of a work that was titled Madhyamakahrdaya. The cited passage presented us with several problems. First, it was evidently written in prose. The cur¬rently available Madhyamakahrdaya texts, mh[s] and mh[t], are written in verse. Second, it mentions the alayavijnana, a word that does not occur in anyofthe Madhyamakahrdaya texts, although we do find itmentioned in the first and fifth chapters of the Tarkajvala, the latter of which deals with the tenets ofthe Yogacara school. Without stating that it was part ofthe Madhya- makahrdaya, Bhaviveka himself apparently refers to his *Tattvamrtavatara in the *Hastaratna, which he composed in a “mixed” style of verse

and prose. There he refers to the text for further details on the notions of tathata and paratantra.Wewriteandunderscore“apparently,”becausehedoesnotsaythat the Tattvamrtavatara to which he referred consisted of a chapter from what may have become a larger work with the title Madhyamakahrdaya. Indeed, according to mh[s], *Tattvamrtavatara was the title of its third chapter, an identification that is absent from mh[t]. It seems very likely that the Sanskrit manuscript on which mh[t] is based, that is, actually the Sanskrit

manuscript on which tj[t] is based in which mh[t] is cited verse by verse, is of a later recension than mh[s]. But there are problems. For example, the contents of the passage cited by Wonch'uk are found at best in mh[s], V, and only one of thetwo citations ofthe Tattvamrtavatara in the *Hastaratna is found in mh[s], III; the other is again retrievable from mh[s], V. From these data, we maycon¬jecture that the Tattvamrtavatara was Bhaviveka's earliest study. It may have briefly circulated as a separate work until it was mined by Bhaviveka himself for the composition ofthe Madhyamakahrdaya, but not before he composed the *Hastaratna! There are many possible sources for his use ofthe word amrta


111 See, respectively, bg, vol. 16, no. 0645, 711, and bg, vol. 58, no. 3083, 227. For the latter, see Watanabe (1988: 130, 137).
112 bg, vol. 62, no. 3116, 1324.
in the title Tattvamrtavatara. It might conceivably be an echo of Mulama- dhyamakakarika, XVIII: 11, where Nagarjuna states anent the Buddha's teach- ing:113

anekartham ananartham anucchedam asasvatam / etat tal lokanathanam buddhanam sasanamrtam //
Without identity, without distinction, not annihilated, not persisting,
The protectors ofthe world, the Buddhas, taught this immortal teaching.


Regrettably, Bhaviveka has nothing to sayaboutthisverse inhis Prajnapradlpa that is of immediate relevance to this paper, not even anent sasanamrtam (bstan pa bdud rtsi)!114 All the evidence points to the factthatthe *Tattvamrta- vatara was a work that was still intact when Bhaviveka wrote the *Hastaratna, which he composed in a mixture of prose and verse. We speculate that he thereafter incorporated its main points in the third and fifth chapters of the larger work that we now know, albeit not necessarily correctly, as the

Madhyamakahrdaya, a treatise that he composed in mnemonic verse on which he ultimately wrote his own commentary, the Tarkajvala. It is no easy mat¬ter to determine where we should place the Prajnapradlpa in his intellectual development. Suffice it for now to say that the Tibetan and Chinese transla¬tions ofthis work suggest that we must reckon with two different recensions. Itis onlythe Tibetan recension in which the title Madhyamakahrdaya is cited, and Bhaviveka's commentator Avalokitavrata had no doubt access to the same recension ofthe Prajnapradlpa that was translated into Tibetan. Except for his mention of the title Madhyamakahrdaya anent the digression in the twenty¬fifth chapter ofthe

Prajnapradlpa, Avalokitavrata also nowhere explicitly cites it or the Tarkajvala or, for that matter, the *Hastaratna. The various digressions in both versions ofthe Prajnapradlpa point to interpolations and to a certain degree of textual instability and deviation from the original version. It will
113 See Ye (2011: 306). We differ slightly from the translation in Garfield (1995: 253).


114 See bg, vol. 57, no. 3080, 1278-1279; pp[c] shows some interesting differences in the corre¬sponding passage in T. 1566 [XXX] 108c11ff. In connection with his comments, Bhaviveka also cites the verse: de nyid shes pas 'di la ni // 'dod chags bral ba ma thob pa // tshe rabs gzhan la 'bad medpar // nges par thob ste lasbzhin no // [“One who has not [yet] attained freedom from cupidity-attachment through understanding what there is (tattvajnana), willcertainlyattainiteffortlesslyinanotherlife-time;inaccordancewiththeworkingsof karma.”], which Buddhapalita had identified earlier, in bg, vol. 57, no. 3079, 657, as reflect¬ing Aryadeva['s Catuhsataka, VIII: 22].
be important to compare the substance ofthese digressions with Bhaviveka's othermajorwritings,aprojectthatisnowunderway.Aconsensusofsortshas been reached in which it is held that Bhaviveka first composed the Madhya- makahrdaya, then the Prajndpradlpa, and then the Tarkajvdld. The cumulative

evidenceprovidedinthisessaysuggestsasomewhatdifferentscenario,namely, that he first wrote the *Tattvdmrtdvatdra and then the *Hastaratna. These two treatises, we nowbegin to speculate even further, were followed by the Madhya- makahrdaya, in which portions ofthe *Tattvdmrtdvatdra, or perhaps this work in its entirety, were incorporated. The Madhyamakahrdaya is obviously earlier than at least a portion ofthe work that is now known as the Tarkajvdld. And we now somewhat impatiently await the retrieval and eventual publication of the Sanskrit manuscript of the Madhyamakahrdaya that has been located in the Potala, for it may shed light on this tract's compositional history. The Pra-

jndpradlpa, the only work of Bhaviveka that is an exegesis of another treatise, was composed in their wake. But we do not know when the aforementioned “digressions” were inserted in the differently filliated manuscripts that were ultimately rendered into Chinese and Tibetan. Kajiyama has unambiguously shown that Dharmapala takes issue with a passage from the “digression” at the end of the Tibetan translation of the Prajndpradlpa’s twenty-fifth chap- ter.115 Since Dharmapala was obviously familiar with it, we can safely say that it must have been inserted by someone in a version of the text during Bha- viveka's lifetime, if not by the master himself. As was indicated, this “digres¬sion” is absent from the 630-632 Chinese translation of the Prajndpradlpa, as is, for example, the one in the fifth chapter of the Tibetan translation.116 But other such “digressions” can be found in both translations.117 Finally, we have so far found no evidence that one ofhis treatises was written in response to a criticism of an earlier work of his, specifically, the one associated with Dharmapala.

115 See Kajiyama (2005: 202); see also Hoornaert (2004: 135).
116 bg, vol. 57, no. 3080,1027, adMulamadhyamakakanka, V: 3, and 1.1566 [XXX] 71C12.
117 See, for example, the notice of a “digression” (zharla bshadpas chog gi/ skabs 'dir bsdu

banbya ste) ad Mulamadhyamakakanika, VII: 16, of pp[t] in bg, vol. 57, no. 3080, 1068, which well corresponds to the passage ad in 1.1566 [XXX] 77bii.
For a useful listing of where the Tibetan and Chinese translations vary, see the chart in Akahane (2013: 1183).


Appendix: The Names of Bhaviveka


We know for a fact that the Tibetans were familiar with the name Bhavya during the earliestphases ofBuddhism's introduction in Tibet. For example, itoccurs in all the Tanjur recensions of Sna nam Ye shes sde's (ca. 800) Lta ba'i khyad par—the extant manuscript from Dunhuang has “ 'Ba' phyva” and “ 'Ba' phya”— and in all versions of this work he is identified as the author of both pp and mh.118 But when the names quoted in these early sources return in the later recensions ofthese verysame texts, theyare notas stable asone mightexpect, and there are many instances in which they were evidently changed by some eagerorover-eagereditor. Acase inpoint thatis relevant for our presentexer¬cise is found in Dbus pa Blo gsal's tract on the philosophical systems. There he quotes a passage from Ye shes sde's text in which he or someone else had evidently substituted “Bha byas / 'Ba'

phya” by Legs ldan 'byed and “Sha nta ra ksi ta / Shan ta rag si ta” by Zhi ba 'tsho.119 While the Mahdvyutpatti is to be sure anearlysource for equating Sanskrit bhavyawith Tibetan skal ldan,120 there is perhaps a unique case, more or less contemporaneous with this work and Ye shes sde's little tract, where skal ldan [= bhavya] is equated with snang bral [= bhdviveka/bhdvivikta]—bhavya has the sense of the “fortunate one” and seems to be a nickname that may very well have been playfully and pun- ningly derived from bhdviveka. The unique instance in question is Kamalasila's Madhyamakdlamkdrapanjikd. Only available ina circa 800 Tibetan translation that was prepared by a consortium of scholars including Ye shes sde, all the availableTanjurwitnessesofthisworkrefertwiceto whattheycallmh, onceas having been authored by Snang bral and, aswe have alreadyseen, once as hav¬ing been

written by Skal ldan.121 We have to wonderwhyhisname should have beengiventwodistinctTibetanrenditions,andtheonlyanswerwecanthinkof at the moment is that Kamalasila had used two different Sanskrit expressions forhisname.122Ofthetworeferencesinquestion,thefirstisquiteexplicitlymh,


118 Seyfort Ruegg (1981: 214, 217).
119 Mimaki (19: 172, n. 466).
120 Sakaki (1962: 243, no. 3495 [= IshihamaandFukuda(1989:176,no.3493)]).
121 Ichigo (1985:232-233,290-291).
122 Compare here also, for example, the *Bodhimdrgadlpapanjikd in bg, vol. 64, no. 3178, 1754—this work has been wrongly attributed to Atisa—where all prints suggest that the authorusedthe very odd construction bha bya snang bral, as ifthatwere one name. But weshouldprobablyinterpretthisjuxtapositionasanexplicativeappositionorpossiblyin the sense that snang bral was an earlier gloss on bha bya which an unknown editor later incorporated into the text.


III: * 12, the full text of which Santaraksita had already included in his Madhya- makalamkara[[[sva]]]vrtti ad Madhyamakalamkara, 70, albeit without signaling that itwas a citation from mh—and the second is mh[t], V: 28c-d. We wrote mh,III:*12,because this verse isnotincludedinmh[s],althoughitisfoundin mh[t] and tj[t];itreads:123
tathyasamvrtisopanam antarena vipascitah // tattvaprasadasikhararohanam na hiyujyate /

yang dag kun rdzob rnams kyi skas // medparyang dag khang payi // steng du gro barbya bani// mkhaslarungba ma yin no//
This verse is followed bymh[s,t], III: 12:

tattvaprasadasikhararohanam na hi yujyate / tathyasamvrtisopanam antarena yatas tatah //
de nyidkhangbzang[s] thog dzegpa//
yang dag kun rdzob them pa i skas // medparrungbama yin pa // ganggiyin pa [phyirba] dephyirro //
Both verses seamlessly follow one another in tj[t], but only mh[t], III: 12 is commented on. A superficial search through the comparative Tanjur edi¬tion [= bg], courtesy oftbrc.org, reveals that verse mh, III: *12, occurs in the *Saptangasadhana by Vagisvarakirti (11th c.), where it is expressly quoted, and in Atisa's Satyadvayavatara, where it is not.124 Neither states that this verse was taken from mh or tj.

It is a bit surprising that neither Snang bral nor Skal ldan nor Legs ldan 'byed or Bha bya figure even once in Kamalasila's Tattvaloka, in his Sarvadha- rmasvabhavasiddhi, or in his Madhyamakaloka, which, after all, are his prin¬cipal contributions to Madhyamaka thought, all of which are only extant in
123 Ichigo (1985:232); see also bg, vol. 58, no. 3082, 9, bg, vol. 58, no. 3083,141-142, and Ejima (1980a: 271), who noted that Haribhadra (ca. 780) cited this verse in his Aloka. However, Haribhadradidnotidentifyhissource.

124 See bg, vol. 22, no. 791, 1095, and vol. 63, no. 3131, 1052. For the latter, see now also Apple (2013: 314-315).
Tibetan translations. Less surprising, of course, is that Candrakirti is wholly off the radar screen, since his Madhyamaka writings appeared to have “gone

underground”shortlyaftertheywerepennedandonlyseemto have resurfaced in the eleventh century. However, ifhe were notthe firstthen he was certainly among the very first Indian Buddhist intellectuals to cite mh, and he did so in his Prasannapadd.125 Curiously, he neitherprovided the title ofthe workfrom which he drew these quotations nor the name of its author. And this is a tri¬fle strange, since he does mention Bhaviveka three times by name in his same Prasannapadd, and then in a critical fashion.

We believe the jury is still out on this, but Kamalasila may have written the Madhyamakdloka posterior to the Madhyamakdlamkdrapanjikd and, as a more comprehensive work on its specific brand ofMadhyamaka thought, one might be inclined to think that he would have wished to include Bhaviveka's [and Candrakirti's] views in his analyses. He evidently did not. We doubt that he was the first to have done so, but Tsong kha pa did conjecture on philo¬sophical grounds that Kamalasila's panjikd commentary was written while he was still “intellectually incomplete” (thugs ma rdzogs pa) and thus prior to the more “mature”Madhyamakdloka.126 He prefaces his conjecture by stating that Dharmamitra had asserted (bzhed)

that Kamalasila had written this work even if he, Tsong kha pa, felt that there were many inconsistencies between it and the Madhyamakdloka. And he added that Chos kyi bshes gnyen [Dharmami- tra] (9th c.) also had claimed that Kamalasila was the author ofthe Madhya- makdlamkdrapanjikd, but he does not saywhere orinwhattreatise.To be sure, the Dharmamitra in question was the author of the Prasphutapadd study of histeacherHaribhadra'sSphutdrthd-commentaryonthe Abhisamaydlamkdra, anditistherethathe does indeed cite atreatise by Kamalasila, which, thecita- tionclearlyindicates,wasnoneotherthanthe Madhyamakdlamkdrapanjikd.127 This Dharmamitra also refersto Skal ldan [= Bhavya] inthe thirdchapterofhis work, where he rather uninformatively writes that the latter and others “had mainly dealt with what is all-pervasively obscured” (... kun rdzob pa gtsor brjod payin ...).128


125 See,forexample, Seyfort Ruegg (2002: 63), where Candrakirti cites mh[s], III: 41a-b, and 27a-b.
126 SeeTsongkhapa(1981:vol.Ba,792).SeealsoBlumenthal(2004:348-349,n.3),albeitwith reservations about his translation and the integrity ofthe Tibetan text he subjoined to it.
127 bg, vol. 62, no. 3115, 923, where reference is made to the passage ofthe Panjikd that can be found in Ichigo (1985: 21).
128 bg, vol. 62, no. 3115, 951.

Finally, Bhaviveka is quoted or referred to severally in eleventh century Indian Buddhist sources whose authors hailed from north-central India as well as [one, so far] from Kashmir. Thus Bodhibhadra, one ofAtisa's teachers, refers to him twice in his Jnanasarasamuccayanibandhana.129 Via an obser¬vation made by P.K. S0rensen, Lindtner indicated that, though attributed to an Aryadeva, the Madhyamakabhramaghata includes a quotation of inter alia mh[s,t], I: 1-3.130 The identity and floruit of this Aryadeva remains obscure. Atisa and Nag tsho Lo tsa ba rendered this little work into Tibetan while in Nalanda at the behestofthe “King ofthe World” *Sukhacarya. This means that they translated it while they were still in the subcontinent, so that it may date from circa 1040. Atisa, too, cites TJ twice in his Ratnakarandodghata, once ad mh[s,t], II: 5, and once ad mh[s,t], II: 36.131 Lastly, Jnanasribhadra (ca. 1050) also briefly mentions Bhaviveka in his large and rewarding commentary on the Lankavatarasutra albeitin connection witha passage thatis notentirelyclear to us.132


Acknowledgments


We should here like to express our appreciation to the Tibetan Buddhist Re¬source Center(tbrc.org) for many things, but especially for the inclusion ofthe searchable Bka gyur [[[dpe bsdur ma]]] and Bstan gyur [[[dpe bsdur ma]]], that is, bg, among its e-texts. This has greatly facilitated the writing ofour essay. The authors should also like to thank Dr. V. Eltschinger for having read ever so care¬fully through this essay and for having kindly indicated a number ofoversights and errors that have now been corrected. Any that remain are, well, our respon¬sibility.

Abbreviations

Works
hr *Hastaratna
mh Madhyamakahrdaya
129 bg, vol. 57, no. 3079, 897.
130 Lindtner (1982: 173).
131 See Miyazaki (2007: 41-42).
132 See Hadano (1983: 231) and bg, vol. 69, no. 3250, 1100.
pp Prajnapradlpa
mrp *Madhyamakaratnapradlpa
tj Tarkajvala

Texts

hr[c] *[[[Mahayana]]]Hastaratna (Chinese translation) mh[s] Madhyamakahrdaya (Sanskrit)
mh[t] Madhyamakahrdaya (Tibetan translation) mrp[t] *Madhyamakaratnapradlpa (Tibetan translation) pp[c] Prajnapradlpa (Chinese translation) pp[t] Prajnapradlpa (Tibetan translation) tj[t] Tarkajvala (Tibetan translation)


References

Abbreviations

bg Bstan 'gyur [[[dpe bsdur ma]]], ed. Krung go'i bod rig pa zhib 'jug lte gnas
kyi bka' bstan dpe sdur khang (Beijing: Krung go'i bod kyi shes rig dpe skrun khang, 2003), vols. 124.
Sdz Shinsan damihonzokuzökyö, Ed. Kawamura Köshö
(Tokyo: Kokusho kanköku 1975-1989), vols. 90.
t Taishöshinshö daizökyö Ed. Takakusu Junjirö
and Watanabe Kaikyoku (Tokyo: Taishö issaikyö kankö-
kai 1924-1932), vols. 85.
Chinese Canonical Works
Äryadeva. Guang bai lun 1.1570 [XXX]: 182a-187a.
Bhäviveka. Panruodeng SSSWT [Prajnapradlpa]. 1.1566 [XXX]: 50C-135C.
Bhäviveka. Zhangzhen [[[lun]]] 'y'iKÉ [*Hastaratna]. 1.1578 [XXX]: 268a-278b.
Dharmapäla. Guangbailun shilun 1.1571 [XXX]: 187a-25ob.
Huaiyuan ®S. Lengyanjingyishushiyaochao SDZ. 267 [XI]: 79a-
164b.
Huizhao Ä)§. Chengweishilun liaoyizheng 1.1832 [XXXXIII]: 659a-
810b.
Taixian ^R. Chengweishilun xueji SDZ. 818 [L]: 27a-126c.
Wönch'ük HW [ = Yuance]. Haesimmilgyöng so SDZ. 369 [XXI]: 171a-405c.
Xuanzang and Bianji ^^. Datang xiyuji AWWMæ. t. 2087 [LI]: 867b-947b.
Xuanzang k/k; and Bianji ^^. Cheng weishi lun bkkkikùk. 1.1585 [XXXI]: 1a-60a.
Yijing O?. 2004. Nanhaijigui neifazhuanjiaozhu [An Annotated
Edition of A Record of Buddhism en Route from the Southern Seas]. Ed. Wang Bangwei
(Beijing: Zhonghua shuju i^№).
Tibetan Canonical Works
Aryadeva (?II). Jñanasarasamuccayanibandhana. bg, vol. 57, no. 3079: 856-903.
Atisa. Satyadvayavatara. BG, vol. 63, no. 3131:1050-1057.
Atisa(?). Bodhimárgadlpapañjika. BG, vol. 64, no. 3178:1650-1783.
Avalokitavrata. Praftapradipatika. BG, vol. 58, no. 3086:859-1622; vol. 59.
Bhaviveka. Prajñápradipa. BG, vol. 57, no. 3080: 905-1486.
Bhaviveka. Madhyamakahrdayakarika. BG, vol. 58, no. 3082:1-103.
Bhaviveka. Tarkajvala. bg, vol. 58, no. 3083: 104-850.
Bhaviveka(?). Madhyamakaratnapradlpa. BG, vol. 58, no. 3082:1487-1567.
Bhavyakirti. Pradlpodyotanavyakhyatlka. BG, vol. 16, no. 0645:563-1326; vol. 17,1-326. Buddhapalita. Buddhapalita-mülamadhyamakavrtti. bg, vol. 57, no. 3069: 443-764. Candrahari pa. Madhyamaka-Ratnamala. bg, vol. 63, no. 3130: 1035-1049.
Kamalasila. Madhyamakalamkarapañjika. bg, vol. 62, no. 3115: 981-1113.
Kamalasila. Madhyamakaloka. bg, vol. 62, no. 3116: 1114-1403.
Mañjusñmülakalpa. Bka’ ’gyur [[[dpe]] sdurma]. Ed. Krung go'i bod rig pa zhib 'jug lte gnas kyi bka' bstan dpe sdur khang, vol. 88 ( Beijing: Krung go'i bod rig pa dpe skrun khang, 2006), no. 0571: 534-1051.
Sahajavajra. Sthitisamuccaya. bg, vol. 26, no. 1132: 253-272.
Vagisvarakirti. Saptangasadhana. BG, vol. 22, no. 0791:1086-1121.
Wónch'úk. 'Phags pa dgongs pa zab mo nges par 'grel pa'i mdo rgya cher 'grel pa. bg, vol. 68, no. 3248; vol. 69, 1-475.
Tibetan Non-Canonical Works
Urga Chos rje Ngag dbang dpal ldan. 1983. Grub mtha' chen mo'i mchan 'grel dka' gnad mdudgrolblogsalgcesnorzhesbyabalasdbumathalranggiskabs.CollectedWorks, vol.5 (New Delhi): 1-401.
Karma pa III Rang byung rdo rje. 2006. Rje rang byung rdo rje'i thugs dam bstan 'gyurgyi dkar chag. Collected Works, vol. Nga (Lhasa): 427-606.
KarmapaIIIRangbyungrdorje.2006a.Bstanbcos'gyurro'tshalgyidkarchag,Collected Works, vol. Nga (Lhasa): 607-730.
Dkar chag 'phang thang ma [and the Sgrasbyorbampo gnyispa]. 2003. Ed. Bodljongs rten rdzas bshams mdzod khang (Beijing: Mi rigs dpe skrun khang).
Skyo ston Smon lam tshul khrims. Mchims nam mkha'grags kyi rnam thar. *Snar thang gser 'phreng. dbu can manuscript. tbrc.org, W2CZ7888: 50 fols.
'Gos Lo tsabaGzhon nudpal. 2003. 'Gos Lo tsa ba Gzhon nu dpal's Commentary on the Ratnagotravibhagavyakhya. Ed. Kl.-D. Mathes. Nepal Research Centre Publications, No. 24. (Stuttgart: Franz Steiner Verlag).
Ngor chen Kun dga' bzang po. 1999?. Bstan bcos gyur ro tshal gyi dkar chag thub bstan rgyas pai nyi ’od. Collected Works, vol. 4 (Dehra Dun: Sa skya Centre): 573-609.
Ngor chen Kun dga' bzang po. 2010. Ibid. Evam bka’ ’bum 7/20. Mes poi shul bzhag, vol.138.Ed.Dpalbrtsegsbodyigdpe rnying zhib 'jug khang (Beijing: Krung go'i bod rig pa dpe skrun khang): 270-304.
Mnga'risChosrjePhyogslasrnamrgyal.2010.Bstanbcos gyurro tshalgyi dkarchag dri med odkyi phreng ba. Jo nang dpe tshogs, vol. 23 (Beijing: Mi rigs dpe skrun khang): 1-177.
'ChadkhabaYe shesrdo rje. 2007. Grubmtha chenmo. Bka gdamsgsung bumphyogs bsgrigs, vol. 11. Ed. Dpal brtsegs bod yig dpe rnying zhib 'jug khang (Chengdu: Si khron dpe skrun tshogs pa and Si khron mi rigs dpe skrun khang): 225-253.
'Jam dbyangs bzhad pa'i rdo rje I Ngag dbangbrston 'grus. Grub mtha i rnam bshadrang gzhan grub mtha kun dang zab don mchog tu gsal ba kun bzang zhing gi nyi ma lung rigs rgya mtsho skye dgu irebakunskong [[[zhes bya ba]] thal rang gi skabs]. Bla brang bkra shis 'khyil print, CollectedWorks, vol.Pha[15]: 613-1091.
Dar ma rgyal mtshan, alias Bcom ldan ral gri. 2007. Bstan pa rgyan gyi me tog. Collected Works, vol. 1 (Kathmandu: Sa skya rgyal yongs gsung rab slob gnyer khang): 57-101.
Darmargyalmtshan, aliasBcomldanralgri. Ibid. dbu medmanuscript: 31fols.
Dar ma rgyal mtshan, alias Bcom ldan ral gri. (2007a). Grub pa'i mtha rnam par dbye ba rgyan gyi me tog. Bka gdams gsung bum phyogs bsgrigs, vol. 53. Ed. Dpal brtsegs bod yig dpe rnying zhib 'jug khang (Chengdu: Si khron dpe skrun tshogs pa and Si khron mi rigs dpe skrun khang): 9-157.
Dar ma rgyal mtshan, alias Bcom ldan ral gri. (2007b). Ibid. Collected Works, vol. 2 (Kathmandu: Sa skya rgyal yongs gsung rab slob gnyer khang): 1-177.
Dar ma rgyal mtshan, alias Bcom ldan ral gri. Ibid. dbu medmanuscript: 145 fols.
Phya pa Chos kyi sengge. 2007. Bde bar gshegs padang phyirolpa'igzhung rnam par byed pa. Bka gdams gsung bum phyogs bsgrigs, vol. 9. Ed. Dpal brtsegs bod yig dpe rnying zhib'jug khang (Chengdu: Si khron dpe skrun tshogs pa and Si khron mi rigs dpe skrun khang): 7-73.
BustonRinchengrub.1971. Bdebargshegspa'ibstanpa'igsalbyedchoskyi byung gnas gsung rab rin po che i mdzod. The Collected Works of Bu ston [and Sgra tshad pa]. Lhasa Zhol print, part 24 (New Delhi: International Academy of Indian Culture): 633-1055.
BustonRinchengrub.2008.Ibid.Vol.24/28/3.Ed.Dpalbrtsegsbodyigdpernyingzhib 'jug khang (Beijing: Krung go'i bod rig pa dpe skrun khang): 807-1414.
Bu ston Rin chen grub. 1971a. Bstan gyurgyi dkar chagyidbzhin norbudbang gi rgyal po i phreng ba. The Collected Works ofBu ston [and Sgra tshad pa]. Lhasa Zhol print, part26 (New Delhi: International Academy of Indian Culture): 401-644.
Bu ston Rin chen grub. 2008a. Ibid. Vol. 26/28/3. Ed. Dpal brtsegs bod yig dpe rnying zhib 'jug khang (Beijing: Krung go'i bod rig pa dpe skrun khang): 569-896.
Bo dong Pan chen Phyogs las rnam rgyal. 1969. Grangs can pa dang bye brag pal tshad ma'i rnam gzhag gtan la dbab pa. Encyclopedia Tibetica. The Collected Works of Bo dong Pan chen Phyogs las rnam rgyal. Ed. S.T. Kazi, vol. 6 (New Delhi: Tibet House): 611-675, 675-712.
BodongPanchenPhyogslasrnamrgyal.1969a.Rigbyedmtha pa i tshad ma dang spyod [read: dpyod] papa'itshadma'irnamgzhaggtan ladbab pa. EncyclopediaTibetica. The Collected Works of Bo dong Pan chen Phyogs las rnam rgyal. Ed. S.T. Kazi, vol. 7 (New Delhi: Tibet House): 2-85, 85-297.
Byang chub rgyal mtshan et. al. 1971. Bstan bcos gyur ro tshal gyi dkar chag yid bzhin gyi nor bu rin po che i za ma tog. The Collected Works of Bu ston [and Sgra tshad pa] [[[Lhasa]] print], part 28 (New Delhi: International Academy of Indian Culture): 343-573.
Byang chub rgyal mtshan et. al. Ibid., 2008. vol. 28/28/3. Ed. Dpal brtsegs bod yig dpe rnying zhib 'jug khang (Beijing: Krung go'i bod rig pa dpe skrun khang): 453-773.
Dbus pa Blo gsal Rtsod pa'i seng ge. Bstan bcos gyi dkar chag. dbu med manuscript. tbrc.org, W2CZ7507: 80 fols.
Tsong kha pa Blo bzang grags pa'i dpal. Drang ba dang nges pa'i don rnam par phye ba'i bstan bcos legs bshad snying po. Collected Works [[[Lhasa]] Zhol print], vol. Pha (Dharamsala: Library of Tibetan Works and Archives, 1981): 443-669.
TsongkhapaBlobzanggragspa'idpal.Dbumargyangyizinbris(rjeranggisgnangba), Collected Works [[[Lhasa]] Zhol print], vol. Ba (Dharamsala: Library of Tibetan Works and Archives, 1981): 783-814.
Tsong kha pa Blo bzang grags pa'i dpal. Yid dang kun gzhi'i dka ba ignas rgya cher grel pa, Collected Works [[[Lhasa]] Zhol print], vol. Tsha (Dharamsala: Library of Tibetan Works and Archives, 1981): 629-742.
Zhwa dmar IV Chos grags ye shes. Dpal ldan bla ma dam pa mkhan chen thams cad mkhyenpadongyisladdumtshannassmostegzhonnudpalgyirnampartharpayon tan rin po che mchog tu rgyas pa'iljon pa. dbu can manuscript. tbrc.org, W1PD90154: 72 fols.
Zhwa dmar IV Chos grags ye shes. 2009. Collected Works-Gsung bum, vol. 1. Ed. Yangs can dgon ris med dpe rnying myur skyob khang nas bsgrigs (Beijing: Krung go'i bod rig pa dpe skrun khang): 483-588.
Se raRje btsunChoskyi rgyal mtshan. 2008. Dbumala jug pa'i rnam bshad dgongs pa rab gsalgyi dka gnad rnams gsal bar byedpa skal bzang mgul rgyan. Se rarje btsun Choskyirgyalmtshangyigsungpodlngapa.Ed.'Jigsmedbsamgrub(Beijing:Krung go'i bod rig pa dpe skrun khang).
Gser mdog Pan chen Shäkya mchog ldan. 1975. Dbu ma rnam par nges pal bang mdzod lung rigs pa i rgya mtsho [las dbu ma thal rang gi gyes tshams dang grub mtha'i gnas bstan pa le u gnyis pa]. Collected Works, vol. 14 (Thimphu: Kunzang Tobgey): 413-563.
Gser mdog Pan chen Shakya mchog ldan. 1975a. Shes rab kyi pha rol tu phyin pa’i man ngag gi bstan bcos mngon par rtogs pa'i rgyan 'grel pa dang bcas pa'i snga phyi'i 'brel rnam par btsal zhing / dngos bstan kyi dka' ba'i gnas la legs par bshad pa'i dpung tshogsrnamparbkodpa/bzhedtshulrbarlabskyiphrengba.CollectedWorks,vol.11 (Thimphu: Kunzang Tobgey): 157-587.
Hasumati/Pa tshab Nyi ma grags. 2006. *Tshig gsal ba'i dka' ba bshad pa. Bka' gdams gsung 'bum phyogs bsgrigs, vol. 11. Ed. Dpal brtsegs bod yig dpe rnying zhib 'jug khang (Chengdu: Si khron dpe skrun tshogs pa / Si khron mi rigs dpe skrun khang): 29-203.
Secondary Sources
Akahane Ritsu ^33W. 2012. “An Examination of the Digression in Chapter 3 of Pra- jndpradlpa.” China Tibetology 2: 8-25.
Akahane Ritsu ^33W. 2013. “On the Digressions of the Prajndpradlpa with a Reevalu¬ation of its Chinese Translation.” Indogaku Bukkyogaku Kenkyu LX: 1182-1188.
Ames, W. 1993. “Bhavaviveka's Prajndpradlpa: A Translation of Chapter One, ‘Exam¬ination of Causal Conditions' (pratyaya).” Journal of Indian Philosophy 21: 209¬259.
Apple, J.B. 2013. “An early Kadam (bka’ gdams) Commentary on Atisa’s Satyadvayd- vatdra.” Journal of Indian Philosophy 41: 263-329.
Bagchi, P.C. 2011. “A Buddhist Monk of Nalanda amongst the Western Turks.” India and China. Interactions through Buddhism and Diplomacy. A Collection of Essays by ProfessorPrabodh Chandra Bagchi. Ed. Wang Bangwei and Tansen Sen (New Delhi: Anthem Press): 105-108.
Bahulkar, S.S. 1994. “The Madhyamaka-Hrdaya-Kdrikd ofBhavaviveka: A Photographic Reproduction of Prof. V.V. Gokhale's Copy.” Nagoya Studies in Indian Culture and Buddhism Sambhdsd 15: i-iv, 1-49.
Blumenthal, J. 2004. The Ornament of the Middle Way. A Study of the Madhyamaka ThoughtofSdntaraksita (Ithaca: Snow Lion Publications).
Eckel, M.D. 1985. “Bhavaviveka's Critique ofYogacaraPhilosophyinChapterXXVofthe Prajndpradlpa.” Miscellanea Buddhica. Ed. Chr. Lindtner, Indiske Studier 5 (Copen-hagen: Akademisk Forlag): 25-75.
Eckel, M.D. 1992. To See the Buddha. APhilosopher's Questfor the Meaning ofEmptiness (Princeton: Princeton University Press).
Eckel,M.D.2008.BhdvivekaandHisBuddhistOpponents.HarvardOrientalSeries,vol.70 (Cambridge: Department ofSanskrit and Indian Studies, Harvard University).
Ejima Yasunori 1980. “Madhyamakaratnapradlpa ni tsuite, Madhyamakara-
tnapradlpa ¡¿Dtt [On the Madhyamakaratnapradlpa]” Indogaku Bukkyogaku Kenkyu XXVIII: 950-947.
Ejima Yasunori SMW^. 1980a. Chugan shisö no tenkai—Bhâvaviveka kenkyu Bhâvaviveka WS [An Exposition ofMadhyamaka Thought. A Study of Bhâvaviveka] (Tokyo: Shunjusha ^Wtt).
Ejima Yasunori 1990. “Bhävaviveka/Bhavya/Bhäviveka [in Japanese].”Indo-
gaku Bukkyögaku Kenkyu EPS^W^^WS XXXVIII: 98-106.
Eltschinger, V. 1998. “Bhâvaviveka et Dharmakïrti sur âgama et contre la Mimämsä.” Asiatische Studien/ÉtudesAsiatiques 52: 57-84.
Frauwallner, E. 1959. “Dignäga, Sein Werk und Seine Entwicklung.” Wiener Zeitschrift für die Kunde Süd- und Ostasiens III: 83-164.
Garfield, J.L. 1995. The Fundamental Wisdom of the Middle Way (New York: Oxford University Press).
Gokhale, V.V. 1958. “The Vedänta-Philosophy Described by Bhavya in the Madhya- makahrdaya”Indo-IranianJournal II: 165-180.
Gokhale, V.V. 1972. “The Second Chapter of Bhavya's Madhyamakahrdaya”Indo-Iranian Journal XIV: 40-45.
Gokhale, V.V. with S.S. Bahulkar. 1985. “Madhyamakahrdayakârikâ Tarkajvâlâ, Chap¬ter I.” Miscellananea Buddhica. Ed. Chr. Lindtner. Indiske Studier 5 (Copenhagen: Akademisk Forlag): 37-62.
Hadano Hakuyu SfflWSK et al. Ed. 1983. Jnänasnbhadra, Lankâvatârasütravrtti. (Sendai: Tibetan Buddhist Text Society).
Hatani Ryôtai SæTS. Tr. 1932. “ ASSJSK [Daijö shöchinron].”KokuyakuIssaikyö. Chûkan-bu 3 HS—(Tokyo: Daito Publishing Co., Inc.
99-138.
He Huanhuan MRR- 2013. Zhongguan xinlunjiqi guzhu Sizeyanyanjiu {{
[A Study of the Madhyamakahrdayakârikâ and the Tarka- jvâlâ].2 vols. (Beijing: Zhongguo shehui kexue chubanshe WBttêWSWætt).
He Huanhuan MRR- Forthcoming. “Dignäga, Bhäviveka, and Xuanzang on the 'Re¬striction of the Thesis' (*pratijndvisesana).”
Heitmann, A.L. 1995. “Momentane (ekaksana) Gnosis ( jnâna) im Sinne der Madhya- makahrdayakârikâ und der in tibetischer Übersetzung vorliegenden Tarkajvâlâ I¬IV.” Asiatische Studien/ÉtudesAsiatiques 49: 391-427.
Heitmann,A.L.2004.NektarderErkenntnis.BuddhistischePhilosophiedes6.Jh.:Bhavyas Tarkajvâlâ I-III. 26 (Aachen: Shaker Verlag).
Hirakawa Akira SPI^. 1997. A Buddhist Chinese-Sanskrit Dictionary (Tokyo: The Reiyukai
Hoornaert, P. 1999. “An Annotated Translation of Madhyamakahrdayakârikâ/Tarka- jvâlâ V.1-7.”KanazawaDaigakuBungakubu HonshuKödöKagakuTetsugaku hen S [Bulletin ofthe Faculty of Letters of Kanazawa University, Behavioral Sciences and Philosophy] 19: 127-159.
Hoornaert, P. 2001. “An Annotated Translation of Madhyamakahrdayakârikâ/Tarka-
jvala V.27-54.” Kanazawa Daigaku Bungakubu Ronshu Kodo Kagaku.Tetsugaku hen 21:149-190.
Hoornaert, P. 2003. “An Annotated Translation of Madhyamakahrdayakarika/Tarka- jvala V.85-114.” Kanazawa Daigaku Bungakubu Ronshu Kodo Kagaku. Tetsugaku hen 23:139-170.
Hoornaert, P. 2004. “The Dharmapäla-Bhävaviveka debate as presented in Dharma- päla's Commentary to Catuhsataka XVI.23.” Kanazawa Daigaku Bungakubu Ron- shU Kodo Kagaku. Tetsugaku hen 24:119¬
149.
Huntington,Jr.C.W.1995.“ALostTextofEarlyMadhyamaka.”AsiatischeStudien/Études Asiatiques XLIX: 693-767.
Ichigö Masamichi —^£1.1985. Madhyamakälamkära of Säntiraksita with his own commentary or Vrtti and with the subcommentary or Panjika ofKamalastla (Kyoto: Buneido XXS).
Ishihama Yumiko and Fukuda Yuichi —. Ed. 1989. A New Critical
Edition ofthe Mahavyutpatti. Sanskrit-Tibetan-Mongolian Dictionary ofBuddhist Ter-minology. Studia Tibetica No. 16 (Tokyo: The Toyo Bunko X>¥X)W).
Jayaswal, K.P. 1988. An Imperial History ofIndia in a Sanskrit Text (Patna: EasternBook House).
Jiang Zhongxin ^X^. 1991. “Fanwen 'Sizeyanjing' chaoben yingyinban ^X ‘SfWê ij_’ [A Facsimile Edition of the Sanskrit Text of TarkajvalâsUtra]” JiXian
linjiaoshou bashi huadan jinian lunwenji [Papers
in Honor of Prof. Dr. JiXianlin on the Occasion of His 80th Birthday]. Ed. Li Zheng et al. (Nanchang: Jiangxi chubanshe A®X®tt): 111-118.
Kajiyama Yuichi SAæ—. 2005. “Bhävaviveka, Sthiramati and Dharmapäla.” Studies in Buddhist Philosophy. Selected Papers. Ed. Mimaki Katsumi (Kyoto: Rinsen Book Co., Ltd.): 177-187.
Kapstein,M.T.2009.“PreliminaryRemarksontheGrubmtha'chenmoofBya'Chadkha ba Ye shes rdo rje.” Sanskrit Manuscripts in China. Ed. E. Steinkellner et al. (Beijing: China Tibetology Publishing House): 137-152.
Keenan, J.P. Tr. 1997. Dharmapala's Yogacara Critique ofBhavaviveka's Madhyamika Explanation ofEmptiness. The Tenth Chapter of Ta-ch'eng Kuang pai-lun shih lun, commenting on Äryadeva’s Catuhsataka chapter sixteen (Lewiston: The Edwin Mel¬len Press).
Krasser, K.2011.“Howto Teach a BuddhistMonk to refute the Outsiders—TextCritical Remarks on Some Works by Bhäviveka.”Dhth 51:49-76.
Krasser, K. 2012. “Bhäviveka, Dharmakirti and Kumärila.” Devadatttyam.Johannes Bronk-horst Felicitation Volume. Ed. F. Voegeli et al. (Bern: Peter Lang): 535-594.
van der Kuijp, L.W.J. 2006. “The Earliest Indian Reference to Muslims in a Buddhist PhilosophicalTextofCirca700.” Journal ofIndian Philosophy 34: 169-202.
van der Kuijp, L.W.J. Forthcoming a. “The Perils of Translation: the 'Old' and the 'New'
Tibetan Translation of the Prasannapada on Mülamadhyamakakârikâ, 1:1.”
van der Kuijp, L.W.J. Forthcoming b. “A Fifteenth Century Biography ofLha bla ma Ye shes 'od (947-1019/24): Part One: Its Prolegomenon and Prophecies.”
LaVallée Poussin, L. de. Tr. 1928-1929. Vijnaptimatratasiddhi: La Siddhi de Hiuan-Tsang. Tome I (Paris: Librairie Orientaliste Paul Geuthner).
La Vallée Poussin, L. de. Tr. 1933. “Madhyamaka.” Mélanges Chinois et Bouddhiques, Tome II (Bruxelles: Institut Belge des Hautes Études Chinoises): 1-146.
Lalou, M. 1953. “Les textes bouddhiques au temps du roi Khri srong lde btsan.” Journal asiatique CCXLI: 313-353.
Lindtner, Chr. 1979. “Candrakirti's Pancaskandhaprakarana” Acta Orientalia 40:87-145.
Lindtner, Chr. 1982. “Adversaria Buddhica.” Wiener Zeitschriftfür die Kunde Südasiens XXVI: 167-194.
Lindtner, Chr. 1984. “Bhavya's Controversy with Yogâcâra in the Appendix to Pra- jnapradtpa, Chapter XXV.” Tibetan and Buddhist Studies, vol. 2. Ed. L. Ligeti. (Buda-pest: Akadémiai Kiado): 77-97.
Lindtner, Chr. 1984a. “On Bhavya's Madhyamakaratnapradipa”Indologica Taurinensia XII: 163-184.
Lindtner, Chr. 1986. “Materials forthe Study of Bhavya.” Kalyanamitraraganam. Essays in Honour of Nils Simonsson. Ed. E. Kahrs (Oslo: Norwegian University Press): 179¬202.
Lindtner, Chr. 1986. “Bhavya, the Logician.” Adyar Library Bulletin [Golden Jubilee Volume] 50: 58-84.
Lindtner, Chr. 1995a. “Bhavya's Madhyamakahrdaya (Pariccheda Five) Yogacaratattva- viniscayavatara.” AdyarLibrary Bulletin 59 [C. Kunhan Raja Birth Centenary Volume]: 37-65.
Lindtner, Chr.1995b.“Bhavya, Legsldan'byed—QuotingandQuoted.”Studiesin Central and East Asian Religions 8: 90-98.
Lindtner, Chr. Ed. 2001. Madhyamakahrdayam of Bhavya. (Chennai: The Adyar Library and Research Centre).
Matsunaga Yukei fê^WW. 1985. “On the Date of the Manjusrlmulakalpa? Tantric and Taoist Studies in Honour of R.A. Stein. Ed. M. Strickmann. Vol. III. Mélanges Chinois et Bouddhiques, vol. 22 (Bruxelles: Institut Belge des Hautes Études Chinoises): 882-894.
Mimaki Katsumi 1982. Blo gsalgrub mtha (Kyoto: Zinbun Kagaku Kenkyusyo
AXWAWAFA Université de Kyoto).
Miyazaki Izumi AAA 2006. “Chugan hotoron ni mirareru bdag gis bkod pa rTog ge ’barbani tsuite bdag gis bkodpar'l'og ge ’barba ¡¿9lA
[On bdag gis bkodparTog ge 'barba in the Madhyamakaratnapradipa] .” Indogaku Bukkyogaku Kenkyu LV: 453-448.
Miyazaki Izumi /I^PI'. 2007. “'Chugan ubadaisha kaihökyö' tekisuto, yakuchu
f + [Madhyamakopadesa-Ratnakarandodghatd. Text
and Notes],” Kyoto daigaku bungakubu kenkyü kiyö 46:
1-126.
Moriyama Seitetsu SXWfc 1992. “Köki chuganha to genshö-shinjitsu-ron, keishökyo- gi-ron: Säkyabuddhi, Prajnäkaragupta, Kambala
^fö^: Säkyabuddhi, Prajnäkaragupta, and Kambala [The Later Mädhyamika and the satydkdra- and alikdkdravddins of the Yogäcära School: Säkyabuddhi, Prajnä¬karagupta, and Kambala].” Indogaku Bukkyögaku Kenkyü XLI:
436-433.
Moro Shigeki 2013. “Xuanzang's Proof of Idealism (Ä®^Ä) and Silabhadra's
Teaching.” Journal of the First International Academic Forum on Maitreya Studies: 167-175.
Nanjio, Bunyiu. 1883. A Catalog of the Buddhist Tripitaka (Oxford: At the Clarendon Press).
Negi, J.S.1993. A Tibetan-SanskritDictionary, vol. 1 (Sarnath/Varanasi: CentralInstitute ofHigherTibetanStudies).
Obermiller, E. Tr. 1932. History of Buddhism (chos-hbyung) by Bu-ston. II. Part. The History ofBuddhism in India and Tibet (Heidelberg: O. Harrassowitz).
Pulleyblank, E. 1991. Lexicon ofReconstructedPronunciation: in Early Middle Chinese, Late Middle Chinese, andEarly Mandarin (Vancouver:UniversityofBritishColumbia Press).
Qvarnström, 0.1989. Hindu Philosophy in Buddhist Perspective. The Veddntatattvavini- scaya Chapter of Bhavya’s Madhyamakahrdayakdrikd. Lund Studies in African and Asian Religions, vol. 4 (Lund: Plus Ultra).
Qvarnström, O. 1999. “Haribhadra and the beginnings of doxography in India.” App¬roaches to Jaina Studies: Philosophy, Logic, Rituals and Symbols. Ed. N.K. Wagle and O. Qvarnström. South Asian Papers, no. 11 (Toronto: University ofToronto Centre for South Asian Studies): 169-210.
Radich, M. 2008. “The Doctrine of Amalavijndna in Paramärtha (499-569) and Later Authorsto Approximately800C.E.”Zinbun 41: 45-174.
Saitö Akira ^U^. 2005. “'Chugan shinron' no shömei to sono seiritsu o meguru shö- mondai '’'EMl-LX.i [Some Problems with
the Title ofthe ‘Madhyamakahrdayasdstra' and Its Formation] .” Indogaku Bukkyö- gaku Kenkyü EPJWW^WX LIII: 832-838.
Saitö Akira 2007. “Daijyö bukkyö no kigen to jitsutai ni kansuru sögoteki kenkyu
[A Comparative Study of the Origin and Status of Mahayana Buddhism].” Tökyö Daigaku kenkyu seika hökokusyo M [Tokyo University Report on Research Results] (Tokyo): 200-270.
Sakai Masamichi 2011/12. “Dharmakirti's Interpretation of the Causelessness
of Destruction.” Wiener Zeitschrift für die Kunde Südasiens LIV: 187-202.
Sakaki Ryözaburö Ed. 1962. Mahavyutpatti. Suzuki Research Foundation
Reprint Series 1 (Tokyo: Suzuki Research Foundation
Schaeffer, K.R. and L.W.J. van der Kuijp. 2009. An Early Tibetan Survey of Buddhist Literature. The Bstan pa rgyas pa rgyan gyi nyi 'od of Bcom ldan ral gri, Harvard OrientalSeries,vol.64(Cambridge:The Department ofSanskrit and Indian Studies, Harvard University).
Schiefner, A. Tr. 1860. DerBuddhismus,seine Dogmen, Geschichte undLiteratur, 1. Theil: Allgemeine Übersicht (St. Petersburg: Kaiserliche Akademie der Wissenschaften).
Seyfort Ruegg, D. 1981. “Autour du Lta ba'i khyadpar de Ye shes sde (version de Touen- houang, Pelliot tibétain 814).” Journal asiatique CCLXVIII: 207-229.
Seyfort Ruegg, D. 1981. The Literature ofthe Madhyamaka School ofPhilosophy in India. A History ofIndian Literature,vol.VII,fasc.1.Ed.J.Gonda(Wiesbaden:Otto Harras- sowitz).
Seyfort Ruegg, D. 1990. “On the authorship of some works ascribed to Bhävaviveka/ Bhavya.” Earliest Buddhism andMadhyamaka. Ed. D. Seyfort Ruegg and L. Schmit- hausen (Leiden: E.J. Brill): 59-71.
Seyfort Ruegg, D. 2002. Two Prolegomena to Madhyamaka Philosophy. Candrakirti’s Prasannapada Madhyamakavrttih on Madhyamakakarika 1.1 and Tson kha pa Blo bzan grags pa/Rgyal tshab Dar ma rin chen’s Dka’ gnad/gnas brgyad kyi zin bris, Translations, Studies in Indian and Tibetan Madhyamaka Thought, Part 2. Wiener Studien zur Tibetologie und Buddhismuskunde, Heft 54 (Wien: Arbeitskreis für Tibetische und Buddhistische Studien Universität Wien).
Shi Ruyuan, “Bhävaviveka's Syllogism as an Initial Step to Enlightenment.” [available on-line]: 58-92.
Sparham, G. [with Sh. Iida]. Tr. 1993. Ocean ofEloquence (Albany: State University of NewYorkPress).
Thomas, F.W. and H. Ui [= Ui Hakuyu 1918. “‘The Hand Treatise,' a Work of
Aryadeva (sic).” Journalofthe RoyalAsiatic Society: 267-310.
Tucci, G. 1923-1925. “Studi Mahäyänici I. La versione cinese del Catuh^ataka di Ärya- deva confrontata col testo sanscrito e la traduzione tibetana,” Rivista degli Studi Orientali (Roma) 10: 521-567
Watanabe Chikafumi SÄ®X. 1988. “A Translation of the Madhyamakahrdayakarika withthe Tarkajvala III. 137-146.” JournaloftheInternationalAssociation ofBuddhist Studies 21: 125-155.
Ye Shaoyong ^^W. 2009. “A preliminary survey of Sanskrit manuscripts of Madhya- maka texts preserved in the Tibet Autonomous Region,” Sanskrit Manuscripts in China. Ed. E. Steinkellner et al. (Beijing: China Tibetology Publishing House t£|3 307-336.
Ye Shaoyong f^W. Ed. 2011. Mulamadhyamakakarika (Shanghai: Zhongxi shuju 1 ,• 'i,Li
Yonezawa Yoshiyasu 1999. “*Laksanatlka. A Sanskrit Manuscript of an Anon¬
ymous Commentary on the Prasannapada”IndogakuBukkyogakuKenkyu EP®^^
XLVIII:i-3.
Yoshimizu Chizuko 1999. “The Development of sattvanumana from the
Refutation of a Permanent Existent in the Sauträntika Tradition.” Wiener Zeitschrift für die Kunde Südasiens XLIII: 231-254.