Some Observations On Manuscript Transmission Of Nyayabhasya

Author(s): Yasutaka Muroya
Publisher: Yasutaka Muroya
Catalog link: https://jainqq.org/explore/269206/1

Page #1

Some Observations on the Manuscript Transmission
of the Nyāyabhāsya*
Yasutaka MUROYA
1. Introduction
Since the editio princeps of the NBh was published by Jayanārāyana Tarkapancānana in the Bibliotheca Indica Series (No. 50) in Calcutta during the years of 1864-1865, more than twenty-five editions of this text have been published, apart from some not yet identified editions. In spite of this large number of editions, many of them are of doubtful value because
* This is a revised and enlarged version of my paper read at the annual meeting of the Association for the Study of the History of Indian Thought, Kyoto, on December 10, 2005. On this occasion, I would like to express my cordial gratitude to those colleagues who kindly made remarks on my paper at and after the conference, especially, em. Prof. Noritoshi Aramaki. Prof. Toru Funayama, Dr. Kengo Harimoto, em. Prof. Masaaki Hattori, em. Prof. Yasuke Ikari, Prof. Kei Kataoka, Prof. Werner Knobl, Prof. Esho Mikogami, Prof. Hojun Nagasaki, Prof. Masanobu Nozawa, Dr. Yasuhiro Okazaki, Prof.
uko Yokochi, and Prof. Kiyotaka Yoshimizu, and furthermore to Prof. Akihiko Akamatsu who kindly took the trouble to arrange the subsidiary support for my stay in Kyoto and gave me the chance to talk about the issue treated in this paper at a session of the VAADA research group (part of the COE Program, Faculty of Letters, Kyoto University) on December 11, 2005. I would also like to acknowledge my indebtedness with sincere gratitude to the following institutions and individuals for their assistance in gaining access to the manuscripts explicitly used in this paper and for permission to obtain copies of them: Asiatic Society (Kolkata), Banaras Hindu University (Varanasi), Government Oriental Manuscripts Library (Chennai), L. D. Institute (Ahmedabad), Oriental Research Institute (Mysore), Oriental Research Institute & Manuscripts Library (Trivandrum). Research Library for South Asian, Tibetan and Buddhist Studies (Vienna), Srihemacandracarya Jaina Jñāna Mandir (Patan), H.H. Muni Shree Jambuvijayaji, Dr. Sung Yong Kang, Dr. Hisayasu Kobayashi, em. Prof. Asko Parpola, Prof. Karin Preisendanz, Mr. P. L. Shaji, Prof. Ernst Steinkellner and Dr. Dominik Wujastyk. I am also very grateful to the Japan Society for the Promotion of Science for granting a scholarship for overseas postdoctoral research which enabled my research in South India in November and December 2002, and to the Austrian Science Fund (FWF) for funding the project "Metaphysics and Epistemology of the Nyaya Tradition" (FWF Project 17244 G-03) directed by Prof. Preisendanz. My cordial thanks are due to Dr. Sung Yong Kang, who has considerably contributed to my textual study of the NBh as a project collaborator and discussed various issues with me. I am also grateful to Dr. Anne MacDonald for reading this paper. I am deeply indebted to Prof. Preisendanz for taking the trouble to read through this paper and conveying a number of critical comments, thought-provoking discussions, and valuable suggestions. However, responsibility for the text remains entirely with me.
For unidentified editions, cf., for example, the "krpäräma Ed." mentioned in Sowani 1920: 88, fn. 12.

Page #2

24
Journal of Indological Studies, No. 18 (2006)
they clearly have not been critically edited on the basis of manuscript materials. There is, however, one edition that calls for our special attention from the viewpoint of the transmission of the text of the NBh composed by Vātsyāyana or Pakṣilasvamin,2 namely, the one that was published as "a specimen volume" for the first adhyaya of the NS by Anantalal Thakur in 1967 in the Mithila Institute Series (hereafter EM). It contains not only the NBh, but also Uddyotakara's NV, Vacaspati Miśra's NVTT and Udayana's NVTP, collectively called the Nyayacaturgranthikā ("the four famous commentaries on the system of Akṣapāda" 3). After the finalization of the editorial work in 1988, the monumental enterprise was completed in the year 1997 with the publication of the NBh by the Indian Council of Philosophical Research, Delhi (hereafter ED).
During the interval of some thirty years between the edition of the first adhyāya and the final edition, several newly discovered commentaries relating to the NS, based upon manuscripts preserved in the invaluable Jaisalmer collection, came into the scholarly world owing to the efforts of the same eminent editor: Aniruddha's NVP (Darbhanga 1969), Abhayatilaka's NA, edited together with J. S. Jetly (Baroda 1981), and Śrīkantha's ST (Calcutta 1986). In addition to these rediscovered Nyaya works, mention should also be made of Bhaṭṭavagīśvara's NTD edited by Kishor Nath Jha (Allahabad 1979) on the basis of a South Indian manuscript.5
These editions of direct and indirect commentaries on the NS not only provide access to more abundant information about the varied interpretation of the NS in their exegetical discussions, but also enhance the possibility of
2 For a recent study of the author's date, confirming "the second half of the fifth century," cf. Franco/Preisendanz 1995: esp. 86; cf. also Franco 2002: 282-283. For the designation "pakṣila," cf. Steinkellner/Krasser/Lasic 2005: 99.
3Cf. Thakur's Preface to EM: vii.
*Cf. Thakur's Preface to the NVTT and the NVTP: vii (respectively).
5 The manuscript which has been used by the editor is GOML(2). Cf. also Jha's introductory note (prāstāvikam kiñcit) to the NTD: ka-kha. For the basic information on the manuscript, cf. Kuppuswami Sastri 1927: 5080-5081: "Transcribed in 1920-21 from a MS. of M.R.Ry. Paliyattu-Acchan, Chennamangalam, Parur post, Cochin State." Basically the same information is given in the colophon (cf. p. 98) of GOML(2). Reference to passages in GOML (2) is made according to the page numbering. "M.R.Ry." is the abbreviation for mahārājarajasri (cf. Grünendahl 2001: 52). On other manuscripts in the Paliyam manuscript library, cf. Winternitz 1928, where "Paliyath Valia Achan" is introduced as the owner of the collection and also as the uncle of P. Anujan Achan whom Winternitz regarded as one of his best students in Santiniketan. The recent research by Preisendanz in February 2006 confirms that a manuscript of the NTD, i.e. ORIML(5), preserved at the Oriental Research Institute & Manuscripts Library, Trivandrum, was previously in the possession by the Paliyam family (cf. footnote 9 below). In all probability, the transcript written in Devanagarī script and preserved at GOML is a transcript of this palm-leaf manuscript written in Malayalam script, unless the family possessed another manuscript of the work. My present collation does not provide any negative evidence as regards this identification.
For the controversies on theoretical issues among the Nyaya commentators such as

Page #3

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 25
reconstructing the historical development of the transmission of the NS in a more concrete way than ever, and of discovering fragments of lost Nyaya works. At the same time, from the text-critical viewpoint, the frequent mention of pratika-s extracted from those commentaries relating to the NS, or the innumerable direct and indirect references to earlier works are of great importance for the reconstruction of the original reading of the concerned text. They allow us to compare the printed texts or available transmitted manuscripts with the text upon which the commentators relied. By way of the above-mentioned series of publications by Thakur and Jha we have therefore come to be faced with the crucial issue of the history of the textual transmission of the classical commentaries on the NS and also with the task of reconstructing the original text of these commentaries. In particular with regard to the NBh, this task will surely require much time and the careful study of the available texts on the basis of manuscript material. A first step towards such a reconstruction is the philological analysis of primary and secondary testimonies as well as of the variants recorded in the printed editions, especially the variants given in Thakur's two editions, because of their diversity and distinction in quality and quantity.
Together with Sung Yong Kang, the present author is currently engaged in a project on the NBh organized by Karin Preisendanz at the University of Vienna, Austria, which aims at preparing a critical edition of the work on a broad material basis. In the following examination, a preliminary attempt will be made to introduce the Trivandrum manuscript of the NBh, which appears not to have been utilized for any editions known to me, and to consider the value of this manuscript as a primary witness. This contribution focuses on clarifying the history of the transmission of the NBh by means of the comparison of the variant readings of the Trivandrum manuscript with readings in other sources such as the manuscripts of the NBh available to the project, printed editions of the NBh, and secondary or independent testimonies; it does not represent an exhaustive examination of the individual variant readings, simply because the collation of the approximately fortyfive manuscripts is still in progress. Thus only a few aspects relevant to the mentioned purpose and demonstrated by the Trivandrum manuscript are being introduced. For practical reasons, the evidence of the trisütribhasya, i.e., the commentary on the first three sūtra-s of the NS, will be the primary
Visvarūpa, Trilocana and Vācaspati, cf. Thakur's Preface to the NVP: v-vii.
For a recent example of the practice of this renewed methodology in studying the Nyāya commentarial works, especially concerning the critical consideration of the readings of the NBh as recorded in the printed editions, cf. Preisendanz 2000, which collates at least six printed editions. For other examples of a critical approach to the text of the NBh, cf., e.g., Ganganatha Jha's two editions of the NBh and the footnotes to his translation, Sudarsanācārya's Prasannapadā, Preisendanz 1994 (e.g., pp. 701-702) and Okazaki 2005. Nagasaki (1968) examines the textual transmission of the NBh that is reflected in Hemacandra's Pramanamimamsă and points out its difference from that of a printed edition of the NBh.

Page #4

26
Journal of Indological Studies, No. 18 (2006)
focus for the present article.
2. The Trivandrum manuscript, alias the Paliyam manuscript
The Trivandrum manuscript is preserved in the Oriental Research Institute and Manuscripts Library, University of Kerala, Trivandrum. I was able to obtain copies of the ms. in January 2003, following research in November 2002 at the institution. The presence of the ms. in this Library was made known in 1995 by the publication of the sixth volume of the mss. catalogue of the institution. Recent field research by Preisendanz in February 2006 has shown that like the ms. of the NTD, the ms. was originally kept in the possession of the Paliyam family in Chennamangalam, Kerala. This fact, in its turn, confirmed our previous assumption regarding the identity of the original of a transcript of a NBh ms. preserved at GOML, also available to the project (see below).
The palm-leaf ms. (hereafter designated as T), which bears the ms, number 14980A, is part of a composite ms.; the text of the NBh is followed by a list of the prakarana-s of the NS, 10 the text of Udayana's Nyāyaparisista (ms. no. 14980B) and that of Vāmeśvaradhvaja's Pañcikā (ms. no. 14980C).
Let me briefly describe T. It is written in Malayālam script, most probably by a single hand, and is undated. The leaves are in bad condition due to damage of the material, such as innumerable worm-holes, and are sometimes broken off at the edges. They are numbered with letter-numerals in Malayālam script according to the so-called nannādi system, which are placed in the middle of the left hand margin of each recto side. However the first leaf is marked with śr, the second with na, and so forth,12 which does not reflect the common nannādi system (cf. Grünendahl 2001: 94); according to the common system, it would start with na already on the
8 Alphabetical Inder of the Sanskrit Manuscripts in the Oriental Research Institute & Manuscripts Library. Eds. K. Vijayan, P. Visalakshy and R. Girija. Vol. VI. Trivandrum 1995. This ms. is considered identical with the ms. of the NBh corresponding to the entry "Paliyam 257(a)" that is already reported in Kunjunni Raja (1978: 279, left column).
9 According to the Register of the institution, the ms. of the NBh was donated by the Paliyam family in 1951, and the ms. of the NTD in 1969. I owe this information to Prof. Preisendanz. Cf. also footnote 5 above.
10 This list of prakarana-s is identical with the list that is printed in the edition of the Nyāyaparisista. The editor identifies its source as the Madras ms. "ma" and regards it as an "additional part having the form of a summary of contents" (visayasarigraharupo 'dhiko bhagah). Cf. NP 1-2, fn. 1. It is evident from the information in the catalogue (cf. Kuppuswami Sastri 1927: 5038-5039 under "Beginning") that the list was part of the ms. of the NP. The Madras ms. is also a transcript of a Paliyam ms.: "Transcribed in 1920-21 from a MS. of the Paliyattu Valiya Accan, Chennamangalam, Cochin State." Cf. Kuppuswami Sastri 1927: 5038-5041 under R. No. 3377.
1 For some examples of this type of pagination, cf. Ikari 1995: 10, fn. 30; cf. also Ikari 1996: 150, fr. 10.
42 For practical purposes, when referring to the text of T, I will refer to the leaf marked with sri as "f. 1" and to the leaf marked with no as "f. 2."

Page #5

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 27
first leaf. T covers the whole text of the NBh, and ends with folio 99r 9. Individual sūtra-s are not numbered, but they are marked with two kinds of punctuation marks, placed before and after a sutra. The mark put before a sütra has an ornamental, stylized form; the mark put after a sütra is a small dot in the middle of the line, and in some cases looks similar to a semicircle. The former punctuation could possibly be related to the "ornamental flourishes" mentioned by Burnell (1878), which, however, normally appear in colophons. 13 As rightly observed by Isaacson (1995: 44), the function of the signs that are "occasional middle dots" in his exemplar, and "placed before or after a sutra," appears to "distinguish the mula text from the commentary" in a more exact way. There are a few exceptional cases where the scribe appears to have omitted them and also unclear cases as regards the function of the sign. The punctuation with the ornamental signs is also used at the end of daily lessons (āhnika) and books/chapters (adhyāya), as well as in colophons. As concerns the features of the sandhi or conjoint forms of akşara-s, Ikari (1996: 13-17) provides various pertinent observations. In this connection, a notable feature in T is the very rare usage of the avagraha, which occurs thrice in the first book of the NBh.14
In relation to T, I should like to refer to another ms. of the NBh, written in Devanāgarī script on lined modern paper with pagination in Arabic numerals on every page. It is preserved in the Government Oriental Manuscript Library, Madras (hereafter designated as GOML(1)). The basic description by Kuppuswami Sastri (1927: 5513) states that it was transcribed in 192122 from a Paliyam ms. As mentioned before, GOML(1) was assumed to be
13 Burnell (1878: 82) regards the mark as "various forms of the word 'Crī'." Isaacson (1995: 44) follows this identification in recording the marks used in his exemplar written in Malayalam script. Ikari (1995: 17) also mentions the "peculiar sign(s)" in Malayalam manuscripts. He remarks that "[t]he mark generally looks to be just a sign without any meaning, although that of N2 looks like a script of om in Malayālam." Maas (2004: Lxxxviii) regards it as om, following Grünendahl (2001: 52, 92); cf. also his forthcoming edition of the first chapter of the Yogabhāsya (Philipp André Maas. Samadhipäda. Das erste Kapitel des Patañjalayogaśästra zum ersten Mal kritisch ediert. The First Chapter of the Patañjalayogaśāstra for the First Time Critically Edited. Aachen: Shaker, 2006). On this occasion I would like to express my gratitude to Prof. Ikari and Dr. Maas for the discussions relating to this issue, and for sending digital pictures of the mss. where the sign in question appears. Because of their more or less different appearances, the question remains whether the signs described by them can be regarded as identical with the one used in T, even if the sign in question also appears in the colophons of T. One may say that they are utilized with the same function in spite of their apparent variations. I do not call them sri or om in the present article, mainly because the scribe of T uses substantially different aksara-s to denote sri and om. I owe this present decision to the discussion with Prof. Ikari. My cordial thanks are also due to Dr. Ram Manohar and Prof. Tsutomu Yamashita for their comments on this ms.
14 This contradicts the common observation that mss. written in Malayālam script do not at all employ the avagraha sign. Cf. Ikari 1996: 16; Grünendahl 2001: 92.
16 Cf. Kuppuswami Sastri (1927: 5513): "Transcribed in 1921-22 from a MS. of M.R.Ry. Paliyattu Valiya Acchan, Chennamangalam, Parur post, Cochin State." Cf. also the

Page #6

Journal of Indological Studies, No. 18 (2006)
a transcript of T, on the basis of the English colophon, the consistent coincidence of scribal errors and other variants, the places of lacunae marked by series of dots, and the selection of individual sutra-s marked by new paragraphs and preceded by "sū" accompanied by double danda. Given that GOML(1) has now been confirmed to be most probably a direct copy from T, the ms. is very useful when it comes to restoring lost or damaged portions of T, whose condition has deteriorated after more than three quarters of a century.
3. Sūtra-s in the Trivandrum manuscript
In the following, I would first like to refer to the wording and the selection of sutra-s in three cases: NS 1.1.2, NS 1.1.5 and a passage normally regarded as part of the commentary on NS 1.1.5. (In the following, the abbreviation "NS" is not always added to the corresponding number of the sūtra in question.) In treating the signs introducing a sutra in this section, I start from the hypothesis that the phrases or sentences marked as such in T were regarded as sutra-s in a certain tradition of the ms. transmission of the NBh or in a specific historical or regional Nyāya tradition. In other words, I will tentatively accept those texts marked as sūtra-s as such. But the question of whether some of them could be classified into types of text other than sutra-s, for example, grahanakavākya, remains for future study and will not be discussed in the present article. Another question which has to be borne in mind is whether the marking of sutra-s in T can be traced back to the original exemplar from which the ms. was copied, or whether a pair of specific marks were additionally placed before and after certain sentences for the first time by the scribe of T. This question is under my examination, but will not be taken into account here.
As regards the conventions for transcribing the text of mss. which will be treated below, I add word divisions, ignore string-holes, report the text diplomatically with its sandhi, and introduce some symbols for the practical indication of akşara-s or signs: a virāma or a special halanta form is marked by an asterisk after the aksara in question, an ornamental sign placed before a sutra by 'e,' a sign placed at its end by 'o,' and a missing or damaged aksara or part of an aksara by '+'; otherwise I have made no further changes or additions to the ms. evidence.
3.1 NS 1.1.2
T has a substantially different reading for NS 1.1.2 than the common one which reads (cf. EM 150,3–4): duḥkhajanmapravsttidosamithyājñānānām colophon of GOML(1): p. 220. Reference to passages in this ms. is made according to the page numbering.

Page #7

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 29
uttarottarāpāye tadanantarabhāvād apavargah. T reads (I also cite the introductory part before 1.1.2; cf. f. 3r 6–7): kin tarhi tatvajñānāt* duhkhajanmapravsttidosamitthyājñānānām uttarottaräpäye tadanantarābhāvāt* o niśreyasādhigama iti. The sign o' shows that the sūtra ends with tadanantarābhāvāt. This ending disagrees with the commonly accepted text of 1.1.2 where apavargah, as cited above, should be the concluding word after bhāvāt. 16 The reading of T may also suggest that an original nisreyasädhigamah was replaced by apavargaḥ for some reason, or vice versa.17 However, it should be emphasized that T excludes the word from the sutra. As for this exclusion, none of the mss. of the NBh available to us supports such a version of 1.1.2, nor do they place any sign of punctuation, such as a danda, before the uniformly accepted apavargah. T's termination of 1.1.2 with bhāvāt, on the other hand, is supported by several independent testimonies, such as the NM, 18 the three sūtrapātha mss. written in Malayālam script,19 and a direct commentary on the NS, namely the NTD 20 Further
16 For observations on the "original" text of the NS that ends with the ablative and also on its relation to the part of the NBh introducing a sutra, cf. Preisendanz 1994: 412-414, 610-611.
17 This variant of T reminds us of a critical note in EM: "ontarāpāyān nihsreyasādhigama iti bahutra" (cf. EM 150, fn. 3). This mysterious remark indicates that the variant, which is different from tadanantarābhavad apavargah adopted by Thakur, is common in many texts (bahutra); however, this variant Tihśreyasādhigamah instead of apavargah is by no means attested in the printed editions of the NBh. This critical note is not reported in Thakur's recent edition of the NBh, but instead moved to NVTP 102, fn. 5; Thakur informs us elsewhere that "none of the MSS used here (i.e., in his edition of the NVTP) contains the sūtra texts" (supplement by me; cf. Thakur's Preface to the NVTP:x). Thus it is evident that the variant is not related to the mss. of the NVTP used by him, but most probably to those of the NBh. As Thakur mostly reports the variants which deviate from the text as constituted in his editions, it is possible that nihfreyasādhigamah is a variant of the Jaisalmer ms. of the NBh used by him; Thakur does not report that he consulted other mss. of the NBh for his editions. Concerning the originality of tadanantaräpäyát, K. N. Jha, providing many independent testimonies, maintained the position that pāyāt is better and the original; for his discussions, cf. NTĀ 495-497, where most of the parallels are given. He also used a sutrapatha ms. from ORIML, but did not provide information on which ms. he consulted, nor on the absence of apavargal
18 Cf. NM(V) 513,3-6: uktam eva bhagavatā sutrakärena - duḥkhajanmapravrttidosamithyāmānānām uttarottarāpāye tadanantarābhāvād iti. The absence of apavargah after
bhāvāt is supported by the mss. of the NM; cf. BHU(1) f. 55v 14 and MORI(1) f. 224v 16-225r 1 (the latter with the corrupt reading "rābhāvädi"). The Mysore edition of the NM reads tadanantarāpāyād apavargah. The presence of apavargah here may be a silent emendation of the editor's for the sake of adapting the sutra to the common reading, further by replacing bhāvād with pāyād, and omitting iti; no variant has been recorded here by him (cf. NM(M) II 440,3-5).
19 Cf. ORIML(1) f. 1r 4, ORIML(2) f. 1r 2 and ORIML(3) f. lr 3; only ORIML(2) reads pāyāt* for bhāvāt.
20 The NTD reads duḥkhajanmapravsttidosamithyajñananam uttarottaräpäye tadanantaräbhavat. Cf. NTD 2,12. However, the printed edition adds the common version of 1.1.2 in bold face before the text of this commentary; cf. NTD 2,10-11. This addition has to be considered as an editorial change, since neither the transcript GOML(2) utilized by the editor nor its original ms. ORIML(5) has this passage. Jha remarks on the absence of

Page #8

30
Journal of Indological Studies, No. 18 (2006)
more, these testimonies do not affirm that the text of the NBh ends with nihśreyasādhigamah after bhāvāt.
As regards the supplementation of the sutra with nihśreyasādhigamah, Gambhīravamsaja's Nyāyasūtravivarana (hereafter NSV(G)), is worthy of our attention. 21 The ms. of the NSV(G) written in Grantha script reads tatuajñānād iti o duhkhajanmapraurttidosamitthyajñānānām uttarottarāp++++nantaräpāyāt* nisreyasādhigama iti tena sambandhah.22 The last phrases nihśreyasādhigama iti tena sambandhah may indicate that 1.1.2 should be connected with the last word of 1.1.1, i.e., nihśreyasādhigamaḥ.23 A further testimony for the reading in question, namely, nihśreyasādhigamaḥ in place of apavargah, is Akalanka's Tattvärthavarttika.24
How was the wording of 1.1.2 treated by the commentators on the NS in the medieval period, and how was the question of whether 1.1.2 ends with bhāvāt or apavargah discussed by them? Vācaspati Miśra I inserts a brief, noteworthy digression occasioned by anonymous opponents in the
apavargah in an appendix to his edition: "If the word apavargalh is employed subsequently to tadanantarābhāvāt in the first line of the commentary here, then the incoherence of the meaning (of 1.1.2) is removed" (iha vrtteh prathamāyām pamktau "adanantarabhāvāt" ity anantaram apavarga iti padam yadi yojyate tadarthasya visargatir apahrtā bhavati; cf. NTD parisistam (1), p. 1 under 1.1.2).
21 This text was first published in 1992 by Anandateertha V. Nagasampige. According to the editor, the author's date is unknown (cf. his preface to the NSV(G): xv-xvi); in the margalasloka, the work is called "Nyāyavärttikasamgraha" by the author himself. The NSV(G), or probably more correctly the Nyāyavārttikasamgraha, is rich in quotations from both the NBh and the NV.
22Cf. ORIML(4) f. lv 9-11. I take the reading provided by ORIML(4) as better than the text of the printed edition. This ms. was not utilized by the editor of the work, A. V. Nagasampige, although the presence of the ms. is noted in R. G. Mālagi's Introduction (prastāvanā); cf. his Introduction to the NSV(G): v, fn. 1. As for the basis of his printed edition (cf. his prārinivedanam "prior announcement"), Nagasampige has mentioned two mss. written in Grantha and Kannada scripts respectively, and preserved at the Oriental Research Institute, Mysore (mss. nos. P. 4071/B and A. 743/2 respectively; the former is a palm-leaf ms.). Cf. also R. S. Shivaganesha Murthy's Preface to the NSV(G) (p. ii), which states that the edition is based upon a single ms., most probably the former one.
As for the reading of the printed edition, cf. NSV(G) 8,10-13, which, probably on the basis of the ms. mentioned above, reads: duḥkhajanmapraurttidosamithyajñananām uttarottarāpāye tadanantarābhāvät, nihśreyasādhigama iti sütrena gatena sambandhah. I find it difficult to construe nihśreyasadhigama iti sütrena gatena sambandhah, which may literally be understood as "a connection (of 1.1.2) with the elapsed sutra (running) 'the attainment of the highest good' (should be effected)."
Cf. NS 1.1.1: pramānaprameyasamsayaprayojanadrstāntasiddhāntāvayavatarkarirnayavādajalpavitandāhetvābhāsacchalajātinigrahasthānānām tattvajñanan niņśreyasādhigamah (cf. Ep 2,7-9).
24 Cf. TAV 12,8-9: duhkhajanmaprausttidoşamithyajiananám uttarottarapaye tadanantarābhāvān ninfreyasādhigama ity anyesām darśanam. Akalanka's reference to "the view of others" (anyesäm darśanam) does not make clear whether this view is concerned with the sutra only, or whether it relates to its text as embedded in the NBh. Obviously he does not presuppose the commonly accepted text of 1.1.2. A further independent testimony is ASTV II 628,2-3, which runs yad uktam parena duhkhao ... tadanantarabhāvan nisreyasa iti (ellipsis by me). I owe this reference to Mr. Himal Trikha, M.A.

Page #9

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 31
beginning part of his commentary on 1.1.2.25 Opponents whom he calls "some (scholars or commentators?]" (kecit) assert that the sutra should be divided into two parts by virtue of "division of a rule" (yogavibhāga) or "division of a coherent connection." 26 The first "rule" or coherent connection runs duḥkhajanmapravrttidoşamithyājñānānām, the second uttarottaräpāye tadanantarābhāväd apavargah.27 The former states the causal relationship (kāryakāraṇabhāva) among the five soteriologically relevant elements which are mentioned next to each other, e.g., pain (duhkha) is an effect and birth (janma) its cause. The "mutual connection" (itaretarayoga) of the five elements, as expressed by means of a type of dvandva-compound, implies the relation of cause and effect by virtue of "suitability" or "appropriateness" (yogyatä).28 This first coherent connection makes logically possible the second that illustrates the gradual annihilation of the mentioned elements in reverse order. This second connection is made comprehensible because the causality relating to these elements has already been established by the preceding coherent connection. Thus the annihilation of each following element (as cause) brings forth that of each immediately preceding element (as effect); for example, false knowledge (mithyājñāna) as a cause vanishes and then the faults (dosa) as its effect also vanish.29 The opponents' view is
25 Cf. Perry 1995: 74, fn. 158.
26 Cf. NVTT 62,21: atra kecid yogavibhāgam icchanti. On yogavibhāga in the grammatical tradition, cf. Abhyankar (1986: 318r) and Renou (1957: 256-257). Cf. also Apte (1957: 1318, s.v.): "separation of that which is usually combined together into one; especially, the separation of the words of a Sūtra, the splitting of one rule into two or more." It is remarkably unusual for a commentary on the NS to apply the grammatical terminology yogavibhāga to the exegetical procedure for the interpretation of the NS. As the sutra-s in the Nyaya tradition are not normally treated as laying down a yoga 'rule' (cf., e.g., Abhyankar 1986: 3181), it may not be reasonable to construe yoga as a rule. NS 1.1.2 describes the ordered sequence of soteriologically relevant elements and their annihilation leading to liberation. They constitute a single coherent totality as a statement. Thus yoga, literally "connection" or "what is connected or united together," as it is terminologically employed in this discussion, is to be understood as the coherent connection that involves soteriologically relevant elements whose annihilation leads to the liberation. I tentatively employ the English equivalent "coherent connection" for yoga, and "division of a coherent connection" for yogavibhaga.
27 As regards the reading bhāvād, I follow EM (157,17), which is supported by the Jaisalmer ms., whereas NVTT (63,3-4) reads tadanantarāpāyād instead of tadanantarābhāvād. Cf. also footnote 29 below.
28 Cf. NVTT 631,1-2 = EM 157,15-16: duhkhajanmapravrttidoşamithyājñānānām ity eko yogah. atra kila samäsäd eteşām itaretarayogo 'vagamyate. sa ca yogyatayā kāryakaranabhavah.
29 Cf. NVTT 63,3-5 = EM 157,17-18: atah siddhe kāryakäranabhäva uttarottaräpäye tadanantarabhavad" apavarga ity anena yogena kāranocchedakramena karyocchedakramapratipädanenāpavargah pratipadyate. ( bhāvād "J" in Ep, Em; pāyād NVTT.) In Srikantha's ST, the second coherent connection as presented by the opponents is not identical with that adduced in the NVTT. Srikantha seems to presuppose that it ends with
bhävät, and thus his analysis is considerably different from Vacaspati's: uttarottaräpaye tadanamtarabhāvād ity asmin* dvitiye yoge ... duhkhajanmapraurttidosamithyajiananam ity asmin* prathamayoge (ellipsis by me). Cf. LDI(1) f. 42r 6-7, a passage which is

Page #10

32
Journal of Indological Studies, No. 18 (2006)
dismissed by Vacaspati who appeals to Uddyotakara's explicit mention of the sutra in the singular as invalidating evidence.30 It should be noted that in this digression both the opponents and Vacaspati appear to regard 1.1.2 as ending with apavargah, the reading which represents the generally accepted text.31 If this reconstruction is accepted, it has to be assumed that Vacaspati actually commented upon a version of 1.1.2 different from that of Jayanta Bhaṭṭa and other commentators. 32
There are some places where Vatsyayana refers to 1.1.2. His mention of the sutra in his commentary on 4.2.1 can be adduced as internal, problematic evidence which speaks against the T version of 1.1.2. In the context of discussing the faults (dosa), which are the three elements causing karma
not available in Thakur's edition according to ST 69,20, since "one complete folio escaped the camera" (cf. Thakur's Preface to the ST). Furthermore, it can suggest that Śrīkantha regarded 1.1.2 as ending with bhavad, but this assumption evidently contradicts the discussion adduced by Vacaspati which Śrīkantha supposedly comments upon. In fact, Śrīkantha's mention of these two yoga-s is placed in the part of his commentary on Udayana's corresponding discussion (cf. LDI(1) f. 42r 6: "une" abbreviating "Udayane"). It is totally unclear how he could keep silent about the possibility of the charge of contradicting the NVTT.
30 Cf. NVTT 63,7-8= EM 157,20-21: tam imam sutravibhāgam amṛṣṇamāno vārttikakṛd aha idam sūtram. ekavacanena bhedam vyavartayati. Vacaspati further adduces as the argument against the theory of yogavibhaga the fault of the "splitting of a statement" in 1.1.2 (cf. vakyabheda). Cf. NVTT 63,8-9 EM 157,21-22: na hi samucchedakramapratipadanenäpavargaparatayaikavakyatve sambhavati vakyabhedo nyayyaḥ. ("The splitting of a statement, indeed, cannot be reasonable, because [sutra 1.1.2] constitutes a single statement inasmuch as it is devoted to [the explanation of] liberation by means of demonstrating the order of the complete destruction [of pain and the others].") In Vacaspati's view, the sutra constitutes a single statement (ekaväkyatva), insofar as 1.1.2 has "liberation" as its main objective to be presented (apavargaparataya). Cf. NVTT 63,8-9 EM 157,21-22; NVTP 108,7-9 EM 173,22-24. For ekavakyata, see Preisendanz 1994: 204-207; Kane 1962: 1297-1298. For vakyabheda, cf. Kane 1962: 1299-1303.
31 For another problematic instance, cf. SDS(BI) 115,8-10: kintu tattvajñānād duḥkha ... bhava iti, which seems to silently quote the corresponding portion of the NBh. There is also an explicit reference to 1.1.2 in the same compendium ascribed to Madhava; cf. SDS(BI) 116,9-11: tatha ca paramarṣam sutram duḥkhajanma... bhavad apavarga iti (ellipsis by me). Abhyankar's edition, however, provides a different reading in both places, namely, the one ending with payad apavargaḥ. Cf. SDS 245,7-9 and 246,16-18.
32NBhüş 72,15-17 (= HJJM(1) f. 16r 7-8), which runs parallel to the introductory part of the NBh on 1.1.2, does not support niḥśreyasadhigamaḥ of the T version: tat khalu vai tattvajñānam kim ātmalābhānantaram eva niḥsreyasam sampadayatiti. ucyate
na, kim tarhi tattvajñānād duḥkhajanmapravṛttidoṣamithyajnananam uttarottarāpāye tadanantarābhāvāds apavargo bhavatiti vakyaśeṣaḥ. (°bhāvād HJJM(1); °pāyād NBhüş, where the editor seems to have corrected the text of 1.1.2 to the common one.) With regard to the question whether Bhāsarvajña regarded 1.1.2 as ending with bhävät or with apavargaḥ, it depends on the interpretation of "the rest of the sentence" (vakyaseṣaḥ). Yogindrānanda, the editor, places a danda after apavargaḥ, which indicates that he regards bhavati as the rest of the sentence. However it is also possible to assume that the words apavargo bhavati are intended as that which is to be supplied. The latter assumption entails that Bhāsarvajña regarded 1.1.2 as ending with bhāvāt. On Bhāsarvajña's discussions on NS 1.1.2, cf. Yamakami 2001: 13-18.

Page #11

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 33
such as attachment (rāga), aversion (dveṣa) and delusion (moha), he refers to the previous discussion, stating evam ca kṛtva tattvajñānād duḥkhajanmapravṛttidoṣamithyajñānānām uttarottaräpäye tadanantarapayad apavarga iti vyākhyātam iti.33 As Thakur typographically marked the text duḥkhajanma ... pāyād apavargaḥ by inserting a new paragraph for it, one might consider that NS 1.1.2 is directly quoted by the oldest commentator. However, attention should be paid to the words iti vyākhyātam: in a commentarial text, the verb vya-khyā normally designates the commentator's own activity, and does not refer to the basic text upon which commentary is provided.34 Under this assumption it is conceivable that Vatsyāyana does not directly refer to 1.1.2, but to his own previous commentary on it. In this connection, there would be at least two possibilities to be considered. (1) If Vatsyāyana would have supplied niḥśreyasadhigamaḥ after bhāvāt in his commentary on 1.1.2, he would have given a reformulated text in his commentary on 4.2.1; (2) if he would have read the commonly accepted text in 1.1.2, he would merely have repeated the passage in 4.2.1. The resolving of this issue depends, to some extent, upon a stylistic analysis of his technique of composition. I would like to leave the issue open to question in the present article.
This second sutra has repeatedly attracted the attention of scholars and has been the focus of recurrent discussions with regard to its philosophical and soteriological implications, the literal understanding of the sutra, and its relation to the first and ninth sutra-s, the so-called prameyasutra, or to other sections of the NS.35 Amongst scholars who have discussed the sutra, Slaje (1986) points out the unique occurrence of the expression niḥśreyasa in the NS as well as the remarkable terminological inconsistence ("auffällige terminologische Inkonsequenz") and the alleged synonymity of nihśreyasa in 1.1.1 and apavarga in 1.1.2.36 Under the supposition that the T version of 1.1.2 and the supplementation with niḥśreyasadhigamaḥ in the NBh are original,37 such an apparent terminological inconsistency would have to be considered irrelevant because there is no immediate connection of apavargaḥ to 1.1.2. The T version suggests, furthermore, a possible inter
33 Cf. ED 221,12-15. Further alleged references to 1.1.2 in the NBh that need to be carefully examined are ED 248,17-21 on 4.1.59 (Ruben) 4.1.55, and ED 259,7-10 on
4.2.1.
34 The vya-khyā and its derivatives do not occur in the NS, as opposed to the VS(C), e.g., 1.1: athato dharmam vyäkhyasyamaḥ. The style of Vätsyäyana's references to sūtra-s requires a comprehensive study.
35 Cf. Strauss 1930; Biardeau 1964: 101-102; Oberhammer 1964; Slaje 1986: 164; Akamatsu 1989; Perry 1995: 29-81; Akamatsu 2000.
36 Cf. Slaje (1986: 164-165): "[W]arum denn nur hier in den ersten beiden Sütren verschiedene Termini verwendet wurden." Perry (1995: 70-81) critically reviews Slaje (1986).
37 It remains unclear how the iti appearing after niḥśreyasadhigamaḥ in T functions, especially in relation to the sutra.

Page #12

Journal of Indological Studies, No. 18 (2006)
pretation of Vātsyāyana's previous statement in the commentary on 1.1.1: ātmādeh khalu prameyasya tattvajñanan nihśreyasādhigamah, tac caitad uttarasütreņānüdyate ("To be sure, the attainment of the highest good (arises) due to the adequate knowledge of the [twelve) objects of valid cognition beginning with the soul (as enumerated in 1.1.9). And this is restated by means of a posterior sūtra.").38 It may be supposed that in the following Vātsyāyana actually introduces 1.1.2 by adding the two phrases tattvajriānāt and nihsreyasādhigamah in accordance with his own previous announcement.39 If this were the case, the T version of 1.1.2 could be understood to be tattvajñānād 1.1.1) duhkhajanmapravsttidosamithyajñanānām uttarottarāpāye tadanantarābhāvād (nihśreyasādhigamah 1.1.1). It may also be observed that Uddyotakara provides his second theory on the classification of nihśreyasa, namely, into a higher (1.1.2) and a lower nihśreyasa (1.1.1), just before introducing 1.1.2.40 Under the above assumption, this would have the effect to withdraw the focus from the term apavarga and put it on nihśreyasa as the term to be supplied in 1.1.2, or of demonstrating the clear distinction between the two types of nihśreyasa which he discerns as intended in 1.1.1 and 1.1.2.
3.2 The atha in NS 1.1.5
As is well known, the generally accepted text of NS 1.1.5 begins with atha tatpūrvakam anumānam.41 In his critical notes on the NTĀ, Jha has rightly observed that some secondary testimonies do not read atha before tatpūrvakam in 1.1.5,42 but he leaves the issue open whether atha was originally contained in 1.1.5 or not. In the T version, too, the text does not contain atha at the beginning of 1.1.5. Instead, it reads (cf. f. 5v 7-8): athānumānam* tat*pūruuakan trividham anumānam*. Here, atha is part
3 For translations of this passage, cf., for example, Perry 1995: 33, 75, 186, etc.
39 Instead of taking anu-vad in the sense of 'restate' or 'confirm,' some commentators interpret it literally and etymologically in the sense of 'state afterwards.' Cf. NVTT 32,19 = EM 47,3: nihsreyasahetubhāvābhidhänasyanu pascad udyate 'nūdyate. Cf. also Perry 1995: 38, fr. 33. Cf. further ST 39,5-6: tīkāyām nihfreyasetyadi. nätraikasyaivarthasya dviruccäranätmako nuvädo grhyate. ("[It is said) in the Tika: 'the highest good,' and so forth. Here in the NBh the author] does not refer to anuväda that has the nature of stating only one thing twice.") On Vätsyāyana's exposition of the technical usage of anuvada, cf. Oberhammer/Prets/Prandstetter 1991: 62-63
40Cf. NV 10,19 = EM 152,6: nihsreyasasya paraparabhedät, which is to be compared with his first theory on the classification of nihśreyasa. Cf. NV 2,2-3 = EM 6,14: tac chreyo bhidyamanam dvedhã vyavatisthate drstädretabhedena. Cf. also NV 10,19 = EM 13,14: nihśreyasam punar drstādrstabhedad dvedha bhavati.
4 NS 1.1.5: atha tatpurvakam trividham anumanam purvavac chesavat samanyato drstam ca. Cf. Ep 12,2
42 Most of the relevant testimonies citing 1.1.5, some of which I mention in the following, are already given in NTA 488 on 1.1.5. Ruben (1928) does not mention the absence of atha in some testimonies.

Page #13

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 35
of the NBh, i.e., the sutra is preceded by Vātsyāyana's introductory words atha-anumānam ("[After the characterization of perception, now inference is characterized)."). 1.1.5 thus begins only with tatpūrvakam. The text of 1.1.5 lacking atha is not supported by other mss. available to us, nor by any printed edition, but is supported by some secondary testimonies such as Dignāga's PSV,43 Jinendrabuddhi's PST,44 the NM (adding an enclitic ca which can be ignored in the present discussion), the NBhüş, 6 the NTD,47 and all three mss. of the sūtrapātha from Trivandrum.48
Uddyotakara's introductory commentary to 1.1.5 does not allow us to determine the status of atha as he perceived it, or whether he had before him an introductory sentence of the NBh as found in the T version of this text. Uddyotakara begins to comment upon 1.1.5 with the following words: atha tatpūrvakam trividham anumānam iti. athety anantarye. anumanavisesaņārtham sūtram.49 Uddyotakara does not explicitly specify
43 Cf. PSV(V) 33b5-6: rigs pa can mams na te de srion du son ba can gyi rjes su dpag pa ni mam pa gsum ste, sria ma dari Idan dari, lhag ma dan Idan pa dari, spyir mthon ba can no zes zer ro.; PSV(K) 115a3-4: rigs can rams ni de srion du 'gro ba can gyi rjes su dpag par ni mam pa gsum ste. sria ma dari Idan pa dari, lhag ma dan Idan pa dari spyi mthor ba'o íes zer ro. (Cf. also Kitagawa 1965: 563); VS(C) 215,9-10 (reconstruction): naiyāyikānām api tatpūrvakam trividham anumānam purvavac chesavat samanyato drstam ceti.
44PST Ms.(B) f. 75v 3: naiyāyikānām [!] ityādi pratyakşārumānopamanasabdāh pramānānīty uddiśya pratyaksalaksane bhihite tadanantaram prāptāvasaram anumānalaksanam āha tatpūrvakam ityādi. I owe this reference to Prof. Steinkellner and his seminar. The relevant part in Ms.(B) of Jinendrabuddhi's PST is currently being critically edited at the Institute for the Cultural and Intellectual History of Asia, Austrian Academy of Sciences. Cf. Steinkellner/Krasser/Lasic 2005.
45Cf. NM(V) 109,21-22 = NM(M) I 282,6-7: tatpūrvakam ca trividham anumānam purvavac chesavat sámányato drstam ca.
46 Cf. NBhūş 189,3-5: athedānim anumānasvarüpam vicāryate - tatpūrvakam trividham anumanam purvavac chesavat sämänyato drstam ceti sutram.
**Cf. NTD 3,19: tatpūrvakam trividham anumanam pūrvavac chesavat sämānyato drstam ca. Jha suggests the supplementation of atha before tatpūrvakam, but GOML(2) (p. 5,1) does not have atha. Thus, the absence of atha is to be favoured as original.
48 ORIML(1) f. lv 3; ORIML(2) f. 1r 4; ORIML(3) f. Ir 7.
*8*After the characterization of perception follows) inference, which is preceded by that (i.e., perception and others), (and) of three kinds. As regards this phrase, the word] 'atha' fis used in the sense of an immediate sequence. The sutra has the purpose of specifying inference." C. NV 41,3-4 EM 292,11-12. The translation is tentative, because the inclusion of atha in 1.1.5 affects it, whereas the other sutra-s defining the means of valid cognition, i.e., 1.1.4, 1.1.6, and 1.1.7, do not pose structural problems. As for the translation of the commonly accepted version of 1.1.5, I have difficulty in following some previous translations that construe tatpūrvakam and trividham as attributes of anumanam; cf. for example, Ruben (1928: 3): "Darauf folgt die auf der Wahrnehmung) beruhende dreifache Folgerung," a translation followed by, e.g., Oberhammer/Prets/Prandstetter (1991: 43r). I would prefer to construe tatpūrvakam as a predicate of anumānam, since the former is the definiens and the latter the definiendum. Cf. NVTT 127,14 = EM 303,4-5: laksyam pramāṇabhedam anumānam anüdya tatpūrvakam iti laksanam vidhatte. Jha (1915: 153) puts tatpūrvakam in a relative clause specifying anumānam: "After Perception comes Inferential Cognition, which is led up to by Perception; it is of three kinds."

Page #14

36
Journal of Indological Studies, No. 18 (2006)
here whether atha is contained in the NS or part of the NBh. Thus the question remains open whether Uddyotakara regarded atha as part of 1.1.5 or not. It should be noted that atha and tatpūrvakam immediately follow upon each other, and that there is no specification of atha by a phrase such as iti bhāsyam. What about Uddyotakara's subsequent commentary on the definitions of the other two means of valid cognition, namely, analogy (1.1.6) and verbal testimony (1.1.7)? In these places, he silently quotes the introductory phrases of the NBh, namely athopamānam (NV 53,19 = EM 356,1) and atha sabdah (NV 54,16 = EM 365,18), in clear contrast to the case of 1.1.5. It leaves a general impression that Uddyotakara did not have before him the T version of the introductory phrase in the NBh, and that he took atha as part of 1.1.5.50
As regards the two introductory phrases of the NBh on 1.1.6 and 1.1.7; Vācaspati explicitly specifies the texts as pertaining to the NBh: athopamānam iti bhāsyam (NVTȚ 161,21 = EM 356,18) and atha sabda iti bhāsyam (NVTT 166,5 = EM 367,21). In the case of 1.1.5, however, the pratika' of the beginning part of the sūtra adduced by him confirms that he regards the sūtra as beginning with atha, and thus reflects the commonly accepted text; most probably athanumānam in the T version was unfamiliar to him. He introduces 1.1.5 with pratyakşalaksaņānantaram anumānalaksanaparam sutram pathati - atha tatpūrvakam trividham anumānam iti51: "He52 recites (i.e., repeats) the sutra which is devoted to the characterization of inference subsequent to his recitation of] the characterization of perception, saying "After the characterization of perception follows) inference which is preceded by that i.e., perception), and of three kinds." Vācaspati even interprets atha as implying that "perception is the cause of inference," and
ONSV(G) has atha as part of 1.1.5. The author also quotes the corresponding passage of the NV. in contradistinction to the case of 1.1.2, where his commentary is literally based on the NBh. Cf. NSV(G) 16,10-12: athety anantarye. anumanavisesanārtham sūtram. Gambhiravamśaja's commentary on 1.1.5 is evidently based on the NV.
S1Cf. NVTT 127,4-5 = EM 302,20-21.
52 It is understandable that one is inclined to take the subject of path 'read, recite' as the author of the NS, Gautama/Gotama. But it is also possible to take it as one of the authors of the commentaries upon which Vācaspati provides his commentary, namely Vātsyayana or Uddyotakara. For Vātsyāyana as the subject of path, cf. NVTT 424,14 on 2.2.53 = (Ruben) 2.2.51: bhāsyakāro 'traivārthe sutram pathati; for Uddyotakara as the subject, cf. NVTȚ 327,19 on 2.1.22 = (Ruben) 2.1.22: tad etad värttikakāro bhāsyam anubhāsya pūrvapaksasutram pathati. Udayana comments on sūtram pathati that appears in the NVTT on 1.1.5 (cf. NVTȚ 127,4 = EM 302,20), and suggests the supplementation of "in order to expound (it)" (vyākhyātum). Cf. NVTP 184,3 = EM 331,14: sutram pathati. vyakhyātum iti sesah. In the above translation, one can take the subject as either Vätsyāyana or Uddyotakara. It may also be noted that if the subject of path is Uddyotakara, it is still conceivable that Vācaspati could have had the T version of the introductory phrase in the NBh before him. In this case one has to assume that Vācaspati included atha in sutra 1.1.5, just as Uddyotakara did, not following Vätsyāyana, and without making mention of the different wording of 1.1.5 as possibly presupposed by Vātsyāyana.

Page #15

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 37
states that "now then (athedānim) inference is explained as possessing a cause (i.e., perception)" is meant.53 This fact that atha is firmly established as part of 1.1.5 by Vācaspati is clearly contradictory to T and the other secondary testimonies adduced above. It may safely be said that the inclusion of atha in 1.1.5 had already been established by the time of Vācaspati Miśra I, or in the commentarial tradition to which he belonged. It can also be hypothetically assumed that the presence of atha in 1.1.5 would have been accepted by the commentators on the NS following the NVTT or its commentarial tradition. For example, in the NTĀ of Vācaspati Miśra II, who clearly incorporates atha into 1.1.5, atha is glossed as denoting the causal relationship between perception and inference, just as in the NVT?.54 Should the introductory part of the T version be original, it would have to be supposed that the direct or indirect influence of Vācaspati would have caused scribes or later Naiyāyikas to correct the transmitted text of the NS, either on purpose or inadvertently. It remains open whether the introductory part, i.e., athānumānam, which only T has, was also accordingly corrected in the transmission of the text of the NBh. Vācaspati's explicit mention of the words and phrases to be commented upon could have directly or indirectly influenced the textual transmission of the NS and the NBh, as may also be reflected in the case of 1.1.2.
Further mention should be made of Keśavamisra's brief reference in his GSP to a (fictive?) opponent who suspects that "[the word) atha is placed outside the sutra," and points out the "contradiction among Bhāsya, Vārttika and Tīkā."55 Keśavamiśra does not go into the issue and therefore the issue remains vague. Udayana and others are silent on the issue. 56
3.3 traikālyagrahaņāt in NS 1.1.5
The signs used in T to regularly distinguish a sütra from the commentary point at possible further sūtra-s embedded in the text of the NBh. For example, in the commentary on 1.1.5, T reads (cf. f. 6r 7): traikālyagrahanat* o. The scribe clearly understands this phrase as a sūtra because of his usage of the common pair of signs. The phrase appears in a context where Vātsyāyana differentiates inference from perception in view of the distinction
53 Cf. NVTT 127,15-16 = EM 303,5-6: athety anantarye, uktam pratyaksam anumānasya hetuh, athedānīm anumānam hetumad vyutpadyata ity arthah.
54 Cf. NTA 69,22-23: athoddesakramasamgatyanumanalaksaņāya sutram. atha tatpuruakam trividham anumānam. Cf. also NTĀ 69,25: atheti hetuhetumadbhāvasūcanāya; GSP 5,25: atheti hetuhetumadbhāvapratipädanärtham. 56 Cf. GSP 5,25-26: nanv atheti sutrabahirbhutam, bhasyavärttikatikavirodhat.
There is no corresponding pratika or explanation in the NVTP, or in the NNP. The text of NS 1.1.5, inclusive of atha, printed in the editio princeps of the NVTP (cf. NVTP(BI) 654,1-2), seems to be an editorial supplement. Cf. also footnote 17 above.

Page #16

38
Journal of Indological Studies, No. 18 (2006)
of the time to which their objects pertain.57 Neither Ruben's critical edition nor any other edition mentions even the possibility that this phrase could be regarded as a sutra. Moreover, it is to be noted that neither Uddyotakara nor Vacaspati specifies it as a sutra.58 The selection of the phrase as a sutra, on the other hand, is supported by a ms. of the sutrapatha from Kolkata and the three mss. of the sutrapatha from Trivandrum.59
As to the possibility that it can be regarded as a sutra, Bhaṭṭavägiśvara's NTD provides intriguing evidence:
traikalyagrahanat. na kevalam lakṣaṇataḥ, kim tarhi viṣayabhedad apy anumanam pratyakṣabhinnam, trikalaviṣayatvät. vartamanaikaviṣayam pratyakṣam trikalaviṣayam anumanam iti süträrthaḥ.6 [Inference is distinct from perception] because [the objects] belonging to the three phases of time are apprehended [by means of it]. [That is to say,] inference is distinct from perception, not solely from [the point of view of their] definitions [in 1.1.4 and 1.1.5], but also due to the distinction of the objects [apprehended by them], because inference has as its objects [things] in the three phases of time. Perception has as its objects only [things] which are present; inference has as its object [things] in the three phases of time.61 This is the meaning of the sutra.
60
It is evident that Bhaṭṭavägiśvara's treatment of traikalyagrahaṇāt basically follows Vatsyāyana's intention to differentiate inference from perception from the temporal perspective. 62 Although the printed edition of the NTD does not formally acknowledge the phrase in question as a sutra, the last phrase iti sūtrarthaḥ can hardly refer to 1.1.5, since the content of the
57Cf. ED 13,5: sadviṣayam ca pratyakşam, sadasadviṣayam canumanam. kasmāt. traikalyagrahanät. For recent translations of the relevant passages, cf. Oberhammer/Prets/Prandstetter (1991: 51) and Okazaki (2005: 168).
58 For the elaborate explanation of the corresponding passages in the NV, cf. Okazaki
2005: 168-174.
59 ASC(1) f. 1r 5; ORIML(1) f. 1v 4-5; ORIML(2) f. 1r 5; ORIML(3) f. 1r 8.
60 Cf. NTD 4,17-19 = GOML(2) p. 6,8-10. For unknown reasons, the edition places the word pratyakṣam in square brackets. In the transcript, traikalyagrahaṇat* appears in a new line and is put between double danda-s, which suggests that the phrase is regarded as a sutra by the scribe of the transcript. The original ms. ORIML(5) has a short danda before and after the phrase; but these danda-s seem to be added secondarily by the scribe himself or, more probably, by another hand, and the color of ink used for the danda-s is different from that of the text.
61 Cf. also NVTT 152,5-6 EM 323,17-18: pratyakṣam hi laukikam vartamanaviṣayam eva. anumanam tu traikalyaviṣayam.
62 Cf. footnote 57 above. For a similar explanation that the distinction of inference from perception is due not only to the distinction of their definitions, but also the distinction of their objects, cf. NVTT 152,3-4 EM 323,15-16: evam taval lakṣaṇabhedad anumanam bhinnam pratyakṣäd darsitam. bhāṣyakāras tu viṣayabhedad api bhedam aha. (a lakṣaṇabhedad anumanam NVTT; lakṣaṇabhedanumanam Em, which seems to be a misprint.)
=

Page #17

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 39
iti-clause, as clearly formulated by Bhattavāgīśvara, relates to the distinction of inference from perception, which is not addressed in 1.1.5. It rather seems likely that the word sutra- in the compound sutrārthaḥ refers to the phrase traikālyagrahaņāt, which corroborates its formal treatment as a sutra in T.
Vācaspati refers to the same idea as Bhattavāgīśvara does, although without explicit mention of the phrase traikālyagrahaņāt, and he clearly ascribes the idea to Vātsyāyana.63 Jayanta, on the other hand, directly quotes the expression in question, but does not specify it as a sūtra.64
A further occurrence of the phrase may be noted. In the second chapter of his PSV, Dignāga takes up the phrase *traikalyagrahanāt (V: dus gsum la 'dzin pa'i phyir) when criticizing the Naiyāyikas' definition of inference presented in 1.1.5.65 However, the Naiyāyika referred to by Dignāga does not appeal to the phrase traikālyagrahanāt in order to distinguish inference from perception as Vätsyāyana does, rather to justify the qualifying element trividham in 1.1.5 and to give additional grounds for the threefold division of inference.66 In his Japanese translation and exposition of the relevant phrase in the PSV, Kitagawa (1965: 378) makes the assumption that Vātsyāyana does not seem to have been aware of the theory of the threefold classification of inference according to the three phases of time, i.e., a classification implied by the phrase trividham.67
As briefly shown above, the interpretation of traikālyagrahaņāt as additionally corroborating the threefold classification of inference is not justified in the extant Nyāya commentaries on 1.1.5, and at the same time the original
63 See footnote 62 above.
**NM I 359,5-6: tad ucyate - trikālavisayam anumānam iti. kasmät. traikalyagrahaņāt.. trikālayukta artha anumanena grhyante. For similar formulation, cf. NV 239,3 on 2.1.39 = (Ruben) 2.1.37: trikalavisayam anumanam, traikalyagrahanād ity uktam, where it remains obscure whether Uddyotakara quotes a sútra or the NBh with ity uktam. For a parallel to the relevant passage of the NBh (cf. footnote 57 above), cf. further NSV(G) 18,11-12: sadvisayam ca pratyaksam. sadasadvisayam anumanam. kasmát. traikālyagrahanät.
65 On Dignāga's refutation of NS 1.1.5, cf. Wezler 1969a, in which the discussion in question is not taken into consideration. A further contribution announced by Wezler (1969a: 836, fn. 1) has not yet been published.
66 CE. PSV(K) 116a6-7: gari yari sria ma dari ldan pa kho na mam pa gsum yin te dus gsum du 'dzin pa'i phyir to, res par gzun ba de ni mi rigs te gari gi phyir to.; PSV(V) 34b8-35al: gari yari sria ma dari ldan pa'i rjes su dpag pa kho na ruam pa gsum du 'gyur te, dus gsum la 'dzin pa'i phyir ro íes pa'i res par gzuri ba de yar rigs pa ma yin te. (cf. also Kitagawa 1965: 567-568); VS(C) 217,10-11 (reconstruction): yac ca pürvavad (V: anumanam eva trividham traikālyagrahaņād ity avadhāraṇam, tad [V: api) na yuktam, yasmāt sarvam trikālavisayam. The phrase traikälyagrahanāt, translated into Sanskrit by Muni Jambuvijayaji, is attested by way of indirect mention by Jinendrabuddhi in his PST (Ms.(B) f. 78v 5): ayam eva traikālyagrahanād iti.
In the context of the interpretation of NS (Ruben) 2.1.35-36 = NS (ED) 2.1.37-38 = NS (EJ) 2.1.38-39, a similar assumption has been made; for modern secondary literature, cf. Wezler 1969b: 192, fn. 10. Cf. also Hattori 1979: 351, fn. (7).

Page #18

40
Journal of Indological Studies, No. 18 (2006)
work of the unnamed early Naiyāyika consulted by Dignāga is deplorably lost to us. However, in spite of the "strong doubt" expressed by Wezler (1969b), the fact that the expression traikālyagrahaņāt is explicitly mentioned by Dignāga as a Naiyāyika's statement additionally supporting the qualification trividham in 1.1.568 may lead us again to consider whether the expression as such played a certain, possibly supplementary, role in relation to 1.1.5 in the early history of the Nyaya school.09 The question also remains whether an additional sūtra was adduced here as such by Vätsyāyana, as most probably selected by Bhattavāgiśvara, whether the phrase is a kind of grahaņakavākya of the NBh, or whether the selection of this expression as a sūtra was secondarily developed in a certain Nyāya tradition.
4. Text of the NBh according to the Trivandrum manuscript
As is well known, Thakur's editions of the Nyāyacaturgranthikā are substantially distinguished from other editions of the four works due to the fact that he was able to gain access to the mss. preserved at the Jaisalmer Jain Bhandar in the form of "complete photographic copies"; in the case of the NVTP it included the extended critical edition up to the first adhyāya when published in 1967, and was succeeded by the monumental publication of the edition of whole work in 1996. In his preface to EM, Thakur states that those "manuscripts offered better readings, filled up lacunae and supplied long passages left out in the published editions," so that he "could solve a number of textual problems." 70 Unfortunately, Thakur provides only scant information concerning the Jaisalmer mss. utilized for editing the Nyāyacaturgranthikā; it is therefore not easy to identify the materials used by him with the mss. known from the published catalogues. It is by no means sure whether he consulted all the mss. of the Nyāya works concerned that are preserved at the renowned Bhandar, or only some of them. However, it can be presumed that his collation of the text of the four classical commentaries on the NS was undertaken on the basis of a single Jaisalmer ms. respectively, if the following statement that was presented in Thakur (1968) with regard to the mss. concerned can be applied to his recent editions: "[T]hey were copied by the same scribe in or around Samvat 1501, the date given at the end of the
The assumption of a corroborative function of the expression presented by the Naiyāyika in the PSV should be differentiated from the position that trividham originally intends the threefold division of inference according to the three phases of time, in regard to which Wezler (1969b: esp. 192-194, 196) raises doubt.
69 For the suggestion that the original meaning of 1.1.5 be related to the threefold division of time, cf., e.g., Ruben 1928: 188, n. 127; Randle 1930: 152; Schuster 1972: 354; Hattori 1979: 350, 351, fn. (7).
70 Cf. Thakur's Preface to Em: vii. For a similar remark on his edition of the NV, cf. Thakur (1968: 380): "My collation of the readings has enabled me to fill up many a blank and to restore the original words of the Vārtika in places of the imaginary ones that crept into the text."

Page #19

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 41
Tātparyaṭīkā MS."71 It can therefore be surmised that other relevant Nyaya mss. remained unused.72 In the following discussion, the variant readings designated as "J" in Thakur's editions ED and EM are abbreviated to JD and JM respectively. Under the above presumption, which still has to be confirmed, I tentatively refer to the variants as being found in one Jaisalmer ms. only, and do not discuss the other scenario, namely, that the variants have been recorded from more than one Jaisalmer ms.
4.1 A different transmission of the text of the NBh
Our present collation of the trisütribhāṣya tentatively suggests that most of the mss. of the NBh available to us, apart from T as well as the Jaisalmer ms. in the form of the variant readings recorded by Thakur, can be divided into two major groups." .73 In the following they are called groups A and B (hereafter designated as MSSA and MSSB respectively). Attention cannot be paid to the various subdivisions of each group and the relation between these subdivisions, since this would be beyond the scope of the present article. Thus, the relation between groups A and B, ms. T and the variants of the Jaisalmer ms. reported in JM and JD will briefly be treated.
In comparing these variants with those of the printed editions, the following four editions, besides EM and ED, have been utilized: the editions by Gangadhara Shastri Tailanga (Varanasi 1896), Phanibhusana Tarkavagisha (Calcutta 1917-1929), Ganganatha Jha (Poona 1936-1945) and Taranatha Nyayatirtha who was responsible for the first ähnika of the first adhyaya of the NBh (Calcutta 1936), respectively designated as 'EG,' 'EPH,' 'E' and 'ET' 74 I do not provide the corresponding page and line numbers in these printed editions; variant readings found in the mss. collectively designated as MSSA and MSSB are not reported diplomatically; minor distinctions, such as sandhi variants, scribal errors or corruptions in individual readings of the
71 Cf. Thakur 1968: 380. For the mention of the Jaisalmer material in a singular form, cf. Thakur's Preface to ED: xiii: "[I]n the preparation of the present edition of the Nyayadarsana and Nyayabhāṣya, I mainly depended on the photocopy of the manuscript received through my late lamented friend Dr. J.S. Jetly."
72It is probable that the mss. utilized by Thakur correspond to Pothi 5, serial no. 67 for the NBh (57 fols.), serial no. 68 for the NV (142 fols.), Pothi 6, serial no. 69 for the NVTT (201 fols.) and serial no. 70 for the NVTP (165 fols.). Cf. Punyavijayaji 1972: 188-189; Jambuvijayaji 2000: 50. Concerning the NBh, another possibly untapped ms. would be the ms. Pothi 65, serial no. 1274(3) (70 fols.), dated samvat 1279. Cf. Punyavijayaji 1972: 356; Jambuvijayaji 2000: 110.
73 There are a few mss. whose readings are difficult to classify on the basis of the "test passage." They are excluded from examination in the present article because they are not of high relevance for determining the relationship of the Trivandrum ms. with other mss. of the NBh.
74 This selection of the printed editions is partially based on the results of the collation of a larger number of printed editions of the NBh prepared by Mr. Christian Ferstl, Ms. Heidrun Jäger and Mr. Gautam Liu, M.A., and also based on their comments on it.

Page #20

42
Journal of Indological Studies, No. 18 (2006)
mss., and the readings pertaining to the subgroups, are not, in principle, taken into account. Accordingly the readings provided below are generalized, unless the reading of a particular ms. or edition is discussed, especially of T.
1. Concerning Ep 1,15 on NS 1.1.1, catasrsu caivamvidhāsu tattvam pari
samāpyate ("And in these four kinds (such as pramana and so forth the true nature is accomplished."), MSSA and T read tattvam, agreeing with EM. Eph reads closely to MSSA: catastsv evamvidhāsu tattvam parisamāpatye.75 MSSB read arthatattvam, agreeing with EG, Ej and Et. In corroboration of tattvam, Thakur refers to Prajñākaragupta's PVBh.76 The reading tattvam is also supported by the NM, though not in an exact quotation, and by the NBhus. JM and Tread cedam for caivam, which is supported by the NBhūs; the variant of JM is not adopted in Ed. In this case, idam would have to be construed with the following, but separated tattvam; such a construal is syntactically unusual. As regards arthatattva, the first member artha- appears to be an extension; the reading lacking artha- is also corroborated by the immediately following question "What is then the true nature?" (kim punas tattuam?).78
2. In the case of Ep 4,6-7 on 1.1.1, năstikaś ca drstantam abhyupagacchan
nästikatvam jahäti ("And a nihilist, inasmuch as he admits a generally accepted) example, abandons (his nihilism (or 'the state of being a nihilist'?)."), the text adopted in Ep agrees with MSSB and the five printed editions mentioned above, whereas MSSA read nästikaś ca drstantam abhyupayan nästikatuam (or nästikatām) jahyāt. The readings abhyupayan and jahyāt in MSSA are supported by JM and JD. T (f. 2r 10) agrees with MSSA except for nāstikatvam: nāstikaś ca drstāntam abhyupayan nāstikyan jahyāt*.79 Interestingly, the NBhūş preserves a conflated text of MSSA and MSSB: It supports jahyāt as
found in MSSA and abhyupagacchan as in MSSB.80 The optative form 75 parisamāpatye in Eph 12,1-2 has to be corrected to parisamäpyate.
" Pramanavārttikabhäsya 401,19--20: catas?su caiuamvidhäsu tattuam parisamāpyate - pramātā prameyam pramanam pramitir iti. Cf. Ep 1, fn. 5 and Em 1, fn. 6; cf. also EPH 12, n. *
77 Cf. NM I 32,6-7: evam ca yad ucyate - pramātā pramanam prameyam pramitir iti catastşu vidhäsu tattvam parisamāpyata iti, tad vyāhanyate. Cf. also NBhuş 580,18-19 = HJJM(1) f. 145v 2: pramāṇopapattau pramātrādibhedasiddhiḥ. tathā coktam - catastsu cedam vidhäsu tattvam parisamāpyata iti. 78 Ep 1,16 = EM 1,13.
For evidence for the secondary derivative "nastikya" formed with the syan-suffix (-ya), cf. Ep 7,3-4 = EM 150,19-20, where năstikyam "anti-Vedic heterodoxy" (cf. Halbfass 1991: 73) is mentioned as a type of bad mental deeds or activities (papātmikā praurttih) causing demerit (adharma). The instance of năstikatva adduced above is the only occurrence in the NBh.
80Cf. NBhūş 64,21-65,1 = HJJM(1) f. 14r 9: nästikaś ca drstāntam abhyupagacchan

Page #21

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 43
of ha 'abandon' (3rd sg.) is paralleled by the corresponding verbal predicate upalabheta that is also in the optative form, more specifically, in a potential sense." 81 The reading abhyupayan is the lectio difficilior 82; the two variants jahyat and nastikyam that contain the more difficult conjunct letters hya and kya, in contradistinction to simpler ha and ka in MSSB, may not be regarded as corrupt.
3. In the case of ED 4,13-14 on 1.1.1, tasya pañcāvayavāḥ pratijñādayaḥ, samuham apekṣyavayavā ucyante ("To this [aggregate of statements (sabdasamuha)] pertain the five members, [namely,] the thesis and so forth, [and they] are called "members" in correlation with the aggregate."), the text given by Thakur agrees with MSSB and all the editions mentioned, whereas MSSA read samuham abhyupetyavayavā ity ucyante. The reading iti in MSSA is supported by JM, though the variant of the Jaisalmer ms. is not adopted in ED. abhyupetya 'after having admitted' seems to be a secondary modification (the motive is unclear, unless it is differently rendered); it is difficult to construe the absolutive with its subject, namely pañcavayavaḥ or pratijñādayaḥ. iti seems to serve as syntactical clarification and should probably be regarded as an addition.
T (f. 2v 2) reads vākyasamuham apekṣyāvayavā ucyante, agreeing with MSSB on two points, apekṣya and the lack of iti. The first member vakya- of the compound vakyasamuha 'aggregate [of parts] that constitutes a statement'(?) should be regarded as an extension; vakyasamüha ("aggregate of statements") is the term employed in Vätsyāyana's exposition of vitanda ("contentious debate") and vāda ("amicable debate").83
4. In the concluding part of the commentary on 1.1.1 which contains a verse allegedly adopted in modified form from Kautilya's Arthasastra,84 ED (5,18-6,3) reads85: seyam anvīkṣiki... prakīrtitā iti. tad idam tattvajñānam nihḥśreyasadhigamaś ca yathavidyam veditavyam. iha tv
nastikatvam jahyat, anabhyupayan kimsadhanaḥ param upalabheta? The reading anabhyupayan in the NBhüş agrees with MSSA, JD and JM.
81Cf. ED 4,7 EM 4,1-2: anabhyupagacchan kimsadhanaḥ param upalabheta? ("If he does not admit [any generally accepted] example, through what means could he refute an opponent?")
82 However, in his gloss on this passage, Uddyotakara employs the nominal form abhyupagama derived from abhyupa-gam, which, one may assume, tends to verify the reading abhyupagacchan found in MSSB. Cf. NV 14,18 EM 16,18: nästikasya ca vyaghato 'bhyupagame 'nabhyupagame va. tad uktam bhāṣya iti.
83 For vitanda, cf. ED 3,21; for vada, cf. ED 5,10-11.
84 Cf. Preisendanz 2000: 226-227 and 228, fn. 31. For the translation of the verse as found in the NBh, cf., e.g., Preisendanz 2000: 227-228. For the translation of the verse as it appears in the Arthasastra, cf., e.g., Halbfass 1991: 27.
85 Cf. Preisendanz 2000: 229, fn. 37.

Page #22

Journal of Indological Studies, No. 18 (2006)
adhyātmavidyāyām ātmādijriānam tattvajriānam, nihśreyasādhigamo 'pavargaprāptir iti ("Therefore this investigative science has been proclaimed (or 'praised') (as ...). Therefore this (above-mentioned) adequate knowledge as well as the attainment of the highest good has to be understood according to the specific science. But here in this science concerned with the Self, adequate knowledge consists in the knowledge of the Self and so forth. The attainment of the highest good consists in the attainment of liberation.").86 Apart from the final iti after apavargaprāptih, this text is supported by MSSA, agreeing with Eph and EM MSSB, on the other hand, read: ... prakirtitā. tad idam tattvajñānam nihśreyasādhigamartham yathāvidyam veditavyam. iha tv adhyatmavidyāyām ātmäditattvajñānam nihśreyasādhigamo 'pavargapräptih, a text which is represented by EG MSSB disagree with MSSA on three points: MSSB lack iti after prakirtitā; they read nihśreyasādhigamartham instead of nihśreyasādhigamas ca®; and they read ātmādi instead of ātmādijñānam, compounding it with tattvajñiānam.88 In EJ and ET, the first two readings of MSSB are preferred. Concerning the problematic last reading, Ej reads ātmaditattvamanam tattvajñanam, but this reading does not have any support from the mss. available to us89, ET favors the version of MSSA, and thus has a conflated text. The iti concluding the commentary on 1.1.1 is favoured only by Eph and EM.90
The MSSA version is evidently supported by the two unmarked quotations in the NV: tad idam tattvajñānam nihśreyasādhigamaś ca yathāvidyam veditavyam (cf. NV 20,12 = EM 21,9; cf. also NVTT 59,1-2 = EM 68,9) and iha tv adhyatmavidyāyām ātmādijñānam tattvajñanam nihśreyasādhigamo 'pavargaprāptir iti (cf. NV 20,20-21 - EM 21,1617). The coordinate structure of tattvajñāna and nihśreyasādhigama is
86 For the exposition and the analysis of this pasage, cf., e.g., Preisendanz 2000: 226229. For another plausible understanding of the structure of the second sentence, cf. Preisendanz 2000: 228, fn. 34: “The following, namely, ... and ..., is to be understood
...."
87 For the syntactical analysis and difficulty) of the sentence containing the relevant expressions, cf. Preisendanz 2000: 228, fn. 34.
88 On the MSSB version of the text, cf. Perry's (1995: 42) critical comments.
89 Cf. Preisendanz 2000: 229, fn. 36. In his edition of the NBh, Laksmana Sastri Jatapathin gives this text using parentheses: "atmaditattuajānam (tattvajriānam)" (cf. NBh(KSS) 7), with a footnote, implying that the text in parentheses has no basis in the mss.?): "O etaccihnamadhyasthapatho năsti."
90Cf. EPH 60, fn. *. Phanibhusana argues that it is necessary to denote the completion of a sutra (samāptisücak), referring to Vācaspati's gloss on iti. Cf. NVTT 59,14: itin sütrasamäptih. iti clearly appears in the corresponding commentary in the NV where the unmarked quotation of, or implicit reference to, the last part of the NBh, i.e., iha tu adhyatmavidyāyām ... 'pavargaprāptir iti (ellipsis by me), is made (cf. NV 20,20-21). ED (cf. 6, n. 2: "ti Omn 0") and Em (cf. 5, fn. 13: "ititi rasti C") refer to the lack of iti in Et. This explicit mention of Er implies that JD, JM and EpH also have iti.

Page #23

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 45
evidently better than the MSSB version, considering the significant role both terms play in 1.1.1 and also indirectly in 1.1.2; the MSSA version should thus be regarded as original. The reading "dhigamasca may have caused a syntactical difficulty in the ms. transmission; for example, disagreement in the grammatical gender, as a result of construing a neuter noun jñānam and a masculine odhigamah with a neuter predicate veditavyam.91 The text ātmāditattvajñanam of MSSB may be the result of an omission of jñānam after ātmādi", due to eyeskip. T (f. 3r 4-5) reads in partial agreement with MSSA: ... pariksiteti tad idan tatvajñānan niśreyasādhigamaś ca yathāvidyam* veditavyam* iha tu atmavidyāyān tatvajñānam ātmādijñānan niśreyasādhigamo pavarg9aprāpti+. The distinctive reading parīksitā instead of prakirtitā will be addressed below in relation to the Jaisalmer ms."2 Besides, the reading atmavidyāyān instead of adhyātmavidyāyām and the different word sequence of tatvajñānam ātmādijñānam* are unique to T and in contrast to the evidence of the NV.93 As regards the reading adhyātmavidyāyām instead of atmavidyāyām,94 it may possibly be regarded as the standardized form on the basis of the corresponding unmarked reference given in the NV; however, the possibility of the loss of adhy- due to eyeskip cannot be ruled out. According to the text transmitted to T, Vätsyāyana again uses the term atmavidya in his commentary on NS (Ruben) 4.2.46 = ET 4.2.47.95
5. For Ep 7,1-2 on 1.1.2, rāgadvesādhikarāc cäsatyersyäsüyāmānalobhada"Cf. Speijer 1886: $ 28, b.), Rem. Cf. also footnote 86 above. 92Cf. no. 4 on page 50 below.
93 The inverse order of words in T amounts to a syntactical distinction from the version in MSSA and the NV: In the T version, the subject (tattvajrianam) comes first and then the predicate (ātmādijñānam); the same sequence can subsequently be observed with nihsreyasādhigamah as subject and apavargaprāptih as predicate (i.e., ). In contradistinction, in MSSA and in Uddyotakara's version, there is the sequence . In order to determine the original reading, a more extensive syntactical analysis of Vätsyāyana's text is required.
* The former term adhyatmavidyā is used by Vātsyāyana in a preceding passage where it refers to the Upanişads (cf. Ep 2,20-3,1 = EM 2,17-18). It is not clear to what extent there is a difference between atmavidyā "science of the self" (cf. Halbfass 1991: 24) and adhyātmavidyā "science concerned with the Self" (cf. Preisendanz 2000: 229). The use of atmavidyā would be in opposition to Uddyotakara's explicit mention of adhyatmavidya in the present context.
9 Cf. T f. 89r 3: atmavidyāddhyātmaśāstram*, which comprises the two phrases atmavidyā (f., nom.sg.) and adhyatmasastram (n., nom.sg.). At the same place, although it reads closely with T, ED (cf. 280,11 on 4.2.47) has adhyatmavidyā-adhyatmaśāstram, presumably a reading of the Jaisalmer ms., because the omission of this reading in ET and Eph is reported in the corresponding critical note of Ep. It is evident that there is no occurrence of atmavidya lacking the prefix adhi in Ep and Em, whereas in T there are two instances where the expression atmavidyā is used to designate the Nyaya system. For the variant atmavidyāśāstram in compound form, cf. ET 1097,3 on 4.2.47, which is also supported by some mss. checked by me.

Page #24

Journal of Indological Studies, No. 18 (2006)
yo dosā bhavanti ("And due to the governance of attachment and aversion, the faults, such as falsehood, envy, deception, greed, etc., arise."), the majority of MSSB reads rāgadveşadhikarāc casuyersyamāyālobhādayo doşā bhavanti.96 rāgadveşādhikarāc is adopted in all printed editions. As for the enumeration of the dosa-s, the text of MSSB is also found in EJ; EG, Eph and Er read cāsatyersyāmāyālobhādayo.97 The text adopted by Thakur is supported neither by the mss. available to us nor by other printed editions. MSSA, on the other hand, read rāgadveşādhikaranās cāsüyersyāmānalobhādayo dosa bhavanti; the reading dhikaranāh (m., nom.pl.) is supported by the Jaisalmer ms. (cf. JM and JD).98 A further distinction of MSSA from MSSB is omānao instead of māyao in MSSB. Though the reading omāyā° is predominantly found in MSSB and accepted by all printed editions except for Em and Ed, the reading manao is compatible with the list of various delusions (moha) given by Vātsyāyana in his commentary on 4.1.3, where māyā is not referred to.99 According to Vātsyāyana, false knowledge (mithyājñāna) is contained in the subdivision of moha, which brings forth both of the psychological elements rāga and dveşa (cf. footnote 101 below).
T reads differently: rāgadvesā+ikaranā dvesyāsüyāmānamadamatsaralobhādayo dosāh prādurbhavanti. dvesyao should be a corruption, which can be corrected, for example, to scersyao. The beginning part of the emended text, i.e., Ścersyāsüyāmāna, corresponds to the reconstructed reading of the Jaisalmer ms. (cf. footnote 98 above). The text of T seems to be an extended enumeration with a view to patterning the dosa-s in pairs of two concrete elements for each "ag
Chattopadhyaya/Gangopadhyaya (1967: 28) render rägadvesādhikärät as "[u]nder the influence of attraction and repulsion," and māyā as "deception." Sudarsanācārya paraphrases adhikarät as due to the predominance" (udrekat) in his Prasannapada (cf. NBh(BBS) 14,36), whereas Jha glosses it with "by their virtue" (tadvaśāt). Cf. EJ 8, fn.
3.
97 The reading of Eph as it might be reconstructed from the relevant critical note in Ep ("māyā for mäna TC") is "cäsatyersyāsüyāmāyālobhādayo"; in fact, this is not found in Eph. However, the reading that can be constructed from the corresponding note in EM ("astyao om māyā for māna CT") is identical with that of Eph. Accordingly, the critical note "asūya Om C" in Ep (cf. p. 7, fn. 2) has to be corrected, for example, to "asuyā Om CT"
98 The variant cersyāsüyamanalobhadayo, which is reconstructible as the text of the Jaisalmer ms. from JD(" dhikaranas cersyao"), is not attested by any of the mss. available to us.
For māna as a subtype of moha, cf. Ep 220,7-8 (cited in footnote 100 below). It should be noted that in Visvanatha's Nyāyasútraurtti māyā is classified into the "attachment side" (râgapaksa). Cf. Er 925,30 on 4.1.3, which enumerates the subtypes of attachment more extensively than the NBh. Cf. also Sinha 1961: 92. Visvanātha's classification would allow for the presence of māya in the enumeration of faults in a textual transmission of the NBh, namely, MSSB

Page #25

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 47
,, 100
gregate." pradur-bhu instead of bhu explicates the causal aspect of the actualization and the "manifestation" of dosa-s; it should probably be regarded as (a secondary) clarification; however, the possibility of the omission through eyeskip over pradur- may not be ruled out.
The reading ragadveṣādhikaraṇaḥ of MSSA is supported not only by the Jaisalmer ms., but also by T. As a secondary testimony for it, the pratika and following gloss in Abhayatilaka's NA should be noted: ragadveṣādhikarana iti na tadaśrayatvam tadadhikaranatvam iha vivakṣitam, api tu tadabhinnatvam. "[The faults] 'having attachment and aversion as their locus': It is not intended here that the state of having them as locus is the state of having them as substrate, but rather the state of being identical with them." 101 Abhayatilaka's gloss provides
100 For another enumeration of dosa-s, cf. Ep 220,3-4 on 4.1.3: tatha ceme manerṣyāsuyavicikitsämatsaradayaḥ. The variant of T enumerates the various faults according to a discernible order, with the exception of mada: (1) irṣya ("envy") and asuya ("malice, jealousy"), belonging to the dueşapaksa ("aversion side"); (2) mana ("self-conceit, pride") belonging to the mohapaksa ("delusion side"), and mada ("conceit"), although it is not referred to in the NBh (see below); (3) matsara ("selfishness, jealousy") and lobha ("greed"), belonging to the ragapakṣa ("attachment side"). According to Vätsyāyana's programmatic scheme, faults that are the eighth object of valid cognition (cf. NS 1.1.9) can be allocated to three types of "aggregate" (trayo rasayaḥ) or "sides, wings" (pakṣaḥ) of psychological, karmically effective states and attitudes, namely, attachment, aversion and delusion. This allocation or distribution of doșa-s is reflected in the nominal enumeration in the passage above. On the threefold classification (trairasya), cf. ED 220,6-8 on NS 4.1.3: tesam doṣaṇam trayo rasayas trayaḥ pakṣaḥ. tatra ragapakṣaḥ kāmo matsaraḥ spṛha tṛṣṇa lobha iti. dveṣapakṣaḥ krodha irsyä asuya droho 'marṣa iti. mohapakṣaḥ -mithyajñānam vicikitsä mänaḥ pramada iti. For an unmarked parallel passage in the NV, cf. NV 424,10-12 on 4.1.3. Cf. also NSV(G) 195,2-3, which classifies mada into the mohapaksa and which, in this regard, corroborates the T version of the enumeration. For a further discussion, cf. footnote 101 below. Cf. also Sinha 1961: 91-94; Junankar 1978: 426.
101 Cf. NA 40,23-24. In his commentary on 1.1.2 (cf. ED 7,1-2), Vatsyāyana singles out twofold dosa-s, namely, attachment (räga) and aversion (dvesa) that are a fixed pair causally preceded by mithyajñana; cf. ED 7,1: etasman mithyäjänäd... ragaḥ... ca dveṣaḥ (éllipsis by me). On the other hand, in Vätsyayana's commentary on 4.1.3 false knowledge (mithyajñāna) is regarded as a type of moha, namely, the third "aggregate" that is evenly correlated with attachment and aversion (cf. also footnote 100 above). He also states that attachment and aversion have delusion as their source (yoni) (cf. ED 221,11), which is associated with his soteriological thought in 1.1.2: tav imau mohayoni rāgadveṣāv iti. Furthermore, both items as a fixed pair are mentioned in his commentary on 1.1.18 where dosa is defined (cf. ED 20,3-4): jñātāram hi rāgādayaḥ pravartayanti punye pape vā. yatra mithyājñānam tatra ragadveṣāv iti. ("Indeed, attachment and the others make the agent of knowledge become active towards good or bad [deeds] (cf. ED 7,2-6; 19,1314). Where there is false knowledge, there is attachment and aversion.") Here Thakur suggests the emendation of ragadveṣamoha iti instead of the reading ragadveṣäv iti (cf. ED 20, fn. 3). If one takes into account Vätsyäyana's analysis of the causal relationship of the three fundamental dosa-s as pointed out above, such an emendation is unnecessary. Vätsyāyana's exposition in 1.1.2 appears to presuppose that fundamental attachment and aversion, for their part, are the basis (cf. adhikarana) for their own concrete varieties and for the concrete varieties of delusion.

Page #26

Journal of Indological Studies, No. 18 (2006)
somewhat convincing evidence for accepting the reading of MSSA, T and the Jaisalmer ms.; moreover, the optical confusion of śca with cca (for example, in Saradă script) should be regarded as one of the major causes for the textual corruption in MSSB. However, further examination of Vātsyāyana's usage of adhikarana and adhikara is required.
The divergent readings adduced and discussed above as representative of the two groups of mss. are only samples of other instances known to us. As already mentioned, the discussion of further variants found in the two traditions of transmission is beyond the scope of the present overview. Although it can be said that in general MSSA and MSSB read rather closely, there are some noteworthy cases where their divergence does not concern trifling variants, but rather affect the basic understanding of the text, as seen especially in the fourth and fifth examples presented above. These instances indicate that the transmission of the text in MSSA enjoys more support from secondary testimonies as well as from the Jaisalmer and Trivandrum mss. than the transmission in MSSB. The designating of MSSA as the "better" group that preserves more original readings is, in my opinion, premature. In the above I have merely adduced some evidence which suggests that these two groups may reflect two separate streams of the textual transmission of the NBh. Further examination beyond the trisütribhāsya is required. It is furthermore to be noted that MSSA play a significant role in evaluating the variant readings found in the Jaisalmer ms. and often support them.102
4.2 The relation of the Jaisalmer and Trivandrum manuscripts
In the previous section, a substantially positive aspect of Thakur's two editions has been brought out, namely, that they allow us to show that the phenomenon of the striking deviations of the Jaisalmer ms. from the printed editions should not be understood as mere evidence for the ms.'s peculiar identity, and that the only apparently isolated variants of this ms. are indeed shared by a whole group of primary testimonia that have not been considered until now, namely MSSA.
In this section, attention will be paid to another aspect of the Jaisalmer tradition, and instances provided which show that some deviations of the Jaisalmer ms., as reported in Thakur's two editions, find, from among the primary witnesses, their only support in the Trivandrum ms. Such instances of mutual agreement, complete or partial, can frequently be noticed in the Trivandrum ms. Although they cannot exhaustively be mentioned within the scope of the present article, a few instances of this agreement should suffice to demonstrate the point.
102 Another tendency that has been observed so far may be pointed out, namely, that the transmission of MSSB is closely reflected in the readings in the printed editions. It is rather probable that most of the printed editions are based upon the mss. pertaining to the recension of MSSB.

Page #27

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 49
1. In the introductory part of the commentary on 1.1.1, there is a salient
deviation of JM and JD from the commonly accepted text: so 'yam prānabhrmātrasya vyavahāraḥ, pramāņenārtham upalabhamānas tam artham īpsan (or abhipsan JD) va jihāsan vā samīhamānas tam artham āpnoti vā jahāti vā. 103 This text as an additional passage is placed between duḥkhahetuś ca and so 'yam pramānārthah.104 T (f. 1r 3-4) has similar additional text at the same place: so yam* prāṇabhnmātrasya vyavahārah pramāne+ārttham upalabhyamānam* samarttham ipsan* jihāsan* vă samihamānas tam arttham āpnoti jahāti veti. This passage concerning the "everyday practice of all (creatures) who breathe" (prānabhrnmätrasya vyavahärah) is located in Vātsyāyana's own exposition of the very first statement (ādivākya) of his work.106 It may be noted that Uddyotakara also briefly refers to the "everyday activity" (lokaurtta) in connection with the same first statement.100 Uddyotakara's mention of the everyday activity can also allude, to a degree, to the relevancy of the discussion in the same context. The possibility of loss of text due to homoeoarchy, namely the similarity of the beginning part so 'yam prä/pra- cannot be ruled out.107 Yet this does not constitute convincing evidence for the additional passage in JD, JM, and T. Even if it represents an original text, the cataphoric usage of the demonstrative pronoun ayam in so 'yam prānabhrmātrasya vyavahārah ("Therefore this (followingl is the everyday practice of all living beings") seems unusual, and it is difficult to correlate it with the preceding passage as to the fourfold objects/objectives (unless it is differently construed).108 The passage in question should tend to be regarded as an insertion of a marginal or interlinear gloss.
2. As compared to Ep 1,14 on 1.1.1,109 sa yenārtham praminoti vijānāti
... yo 'rthah pramīyate jñāyate (ellipsis by me), other printed editions such as Eph, EG, EJ and Et do not have vijānāti and jñāyate. In ED and Em, only vijānāti is recorded as being in Jp and JM, but jñāyate, too, is assumed to be a variant of the Jaisalmer ms. T (f. 1r 6) also reads very closely to it: sa yenārttham* pramiņoti vijānāti ... yo
103 Cf. Ep 1, fn. 3 and EM 1, fn. 3. 104 Cf. Ep 1,10 = EM 1,6. 108 On the designation "ādiväkya," cf. NVTT 3,16, etc. 106 Cf. NV 3,16-17: lokavyttänuvädo vā. sarvah pramätā pramāņenartham avadhārya pravartamanah phalam upalabhata iti lokavsttam tadvākyenānūdyata iti. ("Or (the purpose of the first statement is the confirmation of everyday activity. That is,) every agent of cognition, inasmuch as he takes action after having determined an object by way of a means of cognition, obtains a result. Thus everyday activity is confirmed by this statement.") Perry (1995: 129, fn. 31) considers this gloss by Uddyotakara as probably "recording interpretations of other commentators on the Bhasya."
107 Cf. Ep 1,10-11: so 'yam pramanartho 'parisamkhyeyah. 108 Cf. Ep 1,10 = EM 1,5-6: arthas tu sukham sukhahetus ca duhkham dunkhahetus ca. 109 Cf. also Em 1,10-11.

Page #28

Journal of Indological Studies, No. 18 (2006)
rtthaḥ pramiyate vijñāyate (ellipsis by me). The reading of T and the Jaisalmer ms. suggests the synonymity between pra-mä and vi-jñā.110 This paraphrase of pra-mā with vi-jña and the idea of their equivalence are also reflected in Vātsyāyana's gloss which states that pramiti ("the result of cognition") is equivalent to arthavijñāna ("the cognition of
object"), cognition in the sense of nomen acti111 3. In comparison to ED 4,13-14 on 1.1.1, tasya pañcavayavah pratijñā
dayah ("To this (collection of statements (sabdasamüha)] pertain the five members, (i.e.,) the thesis and so forth."), T (f. 2v 2) reads: pañca bhāgāh pratijñādayo. The reading pañca bhāgāh instead of pañcāvayavāh agrees with JM and JD. Cf. also NV 15,6 = EM 17,2-3: tasya (scil. vākyasya) bhāgā ekadeśā iti. Uddyotakara's paraphrase
indicates that he commented upon the expression bhāgāh. 4. As mentioned before (cf. no. 4 on page 45 above), in the d-pāda
of the verse (cf. Ep 5,19-20 on 1.1.1) allegedly adopted from the Arthasāstra, T (f. 3r 4-5) reads pariksitā instead of prakirtita. This reading agrees with JM, whereas the variant in JM is not adopted in Ep. Discussing the two readings, Preisendanz (2000) adopts the reading parīkṣitā "[wlith some hesitation." T's variant corroborates her suggestion. 112 Uddyotakara, Vācaspati and Udayana are silent on the word in question. It has to be noted that Vācaspati Misra II quotes the verse in question ending with prakirtitā, instead of pariksitā.113 If this reading is original in the NTA, then it has to be supposed that the text of the NBh known to Vācaspati Miśra II or the verse as a well-known saying recollected by him or others would already have had prakirtitā instead of pariksitā by his time, namely by the 15th
century. 114 4.2.1 The case of adhigantavyah in the NBh on NS 1.1.1
There is a passage in Vātsyāyana's commentary on 1.1.1 that poses considerable problems regarding its coherence. After explaining the structure and literal meaning of the first sūtra, Vātsyāyana presents the soteriological interpretation of the "attainment of the highest good": heyam tasya nirvartakam hānam ātyantikam tasyopāyo 'dhigantavya ity etāni khalu 15
110 On mã and praminoti, cf. Preisendanz 2000: 225-226, fn. 22. Cf. further Werba (1997: 310-311, no. 331): pra-ma 'ermessen/kennen'.
111 Cf. Ep 1,15 and EM 1,11: yat tadarthavijnanam sa pramitir iti. tad after yat and the final iti find no support in the mss. available to us; the former does not seem to be necessary, whereas the latter is problematic. (I do not go into this problem here.) 12 For her discussion and other relevant parallels, cf. Preisendanz 2000: 227, fn. 29. 113 Cf. NTA 32,10-11. 114 On the date of Vācaspati Miśra II, cf. Preisendanz (1994: 1-2): "ca. 1420-1490." Cf. also Jha's Preface to the NTĀ: (6)-(7). 115 MSSB omit khalu, as do the printed editions such as EG, Eph, Ej and ET.

Page #29

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 51
catvāry arthapadāni samyag buddhvā nihśreyasam adhigacchati ("One attains the highest good, after having rightly comprehended these four arthapada-s which one should know, namely, heyam ...").116 What is problematic here is the enumeration before iti. MSSA and MSSB agree with all printed editions in reading this part as quoted above and do not indicate any textual problem. Jha (1915: 37-38), for example, translates the corresponding part in the following way: "(a) that which is fit to be discarded (e.g. pain) along with its causes, i.e., ignorance and desire, merit and demerit), (b) that which is absolutely destructive (of pain, i.e., true knowledge), (c) the means of its destruction (i.e., the scientific treatises), and (d) the goal to be attained (i.e., Highest Good)." As Jha clearly suggests, the "soteriologically significant topics" are divided into four by connecting the second element tasya nirvartakam to the first heyam, 117 even though they, at first glance, consist of five elements. On the other hand, Uddyotakara explicitly refers to the four significant topics (catvāry arthapadāni) by virtue of the truly fourfold distinctions such as heyam, hānam, upāyaḥ and adhigantavyah (m.), 118 seemingly bracketing the second element tasya nirvartakam.119 However, he includes it, just as Vātsyāyana does, in his following paraphrase after the first element. 120
Concerning the possible discrepancy between Vätsyāyana and Uddyotakara as to the interpretation of the four relevant matters, Wezler (1984) refers to Vätsyäyana's equation of tattvajñānam with tasyādhigama-upāyaḥ ("a means to the attainment of this (liberation)") in his commentary on NS 4.2.1.121 He then points out Uddyotakara's "contradiction to this statement of the Bhāṣyakära's" (325), namely, his equation of tattvajñāna with hāna
116 Cf. Eo 2,15-16 - EM 2,10-12. On the English equivalent for arthapada, cf. Wezler (1984: 325): "right statements with reference to atthapada in Pali, meaning "'a right or profitable word (often referring to the holy texts)', i.e. a word that is to the advantage of another person." Halbfass (1991: 247), however, takes the expression to refer to a "set of important topics or significant terms"; cf. also Chattopadhyaya/Gangopadhyaya (1967: 8): "human concerns" and "lit. 'the basis of the human end'"; Halbfass (1990: 276): "relevant matters"; Halbfass (1991: 260, fn. 24): "fundamental topics"; Perry (1995: 186): "cardinal entities." On the other hand, Vacaspati paraphrases arthapadani as puruşarthasthanani ("bases of human purpose"), and there "word" (pada) is paraphrased as "basis" (sthana). Cf. NVTT 33,2 = EM 47,9; cf. also Hattori 1979: 336. **Cf. also Chattopadhyaya/Gangopadhyaya (1967: 8). 118 Cf. NV 11,10-11 = EM 14,1-2: heyahānopayadhigantavyabhedac catvary arthapadani samyag buddhva nifreyasam adhigacchatiti. On the masculine of this gerundive, cf. Wezler 1984: 326, fn. 105a.
11 Cf. Perry (1995: 187, fn. 7): "Paksilasvāmin seems here ... to mention a fifth arthapada, 'that which produces it' ..., unless this is to be taken parenthetically" (ellipsis by me).
120 Cf. NY 11,11-12 = EM 14,2-3: heyam iti. heyam duhkham. tasya nirvartakam avidyātrsne dharmādharmāv iti.
121 Ep 259,1: apavargo 'dhigantavyatayā, tasyadhigamopāyas tattvajñānam. Cf. Wezler 1984: 325, fn. 105.

Page #30

52
Journal of Indological Studies, No. 18 (2006)
and of upāya with śāstra 122 As a structural understanding of the iti-clause in the NBh (not in the NV), Wezler (1984) proposes that adhigantavyaḥ should not be understood as an independent element in the enumeration of the "four right statements" (catvāry arthapadāni) starting with heyam, but rather as a nominal predicate relating to four grammatical subjects. 123 Furthermore, as a "necessary correction of Uddyotakara's interpretation," Wezler (1984: 326) suggests that "what is meant by the expression hāna here is not 'means of avoidance,' but 'avoidance' itself." His suggestion seems to presuppose the twofold interpretation of the lyut-suffix (-ana) added to the verb hā.124 The suffix serves to denote an action (bhāva) and also an instrument (karana), forming a nomen actionis and a nomen instrumenti respectively (cf. Astädhyāyi 3.3.115 and 3.3.117). In his NVTP, in fact, Udayana presents the twofold interpretation of hāna, namely, the equation of hāna "means of abandonment" with tattvajñana, on one hand, and of hāna "absolute) abandonment” with apavarga, on the other, obviously in order to reconcile the interpretational discrepancy. 125
The exclusion of adhigantavyaḥ from the enumeration of the four important topics (arthapada), as maintained by Wezler (1984), is also corroborated by the explanation in Bhāsarvajña's NBhūş. 126 It should be noted, furthermore, that the word is omitted in Jp and JM. This resulting reading in the Jaisalmer tradition is again supported only by T among the available mss. of the NBh, and most probably by evidence of the NBhüş.127
Further extremely interesting evidence for the omission of adhigantauyah is provided by Śrīkantha, a commentator chronologically located between Udayana and Abhayatilaka 128; his comments actually illustrate the textual criticism as developed in the medieval Nyāya tradition.
122 NV 11,11-13: heyam duḥkham, tasya nirvartakam avidyātrsne dharmadharmāv iti, hānam tattvajñānam, upāyaḥ śästram, adhigantavyo mokşah. Wezler (1984) assumes that Uddyot akara was not satisfied with the strikingly similar fourfold classification of soteriologically significant topics (caturvyuhatva) as presented in Yoga texts. Cf. Wezler 1984: 325-326. Cf. also Oberhammer 1964: 312-315.
123 The translation Wezler (1984: 325) suggests is as follows: "One has to understand that which is to be avoided, that which brings it forth, [its) absolute avoidance and the means [leading to it." Regarding the passage in question in the NBh, Wezler (1984) does not go so far as to explicitly equate hānam atyantikam with mokşa. Cf. also Wezler 1984: 293, 302, Table. 124 Cf. Werba (1997: 331, no. 374): 'ver/zurücklassen' and 'aufgeben.' 128 NVTP 72,16-17: karanavyutpattim aśrityānena tattvajñanam vivaksitam. bhāvavyutpattyā tu ätyantikapadasamabhivyähärad apavarga ity arthah. Cf. ST 39,23-31. Cf. also Perry 1995: 41 and 42, n. 44.
128 Cf. NBhūş 436,15-16: tac (scil. prameyam) caturvidham heyam tasya nirvartakam hānam atyantikam tasyopaya iti. For an English translation of the relevant passage, cf. Wezler 1984: 327. Wezler (1984) does not refer to the absence of the word adhigantavyah in the NBhūş. 127 Cf. footnote 126 above.
128 Cf. Thakur's Introduction to the ST (cf. xiii): "Hje flourished in western India between Udayanacärya (c. 1070-80 A.D.) and Abhayatilaka Upadhyāya (1263 A.D.)."

Page #31

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 53
une. atra ca heyam ityādi. heyam duhkham tasya nirvartakam avidyatrsne ityādi vārttikam keşu cit pustakeșu na drśyate. tato na bhavaty eveti nāśarikaniyam, tikäkrtā heyam itīty ullekhena grhnatā siddhavad upasthāpitatvāt. ayathābhāşyetyādi. heyam ityādau bhāsye 'dhigantavyavärtäpi na srüyate. värttike tv adhigantavyo moksa ity uktam. ato 'yathābhāsyatā vārttikasya. iyam ca kutah. ucyate. arthānuvädatvāt. bhāsyaparamartha evāyam vārttikakrtā 'nūditah, ananubhāsyākşaravyākhyānam krtam ity arthah.129 ( eveti LDI(1); eva ŚT. b 'yathābhāsyatā vārttikasya LDI(1); 'yathābhāsyavärttikasya ST.) In Udayanal's NVTP (72,8-9) it is said, "and here what is to be abandoned" and so forth. (Opponent: The passage in the Värttika ((NV 11,11-12)] beginning, "what is to be abandoned is pain; what brings forth this pain) is both nescience and desire," is not found in some manuscripts (pustakesu). Therefore (the passage) is definitely not present [in the NV). (Reply:) (This) should not be suspected, because (the passage in the NV) is presented130 as established by the author of the Țikā, inasmuch as he employs the expression "what is to be abandoned" as an allusion to this passage in his NVTT (32,21)]. (In his NVTP (72,10) Udayana says, "there is nonconformity to the Bhāsya" and so forth. (To explain: In the Bhäsya beginning with "what is to be abandoned," there is no mention of (lit.: talk about) "what is to be attained." In the Vārttika, however, it is said that "what is to be attained is liberation." Hence (one may suppose that the Vārttika is not conformable to the Bhāsya. And whence this (nonconformity)? [In answer to this question:) It is said: Because (the passage in the NV (11,13)) is a restatement of the meaning intended in the NBh). This ultimate meaning of the Bhāşya only has been restated by the author of the Värttika. (Thus) an explanation of a word not following the Bhāsya has been given. This is the meaning (of Udayana's remark).
In the quoted passage, Srīkantha makes two text-critical remarks: first the alleged lacuna in some manuscript(s) of the NV, and secondly Uddyotakara's nonconformity to the NBh. Concerning the first point, Srikantha briefly expounds Udayana's pithy statement, in which it is stated that one should not suspect that the passage beginning with heyam is actually absent in the NV. Udayana makes mention of the possibility (upapatti) of the "absence of some writing" (lipyabhāva) and ascribes it to the "fault of a scribe" (lekhakadosa) of some NV manuscript(s). Even though he considers this possibility (upapatti), he affirms the existence of the relevant passage in the NV, turning to Vācaspati's authority as a direct commentator of the 129 Cf. ŚT 39,9-12 = LDI(1) f. 23r 3-5. 130 In place of Srikantha's upasthäpitatvāt, the NVTP has utthäpitatvāt. Cf. footnote 131 below.

Page #32

Journal of Indological Studies, No. 18 (2006)
work. 131 Śrīkantha elaborates on suspicion referred to by Udayana and his refutation of it: The phrases constituting the important topics beginning with "what is to be eliminated is pain" are not found in certain mss. (kesu pustakeşu, pl.), i.e., the "absence of some writing" mentioned by Udayana refers to a lacuna in part of the ms. transmission of the NV. If this interpretation by Srikantha is accepted, it adduces evidence for the fact that Udayana appears to have had access to some manuscript(s) of the NV which contained a lacuna at this point.
Secondly, with regard to Uddyotakara's "nonconformity to the NBh" (ayathābhāsyatā), Udayana does not provide any concrete explanation apart from the vague reference to "restatement of the meaning intended in the NBh).” 132 It remains unclear how unconformable the NV is to the NBh and which precise phrase(s) are concerned. Srikantha clarifies the situation and does not hesitate to point out the problem involved. According to his gloss, Uddyotakara's nonconformity to the NBh consists in the additional statement of "what is to be attained" (adhigantavyah). Srīkantha thus clearly presupposes that the word adhigantavyah is not present in the NBh, but
131 NVTP 72,9-10 = EM 125,25-126,1 = NVTP(BI) 238,7-8 = LDI(2) f. 26v 67: atraca heyam ityädy" anuvādavārttikam nästy evets nāśarkaniyam, tīkākrtā siddhavad utthapitatvät, kvacil lipyabhāvasya lekhakadosesiapy upapatteh. ("heyam ityady LDI(2); heyatvädy NVTP, EM; heyetyädy NVTP(BI). eveti näsarkaniyam LDI(2); evety anāśarkaniyam NVTP, EM, NVTP(BI).) The variants of LDI(2) are not reported in Thakur's two editions of the NVTP. Although Thakur did not specify the ms. "Ahmedabad (A)" (cf. his Preface to NVTP: vii) and the "Ahmedabad Palm-leaf MS" (cf. Abbreviations in EM) utilized for his editions of the NVTP, I currently assume that LDI(2) is identical with his exemplar that is assigned the siglum "A".
132 All three editions of the NVTP read anyatha bhāsyatātparyārthānuvádakatvat with no variants recorded (cf. NVTP 72,10 = EM 126,1 = NVTP (BI) 238,3-4); this reading is also supported by the ms. (ms. no. PM 1491: f. 49r 1-2) preserved at the Adyar Library, Chennai, which is assigned the siglum "M" in Thakur's editions. I owe this information about the reading as found in the Adyar Library ms. to Prof. Preisendanz. Against the text adopted in the printed editions and recorded in the Adyar Library ms., I read ayathābhāsyata tu arthānuvädatvät, based upon the reading of LDI(2) (f. 26v 7), which is supported by some secondary testimonia: SȚ 39,11-12: ayathabhasyetyadi; NA 31,23: ayathābhāsyatā tv ityādi. As an indirect reference to the phrase in Vardhamāna's commentary on the NVTP, cf. NNP 238,7-8: atra bhāsyanuvadatāyam ayathābhāsyatā na yujyata iti varttikam evaitan nastīty asarikyaha - atra ceti. Vardhamana's (fictive) opponent appears to argue in favour of the absence of the corresponding passage in the NV. This argument implies that the opponent justifies the absence of the passage on the ground of Uddyotakara's nonconformity to the NBh, inasmuch as he does not "restate" and confirm the NBh (cf. bhäsyänuvadatā). It could also imply that there was a (historically preceding?) editorial movement toward, or a controversy relating to, intentionally adapting the text of the NV to that of the NBh lacking adhigantavyah, namely, removing the relevant phrases, inclusive of adhigantavyah, from the NV. This presumable movement might have been reflected in the lacuna in some manuscript(s) of the NV reported in the NVTP. Furthermore, Udayana's argument that Uddyotakara "restates" the intention of the NBh (cf. arthānuvädatva), not the NBh itself, might have been effective in invalidating the opponent's argument and securing the presence of the phrase in the NV as it is.

Page #33

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 55
only occurs in the NV.133 This statement by Śrīkantha is in contradiction with the bare fact that most of the NBh mss. available to us transmit the term as part of the four important topics. If Śrīkantha's gloss on the term correctly reflects the problem Udayana was facing, and if Udayana was aware of the textual discrepancy in the sense conveyed by Śrīkantha, we may infer that the evidence of the Trivandrum and Jaisalmer mss., as well as of Bhāsarvajña's indirect reference, provides us in this case with a text of the NBh as it was still existing in the period of Udayana. The additional expression adhigantavyaḥ would have crept into the text of the NBh some time after Udayana, or even Śrīkaṇṭha, most probably under the influence of Uddyotakara's philosophically motivated re-interpretation and modification of the words of his predecessor, Vatsyāyana.
5. Consideration of the textual transmission of the NBh
To conclude the present examination, I would like to offer a summary with some additional remarks. This article is meant to introduce the unique features of the Trivandrum ms. of the NBh, previously in the Paliyam collection, or of what we may call the Kerala tradition of the text of the NBh. In this introductory attempt I have not been able to fully discuss the textual problems of the variant readings and their historical implications; of course, some of them require further analysis and deliberation. However, fundamental text-critical observations show that the Trivandrum ms. often preserves original readings not found in the majority of mss. available to us, or readings which are closer to the original than those provided by them. Furthermore, as shown in Sections 4.2 and 4.2.1, the close affinity of the Kerala tradition to the Jaisalmer ms., together with the support of some of their substantial readings by earlier secondary and independent testimonies allows the hypothesis that the Paliyam ms. and the Jaisalmer ms. belong to a state of the ms. transmission of the NBh which is closer to the original text than the other available mss. Among the latter, MSSA stand out through their frequent agreement with the evidence of the Trivandrum and Jaisalmer mss., and thus, compared with MSSB (and all printed editions except Thakur's), also preserve more original readings of the text. Because of the reasons stated above (cf. pages 25-26), this hypothesis must currently be limited to the transmission of the trisutrībhāṣya.
Apart from the above, the examination of variant readings, as presented in Sections 3.2 and 4.2.1, shows that the text of the NV or Uddyotakara's ideas consequently influenced the textual transmission of the NBh and probably provided some motivation for "correcting" its text, purposely or unintentionally, if the copyist was somehow familiar with the text of the NV. This is why, in my opinion, wherever there are divergent readings in the mss., one 133 This is corroborated by Abhayatilaka. Cf. NA 31,23-25.

Page #34

56
Journal of Indological Studies, No. 18 (2006)
should consider cases of identical wording in the text of the NBh and the NV with caution, bearing this possible influence in mind. Cases of identical wording should therefore be treated differently than explicit references to the NBh or pratika-s in the NV.
Furthermore, as shown in Sections 3.1 and 3.2, the divergency of the text of the NS in the Kerala tradition from the one represented by Vācaspati Miśra I suggests that the transmission of the text of the NS appears to have undergone some modifications already by his time. This naturally causes the suspicion that the text of the NBh utilized by him might also have already become to some extent divergent from the original text. 134 Uddyotakara and Vācaspati often refer to the text of the NBh; their treatment of it should also be investigated with a view to determining their attitude towards divergent commentarial and philosophical traditions which may still be available or lost to us. Careful consideration of their treatment of the text of the NBh is especially important when there are substantial variants in the mss., in independent testimonies and in the texts of the NV and NVTT referring to the NBh, as shown in Section 4.2.1, where one may suspect that the two philosophers' complete silence on their deviation from Vätsyāyana gave rise to confusion in the transmission of the text of the NBh.
6. Bibliography
Abbreviations for the Manuscripts Consulted
ASC(1): Asiatic Society, Kolkata, "Nyāyasūtra," Ms. No. I.M. 613. BHU(1): Benares Hindu University, "Nyāyamañjari," Ms. No. C1015. GOML(1): Government Oriental Manuscripts Library, Chennai, "Nyāyabhasyam," Ms. No. R. 3725. GOML(2): Government Oriental Manuscripts Library, Chennai, "Nyāyasūtratātparyadipika," Ms. No. R. 3405. HJJM(1): Śrīhemacandrācārya Jaina Jñāna Mamdira, Patan/Pāțaņa, "Nyāya Bhusana Sāra Samgraha Vārttika," Ms. No. 10717. LDI(1): Lalbhai Dalpatbhai Institute, Ahmedabad, Srikanthiya Tippana," Photocopy Ms., Register No. 71182.
134 In this connection, it has to be noted that Vācaspati makes suggestive mention of a variant reading of the text of the NBh. In the introductory part of his commentary on NS 1.1.23, he adduces the variant sthānavata eva tarhi as a kuacit pathah different from the sthānavata etarhi quoted by him as a pratīka. Cf. NVTT 204,8-9.= EM 475,14-15: samsayalaksanavatāranaparam bhāsyam sthānavata etarhīti. ... kuacit pāthah sthānavata eva tarhīti. Our current collation attests the both readings, i.e., etarhi and eva tarhi, as the substantial ones.

Page #35

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 57
LDI(2): Lalbhai Dalpatbhai Institute, Ahmedabad, "Tatparyapariśuddhivṛtti," Photocopy Ms., Register No. 71181.
MORI(1): Oriental Research Institute, Mysore, "Nyayamnanjari," Ms. No. C1374. ORIML(1): Oriental Research Institute & Manuscripts Library, University of Kerala, "Nyāyasūtrāni," Ms. No. 974E.
ORIML (2): Oriental Research Institute & Manuscripts Library, University of Kerala, "Nyayasutram," Ms. No. L.1251P.
ORIML(3): Oriental Research Institute & Manuscripts Library, University of Kerala, "Nyāyasūtram," Ms. No. 22615A.
ORIML(4): Oriental Research Institute & Manuscripts Library, University of Kerala, "Nyayasutram Savivaranam," Ms. No. 19866.
ORIML(5): Oriental Research Institute & Manuscripts Library, University of Kerala, "Nyāyasūtratātparyadipika," Ms. No. 14670.
T: Oriental Research Institute & Manuscripts Library, University of Kerala, "Nyayabhāṣyam," Ms. No. 14980A.
Primary Literature
ASTV: Nyayavisärada-nyāyācārya mahopadhyaya Śriyasovijayaganivaraviracitam Aṣṭasahasritatparyavivaranam. Ed. Vairagyarativijaya. Śrīvijayamahodayasuri
granthamālā 15. Pune 2004.
EG: The Nyayasutras with Vätsyāyaṇa's Bhasya and Extracts from the Nyayavarttika and the Tatparyatika. Ed. Gangadhara Sastri Tailanga. 1st ed. Varanasi 1896. Sri Garib Dass Oriental Series 12. 2nd ed. Delhi 1984.
EJ: Srigautamamahāmunipranītam Nyayasutram. (Nyayasutra of Gautama: A System of Indian Logic). Ed. Ganganatha Jha. Poona Oriental Series 58. Poona
1939.
ET: Nyayadarsanam with Vätsyāyana's Bhāṣya, Uddyotakara's Varttika, Vācaspati Miśra's Tatparyaṭika & Viśvanatha's Vṛtti. Eds. Taranatha Nyaya-Tarkatirtha and Amarendramohan Tarkatirtha. Calcutta Sanskrit Series 18-19. 2 vols. 1st ed. Calcutta 1936-1944. Reprint, Kyoto 1982. Reprint, New Delhi 1985.
ED: Gautamiyanyayadarśana with Bhasya of Vätsyayana. Ed. Anantalal Thakur. Nyayacaturgranthika Vol. 1. New Delhi 1997.
EPH: Nyayadarśana (Gautamasutra). Vätsyāyana Bhāṣya. Ed. Phanibhusana Tarkavagisha. 5 Vols. 1st ed. Calcutta 1917-1929. Reprint, Calcutta 1981-1989. EM: Nyayadarsana of Gautama, with the Bhasya of Vatsyayana, the Varttika of Uddyotakara, the Tatparyaṭika of Vacaspati & the Parisuddhi of Udayana. Ed. Anantalal Thakur. Mithila Institute Series, Ancient Text 20. Vaisali, Muzaffarpur 1967.
GSP: Kesavamiśrapranitaḥ Gautamiyasütraprakāśaḥ. Ed. Kishor Nath Jha. Allahabad 1978.
JD: Variant readings recorded in ED.
JM: Variant readings recorded in EM.

Page #36

58
Journal of Indological Studies, No. 18 (2006)
TAV: Tattvartha-Värttika (Rajavartikal of Sri Akalarikadeva. Ed. Mahendra Kumar Jain. Part 1. Moortidevi Jain Granthamala: Sanskrit Grantha 10. 1st ed. Varanasi 1953. 6th ed. New Delhi 2001. NA: Nyāyalarikära (Pañcaprasthānanyāyamahätarkavisamapadavyakhyā). A Commentary on the five classical tests of the Nyāya philosophy of Abhayatilaka Upādhyāya. Eds. Anantalal Thakur and J. S. Jetly. G.O.S. 169. Baroda 1981. NTĀ: Nyāyatattväloka. A Commentary on the Nyayasutras of Gautama by Vācaspati Misra (Junior). Ed. Kishor Nath Jha. Allahabad 1992. NTD: Bhattavägisvarapranītā Nyāyatātparyadīpikā. Ed. Kishor Nath Jha. Allahabad
1979.
NNP: Nyāyanibandhaprakāśa by Vardhamana. In: NVTP(BI). NP: Udayana's Nyāyaparisista with Pancikā of Vāmesvaradhvaja. Ed. S. N. Srirama Desikan. Kendriya Sanskrit Vidyapeetha Series 25. Madras 1976. NBh(KSS): Nyāyadarśanam. Srigautamamunipranitam. Srivātsyāyanamunipranitabhāsyasahitam. Ed. Laksmana Sastri Jatapathin. Kashi Sanskrit Series 43. Benares 1920. NBh(BBS): Mahāmuni Vātsyāyan's Nyāya-Bhāsya on Gautama-Nyāya-Sutras with Prasanınapadā Commentary of Pt. Sudarsanācārya, śāstri. Ed. Dwärikädās Šāstrī. 1st ed., by Sudarsanacharya Sastri Punjabi. Bombay 1922. Reprint, Sudhi Series 10, 1986. Unique ed., Bauddha Bharati Series 38, Varanasi 1998. NBhūş: Srimadācāryabhāsarvajñaprañitasya Nyayasärasya svopajñam vyākhyānam Nyāyabhusanam. Ed. Svāmī Yogindrānanda. Varanasi 1968. NM(M): Nyāyamanjarī of Jayantabhatta with Tippani – Nyāyasaurabha by the Editor. Ed. K. S. Varadacharya. Oriental Research Institute Series 116 and 139. 2 vols. Mysore 1969, 1983. NM(V): The Nyāyamañjari of Jayanta Bhatta. Ed. MM. Gangadhara Sastri Tailanga. Vizianagram Sanskrit Series 8/1. Benares 1895. NV: Nyāyabhāsyavārttika of Bhāradvāja Uddyotakara. Ed. Anantalal Thakur. Nyāyacaturgranthikā Vol. 2. New Delhi 1997. NVTT: Nyāyavārttikatātparyatīkā of Vācaspatimiśra. Ed. Anantalal Thakur. Nyāyacaturgranthikā Vol. 3. New Delhi 1996. NVTP: Nyāyavārttikatātparyaparisuddhi of Udayanācārya. Ed. Anantalal Thakur. Nyāyacaturgranthikā Vol. 4. New Delhi 1996. NVTP(BI): Nyāya-Värttika-Tätparya-Parisuddhi by Udayanachärya, With a gloss called Nyaya-Nibandha-Prakāśa by Varddhamanopädhyāya. Eds. Vindhyesvarī Prasad Dvivedin and Lakshamana Sāstri Drāvida. Bibliotheca Indica Series 205. Calcutta 1911f. NVP: Nyāyabhäsyavärttikatikāvivaranapanjikā (II-V] of Aniruddhācārya. Ed. Anantalal Thakur. Mithila Institute Series, Ancient Text 19. Darbhanga 1969. NSV(G): Nyāyasūtravivaranam of Gambhiravamsaja. Ed. Anandateertha V. Nagasampige. Mysore 1992. PVBh: Pramānavārtikabhāshyam or Vārtikālarikārah of Prajñākaragupta (Being

Page #37

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 59
a Commentary on Dharmakirti's Pramanavārtikam). Ed. Rahula Sankrityayana. Patna 1953. PST: Pramāsasamuccayatīkā by Jinendrabuddhi. On Ms.(B), cf. Steinkellner/Krasser/Lasic 2005. PSV: Pramāsasamuccayavrtti by Dignāga. PSV(K): Tibetan Translation of the PSV. Trans. Kanakavarman (gSer gyi go cha) and Dad pa's ses rab. Peking ed., Vol. 130. No. 5702, ce 93b4-177a7. PSV(V): Tibetan Translation of the PSV. Trans. Vasudhārarakṣita and Sen rgyal. Peking ed., Vol. 130. No. 5701, ce 13a6-93b4. VS(C): Vaiseșikasūtra of Kanāda with the Commentary of Candrānanda. Ed. Muni Sri Jambuvijayaji. G.O.S. 136. Baroda 1982. ŚT: Nyāyadarsane Srikanthatippanakam. (A Commentary on the Major Nyāyatexts) by Srikanthäcārya. Ed. Anantalal Thakur. Calcutta 1986. SDS: Sarva-Darsana-Samgraha of Sāyana-Madhava. Ed. Vasudev Sastri Abhyankar. Government Oriental Series Class A, No. 1. 3rd. ed. Poona 1978. SDS(BI): Sarvadarśana Sarigraha; or an Epitome of the Different Systems of Indian Philosophy. By Madhaváchārya. Ed. Iswarachandra Vidyāsāgara. Bibliotheca Indica Nos. 63 and 142. 1st ed. Calcutta 1853-58. Reprint, Calcutta 1986.
Secondary Literature
Abhyankar, Kashinath Vasudev. 1986. A Dictionary of Sanskrit Grammar. Baroda. Akamatsu, Akihiko. 1989. "Uddiyōtakara no Shiso - NV Kenkyū (1) - tattvainanan nihśreyasādhiqamah" (= A Study of the Nyāyavärttika (1) tattvajñānān nihśreyasādhigamah), Studies in the History of Indian Thought (= Indo-Shisõshi Kenkyū) 6: 67-76. (Japanese) Akamatsu, Akihiko. 2000. "Uddiyötakara no Shiso -- NV Kenkyū (3) - NS 1.1.2 no Gedatsu-ron wo megutte" ( The Thought of Uddyotakara - Study on the NV (3) - with a reference to the theory of salvation in NS1.1.2). In: Akamatsu Akihiko (ed.). Indo no Bunka to Ronri: Tosaki Hiromasa Hakase Koki Kinen Ronbunshū (= Indian Culture and Logic. A Volume in Commemoration of Dr. Hiromasa Tosaki on his Seventieth Birthday). Fukuoka. 667-683. (Japanese) Apte, Vaman Shivaram (ed.). 1957. The Practical Sanskrit-English Dictionary. Revised and enlarged by P. K. Gode and C. G. Karve. Poona. Reprint, Kyoto 1986. Biardeau, Madeleine. 1964. Théorie de la connaissance et philosophie de la parole dans le brahmanisme classique. Le Monde d'outre-mer passé et présent, première série: Études XXIII. Paris. Burnell, Arthur Coke. 1878. Elements of South Indian Palaeography, from the Fourth to the Seventeenth Century A.D., being an Introduction to the Study of South Indian Inscriptions and Mss. Second enlarged and improved edition. London. Chattopadhyaya, Debiprasad and Mrinalkanti Gangopadhyaya (trans.). 1967. Nyaya Philosophy. Literal Translation of Gautama's Nyāya-sutra & Vätsyāyana's

Page #38

60
Journal of Indological Studies, No. 18 (2006)
Bhasya, along with a free and abridged translation of the Elucidation by Mahamahopadhyaya Phanibhuṣana Tarkavägisa. Part I: First Adhyaya. Indian Studies Past & Present. Calcutta.
Franco, Eli. 2002. "A Mīmāmsaka among the Buddhists. Three fragments on the relationship between word and object." In: J. Braarvig, et al. (eds.). Buddhist Manuscripts in the Schøyen Collection. Vol. 2. Oslo. 269-285.
Franco, Eli and Karin Preisendanz. 1995. "Bhavadasa's Interpretation of Mīmāmsăsutra 1.1.4 and the Date of the Nyayabhāṣya," Berliner Indologische Studien 8: 81-86.
Grünendahl, Reinhold. 2001. South Indian Scripts in Sanskrit Manuscripts and Prints. Grantha Tamil Malayalam Telugu Kannada Nandinagari. Wiesbaden.
Halbfass, Wilhelm. 1990. India and Europe: An Essay in Philosophical Understanding. Albany 1988. Indian ed., Delhi.
Halbfass, Wilhelm. 1991. Tradition and Reflection. Explorations in Indian Thought. Albany.
Hattori, Masaaki. 1979. "Ronshō-gaku Nyumon" (= Introduction to the Dialectic). In: Gadjin Nagao, et al (eds.). Baramon Kyōten, Genshi Butten (= Brahmanical Scriptures and Early Buddhist Literature). Originally ed. in 1969. 1st ed. Tokyo. 4th ed. Tokyo 1987. 331-397. (Japanese)
Ikari, Yasuke. 1995. "Vadhula Śrautasutra 1.1-1.4 [Agnyadheya, Punaradheya] A New Critical Edition of the Vadhula Śrautasūtra, I" Zinbun 30: 1-127. Ikari, Yasuke. 1996. "Towards a 'Critical' Edition of the Vadhula Śrautasūtra - A Report on the New Manuscripts," Studien zur Indologie und Iranistik, Festschrift für Paul Thieme 20: 145-168.
Isaacson, Harunaga. 1995. Materials for the study of the Vaiseṣika system. Dissertation (Rijksuniversiteit te Leiden). Unpublished.
Jambuvijayaji, Muni Shree (ed.). 2000. A Catalogue of Manuscripts in Jaisalmer Jain Bhandaras. Delhi.
Jha, Ganganatha (trans.). 1915. The Nyaya-Sutras of Gautama with the Bhasya of Vatsyayana and the Vartika of Uddyotakara. Vol. 1. 1st ed. Indian Thought Series 7. Allahabad. Reprint, Kyoto 1983. Reprint, Delhi 1984, 1999.
Junankar, N. S. 1978. Gautama: The Nyaya Philosophy. Delhi.
Kane, P. V. 1962. History of Dharmasastra (Ancient Medieval Religious and Civil Law in India). Vol. V, Part II. Government Oriental Series, Class B, No. 6. Poona. Kitagawa, Hidenori. 1965. Indo Koten Ronrigaku no Kenkyu. Jinna (Dignaga) no Taikei ( A Study of Indian Classical Logic - Dignaga's System). Tokyo. (Japanese)
Kunjunni Raja, K. 1978. New Catalogus Catalogorum. An Alphabetical Register of Sanskrit and Allied Works and Authors. Vol. 10. Madras.
Kuppuswami Sastri. 1927. A Triennial Catalogue of Manuscripts, collected during the Triennium 1919-20 to 1921-22 for the Government Oriental Manuscripts Library, Madras. Vol. IV, Part I, Sanskrit A. Madras.

Page #39

Some Observations on the Manuscript Transmission of the NBh (Y. Muroya) 61
Maas, Philipp André. 2004. Samādhipăda. Das erste Kapitel des Patañjalayogaśāstra zum ersten Mal kritisch ediert. Dissertation (Rheinische Friedrich-WilhelmsUniversität zu Bonn). To be published in 2006 (cf. footnote 13). Nagasaki, Hojun. 1968. "Pramāņamīmāmsă ni araware taru Shori-Gakuha no Bunken" (=A Study of the PramanamīmāmsāQuotations from Nyāya Works --), The Otani Gakuho, The Journal of Buddhist Studies and Humanities 48.1 (=177): 61-74. (Japanese) Oberhammer, Gerhard. 1964. "Paksilasvāmin's Introduction to his Nyāyabhasyam," Asian Studies 2(3). University of the Philippines, Institute of Asian Studies: 302332 Oberhammer, Gerhard. 1966. "Zur Deutung von Nyāyasūtram 1,1,5," Wiener Zeitschrift für die Kunde Süd- und Ostasiens 10: 66-72. Oberhammer, Gerhard, Ernst Prets and Joachim Prandstetter. 1991. Terminologie der frühen philosophischen Scholastik in Indien. Ein Begriffswörterbuch zur altindischen Dialektik, Erkenntnislehre und Methodologie. Österreichische Akademie der Wissenschaften, Philosophisch-Historische Klasse Denkschriften, 223. Band. Wien Okazaki, Yasuhiro. 2005. Uddoyotakara no Ronri-gaku. - Bukkyo-ronri-gaku tono Soukoku to sono Toutatsu-ten-(=The Logic of Uddyotakara - The Conflict with Buddhist logic and his Achievement --). Kyoto. (Japanese) Perry, Bruce Millard. 1995. An Introduction to the Nyāyacaturgranthika: With English Translations. Dissertation (University of Pennsylvania). UMI No. 9532256. Preisendanz, Karin. 1994. Studien zu Nyayasūtra III.1 mit dem Nyāyatattväloka Vācaspati Miśras II. Alt- und Neu-Indische Studien 46,1. Stuttgart. Preisendanz, Karin. 2000. "Debate and Independent Reasoning vs. Tradition: On the Precarious Position of Early Nyāya." In: Ryutaro Tsuchida and Albrecht Wezler (eds.). Harānandalahari. Volume in Honour of Professor Minoru Hara on his Seventieth Birthday. Reinbek. Preisendanz, Karin. 2005. "The production of philosophical literature in South Asia during the pre-colonial period (15th to 18th centuries): The case of the Nyāyasútra commentarial tradition," Journal of Indian Philosophy 33: 55-94. Punyavijayaji, Muni Shree (ed.). 1972. New Catalogue of Sanskrit and Prakrit Manuscripts. Jaisalmer Collection. L. D. Series 36. Ahmedabad. Randle, H. N. 1930. Indian Logic in the Early Schools. A study of the Nyāyadarsana in its relation to the early logic of other schools. Reprint, New Delhi 1976. Renou, Louis. 1957. Terminologie grammaticale du Sanskrit. Paris. Ruben, Walter. 1928. Die Nyāyasūtra's, Tert, Übersetzung, Erläuterung und Glossar. Leipzig. Schuster, Nancy. 1972. "Inference in the Vaišeşikasütras," Journal of Indian Philosophy 1: 341-395. Sinha, Jadunath. 1961. Indian Psychology. Emotion and Will. Vol. 2. Calcutta. Reprint, Delhi 1986, 1996. Slaje, Walter. 1986. "Niḥśreyasam im alten Nyāya," Wiener Zeitschrift für die

Page #40

62
Journal of Indological Studies, No. 18 (2006)
Kunde Südasiens 30: 163-178. Sowani, V. S. 1920. "The history and significance of upama," Annals of the Bhandarkar Institute 1: 87-98. Speijer, J. S. 1886. Sanskrit Syntax. 1st ed. Leiden. Reprint, Delhi 1988. Steinkellner, Ernst, Helmut Krasser and Horst Lasic. 2005. Jinendrabuddhi's Viśālāmalavati Pramāṇasamuccayaţikā. Chapter 1. 2 vols. Beijing-Vienna. Strauss, Otto. 1930. "Die Anordnung der Reihe in Nyayasutra 1, 1, 2," Journal of the Taisho University 6-7(2): 13-19. In: Friedrich Wilhelm (ed.). Kleine Schriften. Wiesbaden 1983. 304–310. Thakur, Anantalal. 1968. "Textual Studies in the Nyāyavārtika," Wiener Zeitschrift für die Kunde Süd- und Ostasiens, Festschrift für Erich Frauwallner 12-13 (1968/1969): 379-387. Werba, Chlodwig. H. 1997. Verba Indo Arica. Die primären und sekundären Wurzeln der Sanskrit Sprache. Pars I: Radices Primariae. Wien. Wezler, Albrecht. 1969a. "Dignāga's Kritik an der Schlusslehre des Nyāya und die Deutung von Nyayasūtra 1.1.5," Zeitschrift der Deutschen Morgenländischen Gesellschaft, Supplementa I.3: 836-842. Wezler, Albrecht. 1969b. "Die 'dreifache Schlussfolgerung im Nyayasutra 1.1.5," Indo-Iranian Journal 11: 190-211. Wezler, Albrecht. 1984. "On the quadruple division of the Yogaśāstra, the caturvyūhatva of the Cikitsāśāstra and the Page #41

Studies in the History of Indian Thought (15 Nos., 1981-2003) was renamed Journal of Indological Studies in 2005.
September 15th, 2006 Association for the Study of the History of Indian Thought clo Department of Indological Studies Graduate School of Letters, Kyoto University Yoshida-honmachi, Sakyo-ku, Kyoto 606-8501, Japan
Printed by Showado E-press Co. Ltd., Kyoto, Japan All rights reserved. No part of this publication may be reproduced
without permission of the Association

Page #42
Number 18 / 2006 Journal of Indological Studies (New title for Studies in the History of Indian Thought) Articles Masato Kobayashi, Panini's Phonological Rules and Vedic: Astadhyayi 8.2..........1 Yasutaka Muroya, Some Observations on the Manuscript Transmission of the Nyayabhasya ...... ............................... 23 John Taber, Kumarila's Interpretation of Mimamsasutra 1.1.4 ..................... 63 Kengo Harimoto, The Date of Sankara: Between the Calukyas and the Rastrakutas . . 85 Taisei Shida, On the Causal Factor for Validity at the Origination of Cognition: What are the guna and the general cause of cognition in Naiyayikas' paratahpramanyavada?.......... . . . . . . . . . ....113 Muneo Tokunaga, Buddhacarita and Mahabharata: A new perspective ............ List of Contributors Association for the Study of the History of Indian Thought c/o Department of Indological Studies Graduate School of Letters Kyoto University