Vaisesika Vakya And Bhasya

Author(s): Johannes Bronkhorst
Publisher: Johannes Bronkhorst
Catalog link: https://jainqq.org/explore/269577/1

Page #1

1. A long period of time separates Prasastapāda's Padarthadharmasangraha from the oldest kernel of the Vaiseṣika Sutra whose system of thought it aims to explain. Vaiśeşika works were composed in the intervening period, but they have not survived, with the possible exception of Candramati's *Daśapadarthasastra, which survives only in Chinese translation.
THE VAISESIKA VAKYA AND BHAṢYA1
By
Mallavadin's Dvādaśāra Nayacakra, a Jaina work, refers on two occasions to a Vaiseṣika vakya and bhāṣya, that were apparently known to Prasastapāda. The first and most important passage that contains information about these two reads, in the admirable reconstruction of Muni Jambuvijaya :
K 1
JOHANNES BRONKHORST
DNC p. 508-09 and 512-13:
yad api coktam :
vk 1
T1
"vikalpatrayānāśrayād vikalpāntarāśrayaṇāc ca vikalpānupapatteḥ iti na dosaḥ,
niş thāsambandhayor ekakalatvāt |
niṣṭhā kāraṇasāmagryavyāpārakālaḥ prāgasato vastubhavaḥ niṣṭhānam samāptiḥ... / sambandhaḥ svakāraṇasattāsamavāyaḥ / tayor ekakalatvam, svakāraṇasattāsambandha eva niṣṭhākālaḥ, kutaḥ? samavāyasyaikatvāt, yasminn eva kāle pariniṣṭhām gacchat kāryam kāraṇaiḥ sambadhyate, samavayasambandhena ayutasiddhihetunā - tasminn eva käle sattādibhir api, tasmäd apravibhāgāt prāk kāryotpatter asataḥ sadadir anaspado vikalpaḥ/" etad api na ...
asatsambandhaparihārārtham ca niṣṭhāsambandhayor ekakālatvāt ity etad eva vākyam sabhāṣyam prasasto 'nyatha vyacaṣṭe:
"sambandhaś ca sambandhaś ca sambandhau, nisthāyāḥ sambandhau niṣṭhāsambandhau, tayor ekakalatvāt / niṣṭhitam niṣṭhā, kāra
1
I would like to thank A. Wezler and J. Houben, who read an earlier version of this article and made valuable suggestions.
19 Annals, BORI [ A. M.]

Page #2

146
ABORI: Amptamahotsava Volume
kaparispandad vastubhavam apannam avyapadeśyādhāram kāryam nisthitam nistha ity ucyate, tasya svakaraṇaiḥ sattaya ca yugapat sambandhau bhavataḥ / bhāṣyam api pariniṣṭhām gacchad gatam ity etam artham darśayati, vartamănasămipye vartamānavad vā (P. 3. 3. 131) iti / yathā kārakantaram utpadyamānam dṛṣṭam kārakavyāparād vastubhavam apaṇnam avyapadesyadharam nirvṛttam sat svakaraṇaiḥ sattaya ca sambadhyate tatha paṭakhyam /" tad api na...
The phrase nişṭhāsambandhayor ekakalatvāt is here called a vakya'. This vakya is twice explained, in the first and in the second half of the above passage respectively. The second explanation (T1) is, Mallavädin tells us, a reinterpretation of the vakya and its bhāṣya by Prasasta. The first explanation (K 1) must therefore be its bhasya. And indeed, T 1 quotes the words pariniṣṭhām gacchad from the first explanation, and states that it belongs to the bhāṣya.
There is no reason to doubt that Prasasta - or Prasastamati, as Mallavadin and his commentator Simhasuri call him elsewhere (see below) - is the same person as Prasastapāda, the author of the Padarthadharmasangraha. All of these names, and various others, have been used by different authors to refer to the author of the Padarthadharmasangraha." This Prasasta, it is plausible to conclude from the above passage (and the following passage to be considered confirms this), commented both upon the vakya and upon the bhāṣya, which we will jointly refer to as Vakya-cum-Bhāṣya'. As Mallavādin points out, Prasasta felt free, where necessary, to interpret this Vakya-cum-Bhāṣya' the way he considered correct. The quotation from his commentary (T1) makes also clear that he would none-the-less try to show that his interpretation agreed with the bha sya.
Passage K 1 throws some light on the nature of the bhāṣya. It apparently contained and commented upon individual vakyas. A vākya is here - judging by the one vakya we now know - a short nominal sentence, which is explained in the Bhāṣya in normal Sanskrit. This is in no way surprising. Several works of a similar nature-written in the so-called Varttika-style' - are known from the period round the middle of the first millennium C. E., and the term vakya' used to refer to the sutra-like phrases in them is quite
2 The sentence preceding the vakya may be no more than Mallavadin's restatement of what precedes in the Bhasya, and may not be a literal quotaiion.
3 See Chemparathy, 1970.

Page #3

BRONKHORST : The Vaišeșika Vākya and Bhāşya
:
147
common. The example of these works appears to have been the (Vyākaraņa-) Mahābhāşya of Patañjali. We shall return to this subject below.
The second passage in the Drādaśāra Nayacakra that mentions the vākya and the bhāşya reads (p. 516-17):
vastutpattikāle eva iti vākyakārābhiprāyo 'nusrto bhāșyakāraih / siddhasya vastunaḥ svakāraṇaih svasattayā ca sambandha iti prāśa
stamato 'bhiprāyaḥ / This passage does not appear to quote either a vākya or from the bhāşya. It rather sums up the positions expressed in K 1 and 1. The passage sugge. sts at first sight that in Mallavădin's opinion vākya and bhāşya had different authors. (The plural ending of bhāsyakāraih may express respect, and does not necessarily entail that there was more than one bhāsyakāra.) Both these authors held that connection with the universal 'existence? (sattāsambandha) occurs simultaneously with the origination of the thing. This view is contrasted with the one of Prasastamati, who thought that both connection with existence and connection with the own causes' take place when the thing is already there... But is Mallavādin correct in thinking that there were two authors ? Or perhaps : do we understand him correctly?
It would be surprising if there actually were two or more ) authors of the Vakya-cum-Bhasya. As stated above, several works are known that date from around the middle of the first millennium C. E. and that consist of vākyas and their explanations. All known examples, however, have one single author. Indeed, works that display this so-called Vārttika-style' appear to owe their inspiration to the (Vyakarana-) Mahābhāşya, about the authorship of which very different ideas reigned from today.
The evidence from Bhartshari's commentary on the Mahābhāsya shows that we do not of necessity have to interpret Mallavādin's statement to mean that he believed in two or more authors of the Väkya-cum-Bhäşya. Bhartshari, too, speaks about a Vākyakāra and a Bhāşyakāra, be it that he does so while referring to the Mahābhā sya. Yet he appears to have thought that the vākyas of the Mahābhāşya (i. e., its vārttikas') were written by the very person who also explained them in the Bhāșya.
4 See also Halbfass, 1986 : 281 f. 5 See Bronkhorst, 1990; also Lang, 1988. 6 This is argued at length in Bronkhorst, 1990. The main evidence can be summarized
as follows: (i) Bharthari regularly uses the word 'vārttika' to refer to bhäşyapassages, the Yuktidipikā does so on one occasion; (ii) I-ching gives evidence that no distinction was made between vārtikas and the Mahābhāşya (this had already been pointed out by J. Brough ).

Page #4

148
ABORI: Amṛtamahotsava Volume
It is further important to realize that the terms vakya and bhāṣya are not necessarily titles of works, especially not in a work that drew its inspiration from the Mahabhāṣya. Bhartṛhari's Vakyapadiya (ed. Rau, 1.23) speaks, for example, of bha syas, in the plural, and there can be no doubt that portions of the Mahabhāṣya are meant. Bhartṛhari's Mahābhāṣya Dipika, moreover, speaks twice of this bhāṣya', meaning this portion of the Mahabhāṣya' (Ms 9d7, AL 29. 11, Sw 35. 3, CE I. 24, 15-16;- Ms 97a8, AL 278. 19) and once of this whole bhāṣya' (sarvam idam bhāṣyam; Ms. 44d2, AL 135. 22-23, Sw 158. 5, CE IV. 22.17) in the same sense. It is conceivable that Mallavadin, too, when mentioning a vakya and a bhāṣya, meant portions of a work that, as a whole, carried a different name.
Simhasüri mentions two, or perhaps three, Vaiseṣika works: the Kalandi, Prasastamati's Ţikā, and a Țika without further specification;. sometimes Prasastamati is simply referred to as Ṭikākāra' (p. 516, 517). The Vakya-cum-Bhasya, or any other. Vaiśeşika Bhāṣya, is never mentioned by Simhasüri, except in the context of the two passages discussed above, where he follows Mallavadin and where the discussion concerns the differences or agreements between a vakya and its bhäṣya. Is it possible that the Kaṭandi is the same work as the one we call Vakya-cum-Bhāṣya? Several indications support this supposition.
On p. 458 we learn about an opinion that has been rejected (purvapakṣita) in the Katandi and in the Tikā (kaṭandyām ṭikāyām ca). The juxtaposition of these two names creates the impression that the Tika is a commentary on the Katandi. If here too the Tika is Prasastapāda's Ţikā - and there is no reason to believe otherwise the Kațandi can hardly be anything but our Vakya-cum-Bhāṣya.
This conclusion is confirmed by the following. The opinion which, according to Simhasuri, figured as purvapakṣa in the Kaṭandi and its Ṭikā, and which is accepted by Mallavädin, is summarized by the latter in the following words (p. 459): tasmad vikalpānupapatter na sattasambandho 'bhidhānapratyayahetuḥ. It is precisely this purvapakṣa that is answered in K 1, the only passage that is explicitly attributed to the Vakya-cum-Bhāṣya by Mallavadin and Simhasuri. The introductory sentence of K 1, it will be recalled, reads: vikalpatrayānāśrayaṇāc ca vikalpānupapatteḥ iti na doṣaḥ.
Simhasūri ascribes another passage to the Kațandi on p. 499. Since this passage rejects the second alternative introduced in a quoted passage on p. 490-491 and thus fils a lacuna left open in the earlier passage, it seems reasonable to assume that both belong together and formed part of the Kaṭandi. Together they read:

Page #5

K 2
DNC p. 490-491:
vk 2
BRONKHORST: The Vaiseṣika Vakya and Bhāṣya
vk 3
149
na, vikalpänupapatteḥ | [kimi yenaivātmanā sat tenaiva asat, ahosvit āpekşikam sadasa'tvam anyenātmanā mṛdādinā prāk sad ghaṭādi kāryam] ghaṭātmana casat ? na tävad [ yenaivātmanā sat tenaivātmana asat, sadasator vaidharmyat / yad uktam sadasator vaidharmyat karye sadasatta na (VS 9. 12), sattvapratipakṣo 'sattvam ] asattvapratipakṣaś ca sattvam/ sat sopakhyam asan nirupākhyam, [ tayor vaidharmyat ekasmin kärye ] sadasattvam na bhavatity arthaḥ / DNC p. 499:
yad, ucyate saiddharthiyaiḥ upādānaniyamadarśanāt sat kāryam tilatailavat tatkriyadyasattvadarśanad asat, dṛṣṭam tavat [paṭārthitāyām tantūnām evopādānam na tu pāmsvādinām, evam paṭārthaś ca kuvindasya vyāpāro dṛṣṭaḥ, J' itaratha tantupāmsvadiṣv avise ṣaḥ prāg ipi vyāpārābhāvaś ca syat, dṛṣṭā tu kriya [ paṭārtbā kuvindasya tantünām eva copādānam; tasmād upādānaniyamatadarthavyapārābhyām sadasat käryam, ] ubhayaikante doșadarśanāt sad evāsad eva veti cayukta ekāntaḥ, sadasadātmakatvāt kāryasya upādānaniyamaḥ kriya ca yujyate' iti, tan.
āpekṣikam sadasattvam, prag utpatteḥ mrdatmanā sat kāryam ghaṭātmană căsat, niṣpanne 'pi ghate mṛttvadarśanad mrdupādānopapattiḥ, ghaṭātmană căsattvad ghaṭārthakriyopapattir ity evam kila arhata äha / atrottaram
na, asatkāryatvasiddheḥ | Sevam tarhi mrdatmanaḥ kartavyatväbhāväd ghaṭātmanaḥ kartavyatvad asad eva käryam / tasman na prāg utpatteḥ sadasat karyam /
The second half of this quotation is found again on p. 503.
The Värttika-style which characterizes also this passage confirms us in our idea that the Kațandi is indeed identical with the Vakya-cum-Bhāṣya'. One final quotation in the Dvadaśāra Nayacakra that is attributed to the Kațandi shows that this work did not consist exclusively of vakyas and their explanations. This third passage explains a Vaiśeşika sutra:
7 Here and in following quotations from the Nayacakra, square hooks enclose tentative phrases proposed by the editor in footnotes to fill lacunae in the text.
8 What follows is vyäkhya according to Simhasuri.

Page #6

150
ABORİ: Amstamahotsava Volume
K 3
DNC p. 498-499 :
sadasator vaidharmyāt kārye sadasattā na (VS 9. 12), sadasacchabdārthayor virodhād ekasminn eva kārye sadasacchabdayor ekādhikaranabhāvena prayogo năsti, sad eväsat' ity anusandhanan nästy ekādhikaraṇabhāvena iti saptamyabhidhānena darśayati
If the reasoning presented thus far is correct, it follows that all the quotations from Vaise şika works that are identified by Mallavādin or by his commentator Simhasüri, belong to the katandi or to the Tikā written on it by Prasastapada." The Kațandi, furthermore, was then at least partly written in the Vārttika-style, which contains vākyas and bhāsyas..
It seems reasonable to assume that more quotations from the Katandi and from its ȚIkā occur in the seventh Ara of the Dvādaśāra Nayacakra. We might furthermore be tempted to think that all quotations that clearly derive from a Vaiśesika work and that exhibit the Värttika-style, are quotations from the Kațandi. This latter assumption, however, has to be treated with much caution, for the following reason :
We have been able to identify one passage from Prasastapāda's Tikā. in Ţ I, above. Mallavādin rejects the opinion expressed in that passage, saying (p. 513):
tad api na, samavāyikāraṇatvavirodhāt svavacanābhyupagamavirodhau
It appears that this objection had been foreseen by Prasastapāda, for Mallavädin quotes the following reply, which must, therefore, belong to Prasastapāda's Tikā:
T 2 DNC p. 514 :
(samavāyikāranatvanivȚttir iti cet) na, anyatrāsamavä yät / Pyadi tasya ( anyatra samavāyo 'bhyupagamyeta syād ayam dosah, na tu
tathābhyupagamyate, 1 tasmād adoşah / Another objection raised by Mallavādin runs (p. 513):
9 Sinhasūri introduces the explanation with the words; tadvyakhyanam

Page #7

BRONKHORST: The Vaiseṣika Vakya and Bhāṣya
kim ca, nisthitasya karyasya kārapaiḥ sattaya ca sambandho yutasiddhasambandhaḥ, kāryasya kāraṇebhyo 'nyatra pariniṣṭhitatvāt / This objection, too, must have been taken from Prasastapāda's commentary, for it is answered in the following quoted passage:
T 3
DNC p. 516:
151
na, asyasamyogāt / 10na hi kāraṇasambandhibhiḥ karyasya samyogo 'sti /
As is clear from these two quotations, it looks as if Prasastapāda's Tika, too, contained vakyas. In fact, there is no reason to assume that his Tika was written in the Varttika-style. The two vakyas which occur in the above two quotations from his work answer objections, and a short nominal phrase, subsequently explained, in such a position is not to be confused with the consistently used Varttika-style. The latter does not only express the answers to objections in subsequently explained vakyas, but normally also the objections themselves. We do not know whether the Kaṭandikara used this style consistently in each and every case (the above quotations from his work suggest he didn't), yet the way Mallavadin refers to him allows us, at least tentatively, to assign any quotation in full' Värttika-style to his work. In the context of our purpose - identifying quoted portions from the Kaṭandi - this means that there where we have no other indications but the style, we can only be reasonably sure that a passage belonged to the Katandi if both its objection(s) and answer(s) take the form of a vakya plus explanation.
In the case of one such passage we have independent evidence which confirms our belief that it must belong to the Katandi. The passage fulfils our primary requirement that the objection too be expressed in a vakya that is then explained. The extra reason to believe that it derives from the Kațandi is that elsewhere in the seventh Ara Mallavādin appears to quote Prasastapāda's commentary, or a paraphrase thereof, on at least part of this quoted portion. The passage reads:
K 4
DNC p, 486:
athavā vise ṣaṇasambandham antareṇāpi vastumātrāṇām parasparātiśayo 'sti tena
viseṣaṇasambandhaniyamasiddhiḥ /
10 Simhasüri introduces the explanation with the words: tad vyacaṣṭe.

Page #8

152
ABORI: Amstamahotsava Volume
vk 4' katham parasparātiśaya iti cet / katham prāk [ sattāsambandhad
dravyagunakarmaņām parasparato ] atiśayah syat / vk 5 na, drstāntāt | yatbā parapakşe ( sattasambandhăd şte 'pi sattva
rajastamasām parasparato ] atiśayas tathehāpi syāt / vk 6. Sämänyādivad vā yathă sămānyādi svata evästi arthäntarasam
bandhanirapekşam tathā dravyādy api ) svata eva syāt / Simhasűri ascribes the following lines to Prašastamati :
....
DNC p. 462-463 :
na ca tad api nirātmakaṁ Śaśavisāņavat, sattāsambandhad rte 'pi yathā para pakse pradhānādinām sātmakatvam tathehāpi syāt (tvatpakşe drstäntābhāva iti cet, sämānyādivad va, ... Sāmānyadi
vad eva sätmakam na ghatādivat sätmakam Note that vk 6 is repeated in this passage.
The next passage that appears to have been quoted from the Katandi needs some introductory remarks. It was stated above that vk 3 and its explanation are repeated, and refuted, on p. 503. This page, and the ones following it, contain a debate between Mallavädin and an opponent who is, apparently, the author of the Kațandi. In this debate a passage occurs which, even though it has Mallavādin's agreement, is written in the full'. Vārtika-style. This should not confuse us. Mallavādin agrees at times with the purvapakşa of the Kapandi ( see above), and this latter text contained evidently elaborate arguments. It is at least conceivable that Mallavādin borrowed here too a pürvapak sa of his opponent and presented it as his own view. The passage reads : K 5 DNC p. 504-515 vk 7 samarthasya karane 'dhikāraparigrahāt sa iti cet samarthasyaiva
kāryakriyayam [ adhikāraparigrahād upādānaniyamah, tasmād]
asad eva kāryam / vk 8 ekibhāvagatārthasya samarthatvāt satkāryat vam eva / atha katham
tantuturyāder eva kār:ņatvenopādānam pațanirvsttau, na pāmśuvāsyādeh ?11 tasyaiva samarthatvāt tathā tathā pašādikāryam tantuşu
11 The editor thinks that something is wrong with this sentence and proposes, in a note,
the following alternative: atha katham tantuturyadi eva pațanirvettau karaṇam, na pāmáuvāsyadi.

Page #9

BRONKHORST : The Vaiseșika Vākya and Bhāşya -
153:
vartate tatha amśu su tathā pakşmatitirenuparamănuşu tatha turya
dişv api, pämsvādişv api ca, kāraṇakāraṇatvād aņuvat/
Summing up our findings of this section, it can be stated that the seventh Ara of the Dvādaśāra Nayacakra appears to contain a long discussion with a Vaišeşika text called Kațandi and with its commentary, the Țika by Prasastapāda. No other Vaišeşika works would seem to be referred to. The Kațandi, which had a single author whose name is not mentioned, was written in the Värttika-stýle that characterizes a number of works of around the middle of the first millennium C. E., a style in which vākyas and their explanations (bhäşyas ) play a predominant role. Mallavādin, in his discussion with the Kațandi and its commentary, quotes frequently from these two texts. Simhasūri's comments, as well as the recognizable style of the Kațandi, allow us to identify a number of its quoted passages. The fact that Mallavādin left, at least in some cases, the Vārttika-style unchanged, suggests that he, if he changed his quotations at all, did so to a but limited extent.
Before we turn to the next section, which will study the possible link between the katandi and the Padārthadharmasangraha, it will be useful to list here the vākya referred to by Vyomasiva in his Vyomavati (p. 351. 27-28 ), to which attention was drawn by H. Isaacson (1990 : 85): vk 9 pūrvāparādipratyayānāṁ kärane digākhyā
2. We have seen that Prasastapāda wrote a Țikā on the Kațandi. This Tikā, like the work on which it commented, is now lost. Prasastapāda's Padārthadharmasangraha, on the other hand, has survived, and is indeed considered to contain the classical exposition of the Vaise șika system. It seems no more than reasonable to believe that the Padārthadharmasangraha was profoundly influenced by the Kaçandi. The question to be posed in this section is whether traces of this influence can actually be found in the text.
Note first that the seventh Ara of the Dvādaśāra Nayacakra contains at least one quotation which corresponds almost word for word to a passage of the Padārth-dharmasangraha.! Our criteria do not permit us to determine whether this quotation originally belonged to the katandi or to the Tikā, but either way our expectation is strengthened that the Padārthadharmasangraha may owe a great deal to the now lost Katandi.
In the case of the Padārthadharmasangraha our main criterion for identifying a passage as a quotation from the Katandi, is the Värttika-style.
12 See Jambuvijaya's edition of the Dvādaśara Nayacakra p. 524 n. 3,
20 Annals BORI ( A.M.

Page #10

ABORI: Amṛtamahotsava Volume
The Padarthadharmasangraha as a whole is not written in this style, but some passages, usually dealing with the elaboration of rather obscure points of doctrine, are. Those that seem to be unacknowledged quotations from the Katandi will be enumerated and, where necessary, briefly discussed.13
154
The first of these is in a certain way also the most remarkable. It is a vakya along with its explanation. The vakya appears to counter a preceding proposition, which, however, is not found in the Padarthadharmasangraha. The vakya therefore hangs in the air. Its explanation, on the other hand, can be read as a continuation of the exposition that started before the vakya. The only explanation of this extraordinarily strange state of affairs appears to be that Prasastapāda borrowed an appropriate passage from another text, but quoted along with it its introductory vakya, even though that vakya was out of place in its new surroundings. There is no reason to doubt that this, other text was the Kaṭandi.
I reproduce first the preceding passage of the Padarthadharmasangraha, followed by the presumed quotation from the Kațandi:
N p. 69; Ki p. 84; tr. Jha p. 152
K 6 vk 10
atmatvābhisambandhad ātmā / tasya sauksmyad apratyakṣatve sati karaṇaiḥ śabdadyupalabdhyanumitaiḥ śrotrādibhiḥ samadhigamaḥ kriyate väsyādinām iva karaṇānām kartṛprayojyatvadarśanāt / śabdādiṣu prasiddhya ca prasādhako 'numiyate/
na, sari rendriyamanasām ajñatvāt 1 na śarirasya caitanyam ghatādivad bhutakaryatvān mṛte casambhavat / nendriyāṇām karaṇatvāt upahateṣu viṣayāsānnidhye canusmṛtidarśanāt / napi manasaḥ karaṇāntarānapekşitve yugapad alocanasmṛtiprasangāt svayam karaṇabhāvāc ca / pariseṣād ātmakāryatvād ātmā samadhigamyate /
The following passages presumably quoted from the Kaṭandi are identified exclusively by their style:
13 The Padarthadharmasangraha appears to contain one acknowledged quotation from the Kaṭandi, which will be discussed below.
11. Many commentators (Vyomaśiva, Udayana, śridhara, Padmanabha Miśra) try to make sense of this vakya by supplying the word caitanyam from the following sentence for its interpretation.

Page #11

BRONKHORST : The Vaiseșika Vakya and Bhasya
155
K7
N p. 112-113; Ki p. 133-135; tr. Jhā p. 243-245
Sobhanam etad vidhānam vadhyaghātakapakse / sahānavasthanaiakşane tu virodhe dravyajñānānut pattiprasangan katham/ guņabuddhisamakalam apeksābuddhivināśād dvitvavināśe tada peksasya
dve dravye iti dravyajñānasyānutpattiprasanga iti vk 11. Laingikavaj jñānamätrād iti cet syān matam yathā "abhūtam
bhūtasya' (VS 3.1.8) ity atra lingābhāve 'pi jñānamätrăd anumănami tathā guṇavināśe 'pi guņabuddhimatråd dravyapratyayaḥ syad iti .
. vki 12 na, višeşyajñānatvāt na hi viseşyajñānam sārūpyād višeşapasam
bandham antarena bhavitum arhati / tathā cāha sütrakāraḥ samavāyinab śvaityāc chvaityabuddheh svete buddhis te kāryakārapa
bhūte iti na tu laiigikam jñānam abhedenotpadyate tasmad visamo a 'yam upanyasah, na āśutpatteh yatha sabdavad akaśam iti atra to triņi jñānāny aśūtpadyante tathā dvitvādijñānotpattāv.ily adosaḥ / vk 13 vadh yaghātakapakse 'pi samāno dosa iti cet syān matam / nanu
vadhyagbātakapakşe 'pitarhi dravyajñānānutpattiprasangah katham dvitvasāmānyabuddhisamakālam samskārād apeksā - buddhivināśād iti / na, samahajñānasya samskārahetutvāt samühajñānam eva sanskārakāra nam nālocanajñānam ity adoṣaḥ / jñānayaugapadyaprasanga iti cet / syān watam / nanu jñānānām
vadhyaghātakavirodhe jñānayaugapadyaprasanga iti / vk 16 na, avinasyator avasthānapratisedhāt / jñanayauga padyavacanena
jñānayor yugapad utpattir avinaśyatoś ca yugapad avasthānam pratişidhyate na hi vadhyaghātakavirodhe jñānayor yuga pad utpattir avinaśyatoś ca yugapad avasthānam astiti /
K8 N p. 292-94 ; Ki p. 263-65; tr. Jhā p. 620-623 Yk 17 kar manām jātipancakatvam ayuktami gamanāvišeşāt / sarvam hi
kşapikam karma gamanamātram utpannam svāśrayasyordhvam adhas tiryag vāpy aņumātraiḥ pradeśaiḥ samyogavibhāgan karoti sarvatra gimanapratyayo 'visiştah / tasmad gamanam eva sarvam iti /

Page #12

156
ABORI: Amrtamahor sava Volume
•
vk 18 na, vargasah pratyayānuvrtivyāvíttidaršanāt, ihotkşepaņam para
träpak sepa nam ity evamādi sarvatra vargasah pratyayānuvrttivyāvrtti drste / taddhetuḥ sāmānyaviseşabhedo 'vagamyate / teşām udädyupasargaviścsāt pratiniyatadigvisistakāryārambhatvād upala
kşanabhedo 'pi siddhah vk 19 evam api pañcaivety avadhāraṇānupapatih / niskramanapraveśanā.
dişv api vargašal pratyayānuvíttivyāvsttidarsanät / yady utkşepaņādişu sarvatra vargaśaḥ pratyayānuvrttivyāvrttidarśanīj jātibheda işyate evam ca nişkramaņapraveśanädişv api / kāryabhedāt teşu pratyayānuvettivyāvetti iti cet na, utkşepaņādişv api kāryabhedad eva pratyaya nuvrttivyāvsttiprasangaḥ / atha samäne vargaśaḥ pratyayānuvettivyavrttisadbhāve utkşepaņådinām eva jātibhedo na nis
kramaņādinamily atra viś şahetur astiti vk 20 na, jätisarkaraprasargāt, niskramaņādinām jätibhedät pratyayā.
nuvyttivyāvrttau jātisankaraḥ prasajyate / katham/ dvayor drastror ekasmad apavarakad apavarakāntaram gacchato yugapan nişkramanapraveśanapratyayau dřstau tathā dvärapradeśe pravišati nişkrāmatiti ca / yada tu pratisirădy apanitam bhavati tada na praveśanapratyayo näpi nişkramaņapratyayaḥ kintu gamana pratyaya eva bhavati/ tathā nälikäyām vaniśapatrādau patati bahunam drastroam yugapad bhramanapatanapraveśanapratyayā drstā iti jätisan karaprasangah na caivam utk cepaņādişu pratyayasankaro drstah / tasmād uiksepanādinām eva jātibhedā! pratyayānuvsitivyāvștti nişkrama.
ņādinām tu kāryabhedād iti / yk 21 kathani yugapal prat yayabheda iti cet / atha matam / yathā jātisan
karo nästi evam anckakarmasamăveso 'pi nästity ekasmin karmani yugapad drastiņām bhramanapatanapraveśanapratyayāḥ katham
bhavantiti atra brūmaḥ / vk 22
na, avayavāvayavinor digvisistasani yogavibhāgänāni bhedāt / yo hi drastā avayavänäm pärśvalah paryāyeņa dikpradeśaih samy ogavibhāgān paśyati tasya bhramaņapratyayo bhavati / yo hy avayavina ürdhvapradeśa ir vibhāgam adhah samyogam cāveksate tasya patanapratyayo bhavati / yaḥ punar nālikāntardese samvogam bahirdese ca vibhāgam paśyati tasya praveśanapratyayo bhavatiti siddhaḥ
kāryabhedān nişkramaņādinām pratyayabheda iti K9 N p. 140-41; Kip. 1-8; tr. Jhā p. 363-304 vk 23 nästy ajali saniyogo nis yaparimandalarat prthag anabhidhānāt /

Page #13

BRONKHORST : The Vaiseșika Vakya and Bhāşya
157
yathā caturvidham parimāņam utpădyam uktväha nityam parimaņ.. dalam ity evam anyatarakarmajādisaryogam utpădyam uktvā prthan nityam brüyāt / na tv evam abravit / tasmān nästy ajaḥ samyogah /
Before we turn to the last and most important unacknowledged quotation from the Katandi, we briefly consider what may be the only acknowledged quotation from that work in the Padārthadharmasangraha. It is contained in the following passage (Kip: 235, N p. 239, tr. Jhā p. 509-510):
nanu cāyami visesah 'samśayahetur abhihitaḥ sästre 'tulyajātlyeşv arthāntarabhutesu (ca) viśezasyobhayathā drsțatvād' (VS 2. 2. 26 ) iti, na, anyārthatvāt /
K 10
$abde visesadarśanāt samsayānutpattir ity ukte, nāyam dravyādinām anyatamasya višeşah syāc chrāvaratvani kintu sāmānyam eva sampadyate kasmāt / rulyajātiyeșy arthantarabhūteņu dravyādibhedänām ekaikaŝo višesasyobhayathā di statvād ity uktam (v. 1. ukte), na samśayakaranam / anyathā $utsv api padarthcsu samśnyaprasangät / tasmāt samanyapratyaksād (v. 1. - pratyayad ) cva sansaya iti /
Objection: A specific feature ( visesa ) is stated to be a cause of doubt in the Sastra ( in sutra 2. 2. 26 ) :'[With regard to sound there is doubt whether it is a substance, an action, or a quality, 315 because its specific feature (viz., audibility) is found both in objects) that have the same universal, and in other objects.'
[Reply:] (This is not correct), for the sūtra ] has to be interpreted differently. (A specific feature can not be a cause of doubt, ( for the following reason:] Having stated : No doubt arises in the case of sound, for we know its specific feature, it is then stated : Audibility is not the specific feature of any one of the categories ) substance etc. It is, on the contrary, common ( 10 these ). Why? Because in each of the categories ) substance etc., we find the specific feature, both in objects that have the same universal and in other objects. If it were otherwise, there would be doubt even in the case of the six categories. For this reason doubt can come about on the basis of perception of a general feature only.
It is conceivable that Prasastapāda quotes here from another work the words reproduced in italics, and perhaps also the two concluding sentences of this
16 This translates the preceding sutra 2. 2. 25: tasmin dravya din karma guna iti sa risayah.

Page #14

158
-
ABÓRÍ: Amstamahotsava Volume
passage. It is true that the quoted passage does not contain the features of the Vārttika-style, but we have seen that this by itself constitutes no reason to believe it did not form part of the Kațandl.
A confirmation that this passage does indeed derive from the Kațandi is provided by the fact that. Dignāga knows the opinion according to which the specific feature ( viśeşa ) audibility is really a common feature (sāmānya ). In his. Pramāṇasamuccaya he cites and refutes those who say : visesa ubhayaira drştatvād urdhvatvādivat samanyam eva 16 His commentator Jinendrabuddhi specifies that the reference is to some Vaiseşika(s). We shall see below that there is reason to believe that Dignāga knew the Kațandi.
One case remains to be considered.' It differs from the preceding ones in that the lines followed by a more elaborate explanation are not vākyas, but verses, the only two verses that occur in the Padārthadharmasangraha. We shall see that there are independent reasons for believing that these verses: were quoted from an earlier work, and the fact that they are explained the way the vākyas are explained makes it reasonable to assume that they too derive from the Kațandi. The verses, along with their explanations, read :
K11
N p. 200-04; Ki p. 193-95: tr. Jhā p. 421-431
lingam punaḥ yad . (st. i) anumeyena sambaddhani
prasiddhani ca tadanvite tadabhāve ca nāsty eva
tal lingam anumāpakam // (st. ii) vipari tam ato yat syād
ekena dvitayena vā viruddhāsiddhasandigdham
alingan kāśyapo 'bravit //
yad anumeyenārthena deśaviśc se kālaviše se vā sahacaritam anumeyadharmānvite canyatra sarvasminn ekadese vā prasiddham anumeyaviparite ca sarvasmin pramāṇato 'sad eva tad aprasiddharthasyāpumāpakam lingam bhavatiti
16
This is Jambuvijaya's Sanskrit rendering (1961 : 199 ) of the Tibetan translation, which reads, in its two versions : (1) khyad par ni gnyi ga la mthong pa'i phyir
bred ba bzhin du sphyi kho na yin no; and (2) bye brag gnyis ka la mthong pa'i phyir de sphyi nyid yin te.

Page #15

BRONKHORST : The Vaiseșika Vakya and Bhāşya
159
yat tu yathoktāt trirupāl lingād ekena dharmeņa dvăbhyam vā viparitam tad anumeyasyadhigame lingam na bhavatity etad evāha satrakārahaprasiddho 'napadeśo 'san sandigdhaś ca' (VS 3. 1.
10-11) iti There is an obvious problem connected with the first of these two verses : it requires, but does not contain, the relative pronoun yad.17 Is it possible that either Prasastapā da or someone before himn changed the verse, most probably in order to make it agree with his own views ? Some observations support this supposition: .
The first pāda of the verse reads : anumeyena sambaddham. Regarding the word anumeya Masaaki Hattori (1972) has made some interesting observations. Dignāga, he points out, criticizes in his Pramāṇasamuccaya the word sadhya in the context of inference, which he finds used in some unspecified Vaiśesika text, and proposes anumeya instead: Prasastapāda, most probably under the influence of Dignīga, uses the word anumeya throughout.
This piece of information may provide us with the solution of the riddle of the first verse quoted in the Padārthadharmasangraha. It contains the word anumeya in its problematic part, and we may hazard the guess that in its original version it contained the word sādhya instead. Metrically acceptable reconstructions are not difficult to find : lingam sādhyena sambaddham is possible; or, with the relative pronoun yat : yac ca sādhyena sambaddham or the like. Both these reconstructed readings give a satisfactory meaning, as may other reconstructions. It is not, in the present context, necessary to choose the correct reconstruction. The main point is that an original reading may have been changed in order to replace original sādhya with anumeya.
If this reasoning is correct, the katandi must have been written before Dignāga's Pramānasamuccaya. Is it possible to find out more about its date?
The first of the two quoted verses - also in its supposedly original form, with sādhya instead of anumeya - enumerates the three conditions that an inferential mark (linga ) must satisfy. These conditions were laid down in Vasubandhu's Vādavidhi18 and Vädavidhāna, 19 and in the anonymous TarkaŠāstra, which too may have been written by Vasubandhu. We may assume
11 This was already observed by the commentator Udayana. 18 Frauwallner, 1957: 16-17 ( 730-731, ) 33-34 ( 747-748). 19 Frauwallner, 1933: 301 ( 480 ) Fragment 7a. 20. T. 1633, vol. 32. p. 30c 1. 20-21, p. 31a 1. 11 f. : Sanskrit translation in Tucci, 1929
p. 131. 16-17, p. 14 1. 20 f.

Page #16

160
• ABORI: Amstamahotsava Volume
:
that the Kațandi borrowed these conditions from these Buddhist logical texts. Borrowing in the opposite direction is harder to accept. The Tarkaśāstra, Vădavidhi and Vadavidhāna were works in which the discussion of logical issues occupied a central place; the katandi, on the other hand, was primarily a commentary on the Vaise șika Satra, in which logical questions could not but play a sccondary role,
The dependence of the Kațandi upon the Buddhist logicians seems confirmed by the second verse. This verse enumerates three fallacious reasons (alinga ) : viruddha, asiddha and sandigdha. The explanation of the verse, on the other hand, mentions aprasiddha, asat and sandigdha, terms which occur in, and are here quoted in the context of, VS 3. 1.10-11. Where did the author of the Katandi find the terms viruddha and asiddha ? They occur, together with the third term anaikāntika, in the Tarkaśāstra,32 and in the VĀdavidhi. The second verse and its explanation suggest that the author of the Katandi borrowed the two types of fallacious reason called viruddha and asiddha from the Buddhist logicians, but hid this fact by identifying them with ideas already found in th: Vaiśesika Sūtra. Had the new fallacious reasons constituted a development within Vaiseșika, without influence from without, the terms found in the Vaiseșika Sūtra would most probably have been maintained.
It would appear, then, that the katandi was written sometime in the period before Dignāga's Pramāṇasamuccaya, but after the discovery of the three conditions of the inferential mark, which was perhaps made by Vasubandhu, and which it borrowed without acknowledgment..
3. Did Dignāga know the katandi? We have seen that in at least one case Dignāga was acquainted with an opinion which we had reason to ascribe to the Kațandi (K 10). But there is more, and more convincing evidence. On a few occasions Dignāga's Pramāṇasainuccaya Vrtti quotes directly from a Vaiśesika work different from the Vaise sika Sutra. From the beginnings of the third and fourth Paricchedas it is clear that Dignäga knew the following lines (Jambuvijaya, 1961: 197, 201, 207; Hattori, 1972: 169-170):
K 12
sādhyābhidhānam pratijñā / tadvaddharmasya hetuḥ / ubhayaprasiddho drsantaḥ /
81 See further Frauwallner, 1955: 71 ( 208 ) f. 22 T. 1633, vol. 32, p. 36a 1. 7-16: tr. Tucci, 1929: p. 40 1. 10-22. 23 Frauwallner, 1957; 17 (731), 34-35 ( 748-749 ) n. 7,

Page #17

BRONKHORST : The Vaiseșika Vākya and Bhäsya
161
These sentences define some parts of an inference. The first one is also found in Vasubandhu's Vadavidhi ( Hattori, 1972: 172 ), and this is of course most easily explained if we assume that this Vaiseçika work had borrowed from Vasubandhu, as had the Kațandi. What is more, it supports the idea that the work from which Dignāga quotes is the Kațandi.
Further supporting evidence is obtained as follows. The fact that, in matters logical, the Kațandi was strongly influenced by the Buddhist logicians, did not leave much for Dignāga to criticize in it; for he himself continued and enriched the tradition of Buddhist logicians. If Dignāga were to criticize the logical ideas of the Kațandi at all, we might expect this criticism to be directed against two aspects in particular: (i) points in which Dignāga deviates from his Buddhist predecessors; (ii) points in which the Kafandi tries to hold on to Vaiśesika traditions. Both these aspects are represented in K 11, and part of Dignāga's critique might very well be directed against this very passage. The one respect in which K 11 agrees with Dignāga - the three conditions of the inferential mark -- he passes over in silence, as was to be expected. Let us now look more closely at his points of criticism : 24 (i) Dignāga's criticism of the use of the word sādhya is directed as much against his Buddhist predecessors as against the Vaišeşikas. As we have seen, the Vādavidhi defines the proposition (pratijñā ) as : sādhyābhidhānam pratijñā. (ii) The Kațandi, as we have seen, borrowed the fallacious reasons viruddha and asiddha from the Buddhists, but claimed that they are the same as the ones called asat and aprasiddha in the Vaiseșika Sutra. It did not borrow the term anaikāntika, but held on to the Vaiścsika term sandigdha instead. By doing so, it invited criticism directed against the types of fallacious reason enumerated - or presumed enumerated - in VS 3. 1. 10-11. Such criticism is indeed found in Dignāga's Pramāṇasamuccaya. This text cites the sūtras concerned, then points out that none of the possible interpretations of aprasiddha are suitable to denote a fallacious reason. The designation asat is not acceptable either, the correct term - in view of the example in VS 3. 1. 12 (vişāņi tasmād aśvo ... ) - is viruddha. Sandigdha, finally, covers according to Dignāga only what he calls adhāraṇānaikāntika, whereas asādhārana and viruddhāvyabhicärin are not mentioned by the Vaiśeşikas. (Note that Prasastapāda introduces a fourth fallacious reason, anadhyavasita, to cover these two cases.)
It can be seen that Dignāga cites and criticizes the Vaišeşikas in the context of logical theory where the katandi appears to be susceptible to such
24' For a detailed discussion, see Hattori, 1972. For a Sanskrit translation of Dignaga's
criticism of the Vaisesikas, see Jambuvijaya, 1961; 197 f. 21 Annals BORI (A. M. 1

Page #18

162
ABORI: Amstamahotsava Volume
criticism. His criticism, moreover, is confined to these points. We can with certainty conclude from this that Dignāga's Vaisesika opponent agreed with him on certain essential points, most notably on the three conditions of an inferential mark, also mentioned in the Kațandi.
Dignāga characterizes perception as the simple presentation of the object ( vişayālocanamätra ", which is not“ preceded by conceptual constructions ( vikalpapūrvaka ) ".:5 The former of these two expressions (to be precise, the part alocanamātra) occurs in the Padārthadharmasangraha, in the context of what the commentators call nirvikalpaka perception. This does not, as Hattori (1968: 136 n. 4. 10) rightly pointed out, allow us to infer that Dignāga knew the Padārthadharmasangraha. It does, however, lend additional support to the idea that Dignāga knew Prasastapada's main source, the katandi.
We may conclude that a number of different factors - the date of the Kațandi (considered above), the direct quotations in the Pramāṇasamuccaya Vrtti, the nature of Dignāga's criticism of the Vaişeşikas - all support the conclusion that Dignāga knew and reacted against the Kafandi.
At one point tha Pramānasamuscay, Vrtti distinguishes two contradictory opinions, both of which were apparently held by certain Vaiśeşikas. The passage reads, in Hattori's translation (19:8:42):
Some of the Vaišeşikas) consider that the cognition as a resule (phala) is distinct from the pramāņi, the means of cognition. They claim that the contact between sense and object (indriyārthasamnikarsa ) is the means of cognition since it is the specific cause (asādhāraṇakārana) (of perceptual cognition ). But there art others of the Vaišeşikas ) who hold that the contact between soul and mind (atmamanahsamnikarsa ) is the means of cognition since
it is the predominant ( cause (pradhana ). This passage occurs in the section of the Pratyak sapariccheda which deals with the Vaiseşika view of perception; there can therefore be little doubt that
25 Hattori, 1968: 42: Jambuvijaya, 1961 : 170. 26 See Schmithausen, 1970. 29 The Tibetan reads : (1) kha cig ni tshad ma las don gzhan du 'dod de, thun mong ma
yin pa'i rgyu pa'i phyir dbang po dang don du phrad pa tshad mar rtog par byed do. gzhan dag ni gtso bo yin pa'i phyir bdag dang yid du phrad pa tshad ma'o zhes zer ro. (2) kha cig ni tshad ma las' bras du don szhan du 'dod de, thun mong ma yin pa'i rgyu yin pa'i phyir dbang dang don phrad pa tshad mas rtogs par bya'o zhe'o. gzhan dag ni gtso bo yin pa'i phyir bdag dang yid phrad pa tshad ma yin no zher ro.

Page #19

BRONKHORST: The Vaiseşika Vakya and Bhāṣya
indeed different representatives of the Vaiśeşika philosophy are here referred to. This in its turn justifies the conclusion that Dignaga knew several Vaiseşika works, or, at the very least, that the Vaiseṣika work he used contained references to alternative (Vaiśeṣika) views.
163
Dignaga's commentator Jinendrabuddhi ascribes the two views expressed in the above passage to different authors: the first one to Śrāyaska and others, the second one to Ravana and others. The name Śrāyaska appears to be unattested elsewhere. Hattori points out, however, that the view here ascribed to him is found in the Nyaya Sutra and Bhāṣya (1. 1..4; 2. 1. 25-26).
The second view ascribed to Ravana by Jinendrabuddhi is more, interesting in the present context, for it occurs in the Padarthadharmasangraha (Ki p. 184, N p. 186):
samanyaviseradravyagupakarmaviseṣaṇāpekad
atmamanahsanni
karṣāt pratyakṣam utpadyate sad dravyam pṛthivi viṣāņi śuklo gaur gacchatiti /
Since we have come to think that the Padarthadharmasangraha is heavily indebted to the Katandi, and that Dignaga knew the Kaṭandi, it is tempting to think that Ravana' is the name of the author of the Kaṭandi.
This supposition is strengthened by the fact that later sources describe Rāvana as the author of the, or a, Bhasya on the Vaise şika Sutra. In Murari's play Anargharaghava the character Ravana describes himself as Vaise şikaKatandi-pandita. The Bhāṣya mentioned in Udayana's commentary Kiraṇāvali on the Padarthadharmasangraha is ascribed to Ravana by Udayana's subcommentator Padmanabha Miśra. Govindānanda, in his subcommentary on Sankara's Brahmasutra Bhāṣya, mentions a Bhasya of Ravana in the context of the Vaiśeşika philosophy.29
4. The preceding two sections have given us reasons to think that the now lost Kațandi profoundly influenced the Padarthadharmasangraha. This is hardly to be wondered at, in view of the fact that Prasastapāda himself appears to have written a commentary on that combined text. By way of conclusion we must mention the possibility that the Kațandi, or rather its vākyas, may also have influenced the surviving texts of the Vaiścṣika Sutra. Vākyas are hard to distinguish from sutras - both are short nominal phrases - and the Kaţandi constituted a commentary on the Vaise şika Sutra. Someone who
21
See Hattori, 1968: 135; Jambuvijaya, 1961: 174.
29 See Jambuvijaya, 1961: 150 n. 1, and Thakur, 1961: 12 f.

Page #20

164
ABORI: Amṛtamahotsava Volume
would try to extract sutras from manuscripts of the Kaṭandi - which contained sutras, vākyas, and bhaṣyas - would be in danger of mistakenly including some vakyas.
Of course, it would be hard, perhaps impossible, to prove difinitely that the three surviving versions of the Vaiseṣika Sutra derive from a common source which is the Kaṭandi. It must however be recalled that cases of early Indian texts that have at some time of their history been peeled out of a commentary are known.33 All we can do in the remainder of this article is briefly consider two points which, to say the least, do not contradict the assumption that our versions of the Vaiśeşika Sutra do indeed derive from the Katandi. A close study of the available evidence may further support, or disprove, the above assumption. Such a study is however beyond the scope of the present article.
VS 1. 1, 4, in the version of the Vaiśeşika Sütra commented upon by Sankara Miśra, reads:
4. dharmaviśesaprasütad dravyaguṇakarmasāmanyaviseṣasamavāyānām padarthänām sädharmyavaidharmyābhyām tattvajñānān niḥśreyasam
An enumeration of the six categories at the beginning of the Sutra-text seems, as Frauwallner (1984: 37 n. 5) observed, essential. Yet this fourth sutra' is absent from the other two surviving versions of the text. How to explain this? The easiest solution seems to be that it was there, but was not recognized as a sutra. This, of course, is only possible if the sutras were extracted from a work that contained more than only sütras, most probably from a commentary. The fact that 'sutra 4' is much longer than sutras 1-3 may explain that it was not so easily recognized as such.
Supposing now that the sutras were all taken from a commentary on the Vaiseşika Sutra, is there any reason to think that this commentary was the Kaṭandi? The resemblance of sutra 4 to a portion of the Padarthadharmasangraha may constitute such a reason. The following passage from the Padarthadharmasangraha expresses almost the same contents in but slightly differing words (N p. 6-7: Ki p. 4):
30 See Bronkhorst, 1988: 121 f., where it is shown that the first two Kandas of Bhartṛhari's Vakyapadiya were peeled out of the Vṛtti, a commentary whose author - different from Bhartṛhari - is not known. It seems, moreover, that the Yoga sutras were collected by their first commentator, the author of the Yoga Bhaṣya; see Bronkhorst, 1985a.

Page #21

BRONKHORST: The Vaiseṣika Vakya and Bhāṣya
165
dravyagunakarmasāmanyaviseṣasamaväyānām padarthānām sādharmyavaidharmyatattvajñānam niḥśreyasahetuḥ / tac ceśvaracodanabhivyaktad dharmad eva /
It is more than likely to conclude that the fourth sutra' was known to Prasastapāda, and was therefore in all probability part of the Kațandi.
The opinion has been expressed that the above cited fourth sūtra' was not created before, but rather under the influence of, and therefore after, the Padarthadharmasangraha.31 In response to this objection it could be pointed out that there is one major difference between the fourth sutra ' and its corresponding passage in the Padarthadharmasangraha: the latter adds God (iśvara). God played henceforth a central role in the Vaiseṣika system. The fact that the fourth sūtra' - like all the other Vaiseşika sutras - ignores God, can be taken as an indication that the fourth sūtra' is older than Prasastapāda, and was not composed under the influence of his Padarthadharmasangraha. 33
There is a second indication that the surviving versions of the Vaiseşika Sutra may derive from the Kațandi. It is the use of the term Āhnika to designate the sections into which the Sūtra-text is divided. This term - which means daily, hence what may be studied on one day' is primarily used to designate the subdivisions of the Mahabhāṣya. As such it has nothing to do with the sutras of the Aṣṭādhyāyi, on which the Mahābhāṣya comments. Works that imitate the style of the Mahabbaṣya may also imitate its division into Ahnikas. And indeed, the Nyaya Bhasya, which is partly written in Värttika style (Windisch, 1888: 15 f.), is divided into Ahnikas. Also the Nyaya Sūtra is divided into Ahnikas, but this division is obviously secondary and derives from the Bhāṣya. In the case of the Vaiśesika Sutra we have come to think that it had a commentary that imitated the style of the Mahābhāṣya. We also know that the Sutra is divided into Ahnikas, in each of its three surviving versions. Nothing seems more natural than to assume that this division, here too, is secondary, and derives from the Kațandi, just as the three versions of the Vaiścşika Sutra themselves derive from the Kaṭandi.
31 This was Frauwallner's opinion (1984: 39-40).
See Thakur, 1957: (16).
83 On the provenance of God in the Vaiseṣika system, see my forthcoming article "God's arrival in the Vaiseṣika system ".
31 Adhyayas 8, 9 and 10 are not divided in ähnikas in the version known to Candrānanda, and in that known to the author of the Sarvadarśanasangraha (Thakur, 1961: 21). Also the version of Adhyayas 9 and 10 found and discussed by Thakur (1966) does not divide these Adhyayas into Ahnikas. The other versions do.

Page #22

166
ABORİ : Amệtamahotsava Volume
Is it conceivable that all non-authentic sütras in the surviving versions of the Vaiśesika Sutra derive from the Kațandi? Or do we have to assume also other sources of inauthentic sūtras? It is difficult to answer this question, because the katandi is almost completely unknown to us. Nor do we know the original contexts of inauthentic sūtras, even if we suppose that we are at all able to recognize them as such.
We do, however, know some things about the Katandi. We have seen, for example, that its logic stood most probably under the influence of a Buddhist logician, most probably Vasubandhu, but not yet under that of Dignāga's Pramanasamuccaya. This helped us in determining the approximate date of the Kațandi. It will now help us to show that at least some sūtras were added to the text of the Vaiścsika Sutra before the Kagandi. 35
VS 2.1. 15-16 and 3. 2. 6-7 distinguish two kinds of inference: that based on something seen (drsta ), and that based on something seen in general (sāmānyato drsta ). This cannot but be the same distinction as that between visesato drsta and sāmänyato drsta, current in Sāmkhya, and introduced by the Samkhya teacher Vindhyavāsin.: Vindhyavasin lived around 400 C. E. (Bronkhorst, 1985: 171 ). These sūtras, therefore, appear to have been inserted after that date, but before the katandi which represents the next stage in the development of logic within the Vaiseșika school.
References Bhartrhari : Mahābhāyadipikā. 1) Edited by K. V. Abhyankar and V. P. Limaye. Poona : Bhandarkar Oriental Research Institute. 1970. ( PostGraduate and Research Department Series No. 8.) 2) Partly edited by V. Swaminathan under the title Mahābhāsya Tikā. Varanasi : Banaras Hindu University. 1965. (Hindu Vishvavidyalaya Nepal Rajya Sanskrit Series Vol. 11.) 3) Manuscript reproduced. Poona : Bhandarkar Oriental Research Institute. 1980. 4) Critical edition'. Poona : Bhandarkar Oriental Research Institute. Ähnika 1, edited and translated by Johannes Bronkhorst, 1987: Ähnika 2, by G. B. Palsulc, 1988; Ahnika 3, by G. B. Palsule, 1983; Āhnika 4, by G. V. Devasthali and G. B. Palsule, 1989; Ahnika 5, by V. P. Limaye, G. B. Palsule and V. B. Bhagavat, 1984; Āhnika 6 part 1, by V. B. Bhagavat and Saroja Bhate, 1986: Ahnika 6 part 2, by V. B. Bhagavat and Saroja Bhate, 1990; Ähnika 7, by G. B. Palsule and V. B. Bhagavat, 1991.
35 What follows is essentially based on Frauwallner, 1955: 75 (212) f., esp. 79 (216) n. 30. JU According to Kumarila's Slokavårttika 8 (Anumanapariccheda ), v. 143.

Page #23

BRONKHORST : The Vaiseṣika Vakya and Bhāṣya
Bharthari: Vakyapadiya. Critical edition by Wilhelm Rau. Wiesbaden: Franz Steiner. 1977. (Abhandlungen für die Kunde des Morgenlandes XLII, 4.)
167
Bronkhorst, Johannes (1985): "On the chronology of the Tattvartha Sūtra and some early commentaries." Wiener Zeitschrift für die Kunde Südasiens 29, 155-184.
Bronkhorst, Johannes (1985a): "Patanjali and the Yoga sūtras." Studien zur Indologie und Iranistik 10 (1984), 191-212.
Bronkhorst, Johannes (1988): "Études sur Bhartṛhari, 1. L'auteur et la date de la Vrtti." Bulletin d'Etudes Indiennes 6, 105-143.
Bronkhorst, Johannes (1990): "Varttika." Wiener Zeitschrift für die Kunde Südasiens 34, 123-146.
Chemparathy, George (1970): "Prasastapāda and his other names." Indo-Iranian Journal 12, 241-254.
Frauwallner, Erich (1933): "Zu den Fragmenten buddhistischer Logiker im Nyāyavārttikam." Wiener Zeitschrift für die Kunde des Morgenlandes 40, 281-304. Reprint: Kleine Schriften (Franz Steiner, Wiesbaden, 1982) pp. 460-483.
Frauwallner, Erich (1955): "Candramati und sein Daśapadarthaśāstram." Studia Indologica. Festschrift für Willibald Kirfel. Bonn. (Bonner Orientalistische Studien, 3.). Pp. 65-85. Reprint: Kleine Schriften (Franz Steiner, Wiesbaden, 1982) pp. 202-222.
Frauwallner, Erich (1957): "Vasubandhu's Vädavidhi." Wiener Zeitschrift für die Kunde Süd- und Ostasiens 1, 104-146. Kleine Schriften (Franz Steiner, Wiesbaden, 1982) pp. 716-758.
Frauwallner, Erich (1984): Nachgelassene Werke. I. Aufsätze. Beiträge, Skizzen. Herausgegeben von Ernst Steinkellner. Wien: Verlag der Österreichischen Akademie der Wissenschaften. (Österreichische Akademie der Wissenschaften, philosophisch-historische Klasse, Sitzungsberichte, 438. Band. Veröffentlichungen der Kommission für Sprachen und Kulturen Südasiens, Heft 19.)
Halbfass, Wilhelm (1986): " Mallavadin and early Vaiseṣika ontology." Adyar Library Bulletin 50 (Golden Jubilee Volume), 271-286.
Hattori, Masaaki (1968): Dignaga, On Perception, being the pratyakṣapariccheda of Dignaga's Pramāṇasamuccaya. From the Sanskrit frag

Page #24

168
ABORI: Amstamahotsara Volume
ments and the Tibetan versions translated and annotated. Cambridge, Massachusetts: Harvard University Press,
Hattori, Masaaki (1972): "Prasastapāda and Dignāga : a note on the development of the Vaise sika theory of anumāna." Wiener Zeitschrift für die Kunde Südasiens 16, 169-180.
Isaacson, H. (1990): A study of early Vaišesika. The teachings on perception. Groningen: Unpublished thesis.
Jambuvijaya, Muni (ed.) (1961): Vaisesikasutra of Kanāda, with the commentary of Candrānanda, Baroda : Oriental Institute. (Gaekwad's Oriental Series, No. 136.)
Jhā, Gargā nåtha (tr.) (1915): Padārthadharmasangraha of Prasastapāda, with the Nyāyakandali of Sridhara. Varanasi - Delhi : Chaukhambha Orientalia. (Chaukhambha Oriental Studies, 4.) 1982.
Lang, Karen (1988): "On Aryadeva's citation of Nyāya texts in the *Sataka." Wiener Zeitschrift für die Kunde Südasiens 32, 131-140.
Mallavādin : Dvādaśara Nayacakra. Edited, with the commentary Nyāyāgamānusăriņi of Simhasűri, by Muni Jambuvijaya. Bhavnagar : Sri Jain Atmanand Sabha, 3 volumes. (Sri Ātmānanda Jaina Granthamāla Serial No. 92, 94, 95.) 1966, 1976, 1988.
Mesquita, Roque (1980): "Yamuna's Vedanta and Pancarätra: a review." Wiener Zeitschrift für die Kunde Südasiens 24, 199–224.
Nozawa, Masanobu (1976): "The Vaišeşikasūtra referred to in the Padharthadharmasangraha." Journal of the Buddhist Studies 24, (32)-(38) ( = 1006-1 00).
Prasastapāda : Padārthadharmasangraha. 1) Edited, with the commentary Kiranivali of Udayanācārya, by Jitendra S. Jetly. Baroda : Oriental Institute. 1971. 2) Edited, with the commentary Nyāyakandali of Sridhara, by Vindhyesvari Prasad Dvivedin. Reprint. Delhi : Sri Satguru Publications. (Sri Garib Dass Oriental Series, 13.) 1984. 3) Edited, with the commentaries Sakti, Setu, and Vyomavati, by Gopinath Kaviraj and Dhundhiraj Shastri. Second edition. Varanasi : Chaukhamba Amarabharati Prakashan. (Chowkhamba Sanskrit Series, 61.) 1983.
Schmithxusen, Lambert (1970): “Zur Lehre von der vorstellungsfreien Wahrnehmung bei Prasastapid.” Wiener Zeitschrift für die Kunde Südasiens 14, 125-129.

Page #25
BRONKHORST : The Vaisesika Vakya and Bhasya . . 169 Sucaritamisra : Kasika. In: The Mimamisaslokavarttika with the commentary Kasika of Sucaritamisra, edited by K. Sambasiva Sastri. Trivandrum 1926 ff. (Trivandrum Sanskrit Series 90, 99, 150.) Thakur, Anantalal (ed.) (1957): Vaisesikadarsana of Kanada, with an anonymous commentary. Darbhanga : Mithila Institute of Post-Graduate Studies and Research in Sanskrit Learning. Thakur, Anantalal ( 1961): " Introduction." = Jambuvijaya, 1961 : 1-23. Thakur, Anantalal ( 1966 ):"Studies in a fragmentary Vaisesikasutravstti." Journal of the Oriental Institute, Baroda, 14 (1965/66 ), 330-335. Tucci, Giuseppe (1929): Pre-Dinnaga Buddhist Texts on Logic from Chinese Sources. Second edition. Madras : Vesta. 1981. Vaisesika Sutra. Edited by Muni Jambuvijaya. Baroda : Oriental Institute. (Gaekwad's Oriental Series, 136.) 1961. Windisch, Ernst (1888): Uber das Nyayabhashya. Leipzig : Alexander Edelmann. Abbreviations AL CE DNC к Ki Ms Mahabhasyadipika of Bhartshari, ed. Abhyankar / Limaye Mahabhasyadipika of- Bhartshari, critical edition Dvadasara Nayacakra of Mallavadin presumed passage from the Katandi Padarthadharmasangraha, ed. Jetly Manuscript of Bhartshari's Mahabhasyadipika Padarthadbarmasangraha, ed. Dvivedin Paninian sutra Mahabhasyadipika of Bhartshari, ed. Swaminathan Taisho edition of Buddhist canon in Chinese presumed passage from Prasasta pada's Tika on Katandi vakya Vaise cika Suira. Sw Vk VS 22 Annals, BORI (A, M.)