Concluding Verses Of Bhartrharis Vakya Kanda

Author(s): Ashok Aklujkar
Publisher: Ashok Aklujkar
Catalog link: https://jainqq.org/explore/269290/1

Page #1

THE CONCLUDING VERSES OF BHARTṚHARI'S VĀKYA-KÄNDA*
By
ASHOK AKLUJKAR
1.1 In this paper I wish to offer some observations on verses 4814901 appearing at the end of the second book of Bhartṛhari's Trikāṇḍi or Vakyapadiya. The verses have been studied, primarily or incidentally, directly or indirectly, in a number of publications: Goldstücker (1861), Weber (1862), Kielhorn (1874, 1875, 1876, 1885), Peterson (1885), Thieme (1956), Yudhiṣṭhira Mimāmsaka (samvat 2020), Sharma (1968), Upadhyāya (1968), S. Iyer (1969), Scharfe (1976), Joshi (1976), and Cardona (1977; in this volume). My objective here is neither to review what has been said about them, nor to pronounce judgements on all the controversies they have given rise to. I wish rather to put forward a few considerations that have not so far appeared in print and to refute a few interpretations that have so far gone unrefuted.
1.2 In order to reach the goal I have set for myself, I shall naturally need a critically established text of verses 481-490. Hence I shall proceed
* This is an enlarged and significantly revised version of a paper I read at the 1972 annual meeting of the American Oriental Society. I am very grateful to Professor Wilhelm Rau for the access I had to the typescript and proofs of his critical edition of the Vakyapadiya | Trikandi-kärikäs and for the copies of Tika manuscripts that he so promptly provided. Professors K. A. Subramania Iyer and M. A. Mehendale exerted themselves considerably to make available to me a copy of the Vakya-kaṇḍa-vṛtti manuscript at Patan. I am greatly indebted to them. To Professor D. H. H. Ingalls goes the credit of making me think more about verse 487. The financial support necessary for the acquisition of manuscript copies etc., so vital to research of the present type, was given by the University of British Columbia, the Canada Council, and the American Council of Learned Societies at various stages during 1969-1975.
1. (a) In the present and following publications I shall follow Rau's (1977) enumeration of the Trikandi karikas. It is the only flawless enumeration we have at present that enables us to refer to a tradition of the Trikandi text (the karika manuscript tradition) in a form determined by objective textual criticism. It will be highly convenient if the Trikandi text as preserved in the other (Vrtti and Tika) traditions is critically established by following Rau's enumeration. This I advocate simply as a procedure that will facilitate future text-critical research concerning Bhartṛhari. I do not hold that the karika manuscripts give us the oldest accessible form of the Trikandi text. See Aklujkar 1971, 1978.
(b) The text of verses 481-490 given below is based on a consideration of all known manuscript traditions. In the case of kärikā manuscripts I have simply followed Rau's lead. It is only the collection and evalution of the evidence of the Vrtti and Ţika manuscripts that I have freshly attempted.
2 Annals [D. J.]

Page #2

ABORI: Diamond Jubilee Volume
by presenting those verses as they will appear in my proposed edition of the Trikandi text. For the sake of simplicity of presentation, however, I shall not refer to all the variant readings and their sources. Another clarification in order is that my choice of readings is based on a consultation of all available manuscript traditions: kārikā, Vṛtti, and Tikā. Although such a consultation does not yield any startlingly new readings in the present case, it serves to establish the original as objectively as possible :
prayena samkṣepa-rucin alpa-vidya-parigrahān |
samprāpya vaiyākaraṇān saṁgrahe 'stam upagate ||481// kṛte 'tha patanjalina guruṇā tirthadarsinā
sarveṣām nyāya-bijānāṁ mahābhāṣye nibandhane ||482// alabdha-gādhe gambhiryād uttāna iva sauṣṭhavat | tasminn akṛta-buddhināṁn naivāvāsthita niscayaḥ ||483// vaiji-saubhava-haryakṣaiḥ śuṣka-tarkānusāribhiḥ/ ārṣe viplavite granthe samgraha-pratikañcuke1 //484 // yaḥ patanjali-siṣyebhyo bhraṣṭo vyākaraṇāgamaḥ| kāle sa dākṣinātyeṣu granthamātre" vyavasthitaḥ //485// parvatad āgamam labdhvā bhāṣya-bijānusāribhiḥ | sa nito bahu-śākhatvaṁ candrācāäryädibhiḥ punaḥ //486//
2. Although pata is attested in all manuscripts of the karika tradition and in some important manuscripts of the Vṛtti and Tika traditions, I have decided to accept the reading pata. This is in view of the following facts: (a) There is no other reliable occurrence of patanjali as the name of the author of the Mahabhiṣya. (b) The relatively more reliable manuscripts of the Vrtti and the Tika read pata, These are also the manuscripts far removed from each other in terms of location of writing. (c) Even those. Tika manucsripts which read pata' in the karika portion almost always read pata in the Tika pertaining to the karika, indicating thereby that the author of the Tika knew the reading to be pata.
3. vaiki is found in the place of vaiji in one Vṛtti manuscript, and baidri in one Tika manuscript. The reading baiji, although attested in only two usually reliable sources. can be accepted instead of vaiji, as manuscript writers do not always distinguish v and b, It should also be noted that whereas baiji can be easily derived from bija and thus given some etymological significance, no straightforward etymology seems possible for vaiji. See 5. 1 and footnotes 27 and 31 below.
4.
5. Since the best manuscripts of the karika tradition read pata, at one point I was uncertain about the reading adopted here. Hence the reference in Scharfe 1976: 276 footnote 20 to a letter from me. An examination of the Vrtti and Tika manuscripts has now convinced me that the reading pata is clearly preferable on objective criteria. Contextually too, there is no reason why a taddhita formation like pitanjali should be employed.
6. This reading of Vitti and Tika manuscripts has an objectively stronger claim to being genuine than kalena of karika manuscripts. If the latter reading is accepted, one must either understand saḥ in 485cd or assume that 485 and 486 together constitute one sentence. The second alternative is rather difficult to justify in view of the intervening 486ab.
7. The reading matro of the generally better karikā manuscripts is not corroborated by the Vrtti and Tika manuscripts.

Page #3

AKLUJKAR : The Concluding Verses of Bhartṛhari's Vākya-Kāṇḍa ii
nyāya-prasthāna-margāms tān abhyasya svam ca darśanam [ pranito guruṇāsmākam ayam agama-samgrahaḥ //487// vartmanām atra keṣāṁcid vastu-matram udahṛtam | kände tṛtiye nyakṣena bhaviṣyati vicāraṇā ||488// prajñā vivekam labhate bhinnair āgama-darśanaiḥ kiyad vā sakyam unnetum sva-tarkam anudhāvatā ||489|| tat tad utprek ṣamāṇānām purāṇair āgamair vină anupasita-vṛddhānāṁ vidyā nätiprasidati //490//
2.1 It has so far been assumed that these verses are a composition of Bhartṛhari. I wish to question this assumption. The natural meaning of verse 487 is: "Having frequently reflected upon those nyāya-prasthanamärgas and his own view, our teacher composed this compendium of traditional knowledge". If we suppose that it was Bhartṛhari who made this statement, it follows that his teacher, and not he, composed the kärikäs and Vṛtti up to 480. However, such a conclusion would go against all the evidence we have in the manuscripts and the impressively long and consistent tradition of Bhartṛhari's authorship. On the other hand, if we suppose that it was some student of Bhartṛhari who wrote 487 and the group of verses to which 487 belongs, we shall have shown due regard for the available evidence regarding the authorship of the portion up to 480. To be taken into account in this connection is also the thesis I wish to put forward in a forthcoming article with what I hope to be adequate justification. It is that Bhartṛhari planned to write a vṛtti for the third kända, but could not write it for some reason that he either died or was incapacitated before he could turn to writing it. Now, if that is what actually happened, we should not at all be surprised to find a student of Bhartṛhari writing a few appropriate verses at the end of the Vṛtti of the second kända to mark the point where Bhartṛhari
8. In the writings of Bhartṛhari and those close to him, the word nyaya commonly stands for principle helpful in arriving at a logical or contextually justifiable view' (cf. Cardona's paper in this volume). I take prasthana to mean foundation, basis, source ' (compare the usage prasthana-trayi). A literal translation of the compound expression nyaya-prasthana-marga will, therefore, be the ways of the source of principles helpful in arriving at justifiable views'. In the light of sarveṣam nyaya-bijanaṁ mahabhaṣye nibandhane in 482 and bhasya-bijānusāribhiḥ sa nito bahu-sakhatvam in 486, this amounts to saying the ways of the Mahahhiṣya'. Thus, I am essentially in agreement with Raghunatha Sharma's (1968: 575) explanation tan mula-bhūta-vyakaranagamato jñatan bhasye 'vasthitan nyaya prasthana-margan. I do not object also to the interpretation ways of the nyaya-prasthānas such as Mimāṁsā and Vyakaraṇāgama'. While Weber's (1862 161) translation der Schluss, Vorgang, und Weg' seems incorrect to me, I find the translations the ways of logical discussion' (Goldstücker 1861:238) and the various other systems' (Raghavan Pillai 1971 146) less than exact.

Page #4

12
ABORI : Diamond Jubilee Volume
stopped. In fact, the relevance of the mention of the third kāņda in verse 488 and of the description of that kāņda (devoted to a detailed examination of various relevant vartmans or āgamas ) in verses 489–490 can be explained most straightforwardly if we ascribe the verses 481-490 to a student of Bhartphari. The student, writing epilogue-type verses as he was, must have been anxious to avoid giving the impression that Bhartshari's work came to an end with the kārikās and Vrtti of the second kānda, and must have felt the need to connect the first two kāņdas with the third kāņda. Perhaps the plural form asmākam in verse 487 is also significant from this point of view. It is a form that would naturally occur to anyone writing as a representative of a number of students.10 Furthermore, there is nothing in verse 487 that applies only to Bhartphari's teacher Vasurāta and not to Bhartphari. The latter also had obviously studied the nyāya-prasthāna-mārgas" and had a view, a philosophy, of his own." Therefore, I tend to believe that verses 481-490 are not a composition of Bhartshari.13
9. What I say here entails : (a) Bharthari composed first the kärikäs and then most of the available Vrtti, i. e., those portions of the Vitti which are not syntactically related to the karika's (Aklujkar 1972 : 190-193); or, Bharthari first finalized the karika text and then proceeded to give final form to the Vrtti, which latter activity he could carry out only to the end of the second kanda. (b) The karikas were separated from a kārika +vytti composition and a tradition of karika manuscripts was begun after Bhartr-hari's time. Otherwise, one would not have found in the karika manuscripts verses 481-490 written by a different hand.
10. The Tika explains the use of the plural in two ways : asmakam iti bahu-vacanad anyesam api sahadhydyinam grahanam. atha va mayā tu tad-anucchedayayam upanibandhaḥ krta ity atmano bahu-manah prakafitah. The second of these explanations is misunderstood by Raghunātha Sharmā (1968 : 575 ) when he ,comments : atha vdsmakam ayam agama-sangrahaḥ = matkartykatvena prasiddho 'yam agama-samgrahovákyapadiyakhyo gurund pranito na tu maya maya tu tad-cauccheda yāyan grantha-rūpenopanibandhaḥ kȚta ity átmano gurau bahu-manaḥ prakațita iti. The intention of the Tika is clearly to say that Bhartphari attaches bahu-mana to himself, for he, among all students of Vasurāta, gave a lasting, written form to the collection of traditions that Vasurata had imparted; the Tika does not speak of attaching bahu-mana to Bharthari's guru, at least in the particular remark under consideration. Secondly, it is precisely a statement to the effect that Bharthari's teacher composed the Vakyapadi ya that the Tika avoids making. I also fail to understand what the written composition (grantha-rupa upanibandha) authored by Bhartshari would be, once the composition commonly ascribed to him, the Vakyapadi ya, is attributed to his teacher.
11. The statement holds true under any sensible interpretation of nyāya-prasthanamärga. Bhartphari's knowledge of such systems as the Mimāmsä and Vaiseçika is evident from his commentary on the Mahabhas ya as well as from scores of passages in the Trikandi.
12. The most eloquent testimony to Bharthari's intellectual independence is provided by as early an author as Malla-vādin (pp. 581, 594-595 ):.. iti bhartyharyadi. matam. vasuratasya bhartyharyupadhyayasya matam tu.. evar lavad bhartyharyadi-darśanam ayuktam. yat tu vasurato bhartyharer upådhyayah..ity aha..
13. (a) The importance of 481-490 for the history of Sanskrit grammar is in no way diminished if they are not ascribed to Bhartphari. As the work of a junior contemporary of Bhartghari they remain almost as ancient and as reliable as they have so far been held to be.
(Continued on the next page )

Page #5

AKLUJKAR : The Concluding Verses of Bhartrhari's Vākya-Kāņda
13
2.2 Is it the case that the difficulty I perceive with regard to 487 has not occurred to others who studied it? I find it hard to suppose so. The very fact that attempts have been made to assign secondary, non-literal meanings to the verse indicates that some uneasiness has been felt regarding what it literally says. Let us now examine whether these attempts are justified. If they are justified, the assumption behind them that Bhartphari is the author of 481-490 must be deemed acceptable ; on the other hand, if they lack justification, the assumption must be set aside.
2.3 The Tikā offers the following comment on 487 : atha kadācid yogato vicāryal4 tatra[ bhavatā 715 bhagavatā vasurāta-guruņā mamāyam āgamaḥ samjñāya vātsalyāt pranita iti sva-racitasyāsya granthasya gurupūrva-kramam abhidhātum äha. (At this point, the text of 487 as given above is found.)... anena guruņā sarjñāyal na tathā mamāyam āgama-samgrahaḥ pranito yena samdeho bhaved api tu sāvadhānenety uktam bhavati. Here the intention is clearly to make the verse say, not “my teacher composed this ", but “I composed this because of the affectionate personal attention (note vātsalyāt, sāvadhānena) that my teacher gave to me".17 However, there is
Continued from the last page) (b) In the light of what I have argued here, point 2.1 (a) on p. 548 of my 1969 article on the title Vakyapadiya should be dropped. If the verses at the end of the second kanda are not written by the author of the Vakyapadiya, I cannot use their existence as a proof of his intention to divide the Trikandi into two parts. However, my view regarding the title stands as it is on the basis of the other considerations recorded in the same article.
14. I do not know what precisely is meant by yogato vicarya.
15. The constituent bhavata seems to have been dropped through haplography in the Tika manuscripts. That one must supply it is clear from the lack of connection bet. ween tatra and bhagavata and from the fact that the honorific tatra-bhavat was rather closely associated with the line of scholars to which Bharthari and Vasurāta belonged (Aklujkar 1972 : 186-188).
16. One gets the impression that the author of the Tika glosses samhjaya with vatsalyat and sdvadhanena. However, if that is the case, two difficulties arise : (a) How does one derive the meanings vatsalyat and savadhanena from samjaya 'having known together/completely, having ascertained'? (b) Where can the expression sarhjaya be accommodated in the verse? Obviously the author of the Tika would not gloss his own words in this manner, and he leaves no doubt that he reads the verse precisely as we do. In view of these considerations, I conclude that samjïiaya has not in fact been glossed. It is simply a short expression for tàn nya ya-prasthana-margan suami darsanam ca abhyasya. The expressions that follow it, vatsalyat and na tatha.. api tu sävadhanena, are meant to bring out the spirit of the verse as the Tika understands it to be. They provide more details concerning how Vasurāta imparted the agama or agama-samgraha to Bhartphari. For this as well as other reasons I find Raghunātha Sharma's change of the second sathjia ya to asanjaya quite unnecessary.
17. The summary verses of Punya-raja appearing at the end of the Tika add one detail ( guru-nirdistad bhas yan nyayaviluptaye, which is probably to be read as guru-nirdesad bhas yamnayaviluptaye ) to this interpretation : acārya-vasuratena nyaya.märgān vicintya ca pran to
(Continued on the next page)

Page #6

14
ABORI : Diamond Jubilee Volume
not a single word in the verse that would justify the addition of the element “ because of the affectionate personal attention ” so crucial to this interpretation. Furthermore, it is apparent from the clauses guruņā mama ayam āgamah prañitaḥ and guruņā mama ayam āgama-samgrahah prañitah that the author of the Tikā construes asmākam in the verse with praạitah, understands -pranitah in the sense of some such word as pratipăditah ('stated, explained, delivered, given'), and interprets asmākam as a genitive substitute for asmabhyam (=mahyam, in this instance ). However, the natural connection of asmākam is with guruņā. If the guru is not related to the author of the verse, i. e. to the person referred to by asmat, there is no justification at all for bringing him in; the expectancy whose teacher ?" must be satisfied. Secondly, the word pranita, at least in the writings of Bhartphari and his near contemporaries, does not ever seem to have been used in the sense the author of the Tikā seems to assign to it.18 And even if we assign that sense to pranita, we do not get past difficulties. If we say asmäkam guruņā ayam āgamaḥ pratipăditah, we are guilty of overlooking the constituent sangraha and the obvious reference of ayam to the work Väkyapadiya. On the other hand, if we say asmākam guruņā ayam āgama-samgrahah pratipăditah, we make a statement that goes against the massive evidence favouring Bhartshari's, and not his teacher's, authorship of the āgama-samgraha called Vākyapadiya. Thus, the Tikā explanation is far from satisfactory.
2.4 Raghunātha Sharmā (1968 : 575) mostly follows the Tikā. If he is aware of any of the difficulties pointed out above, he does not say so. The only significant addition made by him , ayam agama-samgraho guruņāsmākan krte pranita iti vā yojanā, suffers from lack of evidence ; neither the manuscripts of the Vāk yapadiya nor any of the known medieval works "attribute the authorship of the āgama-sangraha called Vāk yapadiya to Bhartphari's teacher. Besides, there is no justification for supplying krte.19
(Continued from the last page) vidhivac cayan mama vyakaranagamah // mayapi guru-nirdistad bhasyan nyayaviluptaye kanda-traya-kramenāyan nibandhah parikirtitaḥ //. In my view, S. Iyer (1969: 3) offers an unjustifiably specific meaning (see 2.5 below ) to these verses in his remark : "..Vasurāta gathered together the traditions in a composition for the sake of his disciple Bhartphari and instructed him to write his own work on the basis of that."
18. Cf... yaih pratyakşa-dharmabhis tatra tatra pravacane sūtranutantra-bhas yāņi pranitani tair eva sistaih .. (Vriti 1.23d. p. 63. 9). Note also the use of pranety in 1.23d, Vytti 1. 148150 pp. 203-205, and Tripadi p. 214. 4, and of pranayana in Tripadi p. 37. 17-18.
19. I shall leave out of consideration the translations by Goldstücker 1861: 238 (".. my Guru .. taught me the compendium of this grammatical work”) and Weber 1862: 161 ("Von meinem Lehrer .. ward mir gelehrt hier dieser samgraha des Texts"). They are more arbitrary than some of the interpretations rejected here.

Page #7

AKLUJKAR : The Concluding Verses of Bhartrhari's Vāk ya-Kānda
15
2.5 While the Tikā explains 487cd by saying, in effect, that Bhartphari was very much indebted to his teacher for the contents of the Vākyapadiya and for inspiration and guidance in writing the Vāk yapadiya, Baladeva Upādhyāya ( 1968 : kha) and S. Iyer (1969: 3, 69 ) go a step further. Instead of tracing Bharthari's indebtedness to Vasurāta's oral instruction, they trace it to a book by Vasurāta, and offer that rather specific indebtedness as the basis for the apparent attribution of the Vākyapadiya to Bhartshari's teacher (Since this compendium draws upon the book of my teacher so heavily, you may say that it is actually my teacher who has composed it'). Their interpretation thus abandons the clever and, I am sure, deliberate ambiguity of the Tikā interpretational and does more violence to the text of 487. It is quite evident that that verse does not contain a statement on the authorship of two works by two individuals (Bhartphari and his teacher Vasurāta ). Contextually it can refer to the genesis only of the work that precedes it and is before us, namely, the Vāk yapadiya of Bhartphari. The word ayam in it cannot refer to any other work. This is clear also from the immediately following verse. There we find atra, related to ayam, and that characterization of the āgama-samgraha which entirely fits the Vākyapadiya. 21
2.6 At this point it may be said that there exists another way of understanding 487cd which is free from problematic construing, retention of ambiguity in the case of pranitah etc., and unjustifiable bifurcation of the reference of ayam. Take pranito guruņāsmākam ayam āgama-samgrahah to be an expression of Bharthari's reverence and humility; conclude that, out of gratitude, Bhartphari offered the authorship of his work to his teacher; the remark my teacher composed this is simply an hyperbolic expression for I could not have composed this without the help of my teacher and hence this really belongs to him.' True, this alternative has the merit of not doing any violence to the syntax and literal meaning of 487cd ; but it nevertheless forces one to accept something of which there is no corroboration in the tradition, namely that Bhartshari ascribed his own work to his
20. The Tika see the passage cited in 2. 3 above) does not explain pranita. It also seems to pretend that the constituent samgraha in agama-samgraha does not exist.
21. (a) I do not wish to deny the possibility of Bhartphari's teacher having com. posed an agama-samgraha or of Bhartphari's being indebted to that agama-samgraha. What I object to is the inference of either possibility on the basis of verse 487.
(b) It is not surprising that the author of the pika and Vrsabha (p. 1. 1922 ), not being aware of all the ways in which autographs change, could not see the possibility of there being another hand behind 481-490, and that they consequently read them as a continuation of what precedes. That modern scholars equipped with the science of textual criticism did not realise or explore the possibility is puzzling.

Page #8

16
ABORI : Diamond Jubilee Volume
teacher. Moreover, a figurative or secondary meaning should be resorted to, especially in historical research, only when the literal meaning cannot be accommodated. As shown in 2.1 above, this is not the case with the passage under consideration.
2.7 Another possible way of circumventing the conclusion that it was Bhartphari's student who authored verses 481-490 would be as follows: There is no reason why 481-490 must be taken as marking the conclusion of both the kärikā text and the Vrtti text; it is possible that they indicate the end of only the Vrtti text and constitute a statement of the author of the Vịtti alone. If that is so, then verse 487 can easily be understood as a remark by the author of the Vrtti, Bhartphari, to the effect that it was his teacher who composed the āgama-sangraha in the form of the Vākyapadiya kārikās." However, as I have argued elsewhere (Aklujkar 1972 ), the kārikās and the Vrtti must be thought of as coming from the pen of one and the same person. Besides, we have no evidence to credit Bhartphari's teacher, whether he is Vasurāta or anyone else, with the authorship of (at least most of the kārikās.
3.1 As for verses 481-482, two possibilities need to be considered. Are we to read 481 as giving the context of 482, or are we to read 481 and 482 together as jointly providing the context of 483 228 In the former alternative, atha 'then, subsequently' will retain its most common meaning and need not be understood in the sense and, moreover ',24 but the implication will be that the Sangraha was unavailable or was mostly unavailable (the latter if one construes prāyeņa, not with samk sepa-rucin, but with astam upāgate ) to the author of the Mahābhāsya. This implication is contradicted by the description samgraha-pratikañcuke (see 5.4-6 below) of the Maha
22. (a) This alternative leads to, but does not require an answer to, the question: Who composed the karikas of the third kanda ?
(b) It is possible to phrase the alternative by assuming Hari-vrsabha to be the author of the Vetti and Bhartrhari to be the author of the karikas. However, as I have suggested elsewhere (Aklujkar 1972: 182-183 footnote 2), Hari-Vrsabha does not really exist in the context of the Vakyapadiya. Besides, saying Hari-vrsabha wrote 481-490 implies acceptance of the view that someone other than Bharthari wrote 481-490.
23. The Tika and Raghunātha Sharma's Ambakartri accept the first possibility (note .. samgrahabhidhanam nibandhanam .. astam upagatam... astath yatah samgrahah. ). Thieme (1956: 18-19), while entertaining the first possibility ("the Samgraha had perished"), is quick to realise that 481d can also be taken to mean:"..the Saṁ graha had .. ceased to be studied ". Yudhisthira Mimāṁsaka ( samvat 2020:278 ) clearly distinguishes between the loss of the text of the Sangraha and a break in the tradition of studying the Sangraha. He takes 481-482 as indicating the latter. So does S. Iyer (1969: 3 ).
24. Use of atha in the sense of ca cannot be said to be uncommon, especially in metrical compositions.

Page #9

AKLUJKAR : The Concluding Verses of Bhartrhari's Vākya-Kānda
17
bhäsya and also by Patanjali's statements concerning the Sangraha ;25 it does not seem to be the case that the Sangraha was lost or was substantially lost at the time of Patañjali. Thus, the only interpretation justifiable in view of the available evidence will be the one in which 481-482 are understood as jointly stating the context of 483. In other words, what the verses precisely state is the following: (a) Men of immature intellect could not determine the nature of the views expressed in the Mahābhāsya or could not determine the views acceptable to the author of the Mahābhāsya once the Saṁgraha went into oblivion. (b) This was due, in part, to the fact that the Mahabhāşya took so much of the Sangraha for granted. Its apparently simple statements were based on discussions of great depth and length that were carried out in the Saṁgraha. (c) The imprecise understanding of the Mahābhāsya on the part of students with unseasoned intellects 26 was also due to the fact that Patañjali, as a person well-versed in different branches of learning (tirtha-darsin), used all principles of interpretation and thus placed tħe grammarians to follow in the not too comfortable position of being required to comprehend the principles of interpretation that existed in seed-form in his work. (d) The passing of the Samgraha into oblivion, in turn, was due to the fact that it encountered grammarians who failed to appreciate its worth. There was no proper appreciation of the discussions from the points of view of many systems which the Samgraha contained ;
25. (a) sath graha etat pradhanyena pariks tam nityo va syat karyo veti. tatrokta dosah prayojandny apy uklani. tatra tv esa nirnayo yady eva nityo 'thapi karya ubhayathapi laksanath pravartyam iti. (Mahabhasya 1.1.1 p. I. 6)
(b) sangrahe tavat karya-pratidvandvi-bhavan manyamahe nitya-paryāya-vacino grahanam iti. ( Mahabhasya 1. 1. I p. I. 6)
(c) sobhana khalu dakșa yanasya saṁgrahasya krtih. Sobhana khalu dakşayanena sathgrahasya kytir iti. (Mahabhasya 2. 3. 66 p. I. 468 )
(d) Possibly: kiratimh carkaritintam pacatity atrayo nayet I praptijñam tam ahath manye prarabdhas tena sathgrahah II ( Mahabhas ya 7. 4. 92 p. III. 359)
26. Thieme. translates akyla-luddh nam as “not exercising their intellect". I think a translation like those whose intellect is not sharpened or made mature by the study of fastras" will be closer to the original. It is repeatedly indicated by the author of 481-490 that only a person having a background in several vidyds / agamas / agama-darśanas can understand the apparently simple statements of the Mahābhas ya in a satisfactory way. Besides this contextual indication of the thrust of the adjective, we have its use in Gita 18. 16 ( .. akrta-buddhitvan na sa pasyati durmatih ), which is explained by Sankara with the expression asanskrta-buddhitvat. The compound krta-buddhi is evidently analogous to kyta - -hasta ( Trikandi 3. 14.558) one of trained hand' and kytatman (Gita 15.11, which, in fact, has the negative akrtatman) one of cultivated self'. The root kr in it signifies modification or perfection (compare the use of “do" in English sentences such as " Have you done your hair?"). The Tika glosses it accurately as krta vyutpattyā prakarsam prapta mahati buddhir yesar te. Of course, one cannot be a kyta-buddhi unless one exercises one's intellect first. To that extent, Thieme's translation, although not contextually exact, is indeed justified.
3 Annals (D. J.)

Page #10

18
ABORI : Diamond Jubilee Volume
there was also a demand for works that would provide an overview or gist of the system of grammar.
4.1 The above analysis was prompted mainly by the consideration that the interpretation of 481-482 given in footnote 23 should not be accepted uncritically—that it should be recognised that there are two possibilities of interpretation leading to two significantly different depictions of the history of the Pāninian tradition. The analysis is motivated also by a desire to determine the sense of verse 484. Thieme ( 1956 : 19 footnotes 45-46) has expressed the view that 484c, ārse viplāvite granthe, refers to the Astādhyāyi of Pānini. My analysis of 481-483 should serve to indicate that this cannot be the case. The relationship mentioned in them is that between the Mahābhāşya and the Sangraha. There is no reason why their author should suddenly switch from a consideration of the Mahābhāșya to that of the Astādhyāyi. Thieme's interpretation creates a problem also for what follows 484. If we go along with it, the logical link between 483 and 485 is lost. The former tells us that the grammarians who flourished after Patañjali did not exhibit either the patience or the intellectual ability needed for determining the acceptable views in a work like the Mahābhāşya. From the latter we learn that the traditional (interpretive ) lore of grammar slipped from the hands of the disciples of Patañjali. In between we need a statement saying in effect that the understanding of the Mahābhāsya became distorted as a result of the variety of interpretations and that the confusing variety of interpretations discouraged prospective students. Only verse 484 can provide that link, and that too only if it is interpreted as "speaking of the Mahābhāsya.
4.2 Every expression that occurs with ārse viplāvite granthe in 484 indicates, directly or indirectly, that the reference of the verse cannot be to the Astādhyāyi : (a) From the place at which the names of Vaiji, etc. occur it is clear that those grammarians, pseudo-grammarians, or anti-grammarians lived after Patañjali. The target of their activity, or at least the primary target of their activity, therefore, is more likely to be Patañjali's work than Pāṇini's work. (b) śuşka tarka is characterized in the Vrtti of 1. 153 as sabdaŚakti-rūpāparigshitah..sādharmya-vaidharmya-mātrānusāri sarvāgamopaghāta-hetutvād anibandhanah and in the Tikā on 2.484 as anya-śāstra-parimalarahitaḥ. Thus, the expression suska-tarkānusāribhiḥ specifically points out the failure of Vaiji and others to take into consideration the related branches of knowledge and to realise that the words employed in the tradition of grammar are to be understood in a contextually sensible way. Now, I think it is evident that the Asțādhyāyi does not so directly demand of its readers a knowledge of the principles employed in other systems as does the Mahā

Page #11

AKLUJKAR : The Concluding Verses of Bhartshari's Vākya-Kāņda
19
bhāsya. Contextually too, it is in the case of the Mahābhāsya that proper understanding is indicated to be dependent on a knowledge of other branches of learning; cf. tirtha-darsinā, sarveşām nyāya-bijānām, alabdha-gādhe gāmbhiryāt, and akrta-buddhinām. (c) Even if we go along with Thieme and understand samgraha-pratikañcuke as meaning “ of which the defensive armour was the Sangraha”, the word does not seem to be so appropriate as a description of the Astādhyāyi as of the Mahābhāsya. From the indications provided in 481 (samk sepa-rucin, alpa-vidyā-parigrahān) and from the known references to the Samgraha (Yudhisthira Mimāmsaka sarvat 2020 : 270-277) one can easily conclude that the Sagraha was very ambitiously planned, that it involved study of many vidyās, and that it was predominantly a work on grammatical theory and linguistic philosophy. It does not seem to have been a commentary to the Astādhyāyi in the usual sense or a defense of the Astādhyāyi per se. On the other hand, as shown in 3. 1 above, the Mahābhāsya drew heavily upon the Samgraha and could be misinterpreted in its absence.
5.1 So far four (see footnote 29 below) different explanations of the expression saṁgraha-pratikañcuka have been proposed. The Tikā seems to be unsure about it, as it neither cites the expression as a pratika, nor provides any explicit gloss. However, there is room to believe that its author decided to read 484d as saṁgraha-pratikañcukaih and to understand it as meaning saṁgraha-pratipaksa-bhūtaiḥ. 27 Goldstücker (1861 : 257-258) adopted the same reading, but assigned it the exactly opposite meaning “ who were partisans of the Sangraha". Among those who are aware of the reading samgraha-pratikañcuke as an adjective of granthe, we have Kielhorn (1876 : 244), who provisionally takes it to be a tat-purusa ( samgrahasya pratikañcuke) meaning “preserving the (contents of the ) Sangraha ", and Thieme, who takes it to be a bahu-vrihi (samgrahaḥ pratikañcukaṁ yasya tasmin ) meaning "of which the defensive armour was the Samgraha” or “whose counter-armour is (was ) the Samgraha ".
5.2 Goldstücker's rendering is clearly inappropriate. We have no reason to suppose that there was a group of Saṁgraha partisans in existence after the Mahābhāsya had gained currency or to suppose that that group was interested in making a case against the Mahābhāsya. The high regard for the Sangraha shown by Patañjali and by the followers of Patañjali such as Bhartphari indicates that the followers of the Mahābhāsya were also followers of the Saṁgraha, not a rival group. Secondly, there does not seem to be any
sangraha-pratipaksa-bhūtair acaryais tarka
27. Note the introduction to 484: tatha ca -vidyamatra-Dedibhih..

Page #12

ABORI : Diamond Jubilee Volume
satisfactory way of arriving at the meaning partisan' from the basic meanings of prati and kañcuka. The same difficulty exists in the case of the Tikā explanation. Thieme ( 1956: 19 footnote 46) suggests that the author of the Tikā “understands pratikañcuka as a bahu-vrihi, to be analyzed : pratinaddhaṁ kañcukam yena", he who has fastened his armour, he who is ready for battle'. However, such a compound would be unique in two ways. The root nah is not known to have been used with the prefix prati, not at least in a sense useful for the Tikā author's derivation. Secondly, the bahu-vrihi compounds with a suppressed participle, praparņa etc. (Trikāņdi 3. 14. 52), are always explained with gata, krānta or a synonym thereof; naddha is unparalleled as an implicit or latent member of a bahu-vrihi. Another difficulty with the Tikā interpretation is that it does not explain how the Sangraha could produce hostile reaction even after Patañjali's time or why the opponents of the Sangraha were interested in expressing opposition to the Mahābhāsya. It is, of course, possible that the Mahābhāsya was viewed unkindly because it was based on the Samgraha. 28 But even then it is puzzling that the Sangraha should have given rise to a long line of opponents so dead set against it that even a work based on it was a target for vehement attacks. Finally, one must note as a problem common to the explanations given by both Goldstücker and the author of the Țikā that the instrumental reading samgraha-pratikañcukaiḥ is not attested in any of the manuscripts known so far.
5.3 The sense counter-armour, defensive armour' attributed by Thieme to the word pratikañcuka is etymologically plausible and, when accepted as a part of a reference to the Mahābhāsya, contextually suitable (see 4.1-2 above). However, one wonders whether prati is really called for if that is what the compound word means. As kañcuka or armour is meant to be a protective, defensive covering, prati adds nothing of significance when taken in the sense 'counter-, defensive '. Secondly, the meaning given by Thieme proves to be partially or entirely unsuitable in the other contexts known so far. As Thieme notes, pratikañcuka occurs in Aryabhasiya, Gola-pāda verse 50 as sukstāyusoh praņāśam kurute pratikañcukam yo 'sya.29 In this concluding verse of his work, Āryabhata clearly wishes to warn the reader that a certain type of activity should not be undertaken with
ever, one the Mahabhically plau
28. Cf. Tika on 483: etena sangrahānusarena bhagavatā patañjalina sathgraha-samkşepa-bhutam eva prayaso bhasyam upanibaddham ity uktar voditavyam.
29. The commentary on the Aryabhatiya explains this as asya sastrasya yaḥ pratikañcukam kurute, doşotpadanena tiraskaranam ity arthah, tasya sukyta yuşoh pranasah syat. Thus, , making something obscure by finding faults in it' is the fifth meaning proposed for pratikañcuka. Since it fails to be applicable in 484 and in the passage from Kumarila eited in 5. 4, it must be rejected.

Page #13

AKLUJKAR : The Concluding Verses of Bhartrhari's Vākya-Kānda
21
respect to his work; if undertaken, it would lead to the destruction or waste (praņāśa ) of the reader's good karman (sukrta) and life (āyus). Now, this activity cannot certainly be the making of a defensive armour' or protective covering; there is no reason why the possibility of anyone's attempting to provide more protection (in a literal sense or in the figurative sense of justification, bolstering with arguments, etc.) to Aryabhata's work should disturb Aryabhata to the extent of uttering an imprecation. Realising this, Thieme proposes that we should read apratikañcukam, i. e. supply an avagraha after kurute, in the Aryabhațiya verse, take apratikañcukam as an adverb, and translate the line as follows: “He causes perdition of his good deeds and his life so that there can be no defense (counter-armour), who ( causes perdition ) of this ( work, the Āryabhatiya ).” However, such a translation is possible only if we repeat the phrase praņāśaṁ kurute as yah asya ( āryabhațiyasya ) praņāśam kurute [ saḥ) sukrtāyuşoh apratikañcukam praņāśam kurute. It does not seem likely from the placing of praņāśam and pratikañcukaṁ / apratikañcukam in the verse line that Āryabhata had in mind the connection of praņāśam with yaḥ asya and of pratikañcukam / apratikañcukam with praņāśam kurute. Secondly, although a statement like 'He causes perdition of his good deeds, so that there can be no defense' is sensible, a statement like 'He causes perdition of his life, so that there can be no defense' is hardly sensible; when life is gone, there is no need for defense.
5.4 Recently I have come across an occurrence of pratikañcuka tḥat seems to have eluded all those who previousiy discussed the problem of 484d. It is in the Tantra-vārttika (on 1. 3. 7, p. 122 of the Anandāśrama edition of 1970) of Kumārila : pratikañcuka-rūpeņa pūrva-śāstrārtha-gocaram/ yad anyat kriyate tasya dharmam praty apramāṇatā || tathā ca prāyaścittādidāna-kāle yo vāk yam ātmiyam anya-kavikrtam vā ślokam voccārya mānavādiprāyaścittam dadyān na kaścid [? kañcid] api dharmarthaṁ pratipadyeta. This passage, especially in the gloss it contains, serves to establish that pratikañcuka cannot mean what Thieme thinks it means. What is more important is that, of all the relevant passages known so far, it provides the clearest indication as to what pratikañcuka must mean. Kumārila's point is as follows: If one were to replace the scriptural sentences employed in religious activities with newly composed sentences of similar import, one would not acquire dharma by performing those activities; it is the scriptural sentences that are a valid means of dharma, not their recasts. Thus, the Tantra-vārttika passage leaves no doubt that pratikañcuka stands, in the context of literature, for incorporating contents, expressing the same matter in another composition'. It can be easily seen that this meaning fits the other two contexts in which the word occurs. What Aryabhata is really con

Page #14

22
ABORI: Diamond Jubilee Volume
cerned about is the possibility of plagiarism. He does not wish that anyone (probably, from among his contemporaries) should expropriate his thought - his findings. Therefore, he seeks to deter prospective plagiarists by writing the stern words: "He who prepares a pratikañcuka (work having the same contents) in the case of this (Aryabhaṭiya) causes perdition of [his] good deeds and life ".30 As for the Vakya-kāṇḍa passage with which we are immediately concerned here, the suitability of the sense of pratikañcuka gathered from the Tantra-värttika is even easier to see. The Mahabhaṣya, as a recast or adaptation of the Samgraha, made use of the contents of the Samgraha (see footnote 28 above). Hence it has been described as samgraha-pratikañcuka. Thus, Kielhorn showed a remarkable sensitivity to the drift of 481-490 when he suggested that 484d be translated as "preserving the (contents of the) Samgraha " 31
5.5 How did pratikañcuka acquire this figurative sense of old wine in a new bottle'? I think pratikañcuka is a compound of the type of pratikṛti replica', praticchāyā reflection, mirror-image', pratinidhi 'substitute, representative', pratibimba reflection, mirror-image', and pratirūpa 'counterpart'. The prati in it carries the sense another similar, the one on that side which agrees with what we have on this side'. In other words, there is an implication in it of bodily difference (physical distinctness) as well as of inner or substantial correspondence. Its remaining constituent, kañcuka, is most probably intended in the commonly noticed meaning cloak, robe'. Thus, the etymological meaning of pratikañcuka seems to me to have been another dress, another garb, disguise', the implication being that the substance is the same in spite of the change in appearance. I think that the figurative sense given above emerges naturally when this etymological sense is restricted to the context of literature-to the context of composing works or passages.
30. This interpretation requires a repetition of kurute. However, the repetition is not as strained as in the case of Thieme's interpretation, for kurute is placed in the verse between pranasam and pratikañcukam with which it is connected. There is also the possibility that the original wording of the Aryabhatiya line was sukṛtayusoḥ pranašaṁ kurute kurute pratikañcukam yo 'sya, and that one kurute has been lost through haplography. The äryä metre is not disturbed in either reading.
31. Since it involved a major change in the reading furnished by all accessible manuscripts, I did not give the benefit of the following possibility to the author of the Tika in writing 5. 2. It is possible that his remark introducing 484 (see footnote 27 above) is a result of corruption through haplography-that it was intended to be read as: tatha ca samgraha[-samkṣepa-bhutamh] pratipaksa-bhatair acaryais tarka-vidyamatra-vedibhih See the Tika passage quoted in footnote 28 above. If my guess is correct, the meaning pratipaksa or pratipakṣa-bhūta assigned to the word pratikañcuka by the compilers of dictionaries on the authority of the Tika must be said to be the result of an unfortunate error.

Page #15

AKLUJKAR : The Concluding Verses of Bhartrhari's Väkya-Kānda 23
5.6 That the word kañcuka had transcended the meaning cloak, dress' and was capable of extended use can be seen from a number of texts. In the fifth act of the Abhijñāna-śākuntala, we come across the expression dharma-kañcuka. One of the lines (1. 843cd ) in Abhinava-gupta's Mālini(-vijaya- ) vārttika runs thus : itthaṁ ke 'py abhimanyante sāṁkhya-kañcukasamśrayāt. According to Vidyabhusana ( 1921 : 519 footnote 4), panditakañcuka is found in the Brhat-svayambhū-purāņa (Hara Prasad Sastri's edition, vi. 321). Several texts of the Kashmir Saiva tradition (Siva-sūtra 3. 42; Kșema-rāja's Śiva-sūtra-vimarsini on 3.3, 3.42; Abhinava-gupta's Mālinis -vijaya- )vārttika 1.652, 1. 836, 2.215, for example ) regularly employ kañcuka in the sense of the covering of the true self or soul that the senses etc. form.
6.1 As we saw above in 2.1-7, the problem that verse 487 gives rise to has not been squarely faced in the literature on the Vākyapadiya. The treatment accorded to 486 has been even more superficial in one sense. Although attempts have been made to identify the parvata mentioned in it and although the precise nature of what Candrācārya and others did has been frequently discussed (see George Cardona's paper in this volume ), no modern scholar seems to have realised that the verse contains a textuaj problem. The construction āgamaṁ labdhvā sah bahu-śākhatvaṁ nitah is as strange as maņiṁ labdhvā sah bahudhā bhinnah. Normally one would say in such cases either agamaḥ labdhvā bahu-śākhatvam nitah (compare vişa-vrkṣaḥ sarvardhya chinnah )32 or āgama labdhvă tar bahu-śākhatva nitavantaḥ. That is, either the accusative āgamam or the nominative sah must be given up if 486 is to contain a construction worthy of a grammarian author. Now, it is obvious that changing saḥ and thereby opting for a reading like tam bahuśākhatvam nitavantah would amount to a complete rewriting of the verse; any attempt to introduce a standard construction in the present text of the verse must be made without disturbing the passive phrasing candracāryādibhiḥ..nitaḥ. Thus the only course open to a text critic is the one of altering agamam to āgamah (that is, to āgamo in the given phonetic situation ). But such an emendation, although metrically possible, is incapable of removing a further problem. The ägama of 486 can be either the vyākaraṇāgama mentioned in 485 or the mūla-bhūta vyākaraṇāgama ascribed to Rāvaņa as the Tikā says. If it is the former, there is no need to use the word āgama; it can be easily and simply referred to by a demonstrative
32. I am aware that the sentence the Sanskrit grammarians actually discuss is visa-vīkso 'pi samvardhya svayath chettum asampratam (Kālidāsa, Kumāra-sambhava 2. 55). I have simplified that sentence in order to bring into sharp focus the considerations involved in discussing 486.

Page #16

ABORI : Diamond Jubilee Volume
pronoun. Re-employment of the word āgama hardly fits the elegantly tight style of 481-490. Besides, if a Sanskrit author repeats a substantive at all in such a context, he introduces the corresponding form of the demonstrative pronoun before it; that is, if the author had intended to use the word āgama again by overlooking the possibility that it could be easily understood from 485, he would have fitted in 486а the phrase saḥ āgamah (or the phrase tam āgamam, to assume, for a moment, that the accusative is permissible). It is perhaps an awareness of these considerations that prompted the author of the Tikā to distinguish, rather obliquely, the āgama referred to by ägamam from the āgama referred to by saḥ, i. e. to speak of a mūla-bhūta vyākaraṇāgama distinct from the vyākaraṇāgama that the successors of Patañjali lost. He seems to have reasoned that if such a distinction is made the construction will not be ungrammatical (vajrań labdhyā maṇiḥ bahudhā bhinnah and vişa-vrk sam saṁvardhya āmra-vrk şah chinnaḥ are acceptable constructions), and the recurrence of the word āgama can also be explained. However, there is no justification in the given context to make the verse say parvatāt rāvaņa-viracitaṁ mūla-bhūta-vyākaraṇāgamam labdhvā candrācāryādibhiḥ sah patanjali-śişyebhyaḥ bhrastaḥ. . vyākaraṇāgamaḥ bahu-sākhatvam nitaḥ. Any author who can write as perspicuously as the rest of the verses bear out is not likely to use the general word āgama for a contextually absent and unexpectedly specific thing, especially when that word can be easily (mis )understood as referring to vyākaraṇāgama. If the author of 481-490 had the mysterious rāvana-viracita müla-bhūta vyākaraṇāgama in mind, he would have either written an additional verse or used a distinct expression indicating the distinction of two āgamas. Thus, āgamain does not appear to be the original reading in 486a. It is also possible that not only agamam but the entire phrase parvatād agama.n is a result of textual corruption. Of what it could be a corruption I am unable to determine at present.
BIBLIOGRAPHY
Abhinava-gupta. Śri-Malini-vijaya-vārttika. (Ed.) Madhusudan Kaul
Shastri. Kashmir Series of Texts and Studies, no. XXXI. Srinagar. 1921. Aklujkar, Ashok. 1969. “Two textual studies of Bhartshari”. Journal of
the American Oriental Society, 89: 547-563. ...................... 1971. “The number of kārikās in Trikāndi, book I".
Journal of the American Oriental Society, 91 : 510-513. .................... 1972. “The authorship of the Vākyapadiya-vrtti". Wiener Zeitschrift für die Kunde Sudasiens, 16: 181-198.

Page #17

AKLUJKAR: The Concluding Verses of Bhartṛhari's Vakya-Kända 25
"The authorship of the Vakya-kāṇḍa-tikā ". Charu
1974.
Deva Shastri Felicitation Volume, pp. 165-188. New Delhi.
.........
1978. "The number of kārikās in Trikāṇḍī, book II " To be published in the Adyar Library Bulletin. Aryabhaṭa. The Aryabhatiya with the commentary Bhata-dipikā of Parama
disvara. (Ed.) Kern, H. Leiden: E. J. Brill. 1874.
Bhartṛhari. Trikāṇḍi: (a) kāṇḍa 1. (Ed.) Subramania Iyer, K. A. Vakyapadiya of Bhartṛhari with the Vṛtti, and the Paddhati of Vṛṣabha-deva. Deccan College Monograph Series, no. 32. Poona: Deccan College. 1966. (b) kāṇḍa 2. (Ed.) Mānavalli, Gangadhara Śāstri. Vākyapadiyaṁ... Śri-Bhartṛhari -viracitam Śri-Punyarāja-kṛta-prakāśākhya-ṭīkā-yutam. Benares Sanskrit Series, nos. 11, 19, 24. Benares Braj B. Das & Co. 1887. (c) kāṇḍa 3: (Ed.) Subramania Iyer, K. A. Vakyapadiya... with the commentary of Helārāja. Deccan College Monograph Series, no. 21. Poona : Deccan College. 1963. Vakyapadiya... with the Prakirnaka-prakāśa of Helārāja. Poona Deccan College. 1973. See footnote 1 above.
Goldstücker, Theodor. 1861. Pāṇini: his place in Sanskrit literature. London. Reprint : Chowkhamba Sanskrit Studies, vol. XLVIII. Varanasi Chowkhamba Sanskrit Series Office. Also, Osnabrück: Otto Zeller.
...
Joshi, S. D. 1976. "Sanskrit grammar ”. Ramakrishna Gopal Bhandarkar as an Indologist, a symposium. (Ed.) Dandekar, R. N. Poona : Bhandarkar Oriental Research Institute.
Kielhorn, F. 1874. "The concluding verses of the second or Vākya-kāṇḍa of Bhartṛhari's Vākyapadiya". Indian Antiquary, 3 : 285-287. 1875. "Note on Rajatarangini I 176". Indian Antiquary,
4 107-108.
1876. "On the Mahābhāṣya”. Indian Antiquary, 5: 241-251. 1885. "Der Grammatiker Panini'. Göttinger Nachrichten,
pp. 185-199.
The preceding four are reprinted in Franz Kielhorn Kleine Schriften. (Ed.) Rau, Wilhelm. Wiesbaden: Franz Steiner Verlag. Kṣema-rāja. Vimarsini on Vasu-gupta's Śiva-sutras. (Ed.) Chatterji, J. C.
Kashmir Series of Sanskrit Texts and Studies, no. 1. Srinagar. 1911. Malla-vädin. Dvādaśāra-naya-cakra. (Ed.) Jambu-vijaya, Muni. Bhavnagar; Jain Atmananda Sabha. 2 volumes. 1966, 1977.
Patanjali. Vyākaraṇa-mahābhāṣya (Ed.) Kielhorn, F. Bombay. 1878-1885. Poona Bhandarkar Oriental Research
3 volumes. Third edition.
Institute. 1962-1972
Peterson, P. 1885.
4 Annals (D. J.)
"Note on the date of Patanjali". Journal of the Bombay

Page #18
26 . ABORI : Diamond Jubilee Volume . Branch of the Royal Asiatic Society, no. XLIII, vol. XVI : 181-189. Raghavan Pillai, K. 1971. (Ed., tr.) The Vakyapadiya [ Books 1-2]. Delhi : Motilal Banarsidass. Rau, Wilhelm. 1977. (Ed.) Bhartshari's Vakyapadiya (mula-karikas ). * Monograph Series of the Deutsche Morgenlandische Gesellschaft, no. XLII, 4. Wiesbaden : Franz Steiner Verlag. Scharfe, Hartmut. 1976. "A second index fossil of Sanskrit grammar; ians ". Journal of the American Oriental Society, 96: 274-278. Sharma, Raghunatha. 1968. (Ed.,.comm.) Vakyapadiyam Part II (Vakya..kandam). Sarasvati Bhavana Grantha-mala, no. 91. Varanasi : * Varanaseya Sanskrit Vishvavidyalaya. Subramania Iyer, K. A. 1969. Bhartyhari : a study of the Vakyapadiya in . the light of the ancient commentaries. Deccan College Building Centenary and Silver Jubulee Series, no. 68. Poona : Deccan College. . Thieme, Paul. 1956. " Panini and the Paniniyas". Journal of the American Oriental Society, 76: 1-23. Reprinted in Paul Thieme Kleine Schriften, vol. 2, pp. 573-595. Wiesbaden : Franz Steiner Verlag. 1971. Tika. Usually ascribed to Punya-raja. Available in Manavalli's (see under Bhartshari above) and Raghunatha Sharma's editions: See Aklujkar 1974 for the view that Hela-raja is probably the real author. Trikandi. See Bhartphari. Tripadi. (Eds.) Abhyankar, K. V. and Limaye, V. P. Mahabh@sya-dipika of Bhartrhari. Post-graduate and Research Department Series, no. 8. Poona : Bhandarkar Oriental Research Institute. 1967-1970.. Upadhyaya, Baladeva. 1968. See introduction of Raghunatha Sharma 1968. Vidyabhusana, Satis Chandra. 1921. A history of Indian Logic. Calcutta. Reprint 1971. Delhi : Motilal Banarsidass. Vtsabha. See Bhartphari (a). Yrtti. See BhartThari (a). Weber, Albrecht. 1862. " Zur Frage uber Zeitalter Panini's" Indische Studien, 5:1-176. Berlin : Ferd. Dummler's Verlagsbuchhandlung. Harrwitz und Gossmann. Yudhisthira Mimamsaka. samvat 2020. Samskrta vyakarana-sastra ka itihasa. Revised edition of volume one. Ajmer : Bharatiya-pracyavidyapratisthana.