Some thorny problems in Panini studies: a Japanese scholar remembered.

• Author: Scharfe, Hartmut
• Position: Yutaka Ojihara - Critical essay

This volume of selected writings by the Japanese scholar Yutaka Ojihara (1923-1991, Professor at Kyoto University 1957-1986) emphasizes his work on Sanskrit grammar, which he studied as a student and collaborator of Louis Renou; a few articles are also a reminder of his wide interest in classical Sanskrit literature and Buddhism and his ability to interpret difficult passages in poetry and scholarly literature. However, his editions of two Sanskrit dramas, Japanese translations of Sanskrit literary works, and his text-critical and exegetic notes on the Buddhist Dharmasamuccaya are not included in this volume. He played a major role in linking Japanese and Western Indological scholarship after 1945.

Together with Renou, who was his most influential teacher, he produced a richly annotated French translation of the first pada of the Kasika-vrtti, the oldest preserved running commentary on Panini's grammar. He followed it up with several studies on inconsistencies in this text, attempting to differentiate the contributions of the two authors of this commentary. Jayaditya is usually credited with the first five books, Vamana with the rest; indeed, the contradictions noted by Ojihara are all between the commentary of the first five books on the one hand and the later books on the other, supporting the prevailing opinion. There are indications, though, that both authors may also have commented on the whole Astadhyayi, as Ojihara has pointed out.

Much of his work centered on the Gana-patha, the 261 lists of words (e.g., amsv-adayah "amsu, etc.") attached to Panini's rules, later bundled together as an appendix to the grammar proper. The question has been debated for nearly two millennia, or at least since Bhartrhari, whether these lists predate Panini's rules, are Panini's own work, or are later elaborations. Bhartrhari suggested in his Mahabhasya-dipika that the gana-s whose members are "fixed" (where adi indicates vyavastha "definite arrangement") predate Panini, whereas those that merely indicate a prakara "type" are later elaborations. The latter are often called akrti-gana-s, "lists showing a type." (1) Ojihara's papers, originally published half a century ago, were at the forefront of research at their time, and his detailed translations of the relevant texts were a first. It would not be fair to judge them on the basis of the extensive research conducted since then. But the work of a prominent scholar like Ojihara is, I believe, best honored by a critical evaluation of his contribution to the ongoing debate--and where we stand today.

The debate was triggered by a remark of Katyayana and the ensuing debate in the Mahabhasya. (2) In Sanskrit, as in Latin, Greek, and Germanic, pronominal inflection has encroached on the inflection of nominal forms; (3) but unlike these other languages, in the case of Sanskrit the encroachment was limited to adjectives such as sarva, para, etc. These forms are covered in Panini's grammar in the rules I 1 27-36. The general rule that extends to these words the label of "pronouns" (I 1 27) is followed by several rules that restrict again that label to certain combinations, meanings, or case forms, or make it optional. The last three rules in this set have been controversial from the time of the earliest authorities in the Paniniya tradition, and the controversies continue to our times. The argument started with the observation that these three rules, I 1 34-36, occur in exactly the same form also in the Gana-patha or, more precisely, in the list "sarva, etc." that is attached to rule I 1 27 sarvadini sarvanamani "sarva, etc., are [called] pronouns." Why would Panini be so prolix as to give the same statement twice? Both Katyayana and Patanjali (discussing Katyayana's varttikas), the authors of the two oldest surviving treatises on Panini's grammar, referred three times to word lists (gana) but avoided speaking of a body of lists (gana-patha). In fact, only in the first instance (to be discussed below) is the reference to a passage in what was later called Gana-patha. The other two refer to sections of the Dhatu-pathas "Lists of Roots."

Thus in his varttika 2 on VI 1 6 Katyayana declared that the term abhyasta 'reduplicated' that applies in VI 1 6 [5 abhyastam] jaksity-adayah sat "The six [roots] jaks, etc., [are called reduplicated]" runs to the end of the gana (second-class roots ad-adayah) and therefore needs no (amended) enumeration that identified seven such roots. (4) Actually, they are still followed by four Vedic roots. On VII 4 75 nijam trayanam gunah slau "guna is substituted [in the reduplication syllable] of the three [roots] nij [etc.] before the suffix slu," Katyayana remarked that the word "three" was not necessary, "since it is at the end of a gana." (5) Indeed nij, vic, and vis stand at the end of a section of the third-class roots (juhoty-adayah), followed by Vedic roots. Even though both Katyayana and Patanjali spoke here of gana-s, these gana-s are not listed in the Gana-patha printed as an appendix to the Siddhanta-kaumudi, and only jaks-adayah (not jaksity-adayah!) is cross-listed in Bohtlingk's edition. (6)

It is quite clear from this that the Gana-patha was not an independent text to begin with, but an elaboration of a sutra where Panini had used an abbreviated formulation with "etc." (e.g., -adayah, -prabhrtini); it was part of the vrtti and presented as such in the Kasikavrtti. Patanjali did not use the word gana at all in the technical-grammatical sense, except in his commentaries on varttikas of Katyayana and a slokavarttika that contained the term. (7) In three instances (8) he discussed the content of a gana in his comment on a varttika of Katyayana and twice (9) without such prompting--and always without using the word gana. The constitution of a Gana-patha as a separate text is late. While Panini must have had such lists in mind when he formulated rules like sarvadini sarvanamani, he need not have taught fixed lists; the idea, in modern times voiced perhaps first by I. S. Pawate (10) and elaborated by S. M. Ayachit, (11) Robert Birwe, (12) Ojihara Yutaka, (13) and accepted by Louis Renou, (14) that the Gana-patha existed as a text that is older than Panini's Sutra-patha and influenced it, thus loses, in my opinion, all probability.

There is yet another aspect that has not received the attention it deserves. While we modern scholars can conceive of a development in which...