Jatarupa's Commentary on the Amarakosa, pts. 1 and 2.
Amarasimha's Namalinganusasana, better known as the Amarakosa, is surely the most famous and oldest extant lexicon of the Sanskrit language. The very large corpus of complete manuscripts of the text that has been handed down attests to its great popularity, yet next to nothing is concretely known about its author. Even an edition with a suitably critical apparatus is still outstanding. Writing in circa the sixth century A.D., Amarasimha begins the substance of his work not with one or the other Hindu deity, but with an enumeration of twenty-five different terms for the Buddha, from sugata to mayadevisuta. For this and two other reasons, namely that the first entry of its listing of trees is the Ficus religiosa (bodhivrksa), the tree under which the Buddha achieved his enlightenment, and that the opening stanza is suggestive of a Buddhist religious environment, he is usually considered to have been a Buddhist. The fact that the Amarakosa a is by no means obviously or stridently Buddhist in tenor must no doubt have contributed to its enormous popularity throughout the subcontinent. It is organized into three kanda-sections comprising twenty-five chapters (varga). The first two comprise ten chapters each. The third has five chapters, of which the last deals in forty-six stanzas with a grammatical summary of the rules of gender (lingadisamgraha). It is estimated that more than eighty commentaries were written on it. The well-known histories of Sanskrit lexicography by C. Vogel (1979) and M. M. Patkar (1981) list a good portion of this interpretive plenum, the vast majority of which remain as unstudied as they are unedited. The Amarakosa was also translated into a host of Indic languages, including Sinhalese. A Moggallana (ca. twelfth century) relied a great deal on it when compiling his Abhidhanappadipika, the earliest extant lexicon of Pali. It also exerted no uncertain influence beyond the subcontinent, for it was rendered into Burmese, Nevari, Tibetan as well as Mongolian.
The work under review, Mahes Raj Pant's study and edition of the oldest extant exegesis of Amarasimha's work, titled simply Amarakosatika, by the hitherto barely known Jatarupa, represents in every respect a milestone in the field of Amarakosa studies and Sanskrit lexicography in general. It is based on the author's discovery of two Sanskrit manuscripts that were copied in Nepal. One is housed in the Kaisher Library, Kathmandu, and is dated 1119 [= A]; the other forms part of the collection of the National Archives, Kathmandu, and is dated 1755 [= B]. They were filmed by the Nepal-German Manuscript Preservation Project under, respectively, Reel nos. C 121/1 and A 1031/10 (and once again under B 266/15). Unfortunately, both manuscripts are incomplete. Manuscript A originally consisted of 153 folios, but some ninety-three folios of the text are missing; manuscript B has but twenty-five folios and ends with the seventh varga of the first kanda. Part 1, pp. 57-282, of Pant's work is an incredibly detailed examination of these two manuscripts and their distinguishing features; part 2, pp. 3-325, forms his meticulously executed edition of both manuscripts, to which he has added five appendices (pp. 327-408) and no less than twenty indices (pp. 409-512). Needless to say, both parts are informed by the impressive range of the author's erudition, as indicated in the long bibliography (part 1, pp. 7-53) and by his copious annotations. The latter are sometimes short essays in themselves, as is for example note 21 on the term upakarika, which occupies some nine pages (1: 389-97).
Pant details "Jatarupa: His Time and Place" in part 1, pp. 283-308. Manuscript A is dated 1119 and this forms a convenient terminus ad quem for Jatarupa's work. The fact that, with his stunning knowledge of a wide range of Sanskrit literature, the author was able to trace an unidentified quotation in Jatarupa to Rajasekhara's (early tenth century) Viddhasalabhanjika drama (2: 70) leads him to conclude that Jatarupa must have flourished in the second half of the tenth century, at the earliest. In the absence of a sufficient number of exact dates, one method that needs to be used for arriving at a more or less satisfactory dating of Jatarupa and his commentary is to establish a relative chronology among the oldest known Amarakosa exegeses; these are: (1) Sarvananda's 1159-60 Tikasarvasva, (2) Ksirasvamin's undated Amarakosodghatana, and (3) Subhuticandra's equally undated Kamadhenu (or, as Pant prefers, Kavikamadhenu). Before continuing, we must note an observation made by G. Cardona that merits repeating. Manuscripts of texts travelled across India at a much faster pace than is usually assumed, and there is evidence that two decades or so after composition was often sufficient for one to reach either extremity of the subcontinent. Jatarupa is cited by Sarvananda and, in Pant's view, by Ksirasvamin as well, so that he would also antedate the latter. Contrary to the dating of Kasirasvamin proposed by C. Vogel (namely, the first half of the twelfth century) and giving a more convincing interpretation of the author's colophon of Ksirasvamin's Ksiratarangini, Pant argues that Ksirasvamin flourished one century earlier and that, in fact, it indicates he was a contemporary of King Bhoja (ca. 1000-1055) of Malava. Beginning with Ksirasvamin, the Amarakosa commentators regularly and variously quote from this king's dense studies of grammar, literary criticism, and poetics, such as the Sarasvatikanthabharana and the Srngaraprakasa. Only Jatarupa does not do so in his extant comments on the corresponding Amarakosa entries. This detail provides the first bit of circumstantial evidence that he lived before Bhoja's literary treatises. Pant also demonstrates that Jatarupa was familiar with Sridhara's Nyayakandali of 991-92, and thereby establishes this as the terminus a quo for his commentary. He shows furthermore, on the linguistic evidence of the commentary itself, that Jatarupa was most probably a native of Bengal. As is known, the term "Gauda" is often used in the sense of Bengal. Ksirasvamin critically refers to an exegete by the name of Gauda on some seventeen occasions, of which nine occur in the relevant Amarakosa entries for which we have available Jatarupa's comments. Six of these give the strong impression that he targeted Jatarupa, while the remaining three tell us nothing in this regard. The author concludes on the strength of this corpus of circumstantial evidence that Jatarupa lived around the year 1000 and that he was anterior to Ksirasvamin. Though not airtight, the evidence he provides points to the very high probability that he is right on this score. Lastly, Sarvananda cites both Jatarupa and Ksirasvamin, but not expressly Subhuticandra. Pant adduces in part 2 parallel and identical passages in the comments on Amarakosa I. 1 for Jatarupa and Ksirasvamin (pp. 355-58) and Jatarupa and Sarvananda (pp. 371-82).
The relative chronology of Jatarupa and Subhuticandra is uncomplicated, for the former is doubtless anterior to the latter. Fairly well known in India, Subhuticandra and his work were also quite familiar quantities in Tibetan scholarly circles, and I believe there is very little reason to doubt that he is the very same Subhuticandra who is mentioned in the long and convoluted translator's colophon ('gyur byang) Pa tshab Lo tsa ba ("Sanskritist cum translator") Tshul khrims rgyal mtshan appended to his Tibetan translation of the enormous Aryasaddharmasmrtyupasthanasutra. (1) There we learn that Pa tshab had studied the sutra in Nalanda monastery under, among others, mahapandita Abhayakaragupta, and that he had received instructions on the same from mahapandita Subhuticandra while at Vikramasila monastery. He says of the latter that he was "a scholar of grammar (sgra, sabda), poetics (snyan dngags, kavya), (2) and the 'modality' of the Sanskrit language (sgra dang / snyan dngags dang / legs par sbyar ba 'i skad kyi lugs la mkhas pa)"; the latter phrase may, but only may, indicate lexicography. The first draft of the translation was completed, as Pa tshab says, sometime during the reign of the Pala King Ramapala, who ruled from about 1072 to 1126. (3) Given that Abhayakaragupta flourished somewhere between roughly 1060 and 1125--his last dated work, the enormous Amnayamanjari commentary on the Samputatantra (and a great deal else besides), was completed in Ramapala's thirty-seventh regnal year (circa 1110)--we can therefore predicate more or less the same of Subhuticandra, who was probably at that time a senior scholar as well, if only because be, too, is styled a mahapandita. Pant dates his Amarakosa commentary to about the third or fourth decade of the twelfth century, but this may have to be pushed back a few decades. The translation of the sutra can perhaps be dated not earlier than the first or the second decade of the twelfth century, and Subhuticandra must already have been a senior scholar at this time. The four manuscripts of the Sanskrit text of the Kamadhenu that have been located so far are all fragmentary. Judging from my (cursory) reading through the 1750-57 (or 1748-56) Tibetan translation of a virtually complete text of the Kamadhenu by Si tu Pan chen Chos kyi 'byung gnas' (1700-1774), Subhuticandra nowhere explicitly refers to either Jatarupa (Tib. *Skye gzugs) or Ksirasvamin (Tib. Zhi ba'i rje). (4) The Kathmandu Valley was Si tu Pan chen's source for several incomplete manuscripts of the Kamadhenu he was able to use for this translation and his other studies of Indo-Tibetan lexicography.
Pant also discusses the few details about Jatarupa and his scholarly practice that can be gleaned from his commentary (1: 409-46). Jatarupa was probably a Buddhist. He points out that Manuscript A begins with a line of homage to the Buddha (namo buddhaya //) and B with one to the Omniscient One (om namah sarvvajnaya ///). He rightly refrains from concluding that this necessarily implies he was a Buddhist. Indeed, the very fact that we have these variant readings suggests that at least one of these reflects the religious sentiments of their unknown scribes. But he does indicate that Jatarupa's comment on the indeclinable particle khalu in Amarakosa III.3.252d (2: 289) reads atha khalu bhagavan, which is "well known from Mahayana texts."
The two manuscripts of Jatarupa's Amarakosatika that entered the Kathmandu Valley at an unknown time, and were then copied there, once again demonstrate the key role the valley has played, and continues to play, in the transmission and preservation of the Indian subcontinent's most valuable literary treasures. It is thus quite fitting that Pant, among the very best of the scholars currently active in the valley, has laid before us such an exquisitely arcane piece of work as this exhaustive study. It is a monument to the kind of scholarship that is now, in the face of unrelenting modernity, progressively and regrettably on the decline.
(1.) The Tibetan Tripitaka, Taipei [= Sde dge] Edition, ed. A. W. Barber (Taipei: SMC Publishing Inc., 1991), vol. 15, no. 287 [#287], 65/7-6/3 [Sha, 228a-9b]. We may note here that, in spite of the fact that Pa tshab expressly states that he made use of standardized orthography and nomenclature, as indicated by his use of the expression bkas bcad, the last post-colophonic line by an unknown editor states: "There also appears some dissimilarities with the obsolete terminology of yore" (yi ge'i brda sngon gyi rnying pa dang mi 'dra ba cung zad kyang snang). Aside from the usage of bkas bcad, "[standards for nomenclature] determined by royal decree," and skad gsar bcad, "determination for new terminology," in connection with the Tibetan translations of Buddhist scriptures during the imperial period, Nyang ral Nyi ma'i 'od zer (1124-92) also uses the latter expression in the ecclesiastic chronicle he wrote towards the end of his life, with reference to its application in the era of the royal monk Lha Bla ma Zhi ba 'od (1016-1111); see his Chos 'byung me tog snying po sbrang rtsi'i bcud, ed. Nyan shul Mkhyen rab 'od gsal, Gangs can rig mdzod 5 (Lhasa: Bod ljongs mi dmangs dpe skrun khang, 1988), 465.
(2.) This expression is not listed in Dbus pa Blo gsal Byang chub ye shes' (ca. 1265-1355) undated Brda gsar rnying gi rnam par dbye ba, a glossary of obsolete terms (brda rnying) and their updated (brda gsar) equivalents; see Mimaki Katsumi, "Dbus pa Blo gsal no 'Shin kyu goi shu' k[??]tei bon shok[??]," in Asian Languages and General Linguistics: Festschrift for Prof. Tatsuo Nishida on the Occasion of his 60th Birthday (Tokyo, 1990), 17-54. However, the cognate Brda gsar rnying gi rnam gzhag li shi'i gur khang of 1536 (ed. Mgon po rgyal mtshan [Beijing: Mi rigs dpe skrun khang, 1991], 16), attributed to Skyogs ston Lo tsa ba Rin chen bkra shis (ca. 1485-?), registers dngags as an obsolete form of the updated ngag. Dpa' ris Sangs rgyas, Das yig rig pa 'i gab pa mngon phyung, ed. Dbang phyug mtsho mo (Xining: Mtsho sngon mi rigs dpe skrun khang, 1999), 447, rightly dismisses the supposition of unnamed others that, whereas snyan ngag denotes verse characterized by the use of poetic figures, snyan dngags refers to a treatise devoted to an exposition of such figures.
(3.) D.C. Sircar, Some Epigraphical Records of the Medieval Period for Eastern India (New Delhi: Abhinav Publications, 1979), 31.
(4.) Ming dang rtags rjes su ston pa'i bstan bcos 'chi med mdzod kyi rgya cher 'grel pa 'dod 'jo'i ba mo, Collected Works, vols. 4 and 5 (Sansal: Shesrabling Institute of Buddhist Studies, 1990), 243-738, 2-421.
LEONARD W. J. VAN DER KUIJP
|Printer friendly Cite/link Email Feedback|
|Author:||Van Der Kuijp, Leonard W.J.|
|Publication:||The Journal of the American Oriental Society|
|Article Type:||Book Review|
|Date:||Jan 1, 2003|
|Previous Article:||Brahmabandhab Upadhyay: the Life and Thought of a Revolutionary.|
|Next Article:||The Haunting Fetus: Abortion, Sexuality, and the Spirit World in Taiwan.|