________________
EDITOR'S PREFACE
on similar topics, the absence of a connecting link from one stanza to the next, makes it impossible to prove by any sort of inner criticism that the line upon which doubt has been cast by the MS evidence are actually later additions; Sukthankar, whose unit was a whole passage, could invariably manage this when stripping off some striking and generally accepted episode from the Vulgato text of the Mahābhārata. The desperate efforts macle by our scribes to include every stanza they believed to be Bhartrlari's forces us to attach far more weight here to omission than to inclusion, but there is ample evidence for an earlier long period of complete neglect in which omissions must presumably have occurred. In the Pancettantra, Hertel's long series of studies had established versions which I. Edgerton (A. O. Series, vols. 2, 3; 1924 ) later combined to form a consolidated text, though we need not stop to discuss whether the critical method is identical with Suktbankar's. For me, there was no available determination of Bhartrhari versions*, perhaps because a false appearance of uniformity had been thrust upon editions in widely separated parts of the country by the accident of their having been based on what I call version W. Bhartrhari's popularity is of a different sort than that of the Mahābhārata, as it lacks the religious appeal and replaces the interminable doggerel of the epic by crisp, polished stanzas in far more elegant metres which necessarily imply a more cultured if restricted audience. On the other hand, the entire collection of three centuries is short enough to be memorized in toto, while its use as a school text has generally fixed many of the verses in the memory of any Indian who makes the loast pretence to classical knowledge. This adds to the editor's worries, in that many stanzas have undoubtedly been contaminated by such mnemonic transmission, and many have been attracted by similitude to others which were probably original. Further, the poetry shows a formative influence on classical Sanskrit, in that our lesicographers generally quota a line of Bhartrhari to illustrate the meaning of a word or a plurase, though an examination of the MS apparatus leads to the suspicion that a solecism or at least a neologism may originally have existed at the point in question. In spite of the smoothing effect of this type of popularity, solecisms do occur in any given version of Bhartrhari which the remarkable flexibility of the Sanskrit language and assiduity of our commentators cannot remove. Worst of all are the ghost readings that appear because of the copyist or reader (pathaka) having learned from another version, I can affirm from my own experience in collation that even the most carefully trained worker finds his tongue or his pen slip into the reading first learned in his student days, so that far more careful checking is necessary for Bhartrhari collations than for a work not generally
#
D. 1. Kosambi: Some Extant Versions of Bhartrhari's satakas: J.B.B. R.A.S., vol. 21, 1945, pp. 17-32. I hope to publish all major versions ind their commentaries in due course. My abbreviations N, S, V denote the nili, frigūra, and vairūgur śataka respectively,
Jain Education International
For Private & Personal Use Only
www.jainelibrary.org