________________
68
INTRODUCTION
the verses seem poor stuff indeed, halting, patched up, without the sting or compression that characterizes the best of Bhartṛhari. The similarity of phrase which seemed conclusive to the first editors are to my jaundiced eye such trivial borrowing as anyone might have picked up. Indeed, far better stanzas in the style and spirit of Bhartṛhari have actually been writen, following a jocular remark of mine, by Pt. Krṣṇamurti Šarmā.
The matter is clinched by a bit of negetive evidence: not one of the Vitavṛtta or the Vijñanasataka stanzas is found in the bulky third group, so that no one in the general tradition could ever have heard of either of these two works as Bhartṛhari's. I particularly recommend my group III to enterprising forgers who have designs upon Bhartṛhari, in the hope that they, at least, will patronize this volume.
3. 6. Discussion of the groups. The first question that will arise after examination of such detailed criticism would be: how reliable is the critical method? Has the grouping any real sense? It may be said at once that the basic portion of version B, particularly in its omissions, was fairly well outlined by the procedure I have adopted, and that the discovery of an ancient survival in version J confirmed a good many of the omissions. The question, however, can only be settled by new discoveries of decidedly older MSS though this is by no means certain. The oldest reference to any specific work of Bhartṛhari is in the final section of Merutunga's Prabandhacintamani [1304 A. D.], "tena Bhartṛharina vairagyaśatakādi prabandhani bhāyāmś cakrire". Nevertheless, the V comes off very badly by the present method, inasmuch as barely 53 stanzas survive, though N and S leave 69 and 71 respectively. Moreover, while the other two "satakas" have five slokas each with starred numbers, the V has no less than 14; comparing it against the other two by the modern statistician's chi-square test, the probability of the the V being formed and transmitted by the same mechanism as the other two is negligible.
A certain amount of supplementary information is available from MSS received after printing off the major portion of the text. Nagpur 1087 omits 16, probably in its haste to include kṣudrah santi [471]; with Nagpur 421, it substitutes 451 for 184*. HU 1387 omits, among others, nos. 56-52, 91–96, 108, 111, 117, 196-199, 224, 233, 300, 332 348; no attention can be paid to this as the original from which the scribe copied the specimen clearly had missing folios. Similarly for the omissions of Bikar 3279: nos. 15-18, [21 shifted to V 157 (47)], 22, 26, 38*, 51, 53*, 265, 285, 328, 331. Bikaner 3281 is obviously of the same version as Bikaner 3278, so that the omission of 10 stanzas after S 10, along with their numbers, means nothing. HU 2144 omits 82, 149, 161, 179*, 187, 319, 349. HU 2145 omits 74* while substituting 291 for 190*. This substitution also takes place in PU 496 [which omits 132* and 301], and Bikaner 3280 which joins HU 2145 in also substituting mālatīkusuma [650] for 34. These are all N MSS. Among those of type Y we find HU 2133 omitting 120 and 328 while it repeats 77 as $ 19 and Ś 71(77); but its S is disturbed so that no attention need be paid to this aberration. Bikaner 3277, supported by fragments from the same locality, omits 165, 209, 226*, 273, 274, 278, 230*, 318, 322, 324*. The Nepal MS, supported for N and V by Jodhpur 6, omits 239, 230*, and 278.
Among the
Jain Education International
For Private & Personal Use Only
www.jainelibrary.org
