________________
64
INTRODUCTION
inserted on margins or just before or after colophons ( which occasionally slip into the following sataka] and frequent additions in the main body of the text proper. Had there been some clearly visible unclear portion to which all the rest could be treated as additions, the critical problem would find an immediate solution. Unfortunately, as far as can be seen, this process of inflation is quite general, and the common portion does not occur in the same order. That the inajor differences are due to addition, not omission, is supported by the increasing divergence of N versions in the chart as one nears the end of any śataka. The obvious principle-that of suspecting large numbers in any Century-cannot be applied directly because of the tendency to preserve the original ending, and additions by similitude. One conclusion is that the collection could not have been promulgated by the author in any such form, and that the work is some sort of an an anthology, not by Bhartphari, but by later editors who believed they were gathering together Bharthari ślokas. It might seem, therefore, that a stanza being omitted from a single codex would be prima facie evidence of its being spurious, or at least a later addition. Unfortunately, this is not true, as there is sometimes omission by inadvertence, particularly when the scribe is copying an unfamiliar text. The first step, therefore, is to see whether an omission can be explained. This is best done when it occurs in some individual MS. of an established version, while the general MSS. of that version include the stanza. Thus, the well determined versions occupy a specially important position in the critical apparatus. We are safe in stating that no importance should be attached to the omission of 194 in A3 or 110 in Ws. Then there are other cases where one can see quite clearly omissions which must be due to the copying of lacunary exemplars. Bikaner 3281, for example, omits § 11-19 with the numbers. GVS 2387 bas dropped something like ten consecutive stanzas around V30, NSi an unknown number at the end, HU 1387 some at the end of every sataka, and Bikaner 3279 about 19 in the first quarter of Niti; each of these, except the first, shows no gaps in the numbering, but parallel MSS of the archetype make it amply clear that the disturbance must be due to missing folios in the originals from which these are copied. One such large gap is shown in F6 at $90. It is not clear whether the process might not have worked in reverse, and many consecutive extra stanzas in texts like ISM Kalamkar 195 might not have come from some intruding folio, as in RASB 11030 whose initial folio krs strayed from the Siddhānta Kaumudi, unnoticed because of identical size and calligraphy. Some unplaced stanzas, like 1, are omitted in composite sources which copy their śatakas from different versions or recensions. But no amount of such juggling can serve to explain any considerable portion of the omissions.
There are other types of omissions which do not arise from such defects in the mechanism of transmission. In particular, we have a certain number of paraphrases in our text. Perhaps the most striking is 291. in mandākrāntā metre for 190 in śārdūlavikrīdita, substituted in F3.5, BU, HU 2145, PU 496, Bikaner 3280; and given besides the other in some other northern MSS with Maharāştrian influence. Similarly for 145* = 244, and 129 = 205. Stanza 34 is so closely approximate by mālatīkusu mxr [650]
Jain Education International
For Private & Personal Use Only
www.jainelibrary.org
