________________
DECEMBER, 1923]
BOOK-NOTICES
377
as other parvans do, and the actual point for con- sideration is as to the actual criterion of oriticism for the discovery of these interpolations. A mere criticism on the best of ideas will not constitute A standard as individual opinions on matters like this, are likely to differ very widely indeed. There fore the best and perhaps, in the present state of study of the Mahabharata, critically the most soceptable course would be to reconstruct the text on the basis of the manuscripts. If this reconstruction brings us close to the enumeration in the parvasangraha parva, we gain at least one step, and that a long step, forward in the recovery of the original text.
The parva sangrahaparvan gives, both the Nagari and tise Southern Recensions happen to be in agreement in giving to the Virafaparvar, 67 adhyayas and 2,050 verses. As against this, the two Nagari editions give 72 adhydyas, and 2,272 verses and 2,376 verses respectively. The most recent houthern printed edition, that of Kum. bhakonam, gives to it 78 adhyiyas and 3,494 verses, the Grantha edition giving 76 adhyayas and 3,281 verses ; thus exhibiting a comparatively small difference in respect of chapters and verses as between the two southern editions. By adopt. ing mainly the ordinary principles of manuscript criticism only, the editor has produced a text of 2,033 lines. The division of chapters is a matter perhaps of later arrangement, and actually is of less importance. Thus the difference between the total of verses according to the parvasangrahaparua and the tentative edition is that the former has 17 more verses. As against these 17, there are 35 half verses, which are all collected in an appendix on the autho. rity of the manuscripts, inost of which happen to be extra lines to the two line stanzas. If this could be taken as the equivalent of 17 elo as the total quantity comes up to be the same with a differ. ence of a half sloka. This ought to be regarded 49 a great success we the new text is vouched for by manuscript authority. and the critical texts applied are within very reasonable limite of indi. vidual opinion. According to the editor, "the passages which are now considered as interpolated on the evidence of the manuscripts are (1) mostly repetitions, or (2) meaningless additions ; (3) those which cannot be regarded as necessary to the texts by any cogent line of argument, (4) passages otherwise considered interpolated and which are absent in the southern recension, and (6) similar pasaages not found in the Bengali manuscripts." The parvasangrahaparva dating back to at least A.D., 500 and the manuscripts most relied on going back to the fifteenth century A.D., a mere manuscript tradition would justify the wasumption that for about a thousand years the manuscript tradition continued to be handed down without much corruption. This position in regard to the
manuscript tradition is confirmed by the fact that the passages which are, from the point of view of the manuscripts themselves, regarded as interpolatod are uniformly omitted from the southern recensions : while there is overy possibility of additions being made for various reasons, anything like e curtailment, it would be difficult to prove if postulated. Comparing the reckoning as contained in the first chapter and the second chapter of the Adiparvan, it is found that the second reckoning refers to a period when the Mahabharata was divided into 18 parvane, while that of the first chapter refers perhaps to a period anterior to that. The concluding portion of the passage in the second chapter makes it absolutely clear, that the 18 parvan Bharata was the edition of Lomaharsha, whereas the previous one was one of a hundred parvans by Vyása, though it is possible that the word parvan is not used in the same pense in the two contexts. The parvasangraha having continued the same in all the recensions, north and south, we have to accept it as the reckoning according to the original editor. It would seem however that there should have been vast additions, and at the same time the chapter which gives the reckoning should have remained the same. There is one explanation possible for this. The expansion, which seems to us very vast in the southern recension, appears to be, most of it, if not all, of the character of the expansion of the original text, the original being swamped by the additions. Since this expansion seems to have been more or less due to the sense of propriety of the redactors of the Mahd. bharata, it seems to have been of the character of a mere exposition of the original texts, and as such even the vast additions were not actually regarded as addition to the substance of the whole work. That seems how it is that the so-called interpolations hre been coming in, and that perhaps accounts for te original reckoning being left uninterfered with.
What is said above in regard to the character of the expansion of the southern recension would perhaps explain why some of the blokas found in the northern recensiou are not found in the southern. This will also satisfactorily account for the Swamping of the portions of the original text and the removal of features which might be regarded as crude and unrefined. Hence the editor prefers inclusion of lines and Sloka found in the three manuscripts whose reliability he has taken pains to demonstrate. Hence it seems justified that some of the blokoe not found in the southern recension are worthy of inclusion in the critical text. The main point in each case will however be whether the idea has not been worked up in any corresponding "interpolated" passago, the working up showing the character of expansion and removal of features that jar upon the taste of the redactor or the editor. The editor finds that out of the 3,494 floreas of the Vindfaparvan in the published Davanagari edition of the southern