Pramāṇa · Search · About · API

Critical Studies in the Mahabharata (V. S. Sukthankar Memorial Edition, Vol. I) c4: Foreword

Critical Studies in the Mahābhārata (V. S. Sukthankar Memorial Edition, Vol. I) c4

Language
English

Contents

Foreword

c4.1

¹ [Ādiparvan, Fascicule I, 1927.]

c4.2

An elaborate introduction containing a comprehensive account of the manuscript material as also a detailed discussion of the principles of Mahābhārata textual criticism will be published with the last fascicule of the Ādiparvan. The following cursory remarks are intended merely to guide the reader meanwhile through the labyrinth of a very complicated *apparatus criticus*.

General Observations

c4.s1.1

In the Mahābhārata manuscript tradition, perhaps as much as in any literary tradition, the textual critic is faced with a bewildering profusion of versions as well as an amazing mixture of versions. Divers elements have been working from the earliest times in favour of the development of different types, on the one hand : on the other hand, there were not wanting elements that operated against the evolution of sharply differentiated types. To understand the phenomenon of this luxuriant growth and indiscriminate fusion of versions, one must appreciate clearly certain details of historical moment, certain special factors in the transmission of the Mahābhārata—traits which distinguish our work from every other known text except the Rāmāyaṇa and possibly the Homeric epopees. Notwithstanding the fact that we know so little that is certain and definite about the early history of the text, we may, it seems to me, with confidence assume that after its composition the great epic was for centuries handed down (in differing forms and sizes) from bard to bard merely by word of mouth. It is moreover extremely probable that even after the text had been written down, large portions of it, especially such portions as were popular, continued to be committed to memory, by itinerant *raconteurs* for purposes of recitation. It is further easy to believe that no great care was lavished on the text by these custodians of the tradition to guard it against partial corruption and elaboration or against arbitrary emendation and normalization : to reproduce the received text with any great precision would be neither attempted by these bards nor required of them. It was then inevitable that the protean oral tradition should in one form or another react on the written tradition and *vice versa*. One important and necessary consequence of such antecedents as these is the impossibility of retracing all extant versions to any fixed and authentic archetype ; since some of the modern editions could not but be descendants of fluctuating oral versions reduced to writing in some distant past, independently of each other, at different epochs and in different circumstances. In Devanāgarī Versions (D). Devanāgarī Version of Arjunamiśra (Da). Da₁ = Poona, Govt. MSS. Collection, MS. No. 30 of A 1879-80. Da₂ = Poona, Govt. MSS. Collection, Viśrāmbāg Coll. I, MS. No. 468.

c4.s1.2

Devanāgarī Version of Nīlakaṇṭha (Dn). Dn₁ = MS. belonging to Sardar Kibe of Indore. Dn₂ = Mysore, Oriental Library, MS. No. 1064. Dn₃ = Poona, Govt. MSS. Collection, MS. No. 234 of 1892-1902.

c4.s1.3

Devanāgarī Version of Ratnagarbha (Dr). Dr₁ = Tanjore, Palace Library, MS. No. 1264. Dr₂ = Tanjore, Palace Library, MS. No. 1199. Dr₃ = Tanjore, Palace Library, MS. No. 1313. Dr₄ = Tanjore, Palace Library, MS. No. 1339.

c4.s1.4

Devanāgarī Mixed Versions. D₁ = Poona, Govt. MSS. Collection, MS. No. 29 of A 1879-80. D₂ = Tanjore, Palace Library, MS. No. 1152. D₃ = Tanjore, Palace Library, MS. No. 1360. D₄ = Tanjore, Palace Library, MS. No. 1126. D₅ = Lahore, Dayanand Anglo-Vedic College, MS. No. 4. D₆ = Tanjore, Palace Library, MS. No. 1223. D₇ = Tanjore, Palace Library, MS. No. 1269. D₈ = Tanjore, Palace Library, MS. No. 1329. D₉ = Tanjore, Palace Library, MS. No. 1176. D₁₀ = Tanjore, Palace Library, MS. No. 1293. D₁₁ = Tanjore, Palace Library, MS. No. 1340. D₁₂ = Tanjore, Palace Library, MS. No. 1373. D₁₃ = Poona, Govt. MSS. Collection, Viśrāmbāg Coll. II, MS. No. 191. D₁₄ = Poona, Govt. MSS. Collection, Viśrāmbāg Coll. II, MS. No. 266.

II. S(outhern) Recension

c4.s2.1

Telugu Version (T). T₁ = Yadu Math Collection MS. (without No.) T₂ = Tanjore, Palace Library, MS. No. 11865.

c4.s2.2

Grantha Version (G). G₁ = Yadu Math Collection MS. (without No.) G₂ = Yadu Math Collection MS. (without No.) G₃ = Tanjore, Palace Library, MS. No. 11823. G₄ = Tanjore, Palace Library, MS. No. 11838. G₅ = Tanjore, Palace Library, MS. No. 11851. G₆ = Tanjore, Palace Library, MS. No. 11860. G₇ = Yadu Math Collection MS. (without No.)

c4.s2.3

Malayālam Version (M). M₁ = MS. belonging to Chief of Idapillai, Cochin. M₂ = Cochin, State Library, MS. No. 5. M₃ = Cochin, State Library, MS. No. 1. M₄ = MS. belonging to Kallenkara Pisharam of Cochin.

c4.s2.4

In addition to the above, two Baroda Library MSS. of the commentary by Devabodha (without the epic text) were collated; the important readings found in this commentary have been cited with the symbol Cd.

General Characteristics of the Manuscripts and Their Mutual Relationship

c4.s3.1

An important advance made in the classification of the manuscripts is the separation of archetype K (which represents Devanāgarī transcripts of the Kāśmīrī or North-western version) from other so-called Devanāgarī versions. The affinity of K is documented by the following agreements :

c4.s3.2

1. 1. 2ᵇ : K ( except K₁) महर्षीन्; D₁₄ महर्षि; the rest ( inclusive of K₁) ब्रह्मर्षीन्.
1. 1. 8 : K लोमहर्षणिरुवाच ; V₁ लोमहर्षण उ°; the rest सौतिरु°; सूत उ° or सूत:.
1. 1. 49ᵇ : K V₁ B₁ m संक्षेपतोब्रवीत् ; the rest संक्षेपमब्र° or संक्षिप्य चाब्र°.
1. 1. 51ᵇ : K विचक्षणाः; V₁ विशेषतः; the rest मनीषिण:.
1. 1. 192ᶜ : K ( except K₆) धृतपाप्मानः; the rest ( incl. K₆) शुभकर्माण:.
1. 2. 23ᵇ : K ( except K₁₋₆) यथावदिह संख्यया; the rest ( incl. K₁₋₆) संख्यातत्त्व-विदो जन:.

c4.s3.3

1. 2. 76ᵈ : K ( K₅ missing) यक्षाणामथ रक्षसां; N₂ दैत्यानामथ र°; the rest यक्षाणां च महौजसां.

c4.s3.4.1

Only K₀,₁ represent archetype K in a *comparatively* pure form. Next to these in purity stands K₂; while K₃₋₆ are nothing but misch-codices, K₃ being conflated with the “vulgata” and K₄₋₆ with some Southern version (cf. 1. 1. 26 and 22*, 25*). K₀,₁ contains a text which is shorter and more archaic than either the Bangālī or the [5] “vulgate.” It is worthy of note that while these two manuscripts (K₀,₁) have more readings in common with manuscripts of the Southern recension than either of the two latter groups, yet they contain not a single “additional” line which could be considered as a characteristic Southern interpolation.—V₁ stands, as is to be expected, nearest to the Bangālī version, but it is noteworthy that in a few cases V₁ and K agree in opposition to all other manuscripts, where it is impossible to consider the agreement as purely accidental (cf. 1. 1. 8, 49ᵈ).—The Bangālī version is slightly superior to the “vulgate,” in so far that it is not interpolated quite so heavily as the latter. Bangālī omits (like K) not only the Brahmā-Gaṇeśa episode in the first adhyāya, but (unlike K) also the short dialogue between Paraśurāma and his ancestors in the second, both unquestionably spurious and both found in the “vulgate.” Occasionally Bangālī manuscripts agree with Southern manuscripts in opposition to K and the “vulgate”. (cf. 1. 1. 22ᵇ, 42ᵈ).

c4.s3.4.2

In these cases I have adopted as authentic the concordant readings of the Bangālī and Southern manuscripts in preference to those of K. Bangālī alone has in a few cases preserved the correct reading as compared with all the other manuscripts (cf. 1. 1. 62ᶜ).—Closely connected with the Bangālī is the version of Arjunamiśra. It not only agrees with the Bangālī in the omission of the Brahmā-Gaṇeśa episode and of the dialogue mentioned above, but it shares with the Bangālī quite a number of isolated readings against all other manuscripts. The (epic) text in the Arjunamiśra codices is frequently contaminated from the "vulgate" and has to be corrected with the help of Arjunamiśra's commentary (cf. 1. 1. 17ᵇ, 22ᵇ).—Ratnagarbha's text is eclectic, standing palpably under the influence of the Southern recension (cf. 24*, 25*, 27*).—Nīlakaṇṭha's version (which may conveniently be styled the "vulgate") presents a smooth text, with an inconsiderable amount of Southern element. It may be noted that in rare cases the Bombay edition (of Gaṇpat Kṛṣṇājī) contains readings which have no manuscript support at all or have at best very weak manuscript support.—Next we have the mixed Devanāgarī group represented by D₁₋₁₄, misch-codices of small trustworthiness and of no special value for critical purposes.

c4.s3.4.3

D₂₋₁₄ contain a very large number of old readings in common with manuscripts of the K group, but D₁₄ shows at the same time some Southern readings and some unique readings not found elsewhere (cf. 1. 1. 50ᶜ, 63ᶜᵈ; 2. 101ᵇ). D₉₋₁₂ are palpably under the influence of the Southern tradition.—Telugu manuscripts have been placed in the Southern recension, but they belong in a sense to both recensions; they are eclectic on no recognizable principles.—It is difficult to define precisely the relationship between the Grantha and the Malayālam versions, which are very closely allied; each of them exhibits nevertheless certain features not found in the other (cf. 1. 1. 184ᵈ). On the whole the Grantha version produces the impression of being less interpolated and more archaic than the other. This version has two sub-groups, G₁₋₃ and G₄₋₆; numerous cross-agreements between the two sub-groups show that our manuscripts are conflated. G₇ does not belong to either of these groups; in point of fact, it is a misch-codex contaminated from the Northern tradition and closely allied to T₁ (cf. 1. 1. 63, 64).—M₁ often stands in antagonism to M₂₋₄, sometimes agreeing with manuscripts of the Northern recension (cf. 1. 1. 32ᶜᵈ, 41ᶜᵈ, 71ᵃ, 128ᵃ). The Southern recension, as already remarked, agrees with archetype K more closely than with any other Northern version.

c4.s3.5

[6] The Southern version of the first two adhyāyas is on the whole shorter than the "vulgate"; but the shortest version of these two adhyāyas is that preserved by K₀.₁, Ko being probably even shorter than K₁. The naïve Brahmā-Gaṇeśa episode, the longest as well as the most obvious interpolation in the text of the "vulgate", has been relegated to the Appendix (cf. 1. 1. 26, 53, 60, 62, 64). Its spuriousness has now been placed beyond the domain of sane criticism through its absence in K₀.₃ V₁ B Da D₁ M₁. In the Southern manuscripts (and in some conflated Devanāgarī manuscripts) Brahmā alone is introduced; in these there is no talk of Gaṇeśa, who is unquestionably a late Northern intruder. The yadāśrauṣam section is also evidently an interpolation, but a considerably older one. Being merely a string of stanzas summarizing some of the most important incidents and episodes of the epic, it lent itself easily to being further interpolated by revisers who wanted to supply the omissions and thus make the summary as complete as possible. To present the oldest form of this section now recoverable, I have deemed it sufficient to accept as genuine only such stanzas as are found verbatim in both the recensions. Here again K₀.₁ have the fewest interpolations. The “table of contents” in the second adhyāya is preserved in two versions, a longer and a shorter. K₀.₁ S (except G₇) have the shorter version, which has been adopted in the constituted text as the authentic one.

c4.s3.6

Since I have not been able to discover any traces of “secondary interrelationship” between archetypes K and S, I consider the agreement between these two archetypes as “primitive”, that is depending upon their primitive connection through the Ur-Mahābhārata. This concord is a factor of supreme importance for the reconstruction of the text. The originality of the agreement is established, in my opinion, by the following considerations. The concordant readings of K and S represent as often as not a *lectio difficilior* (cf. 1. 1. 19ᵃ, 94ᵈ, 158ᵈ). Frequently such a reading best explains the other variants (cf. 1. 1. 1. 2ᵃ, 14ᵇ). Furthermore the “additional” stanzas which are found in the “vulgate” but are missing in K and S have all the appearance of being interpolations, lengthening and weakening the text (cf. 19*, 29*, 35*-38); the same remark applies to the additional stanzas that are found in the Southern recension but are missing in K and the “vulgate” (cf. 21*, 22*, 27*). The high position of K seems confirmed by its being the shortest of the known versions.

The Constituted Text

c4.s4.1

In preparing the constituted text of the first two adhyāyas, I have endeavoured to balance the eclecticism advocated in certain matters with a rigid conservatism insisted on in others. I have been most averse to reject or correct the readings of good manuscripts. Interpretation has throughout been given precedence over emendation; in the first two adhyāyas, no emendation seemed absolutely necessary, nor any absolutely certain. Solecisms, when shewn to be original by a clear agreement on this point between (what appeared to be) independent versions, have been allowed to stand uncorrected (cf. 1. 1. 5ᵃ, 170ᵈ). As a general rule, preference is given to a reading which best suggests how other readings might have arisen. When such a reading was not available the choice fell upon one which is common to (what *prima facie* appeared to be) more or less independent versions and which is supported by intrinsic probability; the presumption of originality in such cases is frequently confirmed by a lack of definite agreement between the dis-cordant versions. Occasionally one comes across variants where the matter is identical but the wording of a large part or of the whole of the line is different; one and the same primitive reading cannot in these cases account for the divergence. In the presence of such alternatives, neither of which can have come from the other and which have equal extrinsic support and equal intrinsic merit, the criteria mentioned above fail to give a positive result. A particular instance of the variation contemplated here is that of a puzzling form of a cross-agreement between the Northern and Southern versions; when, for instance, some Kāśmīrī and Malayālam manuscripts agree in opposition to say, Bangālī and Grantha (cf. 1. 1. 60^c). In such cases, I have, owing to the much greater correctness of the K version, mostly adopted, as stopgaps, the readings of K₀.₁, the manuscripts which present the archetype K in a relatively pure form.

c4.s4.2

I have given in the constituted text whatever in each case appeared to be supported by the balance of probabilities, indicating all the important elements—lines, phrases, significant words and word-parts—of the text that are less than certain by a wavy line printed below them. Insignificant differences of spelling (e.g. Naimiṣa-Naimiśa) are ignored for this purpose.

c4.s4.3

Doubt which cannot be resolved by a consideration of the documentary or intrinsic probability, entailing the use of the wavy line, arises in the following cases: (a) when the transmitted readings appear to be corrupt and no satisfactory emendation can be suggested; (b) when there are several readings of equal merit; in particular where the Northern and the Southern recensions offer two different readings of equal value; lastly (c) when the evidence *pro et contra* of documentary and intrinsic probability is equally balanced.

c4.s4.4

As regards interpolations, the additional lines are so ingeniously fashioned and cunningly fitted in, that in any given case the intrinsic evidence is generally inconclusive. In other words, if we leave out of account the documentary evidence, no convincing proof can in general be brought forward to establish either the originality or the spuriousness of the added lines. We cannot, however, entirely ignore the evidence of tradition. Everything points to the fact that what the epic has suffered from is inflation and elaboration, and not depletion or curtailment. On principle, therefore, lines that are peculiar to one recension, *having nothing whatsoever corresponding to them, at the same point, in the other recension*, are to be viewed with grave suspicion. Unless there is overwhelming evidence to prove their originality, they should be treated as spurious; because, the probability of error is far greater in admitting as authentic such one-recension lines on insufficient evidence of originality (both recensions being placed on an equal footing and treated with impartiality) than in rejecting them on insufficient evidence of spuriousness. It may be added that the presumption of unauthenticity is frequently confirmed by the fact that in the recension in which such lines do occur, they are found inserted in different manuscripts (or different versions) at different points of the text.

c4.s4.5

I am greatly indebted to Shrimant Balasaheb Pant Pratinidhi, Chief of Aundh, for uniform kindness and courtesy; but still more for the absolute confidence he is pleased to repose in me. I must also record my thanks, for help of various kinds, to my colleagues on the Mahābhārata [8] Editorial Board : Prof. VAIJNATH K. RAJVADE, M.A. ; Mr. VISHWANATH P. VAIDYA, BAR-AT-LAW ; Prof. Dr. R. ZIMMERMANN, S.J., PH.D. ; Prof. Dr. V. G. PARANJPE, M.A., D.LITT. ; and Mr. N. B. UTGIKAR, M.A. But I desire to make a special mention of my indebtedness to Mr. VAIDYA and to Rev. Fr. ZIMMERMANN, whose advice and ready help have accompanied my labours from the time I first accepted the responsibilities of the work. Nothing has encouraged me more in this arduous and fascinating task than the unwavering interest with which they have followed it. In connection with the help the Editorial Board has received from collaborators outside the Institute, I have to record the indebtedness of the Board to : Pandit VIDHUSHEKHAR BHATTACHARYA, M.A., Principal of the Viśvabhārati, Bolpur ; Rajaguru Pandit HEMRAJ, Director of Public Instruction, Nepal ; M. R. Ry. SAMBAMURTI ROW, Honorary Secretary, Palace Library, Tanjore ; and Prof. K. RAMA PISHAROTI, M.A., Principal of the Sanskrit College, Trippunitura, Cochin. These gentlemen have been good enough to supply the Institute with carefully prepared collations of manuscripts which are in their charge or which were kindly procured for the purpose by them. The Nirnaya Sagar Press has rendered ungrudgingly every assistance in carrying out the typographical arrangements which appeared to me best suited for the purposes of the work. The illustration accompanying this fascicle is prepared from a water-colour painting kindly supplied by the Chief of Aundh.

c4.s4.6

In conclusion, I may be permitted to remark that the renown of the Bhāratavarṣa; of its Princes and its People, is for all time inseparably linked with the Mahābhārata, which is, in more senses than one the greatest epic the world has produced. It must be manifest to anyone who bestows a thought on the subject that the monumental work of preparing the first critical edition of this colossal encyclopaedia of ancient India could be carried on and completed by the young Institute by which it has been undertaken only if it can count upon substantial aid from other sources and upon co-operation on a much wider scale. If the Princes and the People of India were to associate themselves with this imposing enterprise, they would indeed be supporting a *national* work. On behalf of the Institute which I represent, I appeal to all true Indians to ally themselves with the Institute in supporting the publication of a work which is in a unique manner bound up with the history of the Indian people and the prestige of Indian scholarship.

c4.s4.7

*January 1927.* V. S. SUKTHANKAR.

Postscript

c4.s5.1

Since the manuscript of this fascicle was sent to the press, I was able to make arrangements for securing collations of Śāradā and Nepāli manuscripts of the Ādiparvan. These collations will be published later. Here it may just be remarked that the collations so far received wholly support the constituted text, especially as regards the interpolated stanzas, proving the correctness of the method adopted in setting the text.

c4.s5.2

*May 1927.* V. S. S.

Linked texts (2)