Pramāṇa · Search · About · API

Critical Studies in the Mahabharata (V. S. Sukthankar Memorial Edition, Vol. I) c6: Introduction

Critical Studies in the Mahābhārata (V. S. Sukthankar Memorial Edition, Vol. I) c6

Language
English

Contents

Introduction

c6.1

¹ [to Āraṇyaparvan].

The Critical Apparatus

List of Manuscripts Forming the Critical Apparatus

c6.s1.t1.1

The manuscripts utilized for this edition of the Āraṇyaparvan are as follows :

c6.s1.t1.2

I. N(orthern) Recension.

c6.s1.t1.3

(a) *North-western Group* (v).

c6.s1.t1.4

Śāradā (or Kaśmīrī) Version (Ś). Ś₁ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 159 of 1875-76. (The Śāradā Codex.)

c6.s1.t1.5

Devanāgarī transcripts of the Śāradā (or Kaśmīrī) Version (K). K₁ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 471 of Viśrāmbāg I. K₂ = Poona, BORI, Mbh. Collection, No. 15 (= Institute's Collection, No. 246). Dated V. Saṃ. 1828 and Śaka 1693 (ca. A.D. 1772). K₃ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 184 of 1891-95. K₄ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 57 of 1882-83.

c6.s1.t1.6

(b) *Central Group* (γ).

c6.s1.t1.7

Bengali Version (B). B₁ = Dacca, University Library, No. 495. Dated Śaka 1393 (ca. A.D. 1471). B₂ = Santiniketan, Visvabharati Library, No. 781. Dated Śaka 1183 (ca. A.D. 1261). B₃ = Dacca, University Library, No. 601. Dated Śaka 1678 (ca. A.D. 1756). B₄ = Dacca, University Library, No. 728 B. Dated Śaka 1739 (ca. A.D. 1817).

c6.s1.t1.8

Devanāgarī Versions *other than* K (D).

c6.s1.t1.9

Devanāgarī Version of Caturbhujamisra (Dc). Dc₁ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 32 of A 1879-80. Dc₂ = London, India Office Library, No. 3170 (1908). Date A.D. 1765 (?).

c6.s1.t1.10

Devanāgarī Version of Nīlakaṇṭha (Dn). Dn₁ = Indore, Private Property of Sardar M. V. KIBE of Indore. Dated V. Saṃ. 1839 (ca. A.D. 1783). Dn₂ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 236 of 1895-1902. Dn₃ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 496 of Viśrāmbāg I.

c6.s1.t1.11

Devanāgarī Composite Version. D₁ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 269 of Viśrāmbāg II.

c6.s1.t1.12

D₂ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 229 of 1895-1902. Dated V. Saṃ 1795 (ca. A.D. 1739). D₃ = Poona, Bombay Govt. Collection (deposited at the BORI), No. 168 of 1887-91. Dated in year 1677, probably V. Saṃvat (then = A.D. 1620). D₄ = Baroda, Oriental Institute Library, No. 767. Dated V. Saṃ. 1655 (ca. A.D. 1599). D₅ = Pudukottah, State Library MS. (without number). Dated V. Saṃ. 1712 (ca. A.D. 1656). D₆ = Madras, Adyar Library, No. 36 G. 15.

c6.s1.t1.13

II. S(outhern) Recension.

c6.s1.t1.14

Telugu Version (T). T₁ = Lahore, D. A. V. College Library, No. 3908. T₂ = Tanjore, Saraswathi Mahal Library, No. 11809.

c6.s1.t1.15

Grantha Version (G). G₁ = Melkote, Yadugiri Yatiraj Math Library MS. (without number). G₂ = Poona, BORI, Mbh. Collection, No. 53 (= Institute's Collection, No. 266). G₃ = Tanjore, Saraswathi Mahal Library, No. 11839. G₄ = Pudukottah State Library, No. 322.

c6.s1.t1.16

Malayālam Version (M). M₁ = MS. (without number) belonging to Ponnokkoṭṭu Mana Nambudiripad, Alwaye, Travancore. M₂ = Malabar, Poomulli Mana Library, No. 299.

Detailed Account of the Manuscripts

c6.s1.t2.1

§₁

c6.s1.t2.2

(The Śāradā Codex.)

c6.s1.t2.3

Poona, Bombay Government Collection (deposited at the BORI), No. 159 of 1875-76. Total number of folios, for this parvan, 192 (some frayed and patched at ends), with about 24 lines to a page. Size 12" × 19". Clear Śāradā characters. Birchbark (bhūrjapatra).

c6.s1.t2.4

The codex, which is unfortunately incomplete and fragmentary, must have originally contained at least the first three parvans (Ādi, Sabhā, and Araṇya or Āraṇyaka), written, as far as one can judge, by the same hand. The extant portion contains the Sabhā in its entirety, but only fragments of the other two parvans, the beginning of Ādi and the end of Āraṇyaka being lost. The text of Āraṇyaka begins on fol. 211a, line 1; and ends, or rather breaks off, at the end of a folio marked 194b (sic). The numbering of the pages is most erratic. There are in all 192 folios pertaining to this parvan. The fragment breaks off in the middle of our 3. 253. 18ᵈ, ending with the words : वराज्यपूर्णामिव. The margins are mostly clean, and corrections are few and far between. Erasures have been made in a few places with green pigment. In good many places (e.g., folios numbered 156a, 185b, 186a, 193b, 194a, etc.), the scribe has written dots to denote portions of the text which were either lost or illegible to him: sign of a conscientious copyist. On fol. 220a, some stanzas which had been omitted (? hapl.) in the text have been written on the margin (sec. m.), in a different and inferior ink. The top parts of the four folios numbered 89-92 are broken off and lost, involving a loss of from 5 to 8 top lines on each page. The names of sub-parvans have been sporadically mentioned. The manuscript has been fully described at p. x of the *Prolegomena* to Ādiparvan (q. v.)

c6.s1.t2.5

K₁

c6.s1.t2.6

Poona, Bombay Government Collection (deposited at the BORI), Viśrāmbāg I, No. 471, Folios 486 (numbered 55-540; first 54 folios missing), with about 10 lines to a page. Size 12½" × 5¼". Devanāgarī characters. Thin glazed superior Indian paper.

c6.s1.t2.7

First 54 folios are lost; fol. 55 begins with 3. 32, 4ᵃ. Number 237 is given to two consecutive folios, but the second folio thus marked is subsequently corrected (*sec. m.*) to 238. The missing portion of the text at the end of the first of the two folios marked as 237 is written out in small letters (*? sec. m.*) at the bottom of the page. Fol. 485 is written only on one side; and at the end of this folio is the remark :प्रभूतात्रो अग्रे पत्रं लिखितं ॥ विस्मृतं न ॥. the contents of this folio (485) have been copied out again *sec. m.* (on a separate sheet of modern paper with watermarks), including the final remark प्रभूतात्रो ...॥ विस्मृतं न ॥, with the addition, श्रीरस्तु । कल्याणमस्तु ॥ श्रीकृष्णजी सत्य छे ॥ श्री छ, which shows that the copyist of this sheet was some Gujarati scribe. The handwriting of the original, which is Kashmirian in style is not quite uniform : some folios show very thick, black, broad upright characters, while others are appreciably thinner and somewhat slanting. Frequently in the colophons, and the references to speakers, only alternate letters are written, blanks being left to be filled with red ink, which, however, has not been done. Mere folds, without vertical lines, mark the right and left margins of folios. The ink used is jet black and of a superior quality. The sub-parvan names are generally mentioned.

c6.s1.t2.8

K₂

c6.s1.t2.9

Poona, BORI, Mbh. Collection No. 15 (= Institute's Collection, No. 246). Folios 300, with about 14 lines to a page. Size 13¼" × 6⅔". Devanāgarī characters. Dated V. Saṃvat 1828 and Śaka 1693 (ca. A.D. 1772). Grey Indian paper.

c6.s1.t2.10

This manuscript is described as D₁ at p. VII of the Introduction to the Udyogaparvan, for which parvan it was first used. The name of the scribe, which comes after the date (year) at the end of this parvan, appears as : उदैराम लिषायतं मिश्रजी । श्री सदानंदजी; the day and month come last : मिती आषाढ-प्रथमे १५ गुरौ संपूर्णम्. In the Udyogaparvan, the scribe's name is given as Kāśmīriya Sadānanda. Very incorrect, containing numerous little errors of spelling, such as writing श for स. Adhyāya colophons and margins are marked with red pigment, corrections with yellow pigment. Marginal (vertical) lines of pages are in black and red, separated by a thick yellow line. The manuscript is almost without any marks of punctuation. The *daṇḍas* were probably to be marked with red ink, but the [4] revision was never actually carried out. It may be noted that this is a parvan from an almost complete manuscript of the Mahābhārata written by one hand, and as such important.

c6.s1.t2.11

K₃

c6.s1.t2.12

Poona, Bombay Government Collection (deposited at the BORI), No. 184 of 1891-95. Folios 285, with about 17 lines to a page. Size 10½" × 5¼". Devanāgarī characters. Indian paper.

c6.s1.t2.13

Total number of folios is 294, of which fol. 270, 272-279 are missing. Fol. 152a has only six lines of writing, the greater part of the folio being left blank, and the writing continued on the reverse, without any loss of text. In the references to speaker the word उवाच (resp. ऊचुः ) is generally omitted ; the name of the speaker to be generally given in some abbreviated form like युधि, वैशं, मार्क The manuscript seems to be fairly old. The edges, especially of the last folios, are brittle and worn. Characters are short and broad, representing an old style of writing in. Writing is incorrect on the whole. There are occasional corrections in margin (sec. m.). Adhyāya colophons and names of speakers are generally coloured with red pigment. The manuscript mentions sub-parvans as a rule. It does not distinguish between ब and व; both are again often confused with च.

c6.s1.t2.14

K₄

c6.s1.t2.15

Poona, Bombay Government Collection (deposited at the BORI), No. 57 of 1882-83. Folios 370 (numbered 52-421), with about 10 lines to a page. Size 12¾ × 5½". Devanāgarī characters. Old Indian paper.

c6.s1.t2.16

First 51 folios are lost ; fol. 52 begins with 3. 36. 1ᵃ. The hand-writing of the manuscript, though it seems to be of one scribe, is not uniform. For some portion we have large round letters with flourishes, while for others the characters are thin and upright. Pṛṣṭhamātrās are almost invariably used. Marginal notes, especially variants, are written (sec. m.) with the clear remark that it is a pāṭha ; there are explanatory notes also, written on the margins. Colophons, references to speakers, and śloka endings (the latter sporadically) are marked with red pigment.—The manuscript breaks off at the last folio (421), where it ends with the colophon of the last adhyāya of this parvan. There is no phalaśruti, nor the list of contents, which we often find at the end of Mahābhārata manuscripts.

c6.s1.t2.17

B₁

c6.s1.t2.18

Dacca, University Library, No. 495. Folios 1-323, and 325. Bengali characters. Dated Śaka 1393 (ca. A.D. 1471). Paper. Slightly damaged.

c6.s1.t2.19

The manuscript contains two correction slips, one between fol. 163 and 164, and another between fol. 247 and 248.—Collated at the Visvabharati.

c6.s1.t2.20

B₂

c6.s1.t2.21

Santiniketan, Visvabharati Library, No. 781. Folios 264. Size 20" × 4½". Bengali characters. Dated Śaka 1183 (ca. A.D. 1261). Paper—Collated at the Visvabharati.

c6.s1.t2.22

[5] The manuscript is written in one running hand, and has marginal corrections. It is described as "looking very old", and being extremely fragile. It was collated in 1931. With lapse of years its condition is said to have become steadily worse.—*Collated at the Visvabharati.*

c6.s1.t2.23

B₃

c6.s1.t2.24

Dacca, University Library, No. 601. Folios 326 (fol. 141 wrongly numbered as 142 and the mistake continued up to fol. 240; fol. 241 wrongly numbered as 231, and the mistake continued upto the end; fol. 227 duplicated). Bengali characters. Dated Śaka 1678 (ca. A.D. 1756). Paper.—*Collated at the Visvabharati.*

c6.s1.t2.25

B₄

c6.s1.t2.26

Dacca, University Library, No. 728 B. Folios 280 (fol. 185 wrongly numbered as 285, and 214 as 215). Bengali characters. Dated Śaka 1739 (ca. A.D. 1817). Paper.—*Collated at the Visvabharati.*

c6.s1.t2.27

Dc₁

c6.s1.t2.28

Poona, Bombay Government Collection (deposited at the BORI), No. 32 of A 1879-80. Folios 506, with about 11-13 lines to a page. Size 13¾" × 6". Devanāgarī characters. Indian paper.

c6.s1.t2.29

Text with commentary of Caturbhujamiśra entitled *Vākyadīpikā*. The handwriting is not uniform: it begins with a longish upright style, which gradually becomes slanting towards right, up to fol. 164; fol. 165 begins with a broad, short, round, broken style up to folio 265; from the next fol. (266), the style of writing again becomes long, slanting and thin; and towards the end the lettering becomes a little thick and bold. Right and left margins are marked by double lines in red, and stops in the text and the commentary, are also marked by double daṇḍas in red. The lower borders of folios of this manuscript are a little soiled by damp.

c6.s1.t2.30

Dc₂

c6.s1.t2.31

London, India Office Library, No. 3170 (1908). Folios 502, with about 10-14 lines to a page. Size 15½" × 6½". Devanāgarī characters. Date of writing A.D. 1765 (?). Indian paper.

c6.s1.t2.32

Text with commentary of Caturbhujamiśra, entitled *Vākyadīpikā*. 1765 is the date given by EGGELING in the *India Office Catalogue* (1899), though the manuscript itself appears to be undated.

c6.s1.t2.33

Dn₁

c6.s1.t2.34

Manuscript belonging to Sardar M. V. KIBE of Indore, on loan at the Institute. Folios 523, with about 11 lines to a page. Size 16" × 6". Devanāgarī characters. Dated V. Saṃvat 1839 (ca. A.D. 1783). Thick Indian paper.

c6.s1.t2.35

[6] Text with the commentary of Nīlakaṇṭha, called *Bhāratabhāvadīpa*, The date of writing this parvan is noted on the back-cover as: सं० १८३९ प्रारंभ कार्तिक शुद्ध २ समाप्तिः फाल्गुणशुक्ल नवम्यां ९ बुधवारे.—

c6.s1.t2.36

The first part of this manuscript is described as Dn₁, at p. xvi of the *Prolegomena* to the Ādiparvan, and the details of that description apply to this parvan also.

c6.s1.t2.37

Dn₂

c6.s1.t2.38

Poona, Bombay Government Collection (deposited at the BORI), No. 236 of 1895-1902. Folios 768, with about 9 lines to a page. Size 15¼" × 7". Devanāgarī characters. Thick Indian paper.

c6.s1.t2.39

Text with the commentary of Nīlakaṇṭha, called the *Bhāratabhāvadīpa*. The manuscript is written in a uniformly neat upright handwriting. Right and left margins are marked by two double lines in red. Double daṇḍas in red are used on some folios to mark the stops in the text as well as in the commentary. Adhyāya colophons in both the text and the commentary are also in red ink. On some pages, continuations are written on the right side along the marginal red lines.

c6.s1.t2.40

Dn₃

c6.s1.t2.41

Poona, Bombay Government Collection (deposited at the BORI), Viśrāmbāg I, No. 496. Folios 626, with about 12 lines to a page. Size 16" × 6½". Devanāgarī characters. Indian paper.

c6.s1.t2.42

Text with the commentary of Nīlakaṇṭha, called the *Bhāratabhāvadīpa*. The manuscript has its right and left margins marked with close double lines in red as usual. The first three folios and the last folio have an illuminated border with floral designs. The writing, though incorrect, is bold and upright. Numerous corrections are made in the text by using yellow pigment. Fol. 401-439 are written with ink of a faint black colour. Subsequent portion is written in a shaky style, with short and round letters. There are also marginal corrections, entered *sec. m.*, throughout the manuscript. Adhyāya colophons are marked with red pigment. The last colophon in the commentary is left incomplete, and the last but one fol. (625) breaks off with : श्रीगोविंदसूरिम्. The first śloka of the Virāṭa too, is left incomplete, while the last folio is numbered 626 in one (right bottom) corner and 627 in the other (left top) corner. The usual table of contents is also wanting. All this suggests that the actual fol. 626 is lost.

c6.s1.t2.43

D₁

c6.s1.t2.44

Poona, Bombay Government Collection (deposited at the BORI), Viśrāmbāg II, No 269. Folios 225, with about 16 lines to a page. Size 18" × 6½". Devanāgarī characters. Old Indian unglazed paper.

c6.s1.t2.45

A few folios of this manuscript are missing. This manuscript was written by the same scribe who wrote manuscript No. 266 of Viśrāmbāg II, which is described as D₁₄ at pp. xx of the *Prolegomena* to the Ādiparvan. Most folios of this manuscript have a soiled appearance, while the last (225) reveals at its top left corner signs of burning by fire; some folios are partly damaged. The manuscript is carefully written in a perfectly uniform style and the characters are short and round. The name श्रीनृसिंह is written at the lower right margin above the folio number. The corrections in the text are made by deleting the original incorrect reading with yellow pigment and putting in the correct one in black ink. Adhyāya colophons and references to speakers are marked with red pigment.

c6.s1.t2.46

D₂

c6.s1.t2.47

Poona, Bombay Government Collection (deposited at the BORI), No. 229 of 1895-1902. Folios 234, with about 15 lines to a page. Size 14¾" × 6¼". Devanāgarī characters. Dated V. Saṃvat 1795 (ca. A.D. 1739). Old Indian paper.

c6.s1.t2.48

This manuscript, which contains the first three parvans, is the same as manuscript K₀ of the Ādiparvan, described at p. xi of the Prolegomena. The date of the manuscript is recorded at the end of the Āraṇyakaparvan. The manuscript is very corrupt and full of spelling mistakes. It frequently writes ब for व and sometimes च for व; in other words, it confuses च and व. It likewise confuses त and न. The date is given as : संवत् १७९५ वर्षे भातादपद (read भाद्रपद ) वदि १३ छ (?) गुरु ॥

c6.s1.t2.49

D₃

c6.s1.t2.50

Poona, Bombay Government Collection (deposited at the BORI), No. 168 of 1887-91. Folios 240, with about 11 lines to a page. Size 13¾" × 5". Devanāgarī characters. Dated in year 1677, probably V. Saṃvat (then = 1620 A.D.). Old Indian paper.

c6.s1.t2.51

An old and valuable but fragmentary manuscript with many folios missing, some of which have been subsequently replaced, being written sec. m. Many of the folios are badly mutilated, carelessly copied and full of scribe's mistakes. The last preserved folio but one is numbered 372 and breaks off with our 3.278. 10ab; while the last folio, which has its borders worn away and is pasted on to another sheet of paper, has the last three ślokas of the parvan, and the final colophon but no contents or phalaśruti. The date is given as : १६७७ शमए नाम आश्वनि शुदि त्रीदशी १३ वार गुरवाशर. The specification of the era is lost on a missing portion of the folio, but it is probably the Vikrama era (Caitradi). In that case it would, correspond to Thursday, the 28th September 1620 A.D. The edges of some folios are coloured yellow; while for margins there are four close lines in black, on the right and left. Handwriting is uneven and there are many corrections. Single or double daṇḍas are used to mark the stops. There are corrections entered sec. m. on the margins and at some places in the text also by using yellow pigment. Colophons and references to speakers are marked by red pigment in many places. The scribe writes ब for व, and makes little distinction between च and व, as also between प and य.

c6.s1.t2.52

D₄

c6.s1.t2.53

Baroda, Oriental Institute Library, No. 767. Folios 437, with about 11 lines to a page. Size 12" × 4½". Devanāgarī characters. Dated V. Saṃvat 1655 (ca. A.D. 1599). Old Indian paper.

c6.s1.t2.54

The characters are short and round, which later become thin and upright. Margins are marked by three close lines in red; some pages have black lines; while double daṇḍas, which mark the stops in the text, as also adhyāya colophons and references to speakers are marked by red pigment. There are no marginal corrections, and blanks are left in the text where the scribe could not read the exemplar correctly or the exemplar was defective. The following statement containing the date and some other details comes after the last colophon : संवत् १६५५ वर्षे माघ सुदी ७ सोमवासरे बाई श्रीधनागोत गोलवाली । पुस्तकं वन पर्व (sic) दत्तं व्यास दामोदर है दीधो वासुदीको ।। लीषतं जोसी लक्ष्मीदास वास पाटण जंबुदीपे ।। ग्रंथाग्रंथः श्लोकः संध्या सहस्र १२६९५ प्रमाणं ।। Then follows *sec. m.* : || राज्यश्रीभोज्यराज्यप्रवर्तते व्यासश्रीदामोदरपठणार्थे लीषाइतं राजश्री महेराजी तस्यात्मजा उभयकुलआनंददायनि बाई श्रीधनाजी तया इंद पुस्तकं दत्तं ।

c6.s1.t2.55

D₅

c6.s1.t2.56

Pudukottah, State Library Manuscript (without number). Folios 334, with about 10 lines to a page. Size 14¾" × 5¼". Devanāgarī characters. Dated V. Saṃvat 1712 (ca. A.D. 1656). Indian paper.

c6.s1.t2.57

Many of the folios are coloured yellow on both the sides; margins are marked to the right and left, by two black lines with one red line in the middle. Double daṇḍas in red ink are used to mark stops in the text. There are a few marginal corrections, added *sec. m.* Those in the body of the text are made by using yellow pigment. The characters are uniformly longish and upright. The date is given at the end as : || समासेदं आरण्यपर्व संवत् १७१२ ||. No name of the scribe or place of writing is mentioned.

c6.s1.t2.58

D₆

c6.s1.t2.59

Madras, Adyar Library, No. 36 G. 15. Folios 303 (including a *śodhapatra* bearing number 199), with about 12 lines to a page. Size 5½" × 13¾". Devanāgarī characters. Old Indian paper.

c6.s1.t2.60

Style of handwriting is upright and narrow. Double daṇḍas in red ink are used to indicate stops. Margins have, in places, additional stanzas, glosses etc. added *sec. m.* Many pages are coloured yellow; the right and left margins are marked by fine double lines in black, the intervening space being coloured red. The manuscript has a very old appearance, with worn and frayed folios.—A separate additional folio, numbered 199, written (*sec. m.*?) on both sides is inserted as a *śodhapatra* after fol. 199 with the remark :

c6.s1.t2.61

इदं शोधपत्रमेकोनशतद्वयस्थमेकादशप्कावनकपुटे द्रष्टव्यं ।।

c6.s1.t2.62

T₁

c6.s1.t2.63

Lahore, D. A. V. College Library, No. 3908. Folios 144, with about 13 lines to a page. Size 18¼" × 1¾". Telugu characters. Palm-leaf.

c6.s1.t2.64

Incomplete, breaking off at 3. 297. 60ab.

c6.s1.t2.65

T₂

c6.s1.t2.66

Tanjore, Saraswathi Mahal Library, No. 11809. Folios 146 (numbered 248-393), with about 12 lines to a page. Size 29½" × 2¼". Telugu characters. Palm-leaf.

c6.s1.t2.67

[9] This manuscript contains the first five parvans and is the same as T8 of the Ādiparvan. It is written in small clear letters of good style. The manuscript bears no date, but it appears to be old.—*Collated at Tanjore.*

c6.s1.t2.68

G1

c6.s1.t2.69

Melkote, Yadugiri Yatiraj Math Library Manuscript (without number). Folios 130 (numbered 165-294), with about 15-21 lines to a page. Size 18¼" × 1¾". Grantha characters. Palm-leaf.

c6.s1.t2.70

The manuscript contains the first four parvans and a part of the fifth.

c6.s1.t2.71

G2

c6.s1.t2.72

Poona, BORI, Mbh. Collection, No. 53 (= Institute's Collection, No. 266). Folios 216 (two of which, namely 117 and 125, are missing), with about 12-14 lines to a page. Size 15½" × 2½". Grantha characters. Palm-leaf.

c6.s1.t2.73

The manuscript contains parvans three and four. It belongs to the Institute and was purchased a few years ago from a South-Indian manuscript collector, the late Mr. Rangaswamy Aiyangar of the Oriental Library, Mysore. It is in a very good state of preservation, only one folio (149) being broken. The manuscript is undated, but it does not appear to be very old. The margins are clean; but there are a few interlinear corrections.

c6.s1.t2.74

G3

c6.s1.t2.75

Tanjore, Saraswathi Mahal Library, No. 11839. Folios 277 (numbered 75-351), with about 9 lines to a page. Size 21" × 1½". Grantha characters. Palm-leaf.

c6.s1.t2.76

The manuscript, which contains parvans two and three, seems to be old. At fol. 75, the Sabhāparvan ends and the Āraṇyakaparvan begins thereafter immediately. After the end of the Āraṇyaka, about five ślokas from the Virāṭa are written on the last folio (351), which indicates that the manuscript originally contained some more parvans.—*Collated at Tanjore.*

c6.s1.t2.77

G4

c6.s1.t2.78

Pudukottah, State Library, No. 322. Folios 168, with about 10-14 lines to a page. Size 19" × 2½". Grantha characters. Palm-leaf.—The manuscript contains parvans Nos. 3 and 14-18.

c6.s1.t2.79

M1

c6.s1.t2.80

Travancore, Alwaye. From the private Library of Ponnokoṭṭu Mana. Folios 300, with 8 or 9 lines to a page. Size 17½" × 2". Malayalam characters. Palm-leaf.

c6.s1.t2.81

The leaves have been cut regularly and are of uniform size. The manuscript is undated and is probably not very old. It is in a state of good preservation, having a fresh and clean appearance. The margins are clean. No corrections are noticeable.

c6.s1.t2.82

M2

c6.s1.t2.83

Malabar, Poomulli Mana Library, No. 299. Folios 259, with about 9 lines to a page. Size 21½" × 1¾". Malayalam characters. Palm-leaf.

c6.s1.t2.84

[10] The numbering of the folios begins with the second folio, the first being marked śrī and not counted. The manuscript has an oldish appearance. The leaves are uneven and discoloured in places. The edges are considerably worn and uneven; the central holes are enlarged, apparently from constant use; one corner of the last folio (numbered 258) is broken off and part of text is lost. The margins are clean; corrections, which are interlinear, are few and far between.

Manuscripts Not Included in the Critical Apparatus

c6.s1.t3.1

Besides the twenty-eight manuscripts detailed above, the following two manuscripts were also examined by me, which deserve notice.

c6.s1.t3.2

1. Calcutta, Royal Asiatic Society of Bengal Collection, No. 3312 (4065). Folios 166; size 13" × 10". Śāradā characters. Dated Śaka 1630 (ca. A.D. 1708). Kaśmīrī paper.

c6.s1.t3.3

2. Bombay Branch of the Royal Asiatic Society's Collection, No. 966 (= B.D. 245). Folios 235, with about 15 lines to a page; size 14½" × 6½". Old Devanāgarī characters. Dated Saṃvat 1573 (ca. A.D. 1516). Indian paper. Illustrated.

c6.s1.t3.4

The dated Śāradā manuscript belonging to the Royal Asiatic Society of Bengal is a very fragmentary and fragile manuscript, in an extremely delapidated condition. The custodians of the manuscript would not part with it for collation at the Institute, but allowed it to be collated and photographed on the premises of the Society. The difficulty in the way of collating—as of photographing—it, was that many of the folios were stuck together and could be separated only at the risk of damaging the writing. A partial study of the manuscript showed that the basic text is probably Kaśmīrī, but it is contaminated deeply from the Mid-Indian versions (Bengali-Devanāgarī). The manuscript is briefly described in Haraprasada SHASTRI'S Descriptive Catalogue, Vol. 5, at p. 106. A facsimile of a page from this manuscript, containing the text from 3. 242. 20b to 243. 15a, is given at the end of Volume 4 of this edition. Note the bits of commentary (टीका), wrongly incorporated, in lines 9-10, and in lines 12-13.

c6.s1.t3.5

The other manuscript, which belongs to the Bombay Branch of the Royal Asiatic Society, is, on the other hand, an extremely valuable old manuscript, dated in V. Saṃvat 1573 (= A.D. 1516). It would have been fully collated and used for our edition, but for the circumstance that it is also very fragmentary, nearly one-third of the number of folios being, unfortunately, irrecoverably lost. The original manuscript was written at Yoginīpura (modern Delhi) during the reign of "Sulitrāṇ Sikandara", who appears to be the famous Sikandar Shah Lodi, Sultan of Delhi, for Bhānudāsa Chaudharī—a resident of Kacchauvā (a "water-fortress", jaladurga).¹ The manuscript was constantly [11] consulted by me in case of doubt to see what help it could give. I have cited its evidence occasionally, in the "Critical Notes" at the end of this Book, on important matters. I append here in full the somewhat lengthy post-colophon statement, which is not without interest. The copyist was probably a Jaina. Here follows the statement:

c6.s1.t3.6

¹ This Kacchauvā may perhaps be identified with the modern "Kachola", a town in Udaipur State. The Imperial Gazetteer (1886) states that in former times the town, which stood on the western bank of a large lake, must have been a place of some importance; for all around, the ground is strewn with fragments of extremely fine sculpture, and half way up the hill the ruins of a temple are visible.

c6.s1.t3.7

शिवमस्तु सर्वजगतः परहितनिरता भवंतु भूतगणाः। दोषाः प्रयांतु नाशं सर्वत्र सुखी भवतु लोकः॥॥ संवत् १५७३ वर्षे॥ श्रावण वदि ९ सोमे॥॥ तत्समये योगिनीपुरे। सुलित्राण-सिकंदरविजयराज्य प्रवर्त्तमाणे॥॥ तत्स[म]ये कच्छौवाजलगुर्दगे॥॥ चंद्रापुरीनिवासिनः तत्र निवसः सन्। चौधरी विन। तस्य सुपुत्र चौधरी कन्हा। तस्य पुत्र वैष्णव (1) चौधरी श्रीभानदास। इदं महाभारत आरण्यपर्व आत्मश्रवणार्थ लिखापितं। लिखितं कायस्थगौडान्वय। (sign) श्रीलखणसीसुत (sign)। भवानीदास॥॥ शुभं भवत्॥॥ तैलाद्रक्षे जलाद्रक्षेद्रक्षेत् शिथिलवंधनं। परहस्तगतं रक्षे एवं वदति पुस्तकः॥॥ संक्रोडिवि करचरणं। विमुक्कवयणाइ निम्मला (read णिच्चला) दिदी। तं होउ तुम्हे रिउणा (read ण्णो)। जं सुक्खं पावण लेहो॥॥॥॥

c6.s1.t3.8

The manuscript is briefly described in H. D. VELANKAR’S *Descriptive Catalogue* of the Society’s manuscripts, Vol. 2, p. 292.

Testimonia

c6.s2.1

The testimonia of the Āraṇyaka consist of ancient commentaries and epitomes, of which there are not very many in the case of this parvan.

c6.s2.2

Devabodha’s commentary on the Āraṇyaka has unfortunately been lost, or at least has not been recovered so far. The same is true of the Javanese version. This parvan thus unfortunately lacks two of the oldest—and the most important—of the testimonia of the Mahābhārata,—works which have rendered yeoman service in some of the parvans already edited.

c6.s2.3

As regards other works of that nature, there is, as far as I know, extant at least one complete manuscript of the commentary of Arjunamiśra on the Āraṇyaka. It belongs to the Government Sanskrit College Collection in Calcutta (No. 310); cf. *Catalogue*, Vol. 4, p. 190. There are also incomplete copies of the commentary in the Sanskrit College at Benares (MSS. No. 2279; and 7 16 34). I had the use of a rough copy of the Calcutta Sanskrit College manuscript mentioned above, which was kindly supplied to me, at my request, by the Principal of the College. The copy was however found to be full of clerical errors and was not exactly suitable for collation. It was therefore used by me merely for occasional consultation. The text used by Arjunamiśra is, as usual, of a superior type, and his commentary would repay careful study, if good manuscripts of the commentary could be obtained. A facsimile of the single extant page of a paper manuscript of what proved to be Arjunamiśra’s commentary on the Āraṇyakaparvan, written in Śāradā characters and included by chance in the Śāradā manuscript of this parvan belonging to the Royal Asiatic Society of Bengal, forms the frontispiece of the second part of the Āraṇyakaparvan (Vol. 4). The passage commented on is 3. 133. 20 to 134. 9. Note the consecutive numbering of the lines in the left margin.

c6.s2.4

Another useful testimonium of the Mahābhārata is the *Bhāratamañjarī* of Kṣemendra. Kṣemendra summarizes the Āraṇyaka in about 2000—to be precise 1964—stanzas, divided into the following 51 sections :

c6.s2.5

[12] ( 1 ) सूर्यस्तव ; ( 2 ) विदुरागमन ; ( 3 ) सुरभ्युपाख्यान ; ( 4 ) मैत्रेयशाप ; ( 5 ) किन्दीरवध¹; ( 6 ) साल्ववध ; ( 7 ) कैरात ; ( 8 ) लोकपालास्त्रप्राप्ति ; ( 9 ) इन्द्रलोकगमन ; ( 10 ) नलोपाख्यान ; ( 11 ) पुलस्त्यतीर्थयात्रा ; ( 12 ) धौम्यतीर्थयात्रा ; ( 13 ) अगस्त्य ; ( 14 ) ऋष्यशृङ्गोपाख्यान ; ( 15 ) कार्तवीर्योपाख्यान ; ( 16 ) सौकन्य ; ( 17 ) मान्धातृपाख्यान ; ( 18 ) जन्तूपाख्यान ; ( 19 ) श्येनकपोतीय ; ( 20 ) अष्टावक्रीय ; ( 21 ) यावक्रीत ; ( 22 ) हनुमद्भीमसमागम ; ( 23 ) पुष्कराहरण ; ( 24 ) जटासुरवध ; ( 25 ) मणिमद्वध ; ( 26 ) वैश्रवणसमागम ; ( 27 ) निवातकवचवध ; ( 28 ) लोमशतीर्थयात्रा ; ( 29 ) आजगर ; ( 30 ) ब्राह्मणमाहात्म्य ; ( 31 ) वैन्योपाख्यान ; ( 32 ) सरस्वतीताक्ष्र्यसंवाद ; ( 33 ) मत्स्योपाख्यान ; ( 34 ) मार्कण्डेयसमस्या (sic); ( 35 ) भविष्यत् ; ( 36 ) मण्डूकोपाख्यान ; ( 37 ) इन्द्रद्युम्नोपाख्यान ; ( 38 ) धुन्धुमारोपाख्यान ; ( 39 ) पतिव्रतोपाख्यान ; ( 40 ) आङ्गिरस ; ( 41 ) द्रौपदीसत्यभामासंवाद ; ( 42 ) घोषयात्रा ; ( 43 ) दुर्योधनप्रायोपवेश ; ( 44 ) धार्तराष्ट्रयज्ञ ; ( 45 ) मृगस्वप्नप्रदर्शन ; ( 46 ) व्रीहीद्रौणिक ; ( 47 ) द्रौपदीप्रमाथ ; ( 48 ) रामोपाख्यान ; ( 49 ) सावित्र्युपाख्यान ; ( 50 ) कुण्डलाहरण ; ( 51 ) आरण्य.

c6.s2.6

¹ It is worth noting that our Śāradā Codex *uniformly* writes किन्दीर for our किर्मीर.

c6.s2.7

From this list of contents, it can be seen that almost all the stories and episodes which we now find in our text of the Āraṇyaka were there already in Kṣemendra’s time (ca. 1050 A.D.). The parvan must therefore have had then the same general form and appearance as now. More significant is the fact that all important passages—without exception—that are lacking in our manuscripts of the Kaśmīrī version are likewise missing in the *Mañjarī*. This fact alone cannot prove that these passages were lacking in the Kaśmīrī version of the eleventh century. Some or even all of them could have been omitted by a poet who was making an epitome of an extensive work like the Āraṇyakaparvan : he would be within his rights in doing so. But the *Mañjarī* does not omit any important episode or passage which is not omitted at the same time in the Kaśmīrī version of the Great Epic. Consequently, unless the version of the Āraṇyakaparvan used by Kṣemendra was nearly identical with *our* Kaśmīrī version, the coincidence that our Kaśmīrī manuscripts omit just those passages which Kṣemendra had considered not worth including in his epitome, and *no other*, would be extremely curious, if not miraculous. It seems accordingly most probable that the episodes which are not found in Kṣemendra’s *Mañjarī* were already lacking in the Kaśmīrī version of the Mahābhārata *in the eleventh century*. I think, therefore, that with the help of this testimonium, we can establish for the omissions of the Kaśmīrī version an antiquity much higher than that vouchsafed solely by the age of our manuscripts, which are later by some centuries than the period of the composition of Kṣemendra’s *Bhāratamañjarī*.

c6.s2.8

Noteworthy among such omissions of the *Mañjarī* are the following three lengthy passages : (1) the Temptation of Arjuna by Urvaśī (App. I, No. 6); (2) Karṇa's Conquest of the World (App. I, No. 24); (3) the Visit of Durvāsas to Yudhiṣṭhira (App. I, No. 25). The first of these three interpolations has insinuated itself in the manuscripts of almost all versions and is now lacking only in some manuscripts of the Kaśmīrī version.¹ The second among them is documented only by Devanāgarī manuscripts, excepting a few conflated manuscripts of other versions ; consequently its claim to be recognized as an authentic passage is not very substantial. Finally, the last passage introducing Durvāsas into the story, a passage which is really restricted to the Nīlakaṇṭha version, with the casual support of a few Devanāgarī and some conflated manuscripts, does not really come into question ; because its spurious character is very [13] evident and does not need an elobate proof.²

c6.s2.9

¹ See below, p. 160.

c6.s2.10

² With this story disappears one of the very few episodes in the Mahābhārata in which Śrī Kṛṣṇa is represented as hearing from a distance, as it were by clairaudience or divine omniscience, the prayers of his distressed devotees and as either coming instantly to their help in person or providing invisibly the means of their rescue or safety. The other episode I had in mind, which has likewise proved to be an interpolation, is the well-known scene, in the Sabhāparvan of the disrobing of Draupadī (Draupadī-vastra-haraṇa), when, according to the Vulgate version, Draupadī prays to Śrī Kṛṣṇa, who hears her prayer and comes running to her rescue. (त्यक्त्वा शय्यासनं पभ्यां कृपालुः कृपयाभ्यगात् । B. 2. 68. 45), and she is clothed again and again, miraculously, but presumably by the intervention and grace of Śrī Kṛṣṇa. They undoubtedly represent a later phase of Kṛṣṇa worship.

c6.s2.11

The versions of the commentators Caturbhuja and Nīlakaṇṭha are described below under the Devanāgarī version.

Pedigree of the Āraṇyakaparvan Versions

c6.s3.1

[Figure: Vyāsa's Bhārata Ur-Mahābhārata N S ν γ σ Śāradā K Bengali (Non-K) Telugu Grantha Malayalam (Ś) (B) Devanāgarī (T) (G) (M) (D)]

A Critical Survey of the Recensions and Their Versions

General Survey

c6.s4.t1.1

A description of the forest life of the Pāṇḍavas constitutes the topic of the Third Book, the Āraṇyaka—(Āraṇya-or Araṇya-), parvan, commonly called (for short) the Vanaparvan, whose extent is between 11,000 and 12,000 ślokas (i.e. "granthas"). The actual incidents narrated, pertaining to the life-history of the heroes in this parvan, are indeed few and unimportant compared to the size of the book. The entire narrative has been condensed by Kṣemendra in his *Bhāratamañjarī* into less than 2,000 stanzas, without omitting any of the main incidents. The great extent of the parvan has been made up by the inclusion of episodes of the most diverse kind: stories of ancient kings, sages, and virtuous women; description of places of pilgrimage (*tīrtha-yātrā*); discourses [14] on moral, ethical and philosophical topics, ancient gāthās and anuvaṃśa stanzas. It was a fashion for some time to obelize this episodic matter as late interpolation. But that is a misguided view, originating with a certain type of critics who have exhibited a uniform lack of understanding of the meaning of the Mahābhārata and of the basic plan and aim of the creators of our great epic. The episodical material in the Mahābhārata is, in general, not secondarily introduced; it belongs to the original plan and serves a distinct purpose. As PISANI has pointed out in his paper on the "Rise of the Mahābhārata", the bulk of didactic and episodic matter has been used to fill up the great "temporal hiatuses" in the narrative, viz, in the first place, the twelve years of exile in forest (Āraṇyaka), and then the long interval between the end of the Bhārata War and the last adventure of the Pāṇḍavas (Śānti and Anuśāsana). "This distribution", remarks PISANI, "to fill up temporal hiatuses has not only the scope of not disturbing the course of narration, but also that of helping the reader to pass over irrelevant years without striking against too strong a contrast between periods minutely narrated and others rapidly surpassed. In a not different manner Homer introduces often dialogues and episodic stories when he must conceal the flowing of times without noteworthy events."¹

c6.s4.t1.2

¹ A Volume of Eastern and Indian Studies presented to Prof. F. W. Thomas, etc. (1939), p. 170.

c6.s4.t1.3

The episodic material is largely Purāṇic in character. Many of the Purāṇic stories we find here narrated over again. Moreover, as the references given in the foot-notes to the text and the critical notes at the end of this Book (pp. 1091-1109) will show, there is considerable verbal agreement between the epic and Purāṇic versions of the stories. The Purāṇas which show the closest contact and widest parallelism are the Skanda, Padma and Brahma. The entire story of Skanda (adhy. 213-221), together with the passage containing the 108 names of the Sun (our adhy. 3, stanzas 18-28), recurs almost verbatim in the Skanda Purāṇa. The story of the Bhārgava sage Cyavana and the princess Sukanyā, the prose story of the Frog-girl, and a considerable portion of the Rāmopākhyāna are also to be met with in the Skanda, worded almost identically with the Mahābhārata. The 108 names of the Sun recur besides in the Brahma Purāṇa, which, moreover, has a portion of the story of Mārkaṇḍeya in common with the Āraṇyakaparvan. Furthermore, sections of our tīrthayātrā are to be met with again in the Padma and Matsya. A Bengali manuscript¹ of the Padma Purāṇa contains besides a version of the Ṛśyaśṛṅga legend. In all these cases, the parallelism of verbal expression of the epic and the Purāṇic narratives is so complete and striking, that the possibility of their having originated independently of each other is at once ruled out. Many stray stanzas from our parvan are found in the Kūrma, Varāha, Vāyu, Viṣṇu, Viṣṇudharmottara, and in the Harivaṃśa.

c6.s4.t1.4

¹ The Bodleian MS., Wilson 111-116, mentioned by LÜDERS in “Die Sage von Ṛśyaśṛṅga.” *Nachrichten von der Königl. Gesell. der Wiss. zu Gottingen, Phil. hist. Kl.*, 1897, pp. 88, 94.

c6.s4.t1.5

Our recensions of the Purāṇas are admittedly of a very late date, and it is not to be expected that the Mahābhārata would have borrowed any of its material from our Purāṇas. The claim that the the Mahābhārata owed certain geographical and cosmographical sections of the Bhīṣmaparvan to the Padma Purāṇa, preferred by Luise HILGENBERG,² [15] has been answered and liquidated by Rao Bahadur Professor Dr. S. K. BELVALKAR, who has shown that in reality the converse of this proposition is true.³ This scholar has now shown from internal evidence that the Padma Purāṇa must have written up its geographical section from the information supplied by the Mahābhārata. Similarly, the question of relationship between our Pulastya Tīrthayātrā and a very similar episode in the Padma Purāṇa has been discussed by my assistant, Mr. M. V. VAIDYA, M.A., who has likewise shown from intrinsic evidence that the Mahābhārata passage is the source of the Padma Purāṇa version.⁴

c6.s4.t1.6

² *Die Kosmographische Episode im Mahābhārata und Padmapurāṇa* (= Bonner Orientalistische Studien, Heft 4), Bonn 1934.

c6.s4.t1.7

³ S. K. BELVALKAR, “The Cosmographical Episode in Mahābhārata and Padmapurāṇa”, *A Volume of Eastern and Indian Studies presented to Prof. F. W. Thomas* etc. (1939), pp. 19 ff.

c6.s4.t1.8

⁴ M. V. VAIDYA, “Tīrtha-yātrā in the Āraṇyakaparvan and the Padmapurāṇa”, *A Volume of Studies in Indology presented to Prof. P. V. Kane* etc. (1941), pp 530 ff.

c6.s4.t1.9

It must, however, be admitted that although in most cases the compilers of our Purāṇas appear to have drawn their material from the Mahābhārata, there may be—indeed there must be—a few cases in which both the Mahābhārata and the Purāṇas may have drawn independently upon a third common source. We can also say this with regard to episodes like the Sāvitrī episode, which likewise occurs in the Matsya Purāṇa, where it is narrated in an entirely different manner from that in the epic, and where all traces of mutual relationship are absent or obliterated, except for two or three common stanzas and stray pādas.

c6.s4.t1.10

The reference in our parvan to Vāyu, वायुप्रोक्तमनुत्तम्य (3, 189. 14), is worth considering in this connection. Nothing corresponding to the general contents of the passage where it occurs is to be found in the extant Vāyu Purāṇa, as is remarked by HOPKINS,¹ or for the matter of that in any other Purāṇa proclaimed by Vāyu. There are, however, a few stanzas in the Vāyu describing the dissolution of the world where verbal similarity with some stanzas of this passage is pronounced, even though they are introduced in the Vāyu in a different context. The Mahābhārata, as mentioned in it (3. 189. 14), draws upon a Purāṇa of Vāyu—and indeed, the topic narrated belongs properly to a Purāṇa in its right, a Purāṇa which is older than the extant Purāṇas and which must be presumed to be now irretrievably lost. The evidence for establishing a relationship between the extant Vāyu and its old namesake to which the epic is indebted is lacking, barring these few stray stanzas.

c6.s4.t1.11

¹ The Great Epic of India, p. 48 f.

c6.s4.t1.12

Our parvan also contains an epitome of the legend of Rāma, son of Daśaratha, the hero of the Rāmāyaṇa, which is known as Rāmopākhyāna (adhy. 258-275); but in the Parvasaṃgraha this passage is called रामायणमुपाख्यानं (1. 2. 126). Does this name imply any connection with our "Rāmāyaṇa"? JACOBI has tried to show that the Rāmopākhyāna is indeed an epitome of the work commonly known as Vālmīki's Rāmāyaṇa, a conclusion which has been confirmed by subsequent researchers.²

c6.s4.t1.13

² Cf. V. S. SUKTHANKAR, "Epic Studies VIII", in A Volume of Studies in Indology presented to Prof. P. V. Kane etc. (1941), pp. 472 ff.

c6.s4.t1.14

This book, as already observed, is particularly rich in legends of ancient India. Among them are three worthy of special mention, viz., the Tale of Nala, the Ṛśyaśṛṅga Legend and the Story of Sāvitrī which have been critically edited here for the first time. The first of these, the Tale of Nala, has by way of translations into English, [16] French, German, Dutch, Italian and other European languages, passed into the great stream of world literature. The second, the Ṛśyaśṛṅga Legend, though not so well known, has yet a long and interesting history and shows many modulations and ramifications, as has been convincingly shown by Professor Heinrich LÜDERS in his well-known monograph on the subject, "Die Sage von Ṛśyaśṛṅga."³ The story of Sāvitrī portraying woman in the rôle of the saviour of man, which represents the high-water mark of epic poetry, is in a class by itself, and deserves to be more widely known. It is to be hoped that this first critical edition of the three of the most famous episodes of the Mahābhārata will evoke the interest of Indologists and encourage further study of them.

c6.s4.t1.15

³ Nachrichten von der Königl, Gessell, der Wiss. zu Göttingen, Phil-hist. Kl., 1897, pp. 87-135; see also ibid. 1901, pp. 28-56.

The Two Recensions

c6.s4.t2.1

The text of the Āraṇyakaparvan is, relatively speaking, remarkably smooth, offering no special difficulties for textual reconstruction. There are of course, as usual, discrepancies between the Northern and the Southern recensions, and between the various versions comprising these two main recensions. But it is noteworthy that in the whole of this big parvan, comprising between 11,000 and 12,000 “ślokas”, there are not many transpositions of any consequence and there are also remarkably few “substitute passages”. By the latter I mean parallel versions of fairly long passages in two or more conflicting recensions which agree in general sense but are worded differently in different groups of manuscripts, so that they cannot be co-ordinated line by line,—passages which we find in large numbers in some of the other parvans, notably in the Virāṭaparvan. These substitute passages are very clear evidences of textual corruption—or, shall we say, derangement?—and indeed very embarrassing to the textual critic. As for the transpositions, there is only one that is worth noting here, and that is of the last adhyāya of this parvan. The whole of this adhyāya is bodily transferred in the Southern recension to the next parvan and appears there incorporated in the first adhyāya of that parvan! It is not possible to say with certainty whether this chapter, which in all probability is itself an interpolation, belongs legitimately to our parvan or to the next; but it appears to be distinctly more appropriate in its Northern setting, as its intrusion in the Virāṭaparvan seems to lead to some obvious repetitions as well as disturbance of context.¹

c6.s4.t2.2

¹ See H. LÜDERS, “Über die Grantharecension des Mahābhārata”, Abhandlungen der Königl. Gesell. der Wiss. zu Göttingen (1901), p. 53 f.

c6.s4.t2.3

There are naturally quite a number of large and small insertions in the various versions. But in this matter also the present parvan prepares a surprise for us. From the experience gained from our critical editions of the Ādi, Sabhā, and Virāṭa parvans especially, we are inclined to regard the Southern recension as a highly inflated version of the text. Now, in this parvan—strange to say—the Vulgate version, which is based on the late Northern tradition, is actually considerably longer than the Southern, as may be seen from the extent of this parvan according to the existing printed editions of the Vulgate and the Southern recension: the Calcutta edition has 12,848 ślokas (corrected figure) as against 11,138 ślokas in SASTRI'S Southern Recension.

c6.s4.t2.4

The large excess in the Calcutta-Bombay editions compared with SASTRI'S is explained by the fact that the Vulgate text contains about a dozen adhyāyas which are lacking entirely in the Southern recension; whereas the Southern recension contains only two passages of any considerable length (App. I, Nos. 4 and 23) which are not found in representative manuscripts of the Northern recension, they being peculiar to the Southern,—with the result that the Calcutta edition has 1710 ślokas in excess of SASTRI'S edition of the Āraṇyaka-parvan. Thus we see that though, as a whole, the Southern recension is unquestionably inflated, yet the process of inflation, as is evident from this parvan, is not uniform throughout the epic,—which is a very important and significant fact.

c6.s4.t2.5

Though the best version of the Northern recension—namely, the Śāradā-K version—and the Southern recension agree remarkably well as regards the general extent of the parvan, there are nevertheless numerous specific individual readings, characterizing the two recensions—Northern and Southern--and distinguishing them sharply from each other. Here is a selection of readings peculiar to the two main recensions, the references being to adhyāyas and ślokas.

c6.s4.t2.6

| Northern Recension | Southern Recension |
| 1. 37 घोरं | सर्वाः |
| 1. 41 शुभे | शुश्रूषु |
| 2. 13 थ्रिक्पापान् | द्विषतो ( text ) |
| 2. 31 न लिप्सेत | स्वपक्षेभ्यः ( text ) |
| 2. 34 स्मृता | नृणाम् ( text ) |
| 2. 45 ऐश्वर्य | आरोग्यं |
| 2. 50 ब्रह्मन | विप्र |
| 2. 74 °सम्बन्धात् | °योगाच्च |
| 5. 8 एवं शेषं यदि | एतच्छ्रेयस्तव |
| 8. 2 वचनं राजा | मधुरं वाक्यं |
| 12. 16 न्यमीलयत लोचने | विषादमगमद्द्रशं |
| 12. 59 गृह्य मध्ये वृकोदर | भीमसेनो महाबलः |
| 13. 13 अप(or अव )कृष्णेत्तरासङ्गः | कृष्णाजिनोत्तरासङ्गः |
| 13. 28 महिषीं | रुक्मिणी |
| 13. 50 दश | तथा |
| 13. 78 पुनः सुसान् | ततः पुनर् |
| 13. 83 राजानं दक्षिणेन च | दक्षिणेन युधिष्ठिरं |
| 48. 4 सिंहविक्रान्तौ | युधि विक्रान्तौ |
| 101. 11 परमेष्ठिनमाज्ञाप्य | समनुज्ञाप्य ब्रह्माणं |
| अगस्त्यस्याश्रमं ययुः । | वाहर्णि समुपस्थिताः । |

c6.s4.t2.7

I may add here examples of short passages found in the Southern Recension which are lacking in the Northern, and thus distinguish it from the latter : 6*, 17*, 18*, 100*, 118*, 176*, 197*, 201*, 207*, 208*, 209*, 217*, 236*, 237*, 239*, 241*, 246*, 273*, 289*, 304*, 306*, 314*, 315*, 331*, 341*, 347*, 351*, 367*, 383*, 385*, 386*, 407*, 425*, 471*, 474*, 475*, 481*, 489*, 491*, 496*, 497*, 500*, 503*, 506*, 507*, 509*, 513*, 515*, 518*, 519*, 520*, 523*, 524*, 525*, 526*, 527*, 533*, 534*, 573*, 594*, 598*, 606*, 612*, 615*, 626*, 634*, 645*, 650*, 653*, 709*, 721*, 739*, 740*, 742*, 745*, 751*, 756*, 759*, [18] 781*, 782*, 807*, 808*, 816*, 854*, 935*, 941*, 950*, 954*, 975*, 988*, 995*, 997*, 1002*, 1006*, 1030*, 1092*, 1093*, 1102*, 1114*, 1204*, 1206*, etc. etc., etc.

c6.s4.t2.8

It may be observed that all these additions in the Southern recension (with the exception of only one or two), numerous as they are, are uniformly short, each usually consisting of not more than a couple of lines.

c6.s4.t2.9

As remarked already, the text of this parvan is remarkably smooth, not presenting any unusual difficulties in the way of restitution. When the Śāradā-K version (which is the best Northern version) and the Southern recension are placed *vis-à-vis*, we can in general reconstruct the original with confidence, barring a certain number of minor verbal fluctuations in the shape of synonymous phrasings, which remain indeterminate without affecting the construction or obscuring the sense. The concord between the Śāradā-K version and the Southern recension in point of general content is striking, and forms a sure basis for constituting a single text. Contamination between the Kaśmīrī version and the Southern recension cannot be proved, but contamination between the Bengali-Devanāgarī version and the Southern recension is not impossible. The agreements between the Bengali-Devanāgarī and the Southern recension have nevertheless been as a rule utilized to arrive at a tentative stop-gap, based on the indications of documental evidence. But it should be noted that the Kaśmīrī-Southern agreements have for greater documental authority and probative value than the Bengali-cum-Devanāgarī and Southern agreements.

c6.s4.t2.10

Let me put the matter in a slightly different way. The highest documental probability we can demand and expect is when all manuscripts of our critical apparatus—which is the same as saying, all our different versions—agree on a reading or a feature. We must accept this as the original; at least we do not want to question it, at present. In the absence of such complete concord, the next best combination is the agreement between the Śāradā version and the Southern recension (against Bengali-Devanāgarī). Third in importance is, in my opinion, the concord between the Southern recension and the Bengali-cum-Devanāgarī version (against the Śāradā). Fourth in order stands the agreement between only Northern versions or only Southern versions *inter se*, which I consider, in general, as of equal value. With the proviso that a passage or a stanza or even a line, which is not *necessary* to the context, may be rejected, if it is actually omitted entirely in even *one* of the important versions, since—as experience has shown—the chances of conflation are always very much greater than those of accidental or intentional omission.

First 146 of 310 passages. Read on

Linked texts (14)