Olivelle intro

Introduction to the Critical Edition

This critical edition of Yādava Prakāśa’s Yatidharmasamuccaya is based on sixteen manuscripts and two printed versions. They are described here according to the sigla ascribed to them in the critical apparatus.

Description of the Manuscripts

A1 Adyar Library and Research Center, Madras. Ms. no. 73570. A Catalogue of the Sanskrit Manuscripts in the Adyar Library (Madras, 1926), part I, p. 115. Palm leaf. Grantha script. 54 folios. Approximately 30.5 x 3.5 cms (lengths vary; folio 8 is 22.5 x 3.5 cms). Between 11 and 9 lines per page. Approximately 57 akṣaras per line. Complete. No date. A photocopy was used. Somewhat carelessly written with frequent corrections. Folios 47-48 are partly written in a different hand. A reader appears to have gone through the entire ms and made numerous marginal corrections. This was the original of Schrader’s copy (Schr. 45) described below.

A2 Adyar Library and Research Center, Madras. Ms. no. 73486. A Catalogue of the Sanskrit Manuscripts in the Adyar Library (Madras, 1926), part I, p. 115. Palm leaf. Grantha script. 47 folios. Approximately 39 x 4 cms Between 10 and 11 lines per page. Between 55 and 71 akṣaras per line. Complete. The date in the colophon is given according to the Jupiter cycle, and without additional data it cannot to be translated into a unique date of the common era. Excellent condition. A photocopy was used. Two additional folios written in a different hand and containing astrological material are inserted at the end of the ms. Carefully written with few corrections. This was the original of the variants recorded in Schrader’s copy (Schr. 45). Colophon:

श्रीनारायणगुरवे नमः ।
श्रीमते रामानुजाय नमः ।
श्रियै नमः ।
श्रीमते नारायणाय नमः ।
श्रीगोदायै नमः ।
करकृतम् अपराधं क्षन्तुमर्हन्ति सन्तः।
हेविलम्बि शरत्-पुष्य-मासि प्रथमा-तिथौ ।
मयेदं लिखितं सम्यग्
यतिधर्मसमुच्चयम् ।
यद् अक्षर-पद-भ्रष्टं
यत्-पाद-प्रच्युतं यदा ।
तत् सर्वं क्षम्यतां यूयं
कृपा-पूर-वशं-वदाः ॥

A3 Adyar Library and Research Center, Madras. Ms. no. 71573. A Catalogue of the Sanskrit Manuscripts in the Adyar Library (Madras, 1926), part I, p. 115. Palm leaf. Grantha script. 91 folios (many are in bits and pieces). 38 x 3 cms (difficult to measure because most edges are broken). Between 7 and 8 lines per page. The akṣaras per line cannot be accurately estimated because ends of folios are damaged. No date. Incomplete. A photocopy was used. Extremely poor condition. A significant section of almost every page cannot be read because of broken sections of palmleaf, and after Ch. 7 the ms becomes almost unreadable. The ms begins with Ch. 2. After the conclusion of Ch. 7 the ms inserts a large amount of extraneous material. This addition begins with the passage from Parāśara (Ch. 5.11). Most of this material, however, cannot be identified due to the poor condition of the ms A3 thus could be used only for Chapters 2–7 of the edition.

A4 Adyar Library and Research Center, Madras. Ms. no. VB 157. Descriptive Catalogue of Sanskrit Manuscripts. Vol. 13: Viśvabhāratī Collection—I. Compiled by Dr. E. R. Sreekrishna Sarma (Madras, 1976), p. 150, ser. no. 813E. Palm leaf. Telugu script. 31 folios. 42 x 3.5 cms 7 lines per page. 64 akṣaras per line. Complete. A photocopy was used. Good condition. Carefully written. According to Dr. David Pingree, the date of the Jupiter cycle given in the colophon corresponds plausibly with Wednesday, July 6, 1474. This is a rather early date for a South Indian palm leaf ms. Often, however, mss were copied together with their original dates. Thus, the date in a colophon is not necessarily an indication of the date of the ms in question. Colophon:

श्रीकृष्णाय नमः । श्रीनिवासब्रह्मणे नमः । श्रीमन्निलाद्रिपतये जगन्नाथाय नमः । जयवर्षे नभोमासि कृष्णपक्षेऽष्टमीतिथौ । बुधवारेऽग्निनक्षत्रे योगे व्याघातवर्जिते । श्रीनिवासमुदे तेन प्रेरितेन यथामति । लिखितो वरदार्येण यतिधर्मसमुच्चयः । कृत्वानवद्यं निगमान्तभाष्यं निराकृतं येन भयं श्रुतीनाम् । प्रतारितानामबहुश्रुतैस्तं रामानुजं योगिनमाश्रयामः ॥

A5 Adyar Library and Research Center, Madras. Ms. no. PM 726. Paper. Telugu script. 31 folios. 34 x 21 cms Between 21 and 25 lines per page. 45 akṣaras per line. Complete. No date. A photocopy was used. Written in several hands with frequent errors and corrections. Some of the folios are ruled and the ms appears to be rather recent. This was probably the Telugu original of Schrader’s copy of Chapters 1-6 (Schr. 46).

A6 Adyar Library and Research Center, Madras. Ms. no. VB 471. Descriptive Catalogue of Sanskrit Manuscripts. Vol. 13: Viśvabhāratī Collection—I. Compiled by Dr. E. R. Sreekrishna Sarma (Madras, 1976), p. 150, ser. no. 814. Palm leaf. Telugu script. 64 folios. 39.5 x 3.5 cms 8 lines per page. 58 akṣaras per line. Complete. A photocopy was used. Old and moth-eaten. This is a composite manuscript; it switches to the long recension (Y*) at Ch. 8.30 and back to the short recension at Ch. 10.128.

G1 Government Oriental Manuscripts Library, Madras. Ms. no. D 16207. Palm leaf. Grantha script. 20.5 x 4 cms 41 folios. Between 8 and 15 lines on a page. 30 akṣaras per line. A microfilm was used. The ms, as it now exists, is incomplete and ends at Ch. 8.7. The copy of it made by Schrader (Schr. 47), however, contains the whole text. The ms must have been seriously damaged sometime after Schrader’s copy was made in 1910. I have used a microfilm of Schrader’s copy for Chapters 8–11.

G2 Government Oriental Manuscripts Library, Madras. Ms. no. R 3105. Palm leaf. Grantha script. 50 folios. 40 X 4 cms 6 lines per page. Between 45 and 70 akṣaras per line. Complete. A microfilm was used. Good condition. Carefully and neatly written.

G3 Government Oriental Manuscripts Library, Madras. Ms. no. D 2951. A Descriptive Catalogue of the Sanskrit Manuscripts in the Government Oriental Manuscripts Library, Madras. Ed. by M. Raṅgācārya. Vol. V—Dharma-śāstra (Madras, 1909), p. 2189. Palm leaf. Grantha script. 29 folios. 40 x 3 cms 6 lines on a page. 55 akṣaras per line. A microfilm was used. Several folios, containing the passages 5.19–7.55 and 9.50–60, are missing. After Ch. 10.30 the ms departs completely from the text and appends sundry verses.

G4 Government Oriental Manuscripts Library, Madras. Ms. no. R 3196(g). Palm leaf. Grantha script. 6 folios. 21 x 2.5 cms Between 8 and 9 lines on a page. 33 akṣaras on a line. Incomplete: contains only Ch. 8 and Ch. 9.1–9. The last folios of the ms appear to be loose folios written in a different hand and containing a different text on the procedure of renunciation.

G5 Government Oriental Manuscripts Library, Madras. Ms. no. D 2954. A Descriptive Catalogue of the Sanskrit Manuscripts in the Government Oriental Manuscripts Library, Madras. Ed. by M. Raṅgācārya. Vol. V—Dharma-śāstra (Madras, 1909), p. 2190. Palm leaf. Grantha script. 38 x 4 cms 7 lines on a page. 45 akṣaras per line. Contains only Chapter 11, which begins on p. 9 of the ms. The first eight pages contain a text on penances, beginning अथ संन्यासिनां प्रतिग्रहप्रायश्चित्तमाह. After Ch. 11, the ms contains a text on the procedure of renunciation that begins अथ संन्यासविधिरुच्यते.

G6 Government Oriental Manuscripts Library, Madras. Ms. no. D 2950. A Descriptive Catalogue of the Sanskrit Manuscripts in the Government Oriental Manuscripts Library, Madras. Ed. by M. Raṅgācārya. Vol. V—Dharma-śāstra (Madras, 1909), pp. 2188-89. Palm leaf. Kanarese script. 49 x 3.5 cms 8 lines on a page. 45–50 akṣaras per line. First two pages are lost. The ms begins at Ch. 1.14(21).

R1 University of Mysore Oriental Research Institute. Ms. no. P 2776. Descriptive Catalogue of Sanskrit Manuscripts. Ed. Dr. G. Marulasiddaiah (Mysore, 1979), vol. 3, ser. no. 8959. Palm leaf. Grantha script. 45 x 3.5 cms 61 folios. Between 5 and 9 lines on a page. 62 akṣaras on a line. Complete. Written in at least two different hands. Colophon:

इति यतिधर्मसमुच्चयं समाप्तम् । हरिः ॐ । शुभमस्तु । यादृशं पुस्तकं दृा तादृशं लिखितं मया । अबद्धं वा सुबद्धं वा मम दोषो न विद्यते । The rest is unreadable on the microfilm.

R2 University of Mysore Oriental Research Institute. Ms. no. P 6872. Descriptive Catalogue of Sanskrit Manuscripts. Ed. Dr. G. Marulasiddaiah (Mysore, 1979), vol. 3, ser. no. 3961. Palm leaf. Grantha script. 39.5 x 3.5 cms 60 folios. Between 6 and 8 lines on a page. 70 akṣaras on a line. Complete. Carelessly written in at least two hands.

R3 University of Mysore Oriental Research Institute. Ms. no. P 3002. Descriptive Catalogue of Sanskrit Manuscripts. Ed. Dr. G. Marulasiddaiah (Mysore, 1979), vol. 3, ser. no. 8960. Palm leaf. Grantha script. Approximately 6 lines on a page. 50 akṣaras on a line. Complete. Carelessly written with frequent corrections.

R4 University of Mysore Oriental Research Institute. Ms. no. P 2031. Descriptive Catalogue of Sanskrit Manuscripts. Ed. Dr. G. Marulasiddaiah (Mysore, 1979), vol. 3, ser. no. 8957. Palm leaf. Grantha script. 41 x 4 cms 89 folios. Approximately 6 lines on a page. 42 akṣaras on a line. Complete. Worm-eaten.

**B **Printed version in Devanāgarī characters edited by Srī Bhagavadācārya and published in Baroda in 1937 (76 pages). This is an extremely poor reprinting of the Grantha edition (M). The book is rare; I used the copy in the library of the Bhandarkar Oriental Research Institute, Pune.

M Printed version in Grantha characters edited by Mādapūsi Ramanujacāryāṃ and published in Madras in 1905 (139 pages). The editor states that he used sixteen mss in preparing his edition. These mss are not properly identified, and their variants are recorded only in Ch. 1. From Ch. 2 on the editor records few variants, listing them merely as pāṭhāntara. I have listed all variants recorded in this edition with the siglum M+. This is clearly not a critical edition, and a careful comparison of its readings with those of my mss shows that the editor has created a hybrid version. This edition is also rare; I used a microfilm of the copy in the British Library.

Other Manuscripts

Four mss in the Government Oriental Manuscripts Library, Madras (Ms. nos. D 2949; D 2952; D 2953; R 4957) and one ms in the University of Mysore Oriental Research Institute (Ms. no. P 2080) were too damaged and brittle to be copied or collated.

O. F. Schrader had several mss of the Ysam copied during his residence at the Adyar Library, Madras, early this century. These copies are now in the library of the University of GR_o_diaeresisttingen. Schr. 45 is a copy of A1, described above, prepared by V. Krishnamachari and N. Ramanatha Sastri and dated Dec. 12, 1911; this copy also contains variants from A2 recorded on the margins. Schr. 47 is a copy of G1 (both, for example, share the same lacuna at Ch. 6.3) made by K. Gopalaiyer and dated Nov. 26, 1909. Schr. 46 contains only Chapters 1-6 and is probably a copy of A5. Variant readings from a ms of the short recension are noted on this copy. Schr. 55 is an original ms containing extracts from the Ysam; see Klaus L. Janert and N. Narasimhan Poti, Indische Handschriften (Wiesbaden: Franz Steiner, 1979), 5: 450-51. This ms contains a variety of sundry material listed by Janert. The extracts from the Ysam are found on pp. 90–96 of the ms and follow the long recension (X*); most of the readings follow the subrecension X1*. I list here the chapter and verse numbers from the current edition according to the sequence in which they appear in this ms:

3.65(1, a-d), 62(2), 63, 64, 73; जन्तुनिवारणं जलपवित्रं कार्पासतन्तुवस्त्रमित्यर्थः (cf. variant in A6 at 3.73n), अष्टाङ्गयोगक्रमलक्षणानि वक्ष्यन्ते (cf. 5.47); **5.**51, 52, 53(a-d), 54, 55, 60, 82-91, 106-12; **6.**35, हारीतः, **6.**47, 243, 257, 259, 304; **7.**106, 138-39, अयने विषुवे चैव ग्रसने चन्द्रसूर्ययोः । उपवासमकुर्वाणः यतिश्चान्द्रायणं चरेत् ॥ विष्णुः **7.**22, शातातपः **6.**243, 249, 250(1), देवलः **7.**110, 111(a-b), 73, गर्गः **7.**130-36; सनत्कुमारसंहितै [sic] । प्रपत्तेः किंचिदप्येवं परापेक्षा न विद्यते । सा हि सर्वत्र सर्वेषां सर्वकामफलप्रदा । श्रुतिः । पूर्णस्य पूर्णमादाय पूर्णमेवावतिष्ठते । सर्वं पूर्णं सहों वासुदेवोऽसि पूर्णः । **7.**140; ताम्बूले भर्तृहीनायाः . . . संभूय सुरासमम् [cf. App. 1.8]; **7.**141; इति यतिधर्मसमुच्चये जमदग्निः ।

Genealogical Relation of the Manuscripts

The following stemma codicum, in which Z* stands for the hypothetical original of the Ysam, presents the genealogy of the mss used to constitute the text:

…

Recension X*

This recension contains the long version of the Ysam. Although the mss belonging to X* differ among themselves, their common readings stand in sharp contrast to those of the mss of Y*. I have identified two subrecensions of X*. The mss of X2* have numerous unique readings that differ substantially from those of X1*. Among the mss of X2*, G6, R2, and R3 appear to form a group with many unique readings, while R4 often has readings in common with mss of X1*. The subrecension X2* is also notable for several attempts to shorten the text, as at 2.48, 49; 3.65; and 5.136, 155. It shares this propensity with Y*, and in several places Y* and X2* show identical omissions, e.g., 5.117-19, 130-35, 137-43, 149-50. Among the mss of subrecension X1*, A1, A2, and R1 have many common readings and appear to be derived from a common ancestor. These three mss also share many common readings with Y* (e.g., 5.105; 6.54 and App. 1.4; 6.61; 7.129, 139), and it appears that the editor who made the short version used a ms derived from the same ancestor.

Recension Y*

This recension contains the short version of the Ysam. In contrast to the mss of X*, those belonging to Y* show a remarkable uniformity in their readings. I believe that all can be traced to a common source, which, as I will argue below, is the abbreviated version of the Ysam deliberately produced by an editor.

Composite Manuscripts

Two mss, A3 and A6, are hybrid versions of the two recensions. After Ch. 7, as we have seen, A3 appends a variety of other material from diverse sources regarding yatidharma. The scribe or editor of A3 appears to have used a variety of manuscript material, including some belonging to the two recensions. In the early sections, A3 follows recension X1*, but at 6.139 it switches to Y* until the end of Ch. 7. A similar switch between the two recensions is found also in A6, which follows X1* from the beginning until 8.30, Y* from 8.31 to 10.127, and again X1* from 10.128 until the end. Because the switching between recensions occurs quite suddenly in the middle of chapters, the presumption must be that these mss do not represent deliberate attempts to create hybrid versions. This is certainly true of A6. In all likelihood, some folios were missing in the exemplars used by the scribe, who then used other mss belonging to the short recension to supply the missing sections. The case is less clear with regard to A3, which frequently follows the readings of Y* even in those sections where it agrees overall with X1* (e.g., 2.15–17). It is possible that the scribe/editor of A3 used a ms of Y* to make a hybrid version of the entire text.

Constitution of the Text

The manuscript genealogy I have constructed is tentative at best and on its own does not provide an adequate basis for constituting the text. Most mss of the Ysam are in extremely poor condition, and it is very difficult to obtain a clear picture of the ms tradition from the available evidence. The constituted text presented here is the best that can be done without the discovery of new and significant mss.

The major issue to be resolved in constituting as far as possible the original text written by Yādava Prakāśa is the priority of the two recensions.¹ Numerous Indian texts, such as the Mahābhārata and the Manusmr̥ti, profess to be abridgments of earlier and longer versions. It is, nevertheless, a common scholarly assumption based on good evidence that most Indian texts, especially the Epics, Purāṇas, and Dharmaśāstras, were subjected to alteration principally by addition and accretion.²+++(5)+++ The assumption, therefore, is that longer versions are relatively more recent than shorter ones. The Ysam, however, proves to be an exception. The long recension, X*, in my estimate, is closer to the original than the short recension Y*, even though the priority of X* cannot be proved conclusively. The arguments I present taken together, however, suggest strongly that Y* is a later abridgment of the Ysam.

First, when the general principles of textual criticism, such as lectio difficilior and internal coherence, are applied, X* (especially X1*) most frequently presents the superior readings. An examination of 5.26, where Yādava explains the obscure half-verse 5.23c-d, shows how Y* and X2* have failed to understand the Sanskrit and attempted to emend the text. Likewise, the inability to perceive the reference to a later section (6.5) makes Y* delete the statement at 5.69.

Second, we find omissions in Y* that can be explained as scribal errors on the basis of X*, whereas it is difficult to see how these sections could be deliberate additions of X*. For example, Y* jumps from 3.28(1) to the end of 3.33, where we can see the editor combining the reading of 3.28(1) with the final two words of 3.33. A careful reader can readily see that the reading of X* is far superior.

We have seen, moreover, that A1, A2, and R1 have unique readings that set this group of mss apart from others belonging to X*. According to general editorial principles,³ the divergence of this cluster from the tradition represented by the rest of the mss belonging to both subrecensions X1* and X2* must be presumed to be more recent than the divergence of X2* from X1*. We have also seen that the readings unique to this cluster are shared by Y*, in contrast to the other mss of X*, suggesting that the editor of Y* used a ms derived from the same ancestor as A1, A2, and R1. This feature is difficult to explain if Y* represents the original version from which X* diverged.

The uniformity of the readings of the mss belonging to Y* and their relatively small number further support the more recent origin of this recension.

A couple of pieces of external evidence also support the priority of X*. The only direct citation from the Ysam I have been able to trace thus far is in the Śrīpāñcarātraparīkṣā (p. 136) by Vedānta Deśika (1268–1369 C.E): यथोक्तं यतिधर्मसमुच्चये—फलपुष्पोत्पाटने प्रायश्चित्तविधानादद्भिः कर्तव्यमन्याहृतैर्वा. This passage, found in Ysam 6.73,⁴ is preserved only in the long recension. There is a second reference to a passage of the Ysam in Deśika’s Ybh. He says that Yādava Prakāśa disallows an option between a difficult and an easy course of action. This principle is laid down in Ysam 3.60, which again is missing in the short recension. So Vedānta Deśika, writing only a couple of centuries after Yādava, had before him the long recension. Deśika’s testimony is all the more significant both because he assumed the leadership of the Śrī-Vaiṣṇava church in the early fourteenth century and because he was closely connected to the major figures of its early history. Deśika’s maternal uncle, Ātreya Rāmānuja (1221–95 C. E.) was the teacher of Vatsya Varada, the author of the Prameyamālā and the Yatiliṅgasamarthana (see Olivelle 1986, 23). Varada was the great-grandson of the sister of the great Rāmānuja, the founder of the sect and, according to tradition, initially a disciple of Yādava Prakāśa.⁵ Deśika was thus in a position to obtain the best manuscripts of the Ysam. It is, of course, possible that the long recension was created out of the short before the time of Deśika, but, in view of the internal evidence given above, it is highly unlikely.

A careful examination of Y* makes it clear that it is a deliberately created edition and not the result of unintended scribal errors. Space does not permit me to explore fully the editorial principles followed by the editor of Y*, but several emerge from a close reading of the text: eliminating quotations that repeat what has already been said; deleting passages with difficult readings or where many mss have lacunae (e.g., 7.40; 9.82; 10.82); eliminating most of the explanatory material (e.g., 2.15); and rearranging verses for particular reasons. An example of the last is found in Ch. 3, where the editor places 3.46-50 after 3.56 and omits 3.57-58 (which appear to intrude into the flow of the discussion). Thus, verses dealing with the obligations to carry a staff are kept together, and the verses regarding tying a cord to the triple staff are placed last; the entire passage ends nicely with 3.59, which tells the ascetic not to tie a cord to a single staff.

Between the two subrecensions of the long version, X1* gives consistently superior readings. Among the mss of X1*, moreover, A3, A5, and A6 are the most reliable and give consistently superior readings. The most trustworthy portion of the constituted text is thus that part (until Ch. 6.138, where A3 switches to Y*) for which we have the evidence of these three mss. The constituted text from 6.139 to 8.30 is also by and large reliable, because we have the testimony of A5 and A6. The constitution of the remainder of the text was very difficult, because there I had to rely greatly on the mss of X2* and on A1, A2, and R1, the three least reliable mss of X1*. I am least certain about the constituted text of Ch. 11; here, the mss give widely divergent versions.

As a rule, I follow X1* when all its mss have the same reading. Where X1* presents divergent readings, I normally choose the readings shared by A5, A6, and the mss of X2*; in general, readings unique to A1, A2, R1, and the mss of Y* proved to be the least reliable. In the sections where it does not follow Y*, A6 proved to be the most reliable ms, giving consistently superior readings, especially of difficult passages (e.g., 3.24, 62; 5.26; 6.310; 7.75 the dual *yātaḥ; *7.61-62). Metrical reasons sometimes forced the selection of a reading not supported by the majority of mss (e.g., 5.82; 6.64).

There are times, however, when one has to follow one’s instinct. After working with this genre of literature for a lifetime, and after working closely with this text (including the manual collation of all the mss) for many years, I think I have developed a feel for Yādava’s style—in art history they call it a connoisseur’s judgment. Such judgments should be made judiciously and infrequently, but the connoisseur’s touch is an essential part of an editor’s repertoire.

The two printed versions were of little use in constituting the text. They follow X2* sometimes, the ms tradition of A1, A2, R1 at other times, and sometimes even Y*, as in Ch. 11. Further, they introduce new readings without any foundation in the available mss.

The Critical Apparatus

The critical apparatus lists all the significant variant readings, except obvious scribal errors. Longer variants that could not be conveniently listed at the foot of a page are placed in Appendix 1. Most Indian scribes, but especially those writing in Grantha and Telugu scripts, are altogether inconsistent in their notation of anusvāras, visargas (especially before conjunct consonants), and avagrahas; neither are they consistent in the use of sandhi. These variants are not noted in the critical apparatus.

I have attempted to establish some consistency in these areas, except when the absence of sandhi is obviously used as punctuation.+++(5)+++ The spelling of the term pārivrājya is totally inconsistent, even within the same ms: parivrājya, pārivrajya, parivrajya. I have consistently used the standard pārivrājya.

Due to the exigencies of preparing camera-ready copy on a computer, I have been forced to place the entire critical apparatus for a verse or a prose passage at the bottom of the page on which the verse number appears. Thus, the variant readings of sections of some passages may be listed at the bottom of the following page when those passages span two pages.

Grammatical Peculiarities

We must, of course, assume that Yādava knew his Sanskrit. On the whole, correct grammar may be a reason to prefer a particular reading, yet there are times in this genre of literature, especially in verses quoted from sundry sources, when the established rules of Pāṇinian grammar are violated. I have not attempted to correct any of them; for example, accusative mātrāṃ when we would have expected the nominative (3.22) or spr̥ṣṭvā with instrumental (8.43). The most common such “error" is found in the gerund, where the -ya ending is used with uncompounded verbs: gr̥hya (8.44, 45; 9.34); sthāpya (10.31).+++(5)+++ There are several occasions when “double sandhi" is used in verses for metrical reasons (e.g., 6.156, 266).

Notes