On the Written Transmission of the Pātañjalayogaśāstra

On the Written Transmission of the Pātañjalayogaśāstra∗ PHILIPP A. MAAS 1 The Yogasūtra of Patañjali with its oldest commentary, the so- called Yogabhāùya, is one of the most widely read or, at least, one of the most often copied texts in the field of classical Indian Phi- losophy. I have been able to trace thirty-seven printed editions pub- lished from 1874 to 1992 and eighty-two MSS in public libraries in India, Nepal, Pakistan, Europe and the USA.1 1.1 Not only do these high numbers indicate the popularity of these texts – for which, in accordance with the information pro- vided by the colophons, I use the title Pātañjalayogaśāstra (PYŚ) whenever I refer to them collectively – but also the fact that the PYŚ became the subject of at least three subcommentaries. The most famous, without doubt, is the Yogasūtrabhāùyavyākhyā or Tattvavaiśāradī (TVś) of Vācaspatimiśra I, who must have lived at some time between 890 and 984/985 AD (Srinivasan 1967: 63). Although the exact dating of the PYŚ is not conclusively deter- mined, a considerable gap of time and substantial differences in philosophical views clearly separates Vācaspati from the author(s) ∗ I would like to express my gratitude to Prof. Dr. Albrecht Wezler (Univer- sity of Hamburg), to Dr. Harunaga Isaacson (University of Pennsylvania), and to Prof. Dr. Claus Vogel (University of Bonn) for reading provisional versions of this paper. Susanne Kammüller, M.A. was kind enough to check my English. 1 I am currently preparing “A Hand-list of Manuscripts and Printed Editions of the Pātañjalayogaśāstra and the Commentaries thereon” for publication. 88 PHILIPP A. MAAS of the PYŚ. This is even more true of Vijñānabhikùu’s Yoga- vārttika, which seems to have been composed in the latter half of the 16th century.2 1.2 The third subcommentary is the Pātañjalayogaśāstra- vivaraõa (YVi), which was edited on the basis of a single Malayā- lam MS and published under the title “Pāt[a]ñjala-Yogasūtra- Bhāùya Vivaraõam of Śaïkara-Bhagavatpāda” (Rama Sastri & Krishnamurthi Sastri 1952). Whether or not the famous Advaitin Śaïkara was the author of the YVi is, as far as I can see, not yet decided, and I am not at all inclined to enter into that discussion here. For my present purpose, it may be sufficient to emphasize that the YVi’s importance for the history of Indian philosophy was immediately realized by scholars in Europe, Japan and the USA.3 Even in India, in circles among modern Vedāntins, the first com- plete edition of the YVi was echoed by a reconstruction of the first chapter of PYŚ as it was commented upon by the YVi-kāra.4 1.2.1 To my knowledge, Wezler was the first to stress not only the YVi’s philosophical importance but also its philological value. Almost filled with enthusiasm, he sums up his “Philological Ob- servations” (1983: 32): ... [T]o anyone experienced in dealing with problems of textual criticism it becomes plain that the author of the Vivaraõa knew or had before him a text of the Y[ogasūtra]Bhāùya that is definitely older than that known to Vācaspatimiśra and comes hence much closer to the original. 1.2.2 Wezler is perfectly right in claiming that the YVi-kāra based his commentary on a version of the PYŚ that contained more original readings than the printed editions nowadays available. The 2 For details see Larson & Bhattacharya 1987: 376. 3 See for example (in alphabetical order): Bronkhorst 1985, Hacker 1968–69, Halbfass 1991, Mayeda 1968–69, Nakamura 1980–81, Oberhammer 1977, Schmithausen 1968–69, Vetter 1979, Wezler 1983, Whaling 1977. 4 Vedavrata 1984. The edition adopts many more readings from the YVi than the version of the PYŚ printed together with the first complete edition of the YVi. It lacks, however, a systematical approach. On the Written Transmission of the Pātañjalayogaśāstra 89 relation of these versions to Vācaspatimiśra’s TVś is less clear. We neither possess a critical edition of the TVś nor a critical study of the basic text commented upon. Therefore, Halbfass (1991: 223) rightly demanded that: [M]uch further study of the textual tradition or traditions ... is needed before definite conclusions concerning the relative chronology of the Vivaraõa and the Vaiśāradī ... can be drawn. 1.3 Although we are still a long way from “definite conclu- sions”, our knowledge on the topic at hand has improved. Harimoto has prepared a new critical edition of the first chapter of the YVi considering more textual witnesses than were used for the first edi- tion.5 In preparing a critical edition of the first chapter of the PYŚ, I not only utilized the new critical edition of the YVi for a recon- struction of its basic text, but also had the chance to personally dis- cuss preliminary results with him. In addition to this valuable textual witness, I could make use of twenty-two printed editions and of twenty-five MSS in seven scripts and from different regions of the Indian subcontinent. In the first chapter the witnesses are at variance in nearly 2180 cases, of which about 900 are substantial. 2 The variant readings do not allow us to reconstruct the history of the PYŚ’s transmission in detail, because it is contaminated. While preparing new copies, scribes often did not use a single exemplar but compared several MSS. This process can be proved for a large number of MSS containing so-called “corrections” in the margin of the folio or elsewhere. There is no agreement with regard to the question which of two or more possible readings is the original one, and, in some cases, even corrections were corrected, pointing to a double process of checking one MS against others. 5 Harimoto (1999: 3–14) uses five textual witnesses. 90 PHILIPP A. MAAS Srinivasa Ayya Srinivasan has already assumed that contami- nation did not start in comparatively late times.6 This also holds good for our text, as can be deduced from the fact that the textual witnesses – with the exception of some printed editions – do not form “solid genetic groups”,7 i.e., groups containing a high number of common errors that most probably did not creep into the trans- mission independently. In other words, contamination shows itself by the simple fact that no stemmatic hypothesis can satisfactorily explain the relationships existing among all witnesses (West 1973: 36). 3 Although contamination has been a constant factor within the transmission, its varying degrees have not altogether made stem- matical considerations impossible. There are several groups of wit- nesses discernible by the occurrence of errors shared by their mem- bers in a significant number but not in a regular pattern. The two main groups are the “Northern group” and the “South- ern group”. The first of these is represented by nearly all printed editions and by all MSS from North and Middle India in Deva- nāgarī, Śāradā and Maithilī script. The Southern group is repre- sented by MSS in Telugu-Kannaóa script, in Grantha and in Malayālam script. The basic text of the YVi is also part of this group. Both main groups contain regional subgroups, and some late MSS from the South are difficult to sort into either of the two main groups. This is most probably due to the contaminating influence of the version transmitted by the Northern group which, in the course of time, seems to have gained the status of a normative recension and can, therefore, be designated as the Vulgate. 3.1 Within the Southern group the basic text of the YVi holds a special position, as it does not show close affinities to any sub- group. Although, for example, it exclusively shares a number of 6 Srinivasan 1967, §1.1.11, p. 5. 7 Srinivasan 1967, § 1.1.12, p. 6. On the Written Transmission of the Pātañjalayogaśāstra 91 readings with a fairly old Malayālam MS, the total number of such readings is far lower than one would expect from the fact that all known MSS of the YVi are in Malayālam characters. If one takes into consideration that the basic text of the YVi preserves primary readings that are not shared by any other MS, its most likely posi- tion within the transmission is quite close to the common ancestor of the Southern group. On the other hand, the “extraordinary testimonial value” (Wezler 1983: 32) of the YVi for a critical edition of the PYŚ is unfortunately limited by a number of factors. First of all, the YVi has come down to us in quite a poor state of transmission. Even its archetype (the common ancestor of all known MSS) contained a considerable number of more or less obvious errors. Secondly, the basic text of the YVi seems to have contained errors that are not transmitted by any other witness. Moreover, while judging readings from the YVi we have to keep in mind the possibility that comparatively late versions of the PYŚ have influenced its trans- mission, as scribes may have more or less consciously changed the wording of the YVi according to their knowledge of the basic text. Finally, the YVi-kāra, as a creative writer, cannot be expected to have slavishly stuck to his basic text. We always have to reckon with the possibility that readings of the PYŚ were ultimately in- vented by the YVi-kāra himself, in order to adapt the meaning of the basic text to his own philosophical views.8 Therefore, any re- construction of the YVi’s basic text will always be fraught with a substantial amount of uncertainty that can only be diminished by a careful philological analysis of the YVi, on the one hand, and by 8 In dealing with the YVi we have to keep in mind Steinkellner’s remarks on using commentaries as hermeneutical tools: “On the one hand it is necessary to use those explanations which prove to be useful for an understanding of the basic text, and to distinguish these explanations according to their degree of authority. And on the other hand the extensions and digressions are to be exam- ined with regard to their testimony for a development of the doctrine. Finally, if such development is to be met with, we have to pay attention to what extent this development has influenced the plain explanatory parts of the comments, too.” (Steinkellner 1981: 283) 92 PHILIPP A. MAAS comparing assumed readings with the rest of the transmission on the other. 3.2 It is, of course, hazardous to propose any concrete dating for the time when the transmission was divided into two groups, but some general considerations may not be totally out of place. In any case, we are looking for an early date, as MSS transmitting the Vulgate are found in a vast geographical area comprising the whole of the Indian subcontinent with the exception of the extreme south. The period of time that has passed since this division must be long enough for regional subgroups to have developed. If a critical study of the TVś should support Wezler’s observations and demonstrate that Vācaspati knew or had in hand a version containing typical errors of the Vulgate, the latest possible dating would be towards the end of the 9th century, although nothing prevents us from as- suming a much earlier date. 4 Before discussing a number of variant readings capable of sup- porting the most basic assumptions of this general outline, it may be useful to describe the principles of textual criticism applied for the constitution of the text. A reading, in order to be adopted, has to stand a triple test. It must fulfil each of the following criteria (West 1973: 48): 1. [A reading] must correspond in sense to what the author intended to say, so far as this can be determined from the context. 2. It must correspond in language, style, and any relevant technical points ... to the way the author might naturally have expressed the sense. 3. It must be fully compatible with the fact that the surviving sources give what they do; in other words, it must be clear how the presumed origi- nal reading could have been corrupted into any different reading that is transmitted. These criteria have been developed in the field of Greek and Latin classics, but to me they seem applicable in the field of Indian philosophical texts as well, although we face some difficulties. We usually neither know much about “the author” or “the authors” of a On the Written Transmission of the Pātañjalayogaśāstra 93 given text, nor do we know much about the process of composing philosophical texts. Nevertheless, judging variant readings must, of course, involve considerations of language and style as well as of the context in which they appear. 4.1 With these considerations in mind, we can take a look at PYŚ 1.45. The non-uniform transmission of this passage bears out two stemmatic key facts: 1) The Vulgate is free from errors transmitted by the Southern group. 2) The basic text of the YVi belongs to the Southern group. Table 1 Southern Version of YBh 1.45 Vulgate of YBh 1.45 (simplified) (simplified) pārthivasyāõor gandhamātratā1 pārthivasyāõor gandhatanmātraü sūkùmo viùayaþ; sūkùmo viùayaþ; gandhamātrasyāpi2 āpyasyāpi rasatanmātram, taijasasya rūpatanmātram, vāyavīyasya sparśa- tanmātram, ākāśasya śabdatanmātram, teùām ahaükāraþ, asyāpi liïgamātraü, liïgamātrasyāpy liïgamātram, liïgamātrasyāpy aliïgaü sūkùmo viùayaþ. na cāliïgāt aliïgaü sūkùmo viùayaþ. na cāliïgāt paraü sūkùmam asti. paraü sūkùmam asti. 1) Mag, Myt3, Tvy, Tjg; gandhamātratā{sva YVi}rūpamātrateti EFg, YVi 337,2. 2) gandhatanmātrasyāpi Myt3, Tvt; **trasyāpi Tvy; gandhamā{tanmā EFg}tra {traliügamātra<ü> EF}sva{om. EFg}rūpamātrasyāpi EFg, YVi 337,3. 4.1.1 YS 1.42–45 deals with a series of meditative states called samāpatti. This series consists of four samāpattis which differ from each other by the subtlety of their respective meditative object. YS 1.45 describes the utmost degree of subtlety: sūkùmaviùayatvaü 94 PHILIPP A. MAAS cāliïgaparyavasānam / “Furthermore, having [even more] subtle objects ends with [primordial matter, which is] free from any char- acteristic (aliïga).” The Bhāùya in its Vulgate version explains, in the terms of sāükhya-metaphysics, why primordial matter is the final depth layer of meditative objects:9 pārthivasyāõor gandhatanmātraü sūkùmo viùayaþ; āpyasyāpi rasatan- mātram, taijasasya rūpatanmātram, vāyavīyasya sparśatanmātram, ākāśa- sya śabdatanmātram, teùām ahaükāraþ, asyāpi liïgamātram, liïgamātra- syāpy aliïgaü sūkùmo viùayaþ. na cāliïgāt paraü sūkùmam asti. The subtle object[-level] of the “earthen” gross element is the subtle element smell, and of the ‘watery’ [gross element] it is the subtle element taste; of the “fiery” it is the subtle element of form; of the “windy” it is the subtle ele- ment of touch; of the spacious [gross element] it is the subtle element of sound; their [subtle object-level] is egoity, and the [subtle object-level] of this is characteristic-only (liïgamātra), and the subtle object[-level] of char- acteristic-only is [primordial matter, which is] free from any characteristic (aliïga). And there is nothing more subtle than [primordial matter, which is] free from any characteristic. Within the Southern group this passage is transmitted in two versions. Three Grantha MSS read mātram instead of tanmātram almost consistently, and, moreover, have nābhasasya instead of ākāśasya.10 More important is the common reading of some other MSS belonging to the southern group – including the basic text of the YVi – that read a shorter text: pārthivasyāõor gandhamātratā sūkùmo viùayaþ; gandhamātrasyāpi liïga- mātraü, liïgamātrasyāpy aliïgaü sūkùmo viùayaþ. na cāliïgāt paraü sūkùmam asti. The difference results from a loss of text in an early ancestor of the Southern version. It can easily be explained by the double oc- currence of asyāpi in two neighbouring lines of a common source. Presumably, a scribe slipped from one line to the other and over- 9 For details see Oberhammer 1977: 198–209. 10 These MSS are Mag, Tjg1 and Tjg2. On the Written Transmission of the Pātañjalayogaśāstra 95 looked the intervening text. The surviving asyāpi would, in a sec- ond step, have been changed to gandhamātrasyāpi, in order to im- prove the intelligibility of the sentence. If these assumptions are correct, the common source had around fourty akùaras per line. 4.2 The non-uniform transmission of PYŚ 1.29 bears out one more stemmatic key fact: 3) The Southern group is free from errors transmitted by the Vulgate. Table 2 Southern Version of PYŚ 1.29 Vulgate of PYŚ 1.29 (simplified) (simplified) kiücāsya bhavati tataþ pratyak- kiü cāsya bhavati? tataþ pratyak- cetanādhigamo ’ntarāyābhāvaś ca cetanādhigamo ’py antarāyābhāvaś ca (YS 1.29). ye tāvad antarāyā vyādhi- (YS 1.29). ye tāvad antarāyā vyādhi- prabhtayas, te tāvad īśvarapraõidhānān prabhtayas, te tāvad īśvarapraõidhānān na bhavanti. na bhavanti. svapuruùadarśanam1 apy asya bhavati: svarūpadarśanam apy asya bhavati: “yathaiveśvaraþ “yathaiveśvaraþ puruùaþ śuddhaþ, prasannaþ, kevalo, ’nupa- śuddhaþ, prasannaþ, kevalo, ’nupa- sargas, tathāyam api buddheþ prati- sargas, tathāyam api buddheþ prati- saüvedī saüvedī madīyaþ2 puruùa”, ity yaþ puruùas, tam puruùa”, ity evam adhi{v.l.: va}gacchatīti. adhigacchati. adhigacchati 1) EFg, Tvy. 2) EFg, Mag, Tvy, YVi 281,3. “[T]he [yogin], moreover, acquires, because of this [devotion to Īśvara], the realization of [his] inner consciousness (pratyakcetanādhigama) and the non- existence (or not coming into being) of ‘hindrances’ (antarāya) [on the yogic path] (YS 1.29). Whatever hindrances there be, disease and so on, all these, because of devotion to Īśvara, do not come into being (or: do not exist). [T]he [yogin] acquires even sight (or: knowledge) of his own Self (puruùa): ‘As Īśvara is pure, clear, alone and free from trouble, so also is my Self here that experi- ences [its] buddhi.’ Thus [t]he [yogin] realizes.” 96 PHILIPP A. MAAS 4.2.1 YS 1.29 deals with two benefits the yogin acquires by devotion to Īśvara (īśvarapraõidhāna). It reads: tataþ pratyak- cetanādhigamo ’ntarāyābhāvaś ca. “Because of this [devotion to Īśvara, the yogin acquires] the realization of [his] inner conscious- ness (pratyakcetanādhigama) and the non-existence (or not coming into being) of ‘hindrances’ (antarāya) [on the yogic path].” The Vulgate-Bhāùya is of little help in determining the meaning of pratyakcetanādhigamo. It reads: svarūpadarśanam apy asya bhavati / “[T]he [yogin], moreover, gets sight (or: knowledge) [of his (or: the)] own-form.” This passage is difficult. On the one hand, rūpa “form” goes quite well with darśana “sight” but the exact meaning of svarūpadarśanam here and its relation to pratyakceta- nādhigamo from the sūtra is unclear. Should we assume that sva- rūpadarśana is not a gloss but rather states an additional result of devotion to Īśvara? 4.2.2 Both well-known commentators on the PYŚ, Vācaspati and Vijñānabhikùu, solve the problem in peculiar but ultimately un- satisfactory ways,11 and the YVi does not transmit the passage under discussion. 11 Vācaspati comments: tataþ pratyakcetanādhigamo ’py antarāyābhāvaś ca [YS 1.29] | pratīpaü viparītam añcati vijānātīti pratyak sa cāsau cetanaś ceti pratyakcetano ’vidyāvān puruùaþ | tad aneneśvarāc chāśvatikasattvotkarùasaü- pannād vidyāvato nivartayati | pratīcaś cetanasyādhigamo jñānaü svarūpato ’sya bhavat[i |] (TVś 33,19–21). “From this [devotion to Īśvara results the] realization of a sentient being (cetana) that performs [mental acts] opposingly. [To explain pratyak: ‘it] performs [mental acts] (añcati) opposingly (pratīpaü)’ [means ‘it] knows contrarily [to reality]’. ‘A sentient being (cetana) that per- forms [mental acts] opposingly’ [=] ‘a person (puruùa) possessing ignorance’. With this [expression the author] differentiates [the ordinary person] from Īśvara who is endowed with the perfection of [perceiving his] eternal sattva [and there- fore] possesses knowledge. As a sentient being (cetana) that performs [mental acts] opposingly [t]he [yogin] acquires the ‘realization’ [=] ‘knowledge accord- ing to its own form’.” This passage does not allow to reconstruct the TVś’s basic text in detail. Vācaspati’s interpretation of pratyakcetana may be caused by his difficulties in interpreting svarūpadarśana from the basic text. Vijñānabhikùu, on the other hand, comments (YVā 89,12f.): svarūpa- darśanam iti | asya pratīco jīvasya yat tāttvikam rūpaü tasya sākùātkaro ’pi bhavatīty arthaþ | “‘Sight (or: knowledge) of [his] own form’ means ‘he acquires On the Written Transmission of the Pātañjalayogaśāstra 97 4.2.3 A quite simple solution is to accept a common reading of two MSS, one in Grantha and the other in Malayālam characters, that reads svapuruùadarśanam “sight (or: knowledge) of the [yogin’s] own Self (puruùa)” instead of svarūpadarśanam. The former reading, in my opinion, is a paraphrase of pratyak- cetanādhigamo “realization of [his] inner consciousness” from the sūtra. 4.2.4 There may, however, remain some doubt about whether svarūpadarśana is not, in fact, the more difficult reading, and should, therefore, be regarded as primary. Although it cannot be ruled out entirely that, in the course of transmission, svarūpa was deliberately changed to svapuruùa, it is much more likely that sva- rūpa is simply the result of the loss of the akùara pu. The remaining svaruùa would then have been “corrected” to svarūpa. This correc- tion is quite obvious if one takes into consideration the similarity in North Indian alphabets of ru and rū on the one hand, and of ùa and pa on the other.12 4.3 The following excerpt from the Bhāùya not only supports svapuruùa° as the original reading by supplying a suitable context, but also contains a second error of the Vulgate. 4.3.1 “yathaiveśvaraþ śuddhaþ, prasannaþ, kevalo, ’nupa- sargas, tathāyam api buddheþ pratisaüvedī madīyaþ puruùa”, ity adhigacchatīti. Syntactically, this passage is in the form of a direct construction with iti at the end. The main verb adhigacchati takes up the sūtra’s adhigamo. The sentence before ity adhigacchati de- scribes the content of the yogin’s spiritual realization. ‘As Īśvara is pure, clear, alone and free from trouble, so also is my Self here that experiences [its] buddhi.’ Thus [t]he [yogin] realizes.” 4.3.2 A slip of a scribe’s eye got him to overlook the akùaras madī right behind pratisaüvedī. As a result instead of pratisaüvedī madīyaþ only pratisaüvedī yaþ survived. The second word can be also the realization of the form which is the real (tāttvika) [form] of the inner individual soul (jīva)’.” 12 Bühler 1896: Tafel 4 and Tafel 4a. 98 PHILIPP A. MAAS taken as a relative pronoun, although, of course, it does not fit syn- tactically. In the course of transmission, two scribes, presumably, chose different strategies to solve the syntactical problems caused by yaþ. One scribe changed iti to tam, in order to construe an apodosis (yaþ puruùaþ, tam adhigacchati), the other deleted yaþ. 4.3.3 As in the example discussed above, there is no absolute certainty regarding the original reading. It is also possible, though much less likely, that the original version neither contained the possessive adjective madīyaþ nor the relative pronoun yaþ. I can see no reason why madīyaþ should have been inserted here, and its presumable loss is easy to explain. Moreover, there is a con- stant line of argumentation that leads from the Bhāùya’s gloss of pratyakcetanādhigamo as svapuruùadarśanam to madīyaþ puruùa ity adhigacchati. 5 The analysis of variant readings in PYŚ 1.29 points to what, a priori, could have been regarded as most likely. At first, to quote Wezler once more (1983: 32), “...cases where what can be called ‘secondary transmission’ of a text turns out to be more valuable than all extant MSS. taken together, are not rare in our discipline”, as are cases, I would like to add, where South Indian MSS are more valuable than northern MSS, impressive examples being Rau’s ob- servations on the transmission of Bharthari’s Vākyapādīya,13 as well as the Jaiminīyabrāhmaõa, a re-edition of which is currently under preparation by Fujii and Ehlers (Ehlers 2000). 13 Rau 1977: 30; 1991: 4. On the Written Transmission of the Pātañjalayogaśāstra 99 ABBREVIATIONS AND REFERENCES (a) Texts Pātañjalayogaśāstravivaraõa. Pāt[a]ñjala-Yogasūtra-Bhāùya-Vivaraõa of Śaï- kara-Bhagavatpāda. Critically ed. with introduction by Polakam Sri Rama Sastri and S. R. Krishnamurthi Sastri. (Madras Government Ori- ental Series, 94.) Madras 1952. PYŚ Pātañjalayogaśāstra (YS along with YBh). TVś Tattvavaiśāradī or Yogasūtrabhāùyavyākhyā by Vācaspatimiśra I. Pātañjalayogasūtrāõi. Vācaspatimiśraviracitañīkāsameta-Śrī-Vyāsa- bhāùyasametāni. Āśramasya paõóitaiþ saüśodhitam. 4th edition. (1st ed. 1904). (Ānandāśrama Sanskrit Series, 47.) Puõyapattana [= Pune] 1978. YBh Yogasūtrabhāùya. YS Yogasūtra by Patañjali. YVā Yogavārttika by Vijñānabhikùu. Pātañjalayogadarśanam. Vācaspati- miśraviracita-Tattvavaiśāradī-Vijñānabhikùukta-Yogavārtikavibhūùita- Vyāsa-bhāùyasametam. ŚrīNārāyaõamiśreõa ñippaõīpariśiùñādibhiþ saha sampāditam. Vārāõasī 1971. YVi Pātañjalayogaśāstravivaraõa. In: Kengo Harimoto (ed.), A Critical Edi- tion of the Pātañjalayogaśāstravivaraõa. First Part. Samādhipāda with an introduction. A Dissertation in Asian and Middle Eastern Studies. Philadelphia: University of Pennsylvania, 1999. (b) Manuscripts and Catalogues EFg Digital pictures of a palm leaf MS containing the PYŚ in Grantha script from the Library of the “École Française d’Extrême-Orient, Centre de Pondichéry”, Pondicherry. Shelf No. 287. Mag Digital pictures of a paper MS containing the PYŚ in Grantha script from the “Adyar Library”, Chennai. Running No. 24 (in Cat. Adyar). Shelf No. PM 1420. Myt3 Digital pictures of a palm leaf MS containing the PYŚ in Telugu script from the library of the “Oriental Research Institute”, Mysore. Running No. 35065 (in Cat. Mysore). Shelf No. P 1560/5. Tjg1 Microfilm pictures of a palm leaf MS containing the PYŚ in Grantha script from the “Tanjore Mahārāja Serfoji’s Sarasvatī Mahāl Library”, Thanjavur. Running Nos. 9904 (in Burnell 1880) and 6703 (in Cat. Tan- jore). Tjg2 Microfilm pictures of a palm leaf MS containing the PYŚ in Grantha script from the “Tanjore Mahārāja Serfoji’s Sarasvatī Mahāl Library”, 100 PHILIPP A. MAAS Thanjavur. Running Nos. 9903 (in Burnell 1880) and 670 (in Cat. Tan- jore). Tvt Digital pictures of a palm leaf MS containing the PYŚ in Telugu script from the library of the “Oriental Research Institute”, Thiruvanantha- puram. Running No. 13474 (in Cat. Trivandrum). Shelf No. 11837A. Tvy Digital pictures of a palm leaf MS containing the PYŚ in Malayālam script from the library of the “Oriental Research Institute”, Thiru- vananthapuram. Running No. 14371 (in Cat. Trivandrum). Shelf No. 622. BURNELL, A[rthur] C[oke] 1880. A Classified Index to the Sanskrit Mss. in the Palace of Tanjore. Prepared for the Madras Government. London. Cat. Adyar = AITHAL, Parameswara 1972. Descriptive Catalogue of Sanskrit MSS [in the Adyar Library], VIII: Sāükhya, Yoga, Vaiśeùika and Nyāya. (The Adyar Library Series, 100.) Adyar, Madras. Cat. Mysore = MARULASIDDAIAH, Gurusiddappa 1984. Descriptive Catalogue of Sanskrit MSS [in the Oriental Research Institute, Mysore], I–XV. Vol. 4B, 7–15 ed. by H. P. Malledevaru. Vol. 10: Vyākaraõa, Śilpa, Ratna- śāstra, Kāmaśāstra, Arthaśāstra, Sāükhya, Yoga, Pūrvamīmāüsā, Nyāya. (Oriental Research Institute Series, 144.) Mysore. Cat. Tanjore = S[UBRAHMANYA] SASTRI, P[alamadai] P[ichumani] 1931. A Descriptive Catalogue of the Sanskrit Manuscripts in the Tanjore Mahārāja Serfoji’s Sarasvatī Mahāl Library, Tanjore, I–XIX. Vol. 11: Vaiśeùika, Nyāya, Sāïkhya and Yoga. Srirangam. Cat. Trivandrum = BHASKARAN, T. 1984. Alphabetical Index of Sanskrit Manuscripts in the Oriental Research Institiute and Manuscript Library, Trivandrum. Vol. 3: ya to ùa. (Trivandrum Sanskrit Series, 254.) Tri- vandrum. (c) Secondary Sources BRONKHORST, Johannes 1985. Patañjali and the Yoga sūtras. Studien zur Indologie und Iranistik 10: 191–212. BÜHLER, Georg 1896. Indische Palaeographie von circa 350 a. Chr. – circa 1300 p. Chr. (Grundriss der Indo-Arischen Philologie und Altertums- kunde, 1.11.) Strassburg. EHLERS, Gerhard 2000. Auf dem Weg zu einer neuen Edition des Jaiminīya- Brāhmaõa. Berliner Indologische Studien 13/14: 1–28. HACKER, Paul 1968–69. Śaïkara der Yogin und Śaïkara der Advaitin. Einige Beobachtungen. In: G[erhard] Oberhammer (ed.), Beiträge zur Geistes- geschichte Indiens. Festschrift für Erich Frauwallner aus Anlass seines On the Written Transmission of the Pātañjalayogaśāstra 101 70. Geburtstages (= WZKSO 12–13): 119–148. Wien. (Reprint in Hacker 1978: 213–242.) –––– 1978. Kleine Schriften. Hrsg. von Lambert Schmithausen. (Glasenapp- Stiftung, 15.) Wiesbaden. HALBFASS, Wilhelm 1991. Tradition and Reflection. Explorations in Indian Thought. New York. LARSON, Gerald James & Ram Shankar BHATTACHARYA (eds.) 1987. Sāü- khya. A Dualist Tradition in Indian Philosophy. (Encyclopedia of Indian Philosophies, 4.) Delhi. MAYEDA, Sengaku 1968–69. The Advaita theory of perception. In: G[erhard] Oberhammer (ed.), Beiträge zur Geistesgeschichte Indiens. Festschrift für Erich Frauwallner aus Anlass seines 70. Geburtstages (= WZKSO 12–13): 221–239. Wien. NAKAMURA, Hajime 1980–81. Śaükara’s Vivaraõa on the Yogasūtra-Bhāùya. The Adyar Library Bulletin 44–45: 475–485. OBERHAMMER, Gerhard 1977. Strukturen yogischer Meditation. Untersuchun- gen zur Spiritualität des Yoga. (Österreichische Akademie der Wissen- schaften, philosophisch-historische Klasse, Sitzungsberichte, 322 = Ver- öffentlichungen der Kommission für Sprachen und Kulturen Südasiens, 13.) Wien. RAU, Wilhelm (ed.) 1977. Bhartharis Vākypadīya [1]. Die Mūlakārikās nach den Handschriften herausgegeben und mit einem Pāda-Index versehen. (Abhandlungen für die Kunde des Morgenlandes, 42.4.) Wiesbaden. –––– (ed.) 1991. Bhartharis Vākypadīya 2. Text der Palmblatthandschrift Trivandrum S.N. 532 (= A). (Akademie der Wissenschaften und der Literatur: Abhandlung der Geistes- und Sozialwissenschaftlichen Klasse, 7.) Stuttgart. SCHMITHAUSEN, Lambert 1968–69. Zur advaitischen Theorie der Objekter- kenntnis. In: G[erhard] Oberhammer (ed.), Beiträge zur Geistes- geschichte Indiens. Festschrift für Erich Frauwallner aus Anlass seines 70. Geburtstages (= WZKSO 12–13): 329–360. Wien. SRI RAMA SASTRI, Polakam & S. R. KRISHNAMURTHI SASTRI (eds.) 1952. Pāt[a]ñjala-Yogasūtra-Bhāùya-Vivaraõa of Śaïkara-Bhagavatpāda. Critically edited with introduction. (Madras Government Oriental Se- ries, 94.) Madras. SRINIVASAN, Srinivasa Ayya (ed.) 1967. Vācaspatimiśras Tattvakaumudī. Ein Beitrag zur Textkritik bei kontaminierter Überlieferung. (Alt- und Neu- Indische Studien, 12.) Hamburg. STEINKELLNER, Ernst 1981. Philological remarks on Śākyamati’s Pramāõa- vārttikañīkā. In: Klaus Bruhn & Albrecht Wezler (eds.), Studien zum 102 PHILIPP A. MAAS Jainismus und Buddhismus. Gedenkschrift für Ludwig Alsdorf (Alt- und Neuindische Studien, 23): 283–295. Wiesbaden. VEDAVRATA (ed.) 1984. Śrī-Pātañjala-Yogasūtrabhāùyam. <śrīgovinda- bhagavatpūjyapāda-śiùya-paramahaüsa-parivrājakācarya-śrīśaïkara- bhagavat-kta-vivaraõānusāri.> Hindī-vivti-sahitaü. [Vol. 1: Samādhi- pāda.] Vyākhyātā: Saccidānanda Yogī Sarasvatī. VETTER, Tilmann 1979. Studien zur Lehre und Entwicklung Śaïkaras. (Publica- tions of the De Nobili Research Library, 6.) Wien. WEST, Martin L[itchfield] 1973. Textual Criticism and Editorial Technique Applicable to Greek and Latin Texts. Stuttgart. WEZLER, Albrecht 1983. Philological observations on the so-called Pātañjala- yogasūtrabhāùyavivaraõa (Studies in the Pātañjalayogaśāstravivaraõa I). Indo-Iranian Journal 25: 17–40. WOODS, James Haughton (trans.) 1914. The Yoga-System of Patañjali. Or the Ancient Hindu Doctrine of Concentration of Mind, Embracing the Mnemonic Rules, Called Yoga-Sūtras, of Patañjali and the Comment, Called Yoga-Bhāshya, Attributed to Veda-Vyāsa, and the Explanation, called Tattva-Vaiçāradī, of Vāchaspati-Miçra. (Harvard Oriental Series, 17.) Cambridge, Mass. (Reprint: Delhi 1992.)