Note On Mahabhasya II 366 26 Gunasamdravo Dravyam

Author(s): A Wezler
Publisher: A Wezler
Catalog link: https://jainqq.org/explore/269467/1

Page #1

A NOTE ON MAHĀBHĀSYA II 366.26 :
GUNASAMDRĀVO DRAVYAM
Studies on Mallavādin’s Dvādaśāranayacakra II*
Albrecht WEZLER
. With reference to a particular period in the historical development of the Vaiseșika (as tentatively reconstructed by him), Frauwallner remarks that "just at this time the science of grammar began to penetrate philosophy with its ideas, while the grammarians on their part adopted philosophical ideas, made use of them for their own purposes and developed them further”. Indeed, to say it more generally, the main problem one has to tackle when dealing with 'grammar' within the framework of the history of Indian philosophy is - to use an expression as neutral as possible - the influence on each other of grammar 'and philosophy. Though some work has already been done in this field, last but not least by Frauwallner himself, fellow scholars will agree without hesitation that we are still far from a comprehensive knowledge of the interrelation between these two spheres.
The main source, though of course not the only one, Frauwallner had in view was, no doubt, Patañjali's Mahābhāṣya. A good survey of the relevant studies of this text so far undertaken is given by Cardona, but significantly enough this subsection bears the heading " Discussions of philosophical import" and starts with the simple description that "the Mahā-Bhāsya contains discussions on the threshold of grammar and philosophy ”. Nevertheless, among the studies subsequently mentioned or surveyed by Cardona there are some in which a serious attempt is made to deal also with the historical problems involved. The most perceptive, but at the same time problematical studies of this type are perhaps two articles of Frauwallner's, viz. one devoted to the

Page #2

A. WEZLER
6)
penetration of language theory into the Indian philosophical systems and the other to language theory and philosophy in Patanjali's Mahābhäşya. Especially the latter article has provoked vehement protest from Western Paniniyas', both as regards Frauwallner's understanding of Patanjali's discussion as also his very approach. As indicated by me on an earlier occasion, I, too, should like to dissociate myself from the conclusions arrived at by Frauwallner. In particular I completely fail to understand why such investigations should be inhibited by prejudices about Patanjali's alleged lack of interest in philosophical questions, etc. Instead of being carried away by what looks like an obsession to evaluate, one should rather, perhaps even with sincere gratitude for the information given by him, concentrate on an unbiased and thorough analysis, of the various cases where Patanjali mentions philosophical views or actually takes recourse to them.
2
Yet, one observation of Frauwallner's, though meant as a reproach, is indisputably of no little importance, viz. that Patañjali usually remains silent about the sources drawn upon by him; for, this procedure of Patañjali considerably adds to the difficulties one is faced with when dealing with philosophical tenets in the Mahabhäşya. These difficulties are first of all posed by the scarcity of source material at our disposal which would allow us to gain a clear picture of the stage of development of philosophy in Patanjali's own age. Hampered as we are by this lack of information we would, of course, like Patanjali to have given in each and every case at least a precise and detailed account of his source(s), the title(s) of the work(s), the name(s) of the author(s), etc. That he does not meet these expectations, and, I think, quite understandably not, is no justification for censuring him. Yet, there are even more elementary difficulties one has to grapple with: Often it is by no means easy to recognize a reference to a philosophical view as such, and in this regard the commentators do not always render assistance, or cannot at any rate be relied upon without hesitation. Hence any additional information is highly welcome, especially if it consists in a statement to the effect that a particular passage in the Mahābhāṣya is "taken over" from a certain system of philosophy. It is evident that any such contention is of enormous heu
7)

Page #3

A NOTE ON MAHABHÄŞYA II 366.26
3
ristical importance though it goes without saying that it calls for a careful and critical examination.
Now, there is one recently published text which not only almost abounds in material for a study of the reception of the Mahābhāṣya- and grammar in general in later philosophical literature, but also contains information of the type just mentioned, viz. Mallavādin's Dvādaśāranayacakra (NC) together with Simhasūri's Nyāyāgamānusāriņi (=NAA) commentary on it. This is commendably pointed out by the editor himself, Muni Jambūvijaya, at the end of the subsection "Nayacakre carcitā dārśanikä vädäh" of his "Prākkathanam ".
2
10)
Of the two instances to which the Muni draws our attention it is the second only which I should like to study in the present note. What he says is: "Patanjalamahabhaṣye [5. 1. 119] varṇitam guṇasamdravo dravyam' iti matam api Samkhyad evahytam pratiyate, dṛśyatām Nayacakravṛttau pr" 268 pam 11, pr° 303 ityādi".
2.1.
The first of the passages runs thus (268. 4-13): etad adhuna pariksyate atha atha katham ityādi yavad viyad abhyupagamyata iti/yady anekatmakaikakäraṇatvam işyate evam ekakaranatvapratisedhanantaram puruşädyekakaranatvapratisedhahitasamskäratirodhanakalam apy apratikṣya tvaya katham sabdaika guna pravṛtti viyad abhyupagamyate? abhyupagamyatām tavad rüpädisṛṣṭau vyavahäränupätinām eṣām loke dṛṣṭānām dvitryadyanekasparśarasagandhaguṇānām anekaikatvät tadätmakavayvadisystir) astu nama 1 tvanmatena saha ghatamanam() idam tu na yujyate() sabdena ekaguna pravṛttir ittham bhütenasya) tac chabdaikagunapravṛtti viyat (,) gana guna sankhyane' sabdaikasankhyānapravṛtti ekasmat käranädb havat tvanmatavirodhät, nänekätmakaikasmac chabdasparśadidvitricatuḥpañcagunad bhavad väyvädivat tvanmatävirodhat
11)
12)
13)
15)
14),

Page #4
A. WEZLER prayogas catra- na pravarteta i v a m, sabdaikagunakasam tan na bhaved ity arthah, as a n d rute h, gunasandravo dravyam [Padeg Mao Bhadeg 5. 1. 119] iti laksanabhavat, bahunam hi gunanam ekibhavanam aikyagamanam sandrutih, tadabhavo 'sandrutih, tato 'sandruter na pravarteta, pur u s a v a d van d h ya putra v a d vety etad anistapadanam iti /. The criticism stated in this passage is clearly directed against a specific element of the Sankhya theory of evolution, viz. against the conception of the origination of the 'great elements' (mahabhutas) out of the tanmatras: to be more precise, it is directed against this conception in the form it was given when it had to be modified in order to take into account the so-called accumulation theory which had meanwhile been developed in other circles. According to this modified form already the tanmatras possess in ascending number the characteristic qualities of the 'great elements': that is to say, the corresponding process of evolution can be described with the Yuktidipika as follows: sabdagunac chabdatanmatrad akasam ekagunam / sabdasparsagunat sparsatanmatrad dviguno vayuh / sabdasparsarupagunad rupatanmatrat trigu nam tejah / etc. Thus the basic view held by the adherents of Samkhya is, according to Simhasuri, that of an anekatmakaikakaranatva, i.e. that "[each of the 'great elements', too] has a single cause (viz. the corresponding tanmatra), '[yet a cause] which is in itself of a manifold nature"; and against this view it is stated polemically that it is incompatible with what is taught by the Samkhya about the origination of the first great element', viz. ether ; for, ether is said to be a manifestation of the sabdatanmatra, i. e. of a cause which is, however, by no means anekatmaka or an anekatmakaika, as it possesses the quality of sound only. Thus the criticism consists essentially in pointing out a particular internal contradiction of this Samkhya doctrine. 18) On the basis of the passage of the Vstti as quoted above, Muni Jambuvijaya has reconstructed the corresponding portion of the NC, I think, quite

Page #5

A NOTE ON MAHABHAṢYA II 366.26
5
convincingly, thus (268. 1-2):
20)
atha katham ekakaranatvapratiṣedhanantaram sabdaikaguṇapravṛtti viyad abhyupagamyate? na pravartetaivam, asandruteh, puruṣavad vandhydputravad vä/.
22)
23)
The final sentence has correctly been recognized by Simhasüri to be a prayoga, i.e. a proposition which forms a full-fledged inference (anumana). As there is no reason for distrusting Simhasuri's explanation of the hetu, viz. asandruteḥ, the assumption is fully justified that what Mallavädin on his part has in view here is in fact the etymologizing-definition of dravya found in Mahābhāṣya II 366.26 as it is quoted by the commentator. The gist of Mallavadin's counterargument is hence that ether cannot originate in the manner asserted by the Samkhyas because it does not correspond to their definition of dravya, i.e. because it is not a dravya or rather because its cause, the sabdaguna śabdatanmatra, is not a dravya just like the soul or the son of a barren women. The argument as a whole, the context in which it is put forward and the use of puruşa as one of the two examples-all this warrants the conclusion that this definition of dravya goes back, at least in substance, to Samkhya itself, i.e. that Mallavädin uses here basic elements of the doctrine of his opponents themselves in order to refute them. But why, so it might be objected, should Mallavädin refer to the Mahabhāṣya for a definition of dravya which belongs to Samkhya? Should one not rather expect him to have drawn directly upon a Sämkhya source? Not necessarily, and, to be sure, not only if he did not know (any longer) the original Samkhya source, but also if the definition was formulated in a particular manner which was of importance to him (for his critical aims).
24)
It need hardly be explicitly stated that it is the second assumption which has clearly to be given preference: The definition as found in the Mahābhāṣya is characterized by being essentially an etymology-as was noted already above and will be demonstrated below (p. 18)-and Mallavädin would not have been an Indian had he not taken for this very reason particular delight in it. In view of the methodical significance it should further be emphasized that the testimony of Mallavädin and Simhasüri stands by itself, i.e. that it need.

Page #6

6
A. WEZLER
not necessarily be confirmed by similar statements in other sources: There can hardly be any doubt that both of them were firmly convinced that in terms of its doctrinal provenance the definition gunasamdravo dravyam as found in the Mahabhäṣya belongs to Sämkhya.
Space being limited, I cannot deal in extenso with Simhasûri's explanation. Yet, there are two points I definitely should not pass over in silence. The first is his paraphrase of sandrutiḥ, viz. bahūnām guṇānām ekibhavanam aikyagamanam, according to which this term is semantically equivalent to "the becoming one of many qualities/properties". And in addition it has to be noted that the term guna is here evidently not used in its narrower sense, specific to Samkhya terminology, viz. "constituent of primary matter" (although it is open to such an interpretation also as is shown by the passage in the NC following immediately upon that quoted above), but in that of "quality/property" of a cause, whereas dravya is this cause.
25)
26)
27)
2.2. The second passage pointed out by the editor is found almost right in the middle of the third Ara; that is to say, with it we jump as it were into the controversy between the Isvaravadin and the upholder of Samkhya about the following definitions of the three "constituents" of Samkhya-in which, however, the terms sukha, dukkha and moha are used instead of sattva, rajas and tamas ([NAA] 298. 7 ff.): sukham laghv apravṛttisilam prakāśakam drstam, ...duḥkham calam aprakāśakam pravṛttisilam dṛṣṭam, ...moho gurur aprakāśako dṛṣṭaḥ. When the opponent argues that these three are not really different from each other, the adherent of Samkhya tries to safeguard his theory against this objection by stating (300. 1): näpythagbhütată/ sukham mohad guror anyat, laghutvāt, lohapindad iva arkatalaḥ, etc. What is obviously intended by this counterargument is, and I think on the whole quite correctly, made clear by Muni Jambūvijaya in a footnote which reads thus:
28)
sukham mohad guror anyat, laghutvät, lokapindad ivärkatülah, duḥkhad anyat pravṛttisilät, apravṛttisilatvät, ākāśad iva vayuḥ/ duḥkham mohad anyad acalat, calatvät, parvatäd iva ghaṭaḥ/mohah sukhaduḥkhabhyam agurubhyam anyaḥ, gurutvät, prthivya iva agniḥ/ duhḥkhamohau sukhat
29)

Page #7

A NOTE ON MAHĀBHĀṢYA II 366.26
prakasakád anyau, aprakāśakatvāt, pradipad iva ghaṭaḥ/.
Among the various reasons advanced by the opponent against these inferences there are also the following (301.1-3):
anavadhṛtalaghvädidharmatayam itaratmakam apity atathataivaasajatiyalakṣaṇavyävyttärthaviṣayatāyām asiddham pakṣadharmatvam asmän prati bhavantam ca/.
30)
31)
32)
"In case the properties 'light', etc. [of sukha etc.] are not restricted [to sukha, etc.] exclusively (i.e. if what is meant by the logical reasons adduced by you is not laghv eva sukham na guru na calam vã, etc.), [then sukha, etc.] has also the nature of the other [gunas, i.e. it is also cala and guru]; therefore (i.e. because of the fact that sukha, etc. are hence each of a threefold nature) they are in reality (eva) not so (i.e. are not different from each other). If [these properties are, on the other hand, exclusively restricted to sukha, etc. respectively, and if the logical reasons consequently] refer to an object to which the characteristics of everything different [from it] do not apply, [then] the fact that they are a property of the subject [of the inference] is by no means established for us and for you". In pursuing this argument further the opponent then demonstrates that concepts like 'light', etc. are always relative, and finally states by way of summarizing (302. 4-5): evam evapeksikalohapinḍarkatülagurutvalaghutve 'navasthitaikatve/.
33)
34)
"In this very manner the heaviness of a small iron ball and the lightness of the coma of the seeds of the mudar is not something absolutely determined [once and for ever] as both of them are dependent [on the quantity/weight in each case]"..
35)
To these objections the adherent of Samkhya replies (303. 1-3):
atha mamatra kim? dravyam caitad evam laghugurutväpekṣayā tad eva drstam, na guṇaḥ/ sattvädayo gunds to ime laghutvädilakṣaṇā mayocyante, dravyata teṣam sandrave ekatvapattau bhavati/.
"[If you prefer to consider lightness, etc. to be not absolutely determined, then you are right in stating that it is not at all an established fact for you that sattva and nothing else is light, etc.; but] what has this to do with me? [For, I, on my part, am of the opinion that lightness, etc. are
36)
7

Page #8

A. WEZLER
specific properties of sattva, etc., respectively]. And (i.e. yet) [your own position is ultimately wrong; for] what you refer to (etad) is [in reality] a dravya [and it is this alone) which is experienced in everyday life in this manner as one and the same with regard to lightness and heaviness, and not the guņas. Yet it is these gunas sattva, etc. [and not any dravya] of which I maintain that they are characterized by lightness, etc. The fact of being a dravya is given when these (gunas) attain unity in that they come together [and they are different from each other]”.
At the end of his explanation of the last sentence of this latter passage of the NC Simhasūri again quotes the Mahābhāşya, i.e. adds the remark (303. 16 f.): gunasandrāvo dravyam iti laksaņāt. And again there is no room for any doubt that Mallavādin, too, in fact had in view this definition and regarded it as an authentic element of the Sāmkhya doctrine, especially since here it forms part of an argument of an adherent of Samkhya himself. In addition what deserves our attention are the facts
1. that the term sandrāva is explained in the NC itself by being
juxtaposed with ekatvāpatti, and 2. that the term guna is made clear in a similar manner by letting
sattvādayo precede gunās (tv ime) in the third sentence. But what about guņāḥ at the very end of the second sentence? In this case. Simhasūri says (303. 14 f.): ......gurulaghvādayo guņāḥ sattvarajastamāmsi ca......, i.e. he takes it to refer to the properties at issue in this discussion as well as to the constituents of primary matter. Let us hence make a mental note of the fact that according to him this term as used in the definition gunasandrāvo dravyam may refer equally to the constituents of primary matter and to their specific properties and to the qualities of the tanmātras, etc. Of equal importance is the observation that the adherent of Sāmkhya not only distinguishes clearly between guna and dravya, but also takes the latter to become manifest only in the case of a sandrāva of the former.
40)
41)
2.3. Should there still be any doubt about the correctness of my assertion

Page #9

A NOTE ON MAHĀBHĀSYA II 366.26 that both Mallavădin and Simhasüri were quite sure that the definition gunasandrāvo dravyam belongs to Sāmkhya, it can easily be dispelled, at least as regards the commentator. For, in his Vștti on NC 15. 1 f. he statesto wit, in order to make clear a quite different conception of dravya in its distinctiveness-(15. 20 f.): na tu yathā gunasandrāvo dravyam [Pātañjalamao 5. 1. 119] kriyāvad gunavat samavāyikāraṇam iti dravayalaksanam [Vaio Sū° 1. 1. 15] iti vā /. Considering that according to NĀA 11. 23 the list of philosophical systems--which the “ordinary man" (laukika) wants to reveal as useless-starts with Sāmkhya, the only conclusion one can arrive at is that Simhasūri's first definition of the two diverging ones equally belongs to Sāmkhya
This part of the present study may therefore be concluded by repeating once more
1. that both in the NC as well as in the NĀA there is clear evidence
that the definition gunasandrāvo dravyam was regarded by the two authors to be an integral--and obviously also essential-element of
Sāmkhya, and 2. that Mallavādin and Simhasūri's testimonies constitute information in
their own right.
In a note referring to NC 15. 20 Muni Jambūvijaya not only quotes in extenso from the Mahābhāşya (=M.] on Pāṇ. 5. 1. 119, but he also adds the remark:
Pātañjalamahābhāsyasya Sāmkhyamatānusāritvam 'striyām' [Pā° 4. 1. 3]
iti sūtre Pātañjalamahābhāşyasya Uddyotāt Pradīpāc cāvagantavyam/.
Indeed, if one looks into Nāgeśa's Uddyota on M. II 366. 23 ff., one cannot fail to notice that in connection with a critical remark directed against Kaiyața, Nāgesa himself refers to the discussion on Pāņ. 4. 1. 3 (see below p. 12).

Page #10

A. WEZLER
3.1. "The Mahā-Bhāṣya on rule 4. 1. 3", thus Cardona aptly summarizes its contents, "takes up the question of gender. The term linga, used in the meaning 'gender', denotes, in normal Sanskrit, a mark or characteristic. If the term is understood in this sense in grammar, then a linga is a characteristic of males, females, and things which are neither. A female (stri) would then be characterized by breasts and hair, a male (puruşa) by his body. hair, others by neither. This conception of linga does not work in grammar, it is noted, so that another concept is introduced. Any thing is characterized by different states of constituent elements or properties (guna) and these states constitute the genders of things."
46)
47)
48)
According to the värttika samstyänaprasavau lingam astheyau svakṛtantataḥ on Pān. 4. 1. 3 what has to be regarded as the basis of gender is the act of coagulating, on the one hand, and the act of procreating, on the other, the former being equated to stri and the latter to pums. After having explained this värttika, Patañjali pursues the question further, viz. thus (II 198, 4-5):
kasya punaḥ styänam stri pravṛttir va pumān/ guṇānām/keṣām/sabdasparśaraparasagandhānām /.
Now it is to the answer to the first question, viz. guṇānām, that Kaiyața gives the following explanation (which Muni Jambūvijaya must have had in view) (IV 23 a 12-15):
49)
10
45)
sattvarajastamāmsi guṇāḥ, tatpariṇāmarupāś ca tadātmakā eva sabdadayaḥ pañca guṇaḥ/ tatsanghatarupam ca ghaṭādi, na tu tadvyatiriktam avayavidravyam astiti samkhyanam siddhantaḥ //.
It is hence explicitly stated by Kaiyața that here Patanjali takes recourse to a central conception of Samkhya according to which any thing like a pot, etc. is considered to be nothing but an aggregate of the qualities 'sound', etc., the existence of a substance (lit. "a material whole") as different from them being denied,
50)
Yet, the definition of dravya as consisting of an aggregate of gunas is not only attested in the Pradipa, but is also used by Patañjali himself in the very same discussion, viz. in a passage (following upon Kätyäyana's värtt. 7) to which attention has already been drawn by Seyfort Ruegg and which runs
51)

Page #11

52)
53)
A NOTE ON MAHĀBHĀSYA II 366.26 thus (II 200. 12-15):
kasya tävad bhavān ekam gunam nyāyyam manyate stritvam nāma / dravyasya / dravye ca bhavataḥ kaḥ sampratyayah / yadi tāvad gunasamudāyo dravyam / kā gatir ya ete bhāvāḥ kydabhihitās taddhitābhihitāś ca
cikirşā goteti/. I must confess that I fail to understand why Seyfort Ruegg with reference to the phrase guṇasamudāyo dravyam speaks of “ cette définition du substantiť”; but what he says subsequently, viz. that it “est, il est vrai, introduite comme l'opinion d'un autre ", is, no doubt, correct. Yet, I don't think that it is of particular importance, at least for the problem at issue here, to which of the participants of the discussion this or that statement belongs. For, what this passage is evidently about is the question of the view of an "individual material thing" (dravya) one of the participants holds, and the critical remark that if this participant accepts the definition guṇasamudāyo dravyam, there arise certain difficulties—and they are conveniently explained by Nāgesa. And again both Kaiyața and Nāgeśa are to all appearances unanimously of the opinion that this is a definition of the adherents of Sāmkhya. For Kaiyata e.g. explains gunasamudāyaḥ by stating (IV 30 a 8-9): rūpādisamniveśamātram ity arthah.
There can indeed be hardly any doubt that gunasamudāyo dravyam is but another formulation of gunasamdrāvo dravyam, or vice versa, and that the expressions samudāya and sandrāva are hence considered here to be practically synonymous. Yet, it should not be overlooked that the latter expression is used in the discussion of Pān. 5. 1. 119 in a particular context, i.e. that the definition is introduced there by the remark (M. II 366. 25) anvartham khalv api nirvacanam, whereas in M. 200. 11 ff. the question of the derivation of the term dravya-and of defining dravya by pointing out its etymology-is not at issue. The assumption which therefore suggests itself is that the 'normal'expression is samudāya; and this is strikingly confirmed by the fact that Patañjali in all other instances (viz. I 411. 15 and II 120. 11) uses the term gunasamudāya, and not gunasandrāva.
The result of the foregoing observations and considerations is that both

Page #12

A. WEZLER discussions in the M., i.e. that of Pāṇ. 4. 1. 3 and that of Pāṇ. 5. 1. 119, are thematically connected among other things by the common reference to a definition of dravya which according to the commentators belongs to Sāmkhya.
58)
3.2. Returning now to the M. on Pāņ. 5. 1. 119 let us first take a look at Matilal's analysis of this discussion. Matilal calls attention among other things to the fact that “Nāgeśa is critical of Kaiyata's explanation of the Mahābhāsya passage under Pāṇini's rule 5. 1. 119 and says that Kaiyața uses the Vaiseșika notion of substance to explain Patañjali: Nāgesa, on the other hand, thinks that the Sāmkhya-Yoga notion of substance is much closer to the grammarians' notion of substance".
It is evidently the following passage of the Uddyota Matilal has in view here (IV 299 a 15-17):
Kaiyatas tu vaiseșikādinayānusāreņa bhāșyam vyācakṣāṇaḥ katham 'striyām' [Pān. 4. 1. 3] iti sūtrasthabhāşyeņa na virudhyata iti cintyam ; katham ca 'kim punar dravyam, ke guņāņ' [M. II 366. 14, i. e. on Pān.
5. 1. 119] iti praśnasya na nirdalateti ca cintyam /. Yet, this forms part of Nāgeśa's explanation of M. II 366. 23-25 (quoted below p. 17), that is to say, of a section which is taken by Matilal to be “ Patañjali's second attempt to define substance", whereas the immediately following passage (II 366. 25-26), viz. anvartham khalv api nirvacanam / gunasamdrāvo dravyam /, is regarded by him as "the third attempt of Patañjali to define substance”, and, to be sure, it is this latter attempt to which Matilal refers when reporting Nāgesa's critical rebuke of Kaiyața and when remarking that “there is a striking similarity between Patañjali's definition of substance and the Sāmkhya-Yoga idea of substance". Thus one is rather confused and cannot hence but deem it necessary to reconsider the relevant passages of the original texts. This experiment' yields, so it seems to me, among others the following results:
1. What has prompted Nāgesa to censure Kaiyaţa was evidently the
following passage in the latter's comment on M. II 366. 23-25 (IV 299 a 3 ff.):

Page #13

13
62)
A NOTE ON MAHĀBHĀŞYA II 366.26 ......tasmād asti rūpāśrayo dravyam / etac ca pākajarūpādyutpattau ye dravyavināśam necchanti tanmatena dravyapratyabhijñāśrayeņoktam / ye tu pākajarūpādyutpattau pūrvadravyavināśam apūrvadravyārambham căbhyupagacchanti tanmatena dravyabhede 'pi jāter ekatvāj jātipratyabhi
jñāśrayeņoktam / etc. 2. One cannot but emphatically agree with Nāgesa on this point:
Kaiyața is, as we shall see later, definitely on the wrong track. 3. It does not suffice to state that Nāgeśa differs from Kaiyața and simply to report his own opinion. For Nāgesa does not at all confine himself to merely contending that Kaiyaţa is wrong, but he adduces two different, though ultimately related, reasons which-once more-testify not only to his critical acumen, but also and above all to his remarkable capacities in terms of philological method and its application in interpreting Patañjali's work. The first argument consists in pointing out · that Kaiyața's interpretation of M. II 366. 23-25 contradicts what is said by Patañjali on Pāṇ. 4. 1. 3, viz. in M. II 198. 4 (see above p. 10) and II 200. 13 f. (see above p. 11); and in this case, too, Nāgeśa is quite right; for the agreement between the passages in these two discussions in terms of their doctrinal affinity is, as we have already seen, as plain as it can be, so that it can easily be recognized even without the help of Nāgesa who was on his part well aware of this relationship. His second argument shows, if I am not mistaken with regard to the meaning of nirdalat, and the implications of this argument of his, that Nāgesa starts from the assumption that the two questions of M. II 366. 14 make sense only if they are asked by a participant of the discussion who is of the opinion that on the basis of the 'Sāmkhya' definition of dravya (as taught under Pān. 4. 1. 3) it is not possible to account for what is said in vartt. 5 on Pāņ. 5. 1. 119. For according to Kātyāyana's statement (M. II 366. 10: siddham tu yasya gunasya bhāvād dravye sabdaniveśas tadabhidhāne tvatalau) the suffixes -tva and -tā are added after a base which is a dravya-vācaka to denote a guna; and the 'Sāmkhya' definition of dravya, so one might think, implies that there is

Page #14

A. WEZLER no difference between the samūha (=dravya) and the samūhin (=guņa), as explained by Nāgesa. That is to say, Nāgesa starts out again from the idea that the discussion on Pāṇ. 5. 1. 119 can be properly understood only if it is seen against the background of the discussion on 4. 1. 3; but this second argument is of a slightly different nature in that it amounts to the assumption that the knowledge of the M. on Pāṇ. 4. 1. 3 is presupposed in the M. on Pāņ. 5. 1. 119, and it is precisely for this reason that Nāgeśa can in this case not meet with our full approval as the two questions of M. II 266. 14 can easily be accounted for by simply assuming that they are provoked by vārtt. 5 on its own. In spite of disagreeing in this respect with Nāgesa, I think he can definitely be followed when he starts from the assumption, if this is what is implied by him, that the passages in the two discussions are closely related to
each other as regards their doctrinal basis. There is one more point which I should like to make in the present discussion of Matilal's analysis of the M. on Pāṇ. 5. 1. 119, viz. with regard to the question whether we have really to distinguish between a "second" and a "third attempt of Patañjali to define substance". But as this question cannot be discussed independently of the problem which I want to study in the next paragraph, I shall deal with both of them together.
What I was just now referring to is a question one cannot but ask oneself, viz. if there is perhaps any internal evidence in the M. on Pāņ. 5. 1. 119 itself which would confirm the contention of Mallavādin and Simhasūri, on the one hand, and Nāgeśa, on the other, that the definition gunasamdrāvo /osamudāyo dravyam belongs to Sāmkhya.
4.1. That in fact two different “attempts to define substance” have to be distinguished, at least formally, can hardly be disputed; for the second is introduced by an athavā (M. II 200. 23). The passage with reference to

Page #15

67
A NOTE ON MAHĀBHĀŞYA II 366.26
15 which Matilal speaks of “the third attempt", however, runs thus (M. II 200. 25-26):
anvartham khalv api nirvacanam / guṇasamdrāvo dravyam /. Therefore, what has to be examined first of all, is the question if the particles khalv api are similarly indicative of the fact that still another attempt is made by Patañjali or not. The meaning assigned to khalv api in the small Petrograd Dictionary is “nun auch ", the remark being added that it is used “ beim Uebergange zu etwas noch zu Besprechendem”. This observation, though based on another text, proves quite useful in the case of the M., too.
For, to start with the very first instance where it occurs, what else should be its function in M. I 1. 4: vaidikāḥ khalv api/than to indicate that the person who had answered the question keşām śabdānām / by saying laukikānām vaidikānām ca (M. I 1. 3) is now turning to the explanation of the term which has still to be “talked about ”, viz. vaidikāḥ [śabdāḥ] ?. And the next instances found on the same page (lines 16 and 18) are of exactly the same nature--and this holds good also for those referred to by Böhtlingk. Yet, it is by no means easy to exclude the possibility that a slight contrast is also expressed in these cases apart from calling the attention of the person who had asked the aforementioned question or of the other participant of the discussion) to the other or next topic.
Įt is therefore advisable to examine instead only those instances where khalv api is used in sentences which are, except for their context, identical with that of M. II 200. 25. And there are, if I am not mistaken, just two of them.
The first is found in the discussion on Pāṇ. 5. 2. 29 and 30, viz. M. I 206. 21-25:
kuto nu khalv etad aca ete gunās tatsāmīpyāt tu vyañjanam api tadgunam upalabhyata iti na punar vyañjanasyaite guņāḥ syus tatsāmipyāt tv aj api tadguna upalabhyata iti / antareņāpi vyañjanam aca evaite gunā laksyante na punar antareņācam vyañjanasyoccāraṇam api bhavati/a n varth am k halv a pi nirvacan a m / svayam rājante svarā anvag bhavati

Page #16

A. WEZLER
.68).
vyañjanam iti //. “Yet, on account of which [reason did you right now assert]: ‘These (i.e. the udātta, anudātta and svarita) are qualities of a vowel, but due to its proximity to it a consonant, too, is perceived as being qualified by them', not, however, (the opposite, viz.] that they are qualities of a consonant, but that due to its proximity to it a vowel, too, is perceived as being qualified by them? [The reason is that] even without (i.e. in the absence of) a consonant they are recognized as qualities of a vowel and nothing else, but that, on the other hand, without a vowel a consonant cannot even be pronounced [so that the question whether it is qualified by an accent and, if so, why, does not at all arise]. And the etymology [of the terms svara and vyañjana) indeed also corresponds to [this their] meaning i.e. indicates this important difference between them); [for] 'vowels' [are sounds which] 'shine by themselves / on their own' [and] a 'consonant' [is a sound which] 'follows' i.e. becomes qualified by the accent only afterwards)".
In this case it is hence as plain as it can be that the sentence anvartham khalv api nirvacanam is merely meant to corroborate what is already stated in the preceding sentence, viz. that a vowel by its very nature is a completely independent type of sound, whereas a consonant is in contradistinction to it dependent on a vowel even for being pronounced, not to speak of its being. qualified by an accent: The topic to which in the last sentence attention is called is 'new' only in the sense that what has been stated in the preceding sentence is now further supported by pointing to the etymology of the terms svara and vyañjana themselves which allegedly-confirms the correctness of this phonetic statement.
But what about the second instance? It is found in the M. under Pāṇini's rule 8. 2. 48, or, to be more precise, in the discussion following upon vārtt. 3 in which it is stated that the root añj is in certain cases used añcatyarthe, "in the meaning of the root añc", viz. prakāśana. In concluding this part of the discussion Patañjali says (M. III 408. 24-409. 2):
tathāñjer vyañjanam vyañjanam ca prakāśanam / yat tat snehena madhurena ca jadikytānām indriyāņām svasminn ātmani vyavasthāpanam sa

Page #17

69)
70)
A NOTE ON MAHĀBHĀSYA II 366.26 rāgas tad vyañjanam / anvartham khalv api nirva cana
m / vyajyate 'neneti vyañjanam iti //. " In the same manner vyañjana is [a noun derived] from [the root] añj [used in the meaning of the root añc], and vyañjana means that by which [something] is illuminated / made perceptible'. When the sense-organs, made dull by grease and (i.e. or) [something] sweet, are (again) fixed on themselves, [then] this is a source of delight (i.e. a condiment), this is vyañjana (i.e. this is vyañjana as having among others the meaning "condiment', 'spices'). And the etymology [of vyañjana] indeed also corresponds to [this its] meaning ; [for] condiment' is that by which [the flavour of food) is made manifest'". • It is hardly necessary to state explicitly that regarding the function of khalv api this passage yields exactly the same result as that examined first. Therefore it is not simply legitimate, but imperative to start from the assumption that the sentence anvartham khalv api nirvacanam in M. II 200. 25 forms in a similar manner a statement which is made only in order to corroborate what has been said in the passage preceding it. It is hence practically out of the question that M. II 200. 25-26 should be Patañjali's "third attempt to define substance".
72)
Yet, in order to be able finally to convince oneself of its merely corroborative function, the immediately preceding passage has to be taken notice of in detail (M. II 200. 23-25):
athavā yasya guṇāntareşv api prādurbhavatsu tattvam na vihanyate tad dravyam / kim punas tattvam / tadbhāvas tattvam / tadyathā / āmalakādinām phalānām raktādayaḥ pītādayaś ca gunāḥ prādurbhavanty āmalakam
badaram ity eva bhavati/. This " second attempt to define substance"-and, to be sure, it is at the same time also the last one-has been summarized by Matilal as follows: "Substance is the unchanging state of the objects. Qualities emerge and disappear in order to make room for new qualities. What stays permanent amid all changes is what is called substance. It is the tattva 'essence of an entity.

Page #18

A. WEZLER
Thus, a mango fruit becomes green at one time and yellow and red at another time, but it still remains the same mango or, at least, we can call it *the same mango'. What remains unchanging is the mango-essence, the mango-substance".
Now, if the main, nay even the sole purpose of what is stated subsequently by Patañjali, viz. anvartham khalv api nirvacanam / gunasamdrāvo dravyam/, is to corroborate this second attempt to define substance-and the parallel passages do not leave us any other choice, then it has to be admitted that what Patañjali wanted to say by it is the following: The etymology of the word dravya itself, according to which a dravya is a gunasamdrāva, equa l1 y shows that 'substance' or rather an individual material object is that which "remains unchanging " while different gunas “emerge and disappear". But in what respect is this idea expressed by the etymologizing definition, too? The answer to this question is as simple as it is plausible: Because dravya is in view of the context defined to be always, i. e. at each and every point of time, nothing but a guna-samdrāva, no matter which qualities "come together" to constitute it at a particular point of time. The fact that the gunas change, or even permanently change, does not in the least affect the nature of the dravya as such, i.e. its being nothing but a samdrāva of guņas at every moment of its-finite-existence. To render this concept of dravya by calling it the "unchanging state of the objects” or to apply to it the--anyhow problematical-term “essence" as Matilal does, cannot pass for a really constructive attempt to determine this concept more precisely. According to Patañjali it is the tattva, and this term is in its turn explained by tadbhāva, in accordance with Pān. 5. 1. 119. But as it is the meaning and purport of this very sūtra which forms the subject of this discussion in the M., this paraphrase does not help much. Nevertheless, the context makes it possible to draw the conclusion that 'tattva' is, generally speaking, that which remains the same-which in the present case is likewise that by which an object can be recognized.

Page #19

19
A NOTE ON MAHĀBHĀSYA II 366.26 4.3. Now that it has been established that the sentence anvartham khalv api, etc. of M. II 200. 25-26 is but an additional argument meant to vindicate the preceding 'second attempt' to define the concept of dravya, attention can be focussed for a short while on the preceding passage (quoted above p. 17). What has to be noted first is that in reading it one cannot but recall M. I 7. 21-23 :
athavā nedam eva nityalaksanam dhruvam kūtastham avicāly anapāyopajanavikāry anutpatty avrddhy avyayayogi yat tan nityam iti/ tad api nityam yasmims tattvam na vihanyate / kim punas tattvam / tadbhāvas
tattvam / ākytāv api tattvam na vihanyate //. Which is, by the way, also quoted by Simhasūri (NĀA 21. 22 ff.); yet more important, at least for the question at issue here, is it to remember what has already been observed earlier (see above p. 12), viz. that it is with regard to this passage that Nāgesa (in reprimanding Kaiyața) emphasizes the “Sāmkhyamatānusāritvam ", to use Muni Jambūvijaya's expression.
Indeed, already the verb prādur-bhū points to this direction, though it can admittedly not be regarded as an absolutely unequivocal key word. But we are in this case not at all dependent upon any such indirect evidence or mere conjectures. For, as luck would have it, there are close parallels in Sāmkhya and. Yoga texts themselves that the question of the doctrinal provenance of Patañjali's 'second attempt' can be definitely settled. What I have in view is first of all a verse quoted twice in the Yuktidipikā, viz. 49. 10 f. and 75. 6 f., which runs thus:
jahad dharmântaram pūrvam upādatte yadā param / tattvād apracyuto
dharmi pariņāmaḥ sa ucyate 11. Clearly this is a definition of the term pariņāma, and not dravya, and moreover in it use is made of the terms dharma and dharmin which did certainly not originate in Sāmkhya itself, but were taken over from other schools of thought, probably the Vaiseșika. Nevertheless it is evident that in this verse the same idea is expressed as in the M. passage under discussion; and in spite of the 'modern' terms used the assumption is, no doubt, justified that the idea as
75)

Page #20

A. WEZLER such goes back to much earlier times in the history of Sāmkhya, i.e. that it is a genuine Sāmkhya idea, and an old one at that. It is attested also in the Yogabhāşya, viz. at the very end of what is said on YS 3. 13, but formulated thus :
avasthitasya dravyasya pūrvadharmanivrttau dharmāntarot pattih parinā
ma iti. Any doubts one might still have regarding the doctrinal agreement between the M. passage and the Sāmkhya and Yoga sources quoted just now, are finally dispelled when one takes into account what is said by the author of the Yuktidīpikā in explaining and defending the definition of pariņāma of the verse, viz. (75. 8-11):
yadā saktyantarānugrahāt pūrvadharmān tirobhāvya svarūpās apracyuto dharmi dharmāntareņāvirbhavati tad avasthānam asmākam pariņāma ity ucyate / äha-naitad abhidhānamātram dystāntam antareņa pratipadyamahe / tasmād yathā kim iti vaktavyam / ucyate-yathā palāśam palāśād apracyuta[m] nimittāntarasyātapāder anugrahāc chyāmatām tirobhāvya
pitatām vrajati tathedam drastavyam /. For, the drstānta given here is essentially the same as that found in the M. ; that in one case the leaves of a palāśa tree are chosen as example and in the other case fruits like a mango or a badara is of little importance. What really counts is that in both cases the concept of pariņāma is illustrated by referring to processes of outward change readily observed in nature.
I don't deem it necessary to adduce more material from Sāmkhya or Yoga texts; for, I think, that that already quoted is sufficient proof that the “second attempt of Patañjali to define substance" is indeed based on a corresponding view of Samkhya. Yet, in so far as the etymologizing definition of dravya has been shown to be nothing but an additional argument meant to safeguard this "second attempt ”, it can further be stated that there is a very high degree of probability that it, too, is of the very same doctrinal provenance. There is hence no need to confine oneself to observing with Matilal that "there is a striking similarity between Patañjali's definition of substance and the Sāmkhya

Page #21

79)
A NOTE ON MAHĀBHĀŞYA II 366.26 Yoga idea of substance". One can definitely go a step further and-in view of the internal evidence of the M. as well as of the evidence discussed in paragraphs 2 and 3-simply state that Patañjali has in this case, too, taken recourse to a particular element of the Sāmkhya doctrine in order to offer an alternative solution to a particular grammatical problem. Plausible as these conclusions are by themselves, the clear testimony of a Sāmkhya text for this definition of dravya would, nevertheless, be highly welcome. Although I am unfortunately not able to adduce an exact parallel from any of the extant sources, I should like to draw attention to a passage in the Yuktidipikā which, I think, strikingly confirms this peculiar concept of dravya. At the very beginning of the third Āhnika there is a rather long section devoted to defending the satkāryavāda, which is taught in the next Kärikā, viz. 9. The opponent is clearly an adherent of Vaišeșika who does not, however, confine himself to attacking the Samkhya theory, but deems it also necessary to deal in anticipation as it were with eventual counterarguments brought forward against his own asatkäryavāda. One of these counterarguments is that the product is not, because it is not cognized, asat before its origination. This is refuted by an argument which runs thus (48. 29-32):
kriyāguņavyapadeśāsambhavāt / yad dhi pratyakşato nopalabhyate tat kriyayāstiti samsūcyate / yathā harmyāvasthitānām trnānām udvahanād vāyuḥ, gunena yathā mālatilatā gandhena, vyapadeśena vā kāryādinā yathendriyāni / na tu prāg utpatteḥ kāryasya kriyāgunavyapadeśasam
bhavaḥ / tasmād asat kāryam /. In the course of his refutation the defendant comes back to this topic in order to reject the opponent's statement in the following manner (51. 14-18):
yat tüktam kriyāgunavyapadeśāsambhavād anumānābhāva iti tad anupapannam / kasmāt / prthaktvānabhyupagamāt / kāryakāraṇaprthaktvavādinas tatkriyāguņānām pythaktvam anumātum yuktam ity atas ta (ntva). vasthāne pațakriyāguna-(read: ogunā-) grahaņād anumānābhāva ity ayam upälambhaḥ sāvakāśaḥ syāt / asmākam tu kāraṇamätrasyaiva samghātād äkārāntaraparigrahād vā kriyāguņānām pracitir vyaktiviśeşo bhavatiti bruvatām adoṣaḥ/.

Page #22

A. WEZLER “ ......But [this] reproach is not valid for us because what we teach is that a particular manifest thing originates as the accumulation of movements and qualities on account of the cause and nothing but the cause having coagulated or having assumed another shape". Although the key word dravya does not occur here and besides guna the term kriyā is also used (evidently because it has to take into account, at least argumenti causa, the tenets of the Vaiseșika), it is as plain as it can be that virtually the same concept of individual material things is testified to in this passage, for the replacement of samudāya by praciti is hardly of any importance.
81)
Matilal is, however, quite right in remarking, though on what he wrongly takes to be Patañjali's “third attempt”, that “the implication is that the integrated whole or the 'bundle' should be conceived as different from the constituent parts". For, this is indeed the pratijñā Patañjali wants to prove in view of what is said in vārtt. 5 on Pāṇ. 5. 1. 119 (cf. M. II 200. 15 f. and 18). But I don't think that this is only “implied "; it is rather clearly expressed not only in M. II 200. 23-25, but also by the subsequent etymologizing definition, if only this latter is recognized to be but a corroborative statement and is interpreted accordingly. As Patañjali makes use here of a Sāmkhya view in order to vindicate-once more, though by another argument--the assertion that dravya is something different from (the) guņas, it can be safely taken for granted that this is not merely in keeping with the corresponding Sāmkhya concept, but is rather a faithful rendering of it, especially because this conclusion is confirmed also by the testimony of Mallavādin and Simhasūri (see above p. 9) as well as by that of the Yuktidīpikā.
If we now take a look at this Sāmkhya concept, and definition, of dravya itself, we cannot but feel some hesitation in assuming that the term dravya as such is also of Sāmkhya origin. It seems rather that it was coined in or introduced by other circles and became so important an element of philosophical terminology that the adherents of Sāmkhya were forced not only to take

Page #23

A NOTE ON MAHĀBHĀṢYA II 366.26
83)
notice of it, but also to take it over, at least in the sense of making clear what they on their part considered to be a dravya within the framework of their own philosophical teachings. Yet, be that as it may, their concept of dravya, fortunately preserved by Patañjali, is indeed characterized by distinguishing dravya from guna. Nevertheless, their view is distinctly different from the Vaiseṣika concepts of dravya and guna. For, according to the Vaiseṣikas a dravya possesses or rather is inhered in by, qualities or properties, whereas according to the Samkhyas a dravya consists of gunas, or, to be more precise, is nothing but a gunasamudaya, an "aggregate/ integrated whole of gunas". The gunas may change, or in fact change permanently, but the material object remains always a gunasamudaya, i.e. stays permanent amid all changes. A mango fruit remains a mango fruit, no matter which stage in the process of ripening it has reached.
84)
Yet, the drṣtäntas given in the M. and the Yuktidipikä, respectively, raise two questions: viz. firstly, if the recognition (pratyabhijña) is adduced as a reason for the preceding statement or if, on the contrary, the preceding 'second attempt to define dravya (and the preceding definition of parinama) aims at explaining the fact of recognition; and, secondly, if what is meant by these examples is that the mango remains "the same mango", as Matilal would have it, or not. Now, the first question is clearly a pseudoproblem, for the position of the Samkhyas cannot but ultimately have been that both the alternative propositions are true, i. e. that because of the dravya having this nature a particular object is recognized, on the one hand, and that because of recognizing it we are able to determine the nature of the dravya as such, on the other. Yet the second is not only a real question, but also one in answering which the line of argument has to be based on circumstantial evidence only. If what is meant by referring to the fact of recognition were the numerical identity, one would expect Patañjali to have said instead of amalakam badaram ity eva bhavati rather idam tad evämalakam, etc. The assumption that it is hence the generic identity which is intended, is further. corroborated by the important role the concept of jatyanuccheda plays in very similar Samkhya contexts.
85)
86)
23

Page #24

24
A. WEZLER Resuming the problem, provisionally discussed already above, of what precisely is meant by defining dravya as “that the tattva of which does not change", it can now be stated that it is that due to which an object is recognized at any point of time of its existence as belonging to a particular species (jāti). However, it is hardly conceivable that the adherents of Sāmkhya should in this connection not have taken into account also the problem of the numerical identity, at least with regard to human beings as individuals bearing proper names. Therefore it seems advisable not to rely too much on the drstāntas of the M. and the Yuktidīpikā, but to reckon equally with the possibility that the concept of tattva, at least, in addition included the numerical identity, too. It should also be noted that the expression svarūpa used by the author of the Yuktidīpikā-by way of explaining the term tattva of the verse, as it were-similarly covers both the individual as well as the generic “own / characteristic shape / nature" of objects and living beings. In any case, however, it has to be stressed that it is by far better not to use just the term “substance" in rendering this peculiar Sāmkhya concept of dravya, the meaning of which is rather " individual material object as remaining the same (in spite of all changes it undergoes while being manifest)”.
In connection with explaining the Sāmkhya “Doctrine of change ” P. Chakravarti draws attention to "the Sāmkhya doctrine of immutable eternity and eternity-in-mutation ", for which he refers to the Yogabhāsya on YS 4. 33 where they are stated thus:
dvayi ceyam nityatā kūtasthanityatā pariņāmanityatā ca / tatra kūtasthanityatā puruşasya / pariņāmanityatā guṇānām / yasmin parinamamāne tattvam na vihanyate tan nityam / ubhayatra ca tattvasyāvighātān nitya
tvam /. Yet, the gunas referred to here are, of course, the “constituents of primary matter” as becomes clear also from the explanations of the author of the Pātañjalayogaśāstravivarana. Nevertheless, this is a highly interesting passage not only because of its striking similarity to M. I 7. 21 f., but also because of its contributing to a better understanding of the Sāmkhya concept of dravya qua “individual material object". For, the Sāmkhya definition of dravya has

Page #25

A NOTE ON MAHĀBHĀŞYA II 366.26 been shown to cover, at least first of all, the individual material objects-and the term guna correspondingly to refer to the “ qualities” of the tanmätras, etc., too. Now, in the light of this Yogabhāşya passage it becomes apparent that the Sāmkhya-Yoga concept of dravya as a gunasamudāya bears a strong resemblance to that of the "constituents of primary matter" as being pariņāmanitya : Just as the “constituents of primary matter "-of which each and every phenomenon ultimately consists-are conceived of as eternal, i.e. as remaining the same in spite of the processes of change they permanently undergo, so individual material objects are regarded by the adherents of Sāmkhya and Yoga as something which is of a relative per manence if compared to the ever changing " qualities" an aggregate of which they represent. On the other hand the Yogabhāșya passage confirms that the conception of the guņas=“ constituents of primary matter” is distinctly different from that of the guņas=" qualities" of the tanmātras, etc. : the former remain the same amid all changes whereas in the case of the latter it is only their samudāya which appears in contradistinction to the still much more ephemeral" qualities " as unchanging, albeit for a limited period of time only.
Nevertheless, it is primarily in connection with the peculiar Sāmkhya conception of the three “constituents of primary matter" that the Sāmkhya definition of dravya becomes fully understandable. For, there could hardly be a judgement more mistaken than one similar to that of Jacobi who, although with reference to the M. on Pāņ. 1. 2. 64, has voiced the opinion that “this is a very crude theory about the qualities and one that is very far removed from the refined speculations of the Sāmkhyas and Yogas about the tanmātras and mahābhūtas". On the contrary, besides the fact that the Sāmkhya definition of dravya is rightly taken e.g. by Simhasūri to refer to these entities, too, one cannot but state that the Sāmkhya idea of evolution, the concept of prakyti as sattvarajastamomaya, etc., are such that this definition can even be styled quite typical for this school of thought. For, the concept of the three guņas is essentially characterized by the lack of distinguishing between 'substance and quality' as they are conceived of e.g. by the Vaiseșikas. The salient point of the Sāmkhya conception of the three gunas-and, to be

Page #26

A. WEZLER sure, this is what has caused the controversy about the appropriate rendering of the term guna itself-evidently is that we have to do here with an idea which even we find difficult to understand, not to speak of accepting, viz. that of material qualities', i. e. ' qualities' which are at the same time a/the substance or that which in various and permanently changing combinations constitutes the manifest world in its totality. Seen against this background the definition of dravya turns out to be ultimately nothing but an alternative formulation of what is stated e.g. in the Yogabhāşya on YS 4. 13 thus:
sarvam idam gunānām samniveśaviseşamātram (iti paramārthato gunāt
mānah)/. And, to wit, a formulation by which the adherents of Sāmkhya wanted to make clear first of all their own concept of individual material objects..
The conception of the three guņas can, of course, be explained historically by the plausible assumption that it was developed at a time when Indian thinkers ‘had not yet learned to distinguish between substance as such and its qualities or properties. Yet, the characterization of this view as being
archaic' which consequently seems to suggest itself is not wholly satisfactory. At least, if this epithet is used contemptuously-and, to be sure, it usually is used with this connotation-I for one would not readily subscribe to such an evaluation, and not only because I think that our knowledge of Sāmkhya, its origins and development is still fragmentary.
94)
NOTES
* For “Studies on Mallavādin's Dvādaśāranayacakra I" see my contribution to: Studien zum Jainismus und Buddhismus. Gedenkschrift für Ludwig Alsdorf, hrg. von K. Bruhn und A. Wezler (Alt-und Neu-Indische Studien 23), Wiesbaden 1981, pp. 359-408. 1) Geschichte der indischen Philosophie, II. Bd., Salzburg 1956, p. 100. 2) Cf., apart from the articles mentioned in fn. 4 and Geschichte der indischen Philosohie II,
p. 59. 144. (the recently published) Nachgelassene Werke. I: Aufsätze, Beiträge, Skizzen, hrg. von E. Steinkellner. Wien 1984, pp. 93 ff.; by the introductory (bracketed) remark on p. 93 it becomes clear that Frauwallner had the plan to deal at some length with the grammarians' language theory; see also pp. 65 and 137.

Page #27

A NOTE ON MAHABHAṢYA II 366.26
3) Panini. A Survey of Research. The Hague-Paris 1976, pp. 256-259. For what I looked, in vain however, was a reference to G. Oberhammer's review article in: OLZ LIX (1964), 16 ff. 4) Viz. "Das Eindringen der Sprachtheorie in die indischen philosophischen Systeme" in: Indologen-Tagung 1959, Verhandlungen, Göttingen 1960, pp. 239-243 (-Kleine Schriften, hrg. von G. Oberhammer und E. Steinkellner, Wiesbaden 1982, pp. 279-283) and "Sprachtheorie und Philosophie im Mahābhāṣya des Patañjali" in: WZKSO 4 (1960), pp. 92-118(-Kleine Schriften, pp. 284-310).
27
5) Cf. Cardona, o. c. (fn. 3), p. 59.
6) Cf. my article "Paralipomena zum Sarvasarvätmakatvaväda (I). Mahābhāṣya zu Pän. 4. 3. 155 und seine einheimischen Erklärer" in: WZKS 26 (1982), p. 164 ff.
7) Cf. also below p. 13.
8) I use, of course, Muni Jambūvijaya's edition, Bhavnagar 1966. I quote the text always as
it is printed in this edition, but sometimes I add in parentheses punctuation marks in order
to facilitate the readers' understanding.
9) O. c., pp. 23-25.
10) The first one I shall study in a monograph which is still in preparation.
11) Cf. also the long fn. 1 on p. 268 of the edition in which the editor quotes relevant passages from various commentaries on Samkhyakärikā 22. Nevertheless I do not understand what is meant by rupadi sṛṣṭau, etc.
12) For the use of the instrumental Muni Jambūvijaya here refers to Pap. 2. 3. 21: itthambhutalakṣaṇe.
13) As pravṛtti is used here in the sense of "coming into existence [by a process of parinama]" (cf.fn. 20 below) one would rather expect the reading ekagunāt pra°; cf. also ekasmāt kāranad bhavat in the next but one sentence.
14) The editor adds here square brackets, but leaves the space in between empty as neither in the Paniniyadhätupäțha nor in the Haimadhätupatha an exact equivalent is found (cf. his fn. 3 on p. 208). My own endeavours to identify this quotation were equally futile. Nevertheless what Simhasüri aims at here is clear, viz. to adduce a grammatical authority for his interpretation of guna as "number".
15) Muni Jambüvijaya wonders (fn. 5 on p. 268) whether one should read here tvanmatavirodhat which would in fact be better. But the reading avirodhāt is also acceptable if the expression is taken to mean: "because [this alone] does not stand in contradiction to your [general] position".
16) Note that according to the editor (fn. 6 on p. 268) all the MSS. read gunakāśam. I do not quite understand why he decided in favour of the emendation guṇākāśam and confined himself to printing out in this footnote that the reading of the MSS. might go back to 'gunam akāśam which is clearly to be preferred if one does not want to go even a step further and decide in favour of gunad ākāśam (cf. fn. 15).
17) Cf. also Frauwallner, Geschichte der indischen Philosophie, I. Bd., Salzburg 1953, p. 355 ff. 18) Ed. R. C. Pandeya, Delhi 1967, p. 118 1. 14 ff.
19) He is, no doubt, trustworthy in this regard, too; for, this tenet is evidently based on the Samkhya conception of primary matter as sattvarajastamoma y a (cf. fn. 25).

Page #28

A. WEZLER
20) Note that the term pravrtti—as well as pravarteta, of course is used here in the sense of
"becoming manifest by parināma in the process of evolution"; cf. also NĀA 324. 17. 21) Note that Mallavādin himself uses the expression (guna-) sandrāva at other places, viz. 303.
4 and 73. 4. 22) Cf. nts. 13 and 20. 23) The soul belongs to the sphere of sannidhibhavana, whereas primary matter and its evolutes
testify to quite another form of being, viz. apattibhavana. Cf. the article mentioned in aster
isked (*) fn., p. 364. 24) Though it is highly probable that Mallavādin himself took, like Simhasūri (cf. NĀA 15. 17
and 261. 22), vyäkarana to be a sarvatantrasiddhanta, it would be rather far-fetched, to put it mildly, to assume that this is the reason for his referring to the Mahābhāşya, and not the
manner in which the definition of dravya is formulated there. 25) Viz. NC 268. 3: śabde traigunyam asty eveti cet ... and Simhasūri on it (NĀA 268. 14 f.):
sarvasyoktasukhaduhkhamohamayatvāc chabdo 'pi tadātmā triguna eveti ced ity aśankāyām ... 26) These go, at least in substance, back to a Samkhya source. 27) Though I am unable to deal here with these terms, attention may, nevertheless, be drawn
to the fragment from a Sāmkhya text quoted by Simhasūri (12. 17-22), to NC 265. 1 - where sukha, duḥkha and moha are said to be the atmānah of prakrti - and to NĀA 265. 11 f.: te ... sukhaduhkhamohāh prakāśa (pra] -vrttiniyamātmakah sattvarajastamolakşanā gunah şam yāvasthāyām prakrtih' ity ucyante ... The interchangability of the expressions sukha, etc. and sattva, etc. is also evinced by NC 303. 2 (see below p. 7) versus 300. 1 (see below p. 6), etc. Cf.
also below fn. 81 and Sāmkhyakārikā 12. 28) Cf. NC 297. 6 ff. 29) Instead of prthivyā iva agnih he should, however, have said: agner iva prthivi! 30) Cf. NĀA 301. 13 ff. 31) Cf. NĀA 301. 16 ff. : ... it y atathataiva ananyataivety arthah, iti sabdahetvartha..
tvät tryātmakaikatvad ity arthah. 32) Cf. fn. 31. 33) In view of the passage quoted below it should be noted that the iśvaravādin uses the l.
prs. pl. of the personal pronoun with reference to himself and bhavant with reference to the
adherent of Sāmkhya. 34) arka is commonly taken to be the Skt. name of the shrub Calotropis gigantea R. which is
called in English either "gigantic swallow-wort" or "mudar" (Page #29

A NOTE ON MAHĀBHĀŞYA II 366.26
29 likelihood for the assumption that what the Indians had in view when thinking of the arka in connection with something characterized by extreme cottonlike lightness were neither the silky bast fibres nor the tomentum of the leaves as both these parts of the shrub would first have to be obtained or torn off before they can exhibit the corresponding property. It is hence much more probable that it is the long silky coma which is referred to; for thanks to it the seeds are when ripened easily driven away by the wind. This assumption is confirmed 1) by the fact that e.g. in Yogasutra 3. 42 it is cotton itself (tala), i.e. "the white fibrous substance which clothes the seeds of the cotton plant" (Oxford Dict.), which is adduced as an example for something very light (laghu) and 2) by the observation that it is precisely the idea of being driven away by the wind which one finds closely connected with arkatala in some other cases, too, viz. e.g. in Pārsvanāthacarita 2. 926 (aho mama kva tad geham kva paricchadah / vidhina kváham anttah pavanen arkatalavat // ; cf. also 1. 740) and in Trişaştiśalākāpuruşacarita 3. 1. 52 (v atena tena mahatā sa mahan api väridah / arkatalam ijod dh û ya disodiśam aniyata/1). -As for arka in the proverb arke cen madhu vindeta kimartham parvatam vrajet cf. Nilmadhav Sen, “A Note on akka - A Ghost Word in Sanskrit" in : Sanskrit and Indological Studies. Dr. V. Raghavan Felicitation Volume, ed. by R. N. Dan.
dekar et al., Delhi 1975, 341-349. 35) Cf. NĀA 302. 26 ff. : alpavācini kani (cf. Pän. 5. 3. 85] lo ha pinda kah palamätrapra
manorkatalabharal laghtyan palasatika tula, vimsatis tula bharah' [ ] iti paribhasitatvät / ayaspindo gurur api laghur arkatalo laghur api gurur drsta ity anavasthitaikagurutvalaghutvatattve gurulaghutve, apeksikatvad asman prati na gurulaghutve parasparato 'nye, tato näyaspindarkatula
drstanto 'sti/. 36) Cf. fn. 27. 37) Cf. NAA 303. 12-17. 38) Cf. the immediately following passage of the NC (303. 4 ff.) and Simhasüri's commentary
on it (NĀA 303. 18 ff.). 39) Cf, also the expression ekatvagati NC 304. 1. 40) Cf. also NĀA 306. 6 f. 41) Cf. also NĀA 73. 25 ff. 42) See also below p. 23. 43) Cf. the article mentioned in asterisked (*) fn., p. 370. 44) Viz. "Tippanāni" p. 16. 35 ff. 45) 0.c. (cf. fn. 3), p. 257. 46) It forms part of several ślokavārttika (on which in general cf. Cardona, O. c., p. 247) stanzas
on Pan. 4. 1.3 and is quoted in M. I 245. 25. On M. II 197. 26 Nāgesa remarks (IV 22a 30-32): sāmkh yamatāvastambhenäha-sv a krtanta iti / etena samkhyaśāstrānusäritvam vyakaranasyeti
súcitam /. 47) Neither H. Scharfe's rendering of samstyana by "Schwellen" (Die Logik im Mahābhāşya,
Berlin 1961, p. 147) (cf. also Seyfort Ruegg, who o. c., p. 40, translates it by "expansion ") nor that of H. Jacobi by "congelation ", " The Dates of the Philosophical Satras of the Brahmans" in: JAOS 31 (1911), p. 27. = Kleine Schriften, hrg. von B. Kölver, Wiesbaden 1970, p. 27) nor that of Strauss by "Dickwerden” (“Altindische Spekulationen über die Sprache und

Page #30

A. WEZLER
ihre Probleme" in : ZDMG 81 (1927), pp. 99-151 (see p. 146) - Kleine Schriften, hrg. von F. Wilhelm, Wiesbaden 1983, pp. 220-272 (see p. 267)) can be accepted. The latter can easily be misunderstood, and the former two are apparently due to a misconstruction of the sentence styāyaty asyam garbha(h) (M. II 198. 3); this sentence which is to illustrate the preceding statement adhikaranasadhana loke stri may indeed create in a Western reader's mind the idea that what is referred to is the growth of the foetus, yet Kaiyata and Nägeśa clearly are of different opinion and their explanations can hardly be put aside. The former says (Pradīpa IV 22 b 27-30) : st ya yaty a sya m iti / samghätarupam prapnotity arthah / ... samstyanam iti / tirobhāvah, pravrttir ávirbhāvah, śāmyavastha sthitih, etaś cavasthah sabdagocara evety avaseyam //, and the latter adds the remark (Uddyota IV 23 a 1-2): samghatara pam iti / garbhapadena tatkaranam śukraśonitam ucyate /. Cf. also Nägesa's explanations of M. I 245. 27 f. (Uddyota II 98 a 22-24): garbhaḥ śukram styāyati bonitena sanghtbhavatity arthah/ sa te śukram tyajati yonirūpa adhāre / styānam apacayah / pravrttir vrddhih //. What is meant by samstyana is hence just the opposite of 'growth', and styäyаty asyām garbhah does not refer to the growth of the foetus (though both, Kaiyata as well as Nāgesa, 'would not have denied that it grows), but to the "coagulation" of the father's sperm and the mother's "blood ", i.e. to what we would call the act of fertilization. Cf. also K. A. Subrahmania Iyer's article "The Vaiyakarana Conception of 'Gender'" in: Bhāratakaumudi (Studies in Indology
in honour of Dr. Ratha Kumud Mookerjee), Allahabad 1945-47, pp. 291-307. 48) Note that in both cases, i. e. in M. I 246. 2 and M. II 198. 4, the expression prasava is re
placed by pravrtti which latter is, however, used in what follows (M. I 246. 6 and II 198. 7) in the broader sense of the activity' consisting in the āvirbhävatirobhāvasthitirapaparināma
(cf. Pradipa IV 24 a 9 ff. and Uddyota IV 24 b 1). 49) The edition used in quoting from or referring to Kaiyata's Pradipa and Nägesa's Uddyota
is that produced by Bhargavaśāstri Joshi, Sivadatta Kudala and Raghunatha Sarmă, Bombay
(NSP), 1937-1951. 50) Cf. also Nāgesa's remark (Uddyota IV 23 a 31 f.): na tu ta dvyatiriktam iti/
ata evägre (viz. M. II 198. 5 f.] sarvās ca murtaya evamatmika iti bhāşyakaro vakşyati/. 51) Contributions à l'histoire de la philosophie linguistique indienne, Paris 1959, p. 41. 52) Cf. Uddyota IV 30 a 4 f. : bhasye sampratya ya iti / kimatmakam dravyam bhavan
manyata ity arthah //. 53) O. c., 1. c. 54) Cf., on the other hand, B. K. Matilal, Epistemology, Logic, and Grammar in Indian Philo
sophical Analysis, The Hague Paris 1971, p. 104. Perhaps Seyfort Ruegg was misled by the question (M. II 200. 14 f.) ka gatir etc. which is, however, meant to point out that the gender of certain primary and secondary derivations cannot be accounted for if the definition gunasa
mudayo dravyam is accepted (cf. Uddyota IV 30 a 14 ff.). 55) Uddyota IV 30 a 14 ff. 56) Cf. also Pradipa IV 29 a 21 ff. - For the term samniveśa cf. e. g. Yuktidipikā 48. 6 ff. See
also below p. 26. 57) Matilal (o. c., p. 104) rightly refers also to the M. on Pāņ. 5. 2. 42, ie, II 380. 2; yet, this
reference can also be found in the Uddyota IV 298 b 25 (read, however, tayapsütre).

Page #31

A NOTE ON MAHĀBHĀSYA II 366.26
31 58) O. c., p. 101 ff.; cf. also Matilal's article “The Notion of Substance and Quality in Ancient
Indian Grammar" in: Acta et Communicationes Universitatis, II, 2, Tartu 1973, pp. 384-407, which does not, however, differ from what he has written in his book in analysing the discus
sion on Pän. 5. 1. 119. 59) I take this expression (and the synonymous nirdalatva) to mean literally "the fact of lacking
in anything unfolding itself" > "the being of no avail" > "unproductiveness" or rather "irrelevancy". It is met with also in the Laghuśabdendusekhara with Candrakala, ed. by Gopalaśāstri Nene, Pt. II. Kashi SS 5*, Benares 1924) p. 29 1. 12, 52. 1 and 530. 11 f. as well as in Jayatirtha's Nyäyasudha on Madhva's Anvākhyāna on BS 1. 1. 1 (viz, on verse 79) and
on BS 1. 1. 11 (viz. on verse 141). 60) O.c., p. 103 f. 61) O.c., p. 103 fn. 10. 62) Cf. e. g. P. Thieme's characterization and evaluation of Nägesa in : "Bhāşya zu värttika 5
zu Panini 1. 1. 9 und seine einheimischen Erklärer. Ein Beitrag zur Geschichte und Wür.
digung der indischen grammatischen Scholastik", NGGW 1935, p. 198 f. 63) Cf. also fn. 2 on p. 297 of Vol. IV : ... yasya gunasya bhāvad dravye sabdaniveśas tadabhidhäne
tvatalav iti värttikena dravyavāca kāc chabdad gune tvatalau vidhtyete / tayoh prakrtipratyayarthayor
dravyagunayoh ko bheda iti praśnah //. 64). Viz. Uddyota IV 297 a 19 ff.. 65) Kaiyata can, as has been shown, only partly be regarded as trustworthy. 66) By using the singular with reference to khalv api I want to intimate that it should be re
garded as one compound particle. 67) Viz. his Sanskrit-Chrestomathie, St. Petersburg 1877, 227. 22 (which is Kasika on Pan. 2. 3.
5), 233, 6 (-Kasika on Pan. 2. 3. 36), 239. 5 (-Kāśikā on Pän. 2. 3. 70) and and 244. 3 (
Kāśikä on Pāṇ. 3. 3. 135). 68) Cf. Pradipa II 25 b 15 and Uddyota II 25 b 25 f. as well as V. P. Limaye, Critical Studies on
the Mahābhāşya, Hoshiarpur 1974, p. 78 f. 69) Cf. Nägeśa's remark (Uddyota [Rohtak ed. of the M.] V 401. 28) : bhasye' pi indriya'.
śabdena jihvaiva / vyaktibahutvād bahuvacanam /. 70) Cf. Nāgesa's explanation (Uddyota V 401. 29): sva s minn at man i ti / rasāsvādagraha
katvalakşane svasvarapa ity arthah /. 71) Cf. Kaiyata's explanation (Pradipa V 401. 18): tath eti / vyajyate 'nena rasa iti vyanjanam
Iraga iti / rajyate 'neneti rago vyanjanam eva /, and Nāgeśa's further elucidation (Uddyota V 401. 27 f.) vyaj yate 'neneti / jihvāya jadibhāvanirakaranena raso vyajyata ity arthah /.
Cf. also P. V. Limaye, Critical Studies on the Mahābhāya, Hoshiarpur 1974, p. 731. 72) Cf. Nägesa's remark (Uddyota IV 299 a 13 f.): ...athava yas y eti/ 'sadbhāvāt' iti beşah
/ yasyāvayavasamahasya sadbhāvād ity arthah /. - What immediately follows is the passage
quoted above p. 12. 73) O. c., p. 103. 73a) The historical interpretation of this Mahābhāşya passage apart, one could, of course, state
on material grounds that the etymologizing definition of dravya does not agree really well with the definition it is meant to corroborate, i.e. that the former does not necessarily fol

Page #32

A. WEZLER
low from the latter and vice versa. But that is quite a different matter, and evidently not
what Matilal had in view. 74) As for the expression tattvad apracyuta- cf. also the material drawn upon by P. Hacker,
Vivarta. Studien zur Geschichte der illusionistischen Kosmologie und Erkenntnistheorie der Inder (Akad. d. Wiss. u. d. Lit., Abh. Geistes- und Sozialwiss. Kl. Jg. 1953, Nr. 4, pp. 179. 270) Mainz, 1953, p. 16 and 40 as well as e.g. Helarāja on Vākyapadiya IIII Kriyāsamuddeśa 26 (ed. K. A. Subrahmania Iyer, Poona 1973), p. 21 l. 21 ff. (where the corresponding definition of Yäska, Nir. I 2, is quoted too) and Yogabhāşya on YS 3. 15: ... pindah pracyavate ghata
upajāyata iti dharmaparināmakramah /. Cf. also fn. 76. 75) Cf. Frauwallner, Geschichte..., I (cf. fn. 17), p. 389. Note, however, that Frauwallner ap
parently failed to recognise that the sense in which these terms are used by Samkhya and Yoga authors is significantly different from that it has in Vaiseșika texts; cf. e. g. the pas
sage quoted from the Yogabhäşya in fn. 74. 76) Cf. also Nyayabhäsya on NS 3. 2. 15. P. Chakravarti, Origin and Development of the
Sāmkhya System of Thought, Delhi 1975, p. 258, fn. 2 refers also to Nirukta I 2 (ed. R. Roth.
Göttingen 1852, p. 31 l. 17): viparinamata ity apracyavamānasya tattvād vikāram (scil. acaste]. 77) The edition reads palasam palaśad here; yet one would rather expect palāśam palaśatvad. 78) Cf. fn. 6 above. 79) Cf. fn. 94 below. 80) I do not hence take this passage to be an explanation of the term vyaktipracaya occuring
Yuktidīpikā 48. 11, 49. 4 and 53. 26/31. 81) O. c., p. 104. 82) In connection with this question (which I cannot, however, discuss here) I should like to
draw attention, though in passing only, to the fact that the Samkhya definition of dravya bears a strong resemblance to the Sarvāstivādins' conception of phenomenal things as being nothing but an aggregate of atoms on which latter cf. e. g. E. Frauwallner, Die Philosophie des Buddhismus, Berlin 1958, pp. 64, 96 and 120; M. Hattori, Dignāga on Perception, Cambridge, Mass.1968, pp. 26 and 88f. (fn. 1. 38 and 39), and Abhidharmakośabhāşya ed. by P. Pradhan and A. Haldar, Patna 1975, pp. 24. 18 (see also Sphuţārtha Abhidharmakośavyākhyā by Yaśomitra, ed. by U. Wogihara, Tokyo 1971, Vol. I. p. 68. 4: paramānusañcayasvabhāvā daśaiva...), 34. 1 f., 52. 24 ff. and 475. 1 ff. (particularly 14 ff.) (Sautrāntika criticism of the Vaiseșika concept of atman). This similarity is indeed noteworthy, especially since the basic positions are distinctly different, viz. a marked substantialism in the case of the Samkhyas and a no less clearly expressed antisubstantialism in the case of the Vaibhāşikas. Likewise it can only be mentioned here that in the twelfth Ara of the NC, i. e. in the context of the critical discussion of Dignäga's apoha theory, the expression gunasamudaya is also met with (see 652. 14 together with fn. 1; 668. 5/21; 671. 1/3), though used with reference to that
which is denoted by a proper name like Dittha. 83) "Zum Begriff der Substanz (dravya) im Vaišeşika" in: WZKS 19 (1975), pp. 183-166 and
"The Vaiseșika Concept of guna and the Problem of Universals" in: WZKS 24 (1980), pp.
225-238. 84) Cf. Kaiyata's Pradipa above p. 13.

Page #33
33 A NOTE ON MAHABHASYA II 366.26 85) Cf. also Chakravarti, o. c., p. 260: "...its palasahood remains intact always...". - In this connection it may be noted that the maxim (=Paribhasa 37 in Nagesa's Paribhasendusekhara) ekadesavikrtam ananyavad bhavati refers according to e. g. M. I 136. 9 f. likewise to the generic identity. 86) Cf. the article mentioned in fn., p. 375 et passim. 87) O.c., p. 258. 88) O.c., p. 264 (referred to also in fn. 2, p. 258 f.). 89) Viz. (366. 22 ff.): nanu parinaminam gunanam katham nityata syad iti braviti--y a s min pa rina m a m a ne t att vam na vihan yate / yasya yad rupam pramanenopalabdham, tasya tat tattvam / tad yasmin na vihanyate na vyabhicarati tad api nityam / gunanam api sukha duhkhamohatmata prakasakriyasthitisilata ca tattvam na vihanyate / tasmad ete 'pi parinamanityah /. 90) Quoted above p. 19. 91) See above p. 8.. 92) L. c. (cf. fn. 44 above), p. 27. 93) Cf. Hacker's course of lectures entitled "Grundlagen indischer Dichtung und indischen Den kens ", which was recorded in 1978 and will be edited by K. Ruping in the near future in the de Nobili series published in Vienna). Cf. also Frauwallner, Die Philosophie des Buddhis mus, Berlin 1958, p. 96. 94) Space being limited-and I am afraid that I have already taken too much of it-I cannot deal here also with the parallel to this definition of dravya, found in the Yogabhasya on YS 3. 44, to which attention has been drawn among others by Nagesa (e. g. Uddyota IV 299a 30 f.), and with the passage in Nagesa's Manjusa in which the relation between these two definitions of dravya is discussed. It remains for the future to follow up these references in an article which will form a sort of addendum to the present one. Suffice it to say here that I don't think that the Yogabhasya passage forms an additional and independent testimony which should also be taken into account in determining the doctrinal provenance of the definition referred to in the M.