Studies On Bhartahari And Prakrta Dhvani And Samkhya Tanmatras

Author(s): Johannes Bronkhorst
Publisher: Johannes Bronkhorst
Catalog link: https://jainqq.org/explore/269567/1

Page #1
JOHANNES BRONKHORST STUDIES ON BHARTRHARI, 8: PRAKRTA DHVANI AND THE SAMKHYA TANMATRAS' Bhartrhari distinguishes between the word itself (sometimes called sphota) and the sounds that manifest it. These sounds themselves are subdivided in one passage of the Vakyapadiya into prakrta divani and vaikrta dhvani. These two expressions have puzzled modern scholarship. J. F. Staal offers the following interpretation (1969: 519 [123]): "Bhartrhari distinguishes between (1) the sphota of an expression, which denotes the expression as a single unit conveying a meaning; (2) the prakrta-dhvani of an expression, i.e. the phonological structure assigned to the type it represents; and (3) the vaikrta-dhvani, i.e. the phonetic realization in its particular utterance-token." This interpretation goes back to John Brough (1951). K. Kunjunni Raja (1969: 14-15) gives a somewhat different explanation: "First, we have the actual sounds of the words uttered; this is the vaikrta-dhvani. These sounds reveal the permanent prakrta-dhvani which is an abstraction from the various vaikrta-dhvani-s, or which may be considered as the linguistically normal form devoid of the personal variations which are linguistically relevant." Elsewhere in the same book he describes the prakrta dhvani as an "abstract sound-pattern with the time-sequence still attached to it" (p. 117), as "the phonological structure, the sound-pattern of the norm" (p. 120). Jan E. M. Houben (1990: 125 with n. 17) criticises Brough's view to the extent that the vaikrta-dhvani represents "the individual instance, noted in purely phonetic terms" and observes: "The prakrta-dhvani refers to those phonetic features of the audible sound that are differential in the system of language. The vaikrta-dhvani is not differential in the system of language." He further points out that the verses of the Vakyapadiya only use these terms, without defining them, so that for an interpretation we have to rely on the ancient commentaries. The following interpretation, which obviously tries to do justice to the commentaries, is due to Ashok Aklujkar (1990: 132): "Sphota, though without temporal distinctions, appears to have temporal divisions of two kinds: difference in the form of short vowel or long vowel, and so on; and difference in the form of a quick (druta), medium (madhyama), or slow (vilambita) pace of utterance, due to division in Journal of Indian Philosophy 27: 23-33, 1999. 1999 Kluwer Academic Publishers. Printed in the Netherlands

Page #2
JOHANNES BRONKHORST STUDIES ON BHARTRHARI, there are five, called the tanmatras, that are both prakrti and wiki. because they give rise to five other principles (the five elements), and are themselves derived from the principle ahamkara. The five hundrus carry the names of the five qualities, but are not identical with them. The Samkhya distinguishes therefore a sabdatanmarra (sound'), a sparsatanmatra ("touch"), a rupatanmatra ("colour), a rasatanut ("taste"), and a gandhatanmatra ("smell"). What can be said about them! Samkhya Karika 38 begins with the words tanmatruny avisenal "The ones without specific features' (avisesa) are the fannirus". The Yuktidipika comments: the manifesting sound (dhvani). A part of the sound is the minimum needed for the manifestation of the linguistic units (prakrta dhvani); the remainder, if any, simply keeps the manifestation in effect for a longer time (vaikrta dhvani). The former is related to the distinction conveyed by 'short', and so on, the latter to the distinction conveyed by 'fast, and so on." Madhav M. Deshpande (1997: 46-47), similarly, observes: "The sphora level is said to be beyond temporality, while the primary. manifesting sounds (prakrta-dhvani) have the feature of duration or length. The secondary manifesting sounds (vaikrta-dhvani), which are further reverberations of the primary manifesting sounds, reveal the feature of tempo. Thus, in general, we get concentric circles representing different features." He then comments: "As a production model, do not think Bhartrhari's ideas will rank very high in the evaluation of modem phoneticians. On the other hand, the diagrammatic perception of the various phonetic features as concentric circles moving out from more distinctive to less distinctive offers an interesting view of these features and deserves to be explored further. It is of course well known that it is not without risk to interpret old Indian texts only in the light of modem notions of linguistics, especially where there is no explicit evidence to support such an interpretation. Texts have to be interpreted first of all in the light of notions familiar to their own author(s). It is not obvious that all the interpretations mentioned above fulfil this requirement. Aklujkar's interpretation does try to remain close to the texts. It is, however, strange in that it suggests two succeeding parts of sound with altogether different functions. If Bharthari entertained such notions about sound, where did he get them from? Instead of or rather before - following Houben's advice to rely on the ancient commentaries, I propose to explore a different path: to reflect upon the question what prakrta and vaikrta dhvani could be. Consider the expressions prakrta and vaiksta. Why did Bharthari use these? Prakrta is an adjective derived from prakrti; vaikyta is similarly derived, or can be derived from vikti? The terms prakti and vikyti are particularly popular in classical Samkhya, which divides its twenty-five principles (rattva) under these two headings. Samkhya Karika 3 puts it as follows: "The root-prakrti is no vikti: the seven beginning with mahad are both prakrti and vikti; sixteen are [only] vikti (here the synonym vikara is used); the purusa is neither prakti nor vikti." In our quest for the meaning of prakrta dhvani it is not necessary to enumerate all the twenty-five principles of Samkhya and show their mutual relationship. It is sufficient to recall that among those principles Those indeed are the ones without specific features which are the fun ! These Sabdarandra, sparsanmdira, rapatan dira, rasad u, and whe n Why are they called tantras? Because specific features of the same kind are not possible in them). When there is no difference of kind, eg sound, either specific features such as the accents called udara, and it suit, or the being nasal - are found in it, and that is why it is called) sabda andra approx. sound and nothing but that'). In the same way (there are no specific features such as 'soft', 'hard' etc. in the lanmdtra of touch: (no specific features such as 'white' 'black' etc, in the tanmatre of colour: (no specific features such as 'sweet 'Mour etc. in the fanm.dira of taste; (and no specific features such as 'fragrant' eld in the tanmarra of smell. For this reason only the general feature of each quality is present in the ranimiras), no specific feature, and this is why those ones without specific features are the landiras. We are primarily interested in the sabdatanmatra. It is here presented as sound without the specific features that may accompany sound. It is moreover, different from the quality sound. The quality sound, we may assume, possesses all the specific features which the sabdatanmatu is here stated not to possess. But the quality sound, unlike the sabdatanmatra, does not evolve into other principles. The fact that the sabdatanmatra does do so, justifies it being prakrti, or pruiktu sabda. To avoid confusion, let me point out that the Samkhya texts, as far as I am aware, do not use the expression prakrta sabda. But this would seem to be an insignificant detail. Prakrta means "belonging to the prakrti(s), original", and obviously the sabdatanmatra does belong to the praktis of Samkhya. But if the Samkhyas accept a prakrta sabda, one would expect that they also accept a vaiksta sabda "modified sound". Here however we are confronted with a difficulty that characterises classical Samkhya as it has been handed down to us. None of the usual qualities, and this includes the quality sound, figure among their twenty-five principles Contrary to what one might expect, the tanmdiras do not give rise 10 the corresponding qualities, but to the five elements, in the following

Page #3
26 JOHANNES BRONKHORST manner: the tanmatra of sound gives rise to ether, the tanmatra of touch to wind, the tanmatra of colour to fire, the tanmatra of taste to water, and the tanmatra of smell to earth. It is even stranger that these five elements are stated to be 'specific features' (visesa) in the Samkhya Karika. The Yuktidipika seems to take a different position, for it gives a long enumeration of characteristics (dharma) for the five elements, and concludes: "These [here enumerated characteristics] are called 'specific features (visesa)." The position of the qualities is described as follows in the Yuktidipika:7 From the tanmatra called] "sound", which has sound as quality, ether is born.] which has [that] one quality. From the tanmatra [called] 'touch', which has sound and touch as qualities, wind (is born.] which has [these] two qualities. From the tanmatra (called) 'colour', which has sound, touch and colour as qualities, fire (is born.) which has [these] three qualities. From the tanmatra (called] 'taste', which has sound, touch, colour and taste as qualities, water (is born.) which has [these] four qualities. From the tanmatra [called] 'smell', which has sound, touch, colour, taste and smell as qualities, earth [is born,] which has [these] five qualities. We learn from this passage that the qualities are not derived from the tanmatras, but that they somehow characterise both the tanmatras and the elements derived from them. In other words, the qualities have no place in the evolutionary scheme of Samkhya. They are not derived from anything at all, but they somehow pop up in the company of both the tanmatras and the elements. The situation is even stranger than it may look at first sight. Recall that the tanmatra of smell is free from specific features such as 'fragrant' and the like. But now we learn that this same tanmatra has sound, touch, colour, taste and smell as qualities. It looks as if the tanmatras are here not looked upon as "pure" qualities, as was the case in the description above, but as some kind of "pure" or "pre-"elements. Indeed, the passage just cited is introduced by the remark: "From the elements (bhuta) which have each one more [quality than the preceding one] arise the specific elements (bhutavisesa) which have each one more [quality than the preceding one]." Here the first word 'element' (bhuta) clearly refers to the tanmatras. The Samkhya of the Samkhya Karika and its commentaries is, as the above passages illustrate, a strange knot of doctrines, which it may take long to disentangle. However, there is reason to believe that Bhartrhari was acquainted with an earlier form of the system, which may have been, in at least some respects, less obscure. Some passages in his Vakyapadiya and Mahabhasyadipika indicate that he knew a form of Samkhya in which all material objects were looked upon as constituted of qualities. Citations in the works of other authors STUDIES ON BHARTRHARI, X - among them Dharmapala and Mallavadin confirm that this was at some point a doctrine of Samkhya. There is even reason to think. that these qualities once figured among the principles (tattva), as tinal evolutes, and therefore as vikrris only. They may have been the risess before this term came to be reserved for the five elements. If we assume that at one point in the history of Samkhya rammatras were thought to give rise to the corresponding qualities, as seems likely in view of the way the tanmatras are still described in the much later Yuktidipika. we may have found our vaikrta sabda. In that case the prakrta sabda is the sabdatanmatra, free from adventitious features such as accent. nasalization and the like. The vaikrta sabda is then the quality itself. along with such adventitious features. And the vaikrta sable would then be looked upon as being derived from, or having evolved out of the prakrta sabda. This to some extent hypothetical reconstruction of an earlier phase of the Samkhya system of thought may perhaps help us in coming to terms with Bhartrhari's prakrta and vaikrta dhvani. No importance should be attached, I believe, to Bhartrhari's use of divani in the place of sabda: he often uses sabda as a synonym of sphota, so that this term may have been already used in a different sense. This terminological choice may further be explained by the fact that Bhartrhari uses the expression prakrta sabda elsewhere in order to refer to something altogether different, viz. the/a Prakrit language. It should also not be forgotten that Bhartrhari often uses ideas which he borrows from other systems for his own purposes. His distinction between prakrta and vaikrta dhvani, supposing that he really borrowed these ideas from Samkhya, does not imply that he accepted their other principles and their entire scheme of evolution. With this in mind let us consider the relevant passages of his Vakyapadiya and its commentaries. The terms are used in verses 76-79 of the first Kanda: They declare that the difference of condition tyynit of the spot, which has no difference of duration and which follows the duration of the divani, is due to the difference in accidental features of the grasping." (76) Because there is the sphota] being eternal a difference in nature in the case of short, long, protracted [vowels] and other [sounds), it is figuratively stated that the duration of the prakrta dhvani belongs to the sabda. (77) The prakrta dhvani is accepted as being the cause of grasping the subdu. The ruky [dhvani] becomes the cause of difference of its state. (78) But after the manifestation of the sabda the vaikrta dlvanis bring about a difference of condition; the essence of the sphora is not differentiated by them. (79) This translation is kept rather literal in an attempt not to impose too much of an interpretation. It should further be kept in mind that verse

Page #4
JOHANNES BRONKHORST STUDIES ON BHARTRHARI, X 78 may not really belong to the Vakyapadiya: it disturbs the transition from 77 to 79 (so Rau), and the Vrtti ascribes it to a/the Samgrahakara, It seems however clear that all these verses use the word fabda as a synonym of sphota. The sphota is eternal. One sphota can have a different nature from another one (c.g., u is different from a), but the features (such as length) that allow us to distinguish between them do not really belong to them; they belong to the prakta dhvani. Once the prakrta dhvani has manifested "its" sphota, the vaikyta dhvani may bring about further differentiations, which do not however affect the nature of the sphora. Note that nothing in these verses states that the vaikrta dhvani itself is subsequent to the prakrta dhvani, as are its effects. Recall now what the Yuktidipika had to say about the sabdatanmatra:14 "When there is no difference of kind, e.g. sound, no other specific features - such as the accents called udarta, anudarta, svarita, or the being nasal - are found in it, and that is why it is called) Sabdatanmarra." Moreover, "specific features of the same kind are not possible in them)"." In other words, the sabdatanmatra may be different for different sounds, but it does not contain features that do not differentiate sounds. This, of course, agrees in all details with Bharthari's prakrta dhvani. The Vrtti adds some observations to the above verses of the Vakyapadiya:16 Dhvani here is of two kinds prakta and waikrta. Prakrta (dhvani] is that without which the non-manifested form of the sphola is not distinguished. Vaikrta (dhwani on the other hand is that by which the manifested form of the sphora is perceived, again and again without interruption, for an extended period of time. prakrta dhvani disappears once the sphota has been manifested. Indeed. in this line it uses the mere word dhvani, leaving us guessing what exactly is meant. If we assume - and I repeat that the passage leaves room for doubt - that the vaikrta dhvani comes after the prakta dhivani, we are confronted with a difficulty in the Vrtti on Vkp 76 (75 in lyer's edition). This verse, translated above, appears to speak of the vaikrta dhuni (without mentioning this expression) because it deals with the wrribheda (differ ence of condition) of the sphora, exactly the same expression used in verse 79 to indicate what the vaikria dhvanis bring about. The Vitli on verse 76 (75) contains the following line: "The conditions of the sphota in which we imagine differences - viz. the features quick 'medium', and 'slow', each faster than the following one by one third - are reported to be connected with that grasping that has the sphoru as object, and which is an accidental feature of variable duration." I this means that the vaiksta dhvanis bring about the features quick'. 'medium', and 'slow', we are forced to believe that we are informed about the speed in which a phoneme is uttered by sound that follows the sound that makes us know whether the phoneme concemed is short, long or protracted. This sounds odd, and we would expect the prikru dhvani and the vaikyta dhvani to act simultaneously What would be the Samkhya position in this regard? Do the Turku come into being before the evolutes that derive from them? In one important sense, yes. In the evolution out of original nature (mulapukoti pradhana) each next evolute comes into being after the preceding one But can the same be said about an individual utterance? Does it first produce the sabdatanmatra, and only subsequently its evolute, the quality sound? It is difficult to find a satisfactory answer to this question. However, one thing seems clear. Both the tanmatras and their derivatives are objects of the senses. This we learn from Samkhya Karika 34a, which states:20 "Of the (tenfold external organ) the five sense organs have the visesas and the avisesas as objects." The avisesas, it may be recalled, are the tanmatras. The visesas are the five elements in the classical system. but we have seen that in the system known to Bharthari they may have been the five qualities. The important thing is that the fanmaires are perceivable. Each perception, according to the Samkhya system presumably known to Bharthari, must have primarily consisted of two constituents: "pure" qualities (the tanmarras) and "ordinary" qualities (warts and all). I am not aware of any statement in Samkhya literature to the extent that the tanmatras have some kind of priority in perception, And again, 17 Just as a light, immediately after coming into being, is the cause of grasping a jar etc., but when established (avaristhamana) becomes the cause of the continuation of grasping, in the same way the dhvani that continues once the sabda has been manifested brings about a continuation of the notion that has the sabda as object by adding strength to the manifestation of the object. Therefore, though associated with the vaikrta divani the difference of which is clearly perceived, the essence of the sphora, because no identity is superimposed, does not lead to any usage of difference in duration in the science (of grammar) as do the features) 'short' etc. In these passages from the Vrtti one does get the impression that vaikyta dhvani extends in time beyond prakrta dhvani, that the vaikrla dhvani still resounds when the prakrta dhvani has disappeared. The first passage, to be sure, is not explicit about this. The second passage, on the other hand, speaks of "the dhvani that continues once the sabda has been manifested". Of course, this passage does not state that the

Page #5
STUDIES ON BHARTRHARI. X JOHANNES BRONKHORST NOTES but cannot exclude that some such position was adhered to by at least some Samkhyas. However this may be, it is possible or even likely that the Samkhya scheme of things as known to Bharthari did distinguish Iwo elements in sound, one of them perhaps called prakrta sabda, the other one vaikna sabda." Unlike the latter of these two, the former was free from non-differential features. It is time to turn to the Paddhati of Vrsabhadeva. This commentary is clearly not aware of the possible link of the two kinds of sound with Samkhya. This is clear from the way it explains the terms prakrta and Wiikra, without reference to the Samkhya use of these terms. Since the passage concerned is corrupt in all mss. I will only translate the part more or less plausibly reconstructed by its editor: Regarding the word prikra: On account of the fact that dhvani and splota are nex perceived separately, the plata concerned is thought to be the origin of that dhwani The dhani is called) prikria because it is bom in that. The dhvani that comes after that and is perceived to be different from that is called valkrta, because it is like a moditication of the spot. Or the striking of the organs (or sound is the exrigin of the collection of sounds (dhuni). What comes first into being from that. is praktu, what comes next is uita To conclude. For a correct understanding of Bharthari's prakrta and vaikyta dhvani, his intellectual context must first be taken into considcration. Comparison with theories of modern linguistics is delicate, and should not be made until Bharthari's own intellectual background has been properly explored. It seems likely that the notions of prakrta and vaiksta forms of sound come from Samkhya, where these notions appear to have been current until the revision of that philosophy during which the qualities as final evolutes were replaced by the five elements. This hypothesis explains both Bharthari's terminology and the ideas it covers: both Samkhya and Bhartrhari distinguish between two perceptible forms of sound, the one "pure". the other one "impure". Questions remain as to their temporal relationship: does the vaikrta dhvani come into being after the prukrta dhvani? Neither Bharthari's text nor our limited knowledge about the Samkhya known to him allows us to reach a clear and certain answer to this question. The revision of Samkhya referred to above did away with both prakrta and vaikrta dhvani. Not surprisingly, the commentator Vrsabhadeva no longer understood Bharthari's short and enigmatic passage, and gave it a different interpretation. Preceding articles of this series have been published in the following periodicals and books: Bulletin d'Erudes Indiennes 6 (1988). 105-143 (no. : "L'amour et la date de la Vati): Studien zur Indologie und Irumistik 15 IN), 101-117 "Bharthari and Mimamsa"). Asiutische Studiert wes Ases +59915-IS (no 3: "Bhartrhuri on sphora and universals"): id. 46.1 (1921. 56-8)( Law dans le Vakyapadiya et son lien avec le Madhyamaka: id. 17.1 11993. 75-14 5: "Bharthari and Vaisesika"); Vacasparyam: Process Bhughuli Volume (Pune, 1994, pp. 32-41; no. 6: "The author of the Three Centuries": of the Bhandarkar Oriental Research Institute 76 (1995 1996. 97-10% "Grammar as the door to liberation"). I thank Klaus Butzenberger, Harry Falk and Jan Houben for useful comments. Cp note 23. below. SK 3. malaprakrtir aviktir mahadadyah prakrivikava vapte / sek s in vikaru na prakrtir na vikrtih purusan II + YD p. 117 130-p. 118 1.4 (Pandeya) / p. 224 1.19-p. 225 12 Werler & Motegi): te khaly avisesah / kdini punas tanmdtrdnery wewe subidem . sparsatarimairam rapatanmdirant, rasatanmdiram gandulacrum iti / katha puras tanmatraniti? weyate: tulajdilyavisesdrupapatreh / anye fahdajnhee will visesd udanandanasvaritanundsikadayas fatra na santi/tasme cha m am Ievam sparsatanmatre mrdukashinddayah / evam rapatanunarre suklakridevalt Ievam rasatanmdtre madhurdmladayah / evam gandhatamdire surahhaduw / tad tasya tasya gunasya sdmanyam evdira, na visesa in tanndran de Wisesa 1 The end of this passage reads, in Pandeya's edition, tanmatrasy ete "Wesel follow Wezler and Motegi. SK 38 ... tebhyo bhardni palica paricabhyah / ele smrt widesch an ghores cu madras ca // * YDP 119 1.21 (Pandeya) / p. 227 1.15-16 (Wexler & Motegi) ete wed in ucunda iti YD p. 118 1.14-16 (Pandeya) / p. 225 L.15-19 (Wezler & Motegi): allagande chabdatanimdirdd akasam ekagunam / sabdaspursagular pardaland dig vayuh / faldasparsarapagurad nipatanmarrar triguntam Tejah/sahdaspx/Surarsagundd rasatanmatrac cafurgund dpah / Sallasparsarumasagandhagweld gandhalanmdird! pancagund prohivi / Cited and translated in Bronkhorst, 1994 311. Elsewhere the Yuktidepika (p. 119 1.25-26 (Pandeyal / p. 227 1.27 Wezler & Molegi)) tells us that the farmarras are not appeased, terrible, or folish". and therefore free from the characteristics of the three constitueix e l of matter 'YD p. 118 1.13-14 (Pandeya) / p. 225 1.14-15 (Wexler & Motegik churches bhuteblya ekoftarandi bhatavisesandm wipurih. 10 Bronkhorst, 1994. 11 See Houben, 1994: 3 f., along with note 7. 1 Vkp 1.76-79: sphotarydbhinnakalasya divanikaldwinuh/ guhar i bede weibhedam pracaksare Il svabhavabhedan nirane hrawdivupludswiki dhwaneh kalah sabdasyety upacaryale Il sabdarva grahane her prakti nis isvate / sthitibhedanimittavam vaiktah pratipaute I Mabuswanth al t er viribhedam walkiah / dhvanayah sam pohante d i ir hiduell Or. "due to the specific accidental feature which is the grasping". 14 YD p. 117 1.32-p. 118 LI (Pandeya) / p. 224 1.25-27 Werler & Mineix a Sabdajdryabhede 'pisati vided wdatanudattasvariwdividus / tasmac chabdatanmatram / en un nedan hidh

Page #6
32 JOHANNES BRONKHORST STUDIES ON BHARTRHARI. X 33 Sarma, Raghunatha (1988): Vakyapadiyam part 1 (Brahma-kandam). With the commen taries Svopajnavrtti by Harivrsabha & Ambakartri by... Raghunatha Sarma. Varanasi: Sampurnanand Sanskrit University (Sarasvatibhavana-Granthamala, 91). Staal, J. F. (1969): "Sanskrit philosophy of language." CTL 5. 499-531. = Parrei. 1976: 102-136. Staal, J. F. (ed.) (1972): A Reader on the Sanskrit Grummurians. Cambridge. Massachusetts, and London, England: MIT Press. ABBREVIATIONS " YD p. 117 1.32 (Pandeya) / p. 224 1.25 (Wezler & Motegi): tulyajatiyavisesanupapatteh. 16 VP I p. 142 1.1-3: iha dvividho dhvanih prakrto vaiktas ca / tatra prakrto numa yena vina sphorurupam anabhivyaktam na paricchidyate / vaikstas tu yenabhivyaktam sphotarupam punah punar avicchedena pracitataram kalam upalabhyate / 17 VPI p. 144 1.1-5: tad yatha prakaso janmanantaram eva ghatadinam grahane hetuh, avatisthamanas tu grahanaprabandhahetur bhavati, evam abhivyakte sabde dhvanir uttarakalam anuvartamano buddhyanuvrttim sabdavisayam visayabhivyaktibaladhanad upasamharati / tasmad upalaksitavyatirekena vaikstena dhvanina samsrjyamano 'pi sphotatma tadrupyasyanadhvaropat sastre hrasvadivat kalabhedavyavaharam navaturati 1. Cp. lyer, 1965: 80. 18 VP I p. 141 1.3-5: tena ca sphotavisayena grahanenopadhina bhinnakalena prakalpitabhedah sphotasya drutamadhyamavilambita vrttayas tribhagotkarsena yuktah samakhyayante / 19 Cp. lyer, 1965: 78-79. 20 SK 34a: buddhindriyani tesam panca visesavisesavisayani. Note that the commentaries limit this ability to see the tanmatras to the gods and accomplished yogis. This may find its explanation in the changes the system had undergone. 21 Note that Raghunatha Sarma (1988: 131 1.28-29), who does not mention the link with Samkhya, feels obliged to explain the term vaiksta as synonymous with vikrta: vikrta eva vaiksta iti prajnaditvat svarthe 'nprayayah. 22 VP I p. 142 1.16-21: prakrtasya iti / dhvanisphotayoh prthaktvenanupalambhat tam sphotam tasya dhvaneh prakrtim iva manyante / tatra bhavah prakrtah/ taduttarakalabhavi tasmad vilaksana evopalabhyata iti vikarapattir iva sphotasyeti vaikrta ucyate / dhvanisamghatasya va prakrtih karanabhighatah / tatah prathamato bhavah prakriah, tatas tu vaikrtah/ 23 Implicit reference to P. 4.3.53: tatra bhavah. ABORI Annals of the Bhandarkar Oriental Research Institute. Poona ANISt Alt- und Neuindische Studien, Hamburg CTL Current Trends of Linguistics, ed. Th. A. Sebeok. The Hague EIP The Encycopedia of Indian Philosophies, ed. Karl H. Potter et al.. Delhi 1970 ff. HOS Harvard Oriental Series, Cambridge Mass. IT Indologica Taurinensia, Torino SK Samkhya Karika TPS Transactions of the Philological Society, Oxford Vkp Bhartrhari, Vakyapadiya, ed. W. Rau, Wiesbaden 1977 VP I Vakyapadiya of Bharthari with the Vitti and the Paddhati of Vrsabhadeva, ed. K.A. Subramania Iyer, Poona: Deccan College. 1966 YD Yuktidipika. 1) Edited by Ram Chandra Pandeya. Delhi: Motilal Banarsidass, 1967. 2) Critically edited by Albrecht Wezler and Shujun Motegi, Vol I, Stuttgart: Franz Steiner, 1998. (ANISI 47.) WZKS Wiener Zeitschrift fur die Kunde Sudasiens, Wien REFERENCES Section de langues et civilisations orientales Universite de Lausanne BFSH 2 CH-1015 Lausanne Aklujkar, Ashok (1990): "Bhartrhari." EIP 5, 121-172. Bronkhorst, Johannes (1994): "The qualities of Samkhya." WZKS 38 (Orbis Indicus, Festschrift G. Oberhammer), 309-322. Brough, John (1951): "Theories of general linguistics in the Sanskrit grammarians." TPS, 27-46. Reprint: Staal, 1972: 402-414. Deshpande, Madhav M. (1997): Saunakiya Caturadhyayika. A Pratisakhya of the Saunakiya Atharvaveda, with the commentaries Caturadhyayibhasya, Bhargava. Bhaskara-Vrati and Pancasandhi; critically edited, translated & annotated. Cambridge, Massachusetts and London, England: Harvard University Press (HOS, 52.). Houben, Jan E. M. (1990): "The sequencelessness of the signifier in Bhartrhari's theory of language." IT 15-16 (1989-90), 119-129. Houben, Jan E.M. (1994): "Bharthari's familiarity with Jainism." ABORI 75, 1-24 & 255-256. lyer, K. A. Subramania (tr.) (1965): The Vakyapadiya of Bhartrhari with the Vrtti, chapter 1. English translation, Poona: Deccan College Postgraduate and Research Institute. Parret, Herman (ed.) (1976): History of Linguistic Thought and Contemporary Linguistics. Berlin New York: Walter de Gruyter. Raja, K. Kunjunni (1969): Indian Theories of Meaning. Second edition. Adyar, Madras: The Adyar Library and Research Centre, 1977.