Sambandha And Abhisambandha

Author(s): Ashok Aklujkar
Publisher: Ashok Aklujkar
Catalog link: https://jainqq.org/explore/269561/1

Page #1
ASHOK AKLUJKAR SAMBANDHA AND ABHISAMBANDHA* 1.1. On the first page of his 1967 monograph Word Order in Sanskrit and Universal Grammar, J. F. Staal remarks: "The distinction which is relevant in the present context is that between sambandha 'the relation of one word to another within a sentence (as shown, e.g., by grammatical inflexion' and abhisambandha, anupurvya or anupurvi 'the order or arrangement of words (as occurring in actual utterances). This distinction could be specified within a general theory of language; in India it served to delimit the scope of the science of grammar or vyakarana itself. For, though it is sometimes said that the Sanskrit grammarians were interested in grammar but not in syntax, it would be misleading to interpret this as asserting that they were interested in words (pada) and not in sentences (vakya) (cf. Renou 1960, 66; 1961, 129). It would however be correct to say that the Sanskrit grammarians were interested in sambandha and not in abhisamband[ha, and accordingly in grammatical relations but not necessarily in word order." 1.2. It is evident from this citation as well as his remarks in the following pages that Staal attaches considerable importance to the distinction between sambandha and abhisambandha. In the present paper I wish to contend that no distinction of the kind Staal makes exists, that the distinction had no role in delimiting the scope of Grammar (vyakarana) and that it is wrong to set up abhisambandha as a synonym of anupurvi and anupurvya.3 2.1. I do not know what the source of Staal's distinction is. The commonly used dictionaries of Sanskrit in general and of Sanskrit sastras (the Vacaspatya, Sabda-kalpa-druma, Nyaya-kosa, Mimamsakosa, etc.), particularly of Sanskrit grammar (Abhyankar, Renou), do not contain any observations that will support the distinction. True, the ordinary Sanskrit term sambandha, basically meaning 'binding Journal of Indian Philosophy 17: 299-307, 1989. (c) 1989 Kluwer Academic Publishers. Printed in the Netherlands.

Page #2
300 ASHOK AKLUJKAR together, tying, connecting, relation, can contextually mean 'grammatical relation, syntactic connection' and is naturally and frequently used in that sense, but no sense of abhisambandha distinct from this sense of sambandha has been noticed by the compilers of dictionaries and kosas. 2.2. Nor has Staal pointed out any passages in which sambandha and abhisambandha are juxtaposed and a distinction of their senses is indicated. If the two concepts were important, one would expect them to be set apart at least in a passage or two in the long and rich tradition of linguistic theorizing and observations for which India is justly famous among specialists. On the contrary, we find that while explaining his Trikandi K 1.67, prak samjninabhisambandhat samjna. rupa-padarthika, Bhartr-hari (V 1.67) writes yavat samjnina sarjna na sambaddha tavan na samjnipadarthika and indicates that there is no significant difference between abhisam + bandh and sam + bandh. We also notice that Nagesa quite casually shortens Patanjali's phrase yathestam abhisambandhah to yathestai sambandhah in Uddyota 1.1.58. 2.3. The contexts in which abhisambandha or other derivatives of abhisam + bandh (badh) occur do not support the distinction either. Whether one looks at the older uses such as sistan sistabhisambandhan manino 'navamaninah (Maha-bharata 12.57.23) or baijikad abhisambandhat (Manu-smrti 5.63) in non-technical contexts outside Grammar or at the occurrences in early commentarial literature of Grammar such as the Mahabhasya, B's Mahabhasya-tika or the Tripadi, and the Trikandi or Vakyapadiya, one finds abhisambandha employed in the sense 'connection,' occasionally acquiring, on the strength of the context, a specific shade of meaning such as 'grammatical connection' or 'semantic connection.'5 In fact, in sentences such as vyavaya-sabdasya pratyekam abhisambandhah "The word vyavaya is connected with each (word preceding it in the compound at-ku-pvan-num-vyavaye of Panini 8.3.6),' quoted in the St. Petersburg Worterbuch, it is impossible to assign the sense 'word order to abhisambandha. Generation of an imagined clarificatory construction like ad-vyavaye, ku-vyavaye, pu-vyavaye, ar-vyavaye, num-vyavaye ca is

Page #3
SAMBANDHA AND ABHISAMBANDHA 301 not what we normally mean by "word order." Such a generation amounts only to recognizing the intended relationships of words. 3.1. Perhaps Staal has inferred the distinction on the basis of two passages from Pa which are embedded in his presentation as follows: P. 28: "We need not pay attention to the word order in the sutra, says Patanjali, because: neha prayoganiyama arabhyate 'restrictions on usage are not here clung to.' This he explains as follows: samskrtya samskrtya padany utsrjyante. tesam yathestam abhisambandho bhavati. tad yatha ahara patram, patram ahareti 'words are generated in accordance with grammatical rules, but their order Jabhisambandha) is free, as in ahara patram and patram ahara "fetch the bowl"' (ed. Kielhorn, I, 39, lines 18-9)." P. 32: "... he = Patanjali] quotes in support Katyayana's varttika (- Orttikal ... anupurvyena samnivistanam yathestam abhisabandhah sakyate kartum 'any desired order may be established between words arranged in a particular succession' (ed. Kielhorn, 1, 152, lines 24-5; quoted Renou 1957, 57). "The opponent raises a further objection: na caitany anupurvyena samnivistani 'but these are not arranged in a particular succession'. To which the reply is: ananupurvyenapi samnivistanam yathestam abhisambandho bhavati 'free word order may also be established between words not arranged in a particular succession'." 3.2. Here, Staal does not tell us why an etymologically probable meaning like 'connection' (grammatical or semantic) of the word abhisambandha would not fit the context - why one has to accept 'order' as the meaning. Nor does he explain, in the case of the second passage, what essential difference there could be between grammatical or semantic connecting of words, on the one hand, and establishing "desired order ... between words arranged in a particular succession, on the other 3.3. The context of the first passage is that of accounting for the order of words seen in Panini's rule vrddhir adaic as Staal rightly explains; given Panini's usual practice, one expects the rule to read adaij vrddhih. Pa's words relevant to our discussion are: prayoga-niyamartham tarhidan syat vrddhi-sabdat para adaicah prayoktavya iti. neha prayoga-niyama arabhyate. kim tarhi. samskrtya samskrtya padany utsrjyante. tesam yathestam abhisambandho bhavati. tad yatha ahara patram, patram ahareti. The only explanation of this passage that would make a contextually acceptable sense is: 'Or, this (thought of Panini) that adaic are

Page #4
302 ASHOK AKLUJKAR to be employed after the word vrddhi may be for the purpose of restricting usage i.e., for indicating to the reader that certain sequences are to be followed in the employment of derived expressions).? To this the response is: Restriction on usage (or institution of a certain sequence) is not undertaken here. Sentence-usable words, which have undergone derivation, are released (made available) one after another. They come to be connected as desired as can be illustrated with ahara patram and patram ahara.' In other words, the point of the passage is not the general one that there is free word order in Sanskrit, but the rather specific one that free word order is presupposed in Panini's grammar and hence the specific sequence in vrddhir adaic cannot be used to infer that Panini wanted to connote to his reader that the reader adhere to a specific sequence (or a set of specific sequences) in using the expressions the Astadhyayi would derive. To attribute such an intention to Panini would run counter to Panini's presupposition. Hence, an explanation other than the one assuming a suggestive act of the described kind on the part of Panini should be sought for the deviant order in vrddhir adaic.10 3.4. If Pa's statement rested at asserting free word order in Sanskrit and did not make the further point that such freely ordered words come to be associated with each other as desired and that this is the way Panini presupposes his grammar to function, Pa would have conceded the objector's point and accepted that the specific order seen in vrddhir adaic has no purpose is the result of a whim) and that Panini has deviated from his usual practice for no reason.11 3.5. The preceding discussion should serve to establish that 'order' is not contextually warranted as a translation of abhisambandha in the case of the first passage from Pa quoted by Staal. That abhisambandha cannot mean 'order' in the second passage has already been indicated (3.2).12 If its juxtaposition in that passage is to be proper, it must be something different from anupurvya which has a well-established and etymologically justifiable sense of 'order, a certain sequence, a sequential arrangement,' etc. Staal (p. 32) too apparently recognizes this and attaches the sense 'word order in general' to abhisambandha

Page #5
SAMBANDHA AND ABHISAMBANDHA 303 in the case of the second passage. Then, however, his initial statement declaring equivalance of abhisambandha, anupurvi, and anupurvya must be deemed to be ill-considered (see note 1). Also, the specific sense newly attached to abhisambandha does not fit the first MB. passage. A statement of the form "Of the derived words, word order in general comes about as in ahara patram and patram ahara" either does not make sense or can only be taken to mean that a grammatical or semantic relationship comes about among the words delivered by the Astadhyayi. Furthermore, if there is no evidence to support the understanding of abhisaribandha as 'order' or 'word order,' there is even less evidence to support its understanding as 'word order in general,' unless "word order in general" is simply another way of conveying the notion 'a grammatical or semantic connection." 4.1. I suppose it is abundantly clear from the preceding paragraphs that no technical or grammatical distinction of any kind exists between sambandha and abhisambandha and that it is wrong to specify lack of interest in abhisambandha as something that had a delimiting influence on Grammar. It is more probable that the Sanskrit Grammarians, either beginning with Panini or under the influence of Panini, generally entertained a different notion of what Grammar was supposed to achieve, particularly about the kind of sentences Grammar was expected to derive (Aklujkar 1988: note 5), and that it is this notion which resulted in an absence of detailed and sustained studies. of the phenomenon of word order. 4.2. The appearance of, and to some extent the preference for, adbhisambandha when its equally non-technical colleague sambandha can convey its meaning seems to be due to the sensitivity which early Sanskrit authors had for the shades of meaning expressed by the upasargas or prefixes. As the situation was probably perceived as one in which word 'X' turned to thing 'x' or word 'Y' for effecting a connection, it was perhaps felt that an addition of the prefix abhi, which indicates 'facing' or 'looking in the direction of,' was appropriate. There is some evidence to this effect in Pa's and B's usage. A preference for abhisam + bandh, over the simpler sam + bandh, may

Page #6
304 ASHOK AKLUJKAR be detected in the following sentences which are concerned with situations of identification and in which a movement or turning of the signifier toward the signified seems to have been presumed: nityo hy arthavatam arthair abhisambandhah MB p. 1.7; etad iti canenasyam evabhisambadhyate TP on MB p. 1.61, ahnika 4 p. 142; yo go-sabdah so 'yam pinda ity abhisambandhe sati V 1.55, prak samjninabhisambandhat samjna rupa-padarthika K 1.67, so 'yam ity abhisambandho buddhya prakramyate yada K 2.40, so 'yam ity abhisambandhad rupam ekikrtam yada K 2.128, yenarthenabhisambaddham abhidhanam prayujyate K 2.160, sabdantarabhisambandhenagnir manavako, gaur vahika iti V 2.251, navasyam te 'bhisambaddha sabda jneyena vastuna K 2.333, yathaiva samudaya-sva-rupasya samjninabhisambandhah V 2.356, yadi samjnabhisambandhat pran natvam tad alaksanam/ athordhvam abhisambandhad anityatvam prasajyate // V 2.364, vrddhyadinam ca sastre 'sminchaktyavaccheda-laksanah / akrtrimo 'bhisambandho visesana-visesyavat // K 2.369, samjnaya ... samudayo 'bhisambadhyate V 2.283, etc. NOTES * The main point of this article was briefly stated at the Sanskrit syntax session of the South Asia Language Analysis conference held at Ithaca and Syracuse in June 1987. The few abbreviations employed in the body of the article are explained in the bibliography. ! Contradicting what he says here or extricating himself from a possible charge of contradiction, Staal makes the following remark on p. 32, while discussing Patanjali 1.1.58 (p. 1.152 lines 24-5): "The term anupurvya is here used to refer to a given particular succession of words; the term abhisambandha for word order in general ..." As I point out below in section 3.5., the distinction which Staal makes here between anupurvya and abhisambandha is as arbitrary as their synonymity which he proclaims initially. 2 "Taxonomy, in this sense, corresponds to the Sanskrit abhisambandha, and structural linguistics can in this sense be characterized by its insistence upon abhisambandha." P. 2. "... the distinction Du Marsais made between syntaxe and construction corresponds to the distinction of the Sanskrit grammarians between sambandha and abhisambandha." P. 13. "... the distinction between the relation of words in a sentence (sambandha), which certainly belongs to the deep structure, and the order of words of a sentence (abhisambandha), which may belong to the surface structure." P. 30.

Page #7
SAMBANDHA AND ABHISAMBANDHA 305 ? I have no difference of opinion regarding the other observations made by Staal in the passage quoted at the beginning. 4 One of Staal's references on p. 32 may be taken by some as suggesting that Renou 1957: 57 is his authority for the distinction. However, as I point out in note 6, the distinction is not found in Renou and Staal's reference should be taken literally the way he makes it - as simply stating that the passage quoted by him is also quoted by Renou or as indicating at the most that he became aware of the passage through Renou's quotation. $ Important MB passages containing abhisambandha are discussed below. The word is found in TP ahnika 1 p. 12, ahnika 4 p. 142, ahnika 7 pp. 303, 305, 306; and in TK V 1.55, K 1.67, K 2.17, V 2.17, K 2.40, V 2.40, V 2.128, K 2.160, K 2.186, V 2.204, V 2.223, K 2.246, V 2.251, V 2.270, K 2.329, K 2.333, V 2.356, V 2.364, K 2.369, V 2.383, V 2.391, K 2.432, K 2.441, K 2.460, K 3.75, and K 3.163. Not to belabour the point, only a few of these will be quoted at the end of this article. 6 The source referred to here is Louis Renou's Terminologie grammaticale du Sanskrit, Paris: Librairie Ancienne, Honore Champion, Editeur. In it, Renou explains abhisambandha as "connexion (d'un mot avec telle notion ou avec tel autre mot dans la phrase, etc.)" and translates the MB passage reproduced by Staal as "pour des choses qui se presentent dans un ordre successif, la connexion entre elles peut etre faite arbitrairement." From this, it is evident that Staal's understanding of abhisambandha is not derived from Renou. See note 4. 7 (a) That is, the rule vrddhir adaic would achieve prayoga-niyama by setting an example of itself, by being a jnapaka, that is, by being suggestive, through its formal peculiarity, of a general principle presupposed by Panini in his grammar. (b) Kaiyata and his commentators (Nagesa and Annam-bhatta) do not reveal how exactly they interpreted the first sentence. They do not use any word like jnapaka in their explanations which would support my explanation. The TP is unavailable for this portion. However, if Pa's remark is not interpreted in the way I suggest and is taken to mean 'Or, this (vrddhir adaic or the specific order of words in vrddhir adaic) would be for the purpose of restricting usage through a statement of the form 'adaic should be used after the word vrddhi," then it would not be even a tentative answer to the specified problem. If understood as applicable to the sutra itself, it would amount to simply pointing out what we already know, namely that adaic is used after vrddhi in the sutra; if understood as applicable to a sentence of the object language, what would be the justification for making a regulation like the following: 'If one wishes to use vrddhi and adaic in a sentence, one should put adaic after vrddhi.'? Is the purva-paksin so unsophisticated as to entertain the possibility that Panini may write rules about individual sentences? How would what he says be even an answer to the problem pointed out in the case of a sentence of the metalanguage? Even if it were to be applicable to a rule, would it not amount to an assertion without any supporting argument, to a restatement of the problem itself? (c) The discussion in (b) should suffice to establish that the following translation by Abhyankar and Shukla (1987), and other similar translations, are not likely to be correct: "Well, then, the present sutra (vrddhiradaic) can be said to be for regulating the use of the words 'vrddhi and adaic in such a way that the words at and aic i.e. the vowels a, ai and au must be used after the word vrddhi and not in the other way, i.e. before the word vrddhi." * This does not mean either that the freedom is absolute or that there are no preferred word order patterns.

Page #8
306 ASHOK AKLUJKAR "As deduced by his commentators from a study of the Astadhyayi. 10 A corrollary of what I state here is that ahara patram patram ahara is primarily not an illustration of free word order in Sanskrit but of the peculiarity of generation of sentence constituents in Panini's grammar of the thesis that Panini presupposes free order, that in Panini's view variation in the placing of words has no bearing (of course, within limits) on words getting connected to each other. While stating that prescription of particular sequences of derived words is not undertaken in Panini's grammar, Pa must presuppose that there can be such a prescription that there can be prayoga-niyama. He must, therefore, be said to have noticed and this is no great surprise or an advance over his predecessors - the phenomenon of specific ordering of words. His comment cannot, therefore, be interpreted to mean that, in his view, there is no preferred or common word order in Sanskrit. 12 Cf. patha-kramad artha-kramo baliyan iti yathestam atrabhisambandhah. Kaiyata 1.1.58, that is, on our second passage here. BIBLIOGRAPHY AND ABBREVIATIONS Abhyankar, K. V. and Shukla Jayadev Mohanlal: 1975, Patanjali's Vyakarana mahabhasya Ahnikas 1-3, with English Translation and Notes. Pune: Bhandarkar Oriental Research Institute. Research Unit Publications, no. 1. Aklujkar, Ashok: 1988, "Some Theoretical Observations on Word Order in Sanskrit", forthcoming. B Bhartrhari. Trikandi or Vakyapadiya. I have reproduced the text of the Vrtti from my critical edition (under preparation) of the TK. Those wishing to verify my references to the V prior to the publication of my edition should consult the editions by K. A. Subramania lyer: (a) Vakyapadiya of Bhartrhari with the Vrtti and the Paddhati of Vrsabha-deva. Poona: Deccan College Postgraduate and Research Institute. 1966. Deccan College Monograph Series 32. (b) The Vakyapadiya of Bhartrhari, Kanda II with the Commentary of Punya-raja and the Ancient Vrtti. Delhi, etc: Motilal Banarsidass, 1983. I have followed the enumeration of karikas in: Bhartrhari's Vakyapadiya, Die Mula-karikas nach den Handschriften herausgegeben und mit einem Pada-Index versehen. Ed. Wilhelm Rau. Wiesbaden: Kommissionsverlag Franz Steiner GMBH. 1977. Abhandlungen fur die Kunde des Morgenlandes XLII, 4. Hence the numbers in my edition and those in the editions by Subramania lyer do not always match. However, they are not far removed from each other. See also "TP" below. K = karika. See B. MB Maha-bhasya. See Pa. Pa Patanjali. Vyakarana-maha-bhasya, ed. F. Kielhorn. 1880-5. Revised third ed. K. V. Abhyankar. Pune: Bhandarkar Oriental Research Institute. Staal, J. F.: 1967, Word Order in Sanskrit and Universal Grammar Dordrecht Holland: D. Reidel Publishing Company. Foundations of Language Supplementary Series, vol. 5. TK Trikandi. See B.

Page #9
SAMBANDHA AND ABHISAMBANDHA 307 TP = Tripadi or Maha-bhasya-tika, published under the title Mahabhasya-dipika of Bhartshari. Eds. K. V. Abhyankar and V. P. Limaye. Poona: Bhandarkar Oriental Research Institute. Part I and II, 1967-70. Postgraduate and Research Depart ment Series. No. 8. V = Vrtti, See B. University of British Columbia