Paninis View Of Meaning And Its Western Counterpart

Author(s): Johannes Bronkhorst
Publisher: Johannes Bronkhorst
Catalog link: https://jainqq.org/explore/269635/1

Page #1
Panini's View of Meaning and its Western Counterpart Johannes Bronkhorst University of Lausanne Panini's name is widely known among linguists these days. Many know that this ancient Indian linguist wrote a grammar which has been described as "one of the greatest monuments of human intelligence" by Leonard Bloomfield (1933:11). This admiration concerns Panini's analysis of the Sanskrit language. The semantic aspect of Panini's grammar, on the other hand, is a relatively neglected topic. Yet it is not without importance. It has become clear in recent years that meanings are the input of Panini's grammar (Buiskool 1939:16; Kiparsky & Staal 1969:84; van Nooten 1969:244; Bronkhorst 1980, 1982:303; Joshi & Roodbergen 1980: viiif.). That is to say, Panini's grammar produces correct utterances on the basis of a set of meanings. These meanings give rise to grammatical elements - morphemes if you like which undergo further transformations until, in the end, they are all joined up to form a sentence. I say sentence on purpose, because the end result of a Paninian derivation is not normally a word. Of course, occasionally a sentence consists of a single word, as when other words have fallen prey to ellipsis. To keep the discussion simple, I'll briefly describe, in outline, the Paninian derivation of a single word, dasyami. This word means "I shall give"; it expresses, for Panini, the meanings "giving", "future", "singular number", and "first person". To be exact, the last of these meanings, "first person", is not really a meaning in Panini's scheme. This will become clear in the derivation. As said before, these meanings are the input. The meaning "giving" gives rise to the element da, in accordance with the traditional list of verbal

Page #2
456 JOHANNES BRONKHORST PANINI'S VIEW OF MEANING 457 roots, plus meanings, which accompanies Panini's grammar (Dhp. 1.977). Subsequently the meaning "future" justifies the suffix lrt (P. 3.3.13 with bhavisyati from 3.3.3). Irt in its turn requires the augment sya (P. 3.1.33) and is itself replaced by mi (P. 3.4.78). This last replacement requires the meaning "singular number" (P. 1.4.102) and the presence of the pronoun 'T' - or rather, its Sanskrit equivalent (P. 1.4.101 with 107). It is however specified that this pronoun does not have to be present bodily; in the case of ellipsis of the pronoun the ending mi can be maintained. The whole derivation can be visualised as follows input da "giving" da-lry "future" da-sya-Irt dd-sya-mi "singular number" presence of pronoun T The lengthening of a in dasyami, finally, is a purely phonological process which need not detain us here. As a matter of fact, the above example has been chosen partly because it contains a minimum of phonological transfor. mations, unlike many other derivations. It will be clear that the meanings of dasyami are divided as follows over the constituent parts: da - sya - mi "giving" 1st p. sg. This is in itself not very remarkable. Panini's grammar must be under stood in the context of its time and culture. The Brahmanical texts which we possess from that time, and which or at least some of which were known to Panini, contain many 'etymologies'. These are not etymologies in the modern sense of the term, but rather attempts to reach the essence of the thing denoted on the basis of the constituent parts of the denoting word. To be more precise, it was assumed that similar words denote similar things; similar parts of otherwise different words refer to similar aspects of otherwise different things. This approach to language found its classic exposition in a work called Nirukta, which used this approach to find the meanings of unknown words. The Nirukta dates from roughly the same period as Panini's grammar; most probably it is somewhat younger than the latter. (On the shared concerns of Panini and the Nirukta, see Bronkhorst 1981; on the chronological relationship between the two, see Bronkhorst 1984:8-9.) As said before, Panini's grammar must be understood against the background of the etymologies in the Brahmanical literature of his age. The Brahmanical etymologists looked upon the constituent parts of words as the real meaning bearers; Panini, as we saw, did the same thing I would like to add one more observation at this point. In Panini's grammar the ultimate meaning bearers are not always identical with the constituent parts of the words to be precise, they are not always identical with the constituent parts of the surface forms of the words. In the course of a derivation grammatical elements are not infrequently replaced by others. The ending am in afrausam, for example, replaces mi (as in dasydml) in the course of the derivation. Other elements are dropped in the derivations: the 3rd p. sg. Aor, avddit "he spoke", for example, lost the Aorist markers, which is visible in the plural avadisuh. The level of language, therefore, which directly expresses the meanings in accordance with Panini's grammar, is not the surface form, but a deeper level, information about which is provided by Panini's grammar. I shall return to this point later in this paper. I am not going to say more about Paninian derivations. All I wish to add is that the influence of Panini on later Indian thought on language and meaning has been enormous. Part of later Indian thought on meaning can be considered an elaboration of the basis provided in Panini's grammar: it discusses the relative importance of the Paninian meaning elements in a sentence, and deals with questions like what priority they take in the "ver. bal understanding" (sabdabodha) which results from a sentence. "future" A similar derivation gives rise to the form afrausam "I heard". The meanings are here divided as follows: a - frau - $ - am "hearing" 1st p. sg. "past" It is not possible now to study Panini's treatment of meaning in greater detail. The derivations which we have discussed so far allow us already to discern some of its basic presuppositions. Linguistic utterances, be they sentences or words, are agglutinations of expressive elements. The meaning of the utterance is the accumulation of a number of 'primitive' meanings.

Page #3
458 JOHANNES BRONKHORST It is true that others rejected the individual grammatical elements and their separate meanings and regarded them as mere grammatical fictions. One thinks here in the first place of the linguistic philosopher Bhartthariwho may have lived in the 5th century C.E. The true meaning bearing unit, according to Bhartrhari, is the sentence. In the present paper these and other opinions are of no interest to us. I have briefly introduced Panini's view of meaning and propose now to turn to a Western parallel.. The idea of ascribing meanings to the constituent parts of words is not unknown to the West. We find it already in Plato's dialogue Cratylus, and in a number of later authors down to the 19th century. It is not my intention to discuss the history of these ideas in the West, which were frequently, as in India, connected with some kind of etymologizing. I shall rather turn immediately to one of its last representatives, viz., Franz Bopp (1791-1867). Bopp is best known as the founder of comparative Indo-European linguistics. Due to his influence this branch of linguistics embodied, until the 1860's, ideas which show remarkable similarities to those which we discussed in the context of Panini (Kiparsky 1974). Briefly stated, Bopp came to consider the Indo-European languages, and Sanskrit in particular, as agglutinations of meaningful elements. To quote Kiparsky: "Bopp actually seems to have held that in the proto-language the primitive semantic elements were by and large expressed by separate morphemes." (Kiparsky 1974:176.) "A historical explanation of an inflected form was to him a demonstration that the form was derived from a proto-form in which each of the primitive concepts into which its meaning was analyzable was expressed by a separate morpheme." (Kiparsky 1974:177.) The meaningful elements were divided into two groups: verbal roots and pronominal roots. Together these two types of roots account for the declensions and conjugations of words. The roots, according to Bopp, represent the 'primary ideas'. The question why the primary ideas are expressed by those roots and not by others, Bopp doesn't dare to address; he speaks in this context of 'the secret of the roots' (Bopp 1845:V). But it is clear to him that 'the whole body of fundamental ideas' is expressed by the monosyllabic roots (Bopp 1845:97). In the oldest state of the Sanskrit family of languages, he explains, "the root appears as a circumscribed nucleus which surrounds itself with foreign syllables... whose destination is, to PANINI'S VIEW OF MEANING express the secondary ideas of grammar which the root cannot express" (Bopp 1845:98). 459 Regarding the formation of verbs, Bopp observes: "Languages of a structure similar to that of the Greek, Latin &c." needless to add that this includes Sanskrit "can express by one verb... a whole logical proposition, in which, however, that part of speech which expresses the connection of the subject with its attribute, which is the characteristic function of the verb, is generally omitted or understood." (Bopp 1820:23) Some verbal forms, however, preserve all their essential elements, and Bopp's analysis of asrausam "I heard" illustrates well his ideas (Bopp 1820:51). Bopp divides this form into three parts: - asrau-s-am asrau, according to Bopp, is an organic modification of the verbal root sru "to hear", with the same meaning. s represents the verbum abstractum "to be". am represents number and person, and as Bopp came to think soon - is really a pronoun, a pronominal root. For Bopp, then, the finite verb asrausam is an agglutination of three 'roots', each of which carries its own meaning; their combination means something like "I am hearing". Aggluti nation - Bopp writes to Humboldt in 1820 is really the spirit of the Sanskrit language ("Zusammensetzung, sowohl in den fruhesten Elementen als in spaterer Wortbildung, ist wirklich ganz der Geist der SanskritSprache." Lefmann 1897:7, letter of 5.3.1820; cited by Sternemann 1984:22). Another example is dasyami "I shall give". Bopp analyses it as follows (1820:47f.): da-s--a-mi Here da is the verbal root "giving", s is the copula, mi the personal pronoun, while al signifies "to wish". Which were the influences that induced Bopp to view the Indo-European languages, and the Indo-European proto-language in particular, in this special way? It has been argued "that Bopp's fundamental conceptions are derived from the theoretical-mathematical scientialistic rationalism of which Leibniz was the last representative" (Verburg 1950:248; on Leibniz's ideas about language, see Schulenburg 1973). Others have pointed at the similarity of ideas, with Johann Christoph Adelung (1732-1806), the first volume of whose Mithridates the only volume he could complete himself

Page #4
460 JOHANNES BRONKHORST PANINI'S VIEW OF MEANING 461 - appeared as late as 1806 (Arens 1969:150-51, 178). Bopp's fundamental division of languages seems to point to a connection with Adelung. Bopp distinguishes three classes of languages, the first two of which he describes as follows: [The first class comprises Chinese, where all is hitherto bare root, and the grammatical categories, and secondary relation after the main point, can only be discovered from the position of the roots in the sentence. Sec ondly, languages with monosyllabic roots, which are capable of combina tion, and obtain their organism and grammar nearly in this way alone. The chief principle of the formation of words, in this class, appears to me to lie in the combination of verbal and pronominal roots, which together repre. sent, as it were, body and soul. .... To this class belongs the Sanskrit fam ily of languages... (Bopp 1845:102-103) This classification or better: this part of the classification - finds a close parallel in Adelung's Mithridates, where, especially in the Vorrede and Einleitung, the development of language from single syllables to composite forms is described. Monosyllabic languages - Adelung mentions Chinese and 'the languages of south-east Asia'-are more primitive and more orig. inal than multisyllabic languages. We have seen that Bopp, too, comes close to the idea of a develop ment of language from monosyllabic to composite. Like Adelung, he seems to look upon the languages of India and Europe as developments out of an original monosyllabic language. Panini's influence on Bopp in other respects is generally recognized. Brough, for example, wrote in 1951: It has ... been recognized that the Paninian analysis of Sanskrit into a system of roots, stems, and suffixes pointed the way to the method which has prevailed in Indo-European studies to the present day. It is true that roots and suffixes were not entirely new concepts to Europe, but it remains doubtful whether the method would have been applied with such thoroughness if it had not been for Panini's example. (Brough 1951:402) This means that Bopp took from Panini the latter's method of analysis, as well as his semantic ideas. Bopp refers often to what he calls 'the Indian grammarians', not infrequently to emphasize a point where he disagrees with them. Some passages indicate however that Bopp did not reject all that the Indian gramma. rians had done; according to him, they had gone far, but not far enough. This is clear, for example, from the following passage, which was first published in 1824: The Indian grammarians carefully observed and completely developed the euphonic effects of terminations and suffixes on the final letters of what precedes; but these grammarians didn't go so far as to consider the grammatical forms themselves as arisen or modified under the influence of euphonic rules. The reason is that they didn't occupy themselves with the origin of grammatical forms. It was enough for them, c... to know and to indicate that an designates the third person plural in the different tenses. They didn't ask where this an comes from, what it means in this form. Had they asked, their means would probably have enabled them to find out that an stands for ant ... (Bopp 1972:2) It is not necessary to emphasize the differences between Bopp and Panini. Their analyses of the same word are often far apart. The few examples which we have considered illustrate this sufficiently. But in spite of these differences, and besides them, are the similarities. Both Panini and Bopp search essentially for the ultimate meaning bearers of the Sanskrit language. Both find them in a hypothetical layer which is not in all respects identical to the surface forms of the language. Bopp identifies this postulated deeper layer with an earlier - perhaps the earliest - form of the language, whereas we have no reason to believe that Pasini entertained such views. The similarities between Bopp's and Panini's approaches to the Sanskrit language allow us to count Panini among the influences that worked on Bopp in this essential aspect. In point of fact, from among the It is not my intention to give a verdict on the question which authors influenced Bopp: Leibniz, Adelung, both, or neither. The problem is that we do not know for certain that Bopp read either of these two authors. We do know, however, that he was aware of the work of the Indian gramma rians. He had become acquainted with it through the existing Sanskrit grammars written by Europeans, among which I like to mention the one by Colebrooke (1805). In point of fact, Bopp was the first who "made the grammar (of Sanskrit), formerly a maze of Indian subtlety, as simple and as attractive as that of Greek or Latin" (Martineau 1867:203). We have also seen that from the semantic point of view Panini and Bopp show remarkable similarities. It seems safe to conclude that Bopp's view on the structure of the Indo-European languages was to at least some extent influenced by Panini and his school, the more so since Colebrooke's grammar preserves this feature of Panini's work.

Page #5
462 JOHANNES BRONKHORST PANINI'S VIEW OF MEANING 403 sources that have been proposed to explain Bopp's ideas on semantics, Panini is the one best established, better than Adelung and Leibniz. Note In the paper the following abbreviations are used: Dhp. Dhatupdtha, as found in Bohtlingk (1887). P Papinian sutra, as found in Bohtlingk (1887). References Adelung, Johann Christoph. 1806. Mithridates oder allgemeine Sprachen kunde. Erster Theil. Berlin: Voss. Arens, Hans. 1969. Sprachwissenschaft. Der Gang ihrer Entwicklung von der Antike bis zur Gegenwart. 2 Bde. Frankfurt/M: Athenaum Fischer. Bloomfield, Leonard, 1969 (1933). Language. London: George Allen & Unwin. Bohtlingk, Otto. 1971 (1887). Panini's Grammatik. Hildesheim & New York: Georg Olms. Bopp, Franz. 1889 (1820). "Analytical Comparison of the Sanskrit, Greek, Latin, and Teutonic Languages". Internationale Zeitschrift fur Allgemeine Sprachwissenschaft 4.14-60 (Reprinted from Annals of Oriental Literature 1820, 1.1-65) - 1845. A Comparative Grammar of the Sanscrit, Zend, Greek, Latin, Lithuanian, Gothic, German, and Slavonic Languages. Transl. from the German principally by Lieutenant Eastwick. Conducted through the press by Horace H. Wilson. London: Madden & Malcolm. 1972. Kleine Schriften zur Vergleichenden Sprachwissenschaft. Gesammelte Berliner Akademieabhandlungen 1824-1854 (Opuscula, Band 5). Leipzig: Zentralantiquariat der DDR. Bronkhorst, Johannes. 1980. "The Role of Meanings in Panini's Grammar". Indian Linguistics 40.146-157. - 1981. "Nirukta and Asfadhyayi: Their shared presuppositions". Indo-Iranian Journal 23.1-14. - 1982. Review of Joshi & Roodbergen 1980. Indo-Iranian Journal 24.302-305. 1984. "Nirukta, Unddi Satra, and Asfadhyayi." Indo-Iranian Journal 27.1-15 Brough, John. 1951. "Theories of General Linguistics in the Sanskrit Grammarians". Transactions of the Philological Society 1951.27-46. (Re. print: Staal 1972:402-414. The page numbering follows the reprint.) Buiskool, Herman E. 1939. The Tripadi. Being an abridged English recast of Purvatrasiddham (An analytical-synthetical inquiry into the system of the last three chapters of Panini's Astadhyayr). Leiden: Brill. Colebrooke, Henry T. 1805. A Grammar of the Sanscrit Language. Volume 1. Calcutta: Honorable Company's Press. Joshi, Shivaram D., & J.A.F. Roodbergen. 1980. Patanjali's Vyakarana Mahabhasya, Vibhakrydhnika (P. 2.3.18 - 2.3.45). (Publications of the Centre of Advanced Study in Sanskrit, Class C, 12.) Pune: Univ. of Poona. Kiparsky, Paul. 1982 (1974). "From Paleogrammarians to Neogrammarians". Kiparsky 1982:175-188. - 1982. Explanation in Phonology. (Publications in Language Sciences, 4.) Dordrecht: Foris. - & Johan F. Staal. 1969. "Syntactic and Semantic Relations in Panini". Foundations of Language 5.83-117. Lefmann, Salomon. 1897. Franz Bopp, sein Leben und seine Wissenschaft, Nachtrag. Briefwechsel zwischen Franz Bopp und Wilhelm von Hum boldt (1819-1835). Berlin: Georg Reimer. Martincau, Russell. 1867. "Obituary of Franz Bopp". Transactions of the Philological Society 12.305-312. (Reprint: Sebeok 1966.200-206.) Schulenburg, Sigfrid von der. 1973. Leibniz als Sprachforscher. Mit einem Vorwort herausgegeben von Kort Muller (Veroffentlichungen des Leib niz-Archivs, 4.). Frankfurt/M.: Vittorio Klostermann. Sebeok, Thomas A., ed. 1966. Portraits of Linguists. A Biographical Source Book for the History of Western Linguistics, 1746-1963. Vol. 1: From Sir William Jones to Karl Brugmann (Indiana University Studies in the His tory and Theory of Linguistics). Bloomington & London: Indiana Univ. * Press. Staal, Johan F., ed. 1972. A Reader on the Sanskrit Grammarians. Cam bridge, Mass.: MIT Press. Sternemann, Reinhard. 1984. Franz Bopp und die vergleichende indoeuropaische Sprachwissenschaft. Beobachtungen zum Boppschen Sprachvergleich aus Anlass irriger Interpretationen in der linguistischen

Page #6
Literatur (Innsbrucker Beitrage zur Sprachwissenschaft, Vortrage und Kleinere Schriften, 33.). Innsbruck. van Nooten, Barend A. 1969. "Panini's Theory of Verbal Meaning". Foun dations of Language 5.242-255. Verburg, Pieter A. 1950 "The Background to the Linguistic Conceptions of Franz Bopp." Lingua 2.438-468. (Reprint: Sebeok 1966:221-250.)