Buddhist Theory of Meaning (Apoba) and Negative Statement

Dhirendra Sharma

Philosophy East and West
Vol.18(1968) pp.3-10
Copyright by University of Hawaii





THE STUDY OF THE problem of word-meaning in Indian philosophy has had a very complex and interesting development, much more interesting than is usually realized. It has probably sustained a more continuous polemics than any other philosophical problem.

          Speculation about the nature of word had a kind of mystical awe in the early writings of India, viz., the .Rq Veda, the Braahma.nas, and the Upani.sads. Word was considered eternal and the prime source of knowledge.[l] From Pata~njali (ca. 200 B.C.), however, we learn that there were two schools, one consisting of those who believed that the referrent of word was the particular, and the other consisting of those who maintained that it was the universal.[2] Vyaa.di, for instance, believed in the former, while Vaajapyaayana held the latter view. However, as we shall see, in the later development of philosophical systems most of the non-Buddhist philosophers synthesized the two views and contended that a word meant both the particular and the universal. The Miimaa^msaa held that a word denoted a genus, and only indirectly referred to the individual.[3] _____________________________________________________________________________

1.Cf. R. C. Pandey, The Problem of Meaning in Indian Philosophy (Delhi: Motilal Banarsidass, 1963), pp. 9-10.

2.Pata~njali's Mahaabhaa.sya, with Kaiya.ta's pradiipa, Naage`sa's uddyota and Rudradhara `Sarma's 'tattvaaloka' (Banares: Kashi Sanskrit Series, 1954), pp. 35, 38.

3.Kumaarila Bha.t.ta, Miimaa^nsaa-`slokauaartikam...`sriimatpaarthasaarathimi`ara-pra.niitayaanyaayaratnaa-karaakhyayaa uyaakhyayaanugatam, Taila^nga Raama`saastrii Maanavallii, ed. Chow-kamba Sanskrit Series, Work 3, nos. 11, 12, 15-21, 24 (Benares: Caukhambaa Sanskrit Book Depot, 1898), aak.rti section, verse 4.




But the Nyaaya maintained that it denoted three things: an individual, its class residing in it, and its particular configuration or form.[4]


          Buddhist thinkers of the Dignaaga school, however, advanced a theory called "apoha" which is somewhat similar to the Western view of nominalism. Apoha literally means "differentiation" or "exclusion." Words are the result of mental conceptualization, and therefore they refer to mental images and cannot be directly associated with external realities. Meaning, thus, denotes the referend, the instrument of an act of reference, as distinct from the referent, the object toward which the act of reference is directed.[5] The Buddhist regards it as only a logical concept, not an external entity inherently residing in the individuals. In other words, meaninq means the relation of the word and the image of the object, The word cannot directly be associated with external objects; it cannot, therefore, denote the object. The word has an a priori existence, independent of external objects.


          The Sanskrit term "artha" is as ambiguous as the English term "meaning." Of the two main views of meaning, one is realist and the other, with some reservations, may be called idealist and nominalist. The division is based on the interpretation of the tatpuru.sa compound `sabdaartha.[6] It means "meaning of word." The term "artha" conveys three things: (1) purpose, (2) cause/ means, and (3) object of the senses.[7] The realists seem to take the term in its third sense, while the nominalists prefer the other two meanings. The realist group includes the Nyaaya,[8] the Vai`se.sika,[9] and the Miimaa^msaa[10] systems, all of which maintain that words denote both universals and individuals, and that


4. Nyaaya Suutra of Gautama, II. 2. 68.

5.There still exists some confusion in the terminology of reference, and the term "referend" is used by some authors to denote the "object" instead of the "instrument" of the referential act. Cf. Ledger Wood, Dictionary of Philosophy (Ames, Iowa: Littlefield, Adams and Company, 1959), p. 267.

6.Tatpuru.sa or dependent determinative is one in which the first member of the compound depends on the last, the syntactical relation of the former to the latter being that of an attribute in an oblique case.

7.Cf. Monier Williams, Sanskrit-English, Dictionary (New ed.; Oxford: The Clarendon Press, 1960), p. 90.

8.Cf. n. 4 above.

9.Vai`se.sika Suutra, I. 2. 3. ff.

10.Kumaarilla Bha.t.ta, op. cit., apoha section.




both are real objects to be grasped by the senses.[11] The Buddhist thinkers of the Dignaaga, school vehemently opposed the realist view on the ground that universals are not external facts--they are Post res.

          Buddhist ingenuity is shown in the argument that the efficient reality (arthakriyaakaaritva) belongs to the extreme momentary particular (k.sa.nika) and that it is this momentary reality that is grasped at the first moment of sense-stimulus. What is cognized by the intellect following the first momentary sense-stimulus is the universal; and it is this conceptualized fact that is apprehended in inference and referred to by words. If words mean the objects of the senses, our experiences of language would be the same as those of the sense-object-contact in perception. Then, the mere pronouncement of words, for instance, honey and fire, would produce efficient effects of sweet taste and burning sensation.[l2]


The theory of apoha is designed by Buddhist philosophers to solve the problems of the universal (saamaanya) and the particular error (bhedaagraha), the relation between substance and attribute (dharmi-dharma), and the word and its meaning (`sabdaarthasambandha ).[13] The theory seems to be misunderstood as "a negative approach towards meaning."[l4] The charge of "negativism" appears to have been based on the study of non-Buddhist scholars Uddyotakara, Kumaarila Bha.t.ta, Bhaamaha, and Udayana, who "vehemently criticized" the theory as negative.[15] In fact, the charges of "negativism" have been constantly _____________________________________________________________________

11.J. Sinha, History of Indian Philosophy (Calcutta: Sinha Publishing House, 1956), Vol. I, p. 321.

12.Apoha-siddhi of Ratnakiirti, Haraprasad Shastri, ed. In Six Buddhist Nyaaya Tracts (Calcutta: Asiatic Society of Bengal, 1910), p. 9, 12-13. Another edition of Apoha-siddhi has been published with nine other logical monographs of Ratnakiirti entitled: Ratnakiirtinibandhaaualii, Anantlal Thakur, ed. Tibetan Sanskrit Works Series, III (Patna: Kashi Prasad Jayaswal Research Institute, 1957), pp. 53-61.

13.Cf. Th. Stcherbatsky, Buddhist Loqic (Leningrad: Academy of Sciences of the USSR, 1932). Reprinted in Indo-lranian Journal ('s-Gravenhage: Mouton & Co., 1958), Vol. II, pp. 403 ff.

14.Cf. K. K. Raajaa, "The Theory of Meaning According to Buddhist Logicians," Adyar Library Bulletin, Vol. XVIII, Parts 3-4, p. 11, also pp. 3-13; Sinha, op. cit., pp. 331-333, 883; R. C. Pandey, op. cit., p. 219.

15.Raajaa oP cit., p. 11. Raajaa's Indian Theories of Meaning (Madras: The Adyar Library and Research Centre, 1963), reached me after I had completed this paper. There is an excellent review of the book and a survey of the study of semantics in classical Sanskrit writings by J. F. Staal, "Indian Semantics, I," Journal of the American Oriental Society, 86, No. 3 (July-September, 1966), 304-311. My position, however, remains unchanged, for Raajaa has not altered his view of apoha and Staal shows no concern with the Buddhist theory of meaning.




refuted by Dharmakiirti and subsequent Buddhist scholars.[l6] Here we shall deal mainly with one aspect of apoha, namely, the extent to which it is based on the law of opposition (virodha).[17] We shall also attempt to establish that a purely negative meaning theory would be contrary to the Buddhist metaphysics of extreme momentary particular (svalak.sa.na k.sa.nika).


          The apoha theory is directed primarily against the pluralistic conception of reality according to which universals are considered to be real. For the nominalist Buddhist, a real is the extreme point instant (svalak.sa.na) which is beyond propositional operation.[18] Everything past, future, imagined, absent, mental, notional, and universal--that is, every thought construction--is unreal. Thus, the object of a judgment or expression, that is, the propositional operation, is not the momentary real (k.sa.nika) which is in constant flux.[19] Thus, all verbal and logical statements express "differentiation" (apoha).[20] To the realist argument that it is really the universal which is the object of a proposition, the nominalist rejoins that the universal itself is not real but a logical construct (vikalpa). It must be conceived as the idea of exclusion of a common counter-correlate. For instance, the common counter-correlative of all cows is non-horse. Thus, the concept "cow" can be determined by excluding all other instances of reality from which it is excluded. "The universal is in its very essence (lak.sa.na) exclusion of the other."[21]


          For Buddhist thinkers negation is based on a .priori opposition (virodha) of unique momentary particular entities. All negations, according to Dharma-


16.Pramaa.nauaarttika of Dharmakiirti, the first chapter only with "auto" commentary suau.rtti, and sub-commentary of Kar.nakagomin svav.rtti-.tiikaa, Raahula Saankrityaayana, ed. (Allahabad: Kitab Mahal, 1943), pp. 248-263; J~naana`sriimitra-nibandhaaualii (Twelve Buddhist Philosophical Works of J~naana`sriimitra), ed. with introduction by Anantlal Thakur. Tibetan Sanskrit Works Series, V (Patna: Kashi Prasad Jayaswal Research Institute, 1959), pp. 201-232 on "apoha."

17.An independent study of the Buddhist theory of meaning or apoha by this author is being published by Mouton & Co.

18.Pramaa.na-uaartika-bhaashyam of Praj~naakaragupta, deciphered and edited by Raahula Saankrityaayana. Tibetan Sanskrit Works Series, I (Patna: Kashi Prasad Jayaswal Research Institute, 1953), p.621; Dharmakiirti, loc. cit., p.262; Ratnakiirti, op, cit., p. 6 (14-17).

19.E. Frauwallner, "Die Apohalehre," Wiener Zeitschrift f<u>r die Kunde des Morgen-landes, Vol. 37, 275 and Vol. 40, 63.

20.J~naana`sriimitra, op. cit., p. 201.

21.Praj~naakaragupta, op. cit., p. 200 (III. 30).




kiirti,[22] are rooted in opposition[23] which can be divided into two classes: (1) efficient opposition or incompatibility (sahabhaava-virodha) ; and (2) logical opposition or contradiction (anyonyopalabdhiparihaara-sthitilak.sa.nauirodha). The former is defined in the following passage: "When [one fact] has duration [as long as] the sum total of its causes remains unimpaired and it [then] vanishes as soon as another, [the opposed] fact, appears, it follows that the two are incompatible, [or efficiently opposed], just as are the sensations of heat and Cold."[24]

          The second type of opposition is explained as follows: "There is also [opposition between two terms] when their own essence consists in mutual exclusion, as between the [terms] eternal and non-eternal."[25] Some other instances of the second opposition include such pairs as: reality and unreality, existence and non-existence, affirmation and negation, blue and non-blue.[26] In the first opposition (incompatibility) two facts exist independently without opposing each other. Their opposition becomes efficient only when they are placed together in one time-space relation. On the other hand, in the second opposition' (contradiction) the two opposed facts are so related that neither of the two can be defined or apprehended without excluding the other. The very essence (lak.sa.na) of the one consists in exclusion of the other; for example, blue and non-blue. The first opposition seems to mean negation of terms or entities, as can be seen in the eightfold formula of negative inference explained by Dharmakiirti.[27] The second opposition, contradiction, appears to be designated to refer to the negation of propositions. For the very essence of "non-blue" presupposes the proposition "this is blue," and vice-versa. This propositional opposition, however, poses many logical difficulties which are


22.Dharmakiirti, Svau.rtti, pp. 35-37.

23.In translating "virodha" by "opposition" we have followed Stcherbatsky's exposition of the law of contradiction. See Stcherbatsky, op. cit., p. 187, n. 3.

24.Dharmakiirti, Svau.rtti, p. 35; cf. also Stcherbatsky, op. cit., p. 187.

25.Stcherbatsky's observation thereon is noteworthy. "It is clear that in these words we have a definition of the Law of Contradiction, so much discussed in European Logic from Aristotle through Leibniz, Kant and Sigwart up to the modern logicians. It is therefore of the highest importance to realize the exact meaning of the Indian view. It will.be noticed, first of all, that there is no difference between a contradiction of concepts and.a contradiction between judgments, the terms bhaaua = vidhi = vastu, Tib. yod-pa.sqrub-pa = dnoos-pa being synonymous.... The term 'blue' in logic always means the.judgment 'this is blue', it is a synthesis of 'thisness' and 'thatness', it is contrasted with.the mere reflex of the blue (pratibhaasa), an unascertained reflex which has no place in logic. Thus, in the quarrel between Aristotle and Sigwart on the one side, and Kant on the other, the Indian view will fall in line rather with the first party. The contradiction is virtually between the judgments 'this is blue' and 'this is not blue'" (op. cit., p. 193).

26.Dharmakiirti, op. cit., pp. 36-37.

27.Cf. Kar.nakagomin, op. cit., p. 86.




discussed by the Buddhist philosophers under the theory of determination (apoha) = exclusion = differentiation.[28]


Incompatibility or efficient opposition Contradiction or logical opposition =negation of terms or entities (sa- = negation of propositions (anyonyo-habhaava-virodha), for example, the op- palabdhi-parihaara-sthiti-lak.sa.na-viro-position of the sensations of heat and dha), for example, the opposition of cold (`siito.s.naspar`sauat). blue and non-blue, existence and non-existence (niilaaniila, bhaauaabhaaua), affirmation and negation (vidhini.sedha).

          Furthermore, from a purely logical point of view the Buddhists maintain that every term or proposition is the negation of its own negation (anyaapoha). Even an affirmative proposition entails the exclusion of its contradictory proposition. Dignaaga explains that a term, for instance, "the blue-lotus," not only excludes all lotuses that are not blue but also excludes those blue things which are not lotuses.[29] Thus, it signifies the exclusion of the non-blue and the non-lotus.[30]

          Thus, what is intended by the theory of apoha is neither merely a positive cognition qualified by the exclusion of others.[31] For instance, the terms affirmation and negation, existence and non-existence, or A and non-A are mutually exclusive. The relation between the qualifier and the qualificand is not ultimate (paaramaarthika) but only dialectical (vyaauahaarika).[32] Here we may conclude that the theory of apoha is not a "negative approach" to reality but a dialectical approach based on the law of opposition.


          Buddhist logicians, in apoha, maintained that every term and proposition is discriminatory. This means that the affirmative and the negative are mutually exclusive and so related to each other that the definition of one involves the other. Dharmakiirti said: "There can be no affirmation of a thing (A) which


28.Cf. Stcherbatsky, op. cit., p. 195, n. 2.

29.Prameya-kamalamaarta.n.da of Prabhaachandra, ed. with notes and introduction by Mahendra Kumar Shastri (2nd ed.; Bombay: Nirnaya Sagar Press, 1941), p. 436; see also Kar.nakagomin, op. cit., p. 182 and pp. 260-261.

30.Raajaa, "The Theory of Meaning According to Buddhist Logicians," p. 8.

31.Ratnakiirti, op. cit., p. 3 (6-8).

32.Ibid., p. 16 (1-15) ; also Kar.nakagomin, op. cit., p. 121.




does not exclude the other (non-A); nor can there be a negation of that which cannot be affirmed."[33] Hence, whereas affirmation implies negation, negation presupposes the affirmation.[34]

          Now the question arises: if the Buddhists are correct in asserting that the meanings of all words are both affirmative and negative at the same time, then different propositional forms would be meaningless. Propositions are accepted as of two kinds: (1) positive (vidhi), and (2) negative (ni.sedha or prati.sedha). In Indian logic, the negative form is again divided into two: (a) a simple negative (prasajya-prati.sedha), and (b) negative by implication (paryyudaasa).[35]


Negative (prati.sedha)

Positive (uidhi), e.g., Negative (prasajya), Negative by implication

"this is blue" (nilo e.g., "the jar is not (paryyudaasa), e.g., "p

'yam). here" (iha qha.to naasti). implies not q". If it is

"snow is not black". gray, it is not white. If

X is a k.satriya, he is

not a braahma.na.


          The answer to this is that the Buddhists were concerned with the practical importance of the propositions. Kar.nakagomin[36] explains: "A sentence expressing an affirmative [judgment] asserts a positive [meaning] primarily [and] negation of the other [non-A] by implication (arthaat). (2) And [a sentence] expressing a negative [judgment] asserts a negation primarily [and] affirmation of the other [i.e., the positive non-A = B] by implication. (3) While the sentence expressing'negation by implication' (paryyudaasa), following denial primarily asserts the presence of the other fact. Thus, indeed, is the distinction [between the three forms of statements]."[37]


33.Suau.rtti, p. 253: na hy anuayo uyaa.rttimato naapy ananvayino vyaav.rttih.

34.("Not" belongs to the class of "propositional words." Cf. B. Russell, An Inquiry into Meaning and Truth (London: George Allen & Unwin Ltd., [5th impression] 1956), pp. 70 ff. Strictly speaking, negation is always negation of a proposition. See D. Sharma, "The Paradox of Negative Judgment and Indian Logic," Vishveshvaranand Indological Journal, Vol. II, Part (March, 1964).

35.These negative kinds of statements in Indian Logic were originally introduced by the Miimaa^msakas, who were primarily concerned with the problems of the correct application of the Vedic texts in the sacrificial ceremonies. Cf. Jamini Suutra, X. 8. 1. 15. For a very lucid exposition see J. F. Staal, "Negation and the Law of Contradiction in Indian Thought: a Comparative Study," The Bulletin of the School of Oriental and African Studies [University of London], Vol. XXV, Part 1 (1962), 59.

36.OP. cit., p. 253.

37.The Miimaa^msaka writers were the first to use the terms prasajya and paryyudaasa-in connection with their religious and ritual interpretations of the Vedic negative state-




          According to this explanation: (1) The proposition, for instance, "snow is white," is affirmative in which the negation of the other, "snow is black," is implied. (2) The proposition, for instance, "snow is not black," is negative in which the other affirmative, "snow is white," is implied. (3) There is negation in the proposition "Mt. Everest is the highest mountain in the world." Here we primarily assert that Mt. Everest is the highest mountain in the world but we do so by denying the suggestion that there is any other mountain as high as Mt. Everest.

          However, it should be noted that the Buddhist theory of negative inference (anupalabdhi),[38] corresponds to the simple negation (prasajya), and entails the paryyudaasa negation. On the other hand, the theory of apoha corresponds primarily to negation by implication and involves simple negation. According to Ratnakiirti, negation by implication (paryyudaasa) is found rooted (niyata) in the immediate knowledge of the thing and thus is commonly applied to both affirmative and (simple) negative propositions.[39] This means that, logically speaking, there can be neither a pure affirmation nor a pure negation. This is the reason why one does not go and tie a horse or a dog when asked to tie a cow.[40] Apoha is the basis of discriminatory behavior in everyday life, and differentiation is the prime factor of all reflective thinking. Thus, the Buddhist would conclude that it is differentiation that is manifested by words and reason, and apprehended through language and logic.[41]


ments. For instance, the statement: "One must not kill a Braama.na" (Dvijam na hanyaat)--cf. "Thou shalt not kill"--defiance of which leads to sin and calamity, is.to be considered as a pure negation (prasajya). Whereas if the Vedic text reads that "a particular ritual should be performed in a particular ceremony," it implies that the ritual may or may not be performed in some other ceremony, but must be performed in that particular instance. Cf. Miimsaa^maa-Nyaaya-Prakaa`sa of AApadeva, A. M. Ramanatha Dikshita, ed. (Banares: Kashi Sanskrit Series, 1949), pp.156 ff. However, according to grammarians, the stress is on the construction of the sentence: prasajya is where the negation is essential and the positive element secondary, that is, where negation applies to the verb (but not to the last member of the negative compound). Where the positive element is essential and the negation secondary is paryyudaasa. It is to be understood that this negation applies only to the last member of the negative compound. Cf. a grammatical work called Saarasvata, Nava Kishora Kara Sarma, ed. (1936), verse 490; quoted by Louis Renou, Terminologie Grammaticale du Sanskrit. 3 pts. Biblioth`eque de 1'Ecole des Hautes Etudes . . . Sciences historiques et philologiques, 280-282 fasc. (Paris: E. Champion, 1957). I owe this reference to Dr. D. Friedman of the University of London and am indebted to him for many critical suggestions.

38.Cf. D. Sharma, "Epistemological Negative Dialectics of Indian Logic, Abhaava versus Anupalabdhi," Indo-lranian Journal, Vol. IX, No. 4 (1966), 291-300.

39.Ratnakiirti, op. cit., p. 4 (6-7).

40. Ibid.

41.Jnnaana`sriimitra, op. cit., p. 201: apoha.h `sabda lingaabhyaa^m prakaa`syate.