Pramāṇa · Search · About · API

Indian Philosophy v2.c24.s6: XII Śabda or Verbal Knowledge

Indian Philosophy v2.c24.s6

Language
English

XII Śabda or Verbal Knowledge

v2.c24.s6.2

One of the chief sources of knowledge is authority. We accept many things which we have not observed or thought about on the authority of others. We learn a good deal from popular testimony, historical tradition and scriptural revelation. The logical issues involved in this mode of acquiring knowledge are discussed under śabda, or verbal testimony.

v2.c24.s6.3

We may refer briefly to the Nyāya views about the origin and nature of sounds, the import of words and the structure of sentences.

v2.c24.s6.4

Ākāśa, which pervades all space and not air, is the substratum of sound.¹ Sound can be produced even in a vacuum, though we cannot hear it, since there is no air to convey it. The quality of sound does not depend on air though the loudness, etc., do.² It is, however, produced by the contact of two hard substances. One sound produces another which causes another, until the last sound ceases owing to some obstacle.³ We cannot argue that sound is eternal simply because it has an intangible substratum.⁴

v2.c24.s6.5

¹ "The sound series is perceived even at a time when there is no perception of anything possessing colour and other qualities, which shows that sound has for its substratum a substance which is intangible and all-pervading, and it does not subsist in the same substratum as the vibrations (kampa)" (N.B., ii. 2. 37).

v2.c24.s6.6

² V.S., i. 1. 6.

v2.c24.s6.7

³ N.B., ii. 2. 38.

v2.c24.s6.8

⁴ Several grounds are urged to establish its non-eternality (N.S., ii. 2. 13-38). (1) Sound has a beginning, since it arises from the concussion of two hard substances, say, an axe and a tree. It cannot be said that the concussion aids the manifestation of sound but does not produce it, for the concussion and the sound are not simultaneous. We hear the sound at a great distance even after the concussion which occasioned it has ceased. (2) Sound is not eternal, since it has a beginning and an end. If it were eternal, then it must always be heard, since it is close to the perceiving organ, which is not the case. We know also that the sound ceases on account of known causes. We stop the gong by the contact of our hand with the bell (ii. 2. 32-36; V.S., ii. 2. 26-37). Vātsyāyana says that in the case of every sound there is a series of sounds, and in this series the succeeding destroys the preceding. What destroys the final sound of the series is the conjunction or impact of an obstructing substance (N.B., ii. 2. 34). Later Naiyāyikas modify this account to suit the Vaiśeṣika theory that a quality cannot subsist in a quality and cannot have another quality conjunction. Vācaspati observes that what destroys the sound is the impact with the obstacle of, not sound, but the ākāśa, which is the material cause of the sound. The impact of ākāśa with a denser substance renders it incapable of functioning as the material cause of further sounds, and when the immaterial cause of the sound, namely, the contact of the stick with the drum, ceases, there is nothing to start the series afresh, and thus the final sound is destroyed. (3) Sound is cognised by one of our senses as advancing in a series. It belongs to the genus of soundness, and is therefore non-eternal (N.S., ii. 2. 16). (4) Sound is spoken of as possessing the properties of products. It is described as grave, acute, etc. (5) From the fact that we repeat the sounds taught by the teacher, we cannot argue that sounds are eternal. When they were inaudible they did not exist, and we now simply reproduce them. Even different sounds may be said to be repeated, as we are said to sacrifice twice or dance twice (N.S., ii. 2. 29). (6) From the intangibility of sound we cannot argue to its eternality. Motion is intangible and yet non-eternal (N.S., ii. 2. 22-24).

v2.c24.s6.9

A word is a combination of letters signifying an object, by way of denotation (abhidhā) or implication (lakṣaṇa). Every word has a meaning, which is generally regarded as a relation between the word or the sign and the object which it signifies.¹

v2.c24.s6.10

¹ Cp. Vijñānabhikṣu on S.S., v. 37. The grammarians Bhaṭṭoji Dīkṣita (Vaiyākaraṇabhūṣaṇa, p. 243) and Nāgeśa Bhaṭṭa (Mañjūṣā, pp. 23-26) regard this signifying power as residing exclusively in words, while the Sāṃkhya and the Vedānta maintain that it resides in objects also. Pañcadaśī, viii. 4-15; Nyāyabinduṭīkā, pp. 10-11.

v2.c24.s6.11

The fact of meaning is explained by the grammarians on the theory of the sphoṭa.² According to it, any single letter, c, o, w, or all the letters, "cow," cannot produce the knowledge of a thing corresponding to the word, since each letter perishes as soon as it is produced. Even if the last letter is aided by the impressions left by the preceding ones, a number of letters cannot explain the cognition of a thing. There must be something over and above the letters by which the knowledge is produced, and that is the sphoṭa, or the essence of sound revealed by letter, word or sentence.³ This sound-essence produces the cognition of the thing. A single letter, unless it is a complete word, cannot signify any thing. The advocates of padasphoṭa argue that only a pada, or a word, can signify a meaning, while those of vākyasphoṭa hold that only a vākya, or a sentence, can signify a complete meaning. According to the latter, a sentence is the beginning of speech, while words are parts of sentences, and letters parts of words. Sphoṭa, or sound-essence, is said to be eternal and self-existent, bearing a permanent relation to the thing signified by it. Letters, words and sentences manifest, but do not produce, the eternal meanings. The Naiyāyikas hold that every word is significant,⁴ and we become cognisant of its signification when we hear the last letter of the word. On hearing the last letter "w," we recollect the previous ones, c, o, and grasp the whole word by the mind; and we cognise the object by means of the conventional association between the word and the object.⁵

v2.c24.s6.12

² Pāṇini's reference to sphoṭāyana in vi. 1. 123 indicates that the theory prevailed in his times. See S.D.S., Pāṇinidarśana.

v2.c24.s6.13

³ Deussen identifies the sphoṭa with the notion. Thibaut regards it as a grammatical fiction, and is certain that it cannot be a notion, since it is distinctly called a vācaka or abhidhāyaka, and is said to be the cause of the conception of the sense of the word (Thibaut's E.T. of S.B., p. 204 n.) See also S.B., i. 3. 28.

v2.c24.s6.14

⁴ Śaktam padam (Tarkasaṃgraha, p. 59).

v2.c24.s6.15

⁵ N.V., ii. 2. 55.

v2.c24.s6.16

The relation between the word and its meaning is not due to nature, but to convention, and this view is confirmed by our experience of the way in which we acquire a knowledge of the meanings of words. We get to know the meanings of words through popular usage, grammar, dictionaries; and the Vedānta mentions gesture as well.¹ The convention that such and such a word should mean such and such an object is established by God (Īśvarasaṃketaḥ).² Later Nyāya admits that men also establish conventions (icchāmātraṃ śaktiḥ),³ though the latter are styled pāribhāṣika, since they vary with different people.

v2.c24.s6.17

¹ Siddhāntamuktāvali, p. 15; Nyāyamañjari, p. vi.

v2.c24.s6.18

² N.B., ii. 1. 55. See also Nyāyamañjari, p. 243.

v2.c24.s6.19

³ Tarkasaṃgraha, p. 59.

v2.c24.s6.20

What is the import of words, an individual (vyakti), or form (ākṛti), or genus (jāti), or all these?⁴ The individual is that which has a definite form (mūrti) and is the abode of particular qualities.⁵ It is manifested and perceptible.⁶ The form is the peculiar properties; the collocation of the dewlap is the form of the cow. The genus is the type or class, the general notion underlying the object of a class. It helps us to attain a comprehensive knowledge of things similar to the individual in question.⁷ The Nyāya holds that a word denotes all the three, the individual, its form and its genus, though in different degrees.⁸ In practice we refer to the form. When our interest is in distinction, the word refers to the individual; and, when we try to convey the general idea, we refer to the genus. The word suggests the form, denotes the individual and connotes the genus. There is no such thing as a pure indeterminate attribute. It is determined in some way (avacchinna). Again, form by itself is not enough. A clay model of the cow is not treated as a cow, though it has the form, since it is lacking in the generic qualities. Popular usage supports the theory that words denote individuals.⁹

v2.c24.s6.21

⁴ N.S., ii. 2. 56.

v2.c24.s6.22

⁵ ii. 2. 64.

v2.c24.s6.23

⁶ ii. 2. 65.

v2.c24.s6.24

⁷ Samānaprasavātmikā jātiḥ (N.B., ii. 2. 66). Since we have a definite conception of cowness apart from the idea of individual cows, there must be an objective basis for the former (N.B., ii. 2. 61 and 66). Uddyotakara holds that the class nature subsists in each individual by way of inherence, or samavāya. The question whether the genus subsists in its entirety or in parts in each individual is meaningless, since jāti is not a composite and the distinction of whole and part is inapplicable to it. Jāti, or the eternal essence, is said to be in a necessary relation of samavāya to the individuals whose essence it is and in indirect or temporal relation (kālikasaṃbandha) to the other individuals.

v2.c24.s6.25

⁸ ii. 2. 63. Jātiviśiṣṭavyakti.

v2.c24.s6.26

⁹ ii. 2. 57.

v2.c24.s6.27

Words, according to the Buddhist thinkers, do not represent positive objects, but simply exclude others erroneously recalled to mind. The word "cow" denotes the negation (apoha) of objects which are not cows, such as horses, etc. From this exclusion we infer that the word "cow" refers to the object "cow."¹ Uddyotakara criticises the apoha doctrine on the following grounds.² We cannot conceive of a negative denotation unless we have previously conceived of a positive one. All negation has a positive basis. Bare negation is meaningless, while every specific negation has a positive implication. Though in the case of two contradictory words the denotation of the one may exclude that of the other, such an exclusion is not possible in the case of a word like "all."³ Every word denotes something positive which is not exhausted by its distinction from something else.⁴

v2.c24.s6.28

¹ See Nyāyamañjarī, pp. 303, 306-8, and Pārthasārathi Miśra's Nyāyaratnākara. Early Buddhist works do not contain definite information about this view, though it appears in a modified form in the Apohasiddhi of Ratnakīrti. For him words denote neither positive objects nor negative ones. The positive meaning is not a consequence of the negation of other objects any more than the negative meaning is the consequence of the positive denotation. The essence of meaning consists in the simultaneous cognition of the positive and the negative sides. All determinate objects have a positive nature which excludes others. This theory is certainly more satisfactory, though it is not easily reconcilable with the general metaphysics of the Buddhists. Nor do the Hindu logicians accept it as the Buddhist view.

v2.c24.s6.29

² N.V., ii. 2. 65.

v2.c24.s6.30

³ See Udayana's Ātmatattvaviveka.

v2.c24.s6.31

⁴ Nyāyamañjarī, p. 311. See also Nyāyakandalī, pp. 317-321.

v2.c24.s6.32

It is objected that words cannot denote objects, since they do not co-exist with objects, and are present even when the objects are not present, as in a negative judgment "there is no jar here."⁵ Vācaspati meets this objection by saying that a word denotes the universal, including all individuals dispersed in time and place, and so refers to individuals present as well as past.⁶ Nor can it be said that the word is only an abstract idea, since it cannot signify the different features of the different individuals. The word refers to the distinctive features which are objective. We use words in experience and they lead to success in life. All this would be impossible if the word referred simply to mental images and not to outer objects.⁷

v2.c24.s6.33

⁵ Prameyakamalamārtaṇḍa, p. 124; V.S., vii. 2. 17.

v2.c24.s6.34

⁶ N.V.T.T., ii. 2. 63.

v2.c24.s6.35

⁷ Prameyakamalamārtaṇḍa, p. 136; Vidyānanda's Aṣṭasahasrī, p. 249.

v2.c24.s6.36

Sometimes, it is said, that we cannot conceive of the relation between the word and the object. The word is an attribute, and the object denoted by it is a substance, and between the two there cannot be the relation of saṃyoga (conjunction). Even if the object denoted be an attribute, this relation is impossible between two attributes.⁸

v2.c24.s6.37

⁸ V.S., vii. 2. 14.

v2.c24.s6.38

Words are inactive, and conjunction is based on the movement of one of the members of the relation. The word ākāśa and the object ākāśa are both inactive, and there can be no conjunction between them. Nor do we have between a word and its meaning the relation of inherence. Vātsyāyana admits that the relation between the word and its meaning is not of a productive character (prāptilakṣaṇa). The word "fire" does not produce the object of fire.¹ That is why verbal cognition is less distinct than sense-perception.² But it is none the less cognition.

v2.c24.s6.39

¹ N.B. and N.V.T.T., ii. 1. 50-51.

v2.c24.s6.40

² Prameyakamalamārtaṇḍa, pp. 128-130; Kumārila's S.V., v. 11. 6-8 and 10.

v2.c24.s6.41

A sentence (vākya) is a collection of significant sounds or words. We cognise the constituent words and then their meanings. The cognitions of the words leave behind traces (saṃskāras) which are remembered at the end of the sentence, and then the different meanings are related together in one context. While the ancient Naiyāyikas contend that the chief means (mukhya karaṇam) of verbal cognition is the recollection of objects due to verbal memory, the modern Naiyāyikas argue that the verbal memory is the chief means. The meaning of a sentence depends on (1) ākāṅkṣā, mutual need or interdependence, or the inability of a word to indicate the intended sense in the absence of another word, (2) yogyatā, or compatibility or fitness or the capacity to accord with the sense of the sentence and not render it futile and meaningless, (3) sannidhi, propinquity, juxtaposition or the utterance of words in quick succession without a long pause between one word and another. These insist on the syntactical, logical and phonetical connections of words. A collection of words devoid of interdependence, man, horse and colony, conveys no sense. A sentence like "moisten with fire" (agninā siñcet) conveys no intelligible meaning. Similarly, words uttered at long intervals do not convey any sense. A sentence is made up of words which are interdependent, capable of being construed together and in close juxtaposition. Gaṅgeśa adds a fourth condition, namely, a knowledge of the intention of the speaker. A sentence like "saindhavam ānaya" may mean either "bring the horse" or "bring salt," and we can be sure of its meaning only if we know the mind of the speaker. Fitness of words to express a definite meaning covers this requirement also.¹ While fitness demands formal consistency, tātparyajñāna, or knowledge of the intention, may be said to imply material compatibility.²

v2.c24.s6.42

¹ Bhāṣāpariccheda. See also Vedāntaparibhāṣā, iv.

v2.c24.s6.43

² While the Mīmāṃsakas and the grammarians hold that the words of a sentence centre round the verb, without which they convey no sense, the Naiyāyikas hold that a proposition is but a number of significant words (padasamūha) whose collective meaning is apprehended, whether or not there be a verb in the sentence (Tarkasaṃgraha, p. 59; Jhā: Prabhākara School, p. 63).

v2.c24.s6.44

Propositions are divided into three classes: command (vidhi), prohibition (niṣedha), explanation (arthavāda).³ Śabda, when used as a source of knowledge, means āptopadeśa, or the assertion of a reliable person.⁴ The āpta, or the reliable person is the specialist in a certain field, “one who, having had direct proof of a certain matter, desires to communicate it to others who thereby understand it.” They may be of any caste or race, “ṛṣis, āryas or mlecchas.”⁵ When a young man is in doubt whether a particular river is fordable or not, the information of an old experienced man of the locality, that it is fordable, is to be trusted.

v2.c24.s6.45

³ N.S., ii. i. 63, and Tarkakaumudī, p. 17.

v2.c24.s6.46

⁴ i. i. 7.

v2.c24.s6.47

⁵ N.B., i. i. 7.

v2.c24.s6.48

These trustworthy assertions relate to the visible world (dṛṣṭārtha) or the invisible (adṛṣṭārtha). That quinine cures fever is of the former kind, that we gain heaven by virtue is of the latter. The words of the ṛṣis deal with the latter.⁶ Their statements are to be relied on, since their assertions about the verifiable world have been found to be true. The authors of the Vedas are āptas, or reliable persons, since they had an intuitive perception of the truths, love for humanity and the desire to communicate their knowledge.⁷

v2.c24.s6.49

⁶ N.B., i. i. 8.

v2.c24.s6.50

⁷ ii. i. 68.

v2.c24.s6.51

Later Naiyāyikas, like Udayana and Annaṃ Bhaṭṭa and the Vaiśeṣika thinkers, regard the supreme Īśvara as the eternal author of the Vedas. Udayana sets aside the view that the authoritativeness of the Vedas is to be inferred from their eternality, freedom from defects and acceptance by great saints. At the beginning of new world-epochs there can be no acceptance by saints. The Mīmāṃsaka argument of the eternality of the Vedas is controverted by Udayana, who argues that there is no continuous tradition to indicate eternality, since such a tradition must have been interrupted at the dissolution of the world which preceded the existing creation. Vātsyāyana, however, accepts the continuity of tradition, in the sense that God at the beginning of every epoch recomposes the Vedas and keeps up the tradition.¹ If the Mīmāṃsakas adduce texts in support of their view, that the Vedas are eternal, and that the ṛṣis are not their authors but only their seers (mantradraṣṭāraḥ), other texts are quoted in support of the Nyāya view of the origin of the Vedas.² Besides, the Vedas contain sentences which imply an author.

v2.c24.s6.52

¹ N.B., and N.V.T.T., ii. 1. 68.

v2.c24.s6.53

² Idaṃ sarvam asṛjata ṛco yajūṃṣi sāmāni, etc.

v2.c24.s6.54

Objections against the validity of the Vedas, such as those of untruth, contradiction and tautology, are rejected as untenable.³ Their validity is defended on the ground that their contents form a coherent whole. Acceptance of the Vedas does not mean a resort to blind faith or revelation.

v2.c24.s6.55

³ If we perform a sacrifice for the sake of getting a son, and do not get one, the fault may be in the action and not in the Vedic rule. Injunctions such as "offer the oblation after sunrise" or "before" need not be taken as contradicting each other, since they state alternative courses of conduct. There is no useless tautology (N.B., ii. 1. 57-68).

v2.c24.s6.56

Dignāga contends that śabda is not an independent source of knowledge. When we speak of credible assertion, we mean either that the person who utters it is credible or that the fact that he utters is credible. If it is the former, we have a case of inference; if it is the latter, we have a case of perception.⁴ Though śabda is like inference since it conveys the knowledge of an object through its sign, yet the sign here is different from what it is in inference, indicating as it does whether the words come from a reliable person or not.⁵ The relation between the sign and the thing signified is natural in inference, while it is conventional in verbal knowledge.⁶ If we argue that verbal cognition follows the remembrance of the meanings of words and therefore is inferential, then even doubtful cognition and knowledge by comparison should be regarded as inferential. If a reference to the three periods of time makes verbal cognition inferential, then other forms of reasoning, as tarka, will also be inferential. If it is urged that verbal cognition depends on positive and negative concomitance, to the effect that the word "jar" means the cognition of the object, and there is no cognition of the object where it is not pronounced, then even perception may be regarded as a case of inference, since it is present where the jar is present, and where the jar is absent there is no perception of the jar.⁷ Knowledge derived through words is thus different from that gained through perception, inference and comparison.⁸

v2.c24.s6.57

⁴ Dignāga, however, accepts the sayings of Buddha as authoritative, See Kumārila's Tantravārttika, pp. 169 ff.

v2.c24.s6.58

⁵ N.B., ii. 1. 52.

v2.c24.s6.59

⁶ N.B., ii. 1. 55.

v2.c24.s6.60

⁷ N.V, ii. 1. 49-51.

v2.c24.s6.61

⁸ N.B., ii. 1. 52; N.V., i. 1. 7.

v2.c24.s6.62

OTHER FORMS OF KNOWLEDGE

v2.c24.s6.63

To the four sources of knowledge admitted by the Nyāya, the Mīmāṃsakas add arthāpatti, or presumption, and the Bhāṭṭas and the Vedāntins abhāva, or non-existence. The Paurāṇikas regard tradition and probability also as valid sources of knowledge. The Naiyāyika believes that all forms of knowledge are comprehended by the four pramāṇas.¹

v2.c24.s6.64

¹ N.B., ii. 1. 19.

v2.c24.s6.65

Aitihya, or tradition, is brought under śabda.² If the rumour or tradition is started by a reliable person, then it is as valid as śabda. Arthāpatti, or implication, is getting at a new fact or presuming something (āpatti) on the basis of another fact (arthāt). It is assuming a thing not itself perceived, though implied by another thing perceived or inferred. The fat Devadatta does not eat in the day. The implication is that he eats in the night, since it is impossible for one to be fat if one does not eat at all. The Mīmāṃsakas, who regard it as an independent means of knowledge, view it as a disjunctive hypothetical syllogism.³ According to Gaṅgeśa, it is an example of a negative inference which establishes the absence of the middle term through the absence of the major. According to the Bhāṣāpariccheda, arthāpatti is accomplished through the recognition of a negative relationship between the middle and the major (vyatirekavyāpti).⁴ Sambhava, or subsumption, where we cognise a part from a whole of which it is a member, is a case of deductive inference. It is strictly numerical inclusion.

v2.c24.s6.66

² ii. 2. 2.

v2.c24.s6.67

³ See also Bhāṣāpariccheda, p. 143.

v2.c24.s6.68

⁴ It may be expressed in two different stages :—

v2.c24.s6.69

Abhāva, or negation, is sometimes mentioned as an inde-

v2.c24.s6.70

He who does not eat at all is not fat.
This man is fat.
Therefore this man is not one who does not eat at all—i.e. he is one who eats.

This is Cesare. The next step is :—
He who eats must do so either by day or by night.
He does not eat by day.
Therefore he eats by night.

v2.c24.s6.71

pendent pramāṇa. Though the Nyāya-Vaiśeṣika system admits non-existence as an object of cognition, it does not believe that a special pramāṇa is necessary for its apprehension. We have already seen how existence is an object of perception which is connected with its adhikaraṇa, or locus, by the relation called viśeṣaṇatā (or qualified and qualification). The non-existent thing is of the same order of reality as its locus, which is perceived; otherwise the perception of its absence cannot be implied by the perception of its locus. Absolute negation is inconceivable. The negation which is the object of knowledge is relative.¹

v2.c24.s6.72

¹ For a different view, see Śāstradīpikā, i. 1. 5; Vedāntaparibhāṣā, vi.

v2.c24.s6.73

By means of inference, we can infer the non-existence of things. Abhāva means not mere negation but contrast. It is contrast as between what exists and what does not exist, as when the non-existent rain brings about the cognition of the existence of the connection of the clouds with high winds; since it is only when there is some such obstruction, as the connection of the clouds with high winds, that there is no fall of the raindrops, which would otherwise be there by reason of the force of gravity in the drops.² Of two contradictory things, the non-existence of one establishes the existence of the other. The Nyāya logic proceeds on the principle of dichotomous division. The distinction of homogeneous and heterogeneous examples rests on this assumption. Two contradictory judgments cannot both be false, nor can they both be true. A is either B or not B. One or the other of two contradictories must be true since no other course is possible.³ If we infer the non-existence of a thing from the existence of another, it is only a case of inference.⁴ Vātsyāyana says: " At the time the existent thing is cognised, the non-existent thing is not cognised, that is to say there is the non-cognition of the non-existent, only at the time that there is the cognition of the existent. When the lamp illumines and renders visible something that is visible, that which is not seen in the same manner as that visible thing is regarded as non-existent, the mental process being as follows: ' If the thing existed, it would be seen : since it is not seen, it must be concluded that it does not exist.' "¹ Praśastapāda supports this view. " As the appearance of the effect is indicative of the existence of the cause, so is the non-appearance of the effect indicative of the non-existence of the cause."² Even by means of śabda, we can have cognition of non-existence.³

v2.c24.s6.74

² N.B, ii. 2. 1

v2.c24.s6.75

³ Parasparavirodhe hi na prakārāntarasthitiḥ (Kusumāñjali, iii. 8).

v2.c24.s6.76

⁴ N.B., Introduction.

v2.c24.s6.77

¹ N.B., ii. 2. 2.

v2.c24.s6.78

² P.P., p. 225. See also V.S., ix. 2. 5; Nyāyakandalī, pp. 225-226; and Kusumāñjali, iii. 20, 22 and 26.

v2.c24.s6.79

³ Jayanta mentions eleven kinds of anupalabdhi. See Nyāyamañjari, pp. 56-57.

Cites (18)