Verbal communication is ubiquitous in our lives. Whether it is to pass information, make jokes, tease each other, praise someone, or to express feelings – all of these activities involve using words to form sentences that are then communicated to the hearer. The hearers, on their part, put together these words and their meanings and understand what the speaker said. If what the hearer understood and what the speaker had in their minds are the same, then that particular episode of communication is said to be a successful one. However, if there is a mismatch between the two, something must have gone wrong in this process.
What is/are this ‘something’ and how can we avoid it?
One way of answering this question is to identify the elementary units of communication and then impose certain conditions on those units such that the fulfillment of these conditions will guarantee a successful episode of verbal communication.
The answer to the first part of this problem is fairly straight-forward. The elementary unit of communication is a sentence. We do use several gestures or signs to communicate but for the purposes of a rational reconstruction of the communicative process, we can assume that all of these translate into sentences in the hearer’s mind.
The next step is to think of verbal communication as a causal process. According to Indian philosophers (particularly the Naiyyayika-s or Logicians), every causal process involves an instrumental cause (kara?a) which plays a unique role in producing the effect. It is also called efficient cause, i.e. that component which is most efficient in bringing about the effect. To take a mundane example, if writing is the expected result then, a pen or a pencil would be the instrumental cause. In the classical example of the pot being the effect, the potter is this instrumental cause. In our case, that of verbal communication, it is the knowledge of words. The speaker, in an episode of communication, emits certain sounds. The hearer would not be able to derive any meaning from these sounds unless she/he recognizes them as words of some language. Once the hearer recognizes these words, she/he recollects the meanings of the words and puts together their meanings to derive the meaning of the sentence. Finally, the hearer acquires some knowledge from the sentence. Such knowledge is called Testimonial Knowledge (knowledge from testimony) or ??bda-bodha.
However, in this causal process, there are several auxiliary factors involved without which the end result of understanding or knowing will not be achieved. Merely recognizing the words (and knowing their meanings) is not sufficient for the hearer to derive knowledge from them. These words must be put together in such a way that they satisfy certain conditions that are conducive to a successful episode of communication. As we stated above, the knowledge of words (but according to some other philosophers the words themselves) is the instrumental cause in the production of verbal cognition. It plays the same causal role as the potter or the wheel’s motion in the creation of a pot. These auxiliary factors play the role which is analogous to the water sprinkled on the clay from time to time or the straw that is added for binding the clay.
In this article, we shall examine these auxiliary factors as set out by Indian philosophers of the Ny?ya school. According to them, the knowledge of these factors are individually necessary and jointly sufficient conditions to understand sentences. These factors are ?k??k??, Yogyat?, Sannidhi and T?tparya.
?k??k?? (Syntactic Expectancy)
Definition: “Padasya pad?ntara-vyatireka-prayukta-anvaya-ananubh?vakatvam ?k??k??”: ?k??k?? is the inability of a word to express its meaning (as related to the sentence) due to the absence of another word.
(Tarkasa?graha. See ‘Primary Sources’ under the section ‘Further Reading’ below.)
The first condition which sentences must satisfy is that they must be well-formed. This will happen only if the expectancy that each word has for another in order to convey its meaning to the sentence meaning is satisfied. An example will make this clear. “Bring the pot” is a well-formed sentence which is said to have fulfilled the condition of ?k??k?? since the verb ‘bring’ is adequately supplied with the object of the action ‘the pot.’ If someone were to say ‘close’ there would automatically arise a question in the hearer: Close what? A verb has to be supplied with an object. Similarly, if the speaker were to utter the word ‘the gate’ the hearer would not know what to make of it unless something like ‘open’ or ‘build’ or ‘fix’ is added. The case of a verb requiring an object is only one example. Similarly, we would give other examples related to adjectives and substantives (If some said “the black” we would naturally ask “the black what?”) pronouns and nouns, etc. Likewise, every word “is said to have ?k??k?? for another, if it cannot, without the latter, produce knowledge of its inter-connection in an utterance.” Merely uttering a string of unrelated words (even if the meanings are known to the hearer) will not produce any knowledge or understanding in the hearer and will result is an unsuccessful episode of communication. Only those sentences in which all the words have their mutual expectancies fulfilled can serve as units of successful communication.
We can classify this expectancy into natural (uttihita-?k??k??) and contextual (or potential, utth?pya-?k??k??). The examples given above are all that of natural expectancy – the verb has a natural expectancy for a noun (or an object) and so on. Another natural expectancy that is talked about by some Nayyayika-s is between the nominal stem and its affix. For example, the Sanskrit word gha?am is made of two parts – the nominal stem ‘ghata’ and the accusative suffix ‘-am’ which tells us that it is the object of a verb. Similarly, the word ‘rama?’ is made of the stem ‘rama’ and the nominal suffix ‘-su’ (which following some grammatical operations becomes -?). However, even after a sentence like “Bring the pot” has been uttered – in which the natural expectancy of all components is satisfied – the hearer may yet have the question “Which pot? – the black or the brown or the multi-coloured or leaky one etc.”This kind of expectancy, beyond the grammatical expectancy, is called contextual or potential expectancy. Thus in a sentence like “Ravi met Shruti in Bengaluru” ‘Ravi met’ has a natural expectancy towards ‘Shruti’ (and vice-versa) while they have only contextual expectancy towards ‘in Bengaluru.’
Yogyat? (Semantic Compatibility)
Definition: “Arth?b?dho Yogyat?”: The non-contradiction of meaning is yogyat?.
(Tarkasa?graha. See ‘Primary Sources’ under the section ‘Further Reading’ below.)
The previous condition ensures that a sentence is syntactically well-formed. That is, it ensures that every verb has an object, a pronoun has a noun to refer to, an adjective has a substantive, and so on. But is that sufficient to derive meaning from a sentence? Consider the famous example proposed by Chomsky: “Colourless green ideas sleep furiously.” This sentence fulfills the condition of ?k??k??, yet it is quite meaningless. Why? Because the notion of ‘being furious’ is not compatible with the activity of sleeping. Nor is the concept of ‘green’ (or any other colour) applicable to objects like ‘ideas.’ Hence, there is a mutual incompatibility between the meanings of the words. Consider another example, “There goes the married bachelor.” This again is a meaningless sentence since a person who is married cannot be a bachelor anymore (This is a variant of the classical example “there goes the barren woman’s son”). Hence, to exclude such sentences from being legitimate sources of verbal cognition, this condition of yogyat? has been introduced. The classic example is given to explain yogyat? is “agnin? siñcati” (He sprinkles with fire or He wets it with fire). Like the previous examples, here too there is an incompatibility between the activity of sprinkling and the instrument used for sprinkling (fire).
Observe that incompatibility may be of two kinds: Logical (or conceptual) or Factual. In the sentence “There goes the married bachelor” or “I drew a square circle” the incompatibility is a logical one – the idea of ‘married bachelor’ or ‘square circle’ is logically inconceivable. However, in utterances like “He sprinkles with fire” or “There goes the horned rabbit” there is a factual incompatibility – fires are not generally spoken of as things which are sprinkled and rabbits do not have horns. However, there is nothing in the concept of ‘fire’ which says that it cannot be sprinkled nor is it inconceivable that rabbits have horns. According to the Nayyayika, the condition of yogyat? includes both the sentences as nonsensical. This is because their underlying theory of meaning is a Realist one. Words must have an actual referent in the external world to be meaningful. Since, in the real world, rabbits do not have horns, any sentence which uses ‘horned rabbit’ cannot generate any knowledge in the hearer.
One may object that the very fact that we are passing a judgment that a particular sentence is meaningless proves that we have understood its meaning. This is precisely what the philosophers belonging to the Grammatical tradition (Vaiy?kara?a) say. They make a distinction between the meaning of the word (the concept) and the actual referent of the word (the object). Each and every word in Chomsky’s sentence is meaningful but does not have a referent. So also in the case of “There goes the married bachelor.” Meaning, for the Grammarians is something that is mental. Whether the word has a corresponding ‘thing’ in the world to which it refers is a different matter altogether. They insist that we do understand sentences like “He sprinkles with fire” but we know that they cannot be true. Truth or falsity of a sentence, however, has nothing to do with its meaningful-ness. This is the position of Grammarians.
Do we end our discussion of yogyat? By noting that in cases of metaphorical utterances, invariably, this condition is not satisfied. When Romeo says “Juliet is the sun” there is an incompatibility (of the factual type). Yet, given the context, the hearer may understand the utterance as Juliet being the light in Romeo’s life.
Sannidhi or ?satti (Proximity or Contiguity)
Definition: “pad?nam-avilambena-ucc?ra?am sannidhi”: Uninterrupted utterance of words is sannidhi
(Tarkasa?graha. See ‘Primary Sources’ under the section ‘Further Reading’ below.)
This condition necessitates that the words of a sentence occur in a continuous stream so as to constitute an “uninterrupted utterance.” The sentence that the speaker has in mind may be syntactically well-formed and semantically meaningful. However, if the words are spoken intermittently – ‘Rama’ is said now, after 10-15 minutes ‘killed’ is said and a further 20 minutes later ‘Ravana’ is said – then there is be no verbal cognition. This seems quite an intuitive condition since it will not occur to the hearer that these 3 words constitute the words of a single sentence.
Further, consider the sentence “Girir bhukto vahnim?n Devadatta?” (The hill has eaten fiery Devadatta. The actual example given in Nyaya Siddhanta Muktavali is Girir bhuktam agnim?n Devadattena which translated to ‘The hill has eaten fiery by Devadatta’. I have chosen the former version since its translation helps us to understand the point). This sentence is actually composed of two sentences “The hill (is) fiery” and “Devadatta has eaten.” However, the words of each sentence are interrupted by each other. Hence, we cannot understand what the speaker wants to convey. (We may note that the Sanskrit language, in which this example was originally proposed, has a much more free word order than English i.e. the order of the words does not affect the meaning of the sentence in many cases. Hence, in Sanskrit, this sentence is syntactically well-formed I.e it fulfills the ?k??k?? condition). Another example of such a sentence is “Drink eat milk rice” which is composed of two commands “Drink milk” and “Eat rice.” However, because of the words not being in spatio-temporal contiguity, the sentence fails to generate any cognition.
The condition of contiguity may be extended to meanings of the words too (as latter Nayyayika-s do). The reason for this is as follows. In a Sanskrit verse, for example, the noun may occur in the first quarter and the verb that is related to it may occur in the last quarter. In such cases, if our condition of sannidhi dictates that words must occur in a continuous stream then, we will have to accept that a majority of verses will not satisfy this condition. However, common experience shows that we do understand such verses. For example, in
Of Madhu’s foe
incarnate as a lion by his will,
may the claws, which put the moon to shame
in purity and shape,
by cutting off his devotees’ distress
grant you protection. (Dhvanyaloka 1.1)
the phrase ‘may the claws’ (which is the noun-phrase) occurs in the middle of the verse; ‘of Madhu’s foe’ occurs in the beginning and ‘grant you protection’ (the verb clause) occurs in the end and is interrupted by several other words. Yet, a seasoned reader of poetry can put these clauses together and understand the verse. Hence, sannidhi need not be an uninterrupted flow of words but an uninterrupted flow of meanings of the words or uninterrupted cognition of meaning. Each word when heard leaves a memory impression (samsk?ra) on the mind and when the final word is uttered, all the meanings are put together to give a unitary sentential meaning. As long as this happens, sannidhi condition may be assumed to have been satisfied.
T?tparya (Purport or Intention)
Definition: “Vaktur iccha t?tparya”: The desire of the speaker is t?tparya.(Bh??? Pariccheda; see Primary Sources below)
The final auxiliary factor which is necessary for successful communication is the general awareness of the intention of the speaker. This becomes especially important in cases of ambiguous utterances. To take the classical example – “saindhavam ?naya” (Bring saindhava) could have two meanings based on the two meanings of ‘saindhava’: horse and salt. Judging from extra-linguistic factors the hearer must decide which of the two meanings is applicable here. If the conversation is taking place during a meal, then it is the latter while, if it is when someone is stepping out then it is the former. The same applies to sentences like “I saw a man with a telescope”. This sentence is just as ambiguous as the previous one (the difference is that in the former is a case of, what is called, lexical ambiguity while this latter is a case of structural or syntactic ambiguity). From the context (usually extra-linguistic) the hearer must try and understand what the speaker means. If the hearer is unable to resolve the ambiguity, the result will be an unsuccessful episode of communication.
This does not mean that if the hearer knows the intention of the speaker, then the actual meaning of the words doesn’t matter. For example, the speaker cannot use the word ‘table’ to mean ‘computer’ just because the hearer has understood, for the context of the conversation that the speaker intends to talk about the computer. Words have definite meanings and language must serve as an objective instrument of communication.
The above account is predominantly from a Naiyyayika’s point of view. T?tparya has quite a different meaning in the M?m??sa tradition (whose main goal is to develop interpretational rules for the Veda) which holds that no such knowledge of the speaker’s intention is required for verbal cognition. They hold that the linguistic context itself is sufficient to reveal the true meaning of the sentence. The reason they hold such a view is coupled with their theory of language and meaning. According to them, words have an inherent capacity to reveal their meanings and sentences too, being nothing but a group of ordered words, have a similar natural capacity to produce cognition in the hearer. Hence, there is no need to postulate any such condition on knowing the speaker’s intention to understand the sentence. Hence, though they recognize something called t?tparya, what they mean by this term is the ‘purport’ rather than the ‘intention’ and they use it in the context of arriving at the meaning of a piece of text whose author is unknown (generally the Vedic texts). They use the following indicators (li?ga) to understand the t?tparya (purport) of a passage: “(a) consistency in the meaning between introduction and conclusion, (b) repetition of the main topic (c) the novelty of the subject matter (d) the result intended (e) corroborative and eulogistic remarks as distinguished from the main theme and (f) arguments in favor of the main topic.”
Now, we can go back to the question posed at the beginning of the article: what goes wrong in that episode of communication in which the meaning cognized by the hearer does not match the meaning intended by the speaker. Following our account, we can say that either the hearer doesn’t know some (or all) of the words used by the speaker, in which the instrumental cause of verbal cognition is absent. Or, we may say that one of the auxiliary factors was absent. Maybe the parts of the sentence did not have their syntactic expectancy fulfilled (?k??k?? condition not fulfilled), or the concepts involved were incompatible (yogyat? condition not fulfilled), or the flow of words (or their meanings) may have been interrupted (sannidhi condition not fulfilled) and finally the hearer may not have had the knowledge of the speaker’s intention (t?tparya condition not fulfilled).
To summarize, verbal communication, in the Indian philosophical tradition is seen as a causal process with the knowledge of the words acting as the instrumental cause and the result being the understanding of a sentence. In this causal process 4 auxiliary causes have been identified: ?k??k??, yogyat?, sannidhi and t?tparya. These four conditions cover all aspects of communication: those related to words, those related to meaning, those related to the speaker’s intention, and those related to the hearer’s expectation. Hence, the absence of any one of these causes results in the failure of the communicative episode. Together with the knowledge of words, they form the sufficient condition for successful verbal communication.
Topics in Philosophy of Language are mainly covered in the texts of Ny?ya, M?m??s? and Vy?kara?a school of philosophy (dar?ana). The above article has been written mainly from the Ny?ya point of view with other views being considered when there is a stark philosophical difference.
Tarkasa?graha with D?pika, Annambhatta, Translated into English with notes by Swami Virupakshananada, Sri Ramakrishna Math, Mylapore, Madras, 2nd edition, 1994. This book is considered the entry-level book for Ny?ya Philosophy and it contains rudimentary account of the above concepts. The definitions for ?k??k?a, Sannidhi and Yogyata have been taken from here. Readers can also refer to this book for a basic understanding of the nature of verbal cognition. However, this book does not mention t?tparya.
Bh??? Pariccheda with Siddh?nta Mukt?vali, Vishvanatha, Translated by Swami Madhavananda, Advaita Ashrama, Kolkata, 2004. A more advanced book in Ny?ya. Contains an elaborate discussion of t?tparya. The English translation of the ?abda-kha?da (Verbal Testimony) is Ny?ya Philosophy of Language by John Vattanky S.J., Sri Satguru Publications, Delhi, 1995. The definition of T?tparya has been taken from this work.
M?nameyodaya, Narayana, Translated into English by C.Kunhan Raja and S.S.Suryanarayana Sastri, The Adyar Library and Research Centre, 2nd edition, 1975. This is an elementary treatise in the M?m??sa tradition (Bh???a School). Sannidhi gets more attention in this book than the other two factors. T?tparya is not considered.
Raja, Kunjunni K, Indian Theories of Meaning, Adyar Library and Research Institute, 2nd Edition, 1969. By far the best survey on Philosophy of Language from the Indian perspective. Chapter 4 of this book is exclusively devoted to ?k??k??, Yogyat?, Sannidhi and T?tparya. The book also contains an extensive bibliography of primary resources in Indian Philosophy of Language.
Matilal, B.K, The Word and the World: India’s Contribution to the Study of Language, Oxford University Press, 1990. A book written once again from the Ny?ya point of view by one of the most significant philosophers of the 20th century. Readers may refer to Chapter 6 of this for an account of the causal process of verbal cognition.
Coward, Harold G., and K.Kunjunni Raja, The Philosophy of the Grammarians, Encyclopedia of Indian Philosophies, Volume V, General Editor Karl H Potter, Motilal Banarsidass, Delhi, 1900. Discussion of the above concepts is part of the 5th Chapter (in Part One) ‘Sentence Meaning.’
Palit, Piyali, Basic Principles of Indian Philosophy of Language, Mushiram Manoharlal Publishers in association with Centre of Advanced Study in Philosophy, Jadavpur University, Kolkata, 2004. An extensive and advanced discussion of the above concepts, especially ?k??k?? from the perspective of different dar?ana-s.