Comparing Katz and Searle's Views on Literal Meaning, Study notes of Communication

Download Comparing Katz and Searle's Views on Literal Meaning and more Study notes Communication in PDF only on Docsity! Links & Letters 3. 1996 39-48 What is the Literal Meaning of a Sentence?* Reiko Itani Kanagawa University, Japan June 1995 Abstract In this paper, Katz and Searle's controversy over literal meaning will be discussed in the light of relevance theory. Although their notions are drastically different, these notions have their places in utterance interpretation processes and there is no point in deciding whose notion is right. Katz's notion can be characterised as a relevance notion logical form which is an output of the hearer's knowledge of grammar, while Searle's, propositional form, which is a contextually enriched logical form. 1 will introduce a relevance notion literal interpretation which can not only be contrasted with figurative use, but also with non-figurative loose use (e.g. France is hexagonai). Key words: Literal Meaning, Relevance Theory, Utterance Interpretation. Table of Contents 1. Introduction 4. Literal interpretation in relevance 2. Katz and Searle's controversy over theOry the literal meaning of a sentence 5 . Conclusions 3. Katz and Searle's notions of literal References meaning and relevance theory 1. Introduction Literalness is often talked about in much of the literature. For example, Searle (1 968) identifies literal sentence meaning with the linguistic propositional meaning of a sentence, as opposed to intended speaker meaning which is iden- (*) 1 would particularly like to thank Robyn Carston, University College London, for her help and comments. 1 also thank Links and Letters referees for useful comments and suggestions. This is a revised version of the paper delivered at Linguistic Association of Great Britain, Autumnal Meeting, York, 17th, September, 199 1. 40 Links & Letters 3, 1996 Reiko Itani tified with what the speaker says in uttering this sentence.' Searle (1975) states that what the speaker literally says is sentence meaning while what the speaker actually cornmunicates is speaker's utterance meaning. Further, Grice (1975) makes a distinction between an utterance's literal rneaning and the meaning which is actually communicated by the speaker.2 So it seems that literal meaning cuts across sentence and utterance mean- ing, as here we have literalsentence meaning as in Searle (1968) and literal utterance meaning as in Grice (1975). Both of these literal rneanings are impor- tant in pragmatics as they fall in the explicit content of communication which is contrasted with the implicit content. Undoubtedly people perceive the expli- citlimplicit distinction: e.g. the point in an indirect answer to a question lies with the implicit content rather than the explicit content of the utterance. Then, an attempt to explicate the term literal meaning is in fact an attempt to explicate the explicit content of an utterance, the leve1 which is required for a serious pragmatic theory. In this paper, we will investigate what the literal meaning of a sentence is, as well as whether the speaker is saying the sentence literally or not. The inves- tigation is necessary as the existing ~nderstandin~s of literal meaning are various. We will mainly focus on Katz and Searle's controversy over the literal meaning of a sentence as their notions of literal meaning have their hnctions in utterance interpretation processes explicated in !he light of relevance theory, which is a pragmatic theory of utterance interpretation: 2. Katz and Searle's controversy over the literal meaning of a sentence Katz (1977) put forward the notion of the meaning in azero or null context, and argued that this is the literal meaning of a sentence. According to Katz (1977), the literal meaning of a sentence is deterrnined by the meanings of its component words and its structure. Searle (1 978), however, cast doubt on the very notion of meaning in the zero context, i.e. the gramrnatically determined literal meaning of a sentence. Searle (1978: 208) criticizes the received opinion that the literal meaning of a sen- 1. The propositional meaning of a sentence is a meaning which designates what a sentence says about the world. The same sentence can express different propositional meanings on different occasions. For example, he ir hungry can on one occasion describes the state of affairs that John Smith is hungry at the time of uttering the sentence, and on another, that the male lion in the cage is hungry at the time of uttering the sentence. On the contrary, dif- ferent sentences can express one and the same propositional meaning. For examplc, the sentence I t i Wednesday today uttered on a Wednesday expresses the same propositional meaning as It will be Wednesday tomorrow uttered on a Tuesday. 2. It is interesting to note that the distinction between an utterance's literal meaning and the speaker's intended meaning is the one that most people consider to be the semanticslprag- matics distinction. For example, if someone says Can youpass me thc salt?, the utterance's literal meaning is that the speaker is inquiring about the hearer's ability to pass the salt. But what the utterance really does is to ask the hearer to pass the salt, of course. What is the Literal Meaning of a Sentence Links & Letters 3, 1996 43 is, the determination of the literal meaning of a sentence necessarily involves contextual information. Let us consider the following example: (4) A: I'm biger than you. (5) A: bigger (A,B) B: lin biger than you. B: bigger (B,A) (Olson & Hildyard 1983: 49) Olson & Hildyard (1 983: 49) in their paper Writing and Literal Meaning argue that (4A-B) uttered by two children are not synonymous in terms of their literal meanings as represented in (5A-B) where the references A and B are assigned to Iand you. Now, Katz's notion of context-free literal meaning would have to consider (4A-B) as being synonymous because without contextual information (5A-B) cannot be recovered, i.e., the assignment of reference can- not be done out of the context. Naturally, in the ordinary sense of literal meaning we consider (4A-B) to mean different things literally. And hence, we might want to follow Searle's view of literal meaning since otherwise we have to see the argument benveen the two boys being of redundant nature. Does this, however, mean that we want to completely abandon the con- text-free literal meaning, the notion claimed by Katz? The answer is no. Let us now consider the following: (6) A: Didyour treatmentfor stammering work? B: Peter Piper picked a peck ofpickLedpepper. (Sperber & Wilson 1986: 178) The utterance (6B) is relevant as an answer to A's question. That is, (6B) gives rise to an implicature that the treatment seems to have worked. However, (6B) does not have the sentence meaning that the treatment seems to have worked. Now in Searle's sense of literal meaning, (6B) has to have an application against some contextual background in order for its truth-conditional content to be recovered. It is true that (6B) is processed automatically by the hearer's knowledge of grammar and the hearer might access his background assump- tions concerning pickled peppers. However, these background contextual assumptions are not used to determine the truth-conditional content of the utterance i.e., what Searle claims, the literal meaning. This is because the semantic content of (6B) is not communicated to the hearer and for exarnple no referent can be assigned to Peter Piper. Here we do not want to say that the literal meaning of (6B) is the impli- cature that the treatment seems to have worked, nor do we want to say that the literal meaning of (6B) is that Peter Piper picked a peck of pickled pepper. This is because, as was mentioned, the utterance is chosen not for its seman- tic content but for its phonological properties. That is, the speaker does not 44 Links & Letters 3, 1996 Reiko Itani communicate its semantic content and in fact she could have produced any other tongue twister in order to communicate the same message. However, we cannot deny that (6B) has some semantic content although it does not constitute part of the message communicated. Now Searle's notion of literal meaning cannot capture the semantic content of (6B). However, Katz's notion of context-free meaning can explain the semantic content of (6B) because (6B) is a grammatically correct English sentence, and following Fodor (1983), grammar would automatically process the utterance, deriving a context- free grammatically determined meaning. Now what underlies the different conceptions of Katz and Searle is two different notions of semantic competence. Katz's semantic competence is con- ceived of as a component of linguistic competence, i.e. grammar, like syntac- tic competence, and is autonomous from other mental capacities. Searle's semantic competence, on the other hand, seems to be conceived of as know- ledge of truth-conditions. Knowing the meaning of a sentence is knowing how the world has to be for that sentence to be true. These are very different notions of semantics, but a psychologically ade- quate theory of utterance interpretation will give a place to both of these con- cepts. 1 would now like to give a relevance-based account of how Searle and Katz's notions of literal meaning have their places in utterance interpretation processes. 3. Katz and Searle's notions of literal meaning and relevance theory In relevance theory Katz's notion of grammatically determined context-free meaning and Searle's notion of contextually enriched truth-conditional mean- ing are manifested as logicalform and propositionalform respectively. The lat- ter is a semantically complete logical form and is capable of being true or false, whereas the former is an incomplete logical form which our grammar decodes an utterance into. This incomplete logical form can then be completed into a propositional form on the basis of contextual information. A logical form, whether truth-evaluable or not, is technically a well formed formula which can undergo logical processing, and therefore play an important role in cognition. Grammar decodes a natural language sentence into an inc:omplete logical form, and this is comparable to KatzS notion. Here, we use the term logicalform to mean incomplete logicalform which is then con- textually enriched into a complete logical f ~ r m . ~ Complete logicalform is a truth-conditional propositional form and this is comparable to Searle's notion. 5. This is based on Fodor's modular approach to human cognition. By modular, he means that the mind has a variety of specialized systems individuated by virtue of their compu- tational properties. Fodor calls these special-purpose systems input ystems, which process visual, auditory, linguistic, and other perceptual information. The output of these input systems is fed into the central systems which integrate the input information with back- ground information stored in memory, performing inferences on it, and coming to con- What is the Literal Meaning of a Sentence Link & Letters 3, 1996 45 For example, grammar decodes (4A-B) into a logical form, something like that the speaker is bigger than the hearer, which does not distinguish (4A) from (4B). The hearer, then, using contextual information, pragmatically enri- ches the logical form into truth-evaluable propositional form representations such as (5A-B). Thus, at the leve1 of propositional form, (4A) and (4B) cannot be said to be synonymous. In (6B), on the other hand, the grammar decodes the utterance into an incomplete logical form. And as we mentioned, there is no use of background assumptions here to establish who Peter Piper is. The semantic content of (6B) is not to be pragmatically enriched into a propositional form because it is not an intended meaning communicated to the hearer. However, the utterance (6B), as we mentioned, has semantic content and this can be captured as logicalform in relevance theory. Now Searle's and Katz's notions have their places in utterance interpretation processes. That is, our grammar decodes an utterance into a logical form, Katz's notion, which is then enriched into a propositional form, Searle's notion. There is no dispute to be resolved here; both concepts have a role to play in a full theory of utterance interpretation, and whether one or the other, or neither of these concepts is called literal meaning, is a matter of no interest. The notion of literal meaning is usually employed as a contrast with figurative meaning. So, for example, the meaning of john is v e y clever can be taken to be the literal ascription of cleverness to John, or something quite different if it is taken as ironical. But it is not sentences, linguistic objects, which are loose (to be discussed later) or metaphorical, or hyperbolical or ironic, or figurative in any other way. It is the use of these sentences as utterances, as communication. Thus the notion of the literal meaning of a sentence does not enter into any distinction with any other variety of a meaning that a sentence can have. If the concept of the literal meaning of a sentence makes any sense at all, it is simply as the meaning of a sentence and so it is Katz's notion - linguistic mean- ing. Searle's notion pertains to utterance meaning, not sentence as he is con- sidering sentences in use. In relevance theory, we do not talk about the literal meaning of a sentence but the notion literal interpretation has its place. This notion is important, not because it can be contrasted withjgurative use, but because it can be contrast- ed with loose use of language, which prevails in our use of language. In this theory, literalness is not the norm of language use as is often assumed (e.g. Grice's Maxim of Quality), but rather a limited case of language use. Now we are going to introduce a relevance notion of literal interpretation which is a necessary notion in utterance interpretation. clusions about the world. Now logicalfom is the output of the linguistic input system Le., grammar, and this is fed into the central systems in which inferences are performed on it, completing it into propositionalform with background information.