Relevance Theory

A cognitive theory of pragmatics originally developed in the 1980s by Dan Sperber and Deirdre Wilson. Relevance theory offers a new approach to the study of human communication which is firmly grounded in a general view of human cognitive design. With H. P. Grice, relevance theorists assume that human communication is characteristically intention-based, and so they see verbal comprehension as involving not just the decoding of speech signals, but also the recognition of the speaker’s communicative intentions.

Relevance is usually defined as a potential property of inputs (such as assumptions, thoughts, utterances) to cognitive processes. Sperber and Wilson (1995) advance two principles of relevance embodying two central claims about human cognition and communication: a first or ‘cognitive’ principle of relevance, and a second or ‘communicative’ principle of relevance. The first of this principles states that: ‘Human cognition tends to be geared to the maximisation of relevance’ (1995: 260). The second is the statement that ‘[e]very act of ostensive communication communicates a presumption of its own optimal relevance’ (1995: 260). The communicative principle of relevance does not have the same status as Grice’s Cooperative Principle and conversational maxims, for it is simply a generalisation about ostensive-inferential communication. This generalisation about communication applies without exception: it is not something that speakers ‘follow’ or can ‘opt out’ of, for example.

Relevance theory claims that linguistic communication (and, in fact, all human communication) is relevance driven. Relevance is defined within this framework as a trade-off of two competing factors: cognitive or contextual effects and processing effort. A positive cognitive effect is ‘a worthwhile difference to the individual’s representation of the world’ (Wilson and Sperber 2004: 608), such as a true belief or conclusion. According to relevance theory, an input is relevant to an individual just in case positive cognitive effects result from the processing of that input.

Dan Sperber/

Processing new information in a context may yield three main types of positive cognitive effect. First, it may yield a contextual implication deducible from the combination of new and existing assumptions, but from neither of these alone. Second, it may provide evidence that strengthens an already existing assumption. Third, it may contradict and eliminate information already held. For example, given the contextual assumption ‘If the lights are on, then Mary is home’, an utterance of the sentence ‘The lights are on’ as we approach our house may yield the contextual implication ‘Mary is home’. Contextual implications are the central type of positive cognitive effect. Processing effort, on the other hand, is the effort of perception, memory and inference that must be expended in computing cognitive effects. For instance, an utterance of a wordy and syntactically complex sentence would take more effort to process than an utterance of a less wordy and simpler version of that sentence. Likewise, an indirect answer to a question would require more processing effort than a direct one.

On the basis of these two competing factors, the relevance of an input to an individual may be comparatively assessed as follows: (1) other things being equal, the greater the positive cognitive effects achieved by processing an input, the greater the relevance of that input; (2) other things being equal, the greater the effort required in processing an input, the lower the relevance of that input. The following example, adapted from Wilson and Sperber (2004: 609), can be used to illustrate the comparative relevance of alternative inputs to an individual. Suppose that Peter, a friend of ours, asks us who we phoned last night. Let us assume, furthermore, that we phoned Kim and Sandy last night. Each of the following three alternative utterances would constitute a true and relevant answer to Peter’s question: ‘We phoned Kim and Sandy’, ‘We phoned Kim’, and ‘We phoned Kim and Sandy or 2 + 2 = 5’. However, these answers would not be relevant to the same degree: the first would be comparatively more relevant than the other two. Notice that ‘We phoned Kim and Sandy’ entails ‘We phoned Kim’, and so it is a more relevant answer because it yields the positive cognitive effects of the second utterance and more. The first utterance would be a more relevant answer than ‘We phoned Kim and Sandy or 2 + 2 = 5’ because, although these two utterances are semantically equivalent, the latter is obviously more costly to process. On the whole, when a similar amount of effort is expended in processing alternative inputs, the more relevant of these inputs is the one that yields more positive cognitive effects. Conversely, when similarpositive cognitive effects are derivable from the processing of alternative inputs, the one which is less costly to process is the more relevant.

How much relevance are individuals entitled to expect? According to relevance theory, maximal relevance is an unreasonably high expectation in communication because, for example, our interlocutors might be unwilling or unable to produce information that would yield the most positive cognitive effects for the least processing effort (Higashimori and Wilson 1996). In light of this, Sperber and Wilson (1995) have argued that while cognition tends to be geared to the maximisation of relevance, acts of ostensive communication simply create an expectation of optimal relevance. In other words, for any ostensive stimulus (for example, a verbal utterance) addressees are only entitled to expect a degree of relevance that is sufficient to warrant their effort in processing it, and which is also the highest degree of relevance that their interlocutors are able to achieve given their abilities, goals and preferences.

The presumption of optimal relevance suggests the following general comprehension procedure: ‘Check interpretive hypotheses in order of accessibility, that is, follow a path of least effort, until an interpretation which satisfies the expectation of relevance is found; then stop’ (Carston 2002: 45). Every utterance gives rise to a number of possible interpretive hypotheses that are compatible with the linguistic meaning of the sentence uttered. According to this general criterion, addressees follow a path of least effort in considering such hypotheses, stopping once they reach one that satisfies their expectations of optimal relevance. Notice that the term ‘interpretive hypotheses’ as used in the definition above includes not just the proposition the speaker intended to communicate, but also the contextual assumptions, implicatures and attitudes intended by the speaker.

The relevance-theoretic comprehension procedure does not guarantee that communication will always be successful, of course. Misunderstandings do occur from time to time. Rather, this heuristic offers an account of how addressees select the interpretive hypothesis they are entitled to assume is the one overtly intended by their interlocutors.

Relevance theory rejects the traditional assumption that every pragmatically determined aspect of utterance interpretation other than reference assignment and disambiguation must be an implicature. Central to this framework is the claim that the explicit side of communication should also fall under the scope of a theory of pragmatics (Bach 1994; Carston 2002, 2004b; Sperber and Wilson 1993, 1995). Thus, according to relevance theory, there are two types of communicated ‘assumptions’ (conceptual representations of the actual world): explicitly communicated assumptions, or ‘explicatures’, and implicitly communicated ones (implicatures). When is an assumption communicated by an utterance ‘explicit’? Sperber and Wilson (1995: 182) suggest that an explicature is an inferential development of the propositional template or ‘logical form’* encoded by an utterance. In other words, an explicature involves a combination of linguistically decodedmaterial and pragmatic enrichment. By contrast, implicatures are communicated assumptions resulting from pragmatic inference alone. Let us consider an example of a relevance-theoretic explicature involving the restriction of the domain of a quantificational expression. An utterance of the sentence ‘Every cat has white paws’ does not make a patently false universal claim about cats, but will typically be used to express a more restricted true proposition – for example, a proposition about cats in our local shelter. Thus, an explicature of this utterance would be ‘Every cat in our local shelter has white paws’. As pointed out above, this conceptual representation is a pragmatic development of the propositional schema corresponding to the conventional meaning of the sentence uttered

The class of explicatures of a given utterance within the relevance-theoretic framework includes not just the proposition expressed by that utterance, but also a range of so-called ‘higher-level’ explicatures, which are obtained by embedding that proposition under an appropriate pro – positional attitude* or speech act description (Sperber and Wilson 1993). For example, an utterance of the sentence ‘The boss is coming’ may be developed inferentially into the higher-level explicature ‘The speaker believes that the boss is coming’, or even ‘The speaker is warning that the boss is coming’.

An interesting distinction is made in relevance theory between strong and weak implicatures (and communication, more generally). The strength of an implicature crucially depends on the manifest strength of the speaker’sintention that a specific implication should be recovered. As Sperber and Wilson (1995: 197) put it: ‘Some implicatures are made so strongly manifest that the hearer can scarcely avoid recovering them. Others are made less strongly manifest’. Consider, for example, the utterance ‘I don’t like action thrillers’ as an answer to the question ‘Have you seen the latest Harrison Ford film?’. It is not difficult to see that such a response can be used to implicitly communicate the assumptions that ‘The latest Harrison Ford film is an action thriller’ and, in addition, that ‘I haven’t seen the latest Harrison Ford film’. According to relevance theory, these assumptions are strong implicatures because their recovery is essential to understand the speaker’s intended meaning. In other words, the expectation of optimal relevance that an utterance of ‘I don’t like action thrillers’ gives rise to requires that the addressee take these highly salient assumptions (or very similar ones) as being implicitly communicated. The response above may also implicitly communicate the assumption that the speaker has a general dislike for commercial blockbusters, for example. However, this assumption is a weak implicature because it need not be supplied by the addressee in order to satisfy his expectation of optimal relevance. Indeed, there are many other similar implicatures which may have been derived on the basis of the speaker’s response, for example that the speaker is not really a film enthusiast, or the speaker frowns upon the use of violence. In general, the more obvious the speaker’s communicative intentions, the stronger the communication. Conversely, the wider the range of interpretive possibilities allowed by the speaker, the weaker the communication. The relevance-theoretic approach to the implicit-explicit distinction has led to a significant reassessment of the interface between semantics and pragmatics: the existence of pragmatic aspects of pro – positional content which do not correspond to items present in the syntactic representation, as argued by relevance theory, strongly suggests that context-sensitivity at this level is widespread.

An important development within relevance theory was the recognition, due to the seminal work of Diane Blakemore (1987, 2002), that linguistic meaning can encode constraints on the inferential phase of utterance comprehension. Thus, linguistic meaning may affect the inferential processes that characterise utterance comprehension in two different ways: while a majority of linguistic expressions encode constituents of conceptual representations, there are also expressions which encode inferential procedures, which we could think of as instructions to increase the salience of a particular type of inferential process.

Blakemore justified the distinction between conceptual and procedural encoding in both cognitive and communicative terms. Since, as it is assumed in relevance theory, the interpretation of utterances involves carrying out computations over conceptual representations, it is reasonable to expect from a cognitive point of view that languages encode information about the inferential procedures in which such conceptual representations enter (and not just the constituents of conceptual representations themselves). From a communicative perspective, using expressions which encode procedures for the identification of intended cognitive effects would obviously reduce the processing cost involved in achieving those effects, a result that is in consonance with the communicative principle of relevance. For example, Blakemore (2002) links the use of the sentential connective ‘but’ with the cognitive effect of contradiction and elimination. Hence, according to Blakemore, the use of ‘but’ activates an inferential process resulting in the contradiction and elimination of an assumption which the speaker has reason to believe is accessible to the hearer. For example, an utterance of the sentence ‘Kim is rich, but unhappy’ activates an inferential process whereby the hearer contradicts and eliminates the accessible assumption that wealth leads to happiness.

Research in relevance theory has also made an important contribution to the study of figurative language. Verbal irony, for example, is analysed in this framework as an ‘echoic use’ of language in which the speaker dissociates himself tacitly from an attributed utterance or thought (Wilson 2006; Wilson and Sperber 1992). For example, uttering ‘You really are good at this!’ after a friend has failed to score an easy goal in a quick counter-attack can be construed as an instance of verbal irony because we are tacitly dissociating ourselves from a thought or utterance with a similar content (such as a reassurance that our friend is a skilful footballer) which may have been attributed to us had the circumstances of the game been different. This analysis is a departure from more traditional Gricean accounts,according to which verbal irony constitutes an overt violation of the Quality maxim, thus giving rise to a related true implicature which contradicts the literal meaning of the sentence uttered. The relevance-theoretic approach to verbal irony is in harmony with the framework’s assumption that an expectation of relevance, rather than one of truthfulness, is a standard of verbal communication (Wilson and Sperber 2002).

A line of research in relevance theory explores in detail the idea that the interpretation of words is also highly context-sensitive, and that lexical context-sensitivity is not restricted to indexical expressions. This line of investigation puts forward the hypothesis that lexical-pragmatic processes such as narrowings, broadenings, approximations and metaphorical extensions are the result of a single pragmatic procedure which fine-tunes the conventional meaning of words in communication (Carston and Powell 2006; Wilson 2003).

In sum, research carried out within the framework of relevance theory has yielded interesting and insightful results. First, it has provided a cognitive alternative to Gricean and neo-Gricean theories of pragmatics. It has also introduced (or thrown new light into) a number of important theoretical concepts (explicature, echoic uses of language, strength of communicated assumptions) and distinctions (decoding versus inference, explicature versus implicature, conceptual versus procedural meaning, interpretive versus descriptive uses of language, saying versus implicating) in the study of meaning in natural language. Moreover, it has helped to enhance our understanding of the semantics-pragmatics interface by arguing controversially that the contribution of pragmatics to the propositional content of utterances goes far beyond disambiguation and reference assignment.

Primary sources
Blakemore, Diane (1987). Semantic Constraints on Relevance. Oxford: Blackwell. Blakemore, Diane (2002). Relevance and Linguistic Meaning: The Semantics and Pragmatics of Discourse Markers. Cambridge: Cambridge University Press.
Carston, Robyn (1988). ‘Implicature, explicature and truth-theoretic semantics’. In Ruth Kempson (ed.), Mental Representations: The Interface between Language and Reality. Cambridge: Cambridge University Press. 155–81.
Carston, Robyn (2002). Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford: Blackwell Publishing.
Carston, Robyn (2004a). ‘Relevance theory and the saying/implicating distinction’. In G. Ward and L. Horn (eds), The Handbook of Pragmatics. Oxford: Blackwell. 633–56.
Carston, Robyn (2004b). ‘Explicature and semantics’. In S. Davis and B. S. Gillon (eds), Semantics: A Reader. Oxford: Oxford University Press. 817–45.
Carston, Robyn and George Powell (2006). ‘Relevance theory – new directions and developments’. In E. LePore and B. Smith (eds), Oxford Handbook of Philosophy of Language. Oxford: Oxford University Press. 341–60.
Higashimori, Isao and Deirdre Wilson (1996). ‘Questions on relevance’. UCLWorking Papers in Linguistics 8: 111–24.
Sperber, Dan and Deirdre Wilson (1987). ‘Précis of relevance: communication and cognition’. Behavioral and Brain Sciences 10: 697–754.
Sperber, Dan and Deirdre Wilson (1993). ‘Linguistic form and relevance’. Lingua 90: 1–25.
Sperber, Dan and Deirdre Wilson (1995). Relevance: Communication and Cognition. Oxford: Blackwell. First edition 1986.
Wilson, Deirdre (2003). ‘Relevance theory and lexical pragmatics’. Italian Journal of Linguistics/Rivista di Linguistica 15: 273–91.
Wilson, Deirdre (2006). ‘The pragmatics of verbal irony: echo or pretence?’. Lingua 116: 1722–43.
Wilson, Deirdre and Dan Sperber (1992). ‘On verbal irony’. Lingua 87: 53–76.
Wilson, Deirdre and Dan Sperber (2002). ‘Truthfulness and relevance’. Mind 111: 583–632.
Wilson, Deirdre and Dan Sperber (2004). ‘Relevance theory’. In G. Ward and L. Horn (eds), The Handbook of Pragmatics. Oxford: Blackwell. 607–32.

Further reading
Bach, Kent (1994). ‘Conversational impliciture’. Mind and Language 9: 124–62.
Blakemore, Diane (1995). ‘Relevance theory’. In J. Verschueren, J. Östman and J. Blommaert (eds), Handbook of Pragmatics. Amsterdam: Philadelphia. 443–52.
Blakemore, Diane (2002). Understanding Utterances: An Introduction to Pragmatics. Oxford: Blackwell. Huang, Yan (2007). Pragmatics. Oxford: Oxford University Press.
Wilson, Deirdre (1999). ‘Relevance and relevance theory’. In R. Wilson andF. Keil (eds), MIT Encyclopaedia of the Cognitive Sciences. Cambridge, MA: MIT Press. 719–22.

Source: Key Ideas in Linguistics and the. Philosophy of Language. Edited by Siobhan Chapman and Christopher Routledge. Edinburgh University Press. 2009.

Categories: Linguistics, Literary Criticism, Literary Terms and Techniques, Literary Theory

Tags: , , , , , , , , , , , , , ,

Your feedback helps improve this platform. Leave your comment.

%d bloggers like this: