M&R Ch. 3: Truthfulness and relevance

11 Jan

This is my colleague Chris Cocchiarella’s summary of the third chapter of Wilson and Sperber’s (2012) Meaning and Relevance. We previously posted Chris’s summary of the book’s preface, an intro to some of the terms and concepts of Relevance Theory, and a his summary of the introductory Chapter 1. All the other chapters  before the present one have also received summaries. (Click on the “Relevance Theory” tag link on any of these, and you’ll get the whole list.) I also posted a claim that Relevance Theory matters to rhetoric and TC.

Having explained the meaning of inferential pragmatics in Chapter 1 and word-concept relationships in Chapter 2, Sperber and Wilson move to the topic of truth in Chapter 3.  Any discussion of truth can start with the truism that hearers generally expect speakers to say things that are true.  Does this mean, as some philosophers of language believed (e.g., David Lewis and Paul Grice), that language use is governed by specific rules or conventions of ‘truthfulness,’ which enforce trust among speakers and hearers?  For example, Lewis thought that a ‘convention of truthfulness’ makes hearers and speakers trust one another with respect to what they say, especially what they literally say (if this so-called ‘convention of truthfulness’ sounds vague, S&W will explain why, as we shall soon see).  

Basically, S&W disagree with Lewis in at least two ways.  First, language use is not necessarily governed by some rule or convention of ‘truthfulness’ (but by expectations of relevance, of which truth is a by-product).  Second, truth is created not only by what can be literally said (but also by what can be loosely and figuratively said, depending on expectations of relevance).

First, S&W argue, “language use is not governed by any convention or maxim of truthfulness in what is said” (47).  Instead, they insist that language use is governed by expectations of relevance, and relevance is what makes truth possible, not vice versa: “expectations of truthfulness—to the extent that they exist—are a by-product of expectations of relevance” (48).  What do S&W mean when the say truth is a by-product of relevance?  Simply put, a linguistic utterance is recognized as true when it is already inferred as meaningful or relevant: “an utterance is relevant when the hearer, given his cognitive dispositions and the context, is likely to derive some genuine knowledge from it” (60).  This inferential nature of communication explained by Relevance Theory (RT)[1] is what allows hearers to interpret true conclusions about what speakers say:

This relevance-theoretic account not only describes a psychological process but also explains what makes this process genuinely inferential: that is, likely to yield true conclusions (in this case, intended interpretations) from true premises (in this case, from the fact that the speaker has produced a given utterance, together with contextual information) (66).

Second, as RT also has shown, language can communicate truth not just literally but also figuratively or loosely, depending on how expectations of relevance interact with words and context to produce interpretations.  While literal interpretations are the least dependent on context, “figurative interpretations are radically context-dependent” (51); loose interpretations are in between, since they involve “an expression applied to items that fall outside its linguistically determined denotation,” such as rough approximations (e.g., “Holland is flat”) (54).  In sum, true linguistic expressions are not only created by literal interpretations but also by loose and figurative interpretations.

The fact that truth can be spoken not only through literal but also through loose and figurative expressions (a fact well known by artists and poets) explains why the so-called ‘convention of truthfulness’ is just not true.  According to S&W, “Without such an appeal to literal meaning in the determination of what is said, the claim that there is a maxim or convention of truthfulness in what is said would be, if not vacuous, at least utterly vague” (50).  So there are not necessarily rules of ‘truthfulness’ that govern cognition or language use.  However, there is an inferential process in cognition and language use that ‘maximizes’[2] relevance, and this process creates meaning through a continuum of literal, loose, and figurative language—or degrees of stronger or weaker explicatures and implicatures (see Preface and Chapter 1 summaries).  Truth, in sum, emerges from relevance (or, as S&W say, truth is a ‘by-product’ of relevance).

S&W’s explanation of truth as a by-product of relevance explains why, beyond literal truth, there is loose and figurative truth communicated in quotidian and poetic language.  To repeat the earlier premises of RT, stronger and weaker degrees of explicatures and implicatures create a continuum of literal, loose, and figurative interpretations that satisfy expectations of relevance:

Literal, loose, hyperbolic or metaphorical [i.e., figurative] interpretations are arrived at by exactly the same process [the relevance-guided comprehension heuristic or procedure], and there is a continuum of cases which cross-cut these categories (74-75).

Beyond literally true language, when a speaker communicates via loose or figurative language, more processing effort is required by the hearer to create broader cognitive effects (i.e., wider ranges of meaning), resulting in figuratively true language.

The more metaphorical the interpretation, the greater the responsibility the hearer has to take for the construction of implicatures (i.e. implicit premises and conclusions), and the weaker most of these implicatures will be.  Typically, poetic metaphors have a wide range of potential implicatures, and the audience is encouraged to be creative in exploring this range (a fact well recognised in literary theory since the Romantics) (76).

S&W thus differ from earlier theorists (particularly Lewis and Grice) about the relation between truth and relevance in at least two ways.  First, language use is governed by expectations of relevance, of which truth is a by-product.  Second, since relevance creates a continuum of literal, loose, and figurative language, truth can be literal, loose, or figurative.[3]  S&W conclude:

So, yes, hearers expect to be provided with true information.  But there is an infinite supply of true information which is not worth attending to.  Actual expectations are of relevant information, which (because it is information) is also true.  However, we have argued that there is just no expectation that the true information communicated by an utterance should be literally or conventionally expressed, as opposed to being explicated or implicated (83).

Having explained how truth can go beyond literalism, S&W put RT in a position to directly address rhetoric and figurative language such as metaphor and irony.  The implications of RT for rhetoric, metaphor, and irony are detailed in chapters 4, 5, and 6, respectively.


Wilson, Deirdre and Sperber, Dan.  (2012).  Meaning and Relevance.  Cambridge: Cambridge University Press.

[1] Again, according to RT, the mutual adjustment of stronger or weaker explicatures and implicatures creates positive cognitive effects through minimal processing effort, which satisfies expectations of relevance that result in acceptable and accessible interpretations.  This process is called the “relevance-guided comprehension heuristic” (7) or the “relevance-theoretic comprehension procedure” (74).  See Preface and Chapter 1 summaries.

[2] As I parenthetically opined in my summary of chapter 1, instead of saying that we ‘maximize’ relevance, I believe S&W should rather say that we ‘satisfice’ relevance, since “optimal relevance” is just an interpretation that is “relevant enough to be worth processing” (64).

[3] For readers familiar with the theories of Grice and Lewis, S&W also discuss in chapter 3 how their notions of ‘literal meaning’ or ‘what is said’ (in contrast to implicature) are better rethought as S&W’s notion of ‘explicature’:

In our account [Relevance Theory], we give theoretical status to the notions of explicature and implicature (roughly, the explicit and implicit contents of utterances, but not to the notions of literal meaning or what is said.  Indeed, we introduced the ‘explicature’, on the model of Grice’s ‘implicature’, because we doubt that there is any common-sense notion of what is said capable of playing a useful role in the study of verbal comprehension (77).


%d bloggers like this: