M&R Ch 5: A deflationary account of metaphors

3 Feb

This is my colleague Chris Cocchiarella’s summary of the fifth 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.

In Chapter 4, Wilson and Sperber concluded, “metaphor and irony are not rhetorical devices involving codified departures from the ordinary use of language, but ordinary exploitations of basic processes of verbal communication” (96).   In Chapters 5 and 6, W&S deal with metaphor and irony in more detail.  In Chapter 5, they call their approach to metaphor “deflationary”: “we see metaphors as simply a range of cases at one end of a continuum that includes literal, loose and hyperbolic [or figurative] interpretations” (97).  In other words, metaphor is nothing extraordinary.  It is just an instance of weak implicature.  Weak implicature, moreover, is just one point on a continuum of strong explicature (explicit/literal words), weak explicature (illocutionary force indicators like moods), strong implicature (loose language use), and weak implicature (metaphor). 

W&S make two major points about their “deflationary” approach to metaphor in Chapter 5.  First, they argue that there is no specific cognitive mechanism or brain module behind metaphor.  They do agree, of course, that there is a “comprehension mechanism” that creates expectations of relevance (102).  This mechanism is what W&S called the relevance-guided comprehension heuristic or process—a central postulate of Relevance Theory (RT)—and this process gives rise to strong and weak explicatures and implicatures.  In RT, metaphors naturally “emerge” from relevance-guided comprehension as instances of weak implicature (114).

Second, W&S use their “deflationary” approach to argue against what earlier speech-act theorists called a “norm of literalness,” or the idea that the human mind prefers to make literal interpretations by default, only resorting to metaphoric interpretations if literal interpretations do not work at first.  Since metaphors are just ordinary instances of weak implicature that emerge from relevance-guided comprehension, there is no reason to believe that the mind prefers literal interpretations by default.  The mistaken idea that the mind defaults to literal interpretations comes from the “code model” of communication, which says, “language use is governed by a norm of literalness” (98).  S&W criticize earlier speech-act theorists like Grice and Searle for trying to understand metaphors through a norm of literalness that assumed a code model of communication:

Grice tended to take for granted—and Searle explicitly argued—that when someone uses language to communicate, she is presumed to express her meaning literally. . . .  Metaphors and other tropes, where the linguistic meaning of the utterance is not even part of the speaker’s meaning, are exceptional in this respect: Grice suggested that in metaphor, the speaker is not really saying what she appears to be saying, but merely ‘makes as if to say’ it (101).

In contrast to Grice and Searle’s ‘norm of literalness,’ S&W argue “that verbal comprehension involves no presumption of literalness and no default interpretation, and that metaphors are in no way exceptional” (101).  In place of a ‘norm of literalness,’ S&W evoke RT to put forward the cognitive principle of relevance (cognition tends to maximize relevance) and the communicative principle of relevance (ostensive communication conveys a presumption of optimal relevance).  Relevance, in RT, is defined as creating a positive cognitive effect (e.g., confirming, disconfirming, or modifying assumptions) through minimal processing effort (e.g., minimum strain on memory or mental inference).  According to relevance-guided comprehension, say S&W, “the very same procedure that yields a literal interpretation . . . would yield a non-literal interpretation in others” (105).  Relevance-guided comprehension does not default to literal interpretation but simply tends toward the interpretation—whether literal, loose, or metaphorical—most relevant to the context in which communication occurs.

For example, “Holland is flat” is not intended as a literal statement but as loose talk, a form of strong implicature.  It is a statement of “approximation” (106).  It would actually require unnecessary processing effort to try to interpret that statement literally, and there would be no positive cognitive effect from doing so.  Another example is “Joan is a saint,” which is metaphoric or hyperbolic, a form of weak implicature.  Assuming Joan is not literally a saint, trying to process this statement literally would involve pointless effort, and no relevant cognitive effect would come of it.  In this case, say W&S, “the encoded concept [the word ‘saint’] helps activate contextual implications that make the utterance relevant as expected, and the concept conveyed by the hyperbole/metaphor is one of an outstanding type of kindness characterized by these implications” (110).  In all such examples, the same relevance-guided comprehension procedure may yield a literal, loose, or metaphoric interpretation, depending on which interpretation is most relevant.  For a detailed example of how relevance-guided comprehension can yield a metaphoric interpretation, see Figure 5.2 below (113).

W&S Ch. 5, Fig 5.2

W&S Ch. 5, Fig 5.2

Before concluding Chapter 5, W&S make two minor points about metaphor.  First, since metaphors are weak implicatures—that is, they ‘weakly’ imply a broad range of meanings—they are “probabilifications” and not statements of absolute certainty.  So metaphoric interpretations are probable, not certain.  Second, W&S acknowledge how metaphors are suitable for producing “poetic effects.”  While they stress that most metaphors are just normal instances of weak implicature, they also recognize, “In more creative metaphors, relevance may depend to a much greater extent (or even entirely) on such weak implicatures, in a way that makes it quite appropriate to talk of ‘poetic effects’” (121).  Poetry is full of creative metaphors that implicate broad ranges of cognitive effects—that is, poetic metaphors reveal multiple dimensions of meaning.  For that reason, poetic metaphors usually also require more processing effort—that is, they typically must be read and re-read to disclose their many meanings.


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

%d bloggers like this: