Shared assumptions: semantic minimalism and Relevance Theory

Cappelen & Lepore (2006a) claim that linguistic communication requires shared content and criticise Relevance Theory on the grounds that it makes content sharing impossible. I argue that this criticism rests upon two important errors. The first is a flawed understanding of Relevance Theory, which causes Cappelen & Lepore to apply the perspective of an objective, omniscient third party to parts of Relevance Theory that in fact only make reference to (and only depend upon) the subjective judgements made by the addressee of an utterance. The second is a confusion about different definitions of content. The kind of evidence that Cappelen & Lepore produce in fact involves the communication of what they term Speech Act content, a kind of content that few if any approaches require to be perfectly ‘shared’. Certainly, Cappelen & Lepore’s own (2005) position makes no such requirement. Consequently, their arguments have no more force against Relevance Theory than they would if directed at their own ideas. Having shown this, I compare Relevance Theory with Cappelen & Lepore’s semantic minimalism more generally and conclude that most of the apparent differences between them lie in terminology and rhetorical style. Instead of highlighting these spurious differences, practitioners of the two approaches have a good deal to learn from each other. In particular, I argue that relevance theorists should avoid lapsing into a brand of semantic orthodoxy that Cappelen & Lepore call moderate contextualism, something that requires them to emulate Cappelen & Lepore’s uncompromising minimalism with regard to encoded