Experimental pragmatics and what is said: A response to Gibbs and Moise

Gibbs and Moise [Gibbs, R., Moise, J., 1997. Pragmatics in understanding what is said. Cognition 62, 51-74], present experimental results which, they claim, show that people recognize a distinction between what is said and what is implicated. They also claim that these results provide support for theories of utterance interpretation (such as Relevance Theory) which recognize that pragmatic processes are involved not only in understanding what is implicated but also in working out what is said (the 'explicature'). We attempted to replicate some of these experiments and also adapted them. Our results fail to confirm Gibbs and Moise's claims. Most significantly, they show that, under certain conditions, subjects select implicatures when asked to select the paraphrase that best reflects what a speaker has said. We suggest that our results can be explained within the framework of Relevance Theory (Sperber, D., Wilson, D., 1986. Relevance: Communication and Cognition. Blackwell, Oxford) if we assume that subjects select the paraphrase that comes closest to achieving the same set of communicated contextual effects as the original utterance. When an utterance gives rise to a single strong implicature, subjects tend to select this as the paraphrase that best reflects what is said; in other cases (for example in Gibbs and Moise's stimuli) subjects tend to select the explicature.