Inferential Pragmatics and Epistemic Vigilance

Current research on linguistic communication is grounded on the well-established assumption that speakers typically communicate more than they linguistically encode (Grice, 1989). This raises the question of what sources of information and types of cognitive operation drive the recovery of the communicated meaning (or ‘speaker’s meaning’). In this thesis, I argue for the following two claims: (i) pragmatic interpretation is ‘inferential’ in the sense that it relies on two distinct stages of ‘hypothesis formation’ and ‘hypothesis confirmation’. While hypotheses about the speaker’s meaning are constructed on the basis of linguistic evidence and available contextual assumptions, they are assessed against a criterion of pragmatic acceptability based on consideration of the speaker’s mental states (i.e. her beliefs and intentions); (ii) This two-stage process is underpinned by the interaction of two distinct systems: a relevance-guided comprehension procedure (Sperber & Wilson, 1995; Wilson & Sperber, 2004) and epistemic vigilance mechanisms, which assess the quality of incoming information and the reliability of the individual who dispenses it (Sperber et al., 2010). On the basis of this work, I develop a new model of the relationship between comprehension and epistemic assessment, and discuss its implications for the unfolding of pragmatic and epistemic vigilance capacities in the child’s cognitive development, as well as the place of these systems in a modular view of the cognitive architecture of the mind. Finally, I uncover some significant methodological issues which arise in the current experimental pragmatics literature when the cognitive distinction between comprehension and epistemic assessment is overlooked