Is inference necessary to pragmatics

Abstract

The assumption that pragmatic processes are inferential is standardly held by Gricean and post-Gricean pragmatic theories. Recently, however, it has been challenged by accessibility-based approaches to pragmatics. Recanati (2002, 2004) proposes that primary pragmatic processes (i.e. processes that contribute to the recovery of the explicit content of the utterance) are underpinned by a simple dynamics of activations-and-associations, with no need for any further (specifically inferential) step or ‘confirmatory stage’ to warrant the selected interpretation. Mazzone (2009, 2011) extends this account to secondary pragmatic processes (i.e. processes of implicature derivation): the recovery of the explicit and implicit content of the utterance is the result of a unified associative comprehension process. This paper argues that, on close analysis, the role played by information about the speaker’s mental states (i.e. her beliefs and intentions) indicates that inference is indispensable in an adequate account of pragmatic processing.