Grice, Conversational Implicature and Philosophy

Abstract

The importance of Grice’s theory of conversation and in particular his account of conversational implicature (Grice in Studies in the way of words. Harvard University Press, Harvard, pp. 22–40, 1975; 1978) in the development and current concerns of pragmatics is almost impossible to exaggerate. Whether or not they agree with the details or even the broader framework of Grice’s theory, pragmaticists generally acknowledge the significance of his attempt to give a formalised account of the differences between what our words literally mean and what we intend to communicate in using them. But Grice himself was a philosopher, not a linguist; his work was deeply rooted in the philosophical preoccupations of the mid twentieth century, and he never used the word ‘pragmatic’ in his writings as it is used in present day linguistics. This chapter will address the contrast between Grice’s philosophical motivations in developing his account of conversational implicature, and the linguistic framework in which it has subsequently generally been discussed. It will do so by considering the philosophical context in which Grice was working and some of the specific philosophical problems to which he applied his notion of conversational implicature. It will begin with a review of the dichotomy in twentieth century analytic philosophy that can be summarised as a distinction between ‘ideal language’ and ‘ordinary language’ philosophy, and will discuss Grice’s work as an attempt to demonstrate some fundamental misconceptions in both positions. In doing so, it will compare Grice’s work on conversational implicature with the near contemporary work by Austin on speech acts (Austin in How to do things with words. Oxford University Press, Oxford, 1962a; Sense and sensibilia. Oxford University Press, Oxford, 1962b). Austin’s work shares a number of apparent similarities with Grice’s, but reveals some significant philosophical differences, particularly with regard to the nature of ‘literal meaning’ and the role of theorising. This chapter will then offer an exegesis of Grice’s conception of conversational implicature, including the distinction which seemed necessary to him between ‘generalised’ and ‘particularised’ conversational implicatures, and some of the properties that he identified as characteristic of conversational implicatures. Following this exegesis, this chapter will consider some of the applications and extensions of the concept of conversational implicature suggested by Grice and, in some cases, developed further by his later commentators. Grice found his concept to be fruitful in addressing a range of established philosophical problems, including the viability of his own earlier work on ‘non-natural meaning’ (Grice in Studies in the way of words. Harvard University Press, Harvard, pp. 213–223, 1957), the contrasting claims of realism and skepticism (Grice in Studies in the way of words. Harvard University Press, Harvard, pp. 147–153, c. 1946–1950; pp. 154–170, c. 1953–1958; pp. 224–247, 1961), apparent differences between logic and natural language (Grice in Studies in the way of words. Harvard University Press, Harvard, pp. 3–21, 1967a; pp. 58–85, 1967b) and the debate over Russell’s logical account of definite descriptions (Grice in Studies in the way of words. Harvard University Press, Harvard, pp. 269–282, 1981). Grice introduced the technical term ‘implicature’ into his philosophy of language. It has subsequently become part of the defining terminology of present day pragmatics and is a central concept both for those working in a broadly neo-Gricean framework (Horn in A natural history of negation. University of Chicago Press, Chicago, 1989; Pragmatics. Palgrave, Basingstoke, pp. 158–183, 2007; Levinson in Presumptive meanings: The theory of generalized conversational implicature. M.I.T. Press, Cambridge, 2000) and for those working in relevance theory (Sperber and Wilson in Relevance. Blackwell, Oxford, 1995; Carston in Thoughts and utterances. Blackwell, Oxford, 2002). Recent interest in pragmatics has focussed on the division that Grice drew between ‘what is said’ and ‘what is implicated’, and on the viability of that distinction. This has been a focus of study in both theoretical pragmatics (Borg in Minimal semantics. Oxford University Press, Oxford, 2004; Recanati in Literal meaning. Cambridge University Press, Cambridge, 2004; Cappelen and Lepore in Insensitive semantics: a defense of semantic minimalism and speech act pluralism. Blackwell, Oxford, 2005) and experimental pragmatics (Gibbs, Brain and Language 68: 466–485, 1999; Glucksberg in Experimental pragmatics. Palgrave, Basingstoke, pp. 72–93, 2004; Breheny et al., Cognition 100: 434–463, 2006). This chapter will conclude with an assessment of the significance of Grice’s delineation of ‘what is said’ as a defining opposite of ‘what is implicated’. For Grice himself, although it was central to his original philosophical motivations, this remained one of the most troublesome and least successful aspects of his philosophy of language. For pragmaticists it has proved one of the most enduring, challenging and intriguing topics of debate.