On common knowledge in conversation

Abstract

This paper aims towards a coherent account of the sort of common knowledge that is necessary for making felicitous utterances in a conversation. The hypothesis to that effect, originating from Lewis (1969) and Schiffer (1972) is presented in section 1 along with the standard definition of common knowledge.

However, as Clark and Marshall (1981) observed, that definition leads to a paradox; in section 2, I discuss the possibilities of avoiding it. I argue (contra Clark) in favor of a fixed-point approach à la Barwise (1989), which relies on the notion of coordination devices. Since the argument about the common knowledge paradox resorts to the cognitive plausibility of the notion, in section 3 I try to clarify its empirical status by considering some relevant data.

This being done, the outlook of this paper is sketched in section 4, being a genetic account of coordination devices in terms of some simple games. Its purpose is to supplement the fixed-point approach to common knowledge in parallel with Clark's (1996) psychological account. This concludes with an attempt at a cognitively plausible definition of common knowledge.