Saying and Meaning: A look back at the Gricean Program

Paul Grices project of analyzing speaker meaning (initiated in his paper
Meaning, published in 1957) and his theory of conversation (developed in his
William James lectures at Harvard in 1967) have had, and continue to have, a
major influence on the development of both linguistic semantics and pragmatics
and the philosophy of language. The analysis of speaker meaning generated a
cottage industry of counterexample and patch that flourished for ten years or
so. It then faded away, as such industries tend to do, but the root idea of
Grices analysis still affects the way we think about what people mean. The
theory of conversation helped to shape the development of semantic and
pragmatic theory and provided some conceptual tools that have become standard
resources of philosophical analysis and argument. In this informal talk, I
plan
to look back at the central aims of and motivations for the Gricean
project, as
I understand them, and at some of the problems the project encountered - to
see
how they look in light of what has happened since.. I will consider what
problem the analysis of meaning was trying to solve, and how that problem
relates to the theory of conversation.

The patterns of reasoning brought to light by  Grices discussions of meaning
and conversation are, as many people have noted, a kind of strategic reasoning
- reasoning in which agents interact, basing their decisions about what to do
in part on expectations about the rational actions of others. So Grices
enterprise points naturally to game theory - the theory designed to provide
abstract models of strategic interaction. I will look at some simple games in
which moves are naturally interpreted, intuitively, as acts of meaning
something, and consider whether the facts about what is meant in such acts can
be explained in terms of assumptions about the beliefs and intentions of the
agents. I want to suggest that some of the problems Grice encountered in
developing the details of his analysis of meaning are reflected in problems in
explaining, in the game-theoretic context, what it is for an action to be an
act of meaning, or expressing, something.