- Geach's 'Refutation' of Austin Revisited
A characteristic move of what is known as 'ordinary language philosophy' (OLP), as exemplified by J.L. Austin's discussion of knowledge in 'Other Minds,' is to appeal to the ordinary and normal use(s) of some philosophically troublesome word(s), with the professed aim of alleviating this or that philosophical difficulty or dispelling this or that philosophical confusion. This characteristic move has been criticized widely on the grounds that it rests on a conflation of 'meaning' and 'use'; and that criticism has been quite successful in its effect: OLP is widely held nowadays within the mainstream of analytic philosophy to have somehow been refuted or otherwise seriously discredited. However, that the words in question do indeed have something referable to as 'their meaning,' which is not only conceptually distinguishable from their ordinary and normal uses, but also theoretically separable from these uses,1in a way that renders misguided the ordinary language philosopher's characteristic appeal and validates the traditional concerns OLP set itself out to dispel, has [End Page 41] for the most part merely been presupposed and insisted on, as opposed to argued for, by detractors of OLP.
One exception to this is the so-called Frege-Geach argument. This argument has mostly been discussed in the literature as a serious challenge to 'expressivist' or 'non-cognitivist' theories of moral discourse — a complex issue, which is still very much alive in the literature and which I must here set aside in its entirety. But the argument has also been cited as, and was originally taken by Geach to constitute, a rather powerful, even decisive, consideration in favor of the prevailing conception of (word) meaning, and (hence) against OLP. This paper revisits the Frege-Geach argument, as originally presented by Geach, with the aim of showing that as an argument against OLP and its procedures it fails: it does not show what it has widely been taken to show, and does not undermine OLP, as exemplified by 'Other Minds.' On the contrary, a close examination of Geach's argument will actually validate Austin's general approach — though importantly not all of his specific contentions about how 'know' and cognates function in ordinary and normal discourse.
By way of providing the reader with an initial orientation, let me sketch the basic conflict, as I see it, between OLP as represented by Austin and the traditional perspective that Geach seeks to defend. The basic assumption made by those who accuse OLP of conflating meaning and use, and which Geach's argument was supposed to validate, is that — barring ambiguity, indexicality, etc. — there is first something that the words of a syntactically well-formed sentence, pretty much by themselves and in virtue of something called 'their meanings,' say (or 'express'); and then there is the further 'pragmatic' question of use, which is thought of, roughly, as the question of what a speaker's point might be in saying what his words, anyway and irrespective of his point, say.
Commonly, the meaning of at least most words is thought of by the detractors of OLP to be a matter of their 'reference.' Williamson puts the basic idea thus: '[E]xpressions refer to items in the… world, the reference of a complex expression is a function of the reference of its constituents, and the reference of a sentence determines its truth value' (2007, 281). This is taken to be true not just of words like 'chair' and 'red,' but also of philosophically troublesome words like 'know,' 'mean,' 'understand,' 'cause'…. Thus, 'know' and cognates, for example, are taken to have their meaning in virtue of having come to refer to a particular sort of item in the world — namely, the relation of knowing, which is supposed to sometimes hold between potential knowers and facts (or true propositions). 'Serious and literal' utterances of syntactically well formed indicative sentences featuring these words are accordingly taken to 'ascribe' — barring ambiguity, indexicality, etc., but otherwise [End Page 42] by virtue of their meaning alone — knowledge of some particular fact or proposition to some particular knower.
OLP, as I understand it, questions the prevailing conception of meaning, and...