Wittgenstein's Contextualism

In Coliva (2010a), I have proposed an interpretation of Wittgenstein's criticism of Moore's use of “I know” in relation to his truisms and the premises of his proof, as well as a defense of Wittgenstein's position against John Searle's (1969) allegation that his remarks were based on an “assertion fallacy,” which rely on the decisive role of use in the determination of meaning.

Contrary to Moore and Searle, Wittgenstein did not think that different occurrences of “I know” would always have the same meaning or, indeed, that they would always have a meaning. That is, the same words, by occurring in different contexts of use would change their meaning for him, up to the point of losing it altogether, in some deranged circumstances. By contrast, for Moore and Searle they would always retain the same meaning and could at most be used to make assertions that, depending on the case at hand, would be felicitous or infelicitous. Wittgenstein was therefore a contextualist of sorts. Yet, his form of contextualism was utterly different from contemporary variants of it. Indeed, as we will see, by contemporary lights, his position would come close to invariantism. Still, before venturing into that issue, let us see in more depth what his contextualism amounted to.

The Ordinary Use of “I Know”

As is well known, Wittgenstein accorded pride of place to ordinary language. By observing how we normally use “I know,” he identified several criteria that are operative in the ordinary (or empirical) use of “I know,” henceforth “I Know” with a capital “K.” According to him, only by conforming to these criteria would knowledge claims make sense, in the literal sense of having a meaning. Here they are:

• (i) knowledge claims make sense only when they are based on reasons,

• (ii) and, in fact, on reasons that are either noncircular, or at least stronger than what they are supposed to ground;

• (iii) they make sense only when it is possible to make an inquiry to confirm that things are as one claims to know and yet the possibility that things may not be so is open;

• (iv) and it is therefore possible to say “I don't know.” Finally and surprisingly,

• (v) claims to knowledge make sense only when they are relevant within the communicative exchange.