Summary: Linguistics and the a priori

Abstract

I. The best way to show that a thing can be done is to do it: according to the old maxim, valet illatio ab esse ad posse. (It is legitimate to infer from existence to possibility.) In the following pages I shall draw several philosophical conclusions on the basis of various applications of structural linguistics. Yet, as appears from a good number of oral and written exchanges, there is a strong current of opinion challenging the possibility of such a move. Moreover, and this gives me pause, the opposing voices are not restricted to philosophers who are sceptical towards any kind of linguistic approach, nor even to those who regard appeals to ordinary language with suspicion. Indeed, the new wave of attack has been launched by authors belonging to or influenced by the Oxford School and by followers of the later Wittgenstein — by philosophers, in other words, who are very much concerned with ordinary language. Gilbert Ryle, for instance, in his ‘Ordinary Language’ and more explicitly in ‘Use, Usage and Meaning’, seems to imply that the results of linguistic science have no utility on the level of philosophical analysis (Chapters II and III above). And Stanley Cavell, in his ‘Must We Mean What We Say?’ and ‘The Availability of Wittgenstein’s Later Philosophy’, makes the same claim even more forcefully that Ryle, and from a strictly Wittgensteinian point of view (Chapters VIII and IX above).