I have shown that Dummett simply builds the requirement of
knowledge-that into his theory of meaning. I have presented reasons
against doing this. My goal has been to provide a convincing case that
a Wittgensteinian-inspired picture of language as a skill meets the
challenges set by Dummett, and does so better than Dummett's own
account.
The view defended here explains why language has the wide-ranging functionality which Dummett correctly identifies, including its power to let us make differences ``...to what subsequently happens.''37 Language functions as a vehicle of thought and allows us to communicate with one another, and to do any of the myriad of things we do with it because it is part of the warp and weft of human life. The interplay between linguistic exchange and related action finds a natural place in the practice conception of language. On this conception, language is a practice (more accurately, a set of practices) of a community, and it is only in virtue of this that it can meet Dummett's high, though quite reasonable, expectations. The arguments of this paper show that none of this requires the introduction of knowledge-that. If, indeed, language is to be the anything par excellence, it is the human activity or skill par excellence.