That Dummett's anti-realist position has attracted a good deal of attention is partly due to its revisionary character. We appear to be provided with a purely philosophical argument which shows that ordinary practice is based on a deep misconception about the way we succeed in talking about the world. Perhaps the majority of contemporary philosophers (in the wake of Wittgenstein, Quine and the pragmatists) find revisionism of this sort hard to swallow: it smacks of a foundationalist appeal to a first philosophy. In this chapter we examine both the possibility of the revisionary position itself and the nature and extent of revision contemplated.

One note of warning: Dummett should not be identified with verificationists or positivists, despite his interest in verificationist accounts of meaning. Those views, in decreeing that certain sentences are vacuous or meaningless (typically those for which we lack a means of verification), are revisionary about content. In contrast, Dummett thinks that it would be quite wrong to question first-order practice in this way. His position begins from an acceptance that we understand a certain range of sentences (those which, by ordinary standards, we are competent to use) but then goes on to ask what that understanding consists in. So he couldn't espouse a revisionism in verificationist style. To put the point in terms of the notion of content, Dummett is not revisionary about which contents we grasp, he only suggests that, given our natural inclinations towards realism, we are apt to misconstrue the nature of that understanding.