Abstract

Durnmett's suspicions about classical logic are based upon his view that a theory of meaning is adequate only if it is a use theory. His critics have emphasized, however, that there are a variety of use theories that generate classical logic. To ascertain whether these critics succeed against Dummett, we need to consider a second demand he places on a theory of meaning---what we might call his anti-holism demand. The anti-holism demand says that a theory of meaning is adequate for the purpose of justifying deduction only if the theory does not exhibit a certain type of holism---a holism about the meanings of the logical constants. ;The problem is that Dummett's characterizations of the content and motivation of the anti-holism demand are notoriously unclear. In Chapter Two, I specify a motivation for the anti-holism demand---a specification that is at least inspired by Dummett's writings on logical holism. This allows a characterization of the anti-holism demand that is sufficiently precise for the task of Chapter Three: to present Dummettian criticisms of various justifications of classical logic that employ use theories of meaning. ;Chapters Four, Five and Six contain the positive part of the project---to construct a use theory that satisfies Dummett's anti-holism demand. Brandom rejects Dummett's anti-holism demand because he thinks that any use theory that satisfies the demand will possess a certain defect. And Dummett himself, in moments of self-scrutiny, expresses reasons to be pessimistic about the possibility of constructing a use theory satisfying the anti-holism demand. In Chapter Five, I argue that Brandom's and Durnmett's pessimism is unfounded. The strategy is to construct what I call a dualist theory ---a theory that generates not only the assertion conditions of sentences but also their rejection conditions. Chapter Six provides a more technical presentation of the dualist theory.