The Revision of Logical Laws

Silver Professor and Professor of Philosophy Can there be rational reasons for revising our most fundamental logical rules (for instance, the rule of inferring from A to A or B; or the rule of inferring from A or B and not A to B)? I think there can. Nonetheless, the supposition that there can be rational reasons for such revisions gives rise to a wide variety of puzzles, which have led some philosophers to think that rational revision of logic is impossible. I'll start by briefly mentioning two of them, without going into any detail. (The main body of the paper will be an argument that despite the puzzles, rational revision of logic is possible, and on this I will provide some details.) A first puzzle about rational revision of logic is this: any serious evaluation of anything requires logical reasoning; so in evaluating a proposed revision of logic, you need to use logic. But if you use a logic in the evaluation process, won't it automatically rule against all competing logics? One strategy for responding to this would be to argue that a principle of logic can rationally be revised by an evaluation that uses only other principles of logic. (If that's the best we could do, then if there are parts of logic that must be used in the evaluation of any principle of logic, those parts of logic must be immune from revision.) A better idea, I think, is to try to give an account of reasoning which would allow a logical principle to appear in an evaluation that undermined itself. But to say that is to say very little: an account of just how a logical principle can be used in an evaluation that undermines itself is needed. Still, there seems to at least be room for such an account: it would presumably work by appealing to methodological principles that go beyond the logic itself.