Induction and Modality

If we are to have beliefs about the unobserved, what ought we to believe? Should we expect the next emerald we examine to be green, or grue? We inherit a predicate base which favors green rather than grue. But once we become aware of the existence of other predicate bases, we face the question: Why keep ours? Is there any rational basis for preferring it? It is quite clear how to change predicate bases, if we want to. We simply, for example, use grue and bleen for projection instead of green and blue, and leave everything else unchanged. There is some discussion in the literature about there being difficulties integrating projections of grue and bleen into the rest of our inductive practices. But this seems wrong. There is no difficulty in changing predicate bases. We simply associate grue with whatever kinds we had hitherto associated green with-emeralds, summer leaves, a certain range of wavelengths, etc. And so for bleen and blue.' The real issue is: What is the ground of our association of kinds? Is it the case, as Hume says, that custom and habit are the only ground? If someone says that being entails projecting green rather than grue, the question then simply becomes whether to exchange these predicates-green, blue, and reasonable-for three others: grue, bleen, and reasonable (it being by definition greasonable to project grue rather than green). And to say that considerations of reasonableness or rationality could not apply, or could apply only circularly, to this question would be to endorse Hume's claim: one is left with one's habits, or society's, with nothing more to be said. The question is: Can anything more satisfying be said?