Epistemic Access, Confirmation, and Idealization

Abstract

Now that we have come to have some understanding of the analysis of the concept of idealization and its main logical features we can proceed to considering the conditions under which a given idealizing counterfactual should be rationally accepted, or confirmed. This is, of course, a matter of great importance in the philosophy of science. Certainly not all such counterfactuals are equally well supported by the empirical evidence of which we are aware at some time, and it surely is not the case that all of the members of a set of competing counterfactuals are equally acceptable on the basis of our empirical evidence at some given time. But there are several rather perplexing issues concerning the acceptance and confirmation of counterfactuals that arise in the context of idealizing counterfactuals in addition to more mundane worries about confirming counterfactuals in general. Most obviously, such counterfactual conditionals are contingent truths that purport to tell us something about the behavior of idealized systems in idealized worlds to which we seem to have no empirical access. Apparently, then, the acceptance of such conditionals must either rather mysteriously depend on evidence that we can gather in the actual world, or alternatively must not depend on evidence at all.¹ Insofar as the latter option is to be avoided, we must then come to understand how empirical evidence can be relevant to theoretical claims that depend on idealizing assumptions, for where else could such evidence be gathered?