Causal powers : why Humeans can't even be instrumentalists.

Hume urged that there is no difference between the obtaining of a power and its exercise; others, that there is no difference between its exercise and the result that occurs. This chapter reinforces the reasons, based in the success of the analytic method in a variety of sciences and often in daily life, for taking exercisings to be real and separate from both the obtaining (along per-haps with triggering if needed) of a power and the overall result. If exercisings are real, the chapter urges, then the Mill-Ramsey-Lewis view of laws is in trouble, at least if laws are going to account for much of what happens, since ex-ercisings of powers surely should not be admissible into any ontology the view allows. One might hope to rescue the Mill-Ramsey-Lewis account by retaining the demand that laws and theories cover as much as possible as simply as possible but give up the requirement that they be true. In that case laws could involve exercisings but not as true features of the world, rather as part of an instrument for predicting facts that the view lets in to its ontology. This chapter adumbrates this proposal and argues that it will not work. If laws are to cover much of what we think they do, we will have to have exercisings in our ontology, and, the chapter argues (contrary to a proposal floated by Richard Corry), we can’t have exercisings without powers. So we had best accept that powers have full citizenship in the world that science presents to us