Explanation and Metaphysics

Is the nature of explanation a metaphysical issue? Or has it more to do with psychology and pragmatics? To put things in a different way: what are primary relata in an explanation? What sorts of thing explain what other sorts of thing? David Lewis identifies two senses of ‘explanation’ (Lewis 1986, 217–218). In the first sense, an explanation is an act of explaining. I shall call this the subjectivist sense, since its existence depends on some subject doing the explaining. Hence it is people who, in this sense, explain things. In the second of his two senses, Lewis says, quoting Sylvain Bromberger, that one may properly ask of an explanation “Does anyone know it? Who thought of it first? Is it very complicated?” (Lewis 1986, 218; Bromberger 1965). In this second sense, no subject is needed, the explanation can remain unknown, perhaps for ever. So I call this the objectivist sense. Corresponding to the two senses, there are two tendencies, each emphasizing the primacy of one of the two senses. Corresponding to the subjectivist sense is an anti-metaphysical tendency that will typically deny that there are real explanations in nature. It is people who explain things; they do so by arranging information or beliefs in certain ways, so that certain salient connections may be seen. The saliency of such connections, and so whether an explanation is a good one or not, will depend on the interests of the recipient of the explanation. It cannot be denied that we do talk of one fact explaining another, as when we say that the decaying of the O-rings explained the explosion of the Challenger spacecraft. But this is just a projection of our practice of explanation onto the things we refer to in our explanations. Which is not to say that explanation is independent of the way things are; rather it is to say that it does not constitute part of the way things are. In this regard explanations are akin to secondary qualities, on a projectivist view of the latter.