The Modern History of Scientific Explanation

Abstract

To be a philosopher of science means, among other things, to have an account of what scientific explanation is, or, at the very least, to have a response to various accounts of scientific explanation on offer from other philosophies of science while earnestly working toward what one hopes will be one’s own, original account. One presumption clearly and often lying behind such work is that science provides two kinds of knowledge. There is propositional knowledge, “knowledge that” or “knowledge what,” and there is some other kind of knowledge, something beyond propositional knowledge, usually called “knowing why.” We can know that the moon will have such a phase at this or that time, that home sales will always slump following a rise in interest rates, or that probably no two snowflakes are the same shape, without knowing why the moon will have that phase, home sales will fall as interest rates rise, or no two snowflakes (probably) have the same shape. But science, so the common contemporary presumption continues, fills in the missing knowledge — it tells us why. How science does this, when (if ever) it can’t, and what the nature of this sort of knowledge is are precisely the issues that separate theorists of explanation. There are, of course, deflationary views of explanation, which reduce explanation to other properties or eliminate explanation altogether (e.g., van Fraassen’s pragmatic account of explanation), but these are a decided minority. The vast majority of the work on scientific explanation takes itself to be addressing a certain, distinct, kind of knowledge. This is, moreover, a familiar and introductory point made in philosophical discussions of explanation. I rehearse it here because it has a role to play later, in my discussion of Carl Hempel and Paul Oppenheim’s 1948 article, “Studies in the Logic of Explanation” (hereafter, ‘SLE’).¹