Scientific Explanation

This chapter argues that the notion of explanation relevant to the philosophy of science is that of an answer to a why-question. From this point of view it surveys most of the historically important theories of explanation. Hempel’s deductive-nomological, and inductive-statistical, models of explanation required explanations to cite laws. Familiar counterexamples to these models suggested that laws are not needed, and instead that explanations should cite causes. One theory of causal explanation, David Lewis’s, is discussed in some detail. Many philosophers now reject causal theories of explanation because they think that there are non-causal explanations; some examples are reviewed. The role of probabilities and statistics in explanation, and their relation to causation, is also discussed. Another strategy for dealing with counterexamples to Hempel’s theory leads to unificationist theories of explanation. Kitcher's unificationist theory is presented, and a new argument against unificationist theories is offered. Also discussed in some detail are Van Fraassen’s pragmatic theory, and Streven’s and Woodward’s recent theories of causal explanation.