In this paper, I defend the following theses: (1) Arguments for the truth of scientific theories and appeals to supernatural agency are generically similarboth are legitimate cases of argument to the best explanation. (2) The two are nevertheless specifically differentthere are significant differences in method by and the extent to which each can improve early, reasonably good, explanations. (3) The prospect of conflict between scientists and theologians is diminished by the facts that the situations in which appeal to supernatural agency is theologically most plausible are situations which are of least interest to scientists, and the situations of most interest to scientists are situations in which appeals to supernatural agency are theologically least compelling.
In the first section of the paper, I give a preliminary account of "the scientific method" as argument to the best explanation of observed phenomena. Scientists use such reasoning in a variety of situations. Of particular relevance to this topic is their use to explain diversity and complexity by appeal to history (a prior homogeneous state plus a law of development) and to explain some unusual particular events by appeal to an unusual interaction of distinct regularities.
In the second section, I focus on two questions. The first is whether supernatural agency is ever a good explanation and, if so, under what circumstances it would be the best explanation. Such appeal is always possible, since God can do everything. I argue, however, that supernatural agency is a good explanation only to the extent that (1) there is a precedent for supernatural agency in such cases and (2) there are no plausible natural accounts of the explanandum. Circularity is avoided by seeking the precedent in cases where supernatural agency is certified by revelation. The second question is whether appeal to supernatural agency is ipso facto non-scientific and, if so, what the criteria of scientific explanation that make it so would be. I argue that such appeals are inherently non-scientific because the scientific method includes a procedure for identifying and correcting appeals to the wrong natural agency which, though fallible, is constitutive of the method itself. This corrective procedure will not protect against appeals to supernatural agency, which makes such appeals non-scientific, though not illicit.
In the third section, I show that the conclusions reached in the second section have implications for two important kinds of appeal to supernatural agency. The first kind is the willingness to regard particular historical events as miracles. Here there is no principled reason to be suspicious of appeal to supernatural agency, though each case must be taken on its merits. The second kind is appeal to supernatural agency as a component in an explanation of the origins of something, as is done by some Christian critics of various evolutionary theories. Such appeals, by contrast, give much reason for caution. The fact that theories of origins are important components of scientific work, whereas the explanation of extraordinary cures is not, is a consideration which logically diminishes the danger of science-theology clashes over what is the best explanation of some explanandum of mutual interest.
Copyright © Kenneth W. Kemp