How (not) to judge a theory of causation

Abstract

Philosophical theories of causation are commonly judged by their ability to correctly determine whether there is a causal relation present in intuitively clear example scenarios. If the theories survive this test, they are then used to answer big philosophical questions about causation. This Method of Examples is attractive because it seems to allow us to determine the quality of a theory of causation independently of answering the big philosophical questions; which is good, since it means that we can then non-circularly use the theories judged to be best to answer those questions. However, the current article argues that this virtue of the Method of Examples is an illusion. In particular, I argue that the necessary step of judging whether a proposed analysis of causation is reductive can only be taken after many of the most vexing philosophical questions about causation have already been answered. It is then shown that a rejection of the methodological supremacy of the Method of Examples leads to a more pluralistic method of judging theories of causation, a pluralism that benefits non-standard approaches like interventionism and agency theories.