I propose a new form of epiphenomenalism, 'explanatory epiphenomenalism', the view that the identification of A's mental properties does not provide a causal explanation of A's behaviour. I arrive at this view by showing that although anomalous monism does not entail type epiphenomenalism (despite what many of Davidson's critics have suggested), it does (when coupled with some additional claims) lead to the conclusion that the identification of A's reasons does not causally explain A's behaviour. I then formalize this view and (...) show that it is an attractive position, because it captures the insights of existing forms of epiphenomenalism without their onerous metaphysical commitments. (shrink)