Abstract

The traditional model and the contextual unanimity model are two probabilistic accounts of general causation subject to many well-known problems; e.g. cases of epiphenomena, causes raising their own probability, effects raising the probability of the cause, et cetera. After reviewing these problems and raising a new problem for the two models, I suggest the beginnings of an alternative probabilistic account. My suggestion avoids the problems encountered by earlier models, in large part, by an appeal to singular causation.