The Intransitivity of Causation Revealed in Equations and Graphs

Abstract

[Introduction] We live in exciting times. By 'we' I mean philosophers studying \nthe nature of causation. The past decade or so has \nwitnessed a flurry of philosophical activity aimed at cracking \nthis nut, and, surprisingly, real progress has been made. Two \ndevelopments are especially worthy of note. First, there has been a \nresurgence of interest in the counterfactual theory of causation, given \nits best-known formulation by David Lewis. Second, there has been \nincreasing philosophical interest in the techniques of causal modeling \ndeveloped and employed within fields such as econometrics, \nepidemiology, and artificial intelligence. These two developments \nhave been largely independent and addressed to different sorts of \nproblems. Work in the counterfactual tradition has been primarily \nconcerned with issues involving "token" or "singular" causation, while \nwork in the second tradition has tended to focus on issues concerning \n"type-level" or "general" causation.