How Probabilistic Causation Can Account for the Use of Mechanistic Evidence

Abstract

In a recent article in this journal, Federica Russo and Jon Williamson argue that an analysis of causality in terms of probabilistic relationships does not do justice to the use of mechanistic evidence to support causal claims. I will present Ronald Giere’s theory of probabilistic causation, and show that it can account for the use of mechanistic evidence (both in the health sciences—on which Russo and Williamson focus—and elsewhere). I also review some other probabilistic theories of causation (of Suppes, Eells, and Humphreys) and show that they cannot account for the use of mechanistic evidence. I argue that these theories are also inferior to Giere’s theory in other respects.

Acknowledgements

I thank Leen De Vreese, Bert Leuridan, and three anonymous referees of this journal for their comments on earlier drafts of this article. The research for this article was supported by the Research Fund—Flanders through project no. G.0651.07.

Notes

[1] Leuridan, Weber, and Van Dyck (2008) draw a distinction between manipulative policy and selective policy. Manipulative policy requires causation (in the average effect sense), selective policy uses a specific type of non‐causal information.

[2] Suppes distinguishes a second sense of spurious cause, but that type is not important for my purposes.

[3] Giere’s model is not popular in the traditional philosophy of causation, but the situation is clearly different when we look at scientists dealing with methodological issues in their own discipline. For instance, Morgan and Winship (2007, ch. 2) use a ‘potential outcome model’ of causation which is very similar to Giere’s theory; this model is very common in the social methodological literature on causation.

[4] I do not discuss interventionist accounts (such as Pearl 2000 and Woodward 2003) here because that would bring us too far from the original aim of this paper. Pearl and Woodward do not require context unanimity, and their definitions have a counterfactual nature. So they share the important advantages of Giere’s account discussed here. Giere avoids the concept of intervention which Pearl and Woodward use in their definition of causation. This seems to be an advantage, though Pearl and Woodward insist on a non‐anthropocentric interpretation of the concept of intervention. A detailed comparison may reveal other relative advantages and disadvantages.