1 Perspectives on causes and dispositions. Review-Essay on Toby Handfield (ed.), Dispositions and Causes, Oxford University Press, 2009. Metascience 19 (3) (2010), 403-407. DOI : 10.1007/s11016-010-9392-5 Max Kistler Université Paris 1 Panthéon-Sorbonne, CNRS, ENS UMR 8590 IHPST - Institut d'Histoire et de Philosophie des Sciences et des Techniques The concepts of cause and disposition are closely connected. One cannot say what a disposition is without speaking of what causes its manifestations, and problems encountered in analysing dispositions with counterfactuals (relating triggering conditions to manifestations) resemble problems arising in the counterfactual analysis of causation. Handfield’s introductory chapter explores some relevant connections between ideas developed in philosophy of science on causation and ideas developed in metaphysics and philosophy of language on causal power and disposition. Masks or antidotes that trouble the analysis of dispositions correspond to, e.g., pre-empted back-up causes that trouble the analysis of causation. However, the connection is not straightforward. Consider Billy and Suzy throwing rocks at a bottle. Suzy’s rock smashes the bottle, thereby pre-empting Billy’s rock, which gets there a little later, from smashing the bottle. Handfield shows that one can describe the situation within the conceptual framework of the analysis of dispositions, using the notion of a mask, but only in a quite unnatural way: Billy’s throw masks the disposition of Suzy’s not throwing her rock to leave the bottle intact. The papers can be read independently of each other, although some make explicit reference to other papers in the volume. However, a reader determined to make her way through the whole volume who is not yet an expert in this domain, might prefer to leave the first three papers, by McKitrick, Eagle, and Barker, which are quite technical, to the end. McKitrick thoroughly explores the perspective of reducing either dispositions to causes or causes to dispositions. She finds none of the attempts to carry out the former reduction successful, the main reason being that “causes are active [whereas] dispositions are potentially latent” (63). However, she finds no conclusive reason against the hypothesis that causes might be metaphysically reducible to dispositions. It may be impossible to describe two worlds that agree with respect to dispositions but differ with respect to causes. Yet, in the absence of any explicit conceptual analysis of causation in terms of dispositions, such a reduction remains an open possibility. Eagle argues that there is no valid inference from causal structuralism to dispositional essentialism. Causal structuralism is the thesis that “at least some properties […] have a causal profile that is essential to them” (68/9). Part of dispositional essentialism is “dispositional actualism” according to which 1) the laws of nature are necessary but 2) this necessity is grounded in the nature of actual properties (73). Eagle shows that the truth of counterfactuals cannot be grounded on the intrinsic nature of actual properties because the truth value of a counterfactual depends in general on extrinsic facts: on what happens in the surroundings of the situation described by the antecedent. Causal structuralism entails that each natural property supports a characteristic stimulus-manifestation conditional; yet, such conditionals are not necessarily true. In the relevant possible worlds, circumstances may contain interfering factors that make the consequent false. Therefore, laws are not necessary. The causal role characterizing a property may be expressed by what he calls a “habitual”, a generic sentence expressing a regularity tolerant of exceptions. A weak point in Eagle’s otherwise very careful argument is his hasty rejection of the possibility to consider laws as bearing on dispositions, not actual behaviour. This does not, as he says (82), prevent laws from determining the observable behaviour of objects; it only makes this determination indirect.