Causation, Physics, and the Special Sciences
James Woodward, Caltech

Abstract: What is the relationship between, on the one hand, the sorts of causal claims found in the special sciences (and in common sense) and, on the other hand, the world as described by physics? A standard picture goes like this: the fundamental laws of physics are causal laws in the sense that they can be interpreted as telling us that realizations of one set of physical factors or properties “causes” realizations of other properties. Causal claims in the special sciences are then true (to the extent that they are) in virtue of “instantiating” these underlying causal laws; as it is often put, the latter serve as “truth-makers” for the former. The picture is thus one according to which the notion of cause, as it occurs in the special sciences, is reflected or “grounded” in a fairly straightforward and transparent way in a similar notion that occurs in fundamental physics. This picture was defended by Donald Davidson decades ago and has recently been revived by Michael Strevens, who attempts to use it to it to solve a variety of familiar problems concerning causation (For example, he appeals to this idea to provide a basis for the distinction between cause and correlation.) Recently there has been a good deal of work (by Norton, Frisch, Smith, among others) on the role of causal concepts in fundamental physics. My aim in this talk will be to raise some questions about the implications of this work for the standard picture just described.

One possible view is that causal concepts play no useful role in fundamental physics and that attempts to interpret fundamental physical theories casually just create confusion. (I’ve heard this view expressed informally, although no one, to my knowledge, has said this in print.) If this is correct , the standard picture of the relationship between causal claims in the special sciences and physical law seems deeply misguided. Another, less radical view goes something like this: causal thinking is often very appropriate and useful in the sorts of contexts studied in the special sciences. However, there are contingent empirical presuppositions for the useful application of causal concepts and these are not always satisfied (although they sometimes are) in fundamental physics contexts. To the extent that they are not, causal notions do not have a fundamental or universal role to play in physics. On this view, although it certainly will be true that there always will be an in principle physical explanation for why the causal claims of the special science hold to the extent that they do, there is no reason to expect that the relationship between such claims and the underlying physics will always take the simple form portrayed in the standard picture. For example, the fundamental laws underlying such claims may not themselves always have a natural causal interpretation and the truth makers (or explanation) for such claims may have to do with complex facts about initial and boundary conditions and how these interact with fundamental laws, rather than just with the laws themselves. Interventionist ideas about causation and standard assumptions about the relationship between causation and statistical relationships that have proved useful and illuminating in the special sciences (e.g. the causal Markov condition) may be difficult to apply in fundamental physics contexts . This second view leads to an interesting set of questions that have been relatively unexplored by philosophers interested in causation—just what are the empirical presuppositions for the illuminating use of causal talk and just what are the physical facts from which it follows that these presuppositions are, often enough, satisfied in the sorts of contexts explored in the special sciences?