l. Introduction One of the jobs of philosophers of the special sciences is to connect the local concerns of particular disciplines with those of philosophy in general. The two-way complexities of this task are well-illustrated by the case of causation. On the one hand—from the outside, as it were philosophers interested in general issues about causation are prone to turn to the special sciences for real-life examples of the use of causal notions. On the other hand, from the inside, the special disciplines themselves throw up philosophical puzzles in which the notion of causation plays a role. When does correlation indicate causation, for example? Physics and economics both generate hard cases of this kind. In principle, then, a philosopher of a discipline such as physics or economics occupies a rather exposed position, liable to be called as expert witness about causation from inside and out—by people who know more about the special discipline, and by people who know more about relevant parts of philosophy at large. This dual role may seem a trifle self-contradictory. After all, to address a causal puzzle within physics (say) a philosopher of physics needs to hold fixed a general philosophical account of causation; and therefore hasn’t the luxury of remaining uncommitted about the philosophy of causation in general, the kind of cautious player who awaits the verdict of physics on such matters. But progress is surely possible by a kind of reflective equilibrium, an informed interplay between the two kinds of constraints, philosophical and scientific. The difficult task, the task which falls especially to the philosopher of the special disciplines, is to open and keep open the dialogue. This paper is an attempt at that task. In particular, I want to recommend a particular way of thinking about causation—a kind of pragmatism—to philosophers of sciences such as physics and economics. The term pragmatism means several things in contemporary philosophy, and so I’ll say more in a moment about the sense I have in mind...