ABSTRACT
Traditional metaphysicians took causation to be a modal notion; they held that causes necessitate their effects. Hume attacked this idea. Invoking an empiricist theory of concepts, he claimed that if the concept of causation did involve the idea of necessary connection, the necessity would be an empirically manifest feature of particular causal sequences, and he argued that it is not. Causation, he insisted, is just constant conjunction or regularity of succession. Defenders of the traditional approach respond to Hume in a number of ways. Some (like Kant) reject Hume’s empiricism and insist that causation is an a priori concept. Others claim that Hume’s argument establishes only that causation is not an observational notion; they hold that causation is a theoretical concept. Still others insist that the causal relation is one that can be directly observed. More typical, however, are those philosophers who endorse Hume’s insistence that we provide a nonmodal account of causation. Among recent metaphysicians, some (like J.L. Mackie) continue to believe that a regularity analysis provides the requisite nonmodal account, whereas others follow David Lewis in defending a counterfactual analysis of causation.
