We saw in Chapter 4 that one component of the idea of causation, for Hume, is the idea of necessary connection; and we also saw that Hume holds that the source of that idea is a ‘transition’ in the mind: the transition the mind makes when, on observing one event, we infer or come to believe that some other event will follow. The central interpretative question, when it comes to attributing a view to Hume about the semantics of ‘cause’ and the metaphysics of causation, is this: how do we put these two claims together in such a way that they can be reconciled both with Hume’s theory of meaning and with what he actually says about the meaning of ‘cause’ and the nature of causation? Unfortunately, Hume’s theory of meaning itself is an area of interpretative controversy, as is what he actually says about the nature of causation. (For example, he appears to say both that necessity is all in the mind and that there are secret powers in objects whose nature we cannot penetrate.)

In this chapter, I begin to work through the major interpretative dispute surrounding Hume’s views on causation – the dispute about his semantic and metaphysical views – by focussing on what I am calling the ‘traditional interpretation’. As we shall see, ‘the traditional interpretation’ turns out to encompass a variety of views; but they share a common core, and that is why I have lumped them together under a single interpretative heading. The core consists of a positive claim and a negative claim. The positive claim is that Hume holds that causation in the objects is a matter of temporal priority, contiguity and constant conjunction: our causal talk and thought cannot succeed in describing or referring to any more in the world than these features. The negative claim is that it is illegitimate or incoherent to apply the idea of necessary connection to external events.