Hume is traditionally credited with inventing the 'regularity theory' of causation, according to which the causal relation between two events consists merely in the fact that events of the first kind are always followed by events of the second kind.

Hume is also traditionally credited with two other, hugely influential positions: the view that the world appears to us as a world of unconnected events, and inductive scepticism: the view that the 'problem of induction', the problem of providing a justification for inference from observed to unobserved regularities, is insoluble.

Hume on Causation is the first major work dedicated to Hume's views on causation in over fifteen years, and it argues that Hume does not subscribe to any of these three views. It places Hume's interest in causation within the context of his theory of the mind and his theory of causal reasoning, arguing that Hume's conception of causation derives from his conception of the nature of the inference from causes to effects.