One of the central problem areas of metaphysics. Causation is the relation between two events that holds when, given that one occurs, it produces, or brings forth, or determines, or necessitates the second; equally we say that once the first has happened the second must happen or that the second follows on from the first. However, it is not clear that only events are related by causation: Kant cites the example of a cannonball stationary on a cushion, but causing the cushion to be the shape that it is, and this suggests that states of affairs or objects or facts may also be causally related. In any case the central problem is to understand the element of necessitation or determining of the future. Events, Hume thought, are in themselves ‘loose and separate’; how then are we to conceive of the power that one has to constrain others? The relationship seems not to be perceptible, for all that perception gives us (Hume argued) is knowledge of the patterns that events do actually fall into, rather than any acquaintance with the connections determining the patterns. It is clear that our conception of everyday objects is largely determined by their causal powers, and all our action is based on the belief that these causal powers are stable and reliable. But although scientific investigation can give us wider and deeper reliable patterns, it seems incapable of bringing us any nearer to the ‘must’ of causal necessitation. Particular examples of puzzles with causation, quite apart from the general problem of forming any conception of what it is, include: how are we to understand the causal interaction between mind and body? How can the present, which exists, owe its existence to a past that no longer exists? How is the stability of the causal order to be understood? Is backward causation possible? Is causation a concept needed in science, or dispensable? See also constant conjunction.

http://philsci-archive.pitt.edu/view/subjects/causation.html A bibliography of online philosophical papers on causation