5 - THE DIRECTION OF CAUSAL EXPLANATION

Summary

While causes explain their effects, effects do not explain their causes – this fact is constitutive of the causal relation. If the striking of the match caused it to light, then the match's lighting cannot have been a cause of its striking. If we can explain why the match lit by mentioning the striking, we cannot also explain why the striking took place by referring to the lighting. So much is obvious, but analysts of causation have found the obvious hard to digest.

Take our causal claim: ‘the match's being struck caused the house to burn’. What makes this assertion true? An obvious starting place is the thought that this causal statement is true because a struck match was either necessary or sufficient, or both necessary and sufficient, for the house to burn. To express this in terms of subjunctive conditionals: ‘the house burnt because the match was struck’ is true if, and only if, either ‘if the match had not been struck then the house would not have burnt’ is true, or ‘the house would have burnt given that the match was struck’ is true, or both are true. But this simple suggestion quickly comes to grief.

If there is a law of nature which ensures that the striking of a match is necessary and sufficient for a fire then it will equally be a law of nature that a fire is necessary and sufficient for a striking. If a striking is just sufficient for a fire, then a fire will be necessary for a striking and the absence of a fire will be sufficient for the absence of a striking.