Why P rather than Q? The curiosities of fact and foil

Conclusion

Of the various proposals regarding the fact/foil relationship discussed above my own most closely resembles that of Lipton: a largely similar type of causal history is indeed required for true P and false Q to generate a contrastive why question. What I add is the requirement that P and Q be culminating events of a single type of natural causal process.

While my immediate aim has been to describe the nature of the fact/foil relationship, the greater significance of this project lies in an improved understanding of the nature of contrastive explanation. Temple proposes in his (1988) article that to explain why P rather than Q is ultimately just to explain why P and not Q — but on this account (in the absence of an account of contrast itself) it ought to be possible to explain why Bush won the election rather than life never having arisen in the universe by simply conjoining the explanation of Bush's victory with the explanation of the origin of life. But insofar as one may not sensibly say that Bush won the election rather than life never having originated it ought not to be possible to explain this later ‘fact’: our account of explanation should prohibit explanations where there are, in principle, none to be found. Lipton's own theory of contrastive explanation is that one explains why P rather than Q by citing some cause of P that is not matched by a corresponding cause of the same type for Q; but again, this by itself would allow explanations of insensible contrastive phenomena (a cause of Bush's victory would certainly not be matched by a corresponding cause of the same type of life's not coming into existence — should we then allow some cause of the Bush victory to explain why Bush won rather than life never having arisen?) That Garfinkel, Lipton, Sober and Temple make serious efforts to understand the fact/foil relationship indicates that they each accept the premise of this paper: our account of contrastive explanation will remain incomplete until we understand the nature of contrastive phenomena.