Sullivan links a Tony Blair speech in which Dubya’s biggest fan points out that if, on September 10th, someone had proposed pre-emptive action against al-Qaeda, many of us would have reacted more or less as sceptics on invading Iraq have. We would have said: well, is there enough evidence to justify thinking this is a serious threat? Enough to risk American lives, and maybe even the destabilization of the region on? (Put aside for the moment the fact that the worries about destabilization make a lot more sense with respect to Iraq, and that it’s not clear that intervention earlier on would have prevented a large-scale terrorist attack.) Andrew apparently thinks this is a brutal argument against the war skeptics. Well, maybe we would have said the same thing then. But — and I know this is a hard one to swallow — we probably would’ve been right. Sullivan’s argument makes roughly as much sense as pointing out that we wouldn’t have thought it was rational to put a lot of money on this specific lottery number before the fact, even though it turned out to be a million dollar winner. Be that as it may, it hardly establishes that we should then feel chastened and sink our money into the next lottery number someone suggests betting on.

Sometimes we don’t have enough information to see that there really is a case for intervention in a given instance. If we knew everything, we would’ve intervened, just as, if we knew everything about the physics of a particular roulette spin, we’d bet in advance on the winning number. But given that we don’t know everything, that sort of analogy to hindsight calculation is useless at best and dangerously misleading at worst, especially when invading — like betting — in case of uncertainty may be very costly indeed.