As my contribution on this thread was requested in another thread, here goes.
Let E be the set of relevant empirical observations. Then by the principle of multiple explanations what I think happened is a probability distribution over the set H, where H is the set of hypotheses consistent with E, weighed by complexity and my own gratuitously chosen prior probabilities. In particular the probability mass is concentrated about equally among the "standard" hypothesis and the hypothesis of a minimal MIHOP (ie MIHOP by getting some jihadi's to crash planes into buildings), but of course including every hypothesis consistent with E, such as a small probability that there is no such thing as 9/11 because I'm dreaming right now and I'll wake up again soon in the "real" world.
