The historian is never in a position to do what Pythagoras did: not content with drawing more and more right-angled triangles and demonstrating that the square on the hypotenuse always does in fact equal the sum of the squares on the other two sides, he constructed a theorem to prove that this must always be the case. With history it is not like that. Almost nothing is ever ruled out absolutely; history, after all, is mostly the study of the unusual and unrepeatable. What we are after is high probability; and this is to be attained by examining all the possibilities, all the suggestions, and asking how well they explain the phenomena. It is always possible that in discussing the resurrection someone will come up with the sceptical critic’s dream: an explanation which provides a sufficient condition for the rise of early Christian faith but which, by fitting into post-Enlightenment epistemological and ontological categories, or even simply mainstream pagan ones, causes no fluttering in the critical dovecotes. It is worthy of note that, despite the somewhat desperate attempts of many scholars over the last two hundred years (not to mention critics since at least Celsus), no such explanation has been found. The early Christians did not invent the empty tomb and the “meetings” or “sightings” of the risen Jesus in order to explain a faith they already had. They developed that faith because of the occurrence, and convergence, of these two phenomena. Nobody was expecting this kind of thing; no kind of conversion-experience would have generated such ideas; nobody would have invented it, no matter how guilty (or how forgiven) they felt, no matter how many hours they pored over the scriptures. To suggest otherwise is to stop doing history and enter into a fantasy world of our own, a new cognitive dissonance in which the relentless modernist, desperately worried that the post-Enlightenment worldview seems in imminent danger of collapse, devises strategies for shoring it up nevertheless. In terms of the kind of proof which historians normally accept, the case we have presented, that the tomb-plus-appearances combination is what generated early Christian belief, is as watertight as one is likely to find.
N.T. Wright, The Resurrection of the Son of God, 706-707.