Continuing from part 1, we’ll look at Craig Keener’s analysis of Scottish philosopher David Hume’s views on the testimony required to make a miracle claim credible.
Keener, in Miracles: The Credibility of the New Testament Accounts
Further, some of Hume’s criteria for witnesses’ acceptability are too vague to quantifiably support his case: he insists that witnesses be highly reputable, with much to lose by lying. One may assent to these demands in principle, but Hume appears to implement them in a tendentious way. How highly reputable is highly reputable? How much to lose is too much to lose? If one adopts his criteria for witnesses to the maximal possible extent, one might choose to rule out any historical testimony to any event.
As I shall observe, Hume does in fact rule out highly reputable witnesses with much to lose, as defined by normal standards used in court, suggesting that he applies these criteria tendentiously. Moreover, Hume requires witnesses to be of “unquestioned good sense,” but this standard proves impossible to meet, since Hume appears to question the good sense of anyone who claims to have witnessed miracles. By contrast, if we employ such criteria in the ordinary sense of their everyday usage, we end up with plenty of witnesses that we might consider reputable and sensible, but whom he dismisses as unsatisfactory. If he simply will not deem anyone’s testimony satisfactory, it seems somewhat disingenuous to expect his critics to go to the trouble of evaluating witnesses before he informs them of this caveat.
I, myself, can understand Hume demanding that witnesses be reputable with good sense. If there were only a handful of miracle reports available to us, or even 100 or 1000 such reports, and each time we investigated these reports we found that the individuals involved were gullible fools who would believe anything, then I think Hume would have a case against miracles. But that simply isn’t the situation. As Keener documents, there are literally millions of miracle reports, and a great number of them are reported by people who are not gullible fools.
Keener again explains that Hume and his followers simply argue in a circle:
Again, he seems to employ an a priori definition to exclude the need for examination: defining a miracle as the sort of event “that has never been observed,” he simply dismisses or ignores the perspective of all those who claim to have seen, or believe the claims of others to have seen, such events. Analogously, as noted above, he excludes from being a miracle anything that can be observed to occur in the ordinary course of nature, yet he excludes the possibility of anything that does not occur in the ordinary course of nature. This sort of reasoning simply restates his presupposition rather than offers an argument. This mere reformulation of his own presuppositions is not, as one scientist and theologian points out, the open-minded posture normally appreciated in scientific endeavor.
How does Hume so easily dismiss the eyewitnesses of miracles?
Hume must assume the error or lack of integrity of many eyewitnesses to maintain his theory, yet he lacks grounds independent of his theory to accuse eyewitnesses of deception. (This concern is important in view of the significant number of testimonies collected later in this book and elsewhere.) Hume essentially dismisses all witnesses as “fools or liars,” as one scholar puts it. Yet this suspicion of witnesses is arbitrary, dependent entirely on Hume’s theory and increasingly implausible as the number of normally reliable witnesses increases. His warning that people are prone to credulity and deception does not apply equally to all individuals, so one cannot dismiss all claims without evaluating them on a case-by-case basis. Using this standard, and a priori suspicion of any antecedently improbable information, would undermine ordinary communication.
In fact, Hume’s criteria for witnesses would effectively rule out almost all the testimony we have about our past. More on this in part 3.