For the present let it suffice to bear in mind that there is no limit to the strength of working, as distinguished from abstract, certainty, to which probable evidence may not lead us along its gently ascending paths.
W.E. Gladstone, Studies Subsidiary to the Works of Joseph Butler, 1896, p. 349.
In our article on the resurrection in this volume from Blackwell, my husband and I discuss a confusion that has dogged historical apologetics for hundreds of years: The idea that a person who has "too low" of a prior probability for the miraculous is justified in dismissing evidence for a specific miracle out of hand and accepting any other explanation instead.
David Hume made famous capital out of this confusion in his claim to have delivered an "everlasting check to all kinds of superstitious delusion." Atheists run back to the prior probabilities as a rabbit runs to its hole. Never mind the concrete evidence! If I don't already believe that God exists, why should I even listen to your concrete evidence? There must be some other explanation for it, and that's all!
One of the reasons that an understanding of probability is valuable is that it can dispel persistent errors--the intellectual equivalent of urban legends that crop up again and again.
This is such an error. And it is a case where, unfortunately, some Christian writers agree with David Hume. In the relevant section of our article we instance a quotation from the Protestant writers R.C. Sproul, Arthur Lindsley, and John Gerstner, in which they say that "only on the prior evidence that God exists is a miracle even possible" and that therefore "miracles cannot prove God."
This idea of a prior probability that is "too low" gives impetus to an extremely rigid form of classical apologetics that asserts not merely that it is a strategically good idea but that it is an absolute epistemic necessity to convince a person first that God exists, using the arguments of natural theology, before he can be reasonably expected to receive the specific evidence for a miracle such as the resurrection.
The contemporary probability theorist John Earman has done much, building on the work of 19th century thinker Charles Babbage, to demonstrate the falsity of this view. (Earman, to be clear, is not a Christian nor even, as far as I know, a theist.) Briefly, the problem with the Humean view is this: It fails to recognize that any non-zero probability, however low, can be overcome by sufficiently strong probabilistic evidence. This is simply a fact of probability theory. There is no such thing as a prior probability that is "slippery," such that one could never rationally overcome it. Therefore, it is simply false to say that, if one doesn't already believe that God exists, one should "understandably" dismiss the evidence for any miracle. For, if that evidence were sufficiently strong, it would show that, contrary to what one had previously thought, God does indeed exist!
It's important to distinguish strategy from epistemology. For sociological and psychological reasons, it may be quite reasonable in talking with a particular person to present one's arguments in stages--arguing first that God exists on the basis of natural theology and then moving on from there. The arch-probabilist among philosophers of religion, Richard Swinburne, proceeds in exactly this way. It is also probabilistically tidy to do so, because Jesus didn't walk on earth in Palestine unless life exists, and life doesn't exist unless the universe is life-permitting, so it is perfectly reasonable to present these other items as evidence for the existence of God first if one has such arguments.
But those considerations are a far cry from insisting that the evidence for a miracle such as the resurrection literally does not count as evidence or literally cannot be assimilated as evidence unless one already believes in the existence of God. That strong position is simply and flatly false and can be shown to be false by even a rudimentary understanding of how probabilistic arguments work.
In layman's terms, we can isolate the impact of an item of evidence upon an hypothesis from the prior probability of that hypothesis. We can then look at that evidential impact and at least estimate how low of a prior probability could be overcome by that evidential force. That is the strategy that Tim and I employ in our article about the resurrection.
Here is an important point: The strength of the evidence can often be seen by looking at the lengths to which the skeptic must go to explain away that evidence rather than taking seriously the hypothesis that springs to mind. Hence, rather than take seriously the possibility of the resurrection, the skeptic must hypothesize that the women went to the wrong tomb and the persecutor Paul had some inexplicable fit on the road to Damascus that just happened to make him think Jesus was talking to him and that the Christians were right and the eleven disciples all just happened to have a coordinated mass hallucination of Jesus eating, being tangible, and talking to all of them at once, repeatedly, over a forty day period and James just happened to have a similar hallucination and...You get the picture.
A major problem with saying that it is reasonable or understandable for the skeptic (who doesn't already accept the arguments of natural theology) to dismiss the evidence for a miracle is that one is in that case endorsing massively ad hoc hypothesizing on the part of the skeptic. And that is not reasonable.
One could perhaps respond that the skeptic might might simply not know about the details of the evidence for the resurrection. Perhaps he is simply ignorant. In that case he wouldn't be making such ad hoc arguments because he would just vaguely think that "scholars have shown" that the gospels are a bunch of legends made up long after the fact. But in that case, why should the only trigger, the one absolutely necessary trigger, for his finding out more be his coming to believe that God exists by being taken through a prolegomenon of natural theology? Surely we can think of other things that might make him think again--meeting a smart, respected friend or colleague who is a Christian, for example, and wondering why in the world so-and-so believes such a cockamamie tale.
It is important for Christians wanting to think clearly not to become tied to a particular, rigid order for apologetic arguments. This is important first of all because such a rigid order requirement is not defensible. We should want to know the truth, including the truth at the metalevel about how apologetics "has to" go.
Relatedly, it is important not to become tied to such a rigid order because we shouldn't be encouraging people to throw out or ignore strong evidence. We shouldn't even be encouraging them to ignore evidence until and unless they believe some umbrella hypothesis, such as theism, on independent grounds. That is poor philosophical practice. It is one thing to say that it may be helpful to show them first, on other grounds, that God exists. It is quite another to tell them that they needn't bother about the evidence for the resurrection right now, unless they already believe that God exists, because, if they do not have the proper theistic beliefs already in place, it's perfectly understandable that that evidence looks weak to them. No, it shouldn't look weak to them, and no, that isn't understandable.
There are also concerns about eternity involved here; these are matters of ultimate moment. If an unbeliever is simply ignorant of the strength of the available evidence for a miracle, the important thing is to inform him about the evidence, which speaks for itself. To say instead that it is understandable for him to dismiss any miracle claim if he doesn't already believe in the existence of God is to give him an excuse for not looking through the telescope.
We shouldn't be handing out such "get out of jail free" cards readily, because I can tell you one thing: They won't get anybody out of hell.