David Hume, and many of those who follow his thinking today, want to pronounce miracles impossible, or at least so unlikely that no one could ever rationally conclude a miracle has happened. One such person, a friend of mine has, in many recent discussions we have had, said resurrections are extremely unlikely, so we must conclude, given the normal course of events is that dead men stay dead, resurrections do not happen.
He does not seem to realize this way of thinking rules out anything unlikely from ever being a possible explanation for something. This illustration, based on some discussions I had with Tim McGrew, seemed to help his understanding of this principle.
Let us say you are playing darts and a friend claims they can get a bull’s-eye on the first dart thrown. This is highly unlikely, and you have never seen anyone do that before, so you claim, “That’s not going to happen. The chances of getting a bull’s-eye on the first throw are so low that we might as well declare it impossible.” With that you walk into the next room to grab some more bean dip. Then you hear from the other room, “He got it!” In your mind, as Hume might expect, you think, “No he didn’t, that is not possible.” You know the odds of hitting the bull’s-eye are extremely low, and the probability of hitting anywhere else is high. So the conclusion that he did not actually get the bull’s-eye is reasonable.
However, upon walking into the room you see a dart stuck in the bull’s-eye. Can you still claim it is not possible? Maybe, but the story is different now. Now we are not discussing the chances of hitting the bull’s-eye based on the background probability of hitting bull’s-eyes from the start; instead we have to contend with the fact that there is a dart in the bull’s-eye. The evidence is there. No matter how unlikely it is that one should find a dart in the bull’s-eye, we cannot deny what is right in front of our face. This means the background probability that is so important to Hume is less important now that we have evidence to suggest the unlikely thing did in fact happen. We cannot rule out that hitting the bull’s-eye is impossible on the first shot, because we now seem to have (at least potential) evidence to the contrary. Now we must go about explaining the evidence.
At this point we have two main options to consider: 1) Your friend actually hit the bull’s-eye, or 2) someone walked up to the dartboard and stuck it in there. Since the initial probability of getting a bull’s-eye is so low, as a Humean, you should conclude that your friend is tricking you rather than that he actually hit the bull’s-eye, because you should always conclude what is more likely. However, beyond the background data, we must also contend with other pieces of evidence. Your friend is not the type to lie (L). Two other friends in the room also claim they witnessed it (W). Lastly, and most importantly, another friend who had bet $20 against this event happening is paying up (B). So we must take these pieces of evidence one by one and compare them between the two theories. What are the chances that we would see evidence L given 2 versus the chances of seeing L given 1? The chances they might be playing a trick on you are a bit higher than normal given that this is a party, so 1 and 2 seem about equal given L, though your friend is not usually the type to pull pranks. However, W seems more likely given 1 than if 2 had happened. You may have one person in on a conspiracy, but that two are in agreement that they witnessed it and are not cracking a smile is evidence they are telling the truth. Finally, B seems very unlikely given 2 but likely if 1 is the case. That someone would give away $20 when they in fact won the bet is very unlikely. So even though the prior probability of hitting a bull’s-eye is low, and does enter into the calculation, favoring 2 over 1, when we bring in evidences L, W, and B, it seems most likely to conclude that 1 has happened. Though bull’s-eyes are rare, the evidence that it did happen seems to overcome that background probability.
This goes for resurrection as well. Yes, generally dead people stay dead, but simply because resurrections are improbable given the background information, it does not mean they cannot happen. So rather than reject it outright because it is unlikely, we must contend with the fact that some claim to be eyewitnesses of just such an event. There seems to be a certain amount of intellectual dishonesty in those who insist resurrections cannot happen, based on the fact that it has never happened, when faced with evidence that one has, in fact, happened. We have to go back and contend with the evidence and determine what the most likely explanation of the historical evidences that go along with the claim is. Just as with the dart example, we cannot automatically rule out resurrection. We must look at the evidence surrounding the claim made in the New Testament. The fact that Jesus’ own brother, James, who was an unbeliever, became one of the leaders of the church, and the fact that Paul, who originally persecuted the church, ended up writing half the New Testament, help show the validity of the resurrection claim over and above the background that dead people tend to stay dead. For more on this topic and a though discussion of the evidences and how they show resurrection to be the best explanation of the evidence, I refer the reader to Tim and Lydia McGrew’s article in the Blackwell Companion to Natural Theology.