The Doctrine of Creation (part 18)

December 27, 2008     Time: 00:20:43


Miracles continued.

You remember last time we were talking about Hume’s argument against miracles. You will remember Hume’s argument was that because miracles are intrinsically improbable that therefore no amount of evidence could ever go to establish a miracle. This is the same argument that Bart Ehrman uses in his books and courses, and that he used in the debate. You cannot prove a miracle like the resurrection because miracles are so inherently improbable.

What I explained is that this commits a fundamental logical fallacy with regard to probability theory. If we let R be the resurrection, the probability calculus requires that the probability of R on the specific evidence and the background information be equal to a certain ratio. Here R will be the hypothesis of Jesus’ resurrection. E will be the evidence of the empty tomb, the postmortem appearances, the origin of the disciples’ belief. That is the historical evidence. And B is the background information we have about the world in general. You will remember I explained last week that this is equal to a very complicated ratio which consists of first the intrinsic probability of the resurrection on the background information – how probable is the resurrection of Jesus just given our background knowledge of the world. This is then multiplied by the probability of the evidence given the resurrection and our background information. That is to say, how probable does the resurrection hypothesis make the evidence? This is the explanatory power of the resurrection hypothesis. This is its intrinsic probability; this is its explanatory power. Then I explained you just move everything that is above the line down below the line here. So you just reproduce the probability of R on B times the probability of E on R and B. Then you will add to that the probability of the hypothesis that the resurrection did not occur. That is to say, the probability of no resurrection given our background information times the probability of the evidence given no resurrection and the background information. The dash stands for “not.” This is going to be the probability of no resurrection given the background information and the probability of the evidence given no resurrection. That is, how well can you explain the empty tomb, the postmortem appearances, the origin of the disciples’ faith if there were no resurrection. This will consist of all of the alternative naturalistic explanations of the resurrection – the wrong tomb theory, the conspiracy theory, the hallucination theory, and so forth.

Somebody like Bart Ehrman makes fun of this because, I think, he doesn’t understand it. But this is basic probability theory. This is how you calculate the probability of the resurrection given the evidence and the background information. The fallacy of Hume and the fallacy of Ehrman is in considering only this probability alone. Because the resurrection is, they think, highly improbable with respect to our general knowledge of the world like the laws of nature, therefore they infer that the resurrection is highly improbable. That is just fallacious. The probability of the resurrection given the evidence and the background information isn’t equal to the probability of the resurrection just given the background information. That is obvious. You’ve got to add in the evidence for it to get the probability of the resurrection. What will really be key will be this factor down here. That is the naturalistic alternatives to the resurrection. If they are sufficiently improbable (say greatly less than 50%) then that can outbalance any intrinsic improbability in the resurrection itself. That is Hume’s fundamental failure in his argument, and also the fundamental failure of Bart Ehrman’s argument.

But I want to go further than that. What about this probability here – the intrinsic probability of the resurrection on the background information?[1] Why think that that is intrinsically improbable? Why think that the resurrection of Jesus is improbable relative to the background information? Here I think it is very important that we specify exactly what the resurrection hypothesis is. The resurrection hypothesis is not the hypothesis that Jesus rose naturally from the dead. That hypothesis is fantastically improbable. That all of the cells in Jesus’ dead body would spontaneously come back to life again is so biologically and medically improbable that I think we can say that is really naturally impossible. It is naturally impossible that Jesus would come back alive from the dead. That is why we consider the resurrection a miracle – right? Because it is naturally impossible. So the resurrection hypothesis is not that Jesus rose naturally from the dead which is highly improbable. The resurrection hypothesis is that God raised Jesus from the dead. I don’t see any reason to think that that is highly improbable. Why should we think that it is highly improbable that God raised Jesus from the dead? It seems to me that unless you could demonstrate either that God’s existence is improbable or that God’s wanting to raise Jesus from the dead is highly improbable there is no way you could know that it is improbable that God raised Jesus from the dead. So the person who wants to defend the intrinsic improbability of Jesus’ resurrection would have to show either that it is highly improbable that God exists or that it is highly improbable that God would want to raise Jesus. And I can’t imagine how he would have that kind of knowledge to show that. Unless he can show that I just don’t think there are any grounds for thinking that the probability of R on B is very low. I think we have to at least be agnostic about it. I don’t think we can say that it is improbable.

Someone might say, “But miracles are so infrequent that it is improbable in that sense.” Miracles are rare. They hardly ever happen. So it is improbable in the sense that it is infrequent. But as John Earman points out in his book Hume’s Abject Failure you can’t use frequency as a theory of probability. Earman gives the example of scientists trying to find an event of proton decay in their nuclear research facilities. Scientists, as Earman points out, are investing thousands of man-hours and millions of dollars in trying to detect a single event of a proton decaying into its sub-constituents. Yet this has never been observed. No one has ever observed an event of proton decay. So if you have a frequency model of probability that would mean the probability of that event is 0. But as John Earman points out scientists are not investing millions of dollars and thousands of research hours in trying to detect an event that has 0 probability of occurring. So you can’t just use frequency as a theory of probability – to say something is improbable because it is infrequent.

This is especially evident when you are dealing with the decisions of a free agent. A free agent can choose something precisely because it is rare or unusual. Suppose I go into a car lot to buy an automobile and out of the hundred automobiles in the car lot 99 of them are black and there was 1 red one. In calculating the probability that the car that I will drive out of the lot with is black, you have to take into account more than just 99 out of a 100 are black. In fact, if you know anything about my tastes, you would say it is highly probable that Bill will pick the red car rather than the black one. I will pick it precisely because of its rarity and unusualness. Similarly with regard to God, isn’t it highly probable that in order to vindicate the radical claims of his Son for which he was crucified, that God would pick an event of extraordinary improbability? An event which is so singular that it has never happened before in the universe and will never happen again in the history of the universe – namely he raised him from the dead.[2] It is the very infrequency and unusualness and rarity of the event that makes that event highly probable that God would do such a thing to vindicate the claims of his Son. So I think you can see that when you are working with a free agent and his decisions, you can’t just use frequency as your theory of probability – to say that resurrections are infrequent therefore it is highly improbable.

I really honestly believe the detractor of miracles finds himself in a real bind here. Namely, I don’t see any good way that he can prove that the probability of the resurrection on the background information is low or desperately low unless he has some good argument for atheism or some argument that if God exists he wouldn’t want to raise Jesus from the dead. Of course that takes us into the whole question of things like the problem of evil, arguments for the existence of God, and so forth. I think that the background information here includes things like the origin of the universe out of nothing in the finite past, the fine-tuning of the universe, the existence of objective moral values in the world, and all of the other arguments and evidences for God’s existence. I think that the resurrection of Jesus on the background information is really not at all improbable given the existence of God, which I think the background information renders likely.

In short, I think Hume’s argument is just completely fallacious. First of all, it fails to consider all of the probabilities that are necessary to consider. Secondly, it assumes without justification that an event like the resurrection is highly intrinsically improbable. That is why I think Bart Ehrman’s skepticism is so unnecessary. Given the evidence – the empty tomb, the origin of the Christian faith, the postmortem appearances of Jesus – there is simply no reason for his sort of skepticism with regard to the proclamation of the earliest disciples that God raised him from the dead.


Student: [inaudible]

Dr. Craig: This is one of the issues that came up in the debate. I am glad you asked it. Some thinkers think that it can. For example, Richard Swinburne is a very prominent philosopher at Oxford University. He has written a book on the incarnation and resurrection in which he actually tries to give numbers to these figures. Swinburne is very confident in this. He comes out with a probability for the resurrection of 97%! He says, Don’t put a whole lot of stock in this number. It is based on rough estimates. But I think what he would simply say is the resurrection is highly probable given the evidence and the background information. I myself don’t use this approach to the resurrection when I offer a case for it. Why? Because I think this probability here is really inscrutable. I think we can show with high probability that God exists. But I wouldn’t know how to show with high probability that if God exists he would raise Jesus from the dead rather than do something else. I don’t know how we could show that. While I think we can confidently say this is not low, I’m not sure how to prove that it is high either. So I tend to just leave a question mark here which means you can’t really fill in numbers. I think what we can say is that it is not low. It is not greatly less than 5 which means it is not going to disprove the resurrection.

Rather, the approach to the resurrection that I like to use is what is called inference to the best explanation. You remember when I shared about the resurrection, what I said is you arrange the competing hypotheses and then you assess them in terms of certain criteria like explanatory power (which ones explains best the evidence), explanatory scope (which one explains the fullest range of the evidence), plausibility, degree of being ad hoc (that is, contrived and artificial). In each case I would try to show that the resurrection of Jesus passes these criteria – that it has strong explanatory power, wide explanatory scope, it is not seriously ad hoc, it is plausible given Jesus’ radical claims and ministry, and so on and so forth.[3] So you don’t need to really assess these numbers on the approach that I would use. But I think this is useful against people like Ehrman and Hume who negatively argue that miracles cannot be established by the evidence because miracles are inherently improbable and therefore no amount of evidence can go to establish that. We can respond to that objection, I think, by showing that the critic cannot show the resurrection is improbable.

Student: [inaudible]

Dr. Craig: Certainly. In other words, in this background information here, you want to include things like maybe fulfilled prophecy, the accuracy of the Bible, and other things. That’s right. That can go into the background information. So somebody like Stephen Davis, for example, who is a Christian philosopher and has written a book on the resurrection, he argues that the resurrection is probable for someone who believes that the God of Israel exists. He would say somebody who is a first century Jew confronted with the apostles’ proclamation ought to believe in the resurrection because he’ll pack into that background information all these other facts that suggests that God exists, that he is the God of Israel, and so forth. Good point.

Student: [inaudible]

Dr. Craig: Your point, for those that couldn’t hear, is that it really gets back to the question of atheism and whether or not you believe God exists because obviously if there is no God then the hypothesis “God raised Jesus from the dead” has a probability of zero if there is no God! The argument requires you are at least open to the existence of God. The burden of proof here is on the critic. He’s got to show God’s existence is improbable or that he wouldn’t want to raise Jesus. But if you are at least open to the idea that God exists, this could actually lead you to believe, “Wow, maybe there is a God after all.” Or if you already do believe that there is a God then that would make it all the more likely, “Wow, this God has revealed himself in Christ.” That is why I like, before giving an argument for the resurrection, to give all of the arguments for the existence of a Creator, a Designer, a Moral Law-Giver, and so forth.

Student: [inaudible]

Dr. Craig: Good question. Is Ehrman an atheist now? In a Washington Post interview I saw with him, he said that after he lost his faith in the Bible and in Christ, he also lost his faith in God. So I think he would call himself at best an agnostic – someone who no longer has a belief in God but doesn’t really know whether God exists. I think he would characterize himself as an agnostic.

Student: [inaudible]

Dr. Craig: He should, but he doesn’t. He just pours scorn on this. You will remember in his letter he says, “You should get rid of that ridiculous mathematical proof for the resurrection.” Even though I am not using this as a proof for the resurrection. He is using it as a disproof of the resurrection, and I’m just saying you can’t do that because you haven’t estimated this probability and you don’t know what this probability is. So my position is purely defensive on this score.

Student: [inaudible]

Dr. Craig: I don’t think I would. His point was, every individual person’s birth is a singular event that is rare. But of course births in general are not rare. I am not sure how that would be cashed out. In one sense, every historical event is unique, right? But there are certain kinds of events that occur. As I said, I would just challenge the idea that rarity is a measure of probability at all. Frequency just doesn’t measure probability. They are not the same thing, especially when you are dealing with a free agent.

Student: [inaudible]

Dr. Craig: It is a “not” or a “negative” symbol – same thing. This would be “what is the probability of no-resurrection given our background information” and this is “what is the probability of the empty tomb, the postmortem appearances, the origin of the disciples’ faith given that no resurrection occurred?” In other words, how probable is it that you’d have an empty tomb, postmortem appearances, and so forth if say they just hallucinated or if, say, the women went to the wrong tomb, or things of that sort. What is the explanatory power? I think that is one of the tremendous weaknesses of these alternative naturalistic explanations – they are very weak in their explanatory power of the evidence.


That wraps up what we wanted to say with regard to the problem of miracles. Next time we will move into a further dimension of the doctrine of creation which will be dealing with the doctrine of angels and demons.[4]

[1] 5:04

[2] 10:08

[3] 15:08

[4] Total Running Time: 20:43 (Copyright © 2008 William Lane Craig)