A few months ago, I spilled a glass of water on my daughter's book. I quickly wiped it off and put it down to go clean the area where I spilled. But when my daughter came to look for it later because she needed it for school, it was gone. We searched everywhere to no avail, so I just bought her a new book. It never reappeared, and at this point I would be surprised if it ever does.
How could that book have completely disappeared? Did the water have magical properties? Do we have a gnome infestation? That thought has occurred to me a few times when socks go into the laundry chute and never, ever make it out. Maybe the book went the way of those socks, whose partners have taken up permanent residence at the bottom of the clean laundry basket in the hopes that they will someday return.
But enough about my laundry woes, the point here is that in spite of my speculation, I know that we don't have gnomes. I also know that the water did not make the book disappear. In fact, I know that even if the book never shows up there's a natural explanation for its disappearance. I have an invincible unbelief in gnomes and magic water and I believe that to be a rational position.
I used the words "invincible unbelief" because during a debate between Antony Flew and Gary Habermas on the resurrection of Jesus, they agreed on the relevant historical facts, but Flew said that he had an almost invincible unbelief in the resurrection because it was so wildly different from our experience of how the universe functions. Flew agreed with David Hume's argument against miracles in the article "Of Miracles," which states, in a nutshell, that no matter how improbable a naturalistic explanation, a miracle is even less probable. Hume defines a miracle as a violation of the laws of nature and argues that experience is proof against them.
There are several problems with Hume's argument. First, the definition he gives of a miracle is a poor one. A better definition is the one given by J. L. Mackie: "The laws of nature . . . describe the ways in which the world--including, of course, human beings--works when left to itself, when not interfered with. A miracle occurs when the world is not left to itself, when something distinct from the natural order as a whole intrudes into it." That definition is not theologically precise, but it works for the purposes of this discussion.
The example I gave about gnomes stealing my daugther's book or water causing it to disappear would be a violation of the laws of nature in that the universal laws would suddenly not apply. However, the resurrection would be a miracle according to Mackie's definition because God raised Jesus from the dead; this was not simply a violation of the laws of nature, but an intrusion from beyond nature.
This distinction is significant because it is reasonable to believe that the laws of nature are constant. If they were not, the universe would be incomprehensible to us and science would be impossible. We know from experience that the laws of nature are predictable. However, this does not mean that it is reasonable to have an invincible conviction that nothing exists beyond nature. That is a different question altogether.
Second, Hume's argument may be reasonably applied to the disappearing book, but it is not reasonable to apply it to the resurrection of Jesus, which, if true, would be an argument for a God who exists beyond nature. In other words, if God raised Jesus from the dead, then that is an argument for theism, like the fine-tuning argument, the cosmological argument, and the argument from moral law. The argument is not just that a violation of the laws of nature took place, but that God intervened from beyond nature to raise Jesus from the dead, and He would have to exist to do so. And not only is it an argument for theism in general, but it is an argument for the Christian God.
In fact, I have frequently heard non-believers say that they would believe if they witnessed a miracle. Of course, if God had to perform a major miracle in the presence of every person who ever lived, then miracles would be the rule rather than the exception and therefore not evidence for His existence. Instead, He entered His creation in the flesh and did one major miracle that is supported by historical evidence: He rose from the dead. And His resurrection from the dead is the lynchpin of the Christian faith.
So Hume commits the fallacy of begging the question by rejecting the central claim of Christian theism without even considering the evidence. He dismisses the possibility that we could ever have evidence that a God who exists outside of nature revealed Himself to us within nature by doing something that is normally impossible.
Third, Hume's argument is essentially probabilistic, and it has been refuted by Bayes' Theorem. William Lane Craig explained this in his debate with Bart Ehrman, who used a variation of Hume's argument. Craig sets it up as follows:
Calculating the Probability of the Resurrection:
B = Background knowledge
E = Specific evidence (empty tomb, postmortem appearances, etc.)
R = Resurrection of Jesus
Pr (R/B & E) = ?
As I said before, Hume ignores the specific evidence for the resurrection, and says that the probability of the resurrection is R/B, with B representing our background knowledge about the laws of nature and the likelihood of a violation. Of course the probability is very low, in part because of his definition of a miracle and in part because he does not take into consideration the evidence for the resurrection and the plausibility of naturalistic explanations. Agnostic philosopher of physics John Earman (not to be confused with Ehrman) did not mince words in calling Hume's argument fallacious in his book Hume's Abject Failure.
Since Hume's argument is about the probability of a miracle, Earman and Craig used Bayes' Theorem to refute it. Craig argued that the equation has to look like this:
Pr (R/B) x Pr (E/B&R)
Pr (R/B & E) = ____________________________________________
Pr (R/B) x Pr (E/B&R) + Pr (not-R/B) x (E/B& not-R)
The numerator is the probability of the resurrection given our background knowledge and the specific evidence for the resurrection. The denominator reproduces the numerator and adds the probability and explanatory power of all the naturalistic explanations. So the lower the probability of those explanations, the higher the probability of the resurrection. But if the naturalistic explanations [Pr (not-R/B) x (E/B& not-R)] are plausible and have explanatory power, then the probability of the resurrection goes down.
If that is not clear, Craig spends more time (and does a much better job) explaining it here.
The point here is not to give a probabilistic value to the resurrection but to demonstrate that Hume's argument is fallacious because he oversimplifies the probability of the resurrection. Instead of dismissing the resurrection as always the least probable, we also have to take into consideration the specific evidence for the resurrection and the probability of the naturalistic explanations, like numerous people--including skeptics--having the same hallucination in different situations and at different times, and being willing to give their lives for the truth of what they perceived.
But even if we ignore the evidence for the resurrection and just go back to the original equation Pr = R/B, then Hume's argument still fails because the arguments for theism in general make the inherent probability of the resurrection greater. If the existence of a God who created ex nihilo is the best explanation for why the universe emerged out of nothing, and an Intelligent Designer is the best explanation for the fine-tuning of the universe, then the probability that God raised Jesus from the dead increases. So the background knowledge is not just the probability of a violation of the laws of nature--it is the probability that God raised His Son from the dead, and the probability of that increases if the evidence in cosmology points to a Creator. Pr = R/B is affected by the probability of the existence of a Creator.
In the introduction to "Of Miracles," Hume says, "I flatter myself, that I have discovered an argument . . . which, if just, will, with the wise and learned, be an everlasting check to all kinds of superstitious delusion, and consequently, will be useful as long as the world endures."
Although those of us in the throes of "superstitious delusions" are eternally grateful to Hume for his thoughtful gesture, the problem is that his argument is far too ambitious because it is only useful with respect to true superstition. For example, if I were tempted to plug in my gap of knowledge about the missing book with an explanation like gnomes, then his argument would be a "check" to that kind of a "superstitious delusion." But it tells us nothing about the probability that a God beyond nature would identify Himself to us within nature by superseding its laws, because for Him to be able to do that the laws of nature would have to be predictable in the first place.