Can belief in God be rational? David Johnson is a professor of philosophy at Yeshiva University in New York and author of a terrific book entitled Hume, Holism, and Miracles. Johnson’s book dismantles Hume’s long-venerated argument against the rationality of belief in miracles. It turns out that this highly esteemed argument is viciously circular, and so are all subsequent reconstructions of the argument in John Stuart Mill, John Mackie, Antony Flew, and Jordan Howard Sobel. Summing up Hume’s famous argument, Johnson writes,
I mention David Johnson, first, because his book was published around the time that I was seriously thinking about theism and Christianity. I vividly recall discovering Johnson’s book. It meant a great deal to me at that time, since I was really struggling with the intellectual viability of theism and Christianity (maybe I was struggling with what I would look like saying ‘I’m a theist’). Discovering that Johnson was a theist—truly, a bold theist who wrote in ‘History as a Guide to Eternity’ that there is a body of secular background knowledge relative to which the probability that the Resurrection occurred is roughly .9999 (see chapter 5, Truth Without Paradox)—was a major source of support at the time. He had been among my first instructors in graduate school, a recent graduate of Princeton and teaching courses on induction and counterfactuals. It was phenomenal for me to be taking courses from a student of Lewis, Kripke, and Harmon. Learning much later that a thinker as good, careful, and thorough as Johnson was a theist was intellectually a great relief.
Johnson once said—I can’t recall where, so I hope this isn’t a misattribution—that no one becomes a theist because of a good argument for theism and no one becomes an atheist because of a good argument for atheism. That’s a deep insight. People become atheists largely because of some horrible, inexplicable events in their lives or near their lives. The potency of arguments from evil is due more to the immediacy of concrete evil than to logical cogency. We are moved to atheism not through the understanding, you could say, but through the imagination.
Arguments from evil—all of them, from the logical problem of evil through the modal argument from evil—just aren’t very good arguments. All of them depend on poor analyses of gratuitous evil, for one, and all of them simply assume the ‘standard position’ on God and gratuitous evil. The standard position is that a morally perfect being must prevent every gratuitously evil state of affairs in every possible world. Lot’s of people have voiced the standard position in one form of another, but it is not difficult to show that it is impossible—metaphysically impossible—that any being, omnipotent or not, could prevent every gratuitously evil state of affairs in every possible world. It simply cannot be done. No matter what God does, no matter how much gratuitous evil God prevents, there is some gratuitous evil in some world. And of course God exists in that world, too! So the standard position (the so-called standard position) is flatly false. It is not necessary that a morally perfect being prevent every gratuitously evil state of affairs in every possible world. And that is just to say that it is perfectly possible that God exists in worlds where there is gratuitous evil.
On the other side of the coin, as Johnson observed, people do not become theists because of good arguments for theism. What persuades people to theistic belief is the experience or testimony to the miraculous. There is no obstacle—Hume’s argument is a catastrophe—against rational belief based on direct experience or testimony that the miraculous occurred. Of course, if you take a miracle to be the violation of a law of nature, then you have a conclusive apriori argument against the possibility of miracles—whatever else they might be, laws of nature are exceptionless regularities. It’s conceptually impossible that any law should admit of an exception. But miracles are not violations of laws of nature, they are rather exceptions to something which is extremely well-established, relative to a body of inductive evidence, as being a law of nature. If you like, they’re exception to almost-laws. There is of course no worse reason, given our inductive evidence E, to believe that a purported law L is an almost-law highly confirmed by E than to believe that L is a genuine law of nature highly confirmed by E. Indeed, the former is the more cautious conclusion given E. So what do we say about an event e that appears to be an exception to a law that, say, all A’s are B’s? Do we say that e shows that the evidence for the so-called law that all A’s are B’s is imperfect? Or do we say that e cannot have happened since the evidence for all A’s being B’s is exceptionless? Hume says the latter, and it is hopelessly question begging.
As Hume observed, apparently miraculous events are at the foundation of the great religions of the world. If the Resurrection did not occur, as St. Paul says, your faith is worthless.
Of course, there is testimony to the miraculous throughout the Old and New Testaments. Moses claims to have observed his staff become a serpent and St. John claims to have observed a formerly very dead man prepare breakfast. Neither Hume, nor any of Hume’s successors, have offered any argument against miracles that makes the belief in miracles anything less than rational.
Rational theological belief is instructively compared to rational philosophical belief. There are good reasons for and against every interesting philosophical position, but this does not keep anyone from taking a philosophical position. What makes someone take the position that, say, the B-theory of time is true given the fact that A-theory seems to have commonsense on its side? Isn’t there an absolute present and don’t we inhabit it? The B-theorists know all of the objections, and no doubt feel their force, but they are confident that a fully worked-out B-theory will provide the better account of time. And of course they are hard at work to provide a fully articulated B-theory. No one believes that B-theorists are irrational in their beliefs. Everyone knows they have a steep uphill climb to fully develop their views and address a host of serious objections to it. This is because everyone knows that A-theory is in more or less the same boat. And this phenomenon is perfectly general across every interesting philosophical debates: actualism vs. possibilism, endurantism vs. perdurantism, compatibilism vs. incompatibilism, realism vs. expressivism, and so on. Anyone who knows anything about philosophy recognizes that the arguments and reasons on each side of these debates are at best unstable, changeable, and tentative. Since we do not have decisive arguments—and probably never will have decisive arguments—on either side of these debates, the rational requirement on you is not to simply weigh up the arguments on each side and take a position. The very arguments that you’re weighing up are radically under construction! No sense in weighing anything up when the weights are constantly changing. The rational requirement on you is to contribute to the construction of arguments and replies on behalf of the position you suspect—perhaps on the basis of reliable or plausible testimony—is the correct position.
The situation for rational theological belief is exactly similar. The arguments for and against theism, as with all metaphysical debates, are at best unstable, changeable, and tentative. If you’re trying to weigh the arguments for and against theism, you have to recognize that the arguments are radically under construction. There is really no way to usefully weigh up the arguments since the weights are constantly changing—as a philosophy of religion editor I can tell you that they’re changing daily! The rational requirement on theists is not to try to weigh up changing arguments. The rational requirement on you is to contribute to the construction of arguments and replies on behalf of theistic belief—a belief that you almost certainly acquired on the basis of testimony to the miraculous.