exists, then it is necessary that he exists.
That is the rather breathtaking conclusion of the modal ontological argument. And it is an entirely valid one, at least within the framework of modal logic. (To be specific, it is valid in the system of modal logic known in the trade as “S5.”) As Plantinga correctly observes, “It breaches no laws of logic, commits no confusions and is entirely immune to Kant’s criticism.”
Unlike Anselm’s ontological argument, the modal version does not take existence to be a predicate or a perfection. It does take necessary existence to be a perfection, but that is entirely plausible. Whereas existence is not a great-making quality—everything has it, after all—necessary existence obviously is great-making. To exist necessarily means that your existence depends on nothing else. It could not have been prevented. You are immune from the possibility of annihilation. Finally, and not least among its virtues, the modal ontological argument holds out the hope of answering the question Why is there something rather than nothing? If God is possible, it says, then God is necessary—and hence nothingness is impossible.
Is God possible? Or—to put it in the jargon of the modal ontological argument—is maximal greatness possibly exemplified? Think a bit about what “maximal greatness” means. A maximally great being is one that, if it exists in any possible reality, exists in all of them. It’s analogous to a being that, if it can be found anywhere in the world, manages to be everywhere, including here; or to a being that, if it exists at any moment in history, must exist at all moments, including the present one. A maximally great monarch would be one who, if he had a kingdom anywhere in the universe, would reign over the entire universe. A maximally great man, if he ever lived, would live eternally.
Clearly, maximal greatness is well beyond the realm of the familiar. How then could we know that such a thing is possible? Gödel concocted an elaborate argument to prove that the idea of a maximally great being was not inherently self-contradictory (the way, say, the idea of the largest number is inherently self-contradictory). Hence, Gödel concluded, such a being is logically possible. And since the range of possible worlds covers every logical possibility, there is a world that contains a maximally great being. But if such a being exists in any possible world, it must exist in every possible world—including our own, the actual world.
Unhappily for partisans of the ontological argument, this logic cuts both ways. There is nothing inherently self-contradictory either in the supposition that a maximally great being does not exist. Indeed, Plantinga himself refers to the property of there not being a maximally great being by the term “no-maximality.” So, by parity of reasoning, there must be a possible world in which no-maximality is exemplified—that is, one in which maximal greatness is absent. But if God is absent from any possible world, he is absent from all possible worlds—in particular, he is absent from the actual world.
So which will it be? If, in the framework of modal logic, we accept the premise that God possibly exists, then we are committed to the necessity of his existence. If we accept the premise that God possibly does not exist, then we are committed to the impossibility of his existence. Both can’t be true. Yet, from a purely logical perspective, the possibility of God’s existence seems no more compelling than the possibility of his nonexistence. Should we simply flip a coin to decide which premise to accept?
Recognizing the force of the counterargument, Plantinga has conceded that “a sane and rational man” might well reject the premise that a maximally great God is possible, and that the “canny atheist” will certainly do so. Without that premise, of course, this contemporary version of the ontological argument collapses. Nevertheless, Plantinga advocates accepting the premise in the interests of “simplifying” theology—the way one might accept a wild-sounding premise of quantum theory in the interests of simplifying physics.
Critics of the modal ontological argument will have none of this. “The premise that it is just possible that there should be something unsurpassably great looks innocent,” observed the Oxford philosopher (and staunch atheist) John Mackie. But this premise, he warns, is a Trojan horse: “Anyone who is not already and independently persuaded that traditional theism is true has good reason to reject the key premise” of the modal ontological argument. Thus, although the argument may be “interesting … as a logical peculiarity,” Mackie said, it