Craig on the Ontological and (Leibnizian) Cosmological Arguments

Recently, I’ve been looking over the third edition of William Lane Craig’s Reasonable Faith, as well as the textbook he wrote with J.P. Moreland, Philosophical Foundations of a Christian Worldview. One thing these books have that’s missing from a lot of Craig’s works is an attempt to defend, in some detail, arguments that he usually doesn’t give space to. Today, I want to look at his treatments of the ontological argument and the Leibnizian cosmological argument.

The Ontological Argument

Craig’s version of the ontological argument comes from Alvin Plantinga, and might be better described as the modal argument (“modal” is a bit of philosophy jargon for “having to do with possibility and necessity.”) While most versions of the ontological argument, if successful, would seem to prove that God exists necessarily (could not possibly have not existed), Platinga’s argument is slightly unusual in the way it makes ideas of possibility and necessity central to the argument by assuming, from the start, that God is by definition a necessary being. Plantinga’s central claim is that if it’s possible that a God, so defined, exists, then God must exist.

This claim seems very odd to non-philosophers, but it’s important to see here that Plantinga is talking about genuine possibility, not just possibility in the sense of “possible for all we know.” While many logicians had doubts about the logically validity this pattern of inference (possibly necessarily p, therefore p), it seems to me just obvious that they are valid. If God is supposed to be the sort of being who couldn’t possibly have not existed, then it can’t be that he doesn’t exist but might have. If God, so defined, doesn’t exist, then the concept of God isn’t the concept of a possible being. But that means that if there is in a fact a genuine possibility that God exists, then God must in fact exist.

Now while the pattern of inference here may be logically valid, on the face of it we have good reason not to take arguments in this form seriously. In particular, it’s widely thought that mathematical statements are the sorts of things that are, if true, necessarily true, and so a Plantinga-style modal argument could be made on behalf of any statement in mathematics, but nobody would take seriously a “proof” of a mathematical claim of this form. The trouble is that when we understand the initial possibility claim as genuine possibility and not just “for all we know” possibility, we have no reason to accept that initial claim.

I think Plantinga is sensitive to this sort of work, makes only modest claims for his argument: he says only that, because it’s reasonable to accept the premise, it’s reasonable to accept the conclusion. Even that seems to go to far–even if the premise is reasonable, I don’t see how this enhances the reasonableness of the conclusion. If it did, then it would seem that anyone could become more reasonable in their mathematical conjectures simply by reflecting on modal logic.

Craig’s, on the other hand displays none of Plantinga’s modesty. In Philosophical Foundations (p. 497), Craig proudly quotes Plantinga saying that he believes his ontological is as strong “as any serious philosophical argument for any important philosophical conclusion,” and follows this up with a quote from George Mavrodes asking why, then Plantinga’s argument doesn’t amount to a proof.

The answer to Mavrodes’ question should be obviouis: one might think that proofs are nowhere to be found in philosophy. Theistic philosopher Peter van Inwagen, for example, has claimed the problem of evil is a “philosophical failure,” but qualifies this by explaining he thinks that no philosophical argument is really successful. It would be a mistake to report that, since van Inwagen doesn’t claim the problem of evil is any worse off than any other argument, van Inwagen accepts it as a disproof of God’s existence! Plantinga himself seems to think that all philosophical arguments ever do is establish the rationality of believing the conclusion, and if he’s right about that it would be a mistake to describe any philosophical argument as a proof.

In attempting to build up the modal argument for theism into a compelling argument, Craig claims that we must accept that a necessary God is possible unless the notion is incoherent, and to defend that claim, he argues that other concepts of necessary beings–say a necessarily existent lion–are incoherent. But to argue that, he just says that it seems there are possible situations in which no lion could exist. Here, Craig can be understood as using the inference “possibly, there is no necessarily existent lion, therefore there is no necessarily existent lion.” For this to prove what Craig wants, he needs the further assumption that impossibility entails incoherence, which I doubt, but the really interesting thing is that Craig’s argument against a necessarily existent lion follows the same patter as Plantinga’s modal argument, only run in reverse.

Thus, Craig backs up his claim that the modal argument for the existence of God is successful by claiming that modal arguments against things like necessarily existent lions are also successful. Also, just one can make a modal argument against necessary lions, one can make a modal argument against a necessary God. This means Craig must simultaneously affirm that it’s obvious that there might be a necessary God, and it’s obvious that there might not be any lions, but it’s not obvious that there might not have been a God. I simply don’t understand why anyone would find that conjunction of claims obvious.

Finally, Craig does address one of the most obvious objections that arises with most ontological arguments–the fact that they seem to prove the existence of various quasi-gods, beings with many but not all the properties of god, whose existence the theist doesn’t want to accept. Craig’s response is that a quasi-god would be incompatible with the existence of God, so ontological arguments for quasi-gods beg the question against the existence of God. I can only see this as a non-sequitur.

Traditionally, “begging the question” has meant assuming what is to be proved, and ontological arguments for quasi-gods don’t do this. What Craig means, I suspect, is “uses a premise that no one who rejected the conclusion would accept,” but I don’t see why doing that is a problem. Anytime a premise indisputably entails a conclusion, those who are committed to rejecting the conclusion can be counted on to reject the premise. But this will happen even when the premise seems obviously true. Thus, I don’t see how Craig can find fault with such “question begging” arguments, especially since he himself disparages people who reject a premise simply because they don’t like the conclusion it points to (Reasonable Faith p. 55).

The Leibnizian Cosmological Argument

Craig states the Leibnizian cosmological argument as folloows:

1) Anything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause.
2) If the universe has an explanation of its existence, that explanation is God.
3) The universe exists.
4) Therefore, the universe has an explanation of its existence. (from 1,3)
5) Therefore, the explanation of the existence of the universe is God. (from 2, 4)

Surprisingly, Craig’s primary defense of premise (2) is claiming that atheists already accept it! According to Craig, “Atheists typically assert that, since there is no God, it is false that everything has an explanation of its existence, for the universe, in this case, just exists,” and this is equivalent to accepting premise (2) (Reasonable Faith p. 108). While many atheists don’t find the claim that everything has an explanation of its existence convincing, I know if no atheist who says it is false because there is no God, and Craig doesn’t cite a single example. My own view is that, while the Leibnizian claim that there must be some necessary explanation of contingent things (things that might not have existed) seems initially tempting, on reflection there isn’t any reason to think any candidate for the necessary explanation more plausible than any other, and since some candidates are obviously implausible, we should think that there is no necessary explanation of things’ existence.

Craig does have a second argument for (2), that the universe is by definition all of physical reality, so its cause must be non-physical, and therefore would have to be mental. But this seems to assume that there are non-physical minds, or at least that the idea of non-physical minds is more plausible than non-physical, non-mental forces. I don’t see why anyone would think that, though. Craig sometimes talks as if it’s just obvious that the human mind is non physical (see Philosophical Foundations p. 238, for example). While this claim probably serves him well in public debates when he often wants to patch up his arguments as quickly as possible and then move on, intellectually it is unconvincing.

Leave a comment