September 19, 2015

Reservations about the Ontological Argument

Dear Dr Craig,

I have a reservation regarding the Ontological argument as you defend it.

You identify the first premise, it is possible that a maximally great being exists, as the controversial one. You defend it as being more plausibly true than false with two sub-arguments. The first of these is that the notion of a maximally great being seems to be coherent, and that this implies such a being is possible. The second is an appeal to the other theistic arguments; that their plausibility shows that it is at least possible for a metaphysically necessary being to exist.

We can argue against the first sub-argument, that the notion of a maximally great being seems to be coherent and is thus possible, in the following way. This sub-argument requires that conceivability, or conceptual coherence, implies metaphysical possibility. But we have a good reason for thinking that this is false.

Define A as something that is possible but has never existed, like a phoenix or someone that can bench press 70000kg. Define B as an A that has the additional feature of being actual, of existing in the actual world; so a phoenix that exists in the actual world. There is nothing incoherent in the concept of B, it is just an A that is actual, it is quite conceivable. But it is not possible for B to exist, since A has never existed and a necessary feature of B is that A is actual. So we have something (B) that, while a coherent concept, is not possible.

This objection attempts to show that conceivability is not a guide to possibility. And while this is drastic, since we all of us use such a guide every day, its seems a solid objection and if successful would undermine the Ontological argument substantially.

I would be very grateful if you were to respond to this objection, since I have not been able to think of a satisfactory answer to it. In addition, thank you so much for your work on the Kalam cosmological argument; I find the it fascinating and am very nearly convinced by it. Please pray for my conversion.

Kind regards,


United Kingdom

Thank you, Jason, for your question! I'm encouraged by your openness to God, and in response to your final request I offered a silent prayer for you before writing this reply. May you open your heart to the Lord as He convicts you of your need and draws you graciously to Himself!

Your reservation about my first justification for the possibility of a maximally great being (that is, a being which is omnipotent, omniscient, and morally perfect in every possible world) is that coherence is not a good guide to possibility. Now in one sense this objection is just misconceived: in contemporary parlance to be logically coherent just is to be logically possible. It’s not as though one is a guide to the other; they’re just synonymous terms. No one would say that something that is logically coherent might be logically impossible.

Your reservation is really about the reliability of our modal intuitions. Something might seem to us logically coherent or possible and yet not really be so. But it is no part of the ontological argument to assert that our modal intuitions are infallible or indefeasible. The question is whether your scenario ought to make us call into question our intuition that maximal greatness is possibly exemplified.

Your envisioned scenario is quite similar to the objection of the late philosopher J. Howard Sobel. Sobel invites us to conceive of something which, if it is possible, is a dragon in whichever world is the actual world. This is just like your “phoenix that exists in the actual world.” (So you are in the company of an eminent philosopher in having your reservations!) If such a thing is possible, then a dragon exists. So is it possible? The only way to know, says Sobel, is to look around and see if there are any dragons. If there aren’t, then the notion is not logically possible after all. You can’t just rely on your modal intuitions.

Now we need to be clear on just what sort of a being we are being asked to conceive. By “the actual world” you have to mean “whichever world is actual.” Otherwise, the analogy to the ontological argument falls apart.[1] You have to be talking about the concept of a contingent being which mysteriously manages to track whichever world is actual so as to exist in the actual world while not existing in all possible worlds.

But then I don’t think we have any intuition that such a being is possible. On the contrary, to me it sounds patently impossible. The idea of a metaphysically necessary being, a being that exists in the actual world if it is possible, seems unproblematic because it would exist in every possible world and so, of course, in whichever world happens to be actual. But the idea of a being which is contingent and yet such that if it is possible, it exists in whichever world is actual sounds crazy. There would be no way for a contingently existing thing to guarantee its actuality.

So I don’t think the scenario you describe undermines the reliability of our modal intuitions about the possibility of a maximally great being.


[1] If by “the actual world” you mean our world or this world (call it W*), then a phoenix that exists in the actual world is just a phoenix that exists in W*. But W* need not be actual. If some other world W¢ is actual, then in talking about the possibility of a phoenix that exists in W*, you are just comparing non-actual worlds with each other and can’t conclude anything from that about what actually exists.