Excursus on Natural Theology (Part 23): The Ontological Argument

April 06, 2016

The Ontological Argument

In our excursus on natural theology we’ve talked about a number of arguments for God’s existence such as the argument from contingency, the kalam cosmological argument, the argument from the fine-tuning of the universe, the moral argument. Today we turn to a new argument (and the last that we’ll be surveying in our class), and this is the ontological argument for God’s existence.

In the year 1078 a Benedictine monk by the name of Anselm, who later became the Archbishop of Canterbury, formulated a new and bold argument for the existence of God which has now fascinated philosophers for a millennium. A year earlier in 1077 Anselm had finished writing a treatise called the Monologium in which he presented cosmological and moral arguments for God’s existence. But Anselm was dissatisfied with the complexity of the case for theism that he had developed, and he wanted to find a single argument which would prove that God with all of his attributes in all of his greatness exists. He had pretty much given up on the task when he came upon the definition of “God” in Latin as aliquid quo nihil maius cogitari potest. The Latin is so great. You can learn this phrase and impress your friends when they ask you for a definition of God! That is to say, God is something than which nothing greater can be conceived, or, in more idiomatic English, God is the greatest conceivable being.

Anselm argued in his treatise that followed – the Proslogium - that once you understand the definition of God, once you understand what God is, then (if you’ve really understood it) you will see that God must exist because if God did not exist he would not be the greatest conceivable being. A greatest conceivable being must be an existent being. Otherwise it wouldn’t be the greatest. So God’s existence is inconceivable for anybody who really understands the word “God” and understands what God is.

Anselm says that is why Psalm 14:1 says, “The fool has said in his heart there is no God” – because if that person really understood the word “God” then he would see that God must exist. So he is a fool for saying that the greatest conceivable being does not exist.

Anselm’s argument came to be known as the ontological argument which is from the Greek word ontos meaning “being.” It went on to assume a variety of different forms, and has been defended by some of the greatest thinkers in the history of philosophy, for example John Duns Scotus, Rene Descartes, Benedict de Spinoza, Gottfried Wilhelm Leibniz, and so on.

What is the common thread in all of these different versions of the argument that unites them and makes them ontological arguments? I think the common thread among these various ontological arguments is that they all try to deduce the existence of God from the very concept of God together with some necessary truths. Proponents of the ontological argument in its various forms maintain that once we understand what God is (once you have an adequate conception of God – whether the greatest conceivable being, or the most perfect being, or the most real being) then we will see that such a being must in fact exist.[1]

This argument has tended to sharply polarize philosophers. For example, the 19th century German philosopher Arthur Schopenhauer characterized the ontological argument as “a charming joke.” That opinion is certainly shared by many other philosophers today. On the other hand, the argument has been taken very seriously and in fact defended as sound by quite a number of 20th century philosophers who are of some prominence as well, notably Norman Malcolm, Charles Hartshorne, and Alvin Plantinga.

Since Plantinga’s version of the argument is, I think, the most sophisticated and the most recent development of the argument, we’ll use it as the springboard for our discussion of the ontological argument. In Plantinga’s version of the argument he appropriates the insight of Gottfried Wilhelm Leibniz that the ontological argument assumes that the concept of God is possible. That is to say, the argument assumes that the concept “God” or “the greatest conceivable being” is a coherent concept. Or, using the semantics of possible worlds, it assumes that there is a possible world in which God exists.

For those who are unfamiliar with the semantics of possible worlds, let me just say a word of explanation about this lest we be misled. When we talk about possible worlds, we do not mean planets or even other universes. Rather, a possible world is simply a maximal description of reality. It is a way reality might be. I think the easiest way to think about a possible world is as a huge conjunction of propositions – p and q and r and s and so on. These individual conjuncts (p, q, r, s) are propositions which can be true or false. A possible world is a conjunction which comprises every proposition or its contradictory so that it yields a maximal description of reality. Nothing is left out of such a description. By negating different conjuncts or propositions we can arrive at different possible worlds. So, for example, we could call W1 this description of the world:

W1 = p & q & r & s . . .

But other descriptions of the world could be:

W2 = p & not-q & r & not-s . . .

W3 = not-p & q & not-r & s . . .

W4 = not-p & not-q & r & not-s . . .

These would continue being the range of possible worlds. Only one of these descriptions will be comprised of propositions or conjuncts all of which are true and so is the true description of the way reality actually is. That description we will dignify by saying it is the actual world. One of these will have all true conjuncts, and that will be the actual world.[2]


Student: This has always been the argument for God that I’ve had the hardest time wrapping my head around. I just want to be sure that I am clear so far. This is different than the multiworld hypothesis because we are not saying each of these are actual worlds out there somewhere. We are talking about one single actual world, and these are all the different ways it could have been although only one of them is correct. So it is completely different than the multiworld.

Dr. Craig: Right. This is different than the multiverse hypothesis or the Many Worlds Hypothesis in cosmology or quantum physics. Here, as I say, just think of these worlds as descriptions – just big conjunctions like “Ben Jones exists. Cindy Fox exists. Bryant Wright exists. William Craig exists.” That is all this is. It is just a big description, and the description that is true is the actual world.

Student: In order to get rid of a possible world as a possible world – is the only way that we can do that is to find a contradiction within . . . ?

Dr. Craig: That is a good question. I will say something about that in a minute. These propositions obviously have to be compossible – they have to be able to both be true in the same world. Right? Otherwise that is an impossible world. So that is not a possible world. But they also need to be possible in and of themselves. The proposition itself needs to be possible. I’ll say something about that in a moment. You need to have both each conjunct being possible in itself, and then you need their combination to be possible to be a possible world as well.

Student: I was going to ask “what do you mean by possible?” What are the underlying assumptions of “possible?”

Dr. Craig: This is a really good question, and it is probably impossible to define it! Because it is sort of a primitive concept. The idea would be actualizable or realizable. This is something that really could exist – really could be actual. Does that help?

Student: Given maybe the laws of physics of this universe?

Dr. Craig: No, no. Not the laws of physics. OK, good question. There are different types of modality, aren’t there? One of these would be what we might call physical modality. You say that something would be physically possible if it is consistent with the laws of nature, be physically impossible if it contradicts the laws of nature. But that is a fairly narrow kind of modality. There are things which might not be physically possible but they are still logically possible. The kind of modality that is at play here is, again, this sort of ill-defined type of modality called broadly logical modality. We are talking about broadly logical possibility and necessity. Again, what does one mean by that? One means that something is broadly logically possible if it is actualizable or realizable. If it could really be real then that will be something that is broadly logically possible. Sometimes this is called metaphysical possibility and necessity. That is another label. But it is still the same idea – it is something that is realizable.

Student: As you will probably remember, I never thought this was a very good argument. I guess my core complaint is that other so-called possible worlds are not possible. There is only one world possible, and that is this one. The only way you could get another world would be to get a creator (as a theist) to do that. To try to get God to make another possible world, say that had an orange sky instead of a blue sky or which one of us didn’t exist, I would say is imaginative and fanciful. In other words, just because you can imagine (which is what these philosophers are doing) these other worlds doesn’t mean there is any way that any other world but this one could possibly be a reality. That is why I would never evoke this argument.

Dr. Craig: We’ve been around this block before. I think you just don’t get it. What you are expressing is a kind of a logical fatalism where everything that is true is necessarily true and there are no possibilities.[3] On your view it is impossible, for example, that God might have refrained from creating the world, or it is impossible that Peter not deny Christ three times, or it is impossible that the second person of the Trinity take on a human nature one second later than he did in Mary’s womb. Those just seem to be utterly implausible sorts of assertions and would really deny God’s freedom. It would mean that God has no logical possibilities either. So as long as we are not conceiving of these things as actual worlds some place but simply as ways the world could have been, surely there are contingencies, there are possibilities. Some of these concern God himself. If you deny that you land in a kind of logical fatalism even concerning God where everything that happens happens necessarily. That to me just seems obviously mistaken, and also theologically pernicious. It would make evil necessary, for example.

Student: Those are not my attitudes, and I don’t think that they necessarily flow from my position. I don’t see how that flows – how a logical fatalism flows from my view.

Dr. Craig: But you said there is only one logically possible world – this one.

Student: I don’t know what you mean by logically possible.

Dr. Craig: Right! That’s why I said, I don’t think . . .

Student: Logic is of this world.

Dr. Craig: That is why I said I don’t think you get it here. The idea is, as I said earlier, we are talking here about something that is realizable. It is actualizable. This reality could be that way. Freedom on God’s part would surely necessitate that there are other . . .

Student: Of course my God has the freedom to create any world he wanted. But the fact is he did not. That is what I’m saying. Any of these others he did not do that for various reasons known only to him and his sovereignty. He chose not to do that. I can imagine “what if he did?” or “what if he didn’t?” That’s fanciful to me and not really based in reality.

Dr. Craig: These other worlds are not real. That’s right. But they are ways the world could have been. I think you want to admit that. You want to say God has the freedom to do differently and to have done differently.

Student: I would say they could not have been because he chose not to do it, and therefore they could not have been. Had he chose to do it then yes it would be reality.

Student: My understanding of this is that it is just a concept for philosophical discourse. It is to answer the “what ifs” of your debate opponent. Because otherwise it would not be possible to argue the ontological argument unless you can answer the “what ifs.” If your debate opponent says, “How can you say there is a maximally great being if the world could have made itself?” Or something like that. If we lived in a universe where things popped into existence out of nothing and we could demonstrate that then . . . I’m saying what a debate opponent might say about that. It answers the “what ifs.”

Dr. Craig: It seems to me that whether in debate or even just in your own thinking we entertain possibilities all the time. This is especially true when we deliberate about what we are going to do. We think that there are real possibilities that we have the freedom to actualize or not. Possible worlds, as you say, is just a way of conceptualizing this. I look at it as a sort of heuristic device. It plays the same role that diagrams do in mathematics textbooks. They simply are ways of representing it. To say that something is possible like “It is possible that Christ could have been born one second later” - to represent that you say there is a possible world in which Christ was born one second later. It is just a visual aid, if you will, of thinking about possibilities and necessities. It is a very helpful heuristic device because many of these questions would be very difficult to entertain and think about without these sorts of illustrations.

Student: It is a way of considering counterfactuals.

Dr. Craig: Those play a role in this, too. That is right. Counterfactuals, like you said, “If I were rich then I would do this or that.” Those involve possible worlds. That is right.

Student: I’m just wondering if it is not the use of “worlds” which is tangible causing confusion.[4] It is possible that God created this planet – our existence – without mongooses. He didn’t, but he could have. I think that would be an example of a possible world, but is it because we are using that word that is causing confusion. We are not thinking of it as a concept anymore. We are trying to put reality into it.

Dr. Craig: That is absolutely right. That is why I said when we talk about possible worlds we don’t mean planets or universes. We just mean a description. Just think of this list and these are just propositions. It is a big, long conjunction of everything that is true or false. That is what it is. And you call that “a world.” And one of those is true. One of these conjunctions has all true conjuncts in it, and you call that one the actual world. So it is just a description.

Student: Well, to you it is a world. To me it is a possibility.

Dr. Craig: That’s fine! It is a possible world. Right. In the Zangmeister video that Zangmeister is developing right now for the ontological argument, the way he has chosen to portray possible worlds (in order to avoid this misunderstanding) is very clever. Rather than using circles or globes that look like worlds, he uses doors. He shows how you open one door and inside there is unicorns and pink elephants, and you open another door and there is something else in there. I thought that is very nice because it avoids this idea of universes or multiverses, that kind of thing.

Student: At this point in the discussion, do any of those p’s and q’s and r’s include God’s existence?

Dr. Craig: Yes, quite definitely. Now we haven’t gotten to that yet, but let’s let r equal the proposition “God exists.” In W1 that proposition is true, right? Because it says “r”. But in W3 that proposition is false because it says “not-r.” Here we have a case where in possible world 1 God exists, but in possible world 3 God does not exist. That is quite right. It is a maximal description of reality, so it will include propositions like “God exists.”

Student: How, with the ontological argument, do you stay on track with it being ontological? I think if you go to the worlds thing everybody thinks of it as a physical world. If you go to, say, God exists like for instance in a world designed can’t exist without a designer therefore it is God. But then you have the design argument. If you go into cosmological where there is a universe and not nothing then you have the cosmological argument. I think one of the biggest difficulties that I’ve seen with the ontological argument is that it is difficult to stay strictly philosophical.

Dr. Craig: Well, I hope that when we look at the premises of Plantinga’s argument that that will help us to stay on track because he lists the steps very explicitly. The way the argument moves is that if it is possible that God exists then God does exist. That will be the move that it makes. From the mere possibility of God’s existence it will follow that God actually does exist. That is how it will reason. We’ll see whether or not that is a good argument.

Student: I thought you said earlier that Anselm was saying that God is the greatest conceivable being. Then I looked up what I had written – the translation from Anslem – and it said greater than can be conceived. Did I hear that wrong?

Dr. Craig: God is that than which nothing greater can be conceived. That was his definition. Aliquid quo nihil maius cogitari potest. More idiomatically, a greatest conceivable being. That is what God is. If you can conceive of something greater than God then that would be God. So by definition, Anselm says, God is the greatest conceivable being.[5]

Student: So the conceivability limits God?

Dr. Craig: Yes, insofar as if the concept of God is incoherent then God cannot exist. If the concept of God is like the concept of a married bachelor then God cannot exist. That is the insight that I mentioned that Leibniz had. Leibniz said the argument is assuming that the concept of God is a coherent concept. It assumes that the concept of God is possibly instantiated. There is a possible world in which “God exists” is true.

Student: What I am trying to say is conceivability is a human element. So basically that is the limitation.

Dr. Craig: Here I want to go back to what Bob said because he made a distinction that I think is a nice one between what we can imagine and what we can conceive. We are not saying that God is the greatest imaginable being because that would be limited to human imagination. But the idea of greatest being conceivable is the idea of a being which it is logically impossible for there to be something greater than it. There is a difference between imaginability and conceivability. For example, I can conceive mathematically of a thousand-sided polygon. I can think of a geometrical figure that has one thousand sides. It is not difficult to conceive. But I can’t imagine such a thing. I can’t form any kind of visual image of a thousand-sided polygon. That would be an example of something that is conceivable even though it is unimaginable. We are not saying that God is limited by human imagination, but simply that God is the logically greatest being that can exist. If that concept is incoherent then God cannot exist. We’ll see how that actually plays a very, very crucial role in the argument.




[1] 5:10

[2] 10:00

[3] 15:04

[4] 20:12

[5] 25:07

[6] Total Running Time: 28:14 (Copyright © 2016 William Lane Craig)