Excursus on Natural Theology (Part 25): Q&A on the Ontological Argument

May 03, 2016

Review of a Defense of the Ontological Argument

We’ve been talking about the ontological argument, and I hope today to bring our discussion of that argument to a close. The last time we saw that the crucial premise in the argument is the first one. If you grant that premise then the rest of the argument just follows automatically. If it’s possible that a maximally great being like God exists then it follows necessarily that God does exist and therefore God exists in the actual world. So the question is what warrant there is for that first premise, and I argued that we can see that the concept of a maximally great being (that is to say, a being that is maximally excellent: omniscient, omnipotent, morally perfect in every possible world) is a coherent concept. It’s not like a square circle or a married bachelor. This is a coherent notion that could possibly be instantiated. Therefore we have a prima facie warrant for thinking that that first premise is true that it’s possible that a maximally great being exists.

Then we looked at a number of objections to this premise. For example, we looked at attempts to parody the argument by saying wouldn’t it also lead to the existence of a maximally great pizza or the existence of a necessary lion or things of that sort which are obviously absurd? I think we saw clearly in each case that these were actually incoherent concepts. The idea of a maximally great pizza or necessarily existent lion are incoherent concepts and therefore not possibly instantiated and therefore quite different from the concept of a maximally great being.

Finally we closed by looking at the suggestion that maybe it’s possible that a quasi-maximally great being exists – a being who is very, very great, perhaps omnipotent and morally perfect, but maybe limited cognitively (maybe he doesn't know future contingents like the god of open theism). So he’s only quasi-maximally excellent, not maximally excellent in every world. What I suggested there was that if a maximally great being exists or is possible then it’s impossible that there be a quasi-maximally great being because as an omnipotent being a maximally great being would have to have the ability to annihilate or create anything other than himself including a so-called quasi-maximally great being. So if it’s possible that maximal greatness is instantiated, it follows that it is impossible that there be a quasi-maximally great being because such a being would be independent of the maximally great being and that’s impossible since he’s omnipotent. Since our intuitions about the possible existence of a quasi-maximally great being are parasitic or depend upon our intuitions about the possibility of a maximally great being, it’s plausible that we should think that it is possible that a maximally great being exists but it’s not really possible that there be a merely quasi-maximally great being.


Student: Whenever I talk to a lot of people about the ontological argument lately, one of the things I keep hearing from them is they will say the whole idea even of what makes a maximally great being – great-making properties – is just subjective. How do you really know it has to be omnipotent? What if I say a maximally great being is some kind of . . . insert some other property here that is not omnipotence, omniscient, and omnibenevolence?

Dr. Craig: I don't find that a persuasive objection. I think it's obvious that a being is greater if it has omnipotence than if it’s weak and limited in its power. It is greater if the being is morally perfect than if it’s morally defective in some ways.[1] Similarly it is greater if it knows everything – is all-knowing – than if it's limited in its knowledge and only knows some things. So I am persuaded that people who take that route are just looking for an escape route from the argument rather than honestly appraising whether or not it is greater to be omnipotent than limited in power, whether it is greater to be all-knowing rather than limited in your knowledge, whether it's better to be morally perfect than morally defective. I think those are clearly great-making properties. By contrast, in the debate I had at the University of Cambridge someone suggested from the floor that maybe this would lead to the existence of a greatest smelliest being. And as my colleague Peter Williams pointed out, clearly being smelly is not a great-making property in the way that these others are. So I don't think this is at all subjective.

But never mind that. Suppose it were. Still, that really doesn't matter. Anybody is free to propose what he takes to be the qualities of this being and then ask, “Is this a coherent notion?” And if it is coherent then it's possible, and therefore it exists. So really I don't think that the identification of these properties lies at the heart of the issue because you can just stipulate the properties that you want to go to make up maximal greatness and then ask, “Is this a coherent idea?” And if it is then the argument follows.

Student: You know how it starts with “It is possible God exists” and then from there you get the rest of the argument. There is this other objection which I guess relates to the quasi-maximal being objection but I think Peter van Inwagen calls this “the correct atheist argument.” Let’s say you substitute “It is possible God does not exist” and then you get the argument from there. Then the question would be like . . . surely it seems equally plausible – intuitively plausible – that he could not exist. How do you really break the stalemate between the two?

Dr. Craig: There I want to revert to what I said about the difference between epistemic possibility and metaphysical possibility. Certainly you can say it is possible epistemically that God does not exist. For all we know God does exist or doesn't exist. That's agnosticism, right? But that sort of epistemic possibility goes no distance toward showing that God's non-existence is metaphysically possible. If the notion of a maximally great being is coherent then the idea that this being does not exist is incoherent – it's impossible. Think again about Goldbach’s Conjecture. If it’s true, it’s necessarily true. Only epistemically can we say it’s possible that it’s true and it’s possible that it’s false. If it’s true then it’s impossible that it be false. And I would say similarly it is impossible metaphysically that God not exist.

Student: I think what they try to say to that is that you ultimately can’t get beyond epistemic possibility in the sense that you still have to judge which one do you go with – not possible or possible.

Dr. Craig: That’s right. Here, again, I want to appeal to the evident coherence of the idea of maximal greatness – that we have a kind of intuitive insight into this notion and we can see that this is a coherent notion. So while this isn't a knockdown argument obviously, it gives us, as I say, some prima facie or face-value warrant for thinking that this is possible.

Student: A friend of mine at Georgia Tech and I were talking about this argument. He was having difficulty really spelling out how sure can we be about a maximally great being and coming up with all these different things. He said what if we just said it is possible that a necessarily existing being exists and then just let it go from there? In that way you wouldn’t have all-knowing, all-powerful, all-present, but you would have a necessarily existent being which would kind of be like God.[2]

Dr. Craig: I think that argument would be sound as well. Unfortunately it doesn't tell you a whole lot about the nature of the necessarily existent being. Could it be matter, for example, as Aristotle thought? For Aristotle it was merely temporally necessary that matter exists. He didn’t have the idea that it was logically necessary. You would still then need to know something about the nature of this being. But you’re right. That would be a step in the direction of theism to say contingent reality is not all there is. There’s some necessarily existing reality.

Student: In The Nature of Necessity, Plantinga states that this argument really doesn't prove or even establish its conclusion, but he says that mainly it just establishes that it’s rational to assent to the conclusion. In debate you’ve used the argument as an argument for the existence of God, and so I’m wondering if you differ from Plantinga on that.

Dr. Craig: I addressed that in our previous lesson, and pointed out that in The Nature of Necessity, Plantinga set the bar for success in natural theology far too high. He later retracted that statement. Here's what he later said, and I’ll quote this again:

I employed a traditional but wholly improper standard: I took it that these arguments are successful only if they start from propositions that compel assent from every honest and intelligent person and proceed majestically to their conclusion by way of forms of argument that can be rejected only on pain of insincerity or irrationality. Naturally enough, I joined the contemporary chorus in holding that none of the traditional arguments was successful. (I failed to note that no philosophical arguments of any consequence meet that standard; hence the fact that theistic arguments do not is of less significance than I thought.)[3]

Then he went on also to say,

The ontological argument provides as good grounds for the existence of God as does any serious philosophical argument for any important philosophical conclusion.[4]

That is quite a ringing endorsement of the argument on Plantinga’s part. If you're interested in those references I can give them to you after class.

Student: Last time you mentioned towards the end of class these analogies that people throw out like a maximally great island or pizza or bird. I guess I was left wanting – can we not more forcefully and emphatically reject these analogies by way of saying the second someone posits a physical analogy like an island or a bird or a pizza it is immediately not a good analogy? It’s a horrible analogy because it’s dependent upon space-time.

Dr. Craig: I think you are right. Any physical analogy can be immediately rejected because, for example, as I said it’s possible that the universe exists as a space-time singularity and nothing else, in which case there couldn’t be something like a bird, a lion, a pizza, or anything. Absolutely, I think these physical analogies are really very unscientific. They can be immediately ruled out.

Student: They try to squirm out of it by saying, “I’m talking about a non-physical island.” Then it’s not an island!

Dr. Craig: That is so perceptive. You are absolutely right. Then you have incoherence because what is a non-physical island? This happened in my debate with Lewis Wolpert in London where he says, Maybe it was a computer that made the universe. And I said that’s impossible – a computer is a spatio-temporal object that is material. And he said, I’m talking about an immaterial computer beyond space and time. And I said, Well that’s just another name for God! You’ve got to have coherent concepts if you are going to oppose these two.[5]

Student: It seems to fall down between (2) and (3) in this particular argument as far as I can see. It says here that a maximally great being exists in some possible world. Because it may exist in some possible world doesn’t mean that particular world actually would exist.[6]

Dr. Craig: Right. That's correct. It can be a non-actual possible world.

Student: So if you move on to (3), “If it exists in some possible then it exists in every possible world.” If that possible world didn't exist then it wouldn't exist in every possible world.

Dr. Craig: The reason that (3) follows is because the definition of maximal greatness is “maximally excellent in every possible world.”

Student: But if the possible world doesn’t exist then it essentially breaks the possibility.

Dr. Craig: No, because when you say it exists in a possible world all you are saying is “possibly, a maximally great being exists.” To say it exists in a possible world is just a heuristic device – an illustration of saying possibly a maximally great being exists. So the idea that is if it is possible that a necessary being exists then it exists necessarily in every possible world because that is what the definition is. It just follows from the definition of maximal greatness that if it is possible then it exists in every world.

Student: Maybe I am just not particularly sold on the definition of God as a maximally great being. Maybe this is because I’m a physicist and not a philosopher. But the term seems a bit fuzzy to me because it would seem to me that, say, a Calvinist could say that our version of God isn’t maximally great because he has sovereignty but doesn’t exert it. A Muslim could say our version of God isn’t maximally great because he is not transcendent completely above humanity the way Allah would be. It seems to me that there is this fuzziness in this definition.

Dr. Craig: I think what you are saying is that the concept of maximal greatness that Plantinga lays out is incomplete, and that's true. But Muslims, Calvinists, Armenians would all agree that the idea of God as a being which is omniscient, omnipotent, and morally perfect. And that's the bare-boned definition of maximal greatness. Maximal greatness will go far beyond that, as you say. There there may well be dispute about the nature of the maximally great being. But what Plantinga is giving is some necessary conditions of maximal greatness, but he’s not saying that these are all of the properties. Those properties seem to be pretty clear to me. As I said earlier, they seem to pretty clearly belong to the notion of maximal greatness and would be acknowledged by all the people that you mentioned.

Student: I wonder if part of the reason we are kind of stuck on this is that our Western mindset wants empirical data for everything, and this is a non-empirical argument.

Dr. Craig: It sure is!

Student: I wonder if this argument is more acceptable to people with an Eastern mindset where scientific data is not quite as important.

Dr. Craig: I wouldn’t know. I don’t have enough experience in the Orient to be able to answer that question. Certainly, in Oriental religions like Hinduism, Buddhism, Taoism, you do have the idea of an absolute reality which is, I think, metaphysically necessary.

Student: I wonder though if this argument for God would work better for people who are already thinking in that term as opposed to the Western Enlightenment thought.

Dr. Craig: The difficulty is - here we are just talking sociology, right? - the difficulty for the Eastern mind would be while it resonates with the idea of metaphysically necessary existence, it tends to view that existence as impersonal and so would be reluctant to say that this absolute reality is omniscient or morally perfect because those are attributes of a person, and they don't want the metaphysical absolute to be personal. But then again, I want to revert to what I said earlier. It’s better to be morally perfect than not and that this does plausibly belong to maximal greatness despite the Eastern mindset. But you’re certainly right in saying that this argument would be more appealing to people who think in abstract ways of mathematicians, philosophers, maybe engineers, information technologists, as opposed to people who grub around in physical science.[7]

Student: I think the argument stands whether you can know the attributes or not. The argument still stands. But take omniscience and Molinism. To me, omniscience means he knows all that is known, but he has reserved free will for himself so he does not predestine everything. He is with us in time. Therefore it goes against Molinism, but I think that is a superior maximal being.

Dr. Craig: We are not doing theology here. We are not trying to talk about the difference between Calvinism and Molinism or anything of that sort. We are working with a very bare-boned concept.

Student: Even though it is not compatible with my theistic beliefs, let’s stipulate that worlds outside this actual world exist. We’ve been over that. We have a disagreement about that. Let’s stipulate they exist. Let’s also stipulate that we’ve gotten an atheist to accept premise (1) saying that it is possible for a maximally great being to exist. I would find that remarkable and somewhat surprising, but that is fine. Now let’s go down to (3) which I consider the most objectionable aspect of this whole argument. It is, “If a maximally great being exists in some possible world then it exists in every possible world.” I believe that statement is false on its face. I believe even cursory inspection would say that that would show that that is a false statement. A maximally great being would certainly not exist in every possible world because being maximally good he could exist only in a minority of worlds because he would find the others absolutely appalling.

Dr. Craig: Ah. Let me just ask you a question first. Last time I commented on how, on your view that there is only one logically possible world, you should love the ontological argument because if it is possible that God exists then he does exist because there is only one possible world. What did you think about that? Did you resonate with that point that I was making?

Student: I must admit I didn’t take it to heart. Would you state that again? Maybe I just don’t completely understand it. Say it again one more time.

Dr. Craig: Remember I said that the challenge of the ontological argument is that if there is a plurality of possible worlds and it is possible that God exists in one, the challenge is: how do you show that he exists in the actual world? Let’s let that be W0. That is the actual world. How do you get from God’s existence in W3 to God’s existence in W0. That’s the challenge of the ontological argument. But if you say that there are no other logically possible worlds – that there is only one possible world, the actual world – then it follows immediately that if God is possible then God actually exists because this is the only possible world. That is why I was saying you ought to love this argument!

Student: Well, let’s extend that a little further. Let’s say if any world at all exists then God exists. Because as I say, I’m a theist and I don’t believe it is possible for any world to exist unless God created it. That is why I am saying that we could think about worlds with orange skies and all this kind of stuff. Those worlds were possible up to the moment of creation but God obviously rejected them. Therefore they are forever impossible.

Dr. Craig: OK, good. Thank you. Let me now respond to the question that you raised which I think is a very important question. What you point out is that it seems intuitively that there could be worlds that are possible in which God doesn’t exist. Why? Well, let’s imagine a world in which the highest form of life is rabbits which exist in a state of disease and unremitting misery.[8] That seems intuitively possible. But I think you would argue (and I would agree) such a world is incompatible with the existence of God because God is too good to allow there to be a universe in which there is nothing but rabbits in a state of unremitting misery. You can draw two conclusions from this. You could say, “A-ha! I guess that shows that there are logically possible worlds in which God does not exist,” which is what you suggested, in which case God is not maximally great. Or you could say (and this is what I would say) such worlds are not in fact really possible. They are actually impossible because they are inconsistent with the existence of God, as you quite rightly point out. What this means is that given Anselm’s view of God there will be worlds that are not really possible after all, though at face value (if you don’t think about God; if you put God aside) it would seem like there could be a world like that. It would mean that the existence of God will sort of scramble your modal intuitions and show that some things are not really possible after all. Another example would be, say, a world in which everybody goes to hell and nobody is saved. Surely such a world would be incompatible with God’s existence and therefore he wouldn’t create it. But such a world, I would say, is impossible therefore and cannot exist. It is not really a possible world.

Student: I would say that this concept of other possible worlds would be much more acceptable to atheists because they have no clear idea where the world came from. A lot of times they will just put that off – so, sure, anything is possible. But to a theist, the only way a world can exist is if there is a God to make it.

Dr. Craig. There you are reverting back to the cosmological argument, which I endorse and have defended. So I agree with you on that, but I want to see if this argument also can’t supplement or complement the cosmological argument by saying that this is a coherent idea – a maximally great being – and if it is a coherent idea then it’s possible and therefore it’s actual. While this is not perhaps the knockdown argument that Anselm thought it was, I think that it does have some prima facie plausibility and can be part of a cumulative case for theism.

Student: Well, your reputation is well established, and who am I to give you advice?

Dr. Craig: I appreciate your interaction. That was a very good question.

Student: I would suggest that this one would be kind of at the bottom of the stack.

Dr. Craig: OK. Fair enough! Actually, as you’ll see when I finally get to our conclusion, I’ll say something almost like that.

Student: If people are getting caught up on the idea of maximal greatness, what about Anselm’s idea of something’s existing only in the understanding alone is not as great as something that exists actually in the world? Why can’t we use that?

Dr. Craig: If I understand you correctly, you were asking about Anselm’s argument that something that exists in the mind only is not as great as something that exists in the mind and in the world as well. And therefore God’s existence is real in the world. Plantinga discusses that version of the argument, and I think ultimately it doesn’t go through. You still need this crucial premise that it is possible for something like God to be not only conceived in the mind but to exist in reality. That is the crucial insight that Leibniz had – that Anselm’s argument assumes that it is possible for such a thing to be real, to exist not only in the mind but also in reality. That is why, again, the first premise of the argument is, I think, the crucial one.

Student: I actually really like the argument. I think for apologetics it is just kind of like the pill to give to atheism and it trumps all, but they also have to accept it. I think problematically (semantically or philosophically) it is kind of hard to digest for everybody. The problems that I am running into are how to articulate answers for Kant’s response to it saying that existence can’t be a property or Geach’s saying that you can’t quantify identity by words. I think if you keep it on a philosophical level and you can answer at a philosophical level and you don’t get into the cosmological and stuff like that . . . that is where I am having trouble keeping it when I discuss this argument at a philosophical level. Everybody wants to run into something physical or non-metaphysical or whatever. It is hard to keep their brains on something philosophical.

Dr. Craig: Yes, it does require the ability to do abstract reasoning or philosophical reasoning. With respect to the objection by the German philosopher Immanuel Kant, who was one of the most important Enlightenment critics of natural theology and theistic arguments, Plantinga’s version of the argument is immune to Kant’s objection. That is why I just jumped over several centuries – from Anselm to Plantinga – rather than go through all of the convoluted history of the argument. What Kant said is that existence is not a property. If I have a concept in my mind of one hundred dollars, it is not that a hundred dollars in reality has some additional property – namely, existence – that the hundred dollars in my mind doesn’t. They are exactly conceived the same as to what it is. It is just that the one is real – instantiated in reality – and the other one isn’t. But you can argue about that. Some people say, “Yes, existence is a property. The real hundred dollars has the property of existence and the imaginary one doesn’t.” You can argue about that. But you can just circumvent the whole issue because with Plantinga’s argument it doesn’t assume that existence is a property. What it says is that necessity is a property. I think there it is pretty widely agreed that there is a distinction between something’s being contingent and something’s being necessary. So even if existence itself isn’t a property, contingent existence or necessary existence is a property. What Plantinga’s argument says is that a maximally great being is one that has maximal excellence necessarily in every possible world. He just circumvents Kant’s objection.

I didn’t expect to have questions about Kant in our Defenders class. This group is getting more and more sophisticated all the time!

Student: Let’s go back to the suffering rabbits world. It seems that someone could accuse you of using circular reasoning. You are excluding the worlds in which God doesn’t exist in your definition of possible worlds, and only allowing the worlds in which God does exist to define which worlds are possible. Therefore, of course the argument works. How would you respond to that?

Dr. Craig: I think that is a powerful objection. That is a little bit similar to the objection about quasi-maximal greatness. In other words, I can see that the idea of a world in which the highest life form is rabbits in unremitting misery is possible, but that is incompatible with the existence of God so it is possible that God not exist. I guess what I would say is that any intuitions that you have that such a world is possible are parasitic or depend upon the assumption that maximal greatness is impossible. It admits it is impossible to reconcile with maximal greatness. It just pushes it back to that question again. Is maximal greatness possible or impossible? All you’ve done is just shown a scenario that is incompatible with maximal greatness, but you haven’t shown that maximal greatness is impossible and that therefore this world is really possible. That’s what I would say.

Student: I’m wondering if the distinction between possible worlds and conceivable worlds can make the issue about worlds with suffering rabbits helpful in that it is conceivable that God could exist in such worlds but it is not possible.

Dr. Craig: Yes, this is a helpful point that he is making, though I want to put conceivability on the side of metaphysical possibility. What I would contrast with metaphysical possibility or conceivability is imaginability.[9] I can imagine a world in which the highest life form is rabbits in unremitting pain, but that doesn't show that a world like that is really possible. Similarly, you can imagine that Goldbach’s Conjecture is false. Right? That’s easy to do. I can imagine the mathematicians tomorrow discovering a proof of its falsity. But if it’s true, that sort of scenario is impossible. I can imagine it, but it can’t be real. Similarly, I can imagine things popping into existence uncaused out of nothing. I can, in my mind, picture a rabbit popping into being without a cause. But that goes no distance towards showing that such a thing is metaphysically possible. So I think you are right. We mustn’t confuse the imaginability of such a world with its metaphysical possibility. That is a good point that is helpful.

Student: The ontological argument . . . is there a limit to these possible worlds? With the existence of God as, like you said . . . such a world wouldn’t be compatible with God being the amount of suffering? Or is there actually an infinite amount of worlds that we could pick from?

Dr. Craig: If I understood you correctly, you are asking, “Are there possible worlds which are incompatible with the existence of God?” Is that what you were saying?

Student: Yes, and also going off the ontological argument, is there a limit to these possible worlds?

Dr. Craig: The second one is “no.” The logical possibilities are unlimited. They would be infinite. But then, on traditional theism – wholly apart from the ontological argument, let’s just talk about theism – theists take necessity or aseity to be an essential attribute of God. He exists in every possible world. His non-existence is impossible. Therefore, there are no possible worlds where God does not exist.

Student: Accuse me of being metaphysical (but that’s OK, I think), it seems that this world, which has fallen according to Genesis to sin, has a lot of suffering, a lot of rabbits running around with disease and everything else. And we are promised a world where there isn’t sin and it is perfect. When we look at this world we go, wait a minute, if God is all-great, why do we have suffering?

Dr. Craig: That will be the topic to which we shall turn after we finish the ontological argument! We’ve been looking at arguments for God’s existence, but of course on the other side of the scale will be arguments against God’s existence. And the most important of these will be the problem of innocent suffering. If there is an all-powerful, all-loving, morally perfect being then why is there so much innocent horrible suffering? Here one could go on and on with examples of little children starving to death or dying of horrible disease or in fires and things. That will need to be addressed. With respect to my rabbit example, the key word in the rabbit example was the word “unremitting.” That they are in a state of unremitting suffering. That is to say, God doesn’t do anything about it like send a redeemer or something like that. I think that God will not allow that to be unremitting suffering. A good God will provide a solution. That was key in my example of this world.


OK, that was a very good discussion today.

What we’ve been looking at up to this point in defense of the first premise of the ontological argument are a priori considerations. You can remember what that means by the word “prior” in it. It is prior to any experience you have of the world. So as someone was saying, this is an argument that doesn’t appeal to empirical facts.[10] It is an a priori justification for premise (1). But what Plantinga suggests is that there could also be some a posteriori reasons for thinking that premise (1) is true. Notice there we have the word “posterior”. Posterior means after experience. Maybe in addition to any a priori grounds we have for affirming premise (1), there might be some considerations from experience – a posteriori arguments – that would support premise (1) as well. That would put a very different face on the ontological argument. That is what we will look at next time – some of these a posteriori considerations in support of the key premise.[11]



[1] 5:08

[2] 10:08

[3] Alvin Plantinga, Warranted Christian Belief (New York: Oxford University Press, 2000) p. 69.

[4] Alvin Plantinga, “Reason and Belief in God,” typescript dated October 1981, pp. 18-19. This paragraph was inadvertently omitted in the published version of the essay. Fortunately, a nearly identical paragraph appears in Alvin Plantinga, “Self-Profile,” in Alvin Plantinga, ed. James E. Tomberlin and Peter van Inwagen, Profiles 5 (Dordrecht: D. Reidel, 1985), p. 71.

[5] To see this exchange, skip to the 1 hour and 23rd minute of the Craig-Wolpert debate found here: http://www.reasonablefaith.org/media/craig-vs-wolpert-westminster-hall-london (accessed May 7, 2016).

[6] 15:04

[7] 20:18

[8] 25:12

[9] 35:06

[10] 40:02

[11] Total Running Time: 41:25 (Copyright © 2016 William Lane Craig)