18 Ontological Arguments for God’s Existence
A quite different version of the cosmological argument is presented by William Lane Craig, drawing upon the Islamic philosophers of the 9th-12th centuries such as al-Ghazali (1058-1111), called the kalām cosmological argument. Craig argues that whatever begins to exist has a cause, that the universe began to exist, and that God must be invoked as its cause. Why believe that the universe began to exist? For one thing, it seems that the universe cannot have an infinite temporal duration since the successive addition of finites cannot add up to something infinite. Just as one cannot “count to infinity,” the compounding of the moments that pass in time could not ever add up to an infinite temporal duration. For another, if we make the supposition that the universe has an infinite temporal duration various absurdities arise. Sundays are a subset (one-seventh) of all the days that have ever occurred. A very bored deity would count out six non-Sundays for every Sunday. But if the universe has an infinite temporal duration, then an infinite number of Sundays have occurred. And an infinite number of non-Sundays have occurred. Therefore, the subset is equal in magnitude to the set—an absurdity. So, the universe began to exist. Notice that Craig’s argument avoids referring to necessary beings, or the principle of sufficient reason; Craig’s argument requires only that if something begins to exist, then it has a cause. Supporters of the kalām cosmological argument may also cite scientific evidence to support the idea that the universe began to exist, for instance the Big Bang theory or the idea that if the universe had an infinite temporal duration, then entropy would guarantee that complex matter would not exist presently (Craig 1979).
One key question about Craig’s kalām cosmological argument is whether the cause of the universe must be something like our conception of God, a kind of personal agent. Craig, following al-Ghazali, suggests that the cause of the universe must be timeless, outside of time entirely. Physical causes bring about their effects, as it were, immediately. For example, an effect like the process of water freezing will begin to happen as soon as its cause, a sub-zero temperature, is present. So, if the cause of the universe is timeless and is a physical cause, we would expect the universe to have always existed. But as we have seen, that cannot be. So, the cause of the universe must be non-physical. Aside from physical causes, we sometimes explain effects as resulting from actions—we have the idea that personal agents bring about effects spontaneously as and when they will to do so, in a way that is different than and not entirely determined by physical causes. On this model, plausibly the cause of the universe is the action of a personal, but non-physical, agent. Others have objected, though, that it is difficult to make sense of the idea of a personal agent who acts but is also outside of time, and again that we are having to rely too heavily on our limited repertoire of concepts: for all we know, there might be causes that are neither like the physical nor like personal agency.
Questions to Consider
- It seems that the opponent of the cosmological argument can try to defuse it by denying that the universe has a reason for its existence, or a cause, or by denying the principle of sufficient reason. Are these unreasonable moves? Is there any claim or principle that it would be unreasonable to deny, if the alternative was the conclusion that God exists?
- In theory, could science one day prove that the universe did not begin to exist? What impact would such a finding have on Clarke’s cosmological argument? On Craig’s kalām cosmological argument?
- Is it reasonable to rely on our limited repertoire of concepts, as exemplified in the discussion about whether the cause of the universe is a personal agent? Should we be worried by the thought that reality may be stranger than we can conceive?
The Ontological Argument
“Ontos” being Greek for “being” or “existence,” the ontological argument is unusual in that it has no empirical premises at all; God is not called upon as an explanation for anything. Rather, God’s existence is proven by reflection on the concept of God. This is an extremely unfamiliar way of proceeding, since ordinarily we think that by analyzing the concept of something, we may discover the predicates that will be true of it if it exists, but not that it exists. For instance, if I have a child then the predicate “has a grandfather named Patrick” will be true of it. The ontological argument proposes, in the case of God, to abolish this “if” and proceed directly from the concept of God to his existence. The argument’s first proponent was Anselm of Canterbury (1033-1109). It’s a familiar idea that God is great, the greatest in fact, so great one cannot think of anything greater. Anselm draws on this familiar idea in his Proslogion. There, Anselm characterizes God as “a being than which nothing greater can be conceived” (Anselm 1078). In more modern language, Anselm is saying that God is the greatest conceivable being, that it is part of the concept of God that it is impossible to conceive of any being greater than God. It seems that existence is greater than nonexistence. So, if we conceive of God as nonexistent, then we can conceive of something greater than God: e.g., a shoe, a flea. But God is the greatest conceivable being, so our assumption of God’s nonexistence must have been false, and God must exist. Another way of putting this is that Anselm anticipates Hume’s objection that no being’s existence is necessary (since any being’s nonexistence can be conceived without contradiction). Anselm insists that in this case the idea of God, properly understood, does give rise to contradiction if we suppose his non-existence. “The being which must exist does not exist” seems like a contradiction.
From the outset, the ontological argument has had difficulties heaped upon it. For one thing, although it may seem intuitively right that existence is greater than nonexistence, what does “greater” mean? Better than? Preferable to? More real than? A satisfying characterization is hard to find. Another early objection comes from Gaunilo of Marmoutier (994-1083), who makes the parodic suggestion of an island that is the greatest island that can be conceived. If such an island is to be greater than, say, Corsica, it must exist. Must we then say that such an island exists? Surely not. The difficulty raised by Gaunilo is that it seems that the predicate of existence can be bolted on to any concept illicitly. Anselm responds, however, that his argument applies uniquely to the greatest being that can be conceived (not a given, limited kind of being like an island), since although the imagined island would indeed be greater if it existed, it is not part of the concept of anything except the greatest being that can be conceived that it be greater than everything else, and so for it alone can we infer its existence from its concept. A similar response is that contingency is part of the concept of an island (or dog, or horse, or any other specific, limited kind of being which we are acquainted with), so that a necessarily existing island would simply be a contradiction. Only with the non-specific concept of “a being” in general would contingency not just be included in the concept.
The most historically influential criticism of the ontological argument, however, comes from Immanuel Kant (1724-1804). In his Critique of Pure Reason, Kant argues that existence is not a predicate (Kant 1781). Think about the concept of a banana. We can attribute certain predicates to it, such as “yellowness” and “sweetness.” As time goes by, we might add further predicates to the concept, e.g., “nutritional potassium source.” Now think about what happens to the concept of a banana when you suppose that bananas exist. It seems that the concept is not changed at all. To say something exists is not to say anything about the concept of it, only that the concept is instantiated in reality. But if existence cannot be part of a concept, then it cannot be part of the concept of God, and cannot be found therein by any sort of analysis.
Kant’s argument was widely taken to be calamitous to the ontological argument. However, in the 1960s, the argument was rejuvenated, in a form that (perhaps) avoids Kant’s criticism, by Norman Malcom (1911-1990). Malcolm suggests that although existence may not be a predicate, necessary existence is a predicate. As contingent beings, we are the sort of things which can come into and go out of existence. But if God exists, then he is a necessary being rather than a contingent being. So, if he exists he cannot go out of existence. This is a predicate God enjoys, even if existence per se is not a predicate (Malcolm 1960). Intuitively, “indestructibility” and “immortality” are predicates that alter the concept of a thing. Another modern version of an Anselmian ontological argument is offered by Lynne Rudder Baker (1944-2017). Baker’s version avoids the claim that existence is a predicate (as well as several other traditional difficulties). Instead, Baker notes that individuals who do not exist have mediated causal powers, that is, they cause effects but only because individuals who do exist have thoughts and beliefs about them: Santa Claus has the mediated causal power to get children to leave cookies out for him, children who themselves have unmediated causal powers. In short, to have unmediated causal powers is intuitively greater than having mediated causal powers, so given that God is the greatest being that can be conceived of, God must have unmediated causal powers, and so he must exist (Baker 2013).
A final difficulty that we may mention for these three theistic proofs is whether they prove the existence of the God of Abraham, or the God of classical theism (supposing that the two are the same) — which it is the concern of most theistic philosophers to do. The teleological argument may show a designer, which corresponds tolerably well to the creatorhood of God, but seems to fall short of showing God’s other attributes, like omnibenevolence. Similarly, the world-cause or necessary being purportedly shown by the cosmological and ontological arguments may seem far distant from a personal God who is interested in our affairs. One theistic response is that these arguments may work in combination, or be supplemented by the evidence of revelations, religious experiences, and miracles (See Chapter 3 for a few such arguments), or we may be able to find ways in which one divine attribute implies the others. Bear in mind also that there are many less well-known theistic arguments beyond these three traditional ones (McIntosh 2019). (For some specific examples, see Chapter 3.)
Questions to Consider
- Do we really have a conception of “a being than which nothing greater can be conceived”? Is that something we are able to frame in our minds, or have we just begun to misuse words?
- If existence is not a predicate, why do we treat it as one in ordinary sentences, like “the pecan tree exists”? Further, how do we delineate the domain of fiction? Isn’t our concept of “Homer Simpson” a concept of a character who does not exist? If not, what is it a concept of?
- Even once you grasp it, does the ontological argument seem intuitive to you? Does it seem less intuitive than the cosmological argument? Should you put much weight on your intuitions about these arguments?