The Ontological Argument – William L. Rowe

Rowe begins by making a distinction between two types of arguments for the existence of God that might be given. The first such sort is an a posteriori argument. The defining feature of arguments of this kind is that they rely on actual experiences that we have had of the world. Another kind of argument, an a priori argument, relies on no such experiences. These arguments could be reasoned through even if we had had no interaction with the world; Rowe points out that the Ontological Argument is an example of this second type of argument. The argument is supposed to rely only on an analysis of the concept of God in order to demonstrate that such a being must exist.

In §I, Rowe attempts to clarify important terms and ideas in Anselm’s argument, and ultimately states what he takes the “key idea” to be in the Ontological Argument. This idea is that, for Anselm, existing in reality is “great-making quality.” Rowe reminds us that Anselm distinguishes between existence in the understanding, and existence in reality. The idea is that for a thing to exist in reality is for it to exist in the world in the way that we ordinarily understand objects like desks, chairs, cats, planets, and electrons to exist. Existing in the understanding, though, is much different. Although unicorns are not real objects in the world, they exist in the understanding in the sense that we have an idea of what a unicorn; it exists, we might say, “in our minds.” What Rowe is identifying as Anselm’s key idea is that, for any being that exists only in the understanding, but could have existed in reality, that being could have been greater than it is. Following this, Rowe formulates in §II what he takes to be Anselm’s argument.

In the final sections of the selection, Rowe discusses and evaluates several kinds of objections that have been raised against the Ontological Argument. He first discusses Gaunilo’s objection involving the example of the island, but ultimately defends Anselm against this objection, identifying two problems in Gaunilo’s reasoning. The second objection contends that Anselm’s argument is flawed because it treats existence as a predicate, but it in fact is not. Rowe points out that a crucial point to this objection is the claim that when we say that a being has a certain property, we are thereby implicitly asserting that that being exists. If this were true, then it would not make sense to treat existence as a predicate, as we could only ascribe it to something if we assumed that the thing existed. Rowe contends that we do not treat predication in this way at all, and that the objection therefore loses its force. The last objection that Rowe considers in §III contends that the type of God that Anselm has in mind is actually not possible. Again, Rowe demonstrates that the reasoning behind this objection is flawed.

In §IV, Rowe addresses a final kind of criticism of the Ontological Argument. In the first place, he examines Anselm’s concept of God, and examines whether anything concerning God’s existence follows from that concept all on its own. He argues that although it does follow immediately from that definition that no non-existing thing that could have existed could be God, it does not follow that God actually exists. After all, the concept cannot itself demonstrate that something instantiates that concept; it remains to be seen whether some actually existing thing does in fact have the properties that Anselm has in mind. The analysis cannot stop here, though, as Rowe notes. The criticism loses force once we pay heed to Anselm’s claim that God is a possible thing. Any possible thing will be either existing or non-existing; if it really is true that Anselm’s concept of God rules out the possibility of God being a non-existing thing, then it follow from God’s possibility that God must be an existing thing. In light of this, it looks as if it really must be the case that God exists in reality. Rowe points out, though, that we need not be so quick to grant to Anselm that God is a possible thing. This is because, given his understanding of God and his assertion that existence is a great-making feature, allowing that God is possible is just equivalent to allowing that God is an actually existing thing. And surely this is too much to ask. Rowe allows that Anselm’s argument succeeds only if there is already some existing thing that is maximally perfect, but denies that the argument successfully shows that such being must exist.