The Ontological Argument

William L. Rowe

Rowe begins his article with a detailed explication of Anselm's Ontological Argument, making explicit the key inferences that form its base. Next, Rowe provides an account of several important criticisms of the argument and points out why those criticisms are inconclusive. The first of these is Gaunilo's argument (see previous selection) that Anselm's reasoning can be used to "prove" the existence of an island than which none greater can be thought. If we imagine such an island but suppose that it does not exist outside the mind, and since such an island would be greater were it to exist, we can think of an island greater than the island than which none greater can be thought. Clearly there is no such island, so the inference must be faulty. Rowe points out, however, that Gaunilo's reasoning is not precisely parallel to Anselm's. Specifically, Anselm only argues that there is some being than-which-nothing-greater-can-be-thought; he did not argue that there must be a being of every sort than-which-nothing-greater-can-be-thought. Thus, Anselm's argument only implies that there is some being (not necessarily an island at all) greater than Gaunilo's island.

Rowe next examines Kant's objection to the Ontological Argument, that existence is not a quality that makes something great(er) because existence is not quality in the proper sense at all. When we ascribe a quality like size or color to a thing (e.g., "that rock is large, or gray") we seem to presuppose that that thing exists. But if existence were a quality in the same way as color or size, to assert "tigers exist" would be redundant; ascribing existence to tigers is unnecessary because ascribing anything to tigers presupposes existence. Thus, ascription about size and color is very different from ascription about existence. On this line of reasoning, claims about existence are really claims about concepts; "tigers exist, says something about the concept of tiger, namely that that concept picks something out in the real world. The third criticism of the Ontological Argument that Rowe considers concerns whether or not the being-than-which-none-greater-can-be-thought is a possible object. Compare, for example, the infinite series of positive integers. For any integer in this series, there is a larger integer; thus, the integer-than-which-non-larger-can-be-thought is not possible. Perhaps the same is true of the being-than-which-none-greater-can-be-thought.

Rowe ends the article by articulating a fourth criticism of Anselm's argument. This criticism involves whether Anselm smuggles too much into his argument by defining God the way he does, claiming that existence is a quality that makes a thing greater, and asserting that God is a possible object. In fact Anselm's claim that God (defined as a being-than-which-none-greater-can-be-thought) is possible depends upon the assumption that there in fact exists some thing that is perfect. But this is precisely what he had set out to prove. To see why, consider a 'magican' to be an existing magician, and 'magico' to be a non-existing magician. Clearly, no non-existing magician (e.g., Merlin) can be a magican, and no existing magician (e.g., David Copperfield). Now suppose that no magicians actually exist. Because (a) possible magicians include only existing and non-existing magicians, (b) non-existing magicians are precluded by definition from being magicans, and (c) no magicians exist, it follows that there is no possible object that counts as a magican. However, if we were to simply assume that magicians were possible, it would follow that magicans exist. This is the structure of Anselm's argument. His concept of God, and assumption that existence is a quality, serve to make existence part of the definition of God. Thus, the premise that God is possible grants too much; namely that God exists just because existence follows from granting the possible existence of any entity with existence as part of its definition.

[an error occurred while processing this directive]