Aquinas and Russell Against Anselm

In his essay on the ontological argument in The Blackwell Guide to the Philosophy of Religion, Gareth B. Matthews makes an interesting observation about one of Aquinas’s lesser-known objections to the argument. The objection is fleshed out by reference to Russell’s theory of definite dsecriptions, and forms a pretty solid argument against Anselm.

If we take Anslem’s Fool to be saying ‘there is no God’, we can take Anselm as interpreting negative existentials as being both about something ‘in the understanding’ that does not exist in reality. Given Anselm’s argument (which Matthews takes to be a kind of reductio), such a negative existential, asserted by the Fool, is a contradiction:

Thus the denial of God’s existence is to be understood as the claim that something in the mind, namely, God (that is, something than which nothing greater can be conceived) fails to exist in reality. But now we see that the Fool is claiming is self-contradictory. It is self-contradictory because it is the claim of something in the understanding that it is both (a) something than which nothing greater can be conceived and also (b) something that fails to exist in reality. But anything in the understanding that satisfies (b) will be something than which a greater can be conceived, namely, something that exists in reality. And so the atheist’s claim becomes, it seems, a claim about something in the understanding that it is both something than which nothing greater can be conceived and also something which a greater can be conceived, which would clearly be absurd. (p. 89)

Now this take on the argument is something of a logical take as opposed to a metaphysical take. If the Fool affirms what Anselm says he affirms in the way he affirms it, then he affirms a contradiction, which is, of course, precisely Anselm’s point. This is, I imagine, what Aquinas was picking up on when he says, ‘No difficulty, consequently, befalls anyone who posits that God does not exist. For that something greater can be thought than anything given in reality or in the intellect is a difficulty only to him who admits that there is something than which a greater cannot be thought in reality.’ (from here) There is an air of circularity here, to my mind at least. If Anselm’s game is played according to his rules, then the Fool isn’t going to enjoy a victory. So the trick, then, would be to change up just what is actually affirmed by the Fool (or at least how it’s affirmed).

Matthews enlists the help of Bertrand Russell here. Against Anselm’s understanding of negative existentials, he offers a Russellian understanding:

Bertrand Russell suggested that when we say, “Socrates doesn’t exist,” we can be taken to have a definite description in mind to replace “Socrates” – maybe “the teacher of Plato.” So perhaps “Socrates doesn’t exist” means “the teacher of Plato doesn’t exist,” and what that means is, perhaps, “Nobody fits the description, ‘the teacher of Plato.'”

If we don’t accept the idea of translating

(1) Socrates doesn’t exist

into a sentence that mentions the English phrase “the teacher of Plato”…we could follow Russell’s theory of descriptions and translate (1) into this:

(2) It is false that there is at least and at most one person who taught Plato.

Similarly, we could begin with the Fool’s

(3) God doesn’t exist

and use Anselm’s quasi-definition to get this:

(4) Nothing fits the description, “something than which nothing greater can be conceived.”

If, now, we wanted to get rid of the English phrase in (4), we might come up with this:

(5) For any given things, in the understanding or in reality, one greater than it can be conceived.

Someone might ask how the Fool could possibly know (5) to be true. But here it is well to keep in mind that the form of Anselm’s argument is a reductio. Since the Fool is simply stating, in a dramatic way, the supposition Anselm wishes to show absurd, the Fool doesn’t need to know that (5) is true, or even produce any evidence for thinking it is true. The Fools’s role is simply to state that God does not exist so that Anselm can show that the Fool’s statement leads to self-contradiction. If the fool insists on having his statement of atheism understood as (5), he can avoid, it seems, Anselm’s claim that the Fool has contradicted himself. (p. 93)

With a little linguistic and logical footwork, then, the Fool is free from self-contradiction. Aquinas affirmed something similar in the Contra Gentiles:

And, contrary to the Point made by the first argument, it does not follow immediately that, as soon as we know the meaning of the name God, the existence of God is known. It does not follow first because it is not known to all, even including those who admit that God exists, that God is that than which a greater cannot be thought. After all, many ancients said that this world itself was God. Furthermore, no such inference can be drawn from the interpretations of the name God to be found in Damascene [De fide orthodoxa I, 9]. What is more, granted that everyone should understand by the name God something than which a greater cannot be thought, it will still not be necessary that there exist in reality something than which a greater cannot be thought. For a thing and the definition of a name are posited in the same way. Now, from the fact that that which is indicated by the name God is conceived by the mind, it does not follow that God exists save only in the intellect. Hence, that than which a greater cannot be thought will likewise not have to exist save only in the intellect. From this it does not follow that there exists in reality something than which a greater cannot be thought. No difficulty, consequently, befalls anyone who posits that God does not exist. For that something greater can be thought than anything given in reality or in the intellect is a difficulty only to him who admits that there is something than which a greater cannot be thought in reality.

Aquinas, like I said above, picks up on the fact that if Anselm’s rules are followed, then it follows that his conclusions will be the conclusions reached. Here, however, Aquinas uses Anselm’s definition of God to show that the Fool is in no danger of self-contradiction:

In this passage Aquinas seems to be using Anslem’s characterization of God to express the atheist’s denial of God’s existence. If to be God is, or would be, to be something than which nothing greater can be conceived, then for God to fail to exist is for (5) to be true. And, even if Anselm is right and the Fool who claims of something in the understanding that it is both something than which nothing greater can be conceived and something that fails to exist in reality contradicts himself, still, a more clever fool, one who insists on using (5) to express his atheism, would not contradict himself. (p. 95, The Blackwell Guide to the Philosophy of Religion)

The similarity between Aquinas and the Russellian view that Matthews arrives at can be expressed thus: by accepting, but reformulating, Anselm’s characterization of God, it is possible for the Fool to affirm what Anselm wants him to affirm, but in a way that doesn’t reach the conclusion that Anselm wants the Fool to reach. Aquinas’s objection is a matter of stricter logic than the Russellian, since for Aquinas, it simply doesn’t follow from the understanding of the name ‘God’ that what exists in the intellect necessarily has to exist in reality. The Russellian objection is a linguistic objection that reaches the same conclusion as Aquinas – no self-contradiction in sight – in a more precise way (Matthews calls Aquinas’s objection a ‘prefiguring’ of the Russellian objection). Both objections use Anslem’s definition against Anselm, which perhaps shows a deficiency in Anselm’s definition.