The Ontological Argument was first proposed by St Anselm in 1078. In the Proslogion he tried to demonstrate the existence of God from reason alone, first by defining God as “that than which nothing greater can be conceived of”… as existence “in re” rather than only “in intellectu” makes something greater, God must therefore exist, and then by claiming that necessary existence is greater than contingent existence and so must be a property of God. The Ontological Argument soon attracted criticism, first from Gaunilo of Marmoutiers whose “on behalf of the fool” suggested that it seems like a joke to suggest that something must exist just because it is perfect, and then from Aquinas, who pointed out that “because we do not know the nature of God, His existence is not self-evident to us.” Nevertheless, while most people are sceptical of Anselm’s argument, as Bertrand Russell pointed out “it is easier to feel convinced that it must be fallacious than it is to find out precisely where the fallacy lies.” As it happens, the argument – while containing some logical fallacies – does not depend on these so that they cannot be overcome. It is a valid argument… the question of its soundness depends on one’s worldview.

Firstly, it could be said that both versions of Anselm’s argument depend upon the logical fallacy of bare assertion, as in they assert that “God is that than which nothing greater can be conceived of” without proper argument. Nevertheless, all a priori arguments start with a priori premises, definitions which depend on a priori knowledge (reason alone) and often cannot be argued for using evidence. For example, if I argued that as bachelors are unmarried men and Simon is unmarried, that Simon must be a bachelor, it is not reasonable for you to demand that I demonstrate that bachelors are unmarried men from observations before proceeding. Similarly, if I argued that 2+2 = 4, I must begin with a priori knowledge of the numbers 2 and 4 and the concept of addition. It is not reasonable to ask for an argument that 2=2 and 4=4 before accepting that 2+2 = 4… because any sane person knows what 2 and 4 refer to and what the concept of addition entails. Anselm pointed out that anyone who claims that God is not “that than which nothing greater can be conceived” must be a fool. How can anyone think that there could be something greater than God… if they do, then they have fundamentally misunderstood the concept of God. In this way, while Anselm does assert his premises, he is justified in doing so and this “logical fallacy” is not a serious criticism of the ontological argument. Similarly, Anselm’s argument could be accused of begging the question, meaning that his conclusion of God’s necessary existence is contained within the premises. Yet surely this is the whole point of a deductive argument! Nobody criticises the argument 2+2 = 4 because the concept of 4 contains the concept of 2 twice. What Anselm is trying to do is to clarify that our concept of God includes His necessary existence, so it is unreasonable to expect Anselm’s conclusion not to contain his premises. In both these ways, Anselm’s Ontological Argument does not depend on any logical fallacies that cannot be overcome.

Secondly, it could be said that Anselm’s argument is guilty of being ad hominem and of appealing to authority. Anselm certainly attacks atheists as fools and quotes from Psalm 14:1 as part of his argument. Nevertheless, neither Anselm’s colourful language nor his Biblical allusion are part of his reasoning, so his argument could be stated without either quite easily. More seriously, addition, Anselm could be accused of asking a loaded question of atheists. Is God “that than which nothing greater can be conceived of”? The atheist is railroaded into answering yes, in which case they have already admitted the conclusion, or no, in which case they are a fool… Yet as Bertrand Russell pointed out, asking a question about the properties of a non-existent object is meaningless. If I asked you “is the present King of France bald?” I feel bound to give a yes or no answer, when in fact I can’t give either because there is no present King of France. Similarly, in asking atheists to answer the question “is God that than which nothing greater can be conceived of”, Anselm could be bamboozling the atheist into answering yes or no, when either option would mean that they cede their point. This seems a lot like the either-or fallacy as well, with Anselm excluding options other than yes or no. However, it is clear that everybody, atheists included, have a concept of God as “that than which nothing greater can be conceived of” in their minds, meaning that He undoubtedly exists “in intellectu” in a way that the present King of France, perfect islands etc. do not. As Anselm pointed out in his “Responsio” to Gaunilo, there is a difference between islands and God, in that islands can only exist contingently whereas God exists, if he exists at all, necessarily. This means that Russell’s point about the present King of France is not relevant to the Ontological Argument, as when Anselm asks “is God greater than that which can be conceived of”, he is justified in assuming that the knowledge of God exists a priori in intellectu, when the knowledge of contingent things – whether Kings or Islands – can only be a posteriori and synthetic. Kant is right to say “Whatever, therefore, our concept of an object may contain, we must always step outside it, in order to attribute to it existence…” when it comes to any and all contingently existing things, but as Anselm pointed out, God is not like other things, so the ontological argument could only ever apply to God. It seems that Anselm’s argument survives the accusation of depending on these logical fallacies as well.

On the other hand, Kant argued that Anselm creates the whole category of “necessary existence” to get around Gaunilo’s obvious criticism that what applies to perfect Gods should apply equally to perfect islands, unicorns and such. In this way, Anselm’s argument would depend on special pleading. Kant argued that existence involves having the potential to be and not be, so necessary existence is a contradictory concept like a square circle and so impossible. He reasoned that because existence must include having the potential to be and not be, existence cannot be used as an essential predicate of anything. Later in 1948 JN Findlay went further, claiming that “it was indeed an ill day for Anselm when he hit upon his famous proof. For on that day he not only laid bare something that is of the essence of an adequate religious object, but also something that entails its necessary non-existence.” For Findlay, if there are three options – God is impossible, God may or may not exist or God necessarily exists, then the Ontological Argument serves to show that God must be impossible and necessarily not exist, because if God may or may not exist He wouldn’t be God and necessary existence is impossible. Nevertheless, Hartshorne rejected this, arguing that if Findlay says that necessary existence is impossible, so must be necessary non-existence. Further, Kant’s definition of existence applies to contingent existence only, as does his claim that existence cannot be an essential predicate, necessary existence does not include the potential to exist and not exist by definition and so it could be an essential predicate of God. For Hartshorne, there is nothing impossible about necessary existence. We can conceive of God necessarily existing in much the same way as we can conceive of a three-sided triangle, when we cannot conceive of a square circle. As Alvin Plantinga and William Lane Craig have pointed out, if God’s necessary existence is even possible – in the way that a unicorn or a Gruffalo is possible but a five-sided triangle is not – then God exists necessarily in every possible world. Of course, Kant would reject this, pointing out that we have no experience of “necessary existence”, making it a “cupola of judgement”, being outside our possible existence and entirely speculative. Nevertheless, although Kant’s criticisms are coherent with and conclusively destroy the Ontological Argument within his worldview, Kant’s worldview has been criticised by Quine for depending on dogmas and is not shared by everybody. As Norman Malcolm pointed out, it is clear that “necessary existence” is possible and not contradictory within some “forms of life” and their language games. This suggests that at least within these forms of life, necessary existence is not an impossible or invented category of existence, so Anselm’s argument does not depend on special pleading.

Further, other critics suggest that Anselm’s argument takes advantage of the useful ambiguity in the word “necessary”, thus depending on the fallacy of equivocation. The word necessary can mean de re necessary, in the sense it is used in Aquinas’ third way, meaning that God is self-explaining, doesn’t depend on anything, fully actual. The word necessary can also mean de dicto necessary, in the sense that it means that God’s existence is part of the concept of God so God’s non-existence cannot be asserted without contradiction. For example, saying “this triangle has five angles” would be to assert a contradiction, because the word tri-angle necessarily and by definition entails having only three angles. Could it be that the word “necessary” means two different things and that Anselm shifts from one meaning to the other to bamboozle us with a what Schopenhauer called a “sleight of hand trick?” While the concept of necessary existence is confusing and while the word “necessary” is used in both senses in the argument, the argument does not depend on ambiguity or equivocation because there is what Hegel called a “unity of thought and of existence in the infinite.” While there are two meanings to the word “necessary” these are related in that de dicto necessity refers to concepts and the rules of logic that originate in and depend on God’s de re necessity. Of course, Aquinas’ criticism of the attempt to demonstrate God’s existence from reason alone is apt here. Given that most – if not all – people struggle to “conceive of” God’s nature, how can we analyse that nature to find necessary existence – another almost inconceivable idea – within it? Aquinas as right that while God’s existence may be self-evident, it is not self-evident to us, and therefore that it is better to demonstrate His existence from what is known, observations. Nevertheless, the question asks whether the Ontological Argument depends on logical fallacies that cannot be overcome and the answer to that must be that it does not. There is no equivocation or fundamental ambiguity on which the argument depends.

In conclusion, Russell was right to say that “it is easier to feel convinced that it must be fallacious than it is to find out precisely where the fallacy lies.” The Ontological Argument does not depend on logical fallacies that cannot be overcome. It is a valid argument, but depends for its soundness on the particular worldview or form of life within which it is advanced.