The epistemology of Immanuel Kant (1724-1804) prevented him from deriving a proof of God's existence from an a priori definition of his nature. For Kant, knowledge is based on experience and what can be induced or deduced from it. Knowledge of God and of "necessary existence" is beyond our experience and thus it would be impossible to develop a definition strong enough to be the subject of analysis. This criticism is similar to that put forward by Aquinas. Kant allows for the objection to this criticism, voiced by Scotus (see previous page), by arguing that certainly experience indicates regularity, order, laws but that we cannot prove the existence of the God, the supreme good, which most probably explains them - we can only postulate the existence of such a being. Language is based on experience and can be used to explore it and indeed to arrive at evidence to support the postulate of God - but it cannot be used to provide proof in itself.
Kant first distinguished between synthetic and analytic statements - the first referring to existence and the second making a claim about the relationship between concepts.
1. Synthetic: this ball is red
2. Analytic: unmarried men are bachelors
Kant observed that Descartes moved from the analytic claims "perfection entails existence" and "supreme perfection entails necessary existence" to synthetic claims, "perfect things exist" and "the supremely perfect being exists". Though Kant accepts that these analytic statements are not self-contradictory and thus possibly real by Leibniz's definition of reality, he does not accept Leibniz's definition of reality.
For Kant reality is not just about logical plausibility, it is what we experience. He notes that making a synthetic claim is a different matter from making an analytic one. There are no instances of perfect things or supremely perfect beings to refer to so we cannot conclude that they exist. Kant criticised the argument's concept of "necessary existence". He noted that to argue that there is a necessary being is the same as to say that to deny its existence is self-contradictory, remember, saying "God, who must exist, does not exist" makes you a fool according to Anselm. If this was the case then it would mean that at least one analytic statement must be synthetically true - God exists. But Kant argues that it is logically impossible for any synthetic, existential proposition to be logically, analytically necessary. For Kant, every synthetic, existential proposition must contain the possibility of it being otherwise - this ball may or may not be red, may or may not exist. The concept of necessary existence is thus a contradiction, a "miserable tautology".
Bertrand Russell (1872-1970) made a similar criticism of the ontological argument in his article On denoting (1905). He distinguished between two types of propositions - predicative (those which add to the concept) and existential (those which claim a reference between a concept and a state of affairs) Russell argued that everyday use of language makes it possible to talk about non-existent things with apparent meaning. Although we should say "are there things which match the description of cows?" before talking about cows, in fact we just start talking about cows. He used the example of "the present King of France" - as soon as I start talking about this meaningless entity, even if to state that it does not exist, I imply that the concept is a valid one. For Russell, statements can only be true or false if they refer to a meaningful concept. There is no way that I can affirm the existence of any instance of "necessary existence" before embarking on an ontological argument which seeks to conclude the God's necessary existence is necessary - therefore the whole enterprise is meaningless.
Existence as perfection
Kant also criticised the assumption, made by Descartes, Leibniz and Anselm for that matter, that existence is a perfection – i.e. they had all claimed that it is better to exist than not to exist and thus that the most perfect being would have to exist. Kant rejected this because he argued that existence is a necessary ground for any other perfection to be meaningful rather than just another in a list of perfections. Imagine a job interview. The panel look through the CVs of two well qualified candidates - it turns out that candidate A exists and candidate B is made up. There is no real contest between the two. Candidate B never really had any of the qualities the CV claimed and was nonsense all along. Candidate A's existence is not just another qualification which tips the balance in her favour! Kant makes the point that a statement about existence is not the same as a statement about other perfections of an object thus it is inappropriate to argue that "existence is a perfection, therefore the supremely perfect being must exist." Note that Kant's concept of existence is rooted in experience - it appeals to common sense but does not engage with the rationalist argument on equal terms.
Kant criticised Descartes and Leibniz for seeing existence as a predicate of God. Descartes and Leibniz defined God's perfection in terms of a list of attributes - just as a perfect flower would be colourful, scented, have petals and leaves etc, the supremely perfect being would be all-powerful, all-knowing, have necessary existence. He added existence as another attribute of the object but Kant argued that this is flawed. Existence adds nothing to the concept of the flower; it merely refers to whether there are instances of the concept. Existence is not a predicate; to use it as one is simply bad grammar.
Predicate and subject
Kant made a specific criticism of Leibniz, saying, "If I reject the predicate while retaining the subject, contradiction results; and I therefore say that the former belongs necessarily to the latter. But if we reject subject and predicate alike, there is no contradiction; for nothing is then left that can be contradicted ... the same holds true of the concept of an absolutely necessary being. If its existence is rejected, we reject the thing itself with all its predicates; and no question of contradiction can then arise." He did not accept Leibniz's principle of non-contradiction; although influenced by Leibniz's work as a young man, Kant later came to refer to Leibniz as "dogmatic" and claimed that reading Hume "awoke me from my dogmatic slumbers". For Kant, ultimate reality was not accessed by pure reason and truth was not demonstrated by the coherence of its conceptual parts. God's existence could not be proven by showing that the concept of God as a supremely perfect, necessarily-existing being is non-contradictory.
Real and imaginary
What is the difference between a real hundred pounds and an imaginary hundred pounds? Kant observed that the concept in question is identical whether the notes are in my pocket or not. There is a difference between claiming something about a concept and claiming that a concept exists. He concluded: "The attempt to establish the existence of a supreme being by the famous ontological argument of Descartes is therefore merely so much labour and effort lost; we can no more extend our stock of [theoretical] insight by mere ideas than a merchant can better his position by adding a few noughts to his cash account."
Essentially then, for Kant:
• Whatever adds nothing to the concept of an essence is not part of that essence,
• Existence adds nothing to the concept of an essence
• Existence is not part of the essence of a thing, it is not a perfection.
Experience and semantics
Whereas Descartes argued that the conceptual relationship between perfection and existence within the idea of God rendered his necessary existence de dicto necessary, Kant observed that such analytic truths are simply semantic. Certainly the relationship between "unmarried man" and "bachelor" is undeniable but the sort of undeniable truth that it is extends only to how language should be used. For Kant, synthetic claims, claims about existence such as "this ball is red", must refer to a state of affairs which can be experienced if they are to be meaningful. It is the experience of seeing the ball that carries the meaning, not the words we use to describe that experience. If I look at the ball and say "that ball is red" and Francois, having the same experience says "ceci, c'est rouge!" we know we mean the same thing because we are experiencing the same thing. Analysing statements about things we cannot experience does not really get us anywhere. For Kant all synthetic propositions can either be true or false in relation to an actual state of affairs, a state of affairs within our experience. There can be no analytic demonstrations concerning the existence of anything.