Before moving to consider the various arguments for the existence of God, it is worth asking the preliminary question How likely is it that God exists? Our preconceptions on this issue are likely to colour our assessment of whatever evidence for (or against) God’s existence we encounter. It is therefore worthwhile to attempt to establish the intrinsic probability of theism, the a priori probability that God exists.

If we begin with the thought that God’s existence is highly unlikely, then it is going to take very strong evidence to persuade us that he does indeed exist. Whatever positive evidence for God’s existence we encounter, if we begin with a presumption of atheism then we will expect that evidence to be flawed. We may, as a result, view purported theistic proofs with greater suspicion than we otherwise would.

If, on the other hand, we begin our inquiry with an intellectual openness to God’s existence, then we may find persuasive arguments that others would not. Inconclusive evidence may be deemed acceptable on the ground that it confirms a suspicion that we already had. The issue of the intrinsic probability of theism will thus have an effect on the way that we approach any argument on either side of the debate concerning God’s existence. It is rational to take the probability of God’s existence into account when considering such arguments.

The Improbability of God

It is tempting to think that God’s existence is about as unlikely as anything could be. God, if he exists, is infinite in his attributes; in power, knowledge, and love—in his whole being—God is unlimited. Ockham’s razor, then, which tells us that where either of two explanations will do we should always prefer the simpler explanation, recommends that wherever possible we should avoid postulating the existence of God to explain evidence. If there are two explanations of a set of evidence, one invoking God and the other not invoking God, then the explanation that doesn’t invoke God will always be the more economical of the two; it is more economical to postulate any number of finite beings than it is to postulate one infinite being. The hypothesis that God exists, then, seems to be as intrinsically unlikely as it is possible for a hypothesis to be. Prejudice against theism, it seems, is justified.

It might even be thought that the existence of God goes beyond mere improbability, that it is impossible. Certain of the tradition doctrines concerning God’s nature appear to be self-contradictory, while others appear to contradict each other; several of the arguments for atheism seek to exploit this appearance. If this appearance is to be trusted, then God cannot exist—logical contradictions are not just unlikely to be true; they cannot be true—and we can be confidant that any purported theistic proof contains an error even before we examine it.

The Probability of God

Let us suppose, for the sake of argument, that the alleged contradictions in God’s nature can be resolved. If this is correct, and God’s existence is possible, then the theist can offer a counter-argument to case for the improbability of God’s existence set out above. This counter-argument is offered by Richard Swinburne in The Existence of God [p94].

Swinburne observes that it is simpler to postulate an unlimited force than a limited force. If one postulates a limited force then one is postulating two things, the force and whatever constrains it. If one postulates an unlimited force, then one is only postulating one thing, the force; there is, by definition, nothing that constrains an infinite force.

For this reason, scientists constructing theories will, unless there is good reason not to, prefer to use zeroes or infinities in those theories. The speed of light, for instance, was assumed to be infinite until experimental data disconfirmed this. Scientists recognise that an infinite force is intrinsically more probable than any great but finite force.

This methodology, Swinburne suggests, can be generalised; an infinite being, he urges, is the most probable kind of being. Ockham’s razor, if he is correct, far from implying that God’s existence is less likely than any other explanatory hypothesis, implies that it is more likely than any other explanatory hypothesis; the intrinsic probability of theism is relatively high.