Swinburne regards The Existence of God as his central book. With it, he seeks to turn the tide of sceptical arguments against belief in God that have held sway since Hume and Kant. He adapts his approach to modern science in two respects: by providing inductive arguments for the existence of God, and by arguing for the probability(no more) of God's existence. Indeed he thinks that there are no good deductive arguments for the existence of God (p. 330), and that the Five Ways are one of Aquinas's ‘least successful pieces of philosophy’(p. 136), forgetting that these arguments stood in a long tradition of Greek, Jewish and Arab philosophy preceding Aquinas. Swinburne dismisses the teleological argument of Aquinas quite briefly, without giving any good reasons (p. 155). In this new edition, Swinburne has especially revised the chapters on the teleological argument and consciousness. Let us look at these two areas:

Swinburne's starting point is anthropic: human beings exist with free will. He argues like this: If there is a God, it is more probable that he would make such creatures (p. 113). There are such creatures. So it is probable that God exists. Swinburne first thinks what God must be like (a man‐made image) and then finds that this world is just the sort of world you would expect such a God to produce. Free human agents require a physical universe to provide the conditions for exercising their freedom within limits (we cannot all have unlimited freedom). He does not consider angels, who are free agents but do not have physical limitations. He thinks that the chance that God would create a physical universe is quite high, in fact about half (p. 151).

Swinburne appeals to the argument of simplicity: it is simpler to suppose the existence of God than of many universes in order to make the existence of this universe probable (pp. 165, 185). In Swinburne's view, God is ‘one additional entity’ to the universe. Indeed he thinks that there is no ‘absolute explanation’ of the universe (p. 148), just a more probable one. It is more probable that God explains the universe as it is with its laws. Nowhere does Swinburne explicitly make the point that laws presuppose a mind. He misses the point of Aquinas's argument, that there could not be a universe of any kind without God. Throughout he talks of ‘explanations’ rather than causes.

Swinburne's arguments from the human consciousness are not conclusive, because he does not have a sufficiently strict criterion of the immaterial. For Swinburne, mental events are immaterial because they are beyond scientific explanation. But images are purely mental events (p. 195), although we get images through the senses, and brain events cause beliefs. The arguments of John Haldane in Atheism and Theism(with J C Smart) are a much more sustained attempt at showing that human thought does not have a material explanation. Not everyone will like Swinburne's out‐and‐out dualism: we are ‘pure mental substances’(p. 194). Swinburne discusses an argument from providence, which requires some evils. He then tries to decide whether the amount of suffering in the universe can be justified. He finds that, on balance, the evils are not so great that they make the existence of God improbable. Surely, one of the lessons to be learnt from the Book of Job is that we cannot go in for this kind of weighing up of the evils in the world against good; it is beyond our capacity.

It comes as a surprise that Swinburne thinks the strongest argument for the existence of God is religious experience (chapter 13). This is rather the least certain argument, for it depends on subjective judgement and the reliability of other people's reports of their own experience. It is particularly open to objections from the natural scientists. Human religious experience is anyway too varied: one first has to agree on a common set of principles for assessing all its varieties.

Swinburne concludes that, on balance, theism is more probable than improbable. One may ask how much is achieved by arguing only that God probably exists.