"The starting points of cosmological arguments are evident facets of experience. There is no doubt about the truth of statements that report that they hold. It seems to me equally evident that no argument from any of such starting points to the existence of God is deductively valid. For, if an argument from, for example, the existence of a complex physical universe to the existence of God were deductively valid, then it would be incoherent to assert that a complex physical universe exists and God does not exist. There would be a hidden contradiction buried in such co-assertions. Now, the only way to prove a proposition to be incoherent is to deduce from it an obviously incoherent proposition (for example, a self-contradictory proposition),but, notoriously, attempts to derive obviously incoherent propositions from such co-assertions have failed through the commission of some elementary logical error. Furthermore, it seems easy enough to spell out in an obviously coherent way one way in which such a co-assertion would be true. There would be a complex physical universe and no God, if there had always been matter rearranging itself in various combinations, and the only persons had been embodied persons; if there never was a person who knew everything, or could do everything, etc. Atheism does seem to be a supposition consistent with the existence of a complex physical universe, such as our universe. Of course things may not be as they seem, but, in the absence of any worthwhile argument to the contrary known to me, I shall assume that the non-existence of God is logically compatible with the existence of the universe, and so that the cosmological argument is not a valid, and so not a good, deductive argument. Our primary concern is however to investigate whether it is a good C-inductive or P-inductive argument, and just how much force it has."