I posted this over at Vox Day's site and decided it was worth repeating here, as well, because I find that much of the problems in discussing evolution have to do with the failure of my critics to understand the nature of science. This is not a post about God's existence per se, just a discussion of why science can't say much about such claims. . . .
Evidence is not the truth, with a capital 'T', but simply observations reported by (we hope) competent observers. The more rigorous the observation, the more times said observation is replicated, the more confidence we have in that observation and so we tend to refer to such-and-such an observation as 'true.'
Do people sincerely report religious experiences? Indeed they appear to do so, so the fact that people report religious experiences is a phenomena that requires explanation, a phenomena that can be reasonably described as 'true.'
We reserve the right, however, to revise or reject such claims based on improved data, so these kinds of claims are never Truth with a capital 'T', the sort of eternal truths sought through philosophy or religion.
Because of this, scientists are not in the 'truth business.' We are in the model-building and model-testing business. When a model seems to do an especially good job of describing, explaining and predicting many phenomena we have the tendency to call such a model a 'theory', as opposed to a mere 'hypothesis.' But we never claim that theories are Truths.
Now, the problem from this standpoint with claims such as the existence of God is not that they may or may not be true, or that the existence of God is an eternal Truth, etc. The problem is that we can't ever move that hypothesis to the 'theory' stage because we can' test its ability to describe, explain or predict phenomena. The God Hypothesis that fails a particular test can always be 'resurrected' by amending it with claims about God's intent, God's nature or the invocation of the supernatural. It is not the possibility of the supernatural's existence that scandalizes the naturalist, but the fact that within science we have no procedure to test the evidential basis of supernatural claims.
We therefore exclude them without attempting to rule one way or the other on the claim. The best we can do is check out the testable consequences of such claims, whether or not the hypothesis produces the (natural) phenomena that is expected.
