Believers tend to be confused about what constitutes evidence for the existence of God. The believer might say something like, “My mother had advanced-stage cancer and it spontaneously went into remission. How else could that have happened? It must have been God.”
The problem is that his mother’s cancer going into remission is not evidence for God any more than it’s evidence for any other explanation. It is an observation, and the God explanation is one hypothesis about why it happened. There may be other hypotheses. The fact that the believer can’t think of any doesn’t mean that there aren’t any. It’s a statement about his knowledge of the event, not the event itself.
Yes, there may very well be alternative hypotheses. Suppose that she had a type of cancer that could undergo a random mutation that shut off the cancerous cell line. Suppose there were a test available to determine if such a mutation had occurred. Suppose that test came back positive. We would have confirmed an alternative explanation.
Of course, the test may have come back negative. In that case, we would have ruled out one explanation, but there may be others. In each case, further testing, further experiments, further observations are necessary to confirm or disconfirm the hypothesis. As you can see, the initial observation is not sufficient in and of itself to confirm the God hypothesis, or any other hypothesis for that matter.
The God hypothesis, in order to be accepted, must be testable. It must be falsifiable. It must be amenable to confirmation or disconfirmation. Without evidence to establish a causal link between the observation (remission of cancer) and the explanation (God performed a miracle), the explanation is merely a non sequitur.
It could be used to explain anything:
My car won’t start. God sabotaged it.
I failed that test. God doesn’t like me.
I got that great job. God likes me.
Or maybe the car battery is dead. Maybe you didn’t study hard enough. Maybe you were the most qualified for the job.
All of these explanations are, at least in principle, testable. You can check the battery. You can change your study habits to see if you improve. You can, at least in principle, compare your job application with the others.
The God hypothesis, however, is not testable, so it is not a satisfactory explanation of any event. I should amend that statement, however. Earlier I said that the fact that a believer can’t think of an alternative explanation doesn’t mean that an alternative explanation doesn’t exist. Perhaps the fact that the nonbeliever can’t think of a test for God’s existence (or handiwork) doesn’t mean that such a test doesn’t exist.
This is true. The God hypothesis is not completely ruled out. We can’t assign a truth value (a probability of being true) of 0. No Bayesian rationalist would assign a probability of 0 or 1 to any belief, anyway. But we can refrain from accepting the hypothesis until a proper test is devised and proper evidence is collected.
We shouldn’t accept non sequiturs. We shouldn’t jump to conclusions. We should stick with our null hypothesis — we don’t know — until the weight of evidence compels us to accept an explanation.