Testing the supernatural

On Ophelia Benson's blog, Butterflies & Wheels, there was a very amusing discussion (two, actually) running about whether we can scientifically measure God or God’s interaction with the world.

Why can’t we scientifically measure God or God’s interaction with the world? One reason could be because god is not there. Another reason could be because god is especially hard to measure for some reason. If it’s the latter we just need better instruments. I think the reason De Dora is suggesting is that god is in principle incapable of being measured. But if that’s the case, De Dora needs to explain further – how he knows that, why it is the case, what it implies for claims about god, and similar.

I'd say that we can test some specific hypotheses that involves divine intervention, but not necessarily the existence of God (even if we ignore the boring deism). For example, we should in principle be able to test a hypothesis such as "by miracle, God heals episcopalians from cancer more often than he does the Greek orthodox", or "the flagellum could only have been designed by God". But there are other hypotheses that I would venture cannot be tested.

I can think right now of two ways in which God could interact in the natural, non-supernatural world without the possiblity of scientific testing: 1) Personal revelation. This cannot really be tested, because if we show that such events can be caused by some brain dysfunction, then we still cannot rule out that some events are caused by God. 2) If God is very powerful (to the point of being able to do anything), then could he perhaps not also cause miracles and interventions in the natural realm and make it look like he didn't? This seems to be akin to what Ken Miller is suggesting, when he says that God works through changing the probabilities of quantum interactions.

But we can come up with better explanations, of course. For example, we cannot really test if God created the flagellum, because even when we show how it could have evolved, we have not shown that it did evolve. Rather, an evolutionary explanation makes it plausible that God did not have a hand in the matter. This is the best science can do about historical inferences, and we should be quite happy with that, because the same kind of inference is what makes science work in every other case, I think (I'm going to be a little cautious here - correct me if I'm wrong). The inference that the flagellum evolved doesn't have any practical use, as far as I know, but many other scientific discoveries based on the same method do. For example...