I went back to your original comment, the nub of which I take to be this: you tolerate as self-consistent the belief of some groups of theists, on the grounds that their beliefs have no empirical consequences, and that is precisely what marks these beliefs as "faith".

I was responding to the idea that theists are involved in blatant double-think where they anticipate ways that their beliefs can be empirically refuted and find preemptive defenses. The idea of "separate magisteria" may have been one such defense, but it is the last: once they identify God as non-empirical they don't have to worry about CO2 detectors or flour or X-ray machines -- ever. I think that the continued insistence on thinking of God as a creature hidden in the garage that should leave some kind of empirical trace reveals the inferential distance between a world view which requires that beliefs meet empirical standards and one that does not.

You suggest that the "separate magisteria" is a pretense. This appears to be along the lines of what Eliezer is arguing as well; that it is a convenient 'get out of jail free' card. I think this is an interesting hypothesis -- I don't object to it, since it makes some sense.

We base our views about religion on our personal experiences with it. I feel like I encounter people with views much more reasonable than the ones described here fairly often. I thought that 'separate magesteria' described their thinking pretty well, since religion doesn't effect their pragmatic, day-to-day decisions. (Moral/ethical behavior is a big exception of course.) I've encountered people who insist that prayers have the power to change events, but I don't think this is a reasonable view.

I thought people mostly prayed to focus intentions and unload anxiety. Some data on what people actually believe would be extremely useful.