SL: I followed some links on your recent thread to do some reading on this. I find myself very much in agreement with your approach. Loftus wants to restrict the test to only religious matters, but this seems completely arbitrary. Moral beliefs would surely fail this test too. Or would they (I have chapter 3 of the Abolition of Man in mind)? If they wouldn’t we have a first premise of the moral argument.
Like you I also think it’s very unclear what the test really is and that Loftus seems to have a pre-theoretic commitment to the outcome. Unfortunately for Loftus I think he’s on very dodgy ground historically and sociologically. I assume he thinks atheism survives such a test? But atheists make up a very narrow slice of the population both currently and historically. If Loftus had been born 500 years ago, 1000 years ago, 2000 years ago ... or in another culture etc, etc. Being such a narrow subset of the global-historical population it seems much more likely that these atheistic beliefs are a product of cultural oddities. Theistic views have flourished in many cultures in many epochs, and therefore don’t seem especially tied to cultural mores.
Now what I do think is that our normal idea of knowledge is that it should be something which “tracks” the truth; It should be in some way sensitive to the facts. (I think these ways of putting it are Nozik’s.) The outsider test doesn’t provide a reason to think belief is not sensitive to the facts ... the basis of our beliefs would be different / non-existent were the central tenets of Christianity false, and then we wouldn’t hold the belief. This is the line I take when I defend the AfD from the genetic fallacy. If God didn’t exist, we wouldn’t desire him, and wouldn’t believe in him. That is true and relevant even if we don’t reason from the desire to God, but the desire is merely an occasion for the belief. At least, if we accept some appropriate form of reliabilism/proper-function account.
A rational system of belief must be able to account for the existence of its adherents in a way that positively relates the truth and the belief. Similarly it must be able to account for the existence of its detractors without requiring such a positive relation to the content of their beliefs. The OTF brings attention to the possibility that this positive relationship may not exist. Fine. We can doubt that, but when we investigate the relevant evidence, it looks good to us ... what else are we being asked to do?
Only my first thoughts. If I have any more, I’ll let you know.