So, despite the fact that Loftus seems to consider any criticism of his Outsider Test to be out of bounds if you don’t have a better solution to the problem of religious diversity, I’m going to start criticizing it without proposing an alternative. I gave the reasons why in my first post on his book. In looking at the test itself, I’m going to essentially rely on the Outsider Test being the list of four main attributes of informed skepticism that he lists on page 21 of the paperback. I’m also going to ignore the last two since while there may be some issues with them, they’re actually fairly reasonable. So, I’m going to start, in this post, with the second one, which is as follows:
… it adopts the methodological-naturalist viewpoint by which one assumes there is a natural explanation for the origins of a given religion, its holy books, and it’s [sic] extraordinary claims of miracles.
There are two problems with this, both of which are so serious that if the Outsider Test contains them then there is no way that it can be considered a fair and proper test that merely, as Loftus insists, gets religious people to treat their religion the way they treat others. Let’s start with how he explicitly does that.
Religious people, in general, will not be either metaphysical naturalists nor methodological naturalists. Since God is generally considered to be a supernatural being, if they were either of those they wouldn’t be religious anymore. Thus, it is not likely that they look at other religions and reject them because they posit supernatural things. Additionally, there aren’t that many religions — at least in the Judeo-Christian line — that posit that the only supernatural events or entities are God. The Bible, for example, explicitly recognizes other forms of magic (see, for example, the story where Moses competes with the Pharaoh’s magicians and bests them), and many religions also have notions of witchcraft which explicitly references supernatural magic that doesn’t come from God (and is, in their minds, evil). So there is no reason for almost any religion to assume that there are natural explanations for the purported supernatural claims of competing religions. All they have to say is that those supernatural events do not come from God, or anything that is worth holding as either a god, or as the all-powerful God that they worship. So right here he is insisting on adopting a viewpoint that no religious person need adopt nor does adopt when dealing with other religions. At this point, this clearly becomes a normative claim: we ought to adopt a methodological naturalist stance in determine which religion is the right one.
But this clashes with his own definition of religion. He insists in a number of places that what makes religion religion is its link to the supernatural. So every religion that is properly called a religion makes supernatural claims. What Loftus want to do here, then, is start with the assumption not merely that the religion has the burden of proof — that’s his first point — but that the religion is wrong. So, in order to fairly assess whether or not any religion is true, let’s first assume it to be false and then, maybe, if it provides overwhelming evidence that it is true, then maybe we can accept that it is true. He’s not starting with the assumption that it is false and showing that that leads to a contradiction, because he’s not doing that directly for a specific religion. Instead, he’s building in a structure that forces us to assume that the central defining feature of any religion is actually expressing something that is not true. I can’t see how you could possibly have a fair test that doesn’t merely ask you to consider that it might be false, but also to assume it false.
And the thing is that there should be no need for such an assumption. The closest you can get to a reasonable challenge here is to ask people to consider that there might be a natural explanation that makes as much if not more sense than the purported supernatural one. Since most atheists think that there definitely is, you could at least get to doubt that the religious claim is true if not argue that it is most likely that it isn’t true, using normal methods. If the naturalistic explanation is the epistemically superior or more likely one, argue that the most likely explanation is natural. If it’s just roughly equal, try to use Occam’s Razor to slice the religious claims away. The only time this assumption comes into play is if you can come up with a naturalistic explanation for the supernatural claim but it isn’t objectively as good or better than the supernatural claim. But, at that point, we probably have fairly good reason to choose the supernatural claim over the natural one if we’re doing anything even remotely like proper reasoning and epistemology. So this assumption unfairly stacks the deck against religion and ought to be unnecessary if Loftus really does have as good reasons as he thinks he has against religion.
There is no saving this assumption. Even weakening it to “Consider naturalistic explanations” only makes it redundant because considerations of sufficient evidence should already bring that into play. It only works as a way to privilege naturalistic explanations over supernatural ones … and Loftus should not need that in order to make his case.