Reflections on atheism and the "Burden of Proof Principle"

Just a quick post before I leave for classes today:

In this essay, I give some brief reflections on the oft-repeated epistemic claim that I will call the "Burden of Proof Principle". To be honest, epistemology really isn't my field, and I have yet to read a professional philosophical article that explicitly mentions the Burden of Proof Principle. However, references to it occur quite frequently during informal discussions of religion.

Anyway, to begin...

Here's a common argument against strong atheism (i.e. the belief that God doesn't exist):

The Unprovability Argument: Strong atheism is an untenable position, because no one can prove that God doesn't exist. Therefore, if you don't see evidence for God's existence, then you should be an agnostic (i.e. one who lacks a belief in God without believing that God doesn't exist), not a strong atheist.

A usual reply to the Unprovability Argument goes something like this:

We don't reason that way about other hypothetical entities. I can't prove that fairies don't exist, but that doesn't mean that I should be agnostic with regard to fairies. If I see no good evidence for fairies, then I shouldn't just lack a belief in fairies; rather, I should believe that fairies don't exist!

I think there's something right about this reply, but we must be careful when spelling out the principle behind it. The principle is often spelled out as follows:

The Burden of Proof Principle (BPP): If X claims that Y exists, then X bears the burden of proof: if X can't prove (or give good evidence) that Y exists, then others are justified in believing that Y doesn't exist.

How is the BPP supposed to refute the Unprovability Argument? Presumably, the reasoning is as follows: the theist claims that God exists; therefore, by the BPP, the theist bears the burden of proof; that is, if the theist can't prove (or give good evidence) that God exists, then others are justified in believing that God doesn't exist.

Perhaps I'm interpreting the BPP a bit too literally, but I've always found it unpersuasive. Framing philosophical issues in terms of "burdens of proof" strikes me as singularly inappropriate. Look, if X accuses me of committing a crime, then X "bears the burden of proof": if X can't prove that I committed the crime, then the courts are required to regard me as innocent. But if X makes a philosophical claim, then it would be the height of fatuousness for me to say that X "bears the burden of proof", such that I'm justified in regarding X's claim as false until X can prove that it's true. The universe doesn't care whether X can prove his claim: if the claim is true, then it's true regardless of whether X has succeeded in proving it. From the fact that X can't prove P, it doesn't follow that I'm justified in rejecting P.

On the other hand, I accept, as a general epistemic principle, Occam's razor, which demands that I not multiply entities unnecessarily. If I myself currently don't know of any good evidence for Y's existence, then, all else being equal, I ought to regard Y as nonexistent. Note that this is utterly independent of whether someone else has failed to provide proof of Y's existence. My justification for rejecting Y's existence has to do with my own lack of evidence for Y's existence, not someone else's failure to satisfy some "burden of proof".