I’ve been thumbing through Richard Dawkins’ slightly curmudgeonly but generally interesting new book The God Delusion, which doesn’t add a great deal that’s new to the debate over theism (is there anything new to add?) but is, at any rate, a nice roundup of the relevant arguments, even if the tone makes me skeptical of its likelihood to win many converts. I did get to thinking, though, about Dawkins’ argument against one putatively circular sort of justification for faith in one’s favorite religious text.

The idea here is that many people come to think the Bible (Koran, Upanishads, whatever) is so full of moral wisdom that there must be something to it after all, and moreover that we need this revelatory source of moral guidance. The problem, as Dawkins points out, is that either your concept of morality is just directly and uncritically Biblically derived, in which case you’re not offering any good external reason for belief, or else you’re talking about moral principles we can recognize as valid through our own independent reflections, in which case you’re just concealing the fact that your own individual moral judgment is actually doing all the heavy lifting. This is why I’m always puzzled when people fret that we won’t be moral without religion: If you can recognize that certain kinds of amorality would be bad antecedent to belief—which you’ve got to if this is supposed to be an argument for belief—then what work is religion actually doing?

That said, I think I can imagine a kind of case where this sort of argument might work after all. Let’s say we found and decoded some tome of mathematics and physics written by a super-advanced alien civilization. (Or, if you prefer, God.) Some of it consists of theorems and formulas we’ve developed ourselves—though maybe in some cases the proofs look a little weird. Some of it is new theorems with ingenious proofs we can verify as valid once we’ve seen them, though we hadn’t gotten there yet ourselves. Further on, we get to ideas where the method of reasoning is (so far) incomprehensible to us, but which produce formulas we can verify seem to be correct. And finally, we hit a series of claims about the universe, connected to the others as part of a larger theory we can’t fully comprehend, such that we can’t directly verify them, nor can we really figure out the proofs. In a case like that, it seems like it would be reasonable enough to suppose that the stuff in the last couple categories has a high probability of being true, though beyond our ability to rationally evaluate and confirm, by dint of its connection to a number of other truths we can evaluate and confirm.

Of course, this just points back to another rhetorical question Dawkins offers: If God wants to convince us of the veracity of this or that text, why not pepper it with somewhat detailed truths about either physics or mathematics that we can only later independently confirm? (Though this does point up another problem with evidence for the existence of God: Since we ourselves are finite and fallible beings, the best evidence could only establish the existence of another extrmely powerful and knowledgable being.) Presumably, because God has a rather strange pattern of preferences that involves an intense concern with whether his creations believe in his existence coupled with the desire that this belief not be based on anything resembling strong evidence.