My friend Todd Seavey, who recently launched his own blog, has a post up rehearsing some oldie-but-goodie atheist arguments, and while they’re all pretty familiar, something stuck me on this read-through about the phrasing, the specific assertion that belief without evidence (faith) is “irrational.” Maybe this sounds less than compelling to lots of people because, of course, all of us do this all the time, in ways that clearly aren’t irrational.

I don’t mean this in the sense of the mushy relativist everything-is-faith position that seems to be the last refuge of the desperate apologist. (I may actually die of embarrassment on behalf of the next person whose mouth disconnects from his cortex for long enough to assert that belief and unbelief require equivalent “leaps of faith.” ) I just mean that we routinely believe things on the basis of no better evidence than the say-so of other people we take to be authority figures. There are plenty of things I “know” just because I read it in a book that seemed authoritative, or because my parents, teachers, or other folks I regard as reliable and well-informed told me so.

Of course, it is irrational to believe literally everything you’re told, even by your parents. (Santa Claus never seemed believable, for instance.) Some claims are so implausible that they ought to set off the bullshit-detectors of anybody doing basic epistemic due diligence. For others, whether it’s “irrational” for a particular person to believe them will depend on what else they know, and how much thought they give the topic.

Take the old factoid about how people only use ten percent of their brains. (This is, of course, nonsense.) It’s repeated so much that it’s not really irrational for the average person to suppose it simply must be true. But what if someone knows a bit about evolution, and about what a resource-hog the brain is? This person, at least, should be a little skeptical about the likelihood that natural selection wouldn’t have pared down such a wasteful organ. Even then, though, it might not seem appropriate to call such a belief “irrational” in the colloquial sense—as opposed to by some unrealistic standard of perfect integration of all available information—if the person hasn’t given it much thought. My belief in the ten percent factoid is not, in this sense, necessarily “irrational” just because I once, years ago, learned just enough about evolutionary biology or anatomy that I could in principle reason my way to the conclusion that it was, at the least, dubious and needed investigating.

Part of the reason for this is that there’s an implicit evidentiary chain we construct in these cases: We don’t know all the evidence on which the claim is based, and neither, perhaps, does the person who told us, or the person who told her. But somewhere down this line of Chinese whispers, we suppose there are people who do have good reasons for accepting the proposition, and that our social error-filtering mechanisms are reasonably reliable about getting false claims out of circulation within a relatively short time in most cases. So it might often be perfectly rational to make your default the acceptance of widely-held beliefs—or at least beliefs widely held by some properly authoritative group—on the supposition that there’s more substantive evidence further down the chain.

This becomes irrational only if we don’t withdraw that provisional belief when, upon investigation, it turns out to be tortoises all the way down—either because the purported evidence at the end of the chain doesn’t turn out to properly support the claim, or because the chain itself is suspect. Believing once we know (or reasonably ought to) that there’s no good evidence, or that the reasoning from the evidence to the claim is faulty, can be fairly characterized as irrational, though. So, somewhat counterintuitively, a belief of this sort may become more irrational the more thought you’ve given it.