More on atheism and certainty.

Last week, there was a brief but verbose exchange in the comments here about what atheism is, and whether it’s a rational position, or one of faith. It turns out that Scott Adams has been blogging about something similar recently.

His central argument is basically Pascal’s Wager, which he accepts and defends (badly) in a later post. I’m not really interested in going over the arguments for and against Pascal’s wager here - if you’re interested, Google the term - but I do want to take issue with this ridiculous and pervasive idea that somehow atheism requires an impossible degree of certainty.

It’s true that we can never be 100% certain about anything, but why should that be any more relevant when you say “I don’t believe in God” than when you say “I believe the sun will rise tomorrow morning,” or “I do believe in God”? Scott sarcastically asks if there are any atheists out there, but he might as well ask if there are any Christians, or any fans of Star Wars (I mean, I _think_ I like it, but I can’t be sure, right?) Hell, he might as well ask if there’s anyone out there who knows their own name. Am I 100% certain that there’s no God? No, of course I’m not, but I’m as close to certain of that as I am of anything. I’m not 100% certain that I’m not going to float off my chair and hit my head on the ceiling in the next half an hour either, but no-one calls my belief in gravity irrational or a “faith-position”.

So yes, if you define “agnostic” as “having any uncertainty on the subject of supernatural causes” or something similar, then Scott’s dead right, it’s the only intellectually defensible position, but it’s also tautological and therefore meaningless, since by Scott’s own argument, it’s not just the only intellectually defensible position; it’s the only position. That makes it a fundamentally useless definition; it’s like insisting that people answer every question by saying “I don’t know,” and prefix every statement with “it might be possible that…” Technically it’s true, but it doesn’t help anyone in practice, and in fact it just muddies the waters because it makes it impossible to tell where in that range of certainties someone is. Do they really not know, or are they making allowance for that 0.000001% margin of human error? The same is true of the word agnostic; if we extend it’s definition, as Scott apparently wants to, to include everyone who might be wrong in their beliefs on supernatural causes, then we include everyone, and the word loses all meaning - what good is an adjective that applies to everyone? Similarly, defining atheism so narrowly as to exclude everyone is of no value at all; there are certainly people who don’t believe in the supernatural, even while acknowledging that they might be wrong, and even if you insist that technically they’re agnostic, you’re going to need some way of referring to them that differentiates them from the agnostics who believe Christ died on a cross for their sins, and the ones that believe Zeus rules Earth from atop Olympus. Why not take their belief at face value and use the word atheist to refer to them? I think you’ll find that’s how we refer to ourselves.

Oh, and what’s the difference between a weak atheist and an agnostic? That’s easy; a weak-atheist doesn’t believe in supernatural causes, because the evidence does not support such a belief, while acknowledging some uncertainty insofar as more and contradictory evidence may come to light. An agnostic, on the other hand, holds uncertainty as regards supernatural causes to be key to his beliefs, either because he believes there are fundamental limits to our ability to know of such things, or because he believes the evidence is currently inconclusive.