March 27, 2004
Statistics prove that God is deadPosted by shonk at 02:00 AM in Science | TrackBack
A while ago Elliot posted a rather scornful entry criticizing Richard Dawkins; while Elliot got a little carried away with the name-calling, I think he made some valid criticisms of Dawkins. Eventually, someone came across it on Google, fired off an email, Elliot cc’d the exchange to a high school friend in grad school, Neil and I got involved and the emailer responded. From that latest email:
There are lots of compelling reasons to reject or dismiss religious claims. I am not someone who comes to the discussion with no knowledge: I am an anthropologist, a student of comparative religion, and someone who has read every major religious text and many minor ones. To me, the best and really the only necessary argument against religion IS ALL THE DIFFERENT RELIGIONS. You are a Christian; great. But most people in the world are not. They have their own religions, that they believe just as fervently as you do yours, and that they “argue” for and find “evidence” for. But if yours is true, theirs are false. But if theirs are true, yours is false. A simple thought experiment: say there are 100 religions in the world. Each has an equal chance of being “true.” That means that each has a 1% chance of being true. That means that each has a 99% chance of being false. A recent study counted 33,000 sects JUST OF CHRISTIANITY. They must differ enough on some points to be different churches or denominations. So, the chances of any one sect “getting it right” is 0.003%. The chances of being wrong, whatever you believe, are at least 99.997%. Or, if there is no god at all, then 100%.
So you see, we atheists do not even need an argument! We bear no burden of proof, since we make no claims; we only question a claim—a claim with no good evidence for it and the odds stacked IMMENSELY against it. Evil schmevil, the point is that a theist believes against all odds.
So, statistically, we’ve proven that God is dead? I suspect even Nietzsche is rolling over in his grave at that.
Elliot promises a response, but I say: what’s the point in responding? Just because most religions are wrong on one count or another doesn’t mean they’re all wrong about everything. If one wants to argue religion based on demographics, one could just as easily point out that all religions share their rejection of materialism, meaning that the overwhelming majority of people and belief systems reject materialism. Hence, if we grant each worldview an equal probability of being “right”, then materialist atheism has a very, very tiny chance indeed of being correct.
I’ve addressed this whole argument from demographics thing before simply by noting that 20 million people can, indeed, be wrong. Although the approach is slightly different this time around, the basic cause remains the same: most people, even highly intelligent people, don’t understand numbers very well, especially in a statistical context. If we reduce an issue, be it buying a car or adopting a belief system, to a simple binary choice, then we’re going to get impressive statistics supporting both sides of the argument. Chevy may advertise that 20 million Americans bought the Lumina, but they conveniently fail to mention that this implies that 260 million Americans didn’t buy a Lumina. Similarly, an atheist can say “there’s 30,000 religions, so the probability that any one is right is, like, 0.003%”, but he usually fails to mention that this implies there are at least 30,001 different belief systems, including atheism, so he, too, only has a 0.003% chance of being right.
Another example of where a lack of quantitative literacy, to use Lynn Steen’s term, creates severe misconceptions arises in the context of the book I was re-reading today, Neal Stephenson’s Zodiac. In it, the protagonist/quasi-eco-terrorist/resident asshole, Sangamon Taylor, rails at several points about major polluters using the analogy of “an eyedropper-full of ‘compounds’ going into a railway tank car of pure water” to explain parts per million of pollutants for the TV cameras. Taylor goes off on a banana-peel-on-a-football-field rap, but he could also have countered by noting that one part per million of some noxious pollutant with a molar mass similar to that of water would mean 3 x 10^16, or 30 quadrillion, molecules of that pollutant in each and every drop of water in that tank car. In the chapter on asbestos in his excellent book A Mathematician Reads the Newspaper, John Allen Paulos points out this counter-argument is also misleading, but the point is that there are scary-sounding statistics supporting pretty much any perspective you like.
Paulos has been trying his best to teach people a bit of basic mathematical literacy over the years, with books like A Mathematician Reads the Newspaper and Innumeracy, but I think it’s fair to say that there are still plenty of people who are easily manipulated by the use of statistics. It’s hard to disagree with Paulos and the likes of Steen et al. in Mathematics and Democracy: The Case for Quantitative Literacy when they argue that “numeracy” or “quantitative literacy” is more important than ever in a society such as ours (even though the ideological slant in Quantitative Literacy is a bit grating).
Anyway, the point is that statistics can be misleading and deceptive even in instances where they are applicable. Getting back to the religion question, however, the larger point is that statistics aren’t even relevant to begin with. All they do is obscure the debate. If there is or isn’t a God, it really doesn’t matter how many people believe in him/her/it. To draw another analogy, just because the majority of people throughout the bulk of human history found slavery acceptable doesn’t mean they were right; by the same token, the fact that the majority of people throughout history were wrong about slavery doesn’t mean they were wrong about loving their children or any number of other beliefs they had. “Right” and “wrong”, if we stipulate that such things exist in the first place, aren’t subject to a vote.