This comes from a critique of an
essay on faith and science by philosopher of religion Mary Midgely that appeared in the Guardian. Midgely accused 'New Atheists' of unreasonably expecting religion to be scientific. PZ Myers' response to that charge is one of the clearest explanations of a science-based rejection of religion I've ever read:
http://scienceblogs.com/pharyngula/2010/06/bumblin_midgley_babbles_again.php...
Here's the logic behind the scientific rejection of religion, which is nothing like the weird version Midgley has cobbled up. The success of science has shown us what an effective knowledge generator accomplishes: it produces consensus and an increasing body of support for its conclusions, and it has observable effects, specifically improvements in our understanding and ability to manipulate the world. We can share evidence that other people can evaluate and replicate, and an idea can spread because it works and is independently verifiable.
Look at religion. It is a failure. There is no convergence of ideas, no means to test ideas, and no reliable outcomes from those ideas. It's noise and chaos and arbitrary eruptions of ridiculous rationalizations. Mormonism, Buddhism, Islam, and Catholicism can't all be true — and no, please don't play that game of reducing each religion to a mush that merely recognizes divinity. Religions have very specific dogmas, and practitioners do not blithely shuffle between them. Those differences are indefensible if they actually have a universal source of reliable knowledge about metaphysics.
Again, this is not a demand that religions must conform to science's methods, only that we should be able to assess whether it works. I can imagine a world where revelation, for instance, actually generates useful knowledge, where people independently acquired specific information piped right into their heads, straight from god. I'd expect, though, that there would be some agreement between all the recipients. It could even be strictly theological information, with no expectation of material support. If a host of people all around the world suddenly heard a gong in their heads, followed by the words (in their own language, of course) "The name of God is Potrzebie", well, then…there's something interesting going on. If these kinds of revelations continued and were consistent across cultures and traditions, I'd be willing to consider that there was something outside the human mind that was communicating with us. I'd admittedly be baffled by it all, but the fact that there'd be growing cross-cultural consensus on very specific claims would be hard to ignore.
As for outcomes, it also doesn't have to be something material — religion wouldn't have to be a tool for making better microwave ovens before I'd believe it, for instance. It could provide a universal moral code, or be an effective tool for improving mental health. If the enlightened people of Potrzebie were demonstrably calmer, more peaceful, and better at coping with stress because of the intermittent revelations, then I'd also have to admit that something was up. It's actually too bad that there isn't any such phenomenon taking place.
Basically, we've learned from the example of science that a way of knowing ought to do what it promises to do. They don't have to promise to do exactly the same thing — architecture and botany, for instance, don't have the same goals or methods, so we wouldn't expect physics and theology to echo each other's answers — but they ought to produce something reliable and true.
The fact that no religion can is damaging to them. Biblical literalism is crazy nonsense, but no more so than transubstantiation or doctrines of salvation or any accounts of what happens in heaven or hell. What drives our rejection of religion isn't that a few bits and pieces of specific religious beliefs, like the literal interpretation of Genesis, have been falsified, but that no consistent knowledge comes out of religion at all…yet every religion claims to provide knowledge about the nature of the universe.
...