Tuesday, December 18, 2007

Philosophy of Science (a Rant)

I read an article today that got my back up a little. Let me give a little background to explain why.

I grew up in the rural, conservatively religious town of Rangely, CO. The first real experience I remember having that set me apart from the majority of the town came in 7th grade science class when we where discussing (you guessed it) evolution. Presented with another in the series of assertions that seems to constitute science education, I blithely bought in. My peers, who had been privately informed that evolution was not to be "believed in", discovered by waywardness, they dubbed me Monkey Boy. Yes, I know. Hilarious.

This was certainly not the last time that I butted heads with conservatives in my town, and it stands at the base of a growing sense of disconcert that I have felt towards organized religion. In the meantime, I went on to study physics in college, and am now a graduate student in astronomy (I do experimental cosmology). It has been surprising to me, now that I have started being introduced as an astronomer, to suddenly to have became an "Interpreter of the Voice of Science" (preacher) for people who want to know how our universe came to be.

The reason the above article got my back up was that I've always been careful, in these situations where I find myself describing the history (as we know it) of the universe, to stress the differences between science and religion. It's confusing terrain because cosmology has only recently come into the realm of science. My "what is science" speech usually ends up something like the following: science isn't a body of knowledge (despite what constitutes science education), or a belief system, or a religion. It is a methodology. Its groundings are philosophical, but that doesn't make its results philosophical. Science started from the philosophy that the universe is predictable: that one can predict the outcome of an experiment if one knows the initial conditions to sufficient accuracy.

Out of this philosophy (which the microscopic, quantum universe has helped us to understand is not strictly correct), a method was devised for arriving at an understanding of the predictability of the universe. Hypothesize, Predict, Experiment, Analyze, Conclude. Elementary school science fair stuff (except that "conclude" now means "peer review"), but still confusing. Science is a level playing field where anyone can make up any theory they want, in the face of all current knowledge if they want, using any strange force or being or Flying Spaghetti Monster, and science has nothing to say about it until you make predictions, devise an experiment to test them, and peer review to make sure it's reproducible. Science works like infants learning shapes--pick a block, try to jam it through the hole, and repeat until you succeed.

The framework that seems to work best for predicting the outcomes of experiments involves math. I'll save a discussion of why this might be for another blog, but it's important to realize that it didn't have to be this way. The universe could have chosen to play by different rules, or perhaps there are even models that can predict the outcomes of all the various experiments we've tried without using math. Math is just a model we have that works. Once you have two models that have equivalent predictive power, science has nothing to say again. Yes, I know about "Occam's Razor", but that's philosophy again, despite its popular portrayal as a pillar of science.

I get frustrated when people (even scientists and cosmologists) misunderstand science to be a body of knowledge, or a set of "transcendental laws", because (it seems to me) it's putting the cart before the horse. Whatever laws and rules we have are only as good as the degree of testing they have undergone, and are subject to revision and replacement as new experiments reveal their flaws. And we know they're flawed. The two most accurate physical models we have--gravity and quantum mechanics--are mutually inconsistent, and neither of them are able to account for an accelerating expansion of the universe (the "Dark Energy" problem) or for motions on galactic scales (the "Dark Matter" problem). So how can a self-respecting scientist hold up a "transcendental law" and claim it is more than our current best model?

My personal opinion is that referring to science as a body of knowledge makes it akin to religion--something transcendental and immutable (and inaccurate) handed down from above. What makes science science are the error bars: little reminders that our models are only as good as the extent to which they have been tested.