I read an interesting NYT article this week. Here's the link:
http://www.nytimes.com/2007/11/24/opinion/24davies.html?em&ex=1196312400&en=5efb2bed27ae7934&ei=5070It reminded me of a conversation I had with a good friend, an atmospheric chemist, many months ago. She told me about a crisis of faith she encountered a year or so into her doctoral program. In short, she realized that all of the laws of chemistry and physics had simply been derived from observation. Why does pV=nRT? Because (pretty much) every time someone ran an experiment in which they calculated the relationship between volume, pressure, mols of monoatomic gas, and temperature, they ended up with the number 8.314472 m3·Pa·K−1·mol−1 left over.
What's so special about 8.314472? Nothing. Why couldn't it be 8.23945720? No good reason; it just isn't. The Ideal Gas Constant, R, is basically something that scientists came up with to fill out an equation. Sure, people went back and derived the Ideal Gas Law theoretically, but this requires simplifying assumptions and relying on R to equal 8.314472 (as well as relying on a host of other empirically-derived equations and constants). Yes, there's a mighty good dose of circular reasoning involved here.
Anyway, the Ideal Gas Law is considered a scientific law. Millions of chemistry and physics students around the world learn it. They base their calculations and lab experiments on it, though most of them have not personally found out for themselves whether or not it is true. And if they bother to ask, there is no answer to the question "why 8.314472?".
I love science. I love the scientific method. I think that it leads to the discovery of truth. However, it can only lead to the discovery of truths that can be empirically tested. Furthermore, honest scientists will admit that the empirical truths they discover can only be considered true in the physical setting in which they have been tested--they are not universal (even if they are, we have no way of knowing until we can run pV=nRT experiments in other universes).
So, I stand with the author of the article. Science is really not that much different than religion--it requires faith, just of a different kind. It requires one to believe that the universe is orderly and predictable and that empirical observations can be trusted and extrapolated from. It requires one to believe that anomalies are caused by some other, higher scientific law. And it only can provide answers to certain kinds of "why?" questions--once you begin to ask deeper "why?" questions, the only answer is that you have to take it on faith.