23 March 2007

How Science Works: Patterns & Predictions

Good news, everyone! Now that my ridiculous Finals Week is over, I can start posting random ramblings again! I know that's what you're all here for, after all. At any rate, as I mentioned a couple of posts ago, I recently took part in an impromptu debate on the philosophy of science. It was in my Community Ecology class, and began as a discussion of whether or not ecology had any general laws (more on that in the second installment of this post). I'm not usually a big participant in class discussions, but one of the other students said something that I just couldn't let go unchallenged. He said something to the effect of "Who cares about general laws, anyway? I mean, it's science, who cares if you're finding patterns or not?" Effectively, his argument was that science is all about exploring the unknown, and that simply going out and describing things (in this case, ecological structures) was sufficient. With all due respect to this student, I would consider this view to be not just mistaken, but dangerously so. Science is, at its core, all about finding patterns. Reduced to it's barest essentials, science is a two-step process: first you make a prediction, then you see if it holds true in the real world or laboratory (that's the Scientific Method, after all: formulating and testing hypotheses). Observation of patterns is essential to both these components. First, how can any prediction be made without some pattern to base it on? To cite an example near and dear to my own heart, a cladogram is a hypothesis of how organisms are related. While it is based on observational data, it is the patterns formed by these data that are of real importance: they allow an evolutionary biologist (by way of PAUP) to predict which organisms are most similar ("Species A and B share features 1 and 2, while Species C does not; therefore, A and B are more closely related."). Second, once your prediction is made, it is tested by observing whether the patterns it predicts are in fact manifested. In keeping with the example cited above, a cladogram might be tested in several ways, most often bootstrapping or jackknifing, which reshape or prune the tree to see of the patterns predicted by the original cladogram still hold. Of course, the student in my class was right that description and data collection are essential to both processes, but they alone are not sufficient. Finding patterns in these data and to use those patterns to construct or test a hypothesis is also necessary.
I probably ruffled a few feathers by saying the student who made the claim that patterns were unnecessary for science was "dangerously" wrong. I'll explain what I mean now, and maybe that will also explain why I take this sort of thing so seriously (it's not just that I'm old before my time and hate people disagreeing with me). I may sound like a conspiracy theorist now, but to me it comes down to that menace that's constantly looming in the shadows, "intelligent" design. The powers that be at the Discovery Institute and throughout the ID world are very, very good at making people think that what they do is science (it's clearly not, but I'll save my lengthy explanations of why for another time). A key component of their campaign is confusing people about what actually constitutes science. At the risk of sounding jaded, I'm never too surprised when they're able to pull wool over the eyes of much of the American public, but
to hear a student at a major university - supposedly one of our educated elite - profess such a profound misunderstanding of the scientific method is extremely worrying. Of course, this has more to do with the generally abysmal state of education in this country than with the agenda of the Discovery Institute per se, but it was still enough to shake me to the point where this has been on my mind literally every day for the last week and a half.
Having said all this and knowing that my audience is, of course, a highly scientifically literate bunch, I'm curious to know what you all think. I realize it's possible that my opinions on the nature of science are divergent from those that you have, and I'd be curious to know if you're with me 100%, agree with some things, or think I'm just a nut job with too much free time on my hands. After all, skepticism, even regarding your own thoughts, is another important facet of science.

1 comment:

Michael said...

This week's Morning Meeting Mini Trainings (sing it with me!) have been about "the" scientific method. (Alice wrote and led this one.) We did real science each morning, making observations, sure, but also forming hypotheses, testing them with experiments, then revisiting our hypotheses and drawing conclusions. While there is no one method for doing science, Alice pointed out, these elements must be present. Doing just one is part of science, but not the whole thing. I'm with you 100%.