Friday, August 29, 2008

Towards More "Elegant" Science


So here I am trying to put the final touches on my dissertation proposal. It's been laid out conceptually, but I've been finding that everything tends to get very complicated very quickly. Control for soil moisture, radiation, or season, or probable date of occurrence or propagule pressure... I think I'm getting the spins.

I was heartened to hear about an experiment wherein a scientist literally looked at the direction cows were facing in pictures in Google Earth (and possibly aerial photos, I'm not sure) and discovered that they seemed to sense the earth's magnetism, because they faced magnetic north a statistically significant amount of time.

How wonderful that such a simple idea- observation, question, simple observational study and discovery of a phenomenon- can lead to a significant discovery in this day and age! And not to mention, a publication in the Proceedings of the National Academy of Sciences (can you see me turning green with envy?). The time of dropping things from the leaning tower of Pisa etc. is long gone, but elegant science seems still possible.

To be precise, elegance is not equivalent to simplicity, rather, it is is the accomplishment of documenting a phenomenon by using a method that captures the essence of what is going on. Often an out-of-the-box reasoning, or simple alternative to traditional methods might be used. Nature teases us ecologists with her elegance, in patterns and processes, and challenges us to find the signal among the chaos.

When I've been designing experiments now, the words (I'm not sure what the citation is here, any help is appreciated) "Simplify, simplify" ring in my head. Also, my 8th grade algebra teacher's favorite expression, KISS, or "Keep It Simple Stupid!"

With all deference to the power and value of complicated experiments, as a student, it is helpful to get back to the question you're asking, and ask, what is the simplest way that I can see if this phenomenon is occurring? It may not approach elegance, but I think it's a start.
.

No comments: