environmental changes. Nature could look random even if we know it is not, and it can be unpredictable even if we know the rules. (He has since served as the British government’s chief science adviser and as president of the Royal Society and is now Lord May of Oxford.) To believe, he says, that physics offers the one pure, rational method of doing science, building a theory by moving from hypothesis to experiment to refutation, is a fiction. There are whole fields, such as astrophysics, without experiments and whole fields, such as string theory, without a single data point. “Ecologists who hanker for the precision of physics don’t have the faintest clue what physics is really like,” he says.

Others, on the other hand, believe that it is not biology and physics that are different but biologists and physicists. The physicist Freeman Dyson wrote that scientists are split into unifiers and diversifiers and that biologists tend to be the diversifiers, “happy if they leave the world a little more complicated than they found it,” and that biology lacks generality because biologists do not look for it.

What Makes a Theory

This is getting out of hand. Trying to work out the differences between biology and physics, and biologists and physicists—and arguing whether biology is in its Copernican, Keplerian, Newtonian, Einsteinian, or whoever-ian phase—is a fun coffee-break conversation for academics. But more productive than asking whether biology is, or should be, like physics is to ask whether there are simple universal laws that apply to biological systems, and whether these laws can be rooted in principles and approaches borrowed from physicists, either directly (such as the theory that network structure controls metabolic rate) or by analogy (the idea that large groups of plants and animals behave like large groups of particles). And if there are, what should we expect from them?

For starters, any law of nature should describe a pattern that would allow us to see generalities and shape our expectations. Kleiber’s rule predicts that if we encounter an unfamiliar mammal, we should expect its metabolic rate to be roughly 70 times its body mass raised to the



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement