CultureLab - Go with the flow system

Bob Holmes, contributor

In Design in Nature by Adrian Bejan and J. Peder Zane a new theory of nature is mooted, but is the idea stretched beyond its reach?

NEW fundamental laws don't pop up every day in science. Yet that's exactly what Adrian Bejan claims to have discovered: a basic principle of nature, overlooked until now, that governs the evolution of everything from river basins to athletic performance and human culture.

Bejan, a mechanical engineer at Duke University in Durham, North Carolina, has been writing about his so-called "constructal law" in the technical literature for over a decade. Design in Nature, co-written with journalist J. Peder Zane, is his attempt to bring his big idea to a general audience.

Bejan's thesis is based on flow systems. In his eyes they cover almost everything: rivers (and trees, too) are flow systems for the water that courses through them; human commerce is a flow system for moving goods around the planet; universities and culture are flow systems for ideas. And his constructal law is simple: flows tend to get better - faster and longer - with time.

The flow of water in a river delta arranges itself into main channels that branch into about four subsidiary channels, which branch again into about four smaller channels and so on, because this arrangement gives the greatest flow rate. Similarly, when a tree branches, the total cross-sectional area of the daughter limbs is equal to that of the parent trunk, giving unimpeded flow of water. Nature doesn't just seek the path of least resistance, Bejan says, it constructs it. The result is not randomness, but design: real rivers and trees look much the way an engineer would draw them. Better yet, we can predict how they'll change in the future: to allow faster, better flow.

So far, so good, although if you want to see detailed proofs you will need to dig into the literature, Bejan and Zane skimp on the nitty-gritty. The constructal law is an interesting idea, and even the light version presented here brings a useful new perspective to ubiquitous natural phenomena. But Bejan is not content to wade in these safe shallows. In pursuit of a theory of everything, he overreaches himself.

Animal life, he claims, is a flow system for moving mass around the planet - and the constructal law says that animals should evolve to move farther, faster. As a result, evolution should go from swimming to running, which is quicker and requires less energy, and thence to flying, which is quicker yet. "This is why animal locomotion first emerged in the oceans, spread onto land, and later rose into the air and not the other way around," he says, in what surely ranks as one of the most bizarre sentences ever written about the evolution of animals. And then he goes one better: eyes and brains, he says, help animals move better and the constructal law, therefore, explains why animals with vision and cognition arose after animals without vision and cognition, not vice versa. Do we really need an explanation of why the earliest animals weren't sharp-eyed, brainy, winged creatures?

Then we're off to the world of human civilisation. The flow of people across the landscape is enhanced by a river-like system of footpaths, side roads and main roads, and our transportation systems have certainly evolved towards farther, faster flows. And just as there are fewer highways than footpaths, Bejan argues that to speed the flow of ideas, there should be just a few great universities - the highways of ideas - and lots of mediocre ones. Maybe so, but I am not convinced.

The book is full of this stuff. The constructal law, Bejan claims, can explain why there are few trees in the desert, why the best sprinters are black, the best swimmers white, and why we prefer paintings that are half again as wide as they are high. Most of these ideas are interesting, and some might even be true. But Bejan comes across as someone who, having invented a hammer, sees everything as a nail. In this case, it's the reader's ability to believe Bejan's ideas that takes a battering.

http://www.newscientist.com/