Friday, February 18, 2011

evolution = chaos?

Kottke points to a graphical variation of the Chinese whispers game whereby an original sign (in this case, a line drawn by a human) is rapidly degraded by multiple repetitions (the more people try to repeat the original line, the less line-like it becomes, eventually degrading into chaos).

A Sequence of Lines Traced by Five Hundred Individuals from clement valla on Vimeo.

Kottke marvels that "The lines get really messy surprisingly fast [...] this is a nice demonstration of evolution."

But is it? Is it the case that evolution leads to chaos*? I don't think so. Evolution leads to variation and change, sure, but chaos? The difference between evolution and this line transformation, I think, is pressures. In evolution there are pressures that greatly effect which changes last more than one generation and hence become permanent stable. But in this game, there are no pressures, as far as I can tell. There is no survival of the fittest because each turn gets to survive for exactly one generation with no pressure to be fitter than another in order to persist beyond one generation. So this exercise, cute as it may be, does not resemble evolution at all, I don't think.

*or messiness in Kottke's phrasing


Chris, The Book Swede said...

There might be a confusion between evolution and natural selection. Evolution, on its own, is just changes in gene frequencies. So accumulating random mutations in a population would be "evolution". That could lead to chaos (*would* lead to chaos).

Natural selection is where bits of chaos, mutations, are added to organisms' designs depending on how well they improve the function of the design, such that it propagates more copies of the genes responsible for that design, relative to alternative -and so competing - genes in the population that also determine that trait.

It's a bit wordy, and crude, but I hope it makes sense. Natural selection is the only non-chaotic (if chaotic means 'random') form of evolution there is, and so the only thing capable of creating complex -- and *functional*! -- adaptations designed to solve specific problems.

Chris said...

Fair point ;)

Putting the Linguistics into Kaggle Competitions

In the spirit of Dr. Emily Bender’s NAACL blog post Putting the Linguistics in Computational Linguistics , I want to apply some of her thou...