Genetic algorithms—do they show that evolution works?

by Don Batten

A genetic algorithm (GA) is a computer program that supposedly simulates biological evolution. GAs have found limited application in generating novel engineering solutions—for example, an electronic circuit that filters out a particular frequency. GAs use mathematical constructs that parallel mutations (random changes in the variables/coefficients), natural selection (elimination of variations in a circuit, for example, that do not move toward the objective of a response to a particular frequency), and even some type of ‘recombination’ (as happens in sexual reproduction). Because of this, some apologists for evolution claim that these programs show that biological evolution can create the information needed to proceed from less complex to more complex organisms (i.e. with more genetic information).

However, GAs do not mimic or simulate biological evolution because with a GA:

For the above reasons (and some of them overlap), and no doubt there are more that could be added, GAs do not validate biological evolution. It does not take long with a decent calculator to see that the information space available for a minimal real world organism of just several hundred proteins is so huge that no naturalistic iterative real world process could have accounted for it—or even the development of one new protein with a fundamentally new function.

Another type of ‘simulation’ is that of antitheist T.D. Schneider.1 Schneider claims that his program simulates the naturalistic formation of DNA binding sites for gene control. This exercise has led to grandstanding by some evolutionists that this proves creationists wrong. However, many of the same problems outlined above also apply to this programming exercise. For example, the selection coefficient is extremely high, the genome is extremely small, the mutation rate high, no possibility of extinction is permitted, etc. For many other problems, see the critique by Dr Royal Truman.

Note that we are not saying that mutations and natural selection cannot generate any information (see Spetner’s book, Not by Chance for example). It’s just that with real world generation times, real-world sized genomes and real-world organisms which have to survive through multi-dimensional adaptive traits, there has not been enough time to generate even a tiny amount of the biological information seen in living things. As Spetner says, look, if mutations and natural selection have generated all the information we see, then we should be able to easily find some examples of some new information (i.e. increase in specified complexity) arising today. The best that anyone has come up with is a GA, which does not simulate real world evolution, for the reasons outlined above.

Reference

  1. Schneider, T.D., Evolution of biological information, Nucleic Acids Research 28(14):2794–2799, 2000. In this paper, Schneider acknowledges the input of fellow atheist Richard Dawkins. Return to text.
Schneider, T.D., Evolution of biological information, Nucleic Acids Research 28(14):2794–2799, 2000. In this paper, Schneider acknowledges the input of fellow atheist Richard Dawkins.

Related Articles