The Canard: Random Mutation cannot produce complex 'finished' systems.
The Response: Mutations in evolution are not so much 'random' as indifferent, or undirected. There is no predictability or purpose to the mutation that, through natural selection, results in evolution...it just happens.
Jerry Coyne's explanation:
On the basis of much evidence, scientists have concluded that mutations occur randomly. The term "random" here has a specific meaning that is often misunderstood, even by biologists. What we mean is that mutations occur irrespective of whether they would be useful to the organism. Mutations are simply errors in DNA replication. Most of them are harmful or neutral, but a few of them can turn out to be useful. And there is no known biological mechanism for jacking up the probability that a mutation will meet the current adaptive needs of the organism. Bears adapting to snowy terrain will not enjoy a higher probability of getting mutations producing lighter coats than will bears inhabiting non-snowy terrain.
What we do not mean by "random" is that all genes are equally likely to mutate (some are more mutable than others) or that all mutations are equally likely (some types of DNA change are more common than others). It is more accurate, then, to call mutations "indifferent" rather than "random": the chance of a mutation happening is indifferent to whether it would be helpful or harmful. Evolution by selection, then, is a combination of two steps: a "random" (or indifferent) step--mutation--that generates a panoply of genetic variants, both good and bad (in our example, a variety of new coat colors); and then a deterministic step--natural selection--that orders this variation, keeping the good and winnowing the bad (the retention of light-color genes at the expense of dark-color ones).
So....if my imperfect understanding of the explanation is accurate, the following happens:
- Populations reproduce.
- Genetic mutation causes changes in offspring during embryonic formation.
- Offspring who 'chance' to have beneficial mutations that favour their survivability may survive to reproduce more than those with neutral or non-beneficial mutations.
- The population of 'beneficially mutated breeding subjects' is more likely to pass on a beneficial mutation to their offspring (along with other mutations).
- Over time the net accumulation of passed on beneficial traits may lead to a beneficial mutation being adapted by a species and becoming a 'common inheritance' of said species.
- Natural Selection combined with Beneficial Mutations may not result in increased complexity, but in fact may result in the reduction of certain complexities if those reductions amount to a net-positive or net-benefit for the species.
Note: I'm becoming quite a fan of Jerry Coyne.