"Why not? There would be a cause for the initial impetus to change and a cause for the particular way of recombining.

That sounds as if you are saying that there is mechanism in a bacterium that says "Aha! I'm being exposed to Methicillin. I'd better make some changes to my DNA. I'll just change that Adenine to a Thymine and that'll sort it out." That sounds dangerously close to Lamarckism."

It doesn't sound like anything of the sort. You are talking in terms of goals and I have comprehensively dismissed this already more than once. The cause and the resultant change are blind to each other. Not difficult.


"Quite. The sequence of bases resulting from the change, whatever the cause for the change, is random; it's natural selection that determines whether it is any use or not or, more specifically, whwther it survives."

The sequence is a result of causes. Absolutely not random. Natural selection is not a determining force: it is blind, without goals. On one island the mutation may be beneficial, on another the self-same mutation may be useless. Natural selection does not work directly on genes, but on favourable (or not) expressed attributes according to the prevailing environmental circumstances. Read your Darwin.