"Natural selection can favor new proteins, and genes, but only after they perform some function. The job of generating new functional genes, proteins and systems of proteins therefore falls entirely to random mutations. Yet without functional criteria to guide a search through the space of possible sequences, random variation is probabilistically doomed."
You dismiss this insight out of hand, Ichneumon. But please consider this:
New proteins and genes are astronomically complex, and one imagines their functions can only arise under such conditions of complexity. Random variation would have to select a whole bunch of molecules/macromulecules all at once for a new protein or gene. Then somehow the organizing principle that could make them "all work together" so as to become a new protein or gene would need to be present. But how could that be a "random" thing?"
The darkness of ignorance, sigh.
BB, please you are usually one of the more restrained ones here. Please don't argue from ignorance.
"New proteins and genes are astronomically complex"
Let's see - how about SNORT?
I suggest you read the papers of E.C.C. Lin on enzyme recruitment. No links - sorry. You'll have to go to a real library and fish them out. But until you do, stop this twaddle about "astronomically complex". You will find it is really quite simple. This stuff goes back more than 40 years. Ignoring it won't make it go away.
It seems to me that you try to establish "astronomical complex" at the protein and gene level so you can sneak design in at the organism level. I'd like to think that I'm not right about this, but that's the way it looks.
It's astronomically complex at a level deeper even than the protein and gene. I am referring to the quantum basis of systems living or non-living. I am not trying to "sneak" anything in here. I'm trying to point out a problem. FWIW.
There are no bounds on self-delusion.