Replies

Well, you don't have to answer if you don't want to. I did review as much of this thread (just your posts) as I could bear--this thread is far too long. It didn't pop out to me. Let me recap, and if the recap is pretty simple it's because no one I believe has really answered a very basic question. Let's try yet another metaphor. Think of a significant somatic change (you would say phenotypic expression) as crossing a river. Now, what "significant" means will eventually have to be clarified, but let's not do that yet. That is, the previous state is the west bank, and the next state is the east bank. Evolutionary theory, as I understand it, says that we can get across the river by hopping across rocks in the river, each of which represents a different genetic state a little different from the last. Each of those rocks have some functional difference, so that the improbability of the mutations between Rock i and Rock i + 1 doesn't really matter--given enough time and enough generations, the mutation happens, and once in existence proliferates, setting the stage for the next iteration. One objection to this theory is the implausibility that you can come up with a chain of functional intermediates to which this story could apply. If in fact there is such a chain, and such a chain in every case, of course the objection falls away. By implication, you and others seem to accept at least the possibility that there could not be a chain of functional intermediates. But you have another argument. You say, consider nonfunctional mutations (or as general re once argued introns), which could accumulate, and at some point become meaningful because of a bitflip, or a transposition. (Btw, whether a mutation is literally single point or something fancier like a transposition doesn't matter for this purpose--it's still intrinsically an unlikely event and still just a single change to the DNA.) In the metaphor, you are saying that the rocks don't need to be so close together--if you wait for nonfunctional mutations, in effect the river crosser can construct one-time use stilts that will give him a bigger stride. I say in response, but that approach depends on how probable it is that the nonfunctional mutations could happen just before the next rock needs to be reached. The probability for that is something like P1*P2*P3..PN, by which I mean proportional to the product of the underlying probabilities of each mutation. (Edsheppa pointed out helpfully that the probability of the mutation may not be the same as the proportion of mutants in the population over time, but as long as the latter is a linear time-dependent function of the former, I think the argument still has force.) So in other words, the farther apart the rocks are, the taller the stilts have to be. So does the concept of nonfunctional mutations get you anywhere? And why do we care what the a posteriori populations of alleles are, when in our thought experiment we are on the west bank and thinking in terms of apriori probabilities?