Monday, May 28, 2007

The Blind Watchmaker - chapter 9

This chapter takes a completely different tone from what has gone before that gives me a feeling of some humanity behind the author. It is about the theory of "punctured equilibrium", but it still talks around many of the key issues.

In the middle of the chapter, we have this honest admission: "Both schools of thought despise so-called scientific creationists equally". That is a harsh statement. I may not be the least impressed by Dawkins ability to form a rational argument, but that doesn't mean I need to despise him as a person. Unfortunately, this statement is true and also extends to many theologians who despise creationists. Near the end of the chapter, Dawkins says "Whatever the motive, the consequence is that if a reputable scholar breathes so much as a hint of criticism of some detail of current Darwinian theory, the fact is eagerly seized on and blown up out of all proportion." Dawkins mentioned control system theory in the previous chapter. Take his two statements together and we have a feedback loop that is guaranteed not only to lock out creationists, but to silence anyone who would dare report data that conflicted with the theory of evolution. This is indeed the current situation in the Ivory Tower.

Punctuated Equilibrium is a theory that arose due to a conflict between the data (fossils) as interpreted by paleontologists and the perceived theory of evolution being one of more or less constant change. In 1972, it was formally acknowledged that most of the major animal groups show up suddenly in the fossil record. This was neither compatible with creation nor with evolution, if I recall correctly from what I was taught about evolution when I was a child. Dawkins tries to show that evolutionists had always believed in the possibility of wide variations in evolution rates so that none of this should have been the least surprising.

What is more important here is that Dawkins notes a gap between the thinking of the molecular biology community (constant evolution rates), the paleontology community (evolution singularities) and the biology community (something in between). The engineering optimization viewpoint has something to contribute to this discussion. The molecular biology community does an excellent job of looking at current mutation rates. This is a very important contribution for immunology. What isn't possible for them to do is to relate genes to fitness. Optimization is also coupled and we are forever looking at derivatives of fitness functions with respect to design variables. Because it is impossible for molecular biologists to relate mutations directly to fitness, they are forced to abandon the survival of the fittest concept in its entirety and look at mutations alone. Constant evolution rate is the only thing left, but it is completely wrong at a macro evolution level. More importantly, peaks and valleys of fitness are irrelevant by the necessarily simplistic presuppositions of current molecular biology, so the "saltation" issues disappear with a hand wave.

A suggested theory is that most of the macro evolution took place in isolated communities before moving back to larger communities and taking over. The problem here is that we actually do have data. When isolated populations from Hawaii, Australia and New Zealand where brought into contact with Africa, Eurasia and America, the result was the exotic species of the isolated communities suffering, not the other way around.

A final point that Dawkins notes is that selective breeding allows for changes to be made quickly (as in dogs), but eventually there is a point where additional change becomes increasingly difficult. His explanation is that these perturbed states of the species need to build up a sufficiently varied gene pool before evolution can be pushed further and that takes large amounts of time. Optimization theory gives us a different explanation: At the peak or valley, the fitness gradients are near zero, so that perturbations are quite easy. Moving further, however, we eventually hit regions of very large fitness gradients so that it is impossible to keep pushing. If we no longer artificially push the species out of equilibrium, then it will revert back to near its original state. Unfortunately for Darwinism, this is indeed the situation and the reason that "saltation" cannot be explained away without telling science to go to hell.

In spite of these comments, I would commend Dawkins for being much more forthright in this chapter than he has been in any of the previous ones.

Links: Chapter 1 & 2, Chapter 3, Chapter 4, Chapter 4b, Chapter 5, Chapter 6, Chapter 7, Chapter 8,


Bunc said...

Looney, again just a quick comment.
Your references to "saltation" appear to suggest that you see saltation as a fact - I may be misunderstanding your intent. Saltationism is in fact a theory of sudden change from one variation to the next which is of an order of magnitude different from normal rates of change.
As there is no evidence for saltation I am not clear why you feel that it needs to be explained away.
There has been soem misunderstanding that Goulds puntuated equilibrium is a "saltationist" theory but Gould himslef has made it very clear that this is a misreading of what he is proposing.
You are in fact crticising evolutionary theory for failing to explain away soemthing for which there is no evidence anyway!

Looney said...

Bunc, the notion of "saltation" being impossible (without a engineer) is something that I would share with Dawkins.

The problem of this chapter is unrelated to what was discussed: No matter how you spin it, all of the major design features of modern animals appear suddenly in the fossil record. The IDers assert that macro evolution is impossible. The paleontologists and Dawkins are pretty much asserting that it will never be possible to show macro evolution via the fossil record.