Thoughts in a Haystack

Dr. Michael Egnor, striving mightily to demonstrate that a brain is not an absolute requirement in order to operate on the things, is back with a double-header … and so soon after Opening Day!

Egnor has "responded" to Mark Chu-Carroll's criticism of his claim that Natural Selection is "a tautology, and therefore useless to modern medicine." Mark has amply demonstrated the emptiness of Egnor's attempt to answer the charge. But I thought I'd throw in a bit more. Egnor claims:

All permutations of "Natural Selection" — "survival of the fittest," "reproductive success," "mechanisms that contribute to the selection of individuals that reproduce," "sexual selection," "gametic selection," "compatibility selection" — reduce quite readily to "successful reproducers successfully reproduce," or colloquially, "survivors survive." The truth of Natural Selection is its logical structure. It can't be false. "Natural Selection" is logically true, but it’s a weak theory because it’s merely a tautology.

What Egnor is leaving out is that those reproducers, through heredity, pass along the very traits that make them able to successfully reproduce in that particular environment. In short, through mutation and selection, information as to how best to survive in that environment is added to the genomes of the descendant organisms.

Contrary to Egnore's assertion, however, that is not logically necessary. There are a number of ways that reproductive success could be decoupled from the genes passed to an organism's offspring. Throughout the 18th Century and into the 19th, Preformationism was a live scientific idea. Depending on the version, Preformationism held that each individual organism was complete in either the egg or sperm and merely developed by growing larger and "unfolding" (the original meaning of "evolution," before Herbert Spencer screwed things up). But such a fully-formed but tiny "homunculus" would, by implication, have all its sperm and/or eggs already, just as its parent before it had. Indeed, this "Russian Doll" situation would mean that all the individuals of a species came into existence when the original organism was created.

This, in turn, would mean that the traits of one individual are not passed on to another. They all come down from the original creation and the fitness of your ancestors had no effect on your biological traits. And that is only one possible way that fitness could be unrelated to the genes of descendants.

Thus, Natural Selection, within the entirety of evolutionary theory, is not a tautology but a prediction of the effect of environmental changes on the genetic makeup of populations and an explanation of (one of) the mechanisms by which such genetic changes come about.

As to Natural Selection not being important to modern medicine, it seems that immunologists disagree, as shown in this article from the Howard Hughes Medical Institute. As the article (one out of many that can be easily found in a Google search) puts it:

First, let's make sure we're on the same page about how resistance develops. The development of antibiotic resistance is like really fast natural selection. When a selective pressure (such as a drug) is applied to a population of organisms (such as bacteria), the "fittest" organisms will be the ones that can survive and propagate in the presence of the pressure. In bacteria, the race is between the rate of mutation in the molecule that the antibiotic targets and the rate at which the bacteria are killed by the drug.

… [T]he longer the bugs survive under the selective pressure of the antibiotic, the greater the chance of developing a resistant strain. But another factor is the size of the population. Genetic mutation is essentially random. A mutation that gives rise to the specific resistance is going to be a rare event -- sort of like winning the lottery. But in this case, you're talking about a rapidly dividing, fairly large population to start with (if you were sick enough to have gone to the doctor in the first place, there's probably a good number of bacteria there already). It's like the difference between buying 100 lottery tickets and buying 1 million lottery tickets. You always have a chance with a few, but the odds of winning go way up with the larger "population" of tickets.

... [I]t's better to take the entire course (or none of it) than to take half of it—better to eliminate all the bugs or not to expose them at all to the drug than to expose them just enough to select for resistant strains.

In other words, doctors want you to complete the course of antibiotics because they hope any mutation will provide only partial resistance and they hope to kill off those mutated individuals before a further mutation will make the resistance too good. Leaving the partially resistant population to breed increases the likelihood of the further mutation.

I wonder if Egnor instructs his patients to complete the entire course of antibiotics he prescribes. If not, a good case can be made that he is committing malpractice. If he does, then what reasonable explanation does he have for why he does, even when an infection is already under control? If Egnor's simplistic version of Natural Selection is right, there is no immunological reason to continue the antibiotics once the infection is controlled.

As for the twinight game, challenged by Orac to produce examples of his claim that the "design inference" has been "of great value" in medicine, he ... um ... threw up a claim that Watson and Crick inferred design in order to untangle the structure of DNA.

They reversed-engineered DNA. They collected physical data about the structure of DNA (X-ray diffraction studies, Chargaff's rules, the physical chemistry of nucleotides, etc), and then they designed a model of the molecule to understand its structure and function.

Egnor is not clear how this infers design but it seems to have something to do with a single sentence he pulled off a Wikipedia page:

Sciences such as biology and physics can be seen as reverse engineering of biological 'machines' and the physical world respectively.

Besides apparently never having heard the word "analogy" in his life, Egnor seems to think every scientific enterprise that collects physical data and then tries to construct physical or conceptual models is using a "design inference."

But then, evolution is also using a design inference and why is Egnor so against it? ... Other than his utter confusion, that is.

____________________________________________


And now for the real purpose of this post: to further propagate links to refutations of Egnor's proud ignorance and make sure anyone with a modicum of curiosity about his claims won't be mislead into thinking they are based on his knowledge of biology:

Egnor's latest kook-fests

Michael Egnor, Reverse Engineering and Genetic Algorithms

The latest from the World of Egnor

Egnor's awful example of "design inference"

Brain surgeon knows Jim Watson's mind better than Jim Watson

This statement is not a tautology

Egnor's next trick

reverse engineering, the design inference, and geology

Egnor doesn't know how to think either

Link - Posted by John Pieret @ 8:59 PM