Michael Egnor is at it again. The guy is pretty much the energizer bunny of anti-evolution bullshit. This time, he's purportedly refuting an article by Dr. Steven Novella, a Yale professor of neurology.
So, why am I butting my nose in to a discussion between two doctors? For two reasons:
- First, because once again, Egnor pulls out his gibberish about information theory - and that's definitely my turf.
- Second, because ultimately, the argument that Dr. Egnor makes comes back to the silly way that he reduces to evolution to a tautology. As I've discussed several times before, Dr. Egnor formulates a trite, foolish tautology out of a description of natural selection, and then pretends that the entire theory of evolution is nothing more than his foolish tautology. Apparently, he's convinced himself, and as a result, he creates arguments from it without
ever bothering to consider whether or not they make the slightest bit of sense. This latest
screed of his is the worst example of this that I've ever seen. And that's saying a lot!
Anyway, enough intro. Let's look at Egnor's own words. Nothing I can say could possibly
be as damning as what he says himself. Replying to Novella's description of how
Egnor publicly entered the argument over evolution, he says:
Actually, all I did was ask a question: how much biologically relevant information can Darwin's mechanism of chance and necessity actually generate? I didn't settle for hand-waving or for reassurances that "Darwin's theory is a fact." I wanted a measurement of biological complexity, with empirical verification, in a way that was meaningful to biology. I never got an answer to my question.
As regular readers of this blog will recall, that's not exactly a true statement. Dr. Egnor first invoked information theory, demanded that he be shown a quantitative analysis
of how much information an evolutionary process could generate. And he was answered, with a detailed quantitative analysis of an evolutionary process in terms of Shannon's information theory. That's when he changed his terms, and started demanding a quantitative analysis of biological information - or as he terms is above, "biologically relevant" information. And since then, he has refused all requests for a definition of "biologically relevant" or "biological" information. I've dealt with this problem extensively before - it's a classic shell game, where the questioner always wins, because he's in control of the game. No matter how you respond, he can always say "You're wrong, that's not biological information."
After that, he goes from bad to mind-numbingly, painfully awful:
Dr. Novella is missing a much better example of random mutation and natural selection that's not metaphorical at all. Cancer is a test of Darwin's theory. Cancer is real biological evolution by random mutation and natural selection, writ fast. There's no reason to invoke encyclopedia typos or tractor engines in order to understand what "chance and necessity" can do to a living system. Brain tumors are perfect little Novellian "two-cycle engines" nestled inside the skull, "random mutations" coming out the ears, and "natural selection" like there's no tomorrow (excuse the metaphors). Brain tumors are constantly generating new biological variation, and they are avatars of natural selection. They provide a tremendous spectrum of variation, from "variation jet-engines" like malignant glioblastoma multiforme to "variation tortoises" like benign pilocytic astrocytomas. Cancer wards are full of patients brimming with "two-stroke engines" of evolutionary change.
Dr. Novella, again:
...it is [easy] to imagine how shuffling around information, duplicating, and altering the information could easily result in meaningful and even useful new information.
The best real biological test of "shuffling around information, duplicating, and altering the information" is cancer. According to Dr. Novella's reasoning, brain tumors ought to be generating quite a bit of "meaningful and even useful new information." Better neuroanatomy and better neurophysiology ought to be popping up "easily." Better frontal lobes and cognition, from cancer. Better temporal lobes and memory, from cancer. Better cerebellums and coordination, from cancer. If random mutations and natural selection--Dr. Novella's "two stroke engine"--is the source of all functional integrated biological complexity, brain tumors ought to help our brains evolve in some way.
Perhaps Dr. Novella has data that show real evolutionary improvements in the brain caused by brain tumors. If he has, he should show us.
I'm just a rube, not a Darwinist from Yale. But I've never seen cancer make a brain better.
Dr. Egnor has become a victim of his own foolishness. He's been so aggressive about pushing the idea that evolution is no more than his own reductionist tautology that he's convinced himself that it's really true - even to the point of ignoring everything that
he must know as a medical doctor.
As I've said before, you can reduce any proof to a tautology by simply wrapping all of the premises and inferences together into a single statement. Similarly, you can create
a tautology out of a scientific theory by combining the basic observations and inference steps that led to the theory into the statement of the theory.
Given a theory, you can state it as a tautology in a bunch of different ways. You can do it the correct way - in which case it doesn't sound silly. To do that, you need to include statements about all of the relevant observations, and all of the necessary details of the theory. What you end up with really looks like a detailed argument leading up to the theory.
And then, you can do it the silly way. You can pull out a silly reductionist statement of the form "A therefore A": "Things fall, therefore things fall"; "The individuals that survive to reproduce are the individuals that survive to reproduce".
Egnor continually insists that evolution can be reproduced to that silly reductionist
version: "The individuals that survive to reproduce are the individuals that survive to
reproduce". And that silly version, obviously, drops important parts of the theory - and
that's exactly where Egnor's problem comes from.
Remember, his reductionist tautology is "The individuals that survive to reproduce are
the individuals that survive to reproduce" - a reduction that is full of holes. He leaves
out everything else about the theory, and then falls into the hole that he put there
himself, all the while crowing about how that show's that he's correct.
What is he missing? What does evolution really say that he left out? It's a
complex theory, but the part that's relevant here is: It says that within a population,
changes will occur in the distribution of genes. Individuals with traits that make them
more successful at surviving to produce offspring pass those traits on to their
Egnor's cancer babble goes wrong because he forgets about that crucial "pass those traits on to their offspring".
Cancer does produce a population of rapidly reproducing and rapidly mutating individuals. It's certainly an example of a kind of evolutionary process: cells are reproducing wildly, and the ones that manage to survive to reproduce are the ones that survive to reproduce and pass their traits on to their offspring. But expecting the changes in cancer cells to produce any evolutionary changes in the host organism of the cancer is just incredible stupidity. Because changes in cancer cells don't produce inheritable changes in the host. A brain cancer does not get to pass genes to the children of the host organism.
There's a basic, fundamental distinction that Dr. Egnor must be familiar with - the distinction between germ cells and somatic cells. Germ cells are the cells which contain the genetic material that will be passed on to offspring; somatic cells are the rest of the cells of the body. Genetic changes in a cancer in somatic cells - i.e., pretty much all cancer cells - doesn't affect the genes of the germ cells. The genetic changes of cancer are not heritable in terms of the reproduction of the host! The kind of selection that acts on the cells of a cancer that causes them to change over time isn't the same selective pressure that's working on the cells of the host, because cancer cell reproduction is not host reproduction; cancer cell mutations are not heritable mutations for the host!
In some sense, cancer can be viewed as a distinct species: instead of being a part of the host organism, they're a parasite living inside of the host. The only children that cells of the cancer have are more cancer cells: the only way that a cancer cell passes genes to a child is by reproducing itself. Genetic changes in the cancer cells cannot produce a heritable change in the host organism - because genetic changes in cancer cells don't change the genes of the host organisms reproductive cells. They don't affect the genome of the host organism - only the genome of their own descendants: the other cancer cells that they produce.
When we talk about selection operating inside of a tumor, we're talking about selection of cancer cells, selecting the
cells that are most successful at reproducing in the environment of the host: not selection that makes the survival and reproduction of the host more likely. The cancer cells that are most "successful" will eventually kill the host - so if anything, the success of cancer cells will cause a reduction in the fitness of the host.
And in fact, that's something that we do observe. Organisms do possess traits that make them less likely to develop cancer.
Dr. Egnor really needs to stop his mouth from moving for a little while, and think
about what he's saying. It's really shocking to see an argument like this from someone who must know the difference between germ cells and somatic cells.