Dispatches from the Creation Wars

This is a post I wrote during the Dover trial in 2005. I think it’s important to put it up again.

One of the interesting segments of the Michael Behe cross examination begins on page 42 of the Day12AM transcript, and it concerns a paper that Behe wrote with David Snoke. That paper, called Simulating Evolution by Gene Duplication of Protein Feature that Requires Multiple Amino Acid Residues, was based upon a computer simulation that attempted to answer the question of how long it would take cumulative point mutations in a single gene to produce a new trait – the interaction of two proteins – requiring a change in multiple amino acid residues if there was no selective advantage to preserve any of the individual mutations until they were all present and the final result was fully functional. For Behe, this is a simple example of irreducible complexity:

Thus in order for a protein that did not have a disulfide bond to evolve one, several changes in the same gene have to occur. Thus in a sense, the disulfide bond is irreducibly complex, although not really to the same degree of complexity as systems made of multiple proteins.

This paper has been lauded by ID advocates as an excellent example of ID-stimulated research. The DI has listed it as an example of genuine peer reviewed research that supports ID. William Dembski has declared that Behe and Snoke’s research “may well be the nail in the coffin [and] the crumbling of the Berlin wall of Darwinian evolution.” Unfortunately for them, this paper didn’t hold up well under questioning during the Dover trial.

I’m going to post a very long section of the transcript and I’ll start below the fold. It begins on page 42 with Mr. Rothschild describing the article to Behe and Behe correcting a few minor things and agreeing on what the paper examined and concluded, and with what restrictions.

Q. And let me just ask you a few questions, and you tell me if I’m fairly summarizing the results of your computer simulation. What you’re asking is, how long will it take to get — and please follow with me, I’m trying to do this slowly and methodically — two or more specific mutations, in specific locations, in a specific gene, in a specific population, if the function is not able to be acted on by natural selection until all the mutations are in place, if the only form of mutation is point mutation, and the population of organisms is asexual?

A. I would have to look at that statement closely because there are so many different aspects to it that I don’t trust myself to sit here and listen to you say that and form a correct judgment.

Q. Anything I said about that sound incorrect?

A. If you repeat it again, I’ll try.

Q. I’d be happy to. Two or more specific mutations?

A. Actually, this dealt with one or more.

Q. One or more mutations?

A. Yes. If you notice, in figure — if you notice in figure 3, you look at the x axis, you notice that there are data points there that start at one. So we considered models where there were one, two, and more mutations.

Q. Fair enough. In specific locations?

A. No, that’s not correct. We assumed that there were several locations in the gene that could undergo these selectable mutations, but we did not designate where they were.

Q. In the specific gene?

A. We were considering one gene, yes.

Q. In a specific population?

A. Yes.

Q. Okay. If the function is not able to be acted on by natural selection until all mutations are in place?

A. Yes, that’s what’s meant by multiple amino acid residue, multi-residue feature, yes.

Q. If the only form of mutation is point mutation?

A. Yes, that’s a very common type of mutation, which is probably half or more of the mutations that occur in an organism.

Q. And if the population of organisms is asexual?

A. Yes, we did not — actually, we did not confine it just to asexuals, but we did not consider recombination.

Q. Are prokaryotes an example of the kind of organism that you were studying there?

A. Again, we weren’t studying organisms, but, yeah, they’re a good example of what such a model has in mind.

Q. And to say this very colloquially, you conclude that it will take a large population a long time to evolve a particular function at disulfide bond, right?

A. A multi-residue feature. That’s correct, that’s correct.

Q. And specifically —

A. I’m sorry.

Q. Go ahead.

A. Let me just finish. Depending on — as we emphasize in the paper, it depends on the population size. And, of course, prokaryotes can oftentimes grow to very large population sizes.

Q. And here the conclusion, the calculations you concluded was that, if you had a population of 10 to the 9th power, that’s a population of 1 billion?

A. That’s correct.

Q. To produce a novel protein feature through the kind of multiple point mutations you’re talking about, it would take 10 to the 8th generations, that’s what it says in the abstract, correct?

A. If, in fact, it was — if, in fact, the intermediate states were not selectable.

Q. Okay.

A. And if this is by gene duplication as well.

Q. Okay. So 10 to the 8th generation, that’s 100 million generations?

A. That’s correct.

Q. And yesterday, you explained about bacteria, that 10,000 generations would take about two years in the laboratory, correct?

A. Yes.
Q. So 100 million generations, that would take about 20,000 years?

A. I’m sorry?

Q. 100 million generations, which is what you calculated here, that would take about 20,000 years?

A. Okay, yes.

Q. And those are numbers based on your probability calculations in this model, correct?

A. Yes.

Q. Now it would be true that, if you waited a little longer, say, instead of 10 to 9th generations, 10 to the 10th generations, then it would mean that you wouldn’t need as big a population to get the function that you are studying?

A. That’s right. The more chances you have, the more likely you are to develop a feature. And the chances are affected by the number of organisms. So if you have a smaller population time, and more generations, that could be essentially equal to a larger population size and fewer generations.

Q. So, as you said, so if we get more time, we need less population to get to the same point, and if we had more population, less time?

A. That’s correct, yes.’

Q. Now would you agree that this model has some

A. Sure.
Q. And you, in fact, were quite candid in indicating that in the paper?

A. That’s correct.

Q. And if we could turn to, what I believe is, page 8 of the document. And if you look in the paragraph that’s actually continued from the previous page that says, we strongly emphasize. And if you could —

A. I’m sorry. What page number is that?

Q. It’s page 8 in the document. And it’s up on the screen as well.

A. Yes, okay. I’ve got it.

Q. Could you read into the record the text to the end of the paragraph beginning with, we strongly emphasize?

A. We strongly emphasize that results bearing on the efficiency of this one pathway as a conduit for Darwinian evolution say little or nothing about the efficiency of other possible pathways. Thus, for example, the present study that examines the evolution of MR protein features by point mutation in duplicate genes does not indicate whether evolution of such features by other processes, such as recombination or insertion/deletion mutations, would be more or less efficient.

Q. So it doesn’t include recombination, it doesn’t include insertion/deletion of the mutations?

A. That’s correct.

Q. And those are understood as pathways for Darwinian evolution?

A. They are potential pathways, yes.

Q. This study didn’t involve transposition?

A. No, this focuses on a single gene.

Q. And transpositions are, they are a kind of mutation, is that right?

A. Yes. They can be, yes.

Q. And so that means, this simulation didn’t examine a number of the mechanisms by which evolution actually operates?

A. That is correct, yes.

Q. And this paper, let’s be clear here, doesn’t say anything about intelligent design?

A. Yes, that’s correct. It does imply irreducible complexity but not intelligent design.

Q. But it doesn’t say it?

A. That’s correct.

Q. And one last other question on your paper. You concluded, it would take a population size of 10 to the 9th, I think we said that was a billion, 10 to the 8th generations to evolve this new disulfide bond, that was your conclusion?

A. That was the calculation based on the assumptions in the paper, yes.

MR. ROTHSCHILD: May I approach the witness, Your Honor?

THE COURT: You may.

Q. What I’ve marked as Exhibit P-756 is an article in the journal Science called Exploring Micro–

A. Microbial.

Q. Thank you — Diversity, A Vast Below by T.P. Curtis and W.T. Sloan?

A. Yes, that seems to be it.

Q. In that first paragraph, he says, There are more than 10 to the 16 prokaryotes in a ton of soil. Is that correct, in that first paragraph?
A. Yes, that’s right.

Q. In one ton of soil?

A. That’s correct.

Q. And we have a lot more than one ton of soil on Earth, correct?

A. Yes, we do.

Q. And have for some time, correct?

A. That’s correct, yes.

So let’s review what the article that is going to be the “nail in the coffin” of Darwinian evolution actually proves, according to its author. But first, let’s bear in mind something very important: the evolution of a new binding site between two proteins of the type described here by Behe’s own article is, by his own definition, an example of irreducible complexity:

An irreducibly complex evolutionary pathway is one that contains one or more unselected steps (that is, one or more necessary-but-unselected mutations). The degree of irreducible complexity is the number of unselected steps in the pathway.

And remember, the core of Behe’s entire argument for ID is that irreducibly complex systems cannot evolve. Yet what does he admit under oath that his own study actually says? It says that IF you assume a population of bacteria on the entire earth that is 7 orders of magnitude less than the number of bacteria in a single ton of soil…and IF you assume that it undergoes only point mutations…and IF you rule out recombination, transposition, insertion/deletion, frame shift mutations and all of the other documented sources of mutation and genetic variation…and IF you assume that none of the intermediate steps would serve any function that might help them be preserved…THEN it would take 20,000 years (or 1/195,000th of the time bacteria have been on the earth) for a new complex trait requiring multiple interacting mutations – the very definition of an irreducibly complex system according to Behe – to develop and be fixed in a population.

In other words, even under the most absurd and other-worldly assumptions to make it as hard as possible, even while ruling out the most powerful sources of genetic variation, an irreducibly complex new trait requiring multiple unselected mutations can evolve within 20,000 years. And if you use more realistic population figures, in considerably less time than that. It sounds to me like this is a heck of an argument against irreducible complexity, not for it.