Professor Jerry Coyne Addresses Michael Behe’s “response” to Coyne’s review of Behe’s new book.


Although the line above says that this is a post by Mark Perakh, in fact I only served as a conduit for posting Professor Jerry Coyne’s material.

Professor Coyne has published a review (see here ) of Michael Behe’s new book titled The Edge of Evolution. Along with other reviewers, such as Mark Chu-Carroll, Sean Carroll, Richard Dawkins, and others, Jerry Coyne views Behe’s new book as Behe’s poorly substantiated (and vain) attempt to somehow pull up Behe’s status from the deep pit he finds himself in after the Dover trial, and thus to be re-admitted to the scientific community as a genuine scientist.

Behe responded to Coyne’s review on an Amazon blog (see here).

Now Professor Coyne offers a rebuttal of Behe’s response, which rebuttal I have the privilege of posting.

I can add to Jerry Coyne’s rebuttal just a few words. As it could be expected, Behe’s “response” to Coyne’s critique is typical of Behe’s supercilious style wherein he does not shy away from a self-gratifying delusion regarding his fiasco as an expert witness at the Kitzmiller vs Dover Board of Education trial. Professor Coyne in his brief rebuttal shows the dismal failure of Behe as the author both of his new book and of his “response” to critics.

Read Jerry Coyne’s rebuttal at Talk Reason.


From J. Coyne’s response

It is important to draw the distinction between Behe and his fellow IDers, lest people mistake ID for a monolithic theory accepted by all its proponents. Behe is one of the few intelligent-design proponents who accepts common descent, macroevolution, natural selection, and an old earth. This puts him severely at odds with his other ID colleagues at the Discovery Institute, including William Dembski and Stephen Meyer.

This is something that always puzzled me. In addition to Behe former DI fellow Siegfried Scherer was cited in a German news paper saying that “ID is not science and shouldn’t be taught in schools”. Seemingly, everybody in the ID camp has his very own personal designer. IMO, they see each other as useful IDiots and the only thing that keeps them together is the denial of evolution. I would appreciate if they would teach the controversy where there is a real controversy, i.e. within their own camp. Unfortunately, this is just wishful thinking.

Oh, Crap. I was going to say “Even Behe is still using the evolution=RM+NS dodge. It’s hopeless,” but then I realized he was sort of loosely paraphrasing Coyne. What a mess. Lots of “Darwinian/ism/ist”s dropped in there, too. Sad, really.

The world of ID looks very different under oath :-)

It seems to me that it is possible to lose an important point in arguing over scientific issues. (As well as give the impression that there is a scientific controversy being discussed.)

Nowhere in any of the writings about ID is there any description of what “design” might be. Other than something-or-other which is capable of doing anything, capable of being responsible for anything from fiddling with bacteria, giving them flagella, to fixing the basic physical constants of the universe. It would be easy to dwell on the vast, unsupported extrapolation that this amounts to; but “design” - whatever it might be - is never shown to be capable of doing anything which is not done by natural processes, for the only design that we are aware of operates within, and because of, the laws of nature. There is never any interest shown in thinking about what this vastly more capable “design” might be like, what its limitations (if any) might be, Who, What, Where, When, Why, or How - much less subjecting the hypothesis to any test of its efficacy.

This is a repost of mine from the PZ blog. Behe is wrong on the basic science. Malaria can evolve resistance to chloroquine easily and by multiple different mutations. He also did not take into account the role of stepwise sequential mutations and sexual recombination, the latter a common method for concentrating adaptive alleles from an extended gene pool into one organism.

Below is an example of malaria evolving resistance to a drug combo with a triple mutant in one gene and a double mutant in another, total 5. Whatever the limits of evolution, Behe has vastly underestimated them. Them’s the facts.

This ‘quintuple mutant’ is considered as a molecular marker for clinical failure of SP treatment of P. falciparum malaria.

A quick review of the literature reveals that malaria evolves resistance to drugs routinely. In the case below, resistance to a drug combo is associated with 5 mutations in 2 genes. One is a triple mutant in dhfr combined with a double mutant in dhps. Behe is claiming that double mutants in the chloroquine resistance gene is the edge of evolution. Here we have a simple case of a triple and a double mutation in the same organism.

His assumptions are faulty, his science is faulty, and his conclusions are false. This is your basic Demski-Meyers class fraud. Fraud and lies are the signatures of christian creo thinking, an internal contradiction that insults both the religion and science.

Malaria can accumulate multiple mutations to become resistant to many drugs without breathing hard. Science/medicine can explain these without breathing hard, standard evolution aided by sexual recombination. It’s too bad for the patients. Malaria is like TB, there are strains resistant to most drugs and strains resistant to all drugs are on the horizon. One of these days we may well be hearing about Extremely drug resistant malaria (XDR-M) and some lawyer will pick up a case somewhere and introduce it to the US, :>(.

Acta Trop. 2003 Mar;85(3):363-73. Links High prevalence of quintuple mutant dhps/dhfr genes in Plasmodium falciparum infections seven years after introduction of sulfadoxine and pyrimethamine as first line treatment in Malawi.Bwijo B, Kaneko A, Takechi M, Zungu IL, Moriyama Y, Lum JK, Tsukahara T, Mita T, Takahashi N, Bergqvist Y, Bj�rkman A, Kobayakawa T. Department of International Affairs and Tropical Medicine, Tokyo Women’s Medical University, 8-1 Kawada-Cho, Shinjuku-Ku, Tokyo 162 8666, Japan.

Malawi changed its national policy for malaria treatment in 1993, becoming the first country in Africa to replace chloroquine by sulfadoxine and pyrimethamine combination (SP) as the first-line drug for uncomplicated malaria. Seven years after this change, we investigated the prevalence of dihydropteroate synthase (dhps) and dihydrofolate reductase (dhfr) mutations, known to be associated with decreased sensitivity to SP, in 173 asymptomatic Plasmodium falciparum infections from Salima, Malawi. A high prevalence rate (78%) of parasites with triple Asn-108/Ile-51/Arg-59 dhfr and double Gly-437/Glu-540 dhps mutations was found. This ‘quintuple mutant’ is considered as a molecular marker for clinical failure of SP treatment of P. falciparum malaria. A total of 11 different dhfr and dhps combinations were detected, 3 of which were not previously reported. Nineteen isolates contained the single Glu-540 mutant dhps, while no isolate contained the single Gly-437 mutant dhps, an unexpected finding since Gly-437 are mostly assumed to be one of the first mutations commonly selected under sulfadoxine pressure. Two isolates contained the dhps single or double mutant coupled with dhfr wild-type. The high prevalence rates of the three dhfr mutations in our study were consistent with a previous survey in 1995 in Karonga, Malawi, whereas the prevalences of dhps mutations had increased, most probably as a result of the wide use of SP. A total of 52 P. falciparum isolates were also investigated for pyrimethamine and sulfadoxine/pyrimethamine activity against parasite growth according to WHO in vitro standard protocol. A pyrimethamine resistant profile was found. When pyrimethamine was combined with sulfadoxine, the mean EC(50) value decreased to less than one tenth of the pyrimethamine alone level. This synergistic activity may be explained by sulfadoxine inhibition of dhps despite the double mutations in the dhps genes, which would interact with pyrimethamine acting to block the remaining folate despite dhfr mutations in the low p-aminobenzoic acid and low folic acid medium mixed with blood.

Posted by: raven | June 28, 2007 10:08 AM

If evolution is a car, then natural selection is the engine and mutation is the gas.

I have seen this, or closely similar analogies used before. I personaly edit them to, “If evolution is a car, then reproduction is the engine, mutation is the fuel, and natural selection provides the steering wheel.”

Thanks Mark for posting this and hosting Coyne at TalkReason. I enjoyed reading the exchange between Behe and Coyne.

Coyne’s rebuttal is embarrasingly inadequate.

A researcher by the name of White calculated the 1 in 10^20 probability, not Behe.

In any given year, there are about 10^17 malarial parasite cells in the human population.

Because of fitness disadvantages, host immunities, and sampling, the rate of spontaneous resistance to malarial drugs is 1 in 10^12, and involves one mutation. The probability of a second mutation developing in a cell that has already developed the first mutation (since TWO mutations are always needed–though they need not be the same two mutations), would be 1 in 10^8, based on genome size. That means that the probability of the two needed mutations developing in a human is 10^20 [(1 x 10^12) x (1 x 10^8)]. Since, as noted above, the total population of malarial cells in any given year is only 10^17, then spontaneous resistance isn’t seen. It takes more than one year for the resistance to develop.

So, bottom line, Behe’s statistics hold up. And Coyne (and Miller) should re-examine their critcisms.

And, what Behe says in TEOE, still holds true. In trillions upon trillions of every kind of conceivable mutations that have taken place over the tens of thousands of years in which humans and P.falciparum have engaged in “trench warfare”, where selective advantages are extremely high, are a series of SNPs in hemoglobin. Indeed, Darwinism is very limited in what pathways it can fruitfully trace out.

Here’s the other side to raven’s post:

“A study from Malawi (Kublin et al., 2002) showed correlations between mutation patterns and clincial failures. One of the findings was that a quintuple mutant(carrying DHFR mutations at positions 51, 59, and 108 and DHPs mutations at positions 437 and 540) is associated with SP treatment failure and that the presence of a single DHFR mutation (Arg-59) combined with a single DHPS mutation (Glu-540) accurately predicts the presence of a quintuple mutant. In study II, all patients who still had parasites on day 14 had an adequate clinical response. IN study V, one patient was a quintuple mutant on day 14, but still had an adequate clinical and parasitlogical response. One of these patients had all five mutations on day 14, but still had an adequate clinical response, contrary to the findings in the Malawi-study (Kubling et al., 2002) All our patients had a Glu-540 mutation on day 0; many of these together with the Arg-59 mutation, though only one of them was a quintuple mutant. This contrasts with the findings of Kublin et al. (2002) on the value of these mutations to predict quintuple mutations, but agrees with findings in another recent study in Malawi (Bwijo et al., 2003)

From: Kobayakawa, T. (2003) High prevalence of quintuple mutant dhps/dhfr genes in. Plasmodium falciparum infections seven years after introduction of sulfadoxine

So, the quintuple mutation can be present and you still get SP suppresion of the malaria. And, in many cases, you have two of the quintuple mutations present without the quintuple mutations being present–which throws into question when, and how, two of the quintuple mutations came about.

Blast, your comment is embarrassingly inadequate.

Coyne doesn’t go into the details of Behe’s numbers, because he sees immediately that “Behe’s bizarre and unrealistic assumption that for a protein-protein interaction to evolve, all mutations must occur simultaneously” is wrong. And raven’s and your reference indeed each shows data on stepwise mutations.

Henry J Wrote:


Oh, snap!

[LOLparrot cackles softly on Dembski’s shoulder: “C no crackers. Iz nuts!”]

Coyne omitted what I thought was the most hilarious part of Behe’s rebuttal:

Leave aside the fact that the parts of the opinion Coyne finds so congenial (which are standard Darwinian criticisms of intelligent design) were actually written by the plaintiffs’ lawyers and simply copied by the judge into his opinion. (Whenever the opinion discusses the testimony of any expert witness — for either side, whether scientists, philosophers, or theologians — the judge copied the lawyers’ writing. Although such copying is apparently tolerated in legal circles, it leaves wide open the question of whether the judge even comprehended the abstruse academic issues discussed in his courtroom.)

Note that Behe is still trying to give the impression that the judge did something improper (albeit “tolerated”) by following the standard legal practice of incorporating verbatim those portions of the complaint that he judged to be proved. Of course, Behe’s rather desperate attempt to portray the judge as uncomprehending is understandable, considering that it is quite clear (from those portions of the decision that were not derived from the plaintiff’s complaint) that it was in large part Behe’s own testimony that convinced the conservative, Republican-appointed judge that Intelligent Design is a load of manure.

Just one question. Everybody does this but what is with putting quotation marks around “response”? Was Behe’s response not a real response? It may have been wrong but even if it was wrong it was a response. putting quotes around it seems like a subtle ad hominem.

Torbj�rn Larsson:

Your response is embarassingly inadequate. Where does Behe ever say the mutations have to be simultaneous? Please give me a page number.

Colin (comment 185721) points out that placing the quote marks around the word “response” regarding Behe’s reaction to Coyne’s critique is a subtle form of ad hominem. This remark may sound superficially true for those readers who are not familar with what Behe wrote replying to Coyne. A real response to critique is supposed to address the essence of critical remarks without irrelevant comments designed to denigrate the opponents. If you (Colin or anybody else) read Behe’s post (quoted by Coyne in detail) you’ll see that Behe’s “response” does not meet those minimal conditions to qualify as a real response. Just one example. When Behe asserts that “…Coyne hides behind the judicial skirts of the former head of the Pennsylvania Liquor Control Board,” this phrase comprises two not quite “subtle” ad hominems in one cheap shot, one personally denigrating Judge Jones and the other doing the same to Professor Coyne, and it has nothing to do with the essence of Coyne’s critique. Because of such irrelevant remarks, obviously designed to hurl mud upon his opponents without providing a substantive response to the opponent’s arguments, IMHO the quote marks around the word “response” regarding Behe’s post are fully deserved.

IMHO the quote marks around the word “response” regarding Behe’s post are fully deserved.

After reading this stuff long enough, it becomes clear (as Judge Jones almost said in so many words) that a response to the substantive critiques, meaning an on-topic examination of the issue and an evidence-based defense of the initial position, is simply not possible. This is the case because the creationist claims about objective reality are simply wrong. The creationist must choose between ignoring, evading, or misrepresenting the critique.

A better term than response (which doesn’t occur) might be “react”. Clearly, Behe is reacting to Coyne. His purpose isn’t to defend his falsehoods, but to create the impression that Coyne has been “dealt with”. Underlying such reactions is the general principle that since creationism is based on falsehoods, it is not possible to be honest and be a creationist at the same time. Even Kurt Wise comes out and admits that he chooses to deny reality (as a scientist!) as required. Behe’s reactions are exercises in PR, not substance.

Here’s some more from White’s review: “Antimalarial drug resistance”; White, Nicholas J.; J.Clinical Investigation 113:1084-1093. (April, 2004)

Recent remarkable molecular epidemiological studies in South America, southern Africa, and Southeast Asia have challenged this view. By examination of the sequence of the regions flanking thePfdhfr, it has become apparent that , even for SP, multiple de novo emergence of resistance has not been a frequent event, and that, instead a single parasite ( with a mutation in Pfdhfr at positions 51, 59 and 108) has in recent years swept across each of these continents.

You’ll notice this matches three of the five “quintuple” mutations. Malawi is in southern Africa.


Blast Wrote:

Your response is embarassingly inadequate.

I don’t think that joke works the way you think it works.

Blast Wrote:

Where does Behe ever say the mutations have to be simultaneous?

I quoted Coyne above, to show that you hadn’t adequately shown him wrong.

If you have a problem with Coyne’s professional assessment, you have to present Behe’s models and show where Coyne is wrong.

Coyne’s assessment is not on Behe’s treatment of malaria resistance specifically but about his general treatment of protein-protein interaction. Are you claiming that Behe is inconsistent? Well, it certainly seems so, but I am surprised to hear a creationist admit to that.

Coyne notes further in his review:

Jerry Coyne Wrote:

Not surprisingly, it turns out that getting by mutation a set of three to four amino acids required for only one protein-protein interaction is very low (mutations in the DNA affect the building blocks of proteins, since DNA codes for a sequence of amino acids). It is especially low because Behe requires all of the three or four mutations needed to create such an interaction to arise simultaneously. [Bold added.]

And he goes on to describe that Behe describes a strawman of evolution, which of course is irrelevant to how the biology works. That is why he doesn’t need to look at Behe’s specific numbers, they are GIGO. It is as simple as that.

Could someone explain to me how the following is more palatable than the ideas of Behe et al…thank you

‘To avoid damage…the microorganisms…traveled in the head of an unmanned spaceship sent to earth by a higher civilization which had developed elsewhere some billions of years ago…Life started here when these organisms were dropped into the primitive ocean…We called our idea Directed Panspermia…’

Crick, Francis - LIFE ITSELF, (NY:, Simon & Schuster, 1982) pp. 15-16

or this…

‘Note that chemical evolution is a special case of spontaneous generation…’

Dodson, Edward O. & Dodson, Peter - EVOLUTION: Process and Product, (Boston: Prindle, Weber & Schmidt, 1985) p. 349

Please refrain from the old wives’s tale that Darwinism doesn’t deal with Origins…we all know that he imagined his “ warm little pond” while writing to Joe Hooker…

Torjborn Larsson:

Coyne’s assessment is not on Behe’s treatment of malaria resistance specifically but about his general treatment of protein-protein interaction. Are you claiming that Behe is inconsistent? Well, it certainly seems so, but I am surprised to hear a creationist admit to that.

Behe is not being inconsistent; instead, Coyne is simply failing to follow Behe’s argument.

The simultaneity in question refers to the needs involved in forming a protein-to-protein binding site. Behe points out that such a PTP binding site requires five or six a.a. changes. Of these five or six, in order to be conservative in assessing what is needed of evolution, Behe considers three to be neutral. Behe then indicates that if a mutation is non-neutral, then it would very likely have a harmful effect on the mutated protein, as much as most mutations are harmful. Hence the need, and invocation, of simultaneity.

This has nothing to do with Behe’s calculations, however! Rather, Behe points out that such simultaneaity makes a singular PTP binding site even more improbable than a CCC (presumably because it involves an extra a.a. [CR requires only two a.a. changes] and because of the simultaneous requirement). This gets us back to the CCC calculation, which, again, was made by White, and not Behe, of 1 in 10^20.

So the argument runs this way: it takes 10^20 parasites (=organisms) to develop a two a.a. mutation that resists chloroquine. This, then, is all that NS could produce in the “trench warfare” taking place between man and parasite, viz., in an environment where the selective pressure on the parasite couldn’t be higher. Yet, that is all we can get out of evolution after 10^20 organisms have been produced. And a PTP binding site requires more mutational power! While these kinds of numbers can be overcome by bacteria, fungi, and parasites–which exist in huge populations and are capable of phenomenal growth rates—they don’t work for mammalian species.

That’s the argument–not the strawman version of it that Coyne wants to present.

Please refrain from the old wives’s tale that Darwinism doesn’t deal with Origins… we all know that he imagined his “ warm little pond” while writing to Joe Hooker…

The ToE (Theory of Evolution) deals with relationships among species. What “Darwinism” deals with depends on who’s using that term, since its meaning varies from one speaker to the next.


Re “they don’t work for mammalian species.”

I thought the species under discussion was malaria?


Henry: I thought the species under discussion was malaria?

What was under discussion was the number of organisms that need to be born in order to bring about two SNPs. Malarial parasites are eukaryotes. Mammals are eukaryotes. Humans, caught in the same “trench warfare” with malarial parasites, have come up with only one SNP, as in sickle-cell anemia, or thalassemia.

Re “Malarial parasites are eukaryotes. Mammals are eukaryotes.”

So? It’s mammals that need more time for mutations to accumulate, not eukaryotes in general.


Test post, due to spam queue.

hfd Wrote:

Please refrain from the old wives’s tale that Darwinism doesn’t deal with Origins

Most people aren’t like creationists, willing to lie for their cause. So no, how populations with replicators arose isn’t part of what basic evolutionary theory needs to treat. It concerns itself with development on existing populations.

It is easy to understand from a cursory look at the descriptions of the fact (“common descent”) or the simplest description of a main mechanism (“hereditary variation with selection”). To use physics speak, the boundary conditions are populations in and populations out.

[Missed piece that was to go before the start of the last comment:]

Seems my latest commentaries didn’t pass the spam filter. I will recreate some of it, and cut it to non-offending pieces..

[Cont:] The conditions and mechanisms for biogenesis as a scientific area of its own is an acute interest for astrobiologists.

Panspermia is an interesting mechanism, but of limited value in our universe which has finite history. It is in principle testable too - if we get to the planets in question we can test for common descent on biochemistry et cetera. So yes, it is more palatable, but not a main interest for biogenesis.

The main interest for astrobiologists is to find and define the conditions for habitable planets. IIRC at the current rate of discovery, the expectation is to see an habitable earth analog in 2 +/- 1 years. The discovery of worlds with detectable life (imbalanced atmosphere chemistry et cetera) can come anytime from that point.

AFAIK one of the main researchers claims that within 10-20 years we will likely have a good statistical description of how planets with life looks now and when life started. That will be a good guide for future biogenesis research.

[Now I get what the offending words are! Duh! I do hope PT 2.0 fixes that. Cont:]

hfd Wrote:

we all know that he imagined his “ warm little pond” while writing to Joe H… [Edited for spam filtering.]

Darwin’s idle speculation in a private letter to botanist Joseph H… has a context which your quote-mining removed. He goes on to note:

Charles Darwin Wrote:

It is mere rubbish thinking at present of the origin of life; one might as well think of the origin of matter.

[not linked reference to get through spam queue]

Today we are of course in a position where we can reasonably get answers to both questions. Biogenesis research has started as above, and cosmology/high energy physics has started as well.

Blast Wrote:

This has nothing to do with Behe’s calculations, however!

Exactly my point. Thanks for finally realizing that Coyne is assessing Behe’s general treatment of protein-protein interaction. And he is then more detailed.

He goes through the malaria example at length but no depth, since there is no depth to plumb, and notes that too as “an almost willful misunderstanding of Darwinism”.

As for your discussion of Behe’s PTP model I don’t see where Coyne isn’t following Behe’s argument.


I don’t think you’re being honest with yourself.

This is from one of your posts;

Not surprisingly, it turns out that getting by mutation a set of three to four amino acids required for only one protein-protein interaction is very low (mutations in the DNA affect the building blocks of proteins, since DNA codes for a sequence of amino acids). It is especially low because Behe requires all of the three or four mutations needed to create such an interaction to arise simultaneously. [Bold added.]

It’s quite clear that Coyne suggests the high improbability of Behe’s CCC is due to Behe’s requirement that the mutations be simultaneous. But the low improbability was calculated independently of the PTP binding scenario, and doesn’t, in fact, require simultaneity. If one follows the way in which Chloroquine resistance develops, you would see that the 1 in 10^20 number presumes that one, and then the other mutation (that is, sequentially) develops. The fact that simultaneity is needed for PTP binding sites to develop simply means that its probability is even lower than 1 in 10^20. Coyne entirely misses this point. It’s too bad you don’t want to admit that.

Blast Wrote:

It’s quite clear that Coyne suggests the high improbability of Behe’s CCC is due to Behe’s requirement that the mutations be simultaneous.

You insist on conflating these two arguments. Read my comment above which is a summary of Coyne’s review, or better yet read the review.

And you can’t use Coyne’s rejections of both Behe’s argument separately to argue that he has missed them in combination. It is a ridiculous argument, yet you go there.

You will probably continue to insist that I have misunderstood Coyne, as you continue to insist Coyne has misunderstood Behe. Why break a pattern?

Fortunately other readers will easily pick up on what is correct; which is good because I am tired of repeating myself - and I can break a pattern.

Behe’s talk at DI on his new book will be carried on C-SPAN2 from 4:10 to 5:45 New York time Saturday afternoon July 7.


You insist on conflating these two arguments.

No. Coyne uses the ploy of conflating the arguments; I’m trying to separate them—as they should be.

About this Entry

This page contains a single entry by Mark Perakh published on July 1, 2007 9:20 AM.

Yecke in Her Own Words was the previous entry in this blog.

Haldane’s non-dilemma is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.



Author Archives

Powered by Movable Type 4.381

Site Meter