Behe has written a very bad book, so poorly supported that I don’t want to waste a lot of time taking apart every sentence, but I did want to say a few words about chapter 9, where he takes on evo-devo. I waited a bit because I knew that Sean Carroll was writing a review of the book for Science, and I expected he’d go gunning for chapter 9, too—but no, he didn’t. I guess he felt as I do, that since Behe’s fatally flawed premise was exposed in the first few chapters, there was little point to addressing his incompetent nit-picks later in the book. After all, when the construction crew has built a foundation of tissue paper in a pool of quicksand, by the time you get around to criticizing the roofers for using graham crackers for shingles, you’re about out of outrage.
I’ll briefly note the best parts of Carroll’s review, though, and I’ll try to gather up a few tired shreds of indignation and exasperation to critique some of the more ridiculous canards of Behe’s evo-devo chapter.
Just to refresh your memory: Behe claims that every single species is explicitly and specifically designed, and the measuring stick he uses to make that claim is the improbability of evolving any molecular interaction that involves even a mere two amino acids. The basis for this grand conclusion is clearly bogus and based on some very poor scholarship. This ought to be enough to demolish the value of the whole book, and Carroll reiterates the main problem.
Behe seems to lack any appreciation of the quantitative dimensions of molecular and trait evolution. He appears to think of the functional features of proteins in qualitative terms, as if binding or catalysis were all or nothing rather than a broad spectrum of affinities or rates. Therefore, he does not grasp the fundamental reality of a mutational path that proteins follow in evolving new properties.
Ironically, though, the main premise of the book is an attempt to turn such black-and-white thinking into a quantitative concept. The “math” (just as what the DI does with “science” necessitates hiding the misuse of the term within quotes, so too we’re going to have to start calling this number-diddling they do “math”) is entirely fallacious. Behe wants to argue that evolution of a two amino acid combination is at the very edge of what is evolutionarily possible, and Carroll annihilates that whole idea with a brilliantly simple explanation.
Very simple calculations indicate how easily such motifs evolve at random. If one assumes an average length of 400 amino acids for proteins and equal abundance of all amino acids, any given two-amino acid motif is likely to occur at random in every protein in a cell. (There are 399 dipeptide motifs in a 400-amino acid protein and 20 x 20 = 400 possible dipeptide motifs.) Any specific three-amino acid motif will occur once at random in every 20 proteins and any four-amino acid motif will occur once in every 400 proteins. That means that, without any new mutations or natural selection, many sequences that are identical or close matches to many interaction motifs already exist. New motifs can arise readily at random, and any weak interaction can easily evolve, via random mutation and natural selection, to become a strong interaction (9). Furthermore, any pair of interacting proteins can readily recruit a third protein, and so forth, to form larger complexes. Indeed, it has been demonstrated that new protein interactions (10) and protein networks (11) can evolve fairly rapidly and are thus well within the limits of evolution.
(It’s bad form to cite articles I haven’t dug into myself, but just in case you want to review the citations yourself, here are the citations referenced above.)
9. V. Neduva, R. B. Russell, FEBS Lett. 579, 3342 (2005).
10. Y. V. Budovskaya, J. S. Stephan, S. J. Deminoff, P. K. Herman, Proc. Natl. Acad. Sci. U.S.A. 102, 13933 (2005).
11. P. Beltrao, L. Serrano, PLoS Comput. Biol. 3, e25 (2007).
That’s beautiful. What Behe claims is virtually impossible is easily generated by random chance, without even considering the role of selection.
But now, would you believe that Behe already has an excuse, buried within chapter 9? He has read Carroll’s book, Endless Forms Most Beautiful — but doesn’t seem to have understood it — and he makes an effort to criticize it, feebly. Carroll was probably predisposed to make the explanation above because one of the points he made in that book was that binding sites for transcription factors in the regulatory regions of the domain are short, only a few nucleotides long, and therefore common just by chance alone. Behe makes a note of this.
It turns out that, because the regions they bind are so small, developing a binding site for a regulatory protein is too easy. By chance, any particular six-nucleotide sequence should occur about once every four thousand or so nucleotides. Given the enormous length of DNA, there is a great chance that a binding site might already be near a gene. What’s more, the likelihood of having a site that matched five out of six positions—so that only one mutation would be needed to change the last position to make a perfect match—is even better. There should be one of those every few hundred nucleotides.
What a strange blind spot Behe has. He’s a biochemist. He should know something about the genetic code. The above paragraph is correct (although the comment about the enormous length of DNA is weird and irrelevant), and any particular six-nucleotide sequence should pop up frequently. Or to think of it another way, any two three-nucleotide sequences, which is roughly similar to saying any sequences coding for two amino acids (that’s even more likely than a six-nucleotide sequence because of the degeneracy of the genetic code, though). Note that much of Behe’s book is spent telling us just how damnably unlikely a two-amino acid change is. Does talking about regulatory sequences rather than coding sequences justify a radical change in the conceptual likelihood to Behe?
You might object at this point that I seem to be impossible to pleas: Mutations are too rare, when we look at chloroquine resistance to malaria, but now they are too common, when we look at theoretical possibilities for all these genetic switches. Here’s the problem: So many kinds of switches are so common that, if they were the most important factor in determining whether a gene was turned on, the organism would be an incoherent mess. Instead of a fly or a sea urchin or a frog, a developmental program might at best produce a blob of tissue.
Why yes, and bumblebees ought not to be able to fly. That kind of conclusion ought to be interpreted to mean your understanding of the process is incomplete or wrong, not that you’ve discovered a flaw in reality.
Behe’s error is that he has again imposed his black-and-white mentality on the issue. There are lots of switches, and a switch is either in a good or bad position, and therefore if any one of them is bad, kerbloooiee, the animal turns into a blob…therefore, evo-devo is refuted. It’s so wrong that I had to laugh.
Yes, there are a large number of these switches associated with every gene, but most of these switches will not be ultimate activators or repressors—we already know that a single switch is usually not the most important factor in determining whether a gene is turned on, but they will instead contribute incrementally to the pattern of expression. Carroll uses the example of the BMP-5 gene, which has different switches to regulate its expression in the ribs, the omosterna, the thyroid cartilage, and the outer ear. Mutations to a switch don’t cause global collapse of BMP-5 gene expression, or ubiquitous, constant expression everywhere in the body—they fine-tune the pattern of expression, turning it up or down in small regions.
Think of them as more like a mixing board. The sound control guy at an event has a board with a bewildering array of switches and dials and knobs, and sure, there’s an on/off switch—but most of those controls are for tweaking the sound levels at various frequencies or at various speakers around the room. If one turns up the amp on the left midrange speaker, or amplifies the vocalists microphone a bit more than the bass guitarist’s, it does not mean that the entire sound of the room collapses into incoherence.
“Incoherence” is Behe’s new excuse. He admits that he’s got a bit of inconsistency in his explanations, claiming that simple mutations almost never happen, but then claiming that simple mutations in regulation happen all the time, but he’s simply going to wave his hands and say that the evo-devo explanations therefore only produce incoherence, whatever that is. The obvious refutation, though, is that if regulatory mutations are common, as they are and as Behe is forced to admit, and if embryos develop just fine anyway, with just minor variations (which, of course, are then amenable to selection), isn’t it clear that these changes do not produce “incoherence”? The only thing collapsing into an amorphous blob here is Behe’s position.
One last argument from this chapter that I’ll discuss is a particularly tired one that I’ve heard all too commonly from creationists. It’s the argument from surprise.
The argument from surprise is simple. Search through the scientific literature and find instances where scientists express surprise at an unanticipated result — and such instances are everywhere; one doesn’t become a scientist because one knows all the answers, but because one doesn’t — and then use that to declare that therefore their knowledge was grossly incomplete and their guiding theories inadequate. Good theories would give you perfect predictability, right? Scientists would never have to do experiments!
It’s annoying when rank creationists who claim the literal infallibility of the Bible do it — they, after all, believe they have complete and perfect knowledge — but for a purported scientist like Behe to play this game is simply contemptible. He should know that a science that is partly historical like evolution is going to be rich in contigencies, and that sciences that deal with complexities above a few simple elements, you aren’t going to ever see simple solutions. Ask a physicist about the n-body problem; does an expression of surprise at a particular solution mean that physics must be all wrong?
Behe goes on and on about one of the “surprising” results of molecular genetics: that we’ve found a greater degree of conservation of developmental mechanisms than many scientists expected. For instance, it was not predicted that all animals, from sea anemones to teleosts to beetles, would have a core module of Hox genes that regulated pattern formation along the longitudinal axis. Behe accurately quotes many biologists who expected that the rules of development for different body plans would be very different, and who were then surprised when these fascinating genes that Ed Lewis analyzed in flies were so nearly ubiquitous in the metazoa.
“Surprise” is such a nonspecific word, though. There is the kind of surprise you get when you’re unexpectedly laid off from your job and go home early to catch your spouse in bed with the hot UPS delivery person, and then there’s the kind of surprise that a six-year old gets on Christmas morning when she discovers Mom and Dad got her that terrific Lego kit instead of a pair of socks. Developmental biologists experienced something analogous to the latter, rather than the former, although Behe strains mightily to turn the Hox genes into an ugly pair of socks for us.
Before we actually had the detailed molecular data, many scientists expected there would have been so much churning of genomes with mutations that the similarities between distant species would have been obscured. Strangely, though, Behe claims that “reasoning straightforwardly in terms of Darwin’s theory” is what led them astray, which is a bit incomprehensible—Darwin said nothing about molecular biology and very little about genetics (and what he said there was often wrong). We could not have a theory about the pattern of change at the molecular level until, well, we had the data about the molecules in hand. Once we did, then the theories started to build; prior speculation in the absence of evidence is not an indictment of current explanations which are backed up by data.
Now what we’ve got is the evidence, which shows a strong and enduring genetic relationship between different organisms, and what evo-devo is is an effort to explain the developmental mechanisms behind that observation. Behe should understand that; one of his major points is that evidence for common descent, which he accepts, is not in itself evidence for the mechanisms that generate differences between species. He claims the mechanism is intelligent design, but provides no evidence for that. Evo devo, and all of modern biology, says that the mechanisms are natural processes, and the scientific literature is pretty much a massive compendium of nothing but the evidence.
One last bit of contemptible smugness from Behe, and then I’m going to toss his book into my archive of creationist junk.
…it is fascinating to note that the appearance of Hox toolbox components seems to have significantly predated the appearance of new animal forms. As Sean Carroll remarks:
The surprising message from Evo Devo is that all of the genes for building large, complex animal bodies long predated the appearance of those bodies in the Cambrian Explosion. The genetic potential was in place for at least 50 million years, and probably a fair bit longer, before the large complex forms emerged.
Another surprise to the Darwinists! To an intelligent design proponent such as myself, this is a tantalizing hint that parts were moving into place over geological time for the subsequent, purposeful, planned emergence of intelligent life.
Intelligent Design creationists made no such prediction, and their whole sloppy discipline can say anything, and therefore nothing. It is the height of arrogance for Behe to claim he knew it all along; he’s lying, plain and simple.
Furthermore, the surprise of the developmental and evolutionary biologists is partly the fault of the complexity of our disciplines, and the difficulty of getting the big picture of the whole of biology. Our failures to recognize the possibility of greater preceding complexity isn’t a fault with biology, but our own specialization—the microbiologists have been saying for years that there is immense complexity and sophistication in the bacterial world, and we ought to be paying more attention to it. It’s a fair cop to accuse the evo-devo crowd of some degree of nearsightedness, but that is no indictment of evolutionary biology.
That principle of early complexity is now fairly common, as Carroll states—another example is the recent work finding proteins associated with synapses in sponges. However, unlike the Intelligent Design creationists, we do not make the stupid mistake of assuming bacteria and protists were only bearing this complexity because their metazoan descendants would need it—an assumption that is not only arrogant, but that contradicts basic understanding of how molecular evolution works, since it assumes a lineage would retain a “potential” attribute for 50+ million years without degradation—but instead seek explanations that have some respect for the biology. Sponges and choanoflagellates have “post-synaptic proteins” not because mammals need them, but because sponges and choanoflagellates have functional needs for complex cell-cell associations, and more complex multicellular organisms have merely coopted that general functionality for more specialized purposes.
What Behe needed was some evidence for his “purposeful, planned” mechanism, and one thing that did not surprise me at all, that we’ve come to expect from these frauds, these shallow poseurs, and these superstitious corrupters of education and science, is the utter vacuity of their claims and the complete absence of any kind of evidence for their assertions. What is particularly disappointing from someone with scientific training, like Behe, is that their claims are so routinely contradicting the facts of biology. From his phony arguments about the probability of mutations to his misrepresentations of developmental biology, Behe is in defiance of the evidence. The whole book is fundamentally wrong, and the only two possibilities are that Behe is oblivious to his ignorance, or he is aware of the errors and is cynically selling a work of pseudo-scholarship to that large audience of deluded creationists who will welcome it.
Carroll SB (2007) God as genetic engineer. Science 316(5830):1427-1428.