More Articles on Evolution

H. Allen Orr Responds

Michael Behe, Phillip Johnson, and David Berlinski have pulled off a remarkable feat. They've made squabbling over creationism seem almost intellectually respectable. Previously, an air of the third rate hung over the whole business and any scientist who looked too deeply into creationism had to work hard to shake the feeling of intellectual slumming. No longer.

The cause of the change isn't hard to find: Johnson, Berlinski, and especially Behe pack serious credentials. Their jobs are (or have been) at real universities, their books reside in real bookstores, and their views get aired in real places like the New York Times, Commentary, and Boston Review. There's a swagger in their step and a punch in their prose that was missing from the works of their less sophisticated predecessors. The result of all this is respectability--and revived hopes that those Darwinist materialist killjoys are all wrong about this evolution stuff.

All this is obvious. What is less obvious is that these new-wave anti-evolutionists disagree among themselves on just about everything. They don't go out of their way to broadcast this fact, but take a look: Behe thinks the primordial cell was designed, but that old-fashioned evolution took over from there, producing all the species we now see. But Johnson questions almost all evolution, including common descent and the fossil record that Behe trusts. Behe thinks the data point to an intelligent Designer, while Berlinski claims he's no creationist, he merely questions the sufficiency of Darwinism. Similarly, Behe has no problem with natural selection, while Berlinski seems to have no problem with anything but natural selection.

These differences have two consequences. One is that the popular impression of some united intellectual front that's risen up against Darwin is an illusion. (And one that's been carefully fostered by Behe, Berlinski and Johnson, who routinely write in vague support of each other.) The other is that it's hard to respond to these gentlemen en masse. Criticisms that stick to one won't stick to the others and we're obliged to consider Behe, Johnson, and Berlinski separately. Because my review of Behe's book kicked off this whole affair, and because he's gotten the most attention, I'll focus on him. I'll then briefly turn to Johnson and Berlinski, touching along the way on the comments of several other respondents.

When the author of a book has his errors pointed out to him, he has several options. One is to say, "I was wrong," but this is never terribly popular and may be too much to ask of mere mortals. A more common option is to talk about everything but your critic's main point. This is the tack Behe takes. My main criticism of Behe's book was simple. He claims that many biochemical systems are "irreducibly complex," i.e., all the parts are required for the system to work. I had no problem with this. My heart needs valves and a pump to work and I'm not surprised that the same kind of tight interdependence occurs at the molecular level. But Behe further claims that irreducibly complex systems cannot evolve by natural selection. This is the chief claim of his book and the assertion that propelled him into Newsweek et al. and into the arms of the intellectual right. And it's just wrong. As I pointed out, gradual Darwinian evolution can easily produce irreducible complexity: all that's required is that parts that were once just favorable become, because of later changes, essential. We may well wind up with a structure or a biochemical pathway in which all parts are needed, but every step of the way was Darwinism. Russell Doolittle and Douglas Futuyma, in their responses, provide several beautiful examples of this process, e.g., the evolution by gene duplication of the globin family of proteins, all of which are now required in humans.

So how does Behe respond to this criticism? Well, he doesn't. He eagerly talks about everything else--what evolutionists versus biochemists have achieved, who's the podiatrist and who's the brain surgeon--but when it comes to telling us why the Darwinian explanation of irreducible complexity is wrong, he's astonishingly silent. The closest he gets is to trot out his favorite fancy object, the mousetrap, for our renewed consideration. Parodying my description of Darwinism--we start with some part A that does some job; a part B then gets added on that helps A; A then changes in a way that makes B essential--he writes, "Some part does some job? Which part of the mousetrap is [Orr] talking about? A mouse has nothing to fear from a `trap' that consists of just an attaching holding bar, or spring, or platform, with no other parts." This is the sum total of his response to my chief criticism of his book.

But it's just substantive enough to betray Behe's colossal misunderstanding of Darwinism. For under the Darwinian scenario, all the parts can change through time and there's no reason to think we started with anything like a holding bar, spring, or platform. Indeed this is the whole point of the scenario: no single current part can do the job, so none could possibly represent the ancestral system. Instead most or all of the parts likely changed through time, growing, in the process, more interdependent. Behe's failure to get this point represents a fundamental--and fatal--error which likely explains his refusal to buy evolutionary accounts of biochemistry. We may well see here the source of Behe's whole misguided campaign against Darwinism.

Our argument does not, of course, show that complex objects are never designed (the mousetrap was); it just shows that design is not the sole and so necessary explanation of irreducible complexity. But if design isn't necessary, is it at least plausible? Behe thinks so, assuring us that design "is both natural and obvious." His argument is straightforward: "Just as in the everyday world we immediately conclude design when we see a complex, interactive system such as a mousetrap, there is no reason to withhold the same conclusion from interactive molecular systems," even though such a hypothesis might have "theological implications." I hate to be a party-pooper, but it seems to me there's a pretty good reason why the design hypothesis is a bit more "natural and obvious" when considering a mousetrap than a cell: We know that there are people who make things like mousetraps. (I'm not being facetious here--I'm utterly serious.) When choosing between the design and Darwinian hypotheses, we find design plausible for mousetraps only because we have independent knowledge that there are creatures called humans who construct all variety of mechanical contraptions; if we didn't, the existence of mousetraps would pose a legitimate scientific problem. Needless to say, we have considerably less independent evidence for a Tinkerer who spends His days soldering cells. As it stands, then, mousetraps and cells are far from analogous and the hypothesis of intelligent design of cells remains distinctly supernatural and unobvious.

But if design isn't terribly plausible, is it at least falsifiable? Jerry Coyne argues that it is not: Because Behe accepts some evolution, showing that a particular cellular process evolved cannot put him out of business as he can always claim that some other pathway was designed. But Behe denies this. He now tells us: "To falsify design theory a scientist need only experimentally demonstrate that a bacterial flagellum, or any other comparably complex system, could arise by selection." I find this statement extremely curious as, in his book, Behe plainly admits that some cellular processes could have evolved by natural selection. If all those other cases didn't cause Behe to surrender his pet theory, why should one more?

In the end, though, we mustn't let all this agonizing over necessity, plausibility, and falsifiability distract us from the real point about Behe. For that point is simpler: Behe claimed to find a phenomenon that can't be explained by Darwinism and he was wrong. But I will give him this: Behe knows a great deal about the arcane inner-workings of the cell and he seems to appreciate that, in the long run, it's a theory's ability to explain these nuts and bolts--these hard facts--that counts. Presented with overwhelming evidence, you can therefore at least imagine him changing his mind (though I wouldn't advise any wagers).

Behe's comrade-in-arms, Phillip Johnson, seems, by comparison, past praying for. Johnson's commentary reveals that his opposition to Darwin is entirely ideological. His beef, it turns out, is not so much with Darwinism as with materialism. Indeed his whole view can be summed up in the following syllogism: Materialism is bad; Darwinism is a form of materialism; ergo, Darwinism is bad.

He tries to persuade you of the soundness of this logic by reminding you that Marxism and Freudianism--both materialist--are also bad. But this guilt by association maneuver is silly and the absurdity of Johnson's view can be seen by plugging any other species of materialist science into his syllogism: "Celestial mechanics is a form of materialism; ergo, celestial mechanics is bad." Is Johnson really willing to believe that--in the imminent collapse of materialism--mechanics will go the way of phrenology and we'll all happily return to the pre-materialist view that angels, by flapping their wings, propel the planets through their orbits? If not, why not? Why is it kosher to reject Darwinism on the grounds of materialist dogma, but not celestial mechanics? The answer is, I think, obvious. Darwinism hits closer to home. It concerns the origins of people, not the orbits of planets.

It's clear that Johnson longs for a return to some idyllic pre-materialist culture. But this is escapist nonsense, a cop-out. The fact is we live in a scientific age. To deny this--to wish it away by chanting "materialism"--is to live in a make-believe world. Johnson is of course welcome to reside in such a world, but we, his readers, are left with the slightly incongruous image of a man who, flatly denying half the world and the last four hundred years of history, lectures us on how things really are.

As for Johnson's treatment of my paper with Coyne on the genetics of adaptation, Coyne has said most of what needs saying. Behe, and now Johnson, have grossly exaggerated the "challenge" to Darwinism posed by our work. How the notion that adaptation involves larger-sized mutations than we thought pulls the rug out from under evolution is beyond me. Indeed Johnson misunderstands our claim, getting it all backwards. Coyne and I noted that population genetic theory shows that bigger mutations are the most likely to play a role in adaptation. But Johnson, pointing to our paper, concludes that, although big mutations happen, "they can't climb Mount Improbable again and again," i.e., they can't play a role in adaptation. This is precisely the opposite of our conclusion. Such misunderstandings don't inspire much confidence in Johnson's grasp of evolutionary theory. But this is subtle stuff compared to his assertion that I invoke "a naked hypothesis (Muller's Ratchet) against irreducible complexity." Since Johnson is viewed in some quarters as an expert on evolutionary matters, I feel somewhat obliged to point out that "Muller's ratchet"--though well-known and important--has nothing whatever to do with Muller's explanation of irreducible complexity. Muller had more than one idea in his life and Muller's ratchet refers to the accumulation of deleterious mutations on asexual chromosomes--a long ways from irreducible complexity. Drawing a conclusion here seems almost rude, but too much is at stake for polite silence: the fact is Johnson has little idea what he's talking about.

Last, I must comment on Johnson's astonishing claim that "in fields like paleontology, genetics, and embryology . . . the empirical evidence and the materialist project are going in opposite directions." I sincerely hope no reader is taken in by such nonsense. I work in a department filled with geneticists, embryologists, and paleontologists and not one of them yet has stopped me in the halls, pale and trembling, with revelations of data that undermine materialism. (You should ask yourself what such data would look like.) I'm afraid that, if Johnson is your guide to evolution, paleontology, genetics, or embryology, there's a good chance that you and the empirical evidence are going in opposite directions.

And now for the last of our anti-evolutionists, David Berlinski. Because Berlinski's padlock is the same as Behe's mousetrap--right down to the error of assuming that the first part looked like a current part--I needn't waste your time with it. Berlinski's more interesting claim is that Darwinism is unfalsifiable. Behe makes a similar charge. By claiming that Darwinism can yield irreducible complexity, I seem to claim that Darwinism can explain anything. But, as Berlinski notes, "some objects must be inaccessible to a Darwinian mechanism" if Darwinism is to be taken seriously as science. I absolutely agree. Darwinism must not facilely explain away any imaginable observation--there must be phenomena that are "inaccessible." And there are. Darwin's classic example was a character that increases the fitness of some second species at the carrier species's expense. (And if this isn't good enough for you, there's always J. B. S. Haldane's example: a Precambrian rabbit.) But Behe pointed to no such thing. He pointed to irreducible complexity. And irreducible complexity is manifestly not a phenomenon that is inaccessible to Darwinism. So the answer is simple: Darwinism can be falsified, but Behe hasn't done it. Surely Berlinski can appreciate the distinction.

Although I've touched on the comments of several biologists who largely agree with me, two scientists who more or less disagree with me, Robert DiSilvestro and James Shapiro, were also invited to respond. Although DiSilvestro simply seems to misunderstand the Darwinian account of irreducible complexity, Shapiro's comments are weightier. But before discussing where we part company, it's important to see that Shapiro does not support any creationist account of life. He merely claims that orthodox Darwinism might not be sufficient to explain all molecular biology. The proper explanation will, he thinks, be naturalistic, just having more to do with information theory than with Darwinism.

This is a smaller debate and Boston Review is not the place for it. But Shapiro also accuses evolutionists of a close-mindedness about evolutionary mechanism, and this requires comment. I agree with him that "[d]ogmas and taboos . . . have no place in science." Evolutionary biology is not immune to profound change and it would be sheer hubris to think that evolutionists have answered all or even most of the questions facing us. (It would also be boring. Who wants to work in a field that's all wrapped up?) But there is a related, though mirror-image, danger in thinking that each new fact that gets uncovered poses some profound intellectual challenge to Darwinism. Although Shapiro's litany of technical advances is interesting (though well-known), I doubt that few, if any, of these findings pose much challenge to Darwinism. (For example, the initial wild claims that mobile genetic elements play a novel role in adaptation or speciation have been thoroughly discredited. Surely Shapiro knows that successful Darwinian accounts of these "selfish" mobile elements were offered long ago.)

The real irony is that Shapiro, DiSilvestro and, of course, Behe all think the molecular world poses the most profound challenge to Darwinism. But if one had to point to the one field of evolutionary biology that has been most thoroughly mined, it would likely be molecular evolution. Molecular evolution has dominated--indeed virtually monopolized--the study of evolution for the last forty years. And it has been overwhelmingly successful. (The fact that you don't know about this probably has something to do with Gould and Dawkins not being molecular types.) Indeed there's a growing sense that many of the deep questions that first tempted evolutionists to think hard about molecules are on the verge of getting wrapped up. And there's a similar feeling that many of the questions that got left behind in the rush to molecules--questions about the "phenotype," the way an organism looks and why--remain profound and unanswered. So if you want to see where evolutionary biology faces it greatest difficulties, I, for one, suggest that you ignore Behe et al. and look up, not down.

---