wiki(moderated)
May 31, 2005
Michael Behe
As Ernst Mayr pointed out in One Long Argument (p. 36):
That writers on Darwin have nevertheless almost invariably spoken of the combination of these various theories as “Darwin’s theory” in the singular is in part Darwin’s own doing. He not only referred to the theory of evolution by common descent as “my theory,” but he also called the theory of evolution by natural selection “my theory,” as if common descent and natural selection were a single theory. . . . [Darwin] ascribed many phenomena, particularly those of geographic distribution, to natural selection when they were really the consequences of common descent.
Similarities between structures are, at best, evidence for common descent, but do not themselves say anything at all about RM/NS. In order to show that Darwin’s mechanism of RM/NS is responsible for either a new structure or a new function for an old structure, one has to show, as Darwin insisted, that there is a pathway leading to the structure by “numerous, successive, slight modifications”, each of which is an improvement for the organism. That is a much, much more difficult task than simply showing similarities between features. Mayr showed that, while most early scientists agreed with Darwin’s idea of common descent, few agreed with him on natural selection.
The much greater difficulty of showing how a system might develop by “numerous, successive, slight modifications” versus simply showing similarities in structures is illustrated in the tale of Russell Doolittle and the blood clotting cascade, which I have discussed more fully in other places (for example, here and here). Briefly, Professor Doolittle, an expert on blood clotting, wrote an essay in 1997 in Boston Review in which he asserted there was experimental evidence from laboratory mice refuting my claim in Darwin’s Black Box that the blood clotting cascade was irreducibly complex. It turns out, however, that he had misread the paper he was citing, and the lab mice’s altered clotting system was broken. The most important point to note, I think, from Doolittle’s error is that it showed clearly he did not know how Darwinian mechanisms produced the clotting system. If he did, he would simply have said so in the essay, or else would have cited a paper (or, rather, a long series of papers) where the system had been accounted for by “numerous, successive, slight modifications”. And if an expert such as Professor Doolittle does not know, then nobody knows.
Another important conclusion we can draw from the error is that protein sequence data are insufficient to tell us how a Darwinian mechanism might have produced a complex molecular system. Many of the proteins of the clotting system are quite similar in sequence and structure to each other and also similar to other, non clotting proteins. Professor Doolittle knew all about the structural and sequence similarities. Yet that knowledge did not allow him to say how RM/NS could have put together the clotting system. The structural similarity may point to the proteins being derived by descent from a common precursor protein, but it says nothing about Darwin’s mechanism of RM/NS.
A third important lesson to draw from Professor Doolittle’s mistake is that Darwinists often have extremely low standards of evidence to support their theory. Any remotely, superficially plausible account for the origin of system is taken by the Darwinian faithful as earth-shaking evidence that unintelligent mechanisms can do the trick. And anyone skeptical of the just-so story is often derided as either dumb or as having vaguely nefarious motives.
Although Professor Doolittle made his illuminating mistake a long time ago (eight years now -- time flies!) the lessons continue to be relevant. Two very recent essays make the same, glaring mistake of taking evidence of common descent to be evidence of RM/NS. The first essay was by University of Rochester evolutionary biologist Allen Orr in The New Yorker, who wrote:
Biologists actually know a great deal about the evolution of biochemical systems, irreducibly complex or not. It's significant, for instance, that the proteins that typically make up the parts of these systems are often similar to one another. (Blood clotting -- another of Behe's examples of irreducible complexity -- involves at least twenty proteins, several of which are similar, and all of which are needed to make clots, to localize or remove clots, or to prevent the runaway clotting of all blood.) And biologists understand why these proteins are so similar. Each gene in an organism's genome encodes a particular protein. Occasionally, the stretch of DNA that makes up a particular gene will get accidentally copied, yielding a genome that includes two versions of the gene. Over many generations, one version of the gene will often keep its original function while the other one slowly changes by mutation and natural selection, picking up a new, though usually related, function. This process of "gene duplication" has given rise to entire families of proteins that have similar functions; they often act in the same biochemical pathway or sit in the same cellular structure. There's no doubt that gene duplication plays an extremely important role in the evolution of biological complexity.
However, as I pointed out in Darwin’s Black Box years ago (pp. 89-90), while gene duplication does indeed occur, and similar proteins may be descended from an ancestral one, none of that says anything at all about how a complex system could be put together by “numerous, successive, slight modifications” by RM/NS. That is an entirely separate question which needs much additional evidence to support it. Russell Doolittle knew all about gene duplication, yet couldn’t explain how Darwinian processes produced the clotting cascade. And if Russell Doolittle doesn’t know how the clotting cascade could have been put together in a Darwinian fashion, then Allen Orr certainly doesn’t either, nor does anyone else on earth.
The second recent essay was posted at the Panda’s Thumb blog by Andrea Bottaro, a Professor of Medicine who is also at the University of Rochester. Professor Bottaro pointed to a recent paper which finds some sequence homology between RAG1 (an enzyme which is involved in recombination of antibody genes to generate extensive diversity) and transposases from the Transib superfamily of transposons. (Transposons are elements of DNA that code for enzymes -- transposases -- that can either move the DNA element to a new location in a genome or place new copies of the DNA elsewhere.) That of course is a very interesting result, and may say something about the ancestry of the protein, but it does not say anything at all about RM/NS. Bottaro also cites earlier interesting work showing some mechanistic similarities between RAG1and some transposases. However, there are far greater mechanistic similarities between components of the blood clotting cascade and, as Doolittle’s mistake makes clear, such similarities simply don’t tell how RM/NS could have built such a system.
Professor Bottaro, perhaps sensing that the paper he cites won’t be persuasive to people who are skeptical of Darwinian claims, laments that “Behe and other ID advocates will retreat further and further into impossible demands, such as asking for mutation-by-mutation accounts of specific evolutionary pathways...” Well, yes, of course that’s exactly what I ask of Darwinian claims -- a mutation-by-mutation account of critical steps (which will likely be very, very many), at the amino acid level. But that’s neither a “retreat” (In Darwin's Black Box (page 176)I implied that many small details would be necessary for a real Darwinian explanation) nor is it unreasonable -- that’s simply what’s necessary to actually explain the appearance of a complex, functional system in a Darwinian fashion, to show that it could indeed happen as Darwinists claim. Proteins change single mutation by single mutation, amino acid by amino acid, so that’s the level of explanation that is needed. What part of “numerous, successive, slight” is so hard to understand?
And not only a list of mutations, but also a detailed account of the selective pressures that would be operating, the difficulties such changes would cause for the organism, the expected time scale over which the changes would be expected to occur, the likely population sizes available in the relevant ancestral species at each step, other potential ways to solve the problem which might interfere, and much more. Alternatively, Darwinists could present a series of experiments showing that RM/NS is capable of building a system of the complexity of the adaptive immune system.
Professors Orr and Bottaro seem to think that because Darwinists’ fantastic claims are very difficult to support in a convincing fashion, then they should just be given a pass, and that everyone should agree with them without the required evidence. As Russell Doolittle helpfully showed, Darwinists find it easy to imagine that evolution could proceed along pathways which nature would never allow. Like Calvin and Hobbes, in their imaginations they hop into a box and fly over treacherous evolutionary terrain that nature would have to try to cross on foot. There is no reason for skeptics to trust Darwinists’ imaginations.