Miller and Behe on Origins

Guest response to Ken Miller's review of Darwin's Black Box Published at ARN Forum

Ken Miller is a scientist who has been taking the lead as a critic of ID. Thus, I took a look at his review of Behe’s book, Darwin’s Black Box, found at

There is really nothing in this review that I have not already replied to in [the ARN] forum. But there is one thing worth exploring in more depth. Miller writes:

If Behe’s formulation of intelligent design as science is illogical, his mechanism for how the work of the designer was inserted into living systems is almost laughable. Remember that Behe accepts the validity of the geological ages and the fossil record — an open-minded scientist can hardly do otherwise — and yet he claims that the complex biochemical systems he extols were fashioned by an intelligent agent. When did this agent go to work, and when were the genes encoding them engineered? He has an answer ready:

“Suppose that nearly four billion years ago the designer made the first cell, already containing all of the irreducibly complex biochemical systems discussed here and many others. (One can postulate that the designs for systems that were to be used later, such as blood clotting, were present but not “turned on.” In present-day organisms plenty of genes are turned off for a while, sometimes for generations, to be turned on at a later time.)” (11)

This means that billions of years ago a humble prokaryote was packed with genes that would be turned off for hundreds of millions of years before they produced the eukaryotic cilium, and genes for blood clotting proteins that would pass more than a billion inactive years in genetic “cold storage.” And what happens during those billions of years? As any student of genetics will tell you, because those genes are not expressed, natural selection cannot weed out genetic mistakes. This means that mutations will accumulate in these genes at breathtaking rates, rendering then hopelessly changed and inoperative hundreds of millions of years before Behe says that they will be needed.

As a reviewer and antagonist of Behe, it is ironic that Miller so clearly misses the point (and this misrepresents Behe). That is, just like the creationists are so often accused of doing, Miller lifts an excerpt out of context and makes it say something else. Behe did not offer this speculation as “his mechanism for how the work of the designer was inserted into living systems.” Anyone who has bothered to read Behe’s book knows that Behe’s speculation is raised as a response to one of Miller’s bold claims. Behe quotes Miller as saying: “The theory of intelligent design cannot explain the presence of non-functional pseudogenes unless it is willing to allow that the designer made serious errors.”

Behe then raises a possibility to counter Miller’s “it cannot explain” charge. Miller misses the point entirely and tries to turn this argument into the mechanism Behe proposes.

But the irony only increases when we realize that Miller’s cause for laughter seems to be missing elsewhere, where he makes essentially the same argument to support his views of evolution. Miller published the following argument elsewhere:

No living bird has teeth, and that fact, of course, is behind the old saying that a rare object is “as scarce as hen’s teeth.” Why don’t birds have teeth? A proponent of intelligent design must answer that they have not been designed to have teeth, quite probably because the designer equipped them with alternatives (hard beaks and food-grinding gizzards) that are superior for lightweight flying organisms.

Is this in fact the case? In 1980 Edward Kollar and his colleague C. Fisher decided to test whether or not chicken cells still have the capacity to become teeth. Intelligent design would predict that they cannot, because teeth were never designed into the organism.

Kollar & Fisher’s experiment was simple. They took mouse tissue that normally lies just beneath the epithelial cells that develop into teeth, and put it in contact with chick epithelial cells. What happened? The chick cells, apparently influenced by the mouse tissue, dutifully began to develop into teeth. The produced impact-resistant enamel on their surfaces, and developed into clear, recognizable teeth (Figure 5). The experimenters took great care to exclude the possibility that mouse tissue had produced the teeth, first by making sure that no mouse epithelium was included in the experiments, and second by confirming that the cells in the tooth-producing tissue were indeed chick cells. Their experiments have since been confirmed by two independent groups of investigators.

No plan of intelligent design can account for the presence of tooth-producing genes in chicken cells. Indeed, it would be remarkably un-intelligent to endow birds with such useless capabilities. Evolution, on the other hand, has a perfectly good explanation for these capabilities. Birds are descended from organisms that once had teeth, and therefore they may retain these genes, even if other genetic changes normally turn their expression off. In short, birds have a genetic mark of their own history that no designed organism should ever possess. Designed organisms, after all, do not have evolutionary histories.

Now, it [is] not my intention to build on these experiments [since there is some reason to doubt the veracity of the experiments]. My main point here is that Miller can’t have it both ways. Last time I checked, biologists agree it has been quite a long time since birds had teeth. Yet Miller maintains these teeth-genes have been “retained?” Is this really that different from Behe’s speculation? In one context, Miller writes, “As any student of genetics will tell you, because those genes are not expressed, natural selection cannot weed out genetic mistakes. This means that mutations will accumulate in these genes at breathtaking rates,” yet now we are told that evolutionary theory perfectly explains experimentally-induced teeth genes in chickens as these non-expressed genes have been retained for a very long time. So which is it? Is there something wrong with these chicken experiments or should we begin laughing at Miller’s perfect explanation?

While Miller may laugh at Behe’s speculation, I find such laughter to be quite smothering as we ponder about our complicated and ambiguous reality. For Behe’s speculation is interesting and does receive a little support from the chicken experiments Miller cites. Taken together with Miller’s problem, what we get is an interesting design problem–how do we design an IC system so that it remains intact for long periods of time without the help of selection (functional constraint)? This would constitute an example of front-loading evolution (one of the explanations in the teleological tool kit).

Furthermore, Miller incorrectly assumes that Behe’s thesis entails a “humble prokaryote was packed with genes that would be turned off for hundreds of millions of years.” Instead, we can propose a community of cells. Such cells could have originated as eukaryotes (as some scientists think) or included both eukaryotes and prokaryotes.

Now, such front-loading need not be the general solution to IC’s origin, as it may apply only in certain cases.

Also, we need to remember that the problem of accumulating genetic mistakes can be buffered to some extent by classical notions in biology.

a. Not all nucleotide substitutions result in amino acid substitutions. And not all amino acid substitutions are deleterious. Thus, one could design IC components such that they tolerate changes, explaining why it is rare that IC components have very strong sequence conservation across phylogenetic lines (although they are conserved at a functional and structural level).

b. Secondly, we’re dealing with populations, not isolated genes. Since such mutations would be neutral, they would likely be lost once they appear (standard population genetics). Behe’s thesis does not require that all the unexpressed IC machine genes accumulate no deleterious mutations. No, only one set needs to get through (like a bottle neck). That some might get through might be reflected in a common pattern in evolution – rapid explosions/radiations stemming from a monophyletic origin. That is, if certain IC machines turn on in a physiological context that provides immediate benefit, we’d expect such rapid explosions in accord with standard selectionist views. This would be consistent with the manner in which several biologists think evolution is dictated more by the genome than the environment.

c. A designer could help facilitate this type of evolution using several other strategies.

i. Physically and closely link the non-expressed IC genes to crucial housekeeping genes. This would slow down loss through deletions.

ii. Employ DNA repair mechanisms that are more stringent than what is seen today. In fact, these non-expressed genes could be linked to genetic signals that attract a subset of the repair machinery. This ID hypothesis would predict that DNA repair may have been originally more efficient, but has actually relaxed over evolutionary time (where selection has picked up the slack).

iii. Express some of the IC genes by giving them multiple functions. Pax6 is a potentially instructive example, as it is an essential gene in eye development. Yet it has other functions (even being expressed in worms with no eyes). By giving it a multiple function, selection can retain it until it can be used in another intended context. The interesting thing about this explanation is that it steals the co-option explanation from the non-teleological view and shows again there is more than one way to interpret the data. Actually, I think this is a very interesting interpretation and need to give it more thought. After all, the co-option explanation (like gene duplication) in a non-teleological explanation can only go so far, as some original state must have existed.

Multiple functions may even come in the form of endowing RNA genes with protein-coding sequence. A few months ago, I raised some evidence (albeit weak) concerning t-urf13 to indicate this might have originally been the case.

Although these are thoughts from the top of my head, I find them to be interesting. And that someone like Miller may laugh at them is okay with me. It simply means that while Miller and others continue to force all the data into one explanation, they leave all kinds of opportunity for truly creative and stimulating thought.

For more discussion, go to: