Previously I noted that Stephen Meyer and Charles Marshall participated in a radio debate over Darwin’s Doubt which was broadcast this past weekend. I highly recommend, listening to it. As I wrote yesterday, both participants conducted themselves well in what turned out to be an extremely interesting, civil and thoughtful exchange.
Because the debate lasts nearly an hour and a half, and is well worth listening to in its entirety, I thought it would be useful to provide a short guide to help listeners focus on the main argument of the debate and to follow the thread of that argument as it pops up again and again, amidst many other topics
Recall that in Darwin’s Doubt Meyer argues that intelligent design provides the best explanation for the origin of the genetic and epigenetic information necessary to build the novel forms of animal life that arise in the Cambrian period. In support of that argument, Meyer offered five separate critiques of the ability of the mutation/selection mechanism to generate that necessary information. He also later critiqued six post neo-Darwinian evolutionary mechanisms and made a positive argument for intelligent design by showing that intelligent agents alone have the power to produce large amounts of functionally specified digital information.
In an opening interview with host Justin Brierley at the beginning of the debate, Meyer took time to explain one of his main critical arguments against the creative power of the neo-Darwinian mechanism. In particular, he developed an illustration of a thief desperately trying to find by chance the correct combination to a ten-dial bike lock with a security guard rapidly approaching. Meyer used this illustration to explain why random mutational changes are overwhelmingly more likely to fail than to succeed in generating even a single functional gene or protein in available evolutionary time. Like Meyer’s hypothetical dial-spinning thief, the evolutionary process simply has too many possible sequences to search in the available time to have a realistic chance of generating a desired sequence — a new functional gene or protein — let alone the many functional genes and proteins that would be necessary to build a whole new animal form.
In his review of Darwin’s Doubt in Science, Marshall attempted to rebut this argument, not by demonstrating that any known evolutionary mechanism can produce the information necessary to build the Cambrian animals, but instead by contesting the idea that a significant amount of new genetic information would have been necessary to build these animals. In particular, Marshall claimed that the evolutionary process could have “rewired” the gene regulatory networks that control how other “already existing genes” are expressed during the process of animal development.
As Marshall argued in his review:
[Meyer’s] case against the current scientific explanations of the relatively rapid appearance of the animal phyla rests on the claim that the origin of new animal body plans requires vast amounts of novel genetic information coupled with the unsubstantiated assertion that this new genetic information must include many new protein folds. In fact, our present understanding of morphogenesis indicates that new phyla were not made by new genes but largely emerged through the rewiring of the gene regulatory networks (GRNs) of already existing genes.
During the debate, Marshall amplified this argument by claiming that Meyer’s argument presupposed an outdated “1980s model of the way genes operate” and that his book “confronted a different set of problems that hark back to an older age.” According to Marshall, biologists no longer believe that building the diverse forms of Cambrian animals would require evolving new genes (or, at least, many new genes). Instead, Marshall argued, again, that new body plans could be generated by rewiring networks of already-existing genes, especially those which are part of the developmental gene regulatory networks (dGRNs) that control the timing and expression of pre-existing genes during animal development. Marshall pointed out that animals have far fewer genes than we once expected, and that today it is thought that “animals use essentially the same genes, just deployed slightly differently.” By changing the deployment of those genes — by rewiring their dGRNs — Marshall thinks new body plans can arise.
Meyer had a ready response. After pointing out that he did not presuppose a 1980s gene-centric understanding of animal development, but that he had examined and discussed in Darwin’s Doubt the hypothesis that dGRNs played a major role in the evolution of animals, Meyer noted that Marshall’s proposal does not eliminate the need to explain the origin of genetic information. Specifically, he showed that Marshall’s proposal presupposed, but did not explain, several separate sources of pre-existing genetic information.
In the first place, Meyer cited some of Marshall’s own scientific papers to show that Marshall himself acknowledged the need for what Marshall called “gene novelties” — new genes — for building the specific anatomical structures (and parts) that characterize the Cambrian animals. (Meyer pointed out that building these animals would have required many so-called “taxon specific” genes that are known to be completely unique to specific animal forms.)
Second, Meyer also noted that the developmental gene regulatory networks that Marshall emphasized were themselves made of genes and gene products and that Marshall had presupposed the existence, but not explained the origin, of these regulatory genes.
Thirdly, Meyer pointed out that Marshall’s proposal assumed a Precambrian pre-adapted genome with many of the regulatory genes for building Cambrian animals. Meyer noted that this proposal also does not eliminate the need to account for the origin of the genetic information necessary to produce the Cambrian animals. Rather, it just pushes the problem of the origin of some of that information back into the Precambrian.
Fourthly, Meyer noted that Marshall’s proposed “rewiring” of gene regulatory networks itself requires an infusion of new information. In particular, it would have required a whole host of coordinated genetic mutations (to various regulatory regions in the genome), requiring significant informational inputs. (This need to generate multiple coordinated mutations would have faced many of the problems faced by evolutionary mechanisms which Meyer discusses in Chapter 12 of Darwin’s Doubt.)
Thus, overall, Meyer showed that Marshall’s proposal presupposed, but does not explain, the origin of several key sources of genetic information (and, I would add, that it does nothing to explain the origin of the epigenetic information that Meyer discusses in Chapter 14 of Darwin’s Doubt).
In response, Marshall argued that Meyer had been forced to shift ground by allowing that the biological information could have arisen in the Precambrian, rather than the Cambrian. Meyer readily acknowledged that information necessary to build the Cambrian animals might have arisen in the Precambrian period, but that did not solve the central problem posed by Darwin’s Doubt — which is the problem of the origin of that information, not precisely when it originated. Meyer noted several times that Marshall had simply pushed the problem of the origin of the necessary information back into the Precambrian, without offering any explanation for how that information had arisen.
Meyer also pointed out that experiments on developmental gene regulatory networks in actual animals have repeatedly shown that perturbing the central components of these networks have catastrophic consequences for animal development. Meyer noted that there is no empirical support for the idea that dGRNs could have been labile in the past. Meyer made this same point in his response to Marshall review. In support of this claim, Meyer cited the work of Eric Davidson, whom Marshall had earlier accused Meyer of neglecting in his discussion of how body plans are built.
In fact, Meyer discussed the importance of Davidson’s work extensively in Chapter 13 of Darwin’s Doubt, which makes it odd that Marshall charged during the debate that Meyer “completely missed” this literature. As Meyer wrote:
Indeed, there are no examples of these deeply entrenched, functionally critical circuits varying at all. At the periphery of the hierarchy are gene regulatory networks that specify the arrangements for smaller-scale features that can sometimes vary. Yet, to produce a new body plan requires altering the axis and global form of the animal. This requires mutating the very circuits that do not vary without catastrophic effects. As Davidson emphasizes, mutations affecting the dGRNs that regulate body-plan development lead to “catastrophic loss of the body part or loss of viability altogether.” He explains in more detail: “There is always an observable consequence if a dGRN subcircuit is interrupted. Since these consequences are always catastrophically bad, flexibility is minimal, and since the subcircuits are all interconnected, the whole network partakes of the quality that there is only one way for things to work. And indeed the embryos of each species develop in only one way.”
Thus, contra Marshall, Meyer didn’t “miss” the arguments of Davidson and others about dGRNs and hypotheses about rewiring preexisting networks of genes, nor did he critique a model of animal evolution that’s stuck “in the 1980s.” Rather, he discussed them, he confronted them, and he critiqued them.
As he had done in his review, Marshall again insisted that it was entirely possible that dGRNs were more flexible in the past, and that an ur-animal, an original form of animal life, evolved into the various animal forms that arose in the Cambrian period with their distinct body plans. If we frame Marshall’s views in terms of fitness landscapes, he essentially believes that this ur-animal sat at the bottom of a valley surrounded by perhaps an entire range of adaptive peaks, where it could easily evolve in virtually any direction.
Meyer replied by noting that Marshall’s proposal goes against everything we’ve learned about how actual biological systems work, and thus must be regarded as entirely speculative and counter to the facts of biology. He also questioned the coherence of Marshall’s proposal by noting that dGRNs function as control systems and that control systems by definition specify outcomes both spatially and temporally. He argued that postulating a biological control system that doesn’t specify outcomes (i.e., is malleable) is really a contradiction in terms. What, he asked, would the role of a gene regulatory network have been if it was not constraining the timing of the expression of networks of genes to insure proper body plan development? Thus, Meyer questioned whether Marshall’s proposal of a labile or malleable dGRN was even coherent.
Marshall and Meyer picked up this thread again at the very end of the conversation. Marshall insisted that it was unrealistic to expect that the dGRNs of modern animal phyla would tell us about how dGRNs operated in the deep past. In response, Meyer pointed out that Marshall’s position reversed the requirement of the historical scientific method as pioneered by Darwin and Lyell, both of whom insisted that our present knowledge of cause and effect should constrain our theorizing about the evolutionary past. Meyer said that Marshall’s willingness to jettison that principle reflected his own prior commitment to a materialistic worldview. (Marshall had previously, in his review and in the debate, accused Meyer of allowing his theistic perspective and motivation to color his analysis of the evidence.)
Here’s what Meyer said in reply:
Charles is actually revealing that he has some deeper metaphysical commitments of his own. The move he made in the review, where he said “these developmental gene regulatory networks, yes they can’t be perturbed but they must have been perturbed in the past,” what he ends up doing there is he ends up jettisoning basic principles of the historical scientific method where we are enjoined to look for causes now in operation. Since we don’t have causes in operation that can produce the kind of complexity that we need to build a body plan — the informational complexity, the circuitry, etc. — we say well maybe things could have been different in the past. And I think what that [reveals] is a prior commitment to at least methodological materialism.
And what we’re doing in the ID movement is saying “Hey, let’s not limit ourselves to just materialistic explanations. If we have the right kind of evidence — the kind of evidence which we know from experience always indicates the prior activity of a mind — let’s allow ourselves to consider that as a possibility, in part because it provides a better explanation, but it also may lead to real advantages for science, because if life is a designed system, it’s gonna look different, it will lead to different kinds of experiments.
In the course of the conversation, Marshall did offer two possible explanations for the origin of new biological information.
First, he appealed to self-organizational processes in which altering a few simple rules can produce new complexity. Second, Marshall also invoked thermodynamics, noting how the earth is a closed system that is constantly bathed in sunlight, and this energy can create localized order and “explore improbabilities.”
Meyer critiqued both these proposals, noting that known examples of self-organization produce simple, redundant order but not the kind of complex and specified information (or the digital code) we see in life. Meyer also pointed out that Darwin’s Doubt evaluated many self-organizational scenarios, such as developmental patterning modules, but that inevitably “complexity depends on prior informational complexity,” and thus does not explain the origin of that information. In Meyer’s words, Marshall is “helping himself to informational endowments that make those processes possible.”
Marshall responded by saying “that’s a very good point” and acknowledged that he’s “not trying to deny” it.
At the very end of the debate, Marshall tried to close with a new point — a “gotcha point” — regarding lysyl oxidase, a protein that Meyer had noted is necessary for building the hard exoskeletons of arthropods, which first arise in the Cambrian period. Marshall noted that this protein is found not only in arthropods, but also in other animals, including humans (vertebrates). On an evolutionary view, this would require the gene for lysyl oxidase to have arisen early in the branching of animals, and so it must have been present in the common ancestor of arthropods and humans. In Marshall’s view, this shows that such a gene must “preexist animals by a long shot.”
Meyer concluded the debate with a ready rejoinder:
I think Charles took my use of that protein out of context. I was simply showing that the organisms have parts lists, and that the parts lists have to be built by proteins. And I acknowledged that he wants to push the problem of the origin of information back into the Precambrian. He’s free to do that. But that is not solving the problem. That’s just begging the question.
Notice the question that he didn’t answer, which is what is the origin of the genetic information necessary to build that protein? Some proteins he acknowledges arise with the Cambrian animals, that’s why he talks about the need for “genetic novelties.” Some he says must have preexisted them in this preadapted Precambrian gene set. And that I think is the fundamental scientific issue. What is the origin of genetic information? He doesn’t really answer that question. He just presupposes, but doesn’t explain, two sources of information, the information that arises in the Cambrian, and in the Precambrian.
I want to say that that information problem which is unsolved within a materialistic framework has a ready and obvious solution because of what we know from experience about what it takes to generate information. And that’s intelligent design.
In the course of the debate, Meyer and Marshall discussed many other issues: whether intelligent design would lead to fruitful new research; whether the case for intelligent design constituted a God–of–the–Gaps argument; and their own backgrounds and motivations for doing science and for developing their arguments and perspectives. Nevertheless, the debate did repeatedly return to the central argument of Meyer’s book concerning the origin of the information necessary to produce the Cambrian animals. After having written many responses to critics of Meyer’s book who have refused to address Meyer’s main information argument for intelligent design, I personally found Marshall’s willingness to engage the main argument of Darwin’s Doubt refreshing. One gets the sense listening to Marshall of a first rate scientific mind grappling with the real issues confronting evolutionary biology.
In wrapping up, the debaters reflected on what if anything the conversation had accomplished. I think that Marshall was correct in saying the parties ended right where they started. But Meyer was also correct in saying that that the debate was “clarifying.” In my view, it showed most importantly that Marshall, like other leading evolutionary theorists, presupposes, but does not adequately explain, the origin of the information necessary to build new forms of animal life.
The debate also suggested to me — given Marshall’s status as one of the world’s leading Cambrian paleontologists and evolutionary biologists — that a strictly materialistic evolutionary approach has little left to offer with respect to that critical question. Despite his indisputable erudition concerning the Cambrian fossil record and evolutionary biology generally — clearly evident in the debate — Marshall was unable to offer a compelling explanation for the origin of new biological information and relied mainly on the same question-begging proposal that Meyer had already decisively rebutted in writing (see here). After listening to the debate, I was impressed with Charles Marshall the evolutionary biologist, the scientist, and the thinker, but was also left thinking that if his arguments about the origin of biological information were the best that mainstream evolutionary biology has to offer, then the future of biology definitely belongs to ID.