William A. Dembski
University of Chicago molecular biologist James Shapiro counts himself as neither a design theorist nor a Darwinist. Indeed, in his reply to my piece “Is James Shapiro a Design Theorist?” he claims to provide a “third way” that avoids the conceptual errors of ID and Darwinism, a point he reiterates in his further reply to Ann Gauger and Douglas Axe. Yet his replies suggest that his third way is really nothing of the sort. Rather, it seems more like an equivocation strategy that grants him the best of both worlds, getting the design that Darwinian evolution is incapable of achieving and maintaining the naturalism that keeps him in good standing with the scientific mainstream.
In reading his replies, I was reminded of the recently deceased Lynn Margulis, whose rhetoric against Neo-Darwinism was as sharp as any design theorist’s. Yet when she was awarded an honorary doctorate a few years back in the Galapagos Islands at the World Summit on Evolution, she remarked that she was a Darwinist after all. I fear the same is true of James Shapiro, and that all his talk about “natural genetic engineering” is really window dressing on an approach to evolutionary biology that at root is Darwinian.
How can I say this? Shapiro places at the heart of his third way “natural genetic engineering, horizontal DNA transfer, interspecific hybridization, genome doubling and symbiogenesis,” which he claims “provide solutions to problems recognized to be intractable under the limitations of conventional evolutionary thinking.” Elsewhere in the same conversation he refers to “the revolution that has occurred in our understanding of protein evolution.” I call the reader’s attention to the words “solution” and “understanding.” Let me suggest that Shapiro’s use of these words in these discussions vastly oversells what his third way has actually accomplished.
Neo-Darwinism essentially localizes the creative potential of evolution in genetic mutations. Shapiro rightly sees that this can’t be the main source of evolutionary variation. So he expands it to include “horizontal DNA transfer, interspecific hybridization, genome doubling and symbiogenesis.” Fine, now you’ve got a richer source of variation. But what is coordinating these variations to bring about the increasing complexity we find in biological systems? Shapiro’s answer is “natural genetic engineering.” Cells, according to Shapiro, are intelligent in that they do their own natural genetic engineering, taking existing structures through horizontal DNA transfer or symbiogenesis, say, and reworking them in new contexts for new uses.
But in making such a claim, has Shapiro really solved anything? Has he truly understood the evolution of any complex biological structures? For instance, does his third way now provide some real insights into protein evolution? I’m hammering on these words “solution” and “understanding” because they are easily misused to suggest greater knowledge and conceptual progress than exists.
Natural genetic engineering would actually mean something, providing genuine understanding of and solutions for the origin of novel biological structures, if Shapiro could point to actual, identifiable mechanisms and show how they take existing structures and then refashion them into new ones. But Shapiro doesn’t do this. For him, natural genetic engineering is a magic phrase, a label, that he attaches to hypothesized processes that are opaque to him and yet that he claims result in evolutionary novelty. Note that I’m not talking here about the mechanisms he regularly cites, such as lateral gene transfer or symbiogenesis, which are mechanisms for generating evolutionary variation. Rather, I’m talking about deeper mechanisms that must exist if natural genetic engineering is to be real and take advantage of more obvious mechanisms such as symbiogenesis.
Here’s an analogy to clarify this point. Imagine an antique car restoration company. We see old beat-up cars drive in. We see completely refurbished versions of them drive out. Amazing what “restorative evolution” can do, isn’t it? Antique car buffs, trying to account for the amazing transformation of these cars, introduce the phrase “restorative automobilic engineering.” It’s a nice phrase. It rolls smoothly off the tongue. But it is vacuous unless we can actually get inside the company and see what precisely its mechanics are doing to restore the cars. And perhaps the mechanics are not restoring the cars. Perhaps they are teleporting them to an advanced alien civilization, which in fact is restoring the cars. Far fetched, you say? Well, yes, but until we actually know what’s going on inside the factory, there’s no way to know for sure, especially if what this company is doing with cars is unprecedented in the automobile industry.
And that’s the point: Shapiro doesn’t know the first thing about how natural genetic engineering itself works. What Shapiro knows is the inputs to evolution. Those inputs are richer than the impoverished inputs of Neo-Darwinism, whose main input is genetic mutation. So Shapiro adds symbiogenesis, lateral gene transfer, etc. We are supposed to be impressed. Okay, it’s great that scientists like Shapiro have been able to discover this enriched set of inputs. And Shapiro claims to have discovered a richer transformative principle for these inputs. Darwin gives us natural selection. Again, this is too impoverished for Shapiro. In its place (or, perhaps, supplementing it), Shapiro gives us natural genetic engineering.
But in fact, natural genetic engineering, in the way Shapiro uses it, is no more enlightening than natural selection, which Shapiro to his credit at least admits is bankrupt. But they’re both magic phrases, mantras that claim to provide insight into how evolutionary transformations occur, but in fact offer no real understanding, nor real solutions. We can see this in Shapiro’s latest reply to Gauger and Axe: “well-documented natural processes are more than adequate to explain how protein evolution for new functionalities can occur in a purely natural and combinatorial fashion.”
Come again? “More than adequate”? Such overblown rhetoric ought immediately to set off warning bells. My colleagues and I would be happy with mere adequacy. The problem is that the natural processes Shapiro cites don’t even rise to that level. To explain protein evolution, it’s not enough to point to some known antecedents (genetic shuffling of one form or another) and then merely invoke the label “natural genetic engineering,” as though this explained anything. Far from explaining what needs to be explained, it sidesteps and misdirects from the real question. This becomes evident when Shapiro cites getting new functionalities in a purely “combinatorial fashion.”
As a probabilist, I’ve had to do my share of combinatorics, a branch of mathematics concerned with counting possibilities. The problem is that in genetics and proteomics, the possible gene and protein products are immense, and so the challenge, always, is to find some biologically meaningful path through these combinatorial spaces. So, when Shapiro invokes natural processes that operate in combinatorial fashion, he is in fact explaining nothing about protein evolution but merely restating the problem. The problem is how to navigate through those vast oceans of combinatorial possibility that are genes and proteins. Saying that this happens in “combinatorial fashion” adds no insight.
In the past, I always had the sense that despite my disagreements with him, Shapiro was an independent thinker willing to take on the sacred cows of science. His recent replies suggest someone who is posturing and gesturing to maintain the intellectual high ground. And yet, on closer scrutiny, he is merely playing to the gallery of the scientific mainstream, which is so committed to naturalism that they couldn’t discover a real instance of intelligence in nature if it hit them over the head (which it does repeatedly).
In particular, Shapiro’s comments about intelligent design requiring the supernatural, constituting a theological crutch, and looking to a deus ex machina betray a seemingly willful misunderstanding of intelligent design. Science is not merely about discovering continuities in nature that can be described by a seamless naturalistic story. Sure, areas of science are like that. But science also presents us with discontinuities that resist naturalistic just-so stories. The most widely cited of these is the singularity of the Big Bang. A singularity is a discontinuity for which explanations in terms of ordinary physical processes break down.
It’s not that scientific investigation stops at a singularity. It’s just that the usual way of doing science, looking to ordinary processes that we’ve seen active in other contexts, no longer works. Granted, it is no explanation of a singularity merely to say “God did it.” But a singularity can be studied on its own terms, and the natural forces that may have played a role on either side of it may be studied and their inability to bridge the gap may also be assessed. Such singularities, proponents of intelligent design argue, have happened throughout the history of life. Life presents us with numerous singularities, everything from the Cambrian explosion to the emergence of some (but not all) novel proteins. So we claim. We may be right. We may be wrong. But we are doing science.
As for Shapiro, let me close by suggesting that, despite protestations about a “third way,” he really is much more on the side of Darwin than design. One of Darwin’s favorite quotes was “Natura non facit saltum.” Shapiro, especially in his book Evolution: A View from the 21st Century, seemingly distances himself from this quote, arguing that evolution need not be gradual. But the only reason for him it is not gradual is that organisms can quickly co-opt structures from other organisms and then quickly re-engineer them. But that quickness, for Shapiro, follows simply from deeper evolutionary mechanisms embedded in organisms.
In the past, I’ve criticized Shapiro for not accounting for the origin of these mechanisms, and I still do. Nor does it seem appropriate for Shapiro to dismiss intelligent design given his complete lack of understanding of the mechanisms needed to account for such mechanisms. For Shapiro, as a naturalist, it’s “mechanisms all the way down.” Intelligent design can consider a richer space of conceptual possibilities, one possibility being that the mechanisms that drive evolution (insofar as they exist at all) might themselves be designed.
But then there’s also the question of whether such mechanisms of evolution (the functional capacities or organisms that govern their evolvability) exist at all. I come back to the antique car restoration company analogy. In one respect, this analogy breaks down. We know something about car restoration because many of us are mechanics and those who aren’t have seen car buffs tinkering with their old automobiles. But at this point in the history of science, we really have no idea what a mechanism of evolution looks like. Sure, Shapiro can point to lateral gene transfer, symbiogenesis, gene duplication, etc. But these are merely raw inputs to evolution. A mechanism of evolution would take these raw inputs and rework them, adapting them to new conditions of life and functional requirements of organisms.
Evolution works by borrowing, taking existing structures and reworking them. It’s the reworking that’s the problem and for which Shapiro offers absolutely no insight. What he offers is the magic phrase “natural genetic engineering,” but how cells take the raw inputs of evolution change and then refashion them for new uses is, for Shapiro, a black box. To the credit of Douglas Axe and Ann Gauger’s Biologic Institute, they are running experiments to determine what sorts of processes running in that black box might or might not facilitate evolution.
So, is Shapiro a Darwinist after all? In concluding this piece, let me offer a qualified yes. Shapiro, it seems to me, is a Darwinist in the same way that Lynn Margulis was and that Stephen Jay Gould was before her. For all three, their natural tendency was to follow the evidence where it leads, which was away from Darwin. But as soon as they veered too far from Darwinian orthodoxy, they found themselves uncomfortably close to outright heretics (e.g., proponents of intelligent design) and in danger of losing their standing in the scientific mainstream, so they reflexively offered ritualistic obeisances to Darwinian naturalism — not to Darwinian evolution per se but to the naturalistic approach to science that characterizes it.
Darwinism thus serves as a tether to rein in the freedom of inquiry that would otherwise have allowed them to transcend Darwin. It’s in this sense that Shapiro is a Darwinist.