This is the first part of a review of Sleeping Beauties: The Mystery of Dormant Innovations in Nature and Culture, by Andreas Wagner (2023).

In the introduction to his new book, Dr. Andreas Wagner, a professor of evolutionary biology at the University of Zürich, makes a startling admission followed by a bold claim: 

Experts still debate how Darwinian evolution can bring forth the truly new, because natural selection can select only what is already there. On its own, it cannot create new forms of life. Intelligent design creationists even argue that true innovation is impossible in biological evolution. The examples in this book prove the opposite. 

Apparently, Darwinian evolution doesn’t actually explain life. This may come as news to some people, since from popular media, public schools, university curricula, and the words of certain famous scientists, one would gather the impression that there is no professional controversy about how life evolved and everything has been explained by Darwin’s theory. 

Wagner, in contrast, is comfortable revealing to the public that this is not true, and that evolutionary biologists within his professional circles are still debating how exactly life could have emerged. His confidence seems to stem from his assurance that his own theories can fill in the gaps and put an end to the controversy, keeping the creationists at the gates at bay — if only people will stop ignoring him.  

Ahead of His Time?

If it seems vain or arrogant to suggest (as Wagner does) that one’s ideas are not accepted because they are so far ahead of their time, Wagner at least has the excuse that he is being thematic: his book is all about things that are way ahead of their time. That’s what the title, Sleeping Beauties, refers to. Wagner borrows the term from the discipline of informatics, where it refers to scientific papers that endure a long period of obscurity before bursting into significance, and applies it to — well, just about everything. Wagner sees Sleeping Beauties as an archetype, not just in publishing, but in all areas of life and the world — and not least in the history of life itself. The bulk of the book is dedicated to the Sleeping Beauties of evolutionary history: useful features or adaptations that seem to have served little purpose at first, and then, after perhaps millions of years of dormancy, suddenly became wildly successful. According to Wagner, such surprises are not the exception, but the norm. 

With a thesis like this, perhaps it’s no wonder Wagner feels the need emphasize his naturalist bona fides. After all, sentences like “Each such gene is a solution in search of a problem, which may arise long after the gene’s origin, or never,” could easily be taken to imply … well, foresight. The very existence of Sleeping Beauties might be taken as evidence of planning or purpose in evolutionary history. So Wagner has to take pains to show that, no, what he’s saying actually disproves intelligent design. 

The trouble is, Wagner doesn’t seem to know anything about intelligent design. That would be fine if he weren’t presenting his book as some sort of refutation of it; but if you’re going to try to refute something, you should be sure you know what it is you’re refuting. Otherwise, you’ll just end up looking silly.

A Big Improvement

Honestly, the book would have been easily improved by simply removing the mentions of intelligent design. For the most part, it’s a delightful read. Wagner’s collection of Sleeping Beauties would have been fascinating enough if he had let it simply be that. 

Instead, he gets caught up in the lofty dream of explaining life and reality once and for all, and insists on turning the book into yet another half-baked attempt to prove that intelligent design is wrong, without the courage or patience to actually engage with design arguments. Wagner creates the impression that everything he says in the book is somehow a refutation of the arguments of “intelligent design creationists,” but what those arguments are remains unstated. I suppose that makes them easier to refute — vague, shadowy errors, placed in the back of the reader’s mind by some choice phrases, are easier to do away with than explicit arguments. 

It’s an old trick. But it shouldn’t be allowed to stand. Let’s bring the arguments into the light, look at them clearly, and see whether or not Wagner’s evidence really amounts to a refutation of anything. 

Claim 1: Evolution Must Be Easy, Because It Happens All the Time 

First, we have to ask: Which design arguments does Wagner think his book is refuting? It’s a little difficult to tell, since he never addresses any design argument explicitly, but he seems to have in mind the most famous (and thus most misunderstood) argument for design in biology: Michael Behe’s irreducible complexity argument.

Behe argues that while it may be possible for simple, adaptive features to accrue over time in a Darwinian fashion, what really needs explaining is the existence of beneficial features that only emerge as the result of many parts working together in the form of complex systems and molecular machines. Behe points out that unguided evolution and natural selection wouldn’t have any better odds of creating that sort of arrangement than blind luck would, because natural selection would only start working after the system in question was complete and had started functioning. What’s more, such “irreducibly complex” systems are the norm, not the exception in life. Behe sees them everywhere, much as Wagner sees his Sleeping Beauties everywhere. And Behe says they demonstrate that life did not evolve through merely unguided processes. 

Wagner disagrees. He points out that many (or even most) features evolved not just once but many times over. Latex evolved at least forty times, in disparate branches of the tree of life. Phytophagy evolved at least fifty times. Rectangular molars evolved at least twenty times. Furthermore, he writes, evolution seems to be able to produce astonishing diversity in almost the blink of an eye, geologically speaking, as soon as a new environment opens up to be filled with new forms. The Andes, Hawai’i, and Lake Malawi, for example, were each filled with diverse species within just a few million years. All this, according to Wagner, proves that the evolution of novel features must simply not be as difficult as intelligent design proponents believe. Innovations must not be so improbable after all, if unguided evolution makes them all the time. 

The trouble is, Wagner never explains how he knows that the evolution of the features he mentions was unguided. And for that matter, he doesn’t seem to see the need to give any evidence that the features evolved at all, beyond that fact that they exist. So if you fill in some of the unstated assumptions, this is the argument you get: 

1. Every biological trait that exists arose through unguided evolution.
2. Trait X exists in many unrelated organisms. 
3. Therefore, X evolved by unguided evolution many times. 
4. A trait that has arisen by unguided evolution many times can’t be extremely improbable.  
5. Therefore, X is not extremely improbable, and ID proponents are wrong. 

That’s all well and fine… if you accept the first premise, every trait that exists arose through unguided evolution. But that’s the very thing under question. The argument is completely circular. 

It’s also pretty astonishing that Wagner’s argument hinges on the fact that so many traits evolved multiple times. He takes great pains to emphasize this. He’s quite taken with it. He doesn’t seem to realize that this will inevitably lead a critical reader to ask, “Well, does homology imply common descent, or doesn’t it?” To hear Wagner talk, seeing homologous traits in two organisms (or fifty) must not be evidence at all of common descent, since it happens all the time without common descent. But homology was one of the main pillars of evidence for common descent. You can’t have your cake and eat it too. So not only is Wagner’s argument circular, his evidence seems to actually call common descent into question, rather than intelligent design. 

It’s the same for his other piece of evidence — that diverse lifeforms seem to appear almost instantly in the geological record. Surely Wagner knows that this is typically used as evidence against Darwinian evolution. He’s making a bold rhetorical move by taking standard anti-Darwinian arguments and trying to twist them into pro-Darwinian arguments. But a closer look exposes it as mere rhetoric. 

Fortunately, this is only the beginning of Wagner’s case. He follows this up with a (blessedly) less circular and more empirical argument against intelligent design. 

Claim 2: Evolution Has Been Observed in Action

We know evolution works, Wagner writes, simply because we can see it happening in real time. Evolutionary experiments, like the ones in his own lab, are a “potent antidote to the toxin of creationism,” he says. 

The first problem with this argument might surprise Wagner: intelligent design does not require that evolution doesn’t happen. Intelligent design merely requires intelligent design. Some prominent ID proponents, such as Michael Behe, accept universal common descent. That doesn’t preclude design. 

The second problem is that all intelligent design advocates (as far as I know) believe in Darwinian evolution of some sort. The problem isn’t Darwinian evolution; the problem is that the math does not seem to show that complex and highly ordered features could emerge through such a process. But Wagner conveniently ignores the distinction between complex features that appear to require an injection of information into the system, and simple changes that everyone agrees could occur through random mutation and natural selection. To him, they’re all just “innovations.” 

So he proudly displays examples that a scientist who supports intelligent design wouldn’t blink at. For example, Wagner highlights Lenski’s long-running evolution study, seemingly without realizing that Michael Behe himself has highlighted the results of this study over and over again to show that evolutionary innovations are never genuinely constructive. (Part 4 of Behe’s 2020 anthology A Mousetrap for Darwin is devoted to discussion of the results of the Lenski experiment.) 

Wagner’s failure to grasp the crucial distinction is on display in one of his showcase proofs of evolution from the fossil record: a species of stickleback fish that lost its spines over the course of ten thousand years. Wagner is impressed to no end at how quickly this transition occurred, apparently unaware that the mystery of life is not how things get lost or broken, but how things get built.

I don’t think I’ve ever heard an ID proponent, or even a dinosaur-hating flat-earther, deny that an organism could lose a trait over many generations. Mutations happen; everyone can see that. And it only takes a little imagination to see how some deleterious mutations, like going blind or losing armor, might benefit a creature under certain circumstances. That’s not the issue; constructive mutations are the sticking point.  

It gets better. Wagner explains that the loss of the spines was the “rational, frugal thing to do for evolution” because there were few predators in the lake. He is (unwittingly, I assume) making the point of Behe’s book Darwin Devolves, which is that Darwinian evolution actually tends to reduce the complexity of organisms over time, because a loss of complexity is sometimes adaptive, and because this is much easier for random variation to achieve than an increase of complexity.

Why does Wagner choose such a counterproductive example to make his point?  Because it’s the best he’s got, apparently. As he tells us, other instances of evolution aren’t typically visible in the fossil record. “Fossils do not document the evolution of most species so well,” he writes, “but the fossil evidence we have shows that stickleback evolution is typical: evolution responds lightning-fast to the challenges thrown at it.” 

Interesting. I suppose it’s a mere coincidence that the well-documented stickleback transformation was devolutionary rather than constructive? I guess we’re to believe that evolution can slap together complex structures just as fast as it can delete them, and it just happens to be shy about showing off this amazing ability? 

In the end, Wagner’s evidence that unguided evolution happened amounts to (1) evidence for foresight, (2) rejection of a key element of the argument for common descent, and (3) evidence for the devolutionary power of Darwinian processes.

Of course, even if Wagner had managed to prove that unguided evolution had created complex features, he still wouldn’t have answered Behe’s main argument; we would still need to explain how this took place, in spite of the apparent mathematical difficulties involved. 

To his credit, Wagner does propose an explanation. And it turns out his arguments for how he thinks evolution took place are a lot more interesting and original than his arguments that it took place.  We’ll save those arguments for the second part of this review.