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

After rehashing some old arguments for why Darwinian evolution must have occurred, Dr. Wagner moves on to his own, more original, arguments for how it might have happened. Unlike some evolutionary biologists, he is willing to admit that Darwin’s theory does not explain the “arrival of the fittest.” Instead, he has his own theories. 

But before getting into that, Wagner devotes some time to elucidating what he calls the “old fashioned” (yet post-Darwinian) explanation for the arrival of the fittest, formulated by the French biologist François Jacob.

Argument 3: Co-Option Can Explain the Origin of Complex Structures

Summing up Jacob’s hypothesis, Wagner writes:

[E]volution is like a tinkerer with a huge workshop full of junk, devices in various states of assembly and repair, gizmos with half-forgotten uses, and countless tools just as likely to be working as to be broken. And like a tinkerer, evolution modifies, fiddles and plays with these parts, assembling them into ever-new contraptions, gadgets, and molecular machines.

The junk-shop tinkerer is an (unintentionally) apt metaphor, for two reasons. First, because the re-useable junk was already made by an intelligent designer in the first place (good luck “tinkering around” in, say, a sand pit!); and second, because the act of recombining junk into new machines itself requires intelligent design. 

The co-option argument is a good explanation for how a large number of complex biological machines might be aggregated in a single system, but it doesn’t explain how each of those machines developed. That’s because the basic quality of a complex machine is that it requires the coordination of several pieces to perform a single function — which is why Behe has argued that complex machines cannot arise through a gradual, stepwise process. The pieces could be co-opted from elsewhere, but that would not provide the necessary information regarding how the pieces must interact to make a complex machine, which was the main problem in the first place.    

The fundamental dilemma is this: whenever you co-opt any part from an old system and put it in a new system, there are two possibilities. If (1) the co-opted part itself was complex, then that original complexity is left unexplained; but if (2) it was not complex, then the co-opted part does little to explain the complexity of the system it was co-opted into. For example, neither (1) “The GPS system in this car was taken from a helicopter!” nor (2) “The screws in this car were taken from a helicopter!” does much to explain the complexity found in the car, even if the explanations are true.

There is direct negative correlation at work here: the more complexity in the new system the co-option explains, the less complexity in the source system it explains. In other words, you’re just shuffling information around. And as information theorists like William Dembski can tell you, that will never give you new information — there’s no free lunch. 

So the co-option argument amounts to nothing. It’s like if a child asked, “Where do toys come from?” and got the reply, “From the toy store.” This might be a satisfying answer to a child, because (so they say) children are prone to magical thinking. But it shouldn’t be satisfying to an adult scientist.

Argument 4: There’s More Than One Way to Skin (or Build) A Cat

This is not Wagner’s star argument, however. Although he does agree that evolution is like a tinkerer in a junk shop, he says that Jacob was “dead wrong” in thinking that evolution cannot make new genes from scratch (i.e., “de novo genes”). Wagner writes that Jacob can’t be blamed for his false assumption, because he was writing in the days before genomics. He goes on to say:

One can hardly blame Jacob for arguing against de novo genes, because a gene really is a very special stretch of DNA. It has complex features that Jacob knew well — he had discovered some of them — and whose origin is hard to imagine. [Here follows a lengthy description of the complexity required for a single functioning gene.] …Given these requirements, it’s hard to imagine indeed that a gene could emerge from scratch. 

Hard to imagine, but still true… Comparing an ever-increasing number of genomes resulted in many discoveries, but none more mystifying than this one: every newly sequenced genome contained hundreds to thousands of genes whose DNA was unique, bearing no resemblance to DNA in any other organism. Such genes were called orphans. 

Now, an outside observer might say that this is a classic case of a hypothesis being tested and falling short. The hypothesis of unguided, incremental evolution resulted in the prediction that de novo genes would be impossible. In Jacob’s words, as Wagner quotes him: “The probability that a functional protein would appear de novo by random association of amino acids is practically zero.” So if evolution was unguided, it must have built genes slowly, from other genes; we wouldn’t expect to find any de novo genes. 

It was a reasonable prediction. Yet genomics proved that prediction wrong.

Normally, when a hypothesis fails to make successful predictions, that means the hypothesis should be discarded. But some hypotheses are too big to fail. For these privileged hypotheses, when the predictions don’t pan out, rather than ditch the hypothesis, you think and think until you come up with some explanation for why the prediction failed. 

So what’s the new explanation? Wagner argues that de novo genes might be individually unlikely, but they are not unlikely in aggregate. That’s because there are so many possible genes. Any given gene might be astronomically improbable, but there are also an astronomically large number of possible viable genes — so getting some gene that fulfills a given need is not unlikely. 

This is a clever argument, in my opinion. At least, unlike most anti-Behean arguments, it doesn’t fall apart at the level of basic logic. To see the problem with it, you have to think a bit deeper. 

So let’s return to the junk-yard analogy. Imagine that you are walking through a junkyard with your friend (as one does), and you come across what appears to be a fully functioning car. 

“I wonder why they left a perfectly good car here?” you say.

“Maybe they didn’t,” says your friend. “There are lots of usable parts here, and this is tornado country. Maybe a tornado picked them up and combined them into a car.” 

“That doesn’t seem very probable,” you point out.

“Does it not?” your friend replies. “Think about it: there’s more than one way to build an automobile. It could have three wheels, for example. Or six. Or, why not 1,000 little wheels? Or legs, like a spider. Or it could bounce on springs! When you think about it, the number of ways to make a device of locomotion is probably infinite, or practically infinite at least. With that many options, the chances of making an automobile can’t be very low. And anyhow, we know the odds must not be prohibitively low, because — there’s the automobile!”  

Forgetting for a moment that this is an analogy — would you find this argument convincing, or not?

Of course you wouldn’t. That’s because you intuitively know a strange truth about math: that infinity is not the largest number. 

Even if something has infinite opportunities to occur, that does not mean it is likely to occur. Why? Because each one of those infinite possibilities of success might come with its own sea of infinite possibilities of failure. Infinity squared, or times infinity is much larger than infinity. Think of an infinite plane versus an infinite line; despite both being infinite, one is clearly larger than the other in a very real and practical sense.  

Think of it another way. Imagine you throw a dart at the natural number line. The odds of hitting any integer with infinite precision (that is, 2.00000…000 with zeroes repeating to infinity, rather than 2.00000….00013 or anything like that) is infinitely small. That is true even though there are an infinite number of integers to hit! Each integer is infinitely small and has an infinite number of non-integer possibilities on each side, so the fact that you are aiming at an infinite sea of integers does not make hitting one any more than infinitesimally probable. 

In the example of the automobile from the junk-yard tornado, you intuitively knew that getting any sort of automobile is unlikely, no matter how many potential sorts there are. It does not matter if there are literally infinite ways to make an automobile, because there are (give or take) infinity-squared ways of not making one.

The only relevant difference between this hypothetical scenario and the real scenario in question with molecular machines is that molecule machines are far more complex than automobiles. There is no reason to consider them more likely to form by chance in a cell than an automobile is likely to form by chance in a junkyard. Even the vast age and size of the universe doesn’t make a dent in the unlikeliness of it. 

Wagner has discovered the interesting fact that there are many ways to make a complex system, and is unduly impressed by this fact. It’s not actually that revolutionary of an idea. Anyone could have guessed that there would be nearly infinite ways to arrange, say, a metabolic pathway, or a bacterial motor—there’s always more than one way to skin a cat, as they say. 

But there are far, far more ways to fail to skin a cat. 

The Thrall of the Zeitgeist 

In the later part of the book, Wagner moves on from evolution to examples of Sleeping Beauties in human culture. He writes: 

We heard that innovations come easily to evolution. They come just as easily to culture, and multiple discoveries are exhibit A for this claim. The wheel, discovered in the new and in the old world, is only one among hundreds of examples. Even more ancient is agriculture. It has at least eleven independent origins…The pendulum clock was invented at least three different times, the thermometer seven times, the telegraph four times and the radar six times.

The thing is…cultural innovation is easy because of the intelligent minds behind it. The comparison is (again, unintentionally) quite rich. 

If Wagner had felt like it, he could have just as easily made his book into an argument for intelligent design, rather than against it. Wagner makes no controversy about the basic facts: irreducible complexity, vastly improbably sequences in genes, etc. His arguments against design all amount to rhetorical glosses and strange contortions of reasoning. 

It’s a shame. The book, as I’ve said, is truly delightful, and there’s no reason it had to be marred by naturalist blustering against a theory Wagner can’t be bothered to understand. Wagner is obviously a man of great intelligence, curiosity, and erudition. It’s a shame he doesn’t let those qualities shine through when he’s taking up the task of debating the most important questions of life. 

It’s a shame, but it isn’t shocking. Wagner wouldn’t be the first otherwise intelligent and curious person to suddenly lose interest in rational exploration of ideas when it comes to intelligent design. It happens all the time.

The cause of that isn’t so strange. In fact, it’s something that Wagner highlights in his book. Wagner writes that one reason that Sleeping Beauties stay asleep for so long is that people are prevented from appreciating them by the mind-numbing power of the Zeitgeist, the “spirit of the age.” Every age has its own spirit, and any idea that contradicts the spirit of the age is doomed to obscurity — at least, until the age changes.

What’s the antidote? Once again, Wagner has the answer in his book: stop worrying about what others think, and pursue the object of your passion for its own sake.

It’s good advice, and I hope Dr. Wagner will follow it when it comes to the ultimate question of his field. He seems to want to; he just needs to go all the way. If he does, he may ultimately reject ID or affirm it — but at least he will have to give it a fair hearing, not the cursory glance and mishmash of half-arguments he served up in this book.