When Darwin wrote On the Origin of Species, what was the main piece of evidence that he couldn't explain?
Darwin was concerned about the sudden appearance of major groups of animals in a particular geological layer. At that time, it was called the Silurian period, but later it was renamed the Cambrian. Darwin confessed that this was not something that his theory could adequately explain. He acknowledged that doubt in On the Origin of Species, and he said that it was a "valid argument against the views here entertained." Here's the mystery: Why did the Cambrian animals appear so suddenly if, in fact, the history of life is best represented by a slow, gradually unfolding, branching tree, and if the chief cause of the change depicted by that tree is natural selection acting on random variations?
How does your book build upon this problematic piece of evidence?
The book tells the story of Darwin's doubt and what has become of it. Not only is there the problem of those missing Precambrian ancestors that are expected from Darwin's theory, but there is also the question of cause or mechanism—that is, how do you build a new animal? Darwin understood that his mechanism of natural selection acting on random variations must work very slowly and gradually, but what he saw in the fossil record, and what has been documented by subsequent findings, is the sudden appearance of a great number of new animal groups—many phyla, subphyla, and classes of animals.
If we could step back from evolutionary considerations for a moment, why is the Cambrian explosion such an important event?
The Cambrian explosion is important, independent of one's theory of origins, because it's the place where most of the major animal body plans first come into the fossil record. We did a very careful count for the book and found that there are about 36 total phyla that have existed during the history of life. Of these, 26 are represented somewhere in the fossil record. And of these 26, fully twenty of them first come on the scene in the Cambrian. And this happens within a fairly narrow seam of rock that dates back just five to six million years. In geological history, that is a blink of the eye.
So far, we're talking a lot about fossils. Why does your book also focus on information?
There are really two mysteries associated with the Cambrian. The first is the mystery of the missing ancestral animal fossils. The Precambrian strata are not documenting the existence of the ancestral forms—the transitional intermediates—that you would expect from a Darwinian picture of the history of life. But there is a second mystery—the question of whether natural selection and random mutation could build these animal forms and do so quickly enough to account for the pattern in the fossil record. That mystery became much more acute in the second half of the twentieth century, as we learned more and more about what it takes to build an animal.
The Cambrian period is marked by an explosion of new body plans—new cell types and new ways of organizing cell types into tissues, tissues into organs, and organs and tissues into body plans. This would require many new genes and proteins to service all of the new cell types that are necessary to build the new animal forms. We now know that to build a new animal, you need a lot of new information.
Why has neo-Darwinism had so much difficulty explaining the origin of new information?
One of the problems with the neo-Darwinian mechanism for generating new information is that of "combinatorial inflation"—a term coined by mathematician David Berlinski. It's an excellent way of describing a basic mathematical problem that is associated with any system that stores information—such as a gene or a protein or a section of digital code.
In the book, I use the example of a bike lock. If you've got a bike lock with four dials and ten settings on each dial, you have 10 times 10 times 10 times 10—that is, ten thousand possibilities. So even a small bike lock of four dials presents a big combinatorial problem; many combinations must be searched to find the functional solution.
Imagine a standard bike lock with ten dials, where there are 10 to the 10th (10 billion) possibilities. Now relate that to proteins. Proteins don't have just ten dials. A modest-length protein of 150 amino acids has something equivalent to 150 dials. Yet at each one of those dials, there are 20 possibilities. As you begin to think about the odds of generating a functional protein, you realize that you've got to search an enormous space of possibilities—in this case, it would be 20 to the 150th power, which computes to one in 10 to the 195th power.
Now, in the book, I go into different factors that minimize that problem and/or accentuate it. One thing that minimizes the problem a bit is the recognition that there are many combinations of amino acids that will produce a functional protein. However, my colleague Doug Axe has experimentally determined that the ratio of those functional proteins to all of the possible combinations of a protein of a given length is still miniscule. He got a number of one in 10 to the 77th power. You have to ask whether mutation and selection would be effective in searching a space of possibilities that large.
Here's where the bike-lock analogy helps. Imagine a thief who sees a nice racing bike and decides that he would very much like to have that bike for his own. The bike is secured by a lock with just four dials—ten thousand possibilities. There's a security guard who makes regular rounds, so the thief may have, at best, ten minutes to find the right combination. So the question for the thief is whether he's likely to crack the combination before the guard spots him. This shows that if you want to determine the plausibility of an event occurring by chance, you not only need to know the probability of the event itself, but also how many opportunities there are for the event to occur.
In the case of our thief, it's not plausible that he'll crack the combination lock by chance in the mere ten minutes allowed. In the same way, the sequence spaces that need to be searched by Darwinian evolution are too vast for the time available in the fossil record.
We now know that living organisms don't just rely on genetic information, but also epigenetic information. What is epigenetic information, and why is it a problem for Darwinian theory?
Developmental biologists have known for decades that the information in DNA alone is necessary—but not sufficient—to build an animal. There are other sources of information, including patterns of targets on cell membranes, the arrangement of the cytoskeleton in cells, and a sugar code on the exterior of cells that stores information that's necessary for coordinating cell-to-cell interactions. There's a lot of information beyond the genome that is playing a role in animal development.
According to neo-Darwinism, the source of new variation ultimately comes from mutations in genes. But if genes alone do not produce new animal forms—if we need other sources of information—then you can mutate DNA indefinitely, without respect to probabilistic limits, and you will still never build a new animal form. This is a powerful challenge to neo-Darwinism.
Your book describes a "post-Darwinian" world. What is that?
In the prologue, I describe a huge disparity between the public pronouncements in favor of standard theories of Darwinian evolution and the actual status of the theory in the peer-reviewed technical literature. What you find among leading evolutionary theorists is that there are a few neo-Darwinians still around, but more often you will find people such as the Altenberg Sixteen—a group of evolutionary biologists who met in 2008 to find a new theory of evolution. After critiquing neo-Darwinism in the book, I then look at some of these post-neo-Darwinian theories.
Some of the theories are older ideas, such as self-organization, evo-devo, or punctuated equilibrium, but I also look at new theories such as James -Shapiro's ideas about natural genetic engineering or Michael Lynch's ideas about neutral evolution. In all cases, whatever their other virtues, the theories fail to account for the origin of the information that is necessary for building animal forms. Many of them even beg the question; they identify some real and important biological processes, but those processes presuppose the existence of much -genetic or -epigenetic information. I find these theories ultimately wanting as solutions to the origin of form and information.
There's another "post-Darwinian" model that you mention—intelligent design (ID). Why is ID a superior explanation for the information explosion that occurred in the Cambrian?
My book looks at the reasoning used by Darwin—sometimes called "uniformitarian reasoning"—which provided the foundation for a method of scientific reasoning called by philosophers "the method of multiple competing hypotheses," or the "inference to the best explanation." Scientists of the nineteenth century generated clear criteria for what constituted the "best" explanation: causal adequacy. The famous geologist Charles Lyell said that when reconstructing events of the ancient past, we should invoke "causes now in operation." Causes should be sufficient to produce the effect in question. I found this same criterion at work in Darwin's writings; he sought to find a "vera causa" or a "true cause"—a cause known to produce the effect in question.
In my grad-school years, I became immersed in studying this method of historical scientific reasoning. It turns out that it applies beautifully to the question of the Cambrian. We should be looking for causes that are known from our present experience to have the capacity to produce the effect in question.
In the book, I start with the origin of information. If we ask, "What do we know from experience that is capable of producing functional information in a digital form as we find it in genes?", we immediately realize that there is a cause that's capable of producing that kind of effect: intelligence. Minds are capable of creating information, and nothing else we know of is.
We know from our uniform and repeated experience that conscious and rational activity generates information. So when we find information, absent direct observational knowledge of its cause, we can infer backwards in time to that cause, because we know of only one cause—mind—that is sufficient to produce this key effect.
If we're now living in a "post-Darwinian world," and scientists are indeed searching for new models to replace the failing neo-Darwinian paradigm, why can't ID be on the table for explaining the Cambrian explosion?
Intelligent design is a scientific argument in several ways. It's based on scientific evidence; it uses an established method of scientific reasoning; and it's in theoretical competition with other scientific theories that are attempting to explain the same phenomena. I would even argue that ID explains them best, where "best" has a clear definition: namely, causal adequacy.
So why do many scientists want to exclude ID from consideration? Here we get into an unspoken convention (though sometimes it is spoken) that says that scientists must not seek the best explanation, but rather the best materialistic explanation. That convention is really unfounded, and at bottom it's anti-intellectual because it limits the freedom of scientists to follow the evidence where it most naturally leads.
In the book, I show that the pattern of fossil appearance in the Cambrian looks just as we would expect if, in fact, an intelligent agent had acted. Many hallmarks of the Cambrian event are features that are known from experience to be produced by one and only one type of cause. If we're simply willing to open our minds, we find that we do have a very compelling explanation for features that are otherwise inexplicable. That cause is intelligent agency. •
For more information about Stephen Meyer's book, Darwin's Doubt: The Explosive Origin of Animal Life and the Case for Intelligent Design (HarperOne, 2013), visit www.darwinsdoubt.com.
If you enjoy Salvo, please consider giving an online donation! Thanks for your continued support.