Recently an email correspondent asked me about a clip from Neil deGrasse Tyson’s reboot of Cosmos where he claims that eyes could have evolved via unguided mutations. Even though the series is now eight years old, it’s still promoting implausible stories about eye evolution. Clearly, despite having been addressed by proponents of intelligent design many times over, this issue is not going away. Let’s revisit the question, as Tyson and others have handled it.

In the clip, Tyson claims that the eye is easily evolvable by natural selection and it all started when some “microscopic copying error” created a light-sensitive protein for a lucky bacterium. But there’s a problem: Creating a light-sensitive protein wouldn’t help the bacterium see anything. Why? Because seeing requires circuitry or some kind of a visual processing pathway to interpret the signal and trigger the appropriate response. That’s the problem with evolving vision — you can’t just have the photon collectors. You need the photon collectors, the visual processing system, and the response-triggering system. At the very least three systems are required for vision to give you a selective advantage. It would be prohibitively unlikely for such a set of complex coordinated systems to evolve by stepwise mutations and natural selection. 

A “Masterpiece” of Complexity

Tyson calls the human eye a “masterpiece” of complexity, and claims it “poses no challenge to evolution by natural selection.” But do we really know this is true?

Darwinian evolution tends to work fine when one small change or mutation provides a selective advantage, or as Darwin put it, when an organ can evolve via “numerous, successive, slight modifications.” If a structure cannot evolve via “numerous, successive, slight modifications,” Darwin said, his theory “would absolutely break down.” Writing in The New Republic some years ago, evolutionist Jerry Coyne essentially concurred on that: “It is indeed true that natural selection cannot build any feature in which intermediate steps do not confer a net benefit on the organism.” So are there structures that would require multiple steps to provide an advantage, where intermediate steps might not confer a net benefit on the organism? If you listen to Tyson’s argument carefully, I think he let slip that there are.

Tyson says that “a microscopic copying error” gave a protein the ability to be sensitive to light. He doesn’t explain how that happened. Indeed, biologist Sean B. Carroll cautions us to “not be fooled” by the “simple construction and appearance” of supposedly simple light-sensitive eyes, since they “are built with and use many of the ingredients used in fancier eyes.” Tyson doesn’t worry about explaining how any of those complex ingredients arose at the biochemical level. What’s more interesting is what Tyson says next: “Another mutation caused it [a bacterium with the light-sensitive protein] to flee intense light.”

An Interesting Question

It’s nice to have a light-sensitive protein, but unless the sensitivity to light is linked to some behavioral response, then how would the sensitivity provide any advantage? Only once a behavioral response also evolved — say, to turn towards or away from the light — can the light-sensitive protein provide an advantage. So if a light-sensitive protein evolved, why did it persist until the behavioral response evolved as well? There’s no good answer to that question, because vision is fundamentally a multi-component, and thus a multi-mutation, feature. Multiple components — both visual apparatus and the encoded behavioral response — are necessary for vision to provide an advantage. It’s likely that these components would require many mutations. Thus, we have a trait where an intermediate stage — say, a light-sensitive protein all by itself — would not confer a net advantage on the organism. This is where Darwinian evolution tends to get stuck.

Tyson seemingly assumes those subsystems were in place, and claims that a multicell animal might then evolve a more complex eye in a stepwise fashion. He says the first step is that a “dimple” arises which provides a “tremendous advantage,” and that dimple then “deepens” to improve visual acuity. A pupil-type structure then evolves to sharpen the focus, but this results in less light being let in. Next, a lens evolves to provide “both brightness and sharp focus.” This is the standard account of eye evolution that I and others have critiqued before. Francis Collins and Karl Giberson, for example, have made a similar set of arguments. 

Such accounts invoke the abrupt appearance of key features of advanced eyes including the lens, cornea, and iris. The presence of each of these features — fully formed and intact — would undoubtedly increase visual acuity. But where did the parts suddenly come from in the first place? As Scott Gilbert of Swarthmore College put it, such evolutionary accounts are “good at modelling the survival of the fittest, but not the arrival of the fittest.” 

Hyper-Simplistic Accounts

As a further example of these hyper-simplistic accounts of eye evolution, Francisco Ayala in his book Darwin’s Gift to Science and Religion asserts, “Further steps — the deposition of pigment around the spot, configuration of cells into a cuplike shape, thickening of the epidermis leading to the development of a lens, development of muscles to move the eyes and nerves to transmit optical signals to the brain — gradually led to the highly developed eyes of vertebrates and cephalopods (octopuses and squids) and to the compound eyes of insects.” (p. 146)

Ayala’s explanation is vague and shows no appreciation for the biochemical complexity of these visual organs. Thus, regarding the configuration of cells into a cuplike shape, biologist Michael Behe asks (in responding to Richard Dawkins on the same point):

And where did the “little cup” come from? A ball of cells–from which the cup must be made–will tend to be rounded unless held in the correct shape by molecular supports. In fact, there are dozens of complex proteins involved in maintaining cell shape, and dozens more that control extracellular structure; in their absence, cells take on the shape of so many soap bubbles. Do these structures represent single-step mutations? Dawkins did not tell us how the apparently simple “cup” shape came to be.

Michael J. Behe, Darwin’s Black Box: The Biochemical Challenge to Evolution, p. 15 (Free Press, 1996)

An Integrated System 

Likewise, mathematician and philosopher David Berlinski has assessed the alleged “intermediates” for the evolution of the eye. He observes that the transmission of data signals from the eye to a central nervous system for data processing, which can then output some behavioral response, comprises an integrated system that is not amenable to stepwise evolution:

Light strikes the eye in the form of photons, but the optic nerve conveys electrical impulses to the brain. Acting as a sophisticated transducer, the eye must mediate between two different physical signals. The retinal cells that figure in Dawkins’ account are connected to horizontal cells; these shuttle information laterally between photoreceptors in order to smooth the visual signal. Amacrine cells act to filter the signal. Bipolar cells convey visual information further to ganglion cells, which in turn conduct information to the optic nerve. The system gives every indication of being tightly integrated, its parts mutually dependent.

The very problem that Darwin’s theory was designed to evade now reappears. Like vibrations passing through a spider’s web, changes to any part of the eye, if they are to improve vision, must bring about changes throughout the optical system. Without a correlative increase in the size and complexity of the optic nerve, an increase in the number of photoreceptive membranes can have no effect. A change in the optic nerve must in turn induce corresponding neurological changes in the brain. If these changes come about simultaneously, it makes no sense to talk of a gradual ascent of Mount Improbable. If they do not come about simultaneously, it is not clear why they should come about at all.

The same problem reappears at the level of biochemistry. Dawkins has framed his discussion in terms of gross anatomy. Each anatomical change that he describes requires a number of coordinate biochemical steps. “[T]he anatomical steps and structures that Darwin thought were so simple,” the biochemist Mike Behe remarks in a provocative new book (Darwin’s Black Box), “actually involve staggeringly complicated biochemical processes.” A number of separate biochemical events are required simply to begin the process of curving a layer of proteins to form a lens. What initiates the sequence? How is it coordinated? And how controlled? On these absolutely fundamental matters, Dawkins has nothing whatsoever to say.

David Berlinski, “Keeping an Eye on Evolution: Richard Dawkins, a Relentless Darwinian Spear Carrier, Trips Over Mount Improbable,” Globe & Mail (November 2, 1996)

More or Less One Single Feature 

In sum, standard accounts of eye evolution fail to explain the evolution of key eye features such as:

• The biochemical evolution of the fundamental ability to sense light
• The origin of the first “light-sensitive spot”
• The origin of neurological pathways to transmit the optical signal to a brain
• The origin of a behavioral response to allow the sensing of light to give some behavioral advantage to the organism
• The origin of the lens, cornea, and iris in vertebrates
• The origin of the compound eye in arthropods

At most, accounts of the evolution of the eye provide a stepwise explanation of “fine gradations” for the origin of more or less one single feature: the increased concavity of eye shape. That does not explain the origin of the eye. But from Neil Tyson and the others, you’d never know that.