The Mating Mind: How Sexual Choice Shaped the Evolution of Human Nature (26 page)

It remains to be seen how important the human brain is as an evolutionary innovation. If we became extinct tomorrow, it would count as a micro-innovation characteristic of just one species. If our descendants succeeded in colonizing the galaxy and splitting apart into a hundred thousand species millions of years from now, it would count as a macro-innovation. But an innovation's ability to trigger an adaptive radiation millions of years after its origin cannot explain why it evolved. This raises a serious problem that has remained unsolved ever since Darwin: how can innovations emerge through a gradual process like natural selection? This question has three variants of increasing difficulty.

The easy, most general problem is: how can a qualitatively novel structure arise through gradual, quantitative changes? The answer, of course, is that the whole universe unfolds by processes that turn quantitative change into qualitative novelty. The incremental process of gravitational attraction turns interstellar dust clouds into star systems. The incremental processes of capital investment and education turns poor villages into prosperous cities. The incremental process of growth turns a fertilized egg into a human baby. There is nothing special about evolution in this respect. Every thing in the world that we bother to name is a bundle of qualitatively novel properties emerging from an accumulation of quantitative stuff.

The moderately hard problem is: how can a complex innovation emerge that depends on many parts functioning together? Assuming that natural selection can tinker with only one part at a time, it seems difficult for natural selection to construct multipart innovations. What good is the retina of an eye without the lens, or vice versa? This sounds like a lethal argument against

incremental Darwinian evolution, but it isn't. If it were, the existence of Microsoft would force one to be a Creationist. The

Microsoft Corporation is composed of thousands of employees who must all work together for the corporation to function: management, accounting, personnel, marketing, finance, programming, and so forth. Could Microsoft have arisen through the incremental accumulation of employees, hiring them one by one? It seems logically impossible. If employee number one was a programmer, the corporation couldn't survive, because there would be no one in marketing to sell her product, no one in personnel to pay her, and no one in the legal department to sue software pirates. But if employee number one was in marketing, she wouldn't have any product to sell. And so on. How could a corporation that includes dozens of different kinds of employee possibly have emerged in just twenty years through incremental hiring? The answer is that the early employees were less specialized, and each filled many roles. When Microsoft consisted of just the teenaged Bill Gates and Paul Allen, they split all the corporate responsibilities between them. As more employees were hired, responsibilities were delegated and became more specialized. If one accepts the possibility of growing large, multi-part corporations by hiring one person at a time, perhaps one should not be too bothered by evolution's ability to produce innovations by compiling one genetic mutation after another. As far back as the 1850s, Herbert Spencer was pointing out that gradual growth through progressive differentiation and specialization is the way that both social organizations and biological adaptations must evolve.

The Threshold of Innovation

The really difficult problem is: how can natural selection favor the initial stages of evolutionary innovations when they are accumulating costs but not yet offering any net survival benefits? Darwin worried a lot about this problem. How could natural selection favor proto-eyes or proto-wings before they grow sufficiently large and complex to yield their survival benefits?

Selection is frugal: it penalizes traits that impose costs without offering benefits. If most innovations give net survival benefits only once they have passed some threshold of complexity and efficiency, it is hard to see how evolution could favor them before they reached that threshold. This has always been the single most serious objection to Darwin's theory of evolution by natural selection. It was argued most forcefully by the zoologist St. George Mivart just after
The Origin of Species
was published, and it has been a stumbling block ever since.

Some minor innovations do not suffer from this threshold effect. A giraffe's neck could have evolved to its present length gradually, each increment of length giving an immediate improvement in reaching higher acacia tree leaves. An insect's camouflage could evolve gradually, each step further reducing a predator's chance of noticing the insect. Neck-stretching and color-changing could provide net survival benefits continually throughout their evolution.

Some evolutionary theorists such as Richard Dawkins and Manfred Eigen suggest that the threshold effect is overstated for many major innovations. They think that there are often ways to evolve dramatic innovations along a continuous path where every step right from the beginning yields a new survival benefit. They might be correct. We do not know enough about the evolutionary dynamics of complex traits to know how common the threshold problem is. Most biologists still believe this to be the most significant problem that theories of evolutionary innovation must address. I agree. In my experience with running genetic algorithm simulations on computers, the threshold problem is a very serious obstacle to evolving innovations. If you actually try to evolve something complicated and useful inside a computer using simulated natural selection, you are likely to be frustrated. Simulated evolution often stalls for no apparent reason, gets stuck in a rut for thousands of generations, and shows a perverse tendency to avoid interesting innovation whenever possible. This frustration with simulated evolution's limited innovation ability is fairly common among genetic algorithm researchers.

The threshold effect boils down to this: the evolutionary costs and benefits of innovations work like the economics of pharmaceutical research. The Pfizer Corporation spent over $I00 million and many years developing the drug Viagra before the drug made a single cent of profit. The costs accumulated early, and the benefits came only later. Drug companies can cope with this delayed gratification, and have the foresight to undertake the research that leads to such profitable innovations. But evolution has no foresight. It lacks the long-term vision of drug company management. A species can't raise venture capital to pay its bills while its research team tries to turn an innovative idea into a market-dominating biological product. Each species has to stay biologically profitable every generation, or else it goes extinct. Species always have cash-flow problems that prohibit speculative investment in their future. More to the point, every gene underlying every potential innovation has to yield higher evolutionary payoffs than competing genes, or it will disappear before the innovation evolves any further. This makes it hard to explain innovations.

Sexual Selection and Venture Capital

Let's go back to the Microsoft example. We saw that large corporations could grow from a couple of entrepreneurs by hiring employees one at a, time. Evolution's threshold problem is more of a finance problem than a personnel problem. How did Microsoft grow large enough to reach the threshold of profitability? Like most companies, it survived in the early days through bank loans, venture capital, and stock issues. It didn't grow just from the profits it made. It grew because people were willing to lend it money in the hope that they would get paid back in the future. The problem in growing large corporations is not that you have to hire people one by one—that's the easy part. The problem is that most corporations can't break even until they reach a certain critical mass, and they can reach that mass only by borrowing money against their future profits.

Evolution seems to offer no mechanism to do this when there is

a potential to develop a major innovation. Capitalism depends on
foresight, and evolution has no foresight. The problem of evolu-

tionary innovation boils down to this: evolution needs something
like a venture capitalist. It needs something that can protect the

very early stages of an innovation against the ravages of the competitive market and the laws of bankruptcy, by granting it some line of credit.

Sexual selection works, I think, as evolution's venture capitalist. It can favor innovations just because they look sexy, long before

they show any profitability in the struggle for survival. It can

protect the early stages of innovations by giving them a
reproductive advantage that can compensate for their survival

costs. Of course, this is a risky business. Most innovations may never show any profit, and may never yield any survival advantages. But they don't have to. Venture capitalists can make money when a company floats stock on the stock market, even if the company never sells a single product. Runaway sexual selection can favor evolutionary innovations that never offer a single survival benefit. Both processes work through the magic of runaway popularity. Desire reinforces desire. A confidence bubble grows.

Sometimes the bubble bursts. For every courtship ornament like the peacock's tail that persists, perhaps dozens of ornaments come and go. These ornaments may originate in humble form, become popular for a while, grow a little in complexity and size, and then become unfashionable through various random evolutionary effects, sinking back into evolutionary oblivion. These ornamental fashion cycles may not be good for the species as a whole, but evolution cares no more about the species as a whole than capital markets care about entrepreneurs.

Why Is Evolutionary Innovation Obsessed
with Male Genitals?

If many innovations originate through sexual selection, we would expect most micro-innovations that distinguish one species from another to be sexual ornaments. This contradicts some traditional views of how species split apart, but, surprisingly, this is pretty

much what biologists see. The vast majority of species-defining innovations seem inconsequential for survival. Francis Bacon, father of the scientific method, disparaged the seemingly pointless variety of plants and animals, calling them "the mere Sport of Nature." Darwin was equally perplexed, often wondering why there was so much variety but so little real novelty. If innovations spread through populations because of their survival benefits, why do so few innovations show the survival improvements associated with major innovations and adaptive radiations?

One clue comes from the criteria that taxonomists use to classify specimens into species. Male sexual ornaments and male genitals are the most useful traits for distinguishing most animal species from closely related members of the same genus. If you can't tell whether a beetle is one species or another, look at its color pattern, its weaponry, and its genitals. In his book
Sexual Selection and Animal Genitalia,
William Eberhard emphasized that male genitals are often the first things to diverge when one species splits off from another. Evolutionary innovation seems focused on the details of penis shape. In Eberhard's view, this is because female choice focuses on the details of penis shape, and female choice apparently drives most micro-innovation. In plant taxonomy, the analogous sexually selected traits are the flowers, and they are often most useful in making species identifications. It is often harder to tell what species a female animal is, because the appearance of females diverges much less between species. Bird watchers know this: given a female, you can often only identify the genus, but given a male, you can zero in on the exact species.

The micro-innovations that distinguish species often evolve through sexual selection, as sexual ornaments (or genitals) shaped by mate choice. At one level, this fact simply restates the modern definition of a biological species: a reproductively isolated group of individuals. The commonest kinds of traits that distinguish species must be traits that can work as sexual isolators to keep one group from interbreeding with other groups. Sexual choice is a very efficient sexual isolator for keeping species distinct. As the biologist Hugh Paterson pointed out in the 1970s, species are

basically consensual systems of mate choice. The result is that human taxonomists end up using the same traits to distinguish species that species members themselves use: sexual ornaments. This is why most micro-innovations are concentrated in genitals, ornaments, and courtship behaviors.

Innovation Through Sexual Choice

Sexually selected novelties of this sort could be called "courtship innovations." Most will be nothing more than a slightly novel design for a penis, a minor variation in mating coloration, or a different style of courtship dance. But from these humble origins, a small proportion of courtship innovations and their side-effects may turn out to have some survival benefits in addition to their courtship benefits. They may then become favored by natural as well as sexual selection. Of these survival adaptations, a small proportion may prove significant enough to allow a species to invade many new environmental niches. They produce adaptive radiations, proving themselves over time as major innovations. The ecological success of major innovations may hide the fact that many of them originated as courtship innovations.

The feathered wing may be a good example of a courtship innovation that proved to have large survival advantages in the long term.
Archaeopteryx
fossils from 150 million years ago were first found over a century ago, and paleontologist John Ostrom's 1969 theory that birds evolved from small, fast-running theropod dinosaurs has held up fairly well. However, biologists are still not sure how or why feathered wings evolved on dinosaur-type bodies. Many biologists propose that wings always had an aerodynamic function, even in their early stages of evolution. There is the ground-up theory that wings evolved to help small dinosaurs jump and turn quickly to catch prey, and the trees-down theory that wings helped to break their falls (progressing from parachuting to gliding to powered flight). Other biologists point out that the earliest proto-birds (such as the
Protarchaeopteryx
unearthed in China in the early 1990s) had well-developed wings, but no sign of the lighter skeleton associated with flying, and no sign of the