Book: The Agile Gene

Previous: 8. Conundrums of Culture
Next: 10. A Budget of Paradoxical Morals

A scholar is just a library’s way of making another library.

Daniel Dennett

It is bad enough to be eclipsed on the brink of eternal fame by a competitor, but imagine how much worse it feels if that competitor has been dead for more than a decade and lived his entire life in total obscurity inside a monastery. No wonder Hugo De Vries stares rather unhappily out of my photograph. In 1900 he published a radical theory, for which he felt he deserved the sort of acclaim that had been showered upon John Dalton and was about to be showered on Max Planck. Where Dalton had suggested that matter is composed of atoms, and Planck would treat light as coming in lumps, De Vries too had come up with a quantum theory—that inheritance comes in particles: “The specific characters of organisms are composed of separate units.” He had deduced this by a series of brilliant experiments hybridizing varieties of plants, and he had even hit upon a truth that would take a century to be proved. He speculated that the particles of heredity, which he called the “pangens,” did not obey the species barrier, so that a pangen for hairiness in one plant was also responsible for hairiness in another hairy species of flower.

De Vries, in other words, surely deserved to be known as the father of the gene. But soon after his triumphant account appeared in print, in the French journal Comtes Rendus de l’Académie des Sciences, he was stung by a German bee: Karl Correns. Correns was a mild man but had been driven into an uncharacteristic rage by De Vries’s paper. He had been beaten to a scientific result by De Vries before and was determined to have his revenge. Correns acidly pointed out that though De Vries’s experiments were his own, his conclusion—particulate inheritance—was borrowed, not just in outline but in detail, from the work of a long-dead Moravian monk named Gregor Mendel, even down to the terms De Vries used: recessive and dominant, for example.

Knowing he had been exposed, De Vries grudgingly conceded priority to Mendel in a footnote to the German version of his paper, and settled unhappily for the role of rediscoverer of the laws of heredity. Worse, he had to share even this little credit with two other men: not only Correns, but also a young gate-crasher, Erich von Tschermark, who was good at only two things—persuading the world on flimsy evidence that he, too, had rediscovered Mendel’s laws, and (much later) applying his talents in the service of Nazism. For De Vries, who had a high opinion of himself, this was bitter medicine; to the end of his days he looked on the deification of Mendel with disgust. “This fashion is likely to pass,” he averred, refusing an invitation to the unveiling of a statue of the monk. The trouble was that not many people warmed to De Vries. Fastidious, aloof, touchy, and so misogynist that he was rumored to spit in the culture dishes of his female assistants, De Vries was doomed to see even his terminology eclipsed by that of others. By 1909 the pangen had become the “gene,” a word coined by Wilhelm Johannsen, a professor in Denmark.

Was De Vries a plagiarist? Probably he did discover Mendel’s laws through his own experiments before he rediscovered Mendel’s work in the library: his sudden change of terminology in the late 1890s hints as much. In that sense, he made a great discovery. Probably, too, he thought he could get away with not citing Mendel. After all, how many people would have read 40-year-old volumes of the Proceedings of the Brünn Natural History Society? In that sense, De Vries was a fraud. But it is no surprise when a scientist buries his ancestors, more or less unconsciously downplaying the insights of his predecessors lest they seem to diminish his own breakthrough. Even Darwin was adept, in his humble way, at skating over the contributions to his thinking of others, not least his own grandfather. Ironically, Mendel himself may have borrowed at least part of his main idea from someone else. He made no mention of the English horticulturist Thomas Knight’s paper of 1799 showing how the easily achieved artificial pollination of different varieties of pea could hint at the mechanism of heredity, even down to the reappearance of characters in the second generation. Knight’s paper, translated into German, was in the university library in Brünn (Brno).

So, without taking anything away from Mendel, the irreplaceable genius of the gene, give De Vries his moment of glory as well. Let his concept of pangens, the interchangeable parts of heredity, stand for a moment alone and unique. Just as the different elements are made from different combinations of the same particles—neutrons, protons, and electrons—so the world now knows, as it did not 20 years ago, that the different species are at least in part made from different combinations of very similar genes.


During the twentieth century geneticists used at least five overlapping definitions of the gene. The first was Mendel’s: a gene is a unit of heredity, an archive for the storage of evolutionary information. The discovery of the structure of DNA in 1953 immediately made Mendel’s metaphor literal, by suggesting how genes could make genes. As James Watson and Francis Crick announced with arch understatement in Nature, “It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material.” Merely by following the base-pairing rule that A must pair with T (and not C, G, or A), and that C must pair with G (and not C, T, or A), each DNA molecule in two stages automatically produces an exact digital copy of its unique sequence. It needs a machine to do the copying, called DNA polymerase; but because the system is digital it loses no precision, and because the system is fallible it allows for evolutionary change. The Mendelian gene is an archive.

A second definition of the gene, only recently revived, is De Vries’s interchangeable part. The stunning surprise from the reading of genomes in the 1990s is that the human being has far more genes in common with the fly and the worm than anybody expected. The genes for laying down the body plan of the fruit fly turned out to have precise counterparts in the mouse and the human, all inherited from a common ancestor called the roundish flatworm that lived 600 million years ago. So similar are they that the human version of one of these genes can substitute for its fly counterpart in the development of a fruit fly. Even more surprising was the discovery that the genes flies use for learning and memory are also duplicated in people—and also presumably inherited from roundish flatworms. It is only a slight exaggeration to say that genes in animals and plants are a bit like atoms: standard parts used in different combinations to produce different compounds. The De Vriesian gene is an interchangeable part.

A third definition of the gene starts in 1902 with De Vries’s contemporary, the English doctor Archibald Garrod, who rather ingeniously identified the first single-gene disease, an obscure ailment called alkaptonuria. From him descends the all too common definition of genes by the diseases they cause when broken, the OGOD definition: one gene, one disease. This is misleading in two ways: it fails to mention that one mutated gene can be associated with many diseases, and one disease with many mutated genes; and it implies that the function of the gene is to prevent that disease. This is like saying that the function of the heart is to prevent heart attacks. Still, given that most genetic research is driven by medical necessity, OGOD definitions are probably unavoidable. The Garrodian gene is a disease averter, a health giver.

A fourth definition of a gene is what it actually does. Right from the start, the pioneers of DNA realized that genes had two jobs: copying themselves and expressing themselves through the construction of proteins. Garrod suggested that genes made enzymes: chemical catalysts. Linus Pauling broadened the point: genes made proteins of all kinds. Then, four months before the discovery of the double helix, James Watson suggested that DNA makes RNA, which makes protein, a concept later jauntily dubbed by Francis Crick as the “central dogma” of molecular biology. Information flows out of the gene and not back into it, just as information flows from the cook to the cake and not the other way. Though many details—alternative splicing, junk DNA, transcription factors, and most recently a plethora of new genes that make RNA but not protein, many of which seem to be intimately involved in regulating the expression of protein-coding genes—have complicated the standard picture of the metabolic gene, the central dogma still holds. With very few exceptions, protein does the work, DNA stores the information, and RNA is the link between them, as Watson guessed. So the Watson–Crick gene is a recipe.

A fifth definition of the gene, which can be credited to the two Frenchmen François Jacob and Jacques Monod, is the gene as a switch and therefore as a unit of development. What Jacob and Monod did in the 1950s was to discover how a bacterium in a solution of lactose suddenly begins to produce the enzyme that enables digestion of lactose, and then stops making it when enough has been produced. The gene is switched off by a repressor protein, and the repressor is disabled by lactose. Jacob and Monod had guessed that something like this must happen, floating the then startling idea that genes were turned on and off by the attachment of proteins to special sequences close to those genes—that, in other words, genes came with DNA switches. Now known as promoters and enhancers, these switches are the key to the development of a body from an embryo. Many genes require several activators to attach to their promoters; activators can work in different combinations; and some genes can be switched on by different sets of activators. The result is that the same gene can be used in different species or in different parts of the body to produce completely different effects, depending on which other genes are also active. There is a gene called sonic hedgehog, for instance, which in one context turns neighboring cells into neurons; in another context, it induces neighboring cells to start growing into limbs. This is one reason that it is risky to speak of a “gene for” something: many genes have multiple jobs.

Suddenly, here is a very different way of viewing genes: as a set of developmental switches. All tissues carry the complete set of genes, but the genes are switched on in different combinations in different tissues. Now forget the sequence of the gene; what counts is where and how the gene is expressed. It is in this sense that many biologists now think of genes. To build a human body means throwing a series of switches in the right order, switches that cause the growth and differentiation of the body. And to make things more interesting, the machines that throw the switches—the transcription factors—are themselves products of other genes. The Jacob–Monod gene is a switch.


Yet, to tell the truth, there were legions of scientists who had been merrily using the word gene since it was coined in 1909 without really meaning any of these five concepts. For them, the gene was not the unit of heredity, evolution, disease, development, or metabolism so much as it was the victim of selection. Ronald Fisher first clarified that evolution was little more than the differential survival of genes. And George Williams and William Hamilton, together with their bulldogs Richard Dawkins and Edward Wilson, finally spelled out the full and startling implications of this idea. Bodies, said Dawkins, were temporary vehicles constructed for the replication of genes, exquisitely designed by genes to grow, feed, thrive, and die—but above all to strive for reproduction. Bodies were the genes’ way of making new bodies. This “gene’s-eye view” of the organism was a sudden philosophical shift.

For instance, it immediately explains something that Aristotle, Descartes, Rousseau, and Hume had not even realized needed explaining: why people are nice to their children (or, in Rousseau’s case, not). People are generally nicer to their own children than they are to other adults, other children, or even to themselves. One or two twentieth-century anthropologists had feebly explained this in purely selfish terms—you are nice to your kids in the hope they will be nice to you in your old age—but here, from Williams and Hamilton, was a genuine explanation that did not take the altruism out of parenting. You are nice to your children because you are descended from people who were nice to their children and were therefore better at enabling their children to survive to breed. This they could achieve because there are genes on their chromosomes which built their bodies in such a way that, given a certain environment, they would reliably produce in an adult behavior leading to reproduction and parental care. Targeted niceness could be in the genes.

Here is a definition of the gene that is neither a unit of heredity nor a unit of metabolism nor a unit of development but a unit of selection. It hardly matters for this purpose what this “gene” is made of. It could be a pair of real genes, or a score. It could be a series of genes acting in sequence. It could be a network of genes, regulated by a plethora of RNAs. What counts is that it reliably produces a certain effect. How does it do that? How can there be a gene that says “Take care of your offspring!” in the language of DNA? And if there is such a gene, how can it thereby take care of itself? The whole concept—best known by Richard Dawkins’s term “the selfish gene”—seemed to many people almost magical. They were so used to thinking teleologically that they could not imagine a gene behaving selfishly unless it had the goal of selfishness in mind. Genes, asserted one critic, are just protein recipes; they “cannot be selfish or unselfish, any more than atoms can be jealous, elephants abstract or biscuits teleological.” But that was simply to miss Dawkins’s point. For the sociobiologists, as they came to be called, the point was that natural selection could cause genes to act exactly as if guided by selfish goals: it was an analogy, but a remarkably useful one. People whose genes caused them, however indirectly, to be nice to their children left behind more descendants than people whose genes did not.

It is now quite easy to build a link from the Watson–Crick gene to the Dawkinsian gene in real cases. Here is one, a gene on the northern tip of the Y chromosome called SRY. It is a tiny gene, just 612 letters long in a single exon (paragraph) of text—as simple as genes get. As a Mendelian unit of heredity, it replicates this 612-letter text. As a Watson–Crick unit of metabolism, it is translated into a 204-aminoacid protein called the testis-determining factor. As a Jacob–Monod unit of development, it is switched on in parts of the brain and just one other tissue—the testis—for just a few hours, usually on the eleventh day after conception (in mice). As a De Vriesian interchangeable pangen, it is found in much the same form in human beings as in mice and all mammals, where it performs a similar function—masculinizing the body. As a Garrodian unit of disease, it is associated with various forms of sexual abnormality, most notably people with normal female bodies who nonetheless possess a Y chromosome but lack a working version of this gene, or mice with normal male bodies who nonetheless possess no Y chromosome but have a working version of this gene inserted into them by devious biologists. Broadly speaking, all an embryonic mammal needs to become a male is to have a single SRY gene, and to become a female it merely needs to lack a functioning version of the same gene.

For those readers who like to know how the engine of a car works, SRY probably masculinizes a body by one very simple action: it switches on another gene called SOX9. That is all it does. Genetically male human beings are occasionally born with one of their two SOX9 genes not working, and most of them develop into women with a skeletal disorder called campomelic dysplasia. SRY seems to be the captain of the ship casually ordering SOX9 to bring the vessel into port before retiring to its bunk. SOX9 does all the work, switching on and off all sorts of genes not only in the testis but in the brain as well—genes such as Lhx9, Wt1, Sf1, Dax1, Gata4, Dmrt1, Amh, Wnt4, and Dhh. These genes in turn switch on and off the production of hormones, which alter the development of the body and in turn affect the expression of other genes. Many may prove to be sensitive to external experience, reacting to diet, social setting, learning, and culture to refract the developing masculinity of the person. Yet it remains true that, given a typical middle-class upbringing, all the vast details of masculinity, as expressed in the modern environment—from testes to baldness to a tendency to sit on the couch drinking beer and flipping between channels on the television—stem from this single gene, SRY. It is surely not absurd to call it the gene “for” maleness.

So you can easily see SRY as an archive, recipe, switch, interchangeable part, or health-giver of maleness—depending on which of the twentieth century’s five definitions of the gene you prefer. You can just as easily see it as a unit of selection, a Dawkinsian selfish gene. Here’s how. One of its downstream effects, inseparable from masculinity, is a greater likelihood that the body will take risks, act violently, and die young. As soon as the testosterone of masculinity begins to bite in late adolescence, the premature mortality of males rises inexorably because of four main factors: homicides, suicides, accidents, and heart disease. This is true even in western societies—indeed, the gap between male and female mortality is widening. Of the major causes of death, only Alzheimer’s disease kills more women than men. Nor is this an aberration of modern life. In some Amazon tribes more than half of the men are murdered. The average rate of violent death among men was higher in hunter-gatherer societies than it was in war-torn twentieth-century Germany.

These risks are part of being a man. Risk-taking is in essence male—though it can be tempered by culture, varied by individuality, and muted by technology. Old-fashioned Darwinian natural selection—the survival of the fittest individual—must struggle to explain this fact. A gene whose consequence is higher mortality should head for rapid extinction. The reason it does not is obvious enough. Risk-averse wimps may live longer, but they do not have more children. The best way to reproduce, if you are a male, is to take a few risks, elbow a few other males out of the way, and impress a few females. If you are lucky and have been born in middle-class California, you can do all this without much chance of actual death—you may leave a few bruised egos and bent fenders behind, but you will probably survive. If you are less lucky and were born the son of a Yanomamo warrior, then your best bet for achieving genetic immortality is to kill and not be killed. In that society men who have killed other men have more than the average number of sexual partners. Whichever, there is no doubt that being a male is bad for survival and therefore fails the test of natural selection. The rational way out of this dilemma is to see the SRY gene, through the downstream effects of masculinizing the body and brain, taking care of its own replication into future generations at the expense of the survival of its current body.

This is sexual selection, Darwin’s other, much neglected theory, which urges not survival of the fittest but reproduction of the fittest. Darwin considered it just as important as natural selection, perhaps more so in the case of human beings, but sexual selection spent most of the twentieth century in scientific exile. In its current form, as refined by people such as Amotz Zahavi and Geoffrey Miller, sexual selection theory suggests that the risk-taking of many male animals results from an unconscious ploy by the genes of a female to expose the genes of males to trial by fire so that she can be sure of selecting the best genes for her offspring. (In some species, it is the other way around.) Even if she passively watches males fighting over her, as seals and gorillas do, by mating with the winner she automatically selects fighting genes for future generations. Sexual selection of this kind can breed any kind of male, from a vicious bully to a precious dandy to a gentle caregiver, and it can act upon the female, too, if exercised by the male. In socially monogamous species such as puffins or parrots, each sex has bright colors to impress the other. In the human species, compared with other apes, there is clearly some degree of male selection for displaying youth, health, beauty, and fidelity among females, while there is some female selection for displaying dominance, health, strength, and fidelity among males.

A peahen that selects the male with the biggest, most ornamented train is unconsciously ensuring that the very act of growing a fancy tail is a test that will reveal the quality of the male’s genes. The more females express such a preference, the more males will inherit the capacity to grow the largest tails they can. To put this in corporate terms, peacock genes cannot be content with manufacturing a good body: they must market it. Like a toothpaste company, they have to put a lot into the advertising budget: the tail. Like an advertising budget, the tail seems a costly luxury, but it is vital. Such ornaments and rituals are, like advertising slogans, signals that try to be dishonest (does good toothpaste really improve your confidence?) but in the process help females honestly identify the genetic quality on offer in the mating market.

Miller argues that it is no coincidence that many human talents—from storytelling to art, from jazz albums to sporting prowess to generosity to murder—tend to be displayed with the greatest vigor by young male human beings at the age of mate selection. Miller points out that human beings devote ridiculous amounts of time to cultural practices that can only rarely enhance survival: art, dance, storytelling, humor, music, myth, ritual, religion, ideology. Yet all these make sense as enhancers of reproductive success, of genetic rather than individual survival.

Genes as units of instinct? The concept has traveled far from Mendel’s hereditary particles. Confusion between many different conceptions of the gene has bedeviled the nature–nurture debate. You will no more find “advertise male quality to females” written into the SRY gene than you will find “advertise male wealth” written into the instruction manual of a Ferrari, but that does not mean it cannot be a valid interpretation of what each is for. Ferraris can be exquisite pieces of engineering at the same time as they can be sexual ornaments, and the same is true of genes.


This abstract concept of the Dawkinsian gene as a unit of instinct first became prominent in Edward O. Wilson’s massive book on animal social behavior, Sociobiology. Wilson, at Harvard, was an expert on the ecology of ants, and he was impressed, as all entomologists soon are, by the complexity of instinct. With no opportunity for learning, insects behave with sophistication and subtlety, but in a characteristic way for each species. The most striking aspect of ants’ behavior is the way they delegate reproduction to a queen. Most ants, as workers, never breed. This fact had puzzled Darwin, and it puzzled Wilson too, for it seemed to represent an exception to the rule that animals strive to reproduce. One day in 1965 Wilson boarded a train from Boston to Miami, having promised his wife he would not fly while their daughter was young. Trapped in the train for 18 hours, he turned to a new scientific paper by an obscure young British zoologist named William Hamilton. Hamilton had argued that the reason so many ants, wasps, and bees were social was a quirk of their “haplodiploid” genetics, which left workers more closely related to their sisters than to their daughters. So, in terms of the selfish gene, it paid them to raise the queen’s offspring rather than their own. Hamilton’s aim was broader than explaining ants—he wanted to draw attention to how such precise genetic calculus explains all cooperation between kin, the degree of instinctive cooperation being neatly related to the degree of relatedness. In other words, people are instinctively nice to their children because their genes make them that way, and their genes make them that way because genes that do so survive—through the children—at the expense of genes that do not.

Wilson at first found the paper naive and foolish and tossed it aside after a cursory reading, but he could not quite pin down its flaw. By the time his train was passing through New Jersey, he was rereading the paper more carefully. In Virginia he was frustrated and angry at Hamilton’s presumption. In northern Florida Wilson was weakening. By the time he reached Miami, Wilson was a convert.

Hamilton’s theory—building on ideas from the self-effacing American, George Williams—dropped into the lives of many zoologists like a map into the lap of a lost explorer. Suddenly, they had a criterion by which to judge an explanation of an animal’s behavior: did it favor the propagation of its owner’s genes? Richard Dawkins explored and expanded the implications of the idea in his beautiful book The Selfish Gene, but unlike Wilson he stuck to animals. Human beings, Dawkins said, were largely exceptions to the rule, because their conscious brains allowed them to ignore the dictates of their selfish genes.

Wilson had no such qualms. In the last chapter of Sociobiology he began to speculate about how human behaviors, too, might be products of scheming genes. Was homosexuality a form of nepotism, genetically induced to allow childless “uncles” to assist cooperative breeding? Did ethics need an evolutionary understanding? Could “the social sciences shrink to specialized branches of biology”? Wilson speculated “in the free spirit of natural history,” but at times he slipped into the evangelical language of the Baptist preachers he had heard in Alabama as a youth. To the extent that he had a hidden agenda, he was motivated more by wanting to tweak the tail of religion than by wanting to fight for nature over nurture. Indeed, he thought he was being mild and pluralist in his interpretation of how genes could collaborate with nurture to produce human social patterns. Aside from a few quasi-Marxist remarks about the inevitability of a planned society in the coming century, he had intended to say nothing overtly political. The storm that broke over his head in November 1975 took him genuinely by surprise.

It began with a letter to the New York Review of Books signed by a committee calling itself the Sociobiology Study Group. Among the 16 signatories were two of Wilson’s colleagues at Harvard and (he thought) friends: Stephen Jay Gould and Richard Lewontin. The letter accused Wilson of providing a new version of an old scheme:

a genetic justification of the status quo and of existing privileges for certain groups according to class, race, or sex…. Such theories provided an important basis for the enactment of sterilization laws and restrictive immigration laws by the United States between 1910 and 1930 and also for the eugenics policies which led to the establishment of gas chambers in Nazi Germany.

As the controversy grew, appearing on the cover of Time magazine the next year, it soon fell into the well-worn tracks of the nature–nurture debate, apparently pitting progressive but merciless environmentalists against conservative but hapless hereditarians. Wilson’s lectures were picketed. Leaflets handed to students in Harvard Square accused him of postulating “genes for all social life including war, business success, male supremacy and racism.” Lewontin accused him of reflecting “the ideologies of the bourgeois revolutions of the eighteenth century,” “bourgeois” being a standard term of abuse among Marxists. While he waited to respond to Gould at a symposium in Washington in 1979, Wilson was suddenly splashed with a glass of ice water by a group of chanting activists.

The argument was no less bitter across the Atlantic. Richard Dawkins, despite having largely ignored human beings in The Selfish Gene except to say that consciousness freed people from the tyranny of the genes, found himself accused of lending intellectual support to far-right politicians. Meanwhile, Wilson’s attempts to explain himself at greater length, in two later books, persuaded some but largely failed to satisfy his critics, who were by now polarized into two extremes. He had encountered exactly the same wounded pride as Copernicus and Darwin: human beings do not enjoy seeing themselves removed from the center of the universe. To see human behavior dethroned from its supremacy and described in the same terms as ants’ behavior was as insulting to the pride of the species as to see the Earth demoted to a planet. Perhaps, also, there would have been less vitriol if Wilson had talked about constellations of innate predispositions rather than “genes.” The idea of a single sequence of DNA having the capacity to determine a human social attitude seemed intuitively wrong as well as humiliating.

Many biologists wedded to the concept of the selfish gene failed to come to Wilson’s aid, causing bitterness that lingers to this day. Some felt that Wilson’s human speculations were naive, premature, and asking for trouble. Others were troubled by Wilson’s imperialism: the boast that biology would soon take over the social sciences seemed at the very least insensitive. Others were merely in search of a quiet life: to defend an alleged racist is to incur the label yourself. Indeed, a sharp division between genetically determined animals and culturally determined human beings was a godsend for most biologists because it freed them:

to pursue their research in peace, without having to fear that they might accidentally stumble into or run afoul of highly charged social or political issues. It offers them safe conduct across the politicized minefield of modern academic life.

The authors of this sentence, two other former Harvard scholars, John Tooby and Leda Cosmides, eschewing such safety, attempted a reform of sociobiology from within in 1992. They argued that the expressed behavior of a human being need not be directly related to genes, but the underlying psychological mechanisms could be. So, to take a simple example, the search for “genes for war” is bound to fail, but the contrary dogmatic insistence that war is a pure product of culture written on the blank slate of impressionable minds is equally foolish. There could well be psychological mechanisms in the mind, placed there by natural selection acting in the past upon sets of genes, that predispose most people to react to some circumstances in warlike ways. Tooby and Cosmides called this evolutionary psychology. It was an attempt to fuse the best of Chomsky’s nativism—the idea that the mind cannot learn unless it has the rudiments of innate knowledge—with the best of sociobiology’s selectionism: the idea that the way to understand a part of the mind is to understand what natural selection designed it to do.

For Tooby and Cosmides it is the whole developmental program that evolves, the program for creating an eye, a foot, a kidney, or a language organ in the brain. Each program requires the successful integration of hundreds, perhaps thousands, of genes (many of them pangens used in other systems as well), and the presence of expected environmental cues. This is a subtle mixture of nature and nurture that studiously avoids putting the two in opposition to each other:

Every time one gene is selected over another, one design for a developmental program is selected over another as well; by virtue of its structure, this developmental program interacts with some aspects of the environment rather than others, rendering certain environmental features causally relevant to development…. Thus, both genes and the developmentally relevant environment are the product of natural selection.

But, crucially, the environment is not an independent variable. The design of the developmental procedures specifies the environmental effects that will be used. Royal jelly turns a bee larva into a queen, but it does not turn a human baby into a queen. Genes, for Tooby and Cosmides, are designed to expect certain environments, and to make the most of them.

Despite this renewed emphasis on the environment, Tooby and Cosmides ran into the same political problem as Wilson and Dawkins. The social science establishment, liking their ambitions with regard to its subject matter no better than it had liked Wilson’s, painted them as extreme reactionary nativists. I think this is a radical misinterpretation. For me, Tooby and Cosmides represent a retreat from naive nativism toward an integration with nurture. The subject they helped to found—evolutionary psychology—is as comfortable with nurture explanations as it is with nature explanations. In the hands of Martin Daly and Margo Wilson, for example, it has been used to explain patterns of homicide and infanticide. Daly and Wilson recognize the role of sexual selection in making young adult males the prime perpetrators of murder, for example, but recognize just as strongly the role of the environment in producing the situations that actually elicit murder. The evolutionary psychologist Sarah Hrdy has hypothesized that juvenile human beings are “designed” by their past to expect to be reared communally rather than in a nuclear family. It is impossible to parcel these studies into “nature” or “nurture.” They are about both. As Hrdy has put it:

Nature cannot be compartmentalised from nurture, yet something about human imaginations predisposes us to dichotomise the world that way…. Complex behaviours like nurturing, especially when tied to even more complex emotions like “love,” are never either genetically predetermined or environmentally produced.

The main complaint Tooby and Cosmides have against the social sciences is their desire to insulate themselves from other levels of explanation (to the cry of reductionist!). Durkheim famously declared: “Every time that a social phenomenon is directly explained by a psychological phenomenon, we may be sure that the explanation is false…. The determining cause of a social fact should be sought among the social facts preceding it and not among the states of individual consciousness.” In other words, he rejected all reductionism. Yet other sciences have successfully integrated “lower” levels of explanation without losing anything. Psychology uses biology, which uses chemistry, which uses physics. Tooby and Cosmides wanted to reinvent psychology in such a way that it used genes, not as implacable determinists of an inevitable human nature, but as subtle devices designed by ancestral selection to extract experience from the world.

The beauty of Tooby and Cosmides’s gene, for me, is precisely this. It integrates all the other six definitions and adds a seventh. It is a Dawkinsian gene with attitude (in its dependence on passing the test of survival through the generations); a Mendelian archive (inscribed with the wisdom derived from millions of years of evolutionary adjustment); a Watson–Crick recipe (achieving its effects through the creation of proteins via RNAs); a Jacob–Monod developmental switch (expressing itself only in precisely specified tissues); a Garrodian health-giver (ensuring a healthy developmental outcome in the expected environment); and a De Vriesian pangen (reused in many different developmental programs in the same species and in others). But it is also something else. It is a device for extracting information from the environment.

SRY, the masculinizing gene on the Y chromosome, might seem at first glance to be a genetic determinist of the kind that gives social scientists the vapors. I have suggested that it sets in motion the sequence of events that (usually) leads to men sitting on couches drinking beer and watching football while women shop and gossip. But looked at another way it is the ultimate servant of nurture. Its job, aim, and desire in life—with the help of hundreds of downstream genes—is to extract certain kinds of information from the upbringing and environment of its landlord organism. It extracts the food needed to grow a masculine body, the social cues needed to develop a masculine psyche, the gender cues needed to develop a masculine sexual preference, even the technology needed to express a masculine personality in the modern world (toy guns, say, or remote controls). It—or rather the developmental program it starts—can be steered and adjusted by changes in that environment along the way. Take a baby boy from medieval Europe and transport him through time to modern California for his upbringing, and it is a fair bet that his mind would be fascinated by guns and cars in place of swords and horses. SRY is no more than a glorified nurture-extractor.

Here again is the author’s message of this book. Genes themselves are implacable little determinists, churning out utterly predictable messages. But because of the way their promoters switch on and off in response to external instruction, genes are very far from being fixed in their actions. Instead, they are devices for extracting information from the environment. Every minute, every second, the pattern of genes being expressed in your brain changes, often in direct or indirect response to events outside the body. Genes are the mechanisms of experience.

Previous: 8. Conundrums of Culture
Next: 10. A Budget of Paradoxical Morals