NINE

The Tender War

Et plus le temps nous fait cortège
Et plus le temps nous fait tourment
Mais n'est-ce pas le pire piège
Que vivre en paix pour des amants?
Bien sûr tu pleures un peu moins tôt
Je me déchire un peu plus tard
Nous protégeons moins nos mystères
On laisse moins faire le hasard
On se méfie du fil de l'eau
Mais c'est toujours la tendre guerre.^*

—Jacques Brel, La chanson des
vieux amants, 1967

Visiting Anthropologists

NO MEMBER OF ANY OTHER SPECIES than our own has ever been appointed to a university position in social anthropology. That's a pity in lots of ways. We are far from being the only species with a talent for the study of animal behavior: many predators, for example, have an exquisitely fine understanding of the behavioral weaknesses and idiosyncrasies of their prey. Instead we rely for our understanding of human society on the insights of social anthropologists (as well as psychologists, sociologists, and economists) who are, if not exactly blinded, at least desensitized to many of the truly weird aspects of human behavior by the fact that they belong to the very species they study.

So we have to imagine what anthropologists from another species would make of us. To make it easier, let's stick to the apes: it's too hard to imagine what dolphins might make of our landlubberly clumsiness. Chimpanzees would undoubtedly marvel at our ability to congregate in groups without fighting:

Members of Homo sapiens appear not even to notice, let alone react to, what would seem to us intolerable invasions of their bodily boundaries. Many, even most, individuals are capable of accepting the uninvited proximity of others for an entire day without once coming to blows. Our initially promising field study “Violent Conflicts in the Tokyo Subway” had to be abandoned after weeks of patient observation failed to yield a single recorded incident.

Bonobos would surely be amazed by something else:

It cannot fail to strike the most unbiased observer that there is a startling discrepancy between the vast amounts of time and energy that members of Homo sapiens devote to thinking, talking, and agonizing about sex and the microscopic amounts of time and energy they devote to actually engaging in it. Apparently healthy males and sexually receptive females appear capable of spending hours in each other's company without ever making physical contact, though reliable native informers report that neural activity can become highly agitated under such circumstances. It's not as though they have much else to fill their time: foraging for food requires only two or three hours per day, and the rest of the day is spent in elaborate display activities, involving frequent changes of bodily decoration, that almost never lead to intercourse. Our initially promising field study “Sexual Intercourse among High-Ranked Business Executives” had to be abandoned after weeks of patient observation when the only incidents worth recording were so infrequent and of such short duration that they initially escaped the attention of our field researchers, who were themselves having sex at the time.

Chimpanzees fighting, Liberia. © Clive Bromhall, Oxford Scientific.

What gorilla anthropologists would notice most about Homo sapiens (apart from the rarity and furtiveness of our polygamy) would probably be the way in which adults' lives are ruled by children:

Juvenile members of Homo sapiens expect to be fed and groomed long after they are physically capable of foraging. Social codes entitle and even expect juveniles to object to all kinds of physical contact that we would consider routine; assaults on them are considered undesirable and even shocking, rather than an occupational hazard of their proximity to adult males. This deference to juveniles appears to be part of a systematic wider pattern in which alpha individuals frequently permit themselves to be pushed around by the betas. Indeed, high-ranking individuals frequently adopt a kind of forced meekness in the face of coordinated hostility from groups of low-rankers who believe them to have abused their high rank, an attitude that apparently serves only to encourage the betas.

Female bonobos neglecting their anthropological fieldwork. © Frans Lanting / Corbis.

As professional anthropologists, these apes would take care to avoid expressing any shock or even moral disquiet at our behavior but would do their best to understand it as an adaptive response to the constraints of our environment. A mixed-species research team might well begin to conjecture that these three aspects of our lives were somehow connected: our low levels of intragroup violence; our much greater concern with thinking and talking about sex than with actually engaging in it; and our organization of society around the needs of children. By pooling their insights, they might come up with a story that went something like this:

Homo sapiens has colonized an evolutionary niche that depends on making large and elaborately cooperative investments in its offspring.¹

This niche requires Homo sapiens to engage in complex webs of cooperative activity, in which just about every important activity is the product of continuous teamwork. In recent millennia this has also led Homo sapiens to live in large, densely populated communities to make the most of this cooperative potential.

But the same overcrowding and mutual dependence that make such cooperation possible also create massive temptations for violent and antisocial activity, notably the killing of rivals for scarce economic and sexual resources but including also a vast number of lesser antisocial acts.²

Human societies would therefore implode under the weight of their own social proximity and the complexity of their mutual interdependence unless the actions of their members were continuously supervised and controlled. All human beings live in what are, by ape standards, effectively police states, in which the police are everyone else.

In particular, individuals who transgress against the elaborate social codes that govern acceptable behavior provoke not just individual retaliation but also the collective retaliation of coalitions of others. This applies even (and perhaps especially) to high-status individuals, who, because of their access to powerful means of coercion, would be especially tempted to abuse and eliminate rivals. Their acts are scrutinized minutely by lower-status individuals, none of whom would be credible challengers on their own but who in coalition constitute a formidable countervailing force.³

In theory, such a society could treat abundant sex as a form of pressure-valve against violent tensions and a reward for peaceable cooperation, much as bonobo societies do.⁴ That is a feasible option for bonobo societies because most of the costs of a bonobo pregnancy fall on the mother, who is the one who controls access to the sexual rewards in the first place. She does so because (as in most species, most of the time) males demand as much sex as females are willing to concede them, and the female decision therefore determines how many sexual acts take place.

But the evolutionary niche occupied by early Homo sapiens meant that every pregnancy imposed large costs on fathers, grandparents, neighbors, and siblings, on the whole supporting team that each newborn required for its survival. In a world without contraception and delicately dependent on maintaining the incentives for cooperation, those costs of sex could not be socialized unless the decisions about engaging in sex were socialized too. The human police state was therefore extended to every aspect of its members' sexual lives. The sanctions it imposed were more severe for women than for men because women, as the more selective sex, were the ones who controlled access to sexual rewards.

It's well known that in police states people spend much more time talking and thinking about transgressions than actually engaging in them. Indeed, talking and thinking about transgressions become a form of intense vicarious pleasure that is compatible with high-pressure group living because it usually has much milder social consequences than those of the transgressions themselves.⁵

These features, in a nutshell, are what likely strike our cousin apes about the links between these three quite un-apelike aspects of our otherwise very apelike human societies. Still, even their professional objectivity might face a sore test when they tried to make sense of the Bill Clinton-Monica Lewinsky story, in which for a couple of years in the mid-1990s the world's most powerful country placed its economic policy, domestic policy, and foreign policy second in importance to determining whether the highest-ranking alpha male in the entire world had engaged in consensual oral sex with someone other than his official partner.⁶ Had nobody told Homo sapiens that oral sex doesn't lead to pregnancy? Wasn't this a case where the vicarious pleasures of censoriousness had massively larger social consequences than whatever pleasures the participants might have found in the original acts? Was the sexual police state starting to spiral out of control?

Contraception and Consequences

The widespread availability of contraception in the modern world does not mean that sex now has no consequences—far from it. But it does have very different consequences from those it had in the hunter-gatherer communities in which both our social habits and our emotional reactions to each other's behavior first evolved. And it's important to ask whether those habits and reactions are as appropriate now as (in the story helpfully offered by our ape anthropologists) they might once have been.

We are incorrigibly curious about the sexual behavior of others. We also move at lightning speed to judgment, reserving distaste and disapproval for those who are more sexually active than we are (or more sexually preoccupied, or merely differently so), and condescension and scorn for those who are less active.⁷ These attitudes may even be self-reinforcing, since our disapproval is heightened by the suspicion that those who are more active than we are feel condescension and scorn for us, and our scorn is likewise heightened by the suspicion that those who are less active regard us with distaste and disapproval. Our reactions sometimes seem to be on a hair trigger, so that behavior that provokes admiration in one context may just as easily provoke disgust in another. All of this makes sense in the context of our evolution. In hunter-gatherer societies the sexual behavior of others had potentially massive consequences for us, not just because it might increase the number of mouths to feed but also because sexual behavior led to the forging and breaking of coalitions on which the very lives of our ancestors depended. It's not surprising that the sex lives of others, and the possibilities opened and closed to us by our own sexual lives, should have become an object of intense reflection and introspection.

Not only do we have intense desires, but we also have desires about our own desires and the desires of those we are close to. This is true not just about jealousy. In no other domain is there so much anxiety about how to maintain and stimulate desire. People don't worry about how they can enjoy their food more, or how to savor power. They don't buy pharmaceuticals or concoct ancient herbal remedies to stimulate their desire to watch football, or even to induce their partners to watch football. Our worries about the adequacy of sexual desire are not some dysfunctional by-product of our large brains (it's not because our cortex hasn't got enough to do that we find ourselves contemplating desires about desires). They are the direct result of natural selection. Our sexual partnerships founder without mutual desire, and we are a species for whom life is about partnerships, if it's about anything at all. In hunter-gatherer societies, sexual partnerships were the fulcrum of social cooperation, so of course we care about the desire on which that cooperation hinged.

In a world with contraception, and in which partnerships are built on many more affinities and for many more diverse purposes than in hunter-gatherer society, it seems likely that the intensity of our focus on desires about desires may be misplaced. Infidelity is less serious the less likely it is to lead to pregnancy, and the inadequacy of sexual desire is less serious the more alternative foundations of affinity we have in our social lives. Our inability to treat either of these possibilities less seriously than we used to is likely to be causing us needless distress. Our emotions, which started out as natural selection's way of directing our attention to things that mattered for our fitness, have become the things that matter in themselves, and they still matter even when the fitness landscape in which they used to help us navigate has changed beyond all recognition. As we saw in chapter 3 in relation to the somatic markers hypothesis and the role of emotions in signaling commitment, the relative inflexibility of our emotional responses was not necessarily a disadvantage during prehistory and in some circumstances was positively adaptive. It is likely to mean, though, that our emotional reactions to the circumstances of the modern world are still colored by traces of our prehistoric past.

At the same time, changes in society due to contraception have also brought men and women into proximity in their working environments in new ways. Women have always been at risk of sexual assault from men (domestic servants in particular must have been very often coerced); but the context in which women face sexual coercion has changed significantly. Sexual assault is now a greater risk for women when they are trying to do the same job as men: in particular, they may have fewer other women in a position to act as witnesses on their behalf. Certainly, the conditions of hunter-gatherer societies often kept sexual coercion in check (outside the marital relationship, at least) because women typically forage in groups.⁸ Agricultural societies, with their hierarchy and their dependence on slaves and servants, created massive opportunities for sexual coercion, though it was still rare for men and women to work together in conditions of privacy. Now women are much more likely to find themselves alone in the company of men in a way that facilitates both consensual and coercive sexual encounters. Men can use (or threaten to use) economic power or physical force in order to coerce women, and to keep the fact of coercion secret, in ways that would have been difficult in the intimate setting of a hunter-gatherer group. The same police state that overreacts to consensual transgressions of its norms has great difficulty devising an adequate reaction to the continuing prevalence of sexual coercion.⁹ Indeed, there are many examples in human societies of sexual coercion being explicitly encouraged as a means of enforcing male norms on women and of punishing women who are inclined to challenge these norms.¹⁰

Contraception is not the only technological change that has turned our sexual lives inside out. Another truly radical change is photography, and specifically the technology that has made it possible to transform our public as well as private spaces into a panorama of sexual signals entirely detached from the intentions of those who once sent them. Not only can we imagine ourselves to be receiving sexual signals from people we have never met and who care nothing for us, but we can even be distracted by signals where we would normally never expect them. While driving down a busy street, I may receive what seems like an unmistakable sexual invitation from a smiling woman dressed only in her underwear. It is indeed an invitation; but it's an invitation to go shopping. After repeated exposure, I have now rewired my brain to reinterpret the signal, but it's not surprising that some people find that adjustment difficult. One of the reasons men from more conservative societies have often considered modern capitalism to be revoltingly sexually corrupt is that they repeatedly, and understandably, confuse invitations for shopping with invitations for sex.¹¹ Among the many things they thereby fail to grasp is the extent to which public nudity desexualizes the body. A woman can be represented, or interpreted, as nude and bored, nude and indifferent, nude and oblivious. In the cacophony of the modern street, she is usually all three at once. This doesn't make her images, and the use made of them by advertisers, a sexually trivial matter—but it imposes difficult interpretive demands on our stone-age brains.

Changes in communication technology have had a no less revolutionary impact on our sexual lives. Industrialization radically expanded people's opportunities to meet strangers, compared to their previous lives in agricultural villages (just as agriculture radically changed the opportunities faced by hunter-gatherers). The invention of the telephone added a new dimension to those opportunities, which the Internet and mobile telephony have recently revolutionized once again.¹² Every opportunity to meet a stranger is an opportunity for exchange—of ideas, of economic resources, of sexual favors. It's no wonder our hunter-gatherer emotions find it all puzzling, intriguing, arousing, terrifying. And our responses to the behavior of others—the part we all play in the sexual police state created by our hunter-gatherer ancestors—have now been magnified by those same changes in communication technology and can now resonate around the world.

Lessons from Evolution

Sex need no longer result in pregnancy, and pregnancy need no longer interfere with a woman's capacity to engage in valuable economic activity if she wants to, but many of our instinctual responses to both sex and pregnancy continue to shape our social world. What does that mean for us today? Can we modify these instinctual responses, and should we try?

It's common to claim that whatever we may learn about human evolution cannot determine what we ought to do, because moral prescriptions cannot be logically derived from facts or hypotheses (you cannot derive an ought from an is). While true, that claim is not very interesting, because many scientific facts do indeed give us a basis for action, even if that basis is not one of logical deduction. For instance, the discovery that smoking substantially increases the risk of getting lung cancer gives people a reason not to smoke. It's not that you can logically derive the injunction “Don't smoke” from the factual claim. Instead, the factual claim provides a reason not to smoke for people who already accept a model of reasonable action that puts a good deal of value on staying healthy. Hard to believe though it may seem now, it was once common for cigarettes to be advertised as positively beneficial to health (advertising slogans included “Give your throat a vacation—smoke a fresh cigarette,” “Light an Old Gold instead of a throat treatment!” and “More doctors smoke Camels than any other cigarette”).¹³ Scientific knowledge has made it impossible for such claims to be taken seriously today. As for people who don't place value on staying healthy, it's not clear that there's much to be gained from arguing with them anyway.

Advertisement for the healthiness of smoking, 1946. © Apic / Getty Images.

The evolutionary history of Homo sapiens yields very few conclusions about social behavior that are anywhere near as firmly established as the hypothesis that smoking causes lung cancer. Nevertheless, it has led to an accumulation of evidence that casts into serious question several commonly held beliefs about the relations between men and women that continue to shape how we think about our personal lives and about public policy. These beliefs fall into two main groups, which we can call the model-relationship view and the divided-workplace view. The first consists of beliefs based on the idea that there exists a model form of sexual partnership that can avoid conflict between men and women if we only follow its instructions. The second consists of beliefs based on the idea that men's and women's relationship to the world of work must remain different because of unavoidably different talents, aptitudes, or constraints. Scientific evidence hasn't conclusively proved either idea to be wrong: it has just chipped slowly away at the foundations of their credibility. Let's look at them in turn.

The Model Relationship

Many people believe that nature has made it possible for men and women to have reasonably conflict-free relationships if they only follow certain rules. Many of the world's religions treat marriage as a sacrament conceived and guided by God. Those who think of marriage as a human and not a divine institution often believe that passionate human relationships have been shaped by natural selection in the same way that natural selection has shaped the organs of the human body, to be exquisitely fitted for the purpose of creating and raising new human beings. Such relationships, according to this view, rest on some basic affinities: emotional and intellectual companionship, erotic passion, sexual fidelity, a deep capacity for empathy between partners, and a natural willingness for each partner to take the other's interests as their own. Even if marriage in former ages was an economic contract that gave little space to erotic passion, in the United States and in other prosperous democracies in the twenty-first century it is increasingly seen as both permitting and demanding very high standards of aspiration along many dimensions of the relationship.¹⁴ Single people may take a lot of time to find the right person for a permanent partnership, but once they have found someone, if the partnership is “right,” love will ensure that the fundamental interests of the partners are more or less aligned.

In this view, erotic courtship generates sparks that either die down very quickly or ignite the flames of true love: there is no intermediate state. If the fire really does take off, its very uncontrollability, the sense it gives of sweeping all before it, makes it easy to imagine that passion has flamboyantly driven out reason and that true love inhabits a world far removed from the charmless compromises of daily self-interest. We can no more subdue our erotic passion than we can will our heart to stop beating. It's tempting, then, to think of erotic passion as having a design as intricate and as minutely adapted to our survival as the physiological design of the human heart.¹⁵ It's tempting also to think that negotiating and compromising in the pursuit of passion would entail placing our own petty interests above those of our lovers, which would be tantamount to demonstrating a failure of passion itself. If serious conflicts of interest later arise, if there is infidelity, if mutual erotic desire disappears, if the partners are less than fully honest with each other, the relationship is faulty and quite probably was not based on “real” love in the first place. Some (though not all) of those who hold this view also believe that the ability to make a passionate relationship work is a litmus test of honesty and decency: we can therefore judge people's fitness for public office in part by the success of their marriages or civil partnerships.

The biology of human evolution strongly suggests that this model-relationship view is mistaken and that countless couples may be suffering needless unhappiness because they continue to believe in it despite the evidence to the contrary. Nothing in the logic of natural selection is incompatible with sexual conflict. On the contrary, conflict is written into the very heart of the sexual relationship, and it occurs because of, not in spite of, the huge potential for cooperation on which that relationship is founded. Natural selection does not fashion optimal relationships, not even in the limited sense in which it has fashioned optimal physical hearts. In particular, nothing in evolution would have selected for relationships that last a lifetime. If relationships do last a lifetime, it is because the parties can be lucid and constructive about reconciling their conflicting interests. Sexual infidelity is widespread in nature, among females as well as males. That in itself says nothing about how good or bad it may be (murder is also widespread in nature). But the fact that infidelity occurs among individuals who love and want to stay with their partners, as well as among those who do not, implies that it is not incompatible with relationships that last. Couples who wonder why conflict persists between them and suspect it is because they are not “really” in love; single people contemplating whether to commit to marriage or a permanent relationship and trying through introspection to decide whether what they feel is “the real thing”; people whose partners have been unfaithful and who think their relationships have been irremediably tarnished; couples who think that waning erotic desire must be a signal that something invisible is going seriously wrong in the rest of their relationship— all these people may be perplexed or unhappy, and for valid reasons. But they may be making their unhappiness far worse than they need to by comparing themselves to a standard they falsely believe to be natural and possible.

Love may be a many-splendored thing, but its splendors do not include the abolition of conflict, so the persistence of conflict tells us nothing about the nature of love. The fact that your partner is completely unpersuaded by your political views says nothing about how much he loves you: love may make him want to give you his body but not to rent you his brain. If you've convinced her to become a vegetarian for your sake, the fact that your partner continues to crave meat tells you something about the robustness of her appetite but nothing at all about the robustness of her love. If the fact that she continues to be drawn to other sexual partners means you can no longer bear to live with her, leave her if you must, but don't claim it's because you've realized her love for you isn't real. If he's less than honest with you on something you care about, reproach him for his dishonesty, but don't presume he can't really love you: be lucid enough to realize that the greater his love, the greater his incentives to exaggerate or dissemble in its pursuit.

Although jealousy, especially male jealousy, is sometimes regarded as a violent and primitive emotion that is not under voluntary control, it is highly sensitive to social context, and we know that people who in one setting would be “overcome” by jealousy may find themselves capable of mastering their emotions when it is in their interest to do so.¹⁶ As we saw in chapter 4 when discussing the evidence that females had multiple partners during prehistory, various cultural practices can help to soften the disruptive potential of male jealousy. Indeed, male emotional responses to a woman's multiple sexuality are complex, involving not only insecurity and potential aggression but also some element of arousal and attraction (after all, much pornography aimed at men depicts women engaging in sex with other men, and its purpose is to arouse men sexually rather than to provoke them to violence). Different ways of portraying women's potential for multiple sexuality can tip the balance delicately between seeing it as threatening and seeing it as glamorous or charming, as in this commentary from Sam Wasson's book about Audrey Hepburn and the filming of Breakfast at Tiffany's. What exactly was it about the famous “little black dress” that Hepburn turned into an unforgettable symbol of chic glamour? “Centuries ago, black dye was affordable only to the very rich.…[I]n the Victorian era…it was worn almost exclusively by those in mourning.…[I]t was a surefire sign of widowhood. To the men passing by, it signified the wearer's knowledge of sex. It meant experience. No wonder the flappers of the 1920s were drawn to it.” Wasson observes that to Hubert de Givenchy, designing a dress for Audrey Hepburn to wear in Breakfast at Tiffany's, “seeing how this was a dress to be worn by Audrey Hepburn—and not at night, but in the very early morning—it was unusual to say the least. Because it's Audrey—wholesome, wholesome Audrey—there is irony in her endorsement of a color heavy with unchaste connotations.…[T]he contrast is sophisticated. Black on Audrey Hepburn gives her an air of cunning…that's the essence of glamour.”¹⁷ In short, in the right cultural context and with a sufficiently deft touch, male jealousy can be channeled into something a lot more playful and a lot less threatening.

Audrey Hepburn and the “little black dress” designed by Hubert de Givenchy for her to wear in the film Breakfast at Tiffany's, 1961. © Hulton Archive / Getty Images.

The model-relationship view also has a powerful impact on politics in many countries, making candidates for public office feel obliged to project images of themselves as good spouses and parents. In some countries (notably the United States), the breakdown of a marriage can often damage, and sometimes end, a person's political career. I'm not aware of any scientific study that has shown that people whose relationships last, and who avoid infidelity, are any better at serving in political office than people who don't, and it would be hard to imagine how such a study would be conducted. Since even the pretense of scientific evidence is almost never offered in such cases, it may be that those who join the hue and cry after politicians whose relationships break down are not really persuaded by evidence at all. They may merely be indulging the pleasure of censoriousness that we've already seen to be a highly evolved part of our nature, as well as using against political opponents the notoriously hair-trigger character of our disgust at other people's sexuality.

In any case, if there were such evidence, it would almost certainly indicate that the case for resigning from politics when your marriage breaks down should apply also to business, journalism, and indeed most other professions. The toll such resignations would take on productive organizations is terrifying to contemplate. In the absence of evidence, it's probably best to consider the periodic outbursts of public indignation at the relationship misfortunes of others as a kind of blood sport. Our delight at interfering in the consensual relationships of others would be amusing if it did not divert our attention from the much more serious (and much more difficult) problem of sexual coercion.

In short, the model-relationship view may represent some people's ideal, but an ideal has to be attainable if it is not to paralyze with self-reproach the people who fail to live up to it and make it harder to be lucid about living with relationships that do not conform to the model.¹⁸ The biology of human evolution indicates clearly to us that the persistence of conflict among couples need not mean the end of love, that deception or infidelity need not mean the end of a relationship, and even that the end of a relationship need not mean the end of someone's world. To say this is much more than to repeat the truism that no relationship is perfect: it is to reject a specific biological model of relationships as being as well adapted as organs of the human body. A cardiologist trying to repair your physical heart needs first to understand how the model heart works. The repair of emotional heartbreak needs a different approach, one less wedded to an ideal model of common interest and more open to the lucid expression of what the partners want and need from each other, even and especially when these wants and needs are in conflict.

The Divided Workplace

The persistence of gender disparities in salaries in many occupations, as well as the persistent underrepresentation of women in powerful positions in business and public life, leads certain people to conclude that there are things about modern working life and about being a fulfilled woman that do not quite mix. We saw in chapter 4 that in prehistory, there was a deeply ingrained division of labor between men and women that resulted in entrenched economic inequalities. The economic basis for that division of labor has now completely disappeared, but the inequalities, though much diminished, are still with us. We saw in chapters 5 to 7 that there is no evidence that these disparities have anything to do with differences in talent between men and women. Instead, there is substantial evidence suggesting they are due to a combination of two factors:

1. Differences in preferences for which women pay a high (and probably unreasonably high) price. Taking career breaks and working fewer hours than many men appear to incur particularly large disadvantages. These choices have an adverse effect on women's advancement not just in the child-rearing years but for decades afterward. It makes a difference also in that men are rewarded, as women are not, for more aggressive bargaining strategies in the professional environment.

2. Subtle differences in networking strategies between men and women that can operate to make the talents of women less conspicuous to potential colleagues and employers than those of equivalently talented men.

These two factors are not alternative explanations but complementary ones, in that the second factor explains how the first can be so powerful and persistent. It is the differences in networks that explain why entrepreneurs do not easily move to take advantage of the opportunities offered by talented but underrewarded women in the modern workforce. The reason why women pay such a high price for their choices appears to have something to do with a signaling trap, in which it is difficult for women to signal their talents and their motivation consistently with organizing their lives to accommodate the many goals they can and should want those lives to accomplish. Men often pursue a less diverse set of goals, partly because of social conventions that encourage them to do so, and a by-product of this more single-minded strategy is that their signals are easier for employers and colleagues to interpret.¹⁹

Can greater understanding of the causes lead to a less divided workplace, and should we want it to? Obliging or pressuring women to act against their preferences is hardly to respect their dignity or their autonomy. Two things suggest that a better understanding of the causes of the divided workplace may lead naturally to some softening of the divisions, without any compulsion or bullying involved. First, some of the apparent differences in preferences between men and women are likely to reflect differences in their constraints that may erode over time as the true costs of those constraints become clear. For instance, the professional single-mindedness that appears to be preferred by many men may be due partly to their being more able than women to find flexible partners who accommodate them. Some professional women might indeed be as single-minded as their male colleagues if only they could find partners like theirs. As the title of Terri Apter's book neatly expresses it, working women don't have wives.²⁰ If the ability to pursue a career single-mindedly without sacrificing a home life becomes something that any individual needs to negotiate openly with a partner, rather than presuming it as so many men have done in the past, gender differences in these choices may persist, but there may be less stark differences than we see now. And those differences that do persist may be easier for both parties to live with if they are chosen rather than imposed.

Second, merely understanding better the causes of the divided workplace may lead to more imaginative initiatives on the part of employers and colleagues. For instance, understanding the fact that individuals who have taken career breaks are underrewarded compared to similarly talented men may lead more employers to seek out such women systematically. They have a particular incentive to do so if they realize that the lower rewards such women receive have everything to do with their lack of conspicuousness once they drop out of male networks and rather little to do with any lack of talent or commitment revealed by their career breaks per se. They may also come to see the virtue of recruiting outside the networks of people who have spent their entire lives working for the same kind of firm without a break, and who may therefore be blind to the complacent assumptions that industry insiders routinely take for granted. True, corporate cultures are intrinsically conservative, and corporate recruiters are not very good at looking systematically for recruits who differ in important ways from themselves. That's why competition—and especially openness to entry from firms outside the industry—is important in stimulating innovation in all dimensions, in recruiting procedures no less than in products. But in a signaling trap, expectations are important too. If people who work in business routinely expect to see all-male boardrooms, that makes it easier for them to justify making little effort to recruit outside the circles they know. If they start to understand—from the way shareholders talk to them at shareholders' meetings, from the reactions of clients and customers, from what their colleagues in other firms report, from what they read in the press—that there's something a little odd about choosing your senior executives from only half of the available talent pool, then even the most complacent recruiters might begin to feel uncomfortable with the status quo.

Coupled with the increase in the educational attainments of women compared to men in recent years (described in chapter 8), innovative recruiting initiatives may mean that over the next two or three decades, the differences in economic power between men and women will largely disappear.²¹ Still, among the talent pool of individuals in their forties and fifties from whom today's corporate boardrooms are recruited, women are already slightly more educated, on average, than men.²² So if the startling disparity in their representation in senior positions (women make up only 2.4 percent of Fortune 500 CEOs, for instance) can persist in these conditions, there's no guarantee that it cannot hold out against the even larger disparity in educational attainment that will appear over the next twenty to thirty years. The evidence reported in chapter 8 suggests also that there might even be some tendency to appoint women to nonexecutive positions to avoid addressing the imbalance in appointments to executive jobs, where the real power in organizations is exercised. So even an advance in women's representation in visible positions, such as nonexecutive directorships, might not do much to shift the balance in the executive positions.

Can or should public policy realistically do anything about this? I suggest in chapter 8 that a fundamental challenge for anyone in a modern economy is to attract and keep the scarce attention of potential employers, clients, and colleagues in a world of information overload. If women are to compete equally with men, they need to overcome what appears to be an attention deficit, to make their talents as conspicuous as those of equivalently talented men. I've already suggested that many of the most appropriate actions to redress this deficit should be undertaken through individual initiative rather than public policy. It's certainly appropriate to be wary of wielding clumsy policy instruments to deal with a very subtle and elusive problem. Still, two kinds of public policy have often been suggested to complement the kinds of antidiscrimination provisions that are already part of the legal framework in many countries, and it's worth examining both of these in greater detail.

The first type of policy aims to address imbalances in the attention of recruiters. Requiring simple parity in employment is unworkable: it is not only unrealistic to require that firms employ equal numbers of men and women, but even if it could be achieved, it would be at a large cost in flexibility and adaptability, virtues that firms in the twenty-first century need more than ever. But it's much less unrealistic, and much less intrusive, to require firms to choose the men and women they employ from short lists that reflect a reasonable balance of the sexes. The reasonable balance would vary according to the occupation: when only 1.3 percent of airline pilots are women, it won't be feasible to insist there be 50 percent of women on every short list for recruitment of pilots, but it might be feasible to insist that every short list contain at least some women (unless it can be shown that in a wholly open recruitment process, none have applied). Balanced short lists also have a major advantage over gender quotas: nobody need feel they got their job only because of the quota. Even if it's true of some women that they got the job because the quota ensured they made it to the short list (which will necessarily be the case, otherwise the policy would be ineffective), it will still be true that they beat the other candidates on the short list in a fair fight.

Of course, such a policy would be completely ineffective if firms systematically treated it as window dressing, adding women to their short lists but not appointing them to jobs in any greater proportions than before. But the evidence we've discussed up to now does not suggest that firms are systematically trying to exclude women: their actions are having that effect largely through a massive failure of imagination, a failure to spot the pool of talent that exists just outside the field of vision of their insular, self-perpetuating networks. If that's so, requiring more women on short lists would help to shift more of that talent inside the corporate field of vision, and the evidence suggests that this might substantially alter recruitment outcomes. But it's important to push the argument to its logical conclusion. If such a policy worked, it would not need to be compulsory: it would be in companies' own interest and could therefore equally well be implemented through voluntary codes of practice and through certification procedures that signaled to the outside world that firms take the gender balance of their recruitment procedures seriously. A quality label that firms could attain only by systematically gender-balanced short-listing might well be a sufficiently attractive signal in the labor market that firms would want to make such commitments without any legal compulsion.

A second type of policy aims to undermine the signaling trap by ensuring that career breaks taken for the purpose of raising children are not interpreted as a lack of motivation or talent for employment. Compulsory paternity leave that precisely matches the availability of maternity leave is one such measure.²³ There is a rationale for paternity leave that has nothing to do with its effect on the gender balance in employment: namely, the idea that parents have a responsibility to the rest of society for the way they bring up their children, because of the large externalities, positive or negative, that those children will create for their fellow citizens as they grow up. It's therefore quite reasonable for society to insist that bringing up children, which is an area where private decisions have wide social repercussions, is a paternal responsibility as well as a maternal one, and the provision of paternal leave is one way to make that point clear. Although public policy cannot ensure that fathers use their paternity leave wisely or productively, it can establish a convention that parenthood involves career interruptions regardless of the talent and motivation of the person concerned. Fathers who believe that their paternal obligations can be fulfilled without any diversion whatever of time from their professional to their domestic responsibilities have misunderstood the nature of what their fellow citizens require from them.

Although the justification of such a policy rests on the fact that paternal responsibilities have repercussions for the rest of society, there may be an additional benefit for gender representation, one that arises from the impact on the signaling trap. Once firms wake up to the fact that most of their potential CEOs will be people who have taken parental career breaks, they're likely to stop considering the decision to do so as revealing qualities that are undesirable in a CEO. It won't stop some parents from taking longer career breaks than others, and it probably won't equalize the gender distribution of those career breaks any time soon. It is not easy to find evidence about the likely impact of such policies, because in the few countries that have tried anything like them, the take-up of paternity leave has been largely voluntary. This makes it hard to compare the behavior of those who take it up with those who do not, since many other factors (including the motivation of the individuals concerned) are likely to be varying at the same time. However, one study investigating this question was conducted in Norway (where paternal leave was made highly advantageous in a reform in 1993, so that take-up of leave by eligible fathers increased from 4 percent before the reform to over 60 percent within a couple of years). Interestingly, this study suggests that compulsory paternity leave may increase the tendency of women to take time off work to rear children, perhaps because it becomes more pleasant to do so once the task is shared.²⁴ The idea that paternal career breaks might simply substitute for maternal breaks is not borne out by this experiment. If anything, the message is that both men and women can put a higher weight on their domestic responsibilities when they can do so at a lower cost to their careers. Nevertheless, such a measure may help, through changing the signaling equilibrium in employment, to ensure that those who take career breaks pay a price that more nearly reflects the balance of costs and benefits to society as a whole.

As both of these examples probably suggest, the role of formal legislation in redressing the gender imbalances that I've highlighted is only a part, and probably a very minor part, of the solution. Far more important will be changes in norms and conventions among men and women alike, as we come to realize that methods of organizing work that penalize women for their career choices are not just inequitable—they're also profoundly stupid. The evidence we've looked at in the last few chapters suggests it is likely (not certain, but likely) that there are talented women everywhere who are being paid less and given fewer opportunities than equally talented men. If so, it's time for everyone, women and men together, to wake up to what an opportunity these women represent for their potential employers and colleagues.

Many working practices, such as spending long hours in the office doing tasks that could just as efficiently be performed at home, are a form of wasteful display. They are a peacock's tail of individually rational but collectively pointless effort that happens to penalize women, on average, more than men in terms of professional advancement but penalizes men more than women, on average, in terms of their domestic and family fulfillment. In fact, these costs to both men and women are probably a more important reason for taking action than the remaining inequalities in themselves. What really matters is not that some already highly paid women are paid less than their male counterparts but rather that both men and women are failing to find the fulfillment that a modern productive economy ought to make possible for them.

Escaping the logic of the peacock's tail will not be easy: a peahen that chose to ignore peacocks with large tails would not increase her fitness by doing so. But that's because she would have no other way of discriminating the fitter males from the less fit. If she could find such a better basis for diagnosing male fitness, ignoring those large tails would be a very smart move indeed. If Homo sapiens is to wring any collective benefit from the fact that we have larger brains than peacocks, it must surely include the possibility of devoting our ingenuity to escaping such signaling traps by devising less wasteful ways to reveal our talents and motivations to each other. Technology is transforming the physical constraints of our working lives; it does not seem unrealistic to hope that it can reorganize some of the social constraints as well.

The revolution in contraceptive technology, and the revolution in society that has accompanied it, have made it possible for the first time in the history of our species to separate the sexual collaboration between men and women from their more general economic relations. That will not mean, and nor should it, that sex is banished from the workplace. But sexual relations and economic relations can achieve more when each is liberated from the shadow cast by the other. Just as sex freed from economic dependence is usually better sex, economics freed from dependence on sex is likely to be better economics too.

^* The longer time keeps company / With us, the more it tortures us. / The very worst of lovers' traps / Would be to try to live in peace. / Of course, you cry less easily / These days, of course I crack less soon. / Less jealous of our mysteries, / We leave less in the hands of chance. / The current makes us careful now, / But still the tender war goes on.