Testosterone Rex

by Cordelia Fine

  One explanation might be that these represent cases of desperate times calling for desperate measures. Yet Wood and Eagly conclude that the “evidence that men and women sometimes engage in gender-atypical activities suggests a flexible psychology that is not vividly differentiated by sex.”71 This brings us to an evolving controversy: just how different—or similar—are women and men? A Live Science headline—“Men’s and Women’s Personalities: Worlds Apart or Not So Different?”—neatly captures the range of possible views. And of course the reason we argue about this more than, say, whether dogs are from Saturn and cats are from Neptune, is that it seems fundamental to the question of how we should feel about the status quo.

  What is relatively uncontroversial (although the memo has yet to make it to a number of popular commentators) is that the majority of sex differences in the basic building blocks of behaviour—cognition, communication, social and personality traits, and psychological well-being—are relatively small. University of Wisconsin–Madison psychologist Janet Hyde drew attention to this important point in a now classic paper proposing the “gender similarities hypothesis.”72 This was based on a synthesis of forty-six meta-analyses of sex differences in those fundamental building blocks. A meta-analysis is a statistical compilation of published and unpublished studies looking at a particular research question that, by pooling data, yields a more reliable estimate of what’s going on. From this, researchers calculate a useful statistic known as the “effect size.” It not only tells whether there is a difference between two groups but how much of a difference, if one exists. The bigger the effect size, the bigger the difference. A sex difference, after all, could mean anything from “almost all women scored higher than all men” (the situation when there is an effect size of about 3) to “there’s a 56 per cent chance that a woman picked at random will score higher than a randomly selected man.” This far less impressive difference, reflecting much greater overlap in the scores of women and men, would reflect an effect size of about 0.2.

  What Hyde found is that more than three-quarters of the sex differences that emerged from these meta-analyses were either very small (0.1 or less) or small (0.35 or less), meaning that about 40 per cent of the time, at least, if you chose a woman and a man at random, the woman’s score would be more “masculine” than the man’s, or vice versa.73 (If there were no average sex differences, this would happen 50 per cent of the time.) These included skills like mathematical problem solving, reading comprehension, and characteristics like negotiator competitiveness and interpersonal leadership style. A recent ten-year follow-up of Hyde’s landmark paper that synthesised 106 meta-analyses of sex differences confirmed the gender similarities hypothesis no less emphatically.74

  A more recent turn has been to look not just at the sex difference in a single variable but the pattern over sets of variables. In one recent study, Bobbi Carothers and University of Rochester psychologist Harry Reis found that people often score in a stereotype-consistent direction on some variables, but in the opposite direction on other, related ones. In other words, they can’t be tidily sorted into “masculine” and “feminine” categories but are instead spread across a continuum.75 As the researchers put it, “Although there are average differences between men and women, these differences do not support the idea that ‘men are like this, women are like that.’” Instead:

  These sex differences are better understood as individual differences that vary in magnitude from one attribute to another rather than as a suite of common differences that follow from a person’s sex.76

  Another evolving aspect of this debate is that old arguments that sex inequalities are explained by women’s intellectual inferiority have shifted towards claims instead that these inequalities are due to sex differences in values and interests. It’s not that a woman can’t behave like a man; it’s just not in her nature to want to. Yet contrary to the Testosterone Rex perspective, sex differences in both “masculine” values (social status, prestige, control and dominance of people and resources, and personal success) and the “feminine” value of caring for loved ones are also small.77 Nor are such priorities set in stone. The Pew Research Center in the United States, for instance, recently reported that young women have now overtaken men in the importance they place on success in a high-paying career, and the sexes are equally likely to count being a good parent and having a successful marriage as more important than lucrative workplace success.78

  But perhaps, you may be thinking, women’s equal work ambition is all very well and good, but only men tend to have the requisite ruthlessness to get ahead. A well-worn story is that, since evolutionary dynamics dictate that the nasty guy dominates and therefore gets the girl, men tend to be inherently more aggressive than women. In fact, there are some serious question marks hanging over this chain of assumptions.79 But even setting these aside, the argument doesn’t really work. The largest sex difference in aggression is, unsurprisingly, in the physical variety. (You’d have about a two in three chance of correctly guessing whether someone was male or female, based on whether they were below or above average in physical aggressiveness.)80 But here are two facts about your own occupation that, with a few exceptions, are almost bound to be true. First, men predominate at the most senior or prestigious levels. Second, they didn’t get there by virtue of their greater propensity to punch people in the nose. That’s not to say that everyone plays nice at work. But meta-analytic findings for sex differences in verbal aggression fall into the very small to moderate range. (And in some places, like the Gapun village in Papua New Guinea, the typical sex difference is reversed, with women renowned for their colourfully aggressive tirades towards those who have displeased them.)81 As for indirect aggression—the aim of which is “to socially exclude, or harm the social status of, a victim”82 without getting blood on one’s suit—if anything, the scale is tipped towards greater female aggression.83 In short, sex differences in aggression don’t get us very far in explaining the occupational status quo.

  True, sex differences in occupation-related interests are larger. (I took a close look at the evidence supposedly showing the “hard-wired” basis of this in my previous book, Delusions of Gender.) According to compiled findings from one much-used inventory, more than 80 per cent of men report greater interest in “things” than the average woman, who has a greater fascination with “people”-inclined activities,84 and this seems to be reflected in the kinds of occupations into which women have made the least inroads over the past three decades.85 However, it’s worth noting psychologist Virginia Valian’s observation that simply labelling a dimension “things” or “people” doesn’t make it so. For example, the three subscales of the inventory that make up the “thing” dimension require “thing” to be interpreted so broadly—including “the global economy, string theory, mental representations, or tennis”—that the term becomes “vacuous.”86 Valian also suggests that preconceptions about which sex does stuff with things have influenced the creation of the items. Why, for instance, don’t activities like “Take apart and try to reassemble a dress” or “Try to recreate a dish tasted in a restaurant” appear on such scales?87 But also, as Valian observes, the sexes are artificially divided when they are categorized as either “thing people” or “people people.” In fact, being interested in things doesn’t stop you from being interested in people, and vice versa. Many men and women are, of course, interested in both and would be pretty awful at their jobs if they weren’t. For instance, I wouldn’t care to have blood taken by a nurse, however sympathetic, who was completely uninterested in the mechanics of the syringe. Nor would I want to hand over the renovation of my house to a builder who had no interest in understanding or managing the delicate psychology of the tradesperson.88

  One counter-response to claims that a person’s sex tends not to be a very good guide to whether they will be “masculine” or “feminine” on a particular trait is that these usually modest differences nonetheless add up to something rather substantial. Th
e neurobiologist Larry Cahill, as we saw in the Introduction, suggests that the argument that the sexes are similar because most differences are small is “rather like concluding, upon careful examination of the glass, tires, pistons, brakes, and so forth, that there are few meaningful differences between a Volvo and a Corvette.”89

  However, there’s a problem with this line of reasoning. For many decades, researchers supposed that masculinity and femininity are polar ends of a single dimension: someone high in masculinity is therefore necessarily low in femininity, and vice versa. In fact, this assumption was built into the very design of the first systematic attempt to measure masculinity and femininity: a brisk 456-item questionnaire with the carefully obscure title, The Attitude Interest Analysis Survey.90 The survey yielded a single score that placed each individual at a particular point on a single masculinity-femininity line. So if, for example, you felt that the word “tender” went most naturally with the word “loving” or “kind” then you lost a point (naturally!) for being feminine. By contrast, if your mind leapt unsentimentally from “tender” to “meat,” then you may have had trouble getting second dates, but you did at least gain a point for being masculine.

  It wasn’t until the 1970s that this assumption was overturned by the development of two new scales.91 Still in use today, these separately assess “masculine” traits of “instrumentality” (qualities like self-confidence, independence, and competitiveness) and “feminine” traits of “expressiveness” (such as being emotional, gentle, and warm and caring towards others). This revealed that it is possible to be both instrumental and expressive, or neither. To put it in terms of Cahill’s car metaphor, one can have the safety, reliability, and room-inthe-trunk-for-the-weekly-groceries-ness of the Volvo and the power, status, and thrill of the Corvette. Or—and here I stress to any Volvo and Corvette owners that the comments that follow are made solely for pedagogical purposes—one can have the sluggishness of a Volvo and the expense of a Corvette. But even this two-dimensional model of gender is now known to be too simple. Correlations among masculine traits and among feminine ones are often weak or nonexistent. Having one masculine trait doesn’t imply you have another, and likewise for feminine traits.92

  In other words, differences between males and females may not “add up” in a consistent way to create two kinds of human nature; but rather, as with sex differences in the brain, create “mosaics” of personality traits, attitudes, interests, and behaviours, some more common in males than in females, others more common in females than in males. Joel and colleagues tested this idea, drawing on three large data sets, and using the same approach as they did for brains. Even looking at only twenty-five behaviours with at least moderate sex differences (this included attributes like communication with mother, being worried about one’s weight, and delinquency, as well as strongly sex-stereotyped activities like playing golf and using cosmetics), between 55 and 70 per cent of people (depending on the sample) had a mosaic of gender characteristics, compared with less than 1 per cent who had only “masculine” or only “feminine” characteristics.

  This makes the notion of female natures and male natures as unintelligible as that of female brains and male brains. Which of the many combinations of characteristics that males display should be considered male nature? Is it a profile of pure masculinity that appears to barely exist in reality? What does it mean to say that “boys will be boys,” or to ask why a woman can’t be more like a man? Which boy? Which woman, and which man?

  These findings and patterns are awkward for those who want to argue that the sexes “naturally” segregate into different occupations and roles because of their different natures, or because of a slight advantage of one sex over the other, on average, on a particular trait. Job performance, paid or unpaid, depends on a suite of different skills, traits, interests, and values. People simply don’t develop a career doing one thing really well, like identifying facial expressions of emotion, being sympathetic, or banging a fist on a board-room table. What’s more, for most jobs, there isn’t one, single ideal combination of characteristics, skills, and motivations, but a range that could all fit the bill equally nicely. That’s why not everyone at your level, in your role, in your occupation, is just like you. So if you want to, say, trot out the argument that women are just more psychologically suited to taking care of small children, you’re committing yourself to the claim that women’s hugely variable gender mosaics far more often match the many possible mosaics for caring for young children than do men’s hugely variable gender mosaics. I don’t say this kind of argument can’t be successfully made. But I’d ask to see how you worked it out.

  “THERE IS NO DOUBT that biology, via evolution and genetics, has made men and women significantly different.”93 So concludes Wolpert’s book, in answer to the question posed by its UK and U.S. titles, Why Can’t a Woman Be More Like a Man? and Why Can’t a Man Be More Like a Woman? But as Valian astutely notes in Nature, “both titles suggest the retort: each can be.”94 And as we’ve also now seen, although there are certainly average sex differences, the very phrases “like a woman” and “like a man” make little sense at the level of brain and behaviour.

  This isn’t mere semantics or academic nitpicking. When young children and adults are asked to explain statements like “Boys have something called ‘fibrinogen’ in their blood,” or “Boys are really good at a dance called ‘quibbing,’” the kinds of explanations they come up with are different from the ones they create for statements like “This boy has fibrinogen in his blood,” or “This boy is really good at quibbing.” The generic statements more often trigger explanations grounded in assumptions that having fibrinogen in the blood, or being good at quibbing, or whatever it is, is fundamental—part of the true nature of being a male or a female. “I think it’s a hormone that boys have because it is transcribed from male DNA,” was one undergraduate’s explanation of fibrinogen in the blood. “Boys are generally stronger than girls, and quibbing sounds like it requires some strength.” By contrast, non-generic statements brought about relatively greater numbers of “non-essentialized” explanations that saw these characteristics as a one-off, even a problem, or due to external causes like practice or training. “A style of dance that the boy has practiced and is good at.”95

  When mosaics of mostly small average differences are carelessly squished into uni-dimensional generic claims—men are like this, women are more that—the natural inference is that we are talking about universal characteristics that are “central, deep, stable, inherent—in a word, ‘essential.’”96 When we say, think, or write statements like “Males are higher in competitiveness, dominance-seeking, and risk-taking, while females are higher in nurturance,”97 it’s tempting for the mind to turn to the almighty T and the omnipotent Y of Casey’s review—to sex—as a principal, powerful cause that sets us all on one of two divergent paths. But the overlapping, shifting, multidimensional, idiosyncratic mosaics formed by patterns of sex differences instead point to the combined and continuous action of many small causal influences.98 Sex doesn’t create male natures and female natures, and the next chapter turns to risk taking and competitiveness to complete the case.

  CHAPTER 5

  SKYDIVING WALLFLOWERS

  MY ELDEST SON HAS LONG BEEN IRRESISTIBLY DRAWN TO danger. At six months old he rolled across the entire expanse of the living room in order to more closely inspect the drill that his father—forgivably assuming that five yards was a safe distance to place a power tool from a baby who couldn’t yet crawl—had put on the floor. On one memorable toddler playdate, within five minutes he had located the drawer of sharp kitchen knives that his little host Harry had failed to discover in his two years of life, and begun juggling with its contents. As a preschooler, whenever I took him to an indoor play centre—those brightly coloured monuments to the eradication of risk from childhood—my son would nonetheless routinely manage to imperil himself. At the age of ten, I left him happily engaged in the normally hazardless acti
vity of assembling a cake batter, only to return five minutes later to discover him about to plunge a roaring hairdryer into the mixture. As he calmly explained, he had forgotten to melt the butter before adding it to the bowl, and was therefore trying to do so retroactively.

  I admit that at times like these I have occasionally wondered why it was my lot in life to have a child so blasé about risk, and whether ultimately this will prove to be a blessing or a curse. On optimistic days I imagine him reaping enormous benefits: the invention of a time machine, say, after decades of dangerous experimentation. But in darker moments, I foresee much bleaker fates featuring mortuary lockers. While proponents of the Testosterone Rex perspective obviously don’t share this fascination with my firstborn and his future, they do, as we saw in the Introduction, have a strong interest in the idea of risk taking as an inherently male trait. They would regard each of my son’s perilous follies as successful manifestations of evolutionary pressures: a pitiful consolation, I can assure you, when you are trimming singed hair from your child’s bangs and hoping the other guests at the barbecue won’t ask too many questions. Economists Moshe Hoffman and Erez Yoeli recently spooled out the familiar chain of assumptions in the Rady Business Journal:

  When males take on extra risk in foraging for food, ousting rivals, and fighting over territory, they are rewarded with dozens, even hundreds of mates, and many, many babies. A worthwhile gamble! Not so for the females.1