Anne Campbell
The past 25 years have been revolutionary for our understanding of the psychology of sex differences in aggression. Prior to this, the social science orthodoxy was that these differences emerged as a result of early socialization by parents and later conformity to society's gendered division of labor. The fact that these sex differences are early-appearing, universal, and similar to those seen in other species was dismissed (Tieger, 1980). With the publication of Daly and Wilson's book Homicide (1988), social scientists were introduced to an evolutionary viewpoint: Across sexually reproducing species, the greater parental investment made by females leads to a male-biased operational sex ratio and heightened competition between males for access to fertile females. Successful males copulate with more partners and leave more sons who carry their fathers' aggressive genetic legacy. This formed the initial framework for an impressive body of supporting research on male aggression but some issues remain to be resolved. Are females exempt from intrasexual competition? If so, is their role merely to act as quality controllers of male genes? When and how did monogamy evolve (Henrich, Boyd, & Richerson, 2012)? What consequences did it have in terms of creating two-way sexual selection and competition (Clutton-Brock & Huchard, 2013)?
These questions inform this chapter. So dazzling were male courtship displays and so ferocious was male aggression that evolutionary biologists have only recently turned their attention to female forms of competition (Stockley & Campbell, 2013). Evolutionary psychologists have a special contribution to make in identifying sex-linked psychological mechanisms that mediate behavioral differences between the sexes. Some of these we share with our phylogenetic cousins. Others may be unique to humans with their capacity for representational thought, reflective control over lower-level reflexive responses, and cultural transmission.
For aggression to evolve as a strategy, the rewards must exceed the costs (both measured in terms of reproductive success). For many years it was assumed that because females did not need to fight for copulations, there was nothing of consequence for them to fight about and therefore, there were no rewards for female aggression. But copulation is only the start of reproductive success—offspring must be nourished and protected until they are able to reproduce themselves. Pregnant women need an extra 300 calories every day, and 500 more when they are lactating (Institute of Medicine, 1990). Once on solid food, a toddler needs to consume 1,300 calories a day. Like most primates, humans are a group-living species, which means that food can be a contested resource. In addition, mothers must supervise and protect their offspring from ecological dangers, infanticidal males, and harassment by other females. Now the rewards for female competition become more apparent: Provisioning and protecting offspring are tasks that are easier for a dominant female than a subordinate one. Status elicits deference and compliance from those of lower rank. Dominant female primates produce offspring more rapidly than subordinates and their offspring have higher survival rates (Pusey, Williams, & Goodall, 1997). This makes it all the more puzzling that competition for dominance and status among females is so much less evident than it is among males. Dominance hierarchies are chiefly restricted to female-bonded species (Sterck, Watts, & van Schaik, 1997) in which they are organized around matrilines and rank is inherited rather than fought for. Females rarely risk their lives to achieve dominance (Chapais, 2002). Dominance offers substantial rewards, yet competitive striving for dominance was not strongly selected in females when it entailed direct combat. This suggests that the rewards were offset by higher costs.
One cost that constrains the evolution of female aggressiveness is hormonal. Testosterone is associated with male aggression, but it is even more closely associated with competition for dominance (Johnson, Leedom, & Muhtadie, 2012). In many species, male testosterone levels rise during the breeding season and in response to challenge (Archer, 2006). Hyenas are atypical mammals in which levels of female dominance and aggression match or even exceed those of males. During pregnancy, maternal androgen levels are raised, creating a uterine environment that masculinizes female fetuses. Gestational androgen levels are higher in dominant females, and the fetus's androgen exposure is correlated with greater aggression in later life (French, Mustoe, Cavanaugh, & Birnie, 2013). However, this exposure virilizes the female genitalia, resulting in a 7-inch clitoris with a diameter of 1 inch through which the adult female must deliver a 2-pound cub. These maladaptive side effects likely limit the extent to which female dominance and aggression can evolve by hormonal means (Clutton-Brock, 2007). Testosterone carries other costs in terms of compromising the immune system (Schroderus et al., 2010), which may partially explain males' earlier senescence and death relative to females. In young men, through its effects on aggressiveness and risk taking, testosterone also increases deaths from external causes. This cost is particularly relevant for females because, as I argue below, reproductive success depends on mothers' continued survival.
The prospect of death (or severe injury) is not appealing to either sex. But the sexes differ in the impact of mortality on reproductive success measured in terms of surviving offspring. For a male, death removes the possibility of future matings, but is less likely to compromise the reproductive success he has achieved to date because he can rely on the offsprings' mother to ensure their survival. A father's death may be a tragedy for the child, but “the consequences of losing a mother very early in life are catastrophic” (Sear & Mace, 2008, p. 5). Sear and Mace (2008) examined the impact of parental death on offspring survival in populations ranging from 18th-century China to 20th-century Nepal, from Burkina Faso to New York state. In every case, a mother's survival reduced the likelihood of her children dying. The percentage of children surviving a mother's death ranged from 2% to 50%. The beneficial effects are stronger before children are weaned: In rural Gambia, a mother's death multiplied the odds of her child's death by 6.2 in infancy, 5.2 in toddlerhood, and 1.4 in childhood (Sear, Steele, McGregor, & Mace, 2002). Pavard, Gagnon, Desjardins, and Heyer (2005) examined data from 17th-century Quebec, controlling for a range of variables including the possibility of transmitted infection and shared genetic vulnerability. If a mother died while her infant was still a neonate, the odds of the child dying in the neonatal period were multiplied by 5.52, dropping to 1.27 when the child was aged 5 to 15 years. The effect of a mother's death was consistently more serious than that of any other relative. Their importance to offspring survival increases the selection pressure on females to safeguard their own lives (Campbell, 2013). Sex differences in aggression are the result not only of selection pressures on males to compete for dominance, but selection pressures on women to avoid dangerous competition.
The results of many hundreds of studies of sex differences in aggression using a range of techniques, including laboratory experiments, observation, personality assessment, and self- and peer-reported behavior, fit a clear pattern: The more dangerous and risky the form of aggression, the larger the sex difference. This appears to be true cross-culturally (Archer, 2009). For physical acts such as hitting, punching, and kicking, the effect size lies between d = 0.59 and d = 0.91, while for verbal acts such as abuse and threats, the effect size is between d = 0.28 and d = 0.46 (Knight, Fabes, & Higgins, 1996; Knight, Guthrie, Page, & Fabes, 2002). Indirect aggression includes acts such as spreading stories, excluding, and stigmatizing by which the aggressor can remain anonymous and the possibility of retaliation is consequently reduced. Here the sex difference is negligible and nonsignificant, d = −0.02 (Archer, 2004).
The sex difference in risky aggression might be mediated by men's greater attraction to reward, reflecting a stronger approach motivation and manifested at the proximate level in the emotion of anger, or it might be the result of women's greater sensitivity to costs, reflecting a stronger avoidance motivation and manifested in greater female fear. A sex difference in the threshold for experiencing one or both of these emotions might therefore represent a proximate psychological mechanism for the sex difference in aggression.
Anger is a universal emotion, recognized in all cultures and visible early in life. Anger signals goal obstruction: The goal may be respect, resources, or survival. It is only recently that lateralization studies have established unequivocally that anger triggers approach behavior. As evolutionary psychologists would expect, lateralization of emotion is better understood in terms of functionality (approach or avoidance) than valence (pleasant or unpleasant). Results clearly show that anger is left-lateralized, as are other approach motivations (for review, see Carver & Harmon-Jones, 2009). For example, Harmon-Jones and Sigelman (2001) provoked participants by gratuitously insulting them and then allowed them to respond aggressively. Insulted participants showed greater left frontal activity than the control group and the strength of left-hemisphere activation positively correlated with both their reported anger and level of aggression. Other studies have used transcranial magnetic stimulation paradigms to reach the same conclusions (d'Alfonso, van Honk, Hermans, Postma, & de Haan, 2000). Beyond the laboratory, anger correlates positively with psychometric scales measuring approach motivation and negatively with avoidance motivation (e.g., Smits & Kuppens, 2005).
However, there is little evidence that men exceed women on anger. Archer (2004) performed a meta-analysis of 46 studies of psychometric anger inventories and found no sex difference. National and international surveys find either no sex differences in anger frequency, with both sexes reporting anger about six times a week (Brebner, 2003; Fischer, Mosquera, van Vianen, & Manstead, 2004), or that women experience anger more often (Mirowsky & Ross, 1995). The intensity of experimentally evoked anger does not vary between the sexes (Kring, 2000), nor is men's anger greater in response to hypothetical or remembered provocation (Milovchevich, Howells, Drew, & Day, 2001). In short, data do not support lower anger in women as an adequate explanation of sex differences in aggression. Nor is the proposal satisfactory at a conceptual level: A higher threshold for anger in women might protect them from aggressive confrontations but not from other forms of risky behavior. Yet there is ample evidence that women are more risk averse than men.
By contrast, there is much evidence that the sexes differ in the frequency and intensity of fear (Else-Quest, Hyde, Goldsmith, & Van Hulle, 2006). The fear system is “designed to detect danger and produce responses that maximize the probability of surviving” (LeDoux, 1998, p. 128). Extreme fear triggers freezing: Inhibitory neural connections allow anticipatory activity in the fear system to suppress behavioral approach that might lead the organism into a harmful situation (Derryberry & Rothbart, 1997). Developmentally, girls express fear earlier than boys, and in a large longitudinal study of personality development, more girls than boys were on a high fearfulness trajectory (Cote, Tremblay, Nagin, Zoccolillo, & Vitaro, 2002). Among adults, women experience fear more intensely than men (Gullone, 2000). International studies find significant sex differences in the frequency, intensity, and duration of fear (Brebner, 2003; Fischer & Manstead, 2000). Women express their fear more intensely than men, both verbally and nonverbally (see Madden, Feldman Barrett, & Pietromonaco, 2000). While women are superior to men in accurately identifying emotions, they show a greater accuracy for decoding fear than other emotions (Hall, Carter, & Horgan, 2000). In response to physically threatening scenes, women show greater increases in skin conductance and a more marked startle reflex (McManis, Bradley, Berg, Cuthbert, & Lang, 2001). A single dose of testosterone administered to women significantly reduces their potentiated startle response to anticipated electric shock (Hermans, Putman, Baas, Koppeschaar, & van Honk, 2006). These findings extend rodent research that has robustly established that the hypothalamic pituitary axis response to stress is inhibited by androgens and enhanced by estrogens (Lund, Munson, Haldy, & Handa, 2004). Taken together, these findings suggest that gonadal hormones cause the fear system to develop and function differently in males and females.
This sex difference in fear may explain why men make riskier decisions than women. This is especially marked when the risks are life threatening and when actual risky behaviors, rather than hypothetical choices, are examined. In a review of risk-taking studies, Byrnes, Miller, and Schafer (1999, p. 378) conclude that “fear responses may explain gender differences in risk taking more adequately than the cognitive processes involved in the reflective evaluation of options.” Women's lower participation in risky real-world activities is best explained by their stronger anticipation of possible negative consequences and by their higher ratings of the severity of those consequences should they occur (Harris, Jenkins, & Glaser, 2006; Wang, Kruger, & Wilke, 2009). Women's risk taking, but not men's, is especially reduced when their risky decisions have consequences for infants (Fischera & Hills, 2012). A meta-analysis (Cross, Copping, & Campbell, 2011) revealed that although women and men do not differ in their sensitivity to reward, women are consistently more sensitive to punishment than men. This dovetails with the proposal that women have evolved greater sensitivity to negative outcomes than men, manifested in their lower threshold for fear. This lower threshold has direct consequences for aggressive behavior. Two independent meta-analyses have found that women evaluate the same objective situation as more dangerous and more fear-provoking than men, and these appraisals significantly explain the sex difference in aggressive behavior (Bettencourt & Miller, 1996; Eagly & Steffen, 1986). Fear acts as a brake on aggression, and women's brakes respond to danger more sensitively than men's.
Increasingly, we are able to access internal affective states through the use of functional magnetic resonance imaging. The chief focus of such studies has been the amygdala, an almond-shaped subcortical structure (composed of more than 10 nuclei) in the temporal lobe. LeDoux's (1998) pioneering work implicated the amygdala in the registration of fear, although it is now thought to process other emotions associated with salient or unexpected events (Sergerie, Chochol, & Armony, 2008). Its role seems to be to rapidly detect stimuli that are biologically relevant (Sander, Grafman, & Zalla, 2003), especially where they may require an immediate response, which is often the case when they are unexpected, threatening, or dangerous (Adolphs & Spezio, 2006). Afferent sensory inputs to the lateral nucleus of the amygdala are coordinated with efferent outputs from the central nucleus that control behavioral, autonomic, and endocrine fear responses.
We would expect to see a stronger amygdala response to threat in women reflecting their greater fearfulness. Meta-analyses generally conclude that women show greater activation to threat in the limbic system, especially the amygdala (Stevens & Hamann, 2012; Whittle, Yucel, Yap, & Allen, 2011; but see also Sergerie et al., 2008). This suggests that women may register external threat more strongly and more persistently than men. However, because the majority of neuroimaging studies use participants of only one sex, meta-analytic conclusions are based on comparisons of men's and women's neural responses to different stimuli.
In many studies, researchers use facial expressions of fear and anger as “threat” signals, collapsing these stimuli in their analyses. This makes it hard to tease apart the neural and emotional response to being personally threatened (viewing an angry face) as compared to being alerted to environmental hazard (viewing a fearful face). We might expect that these two social messages would activate somewhat separate circuits, and that the sexes might differ in their response to them. Both men and women would be expected to show a similar response to a fearful face (indicating nonspecific local danger). However, an angry face (indicating a possible aggressive attack) might trigger hostility in men and fear in women. McClure et al. (2004) compared men's and women's reactivity to angry and fearful faces. The relative engagement of the amygdala bilaterally to angry faces was greater in women, suggesting that women react more fearfully than men to unambiguously threatening (angry) faces. Relative to baseline fixation, women showed significantly greater activation than men over the whole “fear circuit” (amygdala, orbitofrontal, and anterior cingulate cortex) to angry but not fearful faces. By contrast, men showed a less specific pattern of increased orbitofrontal (but not amygdala) activation to both stimuli. There is, then, some support for the proposal that amygdala activation may be more closely associated with fear in response to aggressive threat in women, while men show a less differentiated pattern of reactivity to “threatening” stimuli in general.
These problems of interpretation (the amygdala can respond to facial expressions of both fear and anger, and amygdala activation may correspond to the registration of both these emotions) become particularly evident in research that examines the effect of gonadal hormones. As with other regions that are sexually dimorphic in size, the amygdala contains a high concentration of sex hormone receptors. On one hand, it decreases fear, and one study reported that, in men only, endogenous testosterone levels were negatively correlated with amygdala reactivity to angry faces (Stanton, Wirth, Waugh, & Schultheiss, 2009). On the other hand, testosterone levels have been linked to increased amygdala activation in response to threat, with this activation interpreted as reflecting anger and approach motivation. In both sexes, endogenous testosterone levels are positively associated with amygdala response to threatening stimuli (van Wingen, Ossewaarde, Backstrom, Hermans, & Fernandez, 2011). Administration of testosterone to young women is associated with increased persistence of amygdala reactivity to angry faces (Hermans, Ramsey, & van Honk, 2008). In one study that attempted to disambiguate men's neural response to angry and fearful faces (Derntl et al., 2009), amygdala reactivity did not differ significantly between the two, and their endogenous testosterone levels were equally correlated with their amygdala responses to both types of stimuli.
Underpinning much of this research is the questionable assumption that testosterone has similar effects in men and women. In men, but not women, circulating testosterone interacts with a brain that has been prenatally primed by androgens. In addition, testosterone effects may differ between the sexes as a result of sexually dimorphic gene expression. Testosterone may trigger the expression of autosomal genes in one sex but not the other, or different genes in the two sexes. In the dark-eyed junco, 651 genes in the medial amygdala differed in expression between males and females, and testosterone administration altered the expression of different genes in the two sexes (Peterson et al., 2013). Many genes are subject to correlated expression and may be coregulated, and this functional modularity may allow suites of responses to differ in men and women in response to the same hormonal milieu (Rosvall, 2013b). (A recent vivid example of the dangers of assuming common or sex-neutral effects comes from studies of the peptide hormone oxytocin. S. E. Taylor and colleagues, 2000, proposed that the calming effect of oxytocin was responsible for women's “tend-and-befriend” response to threat. This was supported by studies showing that administration of oxytocin did indeed have anxiolytic effects [e.g., Kirsch et al., 2005]. However, these studies employed male participants. When female participants were examined, oxytocin increased rather than decreased amygdala responsiveness to threat [Domes et al., 2010; Lischke et al., 2012]). Further studies are needed to examine the circuitry of fear and anger, identifying common and unique pathways. This will allow examination of the role of these emotions in explaining sex difference in aggression, as well as the effects of gonadal hormones on their relative activation.
These fundamental affective tendencies to approach or avoid stimuli are located in the limbic system and are part of a “reflexive” behavioral control system sculpted in many animal species by evolutionary forces. But in humans, these tendencies are subject to higher-level “reflective” control. Emotional intensity and behavioral response can be modulated by the prefrontal cortex, especially the orbitofrontal (OFC) region, which has direct connections to the amygdala. In neuroimaging studies, negative correlations are found between amygdala and OFC activity in impulsively aggressive individuals (Coccaro, McCloskey, Fitzgerald, & Phan, 2007). When participants are instructed to imagine aggressing against (Pietrini, Guazzelli, Basso, Jaffe, & Grafman, 2000) or harming another person (Decety & Porges, 2011), deactivation of the OFC has been found. Given the modulatory role of the prefrontal cortex, studies have looked for sex differences in these regions. Women have a larger ventromedial prefrontal cortex and right lateral OFC (Welbourne et al., 2009). A meta-analysis of 88 studies reported greater OFC activity in women to facial stimuli depicting negative emotion (Stevens & Hamann, 2012). This suggests that women may be more efficient in spontaneously regulating emotional responses.
Testosterone reduces functional connectivity between the amygdala and the prefrontal cortex, while leaving connectivity to the brain stem unaffected (Manuck et al., 2010; Bos, Hermans, Ramsey, & Van Honk, 2012). (Progesterone has the opposite effect, enhancing amygdala-frontal connectivity; van Wingen et al., 2008). Serotonin (5-HT) plays a key role in the functional connectivity between the PFC and the amygdala. There is a dense concentration of 5-HT receptors in the limbic system (including the amygdala) with projections to the prefrontal cortex. Dietary tryptophan depletion (which reduces 5-HT levels) reduces connectivity in the prefrontal-amygdala circuitry, specifically when viewing angry faces (Passamonti et al., 2012). Women have higher 5-HT transporter availability, and because this regulates 5-HT neurotransmission, baseline serotonin may be higher in women than men. Studies have reported a higher density of 5-HT1A receptors in women in areas including the amygdala and medial and orbital PFC (Parsey et al., 2002). Receptor density in these areas is significantly negatively correlated with lifetime aggression. In animal research, 5-HT receptor density is also negatively correlated with testosterone. Although this has not been replicated with humans, men (but not women) with high levels of aggression are characterized by a combination of high T and low 5-HT (Montoya, Terburg, Bos, & van Honk, 2012). Reduced availability or uptake of serotonin, associated with high T, may explain men's diminished prefrontal control over emotion-driven behavior.
Female competition and aggression, once considered a fascinating but inexplicable anomaly, is now a documented fact (Stockley & Campbell, 2013). But what are females competing for? The traditional model of sexual selection makes it clear that it cannot be copulations. Under polygyny, it is males not females who must compete for sexual access, hence their gaudy plumage (the better to advertise their genetic quality) and combative attitude (the better to deter and intimidate rival males). Yet in our own species, a cursory inspection reveals that women expend a considerable amount of their energy and resources on increasing their attractiveness to men via cosmetics, surgery, toxin injections, dieting, and clothing. This behavior bears the hallmark of two-way sexual selection. Like men, women are actively competing to obtain the best mates, which suggests that our species is less polygynous than has been assumed. But if polygyny is the optimal strategy for male reproductive success, why would men forego it in favor of committing themselves to a single woman and costly parental investment?
According to the traditional Bateman model, gestation and lactation remove women from the mating pool, creating a male-biased operational sex ratio. This increases competition between males for access to reproductively available females. But the logic of this has been questioned by Kokko and Jennions (2008), who argue that a male-heavy operational sex ratio should generate frequency-dependent selection favoring increased parental care by the sex that faces more intense competition. In short, there is no logical reason why male competition should generate a positive feedback loop over evolutionary time. As competition becomes more intense among males, there is selective advantage for those males who opt out of mating competition in favor of infant care.
Explanations of the evolution of paternal care predict that it should occur where the number of surviving infants with paternal care is greater than the number of surviving infants without such care multiplied by the number of females that a bachelor male can inseminate (e.g., Clutton-Brock, 1991). It has been suggested that paternal care is especially important in humans as a result of the infants' long period of dependency, which suggests that men might enhance their reproductive success more effectively by paternal investment and attendant monogamy than by polygyny. Yet there is evidence that paternal care does not improve offspring survival. Sear and Mace (2008) examined the impact of parental death in 28 hunter-gatherer and foraging populations. In 68% of cases, a father's death had no impact on the survival chances of his child. In 32% of cases, it actually improved the child's odds of surviving. Even among the Tsimane of Bolivia, who have low divorce rates and high levels of paternal provisioning, the early death of a father had no impact on their children's age of first reproduction, completed fertility, or number of surviving offspring (Winking, Gurven, Kaplan, & Stieglitz, 2009). Across five foraging populations, returns on paternal investment in terms of child survival were less than those obtained by serial mating (Winking & Gurven, 2011). In ecologies similar to the ones in which humans evolved, there are not strong grounds for believing that there was positive sexual selection on men for paternal care. (This is not to say that paternal investment in contemporary societies does not bring social, emotional, and financial benefits to children; see Geary, Chapter 20, this Handbook, Volume 1).
An alternative view of the evolution of biparental care focuses not on sexual selection but sexual conflict between men and women. This occurs when a sexual selection pressure acting to augment the spread of a gene-based trait advantageous to one sex is modulated, constrained, or opposed by the coevolution of a counterresponse by the other sex (Arnqvist & Rowe, 2005). Sexual conflict has been studied chiefly at the genetic level, but the basic premise—that males and females are locked in a coevolutionary arms race—can be extended to phenotypic behavior (itself underpinned by gene complexes). Bipedalism and the resulting need for “premature” delivery of babies increased the energetic demands on mothers. The presence of a male partner would have been advantageous, not because it increased offspring survival, but because it decreased the workload on mothers. A synthesis of data from 10 intensive studies of gatherer societies (Kaplan, Hill, Lancaster, & Hurtado, 2000) suggests that between 60% (among the Nukak) and 84% (among the Ache) of the calories consumed are contributed by men. More than half of calories consumed come from meat, hunted almost exclusively by men. With men relieving women of the full burden of provisioning, women were better able to feed their dependent children, sustain pregnancy and lactation, and return to normal cycling more quickly, thus shortening interbirth intervals (Worthman, Jenkins, Stallings, & Daina, 1993). A permanent male assistant was in a woman's best interests, and she had a bargaining chip. By restricting sexual access to men who were willing to make a paternal contribution, women could act as a selecting force countering men's predilection for promiscuity. Due to high male demand and restricted supply, sex has always been a resource that women can trade. Women can use it to obtain short-term and extra-pair mates of high genetic quality (Gangestad & Simpson, 2000; Thornhill & Gangestad, 2008). However, if too many women pursue this strategy, the marketplace will reflect it in a lower value for female sexual access, ultimately diminishing its utility as a bargaining chip. Indeed, female hostility to women who are too sexually available may function to protect the female sex from losing its leverage (Baumeister & Twenge, 2002; Baumeister & Vohs, 2004). Paternal investment offers the best long-term benefits to women by easing the resource costs of motherhood.
Did biparental care arise sufficiently long ago that there has been time for selection to act on it? Neonate size and adult skeletal remains date it to the start of the Homo line, 1.5 to 2 million years ago (Eastwick, 2009). Other archaeological findings place it even earlier, with Australopithecus afarensis, the predecessors of Homo (Reno, Meindl, McCollum, & Lovejoy, 2003). Selection operates faster on sexually, rather than naturally, selected traits: It is estimated that sexual selection can produce a 0.37 standard deviation shift in the average value of a fitness-relevant trait in a single generation of directional selection (Courtiol, Pettay, Jokela, Rotkirch, & Lummaa, 2012). Evidence of an evolved adaptation to paternal investment can be seen in the down-regulation of men's testosterone levels following pair bonding and fatherhood (Gray & Anderson, 2010), and the rise in oxytocin levels in fathers (as well as mothers) after the birth of a baby and during interaction with them (Gordon, Zagoory-Sharon, Leckman, & Feldman, 2010).
The effects of biparental care and pair bonding are profound. Monogamy reduces the ability of men to exploit women by creating greater equality between them in their reproductive output. Holland and Rice (1999) demonstrated this by forcing monogamy on the naturally polygynous fly Drosophila melanogaster. Individual males and females were housed together over 32 generations. Under polygyny, males can exploit females quite ruthlessly without suffering any costs themselves, but monogamy means that anything that hurts a female (prevents her from achieving her reproductive potential) hurts her male partner just as much. After several generations of monogamy, Holland and Rice examined the effects. When the control group of polygynous females were allowed to mate with the monogamous “new males,” the females benefitted from the decreased toxicity of the male's seminal fluid (a side effect of male sperm competition), which is normally harmful to them. Reciprocally, when the newly monogamous females were returned to the polygynous males, a larger proportion of them died compared to the females who had been allowed to coevolve with male polygyny. As a result of imposed monogamy, males behaved in a less exploitative way toward female partners and so the monogamous females did not need to evolve counterstrategies of resistance.
In many species, monogamy evolves where females are spatially dispersed and each male is forced to associate with an individual female. But this scenario is implausible for humans, who are a group-living species. An alternative model proposes that, if females were willing to forego extra-pair matings and preferred males who provided resources, male provisioning would increase, driven chiefly by low-ranking males who would otherwise fail to reproduce at all (Gavrilets, 2012). Because these males outnumber elite males, selection acts more strongly on them and their strategy. This is not to say that either sex becomes completely faithful, as both have something to gain by extra-pair copulation: Females gain “better” genes and males gain increased paternity. But a basically monogamous mating system, through its effects on individual behavior, brings with it societal rewards in terms of reduced crime rates and stronger within-group alliances. For women, it reduces gender inequality and domestic conflict, while increasing child survival (Henrich et al., 2012).
The cost for women is that sexual selection becomes two-way, increasing competition between women to secure the highest-quality males within the marketplace of assortative mating. The currency of female competition is the qualities that men value in a prospective mate. In many domains, men and women are more similar than different in what they seek in a long-term partner (Buss et al., 1990). There are no sex differences in minimum acceptable percentile for qualities such as intelligence, sexiness, exciting personality, creativity, friendliness, sense of humor, easygoing temperament, health, religiosity, desire for children, kindness, and understanding (Kenrick, Sadalla, Groth, & Trost, 1990). Because men and women are locked together under monogamy and biparental care, both sexes seek qualities that will contribute to the day-to-day cooperation and compromise that such an arrangement requires. But men more than women place a premium on youth (Buss, 1989; Kenrick & Keefe, 1992), physical attractiveness (Buss & Schmitt, 1993), and body shape (Singh, 1993). Female intersexual competition for mates often centers on advertising these qualities. While men are more likely to compete with each other by exaggerating their sporting ability, promiscuity, and popularity, women are more likely to compete with each other in terms of their appearance, using aids such as makeup, nail polish, fake tans, and tight clothing (Cashdan, 1998; Buss, 1988). Both parties are concerned about commitment, but men, who must live with the risk of misplaced paternity, are particularly sensitive to the possibility of sexual infidelity (Sagarin et al., 2012).
Appearance and fidelity can become key weapons when women's competition escalates to indirect or relational aggression. These are acts that stigmatize, ostracize, and otherwise exclude others from social interaction and they can be used without direct physical confrontation. Such acts do not eliminate or physically injure the target, but they do inflict stress and diminish the opponents' reputation and social support. A key component of indirect aggression is the use of gossip to undermine an opponent's reputation and decrease their social capital (Owens, Shute, & Slee, 2000). Pejorative comments about other girls' appearance rank high in girls' topics of gossip (Duncan, 1999) and are used to derogate rivals more often by women than men (Buss & Dedden, 1990). These circuitous attacks are directed particularly at attractive young women (Vaillancourt, 2013). But attractiveness combined with a self-confident flaunting of it seems particularly provocative. Girls who advertise their attractiveness or sexuality too overtly through dress, make-up, or demeanor are often targeted (Miller & Mullins, 2006). These girls offend on two fronts: They attract more than their fair share of boys and they communicate their felt superiority over other girls. This becomes a form of “disrespect,” which adds to the antagonism. Women can also benefit competitively by undermining their rivals' sexual reputation and terms such as “slag,” “tart,” or “whore” are powerful sources of reputation challenge among women (Campbell, 1995, 1999; Lees, 1993; Marsh & Paton, 1986). These tactics are as visible among university students (Buss & Dedden, 1990; Milhausen & Herold, 1999) as among girl gang members in deprived inner-city areas (Campbell, 1984; Ness, 2004).
Although women engage in same-sex physical confrontations less frequently and less seriously than men, fights do occur. In the United States, girls account for 33% of arrests for simple assault and 24% of aggravated assaults (Girls Study Group, 2008), with the larger gender gap for aggravated assault reflecting girls' less injurious behavior and lower likelihood of using weapons. Surveys indicate that in the previous year, 40.5% of boys and 25.1% of girls had been in a physical fight (Grunbaum et al., 2004). In the previous month, 60% of girls had called another girl names, 50% had sworn at them, and 35% had pushed or shoved them (Artz, Nicholson, & Magnuson, 2008). Female assaults most commonly occur among 15- to 24-year-olds, predominantly between friends and acquaintances, and the most frequent forms of attack are pushing, shoving, grabbing, tripping, slapping, kicking, and punching (Campbell, 1986; Ness, 2004). The reasons for fighting are often connected directly or indirectly to young men and fall broadly into three categories. The first is defending a sexual reputation. Gossip about a girl's promiscuity gives rise to rumor that may find its way back to the target, triggering an attack as a means of reclaiming her threatened identity. A second source of provocation is competition for potential partners. Around the world, access to men and their resources is responsible for the majority of women's fights (Burbank, 1987). Thirdly, jealousy about proprietary ownership of a current partner is a frequent source of conflict.
But the probability of escalating from indirect aggression to outright physical attack is not randomly distributed. There are ecological and demographic factors that concentrate it among the young in the poorest neighborhoods. Men and women show a curvilinear age–violence relationship, with aggression rising in the early teenage years and falling away in the mid-20s. Although male violence is far more prevalent than female violence, the shape of the curve is very similar for both sexes—with one exception. Violence rises and peaks earlier in girls by about 2 years, corresponding to girls' 2-year-earlier attainment of sexual maturity (Campbell, 1995). Early menarche is predictive of girls' aggression. Life history theory forms the basis for expecting that age of menarche should be responsive to cues from the local environment that canalize development toward a “fast” or “slow” reproductive tempo. Resource scarcity, high rates of early mortality, psychosocial stress, low-quality parental investment, father absence, and stepfather presence signal environmental uncertainty and unpredictability, accelerating pubertal timing and reproduction in an adaptive fashion. In deprived neighborhoods, girls may experience many of these risk factors simultaneously. These girls begin their sexual careers earlier, putting them at a significant advantage over their peers. In addition, older girls are acutely sensitive to the entry of younger competitors into the mating arena and this may increase their likelihood of victimization and retaliation. Girls who reach menarche early are more likely to be involved in delinquent and aggressive behavior, and this is especially true for maltreated girls (Negriff & Trickett, 2010) and those living in disadvantaged neighborhoods (Obeidallah, Brennan, Brooks-Gunn, & Earls, 2004).
The operational sex ratio is an index of the availability of prospective mates. Women are most likely to find a partner in the immediate neighborhood, and to the extent that the local male–female ratio drops below unity, competition between women increases. The mortality rate among men is considerably higher than among women, especially between the ages of 15 and 35 (Kruger & Nesse, 2006). At the age of 25, men are 3 times more likely to die from all causes than women, and 4 times more likely to die from external causes. Urban areas are more likely to have a female-biased sex ratio (Edlund, 2005). In the U.S. Black population especially, there is a severe shortage of men. As a result of high male mortality and incarceration rates in the 20-to-29 age group (Tucker & Mitchell-Kernan, 1995), there are 85 men for every 100 women compared to 99 men for every 100 women among Whites. When the local male–female ratio drops below unity, competition between women increases. It also means that men find themselves in an excellent bargaining position. Usually only a very few well-resourced and highly desirable men are in a position to successfully pursue a polygynous strategy, but when men are in short supply, market forces mean that men are in a position to call the shots, effectively enforcing a “short-term only” mating strategy on women (Campbell, 1984). This may be far from ideal from young women's point of view, but the laws of supply and demand mean that such men often get their way, with young women adapting their resource-extraction tactics accordingly. As one young woman advised: “I tell her take all his paper, all of it, 'cause it's just a matter of time and he's gonna do some rotten dog shit on her.… Got to get it when you can. You never know when it's gonna stop and you better get much as you can while you can.… When fellas get tired of your pussy, it's good-bye girl, naw, it's get the fuck out of my life bitch! Next bitch!” (Taylor, 1993, pp. 97, 131).
High variation in men's resources further intensifies female competition. Among middle-class young women, the costs of escalating to direct competition are rarely worth it: The difference between marrying a doctor or an accountant is not sufficiently great. But in deprived areas, the difference between the desperate poverty of “dope fiends” and the conspicuous consumption of “high rollers” is extreme. The desirability of access to material resources means that well-resourced men are worth fighting for. Antagonism can be heightened further after a young woman bears a man's child: Even after the relationship has ended, a BM (“baby momma”) feels entitled to make claims on the father's income and to repel rival women who threaten to divert his resources (Ness, 2004). Young men's preference for sexual novelty is a constant threat to relationships (Symons, 1979), and young women are especially sensitive to attractive or newly arrived girls: “It's like, if another girl gets attention, she's taking it away from you. It's as if she's saying she's better than you. So you gonna knock her down a notch” (Ness, 2004, p. 40).
Cultural norms support the use of women's violence in deprived inner-city neighborhoods where it is most common (Leventhal & Brooks-Gunn, 2000). For families in such neighborhoods, the frequent absence of a consistent father figure means that mothers (and grandmothers) play a pivotal role. They are strong figures who must cope alone with the daily stresses of subsistence living. Many mothers are themselves involved in fighting, especially in defense of their family's good name. Some become actively involved in their daughter's fights and, in doing so, become role models and allies (Ness, 2004). Mothers' concern for their daughters' welfare translates into tolerance (and sometimes encouragement) of fighting. Most mothers acknowledge that a girl needs to be able to “stand her ground” and “hold her own.” The strength and resilience of women is not seen as incompatible with femininity: Indeed, passivity is viewed as a weakness rather than an asset. As Irwin and Adler (2012, p. 319) noted, “Given the emphasis on female strength, girls lost respect for and even targeted other girls who fell short in fulfilling idealized notions of feminine resilience circulating in the local communities.”
If weakness makes a girl a target, an important benefit of willingness to fight is the avoidance of victimization. Girls' reports of their fights present aggression as a form of self-defense by emphasizing that their opponent “started it.” In some cases, “starting it” refers to a physical assault, but more often to rumor spreading or verbal taunts to which physical aggression is seen as the appropriate response. The slippery divide between physical and verbal provocation is mirrored in the fuzzy distinction between self-defense and reputation enhancement. For many girls, success in a public fight achieves more than the immediate goal of causing an opponent to back off: It promotes a “crazy” or “mean” reputation that will deter others from future attacks (Jones, 2004; Miller & Mullins, 2006). Reputation enhancement involves a disproportionate response to any perceived act of “disrespect,” including pejorative gossip, staring, and a demeanor that presumes social superiority (a girl who “thinks she's all that”). Once established, reputations must be defended against others who are seeking to enhance their own. One response is for tough girls to get their retaliation in first. In this way, a self-reinforcing cycle develops between sensitivity to challenge, self-defense, reputation enhancement, and preemptive aggression. Although many disputes appear to be about securing and defending status-enhancing relationships with desirable boys, the motivations can be a complex mixture of rivalry, jealousy, and reputation management. As one girl explained, “I don't care about the guy or anything but I'm gonna mess that girl up cause she deserves it. The bitch just be asking for it. The way I see it, I ain't fighting over the boy. I'm fighting the girl because she be acting in a way that says she thinks I'm a punk” (Ness, 2010, p. 84). The right to be treated with appropriate respect lies at the heart of impression management theories of aggression (Felson, 1978) and public disrespect is a common trigger for anger and aggression. This is as true for inner-city girls as it is for others (Ness, 2010). Young women's fights are triggered by challenges to personal integrity in a range of domains, but slurs on a girl's attractiveness and sexual continence are potent and pervasive sources of conflict.
Young women's verbal bluster and bravado should not be taken as an indication of fearlessness. A theme that recurs in ethnographic reports is the need to control fear and suppress its expression in order to avoid victimization. This is as true in Philadelphia (“If I seem like I'm scared to fight, some girl is gonna think she can mess with me all the time” [Ness, 2004, p. 38]) as it is in Glasgow (“Cos if you show fear of somebody, they're just gonna walk all over the top of you. If you show fear of them, they always come back tae you” [Batchelor, Burman, & Brown, 2001, p. 130]). Growing up in these communities, fear is an emotion that signals weakness and young women must learn to master its expression.
At a proximate psychological level, evidence suggests that women's lower level of aggression is not explained by their more placid temperament (women experience anger as often and as intensely as men) nor by a sex difference in impulsivity (women can and do act on impulse but are less likely to do so when such acts carry potentially risky consequences; Cross et al., 2011). Rather, it is women's greater fear that restrains their use of overt aggression. Their greater sensitivity to danger is also manifested in their higher levels of subclinical and clinical anxiety, stronger risk aversion, and lower levels of sensation seeking. At an ultimate evolutionary level, the sex difference in the calibration of the fear threshold is predicated on the greater importance of the mother than the father in assuring offspring survival. Reproductively, women are quality, not quantity, specialists. Their output is limited by lengthy infertile periods of gestation and lactation, and truncated by menopause long before somatic senescence. Their investment in each offspring is measured in years as a result of babies' long and demanding period of dependency. Despite this, women are unusual among primates in their short interbirth intervals, resulting in the need to care simultaneously for more than one dependent child. Such a feat requires assistance, and I have suggested that ancestral women traded genetic diversity of offspring for paternal help in resource provision. Because paternal care does not enhance offspring survival sufficiently to compensate for the abandonment of a polygynous strategy, an explanation in terms of male sexual selection is problematic. A plausible alternative scenario is that biparental care arose as a result of sexual conflict in which women granted exclusive sexual access preferentially to men who were willing to invest in their offspring. But with biparental care and monogamy came two-way sexual selection: Men became increasingly choosy in their choice of long-term partners and in response, women competed with one another to advertise qualities that were attractive to men. This is not to deny that mothers created strong bonds of mutual cooperation with one another, despite or because of female emigration from the natal group. (Whether or not such bonds are evidence of “cooperative breeding” depends on one's definition of the term. In most cooperatively breeding species, a single female aggressively monopolizes breeding [Young & Bennett, 2013], a situation that clearly does not apply to humans. Other researchers [e.g., Mace, 2013] treat the term as effectively interchangeable with kin selection since the helpers have a genetic interest in the offspring [e.g., siblings, grandmothers]). Under extreme conditions of resource scarcity, the limits of mutual benevolence between genetically unrelated women would have been sorely tested. Mothers and fathers may be unrelated, but they share a common genetic investment in their offspring, and a substantial history of monogamy has shaped hormonal and neural adaptations for emotional bonding in both sexes (De Boer, Van Buel, & Ter Horst, 2012). The extremity and visibility of female competition is moderated by interlinked ecological and cultural factors. Harsh environments recalibrate life history tempos and increase competition in women as they do in men: Correlations between the sexes' rates of violence across geographic regions exceed .90 (Campbell, 1999). Community tolerance for women's aggression is correspondingly adjusted, so that the concept of “femininity” embraces strength, resilience, and the public denial of fear.
The past few years have seen a surge of interest in female competition by evolutionary biologists (e.g., Rosvall, 2013a; Stockley & Bro-Jørgensen, 2011). To add to their insights, evolutionary psychologists are increasingly stepping up to the special challenges of understanding competition in our own species (Benenson, 2014; Fisher, Garcia, & Chang, 2013; Vaillancourt, 2013) with our capacity for representational thought, symbolic language, and cultural transmission. Yet these impressive human abilities interact with and serve an ancient stratum of affective adaptation, shared with other species, which have been shaped by universal principles of natural and sexual selection.