You are on page 1of 12

Review of General Psychology Copyright 2008 by the American Psychological Association

2008, Vol. 12, No. 1, 86 –97 1089-2680/08/$12.00 DOI: 10.1037/1089-2680.12.1.86

Why Do Men Rape? An Evolutionary Psychological Perspective


William F. McKibbin, Todd K. Shackelford, Aaron T. Goetz, and Valerie G. Starratt
Florida Atlantic University

Rape of women by men has occurred throughout recorded history and across cultures. In
this article, we discuss rape from an evolutionary psychological perspective. Evolutionary
psychology is a powerful heuristic tool that allows researchers to develop and test novel
hypotheses about complex behaviors such as rape. Some researchers have argued that men
have evolved psychological mechanisms that motivate them to rape in specific contexts. We
discuss evidence consistent with this claim, and argue that a more nuanced view of men’s
rape behavior is necessary. We propose that it may be useful to characterize rapists as
belonging to one of several types, distinguished by individual differences as well as by the
circumstances in which they are predicted to commit rape. We discuss research evidence in
support of each rapist type, as well as the need for future research. Finally, we discuss
research concerning women’s rape-avoidance psychology and behavior.

Keywords: rape, rape avoidance, evolutionary psychology

Rape is a fact of life across cultures (Rozée, potheses across all domains of psychology.
1993; Sanday, 1981). In American samples, es- Evolutionary psychology rests on several key
timates of the prevalence of rape vary with the premises (Buss, 2004; Tooby & Cosmides,
population studied, but are as high as 13% for 2005). The first premise states that the complex-
women (Kilpatrick, Edmunds, & Seymour, ity of human behavior can only be understood
1992). Rape is likely more common, however, by taking into account human evolutionary his-
because rapes often go unreported. Researchers tory and natural selection. Second, behavior de-
estimate that 67– 84% of rapes are not reported pends on evolved psychological mechanisms.
(Greenfield, 1997; Kilpatrick et al., 1992). Al- These are information-processing mechanisms
though other forms of rape occur (e.g., male– housed in the brain that register and process
male rape), this article focuses on the rape of specific information and generate as output spe-
women by men. Rape is typically defined, and cific behaviors, physiological activity, or input
defined in this article, as the use of force or relayed to other psychological mechanisms.
threat of force to achieve penile-vaginal pene- Third, evolved psychological mechanisms are
tration of a woman without her consent (Kil- functionally specialized to perform a specific
patrick et al., 1992; Thornhill & Palmer, 2000). task or to solve a specific problem that recur-
Before we review rape from a modern evo- rently affected reproductive success over evo-
lutionary psychological perspective, we provide lutionary history. This premise is often referred
a brief, nonexhaustive review of evolutionary to as domain specificity. Finally, the numerous-
psychology (for a more complete review, see ness premise states that human brains consist of
Buss, 2004; Tooby & Cosmides, 2005). Evolu- many specific evolved psychological mecha-
tionary psychology is a powerful heuristic tool nisms that work together to produce behavior.
that can be used to generate new testable hy- Together with other theoretical tools and heu-
ristics provided by modern evolutionary theory,
these premises are used to generate testable,
falsifiable hypotheses.
William F. McKibbin, Todd K. Shackelford, Aaron T.
Goetz, and Valerie G. Starratt, Florida Atlantic University.
Aaron T. Goetz is now at California State University, Misconceptions About Evolutionary
Fullerton.
Correspondence concerning this article should be ad-
Psychology
dressed to William F. McKibbin, Department of Psychol-
ogy, Florida Atlantic University, 2912 College Avenue, Some scholars believe that evolutionary psy-
Davie, FL 33314. E-mail: wmckibbi@fau.edu chological research is conducted to justify rac-
86
EVOLUTIONARY PERSPECTIVE ON RAPE 87

ism, sexism, or other undesirable “-isms.” For Comparative Psychology of Sexual


example, Tang-Martinez (1997, p. 116) de- Coercion and Rape
scribes a common feminist view that evolution-
ary psychology is, “inherently misogynistic and Evolutionary metatheory has been used to
provides a justification for the oppression of generate the hypotheses that sexual coercion
women.” However, the feminists to whom and rape occur in species in which males are
Tang-Martinez refers are committing what is more aggressive, more eager to mate, more sex-
known as the naturalistic fallacy: the error of ually assertive, and less discriminating in
deriving what ought to be from what is. This choosing a mate (Thornhill & Palmer, 2000).
error can be demonstrated clearly with an ex- Sexual coercion and rape occur in insects
ample: No sensible person would argue that a (Linder & Rice, 2005; Thornhill, 1980, 1981),
amphibians and reptiles (Reyer, Frei, & Som,
scientist researching the causes of cancer is
1999; Shine, Langkilde, & Mason, 2003), fish
thereby justifying or promoting cancer. Yet
(Magurran, 2001; Plath, Parzefall, & Schlupp,
some people argue that investigating rape from 2003), birds (Gowaty & Buschhaus, 1998; Piz-
an evolutionary perspective justifies or legiti- zari & Birkhead, 2000), and primates (Smuts &
mizes rape (e.g., Baron, 1985; Marshall & Bar- Smuts, 1993; Wrangham & Peterson, 1996),
rett, 1990, cited in Thornhill & Palmer, 2000). among others.
Related to the naturalistic fallacy is the false Two species in particular provide clear ex-
belief of genetic determinism: The idea that amples of adaptations in males to rape females.
behavior is unalterable, programmed, or other- A large body of evidence demonstrates that
wise unchangeable. Biologist John Maynard male scorpionflies (Panorpa vulgaris) have a
Smith noted that genetic determinism is, “an notal organ that is designed exclusively to fa-
incorrect idea that is largely irrelevant, because cilitate sexual access to a female in a coercive
it is not held by anyone, or at least not by any fashion, that is, rape (Thornhill, 1980, 1981;
competent evolutionary biologist” (1997, p. Thornhill & Sauer, 1991). Scorpionfly males do
524). No evolutionary psychologist would ar- not always secure copulations through rape. In-
gue that because rape is produced by evolved stead, males display conditional mating strate-
mechanisms, we should accept its occurrence. gies. Males that are able to produce food for the
The goal of evolutionary psychology, like the female are allowed to mate without coercion.
goal of any science, is to further our understand- Males that are not able to do so resort to the
ing of the phenomenon of interest, which in this conditional rape strategy and use of the notal
case is rape. Researching rape from an evolu- organ (Thornhill, 1980, 1981; Thornhill &
tionary psychological perspective does not jus- Palmer, 2000). Thus, male scorpionflies exhibit
tify this heinous act. Our goal is a greater un- evidence of specific anatomical traits that
derstanding of the causes of rape, which may evolved to facilitate rape and of a conditional
help others prevent its occurrence. strategy of sexual coercion.
Male orangutans (Pongo pygmaeus) also de-
Finally, researchers using an evolutionary
ploy conditional strategies of sexual coercion
psychological perspective often frame hypothe-
and rape. Orangutans are unique among apes in
ses in terms of the costs and benefits to an
that they live solitary lives. Females therefore
organism of performing a particular behavior. do not have mates or kin that may deter or
These costs and benefits refer to the effects on prevent rape (Wrangham & Peterson, 1996).
reproductive success over evolutionary time. This fact alone makes rape a more viable strat-
Costs decrease the probability of successful re- egy for male orangutans. Forced copulations
production, whereas benefits increase the prob- account for up to half of all copulations (Mitani,
ability of successful reproduction. These terms 1985; Wrangham & Peterson, 1996). These
are sometimes misconstrued as referring to a forced copulations seem to be performed pri-
more general idea of perceived costs and bene- marily by a subset of males. Wrangham and
fits to the individual or to society. However, Peterson (1996) reviewed evidence indicating
these terms carry no moral or ethical meaning that male orangutans exist as one of two distinct
and are used only in terms of naturally selected morphs or behavioral types. The large morphs
biological functioning. weigh significantly more, move much slowly,
88 MCKIBBIN ET AL.

and are typically able to find females willing to mechanisms. Thornhill and Thornhill (1992; see
mate with them. The small morphs are typically also Thornhill, 1999; Thornhill & Palmer,
unable to find females willing to mate with 2000) have identified several possible rape ad-
them. These small morphs are more likely to aptations. These adaptations are proposed to be
chase down and rape females. This represents a universal features of male psychology that are
conditional strategy. If the smaller males are activated under specific circumstances.
unable to gain sexual access to females through A hypothesized design feature of rape adap-
intrasexual competition and by being attractive tations involves mechanisms that cause men to
to females, they may use the conditional strat- evaluate the sexual attractiveness of rape vic-
egy of chasing down and raping a female. tims differently than for consensual partners.
Comparative evidence indicates that males of Specifically, a rapist might be more successful
many species have evolved strategies to sexu- reproductively by maximizing the chance that a
ally coerce and rape females. Rape in humans one-time forced copulation will result in preg-
must also reflect adaptations constructed over nancy. According to this hypothesis, a would-be
evolutionary time. Although numerous explana- rapist may be more likely to target a highly
tions have been offered to explain rape in hu- fertile woman than a woman who is less fertile
mans (e.g., learning or enculturation, mental (Thornhill & Palmer, 2000). Human female fer-
illness, personality differences, drug and alco- tility (current likelihood of conception per cop-
hol use, and other factors; Bergen & Bukovec, ulation) peaks in the early to mid-20s. There-
2006, Dean & Malamuth, 1997, Lalumiére & fore, if women in this age range are overrepre-
Quinsey, 1996), these factors alone cannot ex- sented in reports of rape, it is possible that this
plain the existence of such seemingly complex reflects a male adaptation that leads to raping
behavior. These factors may increase the likeli- fertile women more often than nonfertile
hood of rape, but cannot explain the complex women. Numerous studies have documented
organized behavior seen in rape. Only two ex- that young women are most often targeted by
planations are likely to be true: that rape is the rapists, and that women of peak fertility are
product of specialized psychological adaptation, overrepresented in reported and unreported
or that it is a byproduct of other adaptations in rapes (Ghiglieri, 2000; Greenfield, 1997; Kil-
the male mind (Palmer & Thornhill, 2003a, patrick et al., 1992; Shields & Shields, 1983,
2003b; Thornhill & Palmer, 2000). What evi- Thornhill & Palmer, 2000; Thornhill & Thorn-
dence supports the hypothesis that rape is the hill, 1983). This evidence does not support ex-
result of an adaptation? clusively rape-specific adaptation, however, be-
cause men exhibit a preference for sexually
Evidence of Human Adaptations for attractive partners in general, not just in con-
Sexual Coercion and Rape texts of rape (see, e.g., Buss, 1994, 2004).
We, like others (e.g. Thornhill & Palmer,
For rape to be produced by evolved psycho- 2000), propose that rape is a conditional strat-
logical mechanisms, it must have recurrently egy that may potentially be deployed by any
generated reproductive benefits for ancestral man. Similarly, Shields and Shields (1983) ar-
rapists. These benefits must have outweighed gued that men use a conditional mating strategy
the costs that men may incur if they attempt or consisting of many mating tactics, including
successfully complete a rape. Despite the costs, rape. At least one-third of men admit they
there is evidence that rape may have increased would rape under specific conditions, and many
the number of women with whom ancestral men men report coercive sexual fantasies (see Mala-
copulated and, therefore, the reproductive suc- muth, Huppin, & Paul, 2005, for a review).
cess of rapist males (Gottschall & Gottschall, Such evidence suggests that rape adaptations
2003; Holmes, Resnick, Kilpatrick, & Best, might be universal features of male psychology.
1996; Krueger, 1988; Shields & Shields, 1983; Empirical support for evolutionary psychologi-
Thornhill, 1999; Thornhill & Palmer, 2000). cal hypotheses of rape has been mixed. For
Men do not have morphological features example, the mate-deprivation model of sexual
analogous to the notal clamp of male scorpion- coercion, in which men with limited or no sex-
flies. Any rape adaptations that men possess are ual access to women rape for lack of other
likely to occur in the form of psychological options, typically has not been supported (Mala-
EVOLUTIONARY PERSPECTIVE ON RAPE 89

muth et al., 2005; but see also below). This domains, with one domain being sexual asser-
mixed support may reflect a lack of appreciation tiveness, which might lead to rape.
that there may be several distinct types of rap-
ists. As a heuristic strategy, we have defined Specialized Rapists
several rapist types. Specifying these types may
generate new insights and testable hypotheses. Another type of rapist may be the specialized
Other researchers have suggested that defining rapist. Men in this group are distinguished by
subtypes of rapists can be potentially valuable being sexually aroused by violent sexual stim-
(Malamuth et al., 2005). uli. These men may possess a psychology that
We hypothesize that rape may represent a produces differences in sexual arousal in re-
conditional mating strategy, present in all men, sponse to depictions of rapes versus depictions
which may result from several qualitatively dif- of consensual sex. Because rape carries high
ferent ancestral contexts combined with indi- potential costs for the rapist, particularly if
vidual difference factors among men. Specifi- caught in the act, rapists with a psychology that
cally, we propose five types of rapists (or con- motivated quicker arousal and ejaculation dur-
texts of rape), (1) disadvantaged men who resort ing rape might have been more successful than
to rape, (2) “specialized” rapists who are sexu- men who did not possess such a psychology
ally aroused by violent sex, (3) men who rape (Thornhill & Palmer, 2000).
opportunistically, (4) high-mating-effort men Support for the existence of this hypothesized
who are dominant and often psychopathic, and group has been generated by investigating
(5) partner rapists motivated by assessments of whether men are aroused by depictions of rape
increased risk of sperm competition. We next versus depictions of casual sex. Meta-analyses
discuss evidence for each type of rapist. indicate that convicted rapists demonstrate
greater sexual arousal to scenes of sexual coer-
Disadvantaged Men cion involving force than do nonrapists (Hall,
Shondrick, & Hirschman, 1993; Lalumiére &
The first hypothesized rapist type includes Quinsey, 1994; Thornhill & Thornhill, 1992).
men who are motivated to rape if they have no Specialized rapists also might possess mech-
other means of securing copulations. This may anisms that cause them to evaluate the sexual
be referred to as the disadvantaged male hy- attractiveness of rape victims differently than
pothesis. This hypothesis also has been referred the sexual attractiveness of consensual partners.
to as the mate deprivation hypothesis (Lalumi- According to this hypothesis, a rapist will be
ére, Chalmers, Quinsey, & Seto, 1996). It is more likely to rape a highly fertile woman than
supported by data indicating that rapes are com- a woman who is less fertile (Thornhill &
mitted disproportionately by men with low so- Palmer, 2000). Research has demonstrated sup-
cioeconomic status (Kalichman, Williams, port for this hypothesis (see above for details).
Cherry, Belcher, & Nachimson, 1998; Thornhill However, it is unclear if this reflects a special-
& Thornhill, 1983). Furthermore, Krill, Lake, ized rape adaptation or a more general male
and Platek (2006) presented evidence that men mating strategy. Future research might test the
convicted of rape display lower facial symme- hypothesis that men evaluate the sexual attrac-
try, an indicator of poor genetic quality. Facial tiveness of rape victims differently than the
symmetry is linked positively with physical and sexual attractiveness of consensual partners by
psychological health (Shackelford & Larsen, examining whether men target for rape repro-
1997), and men with lower facial symmetry are ductive-age women who are in the most fertile
perceived as less attractive and as less desirable phases of their menstrual cycles. Such a finding
mates (Gangestad & Thornhill, 1999; Ganges- would provide stronger support for this rapist
tad, Thornhill, & Yeo, 1994). Deprived of mates type.
by normal means, some men may resort to rape. If a rape is a one-time event, it might make
Identification of such a rapist type, however, adaptive sense for the rapist to inseminate the
would not necessarily imply a conditional strat- woman with an ejaculate that contains a high
egy for rape. One can imagine that when repro- sperm count or that otherwise increases the
ductive opportunities are dismal, some men chance of successful fertilization. Indeed,
might be motivated to take more risks in all Thornhill and Palmer (2000) have hypothesized
90 MCKIBBIN ET AL.

that some rapists may be capable of producing a Opportunistic Rapists


high-sperm-count ejaculate that would increase
the chance of fertilization. Men seem to be The third hypothesized rapist type is the op-
capable of unconsciously adjusting sperm num- portunistic rapist. These men generally seek out
ber in ejaculates, such as in response to a greater receptive women, but might shift to sexual co-
risk of sperm competition (Baker & Bellis, ercion and rape if women are not receptive, or if
1989, 1993), but it is unknown whether rapists the associated costs, such as injury or retaliation
adjust sperm numbers during rape. Evidence for by the victim, the victim’s family, or society,
are particularly low. All rapists are predicted to
this would lend support to the specialized rapist
be attuned to a potential victim’s vulnerability,
type.
but an opportunistic rapist is especially so. The
Researchers have argued that premature ejac-
universality of laws and societal norms prohib-
ulation might have been adaptive ancestrally, iting rape (wife rape being a special exception;
perhaps by minimizing the chances of predation see below) indicates an appreciation that men
or detection by jealous mates (Hong, 1984; see are more likely to rape when the costs are low
also Gallup & Burch, 2004). It also might make (Palmer, 1989; Thornhill & Palmer, 2000). The
adaptive sense for a rapist to ejaculate as soon fact that rapes regularly occur during wartime
as possible after achieving copulation. This has been presented as evidence of the assess-
would reduce the chances of being injured or ment of victim vulnerability and decreased like-
retaliated against. Therefore, it is possible that lihood of detection (e.g., Gottschall, 2004). Men
selection may have acted to minimize the time it in war are likely to have lowered costs of com-
takes for a man to ejaculate during a rape. mitting rape, because punishment or retaliation
Research is needed to test this hypothesis. For is less likely.
example, one might compare the average pre- The evidence for the existence of this type of
ejaculatory copulation length during rape versus rapist, however, is minimal. Theft also is com-
during consensual copulation. mon during war, and for the same reason: pun-
There is indirect evidence corroborating the ishment or retaliation is unlikely. Support for
hypothesis that rapists’ ejaculates are more this hypothesized type may be seen in research
competitive than nonrapists. Gottschall and demonstrating that women with family mem-
Gottschall (2003) estimated that pregnancy bers, particularly adult male family members,
rates resulting from rape are two times that of living nearby are much less likely to be physi-
consensual per-incident rates. That is, approxi- cally assaulted by their partner (Figueredo et al.,
mately 6% of rapes result in pregnancy com- 2001; Kanin, 1957). This suggests that potential
pared to approximately 3% of consensual cop- rapists are attending to the probability of retal-
iation by a victim’s adult male family members.
ulations. Even after controlling statistically for
the age of the woman, the researchers identified
a higher conception rate for rapes than for con- High-Mating-Effort Rapists
sensual sex. This evidence suggests that there
A fourth hypothesized rapist type is the high-
may be something different about rapists’ psy-
mating-effort rapist. High-mating-effort rapists,
chology or the competitiveness of their ejacu-
in contrast to other types, such as disadvantaged
lates. Further research is needed, however. One rapists, appear to be more sexually experienced
promising area of research is the study of semen (Lalumière & Quinsey, 1996). These rapists
chemistry. Burch and Gallup (2006) hypothe- may be characterized as aggressive, dominant,
sized that men may have an adaptation that and having high self-esteem. Such rapists often
functions to adjust semen chemistry to cause may be characterized as psychopathic (Lalumi-
ovulation immediately following a rape. Future ère, Harris, Quinsey, & Rice, 2005). Lalumière
research could profitably test this hypothesis, et al. argue that high mating effort is an impor-
perhaps by comparing chemical constituents of tant facet of psychopathy. They argue that al-
ejaculates produced by men exposed experi- though most men appear to deploy mating strat-
mentally to a coercive sexual scenario with egies according to environmental contexts, psy-
ejaculates produced by men exposed experi- chopathic men deploy a high-mating-effort
mentally to a noncoercive sexual scenario. strategy in most contexts, pursuing many part-
EVOLUTIONARY PERSPECTIVE ON RAPE 91

ners with little investment, and using coercion most likely to occur when a man learns or
and rape when noncoercive tactics fail. There is suspects that his long-term partner recently has
evidence that psychopathic men display lower been sexually unfaithful (Thornhill & Thornhill,
fluctuating asymmetry, an index of overall fit- 1992).
ness (Lalumière, Harris, & Rice, 2001), further Partner rapes comprise a substantial propor-
distinguishing this rapist type from others, such tion of reported rapes (Kilpatrick et al., 1992;
as the disadvantaged rapist. Russell, 1990). Between 10% and 26% of
Research evidence corroborates the plausibil- women report experiencing rape in marriage
ity of this rapist type. Dean and Malamuth (Russell, 1990; Watts, Keogh, Ndlovu, &
(1997), for example, found that men who scored Kwaramba, 1998). Women are particularly
high on a Sexual Experience measure, “were likely to be raped by their partner during a
more likely to report sexual coercion if they breakup instigated by men’s concerns about
were also self-centered as opposed to nurturant” their partner’s infidelity (Thornhill & Palmer,
(p. 74). Premarital sexual coercion is associated 2000). Until very recently in Western society, it
with sexual promiscuity, earlier onset of sexual was not considered a crime if a man forced his
activity, and greater sexual experience (Chris- wife to have sex with him. The right of men to
topher, Owens, & Stecker, 1993; Lalumière et sexual access to their partner was considered
al., 2005). Lalumière and Quinsey (1996) found absolute, and only relatively recently in the
that a strong indicator of past sexual coercion is United States have men been prosecuted for
positive self-perceived mating success and an raping their wives (Bergen, 1996; Russell,
extensive history of casual sexual relationships. 1990).
Finally, the risk of date rape is greater when the Studying men’s psychological reactions to
man initiated the date, spent money on the risk of sperm competition is another method by
woman, and provided transportation (Muehlen- which the hypothesis that men are motivated to
hard & Linton, 1987). Perceived relative depri-
rape their partners under conditions of sperm
vation, in which an individual’s (high) expecta-
competition might be tested. If men display
tions about having sex are not satisfied (Mala-
psychological reactions to risk of sperm com-
muth et al., 2005) also may play a role in the
petition in noncoercive contexts, it is also pos-
sexually coercive behavior of high-mating-
effort men. For example, men who report a sible that they do so in coercive or rape con-
greater likelihood of committing rape tend to texts. Research evidence indicates that men do
endorse statements expressing an increased per- display such psychological reactions. For exam-
ception of mate deprivation, but do not report an ple, men are more aroused by and prefer sexu-
overall fewer number of sexual opportunities ally explicit images that suggest the occurrence
(Glick & Fiske, 1996; Lonsway & Fitzgerald, of sperm competition than by sexually explicit
1995). More research must be conducted to test images that do not suggest the occurrence of
this hypothesized rapist type. For example, re- sperm competition (Kilgallon & Simmons,
searchers might test whether men convicted of 2005; Pound, 2002). Furthermore, men who
date rape or sexual assault score higher on mea- spend a greater proportion of time apart from
sures of psychopathy. their partners since the couple’s last copulation
(and therefore face a higher risk of sperm com-
Partner Rapists petition) report that they find their partner more
attractive, are more interested in copulating
A final hypothesized rapist type includes men with their partner, and believe that their partner
motivated to rape their partners under condi- is more interested in copulating with them
tions of increased sperm competition risk. (Shackelford, Goetz, McKibbin, & Starratt,
Sperm competition is the competition that can 2007; Shackelford et al., 2002). These results
occur between the sperm of different males to a are independent of relationship satisfaction, to-
female’s eggs (Parker, 1970). The outcome of tal time since last copulation, and total time
sperm competition is biased in favor of males spent apart. The psychological mechanisms that
who produce greater numbers of sperm (Parker, lead men to experience greater interest in cop-
1970, 1982; Pound et al., 1982). Rape in re- ulation, and to believe their partner is interested
sponse to increased risk of sperm competition is in copulation with them, also may be part of the
92 MCKIBBIN ET AL.

suite of mechanisms that lead men to coerce or Women’s Defenses Against Rape
rape their partners.
Finally, in a direct test of the hypothesis that Rape is a traumatic event that is likely to have
men may rape their partners under conditions of been a recurrent problem for women over evo-
sperm competition, Goetz and Shackelford lutionary history. Rape often leads to many neg-
(2006) documented in two studies that men’s ative consequences for women and, therefore,
sexually coercive behavior is positively related women may have evolved psychological mech-
to their partner’s infidelities, that is, to the risk anisms designed to motivate rape-avoidance be-
of sperm competition. Men with partners who haviors. There are several reasons why rape is
committed infidelities, or who suspected that traumatic for women. These include disrupting
their partner had committed infidelities (indicat- a woman’s parental care, causing a woman’s
ing increased risk of sperm competition), were partner to abandon her, and causing a woman
more likely to perform sexually coercive behav- serious physical injury (Thornhill & Palmer,
iors, including rape. These findings are consis- 2000) or death (Shackelford, 2002). Aside from
tent with the existence of psychological mech- death, perhaps the greatest cost to women who
anisms that motivate men to commit partner are raped is the circumvention of their mate
rape in response to risk of sperm competition. choice. This is because anything that circum-
In summary, it may be useful to characterize vents women’s choice in mating can severely
rapists as falling into one of several categories jeopardize their reproductive success (Symons,
or types, specifically (a) disadvantaged men, (b) 1979).
specialized rapists, (c) opportunistic rapists, (d) Researchers have speculated that a variety of
high-mating-effort men, and (5) partner rapists. female traits evolved to reduce the risks of
Although future research is needed to test the being raped. Smuts (1992) argued that women
hypothesized types of rapists, prior studies offer form alliances with groups of men and other
some preliminary support for this model. We women for protection against would-be rapists.
have identified potential unique ancestral con- Wilson and Mesnick (1997) proposed and
texts and individual differences that may have found support for the bodyguard hypothesis:
selected for conditional rape strategies. But Women’s mate preferences for physically and
these contexts and individual differences can be socially dominant men may reflect antirape ad-
overlapping. This is to be expected, however, as aptation. Of course, women may form alliances
we argue that all men may possess adaptations or prefer dominant mates for reasons other than
to rape. For example, a high-mating-effort con- to avoid rape. Alliances offer protection from
text and an opportunity context are not mutually such dangers as assault or predation, and dom-
exclusive: a man who devotes much of his time inant mates may possess higher-quality genes,
and energy to gaining short-term matings may for example. Davis and Gallup (2006) proposed
be even more likely to commit rape when cir- the intriguing possibility that preeclampsia and
cumstances (such as wartime) allow him to do spontaneous abortion may be adaptations that
so at decreased cost (e.g., retaliation). function to terminate pregnancies not in the
The existence of adaptations to rape does not woman’s best reproductive interests, such as
mean that rape is inevitable or justified. Like all those resulting from rape. Little empirical work
psychological mechanism, rape mechanisms re- has been conducted to identify specific psycho-
quire functioning genetic and environmental logical mechanisms that evolved to solve the
components. Rape is only predicted to occur recurrent problem of rape avoidance.
under specific environmental circumstances that Thornhill and Thornhill (1990a, 1990b,
activate men’s evolved psychology. Further- 1990c) have demonstrated that the psychologi-
more, because rape behaviors may have a ge- cal pain that women experience after being
netic component does not mean that men cannot raped may be produced by evolved mechanisms
control their behavior. Just as men thwart their designed to focus women’s attention on the
evolved psychology every time they choose less circumstances of the rape, particularly the social
calorically dense food over more calorically cirumstances that resulted in the rape. They
dense food (as when one is on a diet), so too can argue that, like physical pain, psychological
men thwart evolved mechanisms that may oth- pain motivates individuals to attend to the cir-
erwiselead them to sexually coerce or rape. cumstances that led to the pain and to avoid
EVOLUTIONARY PERSPECTIVE ON RAPE 93

those circumstances in the future. Victims of a woman being sexually assaulted, with riskier
rape who have more to lose in terms of future behaviors given higher risk scores. Individuals’
reproductive success will also experience more risky behavior was estimated by taking the
psychological pain relative to women with less summed composite score of all performed ac-
to lose in terms of future reproductive success tivities. Women in the fertile phase of their
(Thornhill & Thornhill, 1983, 1990a; Thornhill menstrual cycle reported performing fewer
& Palmer, 2000). For example, reproductive- high-risk behaviors. There was no difference in
age women are hypothesized to experience the likelihood of performing low-risk behaviors
more psychological pain due to the greater risk between women in their fertile phase and
of conception. Thornhill and Thornhill (1990a) women outside their fertile phase. This research
documented support for this hypothesis, docu- has some methodological problems that prevent
menting that reproductive-age women are more firm conclusions, however. First, the research-
traumatized by rape than are postreproductive- ers used only one method (i.e., the forward-
age women or prereproductive-age girls. cycle method) to assess women’s menstrual sta-
The research conducted by Thornhill and tus. Also, Chavanne and Gallup do not specify
Thornhill focuses on the aftereffects of being how the inventory of risky behaviors was de-
raped and on the psychological pain that may veloped, noting only that a preliminary sample
motivate women to avoid the circumstances of women rated the riskiness of the behaviors.
leading to the rape. Very little research, how- In addition, the dependent variable may be con-
ever, has been conducted to identify the specific founded with diversity of activity. For example,
behaviors women may deploy to avoid being a woman who performed 10 nonrisky behaviors
raped. (each scored as a 1 on the riskiness scale) could
Petralia and Gallup (2002) examined whether receive the same score as a woman who per-
a woman’s capacity to resist rape varies across formed two high-risk behaviors (each scored as
the menstrual cycle. Women in the fertile phase a 5 on the riskiness scale; see Bröder and
of their menstrual cycle showed an increase in Hohmann, 2003, for discussion). Despite these
handgrip strength, but only when presented with methodological issues, this research docu-
a sexual coercion scenario. Women not in their mented a significant decrease in performance of
fertile phase did not show an increase in hand- risky behaviors by women in the fertile phase of
grip strength. Furthermore, women in all other their menstrual cycle. This evidence is consis-
conditions, including women in the fertile phase tent with the hypothesized function of rape-
who were presented with the neutral control avoidance mechanisms, particularly when
scenario, showed a decrease in handgrip women are fertile.
strength posttest. This provides evidence for The Chavanne and Gallup (1998) study was
specialized mechanisms designed to motivate replicated by Bröder and Hohmann (2003) us-
women to behave in ways that cause them to be ing a within-subjects design. Twenty-six
less likely to be raped. Women who experience women who did not currently use oral contra-
increased strength during their fertile phase ceptives were tested weekly for 4 successive
would be better equipped to defend themselves weeks. The results indicated that women in the
from would-be rapists. The research by Petralia fertile phase of their cycle selectively inhibit
and Gallup (2002) provides evidence consistent behaviors that would expose them to a higher
with the hypothesis that women have evolved risk of being raped, despite performing more
mechanisms that motivate rape avoidance be- nonrisky behaviors. These results provide a
haviors. conceptual replication of the results reported by
Chavanne and Gallup (1998) investigated the Chavanne and Gallup. Women perform fewer
performance of risky behaviors by women in risky behaviors when they are fertile, while still
the fertile phase of their menstrual cycles. A demonstrating a higher overall activity level
sample of women were asked where they were (Morris & Udry, 1970). This selective behavior
in their menstrual cycles, and to indicate indicates that women may have evolved spe-
whether they had performed a range of behav- cialized psychological mechanisms designed to
iors in the past 24 hrs. Behaviors were ranked motivate behaviors that decrease the risk of
by women in a separate study according to how being raped. Although this study addressed
likely performing the behaviors might result in many of the issues in the Chavanne and Gallup
94 MCKIBBIN ET AL.

research, there is still no indication of how risky plore these effects in greater detail. For exam-
behaviors were identified. This study also used ple, researchers might ask women to rate the
the somewhat problematic forward and reverse- coerciveness of familiar faces of classmates or
cycle counting methods for identifying the fer- celebrities.
tile phase of the menstrual cycle, both of which In summary, limited previous work suggests
depend on the potentially unreliable self-reports that women may have evolved psychological
of participants (Bröder & Hohmann, 2003). mechanisms that motivate them to avoid being
A recent study by Garver-Apgar, Gangestad, raped. These studies have not assessed specific
and Simpson (2007) tested the hypothesis that behaviors performed to avoid rape. Rather, the
women are more attuned to signs of a man’s results of these studies suggest that women may
potential sexual coerciveness during the fertile have evolved mechanisms that motivate them to
phase and are more accurate at detecting sexu- assess the risk of sexual coercion, such as the
ally coercive men during the fertile phase. A riskiness of walking in a dark parking lot alone,
sample of 169 normally ovulating women and the coerciveness of a particular man.
watched short segments of videotaped inter-
views of men. The women were then asked to Concluding Comments
rate the men on several items that were summed
to create an overall coerciveness score. Average Evolutionary psychology is a powerful heu-
coerciveness scores for each man were com- ristic tool that allows researchers to consider
puted. Finally, women’s menstrual status was rape in a new light. Researchers have argued
estimated using the reverse-cycle counting that men have evolved psychological mecha-
method. The results indicated that women in the nisms that motivate them to rape in specific
fertile phase of their menstrual cycle rated the contexts. Although some evidence is consistent
men as more sexually coercive. This suggests with this hypothesis, more research must be
that women at greater risk of conception may be conducted before we can conclude that men
more attuned to signs of male sexual coercive- have specific adaptations for rape. We propose
ness than women at lesser risk of conception. that a more nuanced view of rapists is needed,
This may represent an evolved cognitive error in which rapists may be thought of as belonging
management bias (see Haselton, Nettle, & An- to one of several types distinguished by the
drews, 2005, for an overview) towards identi- contexts in which they are predicted to commit
fying men as sexually coercive, which might a rape. Researchers also have hypothesized that
serve to protect women from being raped. This women have evolved mechanisms that motivate
research provides more evidence that women behaviors to avoid being raped. Some evidence
may have evolved psychological mechanisms supports this hypothesis. Researchers also must
that motivate behaviors that guard against continue to investigate women’s evolved rape-
men’s sexual coercion and rape. We note, how- avoidance mechanisms before generating con-
ever, that the participants viewed videos of clusions. Future research should continue to in-
strangers. Studies demonstrate that women have vestigate the psychological mechanisms that
a greater fear of stranger rape than of being may motivate men’s rape behavior and wom-
raped by someone they know (Thornhill & en’s rape-avoidance behavior. It is our hope that
Thornhill, 1990b), which suggests that stranger a deeper understanding of the causes of rape
rape was the greater adaptive problem. This is will aid in its prevention.
despite modern patterns of rape, which indicate
that women are more likely to be raped by
someone they know (Kilpatrick et al., 1992; References
Resnick, Kilpatrick, Dansky, Saunders, & Best,
1993). These results may reflect the greater po- Baker, R. R., & Bellis, M. A. (1989). Number of
sperm in human ejaculates varies in accordance
tential costs associated with stranger rape, such with sperm competition theory. Animal Behav-
as a decreased likelihood of investment by the iour, 37, 867– 869.
genetic father of resulting offspring. Would Baker, R. R., & Bellis, M. A. (1993). Human sperm
similar results be found by testing women’s competition: Ejaculate adjustment by males and
coerciveness ratings of acquaintances or other the function of masturbation. Animal Behav-
familiar men? Future research is needed to ex- iour, 46, 861– 885.
EVOLUTIONARY PERSPECTIVE ON RAPE 95

Bergen, R. K. (1996). Wife rape: Understanding the ity, and fluctuating asymmetry. Ethology and So-
response of survivors and service providers. In C. ciobiology, 15, 73– 85.
Renzetti & J. Edleson (Series Eds.), Sage series on Garver-Apgar, C. E., Gangestad, S. W., & Simpson,
violence against women. California: Sage. J. A. (2007). Women’s perceptions of men’s sex-
Bergen, R. K., & Bukovec, P. (2006). Men and ual coerciveness change across the menstrual cy-
intimate partner rape: Characteristics of men who cle. Acta Psychologica Sinica, 39, 536 –540.
sexually abuse their partner. Journal of Interper- Ghiglieri, M. P. (2000). The dark side of man. New
sonal Violence, 21, 1375–1384. York: Perseus Books.
Bröder, A., & Hohmann, N. (2003). Variations in Glick, P., & Fiske, S. T. (1996). The Ambivalent
risk-taking behavior over the menstrual cycle: An Sexism Inventory: Differentiating hostile and be-
improved replication. Evolution and Human Be- nevolent sexism. Journal of Personality and Social
havior, 24, 391–398. Psychology, 70, 491–512.
Burch, R. L., & Gallup, G. G. (2006). The psycho- Goetz, A. T., & Shackelford, T. K. (2006). Sexual
biology of human semen. In S. M. Platek & T. K. coercion and forced in-pair copulation in humans
Shackelford (Eds.), Female infidelity and paternal as sperm competition tactics in humans. Human
uncertainty (pp. 141–172). New York: Cambridge Nature, 17, 265–282.
University Press. Gottschall, J. (2004). Explaining wartime rape. The
Buss, D. M. (1994). The strategies of human mating. Journal of Sex Research, 41, 129 –136.
American Scientist, 82, 238 –249. Gottschall, J. A., & Gottschall, T. A. (2003). Are
Buss, D. M. (2004). Evolutionary psychology (2nd per-incident rape-pregnancy rates higher than per-
ed.). Boston: Allyn & Bacon. incident consensual pregnancy rates? Human Na-
Chavanne, T. J., & Gallup, G. G. (1998). Variation in ture, 14, 1–20.
risk taking behavior among female college stu- Gowaty, P. A., & Buschhaus, N. (1998). Ultimate
dents as a function of the menstrual cycle. Evolu- causation of aggressive and forced copulation in
tion and Human Behavior, 19, 27–32. birds: Female resistance, the CODE hypothesis,
and social monogamy. Integrative and Compara-
Christopher, F. S., Owens, L. A., & Stecker, H. L.
tive Biology, 38, 207–225.
(1993). Exploring the dark side of courtship: A test
Greenfield, L. (1997). Sex offenses and offenders.
of a model of male premarital sexual aggressive-
Washington, DC: Bureau of Justice Statistics, U.S.
ness. Journal of Marriage and the Family, 55,
Department of Justice.
469 – 479.
Hall, G. C. N., Shondrick, D. D., & Hirschman, R.
Davis, J. A., & Gallup, G. G., Jr. (2006). Preeclamp-
(1993). The role of sexual arousal in sexually
sia and other pregnancy complications as an adap-
aggressive behavior: A meta-analysis. Journal of
tive response to unfamiliar semen. In S. M. Platek Consulting and Clinical Psychology, 61, 1091–
& T. K. Shackelford (Eds.), Female infidelity and 1095.
paternal uncertainty (pp. 191–204). New York: Haselton, M. G., Nettle, D., & Andrews, P. W.
Cambridge University Press. (2005). The evolution of cognitive bias. In D. M.
Dean, K. E., & Malamuth, N. M. (1997). Character- Buss (Ed.), The handbook of evolutionary psychol-
istics of men who aggress sexually and men who ogy (pp. 724 –746). Hoboken, NJ: Wiley.
imagine aggressing: Risk and moderating vari- Holmes, M. M., Resnick, H. S., Kilpatrick, D. G., &
ables. Journal of Personality and Social Psychol- Best, C. L. (1996). Rape-related pregnancy: Esti-
ogy, 72, 449 – 455. mates and descriptive characteristics from a na-
Figueredo, J., Corral-Verdugo, V., Frias-Armenta, tional sample of women. American Journal of Ob-
M., Bachar, K. J., White, J., McNeill, P. L., stetrics and Gynecology, 175, 320 –324.
Kirsner, B. R., & del PilarCastell-Ruiz, I. (2001). Hong, L. K. (1984). Survival of the fastest: On the
Blood, solidarity, status, and honor: The sexual origin of premature ejaculation. The Journal of Sex
balance of power and spousal abuse in Sonora, Research, 20, 109 –122.
Mexico. Evolution and Human Behavior, 22, 293– Kalichman, S. C., Williams, E. A., Cherry, C.,
328. Belcher, L., & Nachimson, D. (1998). Sexual co-
Gallup, G. G., Jr., & Burch, R. L. (2004). Semen ercion, domestic violence, and negotiating condom
displacement as a sperm competition strategy in use among low-income African American women.
humans. Evolutionary Psychology, 2, 12–23. Journal of Women’s Health, 7, 371–378.
Gangestad, S. W., & Thornhill, R. (1999). Individual Kanin, E. J. (1957). Male aggression in dating–
differences in developmental precision and fluctu- courtship relations. American Journal of Sociol-
ating asymmetry: A model and its implications. ogy, 63, 197–204.
Journal of Evolutionary Biology, 12, 402– 416. Kilgallon, S. J., & Simmons, L. W. (2005). Image
Gangestad, S. W., Thornhill, R., & Yeo, R. A. content influences men’s semen quality. Biology
(1994). Facial attractiveness, developmental stabil- Letters, 1, 253–255.
96 MCKIBBIN ET AL.

Kilpatrick, D., Edmunds, C., & Seymour, A. (1992). Palmer, C. T. (1989). Is rape a cultural universal? A
Rape in America. Arlington, VA: National Victim re-examination of the ethnographic evidence. Eth-
Center. nology, 28, 1–16.
Krill, A. L., Lake, T. M., & Platek, S. M. (2006, Palmer, C. T., & Thornhill, R. (2003a). Straw men
June). Do “good genes” predict forced copula- and fairy tales: Evaluating reactions to A natural
tion? A test of whether facial symmetry is related history of rape. The Journal of Sex Research, 40,
to sexual battery. Poster presented at the annual 249 –255.
meeting of the Human Behavior and Evolution Palmer, C. T., & Thornhill, R. (2003b). A posse of
Society, Philadelphia, PA. good citizens bring outlaw evolutionists to justice.
Krueger, M. M. (1988). Pregnancy as a result of rape. A response to Evolution, gender, and rape. Edited
Journal of Sex Education and Therapy, 14, 23–27. by Cheryl Brown Travis. (2003). Cambridge, MA:
Lalumiére, M. L., Chalmers, L. J., Quinsey, V. L., & MIT Press. Evolutionary Psychology, 1, 10 –27.
Seto, M. C. (1996). A test of the mate deprivation Parker, G. A. (1982). Why are there so many tiny
hypothesis of sexual coercion. Ethology and So- sperm? Sperm competition and the maintenance of
ciobiology, 17, 299 –318. two sexes. Journal of Theoretical Biology, 96,
Lalumiére, M. L., Harris, G. T., Quinsey, V. L., & 281–294.
Rice, M. E. (2005). The causes of rape. Washing- Parker, G. A. (1970). Sperm competition and its
ton, DC: American Psychological Association evolutionary consequences in the insects. Biolog-
Press. ical Reviews, 45, 525–567.
Lalumiére, M. L., Harris, G. T., & Rice, M. E. Petralia, S. M., & Gallup, G. G. (2002). Effects of a
(2001). Psychopathy and developmental instabil- sexual assault scenario on handgrip strength across
ity. Evolution and Human Behavior, 22, 75–92. the menstrual cycle. Evolution and Human Behav-
Lalumiére, M. L., & Quinsey, V. L. (1994). The ior, 23, 3–10.
discriminability of rapists from non-sex offenders Pizzari, T., & Birkhead, T. R. (2000). Female feral
using phallometric measures: A meta-analysis. fowl eject sperm of subdominant males. Nature,
Criminal Justice and Behavior, 21, 150 –175.
405, 787–789.
Lalumiére, M. L., & Quinsey, V. L. (1996). Sexual
Plath, M., Parzefall, J., & Schlupp, I. (2003). The role
deviance, antisociality, mating effort, and the use
of sexual harassment in cave and surface dwelling
of sexually coercive behaviors. Personality and
populations of the Atlantic molly, Poecilia mexi-
Individual Differences, 21, 33– 48.
cana (Poeciliidae, Teleostei). Behavioral Ecology
Linder, J. E., & Rice, W. R. (2005). Natural selection
and Sociobiology, 54, 303–309.
and genetic variation for female resistance to harm
from males. Journal of Evolutionary Biology, 18, Pound, N. (2002). Male interest in visual cues of
568 –575. sperm competition risk. Evolution and Human Be-
Lonsway, K. A., & Fitzgerald, L. F. (1995). Attitu- havior, 23, 443– 466.
dinal antecedents of rape myth acceptance: A the- Pound, N., Shackelford, T. K., & Goetz, A. T. (2006).
oretical and empirical reexamination. Journal of Sperm competition in humans. In T. K. Shackel-
Personality and Social Psychology, 68, 704 –711. ford & N. Pound (Eds.), Sperm competition in
Magurran, A. E. (2001). Sexual conflict and evolu- humans (pp. 3–31). New York: Springer.
tion in Trinidadian guppies. Genetica, 112–113, Resnick, H. S., Kilpatrick, D. G., Dansky, B. S.,
463– 474. Saunders, B. E., & Best, C. L. (1993). Prevalence
Malamuth, N. M., Huppin, M., & Paul, B. (2005). of civilian trauma and post-traumatic stress disor-
Sexual coercion. In D. M. Buss (Ed.), The hand- der in a representative national sample of women.
book of evolutionary psychology (pp. 394 – 418). Journal of Consulting and Clinical Psychology,
Hoboken, NJ: Wiley. 61, 984 –991.
Maynard Smith, J. (1997). Commentary. In P. Go- Reyer, H. U., Frei, G., & Som, C. (1999). Cryptic
waty (Ed.), Feminism and evolutionary biology (p. female choice: Frogs reduce clutch size when am-
522). New York: Chapman & Hall.. plexed by undesired males. Proceedings of the
Mitani, J. C. (1985). Mating behavior of male oran- Royal Society of London Series B: Biological Sci-
gutans in the Kutai Reserve. Animal Behav- ences, 266, 2101.
iour, 33, 392– 402. Rozée, P. D. (1993). Forbidden or forgiven? Rape in
Morris, N. M., & Udry, J. R. (1970). Variations in cross-cultural perspective. Psychology of Women
pedometer activity during the menstrual cycle. Quarterly, 17, 499 –514.
Sensory Processing, 2, 90 –98. Russell, D. E. H. (1990). Rape in marriage (Rev.
Muehlenhard, C. L., & Linton, M. A. (1987). Date ed.). Indianapolis: Indiana University Press.San-
rape and sexual aggression in dating situations: day, P. R. (1981). The socio-cultural context of
Incidence and risk factors. Journal of Counseling rape: A cross-cultural study. Journal of Social
Psychology, 34, 186 –196. Issues, 37, 5–27.
EVOLUTIONARY PERSPECTIVE ON RAPE 97

Shackelford, T. K. (2002). Are young women the productive efforts. Annual Review of Ecology and
special targets of rape-murder? Aggressive Behav- Systematics, 12, 355–386.
ior, 28, 224 –232. Thornhill, R. (1999). The biology of human rape.
Shackelford, T. K., Goetz, A. T., McKibbin, W. F., & Jurimetrics Journal, 39, 137–147.
Starratt, V. G. (2007). Absence makes the adapta- Thornhill, R., & Palmer, C. P. (2000). A natural
tions grow fonder: Proportion of time apart from history of rape. Cambridge, MA: The MIT Press.
partner, male sexual psychology, and sperm com- Thornhill, R., & Thornhill, N. (1983). Human rape:
petition in humans (Homo sapiens). Journal of An evolutionary analysis. Ethology and Sociobiol-
Comparative Psychology, 121, 214 –220. ogy, 4, 137–173.
Shackelford, T. K., & Larsen, R. J. (1997). Facial Thornhill, N., & Thornhill, R. (1990a). Evolutionary
asymmetry as indicator of psychological, emo- analysis of psychological pain of rape victims I:
tional and physiological distress. Journal of Per- The effects of victim’s age and marital status.
sonality and Social Psychology, 72, 456 – 466. Ethology and Sociobiology, 11, 155–176.
Shackelford, T. K., LeBlanc, G. J., Weekes- Thornhill, N., & Thornhill, R. (1990b). Evolutionary
Shackelford, V. A., Bleske-Rechek, A. L., Euler, analysis of psychological pain following rape II:
H. A., & Hoier, S. (2002). Psychological adapta- The effects of stranger, friend, and family member
tion to sperm competition. Evolution and Human offenders. Ethology and Sociobiology, 11, 177–
Behavior, 23, 123–138. 193.
Shields, W. M., & Shields, L. M. (1983). Forcible Thornhill, N., & Thornhill, R. (1990c). Evolutionary
rape: An evolutionary perspective. Ethology and analysis of psychological pain following rape vic-
Sociobiology, 4, 115–136. tims III: The effects of force and violence. Aggres-
Shine, R., Langkilde, T., & Mason, R. T. (2003). sive Behavior, 16, 297–320.
Cryptic forcible insemination: Male snakes exploit Thornhill, R., & Thornhill, N. (1992). The evolution-
female physiology, anatomy, and behavior to ob- ary psychology of men’s coercive sexuality. Be-
tain coercive matings. American Naturalist, 162, havioral and Brain Sciences, 15, 363–375.
653– 667. Thornhill, R., & Sauer, K. (1991). The notal organ of
Smuts, B. B. (1992). Male aggression against the scorpionfly (Panorpa vulgaris): An adaptation
women. Human Nature, 6, 1–32. to coerce mating duration. Behavioral Ecology, 2,
Smuts, B. B., & Smuts, R. W. (1993). Male aggres- 156 –164.
sion and sexual coercion of females in nonhuman Tooby, J., & Cosmides, L. (2005). Conceptual foun-
primates and other mammals: Evidence and theo- dations of evolutionary psychology. In D. M. Buss
retical implications. Advances in the Study of Be- (Ed.), The handbook of evolutionary psychology
havior, 22, 1– 63. (pp. 5– 67). Hoboken, NJ: Wiley.
Symons, D. (1979). The evolution of human sexual- Watts, C., Keogh, E., Ndlovu, M., & Kwaramba, R.
ity. New York: Oxford University Press. (1998). Withholding of sex and forced sex: Dimen-
Tang-Martinez, Z. (1997). The curious courtship of sions of violence against Zimbabwean women.
sociobiology and feminism: A case of irreconcil- Reproductive Health Matters, 6, 57– 65.
able differences. In P. Gowaty (Ed.), Feminism Wilson, M., & Mesnick, S. L. (1997). An empirical
and evolutionary biology (pp. 116 –150). New test of the bodyguard hypothesis. In P. A. Gowaty
York: Chapman & Hall. (Ed.), Feminism and evolutionary biology (pp.
Thornhill, R. (1980). Rape in Panorpa scorpionflies 505–511). New York: Chapman & Hall.
and a general rape hypothesis. Animal Behav- Wrangham, R., & Peterson, D. (1996). Demonic
ior, 28, 52–59. males. New York: Houghton Mifflin.
Thornhill, R. (1981). Panorpa (Mecoptera: Panor-
pidea) scorpionflies: Systems for understanding re- Received June 10, 2007
source-defense polygyny and alternative male re- Accepted August 18, 2007 !

You might also like