Gender differences in erotic plasticity: The female sex drive as socially flexible and responsive

Psychological Bulletin 2000, Vol. 126, No. 3, 347-374 Copyright 2000 by the American Psychological Association, Inc. 0033-2909/0045.00 DOI: 10.1037//0033-2909.126.3.347 Gender Differences in Erotic Plasticity: The Female Sex Drive as Socially Flexible and Responsive Roy F. Baumeister Case Western Reserve University Responding to controversies about the balance between nature and culture in determining human sexuality, the author proposes that the female sex drive is more malleable than the male in response to sociocultural and situational factors. A large assortment of evidence supports 3 predictions based on the hypothesis of female erotic plasticity: (a) Individual women will exhibit more variation across time than men in sexual behavior, (b) female sexuality will exhibit larger effects than male in response to most specific sociocultural variables, and (c) sexual attitude-behavior consistency will be lower for women than men. Several possible explanations for female erotic plasticity are reviewed, including adaptation to superior male political and physical power, the centrality of female change (from no to yes) as a prerequisite for intercourse, and the idea that women have a milder sex drive than men. of its efforts to exploit and subjugate women (see, e.g., Kitzinger, Sex and mating seem to be accomplished in a fairly straightforward, predictable, even routine manner in many species of ani- 1987). Although social constructionists do not deny that there may mals. Human sexuality, in contrast, has long been recognized as a be certain biological foundations to sexuality, they emphasize rich, confusing tangle, in which biological drives, sociocultural culture and social influence as the decisive factors in explaining meanings, formative individual experiences, and additional un- human sexuality. known factors play powerful roles. Among the most basic unre- Essentialist theories, in contrast, propose that there are true and solved questions about human sexuality is that of the relative definite forms of sexuality that remain constant, even though situa- contributions of nature and culture: Does sexual response depend tional factors may occasionally interfere with or shape their expres- primarily on sociocultural factors such as meanings, context, re- sion. As DeLamater and Hyde (1998) emphasized, evolutionary and lationship status, communication, norms, and rules—or is it sociobiological analyses of sexuality fall in this category, for they mainly determined by hormones, genes, and other biological pro- explain sexuality in terms of innate motivational patterns that have cesses? Even in recent decades, theories about human sexual desire evolved to suit the reproductive contingencies of males and females have differed radically in their relative emphasis on nature and so as to maximize the passing on of each person's genes (see, e.g., culture. To be sure, hardly any theorist goes to the extreme of Buss & Schmitt, 1993). Some of these theories treat culture as a insisting that either nature or culture is totally responsible for system adapted to accommodate the innate biological patterns (see, determining the human sex drive, but the compromise formula- e.g., Symons, 1995). hi any case, biology, not culture, is featured as tions differ widely in their relative emphasis. the main source of causal explanations. The two most influential theories about sexuality have been the The present article offers yet another conceptualization of the social constructionist and the essentialist (DeLamater & Hyde, relative contributions of nature and culture to human sexual desire. 1998). Social constructionist theories have regarded human sexual The point of departure is that there is no single correct answer that desire as shaped extensively by culture and socialization, often holds true for all human beings. Instead, I suggest that female sexu- mediated by language as an ordering principle that is shared in ality, as compared with male, is more subject to the influence of common with other people. These theorists emphasize cross- cultural, social factors. Although male sexuality must frequently make cultural variation to argue for the cultural relativity of sexual desire concessions to opportunity and other external constraints, male desire (see, e.g., Staples, 1973). Who does what to whom sexually is is depicted here as relatively constant and unchanging, which suggests regarded as a product of cultural rules and individual, linguistically a powerful role for relatively rigid, innate determinants. Female sex- mediated decisions rather than as a biological imperative. Social uality, in contrast, is depicted as fairly malleable and mutable: It is constructionist theories have also been invoked by feminists to responsive to culture, learning, and social circumstances. The plastic- depict human sexual desire as shaped by patriarchal society as part ity of the female sex drive offers greater capacity to adapt to changing external circumstances as well as an opportunity for culture to exert a controlling influence. Prom the global perspective of the broader society, if controlling people's behavior is the goal, women's sexual F thank Kathleen R. Catanese for assistance and Keith Campbell, Julie Exline, Catrin Finkenauer, Astrid Schutz, and Dianne Tice for helpful comments and suggestions. Correspondence concerning this article should be addressed to Roy F. Baumeister, Department of Psychology, Case Western Reserve University, Cleveland, Ohio 44106-7123. Electronic mail may be sent to patterns are more easily changed than men's. rfb2@po.cwru.edu. whether biological or social, in order to reserve the term sex for Definitions I use the term gender to refer to maleness and femaleness, 347 348 BAUMEISTER activities leading to orgasm or genital arousal. The term erotic easier to talk a woman into doing something sexual that she does plasticity is used to refer to the degree to which a person's sex drive can be shaped and altered by cultural and social factors, from not really want to do or something that is not good for her, as compared with talking a man into doing something that is com- formal socialization to situational pressures. Thus, high erotic parably contrary to his wishes and needs. The present hypothesis plasticity entails being subject to situational, social, or cultural influence regarding what types of partners and what types of has to do with receptivity to influence, and being receptive to sexual activities one would desire and enjoy. Desiring to perform influence can under some circumstances take on a negative tone (e.g., gullibility). the same act with a new partner does not necessarily constitute Neither of these value-linked effects is likely to be widespread plasticity, for it is quite possible to have a stable, consistent desire to perform certain acts with many different partners. or powerful. Hence, erotic plasticity should not be invoked to argue for the superiority of either gender. The "sex drive" is a hypothetical construct, and research studies actually measure attitudes, behavior, and desire. The term attitude Empirical Predictions is used here to refer to general opinions and abstract rules that encompass broad categories and multiple situations. Desire refers The hypothesis of differential erotic plasticity permits empirical to situation-specific feelings of sexual arousal and wanting to predictions. A first, basic prediction is that intraindividual varia- engage in particular acts with particular partners. Behavior refers tion (i.e., within-person variance) in sexual behavior will be to what the person actually does, such as physically engaging in greater among women than men. If women are malleable in particular sex acts. Desire may contradict attitudes, such as when a person feels an urge to have sex with a partner who is regarded response to situational and social factors, then as a woman moves from one situation to another, her sexual desires and behaviors as off-limits. Behavior can contradict either desire or attitude, such may be subject to change. The lesser (hypothesized) flexibility of as when a person refrains from much-wanted sex or has intercourse with a forbidden partner. men would mean that male sexual patterns will remain more stable Theory: Differential Plasticity and constant across time and across different situations. (Lack of opportunity may be an exception: A man's sexual behavior may depend on whether he can find a willing partner.) Physical changes, such as ill health or major hormonal changes, might well The central idea of this article is that the female sex drive is have a strong effect. But as regards changing social situations and more malleable than the male, indicating higher average erotic different life circumstances, the average man's desires should remain more stable and constant than the average woman's. plasticity. More precisely, female sexual responses and sexual behaviors are shaped by cultural, social, and situational factors to This theory does not extend to making predictions about inter- a greater extent than male. Plasticity could be manifested through individual variations, because these could well depend on innately changes in what is desired (e.g., type of partner, type of activity), in degree of desire (e.g., preferred frequency of sex, degree of or genetically prepared patterns. The men in a given culture may collectively have more variations in their individual sexual appe- variety), or in expression of desire (e.g., patterns of activity). tites than do the women without violating the hypothesis of female Changes in attitudes may contribute to these behavioral changes. plasticity. A familiar example of gender differences in interindi- Because debates about female sexuality are often perturbed by vidual variance in genetically influenced traits is found in research bitter conflicts based on implicit value judgments, it is important to on mental retardation and intelligence: The two genders have nearly identical mean IQ scores, but the males have higher vari- address the value question explicitly. Frankly, I see almost no reason to think that it is better or worse to have high erotic ance, therefore being proportionally overrepresented at both ex- plasticity, and so the present hypothesis does not entail that one gender is better (or better off) than the other in this regard. The tremes (Jensen, 1998; Lehrke, 1997; J. A. F. Roberts, 1945). Such difference may be important for predicting a variety of behavior patterns, attitudes, misunderstandings, and conflicts, but there is no predictions about interindividual variance (although evidence about paraphilias are considered briefly among the possible limi- inherent moral or practical superiority on either side. tations and counterexamples). The present hypothesis concerns patterns are plausible with sexuality, too, and I am not making There are two small exceptions to the value-free tone of my only intraindividual variance: Once a man's sexual tastes emerge, hypothesis. That is, two small value judgments could be made, and they are less susceptible to change or adaptation than a woman's. they point in opposite directions. The first is that it is generally better to be flexible because one can adapt more readily to changing circumstances. The capacity to change is inherently adaptive, and being adaptive is good. In this respect, women may be better off than men if the present hypothesis is correct because their sexuality can adjust more easily and readily. Thus, if changes in social circumstances place equal demands for adjustment on males A second prediction is that specific sociocultural factors will have a greater impact on women's sexuality than on men's. To put this prediction in more precise, statistical terms, the sociocultural variables will have bigger effect sizes in predicting responses of women than of men. Thus, women will vary more than men from one culture to anomer and from one historical period to another. and females, the females will be more successful than the males at Socializing institutions, such as schools and churches, should produce bigger changes in women than in men with regard to making these adjustments, or they will be able to achieve that sexual behavior. success with less difficulty. The other exception is that higher erotic plasticity may render a person more vulnerable to external influences, with the resulting A final prediction is that attitude-behavior consistency (with regard to sex) will be lower among women than men. If female sexual response is malleable by situational and social factors, then possibility that one could end up being influenced to do things that are not in one's best long-term interests. In simple terms, it may be a woman's behavior cannot be easily predicted by her attitudes (especially general, abstract attitudes). In simple terms, her sexual EROTIC PLASTICITY 349 responses depend more on external context than on internal fac- channel, or transform than a powerful one. Women could thus tors, relative to those of males, and so her attitudes are less likely more easily be persuaded to accept substitutes or alternate forms of to determine her behavior. She may, for example, hold an attitude satisfaction, as compared with men, if women's overall sexual in favor of using condoms or against anal intercourse, but situa- desires are milder. tional factors may intrude to cause her to act contrary to those attitudes under some circumstances (and even to desire such Proximal Sources of Plasticity attitude-contrary acts). This prediction is methodologically a useful complement to the first one because it avoids the confound that These root causes may be translated into the actual degree of data on women are somehow simply more conclusive or reliable behavioral plasticity of living individuals either through innate, than data on men. The erotic plasticity hypothesis predicts that genetic patterns or by social learning processes and personal cultural and social factors will show higher correlations with experiences (even conscious adaptations). The nature of the me- sexual responses of women than men—whereas attitude-behavior diating, proximal causes is not easily resolved, but a few specu- correlations will be lower for women than for men. lative suggestions may be offered. How, then, is erotic plasticity actually instilled? Reasons for Plasticity The possibility that it is biologically based must be considered. Many sexuality-based traits are supposedly genetically prepared Why should women have more erotic plasticity than men? I by the X chromosome, of which women have two and men only have three different hypotheses, each of which could offer some one. Having two different sets of relevant genes could allow for potential insight into the gender difference in erotic plasticity. greater flexibility than having only one. Specifically, the two X The first is based on the difference in power. On average, men chromosomes could carry different prescriptions for behavior, and are physically stronger and more aggressive than women, and they hence it would be up to the environment to determine which one also tend to hold greater sociopolitical and economic power. If two would prevail. Males, in contrast, would receive a single and partners' sexual wishes were to differ, the man would have several unambiguous genetic program, leaving less opportunity for the advantages over the woman for getting his way. Greater flexibility environmental influence. on the part of women would be one adaptive response to the Hormone levels provide another plausible basis for differential standard problem of bonding with someone who would be able to plasticity. Research has generally found that testosterone is the impose his desires by means of physical coercion or social power, single hormone that has the greatest effects on sexual behavior in should that ever become necessary (as he saw it). Biologists and both males and females. Because males have substantially more evolutionary psychologists believe that the relative superiority of testosterone than females, male behavior may be more subject to male physical power is strongly linked to male reproductive pat- its causal influence than female behavior. (On the other hand, terns and goals (such as male competition under circumstances of female receptors may be more sensitive to testosterone than are extreme polygyny; Gould & Gould, 1997; Ridley, 1993), and those of males, which could offset the difference in quantity of the feminists emphasize that male political power shapes the sexual hormone.) interactions between the sexes and results in the cultural suppres- In another relevant line of argument, T. Roberts and Pennebaker sion of female sexuality. The present suggestion could be seen as (1995; also Pennebaker & Roberts, 1992) have concluded that men another such process, hi which women became socially malleable are generally better than women at perceiving and detecting their as an adaptation to male power. inner bodily states. They noted that in socially impoverished The second is that flexibility may be an inherent requirement of environments such as laboratories and hospitals, males consis- the female role in sex. The simplest version of this would empha- tently outperform females at estimating their own bodily reactions size that most societies (including other species similar to humans) such as blood pressure, heartbeat, stomach contractions, respira- limit sexual activity by having the female refuse sexual offers and tory resistance, finger temperature, and blood glucose levels. This advances from most males. Of course, if females refused all male gender difference disappears when measure^ are taken hi natural- advances, the species would fail to reproduce. Women are negative istic and meaning-rich settings, in which multiple cues about toward most potential sex partners (i.e., most men) but occasion- sources of feelings are available. Roberts and Pennebaker pro- ally switch to positive. A negative response is the woman's default posed that men judge their emotional and arousal responses based option, as it were, hi practice, this entails that sex generally on direct detection of physiological cues, whereas women rely commences when the woman switches her initially negative stance more on social and situational cues to know how they respond. If to a positive one. That is, when a couple begins having sex, it is this is true generally for all emotions, it would presumably be even mainly because the woman has changed her decision: The woman stronger for sexual responses, because the signs of arousal are initially rejects the man's advances but later changes her vote from much more salient and unambiguous in the male than in the no to yes. The centrality of this change (from no to yes) hi female female. This argument could also explain why testosterone and sexuality requires each woman to have a certain degree of flexi- other inner, biochemical realities have stronger effects on male bility, and the broader patterns of erotic plasticity would follow than female sexuality: If men are more attuned to then: inner bodily from this foundation. Change requires changeability and hence states, then their level of testosterone would exert a stronger effect begets further change. on their behavior. The third possible explanation is based on differential drive Yet another possibility is that males have evolved to be more strength. This would invoke the politically unpopular but theoret- strongly driven by natural and genetic factors. Some authors have ically plausible view that women have a weaker sex drive than speculated that there may be a higher rate of mutations among men. A relatively weak motivation is presumably easier to redirect, males than females. One such speculation is that the Y chromo- 350 BAUMEISTER some (unique to males) might be a popular target of mutations. This dovetails well with a recent theory of sexual orientation put Nature may have targeted males and the Y chromosome for trying forward by Bern (1996, 1998). Bern rejected direct genetic influ- out new mutations because the greater reproductive variance ences on sexual orientation but suggested that genes may affect among males would give more opportunity for natural selection to temperament, which may in turn lead a young person to prefer operate (discussed by Kacelnik, 1999). The difference in repro- either males or females as friends and playmates. Later, the less ductive variance is well established. In human beings, for example, familiar gender creates arousal and thereby becomes the focus of most females produce at least one child, and hardly any woman has sexual attraction. Bern (1996,1998) has specifically suggested that more than 10 babies. In contrast, many men have zero offspring, his theory predicts that female sexual orientation will be more fluid and others exceed 10 by substantial amounts (Gould & Gould, and changeable than male, because little girls are more likely than 1997; Ridley, 1993). Thus, men exceed women at both extremes of little boys to have opposite-sex friends and playmates. Because reproductive outcomes (i.e., more at zero and more over 10). These "women actually grow up in a phenomenologically less gender- differences help determine how long natural selection takes to sort polarized culture than do men" (Bern, 1998, p. 398), men tend to out whether a particular mutation increases or decreases reproduc- be polarized into finding only males or only females sexually tive success. The relatively small variation in female reproductive appealing, whereas women's greater familiarity with both genders outcomes entails that many generations would be required for a enables them to be attracted to either or both. One can extend given mutant to prove itself better or worse than the original. In Bern's argument to propose that this greater bisexual orientation of contrast, a mutation in males might yield bigger effects within women will provide the foundation for other forms of plasticity fewer generations: An adaptive mutation might help a male pro- and change. This extension is similar to the argument I made duce dozens of offspring, and a maladaptive one would quickly be regarding change and the female sexual script, except that the eliminated from the gene pool. Because males thus make more cause of plasticity depends entirely on social factors and early efficient vehicles with which to select and evolve, therefore, male experiences, and any contribution by genetic factors is indirect. sexual behavior may have gradually become more encumbered with such biological influences. Other, more purely cultural arguments could be proposed to account for differential plasticity. These would suggest that culture Evolutionary arguments often invoke differential reproductive teaches men to obey their biological promptings but teaches goals for men and women (see, e.g., Buss & Schmidt, 1993; Gould women to ignore theirs and obey social prescriptions instead. & Gould, 1997; Ridley, 1993). Because women cannot have as These arguments seem relatively implausible in light of evidence many offspring as men, they are presumably more selective about that, throughout history, the prevailing stereotypes have regarded sex partners. Although one could argue a priori that the greater women as closer to nature than men and that in fact when society selectivity could lead to lower plasticity (because the woman can does try to change women's behavior it usually does so by telling ill afford to compromise or take chances), one might also suggest women what is allegedly in their biological nature, as opposed to that selectivity mandates a complex, careful decision process that teaching them to ignore their biological factuality (see, e.g., Mar- attends to subtle cues and contextual factors and that this very golis, 1984). Still, it is conceivable that new, more plausible complexity provides the basis for greater plasticity. versions of these explanations may be forthcoming. The biological and evolutionary arguments suggest searching for erotic plasticity in other species, which is beyond the expertise Causal Processes of the author and the scope of this review. Still, an important recent study by Kendriclc, Hinton, Atkins, Haupt, and Skinner (1998) is Last, it is helpful to consider the possible causal processes, even relevant. In an experimental design, newborn sheep and goats were though these extend the theoretical argument beyond what can be exchanged, so that the sheep were raised by goats and vice versa. tested against the currently available research literature. If the After they reached adulthood, they were reunited with then: bio- balance of natural versus cultural determinants of sexuality differs logical species, and their mating preferences were observed. Con- by gender, then the causal processes that direct sexual behavior are sistent with the hypothesis of female erotic plasticity, the adult also likely to differ. females were willing to mate with either species. The males, in Natural processes are typically mediated by biochemical pro- contrast, preferred only their adoptive species and refused to mate cesses. Hormones such as testosterone are likely to exert strong with their biological conspecifics, even after living exclusively and direct effects. Despite the fact that the exact processes leading with their own kind for 3 years. These results suggest that male from genes to behavior are not fully understood (although this field sexual inclinations are based on a process of sexual imprinting that is one in which substantial advances are anticipated in the next occurs early in life and then remains inflexible, whereas female decade), one assumes that biochemical factors play a crucial role sexual inclinations can continue to change in adulthood. in mediating such processes. In contrast, cultural processes are mediated by meanings, which The hypothesis that male sexuality is subject to an early imprinting process that is irreversible (as opposed to reversible in- is to say informational, symbolic concepts that can be expressed in fluences on female sexuality) suggests that both genetic prepara- language and communicated between group members. Norms, tion and early experiences are relevant. It qualifies the broad attitudes, rules, expectations, and relationship concepts provide hypothesis about greater female plasticity: Perhaps there is a stage contexts from which specific sexual acts and decisions can draw early in life during which male sexuality is highly receptive to meaning. Behavior depends on these meanings. social, environmental influences. After this imprinting, however, The hypothesized gender difference thus predicts that male male sexuality remains relatively rigid and inflexible, whereas sexuality will be shaped more than female sexuality by biochem- female sexuality retains plasticity throughout adolescence and adulthood. ical factors, including genetics and hormones. In contrast, female sexuality will be more meaning-driven than male sexuality, so that 351 EROTIC PLASTICITY context and interpretation shape women's sexual decision-making by multiple factors, I found it necessary to look elsewhere for (and other sexual responses) more than men's. evidence. Hypothesis Formation: The Sexual Revolution The present investigation was initially stimulated by a conclusion drawn by Ehrenieich, Hess, and Jacobs (1986) in their history of the sexual revolution in the United States, namely, that that revolution was mainly a change in women not men. Men's sexual desires and attitudes were pretty much the same after the sexual revolution as before it, although men had more opportunities for finding satisfaction afterward. It was women who changed fundamentally. Indeed, according to Rubin (1990), women changed several times, at first embracing a promiscuous enjoyment of casual sex like men, then shifting toward a more limited permissiveness that accepted sex in affectionate relationships but did not eagerly seek out sex with strangers (see also Robinson, Ziss, Ganza, Katz, & Robinson, 1991). The conclusion that the sexual revolution was primarily a change in female sexual attitudes and behaviors, rather than male, was made by other researchers beyond Ehrenreich et al. (1986). Arafat and Yorburg (1973) and Birenbaum (1970) had already made similar observations. Empirical studies, particularly those that surveyed the same type of sample (e.g., the same college campus) at repeated intervals, consistently found that women's attitudes and behaviors changed more than men's during the 1960s and early 1970s (Bauman & Wilson; 1974; Croake & James, 1973; DeLamater & MacCorquodale, 1979; Schmidt & Sigusch, 1972; Sherwin & Corbett, 1985; Staples, 1973), continuing even into the 1980s (Robinson et al., 1991). Well-constructed national surveys corroborated these conclusions by comparing older people, who had come of age before the sexual revolution, with younger people, whose sexual prime had occurred after the revolution, and these too found bigger differences in women than men (Laumann, Gagnon, Michael, & Michaels, 1994; Wilson, 1975). As one good example, Laumann et al. (1994) provided data on the proportion of respondents who had had five or more sex partners by the age of 30 (an age when most people have married and ceased accumulating new sex partners). For the oldest cohort, who came of age prior to the sexual revolution, 38% of men had had five or more sex partners by age 30, whereas for the younger cohort, the proportion increased slightly, to 49%. For women, the corresponding numbers are 2.6% and 22.4%. The sexual revolution thus increased men's likelihood of having many partners by 11 percentage points, or by about a fourth, whereas it multiplied women's likelihood by a factor of more than eight and by 20 points. Put another way, the sexual revolution produced a modest increase in the number of men having five or more sex partners, reflecting perhaps nothing more than increased opportunity, but it radically transformed many women's lives and created a large category of multipartnered women that had been almost nonexistent prior to that revolution (Laumann et al., 1994). The implication that women were changed more than men by the sexual revolution suggested the broader possibility that female sexuality is more historically malleable than male. The present investigation was spurred by this hypothesis. Given the difficulty of drawing firm conclusions about psychological principles from single historical events, especially highly complex ones influenced Evidence of Female Plasticity The method of reviewing the literature was as follows. I began with the most recently available volume of the Journal of Sex Research (1996 at the time) and worked backward to the first volume, reading all abstracts and all relevant articles. By covering the major journal in its entirety, I hoped to minimize the dangers of selective review and confirmation bias. The Archives of Sexual Behavior then received a similar treatment by a research assistant. These articles offered a useful starting point, and their reference lists were used to find further sources in other publications. The National Health and Social Life Survey (NHSLS; Laumann et al., 1994) was carefully scrutinized, inasmuch as it offers the most comprehensive and scientifically valid survey data (and indeed it is covered in a separate section). Additional sources were suggested by colleagues and by helpful reviewers of a previous draft, and more recently published work was added during revisions. Data on sexuality are often less than perfect, partly because of the ethical and practical difficulties of studying sex. A summarizing discussion of limitations in the data and general critique is provided after the evidence itself is presented. Alternative explanations are discussed at that point, but two of them deserve to be acknowledged at the outset. First, it is conceivable that there are more efforts to control female than male sexuality. This is not actually an alternative explanation in the usual sense, because it is fully compatible with the view of greater female plasticity. If female sexual behavior can be regulated more effectively than male sexuality, then it would make sense for society to focus its efforts on controlling females. Still, it is plausible that some findings regarding greater variation or causal impact among females could reflect variation in sociocultural controls rather than differential plasticity. The so-called double standard may be one example, if indeed it means that society permits or has permitted men to do things forbidden to women. Second, the findings regarding the power of specific sociocultural variables to change sexual behavior have to contend with different baselines in some cases. For example, if education increases the proportion of men who engage in some sexual practice from 70% to 80% while increasing the corresponding proportion of women from 30% to 80%, some readers might refuse to regard this as evidence of greater impact on women: It might be that the effect of education on men was limited by a ceiling effect. Intraindividual Variability The first major prediction is concerned with intraindividual variability. If erotic plasticity is greater among females, then women should show more variation across their individual sexual histories than men. The focus is on whether particular persons exhibit changes in their sexuality across time. One gender difference in intraindividual variability was noted by Kinsey, Pomeroy, Martin, and Gebhard (1953). Although their sampling has been criticized as not up to the best modem standards, that criticism is irrelevant to this finding, and their data on individual sexual histories are among the most detailed ever col- 352 BAUMEISTER lected. They found that some women, but hardly any men, showed changed along the same lines, and these partner reports con- patterns of substantial swings in degree of sexual activity. A firmed—significantly, this time—that the women had changed woman might go through a phase of having a great deal of sex, more than the men: 12% of the husbands, but only 6% of the then have no sexual activity of any sort for months, and then enter wives, reported that their spouse had changed a great deal. into another phase of having a great deal of sex. If a male were to The greater change by women than men in adapting to marriage experience a romantic break-up or a physical separation from his is especially remarkable given some other features of Ard's (1977) sex partner, he would tend to keep his orgasm rate constant by data. When asked about their current frequency of sexual activity resorting to masturbation or other activities, but women did not and their current preferences for frequency of sexual activity, the necessarily do this. "Discontinuities in total outlet are practically wives' answers indicated that their marital practices corresponded unknown in the histories of males," unlike females (Kinsey et al., almost precisely to the amount of sex they wanted, whereas the 1953, pp. 681-682). These discontinuities are thus an important men reported a significant gap between what they wanted and what confirmation of the hypothesis of female erotic plasticity. they were able to have. Thus, men were not getting what they Intraindividual change was the focus of an investigation by wanted on this important measure, whereas women were—yet still Adams and Turner (1985), who compared the reports of current the evidence showed that women had adapted more than men. sexual activity among an elderly sample (age 60-85) with the Possibly, women succeeded better than the men at adjusting their same people's retrospective reports of what they did in young expectations into line with what they were getting, which could be adulthood (age 20-30). Adams and Turner pointed out that most another manifestation of plasticity and would presumably be a studies of the effects of aging on sexuality simply emphasize very beneficial adaptation. reductions in drive and energy and hence decreased sexual activity, Marriage is certainly not the only type of relationship that can and so they looked especially for any signs of increasing activity. produce change in sexual attitudes. Harrison, Bennett, Globetti, Only a small minority of their sample showed increases on any of and Alsikafi (1974) found that women changed their sexual stan- the measures, but this minority was predominantly female. Thus, dards toward being more permissive as they accumulated dating one pattern of intraindividual change over several decades (in- experience. Men did not seem to change as a function of dating creasing sexual activity) was found mainly among females, and experience. One might have predicted that the necessity of com- this pattern is of particular interest because it is not confounded by promise would produce change in attitudes in both genders, but loss of vigor or declining health, which would make evidence of Harrison et al. found change only in the females. Reiss (1967) reduced sexual activity less relevant to the present theory. Some of Adams and Turner's (1985) most interesting data likewise found that women increased their sexual permissiveness after having steady dates or love relationships, whereas the effects concern masturbation. They found that in comparisons of young of such experiences on men were small and nonlinear. Reiss adulthood with old age, women showed remarkable, significant reported that 87% of the females, as opposed to 58% of the males, increases in masturbation (10% to 26%), whereas men showed a had come to accept sexual behavior that initially made them feel nonsignificant decrease over the same age span (32% to 26%). guilty. Of these, far more more females than males cited the Adams and Turner noted that their sample overrepresented married relationship with the opposite-sex partner as the key factor in women, so the change does not simply reflect a shift into mastur- bringing about this change. Thus, again, the data suggest greater bation as the women lost their partners. Even more important, sexual adaptation in relationship contexts by women. Adams and Turner reported that the old men who masturbated In the 1960s and 1970s, consensual extramarital sex increased, were typically continuing a pattern of masturbation that was and researchers were able to examine how people adapted to this present in young adulthood, whereas the women who masturbated unusual behavior (often called swinging). J. R. Smith and Smith in their 20s had typically discontinued that activity late in life. (1970) studied this phenomenon and concluded that "women are (Also, given the increase in overall numbers, the strong majority of better able to make the necessary adjustments to sexual freedom the women who masturbated in old age had not done so in their after the initial phases of involvement than are men" (p. 136). They 20s.) The authors concluded that the masturbation data showed noted that this greater adaptability of women was especially re- that "women displayed more plasticity in behavior than men" markable in light of the fact that it had generally been the men who (p. 134). initiated the involvement in swinging. Although Smith and Smith Undoubtedly, some degree of flexibility would be useful in failed to provide quantitative evidence to back their claim of the adapting to marriage because the requirement of coordinating superior adaptation of women, their observation is noteworthy one's sexual activities with a particular partner over a long period because it confirms one of the presumptive advantages of plastic- of time presumably requires some compromises unless the couple ity, namely, greater capacity to adapt to new circumstances. Some of the best and most useful data on intraindividual vari- is perfectly matched and their desires wax and wane in complete synchrony, which seems unlikely. Data on sexual changes in the ability concern sexual orientation and same-sex activity. Opera- adaptation to long-term marriages were provided by Ard (1977), tionally, this can be studied by investigating whether homosexual who, in a 20-year follow-up of a longitudinal study, asked the individuals have had heterosexual experience, which would sug- individuals who had remained married for over two decades how gest a higher degree of plasticity in their sexual orientation. Be- much they had changed from their early ideas, habits, and expec- ginning with Kinsey's research (Kinsey, Pomeroy, & Martin, tations regarding sex. Wives were somewhat more likely than 1948; Kinsey et al., 1953), many studies have found that lesbians husbands (13% to 9%) to claim that they had changed "a great are more likely to have had heterosexual intercourse than gay males. This effect is especially remarkable given the greater pro- deal," although this difference fell short of significance. Because some people might inflate their self-reported change in order to miscuity of males, although it might have something to do with the look good, Ard also asked people how much their partners had greater sexual initiative exhibited by males (which would mean EROTIC PLASTICITY that heterosexuals would likely approach lesbians more than gay 353 the gay-bi conflict is thus greater among women than men (see, of gay e.g., Rust, 1993). Many lesbians view bisexual women as being in women, but only about half (54%) the gay men, had had hetero- transition and as denying their true sexuality, and they regard them sexual intercourse. In a quite different sample consisting of gay with distrust (Clausen, 1990; Rust, 1993). males). Savin-Williams (1990) found that four fifths youth in New York City, Rosario et al. (1996) found nearly Female plasticity is particularly apparent in the findings that identical numbers: 80% of the lesbians had had sex with men, but some women who enjoy sex with men start having sex with only 56% of the gay men had had intercourse with women. Bell women also and that they do so even after their sexual patterns and and Weinberg (1978) found that lesbians exceeded gay men in all habits are well established. Dixon (1984) reported on a sample of categories of heterosexual experience, including coitus, oral sex, married, heterosexual women who had never felt any attraction to interpersonal masturbation, sex dreams, and marriage. In Whis- women prior to the age of 30 but who, at a mean .age of 37, had man's (1996) sample, 82% of the lesbians, but only 64% of the gay begun having sex with women as well as men. This occurred in the males, had ever had sex with a member of the opposite gender. context of swinging (i.e., consensual extramarital sex) and was Whisman also asked whether the respondent had had a meaningful often encouraged by the husbands. It does not appear to be late heterosexual relationship, and again, the rate of affirmative re- conversion upon awakening of latent lesbianism because the sponses was significantly higher among lesbians (72%) than gay women continued to enjoy having sex with men. men (45%). Men do not appear to exhibit that form of plasticity. Several The nigh rates of heterosexual experience among gay females studies of swinging and group sex found that women, but not men, were confirmed by McCauley and Ehrhardt (1980), who found that commenced same-sex intercourse under those circumstances. In a over half their sample of lesbians had had sex with men. Kitzinger study of mainly unmarried people who took part in group sex, and Wilkinson (1995) described a sample of women who had O'Neill and O'Neill (1970) found that over half (60%) of the become lesbian after a period of adult heterosexuality, often in- women, but only 12% of the men, engaged in homosexual activity. cluding marriage. Rust (1992) described a sample of lesbian A much larger investigation by Bartell (1970) of a predominantly women, 43% of whom had had heterosexual relationships after married sample involved in swinging yielded parallel findings. they had identified as lesbians, even many years after adopting the Looking at a great many episodes in which two married couples lesbian identity. Bart (1993) also found a small sample of lesbians would exchange partners for sex, Bartell found that the wives who entered into relationships with men, a pattern that led Bart to would have oral intercourse with each other about 75% of the time, attribute high plasticity to female sexuality. In a survey of female whereas the husbands had oral intercourse with each other less college dormitory residents, Goode and Haber (1977) found that than 1% of the time. all but one of the women who had had female sex partners had also Fang (1976) concluded that among swingers, same-gender sex- had male ones (and that one listed having sex with a man as ual activity "is rare for males yet is common for females" (p. 223). something she wanted to try). Schafer (1976) reported that lesbians She noted that many women swingers begin having sex with other in Germany were more likely than gay males to have heterosexual women "in order to please their husbands or to be sociable" (p. experience. 223) but then come to enjoy it. Similar findings were reported by the NHSLS (Laumann et al., One additional place to test the hypothesis of intraindividual 1994, pp. 310-313). Multiple tallies over different time spans variability is erotic activity in places where heterosexuality is (e.g., last year, last 5 years, since age 18) repeatedly showed that impossible, such as prison. This test may however be strongly women were more likely to have both male and female partners, at biased against the female plasticity hypothesis, if women have less least if one adjusts for the higher base rate of male homosexuality. sexual desire than men or if they are simply more willing than men Thus, among people who had any same-gender partners in the to forgo sex altogether for a period of time, as the Kinsey et al. past 5 years, half the men, but two thirds of the women, also had findings suggested. Despite this possible bias, the evidence is sex with opposite-gender partners. "The women are more likely largely supportive. Gagnon and Simon (1968) examined homosex- than the men to have had sex with both men and women than only ual activity in prisons and concluded that half the women in prison, same-gender partners" (p. 311). Likewise, the ratio of bisexual but less than half the men (estimates range from 30% to 45%), self-identification to exclusively homosexual identification was engaged in homosexual physical acts, most of which were con- higher for women (.56) than for men (.40; p. 311). Similar ratios sensual. When one considers that (a) base rates of homosexuality (.50 and .32, respectively) were found by Whisman (1996, p. 134), are higher among men than women, (b) men force other men more and in particular, she found the highest ratio (of 2.00, indicating a than women force other women (Propper, 1981; Scacco, 1975), majority of bisexuals) among lesbians who indicated that their and (c) women can live without any sexual contact more easily sexual orientation was a result of personal choice. This last finding than men, these results point toward a substantially greater will- is especially relevant to the plasticity hypothesis because it ex- ingness among women than men to indulge in same-sex activity plicitly links self-perceived erotic plasticity to intraindividual during prison. variability. Another way of expressing this finding is that bisexuality re- One should recognize that data on sexual activity in prison are subject to question on grounds of serf-report biases, lying, skepti- quires greater plasticity than homosexuality. Studies of the gay and cism about research, environmental and organizational-culture dif- bisexual community show a different balance between bisexuality ferences between male and female prisons, and possibly other and homosexuality depending on gender: Of the people who take problems. A solution to some of these methodological problems is part in same-gender sex, more women than men identify them- to ask prison inmates to estimate the degree of homosexuality selves as bisexual. In fact, the relatively large bisexual community among orner inmates. With this method, people do not have to is regarded by the exclusively lesbian contingent as a threat, and report on their own activity but merely give their estimates of what 354 BAUMEISTER others are doing. Ward and Kassebaum (1965) used this method in for both men and women, but the correlations were stronger for both male and female prisons. Their data suggested that far more women: .51 versus .26 for oral sex and .64 versus .55 for non- female than male inmates engage in homosexual activity. Collaps- Hispanic partners. These data suggest that when a woman moves ing the multiple-choice format to look at how many respondents from one country to a different one that has different sexual thought that over half of the inmates at their prison engaged in attitudes, her behavior is likely to change—especially to the extent such activities, Ward and Kassebaum found that a great many that she adopts the values and outlook of the new culture. In female respondents (51%) but relatively few male ones (21%) offered such a high estimate of the prevalence of prison homosexuality. Consistent with the plasticity interpretation, the researchers also found that the vast majority of inmates and staff thought that homosexuality in prison was merely a temporary adaptation to prison life and would not be continued outside of prison. (In fact, Giallombardo, 1966, concluded that most female inmates maintained a strong distinction between true lesbians who would prefer women outside of prison and were therefore regarded as sick and women who merely turned temporarily gay while in prison.) Ward and Kassebaum also found converging evidence by examining the contrast, men tend to remain the same when they change countries, regardless of the degree to which they adopt the values and outlook of the new culture. Education. The effects of education on age of first intercourse were studied by Wilson (1975), using a national sample in a 1970 survey. On this survey, higher levels of education were associated with delays in starting sexual behavior, and these delays appear to have affected women more than men. The proportions of men who were virgins on their 21st birthday varied only slightly from the least educated (19%) to the most educated (25%), but for women, prison records of inmates, in which mention of homosexual activ- the difference between the least educated (18%) and most educated ity was more frequent for female than male inmates. (43%) was substantial.1 On that same survey, an intriguing item asked people whether they believed that there was a substantial difference between what Sodocultural Factors most people did sexually and what they wanted to do. Responses The second major prediction derived from the female plasticity to this may reflect personal experience, observations about others, hypothesis is that sociocultural factors will have a greater impact and projection of own feelings onto others (e.g., see Finger, 1975, on female than on male sexuality. As already noted, the impetus on projection of sexual material). Once again, women varied more for this investigation was the contention that the sexual revolution than men as a function of education. In fact, men's agreement with had a larger effect on women than men and that this was part of a this item was the same from the most educated (69%) to the least broader pattern in which historical changes altered female sexual- (69%), whereas the highly educated women agreed less (51%) than ity more than male. The present section examines evidence as to the least educated women (65%). Thus, the perception of a gap whether socializing influences, cultural institutions, ideology, and between desire and reality in sex depended significantly on a other causes produce larger effects on females. It must be acknowl- woman's level of education, but the man's level of education was edged that although the plasticity hypothesis predicts greater irrelevant. Highly educated women were also twice as likely as causal effects by these factors, the majority of available findings uneducated women to hold liberal, permissive attitudes toward are only correlational. These can falsify the causal hypothesis but cannot prove it, Culture and acculturation. sex, whereas the corresponding difference for men was much smaller. If women are more socioculturally malleable, they ought to vary more than men from one culture to another. A variety of evidence supports this view, although considerably more work in this area is desirable. As one example, Christensen and Carpenter (1962) compared rates of premarital sex across three Western cultures and found much greater variation in the females than in the males. An unusually broad investigation was conducted by Barry and Schlegel (1984), who used the compiled ethnographic data on 186 cultures to compare sexual behavior patterns in adolescence. On all Education is not of course aimed mainly at altering sexual attitudes, so the effects of educational level should be considered by-products. It is useful to consider separately the question of sex education. This was done by Weis, Rabinowitz, and Ruckstuhl (1992). They sampled three college courses on human sexuality and obtained measures of sexual attitudes and behavior both before and after the classes. Perhaps surprisingly, they did not find that the courses produced any significant changes in behavior, and many attitudes (e.g., on abortion) likewise remained impervious to measures of sexual behavior, they found greater cross-cultural the course. However, they did find some changes in attitudes, variation among females than males, leading them to conclude that generally toward greater sexual permissiveness—but only among "variations among the societies in sexual customs are apparently females. These changes were found regardless of whether the greater for girls than for boys" (p. 325). initial baseline (precourse) attitudes showed greater permissive- The greater impact of culture on females than males was dem- ness among males (e.g., on oral sex) or females (e.g., on homo- onstrated in a different way by Ford and Morris (1993). These researchers studied a sample of Hispanic immigrants to the United States and included a (nonsexual) measure of acculturation that revealed how much the immigrant had adopted American culture. The acculturation measure correlated significantly (and positively) with several sexual practices for women but not men, including genital intercourse in the past year, anal sex, and use of condoms. The acculturation measure also correlated significantly with engaging in oral sex and having had sex with a non-Hispanic partner 1 In Wilson's (1975) data, education had a negative effect on sexual activity, whereas most later findings show a positive effect, and so this seems to be a contradiction. Wilson's sample included substantial proportions of people who came of age before the sexual revolution, which may help explain the difference. Prior to the sexual revolution, college was associated with delayed mating and marriage. It may however still be true that intelligent, college-bound individuals begin sex later than others but do in the long run become more liberal and experienced sexually. EROTIC PLASTICITY sexuality). Males did not change, but females did, and the differences are not attributable to baseline differences. Religion. Church attendance and religious belief seem to have 355 experience, 62% associated with reference groups who approved of premarital intercourse, whereas only 17% of the sexually inexperienced women associated with such groups; for men, the cor- a stronger (negative) effect on female than male sexuality. Reiss responding figures were 100% and 64%. Group encouragement (1967) found notably bigger differences in sexual permissiveness made a big difference, too. Over half (55%) the women with coital for females than males as a function of frequency of church experience had peer groups who encouraged sexual activity, attendance. This result holds up independent of the higher base rate whereas almost none (3%) of the virgin women associated with of church attendance by women. such groups. For men, the effect approached significance but was The stronger link between religion and female sexuality (than still smaller (88% to 50%). The correlation between having friends male) was confirmed by Adams and Turner (1985). Among elderly with sexual experience and having sexual experience oneself was women, they found that church attendance strongly predicted not significant for females but not for males. Thus, Mirande's data masturbating (19% vs. 83% for nonattenders), whereas no signif- suggest that the approval, encouragement, expectations, and be- icant effect was found among men. Harrison et al. (1974) found haviors of friends had a bigger influence on women than on men, that religious participation significantly predicted the permissive- although again self-selection of friends may contribute to these ness and sexual standards of rural females but not males. They also findings. found that females who had more experience with steady dating Further evidence was provided by Billy and Udry (1985). They were more permissive, whereas dating experience was irrelevant to were alert to the methodological problem that females might be males' permissiveness. more likely to associate with similar others than males, which Among students at a small religious college, Earle and Perricone would create an illusion of peer influence, but they were able to (1986) found that religion correlated negatively with sexually rule out this confound by demonstrating that there was no gender permissive attitudes for both men and women, but socioeconomic difference in sexual homogeneity of friendships. By collecting data status correlated with those attitudes for women only. Moreover, from the same sample on two occasions separated by 2 years, they when they compared freshmen's versus seniors' attitudes, they were able to ascertain whether friendship patterns at Time 1 found that "the attitudes of women seem to change more during predicted changes in sexual status, which makes causal inferences college years than those of the male peers" (p. 308). more plausible than purely cross-sectional data permit. These Murphy (1992) found that female Catholic clergy were more effects were consistently stronger for White females than for successful at fulfilling their vows of celibacy than were male males. Specifically, a White female virgin at Time 1 who had a Catholic clergy. This held up across a variety of measures (ever nonvirgin best female friend was six times more likely to lose her had sex, how many partners, how often) and appeared to be virginity by Time 2 than a White female virgin with a virgin best broadly true. Thus, female sexuality is better able than male friend. If the data are restricted to stable friendship pairs (i.e., sexuality to conform to highly nonpermissive standards in a reli- people who cited the same person as best friend on both occa- gious context, which again suggests greater plasticity. Peers and parents. The peer group is not as formal an insti- sions), the relationship was even stronger. Males showed no such effect. tution as the school or church, but it too has effects in socializing Parents can also be considered agents of socialization, and they sexual behavior. Effects of peer group attitudes and behavior on are relatively immune to the self-selection bias problem insofar as loss of virginity were studied by Sack, Keller, and Hinkle (1984). children cannot choose their parents. On the other hand, it does not The behavior of the peer group affected both genders: Whether the seem safe to assume that they socialize boys and girls the same respondents' friends were having sex correlated with whether the (see, e.g., Libby & Nass, 1971). Still, one literature review sug- respondents themselves were having sex for both males (r = .47) gested that the weight of evidence indicates that the parental and and females (r = .49). The peer group's approval was more family environment has a stronger effect on daughters than on sons strongly linked to the sexual behavior of females than males, (B. C. Miller & Moore, 1990). Longitudinal research found that however. When asked how their friends would feel about them living with a single parent increased the likelihood of early loss of having sex, females' responses significantly predicted whether virginity for girls but not boys (Newcomer & Udry, 1987). (Pa- they had had sex (r = .53), but the effect for males was not rental divorce during the study was associated with increased significant (r = .26). The authors also reported direct effects that sexual activity by both sons and daughters.) A broader study of were corrected for effects of other variables. The direct effect of multiple family (especially maternal) influences repeatedly found the peer group's approval for males was negligible, .00, but it that daughters' sexual attitudes and sexual behaviors were more remained significant for females, .25. It is also worth noting that closely related than sons to most .social variables, including par- the proportions of virgins versus people who had sex were nearly ents' age, parents' age at their wedding, parental divorce, mother's identical in the two genders, so the results of this study cannot be premarital pregnancy, and mother's attitudes about sex (Thornton ascribed to any restriction of range or floor/ceiling effects. (There & Camburn, 1987). (Specifically, daughters' permissive sexuality is however the possibility that choice of peers was a result, rather was increased by having older parents, earlier parental marriage, than a cause, of sexual intentions and practices. On the other hand, Billy and Udry, 1985, found such selection effects to be bigger parental divorce, mother's premarital pregnancy, and mothers with permissive attitudes.) among males, which would bias the results against the plasticity hypothesis.) of topics that children learned about from their parents as an index A study of sex education by Lewis (1973) counted the number Similar results were obtained by Mirande (1968), who found a of parental information transmission. This index correlated signif- significant link between peer group approval and sexual activity icantly with likelihood of having intercourse and with number of for females but not for males. Of women who had had sexual sex partners for young women (such that more parental education 356 BAUMEISTER predicted less sex and less promiscuity), but the correlations were sexual orientation" (p. 136), but they themselves regard that con- not significant for young men. Lewis also found that marital clusion as premature and prefer to wait for further confirming conflict in the parental home had a stronger effect on the daugh- evidence. If further work continues to have greater success estab- ter's sexual development (leading to more sexual experience and lishing genetic contributions to male than female homosexuality, more promiscuity) than on the son's, although both effects were that will strengthen the view that female sexual orientation is more weak. socioculturally malleable. Both parental and peer influences were studied by Reiss (1967). He found that female permissiveness was more influenced than Personal choice. Another way to approach the question of plasticity is to examine whether people perceive their sexuality as male permissiveness by a broad spectrum of social forces. Both a matter of choice and something that is at least partly under their peer and parental standards had a stronger correlation with the control, as opposed to regarding it as something inborn and un- permissiveness of females. changeable. In an important sense, this approach takes the question Genetic versus environmental factors. A different way of looking at sociocultural effects is to consider the opposite, namely genetic prediction. Research on behavior genetics has occasionally examined sexual factors by looking at correlations between twins. By comparing monozygotic and dizygotic twins, the degree of genetic contribution can be estimated, and the remainder of the variance can be tentatively chalked up to erotic plasticity. This technique was employed by Dunne et al. (1997) with a large sample of Australian twins in the effort to predict age at first intercourse. Among the people bora after the sexual revolution (i.e., those under age 40), the authors concluded that the genetic contribution accounted for 72% of the variance for males but only 40% of the variance for females. This discrepancy suggests that male sexuality is more determined by genetic factors, which in turn 2 implies a greater role for sociocultural factors for females. The behavior genetics approach has also been applied to homosexuality, and indeed, the question of whether sexual orientation is a matter of nature or nurture (i.e., socially influenced choice or genetically/biologically ingrained pattern) remains the focus of considerable political, social, intellectual, and emotional controversy. Some studies do suggest a greater effect of genetic factors on men. Using a twin registry. Bailey and Martin (1993; see Bern, 1996) found that heritability of sexual orientation was significant for men but not for women. Hu et al. (1995) likewise found a significant linkage between chromosome and homosexuality patterns for men but not women. On the other hand, Hershberger (1997) found effects for both genders, and the female effects were stronger. A pair of studies using the less optimal method of starting with self-defined gay people (which may conceal gender differ- of essentialism versus social constructionism in sexuality (see DeLamater & Hyde, 1998) and asks individuals which view seems to fit their sense of their own sexuality. Do they feel they can socially construct their sexuality, or does it seem to be an innate part of their essence? Several studies have explicitly examined whether people perceive their sexual orientation to be a matter of choice. Whisman (1996) interviewed self-identified homosexuals and found that a higher percentage of lesbians (31%) than gay men (18%) described their sexual orientation as having been a matter of conscious, deliberate choice. Using a more nuanced measure, Rosenbluth (1997) found that over half a sample'of lesbians perceived their homosexuality to be the result of a conscious, deliberate choice. Savin-Williams (1990) found that lesbians felt they had more control than gay men over their sexual orientation. In addition, lesbians were more likely to think that they could renounce their gay orientation and less likely to regard their sexual orientation as beyond their personal control. Thus, subjective perceptions of one's own homosexuality suggest that erotic plasticity is higher in females. These data dovetail well with the trend in the genetics research. Although in neither case is the mass of evidence fully rigorous and overwhelmingly solid, the currently available data offer the best guess that male homosexuality is more strongly linked to innate or genetic determinants whereas female homosexuality remains more subject to personal choice and social influence. Political ideology. Consistent with the view that lesbianism can reflect personal choice and social construction, there are re- ences if women are slower to make a firm, definite self- ports that some females became gay for political reasons associ- identification as gay) and examining siblings found stronger ef- ated with the women's movement. Blumstein and Schwartz (1977) fects among the men, but the differences were slight (Bailey & reported that some women became lesbians under the influence of Pillard, 1991; Bailey, Pillard, Neale, & Agyei, 1993). Genetic political ideology that defined heterosexuality as a form of sleep- explanations were also favored by Bailey and Zucker (1995) in ing with the enemy whereas lesbianism was the only politically their review of findings that sought to predict adult sexual orien- correct form of sexuality. They noted in their conclusions that such tation from cross-sex behavior during childhood. They concluded changes raise theoretical questions about the plasticity of sexual from retrospective studies that effects are strong for both genders desire, and thus they anticipated the present argument to some but significantly stronger for males than females. Prospective degree. Kitzinger (1987) summarized the radical feminist view that studies have thus far found strong, significant predictions only for "patriarchy (not capitalism or sex roles or socialization or individ- males. A recent review by Bailey and Pillard (1995) is one of the few to sort the evidence by gender. They concluded that the evidence for genetic contribution to homosexuality was far stronger for males than for females. To be sure, this difference in strength of evidence does not necessarily mean the true effect size is larger for males because there have been more male-only studies and larger samples. Bailey and Pillard said that some experts have begun to conclude "that female sexual orientation is less heritable than male 2 The older sample, which came of age prior to the sexual revolution, showed quite different patterns, and indeed, the genetic contribution for males over age 41 was 0%, which was lower than that for females (32%). A likely guess at explanation would be that age of first intercourse prior to the sexual revolution was a matter of highly restricted opportunity for males, and Dunne et al. (1997) make the same point. Hence, these findings are not relevant to the plasticity hypothesis, but I acknowledge that this finding contradicts the general pattern. 357 EROTIC PLASTICITY ual sexist men) is the root of all forms of oppression [and] that all greater political influence on female than male sexuality, but more men benefit from and maintain it and are, therefore, [women's] evidence is desirable. political enemies" (p. 64). In consequence, the politically optimal choice of women should he to reject heterosexuality. Kitzihger Education and Religion in the NHSLS quoted a woman who asserted "I take the label 'lesbian' as part of the strategy of the feminist struggle" (p. 113). Johnston's (1973) The other sections of this article present evidence from many formulation was blunter: "Feminists who still sleep with the man different investigations, but this section considers only one, al- are delivering their most vital energies to the oppressor" (p. 167). though it is a very large and thorough one. The best data available Similar reports of politically motivated lesbianism are found in on modern American sexual practices are provided in the NHSLS other sources. Pearlman (1987), for example, in discussing the rise (Laumann et al., 1994; Michael, Gagnon, Laumann, & Kolata, of political lesbianism in the 1970s, wrote, "Many of the new, 1994). These data represent a carefully, properly constructed na- previously heterosexual, radical lesbians had based their choice as tional sample, with lengthy individual interviews plus written much on politics as on sexual interest in other women" (p. 318). questionnaires, which had unusual success at securing high re- Rosenbluth (1997) found that 12% of a sample of lesbians (and a sponse rates and thus avoiding the volunteer bias that seriously similar proportion of heterosexual women) cited political reasons compromises the value of many sex surveys (e.g., see Morokoff, as the basis for their sexual orientation and relationship style 1986; Wiederman, 1993). The NHSLS therefore deserves special choice. Charbonneau and Lander (1991) found that a third of their attention. sample of women who converted to lesbianism during midlife Although the NHSLS researchers did not have any apparent cited reading feminist texts as a reason, and they spoke of the interest in the question of differential plasticity (and did not even feminist path to homosexuality in which lesbianism was an out- bring up the issue), the extensive tables reported in the fuller growth of the commitment to feminism. In that sample, moreover, version of their work (Laumann et al., 1994) permit comparison of some women described the change as one of self-discovery, males and females in terms of sociocultural predictors. They whereas others regarded it as an active choice, and the latter found present extensive data on the effects (correlates) of two main the adjustment more difficult (not surprisingly). Whisman (1996) sociocultural institutions, namely, school and church. More pre- likewise found that women, but not men, cited political reasons as cisely, they break their data on many sexual practices and attitudes a reason for choosing homosexuality. Echols' (1984) history of down by educational levels and by religious affiliation. These data feminist sexual politics recorded the lesbian separatists' phase of enable one to compare whether males or females show greater "establishing lesbianism as a true measure of one's commitment to variation in response to these two institutional forces. If female feminism" (p. 56), and other leading feminists denounced hetero- plasticity is greater, the variation across categories should be sexuality as a choice that was in fact coerced by the patriarchal greater among females than males. Because the effects are typi- political system. cally linear, the present coverage can be simplified by considering Although further evidence would be desirable, the finding that merely the uppermost and lowermost categories (exceptions are some women have seemingly exchanged male for female sex noted below). For education, these are the people with the least partners under the influence of political ideology constitutes com- education (less than high school) versus those with the most pelling evidence for erotic plasticity. No such claims have been (graduate or advanced degrees). For religion, the extreme cases made regarding men, and it does seem intuitively doubtful that were what the researchers called Type n Protestants, representing political writings and speeches would persuade some men to give conservative, evangelical, up women and heterosexuality and begin having intercourse with tions, and at the opposite, the people listing "none" as their other men instead. If some women have indeed made such a switch religion. The latter were typically more active in whatever sexual under similar influences, that would confirm the greater sociopo- category was being considered. litical plasticity of the female sex drive. fundamentalist Christian denomina- The information presented here is based on Laumann et al.'s The reports of women with a history of exclusively heterosexual (1994) tables, not (in most cases) their statistical analyses, and so desires changing to have homosexual relations because of political it is not possible to report statistical significance. One can however reasons resemble a finding that was presented as evidence of meta-analyze the directions of effects, and I report such a summary intraindividual variability: When married couples start congregat- analysis at the end of this section. ing for mate swapping, after a while the women begin having sex It was necessary to make a priori decisions about what depen- with other women, often under the encouragement of the men, who dent variables to consider. The main sexual practices that any like to watch this (Dixon, 1984). The reverse pattern is almost researcher would presumably expect to include are oral sex (per- unheard of (i.e., heterosexual men taking up homosexual acts in forming and receiving), anal sex, masturbation, homosexual activ- group settings, especially if the ostensible purpose is to entertain ity, and contraceptive use. (Vaginal intercourse is too standard to their wives). Such adaptations in women provide a salient and be useful; that is, nearly all heterosexual adults who have sex have vivid illustration of erotic plasticity. vaginal intercourse.) Additionally, I report evidence about sexual There is not much evidence on gender differences in degree of satisfaction, frequency of sex, sexual dysfunction, duration of sex, political influence on other aspects of sexuality beyond sexual and fertility; these are more peripheral, and some researchers orientation. DeLamater and MacCorquodale (1979) reported that might prefer not to include them. I present these to avoid charges general political stance, measured either in terms of self-reported of selective reporting and to permit readers who judge centrality liberal versus conservative classification or in terms of reported differently to draw their own conclusions. political participation, predicted sexual permissiveness more strongly for females than males (p. 127). This again suggests Oral sex. Beginning with the main sexual practices, it is clear that the differences associated with education and religion are 358 BAUMEISTER consistently greater for women than men. On the item of whether Results for religion were not reported in their work, presumably the person had ever performed oral sex on a partner, having a high because differences were generally not significant. It is also pos- level of education raised men's affirmative answers from 59% sible to compare responses item by item, but these simply confirm to 80.5% (roughly a one-third increase), whereas women's in- the pattern reflected in the summary means: Level of education creased from 41% to 79% (nearly double; note, though, that there predicted bigger differences in the breadth of women's sexual could be a ceiling effect, given the similarity among highly edu- interests than men's. cated men and women). On the complementary question about whether the person had ever received oral sex, the most and least Homosexual activity. Education and religion were also linked to same-gender activity. Several items were used. Three asked educated men differed by less (81% and 61%) than the most and whether the person had ever had any same-sex partners in the past least educated women (82% and 50%). Thus, on both giving and year, in the past 5 years, or since age 18. These did not show linear receiving oral sex, education level predicted bigger differences in effects of education, nor was there any consistency as to whether women's sexual behavior. With religion, the differences between the most liberal and most conservative categories were again larger for females. The differ- men or women differed more across categories, so they do not furnish useful information relevant to the plasticity hypothesis (Laumann et al., 1994, p. 302). ences on performing3 and receiving oral sex, respectively, were 12 Clearer results were obtained by asking whether the individual and 13 percentage points for men, whereas for women, they self-identified as gay or bisexual. College education doubled were 22 and 19 points. The effects of religion are thus opposite to men's likelihood of becoming gay, whereas for women, the like- education, and indeed, religiosity appeared to make women more lihood increased by a factor of nine. Likewise, a composite of different from men even as education made them more similar. items asking for same-gender desire, attraction, or appeal found Anal sex. Anal sex provides a useful counterpoint because the that with increasing education, the likelihood of men's positive base rates on the NHSLS were quite low, in contrast to oral sex, answers increased by about half (5.8% to 9.4%), whereas for which had high base rates, and so the findings are less vulnerable women, the increase was nearly quadruple (3.3% to 12.8%). In to explanations based on ceiling effects. The difference between determining same-gender sexual interest and activity, the authors high school dropouts and people with master's degrees (or more) themselves articulated the differential effect of education by say- was only 8 percentage points for males but 16 points for females. ing that "education . . . does seem to stand out for women in a way Moreover, the proportional change makes the difference even more dramatic: Education produced only about a one-third increase in males' likelihood of engaging in anal sex (from 21% to 29%), whereas it more than doubled the women's likelihood (from 13% to 29%). Similar patterns were found for religion: Women showed much greater variation than men, and the difference between categories represented more than doubling women's likelihood (from 17% to 36%, or 19 points), whereas men increased by only about half (from 21% to 34%, or 13 points). The difference is even more dramatic if one looks only at incidence of anal sex within the past year, which is probably a more accurate measure and more closely linked to current religiosity: The most and least religious men scarcely differed (7% vs. 9%), whereas the most (6%) and least (17%) religious women showed very different incidence rates of anal intercourse. Masturbation. With masturbation, comparisons are difficult because the base rates differed substantially among males versus females and because the activity is arguably different in the different genders. There seemed to be no clear gender difference in the correlations between level of education and frequency of masturbation. Success at masturbation (measured by likelihood of reporting that one always or usually has an orgasm during masturbation) did vary more as a function of both education and religion in women than men, consistent with the hypothesis of erotic plasticity. The education effect was not large, however. To be conservative, I count the masturbation data as inconclusive. Sexual variety. Next consider sexual interest and arousal in response to novel, assorted sexual practices. The NHSLS research- that it does not for men" (Laumann et al., 1994, p. 309) and that the increase in same-gender sexuality as a function of education was "more pronounced and more monotonic for women" than men (p. 309). For religion, too, the predictive effects on gay/bisexual identification and on the composite interest were larger for women than for men. Women's gay/bisexual identification differed by a factor of 15 (from .3 to 4.6) in comparing conservative Protestants to people with no religion, whereas men's identification differed by a factor of only nine (from .7 to 6.2; note, though, that given the higher base rate, the men increased by an extra percentage point). On the composite item, women increased from 5.5 to 15.8, whereas men increased from 5.6 to 12.9. Contraception. The last of the major sexual practices I con- sider is contraceptive use. Contraception in marriage is too complex and multidetermined an issue to use for present purposes, insofar as people may or may not be having sex in order to have children. Contraception in extramarital or extradyadic activity is far more straightforward, however, because it is reasonable to assume that if one is married or partnered, one does not want to create a pregnancy with someone else. For this item, the researchers restricted their data to people who were having extradyadic sexual partners, which meant that many categories had too few data points to be reported. Still, there were sufficient data to permit comparisons based on education (Laumann et al., 1994, p. 451). The category of people who reported always using contraception with the secondary partner showed a significant rise among ers offered a list of sexual practices and asked people to indicate how many appealed to them. The least educated men expressed interest in 2.3 practices (out of 15), whereas the most educated men expressed interest in only 2.6, so the difference was negligible. For women, however, the difference was from 1.3 to 2.1 out of 14. (The item about active anal sex was deleted for women.) 3 On this item, the most and least religious men did not have the largest differences, contrary to the general pattern. Cathoiic males engaged in slightly more cunnilmgus than the atheists and agnostics. If one looks at variance across ail categories, however, it remains true that women differed more than men. EROTIC PLASTICITY women as a function of increasing education, from 55% to 79%. 359 yielded 12 comparisons showing greater sociocultural effect sizes For men, ironically, the same two educational categories showed a on women and none showing greater effects on men. This may be small trend in the opposite direction, dropping from 65% to 54%, considered a statistically significant pattern, insofar as meta- suggesting that the more educated men were actually more careless analytic combination yields a very small likelihood (p < .001) of about contraception.4 In any case, the correlation with education such a result occurring by chance (Darlington, 1975). was greater for women than men. Among the less central practices, results were somewhat more As noted above, it is also possible mixed. Still, even among these, the preponderance of comparisons to consider some less central aspects of sex. On reporting that one showed greater effects of education and religion on women than on was extremely satisfied with one's partnered sex (in a physical men. Other dependent variables. sense), the difference between the least and most educated women Thus, the findings from the best database available clearly and was greater than the corresponding difference for men, although consistently fit the hypothesis of female erotic plasticity. The two the difference was not large and the progress across educational major cultural factors studied in that investigation, namely, edu- categories was not linear, so this result may not be conclusive. The difference between conservative Protestants and nonreligious peo- cation and religion, were associated with bigger changes in women's than men's sexual behavior. This was true regardless of ple was also larger for women than men on this item. whether base rates were high (as in oral sex) or low (as in anal sex Frequency of partnered sex is of potential interest, but Laumann or same-gender activity). It was also true regardless of whether the et al. (1994, p. 90) reported that neither religion nor education had effect of the institution was generally to constrain sexual activity any relation to frequency of partnered sex. The differences be- (as with religion) or to promote liberal attitudes and broad interests tween the most and least educated in the number of people reporting highly frequent sex were greater for women than for men, ferences could contribute to some findings, but other findings are consistent with the hypothesis, but given the lack of significance overall, this finding probably should not be accorded much weight. (as, apparently, with education). Ceiling effects and baseline difimmune to these problems, and they all point to the same conclusion of greater female plasticity. The authors noted that there was a slight tendency for men with no religious affiliation to be more likely to report highly frequent sex than the conservative Protestants, whereas there was no difference among women in these categories, so this effect would be in the direction contrary to the plasticity hypothesis. Oddly, though, the other religious categories showed greater variation among women than men, so this item departed from the typical pattern in which conservative Protestants and people with no religion constituted the extremes. The variance across the four religion categories was greater for women than men, consistent with the hypothesis of female erotic plasticity. Probably, these numbers just reflect nonsignificant, random variations, and so they too should be discounted. In terms of duration of most recent sexual event, there was a suggestive trend. Education produced a greater variation in the percentage of women than men saying that their most recent sexual encounter lasted under fifteen minutes.5 A subsequent work with the same data set examined influences on sexual dysfunction (Laumann, Paik, & Rosen, 1999). It found that more education was associated with less sexual dysfunction for women, whereas there was no significant effect for men. Thus, Attitude-Behavior Consistency The third prediction derived from the plasticity hypothesis is that women will show lower attitude-behavior consistency with regard to sex. If women's behavior is more malleable by situational forces than men's, then women will be more likely than men to do things contrary to then: general attitudes. With regard to sex, the discrepancy between women's attitudes and behavior has been noted by several authors. Commenting on their study of Black high school girls, Roebuck and McGee (1977) said that "of interest are incongruities in expressed attitudes and behavior" (p. 104). Social class correlated with sexual attitudes, for example, but not with behavior. Antonovsky, Shoham, Kavenocki, Modan, and Lancet (1978) devoted special study to the inconsistency between attitudes and behavior in their study of Israeli adolescent girls, in which they found that a third of the nonvirgins endorsed as important the value of a female remaining a virgin until she married. These researchers found that such apparent self-disapproval was partly maintained by making external attri- again, sociocultural variables were linked to bigger differences in females than males. Last, one may consider fertility, which may or may not be relevant insofar as one considers sex to be aimed at reproduction. Educational level predicted a bigger difference in women's reproductive patterns than those of men, measured in terms of number of children. Religion did too. Unlike other variables, fertility showed directionally similar effects of religion and education: Highly educated women and highly religious women had fewer children than other women. Summary. On the main aspects of sex covered in the HHSLS, it was possible to construct 8 comparisons as a function of education. One (masturbation) was inconclusive, and the other 7 showed greater effects of education on women than men. There were 5 comparisons as a function of religion, and all 5 showed bigger effects on women than men. Combining these thus 4 There was also a category of more highly educated men whose rate of contraceptive use in secondary relationships was about the same as the low-education category. The data for women did not include this category because too few women with that level of education reported secondary sexual relationships. If one uses this category for the men, then there is no change as a function of educational level, confirming the conclusion of greater change among women. 5 There was also a category of people making the doubtful claim that their most recent sex act had lasted more than an hour, the effects of education were not linear, reflecting perhaps some mixture of altered time perceptions and boastfulness. If one ignores these problems and simply compares the most and least educated categories, the difference for males (16.7 vs. 19.9) is slightly larger than for females (13.3 vs. 14.2), which is contrary to the female plasticity hypothesis. I acknowledge it for the sake of completeness, but the discrepancy of two percentage points seems too small to be meaningful. 360 BAUMEISTER butions for their past sexual experiences—yet the girls continued able attitudes toward condoms and the finding that most respon- to engage in sex even when they disapproved of their doing so. In dents did not consistently use condoms during their last experience considering the gap between attitudes and behavior, Antonovsky et with causal sex" (p. 42). The inconsistency was apparently greater al. pointed out that "overt behavior is much more influenced by among women: Women reported a higher intention than men to situational factors than are attitudes" (p. 270), which confirms the use condoms, as well as reporting greater fear of sexually trans- present rationale for using attitude-behavior inconsistency to test mitted diseases, but actual condom use was the same for both the erotic plasticity hypothesis. genders. Because people use condoms less than they say they The attitude-behavior gap was noted by Croake and James should, the behaviors of the women were more inconsistent with (1973). Their research involved multiple surveys of college stu- their attitudes than were the men's behaviors. Moreover, the temp- dents of both genders. Comparing their findings regarding sexual tation factor should have produced the opposite effect: Condoms attitudes with concurrent findings from other work on coital ex- are generally regarded as detracting more from male than female perience, they noted "a much higher percentage [of women] ex- pleasure, and so males should be more willing to betray their periencing sexual intercourse than those in the same age group pro-condom attitudes. who approve of such behavior" (p. 96) as evidence of inconsistency among females. A similar conclusion was found in a cross-cultural investigation by Christensen and The singles bar sample studied by Herold and Mewhinney (1993) also showed inconsistency between women's attitudes and Carpenter (1962). They computed behaviors regarding casual sex itself, which was defined as having an erotic contact beyond hugging and kissing with someone the approval-experience ratio that permitted them to investigate par- respondent had just met that same day. Only 28% of the women ticipation in premarital sex even when it went against one's own said they anticipated sometimes having sex with someone they had values. In all three cultures they studied, these ratios were lower just met, but the majority (59%) had done it. The high rate of for women than men, and in their sample of Americans in the having had sex with a new acquaintance was especially remarkable sexually conservative Utah region, the women's ratio was only .33, in view of the findings that the women reported high rates of guilt indicating that two thirds of the women who engaged in premarital over such activities (72%) and low rates (2%) of saying they sex had done so against their personal values. (The men in that consistently enjoyed them. The authors pointed out that this in- sample had a ratio of .59.) Because base rates of participation in consistency was peculiar to women, although they had not pre- premarital sex differed substantially by gender, these authors did dicted it and had no explanation: "The apparent contradiction find that numerically more men than women acted against their between the negative attitudes expressed by many of the women values. Still, the ratio seems the more meaningful and relevant regarding casual sex and the fact that most of them had engaged in indicator because it is not confounded by base rate, and it suggests casual sex provides more questions than answers" (p. 41). that premarital sex involves a higher rate of attitude violation for women than men. Below, I consider the possibility that one reason for female erotic plasticity is that women's role requires them to participate in Because the base rates of adultery are quite low and attitudes of sex even when they do not particularly wish to do so. Having sex both genders are fairly negative, it is difficult to get good data on without desire is one form of inconsistency (although it may attitude-behavior consistency in that sphere. One creative solu- involve specific desires rather than general attitudes that are in- tion, developed by Hansen (1987), involved looking at dating consistent with behavior). Beck, Bozman, and Qualtrough (1991) couples' involvement in any extraneous erotic activity, such as surveyed people as to whether they had participated in sex without kissing or petting. Sure enough, Hansen found that majorities of desiring it. Although a majority of both genders reported having both men and women in his sample had experienced extradyadic done this, the proportion was higher for women (82%) than for contacts (by self or partner). Despite the high frequency, tolerance men (60%), and the authors noted further that nearly all (97%) of was low, and both men and women expressed some degree of the women past the age of 25 reported having engaged in sex when opposition to such activity. For present purposes, however, the they lacked desire for it. High rates of unwanted sex also emerged crucial comparisons involved whether the attitudes (and other from a study of people in committed relationships: During a predictor variables) correlated with having engaged in extradyadic 2-week period, 50% of the women, but only 26% of the men, activity. The most directly relevant variable was extradyadic permissive- engaged in unwanted sexual activity at least once (O'Sullivan & Allgeier, 1998). ness, that is, the attitude toward such activity. The correlation Homosexuality provides another sphere in which attitude- between tolerating such activity and having participated in it was behavior consistency can be studied. Laumann et al. (1994) ap- stronger for men (r = .48) than for women (r = .31). Other proached this question by calculating the overlap between three variables, including religiosity, sexual attitudes in general, and categories of same-gender sexuality, namely, desire, behavior, and identification with gender roles, also showed stronger correlations identity. Desire consisted of a positive response to questions about for men than women. Combining the effects of all these attitudinal sexual attraction to a same-gender person and about finding the predictor variables enabled Hansen (1987) to account for a third idea of same-gender sex appealing. Behavior consisted of having (33%) of the variance in whether men had strayed, but the same had sex (past age 18) with someone of one's own gender. Iden- predictors accounted for only a ninth (11.4%) of the variance in tity referred to self-identification as gay or bisexual. Attitude- women's behavior. Another area in which attitude-behavior consistency can be behavior inconsistency is perhaps best shown by people who fail to examined is use of condoms. Herold and Mewhinney (1993) adult behavior or identity... [characterized] 59 percent of the conducted a survey of people in a singles bar. The authors re- women and 44 percent of the men" (p. 298). Maximal consistency, marked on the "obvious discrepancy between the reported favor- in contrast, would be exhibited by people who registered positive implement their desire in any way: "Desire with no corresponding 361 EROTIC PLASTICITY on all three indicators. Consistency was greater among men (24% chism (cf. Baumeister, 1989). Still, across many studies and dif- of those who scored positive on any same-gender item) than ferent kinds of measures, women are more likely than men to have submissive and masochistic fantasies. Thus, the masochism data women (15%). The purest attitude measure regarding homosexuality in the resemble the homosexuality data: Women are more likely than NHSLS was whether the person rated same-gender sex as appeal- men to report having such desires and interests, but they are less ing. This item did not depend on actual experiences of desire or likely than men to report taking part in such activities. Hence, for behavior but was a simple rating of attitude in the abstract. This women, there is a larger gap between attitude (or desire) and item also has the advantage that men and women had approxi- behavior. mately equal rates of positive responses to it. Hence the attitude- The masochism data are especially useful because they rule out behavior consistency question can be formulated by examining one further alternative explanation that could apply to the homo- whether these attitudinal responses correlated with current behav- sexuality data. I have said that men mainly show discrepancies ior. Laumann et al. (1994, p. 159) found that less than half the between desire and behavior because of lack of opportunity: Many women who liked the idea of same-gender sex had actually had sex men want to engage in sex but cannot find a willing female partner. with another woman in the past year. In contrast, nearly 85% of the The generally more conservative and selective attitude of women men who found same-gender sex appealing had had sex with a toward sex entails that women are less likely than men to comply man in the past year. Although it is possible that gender differences with requests for sex (Clark & Hatfield, 1989). The difficulty of in opportunity and initiative contributed to this difference, it is finding a female sex partner could conceivably help explain why clear that attitude-behavior consistency was substantially lower women are less successful than men in enacting their same-sex among females. desires. In masochism, however, it is usually opposite-sex partners Attitude-behavior discrepancies regarding homosexuality were who are sought, and so women would be seeking men to dominate also documented by Bell and Weinberg (1978). They found that them. Women ought therefore to enjoy an advantage over men gaps between homosexual feelings (or desires) and homosexual when both are trying to act out then- submissive fantasies, and so behavior was much larger for the lesbians (22 percentage points) this alternative explanation would predict that women would have than for gay males (3 percentage points). Moreover, lesbians were higher attitude-behavior consistency in this sphere. more likely than males to have tried to relinquish their homosex- women again show more inconsistency, which fits the broad uality and "go straight"—which is ironic because lesbians ex- pattern I have hypothesized, namely that women's attitudes and pressed fewer regrets about their homosexuality and were less behaviors are less consistent than men's when it comes to sex. Instead, likely to wish for a "magic pill" that would instantly transform them into heterosexuals. Golden (1987) too was struck by inconsistencies between women's thoughts and feelings regarding sexual orientation. Her sample exhibited remarkable incongruences, Assessing the Evidence and Possible Problems Contrary Evidence including women who identified as lesbians (often for political reasons) but whose sexual behavior had been exclusively hetero- A few studies have yielded findings that seem to contradict the sexual, as well as the reverse pattern of women who identified general pattern of lower attitude—behavior consistency among themselves as heterosexuals but had only had sex with women. women than among men. In a survey of students at a small, private, Consistent with the present hypothesis that females have higher southern, church-related university, Earle and Perricone (1986) erotic plasticity, Golden contended that gay men would have fewer found that women's attitudes toward premarital sex more strongly such inconsistencies, would be less likely than the lesbians to predicted whether they had had premarital sex than did men's regard their homosexuality as elective, and would in general be attitudes. One possible explanation is that the lower prediction less likely to exhibit the "fluid and dynamic as opposed to fixed rates for males reflected lack of opportunity: Many men may have and invariant" (p. 19) patterns of sexual behavior that she charac- wanted sex but been unable to find a willing partner, particularly terized women as having. at a religious institution. The finding that women's attitudes to- To get evidence converging with the homosexuality findings, ward sex were far more conservative than the men's attitudes one can consider research findings on sexual masochism. Males supports this view, and so the finding may not be a meaningful are somewhat more likely than females to engage in this form of exception to the female plasticity pattern. sexual activity (see Baumeister, 1989), and indeed, some research- McCabe (1987) surveyed an Australian sample of people in- ers have focused on male masochism simply out of convenience: volved in serious relationships as to whether they were having sex It is easier to get a sample of participants who have engaged in this and whether they were pleased with having (or not having) it. She activity if one uses males rather than females (see, e.g., Moser & found "a greater congruence between desire and experience for Levitt, 1987; Scott, 1983; Spengler, women than for men" (p. 31). This discrepancy was mainly due to 1977). Studies that have compared the two generally find more male masochists than fe- the imbalance in the category of people who were not having sex male ones (Moser & Levitt, 1987), although the differences are not but wished they were having it. This category of "reluctant vir- large. gins" was almost entirely male. Again, this seems to reflect a lack Yet data on fantasy and desire do not fit the usual pattern of of opportunity for males to act on their wishes. males showing more. If anything, women have more fantasies and DeLamater and MacCorquodale (1979) found higher correla- desires for submission. This was confirmed in a recent review by tions for women than men between personal sexual ideology (i.e., Leitenberg and Helming (1995), although their analytic approach attitudes of approving particular activities) and recent sexual ac- combined being forced into sexual activities with being tied up, tivity within a relationship. This too could reflect lack of oppor- humiliated, made into a sex slave, and other hallmarks of maso- tunity for males. The nature of the measures seems especially 362 BAUMEISTER conducive to such consistency, insofar as the measures assessed these more than females, although there are some ambiguities in sexuality within a relationship and approval of such activity. The the evidence (e.g., certain patterns such as exhibitionism and bulk of the evidence for inconsistency did not refer to relation- bestiality may be more tolerated among females and hence not ship contexts (and much of it was even explicitly outside of regarded as paraphilias; see Amsterdam Sex Museum, 1999). If relationships). males engage in more varied sexual practices, does this constitute A different sort of contrary evidence was found in the NHSLS. a form of sexual plasticity in which males surpass females? Laumann et al. (1994) reported that urban (instead of rural) resi- As already noted, my predictions regarding variance are limited dence was more strongly correlated with male than with female to intraindividual variance. Interindividual variance can well have homosexuality, and this runs contrary to the general pattern of a strong genetic basis. It is possible that the greater variation in greater sociocultural influences on females. To some extent, this male sexual tastes reflects genetic or biological variation rather finding could be explained on the basis of homosexuality inclined than sociocultural plasticity. men moving to big cities in order to find others, but Laumann et Then again, some paraphilias seem incontrovertibly learned. al. found that some difference remained even if they considered Latex, for example, has not existed on the planet long enough to only whether the person was born in an urban versus rural envi- influence evolutionary processes and genetic markers, and so a ronment. Thus, growing up in a big city apparently had a bigger latex fetish seems most plausibly interpreted as something learned chance of influencing men to become gay than women. rather than innate (although it is difficult to rule out the possibility The urban-rural difference was not predicted by Laumann et al. that this fetish is a byproduct of some other genetic, innate ten- (1994), nor did they have any clear idea of what it meant or why dency; moreover, latex fetishes may be popular with both gen- it occurred. Considering the large number of analyses they con- ders). If males are more likely than females to adopt such para- ducted and the preponderance of evidence for greater influence on philias, then certain social and situational variables apparently females, this finding might conceivably be a statistical fluke. Then have stronger effects on males man females. again, it seems necessary to consider the possibility that it reflects Research has not yet provided a clear understanding of the a genuine difference and hence a legitimate contradiction to the causes of paraphilias. For present purposes, it is merely necessary broad pattern of greater plasticity among females. Crucially, how- to suggest some plausible way that males could show greater and ever, it refers to childhood experiences: Where the person grows more varied paraphilias without contradicting this article's main up affects the male more than the female. Cities undoubtedly offer hypothesis about female plasticity. The most plausible suggestion, a substantially higher likelihood of encountering gay people, as in my view, is that males actually do have a brief period of compared with rural life, and if males are subject to sexual im- plasticity during childhood, after which the sexual patterns are printing early in life, the childhood environment could have a reasonably rigid. Such a difference in childhood would not run stronger effect on the male. against the substantial body of evidence reviewed in this article, One reviewer of a previous draft of this article proposed that men have higher erotic plasticity based on two facts: First, men are which has depicted female sexuality as more socioculturally malleable during adulthood. more willing to have sex with many different partners (see, e.g., The experimental evidence on sexual imprinting in sheep and Buss & Schmitt, 1993; Clark & Hatfield, 1989), and second, goats (Kendrick et al., 1998) is consistent with the view that males women have a longer decision process about whether to have sex are sexually malleable during childhood and inflexible during with a particular man. The greater willingness of men to have sex adulthood. Early imprinting effects were strong and irreversible for with different partners is probably not a result of idiosyncratic males but weak and reversible for females, indicating that female personal choice or flexible change, however; more likely, it re- sexuality remained subject to environmental influence during flects a fairly stable aspect of male sexual desire (see, e.g., Buss & adulthood to a much greater degree than male sexuality, even Schmitt, 1993; Ridley, 1993). L. C. Miller and Fishkin (1997) though males were more strongly affected by the childhood learn- asked a sample of college students how many lifetime sexual ing environment. The environmental influence during childhood partners they would like to have, and the mean male response was was unmistakable because male sheep who had been raised by over five dozen (as compared with 2.7 for women), which suggests goats would not mate with their own species but only with their that interest in multiple partners is a stable aspect of sexuality for adoptive species. many men. As for the decision process, it seems likely that a longer, more Current evidence is consistent with such a characterization of human sexuality, too. Recent Kinsey Institute work (Reinisch, complex process is more flexible and more subject to situational 1990) reported that paraphilias are now believed to originate in and social factors, rather than less. It is well established that quick, childhood (see also Money, 1990). Moreover, they appear to be automatic responses tend to be simple and efficient but inflexible, quite difficult to change during adulthood. Professional therapeutic whereas controlled, deliberate processes tend to be slow, complex, treatment relies heavily on hormone treatments or castration, both and highly flexible (see, e.g., Bargh, 1982, 1994, 1997). The of which are strong biological interventions and therefore suggest longer decision process should therefore be construed as support- that purely meaning-based interventions are not effective—a con- ing the hypothesis that female erotic plasticity is higher than male. clusion that would be very consistent with the belief that male sexuality is relatively unresponsive to social and cultural influ- What About Paraphilias? ences during adulthood. Sexual imprinting on male children might well also be used to One important and potentially contrary pattern is found in explain the homosexuality patterns that were presented earlier. The gender differences in parapbilias (also known as sexual variations view mat homosexuality is purely innate and genetic suffers from or perversions). Nearly all sources report that males engage in the implausibility that natural selection would produce genes for a EROTIC PLASTICITY 363 pattern of behavior that precludes reproduction. As noted in the Some findings might be explained on the basis of direct patri- previous section, however, some data suggest that childhood ex- archal control, but others do not fit well. The fact that women periences have a stronger effect on males than females in dictating report more choice than men regarding sexual orientation (Savin- whether one becomes a homosexual (Laumann et al., 1994), and Williams, 1990; Whisman, 1996; see also Rosenbluth, 1997, on these fit a sexual imprinting explanation. Converging evidence was voluntary heterosexuality) runs directly counter to the view that provided by Bailey and Zucker (1995), who reviewed studies that men have all the choices and women are imprisoned by rigid social sought to predict adult homosexuality from cross-sex behavior factors. Likewise, the evidence about women who in midlife start during childhood. They noted that the effects were larger and having sex with other women while still enjoying sex with men stronger for males than for females. Although there may well be suggests plasticity rather than coercion (Dixon, 1984). If the be- genetic factors responsible for both the childhood behavior and the havior genetic data continue to indicate greater genetic influence adult sexual orientation (cf. Bern, 1998), as well as possible on male than female sexuality, they too will be a powerful argu- methodological factors to consider, these findings are consistent ment that the female sex drive is indeed more socioculturally with the view that male adult sexuality is more firmly and irrevo- malleable. cably shaped during childhood than female sexuality. The greater power of childhood imprinting on males is also The patriarchal oppression theory particularly invokes the socalled double standard, under which certain acts are more permis- suggested by recent findings on sexual dysfunction. Using the sible for men than for women. This view has difficulty explaining NHSLS data set, Laumann, Paik, and Rosen (1999) found that many of the modern findings, however, because the double stan- childhood sexual experience (i.e., being touched sexually before dard has been difficult to document in modern research and many puberty) was much more likely to lead to adult sexual dysfunctions researchers have concluded that it has disappeared or is disappear- in males than females and that it also predicted more different ing, especially among women (DeLamater & MacCorquodale, types of dysfunction. Women are not immune to effects of trauma, 1979; Sprecher, 1989; Sprecher & Hatfield, 1996; cf. Robinson et and indeed, adult victimization such as rape has strongly adverse al., 1991). If American college students do not endorse a double effects on women's sexuality, but the childhood experiences have standard, then the many findings based on them cannot easily be a greater effect on men. explained by reference to that standard. A double standard might These findings raise the possibility that there is a brief devel- help explain some of the older data, but even such arguments are opmental window of opportunity during which the male sex drive questionable: For example, T. Smith (1994) reports that national is malleable. A sexual imprinting stage may be biologically man- (Roper) polls found only a small minority endorsing a double dated for males, during which environmental (and thus sociocul- standard in 1959 (8%) and even 1937 (7%). In any case, the double tural) influences can exert a strong effect. The hypothesis of standard cannot account for a large part of the evidence covered greater female plasticity thus may have to recognize childhood here, even if it may have influenced an occasional finding. experiences as an exception. Still, from adolescence onward, it appears to be the females who are more flexible. The greater consensual lesbianism in prison (as compared with consensual homosexuality among imprisoned males) would be interpreted by the selective control explanation as a sign that Selective Control and the Double Standard prison frees women from the compulsive heterosexuality enforced by society. This alternative explanation thus rests on the doubtful Although the preceding sections have offered ample evidence assumption that women are more free in prison than out of it. It consistent with the hypothesis of greater female erotic plasticity, it also suggests that when a woman reverts to heterosexuality after is necessary to consider one major alternative explanation: male prison, she is simply coming back under the control of patriarchy. control over female sexuality as reflected in the double standard. These views stretch the bounds of plausibility. Most situational The essence of this view is that cultural and social factors selec- analyses would conclude that people are less free in prison than out tively target their efforts to control sexuality at women. That is, of it—especially with regard to sexual choices. culture permits male sexual desire and activity to follow their own As already reported, female sexual behavior varies far more than course, whereas it tries to control and stifle women's sexuality, and male as a function of education. The selective control hypothesis that is why many of the present effects were found. In this view, holds that women are no different than men but have been sexually culture is essentially patriarchal, which is to say that it seeks to stifled by patriarchal society and so education creates only the control and exploit women for the benefit of men. It therefore illusion of change insofar as it frees women from their exploited selectively tries to control women, not because women are easier status, enabling them to become like men. Sure enough, in many to control, but because women have less power and because the cases education produces a convergence between men and women, male-dominated culture seeks to shape female sexuality so women so that highly educated women resemble men. In some cases, can best serve men's wishes. though, the effect is not in this direction, and women become As already noted, the argument about superior male power can unlike men when highly educated. Wilson (1975) found that agree- be advanced as either an explanation for plasticity or an alternative ment with the view that sexual desires and sexual realities diverge explanation for the findings. The former is covered in the next was similar between educated men, uneducated men, and unedu- section, along with other possible explanations for plasticity. The cated women, whereas educated women held a quite different issue here is whether it is possible to explain all the findings as a view. Weis et al. (1992) found that sex education produced bigger direct result of patriarchal exploitation of women and its various increases in tolerant, permissive attitudes in females than males, consequences (including emancipation through education and the regardless of which gender started off more tolerant—thus, in sexual revolution)—and hence to reject the hypothesis of greater some cases, females were initially more tolerant and became even erotic plasticity among females. more so after education. The NHSLS found that across increasing 364 BAUMEISTER levels of education, women varied from less to substantially more interested than men in homosexual experiences (Laumann et al., 1994) and that education also made women more different than men in contraceptive use. Religion was also shown to have greater effects on female than male sexuality. Someone might argue that religion is a tool of male oppression (which entails suppressing female sexuality) whereas education liberates women and allows them to discover and pursue their own desires. This explanation has difficulty explaining the powerful historical facts that Christianity has long appealed to women more than to men, both during its rise to power in the Roman empire (see Stark, 1996) and during the transition into the modern era (Cott, 1977), and that even today female church attendance and membership rates are higher than male. The selective control explanation seemingly must propose that women wanted to be exploited and sexually stifled by Christian doctrines (and still do), a stance that seems sufficiently questionable as to call for strong supporting evidence before it can be accepted. Moreover, if religion is a tool of patriarchy that shapes women to serve men, then the highly religious should show the greatest convergence between the genders in practices that serve men, such as fellatio. The evidence indicates the opposite, however: The fellatio gap between men and women is greatest among the most religious people (Laumann et al., 1994). Religious women are least likely to serve their husbands in this and similar ways and fall most short of their husbands' preferences. Similar findings emerge from studies of the influence of peer groups. Evidence indicates that female sexuality is more influenced than male sexuality by the peer group (Billy & Udry, 1985; Mirande, 1968; Sack et al., 1984). If socializing influences reflect the patriarchal male culture, it is necessary to assume that female peer groups are instruments of patriarchy, at least sometimes, and this too seems dubious. Most societies encourage heterosexuality for the sake of reproduction, and certainly the Judeo-Christian tradition is strongly opposed to homosexuality in both genders. These Christianinfluenced societies have apparently been more successful at stifling female than male homosexuality, insofar as there are more gay males than females (Laumann et al., 1994), and this greater success comes despite generally greater condemnation of male homosexuality (see, e.g., Herek, 1988). Moreover, among homosexuals, more females than males engage in heterosexual activity, as already reported. Women thus appear more socially flexible than males on both scores. To conclude: Evidence clearly supports the historical reality that males have generally enjoyed superior sociopolitical power. It also seems plausible, although the evidence is weaker, that males have used this power to constrain female sexuality. To incorporate this insight into the analysis of erotic plasticity, it seems far more plausible to suggest that female plasticity is a result of or response to superior male power than to suggest that male power directly caused all the behavioral and attitudinal effects reviewed here without any need to invoke differential plasticity. In other words, male political power and the double standard may offer an explanation for erotic plasticity, but they are not fully viable as an alternative explanation. Other Alternative Explanations The extent and variety of evidence make alternative explanations rather difficult to propose for the entire body of evidence, although specific findings can be subject to such explanations. The large mass of evidence that sociocultural factors predict women's behavior more strongly than men's might be questioned by suggesting that data for women are more reliable than for men, possibly because women furnish more accurate self-reports or are subject to fewer sources of error variance than men. Such an explanation however would also propose that attitude-behavior consistency ought to be higher among women than men, whereas in fact it was lower. With self-report data, one often worries about the possibility of experimenter bias and demand characteristics, especially if the researcher is not blind to condition. With regard to erotic plasticity, however, the possibility of researcher bias can be almost entirely ruled out, because none of the researchers appears to have been actively looking for greater malleability among females. Most researchers covered in this review paid no attention at all to the gender difference in effect sizes. The suggestion that dozens of different researchers systematically biased their data to provide confirmation of a hypothesis mat they did not even acknowledge seems implausible. A final alternative explanation would be that the effects reviewed here merely show changes in overt behavior rather than inner, psychological changes in women. This view suggests that erode plasticity may be greater for females than males but that the difference is chiefly in terms of behavioral choices rather than inner states. Against this view, it is easy to cite evidence that women's sexual attitudes change substantially and significantly as a function of education, peer influence, and other factors. Whether specific feelings of sexual desire change, however, is far more difficult to assess. The NHSLS found that education had a bigger effect on women's rating of the appeal of various sexual practices, including homosexual activity and various heterosexual activities (Laumann et al., 1994), but it is not entirely clear whether those findings should be considered as manifestations of specific desires or general attitudes. The measurement of sexual desire (e.g., situation-specific sexual arousal) is undoubtedly more difficult than the measurement of behavior or attitudes, and so it has received less study. One recent investigation by Regan and Berscheid (1995) handled the problem by surveying people about their beliefs about the causes of sexual desire, which is at least a valuable first step. Consistent with the plasticity hypothesis, both men and women agreed in perceiving men's sexual desires (far more than women's) as arising from intraindividual forces, which would therefore be relatively independent of the situation and presumably consistent across broadly similar circumstances (assuming the men were healthy). Both genders also agreed in characterizing female sexual desire (as compared with male) as far more dependent on person by situation interaction effects, and these would certainly be much more variable than the intraindividual causes. This study thus provides preliminary evidence that desire conforms to the same patterns of male consistency and female malleability, but much further work is needed. Given the present state of evidence, the gender difference in erotic plasticity is far better supported with respect to attitudes and behavior than desire itself. 365 EROTIC PLASTICITY The causal question is most relevant to sociocultural factors General Critique of Evidence (i.e., the second prediction). The evidence about intraindividual In general, data on sexual behavior fall short of the highest variability is far less compromised by that issue. Likewise, the standards of methodological rigor. It is often impossible to conduct evidence about attitude-behavior consistency is not greatly dimin- laboratory experiments to test causal hypotheses about sexual ished by the limitations of correlational data. Thus, there is a activity. Most findings are therefore correlational. Surveys and substantial amount of evidence that is not affected by issues of interviews rely on self-report data, and with sexuality, there are causal direction. multiple factors that can distort such data, including social desirability biases, wishful thinking, memory biases, and Baseline differences and ceiling effects. With some findings, self- baseline rates are substantially higher among males than among deception. In general, however, these sources of bias do not females, which raises the possibility that ceiling effects concealed provide clear alternative explanations for findings of greater mal- male plasticity. This possibility seems contradicted by many find- leability of female sexual behavior than male. ings in which greater female plasticity was found despite low base Several limitations are relevant and rates overall (e.g., anal sex) and by findings in women that ended pervasive. Most of the findings regarding the sociocultural factors up above the putative male ceiling (e.g., contraception during Direction of causality. are correlational, partly because both ethical and pragmatic diffi- infidelity, desire for same-gender sex). In such cases, clearly, the culties preclude full experimental study of sexual behavior. Alter- effects cannot be attributed to a ceiling effect for males. native explanations could therefore be raised. For example, instead Given that males tend to be more sexually permissive than of concluding that educational and religious institutions have females overall, it is not surprising that many variables that in- stronger effects on female than on male sexual behavior, perhaps crease sexual behavior encounter the baseline problem. It is there- women's sexual inclinations dictate (more than men's) how much fore instructive to examine factors that restrict or decrease sexual education they pursue and how religious they become. Although behavior because, with these, it is the males who have more room such explanations do not seem highly plausible a priori (e.g., why to change. If the evidence for plasticity consisted of artifactual would engaging in anal sex increase a person's likelihood of findings based on ceiling effects and baseline differences, then one earning a master's degree?), they cannot be ruled out with avail- would expect sex-restricting causes to have stronger effects on able data. One study that attempted to disentangle these competing males than on females. Repeatedly, however, the opposite has been causal pathways concluded that adolescent sexuality and religios- found to be correct. I summarize some of that evidence here. ity are marked by reciprocal causal influences: Religious adoles- The most salient sociocultural factor that causes restriction in cents are less likely to have sex, and adolescents who do have sex sexual behavior is religion, insofar as religious people tended to tend to become less religious (Thornton & Camburn, 1989). Why report lower levels of most sexual activities than nonreh'gious sexually permissive women would seek and achieve higher levels people. If men start off more sexually permissive than women, of education is however a mystery, and it seems more plausible to then they ought to have more room to be affected by religion, and propose that education affects sexuality. so religiosity should produce bigger changes among men than On some variables, reverse causal explanations are fairly plau- women. The opposite has been consistently found, which supports sible. The correlations of sexual behavior with peer attitudes and the hypothesis of female erotic plasticity and contradicts the arti- behavior, for example, could well reflect a tendency for a person fact explanation. to self-select similar peers instead of simple direct influence of A determined advocate of the ceiling artifact might dismiss the peers on the person's sexual behavior. Some studies have been findings about religion by suggesting that religion mainly tries to alert to this methodological problem and have had some success control female behavior and is relatively indifferent to male sexual ruling it out, however. Billy and Udry (1985) found no difference behavior. This view is implausible on several counts. As Tannahill in peer group homogeneity between males and females nor any (1980) pointed out in her history of sex, early Christianity was tendency (in either gender) to deselect friends based on discrepant more hostile and restrictive toward sex than any of its contempo- sexual status. As for acquiring new friends based on similarity in rary religions, and its restrictiveness applied to both genders. The sexual status, males were more selective in this regard, which basic Christian view was that "physical pleasure of all kinds is would operate against the pattern of higher self-peer correlations sinful" (DeLamater, 1981, p. 264). This doctrine appears to have among females. Thus, the greater predictive impact of peers on had strong appeal to women, and in fact, the early rise in Christian females than on males cannot be dismissed as an artifact of peer church membership involved a more rapid expansion of female selection patterns. The convergence between peer influence and than male members (Stark, 1996). Celibacy was in fact sought and family influence is also reassuring on this score because children cultivated as a lifestyle by many early Christian women (Mc- cannot select their families. Namara, 1985). It seems most plausible that Christianity sought to One may also consider the possibility that third variables ac- control sexuality of both males and females but succeeded better count for some of the correlations. For example, the link between with females—which again would point toward greater plasticity education and permissive sexuality could conceivably derive from among females. This conclusion is also well supported by the parental openness to new ideas, insofar as parental openness en- evidence about religious celibacy: When identical standards of courages both pursuit of advanced degrees and of sexual adven- sexual purity are held up to both men and women, the women are ture. Still, this analysis does not truly constitute an alternative far more successful at meeting them (Murphy, 1992; see also Sipe, explanation because, to account for the greater link between edu- 1995). cation and sex among women than men, one would have to Another important point is that the baseline and ceiling argu- postulate that parental openness has a stronger effect on daughters ments apply mainly to the evidence about specific sociocultural than sons—which again would indicate greater female plasticity. variables. They do not seem relevant to the intraindividual vari- 366 BAUMEISTER ance or attitude—behavior consistency evidence. Thus, even if the Assessment of Possible Explanations baseline problem were serious, it would only undermine one of the three predictions of the plasticity hypothesis. Three possible explanations for the gender difference in erotic Thus, baseline differences and ceiling effect problems are rele- plasticity are now considered. Although these explanations can be vant to only a small part of the evidence for plasticity. In all considered as competing, they are not mutually exclusive, and it is fairness, it is quite possible that some individual findings of greater possible from an a priori standpoint that more than one could be variation among women do indicate a ceiling effect for males. It correct. does not seem plausible that such artifacts constitute a substantial amount of the evidence for plasticity, given that some findings Male Strength and Power directly contradict this explanation and others are immune to it. Several strong features of the literature The first explanation is that men are generally stronger and more reviewed here are encouraging with respect to the validity of the physically aggressive than women, as well as generally holding conclusions. First, a wide range of methods and populations has superior political, social, and economic power, and so women have Strengths of evidence. been used. Consistency of evidence across multiple methods can to accommodate themselves to men. In this view, men can coerce help substantially in overcoming concerns about methodological women to do what they want, and so as men pursue their sexual weaknesses because a bias or artifact in one method would likely desires, women must go along with what men want to some extent. be absent from some others, so if conclusions are similar, confi- Even if the man rarely or never uses his physical or political dence in them increases. It is unlikely that multiple methods and advantage to get his way, the fact that he could do so remains approaches would all share the same biases or artifacts. implicit and could affect relationships. Second, the consistency of findings is itself persuasive. Across It is common knowledge that men are physically stronger than three major predictions and a wide assortment of methods, the women on average and that men have generally had superior evidence pointed consistently to greater erotic plasticity in the political and economic power. Men also exceed women in aggres- female than in the male. The main exception (which was also siveness (Eagly, 1987; National Research Council, 1993). In ro- consistent) is that male sexual behavior is often constrained by mantic and sexual relationships, men do sometimes inflict harm on lack of opportunity, and so many men would like to have more sex women in connection with various disputes, and in some cases (or more partners or different varieties of sex) than they are able to men use physical force to obtain sex from their romantic partners have. Apart from opportunity constraints, however, female sexu- (Laumann et al., 1994). Men's superior socioeconomic power also ality shows greater responsiveness and flexibility than male seems to give them some leverage toward eliciting sex from their sexuality. wives and partners (Blumstein & Schwartz, 1983). A last encouraging feature is that the evidence for female Thus, gender differences in physical strength and political plasticity remains robust and is perhaps even strongest in the power have been long-standing and have had some effect on methodologically most rigorous work. If the pattern of female sexual relations. Whether these differences could be responsible plasticity were an artifact of sloppy methods, then its evidence for the gender difference in sexual plasticity is far more difficult to should diminish in proportion to the rigor of the methods, but the say. There are at least two ways that this could be true. In one, opposite has been found. Thus, in the NHSLS (which used some of evolution may have made female sexuality more plastic (socially the most rigorous, thorough, and careful methods), a substantial malleable) because of millennia of having to adapt to stronger, number of pointed comparisons repeatedly confirmed greater plas- politically dominant males. In the other, women continue to be ticity in females than in males. Differential critique. To criticize the evidence for the three conscious of the greater power held by males and hence learn to be more malleable and flexible as a result. predictions separately: The evidence about intraindividual vari- One relevant aspect of this explanation that is different than the ability (the first prediction) is sufficiently strong, diverse, and other two is that it is not confined to sexuality. If men's greater consistent to be satisfactory for now, although further evidence physical strength causes women to be more malleable and flexible would be desirable, particularly in regard to issues such as change as a submissive adaptation, this should presumably be true across in prison environments and the like. Second, the evidence for multiple spheres. Women should therefore show greater flexibility greater responsiveness to sociocultural factors is fairly extensive and malleability on multiple measures apart from the sexual and convincing, although it would be desirable to have more direct sphere. In other words, sociocultural factors should have stronger studies of cross-cultural variation and more longitudinal designs or effects on women than men, intraindividual variability across time other methods that can overcome the ambiguity about direction of should be greater in women, and attitude-behavior consistency causation. The evidence regarding attitude-behavior consistency should be generally lower. The question of whether women are is encouraging as far as it goes, but there are many gaps in the more malleable than men as a general principle across the majority range of possible evidence, and conceivably major exceptions or of spheres of behavior is far beyond the scope of this article. Note, even a large contrary pattern could yet be found. The attitude- however, that experts on gender differences have not thus far behavior prediction is therefore the most weakly supported of the recorded any widespread pattern of greater malleability in females, three, although even on that prediction, the evidence is generally which casts doubt on this as an explanation. Although both gender supportive. Still, the attitude-behavior prediction is the least cen- differences and attitude-behavior consistency have been studied tral of the three, insofar as many factors other than basic erotic extensively, I have been unable to find any evidence of a general pattern of lesser attitude-behavior consistency among women, plasticity could affect such data, and so the relative weakness of the literature on that question is least worrisome with respect to the even after contacting authoritative experts in the field (A. H. present theory. Eagly, personal communication, 1998; R. E. Petty, personal com- EROTIC PLASTICITY 367 munication, 1998). There is some evidence that women are more were rated as more typical of men than women. Clearly, these easily persuaded than men under a variety of conditions, although participants associated seeking sex with maleness and refusing sex numerous exceptions and boundary conditions exist (see, e.g., with femaleness. If that is correct, then sex would depend on the Eagly, 1987; Petty & Wegener, 1998). Meanwhile, recent work woman changing from refusal to acceptance. has found that females show more genetic and less sociocultural Further evidence about the female script and the transition from influence on aggression than males (Eley, Lichtenstein, & Steven- no to yes comes from research on erotica and pornography. Cowan son, 1999; see also Christiansen, 1977), which likewise depicts the and Dunn (1994) exposed both male and female participants to difference in erotic plasticity as specific rather than part of a pornographic films that were classified into nine different story general pattern. If future research continues to suggest that women themes, and participants were asked to rate their arousal levels. are not more socioculturally malleable than men across the board, One of these themes, labeled "submission" by the researchers, then the explanations based on differential strength and power will involved a woman who was initially reluctant to have sex but be less plausible than the following two. changed her mind during the scene and became an active, willing participant in sexual activity. Women rated this theme by far the Change and the Female Sexual Script most sexually arousing of the nine (see also Fisher & Byrne, 1978). These studies thus suggest that the woman's transition from A second possibility is that change is an inherent part of the no to yes, as an idea, increases sexual excitement. female role in sex, and so women are required to have some degree A review of the literature on sexual fantasies found that fanta- of flexibility in their patterns of erotic response. In this view, the sies of being overpowered and forced to have sex were far more standard script for sex between first-time human partners depends common among women than men (Leitenberg & Henning, 1995). vitally on the woman changing her mind. In nearly all known In some studies (e.g., Pelletier & Herold, 1988), over half the societies (and in many other primate species as well), females female sample reported fantasies of being overpowered, and other constitute the restraining force on sex. That is, they refuse many research found a third of women endorsing such specific fantasies offers or chances for sexual activity. When sex happens, it is as being a slave who must obey a man's every wish (Arndt, Foehl, because the woman has changed her vote from no to yes. This & Good, 1985). When women are given lists of sexual fantasies to crucial change might be the basis for greater erotic plasticity in choose among, that of being forced sexually is sometimes the first women, because it instills a capacity for change at the center of the or second most frequently chosen one (Hariton & Singer, 1974; female sex drive. Knafo & Jaffe, 1984). In a study of the content of fantasies people There is evidence that the decisive determinant of whether a have during intercourse with a partner, Sue (1979) found that couple has sex involves the women changing her position from no women were significantly more likely than men to fantasize about to yes. It is well documented that in heterosexual attraction, the being overpowered and forced to have sex. Leitenberg and Hen- man is typically ready for sex long before the woman (Buss & ning cautioned further, as other researchers have, that such fanta- Schmitt, 1993). Men are more willing than women to have sex sies do not reflect any genuine desire to be raped, and indeed, the with someone they have just met (see, e.g., Herold & Mewhinney, fantasies often involve the man overcoming the woman's token 1993; see also Oliver & Hyde, 1993). The precise prediction that resistance so as to bring about mutual pleasure and satisfaction. women will change more than men toward a more sexually per- Thus, these fantasies likewise suggest that a particular sexual missive attitude as a function of increasing duration of dating was charge is associated with the woman's changing from no to yes, confirmed by Harrison et al. (1974). under strong male influence. Also relevant is the fact that men fall in love faster than women hi sum, the transition from no to yes appears to characterize the and hence are likely to feel loving affection and the accompanying female role in sex. It also appears to be marked by a special sexual desire at an earlier point in the relationship (Baumeister, emotional charge and high sexual excitement. It is therefore con- Wotman, & Stillwell, 1993; Hill, Rubin, & Peplau, 1976; Huston, ceivable that this transition, requiring as it does a diametrical Surra, Fitzgerald, & Cate, 1981; Kanin, Davidson, & Scheck, reversal in women's attitude toward having sex widi a particular 1970). Studies of adult virginity have found that many more men man, may have some role in the broader phenomenon of erotic than women report that they have remained virgins because their plasticity. romantic partner refuses sex (McCabe, 1987; Sprecher & Regan, A variation on this explanation is based on the observation that 1996). Even apart from virginity, far more men than women cite a there is a chronic pattern of mismatch between when a woman partner's unwillingness as a major reason that they are not having wants sex and when she has it, so she has to be flexible enough to sex (Mercer & Kohn, 1979). Both genders agree that men want and participate positively and competently in sex when she does not expect sex earlier in a relationship than women (Cohen & Shot- particularly want it. This view is well expressed in Wallen's (1995) land, 1996). point that it is essential to distinguish between receptivity (will- Direct evidence about refusing sex was provided by Clark and ingness) versus desire when talking about female sexuality, a Hatfield (1989). In one condition of their study, participants were distinction that is far less important with male sexuality and hence approached by an opposite-sex research confederate who invited with traditional male-centered theories of sex in general. To ex- the participant to have sex that same evening. All the women in plain when and whether a female has sex, according to Wallen, it both studies refused this invitation, whereas most of the men is more useful to understand receptivity than actual, proactive accepted (see also Jesser, 1978). By the same token, Mercer and desire. Kohn (1979) found that both male and female participants rated all Evidence for this theory is based on temporal patterns in sex. different strategies of avoiding sex as more typical of women than Palmer, Udry, and Morris (1982) found that intercourse patterns of men, whereas all the strategies for initiating and obtaining sex for couples had clear daily and weekly patterns but not monthly 368 BAUMEISTER patterns. That is, couples tend to have sex at a particular time of within gender. For example, in an all-male sample, would the men day (usually in the evening) and on some weekdays more than with weaker sex drives be more affected by education, religion, or others (Sundays especially). Other research on female sexual de- situational influences than men with slrong sex drives? This ques- sire suggests however that monthly variations are significant and tion remains for future research. important (Stanislaw & Rice, 1988). In other words, women feel most sexually desirous at a particular point in the menstrual cycle, General Discussion usually one associated with ovulation (Stanislaw & Rice, 1988; also Luschen & Pierce, 1972; see Wallen, 1995). Putting these findings together indicates that women's sexual behavior does not correlate most strongly with their desires. The monthly rhythm of rising and falling sexual desire does not predict their likelihood of intercourse. The implication is that women in general are flexible enough to have sex when they do not most want it. This flexibility points to the importance of receptivity rather than desire and could also provide a basis for a more general The central question addressed in this article has been whether the female sex drive is more plastic and malleable than that of the male, in response to social, cultural, and situational causes. The evidence reviewed here supports the three basic predictions derived from that hypothesis. First, intraindividual variation (personal change) is higher among females than among males. The average woman is more likely to change her sexual patterns over the course of adult life than the average man is, in such areas as pattern of erotic plasticity. discontinuity in total orgasmic outlet, adaptation within marriage, adoption of new activities over the adult years, and changes in Do Women Have a Milder Sex Drive ? sexual preference. Second, sociocultural factors such as education, The third explanation is that women have a milder, weaker sex drive than men and that this difference allows the female sex drive to be more easily molded. It is common knowledge that in taming animals, which is to say bringing their behavior under meaningful religion, political ideology, acculturation, and peer influence generally have stronger effects on female sexuality than on male. Third, females exhibit less consistency between sexual attitudes and behavior on a variety of measures, including attitudes about rules determined by somebody else, that the weaker their urges are, virginity, approval of extramarital or extradyadic sexual activity, the easier they are to tame. If women's desire for sex were less intended condom usage, having sex despite not wanting it, and powerful, less relentless, and less urgent than men's, then as a interest in or desire for same-gender sex. The low attitude- result, their sex drive could well be more malleable. behavior consistency among women presumably occurs because Although a full review is beyond the scope of this article, that evidence does suggest that women have a milder sex drive. sex depends on many specific contexts, circumstances, and other meanings, and so the broad attitudes are poor predictors. Women report spontaneous sexual desire less often than men and Two main exceptions have been found. The first is relatively think about sex less often than men (Beck, Bozman, & Qualtrough, trivial: Men sometimes exhibit low attitude-behavior consistency 1991; Eysenck, 1971; Knoth, Boyd, & Singer, 1988; Laumann et because of lack of opportunity. Many men are unable to find a al., 1994). They have fewer sexual fantasies involving fewer willing partner, and so they cannot act consistently with their partners and less variety of activity (Ellis & Symons, preferences. 1990; Leitenberg & Henning, 1995). Women report less enjoyment of The second exception, however, suggests a theoretically impor- erotica and pornography (see, e.g.. Reed & Reed, 1972; Schmidt & tant qualification to the female plasticity view. Evidence about Sigusch, 1970; Sigusch, Schmidt, Reinfeld, & Wiedemann-Sutor, sexual dysfunction, paraphihas, cross-gender behavior, and locale 1970). They desire less frequent sex and fewer sexual practices of upbringing suggests that childhood experiences have stronger than men (Ard, 1977; Bergstrom-Walan & Nielsen, 1990; Julien, and more lasting effects on male than female sexuality. In adoles- Bouchard, Gagnon, & Pomerleau, 1992; Laumann et al., 1994). cence and adulthood, erotic plasticity is higher among females, but Women initiate sex less often and refuse it more often (Byers & male sexuality may undergo a childhood phase (akin to imprinting Heinlein, 1989; Clark & Hatfield, 1989; LaPlante, McCormick, & in animals) during which social and environmental influences can Brannigan, 1980; O'Sullivan & Byers, 1992). Women desire fewer have a major influence. Ironically, the relative inflexibility of adult partners than men (Buss & Schmitt, 1993; Miller & Fishkin, 1997) male sexuality may entail that these childhood influences have and seek out fewer extramarital partners (Cotton, 1975; Lawson, strong and durable effects. In contrast, the plasticity of adult 1988; Spanier & Margolis, 1983; Thompson, 1983). Women and female sexuality may permit the effects of childhood experiences girls masturbate less often than men and boys (Arafat & Cotton, to be overridden. With regard to sexual abuse and dysfunction, 1974; Asayama, 1975; Laumann et al., 1994; Oliver & Hyde, such plasticity would constitute an important adaptive benefit of 1993; Sigusch & Schmidt, 1973). Women rate their sexual urges as female plasticity. less strong than men rate men's (Mercer & Kohn, 1979). Women The general conclusion from the adolescent and adult evidence are more likely to cite lack of interest and enjoyment as a reason is that the balance between nature and culture is different for the for not having sex (Leigh, 1989). two genders, at least in terms of their sexuality. Men's sexuality Can the relative mildness of female sexual desire explain plas- revolves around physical factors, in which nature is predominant ticity? Once again, it is far easier to establish that something is and the social and cultural dimension is secondary. For women, correct than to establish its link to erotic plasticity. On an a priori social and cultural factors play a much greater role, and the role of basis, it would seem easier to transform a desire for A into a desire physical processes and biological nature is relatively smaller. for B if the desire for A is not as strong. Still, direct evidence of These findings reverse one cultural stereotype, which is that civ- the link is lacking. The most relevant research agenda would be to ilization is male whereas women are closer to nature. In sexuality, examine possible links between strength of sex drive and plasticity at least, women are the creatures of meaning (which invokes the EROTIC PLASTICITY sociocultural contexts), whereas men are the creatures of nature. (Of course, these differences are relative, not absolute.) The large preponderance of supporting evidence thus supports the firm conclusion that the female sex drive is in fact more malleable than that of the male. It must be acknowledged that essentially none of the studies reviewed was intended to provide a direct test of the hypothesis of female plasticity, and so some prospective tests may be warranted. Still, the fact that researchers were not specifically looking to establish differential plasticity lends further confidence to the conclusion because it rules out any concern that their results are due to experimenter bias, demand characteristics, or selective testing of the hypothesis. To put it simply, researchers have repeatedly confirmed that women's sex drives are more malleable even though they did not intend to show this and, indeed, generally failed to suspect that this feature of their data may have fit a more general pattern. It would however be useful for further research to search for boundary conditions, counterexamples, and mediating or moderating factors regarding differential plasticity. Any exceptions to the general pattern of erotic plasticity would add valuable insight into gender differences in sexuality. Attitude—behavior inconsistency is the least thoroughly supported of the three major predictions, and so this may be an area for further work that might either provide useful confirmation of the broad plasticity hypothesis or, indeed, reveal exceptions and boundary conditions that would be theoretically enlightening. Three possible explanations for gender differences in erotic plasticity have been suggested: differences in power and strength, the requirement of change as part of the female sexual script, and a relative mildness in female sexual desire. On the basis of currently available evidence, I conclude that each of these three possible explanations has a valid basis, but at present, there is little conclusive evidence to suggest which of them is actually linked to erotic plasticity. Proximal causes also remain to be explicated. These may include direct genetic influence, such as the notion that having two X chromosomes gives women alternative blueprints for sexual responses, whereas males have only the one. Sexual imprinting may be more influential and irreversible with males than females, so that women continue to have sexually formative experiences throughout life whereas men have them only at one early point. Hormones may affect men more than women, either because men have more of the most influential hormone (testosterone) or because men are more directly attuned to their inner states without any mediation through social cues and information (T. Roberts & Pennebaker, 1995). Another intriguing possibility is suggested by Bern's (1996, 1998) theory proposing that sexual orientation is shaped by childhood social patterns, such that the less familiar gender becomes the more arousing and sexually appealing one (i.e., exotic becomes erotic). Bern's theory could be reconciled with several of the explanations I have suggested. The notion that females have more mixed-gender social groups in childhood than males do leads, in Bern's analysis, to suggesting that female sexual arousal would distinguish less between males and females than would male sexual arousal. This plasticity of sexual orientation could contribute to more general patterns of plasticity (i.e., change begets change), as with the second explanation I proposed. Alternatively, one could use Bern's line of reasoning to suggest that, because 369 both genders are familiar to females, neither is exotic, and hence neither is likely to become erotic—and this could contribute to my third explanation, namely, the weaker female sex drive. In terms of the daily lives and actual experiences of individual men and women, the difference in plasticity may be felt in terms of the relative importance of physical factors versus social meanings. The importance of social, situational, and cultural influences on women suggests that sex depends very prominently on the meanings and interpretations that a given sex act may have. The relative inflexibility of males with regard to sociocultural factors suggests that meanings matter less than simpler, physical aspects of sex. The current status of knowledge may therefore be described as follows. The female sex drive is more plastic and malleable than that of the male; several well-founded explanations for this differential plasticity can be articulated, but there is no adequate basis at present for preferring any of these explanations over the others. It is also possible that all three explanations are correct and that the difference in plasticity is multidetermined. Implications If women are indeed more responsive to sociocultural changes, then forecasting the future shape of sexuality will be less reliable when it comes to women than men. Had someone at the close of the 19th century sought to predict what would happen in the 20th century, he or she might have been fairly accurate at predicting men (because they have not changed much), but predicting the fluctuations and vicissitudes of female sexuality would have been considerably more difficult. It is, in other words, far more difficult to predict what women will want and expect sexually a century hence than it is to predict men's wants and expectations. From the point of view of society, the gender difference in erotic plasticity suggests that it will be more productive and effective to try to control female than male sexuality. It is possible that a society's survival would be jeopardized by historical events that might require more reproduction (e.g., due to war or famine) or less (e.g.. overpopulation) or that the desirability of promiscuity would increase (e.g., if the sex ratio departs far from equality) or decrease (e.g., if AIDS or another venereal disease raises health risks). A society that needs a change in sexual behavior in order to survive or flourish would do better to target its messages and other pressures at women rather than men because of the greater difficulty in changing the sexual desires and habits of of men. From the point of view of the individuals, women will be better able than men to adapt to new social conditions and demands. If social conditions do change in a meaningful fashion, resulting in a need for serious changes in sexual attitudes and behavior, women are likely to make the adjustment better. True, one might argue on the basis of the relative mildness of the female sex drive that sex matters less to women than men and so women might be more willing to accept different circumstances and contingencies. However, even if changes in active desires and behaviors are needed, women should make these more easily. This may be particularly important if the pace of social change continues to accelerate, as is generally assumed to be the trend in modernity. In sex, at least, women should be able to keep up with changing times better than 370 BAUMEISTER Some misunderstandings and potential conflicts between the order to ascertain what they truly desire in sex, whereas men would genders could be affected by the difference in plasticity. Modern have difficulty appreciating how so much introspective exertion norms of egalitarianism and equitable relationships suggest that could be required for the sake of understanding one's own sexu- people should compromise and seek joint, mutually satisfying ality. The sexual consciousness-raising of the women's movement decisions, but the calculation of compromise is rendered more during and after the sexual revolution, which never evoked much difficult by differential plasticity. In simple terms, sexual compro- of an echo among males, could reflect the greater difficulty of mise will be easier for women than men. I cited several findings understanding female sexuality due to its greater plasticity. Con- indicating that women find their sexual relationships more satis- sistent with this view, some recent findings suggest that women are fying than men, which could reflect the women's greater plasticity. less certain than men of what they want in sex and how to get it For example, in response to Ard's (1977) survey, women indicated (Vanwesenbeeck, Bekker, & van Lenning, 1998). that their actual frequency of intercourse was nearly identical to Sexual decision-making is also likely to be a far more complex their desired frequency, whereas for the men a substantial gap and subtle matter for women than for men. If women's sexual existed. One interpretation of this finding is that women have desires and actions are strongly influenced by sociocultural factors, greater power and can dictate the sexual terms of the relationship. then the social context and situation would potentially be able to Another, however, is that women are better able to adjust their alter the desirability of performing a particular sex act with a preferences and expectations to what is actually available to them, particular partner. For men, in contrast, performing a particular sex and so a compromise gradually ceases to seem like a compromise. act with a particular person may be a straightforward decision Conflict and misunderstanding can exist between members of depending on salient, unchanging cues, whereas for women, the the same gender, too, and again differential plasticity could play a answers might fluctuate as a function of a host of intangible social role. Homosexual communities, for example, are in a sense op- and contextual factors. pressed minority groups and ones from which members may be Last, the ongoing debate as to the degree of influence by nature tempted to defect. If people leave such communities and join the or culture could well end up being somewhat artificially prolonged heterosexual mainstream, the survival of the communities could be by the gender difference in plasticity. Feminist analysis has fa- jeopardized. Given the data reviewed here, such defections are vored the social construction of sexuality, whereas the subsequent likely to be a bigger problem and threat for female than male rise of evolutionary theories has been dominated by male theorists. homosexual groups. Sure enough, lesbian communities have on- If women are indeed more socioculturally malleable than men, going and sometimes bitter struggles over defectors to heterosex- then the social constructionist theories would resonate intuitively uality, which may be less of a problem for gay male groups (see with women more than men, whereas the reverse would hold for Clausen, 1990; Rust, 1993). biological and evolutionary theories. There are clear and important implications for clinical practice. The greater plasticity of female sexuality suggests that sex thera- Concluding Remarks pists should be more effective at treating women than men. In particular, cognitive-behavioral treatments and other social inter- Human progress is generally regarded as a matter of either ventions should be much more effective with female than male reforming society so as to improve its capacity to guide people clients. The relative inflexibility of males suggests that sexual toward more fulfilling lives or, at least, allowing people freedom to problems may require more physiologically and biochemically make their own conscious choices and pursue their individual oriented interventions. Some recent evidence fits this view, al- goals. Moreover, it seems highly likely that sex, love, and mating though further research is needed. Laumann et al. (1999) found will continue to play a central part in human happiness and that male sexual dysfunction was more linked to physical factors fulfillment. The question of how much human sexuality can be such as poor overall health than was female sexual dysfunction. transformed based either on Utopian social arrangements or indi- Meanwhile, female dysfunction was more strongly linked than vidual choice is therefore one that has more than abstract theoret- male to sociocultural factors such as education and change in ical implications. If the sex drive is socioculturally malleable, then socioeconomic status (loss of income). The link to broadly mean- there exist many possible directions in which to pursue social ingful context was also evident in the fact that female sexual progress and individual fulfillment. In contrast, if the sex drive is dysfunction correlated more strongly than male dysfunction with fixed and static, then society must ultimately accommodate and broad measures of happiness and quality of life. The main excep- confront those patterns, and individual choice will be a matter of tion to these patterns was that childhood sexual experiences pre- pursuing those innate, inflexible desires. dicted male sexual dysfunction more strongly than they did female The gender difference in erotic plasticity suggests that women dysfunction, which fits the hypothesis that sexual imprinting pro- present a better prospect for achieving cultural progress than men, duces relatively strong and irreversible effects on males but not on at least with regard to sexuality. To be sure, the differences are females. relative rather than absolute, but, on both individual and collective Sexual self-knowledge, meanwhile, should be far easier for measures, there was consistent evidence that women's sexuality males than females to achieve. The male's understanding of his can adapt and change more effectively than men's. To the extent own sex drive is essentially a matter of gathering information that the road to Utopia runs through the bedroom, social engineers about a stable, fixed may find that male entity. In contrast, the female's self- inflexibility presents the greater prob- exploration is to some extent pursuing a moving, shape-changing lem whereas female plasticity represents the more promising target. The evidence for sociocultural influence and intraindividual opportunity. change could itself persuade some women that they require a long Meanwhile, the sexual responses of individuals are likely to period of inner exploration, experimentation, and soul-searching in continue to take shape in different ways, particularly with respect EROTIC PLASTICITY to the relative importance of physical versus sociocultural dimensions. The relatively low plasticity of the male sex drive suggests that biochemical factors such as hormones, age, general health, and genetic predispositions may often he the driving forces, and men's sexual wishes may be relatively indifferent to the social context. For women, in contrast, sex is driven by sociocultural factors, interpretations, context, expectations, and the like. The question of "What does it mean?"—in other words, what does a particular sex act signify and communicate—is centrally important to the female sexual experience, before, during, and after. For men, in contrast, the different possible meanings matter less, and sex might often be a perfectly fine experience even if it hardly means anything at all. These differences could make mutual intuitive understanding between men and women elusive. 371 Bartell, G. D. (1970). Group sex among the mid-Americans. Journal of Sex Research, 6, 113-130. Bauman, K. E., & Wilson, R. R. (1974). Sexual behavior of unmarried university students in 1968 and 1972. Journal of Sex Research, 10, 327-333. Baumeister, R. F. (1989). Masochism and the self. Hillsdale, NJ: Erlbaum. Baumeister, R. F., Wotman, S. R., & Stillwell, A. M. (1993). Unrequited love: On heartbreak, anger, guilt, scriptlessness, and humiliation. Journal of Personality and Social Psychology, 64, 377-394. Beck, J. G., Bozman, A. W., & Qualtrough, T. (1991). The experience of sexual desire: Psychological correlates in a college sample. Journal of Sex Research, 28, 443-456. Bell, A. P., & Weinberg, M. S. (1978). Homosexualities: A study of diversity among men and women. New York: Simon & Schuster. Bern, D. J. (1996). Exotic becomes erotic: A developmental theory of sexual orientation. Psychological Review, 103, 320-335. Bern, D. J. (1998). Is EBB theory supported by evidence? Is it androcen- References tric? A reply to Peplau et al. (1998). Psychological Review, 105, 395- 398. Adams, C. G., & Turner, B. F. (1985). Reported change in sexuality from young adulthood to old age. Journal of Sex Research, 21, 126-141. Amsterdam Sex Museum. (1999). [Exhibit notes]. Amsterdam: Author. Antonovsky, H. F., Shoham, I., Kavenocki, S., Modan, B., & Lancet, M. (1978). Sexual attitude-behavior discrepancy among Israeli adolescent girls. Journal of Sex Research, 14, 260-272. Arafat, I. S., & Cotton, W. L. (1974). Masturbation practices of males and females. Journal of Sex Research, 10, 293-307. Arafat, I. S., & Yorburg, B. (1973). On living together without marriage. Journal of Sex Research, 9, 97-106. Ard, B. N. (1977). Sex in lasting marriages: A longitudinal study. Journal of Sex Research, 13, 274-285. Arndt, W. B., Foehl, J. C., & Good, F. E. (1985). Specific sexual fantasy themes: A multidimensional study. Journal of Personality and Social Psychology, 48, 472-480. Asayama, S. (1975). Adolescent sex development and adult sex behavior in Japan. Journal of Sex Research, 11, 91-112. Bailey, J. M., & Martin, N. G. (1993, September). A twin registry study of sexual orientation. Paper presented at the annual meeting of the International Academy of Sex Research, Provincetown, MA. Bailey, J. M., & Pillard, R. C. (1991). A genetic study of male sexual orientation. Archives of General Psychiatry, 48, 1089-1096. Bailey, J. M., & Pillard, R. C. (1995). Genetics of human sexual orientation. Annual Review of Sex Research, 6, 126-150. Bailey, J. M., Pillard, R. C., Neale, M. C., & Agyei, Y. (1993). Heritable facors influence sexual orientation in women. Archives of General Psychiatry, 50, 217-223. Bailey, J. M., & Zucker, K. J. (1995). Childhood sex-typed behavior and sexual orientation: A conceptual analysis and quantitative review. Developmental Psychology, 31, 43-55. Bargh, J. A. (1982). Attention and automaticity in the processing of self-relevant information. Journal of Personality and Social Psychology, 43, 425-436. Bargh, J. A. (1994). The four horsemen of automaticity: Awareness, intention, efficiency, and control in social cognition. In R. S. Wyer, Jr., & T. K. Srull (Eds.), Handbook of social cognition (pp. 1-40). Hillsdale, NJ: Erlbaum. Bargh, J. A. (1997). The automaticity of everyday life. In R. S. Wyer (Ed.), The automaticity of everyday life: Advances in social cognition (Vol. 10, pp. 1-61). Mahwah, NJ: Erlbaum. Barry, H., & Schlegel, A. (1984). Measurements of adolescent sexual behavior in the standard sample of societies. Ethnology, 23, 315-329. Bergstrom-Walan, M.-B., & Nielsen, H. H. (1990). Sexual expression among 60-80-year-old men and women: A sample from Stockholm, Sweden. Journal of Sex Research, 27, 289-295. Billy, J. O. G., & Udry, J. R. (1985). Patterns of adolescent friendship and effects on sexual behavior. Social Psychology Quarterly, 48, 27-41. Birenbaum, A. (1970). Revolution without the revolution: Sex in contemporary America. Journal of Sex Research, 6, 257-267. Blumstein, P. W., & Schwartz, P. (1977). Bisexuality: Some social psychological issues. Journal of Social Issues, 33(2), 30-45. Blumstein, P., & Schwartz, P. (1983). American couples. New York: Simon & Schuster. Buss, D. M., & Schmitt, D. P. (1993). Sexual strategies theory: An evolutionary perspective on human mating. Psychological Review, 100, 204-232. Byers, E. S., & Heinlein, L. (1989). Predicting initiations and refusals of sexual activities in married and cohabiting heterosexual couples. Journal of Sex Research, 26, 210-231. Charbonneau, C., & Lander, P. S. (1991). Redefining sexuality: Women becoming lesbian in midlife. In B. Sang, J. Warshow, & A. J. Smith (Eds.), Lesbians at midlife: The creative transition (pp. 35-43). San Francisco: Spinsters Books. Christensen, H. T., & Carpenter, G. R. (1962). Value-behavior discrepancies regarding premarital coitus in three Western cultures. American Sociological Review, 27, 66-74. Christiansen, K. O. (1977). A preliminary study of criminality among twins. In S. Mednick & K. Christiansen (Eds.), Biosocial bases of criminal behavior (pp. 89-108). New York: Gardner Press. Clark, R. D., & Hatfield, E. (1989). Gender differences in receptivity to sexual offers. Journal of Psychology and Human Sexuality, 2, 39-55. Clausen, J. (1990). My interesting condition. Journal of Sex Research, 27, 445-459. Cohen, L. L., & Shetland, R. L. (1996). Timing of first sexual intercourse in a relationship: Expectations, experiences, and perceptions of others. Journal of Sex Research, 33, 291-299. Cott, N. F. (1977). The bonds of womanhood. New Haven, CT: Yale University Press. Cotton, W. L. (1975). Social and sexual relationships of lesbians. Journal of Sex Research, 11, 139-148. Cowan, G., & Dunn, K. F. (1994). What themes in pornography lead to perceptions of the degradation of women? Journal of Sex Research, 31, 11-21. Bart, P. B. (1993). Protean women: The liquidity of female sexuality and Croake, J. W., & James, B. (1973). A four-year comparison of premarital the tenaciousness of lesbian identity. In S. Wilkinson & C. Kitzinger (Eds.), Heterosexuality: Feminism and psychology reader (pp. 246- Darlington, R. B. (1975). Radicals and squares: Statistical methods for the 252). London: Sage. sexual attitudes. Journal of Sex Research, 9, 91-96. behavioral sciences. Ithaca, NY: Logan Hill Press. 372 BAUMEISTER DeLamater, I. (1981). The social control of sexuality. In R. Turner & 1. Short (Eds.), Annual review of sociology (Vol. 7, pp. 263-290). Palo Alto, CA: Annual Reviews. men: Correlates and gender differences. Journal of Sex Research, 25, 451-477. Herold, E. S., & Mewhinney, D.-M. K. (1993). Gender differences in DeLamater, J. D., & Hyde, ]. S. (1998). Essentialism vs. social constructionism in the study of human sexuality. Journal of Sex Research, 35, 10-18. casual sex and AIDS prevention: A survey of dating bars. Journal of Sex Research, 30, 36-42. Hershberger, S. L. (1997). A twin registry study of male and female sexual DeLamater, J., & MacCorquodale, P. (1979). Premarital sexuality: Attitudes, relationships, behavior. Madison: University of Wisconsin Press. Dixon, J. K. (1984). The commencement of bisexual activity in swinging married women over age thirty. Journal of Sex Research, 20, 71-90, orientation. Journal of Sex Research, 34, 212-222. Hill, C. T., Rubin, Z., & Peplau, L. A. (1976). Breakups before marriage: The end of 103 affairs. Journal of Social Issues, 32, 147-168. Hu, S., Pattatucci, A., Patternson, C, Li, L., Folker, D., Chemy, S., Dunne, M. P., Martin, N. G., Statham, D. J., Slutske, W. S., Dinwiddie, Kruglyak, L., & Hamer, D. (1995). Linkage between sexual orientation S. H., Bucholz, K. K., Madden, P. A. F., & Heath, A. C. (1997). Genetic and chromosome Xq28 in males but not in females. Nature Genetics, II, and environmental contributions to variance in age at first sexual intercourse. Psychological Science, 8, 211-216. Eagly, A. H. (1987). Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ: Erlbaum. Earle, J. R., & Perricone, P. J. (1986). Premarital sexuality: A ten-year study of attitudes and behavior on a small university campus. Journal of Sex Research, 22, 304-310. Echols, Al. (1984). The taming of the id: Feminist sexual politics, 196883. In C. Vance (Ed.), Pleasure and danger: Exploring female sexuality (pp. 60-72). Boston: Routledge & Kegan Paul. Ehrenreich, B., Hess, E., & Jacobs, G. (1986). Re-making love: The feminizalion of sex. Garden City, NY: Doubleday Anchor. Eley, T. C., Lichtenstein, P., & Stevenson, J. (1999). Sex differences in the etiology of aggressive and nonaggressive antisocial behavior: Results from two twin studies. Child Development, 70, 155-168. Ellis, B. J., & Symons, D. (1990). Sex differences in sexual fantasy: An evolutionary psychological approach. Journal of Sex Research, 27, 527- 555. Eysenck, H. J. (1971). Masculinity-femininity, personality and sexual attitudes. Journal of Sex Research, 7, 83-88. Fang, B. (1976). Swinging: In retrospect. Journal of Sex Research, 12, 220-237. Finger, F. W. (1975). Changes in sex practices and beliefs of male college students: Over 30 years. Journal of Sex Research, 11, 304-317. Fisher, W. A., & Byrne, D. (1978). Sex differences in response to erotica: Love versus lust. Journal of Personality and Social Psychology, 36, 117-125. Ford, K., & Norris, A. E. (1993). Urban Hispanic adolescents and young adults: Relationship of acculturation to sexual behavior. Journal of Sex Research, 30, 316-323. Gagnon, J. H., & Simon, W. (1968). The social meaning of prison homosexuality. Federal Probation, 32, 28-29. Giallombardo, R. (1966). Society of women: A study of a women's prison. New York: Wiley. Golden, C. (1987). Diversity and variability in women's sexual identities. 248-256. Huston, T. L., Surra, C. A., Fitzgerald, N. M., & Cate, R. M. (1981). From courtship to marriage: Mate selection as an interpersonal process. In S. Duck & R. Gilmour (Eds.), Personal Relationships: 2. Developing personal relationships, (pp. 53-88). New York: Academic Press. Jensen, A. R. (1998). The g factor. Westwood, CT: Praeger. Jesser, C. J. (1978). Male responses to direct verbal sexual initiatives of females. Journal of Sex Research, 14, 118-128. Johnston, J. (1973). Lesbian nation: The feminist solution. New York: Simon & Schuster. Julien, D., Bouchard, C., Gagnon, M., & Pomerleau, A. (1992). Insiders' views of marital sex: A dyadic analysis. Journal of Sex Research, 29, 343-360. Kacelnik, A. (1999, March). Sex, mind, behaviour, and evolution. Paper presented at the annual meeting of Eastern Psychological Association, Providence, RI. Kanin, E. J., Davidson, K. D., & Scheck, S. R. (1970). A research note on male-female differentials in the experience of heterosexual love. Journal of Sex Research, 6, 64-72. Kendrick, K. M., Hinton, M. R., Atkins, K., Haupt, M. A., & Skinner, J. D. (1998, September 17). Mothers determine sexual preferences. Nature, 395, 229-230. Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior in the human male. Philadelphia: Saunders. Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). Sexual behavior in the human female. Philadelphia: Saunders. Kitzinger, C. (1987). The social construction of lesbianism. London: Sage. Kitzinger, C., & Wilkinson, S. (1995). Transitions from heterosexuality to lesbianism: The discursive production of lesbian identities. Developmental Psychology, 31, 95-104. Knafo, D., & Jaffe, Y. (1984). Sexual fantasizing in males and females. Journal of Research in Personality, 18, 451-467. Knoth, R., Boyd, K., & Singer, B. (1988). Empirical tests of sexual selection theory: Predictions of sex differences in onset, intensity, and In Boston Lesbian Psychologies Collective (Eds.), Lesbian psycholo- time course of sexual arousal. Journal of Sex Research, 24, 73-89. gies: Explorations and challenges (pp. 19-34). Urbana: University of LaPlante, M. N., McCormick, N., & Brannigan, G. G. (1980). Living the Illinois Press. Goode, E., & Haber, L. (1977). Sexual correlates of homosexual experi- sexual script: College students' views of influence in sexual encounters. Journal of Sex Research, 16, 338-355. ence: An exploratory study of college women. Journal of Sex Re- Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The search, 13, 12-21. Gould, J. L., & Gould, C. G. (1997). Sexual selection: Mate choice and social organization of sexuality: Sexual practices in the United States. courtship in nature. New York: Freeman/Scientific American. Hansen, G. L. (1987). Extradyadic relations during courtship. Journal of Sex Research, 23, 382-390. Hariton, E. B., & Singer, J. L. (1974). Women's fantasies during sexual intercourse. Journal of Consulting and Clinical Psychology, 42. 313- 322. Harrison, D. A., Bennett, W. H., Globetti, G., & Alsikafi, M. (1974). Premarital sexual standards of rural youth. Journal of Sex Research, 10, 266-277. Herek, G. M. (1988). Heterosexuals' attitudes toward lesbians and gay Chicago: University of Chicago Press. Laumann, E. O., Paik, A., & Rosen, R. C. (1999). Sexual dysfunction in the United States: Prevalence and predictors. Journal of the American Medical Association, 281, 537-544. Lawson, A. (1988). Adultery: An analysis of love and betrayal New York: Basic Books. Lehrke, R. (1997). Sex linkage of intelligence: The X-factor. Westport, CT: Praeger. Leigh, B. C. (1989). Reasons for having and avoiding sex: Gender, sexual orientation, and relationship to sexual behavior. Journal of Sex Research, 26, 199-209. EROTIC PLASTICITY Leitenberg, H., & Henning, K. (1995). Sexual fantasy. Psychological Bulletin, 117, 469-496. Lewis, R. A. (1973). Parents and peers: Socialization agents in the coital behavior of young adults. Journal of Sex Research, 9, 156-170. Libby, R. W., & Nass, G. D. (1971). Parental views on teenage sexual behavior. Journal of Sex Research, 7, 226-236. Luschen, M. E., & Pierce, D. M. (1972). Effect of the menstrual cycle on mood and sexual arousability. Journal of Sex Research, 8, 41-47. Margolis, M. L. (1984). Mothers and such: Views of American women and why they changed. Berkeley: University of California Press. 373 emotion: Gender differences in visceral perception. Journal of Social and Clinical Psychology, 11, 199-212. Petty, R. E., & Wegener, D. T. (1998). Attitude change: Multiple roles for persuasion variables. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), Handbook of social psychology (4th ed., Vol. 1, pp. 323-390). Boston: McGraw-Hill. Propper, A. M. (1981). Prison homosexuality: Myth and reality. Lexington, MA: Heath. Reed, J. P., & Reed, R. S. (1972). Pornography research using direct erotic stimuli. Journal of Sex Research, 8, 237-246. McCabe, P. (1987). Desired and experienced levels of premarital affection Regan, P. C., & Berscheid, E. (1995). Gender differences in beliefs about and sexual intercourse during dating. Journal of Sex Research, 23, the causes of male and female sexual desire. Personal Relationships, 2, 23-33. McCauley, E. A., & Ehrhardt, A. A. (1980). Sexual behavior in female transsexuals and lesbians. Journal of Sex Research, 16, 202-211. McNamara, J. A. (1985). A new song: Celibate women in the first three Christian centuries. New York: Harrington Park Press. Mercer, G. W., & Kohn, P. M. (1979). Gender difference in the integration of conservatism, sex urge, and sexual behaviors among college students. Journal of Sex Research, 15, 129-142. Michael, R. T., Gagnon, J. H., Laumann, E. O., & Kolata, G. (1994). Sex in America: A definitive survey. New York: Warner Books. 345-358. Reinisch, J. M. (1990). The Kinsey Institute new report on sex. New York: St. Martin's Press. Reiss, I. L. (1967). The social context of premarital sexual permissiveness. New York: Holt, Rinehart, & Winston. Ridley, M. (1993). The red queen: Sex and evolution in human nature. New York: Penguin. Roberts, J. A. F. (1945). On the difference between the sexes in dispersion of intelligence. British Medical Journal, 1, 727-730. Roberts, T., & Pennebaker, J. W. (1995). Gender differences in perceiving Miller, B. C, & Moore, K. A. (1990). Adolescent sexual behavior, preg- internal state: Toward a his-and-hers model of perceptual cue use. In M. nancy, and parenting: Research through the 1980s. Journal of Marriage Zanna (Ed.), Advances in experimental social psychology (Vol. 27, pp. and the Family, 52, 1025-1044. 143-175). San Diego, CA: Academic Press. Miller, L. C., & Fishkin, S. A. (1997). On the dynamics of human bonding Robinson, R., Ziss, K., Ganza, B., Katz, S., & Robinson, E. (1991). Twenty and reproductive success: Seeking windows on the adapted-for human- years of the sexual revolution, 1965-1985: An update. Journal of Mar- environmental interface. In J. Simpson & D. Kenrick (Eds.), Evolutionary social psychology (pp. 197-235). Mahwah, NJ: Erlbaum. Mirande, A. M. (1968). Reference group theory and adolescent sexual behavior. Journal of Marriage and the Family, 30, 572-577. Money, J. (1990). Vandalized lovemaps. Buffalo, NY: Prometheus. Morokoff, P. J. (1986). Volunteer bias in the psychophysiological study of female sexuality. Journal of Sex Research, 22, 35-51. Moser, C., & Levitt, E. E. (1987). An exploratory-descriptive study of a sadomasochistically oriented sample. Journal of Sex Research, 23, 322- 337. Murphy, S. (1992). A delicate dance: Sexuality, celibacy, and relationships among Catholic clergy and religious. New York: Crossroad. National Research Council. (1993). Understanding and preventing violence. Washington, DC: National Academy Press. riage and the Family, 53, 216-220. Roebuck, J., & McGee, M. G. (1977). Attitudes toward premarital sex and sexual behavior among Black high school girls. Journal of Sex Research, 13, 104-114. Rosario, M., Meyer-Bahlburg, H. F. L., Hunter, J., Exner, T. M., Gwadz, M., & Keller, A. M. (1996). The psychosexual development of urban lesbian, gay, and bisexual youths. Journal of Sex Research, 33, 113-126. Rosenbluth, S. (1997). Is sexual orientation a matter of choice? Psychology of Women Quarterly, 21, 595-610. Rubin, L. (1990). Erotic wars: What happened to the sexual revolution? New York: Farrar, Straus, & Giroux. Rust, P. C. (1992). The politics of sexual identity: Sexual attraction and behavior among lesbian and bisexual women. Social Problems, 39, 366-386. Newcomer, S., & Udry, J. R. (1987). Parental marital status effects on Rust, P. C. (1993). Neutralizing the political threat of the marginal woman: adolescent sexual behavior. Journal of Marriage and the Family, 49, Lesbians' beliefs about bisexual women. Journal of Sex Research, 30, 235-240. Oliver, M. B., & Hyde, J. S. (1993). Gender differences in sexuality: A meta-analysis. Psychological Bulletin, 114, 29-51. O'Neill, G. C., & O'Neill, N. (1970). Patterns in group sexual activity. Journal o}'Sex Research, 6, 101-112. O'Sullivan, L. P., & Allgeier, E. R. (1998). Feigning sexual desire: Con- 214-228. Sack, A. R., Keller, J. P., & Hinkle, D. E. (1984). Premarital sexual intercourse: A test of the effects of peer group, religiosity, and sexual guilt. Journal of Sex Research, 20, 168-185. Savin-Williams, R. C. (1990). Gay and lesbian youth: Expressions of identity. New York: Hemisphere. senting to unwanted sexual activity in heterosexual dating relationships. Scacco, A. M. (1975). Rape in prison. Springfield, IL: Charles C Thomas. Journal of Sex Research, 35, 234-243. Schafer, S. (1976). Sexual and social problems of lesbians. Journal of Sex O'Sullivan, L. F., & Byers, E. S. (1992). College students' incorporation of intiator and restrictor roles in sexual dating interactions. Journal of Sex Research, 29, 435-446. Palmer, J. D., Udry, J. R., & Morris, N. M. (1982). Diurnal and weekly, but no lunar rhythms in human copulation. Human Biology, 54, 111-121. Pearlman, S. F. (1987). The saga of continuing clash in lesbian community, or will an army of ex-lovers fail? In Boston Lesbian Psychologies Collective (Ed.), Lesbian psychologies: Explorations and challenges (pp. 313-326). Urbana: University of Illinois Press. Pelleu'er, L. A., & Herald, E. S. (1988). The relationship of age, sex guilt, and sexual experience with female sexual fantasies. Journal of Sex Research, 24, 250-256. Pennebaker, J. W., & Roberts, T. (1992). Toward a his and hers theory of Research, 12, 50-69. Schmidt, G., & Sigusch, V. (1970). Sex differences in response to psychosexual stimulation by films and slides. Journal of Sex Research, 6, 268-283. Schmidt, G.. & Sigusch, V. (1972). Changes in sexual behavior among young males and females between 1960-1970. Archives of Sexual Behavior, 2, 27-45. Scott, G. G. (1983). Erotic power: An exploration of dominance and submission. Secaucus, NJ: Citadel Press. Sherwin, R., & Corbett, S. (1985). Campus sexual norms and dating relationships: A trend analysis. Journal of Sex Research, 21, 258-274. Sigusch, V., & Schmidt, G. (1973). Teenage boys and girls in West Germany. Journal of Sex Research, 9, 107-123. 374 BAUMEISTER Sigusch, V., Schmidt, G., Reinfeld, A., & Wiedemann-Sutor, I. (1970). Psychosexual stimulation: Sex differences. Journal of Sex Research, 6, 10-24. Sipe, A. W. R, (1995). Sex, priests, and power: Anatomy of a crisis. New York: Brurraer/Mazel. Smith, J. R., & Smith, L. G. (1970). Co-marital sex and the sexual freedom movement. Journal of Sex Research, 6, 131-142. Smith, T. (1994). Attitudes toward sexual permissiveness: Trends, correlates, and behavioral connections. In A. S. Rossi (Ed.), Sexuality across the life course (pp. 63-97). Chicago: University of Chicago Press. Spanier, G. P., & Margblis, R. L. (1983). Marital separation and extramarital sexual behavior. Journal of Sex Research, 19, 23-48. Spengler, A. (1977). Manifest sadomasochism of males: Results of an empirical study. Archives of Sexual Behavior, 6, 441-456. Sprecher, S. (1989). Premarital sexual standards for different categories of individuals. Journal of Sex Research, 26, 232-248. Sprecher, S., & Hatfield, E. (1996). Premarital sexual standards among U.S. college students: Comparison with Russian and Japanese students. Archives of Sexual Behavior, 25. 261-288. Sprecher, S., & Regan, P. C. (1996). College virgins: How men and women perceive their sexual status. Journal of Sex Research, 33, 3-15. Stanislaw, H., & Rice, F. J. (1988). Correlation between sexual desire and menstrual cycle characteristics. Archives of Sexual Behavior, 17, 499-508. Staples, R. (1973). Male-female sexual variations: Functions of biology or culture. Journal of Sex Research, 9, 11-20. Stark, R. (1996). The rise of Christianity. Princeton, NJ: Princeton University Press. Sue, D. (1979). Erotic fantasies of college students during coitus. Journal of Sex Research, IS, 299-305. Symons, D. (1995). Beauty is in the adaptations of the beholder: The evolutionary psychology of human female sexual attractiveness. In P. Abramson & S. Pinkerton (Eds.), Sexual nature, sexual culture (pp. 80-118). Chicago: University of Chicago Press. Tannahill, R. (1980). Sex in history. New York: Stein & Day/Scarborough. Thompson, A. P. (1983). Extramarital sex: A review of the research literature. Journal of Sex Research, 19, 1-22, Thornton, A., & Camburn, D. (1987). The influence of the family on premarital attitudes and behavior. Demography, 24, 323-340. Thornton, A. D., & Cambum, D. (1989). Religious participation and adolescent sexual behavior. Journal of Marriage and the Family, 51, 641-653. Vanwesenbeeck, I., Bekker, M., & van Lenning, A. (1998). Gender attitudes, sexual meanings, and interactional patterns in heterosexual encounters among college students in the Netherlands. Journal of Sex Research, 35, 317-327. Wallen, K. (1995). The evolution of female sexual desire. In P. Abramson & S. Pinkerton (Eds.), Sexual nature, sexual culture (pp. 57-79). Chicago: University of Chicago Press. Ward, D. A., & Kassebaum, G. G. (1965). Women's prison: Sex and social structure. Chicago: Aldine. Weis, D., Rabinowitz, B., & Ruckstuhl, M. F. (1992). Individual changes in sexual attitudes and behavior within college-level human sexuality courses. Journal of Sex Research, 29, 43-59. Whisman, V. (1996). Queer by choice. New York: Routledge. Wiederman, M. W. (1993). Demographic and sexual characteristics of nonresponders to sexual experience items in a national survey. Journal of Sex Research, 30, 27-35. Wilson, W. C. (197S). The distribution of selected sexual attitudes and behaviors among the adult population of the United States. Journal of Sex Research, 11, 46-64. ORDER FORM Start my 2000 subscription to Psychological Bulletin! ISSN: 0033-2909 Received August 31, 1998 Revision received July 6, 1999 Accepted July 6, 1999 Send me a Free Sample Issue Q Q Check Enclosed (make payable to APA) $76.00, APA Member/Affiliate $153.00, Individual Nonmembcr $331.00, Institution Charge my: Q VISA Q MasterCard Q American Express Cardholder Name Card No. Exp. date _ h DC add 5.75% sales lax TOTAL AMOUNT ENCLOSED $ Signature (Required for Charge) Subscription orders must be prepaid. (Subscriptions are on a calendar basis only.) Allow 4-6 weeks for delivery of the first issue. Call for international subscription rates. SEND THIS ORDER FORM TO: American Psychological Association Subscriptions 750 First Street, NE Washington, DC 20002-4242 AMMCAN PSYCHOLOGY ASSOCIATION Or call (800) 374-2721, fax (202) 336-5568. TDD/TTY (202)336-6123. Email: subscriptions@apa.org Credit Card Billing Address ..__ City State Daytime Phone^_ SHIP TO: Name . Address City .State. APA Customer # PLEASE DO NOT REMOVE - A PHOTOCOPY MAY BE USED Zip. Zip GADOO