An evolutionary perspective on Thai sexual attitudes and behavior.

Significant advances in understanding human sexuality have been made by examining behavioral patterns and associated feelings through an evolutionary perspective. Theory and research in the related fields of behavioral ecology, evolutionary anthropology, and evolutionary psychology suggest that men and women have evolved to think and behave differently with regard to sexual relationships. Although environmental conditions and cultural milieu can make male and female behaviors converge, the underlying tendencies of the two sexes differ and rest on fundamental differences in the evolved psychology of being male and being female (e.g., Buss, 1994; Cronk, 1991; Smith & Winterhalder, 1992; Wright, 1994). A variety of approaches and evidence have been used in the literature to assess predictions derived from this perspective, including data from closed-ended survey questionnaires, cross-cultural secondary analysis of ethnographic descriptions, and data derived though experimental psychology studies. Each approach has strengths and weaknesses, but all share the same theoretical base.

Scholars involved in these emerging fields have made little use so far of the recent substantial accumulation of qualitative data on sexual behavior and attitudes that have been generated from research stimulated by concerns about the international HIV/AIDS epidemic. Such data are increasingly available in both the developed and less developed countries around the world. The cultural diversity an complexity across societies call for more in-depth, open-ended types of information on these issues than can be elicited through survey questions. Thus, qualitative approaches are now being advocated and tried (Caldwell, 1993; Kitzinger, 1994; Pickering, 1988; Scrimshaw, Carballo, Ramos, & Blair, 1991; Smith, 1993; Standing, 1992).

We analyzed data on sexual attitudes and behavior of married men and women in Thailand derived through both focus group and individual interview techniques; the project was originally designed to study the influence of wives and male peers on married Thai males' attitudes and behavior with respect to extramarital sex. Unlike survey questionnaires, these qualitative research techniques allow the researcher to elaborate on the questions being asked and encourage study informants to explain their answers, with all the detail being recorded and available for analysis. Moreover, the conversational nature of the approaches fosters rapport, which is crucial when dealing with potentially sensitive sexual matters.

Although in the original project we did not anticipate that the results would be used to explore themes derived from evolutionary theory, a surprising amount of relevant data was generated for this purpose. Our primary goal in the current analysis was to interpret these data in light of hypotheses about male and female sexuality that have emerged from the evolutionary perspective. A secondary goal was to illustrate how, qualitative data derived from focus group discussions and focused in-depth interviews can be a rich complement to the more typical quantitative data used to test hypotheses generated through an evolutionary perspective.

We hope this approach will deepen understanding of Thai sexual patterns; at the same time, our analysis provides an interesting testing ground for the perspective. Thai society and culture are far removed from the West, from which the originators of the perspective come and most testing of the predictions has been based. One notable feature in which Thai sexual practices differ substantially from those prevalent in the West is in the high availability of commercial sex services and the relative social tolerance of their patronage (Boonchalaksi & Guest, 1994). For example, in a recent study of unmarried military recruits, with an average age of 22, 87% reported having had sex with a prostitute at some time (VanLandingham, Suprasert, Sittitrai, Vaddhanaphuti, & Grandjean, 1993). According to the 1990 urban survey by the Deemar Company (1990), 27% of single men (age 15 and older) and 9% of married men reported sex with a prostitute in the prior 12 months. Higher levels were found in a 1991 national survey: 40% of urban and 38% of rural never-married men reported sex with a commercial sex partner during the last 12 months, and 22% of urban and 9% of rural married men living with a spouse reported having commercial sex during the prior year (Sittitrai, Phanuphak, Barry, & Brown, 1992). This feature of the Thai setting adds to the interest of interpreting sexual attitudes and behavior within the framework of evolutionary psychology hypotheses.

The Evolutionary roach to Sexual Behavior and Attitudes

All living organisms have evolved through processes of natural and sexual selection that favor behaviors and traits that are most likely to leave a genetic legacy. These in turn operate through mechanisms such as striving for matings, investing in offspring, or helping other genetic relatives. There is growing evidence that humans are not immune from this principle; to survive and persist, humans must solve the same adaptive problems as all other species. Evolutionists argue that people have evolved to behave in ways that do, or did, contribute to their reproductive success (e.g., Alexander & Tinkle, 1981; Boyd & Richerson, 1985; Chagnon & Irons, 1979; Daly & Wilson, 1983; Durham, 1991; Lumsden & Wilson, 1981).

Evolutionarily, in most species and under most conditions, individuals who specialize in either mating (i.e., seeking sexual partners) or parental effort (i.e., caring for offspring) tend to leave more descendents. In part, this is because the behaviors that make one successful in mating are often mutually exclusive of the behaviors that promote parental success (reviewed by Low, 1993; also Daly & Wilson, 1983). For example, searching or fighting for a mate may involve risk taking, which is antithetical to surviving to nurse a dependent offspring. Thus, the reproductive "return curves" (reproductive success per unit of effort expended and risks taken) differ for mating and parental effort. Mating effort is not only typically risky but has a high fixed cost that must be paid before any reproductive return is possible (getting established, gaining status); parental effort, in contrast, has a per capita cost that must be expended for each offspring. As a result, if one sex (males in most species) specializes in getting mates, and the other (usually females) in investing in offspring, the two sexes will behave very differently (e.g., Diamond, 1992; Low, 1993).

Despite all cultural variation, this underlying difference sets the scene for widespread, predictable sex differences among humans in what they seek in a mate and what they think about sexual matters. Women have typically lost more investment than men if they failed to rear a child successfully, for they provide the bulk of energetic investment during pregnancy and lactation. However, in most societies, from hunter-gatherers to modern industrial nations, children survive and enter society more successfully when their mothers are assisted by an investing spouse. Thus, women who were indifferent to a man's ability and willingness to contribute parental investment suffered in most evolutionary environments. As a result, whenever conditions are such that men can and do invest significantly, women should have evolved to be less interested in short-term pairings; more interested in attracting a wealthy, willing-to-invest mate for the long term; and more risk averse than men. Women's attitudes about men's infidelity are likely to be shaped by the cost to themselves and their children. We expect women to be less tolerant of their mate's long-term relationships with other women if they might divert resources to those women and those women's children and to be more tolerant of their spouse's short-term encounters, which involve little commitment on the part of either partner (e.g., Betzig, 1986).

Men, in contrast, have likely always profited from a dual strategy. On the one hand, like other male mammals, a mating itself takes little effort for a man, and when low-cost, temporary (free of subsequent commitment) liaisons are available, men are likely to be little concerned with the subtleties of child health and survival. In evolutionary history, men have been able to profit from short-term trysts (especially when women have been able to raise children successfully alone) far more than women; men also have lost less than women in time and effort when children died. Because resources are central to children's success in most societies, as men accumulate the resources that make marriage and successful rearing of well-invested, highly competitive legitimate children possible, they will seek wives. Thus, we expect men to be interested in both long-term and short-term relations, with somewhat different preferences in the women they seek for the two types of partnerships (e.g., Buss, 1994; Buss & Schmitt, 1993).

In the context of long-term mating, in which men provide significant paternal investment, they are predicted to be intolerant of any infidelity (short or long term) on the part of their wives, because it places the paternity of the offspring in question and thus produces a risk of investing in the children of another man (e.g., Buss, 1987, 1989, 1994; Cronk, 1991; Low, 1993). Mate-guarding tactics, i.e., practices that prevent their mate from having sexual contact with other males and thereby ensuring their confidence of paternity, should be common (see Buss, 1994; Buss & Schmitt, 1993).

Conscious or logical thought is not required for men and women to behave in typically reproductively profitable ways; rather, this is apparently accomplished through evolved psychological mechanisms (e.g., Buss, 1994; Cosmides & Tooby, 1992). Moreover, as is true of all evolved phenomena, the relative effectiveness of such mechanisms (e.g., male versus female attitudes about different types of sexual liaisons) will be sensitive to the ecological and cultural milieu. In any enduring environment, some patterns of expenditure will be more optimal than others in their effect on survival and reproduction. In some cases, the fact that we humans change our environments so strikingly will render the evolved responses irrelevant or even maladaptive, but in others, new stimuli seem to evoke still-appropriate responses. Similarly, we can ask whether such patterns as resource striving, mate seeking, and mate guarding still "work" in novel contexts, and we should not be surprised if relatively new risks such as HIV provoke appropriate changes in attitude.

Our qualitative data from Thai men and women allow us to examine several key areas of the evolutionary perspective. In particular, we explored three themes: (a) views of male and female sexuality and whether Thai adults consider men's and women's sexual natures to differ in the ways predicted by the differing extent of effort required of men and women to reproduce successfully (i.e., their different reproductive return curves); (b) views about what men and women seek in a long-term mate, to see if gender differences in preferences reflect women valuing men's resources and men valuing women's parental abilities and qualities; and (c) whether and how men and women differ in their tolerance of a spouse's extramarital sexual activities.

Methods

Informants

Data for our analysis were collected through focus group discussions and focused in-depth individual interviews during the latter part of 1993 and early 1994. Fourteen focus groups involving 113 participants were held in Bangkok (with factory workers, residents of organized slums, and persons with middle class occupations constituting separate groups), two provincial towns in the Central region, and villages in the surrounding area of the two towns. Parallel but separate discussions were held with men and women. We also conducted 47 focused in-depth interviews in the same locations with men and women. All study informants had been married, and all but a few were currently living with their spouse. Almost all were between ages 25 and 40. Table 1 indicates the various categories of groups and the number of participants in each group. Table 2 shows the number of in-depth respondents by place of interview.

Table 1

Number of Participants and Categories of the
14 Focus Group Discussions

                             Men       Women
Bangkok
 Slum residents               8          8
 Factory workers              9          9
 Middle class persons         9          7
Province 1 (Lopburi)
 Rural                        9          8
 Urban                        7          7
Province 2 (Kanchanaburi)
 Rural                       10          8
 Urban                        7          7
Total Participants           59         55

Table 2
Number of Respondents in the In-depth Interviews

                             Men       Women       Total

Bangkok                       4          6          10
Province 1
   (Lopburi)
 Rural                        3          6           9
 Urban                        6          5          11
Province 2
   (Kanchanaburi)
 Rural                        4          4           8
 Urban                        4          5           9
Total                        21         26          47

The emphasis on urban sites was deliberate, because opportunities for commercial sex encounters, and probably extramarital affairs with noncommercial partners, are substantially higher in urban settings. Both of these were important concerns for the original purpose of the project. Still, the inclusion of rural informants and the purposive selection of several different social class mileus in Bangkok ensure that the views and behaviors of a fairly wide spectrum of Thai society are represented in the data. Moreover, our male informants appear to be relatively typical with respect to their sexual behavior. Among the 21 men selected for in-depth interviews, 17 reported having sex with a prostitute prior to marriage, and 11 reported visiting prostitutes after marriage (we did not ask women or focus group men about their behavior). This is quite consistent with the proportions who might be expected to have done so previously based on the quantitative studies cited earlier.

Procedures

Data collection. Our selection of study informants was purposive and largely opportunistic, often involving the aid of intermediaries. Informants were provided with small gifts and/or token payments for their time, although they were generally not informed that this would be the case at the time of recruitment. The focus groups and individual interviews were either conducted by the principal investigators of the original projects or by several assistants under their direct supervision. (For details, see VanLandingham, Knodel, Saengtienchai, & Pramualratana, 1994.)

In the case of the focus groups, interaction among participants was encouraged and is considered a major advantage of the method. In contrast, the focused in-depth interviews took place on a one-to-one basis and were usually held in a private setting to ensure confidentiality. Both approaches proceeded on the basis of prepared guidelines that focused on the topics in which we were interested, yet allowed considerable leeway for the informants to express their views of a situation (Frankfort-Nachmias & Nachmias, 1992; Morgan, 1996). Consistent with the objectives of the original project, the guidelines concentrated on issues related to commercial and noncommercial extramarital sex. The focus groups concentrated on determining prevailing norms, attitudes, and general observations regarding this topic, whereas the focused in-depth interviews explored personal behavioral patterns and their underlying rationale as well as attitudes. (See VanLandingham, Knodel, Saengtienchai, & Pramualratana, 1995, for a full replication of the guidelines.) Because of the difference in the nature of the two data-collection techniques, we refer to focus group discussion participants and to in-depth interview respondents. When we wish to refer to both together, we use the term informants.

Analysis. Both the discussions and interviews were tape recorded and fully transcribed and translated. The combined Thai and English transcripts serve as the data for analysis. We used the Ethnograph software program to assist in the systematic review of the content of the transcripts in relation to the topics discussed (Seidel, Friese, & Leonard, 1995). We coded transcript text segments in terms of topics and categories of interest to derive overview grids that provided a systematic summary of contents. This procedure facilitated judging how common particular themes were in the discussions and interviews as well as the extent of consensus and diversity of opinion (Knodel, 1993). When describing findings, we attempted to portray both the predominant opinion and the diversity of views expressed when consensus is lacking. As with most qualitative research, however, we generally eschewed attempts to quantify our findings beyond rough characterizations of whether a particular view or practice was cited frequently, rarely, or not at all. To state findings in more refined quantitative terms would provide a false impression of precision, given the nature of the sampling and data-collection procedures.

Illustrative verbatim quotations from the focus groups or individual interviews are included in the following analysis. Sometimes the comments are extracted from a longer discussion, omitting intervening statements for the sake of brevity.

Results

Thai Views of Sexuality

Thai views of male and female sexuality are consistent with the evolutionary view of general evolved sex differences. In Thailand, women are typically viewed as having weaker and less urgent sexual desires than men. Men are widely seen as having a natural "need" for sex that requires frequent outlet. This opinion was expressed repeatedly in the focus group discussions and the focused in-depth interviews, both spontaneously and as part of discussions about related characteristics (e.g., ideal spouse). Consistent with our findings are results of a 1990 survey of urban Thais by the Deemar Company, in which 80% of adult men and 75% of adult women agree that "it is natural for a man to pursue sex at every opportunity" (Deemar, 1990, Table 84b).

The male urge for sexual intercourse is seen by our informants as a basic physiological need, and as such, requiring an outlet in the form of intercourse. A man's need to release innate sexual tension resembles his need to relieve hunger by eating food. This view of men's sexual craving was contrasted to that for women, who are typically seen both as having considerably less need and desire for sex as well as being in much greater control of their sex urges. Several informants went so far as to assert sex in itself was of little importance to women. More commonly, informants acknowledged that women also had a desire for sex, but there was universal agreement that unlike many men, women could control their sexual feelings and leave them; unexpressed, as illustrated by the following quotations from a group of male factory workers in Bangkok and a group of provincial urban women:

Mr. A: Men feel it (sexual desire)

all the time. Men are like that but

only few women feel it that way.

Mr. B: If women feel it, they can

keep it better than men. They are

reserved.

Mr. C: Women would keep it to

themselves but men would not. If

men feel the desire, they must find

ways out.

Moderator: Isn't it strange that

men visit prostitutes while they

are single?

Mrs. A: No, it's not strange.

Mrs. B: Like when we meet a man,

nothing happens. But when men

see women, this one's pretty, that

one's pretty, they start thinking

further.

In the views of a number of our informants, the sexual needs of men, married or not, include the desire for a variety of sexual partners. Men are seen as becoming bored with having sex with the same woman and thus in need of at least an occasional variation in their partner. This included the need for a married man to have at least an occasional change from his wife, although our informants also recognized that not all men are the same in this respect. Such comments arose most frequently when discussing the reason some married men continue to patronize commercial sex workers or have casual affairs after marriage. As a rural male respondent said when asked why men visit prostitutes even after getting married: "They are probably bored with their wives. It is like you have the same type of food with the same flavor every day. It is boring." A provincial urban woman echoed the same idea: "I think it's in their nature ... Men naturally want to try something new." Indeed, the analogy between men's sexual appetite and appetite for food was often made and is most succinctly embodied in the common expression that married men require a "change of flavor."

No informant suggested that women needed sexual variety; in fact, rarely did women say that sexual satisfaction was important for themselves. In virtually all the women's focus groups and in many in-depth interviews, women stressed that it was important for wives to please their husbands sexually, but almost no mention was made of their own needs. In the discussion of marital sex, women often pointed out that sex was more important for men than for women.

For the women informants, providing sex for their husbands, and trying to please them in doing so, were seen as an important way that they could discourage husbands from seeking sex elsewhere (VanLandingham et al., 1995). Indeed, given the perception of an innate male need for a sexual outlet, failure to provide satisfying sex for the husband was considered by many women as risking his infidelity. For example, when asked if wives have to be good in bed, a Bangkok female clerk responded, "Yes. Husbands would look for some others if they are not sexually satisfied with wives. Men can't live without it but women can." In a similar vein, a rural woman in a focus group said, "Men are more sexually experienced than women, so they take sex as a priority ... If wives can sexually satisfy them, they will go on providing for them ... For women, children must be properly fed and clothed and sexual life is second to children's welfare."

Consistent with these views, commercial sex in Thailand is widely available for men in a variety of venues; patronage is widespread and probably has been, at least in urban areas, for quite some time (Bamber, Hewison, & Underwood, 1993; Sittitrai et al., 1992; Wilson & Henley, 1994). The ready availability of multiple short-term female partners through institutionalized prostitution provides an ample supply of young and attractive partners for men with minimal commitment. Obviously, the fact that Thai prostitutes routinely practice birth control will thwart the reproductive outcome of most of these sexual acts. However, as mentioned previously, because men's behavior responds to evolved cues rather than conscious intentions to maximize their fertility, this is irrelevant to the basic argument. Nevertheless, it is interesting to note that Thai men have largely resisted the use of condoms with prostitutes until very recently, when the HIV epidemic dramatically raised the stakes of unprotected intercourse and led the government to require condom use at brothels (Hanenberg, Rojanalpithayakorn, Kunasol, & Sokal, 1994).

Thai men generally view condom use as interfering with the naturalness and pleasure of the sex act (Knodel & Pramualratana, 1996). Atypical male view, expressed in a provincial male focus group, was "I don't like to use [condoms] . Its not natural ... Something in between." Although it is unlikely that Thai men's resistance to condoms reflects any conscious desire to impregnate prostitutes, their distaste could stem from an unconscious aversion to such obvious interference with the reproductive nature of short-time sex. However, this is not explicitly stated as such in our data.

What Men and Women Seek in a Spouse

An evolutionary approach leads us to expect men and women to seek characteristics in a spouse that differ in systematic ways: women seeking whenever possible resource investment is useful in rearing children, and men seeking qualities like youth, fecundability, and faithfulness. The most common answer, mentioned spontaneously in all focus groups to a direct query about an ideal spouse, is that both men and women want to have a spouse who understands and gets along with them (see also Buss, 1987, 1989). This, however, appears to have different meanings for men and women. Men typically cited their wish that their spouses understand the need for a husband to join male friends outside the home without question or complaint. In contrast, for women, a husband's "understanding" includes his keeping her informed about his whereabouts and plans, especially when he goes out with friends, and more generally, his recognizing the wife's and children's needs for his responsible behavior in caring for and supporting them.

These parallel, if opposing, expectations from men and women reflect Thai men's practice of joining male friends for drinking, which wives often view as conflicting with their desire to have husbands spend more time at home (VanLandingham et al., 1995). As a Bangkok male worker respondent explained: "Wives should let husbands have their way, not interfere with them and let them sort their way out themselves. Men drink when they are over-concerned with work but later, it will be okay." This contrasts to the view expressed by a woman in a rural group: "I want the one who doesn't get drunk, go out to entertain himself and he should stay home with the wife and children."

Within the Thai context, where commercial sex patronage is prevalent with friends, and where other types of extramarital sexual relations (e.g., with minor wives or other noncommercial commercial partners) are not unusual, it is not surprising that a salient component of family responsibility in women's minds involves concerns that a husband spend time at home and not frequently go drinking with friends. When asked what a good husband should be, two participants in a group of rural women said "They should not go out to entertain themselves." "They should be home punctually. All salary should be given to wives." These concerns are linked to women's fears about male infidelity and its consequences. Thus, to many Thai women, marrying a man who will not turn out to be a philanderer is a high priority, as made clear by a woman in the Bangkok factory group: "Say [a woman] is considering [a single man] to be the future husband and he drinks and smokes and goes out, instead of getting 100 points he might only get 50."

In all the focus groups, women also said that good husbands are those who earn and provide support for their families. The importance of material support was sometimes subordinated to a more general wish for a "responsible" mate, i.e., beyond just the financial aspects. Likewise, in the in-depth interviews, women stressed the importance of husbands' emotional and financial support of family as being important to them. When asked what good husbands should be, two women in a rural focus group said "They have to spend some time working to support the families" and "It's enough if they take responsibility toward their families." Even more telling is the comment by a provincial urban woman respondent about her own husband: "He gives me all he earns. That's why I say we are a happy family. He is highly responsible."

Just as men are expected to be the key persons for material resources of the household, providing financial support to wives and children, women are expected to be good housewives and good mothers. Although in some groups, participants also stated that an ideal wife should help in earning income, the main emphasis was the expectation that she would take good care of the house, children, and husband, thus playing the key role in nonmaterial parental investment for the family. As a man in a Bangkok Slum Group said: "Good wives must be at home, prepare meals for husbands and children. They should not go out and be promiscuous." Similarly, a participant in a provincial urban group said "A good wife? I'm thinking of Thai qualities ... Well, like doing housework, working hard, taking good care of the family, the children, not going out."

The importance of virginity never spontaneously arose as a characteristic of an ideal wife or husband by any men or women focus group participants. When directly asked, men generally agreed that they preferred their wife to have been a virgin, although many also indicated that virginity was not considered as among the most important concerns in selecting a wife. In some cases, the concerns about female virginity were largely an issue of pride about being the first to have sex with one's wife. For example, a participant in a group of provincial urban men said when asked why it was important to marry a virgin: "Everybody wants to be lead warrior, to be the first. We don't want to follow anybody. But if that is not possible and you have to be second, you can accept it. You don't have to think too much." Other comments were linked to concerns about being sure that the prior relationships either would not resume or that they were not indicative of possible future infidelities with other men. Only rarely, however, were worries about being sure of paternity made explicit in connection with a woman's premarital sexual history.

Distinctions were sometimes made between women who had a prior marriage and those who had affairs while single, possibly because of the difference in their significance as indicators of how prone the woman might be for affairs after marriage. As another provincial urban man explained in a focus group: "If she separated from her husband and then we like her and she liked us, that's one thing. But if we know that she had somebody before as a lover and she's not married, she had other men, one here and one there, well I've got to think perhaps I shouldn't have taken the leftover from other men."

On balance, Thai husbands in our study are fairly tolerant of women who have had premarital sexual relations as candidates for spouses. Although many articulate their preference for virgin wives, they do not necessarily condemn wives for prior sexual relations, and most would appear not to consider wives' premarital sexual history to be grounds to prevent or terminate a marital union.

Women did not reveal any preference for virginity by their husbands, pointing out that it was unrealistic to expect a man to be sexually inexperienced at the time of marriage. By far the most common reaction of a wife to learning that her husband had premarital commercial sex experience was that she had little right to condemn her husband for behavior engaged in prior to her relationship with him. Only rarely did women express clear disapproval at the premarital commercial sex activities of their husbands. As two women explained in a Bangkok slum group: "Most men are experienced before getting married. There is hardly anyone who is not ... Men would marry, having experiences and visiting prostitutes. They have such experiences. There are many virgin women. But it's hard to find virgin men." "[My husband] told me he took prostitutes. How can I react since he was not mine at that time?"

Many women never inquired about their husband's premarital commercial sexual activity, and most who did only discussed the matter after they were already married. This may reflect that they did not feel they had the right or power to do so. However, it also suggests that they did not consider this as an important, or at least realistic, criterion for choosing a mate. Wives' general detachment to learning that their husband had slept with prostitutes when he was single probably reflects an acceptance of the inevitable, given the reality that premarital prostitute visitation among Thai men is very common.

In general, women were also accepting of a husband's premarital noncommercial sexual relationships, although typically they expressed greater concern than about premarital patronage of prostitutes. Women's main interest in this connection was seeking assurance that the earlier relationships were ended and would not resume. Some also expressed concern about potential lingering attachments or obligations in cases where the prior relationship, which might have been an earlier marriage, involved children. In most cases, as long as these prior relationships are definitively in the past, wives do not seem to be particularly interested. As a provincial male respondent explained: "[My wife and I] have talked. She asked me how many women I had [before we were married]. I told her six. She did not say anything because it was before our marriage. She was only interested in women other than service girls because she was afraid that I would meet them again and have affairs." In a similar vein, a rural woman in a focus group said when asked how will wives feel if they know that their husbands had sex with some other girlfriends before marrying them: "It might be okay as long as they didn't go on with that affair. It was in the past, wasn't it? If he didn't continue the affair, it wouldn't matter."

Sexual compatibility was never spontaneously mentioned as a characteristic of the ideal spouse, and when informants were directly asked, their opinions about the importance of sex within marriage were varied. Some informants felt sex was a crucial feature of a marital relationship, but many others considered it only of secondary importance. As a male middle class focus group participant in Bangkok said: "Marriage and sex life go together. It's natural but sex life is not the top priority . . . It's like adding spice to the meal." When women discussed marital sex, it was almost always from the perspective of the husband. For example, when discussing the importance of sex in marriage in a group of middle class Bangkok women, one participant said "As I listen to the others, and from my own experiences, sometimes it's quite important for a family, because some men need it and if his wife can't give him, he then must go out and seek it."

How Men and Women View Infidelity

The focus group discussions and in-depth interviews make clear that views of extramarital sexual activity in Thailand vary along several dimensions. The most pronounced is the distinction between views regarding the activity of men and women. In the case of men, views depend on whether the activity involves commercial or noncommercial contacts, the frequency of commercial sex patronage, and whether noncommercial contacts are relatively casual or involve more commitment. Views about extramarital sex are also tempered by whether measures for protection against sexually transmitted diseases (STDs) are taken, especially in the case of commercial sex contacts, but also to some extent in the case of casual noncommercial relationships. In addition, views of men and women differ with respect to some of these dimensions.

Female infidelity. Evolutionary theory predicts that men will, under most conditions, be intolerant of any female infidelity, given that extramarital matings by women carry the risk of misplaced paternal investment. Indeed, there is striking consensus among both male and female informants in their views on female engagement in extramarital affairs. They uniformly agreed that it is intolerable to a man if his wife engages in any sort of extramarital relations and that marital infidelity on the part of a woman is unambiguously against societal mores. Many typically assumed that female infidelity would lead to the termination of the marriage by the man, as illustrated by the following exchange among participants in a group of provincial urban men.

Moderator: Is it common for

women to have other men?

Mr. A: Men just can't accept that. If

the woman does that, it means we

have to separate ... Our society

just can't accept this kind of behavior

in women.

Mr. B: Unacceptable. I would want

to divorce. No further discussion.

Mr. A: Men consider themselves as

tigers. Two tigers can't live in the

same cave. One must die.

Reactions to the discovery of female marital infidelity could even lead the husband to violence against the wife, the lover, or both. This view was expressed numerous times, as illustrate by a rural male respondent: "If they do that [extramarital sexual relations], they are bad women. This is forbidden. They can be killed by their husbands."

The total unacceptability of female infidelity was often explicitly contrasted with the more lenient attitudes toward male indulgence in extramarital sex. Unlike men, women are expected to control their sexuality and devote themselves solely to their husbands and families. As a provincial urban male respondent noted: "As for women, if they have affairs, it is unacceptable ... Such behavior of women is against Thai culture. But for men, it is common. The only thing that women must keep in mind is that they have to take good care of the family, make their family happy." The seriousness of female marital infidelity was also contrasted with the lesser importance of virginity in the choice of a wife, as illustrated in the comments of a Bangkok middle class man in a focus group discussion: "It's a matter of pride for a man to have a virgin as a wife. You're proud to have something brand new. But if it's second hand, it doesn't really matter. When she's married, she can't have affairs with anybody else. She has to be true to her husband. If she isn't loyal to her husband, he can't accept it."

There were occasional mentions of situations by women, where women seeking or accepting a sexual relationship with some man other than her husband is understandable, if not socially acceptable. These situations typically involved extreme circumstances where the husband was irresponsible or abusive toward his wife and family.

Male infidelity. Views of male extramarital sex lack the uniform condemnation that characterizes female infidelity. They tend to be more complex, with the degree of approval or disapproval being conditional on the type of partner and to some extent the circumstances of the relationship. There are also clearer contrasts between men and women in their views about male infidelity than in their views about female extramarital sex.

Our male informants were quite mixed in their views about visiting prostitutes after being married. Some completely opposed it, and only a few condoned such behavior outright without qualifications. Still, many viewed variation in sexual partners as a necessary part of their sexuality. Thus, married men patronizing commercial sex are not perceived as abnormal by most other men. As one Bangkok middle class male focus group participant put it: "If anybody says [married men visiting prostitutes] is unnatural, they themselves are abnormal." As indicated previously, about half of the male in-depth respondents admitted having visited a prostitute themselves since being married.

Numerous male respondents in the in-depth interviews stated that "infrequent" patronage was acceptable as long as family needs were no neglected. This is illustrated in the comments made in the group of Bangkok factory men when asked if it is an ordinary practice for married men to take prostitutes: "It might be if they do it once in a while" and "As men, we can't stay home all the time; we need different flavors. Friends may say we are afraid of wives if we don't go with them anywhere."

At the same time, the need to protect oneself (and thus one's family) from STDs occasionally arose. Even more clearly, a substantial proportion of our male study informants viewed frequent commercial sex patronage by a married man as excessive behavior and disapproved of it. A rural male respondent made this clear: "Going frequently is not proper. But if [married men] go only once in a while and protect themselves from diseases by using condoms, it is all right."

Married women have a less tolerant attitude than men toward male extramarital commercial sex patronage. Some viewed any patronage as totally unacceptable. Among our informants, and undoubtedly in the general public, the vast majority of married Thai women clearly prefer that their husbands do not engage in extramarital commercial sex patronage. As a Bangkok middle class woman in a focus group put it: "Hundred percent, women don't like [married men visiting prostitutes] ... But sometimes we can't forbid it since men think it's ordinary." Despite this, some women reluctantly accepted the view that men have a natural need for sexual variety and saw this as a reason to be tolerant of occasional patronage of prostitutes by their husbands. For them, such activity was seen as a form of male entertainment and not as a serious breech of marital trust. Fewer than half of the women in-depth respondents (compared to about two thirds of the men) thought that commercial sex patronage after marriage was ordinary behavior.

This does not imply approval but rather that they saw it as a fact of life. Moreover, wives' tolerance was typically conditioned on the commercial sex visitation by the husbands being discreet and infrequent, precautions being taken against STDs, and the costs not draining heavily on family finances. In the words of two female respondents, the first a Bangkok teacher and the second from a provincial town: "They give us most of the money so we must sometimes let them stray ... Not as a habit. They are allowed to [visit prostitutes] once in a while ... if we are not financially troubled." and "It's okay for me if [married men take prostitutes] now and then or on some special occasions. If they do it as a habit, they are irresponsible and they may get wives infected."

An evolutionary perspective predicts that women's views will differ on commercial sex versus noncommercial sex liaisons by their husbands. Non-marital sex relationships in Thailand are quite varied, ranging from casual encounters to committed relationships. Although the line between commercial and noncommercial sex is sometimes ambiguous, the destination is usually clear to those involved. Commercial sex generally requires on-the-spot payment for each sexual transaction, after which the obligations of the client end. Noncommercial sex partners usually require a greater investment in time and resources on the part of the man and involve a greater emotional investment from both parties. As such, they carry greater risks of resource conflicts for the family. Men who enter into noncommercial relationships typically understand the expectations involved, and so their wives.

A noncommercial sex partner who is supported in full by a married male is typically referred to as a minor wife. Such relationships have apparently been common for centuries in Thailand among the upper strata (Wilson & Henley, 1994. Moreover, as our study informants acknowledged, they are clearly still a salient resource threat to married women. Probably because of the greater economic and emotional threat they pose for a marriage, minor wives receive more attention in the focus group discussions and in-depth interviews than did other types of noncommercial sex partners. The following exchange among women in a provincial urban focus groups makes clear the threat minor wives pose to family resources:

Moderator: Who are minor wives

then?

Mrs. A: Those who are financially

supported by our husbands. They

may have children by them.

Mrs. B: We have to share the income

with them.

Mrs. C: We don't have as much as

we used to have. They are treated

just like wives since they share

everything with us.

In reality, most noncommercial sexual relationships of married men probably are of a less committed nature than that of a full-fledged minor wife. Such more casual relationships were also seen as threatening family security, both in their own right and because they might eventually lead to a more demanding minor wife-type relationship. This is clear in the comment from a woman in a Bangkok slum group when asked what she thought about a married man who sees another woman, not as a minor wife, but just from time to time: "Even if he doesn't think about it, we women have to think about it. We're afraid he might take it seriously."

With few exceptions, women in our study viewed relationships with minor wives, or women who could potentially have a long-term affair with their husband, as either intolerable or as an extremely difficult burden to bear. (Because only one woman in our study acknowledged being a minor wife herself, we cannot comment on how minor wives view their own relationships.) Wives typically feel far more threatened by ordinary women who engage sexually with their husband than by commercial sex workers, who require neither extensive emotional nor resource commitment. As one woman said in a provincial urban group: "To go out [to visit prostitutes] is better [than a minor wife]. I think it's better to be temporary and it can be finished. In a case of minor wife, he must be responsible for her. In fact, he has to divide the money with her that he would provide me." This view was clearly echoed by a rural woman respondent when asked if prostitute visitation by her husband might be acceptable if AIDS were not a threat: "It might be all right. If he was bored with me, he could take one." But when asked if he could take a mistress, she replied: "No, absolutely not. He would spend all on her, and I and my children would be in trouble. He would waste all, money and time."

Married women's reluctant tolerance of occasional prostitute visitation by their husbands must be understood within this context. As one would expect from an evolutionary perspective, Thai wives are less threatened by their husbands' occasional visit to a commercial sex worker than their having an affair with an "ordinary" girl or, even worse, taking a minor wife. Such relationships are seen as threatening the very family core because they represent potential resource reallocation from wife and children to other sources outside of the family. Given the Thai view of male sexuality, which includes the need for variety, many women perceive some male infidelity as likely, if not inevitable. Through prostitution visitation, this need might be met without the risk of serious resource loss to the family. Some married women among our study informants even seem to adopt a conscious strategy of permitting commercial sex visits to prevent the greater threat posed by noncommercial affairs. More generally, it is typically when familial resources are in question, whether the issue is commercial or noncommercial relationships, that the wife objects to a husband's extramarital affairs.

With the advent of the HIV/AIDS epidemic, women's relative aversion to their husband's commercial and noncommercial extramarital sexual contacts may be changing. Thailand has been at the forefront of the spread of the HIV/AIDS epidemic in Asia (Brown, Sittitrai, Vanichseni, & Thisyakorn, 1994). There have been extensive publicity campaigns aimed at reducing risk behaviors, and general awareness of risks is high (Hanenberg et al., 1994; Thailand, National Statistical Office, nd). Virtually all our study informants recognized the serious consequences of being infected with HIV, and no one dismissed it casually. The escalated risks that husbands' patronage of prostitutes poses to wives in the context of the HIV/ AIDS epidemic may be causing women to reevaluate their reluctant tolerance of such behavior. It may also be changing the lesser aversion by wives to commercial sex patronage by husbands relative to noncommercial sex contacts. Although the majority of women still object less to their husband's extramarital commercial sex contacts than to noncommercial relationships, for reasons of resource commitment, a noteworthy minority expressed the reverse view, citing their fears of HIV/AIDS and its effect on the family. As one woman in a provincial urban group explicitly stated: "Since there has been AIDS, many women let husbands have mistresses . . . They don't want husbands to take prostitutes as they might get AIDS from them."

Discussion

Although an evolutionary perspective can be useful in identifying and interpreting common themes in human sexual behavior and thinking, the limitations of such a framework also need to be kept in mind. It cannot be expected to predict particular intrasexual and intersexual differences in mating practices and outcomes. Rather than implying specific mating configurations an patterns (e.g., number of sexual partners a man is likely to have), these perspectives suggest that there are evolved tendencies in mating strategy behavior and that these tendencies will diverge between the sexes. The particular ways these tendencies are expressed, however, will differ in particular environments. Thus, the social context in which mating strategies occur will greatly influence the behavior and practice by both men and women.

With that said, it is nevertheless fair to conclude that our findings concerning Thai sexual attitudes and behavior conform to evolutionary predictions reasonably well in the three major areas we examined: conceptions of sexuality, desirable characteristics of a spouse, and marital fidelity. Moreover, in some areas (e.g., female marital infidelity), men's and women's opinions converge strongly and vary little; in others (e.g., male marital infidelity), men's and women's opinions diverged in predictable ways.

According to the views of our Thai informants about differences in male and female sexuality, men are assumed to seek short-term sexual encounters whenever possible, seek them with many partners, and dislike novel devices like condoms when they will interfere with potentially fertile copulations with minimal commitment. Women are assumed to be unlikely to seek sex for its own sake but rather as a means for retaining an investing man's commitment to themselves and their children. Not every informant agreed in each aspect, and sometimes the evidence for these views is indirect. Nevertheless, overall, Thai views of men's and women's sexuality address remarkably closely the adaptive problems of some environments posed by reproductive constraints: women seeking to secure long-term resource commitment from a willing and capable man, and men seeking to maximize the number of fertile, accessible mates, while minimizing investment in those sought for short-term pairings (Buss & Schmitt, 1993).

Thai opinions about the ideal spouse likewise fit the evolutionary model reasonably well. An ideal husband concentrates his resources on his family, not seriously diverting them through other sexual liaisons. An ideal wife provides nonmaterial care for both her husband and children, although some men also clearly value a woman's ability to contribute to the material well-being of the family. Virginity per se is not always important. Instead, some men worry about prior sexual relationships of a potential spouse in ways that suggest concern over potential future infidelity. This is quite consistent with recent findings that chastity is not a universally held preference in mating partners and theorizing that the overriding concern should be the probability of extra-pair matings (Buss, 1989, 1994). Our results with respect toward attitudes regarding female virginity, however, may be affected by the strong urban skewness of our informants. Attitudes may be more conservative in rural areas, as suggested by at least one recent study in Northern Thailand (Im-Em, 1996). Thai women generally assign little importance to prior commercial sex patronage of their husband, but they are more likely to show concern about a husband's prior noncommercial sexual relationships with other women, as it might threaten the security of their marriage. Such premarital relationships, however, also appear to carry little weight, provided there is assurance that they are ended and carry no continuing risks of resumption.

Sexual compatibility was not singled out as a crucial characteristic for a long-term mate, although most informants acknowledged it was of some importance for the marriage. The relegation of sex to a place of only modest importance for a marital union among Thais may seem strange to observers from current Western cultures, where it is commonly assumed to be paramount. However, cross culturally and through evolutionary time, first-marriage arrangements have typically been made by others on other grounds. Thus, even though in modern Thai society arranged marriages are rare and individuals are relatively free to choose their mate, it is not surprising that the sexual relationship between husband and wife is subordinate to other issues, especially to the socio-economic functions of marriage (Brummelhuis, 1993; Limanonda, 1995).

Although Thai views on marital infidelity are more complex and conditional than views on premarital sexual experience, again they fit an evolutional model well. The strongest, and least variable opinion, expressed by both men and women, was the uniform condemnation of female marital infidelity. In evolutionary terms, female marital infidelity carries the risk of misplaced paternal investment in any society in which men invest in their wife's children, and we thus expect it to be little tolerated.

Women, confident of their maternity, should care primarily about male infidelities that carry substantial resource cost and less about those that do not. If her spouse fathers children in whom he does not invest (e.g., by commercial sex workers), she loses little. In contrast, women should be far more concerned about infidelities that create potential conflicts of interest about resource investment in their offspring. Consistent with this expectation, Thai women were less concerned with husbands' visits to commercial sex workers than with involvement with a long-term noncommercial partner, particularly a minor wife. Male extramarital sex activity was generally considered more acceptable by men than by women and tolerated better by both sexes when it carried little risk of diverting resources away from the family.

This fit is interesting in part because most theoretical propositions of evolutionary psychology come largely from Western scholars. Because the predictions are meant to be general, applications in non-Western societies are valuable. The new sexual risks of HIV/AIDS give some insight into how evolved traits and preferences can interact with and respond to new situations. In contrast to the general majority view, a minority of women said that if their husband must seek sexual liaisons outside marriage, it would be better for him to find a noncommercial partner who is safer (even if more expensive in terms of resource drain) than a commercial sex worker, because the risk of life-threatening disease is less.

We interviewed married men and women from diverse strata of Thai society. However, they represent only part of the spectrum of men's and women's conditions. The data were focused on the benefits of high male investment and long-term pairing for women. There are, however, conditions in which females seek short-term pairings in Thailand, as elsewhere. In other species, the conditions include female ecological independence (e.g., grazing mammals, in which males have little to offer females; see Krebs & Davies, 1991); situations in which females can seek "good genes" in extra-pair copulations while (if successful) getting investment from a primary male (e.g., Gowaty, 1983); and situations in which neither sex controls many resources, but females can trade sexual favors for resources or protection on a short-term basis. Although we do not have direct information, on the expectations of the women who have affairs with married men, there is some evidence that their arrangements often involve some sort of indirect material compensation from the man (Havanon, Bennett, & Knodel, 1992). Exploring the views of Thai women who engage in short-term relations, including both commercial sex workers who receive explicit financial compensation and others who engage in sexual affairs outside the commercial arena, would be a particularly interesting topic for future research.

We believe qualitative data such as those generated by focus group discussions and focused in-depth interviews offer largely unexploited opportunities for examining evolutionary hypotheses about human sexual behavior. They can serve as a valuable complement to the quantitative data, typically derived from responses to fairly simple questionnaire items, that more commonly serve as the basis for such tests. Some patterns we discovered would be difficult to detect with other methods. This approach also offers considerable scope for exploring within-society variation and responses to societal particulars (e.g., opinions about female virginity) and novelties (HIV/AIDS). We are encouraged about the usefulness of carefully collected qualitative data for examining hypotheses such as those explored in the foregoing analysis.

References

Alexander, R. D., & Tinkle, D. W. (Eds.). (1981). Natural selection and social behavior: Recent research and theory. New York: Chiron.

Bamber, S. D., Hewison, K. J., & Underwood, P. J. (1993). A history of sexually transmitted diseases in Thailand: Policy and politics. Genitourinary Medicine, 69, 148-157.

Betzig, L. L. (1986). Despotism and differential reproduction: A Darwinian view of history. New York: Aldine.

Boonchalaksi, W., & Guest, P. (1994). Prostitution in Thailand. Nakhorn Pathom, Thailand: Institute for Population and Social Research, Mahidol University.

Boyd, R., & Richerson, P. J. (1985). Culture and the evolutionary process. Chicago: University of Chicago Press.

Brown, T., Sittitrai, W., Vanichseni, S., & Thisyakorn, U. (1994). The recent epidemiology of HIV and AIDS in Thailand. AIDS 8 (Supplement 2), S131-S141.

Brummelhuis, H. (1993). Do we need a Thai theory of prostitution? Paper presented at the 5th International Conference on Thai Studies, School of Oriental and Asian Studies, London, July.

Buss, D. M. (1987). Sex differences in human mate selection criteria: An evolutionary perspective. In C. Crawford, D. Krebs, & M. Smith (Eds.), Sociobiology and psychology: Ideas, issues and applications (pp. 335-352). Hillsdale, NJ: Erlbaum.

Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavior and Brain Sciences, 12, 1-49.

Buss, D. (1994). The evolution of desire: Strategies of human mating. New York: Basic Books.

Buss, D., & Schmitt, D. (1993). Sexual strategies theory: An evolutionary perspective on human mating. Psychological Review, 100, 204-232.

Caldwell, J. (1993). How do we secure adequate data on sexual networks and the probability of HIV transmission? Health Transmission Review, 3, 205-208.

Chagnon, N., & Irons, W. (1979). Evolutionary biology and human social behavior. North Scituate, MA: Duxbury.

Cosmides, L., & Tooby, J. (1992). Cognitive adaptations for social exchange. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 163-228). New York: Oxford.

Cronk, L. (1991). Human behavioral ecology. Annual Review of Anthropology, 20, 25-53.

Daly, M., & Wilson, M. (1983). Sex, evolution, and behavior (2nd ed.). Boston: Willard Grant.

Deemar Company. (1990). AIDS research in Thailand, Vol. I: Computer tabulations. Bangkok: Deemar.

Diamonds, J. (1992). The third chimpanzee: The evolution and future of the human animal. New York: HarperCollins.

Durham, W. H. (1991). Coevolution: Genes, culture, and human diversity. Stanford, CA: Stanford University Press.

Frankfort-Nachmias, C., & Nachmias, D. (1992). Research methods in the social sciences (4th ed.). New York: St. Martin's.

Gowaty, P. (1983). Male parental care and apparent monogamy among eastern blue-birds (Sialia sialis). American Naturalist, 12, 144-157.

Hanenberg, R., Rojanapithayakorn, W., Kunasol, R, & Sokal, D. S. (1994, July 23). Impact of Thailand's HIV-control Programme as indicated by the decline of sexually transmitted diseases. Lancet, 344, 243-245.

Havanon, N., Bennett, A., & Knodel, J. (1992). Sexual networking in a provincial Thai setting. AIDS Prevention Monograph Series Paper No. 1, Family Health International AIDSCAP Project, Bangkok, Thailand.

Im-Em, W. (1996). Partner relations and AIDS in Chiang Mai villages. Ph.D. dissertation, Australian National University.

Kitzinger, J. (1994). Focus groups: Method or madness? In M. Boulton (Ed.), Challenge and innovation: Methodological advances in AIDS research (pp. 159-175). London: Falmer.

Knodel, J. (1993). The design and analysis of focus group studies in social science research. In D. Morgan (Ed.), Successful focus groups: Advancing the state of the art (pp. 35-50). Newbury Park, CA: Sage.

Knodel, J., & Pramualratana, A. (1996). Prospects for increased condom use in marital unions in Thailand. International Family Planning Perspectives, 22(3), 97-102.

Krebs, J., & Davies, N. B. (Eds.). (1991). Behavioural ecology: An evolutionary approach. Oxford, England: Blackwell.

Limanonda, B. (1995). Families in Thailand: Beliefs and realities. Journal of Comparative Family Studies, 26(l), 67-82.

Low, B. (1993). Ecological demography: A synthetic focus in evolutionary anthropology. Evolutionary Anthropology, 1(5), 177-187.

Lumsden, C. J., & Wilson, E. O. (1981). Genes, mind, and culture: The co-evolutionary process. Cambridge, MA: Harvard University Press.

Morgan, D. (1996). Focus groups. Annual Review of Sociology, 22, 129-152.

Pickering, H. (1988). Asking questions on sexual behavior ... testing methods from the social sciences. Health and Policy Planning, 3(3), 237-244.

Scrimshaw, S. C., Carballo, M., Ramos, L., & Blair, B. A. (1991). The AIDS rapid anthropological assessment procedures: A tool for health education planning and evaluation. Health Education Quarterly, 18(l), 111-123.

Seidel, J., Friese, S., & Leonard, D. C. (1995). The Ethnograph v4.0: A users manual. Amherst, MA: Qualis Research.

Sittitrai W., Phanuphak, P., Barry, J., & Brown, T. (1992). Thai sexual behavior and the risk of HIV infection: A report of the 1990 Survey of Partner Relations and Risk of HIV Infection in Thailand. Bangkok: Program on AIDS, Thai Red Cross Society and Institute of Population Studies, Chulalongkorn University.

Smith, E. A., & Winterhalder, B. (Eds.). (1992). Evolutionary ecology and human behavior. New York: Aldine de Gruyter.

Smith, H. L. (1993). On the limited utility of KAP-style survey data in the practical epidemiology of AIDS, with reference to the AIDS epidemic in Chile. Health Transition Review, 3(1), 1-16.

Standing, H. (1992). AIDS: Conceptual and methodological issues in researching sexual behavior in Sub-saharan Africa. Social Science and Medicine, 34(5), 475-483.

Thailand, National Statistical Office. (nd). The social attitudes towards children survey 1993. Bangkok: National Statistical Office.

VanLandingham, M., Knodel, J., Saengtienchai, C., & Pramualratana, A. (1994). Aren't sexual issues supposed to be sensitive? Health Transition Review, 4, 85-90.

VanLandingham, M., Knodel, J., Saengtienchai, C., & Pramualratana, A. (1995). Friends, wives and extramarital sex in Thailand. Institute of Population Studies Publication No. 222/95, Chulalongkorn University. Bangkok: Institute of Population Studies.

VanLandingham, M., Suprasert, S., Sittitrai, W, Vaddhanaphuti, C., & Grandjean, N. (1993). Sexual activity among never-married men in Northern Thailand. Demography, 30, 297-313.

Wilson, D., & Henley, D. (1994, December 25). Prostitution: Facing the facts. The Sunday (Bangkok) Post, pp. 16-17.

Wright, R. (1994). The moral animal. New York: Pantheon.

Manuscript accepted December 16, 1996

We draw on data collected by the project, "The Influence of Primary Female Partners and Male Peers on Male Extramarital Sexual Behavior in Thailand," funded by the Rockefeller Foundation and conducted jointly by the Institute of Population Studies, Chulalongkorn University, and the Population Studies Center, University of Michigan. We gratefully acknowledge useful discussions with Anthony Pramualratana and Mark Van-Landingham, two project co-investigators, as well as comments from Tony Bennett, David Buss, and Margo Wilson on an earlier draft. Finally, three anonymous reviewers provided careful and constructive comments on an earlier draft.

Address correspondence to John Knodel, Ph.D., Population Studies Center, 1225 S. University Ave., Ann Arbor, MI 48104. Fax: [phone]. E-mail:[e-mail].