The Political Economy of Marriage and HIV: The ABC Approach, “Safe” Infidelity, and Managing Moral Risk in Uganda

Abstract

Research has shown that married women’s greatest risk for HIV infection is their husbands’ extramarital sexual activities. Using 6 months of ethnographic research in southeastern Uganda, I examined how the social and economic contexts surrounding men’s extramarital sexuality and the dynamics of marriage put men and women at risk for HIV infection.

I found that Uganda’s HIV prevention messages may be inadvertently contributing to increased difficulty in acknowledging HIV risk and to newer forms of sexual secrecy and that structural determinants, including persistent poverty, intersect with gender inequalities to shape marital risk. After examining a community effort to regulate men’s sexuality, I suggest that HIV prevention strategies should focus more on endogenous forms of risk reduction while simultaneously addressing structural factors that facilitate opportunities for men’s extramarital sex.

Evidence from around the world shows that sexual intercourse within marriage or with a permanent partner puts many women at risk for HIV infection, most commonly from their husbands’ or partners’ extramarital liaisons.1 Women who are economically and socially dependent on their husbands or lovers have difficulty negotiating condom use and inquiring about their partners’ extramarital liaisons,2 both key components of the widely promoted ABC approach—abstinence, be faithful, and condom use. Moreover, structural factors such as labor migration involving separation of spouses, masculine sexual privilege, expectations of female sexual passivity, and domestic violence exacerbate women’s HIV vulnerability.3

Although Uganda appears to have experienced declining national HIV prevalence rates and population surveys report an overall reduction in the number of people’s sexual partners,4 marital sex continues to pose a particular risk for women.5 I examined the social and economic contexts of men’s extramarital sex in southeastern Uganda with the aim of contributing to discussions of risk reduction for HIV infection within marriage.

As a risk group for HIV, married women demonstrate the limitations of traditional public health models that focus primarily on an individual’s risky behaviors without fully accounting for secondary risk associated with a partner’s behaviors and the social and economic contexts that influence an individual’s sexual decisionmaking.6 To understand marital HIV risk, it is important to move from discussions of risky behaviors toward an analysis of how risky situations shape opportunities for men’s extramarital sex, which makes both men and women vulnerable to infection.

Using 6 months of ethnographic research in Iganga District of southeastern Uganda, I examine 2 aspects of the wider context that facilitates or enables men’s extramarital sex. First, HIV prevention messages may be inadvertently contributing to an increased difficulty of acknowledgment of marital HIV risk by men. The widespread circulation of social, moral, and public health messages about HIV/AIDS promulgated by the Ugandan government, donor agencies, religious institutions, and a wide range of nongovernmental organizations has reconfigured the social landscape of HIV risk. In the context of highly visible medico-moral HIV prevention messages to “be faithful” and popular culture and religious discourses promoting monogamous marriages,7 increased secrecy and discretion about extramarital relationships allow husbands (as well as their wives) to manage their public reputations and maintain the appearance of being modern and moral. This does not mean that secrecy and discretion are new forms of reputation management in Uganda. Rather, recent notions of sexual immorality have altered the social acceptability of infidelity, heightening men’s personal motivation for sexual secrecy to avoid public scorn and domestic conflict. Risk, in this sense, is more about the social risk of getting caught in sexual scandals and less about the biomedical risk of catching a sexually transmitted infection.

Second, the combination of persistent poverty faced by many families in Uganda and the large number of young adults orphaned from the first decades of the HIV epidemic has left a distorted economic demographic in which fewer people (most often men) have sufficient economic means and many young adults lack access to resources. Young men often do menial temporary work, engage in small-scale farming on their parents’ land, and delay marriage. This economic situation interacts with gender inequality in ways that make the uncertainty of source of income, living situation, livelihood, and everyday life worse for young women. For instance, historically, most Ugandan women have limited rights to their parents’ property, and many lack the means to obtain advanced education or training. Relationships with wealthier (frequently married) men provide young women temporary social and economic security, facilitating sexual networking (situation in which a cluster of people have had sexual contact with the same person or people and hence are potentially exposed to illnesses from their partners’ other partners) and increasing HIV risk for young women, husbands, and wives.

The same ethnographic methods that revealed unexpected sources of risk, however, also pointed toward new ideas for community-based prevention. There are locally developed, male-initiated techniques for discouraging extramarital relationships and strengthening the marital bond. This endogenous response emerged among members of the Household Development Program of the Uganda Taxi Owners and Drivers’ Association (UTODA) in Iganga. The transportation industry, including taxi and long-distance operations, has long been recognized as being at high risk for HIV infection. UTODA’s technique of sexual regulation builds on members’ commonly stated goals of economic and affective well-being of their households. By emphasizing alternative notions of masculinity characterized by marital responsibilities, affective bonds, and public interaction between spouses, UTODA’s program attempts to increase partner cooperation and communication and reduce marital HIV risk.

This study is part of a larger 5-country comparative project researching marital HIV risk, in which the same research design and protocol were implemented in all countries.8 All findings and discussions in this article refer solely to the Uganda study.

HIV/AIDS AND WOMEN IN UGANDA

Uganda is touted as Africa’s HIV success story. The national HIV prevalence rate among women attending antenatal clinics peaked in the early 1990s, when it reached 15% (30% in urban sites), declined in 2003 to around 4.1% (14% in urban sites),9 and slightly increased in 2005 to 6.7%.10 Considerable debate exists concerning how to read the data, the degree of the decline, and how best to explain Uganda’s achievement in reducing the prevalence of HIV.11 Nevertheless, observers agree that Ugandan President Yoweri Museveni deserves praise for his swift and bold response to the HIV epidemic and his ability to forge a national coalition behind the epidemic and appeal to international donors.12 Museveni responded quickly to the HIV epidemic threat by establishing an AIDS Control Program in late 1986 and by incorporating HIV messages into public speeches.

Although Museveni’s frank discussions played an important role in controlling the epidemic, changing economic and social conditions cannot be ignored. After 20 years of civil unrest (1966–1986), Uganda’s improving economic conditions and political stability in the southern and central regions helped create an environment that allowed many people greater control over their sexual lives. Improvements in health infrastructure, which had virtually collapsed during the country’s civil wars, have increased access to health information, health care, and treatments. Although gender inequalities remain, advances in women’s economic and educational status, greater political representation, and laws against sexual abuse and rape have offered women and girls some protection.13 These significant structural changes and the increased stability that have accompanied HIV trends call into question the extent to which Uganda’s declining HIV prevalence can be attributed primarily to the ABC approach and its focus on individual behavior change.14

As in other countries in sub-Saharan Africa, women in Uganda have been more severely affected since the beginning of the epidemic; they currently account for 60% of all HIV/AIDS cases among people aged 15 to 49 years.15 Key findings demonstrating ways that women are disproportionately affected emerged from longitudinal cohort studies conducted in the country’s longest-running HIV research sites in Rakai and Masaka Districts of central Uganda, a region occupied by an ethnic group (the Baganda) culturally and linguistically similar to the Basoga group discussed in this article. A 10-year study of epidemiological trends in Masaka (January 1990–December 1999) showed that HIV prevalence rates declined for all age and gender groups except for women aged 30 to 34 (increase from 11.5% to 20.0%, P = .03) and 35 to 39 years (increase from 9.0% to 14.7%, P = .06), age groups that often comprise married or ever-married women.16 Moreover, women in Uganda are infected at earlier ages than are their male counterparts and often by older partners, reflecting the age–gender power dynamic in sexual liaisons.17

Another study found that husbands were twice as likely as wives to bring HIV into a marriage, most likely through extramarital sex.18 Similarly, a Rakai study revealed that HIV infection was higher among young women aged 15 to 24 years in a permanent union than in never-married women of the same age range and that a woman’s risk for HIV infection increased the longer she was married.19 Overall, these studies suggest that being in a permanent relationship increases women’s vulnerability for HIV infection.

MARRIAGE AND GENDER IN IGANGA

The research site for this study was Iganga town, the rapidly growing administrative and commercial center of Iganga District and outlying villages. Iganga lies 120 km to the northeast of Kampala, the capital, along the Trans-Africa Highway, a major international artery populated by truck stops and towns with high rates of HIV infection.20 Before British colonial rule began in the 1890s, the Basoga ethnic group of Iganga relied primarily on agriculture and livestock. To meet the cash needs of an increasingly monetized economy, today most households supplement farming with a combination of low-profit activities such as shopkeeping, performing manual labor, driving bicycle or motorcycle taxis, trading, and performing repair work.

Having a salaried position in a school, the civil service, private business, or a development agency elevates a person’s socioeconomic status, but such positions are more readily available to men. Residents view the availability of modern lifestyles (such as formal education, biomedical services, and commercialized leisure activities) and luxury goods (such as televisions, mobile phones, refrigerators, and cars) as desirable indicators of progress. However, access to cash income in Iganga has not kept pace with changing consumer demands, creating tensions between spouses and increased socioeconomic stratification among households.

The Basoga have an exogamous marriage system that revolves around the exchange of bridewealth (goods or money given by the groom or senior men in his extended family to the bride’s family) and patrilocal residential patterns (in which a son and his new wife establish an independent household on his father’s land).21 As in the past, a new bride today often experiences isolation from her natal family, and insertion into a new kin network can leave her dependent on and subordinate to her husband’s clan. In this patrilineal system, the exchange of bridewealth traditionally transfers rights over a girl or woman’s sexuality (and labor) from her father to her new husband. After marriage, any unauthorized sexual access to one’s wife was traditionally considered a theft committed against her husband by her lover; to a large extent, this holds true today. Conversely, a man’s extramarital liaisons are not considered punishable offenses (although, perhaps, socially objectionable) if carried out with a woman over whom no man (i.e., a father or husband) claims rights.22 This difference between the rights of men and women is an important historical trajectory from which to understand men’s infidelity today.

In earlier ethnographic literature, the Basoga were described as being highly polygynous, which was particularly the case for wealthier men whose land holdings allowed for and required more dependants. The Basuga were also described as having relatively unstable marriages compared with other Bantu groups.23 To account for this instability, scholars pointed to the low average exchanges of bridewealth and sympathetic fathers who allowed Basoga daughters to leave unhappy marriages. As in other places in colonial Africa, the high rate of divorce and the perceived autonomy of women concerned British colonial administrators and missionaries, who, in turn, collaborated with male local leaders to institute laws and policies that gave husbands greater control over wives and that made separation more difficult for women.24

Over the last 50 years, monetization, Christianization, and formal education have altered the expectations and realities of marriage and family life.25 When this study was conducted in 2004, although extended family obligations and ties still influenced marriage, residents spoke about marriage as a nuclear project that centered on raising children (including orphans and other dependants), “developing” the household, and acquiring goods and luxury items. Although marriage remains an important marker of adulthood and a social expectation, many younger people delay marriage for various economic and social reasons, such as education.26 Cohabitation without formal solemnization appears to be increasingly common and is considered a “marriage” by both the couple and the community.27 More recent religious movements, such as the popular charismatic Pentecostal church, have put marriage and sexual restraint at the center of their public agenda. Premarital workshops and retreats for young singles are regular events of these churches, but there is some debate regarding the ability of these movements to alter sexual behaviors, and some view these churches as hypocritical. These changing patterns, expectations, and contexts form a crucial backdrop for understanding marital HIV risk.

METHODS

I gathered data between January and June 2004 with 5 Ugandan research assistants—3 men and 2 women. I also drew on my previous ethnographic research, conducted during different periods since 1996, on changing forms of sexual relationships and sexuality among adults and young people in Uganda. The primary data for this article were obtained from 72 marital life histories conducted with husbands and wives from 34 marital unions28 (Table 1 ▶). To enhance the generalizability of the small sample size, the research team used systematic ethnographic sampling based on previous research findings29 and recruited participants who would provide the greatest diversity along 3 axes: duration of marriage, current socioeconomic status, and mobility and migration patterns.

TABLE 1—

Number of Marital Unions, by Socioeconomic Status (SES) and Husband’s Mobility: Case Study Interviews of 72 People (Representing 34 Unions), Uganda, 2004

	No. of Marital Unionsb
Marital Life Cyclea	Lower SES	Middle SES	Higher SES
Newlyweds (couples married 5 years or less)
Mobile or migrantd	1	2	3
Not mobile or migrant	3	1 (2 wives interviewed)	0c
Adults (couples married 5–15 years)
Mobile or migrantd	1	3	2
Not mobile or migrant	3 (1 union with 2 wives interviewed)	2 (1 with 2 wives interviewed)	1
Elders (married ≥ 15 years, with grandchildren)
Mobile or migrantd	2	2 (1 with 2 wives interviewed)	1
Not mobile or migrant	3	2	2 (1 with husband not interviewed, 1 with 2 wives interviewed)
Total	13 (27 people)	12 (27 people)	9 (18 people)

^aThe marital life cycle stage for polygynous unions and individuals who had been previously married were classified according to (1) the ages of the spouses and the offspring and (2) the community’s perception of that union. Furthermore, polygynous unions were typically classified according to which of the interviewed married women had been in the marriage longest.

^bIt is common for people, particularly adult and older individuals, to have undergone fluctuations in wealth and socioeconomic status over time given Uganda’s tumultuous economic past. Unions were classified according to their current economic status. The person’s socioeconomic level at the time of an extramarital liaison was taken into account during data analysis to understand how socioeconomic status shapes opportunity structures for extramarital sex.

^cI was not able to locate any newly married, wealthy couples who were not highly mobile or migrant.

^dMobile or migrant means the husband spent the night away from home more than 1 night a week.

The research team visited prospective participants’ homes to begin the informed consent process, complete demographic surveys, and become familiar with the domestic space and interaction between the spouses. To prevent discussion from being restricted because of differences in age, gender, or socioeconomic status, each participant was allowed to decide who from the research team would conduct the interview. Husbands and wives were interviewed separately in 3 interview sessions, each lasting approximately 1 to 2 hours. Interviews were conducted in English or Lusoga (the local language), tape-recorded, translated into English, and coded.

Because understanding the context of how marital and extramarital relations are shaped by wider economic, social, and political forces was important, participant observation was employed to construct this wider context. Participant observation involved observing men and women in various settings to examine how notions of sexuality, gender, reputation, and extramarital liaisons were discussed, enacted, and performed in everyday lives (Table 2 ▶). This method also allowed the researchers to understand the difference between what people say they do (ideology) and what people actually do (practice) and to observe how various settings and audiences shape the presentation of sexual selves and activities.

TABLE 2—

Other Ethnographic Methods Used in the Study: Uganda, 2004

Method	Description of Method and Its Relationship to Project’s Objectives	Description of Sample
Participant observation	Six months of observing domestic and social life to obtain a greater understanding of people’s marital experiences, men’s and women’s patterns of socializing, and local discourses and practices surrounding extramarital liaisons	Domestic spaces Evening leisure places Offices and businesses Markets Hotels and lodges Weddings and engagement parties Age- and gender-specific gathering places Official public celebrations Community functions and gatherings
Key informant interviews	Formal and unstructured interviews with people who have particular insight into marriage, general sexuality, or men’s extramarital sexuality	Women in relationships with married men (n = 3) Religious leaders (n = 3) Managers of hotels and discos in Iganga (n = 2) Women’s rights and HIV/AIDS agencies (n = 2) Commercial sex workers (n = 2) Male pimp Iganga district health officer Dispensary nurse in district hospital Owners of shop with local medicine Chief magistrate in Iganga Court Commercial sex counselor Sex education teacher at secondary school Chair of Iganga’s UTODA Manager of bridal shop Editor of gossip and society newspaper HIV-positive widows (n = 4)
Archival research	Collecting written, public speeches and discourses and other materials to understand how sexuality is constructed in the media and in public culture	Popular media (including newspapers, radio shows, magazines, and tabloids) HIV prevention materials Reports on HIV/AIDS Demographic surveys Court records on divorce and domestic disputes Religious materials on marriage, sexuality, and the family

Note. UTODA = Uganda Taxi Operator and Driver Association.

I interviewed key informants about sexual and marital relationships, opposite- and same-sex socializing, opportunities for male extramarital sex, HIV/AIDS and sexual health education, gender expectations, and sexual practices. Finally, we collected archival data pertaining to marriage and sexual relationships from the print media, court records, HIV/AIDS agencies, legal and women’s organizations, popular songs, radio shows, local dramas, and advertisements. In archival research, we were particularly interested in public constructions of marriage, gender, and extramarital liaisons to ascertain how public discourses might influence local understandings of the different expectations, standards, and consequences of men’s and women’s sexual behaviors.

The data were analyzed: (1) to construct local forms of, variations in, and understandings of marital and extramarital relationships; (2) to delineate the relationship between the axes of participant diversity and marital and extramarital relationships; (3) to understand how marital and extramarital relationships map onto the life cycle of the union; and (4) to compare husbands’ and wives’ perceptions of their marriage and of their spouses’ extramarital activities.

SOCIAL MEANINGS AND THE ORGANIZATION OF EXTRAMARITAL RELATIONSHIPS

Discourses of love, trust, monogamy, and partner reduction (i.e., reducing the number of concurrent and serial sexual partners) feature prominently in Uganda’s popular culture and growing evangelical movements. Nevertheless, male extramarital sex remains the norm in Iganga. Most of the husbands in this study reported having an extramarital relationship during their current marriage. Although we did not ask directly the number of extramarital liaisons the men had had in the past year or ever (in our initial field testing of the interview guide, we found that this discouraged openness and interrupted the flow of the interview), we did ask men to describe their most recent extramarital liaison. Many men, particularly younger ones and those who had worked away from home, recounted extramarital liaisons that had occurred in the past 1 to 2 years. Five of the 33 men, mainly of middle or elite socioeconomic status, reported having had no extramarital relationship during their current marriage; over the course of the research, however, we learned of evidence indicating that at least 2 of them did engage in at least 1 extramarital relationship.30

Most wives in our marital life history sample suspected that their husbands either currently had extramarital partners or had partners in the past, and they expressed disapproval of these liaisons. In fact, accusations of infidelity were the basis for ongoing marital conflicts and dissatisfaction among wives. When I asked one participant how AIDS had changed marriage in Uganda, she replied, “Whenever your husband comes to sleep with you, you feel worried that he is going to infect you with AIDS.” Her immediate association of men’s extramarital sex with women’s vulnerability to HIV infection was a common sentiment among women, indicating widespread knowledge about HIV risk from a partner’s sexual behaviors and histories.

Women generally felt that extramarital liaisons were a breach of marital trust (obutalia bwesigwa, literally “being untrustworthy”) and negatively affected the economic as well as emotional well-being of the household. By contrast, men drew less connection between extramarital sex and the state of their marriage. For example, men’s answers to open-ended questions such as “Describe your relationship with your wife when you had the relationship with the woman” suggested that marital happiness or sexual satisfaction did not necessarily deter men from engaging in extramarital liaisons, although marital problems were considered acceptable reasons for seeking outside companionship. Nor did men equate extramarital sex with lack of love, respect, and honor for their wives. In some cases, men felt that having extramarital partners enabled them to respect their wives by not “disturbing” them with requests for sexual intercourse during certain periods such as pregnancy, breast-feeding, illness, or menstruation.

Local understandings of “being faithful” and “risk” differ in significant ways from public health concepts. A man can be “faithful” as long as he continues providing economic support and maintains a respectful public reputation by keeping knowledge of extramarital relations away from people in the home and in the family’s social networks. Whereas all participants recognized the biomedical risks of sexual activity, the social risks involved in having certain people know of one’s affair was of more immediate concern. Managing social reputation requires conflicting actions. Masculine status among social peers is often achieved through suggested or actualized sexual prowess; however, public knowledge of a man’s sexual exploits that extends beyond his peers can mar his reputation within the wider community. The latter is especially true with men who claim to be highly religious or “born again.” Men might therefore use friends as alibis or accomplices for extramarital activities but might also go to great lengths to keep knowledge of extramarital relations away from people in the home and in the family’s social networks.

Men’s extramarital relationships fall into 3 general categories: (1) longer-term relationships that may eventually evolve into secondary households; (2) short-term or 1-time opportunistic encounters near a man’s residential area, frequently in the same town; and (3) relationships occurring while the man is away from home, often because of work. All 3 types were found among men of all ages and socioeconomic groups but varied in terms of opportunities, configurations, and personal meaning. For example, wealthier men and older men appeared more likely to have longer-term liaisons, whereas recently married men’s liaisons frequently involved male socializing and alcohol-related activities, premarital girlfriends, and work-related mobility.

MORAL STIGMA AND REPUTATION MANAGEMENT STRATEGIES

An unexpected finding of our research was that HIV prevention discourses have increased the moral stigma of extramarital sex, inadvertently heightening people’s motivation for sexual secrecy and personal denial that an extramarital liaison puts one at risk for HIV. In promoting fidelity as an HIV risk reduction strategy, the poster shown in Figure 1 ▶ equates infidelity with immorality and depicts married people seduced by potential lovers as weak, immoral, and backward. Although the infidelity-equals-immorality message may discourage some people from engaging in extramarital liaisons, concealment of extramarital liaisons or denial of risky actions offers alternative routes to preserving one’s self-image, particularly for self-defined religious men or members of other social groups that frown upon infidelity.

FIGURE 1—

A Ugandan HIV prevention poster depicting faithfulness as the route to morality and marital happiness.

Note. In this poster, infidelity is equated with immorality, and married people seduced by potential lovers are represented as weak, immoral, and backward.

HIV prevention campaigns also create a distinction between good “safe” people and immoral “unsafe” people. Figure 2 ▶ shows a popular poster from the early 1990s in which people in a boat attempt to save immoral people drowning from risky behaviors, including infidelity. In addition, discourses about the negative social consequences of illicit liaisons are constantly reproduced in the media, further contributing to moral stigmatization and a resulting desire to conceal and deny risky behaviors.

FIGURE 2—

A popular HIV prevention poster from the 1990s that distinguishes good “safe” people from immoral “unsafe” people at risk of drowning.

Note. The title reads, “Do not drown in the AIDS flood; always be on board.” Although the message might lead some people to change their risky behavior, it could induce others to deny it.

The changing social acceptability and patterns of infidelity clearly emerged in our interviews with men and women. Whereas participants recounted numerous stories about problems in their parents’ marriages caused by their fathers’ flagrant extramarital liaisons, we heard fewer current cases of unapologetic flaunting of mistresses. Several participants told how their father would bring a mistress into the matrimonial bedroom, asking the official wife (the participant’s mother) to sleep in a different room. When asked how the problem of infidelity in her parent’s marriage was resolved, 1 middle-class woman in her late 60s explained that her mother, protesting her husband’s repeatedly bringing other women into their bedroom, eventually refused to sleep in her matrimonial bed, choosing to sleep on the floor instead. During an informal discussion, a young woman remembered her father instructing her and her siblings to refer to his short-term mistresses as “mother,” a pattern that repeated itself frequently during her youth. We heard many stories of men in the past making little effort to conceal their infidelities and showing scant regard for their wives’ feelings.

Today, however, infidelity cases that emerge publicly have various social consequences, ranging from marital conflicts to public embarrassment to media coverage if prominent people are involved. Consequently, men want their sexual improprieties to remain secret. During 1 interview, for example, a young middle-class husband said that to be a “good husband” he was “faithful, respectful, and kind.” He explained that his last affair was a year ago, when he was working on the Uganda–Congo border and feeling homesick, lonely, and “weak.” He maintained that he had been faithful since returning to Iganga. A month after our interview, however, the man’s pregnant wife caught him in the outdoor drinking area of town (area where people go for evening socializing and drinking) sharing local brew with his lover. He had been having a relationship with the woman, a teacher at a primary school, for several months. The public scene caused by the enraged wife, the fearful mistress, and the apologetic yet denying husband created a wave of gossip in town and left a publicly humiliated husband too ashamed to visit the drinking area for a few weeks. As people in town saw it, his guilt was less about his extramarital affair and more about getting reprimanded publicly.

Another significant change we found was a shift away from formal polygyny—a process that involved agreement among the husband, the existing wife or wives, and the kin of a new wife—toward the establishment of informal second households, often established without the knowledge of the existing wife or the new wife’s family. In several cases, an extramarital liaison gradually transformed into a more permanent secondary household after the birth of a child or because of growing affective or economic ties between the 2 partners. We found that secondary households were more likely to occur among (1) socially elite men or men who identified themselves as highly Christian, for whom formal polygyny is socially unacceptable, and (2) men who resided away from their marital home and wanted a permanent partner to tend to their sexual and domestic needs. As an example of the former, 1 woman said she was unaware that her husband, a high-ranking local police officer, had 3 other wives residing in neighboring villages until those wives and their children were announced at his funeral rites and wished to claim some of his estate. The secrecy surrounding men’s secondary households inhibits a couple’s ability to openly discuss the husband’s other sexual partners.

Over the last 50 years, assertions of male sexual privilege have been challenged by discourses about the risks of infidelity (such as HIV awareness campaigns) and newer forms of moral regulation (such as evangelical movements and public interest in sexual scandals). Men still engage in extramarital affairs and concurrent liaisons, but managing those affairs has emerged as an important task as these men try to preserve the public face of their marriage and of themselves.

CONDITIONS OF MARRIAGE AS A FACTOR IN MEN’S EXTRAMARITAL LIAISONS

Men’s perceptions of their marriage play an important, yet overlooked, role in their extramarital sexual activities. When participants were asked to explain how their relationship with their spouse differed from that of their parents, a noticeable shift from authoritarian husbands toward joint decisionmaking, cooperation, and communication between spouses was evident. This shift in power has been occasioned by gender equality campaigns and changing economic needs that require the attention of both husband and wife. Although wives generally see this shift as positive, some husbands are more ambivalent. According to 1 man in his late 60s:

Women [in the past] had good discipline. They were not behaving like today’s young wives. There was a lot of respect given to the husband. If the man said, “I want to find such a thing done [when I return],” that would be final and by the time he came back, the thing would have been done. The wife would not have any complaint or anything to add. Marriage then was good, although there was a lot of wife beating.

Comments about the burden of gender equality on the marriage were expressed by men who felt that their power was being attacked by multiple forces: modernity, the women’s movement, and President Museveni. Unlike in the past when local courts and elders’ councils worked to buttress husbands’ authority, husbands today find few places that are sympathetic to their unquestioned authority. Extramarital liaisons provide a venue for men to repossess traditional masculine authority and independence in marriage.

We also found that expectations of modern lifestyles and luxuries require economic means that many men are not able to acquire, yet wealth remains an important part of masculine identity. Engaging in extramarital relationships offers some men an alternative route to masculine identity. Whereas boasting about one’s economic abilities is a typical courtship maneuver, many men are often unable to provide the lifestyles and luxuries they promised before marriage, leading to complaints from wives and marital conflicts. Three years into his second marriage, 1 middle-income shop owner reflected on his wife’s material expectations:

My parents used to respect each other. . . . But for me, there is a big problem in my marriage. My marriage needs money to make my woman look good. We need a lot of counseling in regards to the behavior of my wife; she needs to be more polite.

For this man and others, the trend toward marital communication has, ironically, opened the way for wives to express their grievances and dissatisfaction. Hence, taking extramarital lovers allows some men temporary solace from household economic pressures and their wives’ complaints.

SOCIAL AND ECONOMIC CONTEXTS OF MEN’S EXTRAMARITAL ACTIVITIES

Residents of Iganga recognize that poverty drives women into transactional relationships (i.e., sex for money),31 but they more frequently comment that the growing supply of unmarried, unemployed females poses a greater threat to marriage. The pool of unmarried young women appears to be increasing because of several demographic and economic phenomena: (1) the economic inability of younger men to formally marry and establish permanent residences; (2) a sizable population of young people who have come of age as HIV orphans, leaving many young women without parental economic or social support; and (3) a decline in fathers’ cultural rights over daughters, resulting from fathers’ inabilities to economically provide for older offspring and, as a result, frustrated adolescent daughters securing their own finances. Hence, whereas young men delay marriage to try to secure economic resources, relationships with older or wealthier men provide young women an opportunity for social stability and economic support.

Residents of Iganga echo popular culture when they depict attractive, nicely dressed, leisure-seeking women as preying on married men, a construction that stands in stark contrast to public health and feminist analyses. “There are just too many unmarried beautiful young women without money, and they simply tempt us,” 1 wealthy man remarked, laughing lightly.

Over the last 2 decades, urbanization, education, and changing labor markets have presented new forms of men’s labor-related migration, providing other opportunities for extramarital liaisons. We found important socioeconomic differences in how migration interacts with opportunities for extramarital activities. Younger and poorer married men who migrate in search of work typically leave their wives in rural areas because they cannot afford to relocate with them or so the wives can serve as “place holders” to maintain the household and farm and to care for children. Wealthier men are more likely to relocate with their families to cities, and those who remain in Iganga take frequent overnight business trips without their wives. Men explain that loneliness and the desire for companionship while away from home lead them to seek extamarital relationships. If a man has his own living space in the place of migration, it is almost inevitable that he will invite a woman to live with him to tend to his domestic, affective, and sexual needs. Furthermore, overnight workshops and trainings sponsored by the government, nongovernment organizations, and various development agencies have become a regular part of the lives of wealthier and professional residents. According to interviews and participant observations, these residential workshops (that often offer a generous per diem) serve as potential opportunities for men and women to engage in extramarital liaisons with local residents (such as barmaids or sex workers) or each other, yet these HIV transmission opportunities are often overlooked by HIV risk-reduction efforts.

Finally, it is almost expected that younger men will spend evenings in town socializing, drinking, or mingling before returning home. In explaining this pattern, 1 middle-class public servant stated, “If you go home, what is there for you? Your wife quarrels with you about this and that, wonders why you are not looking for money. Or, she and the children have their chores. So you sit alone with nothing to do.” The more limited mobility of wives and their shrinking social networks after marriage hinder wives’ ability to know about their husbands’ evening sexual liaisons.

Community-Based Regulation of Male Sexuality

Alongside the pervasiveness of men’s infidelity in Uganda is tremendous public anxiety about the detrimental effects these relationships have on society and marriage. Consequently, UTODA developed a peer-based system of sexual regulation in an attempt to hold members accountable to the group and to their wives. The group’s chairperson explained that members felt that the social and economic demands of extramarital liaisons presented a major obstacle to the group’s overall goals of household economic development. He also elaborated on 2 other sets of concerns. First, members of UTODA, like other transportation sectors in sub-Saharan Africa, had witnessed many HIV-related deaths since the beginning of the epidemic and had experienced the economic burden of assisting their colleagues with medical expenses, funerals, and caring for surviving dependants. Second, members of UTODA said they spent a great deal of office time trying to counsel irate women who brought complaints about other UTODA members who the women accused either of excessive “womanizing” or of neglecting to assist with domestic expenses or child support. In many instances, UTODA members did not know if the women were wives or lovers of coworkers, leading to confusion over how to handle such disputes.

UTODA’s development group attempts to replace the diminished influence of the kinship network in settling marital conflict with a newer form of civic association based on occupation. To actively participate in the program, each member is required to bring his wife (or wives) to every meeting, stand with her (or them) and recite her name (or their names) during the opening roll call. At each meeting, a designated couple (chosen randomly) receives a gift of money supplied by the organization’s membership dues. The master of ceremonies hands the money to the husband and wife together, symbolically signifying joint ownership of household assets. At 1 meeting, the announcer jokingly pushed aside the husband and offered the couple’s proceeds to the wife, stating that wives are more responsible with money. A roar of affirming laughter ensued.

Over the 3 meetings I attended, I witnessed unusual jocularity and friendly bantering about spouses, with comments such as, “They are dressed alike today,” and “Asumani is a big man and manages 2 strong women.” Given traditional social norms that limit public socializing between spouses, the public affection and playfulness were notable. Wives became more familiar with each other and less shy about their public role as a wife in social functions. The speeches revolved around the importance of harmonious marital unions, men’s avoidance of kimansulo (a word used to refer to women who dress provocatively and tempt men), and households free of disease, rumors, quarrels, and financial difficulties. The sexual regulation system created by UTODA attempted to emphasize and reward notions of masculinity that centered on being responsible not only for the economic well-being of the household but also for the affective well-being of the marital bond.

CONCLUSION

Ironically, Uganda’s 2 decades of massive HIV prevention efforts have worked to reconfigure landscapes of social morality in ways that present new obstacles to HIV prevention. The widespread circulation of social, religious, and public health messages that present infidelity and polygyny as risky, immoral, backward, and dishonorable have had the unintended effect of creating new motivations and avenues for sexual secrecy. Although almost all men and women in this study recognized the health risks of extramarital liaisons, the risk of getting a bad public reputation by being caught in an illicit relationship presented more immediate cause for concern than the distant, unforeseen effects of contracting HIV. To minimize risks to reputations, individuals employed strategies such as secrecy and discretion to manage sexual relationships that were deemed socially immoral.

There also has been a gradual transformation from formal polygyny, in which households of co-wives were somewhat interconnected, to a pattern of informal secondary households that often remain autonomous and hidden from each other. Whereas in the past women in polygynous unions ideally owed fidelity to their husbands, women of informal or secondary households have no reason not to have multiple male partners. In fact, we found that some had other male partners to provide additional economic and affective support, thus intensifying HIV risk for all sexual partners. Furthermore, we found that longer-term liaisons might put people at risk because condom use tends to decline when relationships are based on emotional attachment and intimacy.

Although men commonly joke with their friends about multiple lovers as a marker of masculinity (“A man with 1 wife is his own co-wife,” goes a saying; “When the wife is sick he has to cook and clean”) or employ the assistance of a friend in carrying out sexual liaisons, most men today prefer to keep lovers and extramarital trysts away from wider social and kin networks for fear of social consequences such as gossip or complaints from wives. The increased moral stigmatization surrounding infidelity, combined with the high value placed on monogamous marriage, may influence respondents’ willingness to report extramarital sexual activity. This might partly explain why men’s reporting of extramarital sexual partners seems suspiciously low in large-scale surveys in which men were asked during rapid questionnaires about partner reduction and the number of nonmarital partners over the last 12 months.32 Although there may be overall partner reduction in Uganda, our data from participant observation and life histories concur with other community-based research studies that report a much higher occurrence of and variability in extramarital sexuality than are captured in rapid surveys.33

On the basis of my findings, I recommend 3 marital HIV risk reduction strategies that emphasize the need to address structures that provide opportunities for extramarital liaisons. First, whereas women in this study acknowledged their own risk for HIV infection, men tended to downplay their own vulnerability. The emphasis of many HIV prevention campaigns regarding girls, women, and commercial sex workers has allowed men to deny their own susceptibility and perhaps their own responsibility in the spread of HIV. Given men’s historical role in transmitting HIV to wives, younger women, and other sexual partners, it is crucial that campaigns highlighting women’s vulnerability be accompanied by programs that address men’s risk behaviors and the economic, social, and demographic conditions that facilitate these behaviors. We found that men’s participation in extramarital sex is frequently structured around work-related mobility and commercialized leisure activities that involve alcohol consumption; collaborations with key players from occupations and leisure places could therefore provide ideal opportunities for developing risk-reduction strategies. These programs could strengthen alternative social activities or notions of masculinity that are currently available in arenas that encourage men not to engage in extramarital liaisons. The UTODA sexual regulation program offers insight into how sexual and marital well-being can be positively integrated into people’s larger projects of household and individual development. Risk-reduction strategies that build on a community’s resources, understandings, and needs have a greater chance of being sustainable and effective than ones that impose external concepts, assumptions, and priorities.

Second, although many women and some men in our study expressed anxiety about their own HIV status and that of their spouse, most participants had never been tested for HIV. Fear of receiving an HIV-positive test result and uncertainty about the location of testing sites were cited as primary reasons for not getting tested. Anxiety over HIV status was worsened by silence within marriage about sexual histories and possible HIV risks. Notably, hardly anyone felt they had had a productive marital dialog—as opposed to accusations and arguments—about possible HIV risks in marriage, including past and current sexual partners. Yet the expressed desire among men and women for better marital communication suggests ideal opportunities for improving voluntary counseling and testing services for married people. In addition, in this study I found that although people knew the biomedical basics about HIV transmission and had firsthand knowledge of the later stages of AIDS-related illness and suffering, many people did not know how to live comfortably with HIV and how to gain access to medical care, which might also inhibit people’s desire to get tested. My preliminary research with HIV-positive groups suggests that increasing people’s knowledge about how to live with HIV and how to obtain care and support increases people’s chances of getting tested for HIV and joining prevention programs for people living with HIV/AIDS.

Finally, the current prevention message of “be faithful”—the marital prevention strategy in Uganda’s globally recognized ABC approach—may be inadequate, unsustainable, and potentially counterproductive when only 1 partner is faithful, certain acts of infidelity are not considered socially harmful, and structural factors including mobility, masculine sexual privilege, and social and economic inequalities support extramarital liaisons. Because extramarital liaisons are embedded within wider structures of inequalities that exist not only between men and women but also within society as a whole, our findings support poverty reduction as an integral element of HIV prevention.

Early in the sub-Saharan epidemic, wealthier men were among the first to become infected because their greater resources provided access to greater numbers of sexual partners. In line with more recent studies, however, I suggest that risk is also an element of poverty and economic gender inequality.34 Specifically, I found that, whereas wealthier men are attractive to young women in search of economic stability and modern lifestyles, lack of money might lead poorer men either to delay marriage (thereby increasing the pool of unmarried young women) or to engage in extramarital liaisons as an alternative route to masculinity. Policies and programs that enhance women’s educational and economic opportunities but that neglect younger and poorer men risk doing a grave disservice to the community at large. Poverty is not the only force driving the epidemic; the tremendous gap between the wealthy and the poor that puts all economic sectors at risk is also to blame. Structural determinants that promote men’s extramarital sex are not addressed, and if both husbands and wives are not included in prevention programs, interventions targeting marital HIV risk will be ineffective and unsustainable. Furthermore, in designing HIV prevention strategies, serious attention should be given to endogenous forms of risk reduction that have emerged in response to locally felt needs and that have the aim of changing existing norms that contribute to negative social, economic, and health consequences.