Sex, Money, and Premarital Relationships in Southern Malawi

Abstract

In this paper, I argue two main points: (1), the purpose of money exchange in premarital, sexual partnerships in rural Malawi extends beyond the alleviation of female partners’ economic constraints, and (2), by elucidating this broader purpose, it becomes possible to recognize where women exert control over their own sexual selves. These findings come from field observations and a rich set of in-depth interviews (N=54), bolstered on occasion by survey data, conducted with young women and men, aged 15 to 24 years, in the Balaka district in the southern region of the country. My research demonstrates that, contrary to typical expectations, money and gift transfers in sexual partnerships are part and parcel of the courting practices of young Malawian women and men. Transfers are as much about the expression of love and commitment as they are about meeting the financial “needs” of women or the acquisition of sex for men. Using narrative information to shed light on the semiotics of the sex-money link, these findings from Malawi offer a new perspective that broadens usual interpretations of “transactional” sex, the understanding of which is critical in fighting AIDS.

Introduction

The current HIV program prevention emphasis on protecting young women in sub-Saharan Africa from risk of infection is understandable. Throughout the region, recent data show that 4.3% of women aged 15 through 24 are infected with HIV compared to 2.1% of men the same age (UNAIDS, 2006). Malawi is no exception, where the 2004 Demographic and Health Survey found similar gender differences in HIV prevalence among the youth cohort (MDHS, 2004). To account for these disparities, explanations have typically highlighted age-discrepant relationships, inability to negotiate condom use in sexual relationships, or unequal access to key resources and the concomitant use of violence or coercion by men over women (Gregson et al., 2002; Jewkes, Levin & Penn-Kekana, 2003; Lary, Maman, Katebalila, McCauley & Mbwambo, 2004; Magnani et al., 2002; Nzyuko et al., 1997; Smith Fawzi et al., 2005). Despite notable exceptions (e.g. Helleringer & Kohler, 2006; Potterat, Gisselquist, & Brody, 2004), the role of poverty and gender inequality in accounting for the AIDS epidemic in sub-Saharan Africa is indeed a common thread throughout much of the literature, policy and media reports (e.g. Dunkle et al., 2004; Maman et al., 2002; LaFraniere, 2006). The practice of “transactional sex”—where women and girls engage in sex for material gain—in particular is the oft-cited mechanism through which poverty and inequality work to bring about high-HIV rates (Hallman, 2004; Haram, 1995; Leclerc-Madlala, 2003; Meekers & Calvès, 1997; Nzyuko et al., 1997). Relatedly, many surveys include a question asking women if they had ever received money or gifts in exchange for sex (e.g. MDHS 1996). But how should we as analysts interpret the “yes” responses? Should we consider these women to be sex workers? Or are they solely victims of circumstance, unable to make choices other than engaging in sex for money? Alternatively, are there situations where money-sex exchange is less clear, the price and the timing of delivery unspecified? Is there room for courtship as well as love and lust? And what about those who are not infected with HIV? In considering the case of Malawi— the site of the present work and where the epidemic began over two decades ago and—national estimates are that approximately 14 percent of women and men are infected, which means of course that 86 percent are not.

Interesting to note is Luke’s (2005) research in high-HIV prevalence Kisumu, Kenya—the only study I know of to have empirically and rigorously assessed the extent to which sugar daddy relationships exist in a population. Their prevalence was much lower than expected, with only 4% of all sexual partnerships in the sample qualifying as sugar daddy. Discourse within another body of literature—largely in anthropology—has provided alternative (but not always distinct) understandings of sex and money exchange. This work has centered on, for instance, the social change brought about by modernity and globalization, which in turn is argued to have altered contemporary sexual relations between women and men (Ankomah, 1992; Leclerc-Madlala, 2003), or on the commanding roles women have over their sexuality (e.g. Bujra, 1975; Cornwall, 2002; Wardlow, 2004), or on the grey area characterizing sex and money linkages more generally (Cole, 2004; Nyanzi et al., 2004; Schoepf, 1993; Wojcicki, 2002). This literature counters accounts of women as invariably vulnerable and begins to acknowledge the cultural milieu in which sexual decisions are made (see also Leclerc-Madlala, 2002). Yet two major shortcomings remain in this work. The first is that while most (66%) sub-Saharan Africans live in rural areas, nearly all of these studies are in peri-urban or urban areas (Population Reference Bureau [PRB], 2006). The second is that while this research tends to interpret money exchange in the realm of sexuality as somehow different from Euro-American ideas of prostitution, assumptions persist that were they less economically constrained, women would have more sexual power (for exceptions see Cornwall, 2002; Wardlow, 2004). Finally, these studies all miss an important opportunity to identify which women are able to avoid infection and promote their own health, and how they are able to do it. In this article, I thus advance upon this previous literature by recognizing that even in high HIV prevalence contexts such as Malawi, most young adults are not infected. I use ethnographic inquiry to explore the exchange of money in premarital partnerships in a poor, rural setting, where a global economy is only beginning to penetrate.

Methods

The foregoing suggests that understanding the money and sex link requires an examination that accounts for how exchange creates and shapes the sexual practices among young people. To do this, I spent several months living in the southern district of Balaka in Malawi in 2004–05. This research is affiliated with a larger study, the Malawi Diffusion and Ideational Change Project (MDICP), an ongoing, four-phase panel study conducted in the three main regions (Balaka, Mchinji, Rumphi) of rural Malawi—with 125 villages in total—that investigates social processes and family planning and AIDS. The MDICP is a joint project between the University of Pennsylvania (PENN) and the Malawi College of Medicine (COM; see www.malawi.pop.upenn.edu). In 2004, the MDICP’s third wave, a new youth sample (N=1001) was added to the initial 1998 sample. I present findings from 54 in-depth interviews randomly selected from a sub-set of this 2004 survey sample, in 17 villages in Balaka.

Malawi, with a population of 12.5 million, is 86% rural, with most people living from subsistence farming supplemented by limited small-scale income-generating activities that are typically more available to men than to women. Agriculture accounts for 39% of its GDP, largely generated from smallholder farmers, with its main products being tobacco, tea, coffee, and cotton. The country is poor, with the GNI mean for Malawi at $160 (World Bank 2004). Its fertility rate is falling, but remains high at 5.9 children per woman. The Balaka district is typical of other Malawian communities; it is rural and poor, and sources of income typify other rural areas in the country (Mtika, 2000). Unlike Malawi’s central and northern regional districts, however, in Balaka most communities follow a matrilineal kinship and lineage system, and residence following marriage is usually matrilocal (Peters, 1997). While women tend to marry young (on average 18.1 years), one implication of this matrilineal/matrilocal characteristic is high levels of divorce, with over 50 percent of first marriages ending rapidly (Reniers, 2003). This study’s villages are dispersed over several kilometers around the Ulongwe trading center. Access to a main, tarred road is available by public transport, which is not always affordable. The urban areas of Zomba and Blantyre are approximately two hours away. Of the present sample, two-thirds are Muslim and ethnic Yaos, and the remainder are ethnic Chewa and some people of Christian denomination. The mean age of this sample is 18.2 years. About a third of the girls and a quarter of the boys were married when interviewed. Slightly more than half of the boys and slightly less than half of the girls were still in school.

Topics included in a loosely structured interview guide were intended to get at aspects of boys’ and girls’ decision-making processes in the sexual realm, and included partnership beginnings and endings, gift and money exchange, and beliefs and practices associated with sexual behavior (eg. condom use, frequency of sex, and personal motivations). The final interview guide was purposively flexible and unobtrusive, so that fuller responses could be elicited and new themes allowed to emerge spontaneously (Massey & Zenteno, 2000). The ordering of the topics was left to the trained interviewer so as to circumvent an atmosphere of formality, considered especially important given the sensitivity of the topics associated with sexuality. Married respondents were asked about current spouses as well as any premarital partners. Interviews took place in respondents’ homes, but away from distractions and in private, such as under a tree apart from the main dwelling unit. Informed consent was obtained from each respondent, and, for those under 18 years of age, from their parents as well. Ethics approval for this research was received from Institutional Review Boards at PENN and COM’s Research and Ethics Committee.

Experienced local interviewers were matched according to the sex, age, and ethnicity of the respondent, and were trained by a Malawian supervisor with several years of field experience in qualitative (and well as survey) research data collection and me. After a careful selection process of local interviewers—largely chosen, in addition to demographic characteristics, because of their writing skills (in English), conversational skills, and a general ease and openness to discussing sensitive issues—an extended training ensued.

During the interviews, recorders were not used, as they were thought a probable source of discomfort to respondents in this local context, and particularly for younger women with little to no sexual experience. Premarital sexual relationships are often secretive (Haram, 1995; Wight et al., 2006), and a sexually active fifteen-year-old girl may feel less comfortable speaking about her relationship were she asked to record her words into an unfamiliar-looking machine. Therefore, to minimize the potential for undue anxiety, and keeping in line with the intended conversational tone of the interview, interviewers instead took notes.

Two steps were taken to compensate for the limitations this method placed on the precision with which the interviews could be reported and thus the potential loss of important, translatable information. First, the interviewers, with the aid of their notes, immediately attempted to reconstruct each interview in writing and by translating into English. I immediately read each interview to cross-check the reconstruction and translation. These steps allowed for quick clarification of any confusion with the translation or with uncertainties over the meaning of any of the interviewer’s writing; more broadly, this process also allowed me to respond more efficiently to general problems with the interviews. The speed of the verification process improved the interviewer’s ability to remember certain details and thus made it easier to for me to address a concern. Despite these steps, I acknowledge that the translated material still depends on what the interviewers could recall. Nonetheless, I maintain that informal methods, like conversational interviews without the use of recorders, provide better access to local understandings and complex processes related to AIDS (Watkins & Swidler, 2006), and that the method is similar to what ethnographers do when writing field notes (e.g., Power, 1994). In addition, during the many months I spent working in Malawi collecting data for this project, I had numerous, informal conversations with key informants and acquaintances, colleagues, and strangers that have also been invaluable to these interpretations.

Results

Premarital Partnerships

In Balaka and in all of Malawi, a sexual partner outside of marriage is known in chiChewa as a chibwenzi (pl. zibwenzi). The word is gender-neutral, and loosely translates to mean a friend but is understood to be a sexual partner, or a boyfriend or girlfriend. Zibwenzi-partnerships can be casual, lasting for one month or less, but can be, and often are, more enduring, possibly ending in marriage. Within premarital partnerships, a chibwenzi may develop into the more serious chitomelo, or fiancée. It may also happen that young people marry while by-passing the chibwenzi or chitomelo-phase altogether, although this appears to be less common. “Hit and runs”, or one night stands, are usually not considered to be zibwenzis, but at times they may be, especially if the sexual encounter occurs more than once. The MDICP survey data show that hit and runs are relatively common—comprising about 15% of all non-marital encounters— but when unmarried youth talk about their sexual partners, they usually mean more committed boyfriends or girlfriends. Most unmarried women and men (54%) in the Balaka qualitative sample have or have had at least one chibwenzi or partner in their lifetime. Men have more lifetime partners (2.6) than women (1.3), but both sexes become sexually active at similar ages (~15 years). Nearly all respondents claiming virginity status were 15 or 16 years of age.

Premarital intimate relationships are largely among peers who are members of the same or nearby communities. Young Malawians meet each other in schools, in nearby villages, at their churches or mosques, or at soccer games. Statements such as, “I was in Standard 5 and he was in Standard 6”, or “I met her at the football field” are common responses to questions about where they first saw or met their girlfriend or boyfriend. To acquire a partner, the approach, in rural Balaka anyway, typically mandates that a boy funsira (“propose”) a girl to be his chibwenzi. This is done by writing a letter and having it delivered by some intermediary, such as a friend or a brother, stating that the young man thinks his hoped-for girlfriend is beautiful, for example. Alternatively, the proposal can be verbal, but must also pass through a “go-between”, again almost always a friend or relative. The proposed-to young woman is then required to either accept or reject the proposal. If she accepts, she then has a chibwenzi, and the couple will engage in intercourse within a matter of days, or within a week at the most. It is not acceptable for young women to propose to men, although on occasion women may make their interest known, such as by coyly flirting with their eyes. A young woman can, and very often does, refuse proposals, but in Balaka, her acceptance is an acceptance of sex.

The interviews illustrate that the choices of whom to propose to and from whom to accept proposals are far from arbitrary. Unmarried women accept proposals from men for a variety of reasons. For some, a man’s popularity, his physical attractiveness, or his level-headedness (when he is considered “cool”) are decisive factors. For others, men known not to “move around” (to have a lot of partners), or who can provide money or gifts, are esteemed partners. Some women attribute their acceptance of proposals to their friends’ wishes; in the data, women expressed feeling pressure by friends to acquire boyfriends, or they themselves willingly emulated “best friends”, who themselves had boyfriends. Affirmative responses were given when suitors were persistent, a behavior young women took to indicate love and the seriousness of a suitor’s intentions. Among the boys and men interviewed, important characteristics sought after in a chibwenzi varied, but included physical attractiveness—such as a beautiful face or a “good body structure”—and “good behavior”. On the latter, definitions differed by respondent, but could mean an unwavering willingness to have sex, an adeptness in cooking and cleaning, or a respectful demeanor toward elders.

Before accepting or proposing, young women and men assess the suitability of potential partners through locally innovative strategies, such as inquiring about the sexual histories of possible mates from friends, relatives and neighbors. Young people are motivated to have zibwenzis for a variety of reasons, including sexual pleasure, and to assess whether the relationship will lead to a marital union. For instance, a 21-year old male respondent with two girlfriends considered having two important to determine which would make a better wife. In another example, 21-year old Abdullah, after two years of dating, now sees his partner as marriageable:

R: My current girlfriend … has been my girlfriend for two years now. I intend to marry her. I: In the initial proposal, did you include the wish to marry her? R: No, I did not include that. I am [saying] that I will marry her because in the two years that she has been my girlfriend, I have seen that she is a good girl, she respects. (Abdullah, age 21)

Unmarried Malawians exhibit varying degrees of emotional investment or love in intimate partnerships, which may depend on a host of factors, including the perceived loyalty of the partner. Relationships may end if one partner is “moving around”. While elders are involved in the marital strategies of young people—with, for example, the tradition of mediation by a marriage counselor (ankhowse) still intact—youth expect to choose their own spouses; over 97% of never-married women and men in the survey sample expect to select their future partner (see also Haram, 1995).

It is in this cultural context of courting rituals that intimate relationships are intertwined with money. As will be shown, the rules and norms surrounding exchange are intricate and even ambiguous. Yet transfers do have unequivocal connotations. One persistency is this: a sexual relationship can not exist without a male to female transfer of money or gift. Restated, without a transfer, there can be no chibwenzi. In the interviews, all women who have or have had a boyfriend received money from that boyfriend at some point during their partnership. And all men who have or have had a girlfriend gave money to that girlfriend at some point. The first transfer of money and corresponding sexual act marks the beginning of the relationship, and, although over time the amount transferred may change, continued transfers sustain it. Money transfers take place in a range of partnership types, even in more emotionally-committed ones that span some time. In the 2004 questionnaire, the MDICP asked all sexually active, never-married women about the type of their current or penultimate partner, and whether the last time they had sex with that partner they received any money or gifts. Of the ever-sexually active women reporting their current or last partner to be “casual” (28%), 61% reported that money/gifts had been exchanged at last sex. Of those reporting their current or last partner to be a “steady boyfriend” (43%), 55% said they had received money/gifts, and of those reporting a chitomelo or fiancée (29%), 59% received money or gifts at last sex.

The Semiotics of Money Exchange

The semiotics of money are reflected both publicly and privately. Publicly, that transfers mean sex will occur is apparent. Men expect that when giving money, they will be having sex, and they expect to have it sooner rather than later. Women also expect to be “giving” sex, but clearly enjoy it, stating that it is “sweet”. For girls, getting money means material benefits; with money, things both needed and wanted are bought, like sugar and soap. Also clear is the responsibility of boyfriends to provide money. One 17-year-old female respondent was asked why her boyfriend bought her soap, rather than her parents, and her reply was simply that it was her boyfriend’s “duty”. Men’s rhetoric reflects similar expectations; boys see it necessary to provide material benefits for girlfriends. Should they fail, female partners have grounds for ending relationships, as the following illustrates:

I: Why did you not tell your girlfriend to ask her parents for soap and lotion? R: Oh, no! I can not do that. The moment one does that then a girl would end a relationship. A boy that does not have money to give his girlfriend risks losing her. (Raffiq, age 19)

Beyond these distinctions, money’s meaning is blurred, with varied understandings conditioned by individual, couple, or circumstance.

In the private realm, money is given meaning by the relationship itself, and creates ties between couples. Money is used to acquire social status, with both sexes benefiting from female partners’ improved social standing. Women use money to buy luxury items, such as body lotions or cookies, bringing them admiration from their peers. Boys encourage their girlfriends to purchase such items, proud to bolster the position of their girlfriends, as 19-year-old Christopher describes:

I was giving her K20 or K30 (20 to 30 cents) so that she could buy a bottle of Fanta at school. But when there was a football match, I could give her K50. There was a need that she gets more money [at] a match, so that she can stand out among her friends, as one who is given a lot of money from her boyfriend. (Christopher, age 19)

In this normative context, money’s meanings invoke a set of rules underlying the structure of contemporary premarital relationships. Where money and gifts are concerned, rather strict guidelines exist for how sexually active young people should behave. In Balaka, despite some uncertainties—such as how much should be given or when it should be given—definitive expectations exist that money transfers must occur in some form or another. As one 15-year-old female respondent poignantly said, “Gifts never fail to exist in a relationship”. These expectations are powerful, and stipulate that men who do not give money send clear messages to their partners: would-be recipients are not really valued.

It’s as Caplow (1984) finds in his study of Christmas gift-giving in Middletown, USA. Each Christmas, a Middletown resident was expected to give a gift of an equal dollar amount to each of his two sisters-in-law (for instance), otherwise the giver communicates to the in-law (and everyone else in the family) receiving the gift of lesser-value that she is not as important or as well-liked or as loved as the sister-in-law who received the higher-valued gift. Swidler (2001) makes a similar argument. Not calling one’s mother on Mother’s Day, or not sending one’s sweetheart a gift on Valentine’s Day, communicates to the rebuffed, non-recipients a message such as, “you are not important enough to remember”. Likewise, in Balaka, were a young woman not to receive money from her partner, she would be insulted, but worse, she would be viewed as ridiculous to her peers (see also Silberschmidt & Rasch, 2000). She would be laughed at. For young men, providing money is an opportunity for heroism and is deeply bound up with identities around masculinity (Ashforth, 1999; Mills & Ssewakiryanga, 2005).

Thus in southern Malawi, premarital money transfers need not signal exploitative or immoral relationships (see also Parry & Bloch, 1989). Transfers are bound with socio-cultural expectations of transfers between kin, or with obligations of husbands to provide for wives (see also Weinreb, 2002). Interestingly, Cornwall’s (2002) work among the Yoruba in southwestern Nigeria shows how wives’ increased earning potential has not altered social expectations that men spend on their wives, noting if a man really loves a woman, he will spend on her rather than use his money elsewhere; it is when he does not love his wife that he gives her no money (p. 977). Cornwall notes that money can buy love, a similar argument made by Zelizer (2005).

In the in-depth interviews, young Malawian adults carefully describe how they regulate their own relationships. Although when men give money they expect sex in the near future, young people often emphasize a separation between the monetary and the sexual. Money can come, in the words of some, “just anytime”:

I: Why did he give you money? R: Because I was his girlfriend.

I: Was that money after sex or not? R: Sometimes just after sex [but] sometimes just anytime. (Patuma, age 18)

“Just anytime” refers to the timing of the transfer; while money is often given either just prior or just after an act of sex (as was shown with the survey data), it is also given when sex does not occur at all. Alternatively, couples will have sex but without gift-giving. Boys and girls use the ‘just anytime’ rhetoric to stress the significance of the relationship, as if to say, “See, this is how much she means to me, I give her money just anytime”, as Abdul illustrates:

For the whole period of our friendship Lydia never asked me for money or gifts, but I gave them to her to show that I trusted her. I gave the gifts anytime, not just after sex only. (Abdul, age 21)

The timing of the transfer is a function of the act of sex, and of trust and love. It is also a function of the availability of the cash, as told by one interviewer about James:

He said he was giving her soap or money but not at regular intervals. Sometimes he could give her money, and after a long period, maybe two or three weeks later, he would [give] her soap. Sometimes if he had a lot of money [he] could give it to her right away, or buy her lotion. (James, age 19)

In addition, young women and men point out when money is not asked for but given voluntarily, as 17-year-old Annie tells it:

He gives these gifts everywhere we meet. I remember one day I got a gift from him when I met with him at a certain grocery. At that time, I went to buy soap while he was just chatting with his friends, and he gave me this gift without me begging him. (Annie, age 17)

Seventeen-year-old Mariam is pleased by her boyfriend’s gift-giving approach:

The gifts are always given, whether or not we have sex. And sometimes I had sex with him but he gave me nothing. That’s why I say that the gifts are given voluntarily. (Mariam, age 17)

Typical constructions of male-female relationships dominant in public health and other disciplines addressing the AIDS pandemic hold that male-to-female money exchange in sexual partnerships has to be something like prostitution (see Chernoff, 2003; Wardlow, 2004). Or, that sexual exchange is coercive insofar as women are poor and have limited opportunities; in this case, money changing hands among sexual partners could be “nothing but” female exploitation (Zelizer, 2005). Yet, carefully articulated narratives about the voluntary quality of the gift and the timing of the transfer signify that premarital partnerships in Balaka are often qualitatively different from those where sex is explicitly transactional. For many female respondents, the subjective meanings of their relationships are tied to avoiding identification with sex workers, hence the “just anytime” and “I didn’t beg for it” idiom. Perhaps as part of wider regional trends where intimacy within premarital partnerships resembles modern notions of romantic love, women yearn to be the object of such love (and desire) from their male partners, and the moral connotations associated with sex work in Balaka threaten the potential to realize these opportunities (see also Hunter, 2002; Nnko & Pool, 1997; Smith, 2004). Men and women made references to love repeatedly, as 16-year old Grace conveys:

The boy gave me money ranging from K100 to K200. I used that money to buy mandasi (doughnuts) at school… He did it according to his will because he loved me a lot. (Grace, age 16)

Men will frequently verbalize their love or affection for female partners (or hoped-for partners), but women may not trust such words when used alone. When they are accompanied by money transfers, however, words of love are reinforced and given weight. In these premarital relationships, then, money is a gift, and, as a gift in this context, money is imbued with emotion. Transfers express and signify that emotion, which may be of affection, of lust, or of feeling possessive. The exchange itself may initiate partnerships, and symbolize love, but the amount given may signal the extent of a male partner’s interest. Money and love need not be situated antithetically, as 18-year-old Saiti claims: “To show her I was the man of real love, I gave her almost K50 every week”. The saliency of the semiotics of money is revealed not when the amount given is deemed high in the context of a Balaka standard of living per se, but rather, when the money is hard earned or nearly all one can give. In these cases, money exchange indicates the existence of “real love”, as 22-year-old Many describes:

R: We started our companionship in September, and in the first days we were in true love. I gave her money almost twice a week. I worked hard in the field to have money, not less than 30 kwacha at once [to give her]. In the first days I gave her no gift but after a month, she asked for an under wear. I tried my best and bought it with almost 100 kwacha. (Many, age 22)

Exercising Sexual Agency

In this qualitative sample, girls and boys enter into partnerships with their peers, who are very close in age. Partner’s age differences are not large, with the average female respondent slightly less than one year younger than her male partner (1.2 years). Men report being on average a little less than one year older (0.9 years). The age differences in the survey data are larger, but remain under five years; never-married women are on average 2.9 years younger than their partners and men are on average 2.4 years older. The male partners of the women in this sample are young and typically have limited economic opportunities. Meeting their girlfriends’ financial expectations is challenging:

I: How often and what amount do you give to your girlfriend? R: I can give her 70 kwacha, or sometimes 100 kwacha, when I have enough money. Sometimes I give her money after two weeks or after one week, but sometimes it takes me a month to give her money. (Lia, age 19)

Finding work, such as ganyu labor (farming labor), is a fortunate job to get. A woven straw mat takes three days to create, and sells for 60 cents, not nearly enough, the men protest, for the expended effort. As 19 year old Mbtamba explains:

I: What are your plans, and what challenges do you face daily? R: I experience lack of money. It is too difficult to find means of getting money. I weave mats and work in people’s fields. It’s too difficult because it takes a long time to get money. (Mbtamba, age 19)

In a context where money transfers are normative, men’s limited earning power increases women’s bargaining power. A man give his girlfriend much of his earnings or risk losing her. If a girl’s chibwenzi isn’t measuring up financially, she may go elsewhere, as described by 24-year-old John:

I: Can it really happen that whenever boyfriends are not helping (rather giving) girlfriends money the relationship can end? R: Yes. I: So you were giving her with the same reason? R: Yes, it’s like buying her, so that she should not look nor accept another one. Because ladies rather girls keep on being proposed. (John, 24 years)

Some men consider themselves lucky to find the “good ones”, those who aren’t the “money lovers out there” (see also Mills & Ssewakiryanga, 2005). Men view women as calculating, describing women as motivated to have sexual partners in hopes of finding a spouse, because they enjoy sex, but also to glean money from men. In addition, unmarried women self-direct action based on their evaluations of their partner’s or potential partner’s risk of infection, as they are aware of the dangers of AIDS. Young women know how to protect themselves—even if they use local methods to modify the formally espoused “ABCs” of prevention (Watkins, 2004). Yet condom sales in Malawi are rising; in 1994, just under 1 million condoms were sold throughout the country, while in 2001 6 million were sold, and 9 million in 2005 (Population Services International [PSI], 2005). The increased availability appears to be making inroads to increased use. Many respondents had not used condoms in years past, when they “didn’t know about those things”. In contrast, in 2004, while the majority does not use condoms regularly, among the sexually-active, use appears to be increasing. In the MDICP survey sample, 44% of young women and 36% of men reported having ever used a condom, similar to findings from other survey data in Malawi (Munthali et al., 2006).

Reasons for non-use by women vary, including their boyfriends’ disdain for them(kufuna kusagwiritsa ntchito), as one interviewer describes:

They were not using condoms because the boy was not happy to use condoms, saying that (sweet wa mpepala sakoma), meaning that you can not eat sweet while it is inside the packet. (Agness, age 20)

While it may be the case that non-use is due to social norms restricting women’s ability to request condoms, other research shows non-use is more about ideals of procreation and intimacy within partnerships than it is about constraints on agency (Chimbiri, forthcoming; Smith 2004). In qualitative interviews, on occasion women report being insulted when their partners wish to use condoms, as recounted by this man:

I: Why did you stop using condoms? R: At first, I used to take two condoms out of their wrappers and [place] them on the bedside in readiness of the sexual act. Then I would go to get her. When she came into the house, I would turn off the light before she sees the condoms. I wore the condoms in the dark and she never realized I was using a condom. Then, one night I forgot to take the condoms out of the wrapper. So I tried to take them out when she was already in the bedroom. Although it was dark, she heard the sound of the wrapper and immediately lit a match. She got upset upon seeing the condoms and asked me: “Why are you doing this? You think I have [sexually transmitted] infections?” I did not know what to tell her, I was not going to use the condoms. From that time, I have never used a condom again. (Yamikani, age 22)

When sexually active, unmarried women do acknowledge condom use, they cite both pregnancy and disease prevention as primary reasons, claiming for instance, they “cannot know what their chibwenzi is doing”. In this sample, condom use and non-use appear unrelated to the amount of money exchanged, at least in a direct sense. This finding contrasts with Luke’s (2005) work in Kenya, where she finds a significant negative relationship between money and condom use. The in-depth interviews do not allow a similar testing between money and condom use, however. Despite this, narratives here suggest that more money and/or better gifts are given by men to women who are the most loved or are expected to marry, as one interviewer tells it:

The third boyfriend was providing soap, lotion, and money for almost K100 but in other days he offered K200 and in other days he was buying clothes like a skirt or a dress. He was doing this because he promised to marry her, among these three guys, the third guy was the one who provided more gifts than the other two. (Annie, age 22)

A dominant discourse shows that female power in the sexual realm involves more than an ability to negotiate safe sex; having power is about more than using condoms. Adolescent girls and young women have considerable decision-making control over the process of relationship formation and termination, and they use it to their advantage. They can and do choose which partners they want, and can and do choose to end partnerships; these decisions are of critical importance in determining what one does with one’s sexual body, which has profound implications for risk of infection. Girls usually assess the attributes of potential partners in some combined form: Attractiveness, peer group behavior, and money are some of women’s considerations when thinking about whether and whom to have as zibwenzis. In addition, many young women choose not to engage in sexual relationships at all, frequently claiming being too young, that they needed to finish school, as having a chibwenzi would, “disturb their education” (Poulin 2006). Many young women prefer to delay sexual activity, with fears of pregnancy and contracting HIV two frequently cited reasons for abstaining:

R: I can have a boyfriend [maybe] he will tell me to have sex while he is not good. I: What do you mean “not good”? R: Maybe [he] will have the disease, so after having sex with him he will transmit it to me … … I: Are you still proposed by boys? R: Up to this time boys still propose me but I always refuse them. (Rose, age 16)

Likewise, women reduce infection risk by selecting partners with care. Informal chatting about sexual partners is common throughout the villages in Balaka, and if a girl’s chibwenzi is rumored in her village to be “moving around”, it is reasonable to her and others that she end that partnership:

I: Is your relationship still continuing? R: No it ended … I: Who ended the relationship? R: I was the one who ended it. I: Why? R: Because the boy was moving around with other girls. I: How did you end the relationship? R: I wrote a letter and the boy did not respond. That was the end of the relationship. (Biba, age 18)

Women’s ability to select or get rid of partners on the basis of perceived risk is recognized as a legitimate mode of action. In addition, because money transfers over the duration of the zibwenzi-relationships were expected, girls who received less than they thought were within their rights to end partnerships; women frequently exploited the man’s financial constraints, using “not enough money” as justification for finding new partners, despite protests from boys that their transfers had not decreased in value. Women perceive these men as less than adequate providers, and not ideal marriage partners (see also Cole, 2004).

Yet, interestingly and ironically, young women who reported being “in love” with a “boyfriend” did not always expect the same value of money or gifts as those who did not refer to being in love. Herein lies an ambiguity: more love may mean more money, as was shown in the above example with Annie—but perhaps with love a girl does not expect as much, where a male respondent says, “Love is love. To know that she loves you, she should not ask for something to give her.”

Conclusions

In the rural south of Malawi, the link between sex and money is nuanced and normative. The in-depth interviews show exchange practices are driven by women’s need for economic resources; they are poor and appreciate the economic value they gain from having sexual relationships. Yet the dyadic interaction of the transfer means more. Girls interpret receipt of money as loving, and boys intend it to be interpreted in the same way. Money exchange is a binding obligation that marks zibwenzi partnerships, where it takes place in all premarital relationships, whether casual and short-lasting or more enduring. These relationships are held amongst peers and are important for acquiring social status, for finding suitable marriage partners, and for sexual companionship. In this context, money takes on semiotic meaning, where money transfers, as gifts, indicate the woman’s value. A sudden absence of or a decline in the relative amount given may show her lack of worth to the man, and, in all likelihood, would be the end of the relationship.

In thinking about the complexities of sexual exchange, it becomes possible to see how girls have decision-making power over their sexual lives. Women can request—or refuse—condom use, to act on sexual desire, and to determine whom to have or no longer have as partners. Narratives about money being given “just anytime” and about girls not “begging” for it indicate material exchange does not preclude women’s ability to act in risk-reducing ways. For men, pride is derived from providing for women. For women, receiving money means maintaining dignity and avoiding ridicule. For both, money marks commitment, love, and movement toward marriage. Unmarried women are sexually agentic, at times using control to reduce risk, albeit in unexpected and not straightforward ways. These findings add a layer of complexity to typical notions associated with ‘transactional sex’, and may assist us in understanding the intersections between gender, power, and sexuality that contribute to the transmission of HIV.