Reconsidering the orphan problem: the emergence of male caregivers in Lesotho

ABSTRACT Care for AIDS orphans in southern Africa is frequently characterized as a “crisis”, where kin-based networks of care are thought to be on the edge of collapse. Yet these care networks, though strained by AIDS, are still the primary mechanisms for orphan care, in large part because of the essential role grandmothers play in responding to the needs of orphans. Ongoing demographic shifts as a result of HIV/AIDS and an increasingly feminized labor market continue to disrupt and alter networks of care for orphans and vulnerable children. This paper examines the emergence of a small but growing number of male caregivers who are responding to the needs of the extended family. While these men are still few in number, the strength of gendered ideologies of female care means that this group of men is socially, if not statistically significant. Men continue to be considered caregivers of last resort, but their care will close a small but growing gap that threatens to undermine kin-based networks of care in Lesotho and across the region. The adaptation of gender roles reinforces the strength and resilience of kinship networks even when working against deeply entrenched ideas about gendered division of domestic labor.


Introduction
When I first met him, Ntate 1 Thabiso was in his mid-40s, though his illness conveyed a frailness beyond his years. He lives in a rural mountain village in the highlands of Lesotho and is the father of seven children, ranging in age from 5 to 25. Until last year, he lived with his wife and five of his children. A year ago, his wife died and he was left with his four youngest children. Shortly after her death, the family decided to send the youngest child, a 5-year-old girl, to live with her maternal aunt in town. But the four boys were left with Ntate Thabiso because their grandmothers were dead. The youngest boy was 10-years-old at the time.
When I asked Ntate Thabiso about his new living situation, he told me, "It is the first time I have done it … It was not difficult." Ntate Thabiso had always provided for his children and worked hard to support them; but, until his wife died, he had not done the daily care work of feeding, washing, clothing, and nurturing them. Fortunately, his sons were older, so were able to do much for themselves, and were helpful around the house, particularly when Ntate Thabiso's chronic foot problems were aggravated. Initially, Ntate Thabiso wanted his younger sister to help him, but he said, I wanted to find someone, so I could go to find a job … But, because of my sickness, kanete [I swear], I was not able to … Because I was sick I felt that I needed to help her [by caring for the children] because I am the person who is not living well.
Ntate Thabiso decided to forego the domestic help in order to allow his able-bodied sister to work. Jobs are scarce in Lesotho, and the cultural preference for female care was trumped by the family's economic needs. Ntate Thabiso explained that if his mother were alive, she would be caring for the children. When I asked him to elaborate, he simply stated: "Because she is my mother." This apparently self-explanatory elucidation denotes the strength of the cultural preference for grandmaternal care, which has in many ways cushioned the impact of AIDS-related orphaning in Africa, and extended the family's ability to maintain kin-based networks of care.
Ntate Thabiso's situation conveys what many other Basotho families have indicated: male caregivers are a last resort. In Lesotho, as in many other contexts in southern Africa, caregiving is widely considered a female task (Upton, 2003). Grandmothers are generally preferred over younger female caregivers as they are typically freed of both mothering and labor responsibilities. Due to high rates of AIDS orphaning, grandmothers are increasingly relied upon because they have not been decimated by AIDS to the same extent as the socalled "missing generation" of sexually active adults (Geiselhart, Gwebu, & Krüger, 2008). Conversely, southern African ideals of masculinity posit that men are patriarchs and providers, but do not engage in daily caregiving (Hunter, 2006;Lesejane, 2006;Mkhize, 2006;Ogden, Esim, & Grown, 2006). In Lesotho, men's ability to fulfill these traditional social roles and obligations has been diminished by a decline in migrant labor opportunities (Spiegel, 1981). Concurrently, labor migration among Basotho women has increased because labor force demands have swung away from male mine work and toward female domestic and factory work (Coplan, 2001;Crush, 2010;Turkon, Himmelgreen, Romero-Daza, & Noble, 2009). While changing economic and social conditions have undermined men's ability to provide for their families, firmly entrenched ideologies about the gendered division of labor that exclude men from the labor of care work still hold firm (Shefer et al., 2008). In a social context where males are viewed as caregivers of last resort, why are men like Ntate Thabiso increasingly performing this vital domestic task?
The answer to this question lies in a social shift precipitated by a population problem. AIDS has devastated communities across southern Africa, including Lesotho, where a quarter of the adult population is infected (UNAIDS, 2010). Elevated adult AIDS mortality has depleted a cohort of caregivers leaving behind a large number of orphans and vulnerable children. In the early years of the AIDS epidemic in southern Africa, studies warned that the steadily growing population of AIDS orphans would bring about a collapse of existing fostering networks (Foster, 2000). Yet, over time, researchers noted that these "apocalyptic predictions" (Bray, 2003) did not come to pass. Instead, kin-based networks, though strained by AIDS, are still the primary mechanisms for orphan care (Adato, Kadiyala, Roopnaraine, Biermayr-Jenzano, & Norman, 2005;Ansell & van Blerk, 2004;Nyambedha, Wandibba, & Aagaard-Hansen, 2003a;Prazak, 2012;Zagheni, 2011). However, much of this burden of care has been left to grandmothers (Ansell & van Blerk, 2004;Goldberg & Short, 2012).
The availability of grandmothers and the resilience of kinship networks are the primary reasons that kin-based networks of orphan care did not collapse over the past two decades when AIDS mortality and orphaning were at their peak (Ankrah, 1993;Nyambedha, Wandibba, & Aagaard-Hansen, 2003b). However, the varied demographic impacts of HIV/AIDS in southern Africa continue to shift. While AIDS mortality has mercifully decreased due to the widespread availability of antiretroviral therapy, HIV-prevalence remains high, as do new infections (UNAIDS, 2014). Grandmothers, who were part of the last generation of HIV-free southern Africans, are aging and dying. At the same time, the migrant labor market has become increasingly feminized. As mothers are living longer, they are finding much-needed wage labor in the lowlands and in surrounding South Africa (Crush, 2010;Turkon, 2009). In the absence of a viable female caregiver, male caregivers are responding to the needs of the extended family. These menmostly fathers and grandfathers, but also uncles and brothersare few in number. However, the strength of gendered ideologies of female care means that this group of men is socially, if not statistically significant. These male caregivers are, of course, impacted by the same population pressures as women of their age group; younger cohorts of fathers and uncles have been depleted by AIDS mortality and elderly HIV-free grandfathers are dying of old age. However, opening the field of potential caregivers to men of all ages increases the possibility for kin-based care within each family. Men will continue to be considered caregivers of last resort, but their care will close a small but growing gap that threatens to undermine kin-based networks of care in Lesotho and across the region.
There are two basic principles operating beneath the surface of this trend towards male care that reinforce the strength and importance of kin ties in Lesotho. First is a preference for female care, underscored by both discourse and practice. Second is the prominence and resilience of kinship networks. Given the strength of the preference for female caregivers, one might expect the preference for female care to trump the importance of kin-based networks of care. Yet, the ethnographic data presented here demonstrate that this is not the case. Instead, the preference for female care is encompassed within the even stronger penumbra of kinship care, and the male caregivers presented here are evidence of the ever strong pull of kinship.
Feminist scholars have long refuted naturalistic assumptions about both gender and kinship, rooted in reproduction, which previously explained gendered divisions of labor (Franklin & McKinnon, 2001). However, the work of Collier and Yanagisako, which calls us to consider the ways in which we continue to "reinvent gendered analytic dichotomies" (1987, p. 49), including those around domestic labor, is still relevant. In contemporary Lesotho, gender is a flexible cultural resource that can be deployed to meet the social desires and obligations of families, even within the context of rigid gender ideologies. Clearly, family based caregiving is both desirable and possible. If we overemphasize men as providers, as Leinaweaver notes, we risk reproducing this trope instead of being able to "identify sites at which it might be contestable" (2014, p. 100). Following Collier and Yanagisako (1987) I emphasize the ways in which feminized discourse used by both men and women in Lesotho masks practices of care. I situate these continuities and changes in the political economic and historical circumstances of contemporary Lesotho.

Men as caregivers
In Sesotho, to care about (tsotella) someone means you have to care for (tlhokomelo) them. When you ask Basotho about what makes a good caregiver, they do not discuss the affective dimensions of care, but rather focus on the physical tasks of caregiving. Likewise, when discussing a poor caregiver, Basotho typically point to the ways in which they have failed to perform the labor of care required of them. For example, the quality of a caregiver is often discussed in terms of the cleanliness of the child. Until recently, men's caregiving activities for children have been restricted to their role as providers, and have not included the labor of care. Yet it is the daily labor of care that is most at stake as a result of AIDS orphaning. In this paper, care does not refer to the ways men provide for children's material needs. Instead, this work focuses on men who engage in the daily labor of care for children in a caregiving climate where this type labor has not, until recently, been performed by men, yet where it is increasingly important.
While men are frequently considered the head of the household, their realms of responsibility typically exist outside the space of the home (Shefer et al., 2008). Characterizations of masculinity from precolonial times portrayed men as patriarchs, protectors, and providers; but the gendered division of labor meant that the daily care work for children, the sick and the elderly were the exclusive responsibility of women (Hunter, 2006;Lesejane, 2006). The social and economic conditions of the postcolonial period in southern Africa, including the rise and fall of migrant labor economies as well as changes in marriage practices, have served to physically distance fathers from their children while simultaneously decreasing their economic ability to fulfil the role of provider (Boehm, 2006;Desmond & Desmond, 2006;Hosegood, 2009;Hosegood, McGrath, & Moultrie, 2009;Mokomane, 2013;Murray, 1980). For much of the twentieth century, migrant mine work meant that fathers were able to make a significant economic contribution to their households through remittances (Kimble, 1982;Murray, 1977;Romero-Daza & Himmelgreen, 1998). In Lesotho in the 1970s, remittances made up 70% of a rural household's income (Spiegel, 1981). However, a massive decline in the demand for male mine workers and increasing opportunities for formal employment by women further undermined the importance of male labor at the same time as female contributions to the economic wellbeing of the household increased (Coplan, 2001;Crush, 2010;Turkon et al., 2009). The intimacy and bond that men may have had with their children in the past is further diminished by postcolonial social economic conditions that reduce the likelihood that men will cohabitate with their children, thus distancing African fathers from the affective role they might have in their children's lives (Hunter, 2006;Posel & Devey, 2006).
Given the historical and contemporary circumstances that limit men's roles as caregivers, it will come as no surprise that women have performed the majority of daily care work in southern Africa (Akintola, 2006;Desmond & Desmond, 2006). Thus far, the limited research that has been done on male care work in the context of HIV/AIDS has concentrated on both formal and informal care for AIDS patients, not care for orphans (Akintola, 2006;Dworzanowski-Venter, 2008;Kipp, Tindyebwa, Rubaale, Karamagi, & Bajenja, 2007). Additionally, research on orphan care, such as two large regional studies using Demographic and Health Survey (DHS) data, note the presence of a small population of children living with their fathers (but not their mothers) in sub-Saharan Africa (McDaniel & Zulu, 1996;Monasch & Boerma, 2004). The focus on co-residence at a population level gives no insight into the daily caregiving tasks that these men perform as it does not indicate what other relatives are living in the household, and therefore does not add to our understanding of the extent to which men might act as a caregiving safety net for orphans. While some studies indicate a greater acceptance of men performing women's work in the domestic realm (Sideris, 2004;Smith, 2014), men are still viewed as the last resort as primary caregivers for orphans or sick family members (Dworzanowski-Venter, 2008;Kipp et al., 2007;Nnko, Chiduo, Wilson, Msuya, & Mwaluko, 2000).
Women in southern Africa are burdened with the brunt of caregiving responsibilities, which are greatly amplified by AIDS illness and orphaning; yet, both men and women perpetuate the feminization of care work that assumes men's exemption from doing such tasks as cooking, washing, and caring for a child (Desmond & Desmond, 2006;Dworzanowski-Venter, 2008). In a South African study of caregiver expectations, fathers reported a much higher degree of openness to caring for their children compared to mothers, who had very low expectations that their husbands would provide care for their children if they died (Freeman & Nkomo, 2006). While fathers were expressing only a theoretical willingness to care, this indicates both women's role in perpetuating gendered divisions of labor, as well as the possibility that men might be open to performing this highly feminized domestic work.

Data and methods
This analysis is based on over two years of ethnographic and archival research conducted between 2009 and 2015. The research took place in the rural district of Mokhotlong, in the highlands of Lesotho. I worked closely with a small, locally run NGO called Mokhotlong Children's Services (MCS) that provides temporary residential and outreach services to orphaned and vulnerable children, and much of this research is facilitated by that relationship.
In-depth semi-structured interviews were conducted in 2013 with 37 caregivers living with one or more orphans, including 8 male caregivers, as part of a larger study about the impact of grandmother death on orphan care. While some of the interviews were conducted with caregivers I had known for many years, others were newly recruited by word of mouth and by referral from MCS. I employed a reverse recruitment strategy by looking for families where there were maternal or double orphans living with any relative other than a grandmother, based on the assumption that such an arrangement would be likely to lead me to families without grandmothers. This recruitment strategy proved to be successful, as almost all of the 37 families had no living grandmother who could care for the children.
This research also utilizes archival data from MCS' client database, which records health and household information. Currently, the database contains longitudinal data on all children who have received MCS services since December 2006. Data were analyzed using both qualitative (ATLAS.ti) and descriptive statistical (EpiInfo) software. The sample for this research was a criterion-based purposive sample (Patton, 1990), whereby participants were chosen based on meeting a specific criterionin this case, they were all involved in the care or placement of orphaned children, with a focus on the absence of grandmothers.

Results
In this section, I will first examine the caregiving characteristics of the study population. Then, I will use ethnographic data to explore the ways in which male caregivers are helping to fill a caregiving gap in available caregivers, with a focus on two elaborated case studies. The strength of the kinship care system in Lesotho is evident in that only two children in MCS' client records were listed as under the care of non-kin, and both of these children were temporarily being cared for by hired help while their mothers were away for school and work, respectively. The exchange of funds for this service reinforces the perception that non-kin have no social obligation to provide care for children.
The preference for grandmother care continues to dominate the rural landscape in Lesotho. Of 327 orphans in MCS' client records not living with one or both parents, 80% were living with a grandmother (see Table 1). The family profiles of orphans living without grandmothers illustrate the strength of the preference for grandmaternal care. I interviewed 23 families where orphans were not residing with a grandmother, and 70% of these families (n = 16) had no living grandmothers. The remaining seven families had only one grandmother living, and in all cases, the grandmother was not mentally or physically able to care for the child, or was geographically prevented from doing so. Two of these living grandmothers worked in South Africa, three were already caring for a number of other orphans, and two were viewed by the extended family as inadequate caregivers. In short, I did not find a single case where orphans were living with another relative when there was a living, capable, and available grandmother.

Alternative configurations of care
Comprehensive, large-scale quantitative studies of orphan care note that fathers are still mostly absent from daily care work (Govender, Penning, George, & Quinlan, 2012). Lesotho's DHS data from 2009, which examined children's residence patterns, found that among children living with a single parent, 33% were living with their mother, while only 5% were living with their father (Lesotho Ministry of Health and Social Welfare, 2010). MCS' client records indicate that only 2% of fathers were listed as the primary caregiver as compared to 68% of mothers (see Table 2). Among the 327 children not living with either parent, only 4.3% were living with a male relative, the majority of whom were grandfathers (see Table 1). The population of male caregivers in the database sample is small (37 of 354 observations, excluding mothers). Also, there is little data with which to compare the frequency of men who provided daily care for children prior to significant historical social disruptions such as migrant labor and HIV. However, the strength of the preference for female caregivers, the testimonies of elderly Basotho, and the experiences of the male caregivers in this study, indicate that male care is particularly unusual in this social context, and that this small population of male caregivers deserves serious attention as a potential safety net for orphans and as a significant shift at least in gendered practice if not ideology.
The strong preference for females to perform the labor of child care is pervasive. As one elderly grandmother told me: The mothers are the ones taking care of the babies. There are few fathers who can … If the orphans can stay only with the father, you would not find them clean. They would always be dirty. But with the mothers, they will always be clean.
If a man is the head of a household, or is left in charge of a child, that usually means he is responsible for finding a woman to care for the childif possible his own wife. It is highly unusual for a man to do the daily care work of bathing, feeding, clothing, and nurturing a child. Occasionally, an older brother might help out with his younger siblings by watching them while his mother or sister works. But, while girls help with cooking and washing from a young age, boys rarely do such tasks. A preference for female care is particularly strong for very young children who require intensive support and are not able to contribute to household work. Yet I know of several men who were left with young babies and toddlers. This type of care work, even when performed by a man, is viewed as maternal. For example, 4-year-old Kotsi asked his grandfather, Ntate Kapo: "Are you my mother?" I inquired as to why Kotsi might have wondered this, and Ntate Kapo told me: It's because he has seen that I am taking care of him, and I'm taking good care, and I'm cooking, and I'm giving him food, and I wash clothes for him, I'm doing everything with him, and I sleep with him in the bed.
Gendered ideas about care are so powerful that in 4-yearold Kotsi's mind the labor of care is equated with motherhood.
The feminization of care work aligns with the importance of social roles over biological ones in southern Africa (Lesthaeghe, 1989). When describing care work or other housework, both men and women drew attention to instances when they were engaged in work that was typically performed by the opposite gender, regardless of the frequency with which these "anomalies" occurred. Male caregivers often discussed their caregiving roles as highly feminized. For example, one grandfather told his grandson, who he had cared for since infancy, "Yes, my son. Your mother left you while you were young. So, I became your mother." In many cases, men were considered unsuitable for the task, and both men and women believed that a child would be disadvantaged by such a caregiving arrangement. For example, one grandfather helped around the house with tasks such as chopping wood, but was not able to do much else because, as one of his in-laws noted, he was "a male person". In several families, potential male caregivers were discounted because they were unmarried, regardless of the caregiving burden or health of the chosen caregiver. One caregiving aunt said of her brother: "He refused to take the child because he doesn't have a wife." These caregivers did not feel the need to elaborate on why being "a male person" or "not having a wife" disqualified the person as a caregiver because it was viewed as self-evident.
Care by women, particularly grandmothers, was viewed as eminently natural. Just as the inadequacy of male care needed no elaboration other than to comment on the "maleness" of the potential caregiver, grandmothers were frequently described as being suitable for care simply because of their age and gender. When I asked why grandmothers were preferred as caregivers, I was told by one woman who had been raised by her grandmother that she was a suitable choice simply because, "she was the mother of my mother". On numerous occasions I heard that grandmothers "have love" [kena la lerato]. Care by elderly kin is also strategic because, in theory, it allows young peopleboth men and womento migrate for work by freeing them of their caregiving responsibilities, which contributes economically to the household (Leinaweaver, 2010). While elderly caregivers are considered ideal, one middleaged aunt caring for her young niece said, "Even the younger people can be with [orphans] as long as it's a female person." In these ways, both men and women participate in the feminization of care work (Dworzanowski-Venter, 2008). As AIDS mortality and labor migration by both men and women persist in Lesotho, gendered roles have become increasingly flexible, yet discourse about these roles remains aligned with their idealized norms.

The emergence of male caregivers
Thus far, I have gone to considerable lengths to demonstrate that men are not considered ideal caregivers for AIDS orphans in Lesotho in a paper that purports to be about the emergence of male care. Both men and women perpetuate the ideal that women, particularly elderly women, are most capable of providing children with the affective support they need. As this aging cohort of grandmothers dies, families will increasingly rely on male caregivers in order to maintain kin-based care networks. The strength of the cultural expectation and preference for elderly female caregivers is what makes the emergence of a small population of male caregivers notable.
Men do not engage in the labor of care unless there are no female relatives available. While young women, particularly aunts, will certainly be the first line of defense for orphans, conditions of "last resort" will continue to increase as demographic conditions, such as grandmaternal death and increased feminized labor, continue to shift. Of the 37 caregivers I interviewed in 2013, 8 were men: one father, two uncles, and five grandfathers. Grandfathers were preferred because, like grandmothers, their age excluded them from the labor force. If they were over 70 years old, their meager government pension of 500 Maloti per month (or about 40 USD) provided a reliable economic foundation for the household. Men hardly ever shared caregiving tasks with a cohabitating female family member. If there were such a family member under the same roof, she would be providing the daily care. Instead, male caregivers in this study found themselves as the only viable caregiver, and were expected to perform caregiving tasks without help and often with little prior experience. The cultural norm that excludes men from daily care work reinforces the belief that men should only conduct such work if there are no other options, yet it also excludes them from building the skills they will need if they find themselves in such a situation. Of course, there are examples of men who do such tasks when they are young. In addition, the men in this study demonstrate that prior experience is not necessary in order to build the skills they need when they are faced with caregiving tasks. Yet the cultural expectation that such tasks are feminine is used as justification for seeking a female caregiver if at all possible. In several cases, an elderly infirm woman was chosen as caregiver over an able-bodied man because she was deemed more suitable for the work despite severe physical impediments.
The following two case studies help to illuminate both the circumstances that lead to male caregivers and the experiences of these men in undertaking the feminized labor of daily care. Ntate Tebelo is a 90-year-old man who lives in a small cinderblock row house in the town of Mokhotlong. When I first met him, he was sitting quietly on the single bed in his dim house, repairing the laces on an old pair of leather boots. He has lived alone with his three grandsons for the past 12 years in his one room home. The boys belong to Ntate Tebelo's son, Ntate Thabang, who has been able to retain his job as a miner in South Africa, despite the drastic reduction in the workforce. All three boys were born in South Africa, but returned to Lesotho when their mother was dying. Ntate Tebelo cared for the boys' mother until her death a few months after their return. The youngest boy, Paballo, is now 14-years-old. But, at the time, Paballo was only 2-years-old and had just stopped breastfeeding. Ntate Tebelo had been married for many years, and had seven of his own children, but his wife and four of his children had died. Ntate Thabang visited his father and sons twice a year during the Easter and Christmas holidays and regularly sent money for his sons' school fees and clothes, but Ntate Tebelo cared for the boys largely on his own. This grandfather had not intended to raise the children alone, but he was the only person available. After their mother died, he implored her parents, the maternal grandparents of the children, to take them. But, he said, "I was asking them because I don't have a wife. So, that [the maternal grandmother] could help me for a while. But, they refused. That's why I lived with [the children]." He based his claim for help on his gender; but the maternal grandparents, who were already caring for a handful of paternal grandchildren, were not able or willing to take the children. When describing the funeral of his daughter-in-law, he drew attention to the lack of female support. He lamented: The family was not there when we buried this person. So, it was me who worked for the funeral with the neighbors. So, the family left there. I found that there was nobody I could talk to. Because I don't have a grandmother for these children. And my mother was not there.
The children visited the maternal grandparents occasionally, and they often came home with food from their fields and gardens, but Ntate Tebelo never renewed his request for help. He noted with pride, "I didn't ask them again. I didn't follow them." Despite his advanced age and his initial concern about his abilities as a male caregiver, Ntate Tebelo did not recall his early years of caring for the young boys as particularly difficult. He drew on previous experiences to explain how he learned to care for them. When Ntate Tebelo's own wife was sick and dying, he did much of the daily care work for her and for his own young children. A few years after his wife's death, he lived for two years with his uncle's children who had been abandoned by their parents. From these experiences, he learned the skills he needed to care for his grandsons. He said: "Here, no one helped me. I am the one who was living with them while they were young, and I put [Paballo] on my back to go to see the doctors. Hmmm." Ntate Tebelo chose to recount carrying his youngest grandson on his back pepa-style, strapped with a blanket. This act of daily care is a fundamentally feminine actone that I have never witnessed any Basotho men perform. While carrying a child on one's back has a practical function, to free up the hands, men often prefer to carry a child in their arms rather than on the back, despite its inconvenience, because of its feminine connotations. Most men claim they do not know how to tie a child on their backs and the type of blankets used for the job are never worn by males. In drawing attention to this mode of transportation, Ntate Tebelo indexed the deeply feminized nature of his role as caregiver. This devoted grandfather is healthy for a 90-year-old man, though his mobility is limited. His grandsons are now old enough to help him around the house. When I asked what would happen to them when he was no longer able to care for them, he told me: "They have their older brother and he is a man now." His acceptance of male care is both positive for his own grandsons' futures in the event of his death, and indicates the possibility that gendered caregiving roles could be changing even as discourse about them remains highly feminized.
While grandfathers are preferred to young men because of their stability and their exclusion from the labor market, young men are also needed as caregivers in the absence of grandparents or other female relatives. Moreover, the migrant labor market is increasingly favorable for women while opportunities for male migrants are in decline. In some cases, this may make males as caregivers and females as migrant laborers a more economically rational division of household labor.
Thirteen-year-old Mantja lived with her maternal grandmother after she was orphaned at the age of one. In 2008, when she was 8, Mantja's grandmother died, so she was left in the care of her 34-year-old, unmarried maternal uncle, Ntate Tieho. He lived in the rural village of Ha Tjope with Mantja, another teenaged niece whose mother worked in South Africa, and two of his brothers, aged 18 and 25. When his mother died, Ntate Tieho took young Mantja to live with her paternal grandmother; however, she was not being well cared for there so they took her back to her natal home. Ntate Tieho (NT) and his brother, Ntate Makashane (NM) described this situation to me: NT: We took this child from her grandmother, the mother of her father. Because the grandmother was not taking good care of the child. The father was gone to Natal [South Africa], so we found that the child was not in a good situation, so we took her. NM: [That grandmother] is a person who does not care. NT: She doesn't care. She is a person who is drinking beer a lot. Who is living that way … But, because of her life, kanete [I swear], we saw that the child was not being cared for well … She also has the children of her children who are living with her. And now, when the grandmother wants her, [Manjta] refuses to live with her now.
In this conversation, the importance of proper care was emphasized over both gendered ideologies of care and patrilineal social norms. Ntate Tieho noted that because Manja's mother was married and bridewealth had been paid, the child belonged to the paternal kin. However, the caregiving burden this grandmother already had, her inability to provide proper care for Mantja, and Mantja's refusal to stay with her, led to Ntate Tieho as her primary caregiver. It is noteworthy that Ntate Makashane called the paternal grandmother "a person who does not care". In this case, care (tsotella) meant "to care about". While he used the affective term for care, her lack of care was manifested in her failure to adequately "care for" (tlhokomelo) Mantja. Just as the examples of men providing care draw attention to the ways in which these idealized gendered norms are cultural, so does this example of poor grandmaternal care reinforce that while gendered norms become naturalized, they are by no means natural.
The heavy caregiving burden has not been easy for Ntate Tieho, who emphasized poverty as the biggest challenge. He lamented, "Sometimes you don't have the money to support them. You can't afford their needs. It is very difficult. Especially if you don't have a way to answer their needs." Ntate Tieho supported the family with remittances from his migrant sister and by selling vegetables from their small garden. They had one field, which did not produce enough maize and wheat for household consumption and they occasionally received food aid from the World Food Program. However, they struggled to make ends meet. As Ntate Tieho poignantly said: "We are covered in poverty." Like Ntate Tebelo, Ntate Tieho drew on prior knowledge to help him with the caregiving tasks required of him. He said he learned most of it from his mother, who expected both her daughters and sons to participate in household chores: Kanete [I swear], our mother, she was helping us. When you put the dish there not washing it, she was angry saying, "You must wash it". That's why when she was not there, we remember the things that she taught us to do. But, sometimes it was so difficult, we were saying that she was scolding us. But, the things that she told us to do … were the things that when she died, we remembered them. So, we did them.
His brother, Ntate Makashane added, "Because our mother was giving us a good upbringing, it's simple to pass that on to [the children]." It is notable that Ntate Makashane characterized these lessons as "good", because it indicates the possibility that gendered caregiving norms could change at the level of socialization. The brothers also said they acquired some of their caregiving skills because they cared for her mother during the last year of her life when she was ill. While both brothers said things were much easier when their mother was alive, her influence as a grandmother and caregiver is felt even after her death.
Despite these hardships, Ntate Makashane told me he was content because he said: "We know that if we have nothing and they also will have nothing. But, if we have something, they will have it too." Ntate Tieho added: "To be family it cannot be simple to be separated." In other words, the strength of their kinship ties drew them together. Even in the absence of a woman to provide care, they did not consider looking for assistance outside of the family. Yet the preference for a female caregiver persists. Ntate Makashane recently got married. Now his wife and their infant son also live in the house, and while the men continue to assist with household duties, this newly married sister-in-law does the cooking and washing for the household, with the assistance of the teenage girls. Thus, male care is occurring with increasing frequency, but at least when it comes to daily caregiving tasks, it is sometimes only a temporary solution to a caregiving problem until a more suitable (female) caregiver can be located.

Discussion and conclusion
Treatment advances and improved access to ART (antiretroviral treatment) have changed the course of the AIDS epidemic in Lesotho. AIDS has recently been characterized as a chronic illness that can be managed in developing countries (Colvin, 2011). As parents live longer, children will not be orphaned at such a young age, reducing the orphan problem as well as the burden on elderly caregivers. However, according to Lesotho's DHS data (Lesotho Ministry of Health and Social Welfare, 2010), persistently high rates of mortality-including maternal mortality, as well as continued high rates of new HIV infections-mean that the discrepancy between available female caregivers and children in need of care is not going to disappear for quite some time. 2 The demographic impacts of HIV are numerous. These shifts include an aging generation of HIV-free grandparents, a decline in AIDS mortality coupled with the persistence of high rates of new HIV infections, and an increase in female labor migration. The population of potential caregivers for AIDS orphans is constantly in flux such that the preference for elderly female caregivers will not be sustainable within kin-based networks of care. Thus, it is important to understand the resilience factors and potentially unexpected ways that orphans and vulnerable children can be protected and cared for within the family.
The naturalistic assumptions that reinforce rigid notions about gender and kinship (Collier & Yanagisako, 1987) have masked the potentially important role that men could play to fill this caregiving gap in rural Lesotho. As the availability of ideal caregivers shifts, kin-based caregiving networks are thus far being sustained by a noticeable shift in practices of care. A slow but visible decrease in the number of elderly female caregivers has facilitated a small but significant increase in the number of male caregivers. There are two plausible and compatible explanations for this shift. First, masculinities (and more broadly, gender) have been shown to be flexible in an African context (as elsewhere). A number of historical and political economic shifts have altered gender relations, including colonialism, apartheid, migrant labor, and urbanization (Hunter, 2004;Mantell et al., 2009;Morrell, 1998). In some ways, gender relations have moved toward greater gender equality. In other ways, the dispossession of African males because of these historical shifts, coupled with the maintenance of rural households, has entrenched and reinforced ideas about African patriarchy and authority (Harrison, O'sullivan, Hoffman, Dolezal, & Morrell, 2006;Sideris, 2004). Yet, as Bozzoli (1983) has noted, the complexity of gender relations leads to what she has termed a "patchwork" of patriarchies. In Lesotho, this patchwork is able to encompass the contradictions that come with entrenched notions of patriarchy as well as an increase in male caregivers, even while discourse about masculine roles in the domestic realm remains contested. For Basotho, gender is a malleable cultural resource on which to draw in responding to the needs of orphans. Notably, the presence and actions of male caregivers are often understood and reframed in the context of idealized female care. Male caregivers providing poor care are often excused because they are not women, whereas those providing good care are equated with women. Male caregivers present a potential resource that could be harnessed to help lessen the strain on elderly and female caregivers, and to protect children from being raised outside of their extended network of kin.
The second explanation for the surprising emergence of small numbers of male caregivers lies in the resilience of kin networks. Rather than being contradictory, these two trends work in concert. Kin networks are fluid and flexible, and have long adapted to changing historical and political economic circumstances. The strength of kinship networks and the preference for kin-based care trumps the gendered ideals about care. In other words, families would rather have a man caring for a child than to have that child cared for outside the web of kin relations. The malleability of gender roles and norms enables and buttresses resilient kin networks under times of duress, and the resilience of kinship as a system is stronger than even entrenched gendered practices.
A strength-based approach, which identifies the strengths and resources of individuals and communities in dealing with adversity (Cohen, 1999), highlights the malleability of cultural resources in responding to the challenges of HIV/AIDS. In this case, idealized and seemingly rigid notions about gender and appropriate models of care were more flexibly employed by families to respond to the needs of AIDS orphans than initially appeared possible. A small but significant group of men caring for orphans, even temporarily, may help to protect kin-based networks of care. Ethnography was critical in uncovering this community strength because examples of male caregivers were originally framed as exceptional. However, local cultural resources were far more adaptable than short-term investigation would have suggested. While the data presented here is not robust enough to suggest lasting cultural change around gendered norms, we know from previous research on African masculinities that changes in gendered norms do occur. We also know that they are slow, and are likely resisted at first. It is thus likely that a change in practice, precipitated by political economic circumstances, would precede a change in gendered ideologies, which are deeply entrenched.
While male caregivers are bolstering families' ability to provide care for orphans, it is important to note that there are few alternatives for families outside of kinbased networks of care. Institutionalized care is not widely available and is less culturally appropriate, so many caregivers do not consider it an option (Drew, Makufa, & Foster, 1998;Ghosh & Kalipeni, 2004). Women will continue to be important in providing kin-based care for orphans. And while men will still be considered caregivers of last resort, they may become increasingly less exceptional and increasingly important as conditions of care continue to shift. Notes 1. Ntate (father) and 'M'e (mother) are widely used when addressing adults in Lesotho. All names are pseudonyms. 2. Adult mortality increased from 11.1 deaths per 1000 years of exposure (Lesotho Ministry of Health and Social Welfare 2004) to 13.6 deaths per 1000 years of exposure (Lesotho Ministry of Health and Social Welfare 2010) for both men and women. Maternal mortality, which is defined as death of the mother occurring during pregnancy, childbirth, or within two months postpartum, also rose during this period from 964 (Lesotho Ministry of Health and Social Welfare 2004) to 1435 (Lesotho Ministry of Health and Social Welfare 2010) maternal deaths per 100,000 live births.