Sources

The article examines eldercare in three historical periods – the 1860s, the 1990s and the 2000s – focusing on Akuapem. For the 1860s, I rely primarily on the records of the Basel Mission, active in southern Ghana, from the Ga, Akuapem and Akyem-Abuakwa presbyteries. These extensive records include reports of Basel missionaries, mainly from southern Germany, and of ministers, catechists and Christians from the Gold Coast. During this period, the Basel Mission was deeply enmeshed in dealing with slavery and debt pawning, because converts and schoolchildren were being pawned or sold and because the Mission obtained labour and schoolchildren by paying off a person's debts. Furthermore, beginning in 1861, African Christians were under pressure to give up their domestic slaves.Footnote ² The records are not particularly focused on ageing, in part because most of the converts were young people, but there is some data on older Christians. Another important source is Jonathan Bekoe Palmer's notebook, which notes disputes in Akropong and Akyem between 1860 and 1867.Footnote ³ I also use the colonial court archives in Accra, focusing on cases concerning slavery and fosterage among Akuapem people from 1869 to 1900; there is very little on ageing. A few court cases from Akuapem appeared in those records, which suggests that most Akuapem people resolved their disputes through means other than the colonial courts, as also indicated by Bekoe Palmer's notebook.

The data from the 1990s comes from life-history interviews conducted in Akropong. In 2008, I completed a household survey in the town of Akropong to understand child fosterage, interviewing ninety-three foster parents. They were teachers and retired pastors, as well as many charcoal sellers and prepared food sellers, people who could pay for their child's secondary school education and people who were barely scraping by. Most of my interviewees were Akan in ethnic origin, either from Akropong or from another town in the Eastern Region, but one was Ewe, one was Ga, and one was Frafra. The data here comes from interviews with foster parents about their own childhoods and migrations. What I discovered in the course of these interviews was how foster care and eldercare were intimately connected. I use this material, originally collected to document the history of foster care, for its insights into eldercare in the late twentieth century.

Finally, I have conducted ethnographic research for the past three summers (2013–15) on home nursing agencies in Accra and the activities of various non-governmental organizations (NGOs) and churches to support elderly people and encourage elder advocacy; I do not draw directly on this research but it informs the questions that propel this article.

Domestic slaves’ involvement in eldercare

The Basel Mission records provide several tantalizing pieces of information which suggest that domestic slaves played an important, if perhaps supplementary, role in eldercare in the past. Slaves were quite common in Akuapem. Gilbert reports: ‘In mid nineteenth century Akuapem, it is estimated that probably one half of the population of Akuropon [Akropong] were slaves and captives from the Ewe wars’ (Reference Gilbert1995: 378, footnote 15). The colonial court records show that, after the Ashanti invasion of Krepi in 1870, many refugees from that area fled into Akuapem or were reduced to slavery in Akuapem during the 1870s.Footnote ⁴ Slaves were captured in war or as refugees but also individually. In Akuapem, Akyem and Krobo, they were primarily used in long-distance trading, the transport of agricultural products such as palm oil, and cash-crop farming (Haenger Reference Haenger2000; Addo-Fening Reference Addo-Fening1980).

Slavery and kinship existed on a continuum (Miers and Kopytoff Reference Miers and Kopytoff1977), in which slaves could achieve a quasi-kin status, including the ability to inherit, although they were not equal members of the family. Kin groups acted as corporate bodies that had rights in and responsibilities to family members, whom they could transfer to another family house or person in return for goods or money. Slaves and masters used fictive kin terms to refer to one another. For example, in 1870, Kwasi Amoako, Gyaashene of Kyebi, referred to his slaves as his sons (Addo-Fening Reference Addo-Fening1980). In a court case in Awukugua, Akuapem in 1871, the ten-year-old child of a slave called his mother's owner his mother, confusing the British officials; the interpreter helpfully explained that it was customary for slaves to call their owners ‘mother’.Footnote ⁵ Slaves and family members ate and worked together, but slaves, more than family members, were used for the most onerous and dirty work, including transporting goods to the coast (Abun-Nasr Reference Abun-Nasr2003; Perbi Reference Perbi2004). Thus, although they used fictive kin terms with one another, the distinction between ‘slave’ and ‘kin’ was maintained.

Slaves enlarged households, with the size of households a determinant of social status and power (Addo-Fening Reference Addo-Fening1980). In 1861, the missionary Mader described households in Akuapem as composed of fifty to a hundred people made up of a family head, the wives and children of the current and previous family heads, and inherited and more recently purchased slaves and their children (Haenger Reference Haenger2000).Footnote ⁶ Although inherited slaves were not so easily sold as individually owned slaves, they could be pawned or sold in case of family debts, protecting kin from a similar fate. In general, although always contingent on other factors, in Akuapem, slaves worked for their masters a few days a week, and could accumulate their own resources on the other days (Haenger Reference Haenger2000). Through such means, slaves could themselves acquire slaves.

In 1862, after heated discussions, the Basel Mission decided that any slaveholder who wished to be baptized must free his slaves first, and that current members of a parish should release their slaves within two years (Haenger Reference Haenger2000: 22). The Basel missionaries were somewhat inconsistent in implementing this regulation, making exceptions in individual cases and allowing slaveholders to make their slaves pay off their price through wage labour, a ‘solution’ that highlighted the similarities between wage labour and slavery. Furthermore, the missionaries themselves, in freeing Christian slaves, expected them to work off their purchase price for a period of time, and, as a result, others around the Mission may have considered the missionaries to be engaged in slaveholding themselves. It is from records of these kinds of dilemmas that we learn of cases in which slaves provided eldercare, although male kin or family heads tended to manage this. Although such cases may overemphasize the role of slaves in providing eldercare, alongside and supplementary to care provided by adult children, they do suggest that, in the past, adult children were not the only ones involved in providing care to frail, elderly people. Furthermore, such narratives suggest that eldercare could be a fraught, negotiated process; the aged could be in danger of being abandoned or neglected; and the aged worked actively to prevent such neglect, not only in the present, but also in the past.

I became interested in the 1860s because of one particular record, in which the missionary David Eisenschmidt in Kyebi, Akyem reported in 1866 on an old man who had been attending Christian services for several years but had not asked to be baptized. When asked for his reasons for the delay, the old man said that he was loath to give up his two slaves (which he would be pressured to do on baptism), because he was worried about who would care for him in his old age. He himself was a house slave, or inherited slave; in other words, his mother or grandmother was a bought slave who had had children with a family member. In a matrilineal society, her descendants would also be considered slaves. However, reported Eisenschmidt, the old man had hit on a solution to his dilemma: to marry a young woman who was a slave!Footnote ⁷ That way, he could return his two slaves to his master. Unfortunately, because the old man is not named, I cannot track him in further records to see whether his strategy was successful. It is also not clear to me whether he was more vulnerable in old age because he was an inherited slave and he could not recruit the care of kin, because he did not have any.

Other documents show that wealthier older men in the town of Akropong, Akuapem also struggled to secure care from their slaves and kin. Elizabeth Gyabisa, a childless Christian slave, fifty years old, was taking care of an elderly and sick relapsed Christian patriarch, Abraham Kwaku Sae, after his two wives left him for younger men (Haenger Reference Haenger2000: 44). These two examples suggest that, while wives were primarily expected to provide care to their husbands, domestic slaves could provide care in their absence, and vice versa.

One possible outcome for Christian men without young and loyal wives and who did not own slaves was manifest in the example of an Abokobi Christian elder, whom the missionaries thought about the age of seventy. In 1866, a Basel missionary noted:

That he had to do his own household chores was clearly beneath his dignity as a Christian male elder and more than he could handle physically, prompting the Basel missionaries to advocate for an exemption in his case to allow him to keep some ‘servants’, a euphemism for slaves.Footnote ⁸

There are cases of daughters providing care to their elderly or invalid mothers. A young woman attending the Girls’ Institute at Abokobi in the Ga Presbytery left for a period of time to attend to her sick mother (Locher Reference Locher1863). Heinrich Yaw, a young Christian carpenter from Akropong, earned money by sawing wood for the Mission to buy back his sister Afua ‘so that his sick old mother would have her support’ (Haenger Reference Haenger2000: 43). Heinrich and Afua's mother was a Krepi woman and slave to Kwaw Kutanku, an important elder and leader of an asafo or warrior company in Akropong; Kwaw Kutanku, a stingy man (reported the missionary Widmann (Reference Widmann1863)), had married her only daughter, Afua. When Afua had had an extramarital affair with a chief, the Krontihene, her husband had sold her to someone else in Accra. Through his wages at the Mission, Heinrich Yaw was successful in redeeming his sister and bringing her back to Akropong.

Another Christian whom the missionaries sought to protect and whose story involves eldercare was Rosine Opo (or Po). Her conflict-ridden story is well documented in Basel Mission records (Haenger Reference Haenger2000; Sill Reference Sill2010). In one incident in 1867, when she was probably in her mid-thirties,Footnote ⁹ she was put in the stocks by her family (Palmer Reference Palmer1860–67: 402–12). The immediate cause was that, instead of looking after her sick father, she went to live with another man, a Christian, whom she considered her new husband. In the airing of the conflict before a visiting government official, a family member told Rosine Opo: ‘We told you, stay with your father and give him water in the morning to wash, and cook something for him to eat.’ The family heir, Kwame Asiedu, stated similarly: ‘I let them put her in stocks if she doesn't stay in Aboasa [a neighbourhood of Akropong] and look after her father.’ She was not simply a daughter to her father, but also a house slave to his family, because her mother was a slave. Furthermore, the family had incurred many debts because of her conflicts with her ex-husband, to the extent that family members, including her own children,Footnote ¹⁰ had been pawned. For all these reasons, the family felt that they could control her residence and labour, including her care-scription for her sick father. They were quite annoyed with her: the family heir threatened to kill her as a sacrifice to accompany her father when he died. When she complained about this threat, another male family member told her to be quiet because she had no brother to support her. A third accused her of potentially causing her father's death because of the court cases she had brought through her previous marriage. Her case, as well as that of Afua, Heinrich Yaw's sister, suggests that daughters who were indebted to their families could be more easily care-scripted – although not in this case! She was kept in the stocks for eight days in July 1867 before the missionaries secured her release (Haenger Reference Haenger2000: 38).Footnote ¹¹ Missionary presence created a route for a slave daughter to refuse care-scription, and for another to be care-scripted by her brother.

Despite the proscriptions against Christians owning slaves, Christians continued to keep and buy slaves to provide care for sick and elderly relatives. In a report on slavery among Christians by the missionary Johannes Schopf in 1894, he documents that in Tsui, a hamlet near Anya and Accra, Johannes, a Christian with leprosy, with neither wife nor children and with a lame adult brother living nearby, bought a boy ‘whom he treated like his own child’ to ‘fetch water or firewood for him or buy food for him in the market’ (Haenger Reference Haenger2000: 179). A catechist within the Ga Presbytery bought a slave girl when he was in a remote station, with an ill wife and small children. He bought the girl to help fetch water. He said, ‘I was driven to it by need’ – a need no doubt exacerbated by his work for the Mission, which placed him among strangers rather than among kin who might help. In a third case, a Christian bought a girl to help his ‘sickening, argumentative, and mentally somewhat disturbed wife’ (Haenger Reference Haenger2000: 179). Atobrah (Reference Atobrah2012–13) similarly argues that Ga women used slaves, pawns and foster children to help with household tasks. One of her informants, a seventy-seven-year-old woman in Central Accra, talked in 2008 about how a slave girl assisted her blind grandmother, probably in the 1940s or 1950s (ibid.: 92). Perbi has noted that one of her informants said that her grandmother, a royal from Akropong, was carried to school every day by a slave (Reference Perbi2004: 101) and that another informant in Dormaa, Brong-Ahafo told her that slaves did the work one became too weak to do in old age (ibid.: 110).

The cases discussed here suggest that, in the 1860s, slaves were care-scripted by men, both family heads and brothers, to provide eldercare in the absence of wives and children. Slaves who were also daughters were particularly coerced, as we see in the case of Rosine Opo, but such eldercare was always negotiated and might be deflected. By the late twentieth century, eldercare was more regularly provided by the oldest daughter.

The eldest daughter as caregiver

In the care-scripts of the late twentieth century, the eldest daughter was the ideal, normative caregiver of an elderly person. Leinaweaver (Reference Leinaweaver2010) talks about the same process in Peru as filling ‘the care slot’. By the late twentieth century in Ghana, an older woman usually filled the care slot, by providing daily care and household labour both to the elderly and to grandchildren left behind by migrant parents in the home town. Furthermore, she maintained a presence in the family home and represented the family at funerals in the home town. A small garden plot provided everyday food; other expenses were met by relatives, such as the care provider's adult children and male siblings, who ideally remitted money to support this home-town household. As the older woman herself became too frail and weak to do care work, the care slot was passed from one woman to another. Usually, the oldest daughter was recruited by others to live with her mother or her mother's siblings, in part because she was most likely to have grandchildren who also required her care, to be divorced, and to be ready to pull back from strenuous work. However, if other daughters or granddaughters were better positioned because of their unemployment or marital instability, then they might be asked in the first daughter's stead. It was these women whom I interviewed in Akropong in 2008; they were not only fostering the children of their family (their grandchildren, and the children of their nieces and nephews), the topic of my research, but also taking care of family elders, or had done so in the past.

Since the 1950s and 1960s, women in southern Ghana have migrated in their late teens and early twenties from their home towns to the urban areas to work as traders (Brydon Reference Brydon1979; Middleton Reference Middleton1979). The expectation of filling the care slot as their mother aged meant that they returned home in middle age. Among the women I interviewed, they returned to live in their home towns between the ages of forty and sixty after a change in their or another's life circumstances. One woman returned to Akropong after her husband's death, which devastated her financially; another woman when the company she worked for went out of business; and a third when she retired from a job in the formal sector. The home town functions as a social safety net for women. One can live more cheaply in one's home town, because one can stay in the family house, built by a father or family head, for which one does not pay rent. Other relatives live nearby, and they may be able to share labour and resources and provide companionship. In her home town, a woman may also have access to family land, where she can tend a small food farm that reduces her need for cash. In their stories of return migration, women's personal circumstances and the decline of an elderly person coincided, which suggests that the reason why they, of all their siblings, came to live with the elderly person or people is that they were available to do so and in need of the safety net of the home town. Although returning to the home town entailed a loss of financial resources, filling the care slot helped female family members justify financial support from their brothers, the children of elderly relatives, and their own adult children whose children were staying with them.

Abena Oforiwa illustrates some of the general features of this care-script. She lived in a family house in her home town with her older cousin, Akosua Yirenkyiwa, whom she considered her sister, along with eight grandchildren, many of whom were toddlers although the oldest was fourteen. In the late afternoon, I often encountered the older children returning from the family plot, laden with firewood and produce. Forty-eight years old in 2008, Abena had gone to work in a sugar factory in a nearby town in her youth, but returned to her home town around the age of thirty when the sugar factory closed. Her mother had been living in the family house and died, leaving her cousin, an elderly woman, alone. Abena explained how she inherited her mother's position:

Akosua Yirenkyiwa, sitting nearby, commented on her sister's narrative: ‘You who are the eldest female will come home, and the [adult] children will go work hard to look for something to make their living.’ The older woman is care-scripted, allowing others – the young and the male – to focus on economic gain and financially support the daily work of care.

Women get care-scripted in part because the economic sectors in which they work make them poorer than men, on average, and liable to experience economic ups and downs in their work as traders or entrepreneurs (Clark Reference Clark2010; Heintz Reference Heintz2005). This makes them both more prone to be asked to help when there is a need for care work and more willing to accept such a request. Women's structural inequality makes them dependent on their brothers and grown children for support; they are more willing to assume the care slot in their family because of their need for the assistance of others. As I have discussed elsewhere (Coe Reference Coe2011b), Abena Oforiwa said that her older brother had recruited her to move to the family house:

Although not mentioned, such persuasion was no doubt accompanied by the older brother's promise to help with her household expenses.

Not all women are as happy as Abena Oforiwa to care for their elderly relatives, because of the loss of personal income that it entails. A forty-three-year-old woman, Yaa Ofosua, had travelled when she was young: she had first apprenticed as a seamstress in the commercial town of Suhum, in the Eastern Region. There, she married and worked as a trader. Later, her husband died, and she became involved with another man in a more short-lived relationship. After that relationship ended, she returned home to take care of her mother, whom she described as having difficulty moving around and unable to prepare food or fetch water for herself. Yaa Ofosua was less happy than Abena Oforiwa about returning home to care for her mother, and she complained about how difficult it was to make a living in Akropong. She altered clothes occasionally and received firewood and foodstuffs from working as a labourer for other farmers. She worked on her own small plot of land. Her mother ran a little bar in the house and her brother sent money once in a while, but not enough to satisfy her. She scrimped and saved to pay her seven-year-old son's school fees. Yaa Ofosua wanted another of her ten siblings to take her place in looking after her mother, so that she could return to a more commercial town or city. In the presence of her mother, she stated baldly:

Yaa Ofosua was not willing to continue filling the care slot, but instead positioned herself as a migrant who could work and as a mother who could leave behind her child in the home town. Such public statements stated so forcefully might well put pressure on her siblings, either to send more money to their mother so that Yaa Ofosua did not have to suffer in such poverty, or to replace her so that she could earn money to send home.

For some women, the return home seems more voluntary, and the timing of personal circumstances coincides with an elderly person's need for care, as for Abena Oforiwa; for others, like Yaa Ofosua, the return home is perceived as an economic hardship. The return migrant's interpretation of her situation may fluctuate over time and be partially dependent on the degree of financial support she receives from her brothers, children, and other migrant relatives.

Women are called upon to be flexible in their physical location because of their economic dependency and because they are associated with the daily and practical tasks of providing care: cooking, cleaning, laundering, marketing. The male eldercare role involves managing care, by sending remittances, visiting occasionally, and organizing funerals (van der Geest Reference van der Geest2002). The gendered nature of providing care is illustrated by the life history of one seventy-year-old woman in Akropong. She explained that she had taken the place of her father and his brothers when she was a child: ‘On my father's side, they were all men, so they didn't have anyone. They didn't have a woman to stay with their mother, so they brought me to live with her.’ Essentially, a daughter of one of the brothers substituted for a non-existent sister to fill the care slot.

What can a woman like Yaa Ofosua do to prevent her care-scription? Women who want to postpone their return to the home town and their conscription to care work (perhaps indefinitely) can substitute another woman or girl: a daughter who has not yet grown up, another sibling who has experienced a divorce or is struggling financially, a more distant relative, or a domestic servant. By doing so, women assume the more male role of being care managers as economic migrants who send remittances. I discuss a case of substitution below.

The fostered child as caregiver

Those daughters who are reluctant to cut their economic migrations short and who have the economic or social means to recruit someone else to fill the care slot find a substitute in a foster child or house help. Fostering has long been common in Akuapem, particularly among older children for the purpose of their education, training and discipline. In particular, children were sent to live with an educated professional – a nurse or teacher – who lived in an urban environment; their school fees were paid for or they helped their foster parents with their business (Coe Reference Coe2013). More recently, it has become more difficult to differentiate between a foster child and a house help, especially when a child who is more extended kin or non-kin is fostered in a household.Footnote ¹² Even in small towns, because adult daughters living in the cities have competing responsibilities, and because they have access to cash, they may delegate the work of daily eldercare to more extended kin or house help, whether an adult woman or an adolescent girl. Thus, the focus on the responsibility of adult children, as illustrated by the play, obscures the care work of extended kin, fostered adolescents and domestic servants.

I encountered the household of Mama Adelaide in Akropong during my fosterage survey in 2008. Mama Adelaide was a seventy-eight-year-old woman who could not walk easily. She was living with Esther, the fifteen-year-old great-granddaughter of Mama Adelaide's maternal uncle. Mama Adelaide called Esther her great-granddaughter; Esther called Mama Adelaide her paternal aunt (sewaa). In exchange for her domestic care, Esther's junior secondary school fees were paid by Mama Adelaide's four adult children (two men and two women) – a trader, an owner of a beer hall, a nurse, and a car alarm installer in Accra. Esther had come to live with Mama Adelaide from a village near Suhum, where her father grew cocoa, and she had gone to primary school there. The fact that Esther was in her second year of junior secondary school at her age indicated that her education had been disrupted or she had been held back a year, an all too common occurrence for Ghanaian students. After school each day, Esther went to the market and cooked the main meal. During my interview with Mama Adelaide one morning during the school holidays, Esther was washing clothes in the courtyard.

Mama Adelaide had herself been care-scripted into the care slot when she was middle-aged. In her fifties, she returned to Akropong to look after her mother's sister, a former seamstress, for eight years, before the older woman died at the age of eighty-two. While living in Akropong, Mama Adelaide had farmed a small plot of land, but she had become too weak to do so. At that point, her children realized that she needed help and recruited Esther into her household.

Esther would prefer to live with her mother and father. Her mother had come to visit her three times in the ten months of her residence in Akropong. Her father's family lived nearby in Akropong, and she occasionally visited them. She said that she did not feel overwhelmed by the care of Mama Adelaide, because she could call the adult children in Accra and they would come quickly. No doubt, her criticism of the arrangement was tempered because it allowed her to continue her schooling, in ways that living with her parents did not.

I was not able to speak to Mama Adelaide's children about why they did not come to live with their mother or send one of their own children to live with her in Akropong. However, other older women in Akropong told me that they were concerned that neither their children nor their grandchildren would come to live with them and help them. One ageing woman thought that her daughter would hire an adult woman to take care of her, rather than moving to stay with her when she needed help. She and others said that it is becoming more difficult to send a child or adolescent to live with an older person because of the longer period of schooling that is seen as a young person's right. Because schools in Akropong are no longer seen as providing high-quality education, those living in urban environments with better schools view sending a child to live with a grandmother in Akropong as a sacrifice of that child's educational future. The care-script that brought Mama Adelaide to her home town did not have the same force for her own migrant daughters, given the changing intergenerational entrustments. Thus, adult children living in the city raise their own children and turn to more extended and poverty-stricken relatives whose children would otherwise not continue their schooling. They offer to pay their school fees in exchange for eldercare, or promise an apprenticeship after some years of service for those who have already completed junior secondary school. They may even pay wages for caregiving. The care-script has been modified, to allow daughters to hire or substitute more vulnerable young women for themselves; by doing this, they can be good daughters by being care managers, rather than care providers.

Conclusions on negotiating eldercare

When I talked to Ghanaians about my research on home nursing agencies and senior care homes, many of my male interlocutors would respond that traditionally family members would take care of their elderly relations. Their responses supported the care-script presented by the aged fellowship group's drama in which adult children were held particularly responsible for eldercare. Women, particularly elderly women, seemed more pragmatic, willing to explore new practices, perhaps because they themselves had been recruited to care for an elderly man or woman in their middle age, or because they had sympathy for an over-extended daughter. Both men's and women's perspectives are reflected in sociological and anthropological studies in Ghana, which describe a crisis in old age, as kin-scripts that recruited adult children – particularly eldest daughters – to care for the elderly no longer have as much persuasive power as they did in the past (Aboderin Reference Aboderin2006; Reference Aboderin2004; Apt Reference Apt1996; Dsane Reference Dsane2013; van der Geest Reference van der Geest1997). As a result, elderly people can be abandoned and neglected. Economic hardship also plays a role, with some adult children lacking the economic wherewithal to support their aged parents, similar to the situation Roth (Reference Roth2007) describes in urban Burkina Faso.

Yet it is clear that the work of caring for the elderly is not provided solely by kin. The flexibility of Ghanaian families means that non-kin can be substituted. Although care-scripts favour the elder daughter's role in providing daily care, others – a granddaughter, or a great-granddaughter of a maternal uncle – can be substituted if no girl was born among a group of siblings or if the daughter is living with her husband and children elsewhere. Thus, we need to expand our notion of kin-work to include those who are ‘fictive kin’ or non-kin who become care-scripted, usually by kin. This means that we should not draw a sharp line around kinship and kin-care, but rather should understand how daughters, fostered adolescents and slaves can be subject to dependencies and entrustments that result in their care-scription. These entrustments change over time, as different obligations – such as schooling or debts from marital disputes – gain force and as new routes to defer commitments emerge. Care-scription is sensitive to changing economic and social circumstances, including the economic migration of women, increased educational expectations, the allure of urban areas in providing economic and educational opportunities, and longer lifespans with more chronic illnesses. Care-scripts shift, and new kinds of people are care-scripted, attesting to the flexibility and dynamism of eldercare arrangements in Ghana and making visible the negotiations around the labour of sustaining families over time.