Child-headed households in Southern Africa

Empirical literature contains very few references to child-headed households which have been abandoned by existing kin networks. Roalkvam’s description of orphan households in Zimbabwe which seem to have been abandoned by their kin, notes that bride-price was not paid, or not fully paid, in respect of their parents’ marriages (Roalkvam, Reference Roalkvam2005; see also Subbarao and Coury, Reference Subbarao and COURY2004). The children were therefore not regarded as belonging to the networks of their paternal family. However, the literature indicates that many child-headed households result from conscious decisions by family members or the children themselves. Siblings, faced with being separated from one another as they are shared out between different relatives, often prefer to live in child-headed households with material and other forms of support from family members.Footnote ⁴ Children and their relatives may also choose to leave orphans in child-headed households to protect their property rights or as a temporary measure while relatives make arrangements to accommodate them (Foster and Makufa, Reference Foster and Makufa1996). In their research amongst the Ovambo in Northern Namibia, Kuhanen et al. (Reference Kuhanen, Shemeikka, Notkola and Nghixulifwa2008) encountered no child-headed households, but found many instances in which children lived with a person of eighteen years and older, who had not yet required the social status of adulthood in the community. Although they were not technically child-headed households, they could be defined as such from the perspective of the community in which the children lived. Extended families appointed these youths to live with the orphaned children to protect them and provide for them.

The existence of a child-headed household is also not an indication that children are completely bereft of adult support or that all child-headed households are equally vulnerable. Vulnerability depends on a range of factors, such as the age of the eldest sibling in the household, the level of contact with kin and the availability of resources (Jones, Reference Jones2005). Even though they live without the presence of adults in the home, many child-headed households do have access to one or more adults to discuss problems and receive advice or material assistance (Jooste et al., Reference Jooste, Managa and Simbayi2006; Donald and Clacherty, Reference Donald and Clacherty2005). Research about childcare patterns in an informal settlement outside Cape Town, where people live in great poverty, shows that child-headed households may receive care from neighbours and material assistance from relatives who do not live with them. Furthermore, relatives may spend periods of time living with children in these households, even though they are not permanently based there (Bray and Brandt, Reference Bray and Brandt2007). Eighty percent of the child-headed households identified by Khumalo (Reference Khumalo, Rosa and Lehnert2003) in Richmond received economic support from relatives. The fluid nature of many Southern African child-headed households is neither unique, nor new. It arises from poverty and the consequent need for adults to migrate in order to access economic opportunities within the region.

Despite the serious economic, emotional and social challenges faced by children in child-headed households, recent research has drawn attention to their remarkable resilience and surprising strategies to cope with their circumstances. Even the most vulnerable children may have, or may develop, capacities such as self-esteem, a sense of purpose, interpersonal skills and other resources which may assist them in surviving under difficult circumstances (Eloff, Ebersöhn and Viljoens, Reference Eloff, Ebersöhn and Viljoen2007). The existence of the child-headed household may in itself provide emotional support from siblings, and maintaining familiar surroundings is regarded as a source of comfort (Subbarao and Coury, Reference Subbarao and COURY2004). A small-scale study by Donald and Clacherty (Reference Donald and Clacherty2005) comparing heads of child-headed households with the eldest children in adult-headed households found the heads of child-headed households to be more altruistic towards siblings and more adept at solving conflicts within their families, displaying emotional maturity not generally found in children living with adults. They note that these children were also able to create and utilise social networks to meet their needs for advice and for emotional, material and educational support. Compared with children living in adult-headed households, heads of child-headed households had better skills in managing money and time, although they had less free time available due to an increased number of responsibilities.

Heads of child-headed households may also have learnt child-rearing skills before the death of their parents. There is evidence that many children, as young as five years old, are expected to contribute to poor households by performing domestic tasks, including the care of younger siblings. Children are proud of these skills and derive from them a sense of self-worth (Bray and Brand, Reference Bray and Brandt2007; Foster and Makufa, Reference Foster and Makufa1996). It is therefore possible that their roles as caretakers of younger siblings may confer similar benefits on the heads of children’s households. Nevertheless, there are also reports that such children suffer stress from shouldering adult parental responsibilities, especially when they are unable to fulfil educational and other needs due to poverty (Subbarao and Coury, Reference Subbarao and COURY2004).

To summarise, the social science literature indicates that even where children live without permanent adult presence in their homes, they may have access to financial and emotional support from family members, neighbours and community members. Child-headed households may be only a temporary solution and adults may live with children from time to time. In addition, many children display remarkable ability to cope with their challenging circumstances and children draw emotional strength from the presence of their siblings and from remaining in their physical and social environments.

Care arrangements for orphans

This section of the article will examine the suitability of the various options for accommodating children who have no parents to care for them. The discussion is based on the South African legal system, which is arguably the best-resourced and most formally developed system in the region. The ‘formal’ options for caring for orphans are formalised foster care, where foster parents are appointed by a court or, more recently, in ‘cluster foster care schemes’, where groups of children are fostered by a group of adults or an organisation.Footnote ⁵ Foster carers are usually entitled to state welfare grants, to facilitate their care of the children. If this is not possible, then the state must provide accommodation in a state institution or a state-funded institution.Footnote ⁶ Another formal care option is that children be adopted, either within the country or by people living abroad. However, alongside all of the formal options of care, which would involve the state, welfare agencies and the courts, the majority of children are cared for in informal fostering arrangements. Southern African states rely heavily on informal solutions to cope with the rise in the numbers of AIDS orphans.

Family status

The limited South African recognition of certain child-headed households raises fundamental questions about the bases upon which families are generally recognised by family law. Clearly, what constitutes a family is culturally and temporally specific. For instance, Minow (Reference Minow1985) points out that nineteenth-century American families included servants and slaves, while in early Roman law, the paterfamilias had potestas both in relation to adult children and slaves (Clark, Reference Clark and Visser1989).

Recently, many legal systems have move moved towards extending family status to hitherto excluded groups like same-sex couples and unmarried fathers, while the family status of unmarried cohabitants is currently contested in certain jurisdictions (Bainham, Reference Bainham, Ebtehaj, Lindley and Richards2006; Bala and Bromwich, Reference Bala and Bromwich2002; Voegli, Reference Voegli and Maclean2005; Fuszara and Kurczewski, Reference Fuszara, Kurczewski and Maclean2005). These moves have been justified by arguments for the protection of individual rights and the need to protect the interests of weaker parties (Voegli, Reference Voegli and Maclean2005; Glennon, Reference Glennon and Maclean2005). Within academic and public debates about family formation, the strength of normative definitions of families could be measured by the level of anxiety which ‘abnormal’ families, such as unmarried mothers and same-sex families, provoke. These anxieties are often articulated as concerns about the level of care which children will receive in non-traditional families (Fineman, Reference Fineman1995). We can therefore say that, while there is an urge to include children within families, they have to be included in particular kinds of families. Furthermore, the existence and proper care of children is often used in arguments to either extend or deny family status to certain groups.

In the remainder of this section I draw attention to certain underlying characteristics of those families which are afforded legal recognition. I argue that they remain based on sexual relationships between adults, who should, according to current thinking, display qualities of independence and self-sufficiency, that families are nuclear rather than extended families and that they are premised on the idea of active, caring adults and passive, care-receiving children. I argue that there are both practical and conceptual advantages to recognising the family status of certain child-headed households. Even where child-headed households are not granted total autonomy, there are ways to give formal recognition to the care provided by members of extended families and other community members.

Nuclear families

The focus on adult sexual relationships as a defining feature of families undermines the importance of other relationships between family members. For this reason, extended family members, despite their social, emotional and material importance, are not central to the definition of family (Bainham, Reference Bainham, Ebtehaj, Lindley and Richards2006). There are, for instance, no legal systems which recognise aunts’ and uncles’ duties to support or care for their nieces and nephews. Even when extended family members take an active role in raising children, they have no automatic legal rights in relation to children, but have to acquire them through formal legal processes as if they were strangers.

This belies the moral and emotional duties which extended families feel towards children, especially in poorer Southern African communities. In these communities, social, economic and emotional ties extend far beyond households of adult sexual couples to include people who do not share the same residence, but whose family status is acknowledged and who recognise and practise obligations of mutual support. Put succinctly (Russel, Reference Russel2003, p. 11):

The strength of mutual obligations of kinship is evident from the voluntary care of Southern African orphans by family members. Family care continues, even when people are not living together. Research by Bray and Brand (Reference Bray and Brandt2007) shows how financial provision by absent family members symbolises emotional and social ties and how neighbours and friends in an informal settlement care for each other’s children from time to time. The reality in these communities is therefore one of mutual obligations, rather than of isolated independent adults, and of care relationships which extend beyond the adult sexual couple and their children to encompass wider kin and community groups.

Legal capacity of children

Having examined some abstract and conceptual problems raised by the existence of child-headed households, I now turn to the more practical and immediately pressing matter of the legal capacity of children who head these families. Like the assignment of family status, legal recognition of status and assignment of legal capacity are political acts which enhance and protect certain interests at the expense of others (Naffine, Reference Naffine2003). The old Anglo–American rule merging a wife’s legal personality with that of her husband is an example. Limits upon legal capacity are generally justified on the basis that they are designed to protect the vulnerable, as is evident in the case of wives, people with mental illness and children (Minow, Reference Minow1985; Eekelaar, Reference Eekelaar1994), but it could likewise be detrimental to those who lack decision-making capacity and advantageous to those who have the capacity to decide on their behalf. The ability of parents to decide, for instance, whom their children may associate with, could protect children against the consequences of their own unwise choices, but it also allows adults to make decisions which suit their own convenience, prejudices and views.

The effect of age on legal capacity seems somewhat arbitrary and haphazard (Archard, Reference Archard1993; Jones and Basser Marks, Reference Jones and Basser Marks1994). Although there are minimum ages at which certain capacities arise, on the assumption that children below these ages are likely to lack sufficient judgement, there are no corresponding maximum ages at which we presume that people lose their ability to judge. We do not, for instance, assume that people aged ninety and over are likely not to be able to make beneficial and independent decisions, although we may have evidence that this is likely to be the case. Moreover, in relation to children, there are graded sets of age limits for different purposes. In South African law, for instance, children below seven lack both criminal and civil capacity. However, thereafter different ages indicate different capacities, such as the capacity to consent to their own adoption (ten), to witness wills (fourteen), to consent to medical treatment (fourteen), to be presumed to have criminal capacity (fourteen), to consent to sexual intercourse (sixteen), to make wills (sixteen), to be a depositor in a bank (sixteen) and so on (Van Heerden et al., Reference Van Heerden, Cockrell, Keightley, Heaton, Clark, Sinclair and Mosikatsana1999). Within the same legal system, the age of majority has been changed from twenty-five, to twenty-one, to eighteen, all presumably based on an evaluation of children’s capabilities and need for protection.

The different ages at which children acquire various legal capacities reflect the general perception that children develop the abilities which are necessary for full legal status over time and as a result of increasing life experience. Recently there has been a move towards granting children greater rights and capacities at earlier ages, especially in relation to expressing their preferences about their own living arrangements (for instance Dwyer, Reference Dwyer2006). Many legal systems therefore allow, and even compel, judges to take account of children’s wishes in legal matters which affect them (Smart, Reference Smart, Dewar and Parker2003; Tapp, Taylor and Henaghan, Reference Tapp, Taylor, Henaghan, Dewar and Parker2003). A second development has been a move away from universally imposed age limits towards more flexible standards which evaluate the competence and capacity of individual children, based on the recognition that statutory age ranges may not accurately reflect the competencies of individual children.Footnote ¹⁴ Related to these trends is the realisation that children may be far more capable at earlier ages than was generally thought. Recent research has therefore explored the competencies and capabilities of children, rather than focussing on their limitations (Smart, Reference Smart, Dewar and Parker2003).

Academic theory on conferring legal competence to children at particular ages is sparse. Archard argues that the Western legal notion of limited, gradually increasing competence based on age reveals much of our basic views and concepts of childhood. It shows that in Western societies the stage of childhood is physically and conceptually distinguished from adulthood, in terms of the activities, space, views and attitudes which are regarded as appropriate to each period. Our definitions of children focus primarily on their lack of the socially valued adult characteristics of rationality, independence and autonomy (Archard, Reference Archard1993). For this reason, children should receive special treatment and protection from society, as articulated by the best-interests principle in most family law systems.

We can see that the same dichotomy between self-sufficiency and dependence, discussed in relation to family status, simultaneously underlies the assignment of legal status and capacity. People who display characteristics of self-sufficiency and independence are competent adults, while children are defined as lacking in independence. Autonomy therefore seems to be the linchpin upon which legal status turns in the context of the family – those who display it are afforded legal rights, and those who are believed not to display it, are ‘protected’ by being denied legal rights and status. Crucially, those who display autonomy are also, at least theoretically, not in need of state assistance, since they can look after themselves and others. In this way, the concepts of adult and family autonomy together with the public/private dichotomy limit state responsibility for dependent family members (Fineman, Reference Fineman2000; Archard, Reference Archard1993).

But how appropriate is autonomy as an organising concept in families and family law? Fineman (Reference Fineman2000, p. 19) asserts that ‘[w]ithout caretaking there could be no society’. We could extend this observation to say that without caretaking there would be no families. Qualities of altruism, sharing and mutual assistance are what enable family members to provide one another with emotional, social and economic support, especially when resources are scarce. Adult family members are not mainly valued for their independence, autonomy and rationality, but for their caring behaviour towards children and other vulnerable kin. For instance, parental autonomy on childrearing is legally permitted on the basis that, and only to the extent that, adults act altruistically in relation to children.Footnote ¹⁵

If caretaking behaviour, rather than autonomy, is taken as the definitional feature of capacity within the family, it could provide a basis for assigning status and capacity to the heads of child-headed households. I have already referred to the evidence that in Southern African families, and possibly also elsewhere, the understanding that autonomous adults provide care for dependent children is inaccurate. Children are often valued members of households, who provide care and even economic contributions to one another and to the adults with whom they live. Care may therefore be a two-way relationship between adults and children, and ‘parenting’ of younger children by their siblings may be relatively common in poorer households, even when parents are alive (Bray and Brandt, Reference Bray and Brandt2007).

A second argument in favour of extending legal capacity is that, where children younger than eighteen have in fact cared for their siblings, they may have demonstrated sufficient autonomous behaviour, rationality and independence to justify having their capabilities recognised. Such a development would be in line with legal rules which already allow children capacity in relation to certain decisions, such as medical procedures, contraception, having sex, consenting to their own adoption and so forth. Second, there are indications that children who need to exercise certain ‘adult’ skills such as managing money and time, social networking, conflict solving and so forth, may develop these capacities to a greater extent than other children (Donald and Clacherty, Reference Donald and Clacherty2005; Eloff et al., Reference Eloff, Ebersöhn and Viljoen2007).

Finally, the purpose of extending capacity to certain heads of child-headed households would be to allow them and the children in their care to access social benefits to which they would have been entitled, but for the fact that they lack legal capacity. In other words, vulnerable children need legal capacity to facilitate state and community support. Children’s rights advocates argue strongly that the inability of children in child-headed households to access the Child Support Grant constitutes a breach of their fundamental constitutional rights. They suggest that ways should be found to pay the grant to heads of these households, even if they are younger than sixteen years old (Goldblatt and Liebenberg, Reference Goldblatt and Liebenberg2004; Rosa, Reference Rosa2004). My suggestion goes wider. In addition to the Child Support Grant, child-headed households may qualify for range of state services, ranging from free water, municipal services, health care, free education and so forth. (On the provision of these social services to children, see Leat and Rosa, Reference Leat and ROSA2005.) Rather than proceeding on a piecemeal basis for each different social entitlement, it may be more useful to grant a wider, general legal capacity to certain heads of child-headed households to access state services, enrol themselves and their siblings in schools, obtain official documentation and perform other basic legal transactions.

The legal capacities needed by heads of child-headed households are not likely to be many, because of the socioeconomic circumstances in which such children tend to find themselves. They are unlikely to buy or sell immovable property, or to make wills or trade in shares. Given the large number of different ages for different competencies of minor children contained in law already, it may be possible to create a statutory range of competencies for the heads of child-headed households, covering the basic transactions and competencies which they would need.

Such a partial extension of legal capacity to the heads of these household would recognise what Sloth-Nielsen (Reference Sloth-Nielsen2005) calls their ‘quasi-adult status’, but it begs the question of whether there should be a lower age limit for these competencies. Should a statute provide, for instance, that children under the age of twelve or ten may only function as heads of households with adult assistance? This is currently envisaged by s 137 of the Children’s Act for children younger than sixteen. Clearly, legal recognition of child-headed households and conferring legal capacity on the heads of such households cannot be feasible where very young children live together. Clear, also, is the tension between the need to protect children from the vulnerabilities caused by their age on the one hand, and the goal of enabling child-headed households to access services and support which would help them survive (Sloth-Nielsen, Reference Sloth-Nielsen2005).

Households consisting of toddlers and headed by very young children need far more adult assistance and supervision than households headed by children of fourteen and older. It may be that young children must be taken into some form of care, at least until the children are older. It may also be that even households headed by older children need structured adult assistance, as suggested by the provisions of the Children’s Act. The exact relationship between the adult assistants and the child-headed households envisaged in the Children’s Act is unclear. These adults could be social workers employed either by the State or by state-supported NGOs. It could also be argued, however, that where there are adult extended family members or unrelated adults within the communities where children live who provide children with some financial and practical support, their positions should be formalised. Extended kin and other community members could be appointed in terms of the Act to assist and supervise child-headed households. Dewar (Reference Dewar2000, p. 66) notes, in the indigenous Australian context, that:

The supervision mechanisms for child-headed households provided in the Act could be an entry point for extending the legal definition of families to members of extended families. This would reflect the reality of child-care arrangements within Southern African families and would benefit children.

Conclusion

Failing to provide legal recognition to child-headed families will not cause them to disappear. It will also not suddenly enable existing state facilities to cope with these children. Like ‘informal fostering’, unrecognised child-headed households will continue to exist, but without the social support which could have ameliorated their problems. Various forms of community involvement and formal mentorship arrangements have been suggested to assist children living in child-headed households.Footnote ¹⁶ Undoubtedly such mechanisms are necessary, especially where the heads of households are very young. In addition to this, I have argued that there are certain conceptual and practical benefits to recognising child-headed households and providing them with legal capacity as a means to survive and possibly even to thrive.

First, the recognition of the family status of child-headed households will help to destabilise the conceptual link between family status and a sexual relationship between adults, which currently characterises family law worldwide. Second, recognition of the financial and other forms of dependence within child-headed families could open the door to acknowledging that such dependencies also affect families headed by adults. This could be mobilised to disrupt the connection, in popular and state discourse, between dependence and moral unworthiness, since child-headed households are clearly not to blame for the poverty of their members. This, in turn, should focus attention on the need for state and community support for families caring for dependent members, whether those families are headed by children or by adults. One of the great inadequacies of current family models is that they remain based on the nuclear family and fail to acknowledge the real, mutual obligations of care which exist between members of extended families. Recognition of child-headed households could start to break down this insistence on the nuclear family by acknowledging the supporting roles of adult kin and parenting as a communal responsibility. A final conceptual benefit would be the recognition of the caring work already done by children within their families, moving Beyond the unrealistic dichotomy of active, caring adults and passive, cared-for children.