SEGAR & WHITE ARTICLE FAMILY MATTERS AND THE STATE: Policy and Everyday Life Julia Segar and Caroline White Introduction The ANC's Working Document 'A Bill of Rights for a New South Africa' (1990) has, under Article 7 'GenderRights', that 'Men and women shall enjoy equal rights... within the family'. Similarly, its 'Constitutional Guidelines for a New South Africa' state that 'the family, parenthood and equal rights within the family shall be protected'. The ANC has not been alone in such formal declarations - in the Freedom Charter it was asserted that 'the laws which break up families shall be repealed'; COSATU in their document - 'Workers' Rights in a New Constitution' - call for a new constitution to 'recognise the equality between men and women in marriage'; and the ANC Women's League includes in a list of proposed campaigns 'a right to family life'. Commentators - especially those concerned with women's rights - have observed that clarification of what constitutes the 'family' is essential. For some, the broad undefined use of the term implies the possible exclusion of family forms other than the nuclear family (eg Ginwala et al, 1990). For others (eg Driver, 1989), guaran- teeing the integrity of the family does not necessarily challenge patriarchal structures that oppress women in the domestic sphere. Those engaged in contemporary debate appear to be arguing for a more flexible and accommodating vision of the family in the 'new South Africa' (see, for example, Sachs, 1990; Ginwala et al, 1991). At the heart of these debates often lie two very different concerns. The first is the desire to be able to reunite families torn apart by apartheid legislation and the exigencies of migrant labour. The second is the need to recognise the growing significance of the fastest growing family form in South Africa: the matrifocal family (Charman et al, 1991:6-10). Driver (1989), in adiscussion of the ANC guidelines, describes the use of the word 'family' as 'extremely opaque' - because there are at least two models of family in South Africa (nuclear and extended). Without entering into the debate as to whether there are in fact two (and only two) clear models of the family in South Africa (or anywhere else) we would agree that it is not helpful to use any stereotyped view of the family for the process of policy-making. Of course it is not clear what form of 'the family' the ANC (and others) are assuming in these policy statements, and this is precisely why its usage is 'opaque'. Nor is the ANC alone in this opacity: the present government, along with most western governments, uses the notion of the family in an equally unproblematic manner (cf Aldous and Dumon, 1990; Barrett and Mclntosh, 1982). What is clear is that in most countries, including our own, many people, perhaps even the majority, do not live in stereotypical nuclear family units. Furthermore, policies influence the TRANSFORMATION 20 (1992) 61 ARTICLE SEGAR & WHITE shape and size of the units within which individuals live, and with which members of their kinship group they maintain relations which can be described as constituting a family. As Mary Mclntosh (1984) has put it: ... there is a whole structure of family law relating to marriage and divorce, the mutual obligations of husband and wife, the duties of parents, illegitimacy, adoption, wardship, inheritance and so forth, and... these laws do much to favour and establish a particular pattern of family life. They favour long-term monogamous heterosexual relationships; they favour the economic dependence of those who have no income (e.g. wives) on those who have (e.g. the male family breadwinner); they favour a situation in which very few children are born into the world without at least a mother and usually a mother and father, to care for them and give them a place in the social structure. In short, particular state policies help toform and maintain particular family based relationships in society. The purpose of this paper is to flesh out the final italicised clause of this quotation with material from other societies and epochs to support our argument that it is of great significance to the quality of everyday life just what kinds of policies govern- ments adopt, whether they are directly designed to influence the shape of the family or not. We would argue that it is imperative for future policy makers to have a vision of the family in the 'new South Africa'realising that the form of this vision will have far reaching consequences for all South Africans. Assumptions About Families As Ginwala, Mackintosh and Massey (1990) have pointed out, [a]ll over the world, there is a tendency to construct economic policy on the basis of a rather simple 'model' of the household unit. This model assumes that the household is: - headed by an identifiable male, who makes the decisions for the whole household; - a unit with common economic interests; - communitarian in its use of resources for the greatest benefit of all members; - consisting of a couple including the male head with identifiable dependants including his female partner. However, in Britain the stereotypical family of husband, dependent wife and two children now exists in only a minority of households. In South Africa to base pojicy on the notion of every household being headed by a male breadwinner is given the lie by UNESCO estimates of 1988 that enormous numbers, if not the majority, of African households are female headed and contain no economically active male (Ginwala et al, 1990; cf Van der Vliet, 1982). The problem of the myth of the male headed household is compounded by the 'pervasive definition of the family as private [which has] often obscured what goes on inside families' (Mclntosh, 1984). On the basis of the 'privacy of the family' the state itself may argue, or various interest groups argue on its behalf, that it has no business passing certain kinds of legislation. On the other hand, certain other kinds 62 TRANSFORMATION 20(1992) SEGAR & WHITE ARTICLE of legislation which it does wish to pass is classified as taxation or welfare legisla- tion, but its action may seriously affect the domain of 'the private'. Moore (1989) points out that assumptions of this nature concerning the family are based on, and perpetuate, ideologies of male and female gender roles. A common trend here is the assumption that women who live with husbands or male partners are dependent upon them. Often the corollary here is that those women who choose economic and social independence from men are not supported by the state (Moore, 1989:129). Occasionally women protest about this assumption of their dependence, particular- ly when their husbands' unreliable behaviour forces them to protest or starve. For example, in Alice Springs in March 1990, Aboriginal women took to the streets in traditional mourning body paint to protest at the fact that family welfare payments were being paid to their husbands as heads of families. They argued that it ought to be paid instead to mothers, as the men often used the money to purchase drink and thus the state was failing to achieve its purpose which was to ensure that the poor, especially children, were being fed (Adelaide Advertiser, 24.03.90). Here in South Africa, aspects of the legal situation regarding the family illustrate clearly the points made above. Two pieces of recent legislation have had important bearing on the rights of women within marriage. These are the Matrimonial Property Act of 1984 and the Marriage and the Matrimonial Property Law of 1988. The former did away with the marital power that a husband had over the person and property of his civil law wife. The latter extended benefits to African women by stipulating that a man may not contract a civil law marriage if he is already married under customary law (see Segar and White, 1989, for further detail). The effect of this legislation has been to give some women rights (from which they had previously been excluded) to enter into legal contracts and litigation. However, it must be remembered that this legislation, whilst most welcome, is deficient in many ways. It does 'not affect the law relating to the position of the husband as head of the family or the law relating to domicile and guardianship' (Matrimonial Property Act of 1984). Furthermore the 1988 legislation only pertains to African marriages contracted after the passing of the law (cf The Annual Survey of South African Law, 1989:87) and is not applicable to citizens of the 'self-governing' territories. However, in keeping with Moore's observation, these laws do not assist those women who fall outside of this particular vision of the family. For example, there is a growing trend amongst urban women to eschew marriage and such women and their children have joined the divorced, deserted and widowed as heads of their own families. Burman (1988) examines the situation of welfare support for such families and concludes that: ... the most disadvantaged families in the most disadvantage section of the community might receive no regular assistance at all from the State.... If welfare payments were originally meant to serve as a safety net to prevent starvation, the original intention was completely frustrated by the existing regulations (Burman, 1988:522). If one examines some of theregulationsmat Burman refers to here, we are faced with something other man an ostensible concern with child welfare, we find a particular ideology of gender. Take for instance this clause pertaining to a parent's grant. TRANSFORMATION 20(1992) 63 ARTICLE SEGAR & WHITE A parent's grant shall... not be paid to woman if - she cohabits with a man with whom she is not married and is in such circumstances as, in the opinion of the Director-General, are not conducive to the welfare of the children (Amendment of Regulations to the Children's Act of 1960, Government Gazette, 19.08.83). The notion mat the man is the legal head of the family and that this is 'morally correct' permeates all of these regulations. In order to obtain a welfare grant for children, the onus is upon the woman to prove that the father of the children cannot provide maintenance and she has to demonstrate that she has undertaken legal proceedings in order to obtain such maintenance. As Burman points out, the complicated regulations, the inefficient bureaucracy and differential treatment for different race classifications usually results in the poorest African single mothers receiving no state aid at all. Finally, the regulations provide for another type of welfare benefit known as a 'family allowance' which is 'paid for the benefit of the family as a whole' and for which the father of a family must apply. In terms of these regulations a family is defined in the following way: (a) A man and his wife who are living together as such and have a child who is the issue of one or both of them and is maintained by one or both of them. (b) A man not having a wife, or not living with his wife, and a child who is his issue and is maintained by him ...(Regulations made under the Children's Act of 1960, Government Gazette, 10.12.76). So a single-parent family headed by a man is truly a family, whilst the female equivalent is not These South African examples begin to illustrate another key point raised by Moore in her discussion of women and the state, and that is that 'state institutions and policies and their relations to women cannot be analyzed as if all women were affected by the state in the same way' (1989:129). She cites race, ethnicity, class and religion as key differentiating factors which need to be examined in historical perspective. The controversy surrounding the US programme of American Aid to Families with Children (AFDC) - in force since 1935 - provides an interesting case and hopefully some useful lessons for South Africa. Some have argued that the programme which provides welfare payments to an increasing number of female-headed household has contributed materially to the increase in number of such households and to the unemployment rate, particularly of young black men. Such accounts blame increas- ing delinquency, drug-taking, illegitimacy and the resultant unemployment on the break-up of the 'family'. It is true that the numbers of recipients of AFDC have risen steadily over the decades and more specifically, the proportion of female-headed families. During 1961-1979 AFDC families headed by women more than quadrupled from 635 000 to 3 million and constituted over 80% of the welfare case load. In addition, black families were more active recipients of welfare (Erie and Rein, 1988). Stack (1974) argued that the political and economic reality for blacks in the USA was often ignored 64 TRANSFORMATION 20(1992) SEGAR& WHITE ARTICLE by those discussing the 'black family' and that joblessness was the single most important factor determining interpersonal relationships between men and women. Her detailed study of kin relationships amongst poor blacks in a mid-western city revealed complex sets of relationships amongst networks of kin which did not conform to stereotyped views of the single-parent family. Whilst women rarely chanced marriage with men not in steady employment, males continued to play an important role in the raising of their own children and those of their siblings. However 'women come to realize that welfare benefits and ties within kin networks provide greater security for them and their children' than does marriage in such an insecure economic context (Stack, 1974:123). Erie and Rein comment that lower class American urban blacks and whites have similar attitudes and aspirations concerning marriage and family life but that the bleak economic conditions facing blacks prevent them from behaving in the same way as their white counterparts (cf Aldous and Dumon, 1990). Erie and Rein's article is particularly interesting in that they go beyond the debate on the causes of women-headed families and examine the increasing gender gap in political support for the Republican administration in the US. They note the extent to which President Reagan's welfare budget cuts in the early- 1980s affected the lives of millions of poor (predominantly black) women and attribute the significant loss of women's votes for the Reagan administration in the 1984 election to these social policies. They conclude that the welfare cuts have had the political effect of 'pitting women against women as class adversaries'. Although they challenge the notion that poverty and unemployment are to be blamed on welfare pay-outs they do note that: ... for women heading families, welfare receipt has encouraged politi- cal passivity. By enforcing dependency and by isolating women from work and one another, welfare institutions reinforce patriarchal norms (Erie and Rein, 1988:190). At the other end of the spectrum, authors cite research which shows that increased employment opportunities for white women and their greater economic inde- pendence account for their higher divorce rates and the sharp rise in their rates of single parenthood (Ellwood, 1988). This suggests that affirmative action legislation, to ensure equal opportunities for women may well provide women with the neces- sary independence to enable them to escape from unhappy marriages. More than that, it could remove from women the necessity to marry in order to afford to have children. In Britain, where some support is paid to individuals even when they have no unemployment benefit entitlement, this support is not paid to a married woman and is taken away if an unmarried woman is discovered to be living with a man (Mclntosh, 1984). In other words, a woman who is co-habiting with a man is assumed to be his dependent and, if she wishes to receive the benefit, she must cease to live with her partner. It is not a case of splitting hairs, therefore, to question what is meant by clauses pertaining to the family in various proposed constitutions. Whatever family form is favoured, the question of what actually constitutes 'the family' needs to be addressed by those concerned with reform. TRANSFORMATION 20(1992) 65 ARTICLE SEGAR & WHITE Farm Sizes and Peasant Households The interplay between economic factors, state policy and family form is often well illustrated in the countryside. In many societies it seems that the size of farms influences the size of the peasant household and that where this is made possible by state policy or by flexible tenancies, peasant households will attempt to create an optimum relationship between household size and fanning area For example, Shanin (1971) writing about the pre-1917 Russian peasantry, says that marriage was the essential pie-requisite for a man to obtain full adult status and signalled the beginning of a new household with its own farm. The farm of newly-weds would be small but, as the growth of the family created pressures on consumption, so the head of the new household would try to expand his farm income by acquiring additional rented or purchased land. Prior to the reforms of 1906, the policy of the Russian state supported the stability and cohesion of the peasant family by imposing on it collective responsibility for the payment of taxes and the good behaviour of the household members. The state also upheld the wide disciplinary powers of the household head. This devolution of authority combined with the fact that the peasant 'family' was held together not by blood ties but by participation in the household, by virtue of sharing in the labour and eating from a common pot, sustained an inclusive and extended family form. Similarly, according to McDonald (1968) in parts of Emilia-Romagna, the cultivators' extended, patrilocal families were wont to pool their properties and appoint a single administrator (and thus) they avoided the uneconomic dispersion of plots between separate heirs. In other parts of Italy, the 1950 Land Reform attempted to put a stop to the fragmentation caused by the equal inheritance of landed property required under the Napoleonic Code. It did this by insisting that beneficiaries of the Reform appoint only one heir, but it failed to deal with the question of compensation for the other children. It also, by instituting owned farms of fixed size, destabilized family structures, because at times in the developmental cycle of the family, the new farms were too small for the family and at others too big. In the past, in central Italy, where a large number of farms of various sizes were available for rent from private landlords, families could move to a rented farm which suited their size at that time. The Italian experience of land reform has been that despite it being expressly forbidden, hiring of labour takes place, as does renting out of excess land in order to try to balance out farm size with the family labour force and consumption requirements. The farm sizes were legislated on the basis of a combination of the land available, the demand for land and a notional family size. The consequences have been a less efficient agricultural system and the forced departure of family members from the land when compared to a system where family size and farm size could be adjusted to one another. In southern Africa, the practice of migrant labour in combination with various aspects of customary law has far-reaching consequences for the status of rural women and the composition of rural households. Colin Murray (1981:chap 7) examines how migrant labour affects women in Lesotho but his findings are also 66 TRANSFORMATION 20(1992) SEGAR & WHITE ARTICLE applicable to other rural areas in southern Africa. He points out that in keeping with customary law, Basotho men dominate women and that even when they are away for 11 months of the year working on mine contracts, they have jural control over their rural homes. Men in Lesotho say that'women are weak, they lack sense'. They are usually not mentioned in genealogies and are regarded as unimportant in perpetuating their own agnatic family (ie the patrilineage). Yet it is the women who keep rural households together as full-time managers, whilst men are for large parts of their lives, visitors and strangers. Running a rural household often entails taking on responsibility for the cultivation of land - making decisions about ploughing, share cropping, labour on the field(s). This is in addition to running and maintaining a homestead, including effecting structural maintenance; caring for children - clothing and feeding them and paying school fees. Usually the only cash income is derived from migrant labour remittan- ces, but women not only have no control over the regularity of such remittances, but often they have no control over how they are to be spent. A similar picture is painted by Cloete (1988) in discussing the lives of rural Transkeian women. She describes the common practice whereby an absent migrant appoints a kinsman to be usipathileni or caretaker in his absence. These men are supposed to take responsibility for the absentee's household and advise the migrant's wife. Cloete observed that caretakers have an important role in so far as they represent households at meetings which only men are allowed to attend (such as meetings about planned resettlement in the area) and they perform rituals on behalf of the household which can only be performed by men. However, it seems that when it comes to manual labour and the day to day work of running a homestead, the women are on their own. Thus absent migrants maintain control over their rural households both through remittances and the presence of caretakers. Cloete's informants were rather disenchanted with their 'caretakers' noting that they con- tributed little in the way of practical assistance and that despite the presence of these men, women still had to perform men's work such as ploughing fields and repairing fences. Neither Murray nor Cloete suggest that women are the helpless victims of men. On the contrary, both portray rural women as hard working, spirited and resourceful. Yet such women face enormous structural disadvantages which Cloete sums up by pointing out that 'by controlling community politics (which "makes the rules") and ritual (which "enforces the rules") as well as the economy... men have the ultimate power in the community' (Cloete, 1988:56). The pattern of de jure women-headed households is not confined to southern African rural areas, but is an increasing phenomenon world wide. These households tend to be amongst the poorest, lacking in material resources and labour. In a discussion of women and agricultural development, Cloud (1985) highlights the need to examine the internal workings of the rural household to determine who makes decisions, who allocates resources and who carries out the work. Her concerns flow from a steadily growing body of research which shows how rural women are effectively excluded from rural development'... there is substantial evidence that TRANSFORMATION 20(1992) 67 ARTICLE SEGAR & WHITE in many rural systems women have less access to land, capital, credit, technology, wage markets and training than men in the same system' (Cloud, 1985:38). To a large degree this situation can be traced to understandings of what constitutes 'the family' and the rights of various family members. It can be argued that rural women carry a particularly heavy burden of oppression, the roots of which can be traced both to pre-capitalist 'chiefly patriarchy' (see Bozzoli, 1983:149-155) as well as to the recent colonial past For instance, when titles to lands have been formalised, women have often lost rights as government representatives (followed in turn by development agents) have assumed that male household heads own the lands. Many social scientists have argued that colonial administrators misunderstood and dis- torted pre-colonial concepts of land tenure and that individual land ownership was a European import In addition, male colonial officials made suppositions about the nature of the household and the role of women, assuming that the 'nuclear family' with a male head was 'normal' and that men controlled and owned land. It seems that a number of chiefs and men have always been willing to go along with such notions if it has been in their personal interests to do so (cf Boserup, 1970; Davison, 1988; Ranger, 1983; Rogers, 1980). Often the 'traditional' role of chieftaincy has been used to justify claims to private ownership of land and it has been pointed out that 'the "typical colonial trend" of individualizing land tenure continues under the present governments, "trained in the spirit of the colonial powers'" (Rogers, 1980:139). Davison (1988) traces this process in Kenya noting that whilst women currently produce 80% of food crops, they own only 5% of the land and points out that whilst much has been written about the impact of colonial land policies 'much less has been said about the consequences of such policies on women's access to and use of land' (1988:164). In Tanzania women remain the principal producers of subsistence and cash crops, whilst access to and control of land remains in the hands of their husbands and so the fruits of their labours accrue to male household heads (Davison, 1988:9). Even in areas where there existed a high rate of absenteeism due to migrant labour and where women very obviously shouldered the bulk of agricultural work, colonial authorities allocated land to men. In 1957 the Rhodesian authorities completed land reforms which resulted in land being allocated only to men and widows. This was despite the fact that 23% of the recipients of land were absentee migrants. Boserup points out that this legislation made women particularly vulnerable as many lost access to land upon the break-up of their marriages. She cites a similar transfer of land rights from women to men in Transkei in 1898 (Boserup, 1970:61). The 'land issue' in South Africa has been on the agenda for debate for some years. However, women's rights over land, as yet do not form part of this debate. It has been argued that in South Africa it makes little difference whether women have jural rights over land, as they will probably continue to do the bulk of agricultural work anyway. Such a view is extremely myopic and presupposes that the current appalling state of 'homeland' agriculture will never improve. Given any improvement in small-scale agriculture, rights to land will become all important 68 TRANSFORMATION 20(1992) SEGAR& WHITE ARTICLE Some Areas for Research and Action Bozzoli (1983) suggests that an examination of what she calls 'domestic struggles' is crucial to our understanding of the continuing subordination of women in South African society. She identifies 'domestic struggle' as taking place both within the domestic sphere as well as between the domestic system and the broader society. This type of approach encourages us to look at what goes on within households arid to recognise the importance of women's degree of control (or lack of control) over such things as labour, property, income and their own sexuality (as well as allowing us to examine more conventional areas such as the role of the rural household in the migrant labour system). Bozzoli points out that - ...struggle between household members may take place over any of the following issues: family income; home-ownership; access to the labour of children in domestic and non-domestic work; distribution of household labour, use of non-kin for domestic purposes; ownership and control of other household property; control over upbringing and future employment of children; family investment; extended family connections; access to and use of physical violence - and so on (Bozzoli 1983:147). Whilst Cloud points to the lack of data on intra-household decision-making in the literature on farm management she also finds that 'by contrast, anthropological studies have consistently documented the patterns of women's responsibilities for management of their own animals and crops' (1985:30). It is precisely in this way that anthropologists can make a strong and informed contribution to debates about 'the family' in South Africa. For instance, Sharp and Spiegel (1990) show very clearly how the nature and degree of women's dependence on male wage labour differs markedly from one context to another (namely Matatiele and Qwaqwa). Differentials in gender relations within the household are traced to control over income, labour, arable land and produce. They show how the key variable which gives women opportunities for relative autonomy is the availability of these resour- ces and conclude by pointing out the effect of differential access to resources on the marriage patterns of young women. • Wages In Britain the notion of the 'family wage' was used both to exclude women from the workplace and from higher paid jobs, and to bargain successfully for higher pay for men (Cockburn, 1986). In South Africa it has been argued by Wolpe and others that the existence of the bantustans has been in the interests of capital because access to land and the labour of women and children on the land in the rural areas served to perpetuate the pretence that capital did not need to pay a family wage. Although one would certainly support the efforts of workers to gain a more equitable wage, there is a danger of wage demands being based on the 'family wage' argument once it is more widely accepted that the 'homelands' rarely provide a living for those members of the family left behind by the migrant labour system, and as more of the male workers' families settle in the cities. The danger of the 'family wage' argument is that it assumes that women and youths are necessarily dependents TRANSFORMATION 20(1992) 69 ARTICLE SEGAR & WHITE of a male 'breadwinner' and that they do not therefore need to be paid either the rate for the job or a wage that will support them regardless of their domestic circumstan- ces. This is important for three reasons. Firstly, wages are not always equitably shared amongst the members of the household (Whitehead, 1981). Secondly, the cir- cumstances, composition and behaviour of household members is subject to change. Couples divorce, children leave home, 'breadwinners' cease being able to work or change their minds about handing over their wages to the wife. Thirdly, the notion that women are 'dependents' of male breadwinners has served to depress wages in those traditionally associated with women. Women and young people should not be dependent on circumstances and 'should be able to earn a wage that does not assume they live as a dependent in someone else's household* (Barrett and Mclntosh, 1982:148). This will require positive action to ensure that women really do enjoy equal opportunities with men to earn in exactly the same way with the same perks as men. As Barrett and Mclntosh point out, men 'will lose some privileges' but they will also win parenting rights and freedom from the responsibilities of the breadwinner role. • Social Security Provision has to be made for those members of the population who are not able to support themselves. Children and the very old cannot be employed, and as long as the present burden of child rearing continues to weigh on women, they will be more often out of paid work than are men. In order not to disadvantage women nor privilege the family form, whatever benefits are paid should not be connected to the type of household in which the needy individual lives, nor based on the assumption that women will have men to support them. However, in the light of discussion earlier in this paper, it becomes obvious that social security and welfare programmes are often contentious issues with the potential for becoming highly politicised. It becomes clear that the future state needs to have a clear vision of what kind of family form it wishes to encourage in the 'new South Africa'. As it stands, current welfare policy assumes a male-headed nuclear family to be the desirable norm. ANC constitutional proposals which affirm the rights of women, clearly challenge such a view. The reality of growing numbers of women-headed families both in urban and rural areas suggests a new more flexible view of the family is needed. Yet a laissez-faire approach to the family will not necessarily be a prudent course. If a new government were to take seriously the position of women as heads of families, then current welfare policies need to be examined and overhauled. An approach that simply recognises the legitimacy of a number of family forms leads to the extension of welfare benefits to a much wider population than currently receives them. It also has serious implications for housing provision (see below) and possible tax concessions. On the other hand, a more positive or pro-active stance towards women-headed families leads to the consideration of a host of provisions such as: day-care facilities for children; the institutionalisation of maternity benefits; incentives for employers 70 TRANSFORMATION 20(1992) SEGAR & WHITE ARTICLE to provide such benefits; tax breaks for the single parent family; education and training for women to give them greater access to resources; the improvement and enforcement of child maintenance laws (such as the current system of gamishee orders on fathers' wages). At present evidence from Africa and the USA shows that women-headed families - both urban and rural - often constitute the poorest in their communities. The question of what can be afforded in the 'new South Africa' must necessarily be part of deliberations on the kind of family form to be encouraged and supported in the future. • Housing and Land Policy In South Africa working class housing has traditionally been rented. Very often the tenancy agreement was made with the male household head and women who wished to leave their husbands or who wererejectedby them often found themselves with no rights to accommodation (Burman, 1984). Even where the only accommoda- tion rights available are to a bed in a single men's hostel, these rights are the preserve of men, and women have to cast themselves as the dependents of the male bedholder in order to have a roof over their heads. Ramphele has shown that dependence on men for even so tenuous a commodity can expose women to abuse (Ramphele and Boonzaier, 1988). Since the first reforming efforts of the Urban Foundation in the arena of housing, home ownership has tended to be seen as the panacea for the housing shortage, squatting and homelessness. The government and local housing authorities have followed this lead in the precipitate rush to sell township housing to existing tenants. Home ownership has its advantages. It gives security of tenure that cannot be equalled by any rental regime - as long as the purchaser is able to keep up the mortgage repayments. But Urban Foundation figures show that the majority of those needing homes in South Africa cannot afford to buy even the most cheaply con- structed house (White, 1984). It also gives greater control to the resident than does renting accommodation. However, as Barrett and Mclntosh (1982) point out, 'it tends to give that control to the man in the household, the one who can get a mortgage and afford to make the repayments'. As they also point out, where there is an acute housing shortage (as in South Africa), the security of tenure only holds as long as a person remains in the same household group. Wives who leave their husbands are extremely unlikely to be able to afford to purchase and are at a disadvantage in obtaining bonds (Burman, 1984). It would be much better to have plenty of rental accommodation of varying sizes available to allow for flexibility in family and household composition. In the rural areas the formal extension to women of rights over land would seem to be an essential prerequisite for addressing the problem of women's dependence on men. Cloud and Knowles (1988) suggest that there are specific areas which are of particular importance to African women and where the extension of legal rights can bring about real and positive changes to their lives. They identify specifically control over property and landholding and call for rapid and substantial strengthen- TRANSFORMATION 20(1992) 71 ARTICLE SEGAR & WHITE ing of these rights. To this must be added a call for the careful reconsideration of the laws that govern marriage and the custody of children upon the dissolution of marriage. It should be noted that often a woman's marital status is closely linked to her eligibility for land and property rights. In 1979, the FAO World Conference of Agrarian Reform and Rural Development recommended that countries: Repeal those laws which discriminate against women in respect to rights to inheritance, ownership and control of property... promote ownership and co-ownership of land to effectively give women with absentee husbands the legal right to make decisions on the land they manage (Section IV, Ai4i) (in Cloud, 1985:39). Conclusion Upon reflection, the fact that family form and state policy are inextricably interlinked should not be surprising. However, this is contrary to the commonly held notion mat the family and the form it takes is part of the private rather than the public arena We must not forget that the intimacies of marriage, child rearing and home building and the choices that confront the individual are shaped and restricted by public policy. At a time of transition in South Africa it is vital that these issues should be recognised as being of national importance. The state's understanding of what constitutes the family affects everyone, but we would suggest that there issues have particular relevance for women. Those who fall outside the existing stereotypes of the family may have difficulties in acquiring state assistance, housing, land rights, etc. The fastest growing family form both inside and outside South Africa is the single woman-headed family and these women and their families are often in particularly vulnerable positions within society. A variety of state policies have far-reaching implications for families - ranging from the forms that they are encouraged to take, to the benefits they may be granted or denied. The material discussed in this paper shows mat questions about the family are by no means unique to South Africa. Indeed we have chosen our examples from a number of countries and historical periods to show that family matters and state policy go hand in hand and that these issues are not simply the prerogatives of either industrialised or 'developing' countries. Most of our examples are based on the work of anthropologists who, in our opinion, have a definite contribution to make to debates concerning family policy in the 'new' South Africa. Anthropologists are already engaged in documenting the rise of women-headed families and various other family forms, and are also particularly sensitive to the degree of differentiation within and between households. Anthropology has much to offer here given the discipline's long history of expertise within the fields of kinship and gender studies. Questions of family policy and the position of women have, so far, not moved far from the realm of rhetoric in South Africa 72 TRANSFORMATION 20(1992) SEGAR & WHITE ARTICLE NOTES 1. A parent's grant is a welfare grant paid to the parent herself for her own maintenance. According to Burman, 'White women generally received the parent's allowance... whereas Coloured and Indian women often did not, and Black women very seldom if ever did' (1988:518). 2. Even Wolpe's critics do not, in the main, disagree with his argument about the depressive effect of wages. 3. Marriage can be contracted under two kinds of law in South Africa - civil law or customary law and under the latter women remain perpetual minors, have no contractual capacity and lose the custody of their children upon the dissolution of the marriage. REFERENCES African National Congress (1990) - 'A Bill of Rights for a democratic South Africa' (Centre for Development Studies, University of the Western Cape). Aldous, J and W Dumon (1990) - 'Family policy in the 1990s: controversy and consensus', in Journal of Marriage and the Family, 52. Annual Survey of South African Law 1988 (1989) (School of Law, University of the Witwatersrand; Cape Town: Juta& Co). Barrett, M and M Mclntosh (1982) - The Anti-Social Family (London: Verso). Boserup, E (1970) - Woman's Role in Economic Development (New York: St Martin's Press). Bozzoli, B (1983) - 'Marxism, feminism and South African studies', in Journal of Southern African Studies, 9(2). Burrnan, S (1984) - 'Divorce and the disadvantaged: African women in urban South Africa', in R Hirschon (ed) - Women and Property - Women as Property (London: Croom Helm). Burrnan, S (1988) - 'Maintaining the one-parent support in South Africa: law and reality', in MT Meulders-Klein and J Eekelaar (eds) - Family, State and Individual Economic Security (Brussels: Story Scientia). Charman, A, C de S wardt, M Simons (1991) - "The politics of gender, a discussion of the Malibongwe Conference Papers and other current papers within the A N C (Paper 30, Women and Gender in Southern Africa Conference/University of Natal, Durban). Cloete, L (1988) - 'Domestic strategies of rural Transkeian women' (unpublished Honours thesis, Rhodes University). Cloud, K and JB Knowles (1988) - 'Where can we go from here?: recommendations for action', in J Davison (ed) - Agriculture, Women, and Land: the African experience (Boulder and London: Westview Press). Cloud, K (1985) - 'Women's productivity in agricultural systems: considerations for project design', in C Oberholt et al(eds) - Gender Roles in Development: a casebook (West Hertford, Conn: Kumarian Press). Cockbum, C (1986) - The Machinery of Male Dominance (London: Pluto Press). Davison, J (1988) - 'Land and women's agricultural production: the context', in J Davison (ed) - Agriculture, Women, and Land: the African experience (Boulder and London: Westview Press). Davison, J (1988) - 'Who owns what?: land registration and tensions in gender relations of production in Kenya', in J Davison (ed) - Agriculture, Women, and Land: the African experience (Boulder and London: Westview Press). Driver, D (1989) - 'Women and language in the ANC Constitutional Guidelines for a Democratic South Africa' (paper presented at IDAS A Women's Constitutional Conference). Ellwood, DT (1988) - Poor Support: poverty in the American family (New York: Basic Books). Erie, SP and M Rein (1988) - "Women and the Welfare State' in CM Mueller (ed) - The Politics of the Gender Gap: the social construction of political influence (Newburg Park: Sage). Ginwala, F, M Mackintosh and D Massey (1990) - 'Gender and economic policy "> a democratic South Africa' (paper presented at the Lawyers for Human Rights Conference: Putting Women on the Agenda, University of the Witwatersrand). McDonald, JS (1956) - 'Italy's rural social structure and emigration', in Occidente, 12. Mclntosh, M (1984) - 'The family, regulation and the public sphere' in G McLennan (ed) - State and Society in Contemporary Britain (Oxford: Polity Press). Moore, H (1989) - Feminism and Anthropology (Cambridge: Polity Press). Murray, C (1981) - Families Divided: the impact of migrant labour in Lesotho (Johannesburg: Ravan). Ramphele, M and E Boonzaier (1988) - "The position of African women: race and gender in South Africa', in E Boonzaier and J Sharp (eds) - South African Keywords: the uses and abuses of political concepts (Cape Town: David Philip). Ranger, T (1983) - 'The invention of tradition in colonial Africa', in E Hqbsbawm & T Ranger (eds) - The Invention of Tradition (Cambridge: Cambridge University Press). Rogers, B (1980) - The Domestication of Women: discrimination in developing societies (London: Kogan Page). TRANSFORMATION 20(1992) 73 ARTICLE SEGAR & WHITE Rosaldo, MZ and L Lamphere (eds) - Woman, Culture and Society (Stanford: Stanford University Press). Sachs, A (1990) - 'The family in a democratic South Africa: its constitutional position', in Agenda, 8. Segar, J and C White (1989) - 'Constructing gender discrimination and the law in South Africa', in Agenda, 4. Shanin, T (1971) - Peasants and Peasant Societies (Harmondsworth: Penguin). Sharp, J and A Spiegel (1990) - 'Women and wages: gender and the control of income in farm and bantustan households', in Journal of Southern African Studies, 16(3). Silverman, S (1971) - 'Italian land reform', in Anthropological Quarterly, 11. Stack, C (1974) - 'Sex roles and survival strategies in an urban black community', in MZ Rosaldo and L Lamphere (eds) - Woman, Culture and Society (Stanford: Stanford University Press). Van der Vh'et, V (1982) - 'Black marriage: expectations and aspirations in an urban environment' (unpublished MA dissertation. University of the Witwatersrand). Whitehead, A (1981) - '"I'm Hungry Mum": the politics of domestic budgeting', in K Young, C Wolkowitz and R McCullagh (eds) - Of Marriage and the Market: women's subordination in international perspective (London: CSE Books). 74 TRANSFORMATION 20 (1992)