Matrilinypdf

Early studies of matrilineal kinship were published in the structural-functionalist vein, when descent and lineage were key features of kinship studies and taken for granted as central organising principles. These studies include a paper by Audrey Richards (1950), an edited volume titled Matrilineal kinship (Schneider & Gough 1961), and a later contribution by Mary Douglas, printed in 1969, by which point structural-functionalist approaches were under attack. Following a lull, interest in matriliny was revived in the 1970s and 1980s as feminist anthropologists began to ask questions about gender construction, women’s roles, and family forms in different societies (see, e.g., Holy 1986; Lancaster 1974, 1976, 1981; Poewe 1981). Since then, occasional interest has been paid to matriliny within regional literatures (see, e.g., Atkinson & Errington 1990; Davison 1997; Flinn 1986). Indeed, examples of matrilineal kinship can be found on almost every continent: in parts of Asia (see, e.g., Agarwal 1994; Kelkar, Nathan & Walter 2003; Tanner & Thomas 1985), Melanesia (see, e.g., Battaglia 1990; Bolyanatz 2000; Weiner 1976, 1992), West Africa (see, e.g., Fortes 1969; Goody 1959; Oppong 1974), and the Americas (see, e.g., Brown 1975; Lévi-Strauss 1963; MacLeitch 2011; Maybury-Lewis 1979). Much of the focus of this entry, however, will be on what has been called south-central Africa’s ‘matrilineal belt’, which extends across western Congo, northern Zambia, central and southern Malawi, and northern Mozambique. It is here, specifically in Malawi, that Peters herself has worked, and from where she makes the case for reopening debates about matriliny in anthropology; it was also the locus of much of the classic anthropological work on the topic.

What is matriliny?

Matriliny is a way of reckoning descent and belonging through the female line. It is generally contrasted with patriliny, where descent and belonging are traced through the male line. Matriliny and patriliny are both referred to as unilineal kinship systems, since in both cases descent and belonging are traced through a single line. Broadly speaking, in European contexts, kinship descent and belonging are reckoned cognatically (bi-laterally), by looking to what might be called ‘both sides’ of a family. To take myself as an example, when I think about my own family and belonging, I trace it out in two directions, through my mother and my father. That gives me two parents who are of equal significance, and four grandparents (I am talking in abstract terms here, so I am leaving aside any divorces and re-marriages that may well complicate this picture, as well as the behaviour of the incumbents of these roles that might qualify them as closer or more distant in social terms, but not in terms of their position in my family tree). In abstract terms I wouldn’t distinguish between my two grandfathers or my two grandmothers as more or less related to me, and nor would I consider myself any more or less the heir of my mother or father in the case of inheritance. Similarly for my various aunts, uncles, cousins, nephews and nieces (and so on), I reckon my relatedness to them through both ‘sides’ of my family, and consider none of them more or less closely related to me on the basis that we are related through one or the other of my parents or siblings. This is obviously a simplification of actual family forms, but my point here is that it can be helpful to reflect on our own assumptions about kinship and family belonging in order to highlight what it is that is different when we look at other ways of reckoning descent and belonging.

To return to matrilineal kinship, I have said that descent and belonging are traced through the female line. This means that you trace your belonging to a larger family unit through what might otherwise be considered to be ‘one side’ of your family. Your mother, her mother, and their respective siblings are key figures in your matrilineage. While your father and his parents and siblings may play important roles in your life, they do not belong to the same matrilineage as you do. This does not mean that you are not considered ‘related’ in some way, but simply that you belong to different groups. Matrilineal belonging may affect such things as who would be involved in arrangements for your marriage, from whom you might claim access to land or resources, where you might consider your true ‘home’ to be, and where you would expect your body to be buried upon your death. David Schneider encapsulates unilineal kinship well when he explains that in a patrilineal kinship system, you are ‘related to females but not through females’ (1961: 3, original emphasis). In matrilineal kinship it is the other way around: you are related to both male and female kin, but matrilineal kinship only continues through females. Fathers thus belong to different lineages from their children, and if they hold property or titles such as chieftaincies, their likely heirs are the children of their sisters (who belong to the same matrilineage as they do). Thus, the relationship between a mother’s brother and his sister’s children is particularly significant.

Importantly, matrilineal societies are not all the same in every respect. Peters thus argues that it is more ‘useful to consider matriliny as a set of characteristics rather than a totality or “system”’ (1997a: 137). In this vein, Edmund Leach posited that for anthropologists the category of ‘matrilineal societies’ is about as useful as the classification of ‘blue butterflies’ is to biologists (1961: 4). We will come back to the issue of conceiving of kinship systems as bounded totalities or coherent types later, but here I want to stress the diversity of norms and practices that can be subsumed under the category of matriliny. With respect to inheritance, for example, in some cases only women hold agricultural land and they pass it on to their daughters (and not their sons); in other places men may hold property and pass it on to their sisters’ sons (i.e. their maternal nephews, but perhaps not their maternal nieces); elsewhere, both men and women may hold land and property, in which case male and female children might inherit from both their mothers and their mothers’ brothers. Quite commonly, different kinds of property and titles may be inherited in different ways, so a young woman may inherit land from her mother, for instance, while her brother inherits a chieftaincy from their maternal uncle. And clearly, such norms and practices are subject to historical change.

Matrilineal societies also differ with respect to residence patterns. The designation ‘matrilocal’ or ‘uxorilocal’ refers to post-marriage residence patterns in which men move to live with their wives upon marriage. ‘Patrilocal’ and ‘virilocal’ are terms referring to residence patterns in which wives join their husbands upon marriage. Matrilocality (uxorilocality) is often, but not always, associated with matrilineal kinship, though some degree of variation in residence patterns ought to be expected in any given locale.

Having looked at what matriliny is, I will now turn to how matriliny has been understood, and indeed misunderstood, by anthropologists since the mid-twentieth century.

Anthropological (mis)understandings of matriliny

1. Matriliny is male authority (patriarchy) in another guise

One might imagine that if matrilineal societies are distinguished by the transmission of descent and belonging through women, then women in matrilineal societies might enjoy considerable authority. However, this has not been the dominant interpretation in anthropological writings on matriliny. The view that matriliny is, at its core, a different configuration of male authority is associated with David Schneider and Kathleen Gough, and strongly articulated in their edited volume Matrilineal kinship (1961). In the introduction to the volume, Schneider writes:

The role of women as women has been defined as that of responsibility for the care of children ... the role of men as men is defined as that of having authority over women and children … Positions of highest authority within the matrilineal decent group will, therefore, ordinarily be vested in statuses occupied by men (1961: 6).

Schneider effectively reassures his readers that matrilineal societies do not entail significant female authority, it is simply that male authority rests in the hands of men as brothers and uncles, as opposed to men in their roles as fathers and husbands.

Such understandings share much with the earlier work of Claude Lévi-Strauss. In his book The elementary structures of kinship (1969), originally published in French in 1949, Lévi-Strauss places the exchange of women by men at the heart of what kinship means. Marriage is the key institution, because it is by means of marriage that men exchange women and thereby form alliances. Lévi-Strauss wrote:

The total relationship of exchange which constitutes marriage is not established between a man and a woman, where each owes and receives something, but between two groups of men, and the woman figures only as one of the objects in the exchange, not as one of the partners between whom the exchange takes place (1969: 115).

If this is how marriage works, and how societies are held together, matriliny, and especially matrilineal kinship systems that favour matrilocal residence, could be seen as posing a fundamental threat to understandings of kinship, marriage, and alliance. Lévi-Strauss tackled this problem directly in his writing, insisting that: ‘the number of matrilineal systems which are also matrilocal is extremely small. Behind the variations in type of descent, the permanence of patrilocal residence attests to the basic asymmetrical relationship between the sexes which is characteristic of human society’ (1969: 117).

Despite the authoritative tone, however, it is not in fact the case that matrilocality is rare among matrilineal groups, and nor has the idea of necessary and universal male authority stood the test of time (see, e.g., Amadiume 1987; Butler 1990; MacCormack & Strathern 1980; Oyěwùmí 1997). It is worth noting that early understandings of matriliny in this mode were somewhat haunted by a concern with whether matriliny meant matriarchy, or rule by women. This anxiety stemmed largely from nineteenth- and early twentieth century evolutionary thinking which had suggested that matriarchy represented an early stage in social evolution. Peters argues that such fears led anthropologists to downplay the implications of matriliny for women (1997a: 133). Since the 1970s, gender relations have been looked at in a more nuanced fashion, and scholars have returned to questions about matrilineal kinship. Importantly, this work has revealed significant scope for women’s authority in matrilineal settings (see, e.g., Arnfred 2011).

2. Matriliny is puzzling

The phrase ‘the matrilineal puzzle’ comes from Audrey Richards’ contribution to the edited volume African systems of kinship and marriage (1950). Richards focused on the issue of exogamous marriage, i.e. the requirement that lineage members marry outside of their own group. As we have seen, one consequence of exogamous marriage is that men in matrilineal societies marry women from other matrilineages and thus fathers do not belong to the same lineages as the children born of their marriages. In Richards’ words:

The problem … is the difficulty of combining recognition of descent through a woman with the rule of exogamous marriage. Descent is reckoned through the mother, but by the rule of exogamy a woman who has to produce children for her matrikin must marry a man from another group. If she leaves her own group to join that of her husband, her matrikin have to contrive … to keep control of the children … The brothers must divide authority with the husband who is living elsewhere. If, on the other hand, the woman remains with her parents and her husband joins her there, she and her children remain under the control of her family, but her brothers are lost to the group since they marry brides elsewhere and they are separated from the village where they have rights of succession (1950: 246).

The problem for Richards then – and the heart of the so-called matrilineal puzzle – was that men were torn between their roles as fathers and as mothers’ brothers. By contrast, the assumption was that there was no such contradiction in patrilineal kinship where the father had authority over his children both as their father and as their lineage elder. Women’s position within patrilineal groups was given much less consideration than men’s position in matrilineal groups, seemingly because women’s authority was not of concern, and because they were thought to integrate more easily with their husbands’ kin, shedding their attachments to their own patrilineages (i.e. their fathers’ descent groups). There was never any suggestion of a ‘patrilineal puzzle’ that would have addressed the contradictions for women entailed in their dual roles as members of their fathers’ patrilineages and as mothers to children who belonged to their husbands’ patrilineages.

Why was the matrilineal puzzle so puzzling? One reason, Peters argues, is that for anthropologists influenced by the structural-functionalist tradition, it was important to identify bounded social structures, so anything that seemed ‘to divide a person’s identification with one group was assumed to create problems’ (1997a: 128). In the period since the heyday of structural functionalism, however, kinship has come to be understood in more flexible terms, open to considerable variation even within a single society. Although early writers did recognise variation between matrilineal societies, their will to construct typologies, and their concern with bounded groups, encouraged a view of matriliny as a totality, rather than a variable set of characteristics.

Another factor that made the matrilineal puzzle so puzzling was the implicit assumption that the nuclear family, based on the marital relationship, constitutes the essential building block of society. As we have seen, this premise in turn relates to ‘the privileging of the male’ (Peters 1997a: 128). Thus, Audrey Richards suggested that matriliny posed a problem for the ‘sentiment attaching father to son’ (1934: 277), and invoked Bronislaw Malinowski’s work on matrilineal Trobriand Islanders, who were said to face a conflict between their legal responsibilities towards their maternal nephews and their ‘natural desire’ (1934) to favour their own sons (see Malinowski 2002 [1922]: 71–2). Strikingly, this focus on the marital relationship and the father–child bond endured despite the fact that ethnographers repeatedly noted the relative fragility of marriages in matrilineal settings, where divorce rates were often high, as well as the greater significance of relationships between siblings and between children and their maternal uncles. Alternative central foci for kinship studies, such as relationships between same or opposite sex siblings, were not given much attention. It can thus be argued that the emphasis on descent in structural-functionalist anthropology obscured the significance of siblingship for the ways in which matrilineal kinship bonds were understood and valued. This is something that anthropologists working in matrilineal areas of Southeast Asia have also pointed out; see, for example, Peletz (1988).

The above discussion suggests the significance of the assumptions brought to bear on studies of matrilineal societies. There is a pertinent quotation from an Ashanti male elder in Ghana, who, when asked by R.S. Rattray (a colonial official and anthropologist) why he had not been aware of women’s significance in Ashanti political life, replied: ‘The white man never asked us this; you have dealings with and recognise only the men; we supposed the European considered women of no account, and we know you do not recognise them as we have always done’ (1923: 84; cited by Peters 1997a: 135). There are countless other examples of early observers, more familiar with patrilineal or cognatic kinship organisation, struggling to understand matrilineal gender relations (see, e.g., Colson 1958; Mitchell 1956, 1959 [1951]; Read 1942; Rowley 1867). Surveying a number of them, Peters concludes: ‘the matrilineal puzzle was not in fact that at all but a gender puzzle’ (1997a: 141): a gender puzzle precipitated by the seeming incomprehensibility of kinship norms and practices that gave ‘greater social and political space to women’ (1997a: 133).

3. Matriliny is doomed

The question ‘is matriliny doomed?’ comes from the title of Mary Douglas’ (1971 [1969]) article in which she attempts to assess what she calls matriliny’s ‘prospects for survival in the modern world’ (1971[1969]: 123). The idea that matriliny might be doomed emerged largely from a sense that it was fragile in the face of economic change. Douglas cites several authors who argued that ‘power, property, and prestige spell doom to the matrilocal principle’ (Murdock 1949: 206-7; cited by Douglas 1971 [1969]: 123) and matrilineal descent, including the work of Jack Goody, who had suggested that ‘disparity of incomes weakens the principle of matrilineal descent’ (Douglas 1971 [1969]: 124).^[1]

Indeed, over the years a range of voices has predicted matriliny’s demise. In the face of growing market economies, increasing differentiation in wealth, and the acquisition of more significant personal property, anthropologists, and other observers, expected matrilineal kinship norms to weaken and a shift towards patrilineal or cognatic kinship organisation to occur.^[2] The idea was that men would strive to overcome the ‘matrilineal puzzle’ by avoiding matrilocal marriage and associated bride service (the work they might be expected to do for their wives’ kin), and favouring bridewealth (involving payment(s) on the husband’s behalf to his wife’s kin so as to secure her residence and rights to their children), all of which would enable them to favour their own sons as heirs rather than their sisters’ sons.

In this vein, Kathleen Gough titled a chapter for the 1961 volume Matrilineal kinship: ‘The modern disintegration of matrilineal descent groups’. Gough lends support to the view that matriliny is vulnerable in the face of economic advancement:

Recent literature has accumulated evidence to show that under economic changes brought about by contact with Western industrial nations, matrilineal descent groups gradually disintegrate. In their place, the elementary family eventually emerges as the key kinship group with respect to residence, economic cooperation, legal responsibility, and socialisation, with a narrow range of interpersonal kinship relationships spreading outward from it bilaterally and linking it with other elementary families … There is … great variation in the degree of change at present experienced both within and between matrilineal societies. Nevertheless, given continued exposure to the same kinds of economic processes, the directions and end products of the change seem to be essentially the same (1961: 631).

Gough was confident that this was a general trend, inevitable given the conditions of global capitalist expansion.

Douglas’ answer to the question ‘is matriliny doomed?’ is slightly different, however. In talking of situations in which matriliny is likely to give way to patriliny, she highlights not wealth inequality but scarcity of resources, arguing that ‘competition in a restricted field causes men to draw in their horns and to concentrate their responsibilities on their nearest kin’, that is, ‘to favour their sons’ (1971 [1969]: 130). It is possible to recognise some of the same assumptions that led to the ‘matrilineal puzzle’ at work in both Douglas’ and Gough’s analyses: namely, the assumption that a man’s closest kin are his children, and that there is something fundamentally salient, even ‘natural’, about the connection between fathers and sons that matrilineal norms of kinship deny.

I have already referred to Peters’ argument that we ought to consider matriliny more as an assemblage of characteristics than a totality or ‘system’. One corollary of that point is that it makes little sense to consider ‘matriliny’ – as a totality – to be more vulnerable to economic change than ‘patriliny’ or ‘cognatic’ kinship, or to consider matriliny and patriliny as bounded systems that come into contact as wholes that must clash and confront one another in such a way that one would inevitably triumph over the other.

Indeed, recent work has demonstrated the limitations that conceptions of kinship ‘systems’ as bounded wholes place on analysis. Thus, Cynthia Brantley (1997), for example, looks back at the published and unpublished work of Margaret Read from the 1930s and 1940s. Read studied patrilineal Ngoni in Northern and Central Nyasaland, now Malawi, before turning her attention to matrilineal Chewa people in neighbouring parts of the same country. Brantley is able to show the effect that Read’s earlier work with patrilineal Ngoni informants had upon the ways in which she conceived of matrilineal kinship organisation. Her article is titled ‘Through Ngoni eyes’, which succinctly sums up the way in which she argues Read was blinkered by her prior experiences.

Read argued that a slow but sure takeover of matriliny by patriliny was occurring in central Malawi as Ngoni norms and practices prevailed over matrilineal Chewa ways. Reassessing Read’s material, Brantley suggests that ‘Read’s evidence, when set in relation to other information, shows a much more complex situation of interaction and modification to mutual benefit … [people] were accommodating and borrowing from the practices of “patriliny” and “matriliny”’ (1997: 165). She concludes that Read was wrong to identify a shift from matrilineal to patrilineal marriage practices. Instead, she points to the development of what were called chitengwa marriages, whereby women moved to live with their husbands in patrilocal marriage, more typical of patrilineal societies, but their children still belonged to their matrilineages. In these marriages men tended to give token gifts to the women’s kin (something relatively new), but they did not pay significant bridewealth as per patrilineal norms. It is such examples of ‘partial accommodation and blending’ (Peters 1997a: 138) that Brantley concludes were invisible ‘through Ngoni eyes’ and, more broadly, were obscured by the pervasive master narrative of matriliny’s inevitable demise.

Contemporary ethnography

Peters has argued for the need to understand how heavily the odds have tended to be stacked against matrilineal norms and practices:

[M]atrilineal groups have been and continue to be a minority in a sea of patriliny and patriarchy. Most groups in all the African regions where matriliny is found follow patrilineal, dual or cognatic descent, and all the major influences entering over the past 200 years and more have been from strongly patrilineal or bilateral, and patriarchal groups. These include, for the Malawi case … Islamic groups, Christian missions, British colonial over-rule and plantation agriculture. The ideas and practices brought by all the former are reinforced in this ‘global’ world because major players (America, Britain, Russia, Japan, China …) all favour patrilineal or bilateral modes of organisation (1997b: 191).

She thus argues that matrilineal norms and practices ought to be seen as ‘remarkably resilient’ (Peters 1997b: 189).

Indeed, while matriliny is not a subject that has received a great deal of attention from anthropologists in recent years, the ethnographic record does allow some insight into the ways in which matrilineal kinship organisation continues to affect the texture of life in post-colonial Africa. On the basis of fieldwork in Malawi carried out in 2009–2010, Jessica Johnson (2012) has written of the significance of matriliny in shaping HIV-positive women’s efforts to rebuild their lives following the roll-out of anti-retroviral therapies. Crucially, Johnson argues that the generally high rates of divorce, and the related social acceptability of female-headed households, coupled with women’s custodianship of land inherited through the female line, profoundly shaped the options available to them as they regained their health and set about re-establishing themselves as productive members of their rural communities. Many HIV-positive women (divorced or widowed) chose, at least temporarily, to remain unmarried and to focus on providing for their children through their agricultural endeavours. They thus drew upon longstanding ‘traditional’ strategies of relative female independence, while explaining their decisions in relation to their desire to safeguard their health and exercise caution as they adjusted to the radically new possibility of living with, as opposed to dying from, HIV/AIDS.

A final example comes from James Ferguson’s Expectations of modernity (1999), a study of the Zambian Copperbelt at a time in which the copper industry had seen enormous decline. Ferguson refers to the 1989 passing of a new inheritance law, which, for the first time, gave the nuclear family legal recognition in matters of inheritance by providing for a man’s wife and children to receive a share of his property. According to Ferguson, the new law was ineffective because the belief that a man’s rightful heirs were his matrilineal relatives – as opposed to his wife and children – meant that the law was largely disregarded. Similarly, pensions awarded to the wives of deceased miners could be a source of tension: wives could be mistreated by their husbands’ matrilineal kin who believed themselves entitled to the payments. As a result, some widows requested that their payments be discontinued, while a number of living men questioned why their wives and children were the automatic beneficiaries and not their sisters’ children. For at least some workers, Ferguson points out, ‘it was not obvious that one’s primary attachment and responsibility was to one’s “own” children’ (1999: 185). He thus refers directly to Mary Douglas’ question: ‘Is matriliny doomed in Africa?”, concluding that apparently, the answer ‘is no’ (Ferguson 1999: 185).

In conclusion, this entry has introduced matriliny as an important topic in the anthropology of kinship. We have come to understand matriliny as a unilineal system of reckoning descent and belonging through the female line, and learnt that the category ‘matriliny’ contains within it significant variation in terms of, for example, patterns of residence and inheritance. The entry has surveyed three dominant anthropological approaches to the understanding of matriliny: namely, that it is simply an alternative configuration of male authority, that it is inherently puzzling, and that it is doomed. I have suggested that all three approaches are problematic, containing within them a number of unfounded, gendered assumptions. Two more recent ethnographic examples from Malawi and Zambia have signalled the ongoing significance of matrilineal norms and practices, as well as the relevance of kinship organisation to other aspects of social life, from health and wellbeing to property and pensions.

References

Agarwal, B. 1994. A field of one’s own: gender and land rights in South Asia. Cambridge: University Press.

Amadiume, I. 1987. Male daughters, female husbands. London: Zed Books.

Arnfred, S. 2011. Sexuality and gender politics in Mozambique: rethinking gender in Africa. Woodbridge: James Currey.

Atkinson, J.M. & S. Errington (eds) 1990. Power and difference: gender in Island Southeast Asia. Stanford: University Press.

Battaglia, D. 1990. On the bones of the serpent: person, memory, and mortality in Sabarl island society. Chicago: University Press.

Bolyanatz, A.H. 2000. Mortuary feasting on New Ireland: the activation of matriliny among the Sursurunga. London: Bergin & Garvey.

Brantley, C. 1997. Through Ngoni eyes: Margaret Read’s matrilineal interpretations from Nyasaland. Critique of Anthropology 17, 147-69.

Brown, J.K. 1975. Iroquois women: an ethnohistorical note. In Towards an anthropology of women (ed.) R. Reiter, 238-51. New York: Monthly Review Press.

Butler, J. 1990. Gender trouble: feminism and the subversion of identity. London: Routledge.

Colson, E. 1958. Marriage and the family among the Plateau Tonga. Manchester: University Press.

Davison, J. 1997. Gender, lineage, and ethnicity in Southern Africa. Oxford: Westview Press.

Douglas, M. 1971 [1969]. Is matriliny doomed in Africa? In Man in Africa (eds) M. Douglas & P.M. Kaberry, 123-37. New York: Anchor Books.

Ferguson, J. 1999. Expectations of modernity: myths and meanings of urban life on the Zambian Copperbelt. Berkeley: University of California Press.

Flinn, J. 1986. Matriliny without conflict: the case of Pulap. Journal of Polynesian Society 95, 221-38.

Fortes, M. 1969. Kinship and the social order: the legacy of Lewis Henry Morgan. London: Routledge and Kegan Paul.

Goody, J. 1956. The social organization of the LoWiili. London: HMSO.

——— 1959. The mother’s brother and the sister’s son in West Africa. Journal of the Royal Anthropological Institute 89, 61-86.

Gough, K. 1961. The modern disintegration of matrilineal descent groups. In Matrilineal kinship (eds) D.M. Schneider & K. Gough, 631-52. Berkeley: University of California Press.

Holy, L. 1986. Strategies and norms in a changing matrilineal society. Cambridge: University Press.

Johnson, J. 2012. Life with HIV: ‘stigma’ and hope in Malawi’s era of ARVs. Africa 82, 632-53.

Kelkar, G., D. Nathan & P. Walter 2003. Gender relations in forest societies in Asia: patriarchy at odds. London: Sage.

Lancaster, C.S. 1974. Brideservice, residence, and authority among the Goba (N. Shona) of the Zambezi Valley. Africa 44, 46-64.

——— 1976. Women, horticulture, and society in Sub-Saharan Africa. American Anthropologist 78, 539-64.

——— 1981. The Goba of the Zambezi: sex roles, economics and change. Norman, Okla.: University of Oklahoma Press.

Leach, E.R. 1961. Rethinking anthropology. London: The Athlone Press.

Lévi-Strauss, C. 1963. Structural anthropology volume I (trans. C. Jacobson & B. Grundfest Schoepf). New York: Basic Books.

——— 1969. The elementary structures of kinship. London: Eyre & Spottiswoode.

MacCormack, C.P. & M. Strathern (eds) 1980. Nature, culture and gender. Cambridge: University Press.

MacLeitch, G.D. 2011. Imperial entanglements: Iroquois change and persistence on the frontiers of empire. Philadelphia: University of Pennsylvania Press.

Malinowski, B. 2002 [1922]. Argonauts of the Western Pacific: an account of native enterprise and adventure in the archipelagoes of Melanesian New Guinea. London: Routledge.

Maybury-Lewis, D. (ed.) 1979. Dialectical societies: the Gê and Bororo of Central Brazil. Cambridge, Mass.: Harvard University Press.

Mitchell, J.C. 1956. The Yao village: a study in the social structure of a Malawian people. Manchester: University Press.

______ 1959 [1951]. The Yao of Southern Nyasaland. In Seven Tribes of British Central Africa (eds) E. Colson & M. Gluckman, 292-353. Manchester: University Press.

Murdock, G.P. 1949. Social structure. New York: Macmillan.

Oppong, C. 1974. Marriage among a matrilineal elite: a family study of Ghanaian senior civil servants. Cambridge: University Press.

Oyěwùmí, O. 1997. The invention of women: making an African sense of western gender discourses. Minneapolis: University of Minnesota Press.

Peletz, M. 1988. A share of the harvest. Berkeley: University of California Press.

Peters, P.E. 1997a. Introduction: revisiting the puzzle of matriliny in South-Central Africa. Critique of Anthropology 17, 125-46.

——— 1997b. Against the odds: matriliny, land and gender in the Shire Highlands of Malawi. Critique of Anthropology 17, 189-210.

Poewe, K.O. 1981. Matrilineal ideology: male-female dynamics in Luapula, Zambia. London: Academic Press.

Rattray, R.S. 1923. Ashanti. Oxford: Clarendon.

Read, M. 1942. Migrant labour in Africa and its effects on tribal life. International Labour Review 45, 605-31.

Richards, A.I. 1934. Mother-right among the central Bantu. In Essays presented to C.G. Seligman (eds) E.E. Evans-Pritchard, R. Firth, B. Malinowski & I. Schapera, 267-79. London: Kegan Paul, Trench, Trubner & Co.

——— 1950. Some types of family structure amongst the central Bantu. In African systems of kinship and marriage (eds) A.R. Radcliffe-Brown & D. Forde, 207-51. London: Oxford University Press.

——— 1982 [1956]. Chisungu: a girl’s initiation ceremony among the Bemba of Zambia. London: Routledge.

Rowley, H. 1867. The story of the universities mission to Central Africa: from its commencement, under Bishop Mackenzie, to its withdrawal from Zambesi. London: Saunders, Otley & Co.

Schneider, D.M. 1961. Introduction: the distinctive features of matrilineal descent groups. In Matrilineal kinship (eds) D.M. Schneider & K. Gough, 1-29. Berkeley: University of California Press.

______ & K. Gough (eds) 1961. Matrilineal kinship. Berkeley: University of California Press.

Tanner, N.M. & L. Thomas 1985. Rethinking matriliny: decision making and sex roles in Minangkabau. In Change and continuity in Manangkabau: local, regional and historical perspectives on West Sumatra (eds) L. Thomas & F. von Benda-Beckmann, 47-71. Athens: Ohio University Press.

Weiner, A. 1976. Women of value, men of renown: new perspectives in Trobriand exchange. Austin: Texas University Press.

——— 1992. Inalienable possessions: the paradox of keeping while giving. Berkeley: University of California Press.

Note on contributor

Jessica Johnson is a Lecturer in the Department of African Studies and Anthropology at the University of Birmingham. Between 2013 and 2016 she was a postdoctoral Research Fellow at the University of Cambridge, where she also completed her doctorate. Since 2009, Jessica has conducted more than two years of fieldwork in a matrilineal area of Malawi. Her doctoral research focused on aspirations for justice in the context of gender and martial relations, and her new project looks at 'everyday justice' in Malawi's magistrates' courts.

Jessica Johnson, Department of African Studies and Anthropology, University of Birmingham, B15 2TT, United Kingdom. j.johnson.5@bham.ac.uk