Beyond 'fluidity': kinship and households as social projects

Working Paper


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Urban and rural households in South Africa are fluid (in that individuals move between households) and porous (in that individuals may be members of more than one 'household' at the same time). One important consequence of this fluidity and porosity is that the effects of AIDS-related disability and death may be mitigated, as households are reconstituted to provide care as well as to protect the welfare of dependents. This paper argues that our understanding of household dynamics needs to go beyond asserting the fact of fluidity (or porosity). Precisely how fluid and porous are households, and (in particular) how 'extended' are families in terms of the obligations and claims that kin can make on each other? Are there moral or normative constraints on the decisions made by individuals with respect to householding and kinship? The paper reviews the extant literatures on these questions in South Africa as a basis for further empirical research. Whilst inconclusive, the secondary literatures suggest that the claims entailed in kinship are more and more often evaded, especially by men (and paternal kin), and that responsibilies and obligations are increasingly conditional. Children - including, but not only, orphans - are readily accommodated by kin. South Africa's social assistance system means that the elderly are financial breadwinners rather than dependents, helping to perpetuate the practice of extended family-households. Perhaps the most striking shift in household and kinship has been the decline of marriage and the crisis of patrilinearity. There is indirect evidence that people not only prioritise the claims of children but also see the elderly and sick as more deserving than able-bodied adults of working age. Overall, 'fluidity' and the claims made in the name of kinship have clear limits. The paper concludes with the suggestion that further empirical research be informed by the methodologies used in studies of kinship in the UK and USA, notwithstanding the substantive differences between householding and kinship in different settings.