Skip to navigation | Skip to content


We need to talk about kinship

I

The situation of kinship studies is paradoxical.[1] Wherever we ethnographers go, we find that most people live in family households, obsess over marriage choices, celebrate kinship rituals, and quarrel over inheritance. And yet remarkably few anthropologists are writing about this sort of thing.

It was once very different. In 1967, Robin Fox wrote: ‘Kinship is to anthropology what logic is to philosophy or the nude is to art; it is the basic discipline of the subject.’[2] He obviously had in mind the grand tradition of comparative kinship studies that had been going for a century, its great figures Morgan, Rivers, Radcliffe-Brown, Murdock, Fortes and Lévi-Strauss. But as it turned out, Fox was the last person who could claim that kinship was in fact ‘the basic discipline’ of anthropology. The very next year, in 1968, precisely half a century ago, David Schneider published his slim study, American Kinship: A Cultural Account. This accessible, provocative, sketchy ethnographic report was his first move in what turned out to be a remarkably successful campaign to bring down the whole edifice of kinship studies. Schneider followed up in 1984 with his Critique of the Study of Kinship, a wide-ranging attack on the classical tradition.

It was obvious what Schneider was against, but perhaps not where he was coming from. He was in fact an unreconstructed if idiosyncratic disciple of the most influential social scientist in the USA at that time, Talcott Parsons.[3] Parsons laid it down that that there should be a division of labour in the human sciences. Anthropologists were granted a monopoly on what Parsons called culture, but only on condition that they didn’t try to be sociologists or psychologists. And by culture, Parsons meant something much narrower than the baggy notion that had been promoted by Tylor and Boas. For Parsons, culture meant a set of ideas and values, expressed in symbols. Schneider signed up to the Parsons party line. Cultural anthropologists should not bother with social or psychological processes. Their business was only with the study of ‘culture’: a symbolic discourse on values.

‘The book is about symbols’, Schneider wrote, introducing American Kinship, ‘the symbols which are American kinship’.[4] And which symbols were those? Schneider ignored religious symbols, for example, surely a big mistake in America, especially when it comes to marriage, family, sexuality, procreation and so on. But even if it is evident which symbols are in question, reading symbolic statements is a delicate business. So Schneider simplified matters. Metaphorical readings were out. There were no shades of meaning. Each symbol had a single and unambiguous referent. Schneider insisted, for instance, that in American English the name ‘father’ stands at once – and equally – for the husband of a ‘mother’ and for a Catholic priest. He had to make this absurd assumption because otherwise he would have had to admit that the primary referent of a ‘kinship’ term is quite commonly genealogical. People know perfectly well that a Catholic priest is only metaphorically a father.

When it comes to translation, the problems multiply. If no genealogical references are allowed, the ethnographer cannot translate native terms into English as ‘father’ or ‘mother’. It is impossible to identify local equivalents for concepts like ‘family’. Obviously, then, comparison becomes all but impossible. That did not bother Schneider. In his view, the very notion of ‘kinship’, mapped in genealogies, is a purely western cultural construct. Only ‘western’ societies can be said to have ‘kinship’. All those ethnographic accounts of exotic kinship systems were fictions. Their ethnographers had forced the beliefs of other people into a western mould. And by the way, sex and procreation are also cultural fictions. ‘There are only cultural constructions of reality’ Schneider wrote. ‘In this sense, then, “nature” and the “facts of life” … have no independent existence apart from how they are defined by the culture.’[5]

All this follows, perhaps, if one accepts Schneider’s starting point, that kinship should be studied simply as a symbolic system. But why go along with that? It leaves out so much that matters in everyday life. Malinowski’s doctoral thesis at the LSE, The Family Among the Australian Aborigines, published in 1913, tackled the venerable idea that the family did not exist among ‘primitive’ peoples. The Australians were exhibit no. 1. After all, they denied the father’s role in procreation. But Malinowski found ample evidence that people in Aboriginal camps lived in family units, and that fathers and mothers raised their children together.[6] Malinowski would later come across a similar myth of spirit procreation in the Trobriand Islands. This was a matrilineal society, and yet here too fathers, mothers and young children lived together. Had Schneider been the ethnographer of the Trobriand Islands, he would not have got beyond the local mythology of procreation.

If the ethnographer has to take the actors’ word for everything, it is impossible to work out what is going on in their lives. (And by the way, which actors? And which words?) A Schneiderian ethnographer will have nothing to say about the chronic tensions of family life, the strains of alliance, the workings of the domestic economy, the strategies of network mobilization. Even their own experiences in the field must be denied. Young ethnographers are adopted, willy-nilly, into local families and kinship networks. David Schneider himself, as a young man, was adopted as a son by his main informant in Yap. ‘I first claimed [Tannengin] for a father surrogate and now he has claimed me for a son’, he wrote in his fieldnotes.[7] But better not to talk about that when you come back. Schneider ended up by claiming that there were no ‘fathers’ in Yap.[8]

II

Despite all the obvious weaknesses in his approach to kinship, Schneider won out. I do not believe that this was down to the logical force of his critique. One reason for his success may be that his book on American kinship appeared precisely in 1968, just as a season of carnival and rebellion kicked off. A year earlier, in his Reith Lectures, Edmund Leach had said that ‘Far from being the basis of the good society, the family, with its narrow privacy and tawdry secrets, is the source of all our discontents’.[9] Surely, happily, the family was on the way out, as Engels had foreseen? And anyway, new reproductive technologies were about to make fathers redundant.

But it was not just a matter of fashion, or politics, or even science fiction. The new kinship studies fitted in with what was to be a new orthodoxy in American anthropology. A combination of ultra-relativism and philosophical idealism, it embraces the old post-modernists and the new ‘perpectivists’. All that matters is how people imagine the world. Our business, as ethnographers, is with symbols, not actions.

Marshall Sahlins was an early convert. In that very same magical year, in 1968, while on sabbatical in Paris, when the young were becoming Marxists, Sahlins gave up on Marx and Hegel and became a neo-Kantian. He embraced Lévi-Strauss on myth. And now he has endorsed Schneider on kinship. In 2014 he published an up-to-date and authoritative account of the new kinship orthodoxy entitled What Kinship Is – and Is Not…

Well then, What is Kinship? Sahlins takes off, like Schneider, from the taken-for-granted premise that kinship is a symbolic discourse. It discourses about ‘mutuality of being’ or ‘common substance’ – the currently ok terms for what used to be thought of as ‘kinship relationships’. Notions of ‘common substance’ have nothing to do with biology or genealogy. In fact, biology and genealogy are precisely What Kinship Is Not. Symbolic play may therefore take all sorts of imaginative forms. Sahlins crams his book with snippets of believe-it-or-not ethnographic examples, all going to show the variety of fanciful beliefs about common substance. Somewhere or other, people may be claimed as kin if they share the same name, or have the same birthday, or live in the same village, or eat the same food, or if they also suffer from ringworm.[10] Any of these mechanisms may make ‘persons’ feel that they belong together.

So kinship is nothing more, or less, than a set of ideas about relationship, expressed in symbols. ‘I take the risk’, Sahlins declares, ‘all means of constituting kinship are in essence the same.’[11] Kinship has nothing to do with genealogical relationships, reproduction, the imperatives of child raising. And yet it is everywhere. Schneider’s demolition job on kinship studies was followed by a miraculous resurrection. Kinship did not disappear. The body was buried, but the spirit survived. And lo and behold, the new, spiritual, symbolic kinship was universal. A generation of ethnographers took Schneider’s ideas into the field. Wherever they travelled, they found ‘persons’ who are symbolically constructed as ‘relatives’ because they share some ‘common substance’.

Virtually all the sources cited by Sahlins are written by such true believers, ethnographers who embraced Schneider’s doctrine. They present ready-made accounts of precisely what Sahlins is looking for, expressed in the same language of high idealism, so that the citations merge seamlessly into the text. They find kinship not in biology – even folk biology – but in symbolic representations of ‘persons’ and ‘common substance.’ And since they are concerned only with symbolic discourses, they pay no attention to the messy day to day business of family life.

The examples that Sahlins gives have another common feature. Almost all are drawn from non-western societies. But not all non-western societies. There is barely a mention of China, Japan, India, Java, the Middle East. Practically all the cases in this book come from what Sahlins himself once called Stone Age Societies. According to a venerable anthropological tradition, these archaic societies were based on kinship groups. Their medium was the gift. Modern societies, in contrast, are made up of individuals and are based on the market. ‘Money is to the west’, Sahlins wrote, in his more Marxist days, ‘as kinship is to the rest’.[12]

This is not a matter of false consciousness. Those other folk out there, ‘the rest,’ have the true kinship faith. The fundamental principle on which this true faith rests is that a person is made up of parts of other people. Lévy-Bruhl got it right, a century ago. Primitive peoples believe that they participate mystically in nature, and in each other’s very beings. We may believe that we are individuals. They realise that they are ‘dividuals’ (a coinage of Gilles Deleuze that has become widely used by anthropologists).[13]

So it is just as the Victorians thought. Primitive peoples are collectivist. Civilised peoples are competitive, individualist and greedy. (‘Bourgeois’, Sahlins calls them for short.) And while we (western natives and ethnographers alike) have a naïve faith that blood is thicker than water, they have poetic ideas about spirits that enter a woman’s womb and make her pregnant, or they conjure up fateful accidents which make one man another man’s brother.

Sahlins concludes that westerners are so out of it, so very bourgeois, with their individualist, biological theories, that they need to learn how to understand kinship from those other peoples. ‘Rather than imposing an ancient Western philosophy as an ethnographic epistemology,’ Sahlins writes approvingly of one contemporary theorist, ‘Viveiros de Castro let the Indian’s ontology come to him, their potential brother-in-law, and he made a comparative anthropology of it.’ And what precisely did he make of it? In the last lines of this book, Sahlins lays it out. ‘Precisely as Viveiros de Castro wrote … “Amazonian kinship ideas are tantamount to a non-biological theory of life. Kinship here is what you have when you do without a biological theory of relationality”’.[14]

For Sahlins, then, as for Schneider, kinship is nothing more than a set of symbolic expressions, the more outlandish the merrier. These symbols speak to a primordial idea of ‘common substance’, and this is apparently the central doctrine of Stone Age Humanism. Any functions that relatives might have in daily life are beside the point.

But this will not do. Kinship matters. It provides the only social security system that most people in the world can rely on. It delivers the daily bread, emotional support, favours, role models and a sense of identity. Wherever ethnographers travel they find that parents and children live together, expect to love one another, and hope that they can call on more distant relatives to help them out. So kinship is a universal, vital principle of social life. Yet the school of Schneider/Sahlins has no interest in how it works.

III

The comparative project of kinship studies has been in the deep freeze for a long time now. I am not sure that it can be brought back to life, although it is evident that a handful of prototypes – of families, marriage arrangements, inheritance rules, and so on – turn up all over the place. And there are indeed Iroquois and Crow and Omaha and Eskimo kinship terminologies, and they really are associated with particular marriage preferences and rules of descent.

But we are not condemned to choose between Schneider and Sahlins, on the one hand, and Radcliffe-Brown or Lévi-Strauss on the other.[15] There is a third way, a long-standing alternative to both formalism and idealism. In 1930, Malinowski protested against what he called ‘kinship algebra and geometry … long deductive arguments, as well as the piling of hypothesis upon hypothesis.’ He insisted that ‘after all, kinship is a matter of flesh and blood, the result of sexual passion and maternal affection, of long intimate daily life, and of a host of personal intimate interests’.[16] Edmund Leach argued that: ‘while kinship is something that we can get hold of as anthropologists, it is really just a way of talking about property and class relations. This feeling came from my childhood experiences. I still felt it in the Kachin Hills. I still feel it today. In the Sri Lanka case … talk about kinship, which went on all the time, was really talk about claims to irrigation water. In the Kachin Hills talk about kinship was talk about the networks of political power. But kinship is not a “thing in itself”; this is where I differ from Radcliffe-Brown, from Fortes, from Lévi-Strauss.’[17] Pierre Bourdieu made a case against the formalism of Lévi-Strauss’s theory of cousin marriage, and like Leach he invoked the strategic choices and emotional bonds that had to be taken into account to explain marriage with paternal cousins among the Kabyle and also the upwardly mobile marriages that were encouraged in his own family village in the Béarn.[18]

The programme of Malinowski, Leach and Bourdieu survives, though barely. There are realistic studies of the modern family in sociology. The history of the family is a flourishing specialism.[19] Ethnographers still produce interesting accounts of family concerns – for a good, up-to-date sample, see the case studies collected in Peter Schweitzer’s Dividends of Kinship and Susan McKinnon and Fenella Cannell’s Vital Relations.[20]

But we need to go further, to connect kinship, marriage and the family to politics and business. Everywhere in the world, however traditional or modern, conservative or revolutionary, power and wealth are likely to be family affairs. Elites, Choice, Leadership and Succession, edited by João de Pina-Cabral and Antónia Pedrosa de Lima, illuminates the power of dynastic family enterprises in societies around the world, with examples from Fiji, Ghana, Macao, Portugal and England.[21] Ruling elites in many, perhaps most, modern states – from China to the USA, let alone Angola, Azerbaijan, Bangladesh, Cuba, Kenya, and Saudi Arabia – go in for mafia-style family arrangements. This applies to political machines, organised crime, and, of course, family businesses, which you will come across in every village and town on earth.

Writing about the new HBO drama series, ‘Succession’, Emily Nussbaum comments in the New Yorker: ‘Among the series’ many pleasures is how well it functions as a blind item about the Murdochs, the Mercers, the Redstones—families that double as brands, which is to say monarchies, which is to say Mob families. They’re groups that run the world and to whom rules don’t apply’.[22] Prince Philip apparently refers to the British royal family as ‘the firm’. In American politics, there was, long ago, the Adams family dynasty. More recently there are the Kennedys, the Bushes, the Clintons, and now the Trumps. (The son-in-law also rises…)

But surely this family stuff is nepotistic, backward-looking, undemocratic, an insult to free will, even a threat to capitalism itself? It must be doomed. In a classic paper published in the Harvard Business Review, back in 1984, Robert Donnelley asked: ‘Is family management contrary to the fundamental American creed advocating free competition, equality of opportunity, and the best man for the job? Does family influence contradict all precepts of professional management?’[23] In fact, it turns out, family ethics work very well, even in modern multinational concerns. Family businesses take the long view on investing, since they think in terms of generations. They command the loyalty of family employees and treat long-standing workers as members of the family. [24] And yet they perform efficiently in the market place, far better than Professor Donnelley supposed half a century ago.

According to Price Waterhouse Cooper’s Family Business Survey for 2007/8:
Family firms are the most common form of business structure; they employ many millions of people; and they generate a considerable amount of the world’s wealth. One measure of their importance is the proportion of registered companies that are family controlled—a figure which ranges from more than 50 percent in the European Union (EU) to between 65 percent and 90 percent in Latin America and over 95 percent in the US.[25]
Far from being in decline, and despite the recent recession, family-controlled firms made up 19% of the companies in the Fortune Global 500 in 2014, up from 15% in 2005.[26] Throw in family farms and corner shops, and the kinship basis of much of economic activity is even more evident. And, of course, virtually everywhere the family is the unit of consumption. This universal tension between the practices and principles of the market and what he calls the ‘house’ is the central theme of Stephen Gudeman’s illuminating Anthropology and Economy.[27]

IV

Yet although there is still a fair amount of realistic research on the family, some of it done by anthropologists, the study of kinship networks and marriage alliances has virtually disappeared. There are signs of a new interest, but this is to be found more among historians. The editors of a recent symposium on the history of European kinship remark that in place of the old obsession with ‘mom, dad, and the kids,’ social historians are making studies of clusters of related families whose members ‘recognize the social, moral, sentimental, economic, and political ties that bind them’.[28]

That used to be our speciality. However, the classics of kinship studies are now rarely taught to students of anthropology. It is true that the old, competitive models of descent and alliance systems were conceived within a sociology of ‘primitive societies’ that is quite rightly defunct. But in many societies, even modern, industrial societies, ties with married brothers and sisters link together a series of families in an enduring network. These kinship networks often operate as economic corporations or political factions. They also shape marriage choices. They matter, greatly.

And we are not just talking here about life in the distant past, or in rural backwaters. Plus ça change, plus c’est la même chose. A global wave of migration is now reshaping our world once again. Ethnographers are studying the new diasporas. And they find … kinship networks. There is a long-standing preference in rural Pakistan for marriage within the biradari (glossed sometimes as a ‘brotherhood’ of extended kin). The Pakistani diaspora in Britain also has a preference for cousin marriages. Perhaps unexpectedly, however, the prevalence of cousin marriage is even higher among Pakistani immigrants in the cities of Britain than it is in rural Pakistan. Alison Shaw reports that the rate of first cousin marriage is yet higher among younger British Pakistanis than it was for their parents or grandparents. Around a third of the marriages of first-generation immigrants to Britain from Pakistan were with first cousins, but well over half the British-born generation married first cousins.[29]

Traditional precepts and religious injunctions are still relevant, but there are more immediate and more practical reasons for these marriage preferences, arising from the process of migration itself. There are debts to family members back home who helped to finance the initial wave of migration. These may be repaid by sponsoring the migration of a nephew or niece. The best way to do that is by way of an arranged marriage. At least until very recently, if a Pakistani married a British citizen, he or she could come to live in the UK. In most of these second-generation cousin marriages, one partner is brought to Britain from Pakistan.

British emigrants once behaved in much the same way. Almost two-thirds of Boston Brahmin men who reached marriageable age during the Revolutionary period in the late 18th century married their cousins or sisters-in-law. Twenty per cent of Protestants who moved from Northern Ireland to the Midwestern USA in the first half of the 19th century married first cousins. Among Highland Scots migrants in New Zealand in the same period, 70 per cent of marriages were between close kin.[30]

I find all that very interesting. And important. The study of kinship networks is not, however, high up on our professional agenda. No doubt there are a number of reasons for this, but ultimately it is because large swathes of contemporary anthropology have been ceded to philosophical idealists. Ethnographers are being encouraged simply to parrot local myths. Kinship is then treated as nothing more than the symbolic representation of ‘relationships’. My plea here is that we should take at least as much interest as our informants do in the day to day affairs of families, kinship networks and marriage alliances.

  1. [1]This paper was presented at the EASA conference, Stockholm, August 2018. I am grateful to Kirsten Hastrup, who chaired the session in which I spoke, for her encouragement, and I appreciate her occasional reservations.
  2. [2]Robin Fox, Kinship and Marriage, Penguin Books, 1967, p.10.
  3. [3]Adam Kuper, 1999, Culture: The Anthropologists’ Account, Harvard University Press. See chapter 4, ‘David Schneider: Biology as Culture’.
  4. [4]David Schneider, American Kinship: A Cultural Account, University of Chicago Press, p. 18.
  5. [5]David Schneider, 1976, ‘Notes towards a theory of culture’, in K. Basso and H. Selby (eds.), Meaning in Anthropology, University of New Mexico Press, p. 204.
  6. [6]For an authoritative review of the issues, see L. R. Hiatt, 1996, Arguments about Aborigines: Australia and the Evolution of Social Anthropology, Cambridge University Press.
  7. [7]Ira Bashkow, 1991, ‘The dynamics of rapport in a colonial situation: David Schneider’s fieldwork in the islands of Yap’, in George Stocking (ed.), Colonial Situations: Essays on the Contextualization of Ethnographic Knowledge, University of Wisconsin Press. Citation pp. 217-218.
  8. [8]David Schneider, 1984, A Critique of the Study of Kinship, University of Michigan Press, p. 81.
  9. [9]E. R. Leach, 1967, BBC Reith Lectures, ‘Runaway World’, The Listener, 30th November, 1967.
  10. [10]According to Sahlins, ‘such practices of participation in one another’s existence are indefinitely many, inasmuch as they are culturally relative. One may be kin to another by being born on the same day (Inuit), by following the same tabus (Araweté), by surviving a trial at sea (Truk) or on the ice (Inuit), even by mutually suffering from ringworm (Kaluli).’ Marshall Sahlins, 2013, What Kinship Is – And Is Not, University of Chicago Press, p. 68.
  11. [11]Op. cit., p. 29.
  12. [12]Marshall Sahlins, 1976, Culture and Practical Reason, University of Chicago Press, p. 216.
  13. [13]See Gilles Deleuze, 1992, ‘Postscript on the Societies of Control’, October 59: 3-7.
  14. [14]Marshall Sahlins, 2013, What Kinship Is – And Is Not, University of Chicago Press, p.89.
  15. [15]See Adam Kuper, 2016, ‘Meyer Fortes: the person, the role, the theory’, The Cambridge Journal of Anthropology, 34 (2): 127-139.
  16. [16]B. Malinowski, 1930, ‘Kinship’, Man, Vol. 30 (Feb., 1930), pp. 19-29. Citations, pp. 19-20.
  17. [17]Adam Kuper, 1986, ‘An Interview with Edmund Leach’, Current Anthropology, Vol. 27, No. 4, pp. 375-382. Citation, p. 380.
  18. [18]Pierre Bourdieu, 1977, Outline of a Theory of Practice, Cambridge University Press, pp. 30-58.
  19. [19]Christopher H. Johnson and David Warren Sabean (eds.), 2011, Sibling Relations and the Transformations of European Kinship, 1300-1900, Berghahn Books.
  20. [20]Peter Schweizer (ed.), 2000, Dividends of Kinship, Routledge; Susan McKinnon and Fenella Cannell (eds.), 2013, Vital Relations: Modernity and the Persistent Life of Kinship, School for Advanced Research Seminar Series.
  21. [21]João de Pina-Cabral, Antónia Pedrosa de Lima, 2000, Elites, Choice, Leadership and Succession, Oxford: Berg.
  22. [22]Emily Nussbaum, 2018, ‘“Succession” ’s Satisfyingly Nasty Family Ties’, New Yorker, September 3rd, 2018.
  23. [23]R. Donnelley, (1964), ‘The family business’, Harvard Business Review, Vol. 42 No .4, pp.94-105.
  24. [24]Gary S. Becker, A Treatise on the Family (Cambridge, MA, 1991), 300-1.
  25. [25]Research by McKinsey, reported in the Economist, 1st November, 2014. (“Business in the Blood”.)
  26. [26]Research by McKinsey, reported in the Economist, 1st November, 2014. (“Business in the Blood”.)
  27. [27]Stephen Gudeman, 2016, Anthropology and economy, Cambridge University Press.
  28. [28]Christopher H. Johnson and David Warren Sabean (eds.), 2011, Sibling Relations and the Transformations of European Kinship, 1300-1900, Berghahn Books, p. 23.
  29. [29]Alison Shaw, 2001, ‘Kinship, Cultural Preference and Immigration: Consanguineous Marriage among British Pakistanis’, The Journal of the Royal Anthropological Institute, vol. 7, no. 2, pp. 315–334.
  30. [30]See the review of these cases in Adam Kuper, 2009, Incest and Influence: The Private Life of Bourgeois England, Harvard University Press, pp. 246-248.

Please join our mailing list to receive notification of new issues