Version classiqueVersion mobile

Émergence et évolution de la parenté

 | 
Jean Lassègue

Revisiting Matrilineal Priority

Chris Knight

Résumé

Au siècle dernier, jusque dans les années 1920, la plupart des anthropologues se rangeaient à l’idée que l’évolution de l’humanité était passée par des « stades » et qu’un stade matrilinéaire avait précédé un stade patrilinéaire. Après une période de critique radicale, la situation a changé aujourd’hui et ce qui apparaissait comme un résultat incontesté de l’anthropologie du xxe siècle doit être précisé, en particulier en ce qui concerne la question des rapports entre avunculat (c’est-à-dire la relation particulière existant entre un homme et le fils de sa sœur et qui peut porter sur des droits de résidence, de partage des repas ou autre) et matrilinéarité (filiation par les mères). Au xixe siècle, l’antériorité matrilinéaire était fondée sur deux hypothèses ethnographiques: (i) aucune société connue n’était passée d’un système patrilinéaire à un système matrilinéaire; (ii) il était possible d’interpréter le cas de l’avunculat, cas particulier à certaines sociétés seulement, comme une trace de priorité matrilinéaire dans des systèmes devenus patrilinéaires.

La première hypothèse fit l’objet d’une critique vigoureuse de la part de Franz Boas: celui-ci avait découvert une société (les Kwakiutl) en train de passer d’un système patrilinéaire à un système matrilinéaire sous l’influence de voisins organisés matrilinéairement. Le schéma évolutionniste standard était donc inopérant. En fait, le cas étudié ne permettait sûrement pas de tirer des conclusions de cette ampleur mais Boas voulait surtout remettre complètement en question l’aspect déterministe et unidirectionnel du cadre évolutionniste hérité du XIXe siècle. Le résultat fut surtout que, parmi les anthropologues américains, la génération suivante essaya d’établir une direction en sens inverse, allant du patrilinéaire au matrilinéaire. Ce n’est qu’à partir du moment où l’idée de diffusion de la culture eut droit de cité en anthropologie que les aspects unidirectionnels du cadre évolutionniste furent progressivement abandonnés. La deuxième hypothèse, portant sur l’avunculat, fut critiquée par Radcliff-Brown: sa définition de l’avunculat lui permettait de se passer des « stades » propres au cadre évolutionniste. Pour lui, le point important pour déterminer la filiation consistait à pouvoir déterminer à qui l’enfant d’une femme appartenait: s’il appartenait au groupe de sa mère, la filiation était matrilinéaire, s’il appartenait au groupe de son père, elle était patrilinéaire. Dans le cas d’une filiation patrilinéaire – et en faisant l’hypothèse que le lien mère-enfant est le seul qui soit naturellement donné –, le mariage sépare la mère de son groupe de filiation d’origine, et en particulier de son frère, tandis que cet effet ne se produit pas dans le cas d’une filiation matrilinéaire, le frère et la sœur devenue mère restant dans le même groupe de filiation. Ainsi la façon dont le lien du mariage est considéré rend-elle compte de la pratique de l’avunculat sans qu’il soit nécessaire de faire intervenir des hypothèses de nature évolutionniste: si le lien mari-épouse est prépondérant dans une société, il se fait au détriment du lien frère sœur et vice versa. Mais si le lien mari-épouse s’explique par la nécessité de l’alliance, le lien frère-sœur, lui, n’est pas rapporté par Radcliff-Brown à la matrilinéarité et reste, tel quel, totalement inexpliqué.

L’exemple des sociétés africaines permet de supposer que les sociétés où se pratique l’avunculat, prioritairement patrilinéaires pour ce qui est de la filiation, doivent cependant demeurer minoritairement structurées selon un principe matrilinéaire, puisque persistent en elles des groupes au sein desquels se côtoient des hommes, leurs sœurs et leurs enfants sur lesquels les pères n’ont de droits que s’ils les achètent auprès du groupe de leur femme (dans le cas de la société bantoue par exemple). Ce rachat peut être interprété comme l’étape finale d’un processus par lequel s’opère la séparation des femmes et de leurs enfants qui, sinon, resteraient dans le groupe de leur mère. En Afrique, c’est le rachat sous forme de têtes de bétail qui fait la différence entre les systèmes de filiation: dans le cas où il n’y a pas de bétail, on a une société matrilinéaire-matrilocale ; à partir du moment où le bétail apparaît et sert à la compensation matrimoniale (départ de l’épouse vers la résidence du mari), on a un système matrilinéaire-patrilocal; quand la compensation devient plus importante, on trouve des systèmes patrilinéaires-patrilocaux dans lesquels le divorce n’existe pas (Harold Schneider). Reste à comprendre les raisons du maintien de liens matrilinéaires dans des sociétés patrilinéaires.

Le lien frère-sœur n’est pas seulement une survivance parce que la grande licence de conduite du neveu à l’égard de son oncle – sans que ce dernier puisse faire preuve d’autorité à son égard – n’existe que dans le cadre patrilinéaire, alors que l’autorité de l’oncle sur son neveu est beaucoup plus présente dans le cadre strictement matrilinéaire. L’absence d’autorité de l’oncle sur son neveu est donc bien plutôt la marque de la transition d’un système matrilinéaire vers un système patrilinéaire. Encore faut-il admettre de relier la nature de l’avunculat à la filiation matrilinéaire, ce que la majorité des anthropologues (y compris les plus fameux comme Evans Pritchard, Lévi-Strauss, Leach ou, dans une moindre mesure, Goody), suivant en cela Radcliff-Brown, se refusent à faire, prenant la relation frère-sœur comme un simple axiome.

Texte intégral

1 Nineteenth century anthropologists widely agreed that early human society was not based on the nuclear family. Lewis Henry Morgan instead championed the matrilineal clan as the first stable institutional framework for human family life. In this, he was supported by theorists who later came to include E.B. Tylor, Friedrich Engels, W.H.R. Rivers, É. Durkheim and S. Freud. Until the 1920s, most anthropologists still accepted a “stages” view of the evolution of kinship, in which matrilineal descent systems universally preceded their patrilineal counterparts.

2When Morgan’s evolutionist schema was discredited early in the twentieth century, it was largely on the basis of two interventions. First, F. Boas claimed to have discovered a Vancouver Island tribe (the Kwakiutl) in the throes of transition from patrilineal to matrilineal descent, reversing Morgan’s supposedly universal sequence. Second, A.R. Radcliffe-Brown offered an explanation of the “avunculate” in terms of universal psychological and sociological principles, claiming that the “matrilineal survival” interpretation of this peculiarity of kinship could therefore be dispensed with.

3In this article, I make no attempt to return to the debate or adjudicate on the substantive issues involved. My more modest aim is to recall key features of the debate, highlighting the motivations and agendas involved at the time.

4If there is one thing on which all schools of social anthropology are agreed, it is that the nineteenth century “mother-right” theory of early kinship is of no more than historical interest. Of all the theoretical conquests achieved by Boas, Kroeber, Lowie, Malinowski, Radcliffe-Brown and their allies during the first decades of the twentieth century, the overthrow of the Bachofen-Morgan evolutionary scheme has appeared the most secure. To the extent that twentieth century social anthropology solved the problems it set out to address, those responsible for this paradigm shift must posthumously be accorded full credit. Should we conclude, however, that twentieth century social anthropology stumbled from crisis to crisis, solving not one of the most basic problems facing it, then a search for the roots of our crisis might return us to that decisive moment in the history of our discipline.

The discovery of matriliny

  • 1 J. Lederer, The Discoveries of John Lederer, in Three Several Marches from Virginia, to the West of (...)

5Matrilineal exogamy was first accurately described in print by an early English adventurer, John Lederer, in an account of his travels in Eastern North America published in 1672. He was writing of the Tutelo, an eastern Siouan tribe1:

From four women, viz., Pash, Sopoy, Askarin and Maraskarin they derive the race of mankind; which they therefore divide into four Tribes, distinguished under those several names. They very religiously observe the degrees of marriage, which they limit not to distance of Kindred, but difference of Tribes, which are continued in the issue of the females: now for two of the same tribe to match, is abhorred as Incest, and punished with great severity.

6What Lederer refers to as a “tribe” is, in modern anthropological parlance, a “clan”. In this case, there are four matrilineal clans within the same tribe, marriage within the same clan being equally prohibited no matter how “closely” or “distantly” related the partners.

  • 2 J. T. Lafitau, Mœurs des sauvages amériquaines, comparées aux mœurs des premiers temps, Paris, Saug (...)

7Fifty-two years later, Father Lafitau described in glowing terms the apparent equality and even (in important respects) superiority of the female sex over the male among the Iroquois2:

Nothing […] is more real than this superiority of the women. It is essentially the women who embody the Nation, the nobility of blood, the genealogical tree, the sequence of generations and the continuity of families. It is in them that all real authority resides: the land, the fields and all their produce belongs to them: they are the soul of the councils, the arbiters of peace and war: they conserve the finances or the public treasury; it is to them that slaves are given: they make the marriages, the children are in their domain and it is in their blood that the order of succession is based.

8In 1767 the great Scottish historian Adam Ferguson was the first to generalize on the basis of Lafitau’s and other missionaries’ and explorers’ reports. Writing on “savage nations” in general he remarked that:

  • 3 A. Ferguson, An Essay on the History of Civil Society [1767], Philadelphia, A. Finley, 1819, p. 126

[...] as the domestic cares are committed to the women, so the property of the household seems likewise to be vested in them. The children are considered as pertaining to the mother, with little regard to descent on the father’s side.3

9This suggestion was not followed up, however, until in 1861 the Swiss jurist and historian of Roman law, Johann Jakob Bachofen, published his “Mother-Right”. Bachofen knew little at this time of ethnology beyond that contained in the literature of ancient Greece and Rome. His knowledge of the classics was sufficient, however, for him to make a persuasive-seeming case for the priority of matriliny from these sources alone. An idea of the kind of evidence he relied on is given by these two excerpts:

  • 4 J. J. Bachofen, Myth, Religion and Mother-right. (Selected Writings), New Jersey, Princeton Univers (...)

Of all records relating and pertaining to mother right, those concerning the Lycian people are the clearest and most valuable. The Lycians, Herodotus reports, did not name their children after their fathers like the Hellenes, but exclusively after their mothers; in their genealogical records they dealt entirely with the maternal line, and the status of children was defined solely in accordance with that of the mother. Nicolaus of Damascus completes this testimony by telling us that only the daughters possessed the right of inheritance.4

10Or again :

  • 5 Ibid., p. 135-136.

[…] I should like to cite a report from Nicolaus of Damascus’ collection of strange customs, preserved by Stobaeus in his Florilegium: “The Ethiopians hold their sisters in particular honor. The kings leave their scepter not to their own children but to their sisters. If no heir is available, they choose the most beautiful and belligerent as their leader”. This last observation is confirmed by Herodotus and Strabo. The favoring of the sister’s children is a necessary consequence of mother-right and is also to be found elsewhere.5

  • 6 Ibid., p. 71.

11From his analysis of the myths, literature and ethnological reports of ancient Greece and Rome, Bachofen reached a conclusion which he expressed in these words6:

[…] mother right is not confined to any particular people but marks a cultural stage. In view of the universal qualities of human nature, this cultural stage cannot be restricted to any particular ethnic family. And consequently what must concern us is not so much the similarities between isolated phenomena as the unity of the basic conception.

  • 7 J. P. McLellan, Primitive Marriage, Edinburgh, Adam and Charles Black, 1865, p. 123; quoted in S. T (...)
  • 8 L. H. Morgan, Houses and House-Life of the American Aborigines, Chicago and London, University of C (...)
  • 9 Ibid., p. 126-128.
  • 10 See M. Harris, The Rise of Anthropological Theory, London, Routledge, 1969, p. 189.

12J. P. McLellan read Bachofen’s Book in 1866, having already the previous year published his Primitive Marriage7 in which he claimed to have proved that “the system of kinship through females” was “a more archaic system of kinship than the system of relationship through males”. It was Morgan, however, who did most to support Bachofen’s case by relating it to evidence concerning the kinship systems of the Iroquois and other Indians of the United States. Morgan related matrilineal clans to the solidarity between sisters living together, and associated this in turn with the idea of “communism in living”.8 Describing an Iroquois long-house he wrote of its immense length, its numerous compartments and fires, the “warm, roomy and tidily-kept habitations” within the long-house, the raised bunks around the walls, the common stores and “the matron in each household, who made a division of the food from the kettle to each family according to their needs […]” “Here, [he commented] was communism in living carried out in practical life […]”.9 In such households, he concluded, “was laid the foundation for that “mother-power” which was even more conspicuous in the tribes of the Old World, and which Professor Bachofen was the first to discuss under the name of gyneocracy and mother-right”. Morgan and Bachofen exerted a reciprocal influence on each other. While Bachofen’s ideas on the priority of matriliny became incorporated into Morgan’s Ancient Society (1877), Bachofen appreciated the comparisons drawn by Morgan between his own classical materials and the matrilineal clan system of the Iroquois. Bachofen expressed his gratitude by dedicating a book of essays to Morgan.10

  • 11 F. Engels, The Origin of the Family, Private property and the State [1884], New York, Pathfinder Pr (...)

13Following some suggestions made by Marx, Engels wrote his The Origin of the Family, Private Property and the State (1884) in order to show the way in which Bachofen’s and Morgan’s positions gave strength to the interpretation of human evolution and history which he and Marx had evolved. For Engels, the base line from which alone the study of kinship-solidarity could begin was provided not by the modal of “the family” in anything like the modern European sense but by the model of the matrilineal clan. Just how important this seemed to him can be judged from the following evaluation of Morgan’s work11:

The rediscovery of the original mother-right gens as the stage preliminary to the father-right gens of the civilised peoples has the same significance for the history of primitive society as theory of evolution has for biology, and Marx’s theory of surplus value for political economy. It enabled Morgan to outline for the first time a history of the family, wherein at least the classical stages of development are, on the whole, provisionally established, as far as the material at present available permits. Clearly, this opens a new era in the treatment of the history of primitive society. The mother-right gens has become the pivot around which this entire science turns; since its discovery we know in which direction to conduct our researches, what to investigate and how to classify the results of our investigations.

  • 12 K. Marx, “Economic and Philosophical Manuscripts” [1844], in D. McLellan (ed.), Karl Marx: Selected (...)
  • 13 K. Marx and F. Engels, “The German Ideology” [1846], in D. McLellan (ed.), ibid., p. 175-208; p. 18 (...)

14This was not a hasty judgment. It should be remembered that even as early as 1844, Marx had written that the “immediate, natural and necessary relationship of human being to human being is the relationship of man to woman”, and that “from this relationship the whole cultural level of man can be judged”.12 The matrilineal clan, based on the solidarity of women within large, collective households, seemed to Marx and Engels merely a concrete embodiment of something in which they had always believed – namely, an original form of social organization in which the sexes had been equally powerful and in which not even the rudiments of private property, class-exploitation or the state had as yet evolved. Marx and Engels wrote in The German Ideology13of “[…] property, the nucleus, the first form, of which lies in the family, where wife and children are the slaves of the husband”. Morgan’s view that private property had emerged through the break-up of the originally communal matrilineal households of an earlier phase of social evolution seemed to dovetail neatly with this idea and was taken up eagerly by Engels. Morgan described how the transition to patrilineal descent had been associated with the breaking of women’s own kinship bonds and their isolation from one another in the home:

  • 14 L. H. Morgan, Houses and House-Life of the American Aborigines, op. cit., p. 128.

It thus reversed the position of the wife and mother in the household; she was of a different gens from her children, as well as her husband; and under monogamy was now isolated from her gentile kindred, living in the separate and exclusive house of her husband. Her new condition tended to subvert and destroy that power and influence which descent in the female line and the joint-tenement houses had created.14

15Engels commente:

  • 15 F. Engels, The Origin of the Family, Private Property and the State, op. cit., p. 68.

The overthrow of mother right was the world-historic defeat of the female sex. The man seized the reins in the house also, the woman was degraded, enthralled, the slave of the man’s lust, a mere instrument for breeding children.15

16He continued, rooting the central features of class society in the original processes through which marriage bonds had undermined the solidarity of women in the matrilineal joint household and clan:

  • 16 Ibid., p. 75.

The first class antagonism which appears in history coincides with the development of the antagonism between man and woman in monogamian marriage, and the first class oppression with that of the female sex by the male.16

Boas and the Kwakiutl case

17The matrilineal priority hypothesis rested on two fundamental props:

  • a number of matrilineal descent systems were known to have shifted towards patriliny in recent times. No case of a society evolving in the reverse direction was known;
  • in certain African societies with patrilineal descent, apparently inconsistent principles could be discerned. These anomalous features – such as the “special” relationship between a child and its mother’s brother – could be interpreted as matrilineal survivals.
  • 17 R. Lowie, “Evolution in Cultural Anthropology: A Reply to Leslie White”, American Anthropologist, 4 (...)
  • 18 F. Boas, Race, Language and Culture, New York, Macmillan, 1940, p. 635.
  • 19 R. Lowie, “Evolution in Cultural Anthropology: A Reply to Leslie White”, art. cit., p. 420.

18The first prop was attacked by Franz Boas. Boas, Lowie reminds us, had begun as a unilinear evolutionist.17 In 1888 he had defended “the current view of a necessary precedence of matrilineal forms of family organization”.18 But then came… “the new facts”. Writing of Boas, Lowie explained in 194619:

[…] his critique of evolutionary schemes is the psychological equivalent of the experimentalists’ critique of “the biogenetic law”. The facts did not fit the theory, hence the theory would have to be modified or discarded… L. H. Morgan teaches that the individual family is an end-product, preceded by various stages including that of a clan organization; Morgan, Bachofen, and Tylor teach the priority of matrilineal descent. Boas found that in the interior of British Columbia clanless tribes with a family organization and a patrilineal trend adopted from coastal neighbours a matrilineal clan organization. Diffusion thus disproved the universal validity of the formula that Boas himself had been defending in 1888.

  • 20 R. Lowie, “Social Organization”, American Journal of Sociology,20, 1914, p. 68-167; reprinted in C. (...)
  • 21 M. Harris, The Rise of Anthropological Theory, op. cit., p. 305.

19These were large claims, and they certainly served their purpose. Less wellknown is the fact that they were stretched to fit the case – as Lowie himself quite evidently realised. Although “the Kwakiutl facts are very interesting”, he was admitting already in 191420, “it is highly doubtful whether they have the theoretical significance ascribed to them”. The Kwakiutl numaym groups were not matrilineal but “mixed”, the reckoning of a child’s affiliation being “indeterminate”, and the descent groups not being exogamous at all. « For these reasons », Lowie regrettably concluded, “the Kwakiutl conditions do not seem to furnish a favorable test case”. Note that this did nothing to prevent Lowie from subsequently downplaying his misgivings and celebrating the Kwakiutl case as the crucial disproof of Morgan and Engels. Summing up the incident, Marvin Harris concludes that the “extreme interest” of Boas’ handling of his material21:

[…] stems from the fashion in which he and his students seized upon this case to destroy the supposed universal tendency for patrilineality to follow matrilineality and at the same time to discredit the entire historical determinist position. On the basis of this one drastically deficient case, there gradually diffused out of Schermerhorn Hall at Columbia, through lecture, word of mouth, article and text, the unquestioned dogma that Boas had proved that it was just as likely that patrilineality succeeded matrilineality as the reverse.

  • 22 Loc. cit.
  • 23 G. Reichard, “Social Life”, in F. Boas (ed.), General Anthropology, New York, Heath, 1938, p. 409-4 (...)
  • 24 J. R. Swanton, “The Social Organization of American Tribes”, American Anthropologist, 7, 1905, p. 6 (...)
  • 25 R. Lowie, Primitive Society, New York, Bini and Liveright, 1920; A. Kroeber, Anthropology, New York (...)

20Harris describes how, following Boas’ sensational claims, a stampede against Morgan and his followers was let loose.22 When the Kwakiutl case was cited in the general textbook that Boas had edited, Gladys Reichard referred to it as « more convincing » than the other possible instance among the Trobrianders ( “also” – as Harris puts it – “highly dubious, however”) because “the details of its cause are more positive”.23 Boas’ speculations concerning a patrilineal to matrilineal transition among the Kwakiutl actually gave rise to an attempt to show that this sequence was the normal one. Thus, John Swanton24 tried to demonstrate that in America it was precisely the matrilineally organized tribes, such as the Haida of the Northwest, the Zuni and Hopi among the Pueblos, and the Creek and Natchez of the Southeast, who were the culturally most “advanced” peoples, while the groups that were organized on a patrilineal and a bilateral basis were deemed “primitive”.24 Lowie, Kroeber and Goldenweiser all eagerly seized upon this idea.25 As Murdock has shown:

  • 26 G. P. Murdock, Social Structure, Macmillan, New York, 1949, p. 189.

This inverted evolutionistic scheme of a bilateral-patrilineal-matrilineal succession in the forms of social organization became an established dogma in American anthropology.26

  • 27 Ibid., p. 185.

21To see how the “matrilineal priority” hypothesis appeared within the perspective of later twentieth century anthropology, it may be useful to turn to the words of G. P. Murdock. During the latter half of the nineteenth century, he writes, authorities such as Bachofen, Lubbock, McLennan, Morgan, Spencer and Tylor (i. e. virtually all those who shaped anthropological opinion for nearly fifty years) were generally agreed that the matrilineal clan was the original form of human social organization,27 that this form of society gave way to patrilineal and patriarchal institutions as the male sex gradually achieved a position of dominance, and that the emergence of bilateral ( “family” in the European sense) kinship and the nuclear family marked a relatively late phase of social evolution. “The hypothesis of the priority of the matrilineate”, comments Murdock.

  • 28 Loc. cit.

[…] was buttressed with a number of extremely plausible arguments – the presumed ignorance of physical paternity in primitive times, the biological inevitability of the association of mother and child, the alleged non-inclusion of the father in the family under early nomadic conditions, the large number of apparent survivals of matrilineal customs in patrilineal societies and the rarity of comparable patrilineal traits among matrilineal peoples, the relative cultural backwardness of matrilineal as compared with patrilineal societies, and the complete lack of historically attested cases of a transition from patrilineal to matrilineal institutions.28

22Murdock continues:

  • 29 Ibid., p. 190.

The most secure prop of the evolutionist theory of matrilineal priority, that which later anthropologists have had the greatest difficulty in removing, is the complete lack of historically attested, or even inferentially probable, cases of a direct transition from patrilineal to matrilineal descent. No such case has ever been encountered [… ]29

23The author adds:

[…] the explanation turns out to be simple. There are no recorded cases of such a transition because it cannot occur […] the direct transition from patrilineal to matrilineal descent is impossible.

24But be this as it may, the nineteenth century believers in the priority of matriliny appeared to be on very firm ground:

  • 30 Ibid., p. 185.

So logical, so closely reasoned, and so apparently in accord with all known facts was this hypothesis that from its pioneer formulation by Bachofen in 1861 to nearly the end of the nineteenth century it was accepted by social scientists practically without exception.30

25Nonetheless, according to Murdock, the Bachofen-Morgan evolutionary schema remains impossible to defend:

  • 31 G. P. Murdock, Culture and Society, Pittsburgh, University of Pittsburgh Press, 1965, p. 132.

The expansion of ethnographic knowledge and the resulting recognition of the role of diffusion in culture change made these unilinear theories increasingly suspect during the early decades of the twentieth century, and they are now, of course, universally discredited by competent scholars.31

26For his part, Marvin Harris simply writes that matriarchies:

  • 32 M. Harris, The Rise of Anthropological Theory, op. cit., p. 85.

[…] were actually postulated as the primordial condition of mankind by various theoreticians who lived in the nineteenth century. Friedrich Engels, for example, who got his ideas from the American anthropologist Lewis Henry Morgan, believed that modern societies had passed through a matriarchal phase during which descent was reckoned exclusively in the female line and women were politically dominant over men. Many modern-day women’s liberationists continue to believe in this myth [… ]32

Radcliffe-Brown and the case of the mother’s brother

27The avunculate is the peculiar relationship which exists in many African societies between a man and his sister’s son. Such a special relationship is not a universal feature of human kinship as such. Obviously, then, it cannot be explained by reference to any assumed universal sociological principle. It is a characteristic of certain kinship systems but not others, and it is the differences which need to be explained.

28The avunculate in this way differs from the mother-son relationship. We feel no special need to explain the existence of a relationship linking a mother and her child. On biological grounds we expect some kind of unity – commensal and so on – between a woman and her young son under normal conditions in virtually any conceivable society. But we do not universally expect the woman’s adult brother to be included in this relationship. In modern western societies, in fact, a man is not united in any special way with his sister’s son. It is this which, even at the most superficial and immediate level, makes us feel the need to “explain” the special features of avunculate when we encounter them.

  • 33 A. R. Radcliffe-Brown, Structure and Function in Primitive Society, London, Cohen and West, 1952, p (...)
  • 34 Ibid. p. 32-48.

29Let us follow Radcliffe-Brown in assuming that, in the tension or choice between matriliny and patriliny in any society, the point at issue is really this: “To whom does the child belong?”33 If it belongs to the mother’s kin, descent is matrilineal. If it belongs to the father’s group, descent is patrilineal. As far as the abstract question of “descent” is concerned, that is all. However, we may further follow Radcliffe-Brown in assuming that in practice, as people actually implement their descent-rule, they find themselves forced to do other things as well.34

  • 35 R. Fox, Kinship and Marriage, Middlesex, Penguin, 1967, p. 37 et p. 40.

30For example, to the extent that the mother-child group forms an indivisible unit, the group possessing the child must include the mother as well. Robin Fox’s words seem relevant here: “The irreducible and elementary social grouping is surely the mother and her children. Whatever else happens, this unit has to survive for the species to survive […] Whether or not father can be persuaded to stay at home is another matter”.35 We would normally expect a mother to be grouped in some way with her young or dependent child.

31In patrilineal systems, not only children but their mothers, too, are in effect incorporated into the descent group of the husband/father. “In effect”, because at first, of course, the exogamy rule prevents women from being members of their husbands’ kin-group. Patriliny requires a wife to change her descent group membership. Matriliny requires neither wives nor husbands to make such achange. This enables us to understand the fundamental social contrasts between patrilineal systems and matrilineal ones. Radcliffe-Brown brings out the contrast with his usual precision:

The contrast between father-right and mother-right is one of two types of marriage. A woman is by birth a member of a sibling group; strong social bonds unite her to her brothers and sisters. By marriage she enters into some sort of relation with her husband. To provide a stable structure there has to be some sort of institutional accommodation of the possibly conflicting claims and loyalties, as between a woman’s husband and her brothers and sisters. There are possible two extreme and opposite solutions, those of father-right and mother-right, and an indefinite number of compromises.

In the solution provided by mother-right the sibling group is taken as the most important and permanent unit in social structure. Brothers and sisters remain united, sharing their property, and living together in one domestic group. In marriage the group retains complete possession of a woman; her husband acquires no legal rights at all or a bare minimum […] Rights of possession over children therefore rest with the mother and her brothers and sisters. It is these persons to whom the child must go for every kind of aid and comfort, and it is they who are entitled to exercise control or discipline over the child […].

  • 36 A. R. Radcliffe-Brown (ed.), “Introduction”, in African Systems of Kinship and Marriage, London, Ox (...)

The solution offered by father-right is opposite. Possession of a wife, and therefore of the children of her body, are surrendered by marriage to her husband and his kin […] The mother’s kin, her brothers and sisters, in this kind of marriage, have no rights over the children, who, in turn, have no rights over them. The jural bonds between a woman and her siblings are severed by her marriage.36

32In other words, matriliny is inseparable from brother-sister unity, while patriliny contradicts it. Insofar as father-right prevails, children are incorporated, along with the mother, in their father’s descent-group, divorcing the children’s mother from her brother. As Radcliffe-Brown puts it in the passage just quoted: “The jural bonds between a woman and her siblings are severed by her marriage”. This is why the avunculate appears puzzling when encountered in association with patriliny. Under matriliny, there is no such puzzle. Brothers are united with their sisters. The bond between these brothers and their sisters’ children follows naturally. In a patrilineal system, on the other hand, women are incorporated, along with their children, into their husbands’ descent-groups. If, despite this, children are considered to belong in some sense to their mother’s brothers, we are faced with a theoretical challenge.

33All of this seems fairly straightforward until we come to Radcliffe Brown’s celebrated 1924 essay, The Mother’s Brother in South Africa. As if unaware of his writings on the subject elsewhere, Radcliffe-Brown found it necessary to deny any logical connection between brother-sister unity and matriliny. His polemical target was Henri Junod’s (1912) argument that the avunculate among the Thonga of South Africa was a matrilineal feature of an otherwise largely patrilineally organized kinship system, suggesting that these people once possessed matrilineal clans.

34To Junod, it seemed obvious that the strong husband-wife bonds of the present-day Thonga owed their origin to the relatively recent lobola system. Instead of having to visit his wife in her village and work for her and her kin, a husband with sufficient cattle could be permitted to remove his bride and take her home with him, in effect exchanging her and any future offspring for the cattle. In this way, her links with her natal village and hence with her brothers were weakened. Nevertheless, it was the indigenous view that despite lobola, the “true” ties of a woman remained with her natal village, and that this applied also to her children. It was everywhere accepted that it was “only the cattle” which took a woman and her children from her brothers and other natal kin.

35In religious sentiment, the child’s bond with its mother’s relatives was conceived as fundamental, while that with the father and his relatives was “only on account of the oxen”:

  • 37 H. A. Junod, The Life of a South African Tribe, Neuchatel, 1912, vol. 1, p. 294.

In fact, as Mankehulu said in his picturesque language, “as regards sacrifices (timhamba), the mother’s relatives mostly perform them. They are the stem. My father is the stem on account of the oxen, my mother is the true stem; she is the god; she makes me grow. Should she die when I am an infant, I will not live. At the village of my mother it is at the god’s (ka mamana hi ko psikwembyen)”.37

  • 38 Ibid., p. 263.

36If matriliny allocates children to the mother’s kin, then in religious sentiment, this principle was still very much alive among the Thonga of Junod’s time. But matriliny remained operative not only in this context of sentiment. It was also present as a constant threat motivating men to keep up the lobola system. Husbands had no automatic rights in their own offspring. On the contrary, men felt that they could acquire such rights only through considerable sacrifice – by making substantial payments to their wives’ kin. Father-right clearly rested on the lobola. Wherever a woman had not been « paid for », her child would bear her lineage name and would continue to live in the mother’s brother’s village. Whenever Junod discussed with men the possibility of government legislation prohibiting lobola payments they would reply: “Who will guarantee to us the possession of our children if lobola is suppressed?”38

37Matrilineal principles also asserted themselves in rights of property inheritance. When a man died and his possessions were inherited, his sister’s sons had first choice. Obviously, this right could hardly have been fully exercised without undermining the entire principle of patrilineal inheritance. If lobola had been paid by the man to his wife’s relatives, then his own offspring were his legal heirs. The contradiction was overcome by a customary stipulation that the “true” (matrilineal) heirs should voluntarily give way to the “legal” (patrilineal) ones. The sister’s sons, according to the formulation of the people themselves, were the original heirs, but they allowed the deceased’s sons first choice, and then made their own choice of implements themselves. “In this way”, wrote Junod,

  • 39 Ibid., p. 257.

ba nyiketa pfindla, they give over the inheritance to the legal heirs This is a most vivid representation of a right which no longer exists, having in fact become obsolete, but which asserts itself however in virtue of an old custom.39

38There is no need here to multiply examples. Plainly, this kinship system embodied no consistent logic. In many respects it was a system of double unilineal descent, even if the matrilineal descent groups were not property-owning lineages. Murdock summarizes the evidence that the Thonga were indeed in the throes of transition from matriliny to patriliny, just as Junod originally claimed:

  1. Avunculocal residence is customary in childhood. Boys and. girls, as soon as they are weaned, go to live in the village of their maternal uncle and remain there for several years, in the case of girls sometimes until they are fully grown. (This is normal in matrilineal societies with avunculocal residence, e. g. the central and western tribes of the Central Bantu).
  2. When a man has no patrilineal heirs to carry on his line, he may require one of his sisters to remain in his settlement. Her children reside in his house, and the males continue his lineage and clan. (This re-capitulates the household pattern of the matrilineal eastern Central Bantu).
  3. If a man has no immediate patrilineal heirs, his sister’s sons inherit his property in preference to remoter patrilineal kinsmen. Even when he has patrilineal heirs, his sisters’ sons can claim certain items from his estate, e. g. his spears. (Inheritance by sisters’ sons is normal in matrilineal societies.)
  4. A maternal uncle has a right to a definite share in the bride-price received for a sister’s daughter. (He can ordinarily claim most of it in a matrilineal society.)
  5. The officiant at all sacrifices in a man’s life-crisis ceremonies is his mother’s brother. (In a matrilineal society this duty naturally falls to a man’s maternal uncle as the head of his lineage.)40

39Murdock argues that all the Central Bantu tribes must once have been matrilineal. A number of them, he writes, still are, and these are the ones that possess few cattle. In his view, the transition to patriliny was in many cases accomplished only recently, as a direct result of the introduction of cattle. If this view is correct, then the severing of the bond between a man and his sister’s children was the final result of a process by which women were separated from their natal kin by means of the lobola system. Each increase in cattle ownership within a particular tribe would have made possible a raising of the level of lobola payments, hence a corresponding intensification of the rights of husbands (and husbands’ kin) over their wives. A concise formulation of this evolutionary argument has been put forward by Harold Schneider:

  • 41 H. Schneider, Economic Man. The Anthropology of Economics, New York, Macmillan, 1974, p. 145

In Africa the most thoroughgoing patrilineal societies seem to be those in which livestock, particularly cattle, are an important asset – that is, where the number of cattle per person is about one or more. Where there is no asset that can be transferred for compensation, we find matrilineal-matrilocal systems. As compensation increases the husband is allowed to remove his wife to his father’s home, so that the system becomes matrilineal-patrilocal, but with the male children moving to live with their mother’s brother when they marry (avunculocal residence), since mother’s brother owns them despite the patrilocality of the marriage. As compensation reaches a higher level the system shifts to patrilineal-patrilocal, and this increases in intensity as the amount paid increases until at the other extreme one gets patrilineal-patrilocal systems in which there is no divorce.41

40At every intermediate stage in this process, there is always something left of the old brother-sister unity, but where it retains its force this can only be despite the increasing patrilocality and patrilineality of the current pattern of residence and descent.

  • 42 D. Aberle, “Cross-Cultural Perspective”, in D. M. Schneider and K. Gough (eds.), Matrilineal Kinshi (...)
  • 43 R. Mace and C. Holden, “Evolutionary Ecology and Cross-Cultural Comparison: The Case of Matrilineal (...)
  • 44 C. J. Holden, R. Sear and R. Mace, “Matriliny as daughter-biased investment”, Evolution and Human B (...)

41In essential conformity with Morgan’s scheme, the rise of alienable property is nowadays widely recognised to be the crucial factor cementing marital bonds at the expense of brother-sister solidarity throughout much of sub-Saharan Africa. As David Aberle put it, “the cow is the enemy of matriliny”.42 Following in the footsteps of Murdock’s cross-cultural comparative work, Mace and Holden’s43 (1999) phylogenetically controlled analysis confirmed a negative correlation between African matriliny and cattle owning. In their most recent analysis of matriliny as daughter-biased investment, Holden, Sear, and Mace comment, “The two factors Morgan identified, heritable wealth and paternity uncertainty, remain central to our understanding of variation in matriliny and patriliny in human social organisation”.44

42At this point, we may turn to the argument put forward by Radcliffe-Brown in his celebrated essay on the mother’s brother. How does he account for the peculiar unity between brothers and sisters which exists even in some of the most strongly patrilineal societies in Africa?

  • 45 A. R. Radcliffe-Brown, “The Mother’s Brother in South Africa”, South African Journal of Science,21, (...)

43Unfortunately, we cannot answer this question because – surprising as it may seem – Radcliffe-Brown offers no explanation. Or rather, he adopts two mutually inconsistent strategies. When discussing matriliny, he identifies this descent rule as inseparable from brother-sister unity. On the other hand, in his essay on the mother’s brother in South Africa, his whole purpose is to demonstrate that an explanation can be found which does not involve matriliny. To what, then, should we ascribe the strong bond found in certain societies between brother and sister? Radcliffe-Brown does not ascribe it to anything. Instead, he treats it as a metaphysical principle – something which, precisely because of its wide distribution, requires no explanation. Brother-sister unity is presented as a particular expression of the “equivalence of siblings », a “principle” which for unexplained sociological reasons is “most commonly adopted in primitive society”.45

  • 46 A. R. Radcliffe-Brown (ed.), “Introduction”, op. cit.; quoted by M. Fortes, Kinship and the Social (...)

44Elsewhere, Radcliffe-Brown writes46: “The unit of structure everywhere seems to be the group of full siblings – brothers and sisters”.46 Fortes writes that this principle is

  • 47 Loc. cit.

one of the few generalizations in kinship theory that, in my opinion, enshrines a discovery worthy to be placed side by side with Morgan’s discovery of classificatory kinship; and, like Morgan’s, it has been repeatedly validated and has opened up lines of inquiry not previously foreseen.47

  • 48 R. Needham (ed.), Rethinking Kinship and Marriage, London, Tavistick, 1971, p. LXXXVI.
  • 49 D. M. Schneider, “The Distinctive Features of Matrilineal Descent Groups. Introduction”, in D. M. S (...)
  • 50 G. P. Murdock, Social Structure, Macmillan, New York, 1949, p. 121.

45Needham, however, observes that this principle has obviously not been “repeatedly validated” – in some societies the brother-sister group is the unit of structure, while in others it is not.48 Schneider notes that the “interdependence of brother and sister” is characteristic of matrilineal descent groups but is not characteristic of patrilineal ones.49 And Murdock attacks the whole idea of using timeless “principles” to explain social facts50:

In the first place, the alleged principles are mere verbalizations reified into causal forces. In the second, such concepts as “equivalence of brothers” and “necessity for social integration” contain no statements of the relationships between phenomena under varying conditions, and thus lie at the opposite extreme from genuine scientific laws. Thirdly, being unitary in their nature, they provide no basis for interpreting cultural differences; they should produce the same effects everywhere.

  • 51 A. R. Radcliffe-Brown, “The Mother’s Brother in South Africa”, art. cit.
  • 52 Ibid., p. 25.
  • 53 Ibid., p. 19-20.

46The nub of Radcliffe-Brown’s claim is that the avunculate reflects brother-sister solidarity but not matrilineal descent. We are informed, to begin with, that “there are certain fundamental principles or tendencies which appear in all societies, or in all those of a certain type”.51 As an example of a “certain type”, Radcliffe-Brown cites “primitive society”, in which, we are told, “there is a strongly marked tendency to merge the individual in the group to which he or she belongs”.52 Which precise “group” Radcliffe-Brown has in mind is not very clear. Is it the husband-wife group, the territorial group, the group of full siblings or what? Radcliffe-Brown apparently means the brother-sister group. Skillfully avoiding any mention of matriliny, he argues that in societies in which “the classificatory system of kinship reaches a high degree of development”, husbands and wives are grouped apart, while brothers and sisters are grouped together.53 The crux of the argument is that in these “certain societies” wives are not “merged” in the groups of their husbands, but are “merged” in those of their brothers.

47Radcliffe-Brown views the avunculate primarily as a reflection of “sentiment”. He shows little interest in the rich institutional details provided by Junod: avunculocal residence in childhood, rights of inheritance, rights in brideprice, initiation, sacrifice and so forth. For him, the basic feature is the sentimental “indulgence” towards the sister’s son on the part of the maternal uncle, together with certain corresponding sentiments on the part of the child. The central question, from this psychological perspective, is not “How or why is the institution maintained?” but “How do the child’s sentiments come to be formed?” Radcliffe-Brown answers by (a) describing certain aspects of the social structure of so-called “primitive” societies and (b) showing that there is some fit with the growing child’s sentiments.

48In patrilineal societies where the avunculate is found, authority over children is in the hands of fathers, while sibling-unity remains strong. Consequently, argues Radcliffe-Brown, paternal authority is to some extent shared between fathers and their mature siblings, male and female. « Tenderness and indulgence », on the other hand, are to be expected from a child’s mother, hence also from her male and female siblings. All this predetermines the formation of sentiments in the child’s mind:

  • 54 Ibid., p. 27-28.

The pattern of behaviour towards the mother… is extended with suitable modifications to the mother’s sister and to the mother’s brother, then to the group of maternal kindred as a whole, and finally to the maternal gods, the ancestors of the mother’s group. In the same way the pattern of behaviour towards the father is extended to the father’s brothers and sisters, and to the whole of the father’s group… and finally to the paternal gods.54

49We are asked to accept this as an explanation for the fact that children after weaning live with their maternal uncles, see their maternal relatives as the “true stem”, belong to their mother’s group if the lobola has not been paid and so on. Radcliffe-Brown’s aim is not just to modify or improve Junod’s interpretation but to discredit it and replace it with an entirely contrasting explanation. In his own words:

  • 55 Ibid., p. 14.

The point of the paper on the mother’s brother may be said to be to contrast with the explanation by pseudo-history the interpretation of the institution to which it refers as having a function in a kinship system with a certain type of structure.55

50Murdock comments:

  • 56 G. P. Murdock, Africa. Its peoples and their culture history, op. cit., p. 378.

The indicated contrast is clear. In the eyes of the present writer, however, it takes the form of an opposition between sound historical scholarship and untrammelled sociological speculation.56

  • 57 H. A. Junod, The Life of a South African Tribe, op. cit., p. 232-233.

51A more generous verdict would be that Radcliffe-Brown inadvertently takes us back to matriliny by another route. In particular, his argument helps explain the difference between (a) the avunculate in a society which has not shifted to patriliny and (b) the avunculate following this transition. Junod himself was unable to explain why, among the Thonga, the avunculate so one-sidedly favoured the child. A boy could “do as he pleased with his maternal uncle, eating his food, flirting with his wife or wives and wandering in and out of his house without even asking, while the uncle himself had virtually no reciprocal rights in or authority over the boy”.57 Such indulgence is not necessarily typical of matrilineal societies, in which the relationship is often much more balanced, a man usually having definite authority over his sister’s son.

52So the avunculate in a patrilineal society cannot simply be treated as a survival. The institution is transformed in the process, and no longer serves quite the same functions. Radcliffe-Brown’s contrast between « authority » and « indulgence » in this context is not without relevance. Let us assume that despite the loss of the mother’s brother’s authority, the child’s relationship to its maternal uncle remains in many respects intact following the transition. It is easy to imagine why a newly assertive father might be less tolerant of the mother’s brother’s residual rights in a boy than of that same child’s reciprocal rights in his maternal uncle. If this is accepted, we can perhaps understand why the transition to patriliny would soften the avunculate – removing from it the element of adult authority – while leaving the child’s rights in other respects intact.

Among the Thonga, as Radcliffe-Brown points out, the father’s sister is a relative very much to be respected and perhaps even feared. Why is this pattern so typical of societies in which the mother’s brother has become a notably indulgent figure? In a transitional society, a woman, despite patrilocal marriage, will still remain closely bound to her brother. Under the earlier, matrilineal, system, she would have been bound to him still more closely, but this would not have given her any authority over her brother’s child. Once father-right has been consolidated, however, a woman’s brother will have begun to exercise authority over his own biological offspring. To the extent that a woman remains in some sense her brother’s equivalent, she must inevitably share in exercising such paternal authority.

Lévi-Strauss, Leach and Goody

53Is it fathers who give away their daughters in marriage, or brothers who give away their sisters? This depends upon the descent rule. In real life, of course, both relatives may be involved, but where fathers traditionally “give away” their daughters, this can only be because these female relatives are in some sense theirs to give. If young women are considered the responsibility of their brothers and mothers’ brothers, then as they are married it will not be their fathers but their male kin who give them away. Whether a bride is given away as a daughter or as a sister has therefore some importance in any discussion of the avunculate.

54Lévi-Strauss obliterates this distinction by writing of marriage as a system in which one man gives another man “a daughter or sister”. This would not matter much, perhaps, were it not for the fact that the essay in which Lévi-Strauss does this is the place where he ventures his own explanation for the avunculate. For Lévi-Strauss, the the avunculate just exists. Men simply do have sisters, and they are obliged to surrender them in marriage. Hence the avunculate.

  • 58 C. Lévi-Strauss, Structural Anthropology,2 vol., Harmondsworth, Penguin, 1968, vol. 1, p. 46.

55Lévi-Strauss fails to explain why the avunculate predominates in certain African and other tribal societies while playing little role in, say, the industrialised west. We are simply told that the relation between a man and his sister is one aspect of the “primitive and irreducible character of the basic unit of kinship”, while this itself “is actually a direct result of the universal presence of an incest taboo”.58 Then follows the passage in which Lévi-Strauss slips in the words « daughter or sister » quite unobtrusively and as if the difference were irrelevant:

  • 59 Loc. cit.

This is really saying that in human society a man must obtain a woman from another man who gives him a daughter or sister. Thus we do not need to explain how the maternal uncle emerged in the kinship structure: he does not emerge – he is present initially. Indeed, the presence of the maternal uncle is a necessary precondition for the structure to exist.59

56This argument is patently weak. First, it can have force only where women are in fact “given away” in marriage as sisters by their brothers. Where they are given away as daughters by their fathers, then the dynamic will be quite different.

57Reflecting this, we find that in many societies there is no “pivotal” or “special” relationship between a boy and his maternal uncle. Meanwhile, in other cases, a boy lives with this relative for several years, or inherits his property or his widows, or can eat or use or “steal” his property or food at any time, etc. Why this difference? We all observe incest-taboos, so the avunculate cannot conceivably be explained as “a direct result of the universal presence” of this taboo.

  • 60 G. C. Homans and D. N. Schneider, Marriage, Authority and Final Causes, Glencoe, The Free Press, 19 (...)
  • 61 Ibid., p. 22.
  • 62 D. M. Schneider, “The Distinctive Features of Matrilineal Descent Groups. Introduction”, in D. M. S (...)

58Homans and Schneider adopt a very similar position.60 That is, they treat brother-sister solidarity as a sociological axiom requiring no explanation. We are told, for example, that a boy behaves towards his mother’s brother in the same way as he does towards his mother “as a result of the identification of the mother with her brother, who is her protector […]”.61 The authors are here referring to “the patrilineal complex”, it should be noted. But in the “patrilineal complex” (as opposed to its matrilineal counterpart), there is evidently no reason to assume that a woman’s brother must act as her protector. Where this does happen, it must be explained: it is not inherent in patriliny, the underlying dynamic of which on the contrary makes the husband a woman’s protector at the expense of her brother.62 Once again, Homans and Schneider assume as axiomatic the very thing which has to be explained, managing to dissociate matriliny from the avunculate only by adopting this polemical stratagem.

  • 63 J. Goody, “The mother’s brother and the sister’s son in West Africa”, in Comparative Studies in Kin (...)
  • 64 Ibid., p. 78.

59Goody takes a woman’s unity with her brother as given, but then does something for which his professional colleagues have never forgiven him.63 He leaves the distinct impression in the reader’s mind that even in patrilineal societies, matrilineal descent and the avunculate are linked. Worse, he even uses a formulation which would seem to make of the avunculate a matrilineal “survival”. He writes, for example, that among the West African Lo Wiili, the “legitimization” of a marriage gives the husband’s patriclan rights in the child of this marriage. In this way, the child is “excluded” from the mother’s group. Its claim to share the property of the mother’s brother or to inherit from him is thereby denied. But father-right is not consistently or rigidly applied. For one thing, a woman remains united with her brother – she continues to belong to her own patriclan, to which her brother also belongs. Moreover, her child, despite exclusion from her group, “has nevertheless a shadowy claim upon the group by virtue of his mother’s position”64. The avunculate is in this way explained.

  • 65 E. Leach, Rethinking Anthropology, University of London, The Athlone Press, 1961, p. 16.

60In a discussion of the avunculate among the Lakher of India, Leach denounces Goody’s position as “only marginally different” from the “totally mistaken” nineteenth century evolutionists’ “doctrine” of matrilineal priority.65 Schneider subsequently supports Leach:

  • 66 D. M. Schneider, “Some Muddles in the Models: Or, How the System Really Works”, in M. P. Banton (ed (...)

Leach’s disquiet over such terms as “shadowy claim”, “submerged rights” and “residual sibling” is not entirely misplaced, in my view… The point is that if the rights of the sister’s son over bits of property held by the mother’s brother, including perhaps the mother’s brother’s wife, whom sister’s son may “inherit”, are rights which are based on the consanguineal tie through his mother, then these rights are transmitted exactly as is descent group membership. These are rights, that is, that are based on descent. If this is so, the problem is whether matrilateral filiation is not in this sense a “descent rule” and so all patrilineal descent systems with matrilateral filiation are by definition double unilineal descent systems.66

  • 67 In this short essay, my focus has been upon the roles played by Boas, Malinowski, Radcliffe-Brown a (...)

61The accusation here is that Goody once again links the avunculate with a trace or survival of matriliny. But ever since Radcliffe-Brown, it has been one of the central purposes of professional anthropologists throughout the West to exorcise this spectre. To me it seems that the more these anthropologists try, the less convincing they are. The more they try to dissociate the avunculate from matriliny, the more entangled in matriliny they become. Perhaps the time has come for anthropologists to question the usefulness of the entire enterprise on which they and their teachers have been engaged, now, for the best part of a century.67

Notes

1 J. Lederer, The Discoveries of John Lederer, in Three Several Marches from Virginia, to the West of Carolina and Other Parts of the Continent, London, Samuel Heyrick, 1672, p. 10-11; quoted in S. Tax, “From Lafitau to Radcliffe-Brown”, in F. Eggan (ed.), Social Anthropology of the North American Tribes, Chicago, University of Chicago Press, 1955, p. 445-481; p. 445.

2 J. T. Lafitau, Mœurs des sauvages amériquaines, comparées aux mœurs des premiers temps, Paris, Saugrain l’aîné, 1724, 1, p. 71-72.

3 A. Ferguson, An Essay on the History of Civil Society [1767], Philadelphia, A. Finley, 1819, p. 126.

4 J. J. Bachofen, Myth, Religion and Mother-right. (Selected Writings), New Jersey, Princeton University Press, 1973, p. 70.

5 Ibid., p. 135-136.

6 Ibid., p. 71.

7 J. P. McLellan, Primitive Marriage, Edinburgh, Adam and Charles Black, 1865, p. 123; quoted in S. Tax, “From Lafitau to Radcliffe-Brown”, in F. Eggan (ed.), Social Anthropology of the North American Tribes, op.cit., p. 445-481; p. 455.

8 L. H. Morgan, Houses and House-Life of the American Aborigines, Chicago and London, University of Chicago Press, 1881, p. 126.

9 Ibid., p. 126-128.

10 See M. Harris, The Rise of Anthropological Theory, London, Routledge, 1969, p. 189.

11 F. Engels, The Origin of the Family, Private property and the State [1884], New York, Pathfinder Press, 1972, p. 181-182.

12 K. Marx, “Economic and Philosophical Manuscripts” [1844], in D. McLellan (ed.), Karl Marx: Selected Writings, Oxford, Oxford University Press,2nd ed.,2000, p. 83-121; p. 96.

13 K. Marx and F. Engels, “The German Ideology” [1846], in D. McLellan (ed.), ibid., p. 175-208; p. 185.

14 L. H. Morgan, Houses and House-Life of the American Aborigines, op. cit., p. 128.

15 F. Engels, The Origin of the Family, Private Property and the State, op. cit., p. 68.

16 Ibid., p. 75.

17 R. Lowie, “Evolution in Cultural Anthropology: A Reply to Leslie White”, American Anthropologist, 48, 1946, p. 223-233 : reprinted in C. du Bois (ed.), Lowie’s Selected papers in Anthropology, Berkeley, University of California Press, 1960, p. 418.

18 F. Boas, Race, Language and Culture, New York, Macmillan, 1940, p. 635.

19 R. Lowie, “Evolution in Cultural Anthropology: A Reply to Leslie White”, art. cit., p. 420.

20 R. Lowie, “Social Organization”, American Journal of Sociology,20, 1914, p. 68-167; reprinted in C. du Bois (ed.), op. cit., p. 28.

21 M. Harris, The Rise of Anthropological Theory, op. cit., p. 305.

22 Loc. cit.

23 G. Reichard, “Social Life”, in F. Boas (ed.), General Anthropology, New York, Heath, 1938, p. 409-486; p. 425.

24 J. R. Swanton, “The Social Organization of American Tribes”, American Anthropologist, 7, 1905, p. 663-673.

25 R. Lowie, Primitive Society, New York, Bini and Liveright, 1920; A. Kroeber, Anthropology, New York, Harcourt Brace, 1923, p. 355-358; A. Goldenweiser, “The Social Organization of the Indians of North America”, Journal of American Folk-Lore,27, 1914, p. 411-436.

26 G. P. Murdock, Social Structure, Macmillan, New York, 1949, p. 189.

27 Ibid., p. 185.

28 Loc. cit.

29 Ibid., p. 190.

30 Ibid., p. 185.

31 G. P. Murdock, Culture and Society, Pittsburgh, University of Pittsburgh Press, 1965, p. 132.

32 M. Harris, The Rise of Anthropological Theory, op. cit., p. 85.

33 A. R. Radcliffe-Brown, Structure and Function in Primitive Society, London, Cohen and West, 1952, p. 40.

34 Ibid. p. 32-48.

35 R. Fox, Kinship and Marriage, Middlesex, Penguin, 1967, p. 37 et p. 40.

36 A. R. Radcliffe-Brown (ed.), “Introduction”, in African Systems of Kinship and Marriage, London, Oxford University Press, 1950, p. 77-78.

37 H. A. Junod, The Life of a South African Tribe, Neuchatel, 1912, vol. 1, p. 294.

38 Ibid., p. 263.

39 Ibid., p. 257.

40 G. P. Murdock, Africa. Its Peoples and their Culture History, New York, McGraw-Hill, 1959, p. 378.

41 H. Schneider, Economic Man. The Anthropology of Economics, New York, Macmillan, 1974, p. 145

42 D. Aberle, “Cross-Cultural Perspective”, in D. M. Schneider and K. Gough (eds.), Matrilineal Kinship, Berkeley and Los Angeles, University of California Press, 1961, p. 655-727; p. 680.

43 R. Mace and C. Holden, “Evolutionary Ecology and Cross-Cultural Comparison: The Case of Matrilineal Descent in Sub-Saharan Africa”, in P. C. Lee (ed.), Comparative Primate Socioecology, Cambridge, Cambridge University Press, 1999, p. 387-405; K. Maddock, The Australian Aborigines. A portrait of their Society, Harmondsworth, Penguin, 1974.

44 C. J. Holden, R. Sear and R. Mace, “Matriliny as daughter-biased investment”, Evolution and Human Behavior,24,2003, p. 99-112; p. 110.

45 A. R. Radcliffe-Brown, “The Mother’s Brother in South Africa”, South African Journal of Science,21, 1924, p. 542-555; reprinted in A. R. Radcliffe-Brown, Structure and Function in Primitive Society, op. cit., p. 15-48; p. 18.

46 A. R. Radcliffe-Brown (ed.), “Introduction”, op. cit.; quoted by M. Fortes, Kinship and the Social Order, London, Routledge, 1970, p. 76.

47 Loc. cit.

48 R. Needham (ed.), Rethinking Kinship and Marriage, London, Tavistick, 1971, p. LXXXVI.

49 D. M. Schneider, “The Distinctive Features of Matrilineal Descent Groups. Introduction”, in D. M. Schneider and K. Gough (eds.), Matrilineal Kinship, Berkeley, University of California Press, 1961, p. 1-29; p. 11.

50 G. P. Murdock, Social Structure, Macmillan, New York, 1949, p. 121.

51 A. R. Radcliffe-Brown, “The Mother’s Brother in South Africa”, art. cit.

52 Ibid., p. 25.

53 Ibid., p. 19-20.

54 Ibid., p. 27-28.

55 Ibid., p. 14.

56 G. P. Murdock, Africa. Its peoples and their culture history, op. cit., p. 378.

57 H. A. Junod, The Life of a South African Tribe, op. cit., p. 232-233.

58 C. Lévi-Strauss, Structural Anthropology,2 vol., Harmondsworth, Penguin, 1968, vol. 1, p. 46.

59 Loc. cit.

60 G. C. Homans and D. N. Schneider, Marriage, Authority and Final Causes, Glencoe, The Free Press, 1955.

61 Ibid., p. 22.

62 D. M. Schneider, “The Distinctive Features of Matrilineal Descent Groups. Introduction”, in D. M. Schneider and K. Gough (eds.), Matrilineal Kinship, op. cit., p. 11.

63 J. Goody, “The mother’s brother and the sister’s son in West Africa”, in Comparative Studies in Kinship, London, Routledge and Kegan Paul, 1969.

64 Ibid., p. 78.

65 E. Leach, Rethinking Anthropology, University of London, The Athlone Press, 1961, p. 16.

66 D. M. Schneider, “Some Muddles in the Models: Or, How the System Really Works”, in M. P. Banton (eds.), The Relevance of Models for Social Anthropology, London, Tavistock Publications, 1965, p. 54.

67 In this short essay, my focus has been upon the roles played by Boas, Malinowski, Radcliffe-Brown and their mid-twentieth century successors. For current debates and in particular for the modern “grandmother hypothesis” in relation to early kinship, see C. Knight and C. Power, “Grandmothers, Politics and Getting Back to Science”, in E. Voland, A. Chasiotis and W. Schiefenhövel (eds.), Grandmotherhood: The Evolutionary Significance of the Second Half of Female Life, New Brunswick, Rugers University Press,2005, p. 81-98.

Auteur

Actuellement Principal Lecturer au sein du Radical Anthropology Group ainsi que Senior Honorary Research Fellow au sein du département d’Anthropologie, University College (Londres). Son dernier ouvrage s’intitule Decoding Chomsky : Science and Revolutionary Politics (Yale University Press,2016).

© Éditions Rue d’Ulm, 2017

Licence OpenEdition Books

Acheter

Volume papier

decitre.fr
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search