Télécharger format PDF
Genres & générations
« Gender Logic and family ties in Moslem societies », Laurent Barry
Anthropologue, Laurent Barry est membre du Laboratoire d'Anthropologie Sociale du Collège de France ; il y est responsable de l'équipe « Parenté » (a). Il travaille depuis plusieurs années à renouveler le traitement de cette question qui, après avoir été le thème privilégié des années structuralistes, est passée de mode chez les chercheurs, alors que dans le même temps, « sur le terrain » – en Occident, à tout le moins – la famille « traditionnelle' » se voyait remise en question quand elle n'était pas concurrencée par des formes inédites, subies ou voulues (monoparentalité, recomposition, homoparentalité...) (b).
Dans ce court article Laurent Barry montre que, dans les sociétés musulmanes, la parenté est largement déterminée par les catégories de genre véhiculées par l'Islam. En effet, la représentation du genre portée par l’Islam tend à contraster fortement – sinon à dissocier – le rôle dévolu à l’un et l’autre sexe dans la parenté. Si l’homme est souvent pensé ici comme le garant de l’identité politique et légale de la communauté, c’est pourtant la femme seule qui la reproduit et la légitime « naturellement ». L’auteur montre alors que l’opposition générale du masculin et du féminin qui traverse tout le domaine de la parenté dans les sociétés musulmanes se retrouve dans certaines de leurs institutions plus particulières, à savoir dans leurs pratiques de mariages préférentiels et dans la terminologie de parenté qu’elles utilisent.
Ce qui confirme – et ce n'est pas sans importance à l'heure où les questions de filiation et plus généralement d'identité se voient reprises de plus en plus en des termes biologiques – que la parenté est irréductible à la consanguinité et plus largement que les fondements de l'organisation sociale sont moins « naturels » qu'idéologiques.
(a) On pourra, pour approfondir, lire son livre récent, La parenté, (Gallimard, 2008). À signaler aussi, son remarquable « glossaire de la parenté », précis, clair et fort utile (PDF).
(b) Sur la question de l'homoparentalité, on trouvera dans La_Revue, no3, les articles du psychanalyste Alain Ducousso-Lacaze et de la sociologue Martine Gross.
Gender Logic and family ties
in Moslem societies
If it is often conceived that the concept of gender can exert a peripheral and indirect influence, although there are few fields in which one can prove that it plays a central and primordial role in the logic of the construction of social categories. However, it is precisely a dichotomic representation of gender such as it is conveyed by certain ideological values of Islam which seems decisive in the construction of two major social institutions for the societies which follow this religious obedience, namely the rules of marriage and kinship terminology.
The cultural diversity in the manners of building the conceptual categories founded on gender is one of the topics best explored these last years by anthropologists. However, they show until now a predilection for the study of certain particular configurations - primarily those systems known as “Dravidians” - where the logic of the major anthropological theories, specially that of structural-functionalism, may apply. In this model, the role of women is limited to that of “objects of exchange”, i.e., “goods” circulating between descent groups. The social actors are groups of men mutually exchanging their daughters or their sisters. This logic exposes the ideology of male domination which prevails in many cultural contexts, but does little to help us to understand how did human societies think about the social construction of the sexes. However, there is a particular configuration that the ethnologists were unaware of until now where the construction of gender categories cannot be perceived any more like a simple testimony of the instrumentalisation of women by men. This configuration associates, in a tripartite pattern, the system of representation of the gender categories of Islam, a marriage pattern founded on the practice of what anthropologists call the “Arab marriage” and a very particular kinship terminology - a manner of naming the relatives - the “Sudanese terminology”.
The association of these three large features, marriage, kinship terminology, and Islam, seems in fact often widely independent of the geographical area concerned. Asian, African or near Eastern, a great number of Moslem societies adopted this model of endogamic marriages - “Arab marriage” - which privileges (generally within patrilineal descent groups) the search of a spouse within agnatic relatives: especially the marriage with the father’s brother’s daughter. Asia however has more diversity than elsewhere where this remarkable association between Islam and this form of marriage knows little exception. Thus, certain Moslem communities in Malaysia or Indonesia which grant a strong “political” role to women - for example the matrilineal Minangkabau of Sumatra (Josselin de Jong, 1980) - preserved forms of marriages in conformity with the adat (the local tradition) grounded on the practice of marriage of cross cousins and the prohibition of those with parallel parents. The exceptions are especially present in Moslem societies of India and Sri Lanka (De Munck, 1988). The pregnancy of the terminological and matrimonial cultural “Dravidian” pattern will often supplant that of the “Arab marriage” in Southern India, and Sri Lanka, whereas a pattern based on prohibition of marriages with all relatives and “Eskimo” terminologies often dominate in Northern India. Sometimes however, even here, the “traditional” strength of the Islamic pattern manages to subvert the Hindu tradition. Thus Moslem minorities of the Hindu kingdom of Nepal are excluded from the Nepalese legislation on marriage which formally prohibits all other castes to marry between “bone relatives” (relatives in agnatic line). Noting the fact that Moslem customs support on the contrary this endogamic marriage, and “to make it possible for these Moslems to practice it, the legislator gave up imposing on them the common rule of agnatic exogamy” (Gaborieau, 1977:227). In the latter case it is therefore the “Moslem” custom rather than the geographical and cultural context which prevail.
However, Moslem societies which adopted these forms of agnatic marriages also frequently adopted a terminology known as “Sudanese”, the only type of kinship terminology which distinguishes the relatives in strictly agnatic line from those in strictly uterine line. In what, then, the association of this matrimonial system and this way of naming the relatives inform us of the values of Islam, and of the manner of thinking the social construction of gender? This can be understood only if we take note of the constellation of facts and values which, in this philosophical and religious system, show a clear distinction of the respective spheres of influence and particular attributes attached to each sex. More undoubtedly than in other religions, Islam indeed puts forward exclusively the social, political, and legal dimensions of paternity which it opposes to a “naturalistic” and “biological” vision of the maternal bond. Thus, for example, the creation of a “fictitious” relationship can go through a man only by one legal inscription - adoption - whereas for a woman the simple fact of nursing a child created between them a kinship tie with the same strength than that resulting from childbirth. This family tie by “milk” (that of the nurse or that of the mother) and more generally the bond “through women”, is not only founded on a “biological” representation but also the only one to involve matrimonial prohibitions. On the other hand, the “social and political” ties which bind together groups of men do not involve any marriage prohibition, on the contrary, they favor those unions since they take place between agnatic relatives. Empirically, more the practices in these societies are based upon the “Arab marriage” within the agnatic line, more we note avoidances - which reach absolute prohibition - for marriages with uterine relatives: especially for the marriage between the children of two sisters which is however legal in Koran (Barry, 1997; 2000; 2008).
The system of ideological representations of Islam as well as the empirical practices of the actors inasmuch as they express themselves in the ways in which they promote or devalue marriage with certain categories of kin therefore converge towards the idea of a dual gender opposition. The men will be the guarantee of the devolution of the legal and political identity of the community but it is through women that will be transmitted the “substantial identity”, which by means of the too great a proximity of beings and bodies which it entails, will prevent any sexual conjunction. On the other hand, where a relationship does not imply a “natural” order, on the men’s side, the marriage will become possible, even preferred.
One thus explains the total congruence which exists between these forms of matrimonial practices and “Sudanese terminologies”. Indeed, an ideology which places at two extreme opposites the respective roles of man and woman in the act of procreation, must necessarily consider that relatives in agnatic and uterine lines cannot and do not have to be thought of, represented, and to be named in identical way. The category of the “children of two sisters” belongs for this reason to the reverse and symmetrical cognitive category than that of the “children of two brothers”. If the first is “naturally” nearest - and it is precisely on this “natural character” of the uterine bond “(silat al-raḥim)” that talks Ibn Khaldûn in the Muqadimma (Conte, 2000) - it is also socially and politically without true importance. On the other hand, if the second is socially first - for the heritage, the statute, the agnatic organic solidarity in the meaning of the “aṣabiyya” - it however does not fit with any emic “biological” representation. However, there is one type of terminology which precisely distinguishes between theses categories, theses two gendered lines of relatives: it is the “Sudanese” terminology.
1998 « Les modes de composition de l’alliance. Le “mariage arabe’’ », L’Homme, 147 : 17-50.
2000 « Le mariage endogame en Afrique et à Madagascar » in Laurent S. Barry (ed.), « Question de parenté » (special issue), L’Homme, 154-155 : 67-100.
2008 La parenté, Paris, Gallimard (“Folio Essais”), 863 p.
2000 « Mariages arabes. La part du féminin » in Laurent S. Barry (ed.), « Question de parenté » (special issue), L’Homme, 154-155 : 279-308.
De Munck, Victor
1988 “The Economics of Love : an Examination of Sri Lanka Muslim Marriage Practices”, Journal of South Asian Studies, 11: 25-38.
1977 « Minorités musulmanes dans le royaume hindou du Népal », Nanterre, Société d’Ethnologie.
Josselin de Jong, P.E.
1980 “Minangkabau and Negri Sembilan. Socio-political Structure in Indonesia”, Leiden, Martinus Nijhoff.
Karim, Wazir Jahan, ed.
1995 “‘Male’ and ‘Female’ in developing Southeast Asia”, Oxford/ Washington D.C., Berg Publishers.
Pour citer cet article
Laurent Barry, « Gender Logic and family ties in Moslem societies », La_Revue, n° 3, www.lrdb.fr, mis en ligne en août 2009.
Creation date : 11/08/2009 11:21
Last update : 11/08/2009 11:21
Category : Anthropology
Page read 4509 times
Print the page