Cai Hua, L’homme pensé par l’homme, Du statut scientifique des sciences sociales (Man thought by Man. About the scientific status of social sciences), Paris, PUF, 2008, 214 pp. Laurent Barry, La parenté (Kinship), Paris, Gallimard, 2008, 567 pp.

Two recent works reintroduce China into the anthropology of kinship. Cai Hua and Laurent Barry are both students of Françoise Héritier, who at the Collège de France has revitalised our understanding of the systems of African kinship.(1) Their two books, similarly constructed, devote considerable space to the systems of the Han. The anthropology of kinship, a discipline founded in 1870 by Lewis Henry Morgan’s Systems of Consanguinity and Affinity of the Human Family, and revived in 1949 by the thesis of Claude Lévi- Strauss, in Les Structures élémentaires de la parenté (The Elementary Structures of Kinship), has in the course of the past 20 years absorbed the shock of new technologies of reproduction and new forms of family relations: medically assisted procreation, cloning, blended families, civil partnerships, homosexual parentage, etc.(2) The scientific positions on these burning issues have been radically opposed to each other: on the one side are those who, following the declarations of the influential Pierre Legendre, have asserted the immutable character of the “symbolic order” rooted in the individual’s unconscious,(3) and on the other those who, in the wake of the analyses of the lamented Yan Thomas, have highlighted the invented and thus malleable character of the legal constructions of kinship.(4) The works of Cai Hua and Laurent Barry present the advantage of taking a step back from these impassioned debates and of offering us a “view from afar” at European kinship via the detour of the Chinese systems.
In 1997, half a century after Lévi- Strauss’ theory first appeared, Cai Hua published his own thesis on the Na society in China,(5) which made waves – in muddied waters, some would say – in studies of kinship. Under the title Une société sans père ni mari (A Society Without Fathers or Husbands), he describes a matrilineal society that practices a tradition of “visits” (furtive or conspicuous) by men that ignores the rules of marriage. These rules were imposed by the Qing empire, but, as observed by the ethnologist, the Na continued to follow their own customs. This observation poses certain ethnographic problems (how regular is the practice of the “visit” conducted? Does it result from a situation where the men emigrate from the village?), but it also raises an important anthropological question: is it possible to conceive of a social order outside the framework of marriage? Have the theories of kinship suffered from an unfounded assumption of the universal nature of marriage? That would drive a wedge into the Levi- Strauss theory of kinship, as Clifford Geertz remarked in the New York Review of Books, providing Cai Hua’s work with great resonance: “The very idea of a ‘kinship system,’ a culture-bound notion if ever there was one, may be a large part of the problem.”(6)
In L’homme pensé par l’homme, Cai Hua returns to the Na in order to compare them with three other “cases.”(7) The Na represent a type of kinship in which consanguinity is transmitted by the women, the man only playing the role of a temporary “waterer” of a fertile soil (as the Na saying goes: “if the rain does not fall from the sky, the grass will not grow from the earth”). In contrast, the Han case (studied by Cai Hua during research in Kunming) understands consanguinity as exclusively masculine, sperm being considered as blood issued from the bone. Set against these two forms of monolateral consanguinity, Cai Hua puts forward two other forms, which he calls bilateral symmetrical and asymmetrical: the French accord to the man and the woman an equal role in consanguinity (the man contributing the sperm and the woman the blood), while among the Samo of Upper Volta studied by Françoise Héritier, the man contributes the permanent blood and the woman a provisional blood.
This tableau of four cases allows Cai Hua to define kinship as the bond between consanguinity (biological bond) and affinity (social bond). To reinforce this definition, he examines the different theories of kinship, from Lewis Henry Morgan to David Schneider, which in his eyes all exhibit the weakness of tracing kinship back either to consanguinity (and thus the biological aspect) or to affinity (and thus the social aspect), with Lévi-Strauss occupying a paradoxical position somewhere between these two poles. Cai Hua does not, however, discuss the analyses of Robertson Smith in Kinship and Marriage in Early Arabia (1885), which linked beliefs about consanguinity to the ritual of sacrifice as a sharing of the blood, and which exerted a major influence on the theory of Durkheim, according to which kinship is exclusively social, or ultimately religious. From this rapid review of the history of anthropology, Cai Hua concludes that the discipline has not been able to imagine this strange alliance of consanguinity and affinity, or, to put it in other words, of a biological fact and a social representation, or even, to use more charged terms, of nature and culture. He consequently proposes a theory of belief aiming to explain that ideas emanating from the imagination can form institutions and act through norms (according to a tradition that he traces back to Mencius). This theory maintains that the belief in the transmission of substances defines the individuals with whom lasting relations are established.
The problem that arises then is the status of the social sciences in relation to these beliefs. If Cai Hua demonstrates that the anthropological theories of kinship have tended to reproduce a Western conception of consanguinity (from Morgan’s assertion according to which Roman law is the most “natural” classification of the systems of kinship, to Lévi-Strauss’ analysis of marriage as an exchange establishing the foundations of the social order), what, then, guarantees that the anthropologist’s discourse does not take up an indigenous belief while projecting it onto other indigenous beliefs? To answer this question, Cai Hua proposes in L’homme pensé par l’homme an epistemology of the social sciences that does not design them on the model of the natural sciences without renouncing their status as a science. It is not certain that Cai Hua provides all the tools to resolve this epistemological problem, a constituent component of these sciences since Durkheim forged the concept of “social fact” in order to give them an object.(8) The domain of “ontology” where he ends up is promising but a little vague, and the distinction between a general ontology and a restricted ontology remains unclear. However, the great value of his book is to pose the problem on the basis of a very particular ethnographic case.
Claude Lévi-Strauss was very conscious of this problem, and all the rest of his work has aimed at resolving it by distinguishing the environmental organisation of societies and the mythological representations that express their contradictions (according to a model of “dialectical reason” borrowed from Marxism, the premises of which Cai Hua surprisingly does not discuss here). In response to the objection of the Marxist anthropologists, he wrote in La pensée sauvage, published in 1962: “I must now confess to having myself unintentionally and unwittingly lent support to these erroneous ideas, by having seemed all too often in Les Structures élémentaires de la parenté as if I were seeking out an unconscious genesis of matrimonial exchange. I should have made more distinction between exchange as it is expressed spontaneously and forcefully in the praxis of groups and the conscious and deliberate rules by which these same groups – or their philosophers – spend their time in codifying and controlling it.” In 1997, in response to the objection of Cai Hua according to which the Na ignore the exchange, Lévi-Strauss wrote in La Republica: “The Na represent an extreme case of a system, other examples of which have been known for a long time, particularly in Nepal, southern India, and Africa. And far from such cases destroying the accepted ideas, the family structure that they illustrate simply offers a symmetrical and inverse image of our own. These are quite simply societies that do not, or no longer, place a regulatory value on kinship and marriage in order to ensure that they function, but leave that to other mechanisms.”(9) All of Lévi-Strauss’ efforts since his Structures élémentaires de la parenté have been invested in moving beyond this ideology of the exchange, which he had appeared to formalise, by directing his research into the mechanisms of thought that govern practices, the inverse effects of which can be seen in matrimonial exchange and the Na system.
The book by Laurent Barry, published at the same time as that of Cai Hua, takes up this Lévi- Straussian intuition again, but proposes to synthesise all the works of anthropology of kinship that have appeared since Lévi-Strauss’s thesis.(10) The affirmation of the founding role of exchange and communication by Lévi-Strauss in 1949 (taking up again the analyses of Mauss on economics and of Jakobson on language) has indeed been subjected to at least two salvos of objections from English anthropologists. On the one hand, focusing the study of kinship on the alliance of marriage — defining whom it is possible to marry — leads one to ignore the mechanisms of descent — defining the members of the group towards whom the individual has rights and duties.(11) On the other hand, highlighting exogamy as an obligation to marry a member of another group leads one to ignore endogamous forms of marriage, which are far more widespread in the societies of North Africa (“Arab marriage”).(12) Furthermore, the Lévi- Straussian analysis of the universal character of the prohibition of incest (at the interface between nature and culture) has been called into question by the research of ethologists into animal societies, which similarly understand this prohibition, and by the study of ancient societies that tolerated incest while giving it various forms. Finally, the study of the complex structures of kinship, which do not prescribe spouses but define which ones should be avoided, leads one to resort to new tools, such as statistics and demography. From now on, the anthropology of kinship has to be conducted, according to Laurent Barry, on new foundations, while affirming less clearly the autonomy of a set of rules and being more attentive to the variety and the malleability of their transformations.
Laurent Barry starts out from a minimal definition of kinship: “It is a taxonomic process that concerns the human and which sorts us into our ‘kind’ and others in the group of individuals who are really or putatively related among themselves by birth or marriage. But this without the communal identity that it postulates ever resulting immediately from filiation, consanguinity, or alliance.”(13) He then proposes defining “groups of kinship” by “properties,” whereby individuals can be said to share the same kinship, and by “principles,” which express the roles of each sex in mating and procreation. This leads him to distinguish four groups of kinship:(14) a uterine kinship, corresponding roughly to the forms known as the “Arab marriage” and granting women an overwhelming role in the generation of beings; a parallel kinship, corresponding to the “elementary structures” studied by Lévi-Strauss in India, Oceania, and Africa, and obliging one to take a spouse from an outside group; a cognatic kinship, corresponding to our system since the Republic of Rome, the transformations of which Barry illuminates through the Christian theme of una caro (“one flesh”) uniting the spouses; and finally, an agnatic kinship, corresponding to the Chinese system of the Han, in which it is the man who constitutes the pivot in the bonds of kinship.
Laurent Barry grants a lot of space to China in his analyses. If he skips a little quickly over the Na case (which he classifies in the field of uterine kinship), (15) he does on the other hand dedicate long analyses to the systems of kinship among the Han. There, he reviews in detail the demonstrations of Marcel Granet in the Catégories matrimoniales en Chine ancienne, a work that appeared in 1939 and which played a central role in the genesis of Lévi-Strauss’s thesis.(16) Barry particularly notes the attention Granet gave to the link between the degrees of proximity in kinship and the number of days of obligatory mourning. This link was expressed in the Book of Rites and then codified in the legal texts intended for the “barbarians” who were unaware of these obligations. The analysis of the transformation of rites into law in China is as fascinating here as the one that precedes it on the transformation of the moral reflections of the Romans into the theological speculations of the Christians – bearing witness to the virtues of a term-by-term comparison of the two traditions that does not lapse into the easy options of Orientalism.
At the end of this impressive work of more than 700 pages, one question has to be asked, however. Returning in conclusion to the ambition of his work, which is to “reunite in a single explanatory scheme a heteroclitic set of matrimonial systems (…) formerly or elsewhere subject to a colourful assortment of local theories,”(17) Laurent Barry recognises that this perspective, “adopted by every science that frees itself from its ideological foundations,” leads resolutely to “the voluntary abandonment of certain ancient philosophical ambitions of anthropology in this domain,” such as the one that aims to formulate the meaning of the human condition through the passage from nature to culture.(18) Here the epistemological reflection of Cai Hua regains its interest: is it possible to formulate a theory of kinship that is entirely “freed of its ideological foundations”? One can have doubts about that because the anthropologist, if not writing in purely formal language, adopts an entire vocabulary (kinship, filiation, alliance, consanguinity, affinity) that is laden with ideology, and above all because the discussion that it engenders is produced by a certain historical situation and addressed to a particular community. The anthropology of kinship raises both intellectual questions and political passions, as is shown by the debates on civil partnerships in France, because it is simultaneously ideological and critical: ideological insofar as it takes up some of our most common intuitions on what brings us close to other people and what distinguishes us from them, critical insofar as it does this via the detour of distant societies where these intuitions are disrupted. Thus the detour via Africa or China allows Laurent Barry to criticise our evidence originating from the Christian theme of “una caro,” just as the detour via the Na allows Cai Hua to criticise the Chinese conception of marriage in a way that he illustrates through the Chinese proverb: “The stone from other people’s mountains can carve our jade.”(19) This balancing of ideology and criticism is what gives anthropology its philosophical dimension. This does not mean that it expresses in systematic terms the fundamental conditions of humanity, as some hasty readers of Structures élémentaires de la parenté believe, but rather that its discourse always lies halfway between the spontaneous philosophy that it criticises and the indigenous philosophies that it rehabilitates. Claude Lévi- Strauss thus concluded his contribution to the edition of L’Homme devoted to “Questions of kinship” in 2000 with these words: “Whether one rejoices or whether one frets about it, philosophy once again occupies centre stage on the anthropological scene. No longer our philosophy, which my generation had asked the exotic societies to help undo, but, in a striking reversal of things, theirs.”(20)
Translated by Nick Oates

Back to top