The kinship I and the kinship other

This work is licensed under the Creative Commons | © Carlos Fausto. Attribution-NonCommercial-NoDerivs 3.0 Unported. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau3.2.018

Book Symposium

The kinship I and the kinship other

La parenté en question (again)

Comment on SAHLINS, Marshall. 2013. What kinship is—and is not. Chicago: University of Chicago Press.

Carlos FAUSTO, Museu National – PPGAS


Commenting upon Marshall Sahlins’ new book is no easy task. Not only because of the amplitude of the question he poses, but also because of the theoretical difficulties that this question imposes. Of all traditional topics in anthropology, kinship is surely the most sensible to the nature/culture divide, and was the first to succumb to its (partial) dissolution. True, kinship studies only succumbed in its traditional form, being somewhat resurrected by the concept of relatedness— admittedly a loose concept but one that should make cross-cultural comparisons possible again.1 Relatedness brought together studies founded on a double premise: that kinship had nothing to do with biology, and that biology (in the guise of reproductive technology) had a lot to do with culture. It gave a name to a field of inquiry that emerged against Rodney Needham’s (1971) skepticism and from David Schneider’s (1984) critic of kinship studies.2

The first question that popped up in my mind while reading Sahlins’ book was: In which sense does the concept of “mutuality of being” differ from “relatedness”? Or, more generally, how does his approach differ from its predecessor? My first reaction was to spot a good dose of continuity between them—after all most of Sahlins’ ethnographic examples come from people identified with the literature on relatedness. However, on a second reading, I noticed that he is rather critical of two central influences for the latter approach; namely Schneider’s dissolution of kinship and Marilyn Strathern’s “dividuality.”

The definition of “mutuality of being” appears exactly at the point in which he takes stock with both, criticizing Schneider for “ignoring the symbolic constitution of social relations” (Sahlins 2013: 16) and Strathern for a “certain confusion between personhood and kinship relations” (25). Through both critiques, Sahlins strives to establish kinship as a “definable object of study” and calls for a “more sociocentric view” of personhood, affirming that the person as “the current idol of the anthropological tribe . . . may itself derive some motivation from the hegemonic force of bourgeois individualism” (27). Partible and relational identities would be a universal feature of kinship rather than a “universal form of premodern subjectivity” (25). The notion of “mutuality of being” has thus a double function: on the one hand, to anchor kinship without grounding it in nature; on the other, to define kinship, again, from a sociocentric point of view.

These are basic differences between Sahlins’ approach and previous ones. These differences bring about some further questions. In what concerns the anchoring of mutuality without recourse to Nature, I would ask what Michael Tomasello’s work is doing in the book. What does the excursion into Human Nature do to Sahlins’ argument? Does he contradict himself? Is it a matter of choosing between “why we cooperate” and “why we fight” type of questions? In fact, if one looks carefully at the book, one sees that Nature lurks around in many of its bends—for instance, in the recurrent distinction between biological and performative, prenatal and postnatal, birth-ascribed and life-achieved kinship, though many of his examples dissolve this difference into just “kinship” with no further qualification. After all, as he says, “all means of constituting kinship are in essence the same” (29). Why then differentiate biological from performative if there is no way to tell them part?

In framing the book as culture vs. biology, nurture vs. nature, the recourse to Human Nature seems unavoidable. The production of knowledge in other scientific areas does matter to us and we have to engage with it. I myself read Tomasello’s work as well as Edward O. Wilson’s recent re-framing of eusociality and inclusive fitness theory as evidences that anthropology got kinship right because culture is a universal fact for humans as much as eating and procreating (meaning that no human can eat and procreate outside culture). What we need to clearly distinguish, however, is when we are using a phylogenetic, an ontogenetic, or a systemic argument. One thing is to ask how kinship developed during human evolution or how a child acquires it during its development, another is to study kinship systems. And this is how I understand part of Sahlins’ reclaiming of a “sociocentric view.”3

From a sociocentric point of view, kinship is a specific human way of establishing a system of differences, of cutting the network of possible connections (“the analogic flow”) into determinate relationships. The network is cut in variable ways across human societies, but in each system as a system, each cut relates to the other cuts, and cannot be defined outside this system. In this sense, Sahlins is right in contesting the extensionist hypothesis: nobody is a “child” to a “mother” without being at the same time a “nephew” to an “uncle,” a “brother” to a “sister” (whatever these categories mean in each system). On the other hand, the relationships indexed by these pairs (“mother-son,” “brother-sister”) do not cut the analogic flow in a purely conventional way. There are logical possibilities that never occur, others that are infrequent, and still others that are so recurrent that make one address phylogenetic questions. The problem with the latter is not that they do not make sense, but that they only accept trivial answers in face of the lack of empirical evidence.

So, regarding the debate between extensionsists and categorialists, I think the way Sahlins poses the question predetermines his answer. He asks: “[D]oes it not follow that the relations derived from procreation comprise the primary ‘code’ or ‘model’ of all human kinship? Or that such ‘true’ relations of genealogy provide the ‘focus’ or ‘type species’ of kinship categories?” (72). Nevertheless, one could phrase the same question in a non-genealogical or naturalist framework: “Is mutuality of being constructed out of relationships of containment and begetting, closeness and feeding, such as the ones prototypically expressed in human and non-human sexual reproduction?” If one poses the question this way, with no recourse to the opposition between pure nature and culture, the answer could be “yes,” could it not? I am, thus, not convinced that the adoption of a prototype theory for kinship categories necessarily implies a procreative and genealogical model of kinship. Grounding kinship in the “hard facts of life” is not the same as adopting Eleanor Rosch’s or George Lakoff’s non-Aristotelian definition of a class.

Sahlins gives us a number of examples that indicates that a prototypical theory should not be conflated with a naturalistic extensionist hypothesis. The salience of a mother-father-child triangle may result from performative kinship and not from natural relations. Many societies culturally produce this salience through focalization. In the case of procreation, it begins much before birth with a number of taboos followed either by the mother or the father (or fathers). Anthropology has usually conceived of such focalizations as a social way for recognizing a natural fact, and particularly as way of producing a social pater out of a natural uncertainty. But this is not quite the case; otherwise no such work would be required in regard to the mother. Could we argue, then, that seclusion, food taboos, and the couvade produce closeness and sameness, containment and begetting, making some relationships prototypical in the definition of “mutuality” and “co-presence”? Relationships devolving around procreation are good candidates to stand as prototypes of certain classes of relations, but they are far from being the only ones.

There is still another scenario in which we could even do without the notion of “performative.” The distinction biological/performative is called upon when one assumes the separate existence of a conceptual world apart from a physical one, and then has to connect the two again. But what does happen when we crosscut the nomos/physis dichotomy, assuming that signs are as much part of our world as, say, sex? I am not just saying, pace Sahlins, that there is no “pre-discursive fact” (3), but also that discursive facts are not only matters of concern, but also matters of fact, to use Latour’s vocabulary—that is, they are facts as much as sex is. They are not only constructions (i.e., human conventional worlds). The problem of constructivism is that in doing without one part of the dichotomy (nature), it assumes entirely the features of the other side (culture), meaning that it stays within this very dichotomy.

How should we thus understand the meaning of the expression “the symbolic constitution of”? In a Saussurean paradigm, such expression conveys an absolute arbitrariness between form and meaning. Culture is, throughout, an arbitrary convention, a conceptual system closed in itself. Another approach (of which a Peircian one is a good candidate) would make room for different ways of relating sign and meaning, word and world. The difficulty resides here, as we know, in defining the nature of reference and avoiding the specter of representationalism. Does conceiving of signs that establish a non-epistemic continuity between language and the world necessarily invoke the mirror of nature? As I understand the expression “symbolic constitution of kinship,” it does imply a certain relation with a world, which is as much cultured nature as natured culture. This is what make it possible for us, as human beings, to recognize something like “kinship” (or mutuality of being) whatever we go, and at the same time be equivocal about it. We know it is there, but sometimes it is not at the place we expect it to be.

In my sense of “la parenté” (which is certainly different from the Anglo-Saxon one), kinship includes as much sameness as difference, as much mutuality as reciprocity. And this is why I feel that mutuality does not cover the whole field of la parenté. Perhaps mutuality’s Latin etymology can make me better accept it. It originally meant a loan and by extension something done in return or in exchange. In this sense, mutuality falls within the framework of the gift, and must include affinity as central to the constitution of kinship as parenthood and siblingship. After all the “kinship I” always poses a “kinship other” (and vice versa).


I thank Luiz Costa for his observations on an early version of this comment.


Mosko, Mark. 1992. “Motherless sons: ‘Divine heroes’ and ‘partible persons’ in Melanesia and Polynesia.” Man (N.S.) 27: 697–717.

Needham, Rodney. 1971. Rethinking kinship and marriage. London: ASA Monographs.

Rumsey, Alan. 2000. “Agency, personhood and the I of discourse in the Pacific and beyond.” Journal of the Royal Anthropological Institute 6 (1): 101–15.

Sahlins, Marshall. 2013. What kinship is—and is not. Chicago: University of Chicago Press.

Schneider, David. 1984. A critique of the study of kinship. Ann Arbor, MI: University of Michigan Press.



Carlos Fausto
Museu Nacional – PPGAS
Quinta da Boa Vista s/n
Rio de Janeiro, RJ
20940-040 Brazil


1. The word “relatedness” works better in English than in Romanic languages. It resonates with the notion of “relative,” also a loose word pointing to members of an extended group including both kin and in-laws. This is exactly the breadth of the notion of “parente” in Portuguese (or in French, pace Dumont), which includes what kinship excludes in the Anglo-Saxon world. “Parente” includes in-laws, as a Brazilian political slogan of the early sixties makes clear. An important political leader was the president’s brother-in-law, and could not present himself to the latter succession. His followers wrote, then, on the walls: Cunhado não é parente, Brizola para presidente (“a brotherin-law is not a kin, Brizola for president”).

2. Needham’s Rethinking kinship and marriage (1971) was translated into French in 1977 with the title La parenté en question.

3. The other part concerns Jørgen Prytz-Johansen’s “kinship I,” which appears as the sociocentric alternative to Strathern’s “dividual.” If I am not mistaken, such opposition was first brought up by Mark Mosko (1992) and re-taken up by Alan Rumsey (2000), who argues that both encompassement (heroic I) and detachability (partible person) are fundamental dimensions of all human language. As I understand Rumsey, he calls for a micro-level, interactionist, and discursive analysis of specific ethnographic situations rather than a general framework based on these two figures.