The order of intersubjectivity

This work is licensed under the Creative Commons | © Klaus Hamberger. Attribution-NonCommercial-NoDerivs 3.0 Unported. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau3.2.020

Book Symposium

The order of intersubjectivity

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

Klaus HAMBERGER, Laboratoire d’Anthropologie Sociale


Marshall Sahlins’ book undertakes to definitively overturn David Schneider’s verdict that, without recourse to biology, we no longer have the means to say what kinship is. The notion of kinship seemed destined to be lost, diluted among countless forms of relatedness and partibility. Sahlins proposes a criterion to single it out again: contrary to other relations (as between teachers and students, or between fans of the same football club), kinship entails an idea of “mutual,” “conjoint,” or “transpersonal” being, an “intrinsic participation in each other’s existence.” Kinsfolk are “one person,” linked by mutual substitutability and solidarity, fundamentally equal and alike, living each other’s lives, responsible for each other’s deeds, belonging to each other.

Considered from a Watchi-Ewe perspective (which in this respect does not much differ from the Ashanti point of view cited at the outset of Sahlins’ book), this definition would be a perfect characterization of uterine kinship. Watchi maternal kin—be they linked to each other by birth or by purchase—mutually belong to and possess each other, may be killed or sold in each other’s place, must revenge each other, and are in all respects considered as “one person.” But precisely for this reason, Sahlins’ definition would be incapable of characterizing Watchi agnatic kinship, be it constituted by procreation or (more frequently) by co-residence. Sharing each other’s house rather than each other’s person, forming part of a greater whole rather than of each other, paternal kin are related by hierarchy and rivalry rather than by equality and solidarity, by closeness rather than by sameness, by contiguity rather than by substitution.

Similar conceptions exist among Fortes’ Tallensi, and among many other African societies where the fabric of kinship is driven by the interplay of agnatic and uterine principles, conceived not so much as genealogical links than as different forms of relationality. In such a context, Sahlins’ definition of kinship leaves us with only two options: either we deny the status of “kinship” to all relations not taken to entail the union of discrete subjects in “one person,” which would put us at odds with the ethnographic record that leaves no doubt that paternal and maternal relations form a whole, or we make the semantic domain of “mutual being” wide enough to encompass all imaginable forms of kinship (Sahlins 2013:29, 86), to end up once again with an empty and amorphous notion of relatedness that is in no way specific to kinship. Schneider’s standard example still seems to hold: while the criterion of “mutual being” may be sufficient to distinguish members of the same family from members of the same football club, as Sahlins claims, it is certainly not sufficient to distinguish them from members of the same church, equally defined, like kinsmen, by “shared substance” and “diffuse, enduring solidarity” (Schneider 1972: 40).

I am afraid that there is no way around Schneider’s demonstration that these (or any other) definitions are of no help for demarcating kinship as a distinct cultural domain. And this may be precisely because kinship is not a type or class of relations at all, but a logic of relations which may be at work in any domain— genealogy or terminology, politics or economy, ritual or food-sharing. Now, for a logic of relations to work, there must be at least two different forms of relationality: affinity and consanguinity, alliance and descent, paternal and maternal filiation, or whatever. To be sure, these relational forms are not simply given; they have no existence outside the logic which contrasts, combines, and transforms them. We may well express this fact by stating, as Sahlins does with reference to Roy Wagner, that they are all fundamentally alike (Sahlins 2013: 29). But by the same token, any strategy of defining them by reduction to a single criterion is doomed to failure, since it is precisely through the interaction of different relational criteria that kinship operates. What turns paternal and maternal, consanguineal and affinal, horizontal and vertical relations into kinship relations is not some mysterious intrinsic quality they have in common, but the fact that they from part of a common logic which differentiates them from each other and frequently opposes them to each other (the classical example being marriage prohibitions).

Clearly, Sahlins also emphasizes the differentiating work of kinship (43). But the differentiation he undertakes does not proceed from the concept of kinship itself; rather, it appears added to it from the outside by a series of secondary specifications. Thus, genealogical priority is reintroduced as “unbalanced” mutual being (37), genealogical distance as “declining” mutual being (53), affinal relations as “conflictual” mutual being (55), and even Fortes’ distinction between descent and complementary filiation reemerges in the guise of “primary” and secondary mutual being (56). While none of these distinctions (largely congruent with those of classical kinship theory) is explained by the theory of “mutual being,” as a whole they appear indispensable to give “mutual being” the particular shape of kinship— most explicitly so where Sahlins states (56) that the concrete form of “mutual being” depends on the nature of the descent and marriage system.

This is not to say that there is no way out of this circle. As Sahlins rightly notes, Eduardo Viveiros de Castro’s conception of Amazonian kinship as another facet of magic and gift-exchange holds strong against Schneider’s deconstruction. But this is not because kinship, magic, and gift-exchange deal with the same basic notion of animistic “influence” (another avatar of “mutual being”), but because they all express the same symbolic logic, which is about how the continuous reproduction of difference (“affinity” in Amazonian terms) conditions the production of similarity (“consanguinity”). This interaction of similarity and difference is just one aspect of the logic of kinship. Another is the interplay of substitutability and contiguity, which, as the “uterine” and the “agnatic” form of relationality, are the fundamental operators of social organization and religious practice in several West African societies. Clearly, as forms of kinship, all these relational forms are not just abstract principles of thought—they link not only symbols and meanings, but beings and perspectives. Still, this would not define kinship as “mutual being,” but rather as the “order” of mutual being (in the sense that Leibniz defined space as the “order of co-existences”1), a standpoint which Sahlins approaches at the end of his first chapter, when he calls Viveiros de Castro’s generalized symbolic economy a “cultural order of intersubjectivity” (60).2

This foreshadowed definition of kinship as the logic of intersubjective relations appears to me much more fecund than any attempt to re-anchor it in arbitrarily extensible notions such as “mutual being,” “participation,” “amity,” or the like. Sahlins’ book can be read in both ways: as an effort to restore pre-Schneiderian kinship on a non-biological basis as a distinct cultural system, or as an appeal to go beyond Schneider’s critique and to conceive of kinship as an overarching cultural logic.


I am grateful to Laurent Gabail, Michael Houseman, and Ismaël Moya for their comments on my comment.


Clarke, Samuel. 1717. A collection of papers which passed between the late Learned Mr. Liebniz, and Dr. Clarke, in the years 1715 and 1716. London: James Knapton.

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

Schneider, David. 1972. “What is kinship all about?” In Kinship studies in the Morgan centennial year, edited by Priscilla Reining, 32–63. Washington D. C.: The Anthropological Society of Washington.



Klaus Hamberger
Laboratoire d’Anthropologie Sociale
52 Rue du Cardinal Lemoine
75005 Paris


1. See Clarke (1717): “Third letter to Clarke (1716).” § 4. Online edition at www.newtonproject.sussex.ac.uk/view/texts/normalized/THEM00230

2. I understand “intersubjectivity” as a relationship between subjects, which does not necessarily imply the idea of a shared mental content (cf. Joel Robbins’ contribution to this debate).