Initially published 30 May 2018

To make a topic from one of anthropology’s principal means and objects of study, investigating relations through relations, is offered in the spirit of reflexive enquiry. The entry is not confined to anthropological works, touching briefly on certain philosophical dimensions and drawing in writers from other fields. However, it is organized around the way anthropologists have refined and expanded the application of relations through their diverse usages. Emphasis is thus on showing how the concept is used, rather than on prescribing particular versions. Attention is paid equally to the relations through which arguments and analysis are pursued and to the subject matter of anthropological investigation as the relational life of persons and things. The entry also notes a long-standing debate between English-speaking and continental European thinkers in the priority they give to terms (the ‘terms’ of a relation: what a relation holds together) or to the relation as an encompassing totality (of which the terms are a part). This one concept thus embraces whole different sets of assumptions about the nature of social life. Its own relations to other concepts are also relevant, as are changing emphases on what it might purport in a changing world. 


The English-language concept of relation is so ubiquitous, is entailed in such a range of applications, there might seem a good case for leaving it to commonsense to sort out what is meant on this or that occasion. But many anthropologists would also claim it as a signature concept for their discipline, and their usages have taken its potential forward in some very specific ways. Although there is no special anthropological definition, there is broad agreement about the privileged place it has both in structures of argumentation and in what are understood as social anthropology’s principal objects of study, and about the way it is often introduced into discussion to signal a critical (in the sense of probing and questioning) move. There is much to be learnt from its role in the framing of anthropological accounts.

Anthropological notions of description, analysis, and theory, above all in the distinctive terrain it has marked out as cross-cultural comparison, take for granted that one’s job is to show relations between phenomena. Thus one may demonstrate the extent to which religious precepts uphold or challenge values promulgated by the state or hypothesize correlations between new technologies and changing senses of the self. That taken for granted status is built into the way scholarly narratives are organized. Most of the time it is indistinguishable from the perception that relations inhere in the object of enquiry, and the observer is drawing them out. The commitment of twentieth century anthropology to the concepts of ‘society’ and ‘culture’ presented the world with what were above all bundles of relations. People’s actions and behaviour were to be described (analyzed, theorized) in the context of the diverse relations in which they were enmeshed. Anthropologists continue to show the logical or functional relations between entities they abstract, such as religion or the state, and create new fields of enquiry by emphasizing the relational nexus of phenomena, a notable case being that of personhood and the entanglements imagined between self and other, individual and collective. But at the same time, they take it as self-evident that everywhere people too are drawn into relations with the things, beings, and entities that form their environment. Above all, the specific capacity of persons to relate to one another is taken as a fundamental truth of human existence. Social life is what goes on between them.

Relations between and within

However, the Latin term relatio, from which ‘relation’ came into English, did not connote that state of ‘betweenness’, and there lies a history about what gets to be articulated. Classically, relatio referred to what was carried back (to someone) as in a reply or report; indeed, it was a substantive for a ‘motion’ (as in a proposal) or narration (producing a narrative). Medieval philosophers used relatio as an alternative for ad aliquid, an inclination ‘towards something’, a disposition, directionality, order (Brower 2015). They drew from Aristotle’s disquisition on categories (for an anthropological comment, see Allen 2000): the idea that such an inclination was a property (‘accident’) inherent in one entity in the way it pointed towards another.[1] Their reflections addressed common linguistic differences, as in the differentiation of absolute and relative terms, the latter arising from the comparison of things. An attendant concern about the way things bore on one another, through (say) correspondence or resemblance, with respect to the role of their own intellectual activity, through (say) comparison, continued to bother European thinkers into early modern times. As for an articulation of how entities, such as intervals, might lie between other entities, it would seem that philosophical discourse lagged behind ordinary usage. It was in terms of how relations could be formally represented that ‘betweenness’ was a relative late-comer. That this might have anything to do with the scientific revolution is a matter of speculation (Strathern 2005: 33-49). But possibly an emerging worldview that rested on explaining discrete phenomena by reference to the forces, logics, or structures that held them together had found in an old term a new one – relations – for that holding together.

This worldview was not uncontested. If this is a development traceable in English, there were early modern continental thinkers who took relations in a different direction. Descombes (2014) rehearses Gottfried Leibniz’s specific objections to the definition of relations proposed in 1690 by the English philosopher, John Locke: the referring or comparing of two things to one another. The German thinker’s famous dictum, 

there is no term which is so absolute or so detached that it does not involve relations and is not such that a complete analysis of it would lead to other things and indeed to all other things. Consequently we can say that “relative terms” explicitly indicate the relationship which they contain (from Leibniz, written in 1704; Descombes’ [2014: 204] emphasis)[2]

suggests that everything participates in a turning towards another. Caught up in a debate about the real and the unreal – or mental – status of phenomena as these thinkers were, Descombes spells out the implications of their arguments for the empiricist view that social relations are exterior to individual entities and the idealist view that social relations are constitutive of individuals. The part of Locke’s thesis relevant here – the suggestion that, as a mental exercise of comparison, relations are external to phenomena – diverges from that of his German critic, which denies that there is any wholly extrinsic denomination because of the ‘real [in the above sense] connections amongst all things’ (quoted by Descombes [2014: 204]); everything combines extrinsic and intrinsic relations.

Although the distinction between external and internal relations was to have a very mixed future in philosophy, it has sometimes been taken in anthropology to reflect a truth about the priority to be given to the already existing and thus discrete nature of entities, not in essence affected by their relations, as against the view that it is only through relations that entities are constituted. These tenets become visible, for instance, in the way anthropologists organize the frameworks of their accounts and thus decide what they think needs explaining. From the perspective of modern anthropology, both positions may stimulate a stance of criticality. 

First, ‘some descriptions of a thing by its [external] relations with its surrounding milieu have a real scope, [in] that they allow us to know the reality of that thing’ (Descombes 2014: 204-5). Putting things into context – seeing the larger picture, showing the implications, effects and outfalls (‘unintended consequences’) between actions, events, structures, assumptions, and so forth – was always the aim of the traditional ethnographic monograph. Thus the reality of Zande witchcraft divination was to be grasped through a relational nexus that included princely politics, how kin are connected, and the logic of cause and effect (Evans-Pritchard 1950 [1937]). Here too lies the force of imagining ‘merographic connections’ (Strathern 1992), a phrase that formalizes what is commonplace in English usage: the fact that nothing is simply part of a whole insofar as another view or perspective may redescribe it as part of something else. Religion and state, for example, may be shown to relate to each other in this or that respect, while the analytical discreteness of each is retained by the fact that either may also be related to quite other segments of social life, as when mystical beliefs (or population statistics) are regarded as part of the one but not of the other. 

Second, assuming relations are always and already everywhere has furnished anthropological discourse with a vocabulary that challenges the kinds of essentialist categorizations that rest precisely on the discreteness of phenomena. Lucien Lévy-Bruhl’s concept of participationhas drawn Sahlins’ (2013) attention: we take it for granted ‘that beings are given beforehand and afterwards participate in this or that relation; whereas, for Lévy-Bruhl, participations are already necessary for beings to be given and exist’ (33-4). Kinship connections are Sahlins’s prime example, in which the difference between kin positions are internalized by or resolved into the mutuality of their being: a mother’s brother exists as such through the existence of his sister’s son (for an ethnographic example, see Bonnemère 2018). A distinctive theory of ‘internal relations’ has been attributed to Karl Marx: the political scientist, Ollman (1976), points to Marx’s notion that things function because of their spatio-temporal ties with other things, and to conceive of things as relations interiorizes this interdependence.[3]

Descombes (2014: 197) summarizes his own view of the problematizations here by observing that a theory of external relations supposes that every change something has in its relations with other things is a change in its world, and not a change in what that thing intrinsically is, while in the case of internal relations every such change is a change that affects the thing itself. It goes without saying that sensitivity to these conceptual usages underlines the interest anthropologists have shown, though all too rarely, in other vernacular concepts of or counterparts to relations (e.g. Corsín Jiménez & Willerslev 2007).

Relations … and terms; relations … and connections  

As they are articulated for analytical purposes, relations evidently occupy a conceptual field along with other substantives. This section enlarges on certain indicative usages. Perhaps it is the juncture to emphasize that I am reporting on various anthropological usages, for example between epistemic and interpersonal relations, and not filtering everything through one lens or another. 

Both the apprehension of already-identifiable phenomena being brought into (external) relations with one another and that of phenomena (internally) constituted by relations may be built into the very definition of relation. Thus a relation-between may be imagined as itself composed of terms and relations (the relation only works with reference to something other, the ‘terms’ it links). Either the term or the relation can then be internally differentiated. Within the term, the conception of an entity’s self-referential ‘identity’ becomes modified when that entity is thought of ‘in respect to’ another, some degree of interdependence implied. This happens in the course of specification, for instance whether the magic one is thinking about refers to witchcraft or to oracles (all three are in the title of Evans-Pritchard’s monograph). Within the relation, there may be reason to distinguish relation from relationship, or relation from connection, as we shall see in a moment.  

These manoeuvres, including imagining alternatives to the terms-plus-relation model, may be deployed with critical intent. Recently translated works of Descola (2013) and Viveiros de Castro (2014) are exemplary here. Considering identification and relationship as fundamental axes of individual and collective behaviour, Descola develops an intriguing theoretical possibility latent in the interplay between terms and relations: the very manner in which specific cosmologies privilege the one over the other. He thus offers a wide-ranging, ‘combinatory analysis of the modes of relations between existing entities’, which is how he introduces his emphasis on external relations between beings and things as opposed to the internal links that pertain between abstract concepts; his criticism of earlier models remains largely implicit. On the other hand, Viveiros de Castro deliberately writes against a formula that depends exclusively on ‘a connection or conjunction of terms’.[4] Adopting Gilles Deleuze’s vocabulary, he states ‘that the future of the master concept of anthropology – relation – depends on how much attention the discipline will end up lending to the concepts of difference and multiplicity, becoming and disjunctive synthesis’ (2014: 170). These alternative coordinates for thinking about relations explicitly challenge the presumption that the primary values to which relations lead are those of binding ties or attachments.

So let us return to some of the ways in which relation has been differentiated. A case has been argued for distinguishing relations from ‘relationships’. Moutu (2012) wishes to get away from an obsession with epistemological understandings of relations, insofar as, in the case of persons, they occlude the ontological character of ‘relationships’. A thinker’s relational practices, such as connection, association, resemblance, comparison, do not touch on the necessity and transcendence that, in his words, give relationships the character of an infinite being.[5] This is the lesson of his Iatmul ethnography. There is nothing contingent about how Iatmul elder brother and younger brother are related as a pair, hence the necessity of their connection; insofar as each is also the other in another form, it is their relationship that transcends both the externality of their relating (if one wants to put it that way) and their identification as self-similar beings. Such relationships never cease; this is partly because of their processual nature.[6]

In other hands, it may seem equally crucial to split relation from ‘connection’ (here, differentiating epistemic [relations] from interpersonal [relationships] drops from view). Although, following eighteenth century English usage, anthropologists (this author included, and in this text) often use connection as a synonym for relation, the distinction yields further critical purchase. Feldman (2011) argues for a difference between relations and connections as methodological constructs in the study of global processes. Unconnected actors (not in direct communication with one another) may nonetheless be related though ‘indirect social relations’, mediated through apparatuses or some ‘variety of abstract mechanism’, such as the surveillance systems, detention centres, and statistical operations that track a migrant’s path. In other words, relations have an effect on – and pose problems for – actors far beyond the scope of their connections. Imagining an extra-terrestrial perspective on the world, one that invokes the potential of cross-world communication, may invite enquiry into a different discrimination between connection and relation. Pondering instead how people can mistake connection for relation, Battaglia (2005) draws a comparison with the envisioning of information networks so dense that they cover for the ‘work of relationality’ – singular acts of connection fantasized as instances of social exchange. In her rendering, social relations and the work they entail are set in apposition to otherwise uninflected contact or encounter, and refer to a specific order of reality.

Social relations

This phrase (social relations) is found frequently in twentieth century British social anthropology. Sometimes it is used to distinguish relations of sociability (the tenor of interactions, transactions, obligations between persons) from relations of an institutional or systemic kind: economic, political, gender relations, as when Douglas (1970) talks of ‘relating’ beliefs to dominant aspects of social structure. On other occasions it summons the totality of social life, whether it is encompassed by the concept of society or, shorn of certain connotations of society, rendered as sociality. Such relations may be imagined as in the first place relations between persons, human implied. A seminal text is Radcliffe-Brown’s homonymous 1940 address on social structure. 

Radcliffe-Brown (1952: [1940]: 188-204) famously defined social structure as a network of actually existing relations. Thus he was at pains to differentiate a non-social entity, such as the ‘individual’, from the entity that could be (analytically speaking) a node in this network, the ‘person’. His reference point was the concrete human being; as a person such a being was ‘a complex of social relationships’. The person was thereby a unit of social structure. A structural point of view, he said, requires studying how social phenomena such as religion or government have direct and indirect relations to social structure, here understood as ‘relations between persons and groups of persons’.[7] Pointing to kinship, an area anthropologists most readily cite as exemplifying internal relations,[8] Radcliffe-Brown asserted that kinship structures consist of numbers of dyadic relations ‘as between a father and son, or a mother’s brother and his sister’s son’. These were the building blocks of society. His emphasis on the dyad, through which he focused on an interplay between two genealogical positions, was to puzzle later anthropologists precisely for its privileging of genealogical thinking, but we can see it as an attempt to clarify just how one might construct persons as the terms (here equivocally external) to a relation.

Now Radcliffe-Brown’s specification of social relations had critical purchase against what in retrospect seemed the random reporting of diverse customs, as exemplified in early twentieth century accounts. Particular instances of behaviour or practices could be put into wider contexts, such contexts invariably consisting of the way relations were organized, a procedure that had long accompanied the analysis of kin terminologies. This assumption about organization (‘structure’) fed the ability to correlate, quite explicitly, numerous dimensions of social life. Goody (1962) offered an extended example from West Africa with respect to descent group formation, inheritance, and funeral practices. West African mortuary institutions were concerned with the reallocation of rights and duties, after death, precisely insofar as a ‘social person’ is defined through the mutual expectations that constitute his or her relationships. 

Finding correlations between social institutions within a society was accompanied by cross-cultural comparison between societies. Under the rubric of the latter, it was possible to compare institutions such as matriliny or witchcraft in terms of their local social configurations. Here, the notion of ‘relations between’ at once facilitated the comparison of discrete phenomena, invariably along the axes of their similarities and dissimilarities (‘differences’ in this sense), and produced as objects of study ‘societies’ and ‘cultures’ in this mould, to be criticized in turn for the presumption of discreteness. Comparison across discrete contexts – disjunctive comparison (Lazar 2012) – emerged as an anthropological practice. We may see critical purchase here being levered against arbitrary evaluations of what was or was not significant as an object of study.[9] However, there is also a sense in which any comparative move creates the potential of a critical outcome, insofar as social or cultural phenomena being brought into conjunction with one another shifts the observer’s perspective. Comparison was elemental in Locke’s definition of a relation; for the medieval philosophers comparatio had been almost more or less synonymous with relatio.

Needless to say, a re-formulation of relations came to Lévi-Strauss’s assistance in his notable quarrel with Radcliffe-Brown, beginning in 1945 (Lévi-Strauss 1963), and its consequences for British social anthropology. Take, for example, the reversal in the visualization of descent groups. What to the latter (Radcliffe-Brownian social anthropology) may have appeared the interdependency of genealogically discrete kin groups upon one another, through marriage alliance and other relations, from a Lévi-Straussian perspective would have appeared like a description of external relations (not his term). Lévi-Strauss’s own folding of affinity within the fundamental atom of kinship was instead a way of showing how such alliances were also presupposed (internally) by the total organization of relations. ‘[A]nalysis can never consider the terms only but must, beyond the terms, apprehend their interrelations’ (Lévi-Strauss 1978: 83). The whole is given before the parts, so one must begin with the whole, that is, with the relations among the parts.[10]

It is entirely possible to insist on linkages and the associational quality of the lives of collectives without explicit attention to the concept of relations (see Latour 2005). Indeed the ethnographic record affords numerous other ways of imagining the entailments or enrollments of all kinds of entities in one another’s circumstances. Of course the observer may gather these up as species of relations even when there is no vernacular counterpart, just as an anthropologist might use the terms ‘culture’ or ‘system’ to describe social configurations that actors conceive otherwise or do not conceive at all. It then becomes a theoretical choice, with every shade along the way, to decide whether relations are articulated in all but name or are being named because of the anthropologist’s discerning apparatus. For where anthropologists do take it as a master concept – as in those English contexts where the invocation of relations is an invocation of the facility to ‘bring together’ entities of any order – demonstrating relations is seen as probing beyond what is immediately accessible.[11] To reveal the relational dimension of this or that can also be empirical criticism of those worldviews that cannot comprehend or else devalue the way phenomena entail one another. 

The compulsion of relations

Emphasized by some present-day anthropologists more than others, the uncovering of relationality – in whatever system or circumstance – may be understood as confronting a positivism that focuses on the intrinsic nature (self-identity) of things. A critical stance is particularly obvious here. It is no surprise that scholars in general, whose business is in the narrational art of relating, deliberately pursue epistemological relations; it is not trivial to add that, for anthropologists who are also ethnographers, this is consonant with a value placed on social relations in particular, not forgetting their engagement with persons as interlocuters. A disciplinary disposition to uncover the significance of relations is thereby broader than the controversial use of cross-cultural ethnography to point up the identitarian bias built into the (Anglophone) anthropologist’s native language.[12]

When anthropologists talk about relations, it is persons who most often come first to mind; that is, beings inevitably enmeshed in a relational world. This holds regardless of whether, in any specific social configuration, people take relations as already there or as endlessly needing to be created.[13] In whatever manner people assume they are parts of the lives of others, they also put in relational work to support, deny, reconfigure, or transform their relations with one another. It is the transformative, or transcendental, nature of interpersonal relations that leads Pina-Cabral (2017) to suggest that they are a bad analogy for the more general condition of being-in-relation or relationality.[14] Rather, interpersonal (‘social’) relations are a special case to the extent that they are inevitably constituted through interaction and recognition, by contrast with relations that are mere affordances.[15] This offers, in effect, a perspective on vernacular usage.

In English ‘relation’ and its pair ‘relative’ are also colloquial terms for kin. This is an idiomatic support or crutch for the tendency of ‘relation’ to connote connection and attachment before it also embraces disconnection or detachment, just as familial ties are normatively embued with positive rather than negative affect.[16] We have already seen that such values bear on the anthropologist’s work practices, notably strong in the positive sense of accomplishment with which relations, ‘between’ or ‘within’ phenomena, are uncovered; to accumulate relations – as in putting entities and beings of all kinds into larger contexts – is interpreted as an incremental activity. This is simply a cultural comment. We may also underscore the tendency of the English phrase ‘kin relations’, so prevalent in anthropological discourse, to elide the analytical conceptualization of relations (close to Pina-Cabra’s general relationality) with the reciprocals or reflexivity implied in interaction between kinspersons. Inevitably, different argumentative positions emphasize relations as lying between kinsfolk as discrete persons, or as pointing to their mutual self-definition, or as some mix of the two. That said, such theoretical heterogeneity may strengthen rather than weaken the force of relations as a general concept.  

One argument for holding on to the anthropologists’ strong vocabulary of relations is that it joins the few languages we have, from the life sciences and elsewhere, for dealing with the present ecological mess. A new sense of the fragility of the world, as a bio-physical-social entity, accompanies a new necessity to apprehend the interdependence of entities and beings of all kinds.[17] An appeal to ‘relations’ is crisp and all-embracing. Indeed, it is relations all the way down. And in every quarter: dispensing with the internal-external axis in which much of this account is couched, Barad (2007) argues that no phenomenon exists apart from the ‘intra-action’ of phenomena. An anthropologist might add there are still too many imagined worlds that ignore such realities, and there is much work of criticism still to be done. For, among other things, what such an appeal to relations does not do is dispatch the spectre of an underlying presumption of similarity (between terms) entailed in imagining terms to a relation. This is relevant to activist dimensions of remedial politics, anthropogenically-speaking (Danowski & Viveiros de Castro 2017). Relations so conceived fail to challenge prevailing Anglophone requirements of political action, namely that it proceed through demonstrating similarity or convergence of interests (‘connections’ in this sense) when parties are brought together. Such requirements cannot deal with those social encounters to which, of all disciplines, anthropology has specialist access, namely those based on the collective work of difference and division.[18] The relation, observes Haraway (2003), is about significant otherness at every scale. Her conception of what relating entails is implicitly political in tenor.

Non-relations and post-relations?

This entry has indicated some of the ways anthropologists have used ‘relations’ in the course of their practice, now taken for granted, now explicitly differentiated for this or that purpose. Those ways both cross other currents in social thought and are given prominence in the discipline’s traditional concern with the collective or associational dimension of people’s lives. Attention has been paid to divergences between views, and the manner in which they recur. One thread through these usages is the critical edge that being explicit about relations has brought to debate. 

A remark attributed to the twentieth century anthropologist and ecologist, Gregory Bateson, is that one cannot not relate.[19] Interesting, therefore, is recent critical writing that challenges how relationality, in a social or interpersonal sense, appears to suffuse anthropological accounts. Two examples must serve.[20] Candea et al. (2015) take up the positive affect attributed to relations as inevitably implying the desirability of close ties between people or the mutuality of engagement. These essays seek to re-evaluate detachment and disconnection in social life, analyzing strategies of separation and distancing – relations from another point of view – for their political and ethical interest. In different vein, Holbraad and Pederson (2017: 242-81) ask what comes after the relation. They suggest that by intensifying it beyond recognition one can develop examples of apparently ‘non-relational’ ethnographic moments to sketch what a ‘post-relational’ shift might look like. In the course of this they uncover a renewed vernacular or indigenous (in their examples, Christian) interest in the individual, a connection-cutting entity, which holds out the critical potential of modifying the concept of the relation itself, such that it is no longer ‘owned by’ or ceases ‘to be about’ social relations. As these narratives imply, there is more still to relate.


Allen, N.J. 2000. Categories and classifications: Maussian reflections on the social. Oxford: Berghahn Books.

Barad, K. 2007. Meeting the universe half-way: quantum physics and the entanglement of matter and meaning. Durham: Duke University Press.

Battaglia, D. 2005. Insiders’ voices in outer spaces. In E.T. culture: anthropology in outerspaces (ed.) D. Battaglia, 1–37. Durham: Duke University Press. 

Bonnemère, P. 2018. Acting for others: relational transformations in Papua New Guinea. Chicago: HAU Books. 

Brower, J. 2015. Medieval theories of relations [revised]. In The Stanford encyclopedia of philosophy, (ed.) E.N. Zalta (available on-line: Accessed 4 February 2016. 

Candea, M., J. Cook, C. Trundle, & T. Yarrow (eds) 2015. Detachment: essays on the limits of relational thinking. Manchester: University Press. 

Carsten, J. (ed.) 2000. Cultures of relatedness: new approaches to the study of kinship. Cambridge: University Press.

Corsín Jiménez, A. Willerslev, & R. Willerslev 2007. ‘An anthropological concept of the concept’: reversibility among the Siberian Yukaghirs. Journal of the Royal Anthropological Institute (N.S.) 13, 527–44.

Danowski, D. & E. Viveiros de Castro  2017 [2014]. The ends of the world (trans. R. Nunes). Cambridge: Polity Press. 

Descola, P. 2013 [2005]. Beyond nature and culture (trans. J. Lloyd). Chicago: University Press.

Descombes, V. 2014 [1996]. The institutions of meaning: a defense of anthropological holism (trans. S.A. Schwartz). Cambridge, Mass.: Harvard University Press.

Douglas, M. 1970. Introduction. In Witchcraft confessions and accusations (ed.) M. Douglas, xiii – xxxviii. London: Tavistock Publications. 

Evans-Pritchard, E.E. 1950 [1937]. Witchcraft, oracles and magic among the Azande. Oxford: Clarendon Press.

Frankenberg, R. 1957. Village on the border: a social study of religion, politics and football in a North Wales community. London: Cohen & West.

Frankenberg, R. 1982. A social anthropology for Britain. In Custom and conflict in British society (ed.) R. Frankenberg, 1–35. Manchester: University Press.

Faubion, J.D. (ed.) 2001. The ethics of kinship: ethnographic enquiries. Lanham: Rowman & Littlefeld.

Feldman, G. 2011. If ethnography is more than participant observation, then relations are more than connections: the case for non-local ethnography in a world of apparatuses. Anthropological Theory 11(4), 375–95.

Goody, J. 1962. Death, property and the ancestors: a study of the mortuary customs of the Lodagaa of West Africa. London: Tavistock Publications. 

Haraway, D. 2003. The companion species manifesto: dogs, people, and significant otherness. Chicago: Prickly Paradigm Press.

Holbraad, M. & M.A. Pederson 2017. The ontological turn: an anthropological exposition. Cambridge: University Press. 

Latour, B. 2005. Reassembling the social: an introduction to Actor-network theory. Oxford: University Press.

Lazar, S. 2012. Disjunctive comparison: citizenship and trade unionism in Bolivia and Argentina. Journal of the Royal Anthropological Institute (N.S.) 18(2), 349–68.

Lévi-Strauss, C. 1963 [1945]. Structural analysis in linguistics and anthropology. In Structural anthropology, vol. 1, (trans. C. Jacobson & B.G. Schoepf), 31–54. New York: Basic Books.  

Lévi-Strauss, C. 1978 [1973]. Reflections on the atom of kinship. In Structural anthropology, volume 2, (trans. M. Layton), 82–112. Harmondsworth: Penguin Books.

Moutu, A. 2013. Names are thicker than blood: kinship and ownership amongst the Iatmul. Oxford: OUP for The British Academy.

Ollman, B. 1976 [1971] Alienation: Marx’s conception of man in capitalist society. Cambridge: University Press.

Pina-Cabral, J. de 2017. World: an anthropological examination. Chicago: HAU Books. 

Rabinow, P. 2011. The accompaniment: assembling the contemporary. Chicago: University Press. 

Radcliffe-Brown, A.R. 1952 [1940]. On social structure. In Structure and function in primitive society, 188–204. London: Cohen & West.

Sahlins, M. 2013. What kinship is – and is not. Chicago: University Press. 

Strathern, M. 1992. After nature: English kinship in the late twentieth century. Cambridge: University Press.

Strathern, M. 2005. Embedded science. In Kinship, law and the unexpected: relatives are always a surprise, 33–49Cambridge: University Press.

Strathern, M. 2017. Persons and partible persons, in Schools and styles of anthropological theory, (ed.) M. Candea, 236–46. London: Routledge. 

Viveiros de Castro, E. 2014 [2009]. Cannibal metaphysics for a post-structural anthropology, (trans. & ed. P. Skafish). Minneapolis, Minn.: Univocal Publishing.

Wagner, R.1975.The invention of culture. Englewood Cliffs, N.J.: Prentice-Hall.

Note on contributor

Marilyn Strathern is a former William Wyse Professor of Social Anthropology, and currently Life Fellow of Girton College, Cambridge. Her ethnographic forays are divided between Papua New Guinea and Britain. Apart from gender and kinship, she has written on reproductive technologies, intellectual and cultural property, and audit culture. 

Professor Marilyn Strathern,

[1] It took time before the Aristotlean conviction that one property cannot belong to more than one subject was left to the side. 

[2] All terms lead to other terms, but relative terms show this explicitly: ‘weak’ and ‘strong’ are relatives terms, as are ‘parent’ and ‘child’, each implying the other. Leibniz’s overall argument was consonant with his objection to Isaac Newton’s idea of space as something in itself, within which other objects move; for Leibniz, space was simply the ‘order’ (another word for ‘relation’), in which celestial bodies move in respect of each other. 

[3] Thus, ‘the relation between capital and labor is treated ... as a function of capital itself’ – capital is a (social) relation (Ollman 1976: 13). 

[4] The full passage reads: ‘Multiplicity is a system defined by a modality of relational synthesis different from a connection or conjunction of terms. Deleuze calls it a disjunctive synthesis or inclusive disjunction, a relational mode that does not have similarity or identity as its (formal or final) cause, but divergence or distance; another name for this relational mode is “becoming”’ (2014: 112, emphasis ignored). Disjunctive synthesis is a difference understood as positive rather than oppositive. Deleuze’s specific debt to Leibniz is mentioned. 

[5] With reference to his field material from Melanesia. Moutu’s (2012: 202) observation extends from the proposition that Melanesians take relationships as the implicit ground of being, by contrast with the Euro-American impetus to see ‘making relations / relationships’ as a matter of social agency. 

[6] In some senses, this anticipates an observation from Pina-Cabral (below). 

[7] Elsewhere in this address he takes ‘relations of person to person’ as simply a part of social structure, the other part being the differentiation of individuals and classes by their social role.

[8] For example, ‘[T]he terms of kinship are inherently linking terms; … they render the self in and through its relation to certain others’ (Faubion 2001: 3). (Self and other is an axis often taken as fundamental to people’s conceptualization of relations.) However, Radcliffe-Brown seems to have something more like external relations in mind (social structure as ‘actually existing relations’ that ‘link together certain human beings’ [1952: 192]). 

[9] To offer just one example, Frankenberg’s (1957) focus on the politics of a Welsh village sprang from then-burgeoning interests in African village politics, a comparative agenda carried through in his posing a social anthropology for Britain (1982). 

[10] As Descombes in his discussion of Lévi-Strauss’s work puts it. ‘Structural holism asks us to practice structural analysis as a form of holistic analysis, i.e., as a search for the relations that ground the system’ (Descombes 2014: 157). His own account develops the proposition that no social interaction takes place without a third term, that is, the taken for granted, instituted meanings of collective life. Thus in gift exchange between persons, the whole is given before its parts in that a ‘gift’ is already following the conventions of ‘gift giving’. 

[11] Whether or not causation is involved.

[12] Controversy lies in the way that last usage is criticized in turn for the implication, from a ‘western’ perspective, that relations flourish in other, invariably ‘non-western’, places more heartily than at home.

[13] Whether of whole cultural orientations or within the dynamics of specific interactions. On the extent to which people do or do not take a relational world as having to be ‘made’, see Wagner (1975); Note v., above. 

[14] Contrast the purpose Carsten (2000) has for the general term ‘relatedness’, an analytical placeholder to avoid pre-empting assumptions about the nature of kinship.

[15] Because of the ontogenetic – ever developing– character of persons. Per contra, Rabinow (2011) sees a transcendental quality in the relational interactions of ‘assemblages’, insofar as any kind of entity has the capacity to be open to another. 

[16]  The oppositional mode of connection / disconnection is not the same as the disjunctive synthesis noted above (Note iv.).

[17] This very phrasing is positivist, but there is some (political) advantage in it being one of the positions accommodated by a portmanteau appeal to relations. Within anthropology, it should be added, there is much present interest, from a ‘human’ perspective, on (variously) human and animal, human and nonhuman, or human and other-than-human, relations. 

[18] The case is argued apropos concepts of personhood in, for example, Strathern 2017. 

[19] Explicitness about present or absent relations can be evidence of relational thinking; however, an enacted relation (anthropologist speaking) emptied of engagement or attachment may be rendered as a ‘non-relation’ in the English vernacular. 

[20] Both volumes point to a wave of twenty-first century arguments, stimulated by diverse theoretical perspectives, about the limits of the relation as an anthropological analytic. As the connotations of relation shifts, so do the terms around it. Thus the individual person, as a logical concept always relationally constructed with respect to other concepts, may be identified as a relational configuration socially speaking, in which individualism is a knowing strategy. 

If you liked what you just read

The Cambridge Encyclopaedia of Anthropology is an open access resource. Please help us keep it that way by making a one-time or a regular donation. 

Support the encyclopeadia