Anthropological reasoning

This work is licensed under the Creative Commons | © Marilyn Strathern. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau4.3.003

Anthropological reasoning

Some threads of thought

Marilyn STRATHERN, University of Cambridge

The interventionist properties of description are considered in relation to two strands of thinking, each as evidently “outside” anthropology as “inside.” In terms of concept formation, the nature-culture dyad seems forever to be subject to critique, reformulation, and re-critique; examples from current debate over clinical practices in South America make the point. In terms of engagement with “human subjects,” anthropology has been as much heir to regimes of audit and self-scrutiny as it has shown their limits; the reflexivity now routine in ethnographic inquiry is shown up in approaches to present-day health policies for Aboriginal people in Australia. Both arenas (nature-culture/self-scrutiny) have contributed at once to anthropology’s self-formation and to the kind of knowledge it makes more widely visible. Both were also topics of huge interest to the European Enlightenment. A suggestion is proffered about the outlines of a newly apparent object of knowledge then, which could have been something of a driver, and seems to have been a driver of anthropological reasoning ever since.

Keywords: health interventions, relations, nature-culture, self-scrutiny, the Enlightenment


Given the ambition of this collection to think about anthropology’s world, its conditions of possibility, and the clues to that contained in what counts as interventions in it, we may wonder about the discipline’s relational practices. What, for instance, might be learned from colleagues? Anthropologists’ descriptions of professionals whose very job it is to make interventions in people’s lives reflect back on their anthropology in interesting ways. Indeed, an ethnographer who worked with a community of health professionals in Australia’s Northern Territory puts into words something akin to the position I wish to take here. Where knowing interventions in the world of affairs rest on describing what is happening, then description itself is an intervention.1

Medicine has always been an arena in which interventions are deliberate, aimed to be effective and to have benign outcomes. Given the endemic ill health of the local Aboriginal population, Territory Health Services address an ever-deteriorating situation with ever more effort at intervening. They apply a huge organizational apparatus, the need to coordinate services proliferating in schemes for “collaboration, cooperation, networking, team-building, [and] information sharing” (Lea 2008: 63). Yet the more the administration tries, the more apparatus is needed: the greater coordination to which the administration aspires, the greater fragmentation it sees.2 This underwrites a bureaucratic project. People wanting to deliver care do their best by the systems of organization that also organize them: intervention through organizational means implies making the organization work.3

One means is through self-description. The organization keeps detailed track of itself, and we should not be surprised to find auditing procedures addressed not only to the outcomes for the health of the population but also to the officials’ organizational effectiveness as service-providers. Implementation plans are rolled out with performance indicators, and targets are given numerical thresholds, alongside annual reports, data summaries, program reviews, workshop recommendations, and so on. An inordinate amount of time, many complain, is taken up with paperwork: that is, with describing what is happening. Yet describing bureaucratic performance itself is seen as a precondition to changing things.4 At the end of every day of their exhausting induction “in the field,” new recruits are presented with evaluation sheets (Lea 2008: 87). Regardless of any external monitoring, and always working to improve the program, the organizers themselves beg for feedback from those they have been inducting on the way they have carried out their job (of induction).

This is the world in which the anthropologist also lives. Lea minces no words about the liberal rationalism that brings bureaucrats and anthropologists into proximity; rather than simply chiding bureaucratized perceptions for their limitations, anthropologists might also want “to comprehend the cultural habits we share with their formulators” (ibid.: 234). What is interesting in Lea’s rehearsal of this stance, which has been the subject of much debate in the discipline at large (“trying to make sense of people whose job it is to produce sense,” as she puts it [ibid.]), is her emphasis on the task of description. “We” anthropologists “are ensnared by the same belief in the power of representation to bring the misconceptions of others to correcting light”’ (ibid.). Her monograph concludes that, for all the problems there are,

it can be liberating to gain some sense of the lived-in, externally driven and consensual limits on our own agency. Or at least that is my hope, a hope which is the ultimate honouring of bureaucratic magic and the faith it sustains in the power of description to amend conceptualizations of how the world “really is,” and with that improved perception to somehow yield a better outcome. (Lea 2008: 237)

To begin with this thoroughly conflicted situation dispenses with the thought that interventions are conflict-free. Yet I have to share the hope that Lea expresses. For someone whose tools of work include making descriptions, it is close to the hope that anthropologists responding to the Japanese earthquake and tsunami of 2011 have powerfully expressed as a matter of retooling (Riles, Miyazaki, and Genda 2012). You don’t abandon your tools: you go back to the tools you already have and reconstruct them according to the current situation. But in the case of those anthropologists dealing with the aftermath of the natural disasters in Japan, you do it with the professionals.5 In a commentary on the Australian material, and with Graeber’s (this volume) contribution in mind, Salmond (pers. comm.)6 poses a similar relational question, this time with respect to anthropologists/bureaucrats and the subjects of their inquiries. There are other ways of interacting, other forms of attentiveness, that might make of the latter something more than a means for improving the internal processes of the former. Of course. Nonetheless, just how relations themselves come into focus as an object of knowledge is of moment in itself.

Inside and outside the discipline

In the invitation to pursue threads of thought across material that lies as much outside anthropology as within it, what strength might a thread have? Rather than as a road to go down or rope to hang onto, and without the regular patterning of warp and weft, I take a thread to be something that can be caught, both caught hold of and getting itself caught onto what is in its vicinity. Although going somewhere because it is coming from somewhere, a thread gains what striated substance it has in becoming entangled with other threads just as (at the moment when) the entanglements (knots) seem to make further tracing impossible. Any particular thread of thought might appear as a singular twist, might seemingly take the form of a genealogy or archaeology, but in truth it was never unknotted from innumerable others.

“Inside” and “outside” are altogether more ambiguous, for there is nothing unusual about anthropology as a knowledge system to observe of it that anything it touches “becomes” anthropology.7 We have to be wary, though, of assuming that we already know the anthropology we are within: postulating an outside is to postulate another perspective. I catch the threads of two apparently distinct, certainly ongoing, discussions to elucidate the point, and which will compose my subject matter. First, in terms of concept formation, the nature-culture or nature-society dyad seems forever subject to fresh criticism and reformulation, and forever to regain its influence. Secondly, in terms of engagement with “human subjects,” anthropology has been as much heir to regimes of audit and self-scrutiny as it has shown their limits. Both arenas have contributed to anthropology’s self-formation and to the kinds of knowledge it makes more widely visible. Both were also topics of huge interest to the English/Scottish versions of the European Enlightenment.8 This leap is a little more artful than clutching at a straw (hardly a thread at all): I am drawn (by stronger threads) in that direction by what I already know, to places in which I have already been entangled. Seemingly separate threads may indeed at one stage have been knotted together.

Thinkers of the eighteenth century Enlightenment cultivated the secular humanism from which a self-acknowledged economics was to emerge, a precondition to the divergences that Gregory (this volume) identifies. Its [the period’s] interest to the present argument is for what it suggests about certain preconditions within anthropological reasoning that have made the two arenas (self-scrutiny, nature-society) so enduringly salient. Familiar to anthropological debates, they are equally familiar to the humanities and social sciences, and beyond academia altogether. To keep discussion brief, I refer to some ethnographic situations where ideas springing from these arenas seem to have been the carriers of interventions, by anthropologists and by others. One such situation has already appeared. The reflexivity that has become routine in both anthropological and ethnographic enquiry in some senses mirrors the self-conscious auditing of present-day health policies for Aboriginal people in Australia. As for nature and society, to keep a parity of sorts examples will come from current discussions of clinical practices in South America, although they will be treated at somewhat greater length.

A short thread: Self-scrutiny

Description seems such a mundane activity for the text-based scholar that it is interesting to come across it as a focus of attention in other fields. Corsin Jimenez’s (2013) account of the strabismic double vision of anthropological knowledge making contains an extended discussion of Alpers’ work on realism in seventeenth-century Dutch painting. The art historian called her book The art of describing (1983).9 We quickly learn it was description in a special sense—the technique of making the observer disappear by dislocating the eye from a single viewpoint. “The mobile eye [of Dutch composition] cuts the world two ways: it multiplies with one eye what it divides with the other, and in doing so opens up a space for a third form of vision: a ‘seeing double’ that is more than one and less than many” (Corsin Jimenez 2013: 54). Among several “scopic regimes of modernity” (after Jay 1988), it is one—and Corsin Jimenez draws attention to the connection—that might have been shared with microscopists and optical experimenters of the time. The world seen microscopically both multiplies the innumerable small elements within a larger body and in enlargement divides a small part from the whole (my paraphrasing of Alpers). Painter and microscopist alike treated the eye as taking in the world. “A split eye signals the birth of . . . an aesthetic ‘which consisted in making something visible, in being a pure apparition that made appearance appear, from a position just on its edges’ . . . (Buci-Glucksmann [1984] 1994: 60)’ (Corsin Jimenez 2013: 55–56, my emphasis).

This is not the place to start undoing the many knots these writers have so skillfully tied. But conceivably we have here some of the conditions of possibility for what was to be reconceived as a splitting of the self.10 In his discussion of the Enlightenment formulation of the self as a rational entity through being divided into two, a management theorist, Keith Hoskin (1996: 270), picks up another thread. The eighteenth-century philosopher David Hume “proposed that the self becomes ethical by considering how one’s actions are viewed by an ‘impartial spectator’” (ibid.: 271). Let me give Hoskin’s argument in full, the line he draws running through Hume from Thomas Hobbes, a century before, to Hume’s contemporary, Adam Smith.

Where Hobbes had looked out, through his perspectival window, onto the foolish and selfish selves in the thought-world beyond, Hume had attempted to construct a sort of reverse viewing, from the thought-world beyond, back on to the self who views. Hume’s formulation of the impartial spectator paints a picture wherein I become ethical by looking through my window on the world and seeing there some moral Other, the impartial spectator, whose surveillance of me ensures that I develop moral sentiments. On this reading, what Smith achieves is a step beyond this, by discovering on the far side of the “window” a mirror self, the self that examines, which stands in for and replaces the Other as judge of self. (Hoskin 1996: 271)

Hoskin articulated a historical trajectory that set the conditions of possibility for new forms of description. This was the description of performance to which measurements could be given, applicable to university examinations and financial auditing alike. The ethical double self was the self-examining self.11 Other (nonpictorial) conventions of description were to develop, and at various times anthropologists have rehearsed all three positions (Hobbes, Hume, Smith), whether following notions of the privileged observer, of the relativity of the familiar and unfamiliar, or of discovering the self in the other. Where philosophers were concerned with ethical formation, for anthropologists these positions point to diverse concerns with the accuracy or appropriateness of the social knowledge so gained—”social” insofar as such knowledge rests on the kinds of relations the writer has with his or her subjects of study and fellow academics. The gap between the Australian ethnographer and the Northern Territory health bureaucrats, as Lea describes it, is also their proximity: the relation lay in the nature of their descriptive endeavors. To see this rapport is itself something of a descriptive intervention on the anthropologist’s part.

Longer threads: Nature and society

Suppose you are in a place where it is far from clear what the ontological status of people’s “descriptions” of themselves might be. Suppose, for example, that people think it is possible to change one’s “race” or “body,” as we know is true in parts of South America. Euro-Americans would not be surprised if the anthropological description of those particular “descriptions” become entangled in the language of nature and society, or nature and culture. In the interventions I am thinking of, such a language pulls at two kinds of threads. On the one hand, ideas about a natural world and its progressive modernization appear to underlie the efforts of some of the health practitioners who treat these people; on the other hand, the anthropologist finds disciplinary theorizing on the topic a crucial analytical tool.

Take the bodily change implied by the process of “becoming white.” Becoming white is becoming criollo or creole, to take as reference points Venezuelan whites in Amazonia (Kelly 2011) or Ecuadorian whites in the Andes (Roberts 2012, 2013). In various ways these whites would describe themselves as moderns. All those activities requiring organizational effort that characterize bureaucracies are equally projects of a global modernity. In the Venezuelan Amazon, they are found in the travails of young doctors on field assignments, undergoing training in a state-provided health service. (Kelly [2011: 13] refers to the doctors’ ideologies as “Western” or “Euro-American.”) In the Ecuadorian Andes, however, the modernization associated with technological interventions sits a bit differently, as we shall see; here the ethnographic context is that of private fertility clinics. The anthropologists describing these moderns either link their efforts to ideas about nature and society, or make a contrast with those who hold such ideas. For both ethnographic situations, the kind of medical care being offered is taken by the recipients or clients as signs of their own increasing ability to enact whiteness. Through receiving this care, they become “white people” This in turn looks like a case of indigenous redescription.

In their civilizing mission, Venezuelan doctors—criollos—working among Yanomami Indians “make bodies,” in the sense of changing what they would see as indigenous lifestyles, and “make society” thereby. “[One] component of the civilizing project is about making society . . . because for criollos, society was missing and needed to be made’ (Kelly 2011: 97). Again, “doctors see Yanomami as part of nature: [above all] disorganized” and “the more disorganized Yanomami appear, the more doctors strive to organize” (ibid.: 137, sentences transposed). It is the criollo nature-society axis that leads Kelly to distinguish what Yanomami and criollo each take for granted, that is, between what is assumed as a given and what is regarded as subject to human intervention. In this respect, criollos participate in the general Euro-American understanding of society as a developmental project in relation to nature. Social conventions that are innate for Yanomami are fabricated for the criollo. However, Kelly (ibid.: 98, 222–3) is at pains to point out the knots here; neither the understandings nor the misunderstandings each has of the other match up.12 Neither should we jump to any easy conclusion about what is involved in Yanomami “redescribing” themselves.

Now the kind of care that Yanomami would like to elicit from the doctors—a wish that is often frustrated because staff are stationed for such brief periods—is the care of kinship. In the comment about what criollos thought was missing, I missed out a phrase: the full phrasing reads: “[One] component of the [doctors’] civilizing project is about making society, not kin” (ibid.: 97, my emphasis), for making kin is a Yanomami aspiration. Like race (whiteness), kinship is to be made. In the case of criollos, outsiders must at least be domesticated into the status of potential in-laws (affines) who will mediate between Yanomami and the outside world (bringing its knowledge and goods, not restricted to medical care). So as much as criollos would change them, Yanomami are trying to change criollos—not to go all the way and make them Yanomami, but to turn them into potential in-laws (there is no intermarriage; the potentiality remains), thus keeping up the flow of the distinctive services, knowledge, and goods that only criollos can bring. At this point in the argument, the descriptive intervention of the anthropologist becomes apparent. In Kelly’s view, if there were an indigenous analogue of nature it would be affinity. He writes, “doctors and Yanomami enter each other’s worlds as forms of the innate,” “innate” being subsequently glossed, “potential affinity and nature, respectively” (ibid: 137, 225). The expository consequences of this intervention, the analogy, are evident. We come to understand that affinity is an innate condition of the world, a world of social convention, which Yanomami take for granted.13

It is the quality of affinal relations that makes sense of the apparent continuum in becoming white. Yanomami describe one another in terms of those who are more and those who are less white; crucially, however, any specific point along this continuum juxtaposes positions that are radically “other” to each other. Otherness is not diminished by the continuum; rather, people assume positions that are now white, now Yanomami.14 This is not seen by the medical staff, who would themselves understand “becoming criollo” as an evolutionary or developmental process. In this sense, the becoming/being white that might have a transcendent value for criollo medical staff (the inevitability of the modernization that they at once represent and that goes beyond their efforts to assist it) holds a very different place in Yanomami thinking. What for Yanomami is prior to and beyond human agency is the social condition of alterity: there will and must always be others. Yanomami are not so much redescribing themselves as distributing themselves at different moments in time and place between already-existing descriptions.

Roberts’ account shifts the axis of analysis. She sustains a running contrast with attitudes derived from the European Enlightenment, such as those found in twentieth- and twenty-first century Euro-American accounts of assisted conception (e.g. Roberts 2012: 94–5, 114). Among them is the kind of agency entailed in a once firmly held nature-society (she talks of nature-culture) paradigm whose starting point was that nature, to which “the body” belongs, is a more or less immutable given. Now whereas the effect of Kelly’s recourse to this topic (nature-society) is to illuminate the way ideas held by the Venezuelan doctors are entwined with those of anthropological discourse, Roberts separates out the threads. What might be true for anthropological accounts of assisted conception in North America (say) is not a truth that holds across all of the Ecuadorian clinics.

Roberts argues that many (and, conversely, not all) Ecuadorian clinicians share with their clientele approaches likely to have their roots in religious categories based on prebiological determinations15 of lineage, which were precursors to the present-day concept of ‘race’ (ibid.: 117). The continuing assumption is that ‘in Ecuador, people can change their race’ (ibid.: 114). Here is another descriptive intervention on the anthropologist’s part.16 Its consequence is the weight to be put on “the multiplicity and plasticity of the ontological world” (ibid.). The phrasing (“people can change their race”) is intended to describe neither logical confusion nor social construction but the ever-present possibility of ontological transformation. These people are participating in Ecuador’s nation making through what Roberts calls a national whitening project, the continuation of earlier “whitening interventions” (2013: 568). Especially in the case of the poor or clients of Indian background, simply to be the object of such attention is transformative. Under the care of whites (as doctors and clinical staff categorically are), the client becomes white. “Through IVF,” Roberts says (2012: 75), “women can become whiter reproducers . . . [that is,] through being cared for the way whiter women are cared for.”17 This is based on a grounding assumption about the malleability of bodies, and produces redescriptions very different from those the ethnographer can accomplish.

There is something to be said about the kinds of social relations at issue too. The epithet “assisted” before “assisted reproduction” takes on unexpected resonances. In this Catholic country, whose church frowns on fertility procedures, God is everywhere, and an immediate not distant presence. God’s giving of life is constantly brought home. But rather than it halting their activities, this presence is harnessed by the clinics. The scientific techniques that assist the procedure are also the means through which clinical personnel assist God: in deliberately evoking common Euro-American notions, Roberts observes that they do not perceive themselves to be “playing God” but rather as being “God’s helpers.” Hierarchically speaking, patients are in turn assisted. The focus of assistance for many patients is less about the modification of the materials of fertility (IVF, egg or sperm donation, embryo freezing) than about the patronage bestowed upon them. There is regional variation, but, generally speaking, “private medical clinics are similar to haciendas18 in that the relations are paternalistic (not consumer oriented)” (ibid.: 57). Patronage is acted out in the clinic through all the social activity that is entailed in attending to the patient’s health and comfort. This relationship of dependence is actively sought by many Ecuadorian women with fertility problems: “Assistance, not autonomy, is the very basis of existence” (ibid.: 212). The clinic demonstrates that the patient is worthy of care and of the patron’s notice.

Is this the kind of pre-Enlightenment modernity that some Europeans of that moment would have recognized? Roberts (ibid.: 116–18) gives her own history of various Creole concepts that eventually became the basis of “race,” and it is a complex one. However, a significant axis, which she clearly sees as a departure from Enlightenment ideas (and she refers to them as such), concerns the character of God’s “intervention” in everyday life, as in enabling people to have children, and thus concerns understandings of nature. A noninterventionist God is another proposition altogether. Roberts quotes Hume to the effect that a miracle would be a violation of the laws of nature. Catching that thread, I elongate it. Rather than marking a section break here, this turn to the (English-speaking19) eighteenth century simply makes this section longer.

The same Enlightenment philosopher who was caught up in ideas about self-scrutiny was of his time in also being committed to elucidating human nature. Hume argued against the political theorist Hobbes on this score too: natural leanings are not to be reviled but to be cultivated.20 We may note that one means of cultivation in the (social) world at large was the cultivation of connections. Connections referred to persons also known as kin, familial or otherwise, or more generally one’s associates. Arguably such connecting first stood alongside and then displaced earlier perceptions of patronage; the new sense of connection could well have brought a new sense of choice in laying claim to those one knew. The very concept itself was different. “Patronage”—as between Ecuadorian clinical staff and clients—described the quality of a specific tie between persons known to one another; “connection,” by contrast, was a generic term (Tadmor 2001: 131–32) of unspecified possibility, out of which one chose to emphasize this or that associate or refuse others. For the propertied and aspiring middle classes of the English-speaking eighteenth century, one could almost say that acquiring good connections was like becoming white.

Nonetheless, there was a crucial difference: Enlightenment notions of refinement and cultivation were constantly tethered back to a view that one should not move too far from nature (one’s own nature or the workings of nature in general). A relation existed here,21 quite as much as the relation between the two selves, in that cultivation (“society”) and nature also entailed each other. In this context, Hume’s project of putting human understanding itself (understanding understanding) on a “scientific” basis was to align description more accurately with what was now being understood, redescribed, by many writers as an autonomous natural world. Further, and specifically, the eighteenth-century idea of desirable connections mobilized the concept of affinity through connotations of similarity. Similarity held in a double sense, both in terms of one of the dominant usages of “connection” to refer to alliances between families through marriage, and in terms of the likeness found among same-status kin or associates. By contrast, patron and client were differentiated not just by social standing but also by a hierarchical relationship. We might recall that Yanomami affines made others “other” to themselves. In Yanomami, affines evoke alterity, not likeness, and the transactions that made one “more white,” apparent interventions along a continuum, in fact enacted at every juncture a radical break. If, in what was taken for granted in relation to criollos, the Yanomami analogue of nature was indeed affinity, then affinity had a quite different location in the kinship universe from that imagined by English-speakers.22 In English, affinity, by derivation connoting “on the borders” or “bordering on,” and thus neighboring or near to others, gives the viewpoint of the speaker as an ego in a center looking outward. Indeed there is, in this respect, no ontological difference in the radiating spheres of affinity and of consanguinity.23

These generic English-language concepts for interpersonal ties all invited calculations as to nearness to and distance from the speaker. Perhaps we could say that they made such calculations into “interventions” in a social world. To solicit or deny a (social, kinship) connection was not to evoke a preexisting alterity, but to produce or create difference or sameness anew (either up or down). Every assertion of or denial of recognition was a social intervention. The same might be said, epis-temologically speaking, of every interpretation or choice of descriptive language, or scientific investigation. Producing such a specific effect (the named, discovered, invented) mirrors back the producer’s viewpoint. Whatever presented conditions of possibility for discussion about society and nature, then, we might want them to account for interpersonal relations quite as much as for ideas about civilization and the physicality of the body.

The knot

The kind of description at which anthropology excels is expository, making an interpretive relationship with its world explicit. However lyrical or empirical the form, or however knowing the doubt (Carrithers, this volume), a view about understanding is being expounded. In other words, the view implies something about to be understood. Yet because different moments or aspects of elucidation come into view at different times, there is nothing new in saying that exposition is also an entanglement of descriptions from many hands. In other words, anything available to description is of course already described, a kind of strabismic precondition of anthropological knowledge making.24 That does not mean one has to put one’s tools aside. It has been helpful to consider as interventions two clusters of concepts (self-scrutiny, nature-society) that animate anthropological debate within as well as flourishing on its outside, germane to the kind of knowledge anthropology makes visible. I have threaded them through three ethnographic loci to make visible another facet of anthropological knowledge making: the descriptive endeavor does not get very far without the invocation of social as well as conceptual relations.

It is a truism that specific times and places yield specific descriptive genres. Yet perhaps there is something particularly interesting for social anthropology in some of the descriptions of the English-speaking world prevalent at about the time of the scientific revolution, and subsequently sedimented in Enlightenment sensibility. Present-day anthropologists are thoroughly aware that much of their social knowledge consists in relating relations. That stance, relating relations, finds echoes in certain kinds (not all) of painterly and experimental optics whose descriptive effect can be explained more generally as making appearance itself appear—as I flagged in the quotation from Buci-Glucksmann above. It seems to be repeated in certain (not all) philosophical musings on human nature that made explicit the notion of understanding understanding. Appearance as an object of viewing, understanding as an object of reflection: through specific ideas of eye and mind they become objects of description. But we have already slipped through the knots by which these two areas of discussion are tied together—there is no need to tie them more tightly. Rather, we might ask if that other thread can be threaded in as well. Should we be asking about other conditions of possibility? Should the anthropologist be pondering on how relations became an object of description, and thus of knowledge, too?

One answer surely has to be through acts of relating, including relating between persons. I do not mean here the social networking that in Europe of the seventeenth and eighteenth centuries informed the verification of experiments or gave writers their audiences, but rather a particular kind of objectification (as in “object of knowledge”) of the concept of relations. It was one that outlined the contours of the notion simultaneously in reference to and distinct from its enactment among persons. Conceivably, for English-speakers, it is there in their usages of affinity and connection.

At this time there was a kind of two-way passage in the connotations of these terms. They acquired, one could say, at once a strength and a striation of sorts. On the one hand, the concept of “affinity” moved from its delineation of kinship relations by marriage and alliance to include epistemically understood resemblances or rapports, perhaps thereby endorsing25 an abstract stress on empathy or likeness. On the other hand, in the opposite direction, there was an expansion of “connection,” from an epistemological or conceptual usage in the seventeenth century to, in the eighteenth, embracing kin as well, although by no means exclusively.26 In fact the introduction of this specific abstraction (connection)27 into the kinship universe mirrored an earlier change in the way “relations” itself was used—from a term for epistemic relations to, in the seventeenth century, a term also used for related persons (in this case restricted to kinsfolk, consanguineal and affinal alike). Might we see here diverse outlines of a newly visible attention to knowledge about persons and their interactions? If so, they would come with injunctions (albeit of etiquette and politicking) about the social— and thus interventionist—consequences of acknowledging people’s relatedness. Variously termed, a relation was already a concept that pointed to (the act of) relating as itself an object that could be described.

Anthropologists might think it banal in the extreme to say that the threads of self-scrutiny and the nature-society debate are tied together in relations, that is, in a presupposed ability to perceive relations at once internal to these concepts and external with respect to others. That this is true of countless occurrences would seem to deflate the interest of any single set. Yet following these particular strands of thought catches, and momentarily catches on, some taken-for-granted banalities that intervene all the time in their/our descriptions.


Jeanette Edwards’ characteristically illuminating comments have since taken me in other directions, but that is not reflected here: this is basically the text on which she commented. My appreciation of Sarah Green’s open invitation should be evident; my gratitude to HAU’S reviewers is expressed in situ (note 6). And I extend warm thanks to Jose Kelly and Elizabeth Roberts for their helpful elucidations of my earlier use of their material.


Alpers, Svetlana. 1983. The art of describing: Dutch art in the seventeenth century. Chicago: University of Chicago Press.

Buci-Glucksmann, Christine. (1984) 1994. Baroque reason: The aesthetics of modernity. Translated by Patrick Camiller. London: Sage.

Corsin Jimenez, Alberto. 2013. An anthropological trompe loeil for a common world: An essay on the economy of knowledge. Oxford: Berghahn.

Holbraad, Martin. 2008. “Definitive evidence, from Cuban gods.” Special issue, “The objects of evidence: Anthropological approaches to the production of knowledge,” edited by Matthew Engelke, Journal of the Royal Anthropological Institute (N.S.), S93–109.

Holbraad, Martin. 2009. “Ontography and alterity: Defining anthropological truth.” Special issue, “What is happening to epistemology?,” edited by Christina Toren and Joao de Pina-Cabral, Social Analysis 53 (2): 80–93.

Hoskin, Keith. 1995. “The viewing self and the world we view: Beyond the perspectival illusion.” Organization 2 (1): 141–62.

———. 1996. “The ‘awful idea of accountability’: Inscribing people into the measurement of objects.” In Accountabiity: Power, ethos and the technologies of managing, edited by Rollan Munro and Jan Mouritsen, 265–82. London: International Thomson Business Press.

Jay, Martin. 1988. “Scopic regimes of modernity,” in Vision and visuality, edited by Hal Foster, 3–27. Seattle: Bay Press.

Kelly, Jose. 2011. State healthcare and Yanomami transformations. Tucson: University of Arizona Press.

Lea, Tess. 2008. Bureaucrats and bleeding hearts: Indigenous health in Northern Australia. Sydney: UNSW Press.

Luhmann, Niklas. 1990. Essays on self-reference. New York: Columbia University Press.

Perry, Ruth. 2004. Novel relations: The transformation of kinship in English literature and culture 1748–1818. Cambridge: Cambridge University Press.

Porter, Roy. 2000. Enlightenment: Britain and the creation of the modern world. London: Allen Lane.

Riles, Annelise. 2013. “Market collaboration: Finance, culture, and ethnography after neo-liberalism,” American Anthropologist 115 (4): 555–69.

Riles, Annelise, Hirozaku Miyazaki, and Yuji Genda. 2012. “Re-tooling: Techniques for an uncertain world.” Translated by Chika Watanabe. Unpublished manuscript, Cornell University Law School.

Roberts, Elizabeth. 2012. God’s Laboratory: Assisted reproduction in the Andes. Berkeley: University of California Press.

———. 2013. “Assisted existence: An ethnography of being in Ecuador.” Journal of the Royal Anthropological Institute (N.S.) 19 (3): 562–80.

Strathern, Marilyn. 2014a. “The academic as examiner.” In Academic working lives: Experience, practice and change, edited by Lynne Gornall, Caryn Cook, Lyn Daunton, Jane Salisbury, and Brychan Thomas, 143–52. London: Bloomsbury Academic.

———. 2014b. “Reading relations backwards.” Journal of the Royal Anthropological Institute (N.S.) 20 (1): 3–19.

Tadmor, Naomi. 2001. Family and friends in eighteenth-century England: Household, kinship, and patronage. Cambridge: Cambridge University Press.

Viveiros de Castro, Eduardo. 2001. “GUT feelings about Amazonia: Potential affinity.” In Beyond the visible and the material: The Amerindianization of society in the work of Peter Riviere, edited by Laura Rival and Neil Whitehead, 19–43. Oxford: Oxford University Press.

———. 2009. “The gift and the given: Three nano-essays on kinship and magic.” In Kinship and beyond: The genealogical model reconsidered, edited by Sandra Bamford and James Leach, 232–68. New York: Berghahn.

Le raisonnement anthropologique: quelques fils de pensee

Résumé : Les proprietes interventionnistes de la description sont considérées a la lumière de deux modes de pensée, tout deux évidemment ‘interne’ et ‘externe’ à l’anthropologie. Du point de vue de la formation des concepts, la dyade nature-culture semble faire l’objet de critiques et de reformulations toujours renouvelees, comme en temoigne le debat actuel sur les pratiques cliniques en Amerique du Sud. En ce qui concerne l’engagement avec des ‘sujets humains’, l’anthropologie a herite des regimes de controle et d’auto-analyse mais en a egalement demontre les limites; on peut voir cette meme reflexivite, qui fait maintenant partie de la routine de l’enquete ethnographique, a l’œuvre dans l’approche adoptee par les politiques de santé conçues pour la population Aborigène en Australie. Ces deux themes (nature-culture/auto-analyse) ont contribué autant à l’auto-formation de l’anthropologie qu’au type de savoir qu’elle rend visible. Tous deux passionnerent egalement les Lumières Européennes. Cet article esquisse ainsi les contours possibles d’un objet de savoir devenu apparent depuis peu, qui pourrait avoir été une sorte de moteur, et qui semble avoir conduit le raisonnement anthropologique depuis son apparition.

Marilyn STRATHERN had the good fortune to receive initial—and indelible—training in Papua New Guinea, which led to work, among other things, on kinship and gender relations. In the United Kingdom she subsequently became involved with anthropological approaches to the new reproductive technologies, intellectual property, audit cultures, and interdisciplinarity. Now retired from the Cambridge Department of Social Anthropology, she is (honorary) life president of the Association of Social Anthropologists. Strathern is currently working on issues in the conceptualization of relations, some of which were sketched out in her 2005 book, Kinship, law and the unexpected: Relatives are often a surprise.

Marilyn Strathern
Girton College
Cambridge CB3 0JG


1. The status of the activity of “description” is key here, and would be central to any comparison with (say) Holbraad’s (2008, 2009) work on the inventive definitions or “infini-tions” that characterize Cuban diviners’ pronouncements; these are imagined not as claims about the world but as ontological interferences in it.

2. Lea describes how “coordination” takes on a social life of its own, producing organizational complexes that succeed in the “unintended consequence of pinpointing the need for more effective coordination” (2008: 63). I do not do justice here to Lea’s sympathetic rendering of bureaucratic lives (see further Strathern 2014a).

3. All the workshops, fact-finding missions, and health instruction programs which the health officers try to bring to the local population rest on the conviction that “better quality and more accurate information will eventually become better self-understanding for the Aborigines” (Lea 2008: 121). By the same token, the better the data are collected and analyzed, the nearer the bureaucrats will be to implementing the “helping state.” Welfare bureaucracies attempt to change the world by orienting the “bureaucratic inhabitant so that she or he conceptualizes the world in terms of reform and intervention” (ibid.: 225). Is this in part the academization of bureaucracy?

4. In which the reviewers and reviewed concur. Added to anxiety about the health workers’ own efforts to improve things for the local population was a quite different set of anxieties about the ethics of intervention: in this particular context, health workers expressed anxiety about causing collateral damage to indigenous culture.

5. The authors’ disciplines cover anthropology, law, and labor economics. I am very grateful to them for permission to cite this work, intended for Japanese professionals—from one set of professionals to another—as yet unpublished in English. “Retooling” is not to be taken lightly, and in evoking a collaborative endpoint Riles (2013) lays out some of the interpersonal and epistemic complexities of this instance.

6. She trenchantly turns the critique of bureaucrats’ concerns with their own self-elaboration back on anthropology itself, a stance with which Lea would sympathize. I am grateful for permission to cite her thoughtful and illuminating comments, which come from a review for HAU that she de-anonymized. Let me add here that I am at the same time grateful to the several anonymous reviewers of the draft, some of whose words I have also borrowed.

7. I am thinking of Luhmann’s (1990) “system,” whose environment is already systematized; reflections on the insides and outsides of knowledge as such would need to take into account Corsin Jimenez’s (2013) recent excursus.

8. “English” to reflect the language; “Scottish” to reflect its seats of learning. I am not here defending the idea of the “Enlightenment,” any more than I intend an informed/ old-fashioned periodization by the references to different “centuries”—each should be taken as deploying the same deliberate clumsiness as “Euro-American”; the clumsiness signals ffd(d), for further definition (should definition be desirable).

9. Which is what perspective entails. The interest of Alpers’ work for Corsin Jimenez (e.g. as he quotes her at 2013: 53) is that, by contrast with the Italian solution, Dutch composition presumed an aggregate of views made possible by a mobile eye, so an optical capacity was added to a perspectival one.

10. I say reconceived, since Hoskin (1995—and Hoskin is discussed by Corsin Jimenez on this point [2013: 57-58]) argues for earlier, preperspectival, models of a split self (through double-entry bookkeeping).

11. Hoskin (1995: 156), who also draws on Alpers, says that by his introducing “the Self into the epistemological space where only the Other has been before . . . I see Smith as doing for epistemological space what the art of describing had done for pictorial space.”

12. “Yanomami are not becoming the criollos we conceive of, nor are the Indians the state wants to reinstate the ones Yanomami take themselves to be” (ibid.: 194).

13. Against (from) which background consanguines have to be purposively carved out (ibid.: 95, quoting Viveiros de Castro 2001: 26).

14. “In becoming nape [Yanomami term for criollo] or civilazado, then, Yanomami are not becoming or seeking to become mestizo or to be assimilated with whites; on the contrary, the relation between Yanomami and nape must remain, for in the nape transformational context, people must be able to alternate from one meaningful position to another” (ibid.: 221, also 137). Nape are non-Yanomami whites, that is, whites in relation to Yanomami.

15. “To this day [’race’] is enacted through profession, language, and level of education” (ibid.: 9). Children of the same family can be of different “race” (ibid.: 121).

16. Roberts quotes several ethnographers who have said the same. In neither this nor the Amazonian case am I suggesting that the intervention is authorially novel; rather, the point is that these descriptions are examples of anthropological interventions, and bril-liantly so.

17. In a chapter on “Assisted whiteness,” Roberts observes that when “IVF patients with limited material resources go about financing and gathering assistance for their reproductive projects, they become whiter. Both reproductive dysfunction [i.e., having the ailment itself] and attempts to alleviate that dysfunction are physiological and economic markers of whiteness” (ibid.: 76).

18. Landed estates allotted to Creoles, which for a long time controlled the labor of Indians; said to have been more widespread in Ecuador than in any other Andean nation, they implemented labor policies to justify hierarchical structures, including programs “for whitening the national racial stock through mestizaje (mixture)” (ibid.: 59).

19. And specifically to the kind of “language community” of readers, letter writers, and diarists that Tadmor (2001: 13) evokes, and of the authors on whom Perry (2004) draws.

20. The phrase comes from Porter’s (2000: 179) placing of Hume’s Treatise of human nature (1739-40), which was to be followed by An enquiry concerning human understanding (1748), in the context of extensive philosophical interest in the “anatomy of the mind.”

21. Porter (2000: 156) quotes Hume to the effect that it is evident that “all the sciences have a relation” to human nature. It was a short step from here to the “science of man.”

22. This comes out of a well-rehearsed theoretical position that Kelly at once makes explicit and further illuminates. Consider, for example, the contrast between Amazonian consanguinity (“constructed”) and affinity (“given”). Affinity at once points to relations based on difference, not identity, and “is the given because it is the ontological condition underlying all ‘social’ relations . . . [belonging] as such to the fabric of the universe” (Viveiros de Castro 2009: 259).

23. What appeared a single viewpoint from one angle could be divided between two, according to whether one took a view from nature or from society, and brought together again in a simple contrast, as in the later-eighteenth-century contrast between ties that were given (“family”) and ties by choice (one of the uses of the term “connection”). There were earlier contrasts too. Perry’s (2004: 230-31) account of the privatization of marriage, as she calls it, notes a complicated triangular relationship between the law and the different marriage strategies of the English landed classes and wealthier middle classes. (In her terms, following Randolph Trumbach, the latter are between “marriage-of-incorporation,” which conserved the “consanguineal family” by incorporating affines, and “marriage by alliance,” which cemented relations between families while making upward mobility possible and putting the interests of the new conjugal unit above those of their parents.)

24. Corsin Jimenez’s concern is with how we grant epistemic status to objects, and, as he says at the outset (2013: 2), thus with exploring the conditions of description as we understand it, when what “appears to be a description of an object is, on closer inspection, turned inside out into a description of epistemic awareness.” He offers “conceptual interventions” of his own, as “an effort at ‘trapping’ the descriptive forms of late liberalism within their own culture of description, and an attempt to place description at perpendicular angles vis-a-vis emerging forms of global public knowledge” (ibid.: 28).

25. The possibility was always there in early usages of the term for (positive) alliance or companionship.

26. As I have noted elsewhere (most recently Strathern 2014b).

27. As Hume and Smith used “connection” in referring (for example) to “principles of connection”—raising of course the further question (doubt) as to what counted as “principles.”