Viveiros de Castro and Goldman: Introduction to Post-Social Anthropology

This work is licensed under the Creative Commons | © Eduardo Viveiros de Castro & Marcio Goldman. Attribution-NonCommercial-NoDerivs 3.0 Unported. ISSN 2049-1115 (Online)

Introduction to Post-Social Anthropology

Networks, multiplicities, and symmetrizations

Eduardo Viveiros de Castro, Museu Nacional, Universidade Federal do Rio de Janeiro
Marcio Goldman, Museu Nacional, Universidade Federal do Rio de Janeiro

Translated by Ashley Lebner


1st session. Openings
Eduardo Viveiros de Castro

This course was invented by both of us as a way to resolve a technical difficulty that arose after I declared my intention to lead, this semester [March-June 2006], the traditional course of the Post-Graduate Program in Social Anthropology “ATI”: Anthropological Theory 1, the introductory course that begins with the origins of anthropology, in the depths of the nineteenth century, and ends with Levi-Straussian structuralism, or a bit before.

We decided, faced with this difficulty, to offer a kind of “Anthropological Theory 2+,” which would discuss what happens with anthropology after the course of Anthropological Theory 2, which follows AT1; that is, the one that goes from structuralism until “today”—although today is always, in the best case scenario, yesterday. Then we thought that maybe it would be more appropriate to classify this course as “AT —1,” Anthropological Theory minus one, seeing that our proposal, here, is to approach anthropology not from beginning to end, according to the progressivist and teleological conception of intellectual history, but by means of another analytical trajectory, which is not historical, even though, or perhaps because, we are concerned with contemporaneity (hence the initial idea of an “AT 2+”). The “minus one” would therefore signify the subtraction from the notion of history as totalizing reason the idea of anthropology as a conceptual practice of multiplicity. 1

I will start, then, by thanking Marcio for having supported me in this course-project, for sharing the task of conceiving its structure and content. Not only will the classes be our shared responsibility, but furthermore Marcio will probably respond to a more substantial quantity of interventions, as I will be traveling a bit during the semester.

The course is connected to a collective project, which has been in existence for a while, called the Abaete Network of Symmetrical Anthropology. Note that it is about a “network,” not a “group,” which connects people who live in different places and who work in different institutions through a website. It is a network whose aim is to help reinvent anthropology—in another era one would have said its aim was “to rethink anthropology,” and no-one would have thought this pretentious, because it would have been said by someone at the top (Leach 1961). Us, on the other hand, who do we think we are to rethink anything, right? We intend to reinvent the discipline, as we said, through the triple problematic generated by a symmetrical, reverse, and reflexive anthropology. These three magic words (in the good sense) will be generously discussed throughout this course. We will not pause on them now; we will only call attention to the fact that our course is one of the first theoretical productions of Abaete, this ex-centric network. One of the Network’s most important nuclei is here, in the Post-Graduate Program in Social Anthropology / PPGAS of the National Museum (Museu Nacional / UFRJ, Federal University of Rio de Janeiro) where we also have the headquarters of the Nucleus of Symmetrical Anthropology / NAnSi (an academic-institutional contingent of the Network).2 Yet the Network is diffuse; it moves through various places that are not located anywhere; in certain moments, it seems more like a state of mind, or spirit. Frequently, like spirits, it disappears. The three words—symmetry, reversibility, reflexivity—connect the three anthropologists whose work forms the axis of our bibliography. The notion of symmetrical anthropology leads us to Bruno Latour; reverse anthropology brings us to Roy Wagner; and that of reflexive anthropology, or auto-anthropology, guides us to Marilyn Strathern. French, American, and British: it seems just like the usual courses of Anthropological Theory, where those three dominant national traditions are expounded, which constitute for us “peripheral anthropologists” (it wasn’t me who invented this classification) a mirror, a model, and an ideal. Only that here we intend to pass onto the other side of the mirror, to mess up the models and have our own ideas on this ideal—decentering the periphery. Anthropological Theory —1.

The aims of the Abaete Network are contained in a manifesto, the “Abaete Manifesto,” which is posted on a wiki.3 Marcio will soon speak to us a bit about the Manifesto.

The site also details the history of the project. To summarize, we would say that it emerged as a semi-joke. First, we invented the name, and then we started building things around it. “Abaete” is a composite pun uniting the acronym ABA, of our Brazilian Association of Anthropology, and the suffix -ete (or -ete), which is of Tupi-Guarani origin and signifies true, legitimate, or authentic. As you know, this suffix is common in the Brazilian language, figuring abundantly in our geographical, zoological, and botanical onomastics: “Tiete” (the main river running through Sao Paulo), for example, is “water of truth,” that is, a large or principal watercourse; “iauarete” is a jaguar “proper,” that is, a spotted jaguar (as opposed to the black jaguar and of course the puma—Yawar or jaguara is the Tupi word for jaguar). Our joke-project was, then, that of creating a “real” alternative to ABA, or, rather, an alternative to anthropology as understood in practice by ABA, seeing that we were somewhat dissatisfied with the so-called directions that the discipline had been taking. I say “we were” and “had been” to place this all in an indefinite, remote, perhaps mythical past. We do not need more enemies than we already have.

But /Aba/ is not only an acronym; it also transcribes an authentic (-ete!) Tupi word: “Aba (awa, ava) means “human being.”4 Abaete indicates, therefore, the real human being, legitimate people, veridical, that is, us. “Us,” that is, “them,” the Tupi. I am not speaking about us, who are here, but about an us identified with “real humanity.” Abaete is who says “Abaete.”

That is the meaning of the word abaete consigned to the dictionaries of Tupi spoken in sixteenth-century Rio de Janeiro and Sao Paulo. But in the Tupi spoken in Bahia, which was practically identical, for some reason (actually, I think it is not known whether it is a homonym or if in fact it is the same word), meant just the opposite of human being: it meant ghost, apparition, monster. From whence the toponym “Lagoon of the Abaeté,” in the sense of enchanted lagoon, lagoon of the specter, that famous lagoon of dark waters sung by Caymmi. And thus Abaeté seemed an interesting word, as it had various connotations. It registered our dissatisfaction with the institutional representation of the métier, which seemed to us to have excluded itself from the political and theoretical vanguard (where it had been for some time) and stopped defending the interests to which anthropologists should be committed. But the name “Abaeté” pointed, above all, to the idea of a “real” anthropology, an anthropology-eté, where the idea of anthropos, of ava, is what is always in question: person or ghost? Model or monster? “Us” or “them”? It is this kind of thing that we are going to talk about in this course, not about the professional order of anthropology. The joke can be resumed thus, then: the Aba-eté Network is where the “real” Brazilian association for the theory and practice of anthropology will be; “association” in the Latourian sense of the word, which we will discuss in a few classes hence. The rest is not a joke.

But let us move from Tupi suffixes to Latin prefixes. The course that we are beginning here is called “Introduction to a Post-Social Anthropology.” It should be offered, then, within an imaginary Post-Graduate Program in Post-Social Anthropology / PPGAPS, and not in our (in the absolutely nonpossessive sense of the pronoun) Post-Graduate Program in Social Anthropology / PPGAS. This is our way of saying that we, the professors of this class, are demanding the right to transmit the theoretical specter of anthropology in another frequency, a little outside—to the left, we would like to think—of the official channels of the PPGAS. Think of it, then, as if you were tuned into PPGAPS, the Post-Graduate Program in Post-Social Anthropology.

This course intends to introduce, among other things, a certain habit of hesitation in face of the discussion of anthropological problems, and of slowness in the suggested responses. We the professors begin from the principle that we still do not know what anthropology is (understood as: what it can be). The students, we suppose, don’t know either: if they do know, they are here to unlearn. Because the objective of the course is to discover problems that current anthropology wants to forget. I am glossing here one of Deleuze’s formulations on the notion of the “left”: “the job of the Left, whether in or out of power, is to uncover the sort of problem that the Right wants at all costs to hide” (Deleuze 1997: 127). We say, therefore, that our intention is to reveal the theoretical problems that contemporary anthropology keeps sweeping under the rug so as not to tire itself thinking; so as not to need to rethink itself by thinking about them. What we intend in this course is not to offer new solutions to old problems, but to accompany the formulation of new problems and, if possible and when necessary, to reproblematize them; who knows, we may even invent entirely new problems.

Another way to define this course then, now negatively, is by asking the question: what is it no longer possible to say in anthropology? There are ways of speaking, ways of writing, ways of thinking, ways of researching, ways of doing fieldwork that are no longer possible, which does not impede them in continuing to exist officially in various places. It is very instructive to begin considering the kind of thing, the kinds of things that it is no longer possible to say in anthropology. In what sense are these things no longer possible—even while, as I said, they continue to exist? This is an issue that will concern us, directly or as a background, during the whole course.

The course will dedicate a certain amount of attention to the socio-anthropology of the sciences, that subspecialty that we see flourishing in various ecological niches of academia. The project of an anthropological consideration of “Science” seems to us to have profound consequences for all of anthropology. It is interesting not merely for those anthropologists conducting research in laboratories and similar contexts; I am speaking also about those who study indigenous villages, rural communities, popular neighborhoods, colonial archives, federal administrations, whatever. And for a simple and strong reason: the modern Western opposition between science and not-science—opinion, superstition, religion, ideology, culture, politics—is an internal repetition of the meta-opposition that separates Western modernity from the “others,” the premoderns, the barbarians, the savages. The duplication of the constitutive operation of Western modernity, its historical dispositif of subjectivation: the identity of the “West” depends on this duplication. Because of this, any attempt to examine the great division between science/nonscience, immediately threatens, in the best sense of the word, the other great division: that between moderns and nonmoderns. Or even the division between humans and nonhumans. What the anthropology of science does is not dissolve the distinction between science and nonscience; rather it multiplies and differentiates this distinction in a cloud of practices with specific demands and obligations. Note that we are dealing with an anthropology of the sciences, in the plural, which immediately puts into question the idea of Science in the singular. So everything happens as if there were a “fractal” reproduction of the great division: it reappears at various levels and scales. It is internal to the West; at the same time, it identifies all of the West with science, in opposition to the outside, the world of superstition, of fetishism, of fundamentalism, of traditionalism; at the same time, it separates humans and nonhumans. “Anthropological” anthropology, such as we know it, consists precisely in determining the criteria that distinguish “man” from the rest of “nature,” that is, nonhuman life; these criteria are very similar to those—when they are not exactly the same—that characterize modern Occidentals in face of the rest of the continent of Culture, that is, the nonoccidental cultures.

But we all know that, don’t we? What we don’t know is where we are. That is already a good place to begin (again). We also don’t know who we are. Because everything divides itself in itself and multiplies itself through the other—Brazil included—or, as some would say, above all. Us and them? Who?

The course intends, as we already said, to transmit in another frequency; not because it is free of traffic, but, on the contrary, by being sufficiently congested by philosophical noises, it will complicate the transmission of words from the old anthropological order. Once we are exploring this frequency, the texts that we will work with are, in principle, “difficult”; they employ a new language, a vocabulary that is sometimes bizarre. We will have to learn to: pronounce and spell out this terminology carefully: fractality, multiplicity, dividual, rhizome, virtual, actor-network, regime of signs, becoming, symmetry, cosmopolitics... we will also come across transparently common words for us like “culture,” only used in a novel, transparadigmatic way, by certain focal authors in the course, like Roy Wagner. We will explore, with Bruno Latour, flat ontologies, ontologies of variable geometry, ontologies that defy our hierarchies and customary classifications and that force us to abandon various conceptual automatisms. We will pause, linger on dense analytical texts, written with a powerfully complicated syntax, for example, the texts of Marilyn Strathern—which, as we will maintain, is not a reason not to read them, very much on the contrary. The idea that texts are difficult is not something that should frighten us. As Isabelle Stengers reminds us, it is very often necessary to use difficult words so that they resist capture, so that they cannot be pronounced freely, with impunity, by the bosses, by the powers that be, by the political, mediatic, or academic cardinalate. And yet this is never guaranteed once and for all.

On the presence of philosophical texts in the bibliography of the course: our idea is to truly use these texts (“for real”), and not as intellectual piercings, extravagant adornments to mark our radicality. But this is not a philosophy course; its professors are not professional philosophers; our reading does not pretend at all to be a philosophical reading. We believe, nevertheless, that philosophical issues should always be explicated in anthropology courses. They are always acting in any case, and it is better to start by coming clean about them. One can observe, for example, that anthropologists demonstrate an accentuated propensity toward Kantianism, whether or not they know that they suffer this affectation—like Mr. Jourdain, Molière’s character who was speaking prose his whole life without knowing it—which manifests itself as much in chronic as in acute forms. Anthropologists tend to be Kantians because many of our wet-nurses spoke in that language (and they knew that they did): Boas, Durkheim, Mauss, Lévy-Bruhl, Lévi-Strauss... Already jumping to conclusions still distant in this course, it seems to us that one of the great problems of contemporary anthropology is precisely that of finding the opening through which we can move beyond Kant—with all due respect. How can we leave behind, for example, the reflex use (in the sense of automatic, not reflected), the ample, general, and unrestricted use of a concept that Kant invented, or else, reinvented: the concept of “category”? Kant must turn over in his grave every time that the term category is pronounced in our disciplinary region, seeing as its meaning is completely banalized as: “those words that are underlined by the anthropologist in the transcription of native speech.” But at the end of the day, ironies apart, Kant’s idea that there exist a priori categories that format experience is something that, whether by means of Durkheim or by Boas, transformed itself into a kind of spontaneous epistemology of anthropologists, in the sense that Althusser (1990) meant when he spoke of the “spontaneous philosophy of scientists.” One of our challenges is how to interrupt this reflex, in order to open the possibility of a non-Kantian, or post-Kantian anthropology, which goes beyond “category” and “representation.”

The architecture of the course is based on a triangle (or pentagon) of main authors: Gilles Deleuze (more specifically the work he co-authored with Félix Guattari), Marilyn Strathern (and her dialogue with the work of Roy Wagner), and Bruno Latour. An “authentic” philosopher, which is Deleuze; a no-less authentic anthropologist, Marilyn Strathern; and, finally, Bruno Latour, a more hybrid character, trained in philosophy, but who does anthropology or sociology, having invented his own authenticity.

Another author will mediate between them, thus meriting a specific class. That author is Gabriel Tarde, the alternative ancestor to Durkheim, whose silencing made it possible to say that today we know that some things are no longer possible to say in anthropology. It is important to remember that Tarde is an author about whom both Latour and Deleuze complain—the latter was the first contemporary author to revive Tarde’s name, which until then had been buried in the deepest crypt by the Durkheimian school. In a famous paragraph in Difference and repetition, Deleuze (1994: 76-7; 313, n. 3 to ch. 2) reclaims Tarde and inserts him in a theoretical and philosophical lineage that we are demanding for ourselves; years later, in A thousand plateaus (Deleuze and Guattari 1987: 218-19) he appears in another guise, thus marking the persistence of the Deleuzian intuition that it was necessary to return to Tarde to begin thinking the social in other terms, according to another image. Latour, as is notorious and public, considers Tarde his real totemic ancestor, and, for this, considers himself—reasonably—as marginal within the social sciences of his country. In effect, the French socio-anthropological academy is genetically, genealogically, geologically really, Durkheimian; Durkheim is where you find its ultimate substrate, its mother rock.

I would like to finish my part of this opening with a quote from an American philosopher, Richard Rorty, who is considered a radical (in the “good” sense). In a text republished in the collection Objectivity, relativism and truth, Rorty made the following declaration: “We, Western liberal intellectuals, should accept the fact that we have to start from where we are and that this means that there are lots of views which we simply cannot take seriously” (1991: 29). This reads: it is time that we, the heirs of Galileo, Descartes, Newton, start assuming our unitary cultural inheritance and stop being duped by an exotic multiplicity—”lots of views”—that cannot take us anywhere at all. If we move Rorty’s phrase to the anthropological plane, it would come out perhaps somewhere close to Evans-Pritchard’s famous statement: “witches, as the Azande conceive them, cannot exist” (1937: 56). Evans-Pritchard wrote a marvelous book to resolve just this problem: seeing that witches can’t exist (as the Azande conceive them), how can the anthropologist take the Azande’s conceptions of witchcraft seriously? How can the anthropologist reconceive (reconceptualize) witches so that they can exist for us? Rorty identifies this same problem, but his solution, if you can call it that, is merely negative; in truth, his solution is equivalent to a perfect antidefinition of anthropology.

In the first place, who are “we, liberal intellectuals”? Rorty speaks here for his internal public (only “liberal intellectuals” exist, it would seem, in the United States), those who are where he is: whereas “we anthropologists” cannot stay where we are, or we cannot speak without stopping to think of those who would say: “we are not you,” or “we aren’t there (at all).” What does it mean, then, the idea that “we have to start from where we are”? Without a doubt we have to. But this says nothing about where we can, should, or want to end up. In truth, it doesn’t say either where exactly we are. We know that we can only start from where we are: nobody speaks while floating on a magic carpet. Even the magic carpet is a little portable territory that we bring with us and that brings us with it. In our sidereal journeys, it does not matter where we fly, we need a vehicle.

Defining what we can’t take seriously as “lots of views” already predefines what it is that cannot be taken seriously. It is enough to call something “views” so that the not-taking-seriously is already justified. “We know” also that “lots of views” is the name of a Pandora’s box that can only contain falsities, deliriums, hallucinations—in sum, diabolical lies. The devil is the Father of lies; lies are many, the truth is one. If there are a lots of views, it follows automatically that, simply, we cannot take them seriously. Nothing more simple or more pedantic than the adverb “simply”—what can/should be read are its two semicontradictory meanings: that of simplicity (it is easy not to take lots of views seriously) and that of finality (it is imperative, urgent not to take them seriously). Finally, what to do with the supreme anti-idea contained in this affirmation, to know, that we cannot take seriously that which is the very object, the very purpose of anthropology? Because anthropology is “simply” that Western intellectual enterprise that is dedicated to taking seriously the question of how to take seriously lots, not only of views in the sense of opinions on things, but of views in the sense of things seen—not other visions of the world, but other worlds of vision, other worlds given to vision (as much as to hearing, touch, smell, taste, and to extrasensory conception). The problem, then, is how to acquire the instruments that will permit us to take seriously that which “Western liberal intellectuals simply cannot take seriously.” And reciprocally—and necessarily reciprocally—almost all the things that we cannot take seriously are not far from or outside of us, rather they are close, near, inside of us. That which we “simply” cannot not take seriously—we should see whether it isn’t the case that we shouldn’t take them.

In sum, anthropology is defined simply by its inability to accept the “liberal” prohibition of Rorty. Its central problematic is how to take seriously what we are prohibited from doing. Moreover, its problem is how to build a concept of taking seriously that is not confused with belief, with allegory, or with Echolalia.

That is the aim of this course.

Marcio Goldman

This course is about an anthropology that, to use one of Guattan’s expressions, is “in the very act of its constitution, definition and deterritorialization” (Guattari 2000: 44). The Abaeté Network has permitted us to test it by means of certain textual experiments, which have increased the possibilities of dialogue between different researchers. Our objective now is to extend the discussions that we have had in the Network a bit more. Eduardo has already spoken about the rationale behind the title of the course (and slightly less about the subtitle) and I would like to add a few more thoughts.

The Brazilian author Guimaráes Rosa (1968) famously wrote that “everything is faked at first; the authentic germinates afterwards.” This phrase exactly reflects the fact that the expression “post-social” was at first a joke, a play on words, but suddenly it began to make sense and we discovered that it was really possible to take it seriously. The expression Post-Social Anthropology has to be taken seriously, for a series of reasons. We could also call it post-cultural anthropology, seeing that what is in play are in effect the notions of society and culture, which oriented the anthropological project for a long time. In the last twenty or thirty years, critiques of these two keywords have proliferated from within and outside of the anthropological field. A good number of these critiques occurred, nevertheless, to benefit what we can call their obligatory accompaniments. That is, the notion of society was critiqued to save the individual; culture was critiqued to save the idea of human nature. Evidently, these are not our concerns. What interests us are slightly more radical critiques, which put into question a whole conceptual field, in which not only the terms society and culture are involved, but also nature, individual, psychology, human, body, etc. The post-social anthropology that we are defending is not, obviously, sociobiology, nor is it an anthropology focused on the idea of individual interest or manipulation. A text that we will approach indirectly, but which is worth reading with attention, is the debate that Tim Ingold (1996) organized to discuss the notion that society was theoretically obsolete. This happened in Manchester, a temple of social anthropology, where the notion of society was cultivated for generations. Paradoxically, those who won the debate were those who were supporting the motion “the concept of society is theoretically obsolete.” One of the participants of the debate was Marilyn Strathern, who had just published The Gender of the Gift (Strathern 1988), and who had already proposed, as we will see, working with the alternative concept of sociality.

Let’s go now to the subtitle “networks, multiplicities, and symmetrizations,” terms to which we can add some others, like reversibility and reflexivity. These five terms designate the conceptual and intellectual horizon on which we will move. Apparently common, these terms might give the impression that there is nothing new here. The notion of network, for example, has been abundantly used in the history of the social sciences, but, in the majority of instances, it appears in the sense of social network, that is, it carries precisely the adjective that is being problematized in this course. The famous social network of British anthropology is nothing more than the sum of individual relations, an outcome, then, of the Radcliffe-Brownian notion of social structure. Individuals appear there as supposed units that enter into relation. We are not interested, in this course, in these social networks, or in the networks of positions or institutions, which are the diverse sociocultural variations of that notion of network. The notion of network that interests us comes from elsewhere, from studies of science and technology, which culminate in “actor-network theory,” of which Bruno Latour is an important exponent. Note that the authors that we bring together here are not always in agreement and, in this sense, we don’t have to agree completely with them either. One of the texts that we will analyze, for example, is “Cutting the network” (Strathern 1996), in which Marilyn Strathern problematizes the notion of network proposed by Latour. All of these problematizations confer on this course the character of multiplicity. The notions that we present here emerge, in general, from a dissatisfaction with certain theoretical and methodological solutions that end up generating other problems and these are what interest us directly. Despite the critiques directed at Latour’s notion of network, its principal merit is the fact that it doesn’t know the difference between social and not-social, between sociotechnical and biological. All of these notions, it is important to emphasize, trace territories of existence that are not merely social.

There is a bad habit in anthropology of confusing the notion of multiplicity with relativism. Roy Wagner (1981: ch. 1) makes an interesting distinction between relativity and relativism, which should help us. Cultural relativism is certainly an anthropological version of one of the oldest ontologies, or images of thought, that the West knows: that which is based on the opposition between the one and the multiple. On the one hand, the One, the true viewpoint; on the other, “lots of views” to take up Rorty’s statement again. In this sense, nature, the real, would be the only thing that is not a vision. What Rorty calls visions a social anthropologist would call representations. We have theories, they have representations; we have ideas, they have ideologies; we have ontologies or sciences, they have views. Let us return to these base asymmetries. If the idea of network appears powerfully in Latour, as much in We have never been modern (1993) as in Reassembling the social (2005), the idea of multiplicity appears above all in Deleuze and Guattari, mainly in A thousand plateaus (1987). On the second plateau, called “One or various worlds?” these authors present a theory of multiplicities. On plateau 9, “Micropolitics and segmentarity,” they pay homage to Gabriel Tarde, one of the first to formulate this theory. We are going to discuss these two plateaus here. Mille plateaux, published in 1980, was a reflection that was responsible for the revitalization of Tarde’s thought, who had lost the battle, but not the war, with Durkheim. Deleuze and Guattari insist that multiplicity is not the multiple, in the sense that it is not opposed to the one. Multiplicity, then, is of the order of intensivity and not of extensivity; it is not reducible to the idea of plurality because every singular point is already multiple in itself.

Wagner, in problematizing relativism, alleges something similar for relativity. Relativity is not the affirmation of “lots of views.” It is not difficult, then, to consider Richard Rorty a relativist. Relativism is perhaps what is most inoffensive in the world today; it affirms the co-presence of different views as long as one maintains the fixed point around which all the others gravitate. Culturalism, relativism, and all their variants depend on a certain image of thought, which is very old, and which is compressed in the opposition between the One and the Multiple. Culturalism and relativism can, then, live together very well with these great divisions, which separate the “real” from those “lots of views.” Latour’s notion of network has a sociotechnical and territorial existential bias, while the Deleuzo-Guattarian notion of multiplicity should be understood in light of the idea of intensivity and cannot be dissociated from the other keyword of Mille plateaux, the concept of rhizome. The correlates of multiplicity are singularities. Multiplicity does not bend to the yoke of the One and the Multiple.

Among the notions that we are dealing with, symmetry is perhaps the one that circulates the most in contemporary anthropology. But it has not avoided misunderstandings. We use the word symmetrization intentionally, but the notion of symmetry is often confused, with good or bad intentions, with that of semblance, unity, equality, and even justice, as if to do symmetrical anthropology is to do justice. The idea of symmetry, which we extract (we do not copy) from Latour, always seemed more linked to the kind of effort at symmetrization, whose intention is not to substitute the great divisions for a transcendent unity, but to recognize the little differences that the great divisions overcodify...

The authors we are bringing together here are connected, then, in a very interesting way, even though there might be, as we have already emphasized, a series of differences between them. Latour read Deleuze and Guattari (who were inspired by Tarde) and extracted the notion of rhizome in order to construct, with other researchers, actor-network-theory. Note that neither Latour, nor Deleuze nor Guattari, are professional anthropologists, although Latour has called himself an anthropologist, has coined the expression “symmetrical anthropology,” and maintains a robust dialogue with contemporary anthropology. Roy Wagner and Marilyn Strathern, two Melanesianists who are still interested in the problematization of an anthropology of “ourselves,” are authors who allow us to replace these discussions within the field of anthropology in which we work—or want to work. Strathern and Wagner have numerous theoretical affinities, which result in very important concepts such those of sociality, fractal person, dividuality, among others. Strathern, as already emphasized, discussed Latour’s theory, sometimes putting herself in a critical mode; Latour, for his part, if I am not mistaken, does not refer to her. The fact is that it is very important if the connection between the authors actually occurs or if we are the ones who propose the link—we bet that something interesting in theoretical and methodological terms can come from that.

To prevent misunderstandings around the terms network, multiplicity, and symmetrization, the only solution is to keep them together, that is, they only make sense when enacted together. What connects these various notions is what I would call a properly anthropological or ethnographic modality, of reflexivity and of reversibility, which we can associate with Wagner and Strathern. If a certain idea of reflexivity was talked about a lot in anthropology, the same cannot be said about reversibility, which was discussed only by Wagner in The invention of culture (1981). In principle, every anthropological undertaking is reflexive, but the problem is to specify whose reflexivity it is and how it is managed. We extract our idea of reflexivity above all from an article entitled “The limits of auto-anthropology” (Strathern 1987b), which we will discuss in this course. Overall, it is not about taking the notion of reflexivity as a synonym of recursivity, in the sense of doing an “anthropology of anthropology.” That would reflect a bad tendency pointed out by Foucault some forty years ago with regards to the human sciences. And even Foucault seems to have a certain difficulty escaping this trap. In that sense, the criticism of genealogy by Deleuze and Guattari in the first of the thousand plateaus could be emblematic of this course. Deleuze always had issues with the Nietzschean and Foucauldian idea of genealogy and wrote, with Guattari, a plateau called “Geology of morals,” which will be discussed by us here. The difference between geology and genealogy is absolutely crucial. The reflexivity maintained by us is more of a geological order than it is of a genealogical or metascientific order, and should not be confused with what I just referred to as recursivity.

But the notion of reflexivity also doesn’t designate a mode of objectivation, as it appears, for example, in the “Introduction to the work of Marcel Mauss” (Lévi-Strauss 1987). There the subject is extricated from itself in smaller and smaller portions, that is, it objectifies itself by means of the study of other societies. The author who took this the farthest is perhaps Bourdieu, who explicitly forged a reflexive sociology. For Bourdieu, however, the sociologist is reflexive in as much as he is more cunning than everyone else—those people who do things without knowing what they are doing, while he always knows what he is doing. It isn’t by chance that his last book was called Sketch for a self-analysis (Bourdieu 2008) and there Bourdieu tries to show that he is even more cunning than himself!

Reflexivity as a mode of objectivation comes to compete, by the middle of the 1980s, with another form of reflexivity, this time as a mode of subjectivation. We are now in postmodernism and we have to remember a joke told, I think, by Sahlins: after one year living with an anthropologist who is doing fieldwork, the natives asked him whether it was time yet for them to start talking a bit about themselves... In this sense, reflexivity ended up being reduced to the idea that the anthropologist needed to include himself both in his investigation and in his narrative.

The notion of reflexivity, which appears throughout the history of anthropology, brings us back to Rorty’s phrase. If there is anything even slightly true about it—although it is very trivial—it is the fact that we inevitably start from a situation in which there are only two terms or two sides. What we do with this situation is our problem. The most conservative solution in the history of anthropology, and that remains dominant, was brilliantly problematized by Pierre Clastres (1979) in a short and luminous text, published in 1968 and entitled “Between silence and dialogue.” Clastres alleges that, in the last instance, anthropology has to choose between being a “discourse on” and a “dialogue with.” The notion of dialogue will reappear, two decades later, in postmodern North American anthropology with a slightly different meaning from that suggested by clastres. After all, in the West, the dialogue hearkens back to a Platonic tradition that uses the term in a falsified way. Because in the dialogue there are never only two, there is always a third, hidden, interlocutor who is in command. When we read a Platonic dialogue, we know from the beginning where we are going to end up, because the technique of the ascendant dialectic conducts us forcefully to the truth, which is either different than either of the other points of view, or else it is confused with one of them which, nevertheless, duplicates itself. That is, it stops being part of the dialogue to become a judge, endowed with the power to decide who is right. The idea of reflexivity that appears in the work of authors like Wagner and Strathern demands that we definitely take seriously the notion of dialogue, that is, that we consider only the two parts without supposing that there is a third, who is external to the game. “We have to start from where we are,” writes Rorty. Apparently there are no problems with this assertion. We are in a dialogue between “us”—the “liberal Western intellectuals”—and the “others” whoever they may be. The point is that this us will soon enough be duplicated, as in a Platonic dialogue. But in the end, who are “we”? What is always in play, as Eduardo reminded us, is the opposition between science and nonscience. We are always divided, a part of us is similar to “them”—remaining on the side of nonscience—and the other is conscious of everything. If we refuse this kind of move, if we refuse this third position, which is a kind of transcendence, what is left? Everything, Deleuze and Guattari might say.

In “Out of context,” Strathern (1987a) writes that we face a “technical” problem: “how to create an awareness of different social worlds when all at one’s disposal is terms which belong to one’s own.” This is not a criticism, it is merely a statement: this is how anthropology works. We are obliged to produce a discourse about other worlds and, in principle, we have only the resources of our world at our disposal. This is our point of departure, but the solution to this dilemma should be, evidently, the inverse of Rorty’s proposal. This is our proposal: we start from where we are and we have to take seriously “lots of views.” The only way to escape where we are is, really, taking seriously these “views.” The reflexivity that we are defending is, nevertheless, counterintuitive, that is, we have to move against the grain we normally move with. Wagner proposes a new notion of culture that follows this counterintuitive challenge. And it is this notion of culture that can be thrown, by Strathern, against the notion of society, which dominated British anthropology. It is in this sense that Strathern and Wagner furnish the link for our connections, for our network of authors and concepts. The last chapter of The invention of culture is called “the invention of anthropology,” because in reinventing culture we reinvent anthropology. But we are reinventing it radically, proposing a new “image of anthropology,” in the sense that Deleuze and Guattari (1994: passim) speak of the “image of thought,” a new comprehension of what it means to think.

Of course in the history of the discipline there exists a series of images of anthropology which are, at their edges, very heterogeneous and, sometimes, not intercommunicable. The first of these images perhaps is contained in the last phrase of Primitive culture, by Tylor (1891): “anthropology is the science of the reformer”—in the sense that it helps extirpate those last vestiges of superstition, ideology, and fantasy in modern society. It is interesting, for example, to contrast this image with the anthropological cognitivism of authors like Sperber. The image of anthropology isn’t of the science of the reformer in this case; it is rather the image of a pure science, an image that makes the anthropologist eliminate all the ethnography in order to get at human nature. Certainly, in this course we are in some way driven by an image of anthropology; there is no way to be neutral about it. But we do not want this image to be either that of the scientists, or of the technocrat.

Based on Latour’s anthropology of science, Stengers defines the principle of symmetry as the injunction that obliges us to draw “conclusions from the fact that no general methodological norm can justify the difference between winners and losers, which creates the closure of a controversy” (Stengers 2000: 7). The controversy between Durkheim and Tarde is, in this respect, emblematic. When we read Durkheim, Tarde appears as the Beast to be confronted. Durkheim was very successful, in the sense that he was more successful in sociology, while Tarde was almost wiped off the map. Usually, we are content with the histories told by the winners about authors like Tarde or Levy-Bruhl. But when we stop to read these authors we are surprised to discover that they really weren’t the idiots we were made to think they were; they thought many relevant and interesting things. This surprise is an experience of thought that we cannot foreclose. Reassembling the social, by Latour, comes from that experience, as Latour begins to meditate on what would have happened with sociology if Tarde had won the quarrel, if sociology were not Durkheimian by definition, if it had not always operated through the idea of collective representations. Reassembling the social is a kind of trial or attempt to develop the social sciences precisely in that direction. Which reminds us that something we should always avoid are the organicist and evolutionist conceptions of the history of ideas; that it is necessary to get away from the conception that ideas are born, grow, reach their apogee, and wither away and die, being buried for the rest of forever. The notion of multiplicity, proposed by Deleuze and Guattari in Mille plateaux, suggests that ideas are always there. The problem is what to do with them. Stengers, for example, emphasized the nonhomogeneity of the scientific field, the fact that it does not always follow the same rules and operations, or that there is never a general guarantee of scientificity at any point on that field. For Stengers it is necessary to think science in its heterogeneity, because theoretical-experimental science (which exerts a certain repressive power on the rest) is not the same thing that Prigogine does, nor is it the same thing, for his part, that Darwin does.

In closing, I would like to recall two things. The first is that it is exactly twenty years ago that Writing culture was published by George Marcus and James Clifford (1986) and was taken as the announcement of a crisis in representation in anthropology, which was interpreted by many as a kind of death sentence for the discipline. If it is not possible to represent others, to have a science of others, we have nothing left to do. Writing culture is a strange book. On the one hand, it seems very radical, but, on the other, it remains very traditional, as it starts from totally banal images of both science and anthropology. Some twenty years before Writing culture, Foucault published The order of things (Foucault 1970), whose subtitle is an archaeology of the human sciences (there are those who say that he almost called it “an archaeology of structuralism”). The book does not exactly close with the “human sciences,” which is the title of the last chapter, but something rather mysterious that Foucault calls the “counter-human sciences,” a rubric beneath which he unites ethnology (or what we would call anthropology), psychoanalysis, and linguistics. It is evident that he is referring to structuralism, to that kind of resurrection of Kant in Western thought. If the human sciences operate under the sign of representation, the countersciences make a certain step back in order to ask about the very conditions of possibility of representation. Lévi-Strauss and Lacan are certainly exponents of this movement for which representation isn’t of interest, while how it is possible to have representation is. As usually happens with Foucault’s texts, many read this chapter as if it were a kind of death sentence for the human sciences, as if they were prisoners of an “episteme” that is at its end, full stop. Twenty years later, without even mentioning The order of things, history repeats itself in Writing culture: anthropology has ended; all that is left is auto-anthropology, in the bad sense of the term.

But something should be said in favor of Writing culture when Marcus and Clifford emphasize that some ethnographies written in the 1970s and 1980s were able to escape the paradigm of representation. This is the challenge for anthropology: how to do ethnography if we do not have representation? Strathern’s text, “Out of context,” which we will also discuss, accepts Marcus and Clifford’s question, takes it really seriously, and gives us two possibilities: we close the doors or we reinvent anthropology. Strathern, of course, chooses the second option.

Let us return now to Evans-Pritchard’s affirmation that Eduardo quoted earlier: “witches, as the Azande conceive them, cannot exist” (1937: 56). It is strange that, thirteen years after having published the book on the Azande, Evans-Pritchard (who in the interim had converted to Catholicism) ends Nuer religion by commenting on our limitations when describing Nuer experiences with supernatural powers: “What this experience is the anthropologist cannot for certain say. . . . At this point the theologian takes over from the anthropologist” (Evans-Pritchard 1956: 322). In thirteen years, Evans-Pritchard moves from science to theology via anthropology. First, science, the only thing that is capable of saying something about the ultimate truths of reality; later, science continues to say things about the real, but what it says is not everything that there is to say, it is necessary to add theology. I think that Evans-Pritchard outlines the dilemma that we always encounter, and brings us back to the issue raised by Clastres: between simply producing a discourse on others and miming the discourse of others (we pretend that we are simply repeating it), what space remains? How is it possible to produce a discourse linked to that of others, but which is neither simply a metadiscourse in relation to them, nor simply a replication of what they are saying? For Evans-Pritchard, only the theologian seems capable of recuperating what the Nuer have to say about religious experience. The central question of this postsocial anthropology, which we are trying to propose by means of a series of new concepts, consists in our bet that it is possible to invent something new through the shock produced by the contact of positions: “from where we are” and these other “views.” Our triumph vis-à-vis assertions that are similar to Rorty’s resides in the weight of the ethnographic knowledge in this process. Indeed, the whole history of anthropology, which we have outlined here in a rather abstract way, can be retranslated, rephrased, and retold starting with precise ethnographic experiences.

This course will begin, then, with two negative observations: why aren’t Africanist models applicable to New Guinea? And why are they not applicable in South America? The first statement was made, somewhat unwillingly, by John Barnes, in the 1960s; the second, by Joanna Overing, in the 1970s. But these two negative inquiries hide something positive, to wit, the resistance that ethnography can present to theoretical models in vogue. It is in this sense, I think, that Strathern affirms that there still exist realities that are sufficiently exterior to us to support us in overcoming our own thought. This “other,” which should be extirpated, reformed, or explained, is converted into a necessary point of support so that our thought can work and, as a result, be modified in the process.

We can, therefore, transmute Evans-Pritchard’s phrase and apply it to anthropology itself: social anthropology, as anthropologists conceive it, does not exist. But this isn’t enough. If social or cultural anthropology has now arrived at this dead-end, without an apparent exit, the alternative, our own as well as that of the authors we will be dealing with, is anthropology’s invention or reinvention. Now it is more than time to reconsider this whole process of self-sabotage that anthropology has lived in the last twenty years, without intending to recuperate a tradition that it is opportune to now leave behind.


Althusser, Louis. 1990. Philosophy and the spontaneous philosophy of the scientists and other essays. Translated by Gregory Elliott. London: Verso.

Bourdieu, Pierre. 2008. Sketch for a self-analysis. Translated by Richard Nice. Chicago: University of Chicago Press.

Clastres, Pierre. 1979. “Entre silence et dialogue.” In Claude Lévi-Strauss, edited by Raymond Bellour and Catherine Clément, 33–8. Paris: Gallimard.

Clifford, James and George E. Marcus, eds. 1986. Writing culture. The poetics and politics of ethnography. Berkeley: University of California Press.

Deleuze, Gilles. 1994. Difference and repetition. Translated by Paul Patton. London: The Athlone Press.

———. 1997. Negotiations 1972–1990. Translated by Martin Joughin. New York: Columbia University Press.

Deleuze, Gilles and Félix Guattari. 1987. A thousand plateaus. Translated by Brian Massumi. Minneapolis: University of Minnesota Press.

———. 1994. What is philosophy. Translated by Hugh Tomlinson and Graham Burchell. New York: Columbia University Press.

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

———. 1956. Nuer religion. Oxford: Oxford University Press.

Foucault, Michel. 1970. The order of things. An archaeology of the human sciences. Translated by Pantheon Books. New York: Random House.

Guattari, Félix. 2000. The three ecologies. Translated by Ian Pindar and Paul Sutton. London: The Athlone Press.

Ingold, Tim. 1996. Key debates in anthropology. London: Routledge.

Latour, Bruno. 1993. We have never been modern. Translated by C. Porter. London: Prentice Hall.

———. 2005. Reassembling the social. Oxford: Oxford University Press.

Leach, Edmund. 1961. Rethinking anthropology. London: The Athlone Press.

Lévi-Strauss, Claude 1987. Introduction to the work of Marcel Mauss. Translated by Felicity Baker. London: Routledge & Kegan Paul.

Rorty, Richard. 1991. Objectivity, relativism and truth. Cambridge: Cambridge University Press.

Rosa, Joào Guimaraes. 1968. Tutaméia; terceiras estórias. Rio de Janeiro: J. Olympio.

Stengers, Isabelle. 2000. The invention of modern science. Translated by Daniel W. Smith. Minneapolis: University of Minnesota Press.

Strathern, Marilyn. 1987a. “Out of context: The persuasive fictions of anthropology.” Current Anthropology 28 (3): 251–81.

———. 1987b. “The limits of auto-anthropology.” In Anthropology at home, edited by Anothony Jackson, 59–67. London: Tavistock Publications.

———. 1988. The gender of the gift. Cambridge: Cambridge University Press.

———. 1996. “Cutting the network.” Journal of the Royal Anthropological Institute 2 (3): 517–35.

Tylor, Edward Burnett. 1891. Primitive culture: Researches into the development of mythology, philosophy, religion, language, art and custom. London: Murray.

Wagner, Roy. 1981. The invention of culture. Chicago: University of Chicago Press.

Eduardo Viveiros de Castro is Professor of social anthropology at the National Museum, Universidade Federal do Rio de Janeiro. He has been Simon Bolivar Chair of Latin American Studies at Cambridge University (1999-98) and Directeur de recherches at the C.N.R.S. (2000-2001). His publications include From the enemy’s point of view (1992), Métaphysiques cannibales (2009), and The inconstancy of the Indian soul (2011).

Marcio Goldman is an Associate Professor of the Post-Graduate Program in Social Anthropology at the National Museum, Universidade Federal do Rio de Janeiro. He is also a researcher of the Brazilian National Research Council (CNPq) and Fellow of the Rio de Janeiro Research Foundation. He is the author of Reason and difference: Affectivity, rationality and relativism in the thought of Lévy-Bruhl (UFRJ/Grypho, 1994), Some anthropology (Relume Dumarä, 1999), How democracy works: An ethnographic theory of democracy (7 Letras, 2006, to be published in English this year by Sean Kingston) as well as co-editor, with Moacir Palmeira, of Anthropology, voting and political representation (Contra Capa, 1996).

Publisher’s note: We are grateful to the authors for allowing Hau to release the first English translation from Portuguese by Ashley Lebner of the introductory chapter (1a Sessào - Aberturas) of an set of lectures (by Marcio Goldman and Eduardo Viveiros de Castro) titled Introdugào a uma Antropologia Pós-Social: redes, multiplicidades e simetrizagòes given in March-June 2006 at the Federal University of Rio de Janeiro.


1. We should repeat here our note to our “Slow motions. Comments on a few texts by Marilyn Strathern (2008/2009)”:

This text was edited from the transcription of the first lecture given by us in April 2006, at the Museu National section of the Federal University of Rio de Janeiro. Naturally, the oral quality of the lectures that we decided to ‘maintain’ was achieved through various rewritings of the original transcription; and the dated character of our statements (‘April 2006’) hides a number of surreptitious insertions made in January 2009. These lectures were part of the post-graduate course ‘Introduction to a post-social anthropology: networks, multiplicities and symmetrisations’. It experimented with new directions in anthropology created by the collision of certain concepts (‘invention/convention’, ‘reversibility’, ‘reflexivity’, ‘actor-network’, ‘symmetry’, ‘partial connection’, and ‘multiplicity’, among others) proposed by authors like Roy Wagner, Bruno Latour, Gilles Deleuze, Félix Guattari and Marilyn Strathern. The course was based on the collective text ‘Symmetry, reversibility and reflexivity: great divides and small multiplicities’ (https://sites.google.com/a7abaetenet.net/nansi/Home/simetria-reversibilidade-e-reflexividade), of the Abaeté Network of Symmetrical Anthropology, which influenced a series of dissertations, theses and other works of participants (and of the network). The expression ‘post-social anthropology’ should be understood as synonymous with ‘pre-X anthropology’ where X is an unknown. We only speak of ‘post-social’ because we are imagining something still to come and we do not know what it is, nor can we. We would like to thank those who taped our classes, with or without our knowledge; to Fâbio Candotti, for the transcription of the material, and to Renato Sztutman, for the preliminary editing of the transcripts.”

2. All acronyms reflect the original Portuguese.

3. http://nansi.abaetenet.net/.

4. In certain Tupi languages, the term applies also, or especially, to humans of the masculine sex (no-one is perfect).