Agency between humanism and posthumanism

This work is licensed under the Creative Commons | © Andrew B. Kipnis. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau5.2.004


Agency between humanism and posthumanism

Latour and his opponents

Andrew B. KIPNIS, Australian National University

Two articles in the special section on knot-work in this journal (HAU 2014, volume 4, issue 3) take issue with the “posthumanism” of Bruno Latour’s Actor Network Theory (ANT). Arguing that Latour’s conception of agency undermines critical attitudes toward capitalism, they insist on an all-or-nothing, accept or reject attitude toward Latour’s work. In this article, I sketch an alternative vantage on questions of nonhuman agency and Latour’s oeuvre, which, though critical, is much less polemic. While proposing an intermediate stance for framing a theorization of agency, I conclude that it is not ANT’s theorization of agency that inhibits critical ethnographers of capitalism but rather habits in its application that derive, in part, from ANT’s insistence on painstaking ethnographic research.

Keywords: Actor Network Theory, Latour, posthumanism, agency, ethnography

Recent articles in this journal (see especially those by Keir Martin [2014], but also Chris Gregory [2014] and, perhaps, David Graeber [2014]) take issue with the “posthumanism” of Bruno Latour’s Actor Network Theory (ANT). While these articles are concerned with much more than Latour per se, they take a completely oppositional stance toward Latour’s work. This insistence necessarily misses some of the usefulness and humanity Latour’s broader oeuvre. In drawing lines in the way that they do, these critiques also steer discussion of Latour’s conception of agency in a direction that is not particularly productive, even for those interested in global inequality, and mark, from a non-European perspective, what seems to be a particularly British or perhaps European stand-off among different schools of anthropology. In the following, I hope to sketch an alternative vantage on questions of nonhuman agency and Latour’s oeuvre, which, though critical, is much less polemic.

[44]The named target of their criticism seems to drift somewhat, but centers on four poles: Bruno Latour, ANT, the ontological turn, and posthumanism. Posthumanism is a particularly vague term engaging a wide range of authors. For the purposes of this article, posthumanism will refer to analytic stances that grant agency to nonhuman entities and that downplay the differences between human and nonhuman agency. I take a stance between Latour and his critics by granting agency to nonhuman entities but emphasizing the difference between human and nonhuman agency.

The involvement of the three authors in a direct critique of Latour varies. Martin attacks Latour head on. Gregory focuses on critiquing Appadurai, Callon, and the “cultural economy” approach but states that “Latour is a theological thinker who has devoted his life to attacking humanist thought” (2014: 49), and concludes that Latour’s premises are so different from his own that no engagement, other than agreeing to disagree, is possible. Graeber never mentions Latour or posthumanism directly, but, perhaps wrongly, is dragged into the fray by Martin, who asks “what happens to politics in contemporary anthropological descriptions, particularly in the posthuman approaches that are so strongly critiqued by both Graeber and Gregory” (Martin 2014: 99); and later states, “For both Graeber and, especially, Gregory, the rise of ’posthuman’ perspectives within anthropology over the past decade is inextricably linked to the continued consolidation of neoliberal hegemony within the academy and the wider world” (Martin 2014: 102). The oppositional framing of this debate by Martin seems to me especially British or Northern European, though of course in this day in age, no academic debate (especially those that take place in English) can be confined to a single locale. I am an Americanborn and trained anthropologist, who regularly attends the American Anthropological Association (AAA) meetings, but has spent the past sixteen years of his career in Australia. It was at the 2013 AAA meetings where I got my first inkling of what seemed to me to be a European take on these issues. At a well-attended double session titled “Anthropology’s Public Engagement with Capitalism: Beyond Gifts Versus Markets” (American Anthropological Association 2013: 294), Don Kalb compared the size of the audience with that at a session on the ontological turn, announcing Marxian political economy and ontology as the two main competing trends in contemporary anthropology and insinuating that though ontology seemed to have the upper hand at certain elite British institutions, the audience at his session suggested that critics of capitalism still could attract an anthropological crowd. At the same meetings, I also attended two large sessions devoted to the “ontological turn,” but the participants in those sessions did not make me think the anthropological world divided so neatly into these two camps. At the first, in addition to noted “ontologists” Eduardo Viveros de Castro and Eduardo Kohn, there were speakers like Ghassan Hage and Elizabeth Povenelli (American Anthropological Association 2013: 118). At the second, in addition to the ontologists Bruno Latour and Philippe Descola, was Marshall Sahlins (American Anthropological Association 2013: 389). The papers and roles of these noted Australian (Hage) and American (Sahlins and Povenelli) anthropologists suggested to me that, at least outside of Europe, it was possible to be simultaneously interested in the ontological turn and critiques of global capitalism. Indeed, the respondent to Graeber’s paper [45]in the HAU special issue, Jane Cowan, begins by pointing out the parallels between Graeber’s and Hage’s approaches to anthropology (Cowan 2014: 89).

Reading Latour’s oeuvre

The criticisms of ANT’s posthumanism, by which Gregory means “a theory of value that attributes agency to things” (2014: 45) misses the broad, and I would argue, broadly humane, balance of Latour’s writing. Latour’s oeuvre begins with works on science as a series of projects of constructing truth by human actors (Latour 1987, 1988; Latour and Woolgar 1979). The extent to which this work riled some scientists is difficult to overstate (see, for example, David Berreby (1994) for an account of enemies made and Latour’s blocked appointment at Princeton). But even in his early work, Latour never argued that because scientific facts were constructed that they were necessarily false. He simply insisted that the objects science investigates do not speak for themselves. The facts that they are held to establish must be made through the efforts of the scientists themselves. He used the concept of a “black box” to designate processes that were assumed to yield “truth” regardless of the extent to which one understood how the process worked. In We have never been modern (Latour 1993), he continues his assault on the separation of the social from the scientific, famously arguing that science and politics have never been separate—how to answer questions like the extent of global warming or at what moment human life begins necessarily involves politicized debates over scientific data. In his work on Actor Network Theory (Latour 2005, 2010), he applies this critique to the social scientists. Rather than constructing black boxes with big concepts like society or class, he asks us to trace out exactly how processes unfold and the place of the nonhuman in these processes. The nonseparation of the political from the scientific has implications for those who study human societies as much as those who study electrons.

One could argue about whether Latour’s theoretical stance evolves over his career. Is the shift from examining the social construction of scientific facts to the social/natural construction of human facts a reversal or a continuity? But social scientists who only read his work on ANT and conclude that he is a neoliberal conservative whose deconstruction of “society” simply echoes Thatcher’s disbelief in the concept are missing his insistence on the social aspects of scientific work. His position is not “antisocial” but it is antidualistic and anti–black box thinking. This misreading of Latour’s work is deepened by accusations that Latour is cynical and dismissive toward those with activist inclinations (Martin 2014: 110–11). Latour certainly can be critical of various forms radical thought and action that he finds misplaced or one-sided. But he is no antiradical polemicist. He dishes out criticism to an enormous range of thinkers and actors including various versions of his own self. Though himself a European man, he constantly quips about the arrogance of Whites who dismiss the idols of colonized people (Latour 2010). Generally he is quite broad in the targets of his criticism. In a book where he discusses his own Catholicism at length, consider what he says about various forms of religious, philosophical, and radical thought:

[46]These factishes are all broken, shattered by the hammers of critical thought whose long history takes us back: to the Greeks who abandoned the idols of the Cave but put Ideas on pedestals; to the Jews who broke the Golden Calf but built the Temple; to the Christians who burned pagan statues but painted icons; to the Protestants who whitewashed frescos but brandished the true text of the Bible from the pulpit; to the revolutionaries who overthrew the old regimes but founded a cult devoted to the goddess Reason; to the hammer-wielding philosophers who put a stethoscope to the cavernous emptiness of the statues of every cult but put the ancient pagan gods of the will to power back in place. (Latour 2010: 30–31)

By comparison, the tone he takes toward political activists or academic radicals on the pages that Martin cites are quite mild.

Equally lopsided are the comparisons of Latour’s thought and theorization of agency to the anti-humanism of Althusser, which denies humans agency in the face of the all-determining structures of history (Martin 2014). Latour is not denying human agency at all. In fact he sees the denial of human agency as simply the other side of the coin of declaring all agency human. Consider his analysis of a cartoon about stopping smoking. In the first frame a middle-aged father is comfortably relaxing in a chair smoking a cigarette. His young daughter asks him what he is doing. He says, “I am smoking a cigarette.” The daughter innocently replies, “Oh, I thought the cigarette was smoking you.” The father then panics, cutting his remaining cigarettes and their package to shreds with a pair of scissors. Latour comments:

Moving from the first to the last scene, we basically pass from one extreme to another: at the start, the father believes himself given to an innocent vice that he has almost completely under control; at the end, he can extricate himself from his shackles only by pulverizing the cigarette, which so totally controls him that his daughter thought she had seen, in this hybrid, a cigarette smoking a man. In the two instances . . . the reader believes that we are talking about control. From the active form “I smoke a cigarette,” to the passive form, “you are smoked by the cigarette,” nothing has changed other than the apportionment of master and instrument. The father alternates too drastically from one position to the other: too comfortable in the first image; too panicked in the last. What if the question rested instead on the absence of mastery, on the incapacity— either in the active or passive form—to define our attachments? How can we speak with precision of what the Greeks call “the middle voice,” the verb form that is neither active nor passive? (Latour 2010: 55–56)

Two pages later, Latour describes what a good “middle-voiced” reply by the father would have looked like:

“Yes, Mafalda, my daughter, I am effectively held by my cigarette, which makes me smoke it. There is nothing in this resembling a determining action, neither for it nor for me. I do not control it any more than it controls me. I am attached to it and, if I cannot hope for any kind of emancipation from it, then perhaps other attachments will come to substitute for this one.” . . . We can substitute one attachment with another, but we cannot move from a state of attachment to that of detachment. This is what a father should tell his daughter! (Latour 2010: 58)

[47]A Latourian analysis does not deny human agency but examines how other things mediate that agency. These things can take many forms. A surfer or a white-water kayaker might say that she struggles to become “one with the water,” to sense through the surfboard or the kayak, the water’s every movement. At the moment when she performs her most difficult maneuver, she could be said to have fully adjusted to the water’s ebbs and flows and, therefore, to be following the water precisely. A violinist or guitarist may talk of the years it takes to learn “what his instrument may do.” This may refer both to the years needed to learn to play the class of instrument in general, and to the microadjustments necessary to get the most out of the idiosyncrasies of the particular instrument he plays. But these examples do not need to only be about nonhuman things. Perhaps our musician is a singer and she speaks of spending months learning how to get the most out of a particular song, to see “where that song can take her.” Or perhaps a person seeks out certain forms of social relationships because she believes that they will transform herself. These transformations could be of an emulative nature. We can only hope that young students come to admire the anthropologists on campus, both students and faculty, and decide to study anthropology in order to spend more time with anthropologists and thus become more like them. But they could also take a less iterative form. A young man could decide to become a nurse at a psychiatric institution because he believes that helping difficult people will transform himself into a better person, not because he wants to become more like the patients. But whether these agentive attachments are between humans and nonhuman objects, between humans and human products like songs or books or artworks, or between one human and another, all human agency takes place through attachments. To become detached is to erase agency.

It is even possible to read the critiques of Latour presented by Martin in a Latourian vein. Martin seems to tell us that reading the works of Latour could make us less moral anthropologists than reading the theories of humanistic Marxists like Lukács. Thinking in “posthumanist” terms such an argument goes, makes us less likely to see the abuse of power by corporations, less likely to politically fight the greed of the 1%, less likely to join the movement to protect the downtrodden masses. It might even be worse than smoking too many cigarettes! Form your attachments carefully, the argument seems to say, for the ethical future of anthropology as a collective discipline and you as an individual anthropologist depends on it.

And are these texts, by Latour or Lukács or Marx, human or nonhuman agents? Clearly it would be wrong to suggest that the effects of the texts are simply the responsibility of the authors themselves. Is Marx responsible for the crimes of the Soviet Union, the People’s Republic of China, and the Democratic People’s Republic of Korea? I think not. Neither is Marx or any other author reducible to his books. Nor would it be correct to say that the readers of texts accept the crimes of the authors. But even if readers do not accept the crimes of the authors they read, or the crimes that others have committed in the name of those books, is it fair to say that they simply impose their own will on the text? Again, I think not. We read to be moved. A text that simply reinforces what we already know, no matter how well written, bores us. Reading is the search for engagement, a give and take, the start of a dialogue that will change the way we think, alter our feelings, transform our outlook. In this sense the text, a hunk of disembodied discourse, like a musical [48]instrument, a hammer, or a song on the radio, is an agentive human artifact. Shaped by humans to be sure, but not human in itself. In their concerns with the ethical effects of reading and applying various forms of theory, the “humanist” arguments of Gregory, Graeber, and Martin also turn on the agency of nonhuman things.

On agency and agencies

While it is possible in the Latourian sense to ascribe agency to anything in the world, it is not necessary to say that all agencies are the same. Putting my own spin on Latour’s framework, I would begin with positing that the divisions in academia reflect the division of the world into types of agents. The physicists deal with ever smaller and smaller particles. The chemists deal with their chemical “agents.” The biologists with living creatures, the art historians with art works, and the anthropologists with humans in their “societies.” The goal of any form of research is to better understand the properties of a given form of agency so that humans in general, or, in this day of privatized research, a particular group of humans can better manipulate those agencies or profit from their manipulation. But it is only by working through the agencies of different things that humans accomplish anything.

The differences between various categories of agents are no doubt up for debate, but laying out some usually accepted starting points allows for discussion. The ways in which physical particles interact with one another are complex and multiple, but at least since Newton (or perhaps Aristotle), we have known that there are no unmoved movers. The agency of one particle must be understood in relation to the agencies of others. With life another factor enters the equation. Life actively seeks to persist over time. Living beings that reproduce sexually exhibit different dynamics of agency from those that reproduce asexually. Beings that move have different limitations than those that are rooted. Beings that can directly utilize the energy of the sun differ from those that need to eat. Moving further toward the human, we might ask what distinguishes beings who are self-conscious, who can think themselves as separate from the worlds that they inhabit and thus consciously make choices about their course of action. Gordon Gallup (1979) famously operationalized self-consciousness by placing mirrors in front of various animals and studying their reactions. Only a few animals (chimpanzees but not monkeys, for example) demonstrated the capacity to recognize that the creature in the mirror was itself rather than another animal. From this fact he posited that a limited range of animals, including but not limited to humans, had self-consciousness. And finally, to demonstrate on my own humanism, I suggest that existing as a self-conscious animal in worlds that include complex symbolic systems which have evolved over millennia and which mediate communication among fellow species members gives a special twist to the dynamics of agency.

This manner of categorizing agencies differs considerably from that put forth by Latour (2013) in his An inquiry into modes of existence. Strictly speaking, Latour does not categorize agentive beings but rather “modes of existence,” each of which has its own “felicity/infelicity conditions” and types of “beings to institute” (2013: 488–89). But he does place the type of agents I refer to—inanimate objects, plants and animals, human products like technologies, artworks and fiction, and human [49]institutions like law, religion, politics and science—into his categories of modes of existence, so some resemblance emerges. But the gaps between our schemes reveal differing attitudes toward human agency, and, perhaps, different degrees of “posthumanism.”

Latour begins his categorization of modes of existence by pointing out that science and law have differing conceptions of truth—differing types of felicity and infelicity conditions—making them different modes of existence. Fair enough. But he then applies the notion of felicity and infelicity conditions to all possible modes of existence and the beings they involve. He portrays technologies, plants, animals, and mountains as having felicity and infelicity conditions. He acknowledges that some readers may object, but insists on proceeding with the extension of the meaning of “felicity and infelicity conditions” away from linguistic notions of truth and toward processes like the ability to reproduce over time. In my view, such an extension of what at base is a concept of truth results in forms of anthropomorphic fallacy typical of posthumanists who wish to flatten the differences between human and nonhuman agency. First such posthumanists deny the importance of what is typically considered unique about human agency—language and self-consciousness. Then they metaphorically extend concepts derived from the study of human language and consciousness to nonhuman agents in anthropomorphic language. Latour, for example, depicts how the “mode of technological beings . . . misunderstands” other modes of existence (2013: 215). Mute forms of life are said to “enunciate themselves (in the etymological sense)” (2013: 285). In Latour’s case, even the distinction between inanimate and animate beings is blurred as mountains are depicted as reproducing themselves in the same way as animals: “a mountain makes its way in order to maintain itself in existence” (2013: 88). Here I object again. In actively seeking to reproduce themselves, plants and animals exhibit a type of agency quite distinct from mountains. Mountains may persist over time but it is not because of their own active efforts. Latour generously invites other scholars to participate in his project by suggesting and arguing for additional modes of existence to add to his framework. But because of the centrality of “felicity/infelicity conditions” to his categorization, I would find it impossible to do so.

Returning to my own categorization of agencies, I would admit that space must be made for revision in the ways in which I draw boundaries between various forms of agency. When chemists discovered that chemical reactions involved the reorganization of the orbital paths of electrons, then particle physics became relevant to chemistry. When socio-biologists insisted that the social lives of non-self-aware creatures demonstrate a biological basis for all social organization, social anthropologists became defensive. But the anthropologists were able to articulate forceful responses to the socio-biologists and the chemists have maintained enough specificity to the objects of their research to maintain a separate discipline.

Consequently, though I will not argue that biologists will never discover greater commonality between humans and certain animals than is currently commonly acknowledged, I see no problem with maintaining a position that humans, like all species, have unique characteristics. I thus see anthropomorphic ascriptions of agency as a fallacy. This fallacy is not one of ascribing agency to things but one of ascribing a human form of agency to things. So when a surfer who wants to be one with the water asks the Ocean Gods to be kind to her, to protect her safety while [50]showing her the most exciting waves and the best ways to ride them, she gives the water a form of intentionality and capacity to make choices that we do not need to accept. When a scientist, frustrated with her complex experimental apparatus and months spent reading its output with no advancement in knowledge, yells out to her instruments, “Speak to me!” we can understand the entanglement of her agency without accepting that the experimental apparatus itself has the capacity for self-recognition and decision making. And when I sit at my computer with writer’s block and invoke the muses, begging them to let the words move through me, to arise in my head, flow through my hands to the keyboard and on to the screen, I do not really believe in the existence of a sentient being capable of such action. But I do acknowledge the complexity of my relationship to language, that the process of academic creation involves the interaction of reading and writing, involves listening to too many other voices to remember to find my own, involves emulating some, resisting others, and unconsciously echoing others still. In short, I think that the language speaks me as much as I speak it, and that our relationship should properly be described in a middle voice.1

To sum up, the position I take here is that many types of agency exist and that agencies emerge through entanglements and attachments—no agency exists as an isolate. But I still see human agency as distinct in some ways from the agencies of other things and creatures. The dilemmas of human agency bring us into the world of ethics and choice, which is the realm on which Gregory, Graeber, and Martin would like us to focus our disciplinary attention.

Human agency and the ethical

In a stellar essay on the relationships between ideas of agency and those of human ethical responsibility (an essay seemingly ignored throughout the special issue on knot-work), James Laidlaw (2010) contrasts ideas of agency arising from Bourdieuian practice theory with those from Latourian ANT. Practice theory led to a conception of “agency” as an abstract capacity held by particular individuals and opposed to the social “structures” that restrain those individuals. This conception places agency and structures in a zero sum game and implies that politically committed anthropologists should devote themselves to “enhancing” the agency of oppressed people so that they can be liberated from their oppressive structures. The agency that is to be enhanced is completely conscious and should arise solely from within a given individual. Such a concept of agency resonates well in a society where individuals are constantly asked to make choices. But such a concept of agency cannot explain how people become agents of a particular type. It ignores the necessity of agency arising through entanglements.

As Laidlaw (2010) points out, oppressed agents who seem to voluntarily submit to their structures of oppression are particularly problematic for such a conception of agency. Thus when Saba Mahmood (2001) or Tanya Luhrmann (2015) discuss pious Muslim or Christian women who exercise their agency by actively submitting [51]to patriarchal religious institutions, they pose a challenge to the type of feminism that insists on a purely “humanist” view of agency. But such a humanism misses the point that to become an agent we must submit to something. Most readers of HAU have submitted to the discipline of anthropology. This submission has meant reading texts under the direction of certain established academics and having one’s writing criticized, corrected, and edited by these same academics. Even after receiving our doctorates, we face a lifetime of criticism and rejection. Our writing must be peer-reviewed and then edited by someone else. And the opportunity to be reviewed and criticized only occurs if we find a venue that does not reject us outright. The politically active among us must submit to a cause. Activists must tailor their voices to the needs of the movement. Protestors must shout their slogans in unison. Whose voice is more regulated than a professional politician who gains electoral success by staying “on message” and sticking to the party line? Power in democratic politics, or in any society where popular legitimacy matters, comes from numerical dominance and the only way to achieve this dominance is to speak in a voice that the majority appreciate. Does a politically active left wing anthropologist really exert or “have” more agency than a piously submissive evangelical Christian? At best we might say that modern humans exert agency by choosing what and whom to submit to in order to modify the shape of their own agency.

Questions of ethics, of the morality of our choices, of the attribution of blame and responsibility for wrongdoing sidestep the question of how a particular type of human agent comes into being. As Laidlaw points out, other concerns arise when humans ascribe ethical blame or responsibility to a human or human-like actor in a wide range of societies. Following Bernard Williams (1985), Laidlaw identifies these concerns as cause, intention, state, and response. To be blamed for something an individual or group of individuals must be held to have caused a negative situation; their blame increases if this result was intended by the actor; their blame might be reduced if their state of mind was induced by an outside actor—a spirit who possessed them, a temporary madness that overtook them or perhaps drugs or alcohol; finally, the actor should have the capacity to make a response. An agent without the ability to pay compensation, or who lives outside the jurisdiction where the relevant authorities can capture him, is simply not worth pursuing.

Of these concerns, state of mind comes the closest to the question of how a human comes to be a certain type of agent, that is, an agent with the proclivity to make certain types of choices over others. Lawyers in a court of law might be tempted to stretch the logic of a state of mind defense; even if a certain murderer did it, intended to do it, was sober at the time of murder, and was caught by the authorities that have jurisdiction, the lawyer might argue that the murderer’s actions were influenced by her genes, the family in which she grew up, or the abuse she suffered in school as a child. The range of potential causes for a human agent to be predisposed to make certain types of choices is enormous. Such arguments could draw upon social, psychological, biological, or even anthropological research. “Culture,” “Social Structure,” or “Habitus” all are possible anthropological conceptions for such an argument. In this sense, Latour’s or the posthumanist focus on nonhuman agency is hardly unique in the way in which it makes the attribution of agency difficult. Almost any serious form of social theory, not just the structural Marxists or the posthumanists, theorizes how certain people become agents of a certain type. [52]Here I find myself sympathetic to Michael Fischer’s (2014) reading of Latour, in which Latour is framed in relation to a series of other Western thinkers rather than seen, for better or worse, as a singular starting point, an unmoved mover, for a wide range of ideas about agency.

But in the courtroom or in everyday life, there are limits to how far such arguments may be pushed. We may never know why the murderer has such violent tendencies, or why a fraudulent banker became so greedy, but we agree that such people need to be punished both to prevent them from committing further crimes and to discourage others from following in their path. In fact, one could argue that a serious Latourian attempt to understand the processes of subject formation through the interaction of mediators could increase the space for human ethical action. In so far as one did understand how certain inclinations came to be, one could make a choice to engage with mediators that increase inclinations for ethical action. But though such a project could attribute agency to a wide range of human and nonhuman entities, it would still need to distinguish between agents that make self-conscious choices and those that do not.

ANT and political critique

The relationship between any form of theory and political action is difficult to assess. The political activism of theorists themselves shows little pattern. Pierre Bourdieu’s depictions of humans as solely preoccupied with the pursuit of capital (of one form or another) did not prevent him from protesting loudly against neoliberalism. Michel Foucault’s “posthumanist” deconstruction of the subject did not impede his activism on behalf of prisoners and against racism. Bruno Latour’s ANT has not kept him from attacking climate change deniers. But the question of the political limitations of certain forms of theory exceeds that of the political inclinations of any particular theorist.

While certain conceptions of ANT could make adequate theoretical room for critical anthropological efforts of the sort that Gregory and Graeber might support, Latour himself rarely moves in those directions. Moreover, careful ANT research by anyone raises difficult methodological issues for critically minded ethnographers. Consider an imaginary case of an ANT researcher investigating the activities of a selfish banker bent on greedily accumulating an even greater share of the earth’s resources than he already controls. Let us imagine that with the help of a team of accountants, lawyers, economists, and IT specialists, the banker comes to control a “derivative machine.” The machine can rapidly undertake complex stock transactions by manipulating many forms of economic data. It is electronically fed with latest information at speeds milliseconds quicker than is available elsewhere. The accountants have told the banker that they cannot guarantee how much profit the machine will make him, but more often than not, each time he sets it into motion, money will move out of other peoples’ brokerage accounts and into his. The economists are not sure of the long-term effects of using the machine, but warn him that using it too often increases the danger of worldwide economic collapse. The lawyers say they have managed to construct a state of legal ambiguity around the use of the machine but still suggest maintaining a [53]state of “plausible deniability” around his relationship to it. The banker informs none of these people of his precise intentions and hides the machine in his office’s private bathroom. Every day, claiming constipation, he spends hours in his bathroom, setting the machine into action on different settings and recording the rate at which his brokerage accounts grow. If the world economy crashes, he reasons, governments will have to intervene but the profits of using the machine go to him and him alone.

Faced with such a subject, an ANT researcher would have many politically critical topics to consider. The Latourian black box of the machine needs to be opened up for understanding. What accounting categories does the machine manipulate? What trading formulas does it use? What legal definitions allow for ambiguity over the process of taking money out of one person’s account and putting into his? How is the machine wired and how does it interact with the technical systems of various banks and brokerage firms? Exactly how does the banker use the machine and what does he do with the data he derives from its use? Does his constant use of the machine and his careful recordings of data do anything to the banker’s own subjectivity? Does it make him even greedier than he was before? None of these questions are without a critical edge.

But of course such a scenario would never occur. The most improbable part of it is not so much the existence of derivative machine itself, nor the existence of a banker ready to use it or a team of accountants, economists, lawyers, and IT specialists willing to help in its construction. Rather the most improbable part of this admittedly contrived scenario is the ability of our researcher to gain the research access necessary to answer her questions. She would never be allowed to see the machine or the bathroom, let alone observe the banker using it and recording his data. She would not be allowed to interview the accountants, lawyers, IT specialists, or economists, and if she did get a short interview with the banker, most of his answers to her questions would be lies.

The methodological problem that ANT poses for politically critical research is not so much that it considers the agencies of nonhuman entities, or that in “posthumanist” fashion it considers agency to arise in networks rather than to reside in the individualized subjectivities of conscious human actors, but rather that it requires researchers to engage in slow, painstaking, and careful ethnographic research. Doing this research properly requires research subjects who are willing to both answer questions as truthfully as they can and to allow us to observe and possibly participate in their practice. While such research may be possible among surfers who want to become one with the water, or even among scientists manipulating complex experimental apparatuses, it is rarely possible among powerful economic and political actors bent on amassing even more money and power. Here I find much in common with Kim Fortun’s (2014) discussion of Latour and the importance of the silences and absences that emerge in ethnographic research.2 A focus on injustice [54]often requires pursuing what our research subjects do not want to show us or tell us, what is hidden in the ideology of their language, or what is latent in the situation that they accept as the ground for their decision making. Pursuing such issues may require relying on the work of historians, cultural studies analysts, investigative journalists, or medical researchers, or adopting their methods for our own ends.

This political problem with ANT is thus the same as that of any ethnographic research. Ethnography works best with cooperative research subjects, and even when our subjects are cooperative, we must use other methods to explore the absences and silences in the narratives they tell us and the processes they show us. But the answer to this dilemma is not to abandon ethnography or ANT but rather to make these methods part but not all of the anthropological toolkit. Doing ANT or politically critical research are not either-or choices. In addition to ethnography, anthropologists can undertake the sort of research critical investigators undertake when they do not have the cooperation of their subjects. If many other disciplines now at least attempt or claim to do ethnography, why should we not avail ourselves of the full range of methods known to social researchers?

It is also true that Latour does not often take interest in political economic inequality. To me this comes across most forcefully in An inquiry into modes of existence: An anthropology of the moderns (Latour 2013), rather than in his famous essay “Why has critique run out of steam?” (Latour 2004), which I read as simply a reiteration of his argument not to reduce the world to a few black box forces like power and capital, and thus to conduct painstaking ANT-style research rather than positing conspiracy. In contrast, the lack of discussion of Capital and Power in his seemingly encyclopedic “Anthropology of the Moderns” makes me cringe. How can capital not be one of the modes of existence for modern humans? Latour does carefully analyze economic calculation as a mode of moral thinking (2013: 443–74), and I have some sympathy with what he says there—who should get how much and why is indeed the most moral of problems. But the problem he ignores is how capital as an instituted agent compels us moderns to make these calculations in an exclusionary manner. When a CEO thinks “screw the workers, the consumers, and the people who live near my factory, my job as a CEO (with many stock options) is to maximize returns for the shareholders,” capital demonstrates its instituted agency. Or, to use a more homely case, when I shop for the best bargain at the supermarket, without considering the agricultural workers who produced the product or the ecological effects of its production, I am making my “moral calculations” purely for the benefit of my own nuclear family. Capital is an instituted agency that pressures us moderns to make our calculations selfishly. It is too important to ignore.

Finally, I find Latour’s relative disinterest in human power relations mirrored in his lack of attention to the concept of power. If I were rewriting his conceptual framework, I would substitute “power” for “felicity/infelicity conditions” as the primary dimension for differentiating types of agents. Power is both a quantitative and qualitative concept. Consider the difference between a lion bounding at you across the savannah and a kitten bounding at you across the living room carpet. [55]The former is no doubt more powerful than the latter, but its power is also of a different type. The lion’s power derives from its size, strength, and desire to eat you. The kitten derives its power from its ability to entice you to care for it, to form an emotional attachment to it. Power takes many forms, but its relative quantity always matters.


Marx’s critique of commodity fetishism is well known. Under the spell of capitalism, people mistake ownership as a matter of a relationship between a person and a thing (a relationship between an owner and the thing owned) rather a matter of a relationship between people (a relationship between those who do and do not have the right to use the thing in question). Capitalists can even purposefully depict ownership as a relationship between a person and a thing to silence discussion of the relationship between classes. Gregory (2014) invokes this critique to imply that ANT threatens us with same form of misrecognition. But I think that Gregory’s invocation unnecessarily slips us into an either-or type of thinking. I own a car. Undoubtedly this ownership grants me a form of privilege envied by some who would like to own a car but cannot afford it. My teenaged son certainly thought this way whenever he requested to use the car but I refused. But the car also affects me directly. Perhaps I walk less often because of the car and have become less healthy. Perhaps I hit a kangaroo with my car and feel sickened by the experience. Perhaps I visit my aged mother more often because of it. Perhaps some of you do not begrudge me my stinking, little car, but do resent the fact that my driving it further pollutes the air that we all have to breathe. In all of these examples, the relationships among (and the agency of) people and things are thoroughly intermixed. Marx rightly criticized those who insisted that the relationship of a person to a thing did not involve other people. But we are not required to take the opposite position and ignore the intertwining and imbrication of people and things. Like Michael Herzfeld (2015: 19), I “would prefer to avoid the rather arid debate about whether objects have agency.”

I have three propositions to prevent such an arid debate from tying us down. First, I suggest granting agency to everything (which is to acknowledge that anything and everything could affect us), but differentiating types of agency. The types of agency we get with life and self-consciousness are particularly worth differentiating and highlighting. Self-consciousness raises the problems of ethics and choice, and our own ethical concerns require us to consider the effects of our agency on other humans above all. Second, I suggest acknowledging that Newton’s law of no unmoved movers applies to human agency as much as it does to objects moving through space. This rather modest position, like Latour’s call for a middle voice, or Marx’s famous quote that “Men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly found, given and transmitted from the past” (cited in Tucker 1978: 595), is hardly a form of antior posthumanism. Every form of social theory between the extremes of volunteerism and determinism attempts to voice such a compromise. Nor does such a position render ethics impossible. We [56]can advocate an ethics of nonviolence or equality, and even argue that murderers or fraudulent bankers should be punished, without denying that something outside of the murderer may have caused him to be so violent or that something outside of the banker may have made her so greedy.

Finally I would like to suggest that we differentiate agency from power in general and human agency from human power in particular. Certainly I can empathize with the political stance of those who hope to “enhance the agency” of relatively powerless people. But I prefer the language of “empowering” them for analytic precision. I see human agency as that contradictory space we experience whenever we attempt to make a decision about the unknown. If I really knew what the outcomes of a particular decision would be—should I read this book or that book, should I pursue this research project or that one—then I wouldn’t have a choice to make. The best path would be obvious to me and my actions would be dictated by that path. The criteria of what I considered to be best would have been dictated by the “circumstances directly found, given and transmitted from the past.” It is only when I don’t know what to do—when I have to flip a coin or go with my gut instinct— that I can say that I am making a choice. It is this feeling of needing to make a decision in the face of unknown consequences that makes me aware of my human agency. Such a feeling or a dilemma is not the same as power. It is not something that one can have more of or less of. Barack Obama or Bill Gates undoubtedly have more power than I do. They get to follow their proclivities or gut instincts across a wider range of contexts than I do and their decisions certainly affect a greater number of people than mine. But this power does not give them more agency than me. Though their gut instincts are more powerful than mine, their relationships to their guts are not “larger” than mine. And if the latest medical research is to be believed, then this relationship may literally be mediated by the bacteria in our guts to a greater extent than any of us previously imagined (see Smith 2015)!


American Anthropological Association. 2013. “Program: 112th Annual Meeting, Chicago, Illinois.” Arlington, VA: American Anthropological Association.

Berreby, David. 1994. “And now, overcoming all binary oppositions, it’s that damned elusive Bruno Latour.” Linguafranca 6 (4): 1.

Cowan, Jane K. 2014. “Struggling over procedures and other interventions: A comment on David Graeber’s ’Anthropology and the rise of the professional-managerial class.’” HAU: Journal of Ethnographic Theory 4 (3): 89–97.

Fischer, Michael M. J. 2014. “The lightness of existence and the origami of French Anthropology: Latour, Descola, Viveiros de Castro, Meillassoux and their so called ontological turn.” HAU: Journal of Anthropological Theory 4 (1): 331–55.

Fortun, Kim. 2014. “From Latour to late industrialism.” HAU: Journal of Anthropological Theory 4 (1): 309–29.

Gallup, Gordon G. Jr. 1979. “Self-awareness in primates.” American Scientist 67: 417–21.

[57]Graeber, David. 2014. “Anthropology and the rise of the professional-managerial class.” HAU: Journal of Ethnographic Theory 4 (3): 73–88.

Gregory, Chris. 2014. “On religiosity and commercial life: Toward a critique of cultural economy and posthumanist value theory.” HAU: Journal of Ethnographic Theory 4 (3): 45–68.

Herzfeld, Michael. 2015. “Anthropology and the inchoate intimacies of power.” American Ethnologist 42 (1): 18–32.

Kipnis, Andrew B. (Forthcoming) 2016. From village to city: Social transformation in a Chinese county seat. Berkeley: University of California Press.

Laidlaw, James. 2010. “Agency and responsibility: Perhaps you can have too much of a good thing.” In Ordinary ethics: Anthropology, language and action, edited by Michael Lambeck, 143–64. New York: Fordham University Press.

Latour, Bruno. 1987. Science in action: How to follow scientists and engineers through society. Cambridge, MA: Harvard University Press.

———. 1988. The pasteurization of France. Cambridge, MA: Harvard University Press.

———. 1993. We have never been modern. New York: Harvester Wheatsheaf.

———. 2004. “Why has critique run out of steam? From matters of fact to matters of concern.” Critical Inquiry 30: 225–48.

———. 2005. Reassembling the social: An introduction to Actor-Network-Theory. New York: Oxford University Press.

———. 2010. On the modern cult of the factish gods. Durham, NC: Duke University Press.

———. 2013. An inquiry into modes of existence: An anthropology of the moderns. Cambridge, MA: Harvard University Press.

Latour, Bruno, and Steve Woolgar. 1979. Laboratory life: The construction of scientific facts. Los Angeles: Sage.

Luhrmann, Tanya M. 2015. “The appeal of Christian piety.” New York Times, July 4, 2015. http://www.nytimes.com/2015/07/05/opinion/sunday/t-m-luhrmann-the-appeal-ofchristian-piety.html?partner=rssnyt&emc=rss.

Mahmood, Saba. 2001. “Feminist theory, embodiment, and the docile agent: Some reflections on the Egyptian Islamic revival.” Cultural Anthropology 16 (2): 202–36.

Martin, Kier. 2014. “Afterword: Knot-work not networks, or, anti-anti-fetishism and the ANTipolitics Machine.” HAU: Journal of Ethnographic Theory 4 (3): 99–115.

Smith, Peter Andrey. 2015. “Gut feelings: Can the bacteria in your gut explain your mood?” New York Time Magazine, June 28, 2015. MM46. http://www.nytimes.com/2015/06/28/magazine/can-the-bacteria-in-your-gut-explainyour-mood.html.

Tucker, Robert C., ed. 1978. The Marx-Engels reader. New York: W.W. Norton & Co. Williams, Bernard. 1985. Ethics and the limits of philosophy. Cambridge: Harvard University Press.

L’agency, entre humanisme et posthumanisme: Latour et ses détracteurs

Résumé : Deux articles de la section dédiée par Hau aux nœuds (Hau 2014, volume 4, numéro 3) s’en prennent au « posthumanisme » de la théorie de l’acteur réseau (ANT) de Bruno Latour. En arguant que la conception de l’agentivité (agency) de Latour sape les attitudes critiques vis à vis du capitalisme, ils laissent le chercheur devant un choix manichéen entre l’adhésion ou le rejet des travaux de Bruno Latour. Dans cet article, j’ébauche une position alternative face au concept d’agentivité non-humaine et à l’œuvre de Latour – une position critique mais bien moins polémique. Si je propose une position intermédiaire dans l’optique de fonder une théorisation de l’agentivité, je conclus que ce n’est pas la théorisation de l’acteurréseau qui entrave les ethnographes critiques du capitalisme mais bien plutôt ses modes d’application, en particulier ceux qui dérivent de l’insistance de l’ANT sur la recherche ethnographique assidue.

Andrew B. KIPNIS is a Professor of Anthropology in the School of Culture, History and Language, College of Asia and the Pacific, at The Australian National University. Two of his books include China and postsocialist anthropology: Theorizing power and society after Communism (2008, Eastbridge) and From village to city: Social transformation in a Chinese county seat (2016, University of California Press). From 2006–2015, he was editor of The China Journal.

Andrew B. Kipnis
Department of Anthropology
School of Culture, History & Language
The Australian National University
Acton ACT 2601


1. Perhaps because of his desire to flatten the distance between human and nonhuman agencies, Latour does not designate language as a category of his modes of existence.

2. Also, like Fortun (2014), the category of “modernity” for me means much more than the separation of science from politics (as in Latour [1993]), or the emergence of certain forms of argument about agency and the iconoclastic rejection of fetishism (as in Latour [2004, 2010, 2013]), but involves the rapid acceleration of processes like industrialization and urbanization. Such a concept of modernity, however, does not require rejecting Latour’s arguments about the intertwining of science and politics or those about iconoclasm. For detail on how I grapple with the category of modernity in my own research see Kipnis (forthcoming).