Further thoughts on sylvan thinking

This work is licensed under the Creative Commons | © Eduardo Kohn. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau4.2.016


Further thoughts on sylvan thinking

Eduardo KOHN, McGill University

Response to HAU Book Symposium on Kohn, Eduardo. 2013. How forests think: Toward an anthropology beyond the human. Berkeley: University of California Press.

In this essay I respond to comments by Marisol de la Cadena, Philippe Descola, Cesar Giraldo and Gisli Palsson, Bruno Latour, and Anand Pandian on How forests think: Toward an anthropology beyond the human, my exploration of how the Quichua-speaking Runa of Ecuador’s Upper Amazon relate to the “living logics” that tropical forests exhibit and how attention to this relation invites us to rethink our understanding of anthropology, what it means to be human, and the ways we might face the global ecological crisis. I wish to thank the contributors for the time, care, and effort they have put into commenting on How forests think. I am immensely grateful. In my response I seek both to highlight what I share with the contributors and to faithfully rearticulate in the strongest of terms their concerns. I hope to address these concerns in a way that can further this project of giving voice to sylvan thought.

César Giraldo and Gisli Palsson worry that by delineating what makes humans distinctive—specifically that we humans are uniquely symbolic beings—I am reinstating the very nature/culture divide I seek to overcome and thereby sabotaging my attempt to take anthropology beyond the human. According to them, my claim that humans are distinctive is not supported by my ethnography nor is it supported by biosemiotics or ethology.

I’d like to respond to their concerns by reiterating more explicitly what I am trying to do. My project is unabashedly ontological by which I mean this: I feel I need to explore what “is” in order to find space for imagining and enacting what could be. Of course, if anthropology has taught us one thing, it is that venturing to make ontological claims of the sort I try to make is a very risky endeavor. Nonetheless, I think it is an undertaking that is necessary in our times. Given that human and nonhuman lives are coming to share an increasingly entangled and uncertain future my goal is to find a form of attending to the world beyond humans that allows us to develop conceptual tools from the properties of the world in ways that can speak to this problem. My method for doing so is to use ethnographic as well as other empirical modes of attention, to harness some of these properties so that they can become manifest as they take us beyond ourselves.

So, I am interested in an ontological exploration of the different properties of this world that lies beyond us because I want to find ways to harness these properties. I am especially interested in those properties that are not necessarily human—properties that often elude us anthropologists, thanks precisely to our tremendous success at being able to speak to the human. Like a surfer trying to ride a wave (who has to get something “right” about waves in order to harness their power) this quest involves getting things right about the world. Although my interest is in the nonhuman, my concern is that if we don’t get something right about the human, we end up projecting human qualities onto the world in ways that don’t allow us to harness some of these other properties of the world. So, although it might seem liberatory to break down those barriers that we assume exist between humans and nonhu-mans (and this is, of course, an important posthumanist strategy), my concern is that by only doing so, we often just reinscribe the human onto the nonhuman. Although one may accuse me of being anthropocentric in focusing so much on what the human is, my intention is to sidestep a greater anthropomorphic trap. That is, I need to say something precise about what the human is to be able to see all the other things that are “not” human that actually also make us who we are, and—if we can just learn to “ride” them—who “we” might become.

In this project I am particularly interested in the properties of biological life. And what continues to draw me to the Ecuadorian Amazon is that there is so much life there (so many more kinds of life forms and kinds of relations among life forms when compared to other parts of our planet). What drew me to Ávila was that the Runa living there, whose economy has depended on harnessing this vast and complex ecology, needed to find ways to “ride” its “waves.” The forest’s many interrelated life forms amplify life’s living logics, and the Runa in their relation to that forest make these logics manifest. Theirs, then, like mine, is an ontological exploration of the forest’s living logics. And these living logics, and how we can learn to better live and think through them, is what continues to hold my interest.

But, as I mentioned, in order to get at this, we need to say something precise about a particular subset of this logic that threatens to occlude the rest. This subset is one that is, on this planet, distinctively human. (I say “on this planet,” not to celebrate its exceptional status but to underscore that the logic of symbolic representation has formal properties that are not restricted to earthly or human contingencies; the symbolic likely makes itself manifest in other parts of the universe.) And I think we can say something precise about what makes humans distinctively human. What makes humans distinctive is our use of a specific kind of representational modality that involves what Peirce calls symbols. By symbolic representation I mean something very specific. I mean a representational modality that refers indirectly to its object of reference by virtue of the way its signs relate to each other in a conventional system of similar signs. Such symbolic representation is not wholly separate from iconic and indexical modes of representation. It is nested within and dependent on such modes, even though its logics and properties come to be different.

Many of the authors that the commentators cite use the term symbol in a different way. For some, this is simply a question of nomenclature. Howard H. Pattee (2001), for example, is primarily concerned with what he calls the “epistemic cut”—a distinction between processes that involve signs and those that do not. He thus uses sign and symbol interchangeably. For my purposes, however, I need to distinguish among kinds of signs. This is because different kinds of signs exhibit different kinds of ontological properties and make themselves manifest in different parts of the world. If I don’t recognize these distinctions, human symbolically informed thought (as I have defined it) risks becoming conflated with the broader semiotic field out of which it emerges with the consequence that we would lose sight of the distinctive possibilities inherent to these other nonsymbolic forms of thought.

In distinguishing between, on the one hand, the symbolic as a specific kind of semiosis, and, on the other, a broader semiotic field out of which the symbolic emerges, my goal is not to celebrate human uniqueness. Rather, my goal is to capacitate the nonhuman (nonsymbolic) in the human and the nonhuman. But my claim is that we can’t capacitate these other relational logics if we don’t first decolonize thought by finding ways to think beyond language-like ways of thinking.

I am in general agreement with the biosemiotic literature that Giraldo and Palsson reference. What, for me, is important about the biosemiotic endeavor, of which I see my work as a contribution,1 is that by redefining biological life as semi-otic we are able to breakdown a Cartesian divide that still relegates all of mind and thought to the human. In doing so, however, we must be careful not to see these other kinds of minds as human minds (again my goal is to capacitate this other form of thought, in part by treating it as thought). So I would agree, following Ray Paton and Koichiro Matsuno (1998), that enzymes and verbs share something in common (they partake in semiotic life). And I would agree with them that the “metaphor” of a text is more appropriate when thinking about enzymes than the metaphor of the machine (1998: 132). But I would not consider enzymes to be symbolic. Words in language have a kind of necessary freedom from the worlds they represent. Their relationship to the world is mediated by the system of other words in which they exist—a system that then relates as a whole to the patterns it traces among its objects of reference. Individual enzymes, by contrast, must specifically “fit” the molecules they break down. This makes them different kinds of relational entities.

Think of amylase, which is found in human saliva and functions as a catalyst to break down carbohydrates. It is because of amylase that chewing manioc is a central part of making manioc beer in Ávila. This practice facilitates the breakdown of complex starches into simpler sugars that can then ferment and yield alcohol. Amylase is certainly part of a metabolic system, analogous to the grammatical system in which a verb functions, but it also has to “fit” the carbohydrates it breaks down in a way that a verb doesn’t—for the fit of a word to the world in a symbolic system is much more fuzzy and distributed.

The same logic applies to the complex system of bee communication (see Crist 2004). Bee dances refer to their object of reference, such as a food source, through an intricate and layered iconic and indexical logic that link them to their object through relations of likeness and correlation, respectively. Bees are certainly engaged in complex, sophisticated semiotic thinking, but they are not thinking symbolically. Again, my goal is not to celebrate what is unique about humans but rather to understand this distinctiveness in order to harness relational logics that are not like it.

My interest in what makes the symbolic distinctive, then, should not be misunderstood as a recapitulation of a dualist argument, “of icons and indexes on the one side [i.e., nature] and symbols on the other [i.e., the side of culture].” Although one might well define culture in terms of symbol use, the symbolic and nonsymbolic-but-semi-otic cannot so easily be separated in a dualist fashion, as I discuss in great detail in my book, because symbolic reference is emergent with relationship to, and thus nested within and wholly dependent upon, other forms of reference (Kohn 2013: 54–62).

Seeing the symbolic as open in this fashion changes the nature of my relation to ethnography. I am involved in an ethnographically inspired ontological exploration of the living logics that tropical forests so beautifully exhibit, and this exploration is not necessarily bounded by the ethnographic context that inspires it. That I venture to say things outside of, and therefore potentially in tension with, ethnographic contexts can create ethical problems, which have been most eloquently and thoughtfully presented by Marisol de la Cadena and to which I will turn in my response to her contribution. I here only wish to underscore how Runa forms of thought resonate with properties of the world in ways that can teach us something about the world. So, Giraldo and Palsson may wish to emphasize how my work seems to be bounded by my own “Western metaphysics” at the same time that it seems to contradict a Runa mode of understanding. However, as I discuss in my response to de la Cadena, although I do not wish to sidestep the serious problems this sort of relativist critique poses, I do want to point out that, although the commentators are careful to avoid the terms, this critique still couches thought within human, that is, sociocultural, contexts.

I have argued at length about how our thinking exceeds such sociocultural contexts (Kohn 2013: 29, 38–39, 90). And in How forests think I explore through detailed ethnography how it is that Runa thought resonates with and amplifies the living logics of the world in ways that are not necessarily bounded by Runa sociocultural contexts. So, my interest in Runa animism is less about how that thought might be different from my own biosemiotic version of animism and more about the precise ways in which the Runa grapple with the kinds of animacies they confront in the world.

It is, then, an oversimplification both of my argument and of the Runa ethnography I present to say, as the commentators do, that I see the moral as human whereas the Runa recognize it in an unproblematic fashion as a property of a ge-nerically animate cosmos: How, for example, dogs should enter an ethical system is as much a problem for the Runa as it is for us (and, on a pragmatic level, it is also a problem for dogs), and this is what my ethnographic analysis seeks to explore.

What is more, even in an “animistic” mode of thought not all kinds of animacy are the same, and Runa reflections on these many animacies capture the different kinds of properties that make themselves manifest in an animate world; for example, spirits for the Runa are very different kinds of animate beings than humans. My interest here is not so much in asserting this but in exploring the specific recurring formal properties that Runa reflections on these reveal (Kohn 2013: 173–78, 216–17).

So, I would summarize the gist of Giraldo and Palsson’s concern by saying that it centers on how my attempt to decolonize thought (that is, my search for a way to partake in a form of thinking beyond distinctively human forms of thinking) stresses and thus privileges human language. Bees, chimps, and enzymes, according to them, think in the exact same way we do. What is more, they add, the Runa, if I could just heed my own ethnography, are trying to tell me this (after all, for the Runa, the commentators seem to be saying, it’s just people all the way down).

Anand Pandian’s commentary furthers this line of critique. And it is to his insightful and lyrical reflection that I now turn. Pandian and I share a similar vision of anthropology as a field that can provide us with tools to help us think outside ourselves to confront, as he so nicely puts it, “the larger problem of how humanity moves beyond the same—the stubbornness of our inherited nature, the lingering potential of its malleability, the forces and channels that might further provoke its becoming-otherwise.” And I think we appreciate the stakes for this today. This is why trying to clarify my thinking vis-à-vis his provocation is so important for me.

Pandian asks why decolonization of thought should stop at human language. He embraces the idea of thinking beyond the human but asks why I can’t take it further to recognize that thought is everywhere and in everything. Why do I stop at biology, which, after all is a field of study defined and delimited by certain humans in a specific time and place and thus one of “our” “provinces” of thought? I may, according to Pandian, have succeeded at provincializing language, but I’ve woefully failed at provincializing life. It is bad enough, as Giraldo and Palsson argue, that the biologists, biosemioticians, and even the Runa are against me, but now, as Pandian points out, and textually supports, so is Peirce himself!

Pandian asks, apropos of How forests think, “Do books think, or do living beings think with them?” By focusing only on the thoughts of living beings, he argues, I miss the thoughts inhering in things, and not just the thoughts beings (human and nonhuman) have about those things. Things themselves—like books, or spilled paint—then, for Pandian, think. And he turns to Peirce for textual support.

Peirce was battling the ontological dualism, which was one outcome of the scientific revolution, whereby mind became relegated to humans (or God) and matter to the mechanical world of efficient cause, with seemingly nothing to stand between the two. Peirce’s solution was a philosophical doctrine of continuism. Matter, for him, was a kind of mind and mind has certain material and energetic qualities. “The difference,” Pandian writes, “between mind and matter for Peirce, therefore, unlike Kohn . . . was one of degree rather than kind.”

Pandian’s way of framing this allows me to clarify something important about novelty and continuity. Following Peirce, and especially Terrence Deacon’s emer-gentist reformulation of Peirce’s continuism (Deacon 2012), my interest is not so much in kinds in their opposition to degrees, but in the ways in which degrees can become kinds. Peirce’s goal was not to show that everything in the world is the same as everything else—the panpsychic position that it’s all mind—but rather to show that there exists continuity among entities despite the fact that these entities come to exhibit different properties and potentials in a nested fashion. The question for Peirce and Deacon, then, is not just how to explain continuity but also how to explain novelty. Novelty, I should note, does not just refer to one-off newness. Rather it refers to the emergence of persistent unprecedented causal dynamics and properties that are nonetheless continuous with those dynamics in which they are nested and from which they stem. So, Peirce wanted to understand not only how things are related (e.g., that matter is a kind of mind and therefore not something totally different from mind), he also wanted to explore the specific properties inherent to different kinds of entities, processes, and relational logics.

Now, on Peircian grounds it is difficult to accept that things think, for, if thinking is glossed as semiosis, then thinking requires an interpreting self. Books, to think on their own, would need to write and manufacture subsequent books that would read them, and these new books would then need to create their own books to read them and to be read by still other books. The same would apply to spilled paint. But I am less interested the question of who or what should be labeled a thinker and more in asking how the recognition that certain kinds of nonhumans think can help us harness properties in the world we might not otherwise notice.

In this regard, my concern is that in universally extending thought to things much of the rich way in which the Peircian approach recognizes unexpected properties of the world is lost. That is, I don’t think that the attempt to commensurate Peirce with Deleuze helps us. To make matter into thought hides those properties of the world that I wish to harness. To make things think is to reduce human and nonhuman thought to affect and vibration. In Peircian terms this results in a flattening of the world such that creative agency becomes limited to the domain of what he called “secondness.” This may well provincialize human “thirdness” (i.e., symbolic representation as exemplified by human language), but too much gets lost in the process. What we specifically lose is an appreciation for all the kinds of thirdness that are semiotic but not symbolic, as well as an appreciation for certain kinds of thirdness that are nonliving and therefore nonsemiotic (which I explore in detail in chapter five under the rubric of “form”).

In sum, if we say that every entity in the world is alive in the same way we lose something. We lose the ability to harness specific properties that an ethnographic ontological exploration beyond the human might reveal. If, on the other hand, one were to counter that the interest is in the different kinds of lives thought might take then we would have to specify the formal nature of those differences, and this would force us into the kind of ontological territory I am exploring.

There is a reason why I feel that this kind of ontological exploration is important. My question is: What form might an ecological politics in the Anthropocene take that harnesses specific ontological properties of the world, which are revealed ethnographically in certain parts of the world (namely in a complex ecosystem like the Amazon forest)2? I am not simply content with a world that we can romanticallylabel animate; I want to explore and harness its many specific properties. My claim is that constitutive absences, selves, and futures, as I’ve discussed them in the book, make themselves manifest in specific ways that are particularly apparent in specific parts of the world. And I want to find ways to think with these in ways that can change our thinking.

Marisol de la Cadena’s polite and thoughtful response, poses, without being agonistic, the greatest challenge to my work. I am very drawn to this and will attempt to respond in kind; for it is my hope that in the vision of an alter-politics (Hage 2012) we both share, attack-and-defend need not be the only form of thought, just as resistance need not be the only form of political agency. I would like to sit with the ethical problem de la Cadena raises with the hope that it will clarify my thinking.

De la Cadena’s concern is that I too quickly equate Runa with human being, and, as a consequence, certain kinds of nonhumans with biological life, and others, with geology. What, she asks, thinking of her own work in the Peruvian Andes around Cuzco, of the tirakuna, the earth-beings, if she were to adopt my own anthropology beyond the human in which, as I bluntly and perhaps insensitively say, “life thinks; stones don’t” (Kohn 2013: 100)? What happens to those tirakuna after this pronouncement? My statement seems to erase their mode of being. And what happens to the Runakuna who live with those tirakuna?

My work, according to de la Cadena, fits into a predictable Western modern “genealogy” where humans can be opposed to nonhumans and biology can be distinguished from geology. But if the nonhuman tirakuna are other than mountains—other than nonhuman, in my terms—then the Runakuna, are other than human as well. De la Cadena is pointing to a serious problem that arises out of my work. Does my ontological exploration, and especially the claims I make based on it, foreclose other modes of being? In short, does my ontological claim that, “life thinks; stones don’t” threaten to silence the tirakuna and the Runakuna? This is a serious concern, and one that I do not wish to brush off lightly.3

Before I address this directly, I first want to note something about genealogy. Throughout the commentary, de la Cadena seeks to identify my thinking as part and parcel of a particular “genealogy” (a Western one that separates humans from nonhumans), and I think this is indicative of an interesting flip-flop in the discipline. If in the past we were accused of ascribing culture only to others (“we,” after all, the thinking went, don’t have culture), now it is we the anthropologists who have genealogy (that is, our thinking is circumscribed by hidden sociocultural contexts), whereas the others, the Runakuna here, simply live with, say, earth-beings.

One major goal of my book, especially chapter one, is to argue for a vision of engagement with the world that recognizes the ways in which all sociocultural contexts are open. If I have a genealogy so do the Runakuna, but more interestingly, neither of our “genealogies” are closed. Both are open to the worlds beyond us with which we engage, and that can make us over. But, of course, to say this I do have to make the following ontological claim: we (humans) are the kinds of beings that have genealogies (or that live by means of socio-cultural contexts, or that think interms of conventional symbolic systems of reference), keeping in mind, however, that these contexts are what I call “open wholes”—that is, despite their closure, they are also open to worlds that are not made in such a fashion.4 The people with whom we work do not necessarily have to grapple with problems created by “our” genealogy of thought, even when they confront the pernicious effects of that genealogy (e.g., strip mining). But they do have to grapple with being the kind of being, which I call “human,” that lives in terms of symbolic context. This is the “difficulty of reality” (Diamond 2008) that we humans face and I think that sensitive ethnography will reveal this.

What I like so much about anthropology is that it provides us with a method (ethnography, broadly construed) for appreciating other forms of thought as well as how such thoughts are made over by the world in ways that can make us over as well. In How forests think, I have chosen to follow how the Runa are made over by an immersion in a dense living ecology whose properties necessarily become apparent in the ways in which they engage with it. The Runakuna of the Peruvian Andes will be made over by other elements of the world. And these will likely capacitate different properties of the world not so visible from my particular ethnography. As these continue to make their way into de la Cadena’s ethnographic thinking, they will keep challenging and expanding my own way of thinking.

I think that the beings in the forests around Ávila that are most analogous to the tirakuna are the spirit masters of the animals (the amuguna). I don’t of course want to conflate these two kinds of entities because each will bring out different ontolog-ical properties. But I would venture to say that each shares something in common: a certain kind of mortality. In Ávila, perhaps, and in other parts of the Ecuadorian Amazon, we are today witnessing the accelerated death of a kind. A dense relational ecology sustained in part by killing (via hunting) is quickly giving way to other economic forms. Some of these economic forms convert complex tropical ecologies into simpler ones (through logging, cattle raising, etc.), others simply seek to subtract the human relational aspect. In an attempt to stem the bush-meat trade and interethnic violence in the Amazon, the Ecuadorian government is stepping up its enforcement of the ban on the sale of artisanal firearms and ammunition. This may stop killing, but it may well result in the death of a kind; in fact, some shamans, it is said, have even seen the spirits dying in their aya huasca visions. It seems to me that the tirakuna face a similar threat, with grave consequences for the possibility of thinking otherwise.

And it is in the face of this threat that I find so helpful Bruno Latour’s lucid articulation of the stakes of the endeavor we share. At a historical moment in which, as he says, “everybody—ethnographers as well as former informants—are pulled deeper and deeper into the same ecological maelstrom,” what should we do? Latour envisions the ethnographic endeavor to be a diplomatic one. Our aim should be to find ways of moving among modes of existence such that certain modes, like those exhibited by thinking forests and the people that think with them, can continue to flourish. To the extent that How forests think is able to find an idiom to understand something about the thoughts of the forest without always passing through humans, it has, I think, for Latour, succeeded. And yet, as he notes, it faces several challenges, especially if we keep in mind that the end of the kind of anthropology we both feel our times demand is one in which our ontological explorations can be taken up in the actual struggles that, say, the Runa and the forests they think with must undertake as they negotiate “what a forest is made of” with other kinds of beings who think very differently about forests in ways that may not involve thinking with forests.

Inspired by Latour, I should note that I am currently extending my ethnographic exploration of sylvan thinking to understand its political life, today, in the face of ecological conflicts as they are making themselves manifest in Ecuador’s Upper Amazon at a moment in which the national government, led by Rafael Correa, both enshrines the “Rights of Nature” in its 2008 constitution at the same time that, under a policy that has been termed “neo-extractivist” (Gudynas 2009), is exhaustively exploiting the very resources this “nature” provides in the name of development. The pull between a nonhuman nature, the beneficiary of rights on par with human persons—a crude but potentially productive legal translation of the animacy that characterizes sylvan thinking—and a policy of exhaustively exploiting this nonhuman nature for human gain, is the fraught situation sylvan diplomacy will have to learn to face if thinking forests and the sylvan thinking that emerges with it are to survive and flourish.

I think that, at the risk of putting it too simply, Latour is asking whether the semiotics of the sort that I espouse is up to such a difficult diplomatic task. He wonders whether semiotics, as a “universal connector,” has the same reach of Actor Network Theory (ANT), and whether it is an appropriate diplomatic idiom. Should we really expect the Runa, and the loggers, oil workers, tourists, government officials, and media channels to adopt its jargon? There’s a reason why semiotics long fell out of style, remaining only as the butt of Woody Allen jokes and the nutty passion of a few fringe academics.

Regarding this concern, I want to be clear that semiotics cannot stand in for all modes of existence; it is limited to life. Although my equation of semiosis with life certainly creates the sorts of diplomatic problems that de la Cadena has raised and that I am still struggling to find ways to address, an anthropology beyond the human need not be confined to life or semiosis. This, as I mentioned, is the point I try to make in chapter five. In that chapter, titled “Form’s effortless efficacy,” I am concerned, not so much with expanding the scope of semiotics but with looking at how the emergent reality of a kind of semiosis beyond language opens on to a little-understood nonsemiotic and nonliving reality beyond it. This is why (in a manner that is unfortunate for the “marketing” of my ideas) I abandoned the label “anthropology of life” (Kohn 2007) in favor of the more open-ended “anthropology beyond the human.”

My call for an anthropology beyond the human does not involve taking some sort of theoretical schema—say my version of semiotics—and arguing for its universal application. Rather, it involves, as Latour puts it, experience. How is it that the worlds we attend to ethnographically make us over? De la Cadena’s engagements with tirakuna will make her over in ways that we have to confront conceptually. And it is this making over that interests me, because from it we can develop tools for the exploration of the realities through which we must move.

If there is, then, a “universal connector” in my work, it is not “semiotics” but rather “emergent reals,” of which semiosis is just one kind. That is, I am interested in the asymmetrical, directional, or nested qualities of so many of the world’s dynamics, whose distinctive emergent properties are hard to grasp when we adopt ANT’s symmetry, or anthropology’s complexity, or even, sensu stricto, the so called ontological turn’s plural worlds as the background axiomatic conditions for inquiry. It is because of my interest in directional processes that I like Latour’s use of the French “sens” as both meaning but also direction or way. To be fair, Latour is well aware of the limits of the symmetrical project (which is not to say that he endorses the kind of asymmetry I have in mind) and, as he himself has said, his Inquiry into the modes of existence (Latour 2013) suggests a way to come to terms with some of these limitations.

But if I am interested in something situated, as opposed to universal, and yet also other than human, it is the life of signs. Let me be clear that the life that signs take does have a generality to it: semiosis will exhibit certain formal properties, regardless of whether or how we name them, or in what galaxy and drawing on what contingent assemblage of materials it comes to be expressed. But this generality is a formal possibility not a universal property: not all entities are or will ever be sign processes.

And it is true that I am particularly interested in the life of signs on our planet. I want to understand the specific semiotic properties of life and how they may be harnessed diplomatically. In How forests think I use the academic idiom of semiotics to resolve academic problems. That is, our academic thinking is colonized by human language and I want to find ways to think beyond this. Accordingly, Latour would retitle my book “How to bypass talking humans” to emphasize that I am trying to develop an idiom that can provide a way of moving among biology, life, the Runa, and the forests with which they engage “without,” as he says, “stopping at” humans.

But thinking with sylvan thinking need not rely on an explicit semiotic discourse. I envision bringing a film camera to these forests and the fragile forms of sylvan diplomatic thinking that are emerging around them, and being made over by the sylvan logics I might thus attend to, in the way I imagine Lucien Castaing-Taylor and Verena Paravel were made over by the monstrosities they brought to the surface in the making of their film Leviathan (Castaing-Taylor and Paravel 2012). And I envision doing so without using one word of semiotic jargon. My book gestures to this in the uncaptioned photos that mark the beginning of each chapter and that provide a nondiscursive imagistic alternative path through a thinking forest. (I take it as the highest compliment that Latour considers How forests think a “work of art.”) But if I were to make such a film, the way sylvan thinking would make me over would nonetheless draw on the same properties that I describe more abstractly. And this is why I hold onto the technical language, and the ontological, for it is the specific describable properties of these realities that we need to harness in our times.

Bruno Latour and I, then, share a commitment to developing a kind of anthropology that can face the ecological crisis. But the question still remains: For whom is this crisis a maelstrom? If this is a threat we share with other kinds of beings, doesn’t this also mean that we share with them a reality (like the shared fact of our mortality, not only as individuals but as kinds)? And how should we square this reality we share on this single Earth with the plural worlds we are asked to envision?

I thank Philippe Descola for his generous and thoughtful commentary and for so lucidly tracing the senderos we share: the posthumanist route, and that much more local path, leading to the Ecuadorian Amazon and the “human and nonhuman denizens” that live there (for these are the inhabitants who have forced us both to question our assumptions about the human in the first place). I would, however, like to emphasize something else Descola and I share in common—something we share with that other Amazonianist, Claude Lévi-Strauss, despite the fact that my semiotics is neither structuralist nor poststructuralist. When Descola writes that the “sources of the plurality of beings and regimes of existence lie at a deeper level than the sociocultural one traditionally studied by anthropology,” he is, as he says, at some level characterizing the entire so-called ontological turn as a departure from social construction. And yet, in his intimation that there is something simpler that generates the apparent diversity around us, he is taking an unpopular position within this turn that I nonetheless share with him.

Descolas version of generative simplicity is of course his elegant four-fold account of the basic precultural set of stabilized ontological orientations one might take regarding others (Descola 2013). I have found this account quite useful in helping me understand the particular ways in which different modes of existence (Latour 2013) might relate to each other. Eduardo Viveiros de Castro has written at length about the particular way that multinaturalism might travel alongside and, in the process, subvert multiculturalism (Viveiros de Castro 2009), but Descolas scheme allows us to understand the dynamics peculiar to many other kinds of actual and possible encounters among modes of existence.

Nonetheless, the kind of simplification to which I am alluding is quite different from what Descola proposes. At one level, of course, I am claiming that the apparent diversity among human and nonhuman life forms can be understood by recourse to a series of simpler generative formal semiotic dynamics. But I am also arguing that generative simplicity is a property intrinsic to living semiosis itself. It is in this sense that I speak of forests in allusion to their generality. While it is true that I am indebted to the Ávila Runa, and the ecological specificity of the complex tropical forest they know so intimately, the choice of the plural forests in my title is deliberate, for I do not wish to lose sight of generality, redundancy, or simplicity. Just as my title is not How the Runa THINK forests think, it is also not How ONE forest thinks. Speaking about one particular forest makes it all too easy to give explanatory priority to the historically contingent circumstances that make that forest what it is—a form of analysis that can easily begin to look something like culture in the hands of the creative anthropologist. The anthropology beyond the human I envision seeks to be both grounded in ethnographic specificities and able to address the general, in part by taking generality itself as its own kind of ethnographic object.

And this brings me specifically to the sort of semiosis I have in mind. In defining in one instance a self as that which has “a point of view” (Kohn 2013: 132), my goal is to find a certain common ground with perspectivism without necessarily adopting the “cannibal metaphysics” (Viveiros de Castro 2009) to which this Amazonian style of thought has given expression. For the record, perspectivism is an ethnographic reality in Ávila, as the case in question involving dogs and their dreams, as well as many other examples from my book, demonstrate. But the conceptual work I do with perspectivism takes me along a different path than the one that is so provocatively explored by Viveiros de Castro. Instead of looking at the sort of sylvan thinking I saw in and around Ávila as a style of Amerindian thought that, when stabilized analytically as multinaturalism, can be used as a basis for critique of our own form of thinking (whose problematic properties are visible once we recognize it, by contrast, as multiculturalism), I have chosen to stay closer to sylvan thinking’s “venatic” foundations.

Viveiros de Castro’s predatory thinking, and Descola’s animism (which, I am quite aware, are not the same thing) allude to but do not explore their resonances with the basic animacy of life that hunting brings to the fore. The Ávila Runa or the Achuar, whose form of economy depend on entering and harnessing a vast nonhuman communicative ecology, cannot avoid the basic fact of this animacy that tropical forests make so manifest. By saying that all selves have a point of view, I am, perhaps problematically, exploiting a bit of linguistic slipperiness to draw attention to the continuity that my position (namely, that all semiotic selves are loci of interpretance5) has with the Amazonian perspectival one (that all persons, human and nonhuman, have and can share a perspective). But this should not be taken to mean that I am dong conceptual work with the perspectival specification that it is the body that accounts for the perspective. This, in part, is because I don’t think of representation as exclusively mental. Although I agree with this treatment of the body as a generalized ethnographic observation (e.g., in Ávila it is the vulture’s body that allows it to “see” rotting meat as savory manioc), my statement that selves have points of view should not be taken to imply that semiosis becomes superfluous.

That is, I am not adopting the phenomenological position that would hold that since the point of view is in the body we no longer need representation to understand relationality. And it is for this reason that I discuss von Uexküll’s tick in the way I do. One could provide a “nonrepresentational” account of the role of, say, butyric acid, in structuring relations among kinds of beings, but, by doing so, we would miss the semiotic “absential” logic (Deacon 2012) that is so central to these relations. That is, it is not enough to note that ticks are affected by the smell of butyric acid emanating from their various hosts, for one must consider the generative possibilities inherent to the fact that in this process there are differences among their hosts that the ticks don’t notice. It is this generative simplifying confusion based on an absence—that which is not noticed—that makes these hosts a kind, the kind of being, for example, through which Lyme Disease can pass (Kohn 2013: 84–85).

In Jorge Luis Borges’ (1988: 490) terms, “thinking is forgetting differences” (“pensar es olvidar diferencias”). And it is the specifically semiotic and generative absential logic that the tick exemplifies in its form of forgetful thinking that is obscured in the move to treat anything that “does something” as an agent. I agree that an anthropology beyond the human cannot just be a semiotic one (and I appreciate that my emphasis on semiosis seems to close down certain possibilities for attending ethnographically to some kinds of nonhumans and their relations with humans) but any anthropology beyond the human, as it finds ways to open itself to the “theater of worldly interactions,” will still have to be true to the semiotic properties that pop up in certain parts of the world and not in others (for part of the problem with semiotics is that we apply it in places where it does not belong).

I agree with Descola that my approach would be richer if I could open my semiotics more fully to ethology. In my megalomania I have considered doing this (in fact I have an elaborate, but, alas, unsuccessful, grant proposal, sitting in a Drop-box folder, out there somewhere on the password protected cloud, that proposes the very project Descola would like to see). But what has stopped me from doing this, aside from my failures with funding agencies, and the growing realization of my limits, is the recognition that although such a study will tell us all sorts of things about what semiotic beings are “saying” to themselves and to others (please excuse the linguistic metaphor), it will likely tell us less that is new about how they “say” it.

For the moment, and given the fact that this fragile form of sylvan thinking is under threat—if the forests are “saying” something, I think it may be this—I want to find ways to continue to cultivate sylvan thinking as part of the ethical practice our times demand. This involves looking ethnographically to moments where such sylvan thinking enters both as object and vehicle of political thought in contexts where its survival is at stake, and it also involves learning to cultivate this form of sylvan thinking in our everyday lives. For these reasons, I am tremendously grateful to the HAU editors and contributors for giving sylvan thinking this forum through which, in conversation, it might continue to flourish.


I wish to thank the students in my Fall 2014 Anthropology Beyond the Human seminar for very productive discussions related to this response.


Borges, Jorge Luis 1989. Obras completas. Vol. 1. Buenos Aires: Emecé Editores.

Castaing-Taylor, Lucien, and Véréna Paravel. 2012. Leviathan. Videorecording. New York: The Cinema Guild.

Crist, Eileen 2004. “Can an insect speak? The case of the honeybee dance language.” Social Studies of Science 34: 7–43.

Deacon, Terrence William 2012. Incomplete nature: How mind emerged from matter. New York: W.W. Norton and Co.

Descola, Philippe 2013. Beyond nature and culture. Chicago: University of Chicago Press.

Diamond, Cora 2008. “The difficulty of reality and the difficulty of philosophy.” In Philosophy and animal life, by Stanley Cavell, Cora Diamond, John McDowell, Ian Hacking, and Cary Wolfe, 43–89. New York: Columbia University Press.

Gudynas, Eduardo 2009. “Diez tesis urgentes sobre el nuevo extractivismo: Contextos y demandas bajo el progresismo sudamericano actual.” In Extractivismo, política y socie-dad. Quito: Centro Andino de Acción Popular y Centro Latino Americano de Ecología Social.

Hage, Ghassan 2012. “Critical anthropological thought and the radical political imaginary today.” Critique of Anthropology 32 (3): 285–308.

Hornborg, Alf 2001. “Vital signs: An ecosemiotic perspective on the human ecology of Amazonia.” Sign Systems Studies 1: 121–52.

Kohn, Eduardo 2007. “How dogs dream: Amazonian natures and the politics of transspe-cies engagement.” American Ethnologist 34 (1): 3.

———. 2013. How forests think: Toward an anthropology beyond the human. Berkeley: University of California Press.

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

Paton, Ray, and Koichiro Matsuno. 1998. “Some common themes for enzymes and verbs.” Acta Biotheoretica 46 (2): 131.

Pattee, Howard H. 2001. “The physics of symbols: Bridging the epistemic cut.” Biosystems 60: 5–21.

Tsing, Anna Lowenhaupt 2014. “Blasted landscapes (and the gentle arts of mushroom picking).” In The multispecies salon, edited by Eben Kirksey, 87–110. Durham, NC: Duke University Press.

Viveiros de Castro, Eduardo 2009. Métaphysiques cannibales: Lignes d’anthropologie poststructurale. Paris: Presses universitaires de France.


Eduardo Kohn
Department of Anthropology
McGill University
855 Sherbrooks Street West
Leacock Room 718
Montreal, QC H3A 2T7, Canada


1. I am not the only Amazonianist to think about the forests and its inhabitants semioti-cally (see Hornborg 2001).

2. Apart from the fact that I am simply drawn to tropical forests, there are specific reasons why I choose them as a site for this kind of exploration beyond the human. Thoughts beyond the human can be found everywhere there is life: in us, in gardens, fallows, and plantations, in cities, and, most importantly, for the kinds of ethical practices we need to develop, in what Anna Tsing (2014) calls “blasted landscapes.” Sylvan thinking, however, although copresent with life wherever it is found, is easier to see in tropical forests. This is because the many layers of nonhuman life in such forests amplify and thus make apparent its strange logics.

3. I should also note that it is a concern that others, such as Elizabeth Povinelli and Hugh Raffles, immersed in “geo-otic” ethnographic worlds, in personal conversation, have raised with me.

4. I demonstrate this ethnographically and formally in great detail in chapter one.

5. I should note here that interpretance need not require a conscious interpreter.