Title: At long last
Author: David Graeber
Date: 2017
Source: Foreword to The Fire of the Jaguar by Terence S. Turner (HAU Books edition, 2017), pp. xix-xl
Notes: Available online at haubooks.org (open access).

For anyone in the Chicago anthropology department in the 1970s,’80s, and ’90s, The fire of the jaguar holds a legendary status. I mean this in the almost literal sense: it was wondrous; it had strange and awesome powers; no one was entirely sure if it really existed. Terry refused to publish it. Or even to show it around. Yet the very fact of its hiddenness made it a kind of talisman of secret potency.

Terry had a peculiar aversion to publishing. There were rumored to be anywhere between three and half a dozen brilliant monographs in his closet, all of them effectively finished, all in a kind of permanent state of final revision.[1] There were many stories as to where this aversion to publishing came from. At Cornell—again, I am repeating the legend here—he had been a close personal friend of his namesake Victor Turner, even though in many ways the two could hardly be more different theoretically, and they had a kind of understanding that they wouldn’t stray too far from one another. When the University of Chicago offered Terry a job as assistant professor in 1968, he said he’d only come if Victor accepted his offer too; they both arrived, and Terry quickly won tenure there on the basis of what was to be his first monograph, hailed by his colleagues as a brilliant work which proposed an entirely new approach to structuralism and the interpretation of myth. This was The fire of the jaguar, and the book had already been accepted and existed in galley form when he submitted it to tenure review. The moment he actually received tenure, he withdrew it from publication. Ever since, the story went, he had been tinkering away at perfecting it, along with anywhere from three to half a dozen other books (it varied with the narrator) he was rumored to have somewhere in his closet, all of them not quite ready for publication.

People used to beg him to just release the books. He always found some reason not to.

Terry’s lectures were mesmerizing. He appeared to have an absolute mastery of social theory, to have read everything there was to read, and—almost uniquely among those with that kind of comprehensive knowledge—whatever the topic, also had something startling and creative to say about it. He had an uncanny ability to listen to another anthropologist deliver a ninety-minute paper, then stand up afterward and say, “That’s an interesting interpretation. But you know, you could equally well see that material from another point of view . . .” and then proceed to take every single ethnographic detail the paper contained and reorganize it into a grand synthesis that seemed—and I’m pretty sure in most cases usually was—ten times more theoretically sophisticated than the presenter’s own.

Needless to say, a lot of people hated him.

He was also notoriously contentious.


I used to say it sometimes seemed as if Terry had spent twenty years coming up with a theoretical synthesis that resolved all outstanding problems in social theory, and now he was going to have to spend another twenty years trying to figure out how to explain it to anyone else. At least, how to explain it in writing. I remember being quite impressed (in a horrified sort of way) when I first encountered two of his essays as an undergraduate. There were plenty of anthropologists who could write sentences I didn’t understand a word of; I knew of a few who could write incomprehensible paragraphs; but here, uniquely, was one who could write entire pages where I simply had no idea what was going on at any point. Therefore, it was all the more startling when I met the man, began taking his classes, and found in person he had a remarkable ability to make the exact same (still extremely complicated) ideas sound like matter-of-fact common sense, and even to render them fairly straightforward. It was putting it on the page that seemed to be an issue. I well remember one seminar when he was explaining an idea—I think it was about polyphony—and a student asked if there was anything more on the subject she could read. “Well, I wrote a paper a few years ago,”Terry said,“but to be honest, it’s a little rough going. I was looking over it the other day and even I couldn’t figure out half of what I was saying.” Terry was occasionally accused of being “Parsonian.” This is a slander: really he took only one idea from Talcott Parsons, that of a generalized symbolic medium; in almost every other respect his approach was the exact opposite. However, he does seem to have absorbed something of Parsons’ impenetrable prose style.

He tried to fight it. These essays, largely unpublished in his lifetime, might be seen as the products of a struggle to render his ideas transparent. He reworked some of them again and again. He did publish quite a number of essays, some for edited volumes, others when friends took over journals and compelled him, but mainly when he felt it would make a political difference, either in Brazil, or, particularly, for the Kayapó. (Thus, from the ‘90s onwards, he was much better known as a writer on indigenous video activism than as a social theorist.) The majority of his most important theoretical essays were never published, but only shared with friends, students, and colleagues—including a few which acquired a legendary status in their own right, like his magnificent 1984 essay, “Value, production, and exploitation in noncapitalist societies”—and floated about, sometimes in multiple versions. At the time, it was possible to place unpublished papers on reserve as course readings at the Regenstein Library at Chicago, and there they’d remain afterward in special file cabinets until the professor found out and usually had them instantly removed and destroyed.[2] Some of us would copy them at the time; others such as myself worked in the library and knew about the file cabinets. As a result, different versions of some of Terry’s unpublished theoretical interventions would sometimes circulate, often in copy-of-a-copy-of-a-copy form, invariably with handwritten headers by the author saying things like “draft: for god’s sake do not quote.” Later they were pdf ’d and exchanged by email. Everyone had their own collection.

These essays did have an impact on the discipline. I am speaking not just of my own work. My first published monograph (the second one I actually wrote), Toward an anthropological theory of value, was largely inspired by Terry’s ideas and, I will now admit, was written with half an eye to coaxing him out—I thought if he saw his theories expressed in another anthropologist’s words, he would immediately say something to the effect of “the fool, the fool, he got it all wrong!” and, as a result, some of the unpublished texts would actually see the light of day.

It didn’t work.

His lectures and published and unpublished essays did, certainly, have a profound effect on anthropologists of many generations—one thinks here of anyone from Dominic Boyer to Michael Cepek, Jane Fajans, Jonathan Hill, David Holmberg, Nancy Munn, Fred Myers, Sasha Newell, Suzanne Oakdale, Stuart Rockefeller, Stephen Sangren, or Hylton White. (Some of them, of course, were just as much an influence on him.) But at the same time, the core concepts have really not become the common coin of the realm in the way many of us felt they should; the overwhelming majority of anthropological theorists active today, in fact, have barely heard of Terry.


The fire of the jaguar is Terry’s most sustained attempt to carry out the structural analysis of a single myth. It may well be the most sustained and detailed analysis of a single myth that any anthropologist has ever carried out. Obviously, any anthropologist dealing with Amazonian mythology must be at least in tacit dialogue with the work of Claude Lévi-Strauss, and, for Terry, this was very explicitly the case. To put it bluntly, Terry felt that Lévi-Strauss had set off from a brilliant set of insights on a project that could hardly be more important for social theory and then went completely off the rails.

What follows is my own take on the matter, but very much inspired by Terry’s (I was, after all, his student.)


Much of Lévi-Strauss’ later work can be seen as a cautionary tale of the effects of extreme hierarchical social arrangements on human thought. The French academy is structured in such a way that there is typically one man (at least, it is almost always a man) on top of the field in any given discipline. Lévi-Strauss became the king of the anthropologists[3] and, while of a modest and unassuming character personally, was entirely comfortable with this role.[4] As a result, in the second part of his career, he remained largely unchallenged by alternative perspectives, which allowed a brilliant creative mind to devote most of its intellectual life to working out the equivalent of crossword puzzles. Contrast here the startling insight of his early essays with the four massive volumes of Mythologiques. While the latter has proved a delight to fellow Amazonianists, other scholars have labored in vain to find a point in them. By detaching myths from social life and rendering them into a series of formal elements, he could rearrange those elements in an endless variety of fascinating patterns, but did anyone learn a single thing of interest to humanity by the process of doing so? Mainly we learned that there was a very powerful French professor who claimed to despise the cult of individualism and creativity, but demanded an individual monopoly of all creative production so he could indulge the fantasy of being engaged in an ongoing dialogue with primitive philosophers on topics of interest largely to himself.

The result of this massive intellectual self-indulgence was predictable: a frenzied cult of personality and attempts to decipher the true meanings of the master’s oracular pronouncements, along with the usual arguments abroad about who was the truest disciple, followed by the inevitable ritual abjuration. The entire project of structuralism was tossed out the window except insofar, of course, as its replacement (“poststructuralism”) was in most important ways exactly the same thing.

I know I am being unnecessarily harsh: Lévi-Strauss was kind and encouraging to his students and can hardly be held personally responsible for either the structure of French academia, or the fate of a movement that included everyone from Jacques Lacan to Pierre Vernant or Edmund Leach. It is, rather, written out of a sense of frustration with what might have been. Terry represented an unrealized alternative form of anthropological structuralism that never quite came into being. Like Lévi-Strauss an Amazonianist, he made himself in many ways his exact structural inversion. Perhaps we can best see this by using a classic Rodney Needham-style binary table:

Claude Lévi-Strauss Terry Turner
painfully effete gleefully embraces manners of common man
delicate athletic
politically conservative politically radical
static models dynamic models
academically all-powerful academically marginal
endlessly prolific never published a book

The power of the structuralist approach is that it provides a uniform set of tools that can allow one to at least begin to put apparently disparate aspects of human culture—kinship and social organization, myths and rituals, economics, poetics, and so forth—on the same conceptual table, as it were, so that each can provide insight into the other. This holism was always part of the special promise of anthropology, and it cannot be denied that its loss would empty the discipline of much of its raison d’être. If we can’t say that it’s impossible to understand forms of musical improvisation on a Greek island without also understanding the structure of their cheese making, courtship rituals, or knife fights, then we might as well throw in the towel and just become sociologists. Since poststructuralism, as I note, actually is a form of structuralism, this has not been entirely lost—but it has certainly been endangered in some quarters, and there has been a noticeable tendency within the discipline to fragment back into subfields.


Lévi-Straussian structuralism never quite answered this promise—or not in the hands of the Master himself. Lévi-Strauss did not, in fact, end up using his techniques to compare different domains of the same social or cultural orders, to come up with the kind of holistic analysis the Boasians, for instance, had always dreamed of but never figured out quite how to produce—or at least he never did so systematically. His interests lay elsewhere. Partly as a result, the structuralist project largely fizzled out, only to be replaced by a poststructuralism that, rather than resolving any of these dilemmas, effectively abandoned them. Poststructuralism, as the discipline knows it now, largely through the works of Deleuze and Foucault, took aim largely at the very ability to render elements comparable, to put them on the same table—or even, really, to say there was a table in the first place. To put the matter bluntly, while Deleuze, its main theoretical avatar, rejected the static models typical of classical structuralism and insisted that he was working in the dynamic, Heraclitean ontological tradition rather than the static, Parmenidean one favored by almost all analytic and most Continental philosophers, his primary philosophical project appears to have been to preserve its core insight (that objects are processes, that individuals are sets of relations . . .) while absolutely rejecting every aspect of the work of the one man most identified with it—Hegel. In the context of the French intellectual left of the late ’60s, it’s easy to see why Hegel would become the particular object of ire and disdain. At the time, it seemed as if all radical thought was trapped between Kojève-inspired master–slave dialectics (whether in its Lacanian or existentialist variety) or some form of slightly more or slightly less dogmatic Marxism. This had become depressing fare. And the political implications were dire.

Deleuze worked his way through almost every available alternative Heraclitean tradition, from Spinoza and Nietzsche to Bergson and (at least tacitly) Whitehead, in order to create his own anti-Hegelian synthesis. It is not at all clear, however, that he succeeded. Obviously he succeeded magnificently in setting the intellectual agenda for fellow academics in the years to come, at least in anglophone countries—most “social theorists” in the United States or the United Kingdom, for example, are familiar with the ideas of European philosophers like Spinoza, Leibniz, Bergson, and many others almost exclusively through Deleuze, and many seem unaware that Deleuze did not invent them. In fact, his political success within academia is so complete that I rather feel like writing what I am about write counts as minor heresy. But let me say it anyway.

The key objection to Hegelian dialectics in Deleuze, but increasingly on the part of almost all French thinkers who came to be identified with “’68 thought,” was twofold. First of all, Hegel’s emphasis on negation, or, in structuralist terms, binary opposition, was seen as denying the real complexity of the play of positive forces that constitutes natural, social, or human life. We are not really talking about subject/object, self/other, nature/culture, and so on—all this is reductionism; we are talking about degrees of pressure, gravitational fields, converging and contradictory flows of matter and energy. Second of all, the notion of subsumption, of the maintenance of the dynamic tension between any such opposition (subject/object, self/other, nature/culture, etc.) as the subordinate moment in a higher synthesis, which could then be part of a further opposition and further synthesis, was denounced as leading inexorably to authoritarian outcomes. Again, it’s not surprising that, in the context of the ’60s Left Bank, radical theorists should have thought this. Subsumption is a hierarchical notion, and it had been put to hierarchical uses: whether by Hegel, to posit the nation as a higher subject encompassing the various contradictions of the classes and factions that make it up, or by various communist parties, to pose themselves as the revolutionary subject. However, the question was how to ditch all this baggage and still retain the key insight, which is that subjects, or objects, are in no sense fixed substances but are really just particular perspectives on processes of action.

I know I’ll likely lose some friends by saying this, but, honestly, I don’t think Deleuze really pulls it off. The advantage of a dialectical approach is that it not only allows one to see what seem to be objects (“forms”) as being composed, on another level, of elements in dynamic tension with one another (their “content”), but it also allows us to realize that, on a different level, those forms are themselves the dynamic content of some higher level of organization or form, and so on. We are all made up of atoms that have a constant patterned motion we know as “matter” (form), but, on another level, we are all ourselves atoms that have dynamic relations with each other that make up something even more concrete—say, a social system. And so forth.

The problem, of course, is that the result is a series of hierarchical layers, with higher and higher forms, where all contradictions would appear to be eventually subsumed and overcome. This not only has disturbing political implications, but it doesn’t correspond to what life is actually like. Contradictions and tensions are not really overcome. To the contrary, the world seems rather a mess. Obviously you can look at the degree to which they do seem to be overcome and say, “Well, that’s the structure,” but then the word “structure” no longer tells you very much—it just means “that tiny portion of reality that seems to make some sort of sense.” Alternately, you can say matters are still in the process of working themselves out. To put this in more formal language: you can posit the results as a formal logical system, but, in that case, there is some ultimate equilibrium where everything is coordinated by the highest level, which is a very conservative perspective with little explanatory power. Or you can, like Hegel in the Phenomenology, or Marx, see the dialectic as a historical progression, with a resolution perhaps to come in some redemptive future. Both have unfortunate political histories, and it’s not surprising that, after May ’68, intellectual rebels were beginning to think about how to move away from them.

Still, it seems to me, all the poststructural rejection of this logic of subsumption really ends up doing, in most cases, is to divide the static forms and the dynamic content into two camps and set them at war with one another. Myself, I just can’t see this is an improvement. Certainly, in the hands of masters like Deleuze and Guattari, the results are always provocative and extremely sophisticated—so much so it allows professional academics in 2017 to propound on concepts that have been circulating for half a century and still feel they’re doing something vaguely naughty. But in the final analysis, it always comes down to the same thing: whether it’s the juxtaposition of open-ended, free-flowing, polymorphous “desire” versus the fixed form of the Oedipal triangle, the dynamic “war machine” versus the bureaucratic state, or rhizomes versus trees, its end result is a rather New-Agey opposition between (good) dynamic energy and (bad) constraining structures. Foucault (who disliked the way Deleuze and Guattari framed desire in Anti-Oedipus for this reason) tried to overcome the tendency to dichotomization by declaring that everything was power and hence dynamic, but this didn’t really solve the problem, since it left him no cogent way to say power was objectionable, and anyway, the bad constraining forms still lingered in his analyses, just pushed into the background, like all those walls and guns and truncheons keeping the prisoners from fleeing the Panopticon.


Terry Turner’s theoretical corpus can be read as an attempt to overcome such predicaments. To do so, he looked to a different, dialectical variation of structuralism for a way to think his way out of this dilemma. We see it as received wisdom now that structuralism means privileging the synchronic “code” over diachronic process. It resembles dialectical thought in that it sees relations as intrinsic and constituting—it’s not as if there are already-existing objects that then come into relationship in one way and not another; these objects are the relations they have with one another—but structuralism departs from it in that it does not see the play of those relations as a dynamic process with the potential of generating higher totalities that can then themselves enter into relations with one another, and so forth. It is, as Bruno Latour (2007) was later to put it in an only slightly different context, a “flat ontology.”

For a Hegelian, this would have meant structuralism was, quite literally, meaningless. Hegel once remarked that reducing everything to equations essentially means reducing everything to tautologies, since all equations can be ultimately reduced to a simple statement that A = A. We already know that A = A. If you want to say something you don’t already know—that is, if you want to begin to think—you have to look at the degree to which terms are not self-identical and thus break out of the level where A = A and generate a higher one. And Turner would entirely agree that structuralism is, in that Hegelian sense, meaningless. In fact, Lévi-Strauss would occasionally admit this too: he was not interested, he said, in questions of meaning, in the classic hermeneutic sense, where meaning is the message that some author or speaker is trying to convey, the intention lying behind a statement. He was interested in langue, not parole; language, not speech; and intentionality, therefore meaning, fell into the latter category. His work was to look at the elements that made meaning possible. Other people could worry themselves with trying to figure out what a given author or text was trying to say.


So Turner’s project was first of all to reinsert meaning—intentional action— into the equation. Which meant to go beyond just equations. He tried to create a different structuralism, which fused together the German tradition, wherein the basic units of analysis are actions, and the insights of classical French structuralism, about working out the possible formal permutations of a set of logical terms (raw/cooked, left/right, matrilateral/patrilateral, etc.). In order to do this, he traced a different theoretical genealogy, originating in Hegel’s Logic (rather than his Phenomenology), proceeding through Marx’s Capital (more than, say, his historical or ethnographic works), and culminating in Jean Piaget’s Genetic epistemology.


Now, the importance of Piaget here cannot be understated, so it’s worthwhile to dwell on it a moment, since his presence might otherwise seem odd. Nowadays, Piaget is remembered as a theorist of child development and one who, however significant his ideas to mid–twentieth-century thought, is now considered somewhat passé, since he tended to downplay both the existence of innate structures of the mind and cultural variation. As a result, he might seem an unlikely savior for anthropological theory. For Turner, though, what was important about Piaget’s work was much less the particular stages of moral or intellectual development he came up with but, rather, the way he went about it and what he thought those stages and structures in general ultimately were. In a way, Hegel’s Logic and Piaget’s Genetic epistemology are very similar books: they are both meant to demonstrate how, even if one starts from nothing else, no presuppositions whatsoever other than an acting subject confronting the universe, it would still be possible to generate all the most sophisticated categories of human thought simply by their interaction. Abstractions arise from the way that we are forced to reflect on the process of our interactions; these allow more sophisticated interactions; those more sophisticated interactions, in turn, allow more sophisticated reflections, and so forth. In the course of describing the process, Piaget manages to develop a genuinely dynamic version of structuralism. This is the model Turner adopts.


What makes Piaget’s structuralism so different from the Lévi-Straussian variety is that the elements that are organized into more and more complex structures, the “content,” as it were, are not ideas or objects but actions. We may imagine that we start with an abstract set of numerals, 1, 2, 3, and so on, and then start adding and subtracting them, but, in reality, numbers do not exist outside the process of counting, adding, subtracting, and so on. Just as no action can take place without thought, all thought is an element in some schema of action. So the materials being organized in a structure are always “operations,” conscious or potentially conscious attempts to transform the world in some way. So whereas in classical structuralism, everything ultimately comes down to a tautological equation, in dynamic structuralism, even equations are really actions. A “structure,” it follows, is a way a particular group of actions coordinate with one another. Hence, structures are forms of “self-regulation” or “self-organization.” Nowadays, most social theorists seem to think the latter term is derived mainly from complexity and chaos theory, but, in fact, in the ’60s, when Piaget was writing, it had already emerged from cybernetics, and while the principle was only beginning to be applied in the natural sciences, it was already the object of experimental applications by social scientists with training in the natural sciences, such as Gregory Bateson or Piaget himself.

Few of these experiments ended up leading to full-blown social theories, because, by the time ideas like self-organization did become dominant in the natural sciences—and they only really began to take off in the ’70s—the most creative branches of anglophone social science, at least, had largely abandoned the idea that they were engaged in science of any kind at all. Social scientists had already begun to redub themselves “social theorists,” drawing largely on Continental philosophers for inspiration and ignoring developments in science (which they increasingly characterized as if it were still stuck in nineteenth-century positivism, so as better to dismiss it.)

So the potential opening of the ’60s was not pursued.


Self-organization sounds like the sort of notion that would be embraced enthusiastically by radical social theorists, and there are occasional, if usually rather wistful, calls to do so. But nothing much ever seems to come of it. The main reason, I suspect, is that the notion of self-organization is inextricably bound up with notions of totality as well as of hierarchy. Both terms immediately raise the suspicions of anyone with antiauthoritarian instincts—who are, of course, precisely those who would otherwise be most attracted to the notion that structures can regulate themselves.

A self-organizing structure has to be a totality with respect to its own self-organization. There may be all sorts of overlapping and contrasting totalities operative in different situations or even in the same one, but to understand something as a structure means to understand it as a whole that is larger than the sum of its parts. You can’t have self-regulation without a self. But that also means a hierarchy between a higher level of “invariants” that coordinate the transformations and a lower level of the transformations themselves. Usually, it means a hierarchy of a whole series of levels in which that invariant structure becomes a mere dynamic element (“abstract content”) in a larger structure, and so forth. The existence of logical hierarchies of this sort in no sense implies the existence of social hierarchies; but one reason I think left-wing scholars have avoided this kind of thinking is the assumption that on some level, one must imply the other. This idea is promulgated on the right, where conservatives like Louis Dumont have had remarkable success in convincing their fellow anthropologists that all conceptual systems imply the superiority of some terms (and hence some people) over others, and on the left, where “hierarchies” of any sort are often treated as equally objectionable. The two positions play off one another, with the typical result (I’ve seen this) a veering back and forth from a kind of extreme poststructural rejection even of spontaneous self-regulating order and a resigned acceptance that even social hierarchies (say, the elaborate administrative chains of command in contemporary universities) are probably inevitable after all.


Piaget agreed with Lévi-Strauss (who, at least in the early part of his career, also drew on scientific models) in seeing structures as, to quote Turner, “groups of transformations bounded by invariant constraints” (p. 209, this volume)—the invariants being the rules that govern the arrangement and rearrangement of the elements. But where Lévi-Strauss was content to see those rules as givens, part of the elementary structures of the human mind, Piaget, who started from action, could not. As a result, as he put it, “the idea of structure as a system of transformations becomes continuous with that of construction as continual formation” (Piaget 1970: 34, original emphasis)—the structure is always building itself, and, as soon as it seems to have reached the top, it always must necessarily create an even higher degree of coordination of which the actors cannot be entirely conscious, because it is the self-regulating mechanism that’s making it possible for them to think about such questions in the first place:

Gödel showed that the construction of a demonstrably consistent relatively rich theory requires not simply an “analysis” of its “presuppositions,” but the construction of the next “higher” theory! . . . The pyramid of knowledge no longer rests on foundations but hangs by its vertex, an ideal point never reached and, more curious, constantly rising! In short, rather than envisaging human knowledge as a pyramid or building of some sort, we should think of it as a spiral the radius of whose turns increases as the spiral rises. (Piaget 1970: 34)

This is why we’re not dealing with some kind of authoritarian, closed system here. Structures are always open. But critically, they are always open at the top. Even those who think they’re operating at the very top of a conceptual (or social) system cannot, by definition, completely understand what they’re really up to. Turner supplemented Piaget’s insights in this regard with those of Soviet developmental psychologist and educational theorist Lev Vygotsky’s notion of “proximal level of development”—that is, that all of us are always necessary operating on one level of sophistication higher than we can consciously explain. This is why, for instance, it is possible to speak in grammatical English sentences even if one is completely incapable of explaining the difference between a past participle and a gerund, or even never actually heard that past participles or gerunds are things that are supposed to exist. It’s obvious why such approaches should be of interest to anthropologists, because, in a way, this is the key question in any cultural analysis. How do people operate with tacit codes of which they are not consciously aware? Structuralism just makes this problem explicit. Even if we are able to demonstrate that a Greek musical performance or courtship ritual is really an exact inversion of the symbolic code on display in a typical knife fight, one still has to eventually get to the question of where this code actually resides. Is it somewhere in the actors’ heads, some unconscious level of the mind? Would that be an individual or collective unconscious? Is it inscribed in the architecture, as it were, so that people absorb the tacit categories and associations by which they live—hot/ cold, wet/dry, high/low, male/female—simply by moving about in culturally appropriate ways through the physical environment? Or is it somehow implicit in their language?

The solution proposed in “The fire of the jaguar”—and the other essays collected in this book—is not just to see structure as emergent from action, as the forms in which action self-organizes, but to see what we call “mythic thought” as the way that the highest level of self-organization appears, as it were, from below. A very simple example might suffice. The moment one does the same thing twice—say, gives food to a child—that is, the moment one not only performs a specific action again, but does so with the understanding that it is “the same” action as one has performed before, one generates, through the repetition (of an action that, like any, has both material and mental dimensions), a kind of hierarchy, since there is a more abstract level at which those actions are both tokens of the same type. But the moment one says a different kind of repeated action is not the same—say, giving food to husband or to a rival at a competitive feast—one is generating a third level, where different types are being compared. At the same time, by defining certain types of action in this way, one is typically generating certain identities (child, husband, rival), kinds of person who typically perform or are the objects of such actions (a nurse and patient, a dishwasher, a heavy drinker, a student, and so forth). This isn’t just a matter of abstract reflection, it’s practical. There has to be a way of arguing about who is a heavy drinker and who isn’t; who’s a real husband or a real child; there have to be ceremonies for matriculation as a student or qualification as a nurse. This brings us into the domain of ritual, since, at least for the really important categories, this is how such transitions are effected. But as anthropologists have long noted, rites of passage, where one passes from one status to another (“status” here defined as a person seen as typically performing or who is allowed to perform certain kinds of action), have a peculiar quality: even if they mark the transitioning from child to adult, there is always a stage in between, where all the usual distinctions (boy/man, girl/ woman, alive/dead, inside/outside, freedom/authority) seem to be thrown into complete disarray, all social rules suspended . . . For Victor Turner, this was a moment of “antistructure.” For Terry Turner, in contrast, it is “metastructure”—this is simply what the proximal level of development, that level which we can never completely understand (at least, without creating a new level which we also won’t be able to completely understand), will always look like. The effect is the same as it would be if two-dimensional creatures were staring at a three-dimensional object; some aspects will simply not make sense. But in this case, even if they could enter into a 3D world, they would be immediately confronted by the fourth-dimensional objects that had allowed them to do so, and so on . . .

This is exactly why myths (such as the fire of the jaguar) so often deal with origins of social institutions. It is easy to understand arranging a marriage or conducting a wedding ceremony as simply something people do. These are human actions that the people involved chose to do the way they did and could have decided to do otherwise. But in arranging marriages in the same way over and over, those same people are also continually re-creating the institution of marriage—which, after all, only really exists as the form of those actions’ self-regulation. Yet once again, it is almost impossible to keep track of this level of social reality—and, of course, the authoritative effect of the ritual largely depends on the fact that we generally don’t. This is why institutions like marriage, chiefship, or the culinary arts are typically said to originate from creative acts not now, but in a one-time mythic past, what Mircea Eliade referred to as the illo tempore, a time of creation characterized by an apparently random kaleidoscopic collection of subject/object inversions, talking animals, and strange powers, in which the social and natural laws we know today appear to have been almost entirely suspended. This is, again, what the ever-disappearing top of the pyramid looks like from below.


The essays collected here are all in one way or another about myth, and one can see them as Turner’s unique effort to come up with a radical—in the sense of politically left-wing—theory of mythology. It is interesting to reflect on the fact that, as academic subjects go, the study of myth has been overwhelmingly dominated by conservatives. The great triumvirate that dominated theory about myth in the mid– to late–twentieth century, C. G. Jung, Mircea Eliade, and Joseph Campbell, all considered themselves right of center in one way or another: Jung was a Burkean; Campbell considered himself a free-market libertarian; and about Eliade, who was a member of the Iron Guard in his youth, probably the less said the better. Georges Dumézil was close to the Nazi party, and the only left-wing theorist who fully embraced the power of myth as a means of revolutionary struggle, Georges Sorel, ended his life an admirer of Mussolini. Lévi-Strauss was an “apolitical” conservative pessimist. There are a handful of exceptions, from feminists like Jane Harrison, to anti-fascists like Karl Kerenyi, to leftist structuralists like Jean-Pierre Vernant and Pierre Vidal-Naquet, but, from the days of William Blake and Percy Bysshe and Mary Shelley to those of Robert Graves, left-wingers entranced by the power of myth have been far more likely to put their hands to creating new myths than interpreting old ones.

I suspect there are good reasons for this. If left-wing thought, whether in its romantic or Marxist variants, has always been a celebration of creativity, then myth poses it a problem. Mythic thought is endlessly creative. The corpus of world mythology is essentially a vast compendium of human creativity. Yet most myth consists of elaborate arguments why we latter-day humans can no longer be genuinely creative. The great foundational gestures were all performed in the misty past; in these lesser days, we are no longer capable of anything truly new. Myth, then, is creativity turned against itself. To celebrate myth as the deep structure of human society or human thought is to say that all the important things have already been established: all heroic narratives, all ways of conceiving gender relations, all conceptions of authority, all are already given, and even history, as Eliade so famously argued, should be conceived as an eternal return of the same archetypal gestures and characters. Obviously it’s possible to avoid this conclusion: to see myth instead as, for instance, ideology, or, in a more positive light, as a well of self-denying creativity that can and should be drawn on to continually revolutionize society. But it’s unsurprising that few of those drawn to dedicate their lives to the study of myth have embraced such an approach.


Terry Turner’s basic question, then, with regard to myth was: Why have so many human societies embraced such conservative conclusions? Certainly this was true of the Kayapó. As Turner writes in “The fire of the jaguar”:

The question becomes this: why should the Kayapó regard the very power to create and maintain their social order . . . as itself, in origin and essence, an asocial (“natural”) power? The answer is that they do not regard the structure of society itself as within their power to change, or, therefore, within their power to create. It follows that the basic forms, that is, the basic transformative mechanisms upon which their society rests, must derive from an extrasocial source. (p. 30, this volume)

Hence his embrace of Marx and the fundamental insight—one seen nowadays as so intrinsically suspicious by poststructuralists—that there is a necessary link between humans’ misunderstanding of the process of their own creativity and forms of authority and exploitation.

The great moral danger of any such approach is (as Bruno Latour, for instance, emphasized) condescension: Are we really prepared to say that the people we study are fundamentally wrong about the workings of their society and that we know better? This sounds like a very serious charge until we consider that, by doing so, we are really just reducing the Kayapó (or whatever group we are analyzing) to the same status as our professional colleagues, whom we accuse of being fundamentally wrong about the workings of society all the time. Turner would no doubt add: while Kayapó folk understandings of their own society are in many ways more sophisticated than those of most social scientists (certainly, than most structuralists), they’re not social scientists, have no interest in becoming social scientists, and Kayapó social order is in no sense an attempt to resolve intellectual problems. (As Terry notes, when he attempted to outline some of the interpretations developed in this book to Kayapó friends, their main reaction was not disagreement, but indifference. They simply didn’t find such questions interesting.)

Finally, there is a degree—already noted—to which such questions can never really be answered anyway.


This might seem somewhat contradictory: How can one both say that myth is the product of an intellectual puzzle and, simultaneously, that it is not an attempt to solve that puzzle? What, for Turner, are myths actually about? Here, at least, he is considerate enough to spell the matter out:

. . . the basic notion of the function of myth put forward in this study [is] that of directly connecting the “subjectivity” of the social actor with the objective structure of the socioeconomic system to which he or she belongs. (p. 146, this volume)

“Subjectivity” here is meant in the literal sense: it is about the formation of the subject, as an entity disposed to act and capable of acting in a certain way. Myths provide those who hear, learn, and retell them not only with tacit models for how to act but, even more, with a tacit guide to how to feel about the process by which we do so, with all its attendant dilemmas, tensions, and contradictions, what it is justifiable to fear and to desire.

This focus not just on the intellectual but also on the “affective” dimension, on “patterns of feeling and motivation,” is, of course, extremely unusual for the structural analysis of myth. Most of those who study myths would never be able to attempt such an analysis, except perhaps speculatively, since they deal with stories told long ago or far away, often in languages no one has spoken for centuries. We would have little way of knowing if there were certain incidents in the story of Inanna and Dumuzi, or the Labors of Hercules, that Babylonian or Greek audiences considered particularly amusing or terrifying. The response is to create forms of mythic analysis where such questions don’t really matter. Terry’s many decades of fieldwork, in contrast, meant that he had heard the same stories over and over from different narrators and, as a result, knew exactly what parts were supposed to be funny, which scary, as well as what was idiosyncratic in any given performance and what essential to the narrative itself. This in turns allows him to read myths in their social context as oriented to shaping desires and sensibilities in a way that more intellectualist readings simply can’t.

Here, too, Terry saw himself as positioning himself in much the same way as did Marx: as synthesizing the best of the French and German traditions. Marx admired French Enlightenment thinkers because they understood one had to see humans as existing in the material world and meeting material challenges; however, since they started by basically plunking down a collection of purposeless humans fully grown into a world of objects, they ended up seeing them as simply reacting, Marvin Harris-like, to material conditions. German Hegelian philosophy started from action and therefore understood humans as creating themselves through their projects: objects were by definition objects of action, even when that action was mere contemplation. This was much better, Marx believed. The problem is that German philosophers tended to forget there even was a material universe. Terry entirely agreed with this assessment. He just carried the same work of synthesis over into the analysis of myth, where his project was to combine a static French theory of signification (Lévi-Straussian structuralism), which admitted it had nothing to say about meaning, with a dynamic German theory of meaning (Schleiermachian hermeneutics), which saw texts as intentional forms of action. In the latter, the meaning of a text was what an author was trying to say.

For this reason, the analysis of “The fire of the jaguar” proceeds on two levels simultaneously: it deals first with structure, the “formal aspects of the logical relations among [a myth’s] symbolic elements”—the level with which all structural analysis necessarily deals—and second, with its subjective meaning to the actors, “the type of message it conveys” (p. 4, this volume). On the one hand, a myth “lay[s] down a pattern of action.” On the other, it is about “knowing and experiencing and deeply feeling that structure of social relations” (p. 146, this volume), which said pattern of action creates. The power of myth, however, does not lie in either one of these two levels. The power of myth lies in the implicit proposition that they are both the same. Ultimately, the meaning is the structure. The structure is the meaning. The inevitable becomes desirable. Hence inevitable.


To demonstrate how this can be the case and what it means in practice, Terry develops his own unique theory of narrative. It bears little resemblance to narratology as it currently exists and, to my mind at least, is far more promising than anything the semiologists have yet managed to come up with. His approach was first outlined in a piece in the classical journal Arethusa, published in 1977, called “Narrative structure and mythopoesis,” which argues that the plots of stories can themselves be seen as self-organizing structures. Ostensibly, it does so through a reanalysis of the Oedipus myth. Unfortunately, the piece is so long and presented in such an obscure style that it seems to have left most classical scholars scratching their heads, was missed completely by anthropologists, and nowadays has been almost completely forgotten.

Still, it’s an important essay, if only for the reason that it introduces Terry’s notion of the minimal episodic unit. This notion of an elementary structural unit actually is key to Turnerian structuralism (if we can call it that) more generally. To understand any structure, Terry held—whether a poem or story, or a social system—one must first identify what he sometimes called, in typically ungainly fashion, its “minimal modular unit” of structure, the smallest unit that nonetheless contained within itself all the key relations operative within the larger whole. In the case of a narrative, mythic or otherwise, this minimal unit is the episode. Each episode that makes up a story is organized around an action or set of actions. A plot is, after all, as Aristotle insisted, “an imitation of action,” the episodes that make up a plot, its minimal units, are each in each case acts in which characters change something (the world, themselves, their social relations with other characters—usually all three at the same time). It’s only over the course of the story that it becomes clear that each episode shares a common structure, which also becomes the principle that regulates the relation of the episodes to each other.

To illustrate, Terry took the Oedipus story, so famously reinterpreted by Lévi-Strauss as a meditation on the relations of eyes and feet, and applied a model of triangular structures inspired by Roman Jakobson’s phonemics, defined by reciprocal transformations of its elements. (This is the same triangular model that reappears in this book.) There are always two key axes, and in every case, one change along one of them will trigger a complementary transformation of some kind: that is, the old king dies, his warrior usurps the throne. With the first episode, the key relevant features (foreign/indigenous, loyalty/ ambition, etc.) might not be entirely apparent, but the moment there is a second episode and other transformations along the same axes recur, then the very comparison that allows them to be seen as similar necessarily generates a higher level of structure, which becomes a “general principle or force responsible for creating the common pattern it manifests” (1977: 142). To put it more simply, each episode marks an action that changes the overall situation, but, as the story continues, a common pattern in those changes emerges, and that emergent pattern becomes the governing principle—or, as Terry once puts it, “cosmic demiurge”—that generates the plot as a whole. So, just as each episode contains a complementary transformation, so does the story as a whole: that is, the narrative begins with Oedipus as an infant, having pins stuck through his feet, and ends with him as an old man, sticking pins in his own eyes. It is similar, in a way, to the hermeneutic circle, where one reads each episode in a work of fiction as a way of understanding how they together form an overall totality, that totality being seen as identical with the intention of the author—the meaning of Hamlet, that which binds all the episodes together, is assumed to be what Shakespeare is “trying” to say. (“Shakespeare,” in this sort of analysis, is not even really a person, but also a demiurge; the author is just conceived as that unifying intentionality.) In a myth, however, there is no single author, even as an abstraction. The story writes itself.

True, the audience doesn’t typically notice this, instead following the apparent back and forth of episodes with apparently contradictory messages as the plot weaves between them, but it’s the emergence of this “demiurgic” power of self-regulation that allows the reader to feel that a satisfying story has been told. And doing so allows the audience to not just think through, but feel through, the quandaries and contradictions of family life—in each case (the fire of jaguar, the Oedipus myth) in a way sufficiently compelling that the story has been repeated for thousands of years.


Some stories endure. Most theories tend to be a lot more ephemeral. I hope this book will prove an exception.

The fire of the jaguar should, in my opinion, be considered one of the great achievements of anthropological theory. It deserves a place among the classics. It was a book that had the potential of opening doors that no one has been able to walk through, since the doors were dangled in front of us only in potentia, like the kind of shimmering dimensional doors one might see in a science-fiction story, always lingering ghost-like above our heads. One such door has now materialized. Will anyone now choose to pass through it? Has it materialized too late? Does anyone even now care about the possibility of a truly dynamic structuralism?

Well, pendulums do swing. It’s possible that the current adamant hostility to the Lévi-Straussian project, the rejection of any dream of reconciling advances in scientific understanding with social understanding, might be showing signs of giving way. Perhaps the belated appearance of The fire of the jaguar will encourage anthropologists to think about such big questions once again.


Latour, Bruno. 2007. Reassembling the social. Oxford: Clarendon.

Piaget, Jean. 1970. Structuralism. Translated and edited by Chaninah Maschler. New York: Basic Books.

Turner, Terence S. 1977. “Narrative structure and mythopoiesis: A critique and reformulation of structuralist concepts of myth narrative and poetics.” Arethusa 10 (11): 103–64.

[1] I know three definitely existed: The fire of the jaguar, a collection called Critique of pure culture contracted to Berg but endlessly delayed, and The Kayapó of eastern Para, a manuscript prepared for “Cedi, Povos Indigenas do Brasil, Volume VIII” of which I still have a copy of the first 56 pages—I can’t for the life of me figure out what happened to the rest of it. Other rumored volumes may or may not be mythical.

[2] I once got my hands briefly on a draft of Marshall Sahlins’ “Peloponnesian and Polynesian Wars” book this way, but the manuscript was so enormous that my library wages were not adequate for me to be able to afford the costs of photocopying it all. I was already living on ramen noodles at the time there were no more corners to be cut.

[3] This is why Pierre Bourdieu had to move from anthropology to sociology, as there was basically no room for another theorist, and anyway, Lévi-Strauss did not approve of the theoretical direction he was taking.

[4] Terry insisted to me he’d once heard Lévi-Strauss actually say that he was entirely comfortable with an arrangement where other French anthropologists would work primarily to gather and organize data, and he would interpret it. I’m just reporting. Terry’s memories were not always entirely accurate, but sometimes they were.