Lévi-Strauss and Philosophy

Justin E. H. Smith

6a00d8341c562c53ef0120a65e789c970b-320wi I.

In his Tristes Tropiques, composed in 1955, Claude Lévi-Strauss writes with characteristic humor of his decision, some years earlier, to study philosophy:

When I reached the top or 'philosophy' class in the lycée, I was vaguely in favour of a rationalistic monism, which I was prepared to justify and support; I therefore made great efforts to get into the section taught by Gustave Rodrigues, who had the reputation of being 'advanced'… After years of training, I now find myself intimately convinced of a few unsophisticated beliefs, not very different from those I held at the age of fifteen. Perhaps I see more clarly the inadequacy of these intellectual tools; at least they have an instrumental value which makes them suitable for the service I require of them.

Later, in a 1972 interview, he confesses that his decision to study philosophy was motivated by a sense that this discipline, more than any other, would enable him to remain non-committal, to continue to develop all his other interests, under the big-tent of a vaguely defined cluster of intellectual projects called 'philosophy'. This understanding of philosophy, I think, remains significant for our assessment of Lévi-Strauss's intellectual legacy.

For better or worse, while his approach may have made sense in the Paris of the 1920s, as I can personally attest it certainly would not in the New York or California of the 1990s (when I was a student of philosophy). Here, a different conception of philosophy and its boundaries reigned, and still reigns. As Jason Stanley recently reflected at the Leiter Report blog:

Many academics use the term 'philosopher' not as a description of the people working on the set of problems that occupy our time, but rather as a certain kind of honorific. As far as I can tell, on this usage, a philosopher is someone who constructs some kind of admirable general theory about a discipline – be it cultural criticism, history, literature, or politics. So while it would be odd for a philosopher to call themselves a literary critic because they work on interpretation, it is not unusual for English professors to describe themselves as philosophers. In contrast, we philosophers do not regard the term 'philosopher' as an honorific. We tend to think that there are many people who are really truly philosophers, but are pretty bad at what they do. We also think that there are many brilliant thinkers who are not philosophers.

Stanley argues in another post that his own philosophical tradition may be distinguished from a rival tradition, represented by Walter Benjamin, that might better be called 'anthropology' than 'philosophy':

Benjamin isn’t at all confused about metaphysics or the problem of intentionality. He just finds no interest in the question of how, by the use of language, one person can communicate something about the world to another. What’s interesting to him is how language is represented in human mythology, and what that reveals to us about the cultural significance of our practice of naming. This kind of question is one that is not apt to be taken up by a philosopher in the analytic tradition. Someone in my tradition might say that the issues that interest Benjamin are questions of anthropology rather than philosophy. Someone in Benjamin’s tradition might say that the issues that interest me are bourgeois.

Stanley makes two claims in these passages that interest me: first, that not just any abstract or broad-focused thinker may appropriately be called a 'philosopher', and, second, that much of the thinking that is called by some people 'philosophy', might better be called 'anthropology' to the extent that it is principally interested in questions of culture rather than, I take it, in transcultural features of the human mind and its connections to the world.

While I certainly know Stanley is a first-rate philosopher, I do not at all share his conception of what philosophy itself is. If anything, my own understanding of the meaning of 'philosophy' is the one at work not in the United States today, nor in France in the 1920s, but rather in the title of the distinguished journal of the Royal Society of London, the Philosophical Transactions, which, since 1666, has been featuring articles on everything from the reproductive organs of eels, to the smelting of metals, to the causes of comets, to the nature of the passions, to the existence of God. It seems to me that if Stanley wants to make the case for a narrower conception of philosophy, he needs not only to argue against the misguided deployment of the word that we've certainly all heard at dinner parties, but also to explain why the very most recent self-description of a certain academic discipline in a certain part of the world should be permitted to cancel out so much accrued meaning in a word that has migrated and mutated across so many centuries, languages, and continents.

One problem with Stanley's conception of philosophy is that it leaves out a great number of figures who lived and thrived before the academicization of philosophy in the 18th and 19th centuries, yet who are universally recognized by current philosophers as having contributed something important to the tradition we've inherited. Descartes, Leibniz, Boyle, Gassendi, Digby, Huygens, Newton: these figures and many more are recognized as having contributed something to what we today call 'philosophy', yet they also contributed a great deal more that cannot be fitted into the current range of interests of philosophers, such as their work on the causes of spontaneous generation, the role of the animal spirits in physiology, the moral benefits of chemical experiments, the question whether corpuscles have gender, and so on. Now, if I were capable of going back in time, I would not be at all eager to tell any of these great thinkers that only part of what they were doing –i.e., the part that would be recognized and valued in the early 21st century– is in fact philosophy, while the rest is just culture, or tradition, or garbage. I would seek to understand how it all fits together, why what would later be valued was presented together with what would later be ridiculed or dismissed. That's what I conceive my task to be as an historian of philosophy.

It may be that on the prevailing understanding of philosophy a certain kind of historian of philosophy (my kind) is simply not a philosopher. Fine, then, you won't hurt my feelings if you call me a historical anthropologist, studying what is really just another species of pensée sauvage, an enchanted, ridiculous world of substantial forms, hylozoic archaei, and other nescio quods. But quite apart from labelling issues, I find Stanley's conception of philosophy hard to accept because it seems obvious to me that philosophers, good philosophers in the sense understood in mainstream, non-historical, Anglophone philosophical circles today, regularly make use of data furnished to them by other disciplines, and I don't see any prima facie reason why anthropology, even cultural anthropology, should not be one of these disciplines. Stanley is at pains to clearly demarcate the border philosophy shares with other, lower-status humanities disciplines, the ones that threaten to pollute, and so to change the character of, our pristine realm. It's not clear however that he would want to guard the other border, the one it shares with the natural sciences, nearly so vigilantly. But it's also not clear, to me, on what grounds an open border should be permitted with the natural sciences, but not with, e.g., paleography, historical linguistics, biblical exegesis, and even cultural anthropology. Relatedly –and here I'm starting to come back to Lévi-Strauss after a digression that was much longer than I intended– I don't see why the study of culture should not be expected to furnish insights into the transcultural nature of the human mind and of its fit with the world. And if doing that is not philosophical, then I really don't know what is.


Does anthropology in fact furnish such insights? If it does not, at present, this may be a result not of its intrinsic disciplinary limitations, but rather of our broader, transdisciplinary intellectual climate, which seems divided, broadly speaking, between two camps. On the one hand, there are those represented in crudest form by Richard Dawkins, who think that to believe in fairies, ghosts, river-spirits, God, creation, etc., is simply wrong, and therefore unworthy of interest (Daniel Dennett has by contrast argued that religion should be studied as a natural phenomenon, yet seems largely uninterested in the work that has already been done in this area by anthropologists, and seems oblivious to how much more religion involves than ontological commitment to a transcendent entity, e.g., eating this part of the walrus's body but not that part if you are an Inuit shaman, or eating olive oil on Tuesday but not Wednesday if you are a Romanian Orthodox Christian). Dawkins has said that he declines to learn anything about theology for the same reason he declines to learn anything about fairy-ology: you don't need to know anything about the doctrine, he thinks, in order to know that fairies do not exist. But when did we become so intellectually stunted that the question of the existence of fairies came to be the only relevant branch of fairy-ology? When did we lose sight of the interest –I mean, the scientific interest– in the variety of ways people conceptualize the world, from which organ meats are taboo, to the causal influence of dead ancestors in our worldly affairs, to the matter of the ultimate transcendent ground of the world?

On the other hand, there are those –for the sake of convenience let us call them all 'post-structuralists'– who think that believing in fairies and such like entities, under certain circumstances, amounts to defiance of the hegemony of Western scientific rationality, and is therefore 'good'. What gets lost in this rift is the approach that says: Believing in fairies (or in the taboo charge of a certain bit of meat, etc…) is something people do naturally; therefore, the study of beliefs about fairies is the study of something in nature, and is thus worthy of interest quite apart from either considerations of the truth of these beliefs, or of the defiance these beliefs pose to truth-mongering (which post-structuralists believe is always also power-mongering).

It is in this connection that I truly think –rather than just thinking now, for the moment, as a result of my unexpected sentimentality in the days following his death– that the spirit of the work of Lévi-Strauss, if not all of its conclusions, is worth reviving today. This is a spirit that is there in much human science of the early 20th century: in Vladimir Propp, Émile Durkheim, Marcel Mauss, to name a few luminaries. When Propp sought to spell out the morphology of the folk tale by analyzing and formalizing the themes of a number of bits of Russian folklore he had accumulated, this is not because he thought Russians were particularly folksy and needed to be celebrated for it, or because he thought their folk tales were uniquely suited to formal analysis. He fed Russian folktales into his Proppian machine because he wanted to understand the range of possible folktales, and in so doing to understand the range of possible ways different cultures make sense of themselves. This is at the outset a rather tedious, empirical exercise, but the aims are philosophical: to understand how people think, to understand what it is to be a thinking human being in the social and natural world. Lévi-Strauss's life work, the effort to establish a “culturally informed science of mind,” in the words of Marshall Sahlins, is, I believe, the pinnacle of this philosophically motivated approach to the human sciences.

I can't hope to give a comprehensive account of his work here; what I want to do rather is to suggest that, in general, Lévi-Strauss shows us an approach to knowledge systems that ought to be of interest to anyone concerned with figuring out how any future philosophy of mind might come forward as a science. To some extent, structuralism might be said to overlap with what is called 'contextualism' in epistemology, according to which beliefs need to be assessed in relation to the background of other, available beliefs in a given community. The principal difference however, between the project of a culturally informed science of mind, on the one hand, and philosophical accounts of belief and meaning, such as contextualism or holism, is that the former wants to empirically survey the range of contexts, and from there to draw lessons about what they all have in common. Consider the well-known example from Lévi-Strauss's Mythologiques cycle, of the Peruvian tribal chief who, following upon some misfortune or other, demanded that all the twins in the community repent for having caused this misfortune. Lévi-Strauss wants to know by what rationality that response could have been meaningful, and he believes that in the end this rationality, coming out of brains that are no different from our brains, is going to be governed by the same laws that govern our mental and social lives.

Today, again, one might say that there are only two possible approaches to the cultural significance of twins as harbingers of misfortune: denunciation of it as scientifically misinformed; or defense of it as 'other' and therefore as being as worthy of absolute, solemn respect as it is incomprehensible to the outside observer. Few seem interested anymore in simply trying to decipher what those people are thinking. Again, whatever it is they are thinking is something that is being thought by human minds/brains, and so to map the structure that makes this way of thinking meaningful, along with all the other, different structures available from around the world, seems to me like a pretty good way of empirically establishing how the human mind works.


If sensitivity to culture and the way it is structured by a hidden rationality is potentially useful for resolving questions about human cognition, it is arguably necessary for making progress in many debates in social and political philosophy. Take arguments concerning gay marriage. Philosophers tend to argue for or against opening up the institution of marriage by appeal to supposedly transcultural notions such as justice and equality. An anthropologist of the Lévi-Straussian stripe by contrast, who takes it for granted that meaning in cultures is constituted out of sets of oppositions, and that kinship is among the most elementary forms of social structure in which these oppositions are expressed, will want to know what sort of effect, real or perceived, the decision to abandon a long-held opposition will have on the structure of the society as a whole. If there is any truth to Lévi-Strauss's controversial argument in his Elementary Structures of Kinship of 1949 that kinship systems evolved in the first place as a mechanism for the 'exchange of women', one might predict that it would be hard to convince everyone that it's now time to switch to an entirely gender-indifferent conception of kinship. (It would take a separate essay to make adequate sense of Lévi-Strauss's very complicated account of kinship as exchange of women; suffice it to say here that in providing this account Lévi-Strauss was neither 'endorsing' woman-exchange –as if endorsement had any place in science!–, nor was he saying women are commodities; he was, in any case, saying that the opposition male-female plays an ineliminable role in kinship systems, and this much of Lévi-Strauss's account has been endorsed by at least some of his feminist critics, including, if I've understood her, Françoise Heritier.)

This is not at all to say that opening up the institution of marriage to include same-sex couples is not right and just (if I may attempt to channel Stanley Fish, I've said nothing in defense of one side or the other); it is only to say that the inability of arguments that appeal to the right and the just to immediately win over the whole world might have something to do with rather deeply-seated mechanisms at work in the structure of society, mechanisms that it's not going to be at all easy to weed out. It might in fact be the case that it would be easier to dispense with monogamous pair-bonding than with the gender rules governing who may be whose affine, and in this sense it is hard to think of the goal of building a society out of monogamous couples of any gender combination at the heart of nuclear families as something that is timelessly right, and of alternative conceptions of society as motivated simply by hate or intolerance. If it ends up working, for the long term, this might be for hidden reasons, such as that North Americans and Western Europeans just happen to have what Louis Henry Morgan would call an 'Eskimo' kinship structure, as opposed to, say, Hawaiian or Sudanese, which involves relatively fewer differences in the rules governing patrilineal and matrilineal descent. But one could also predict that different obstacles would present themselves in different kinship systems, a surface expression of which, but not a root cause, might be intolerance. We should perhaps keep pursuing the project of modifying our kinship system, but we should also attempt to understand, scientifically, the nature of the resistance, just like we should keep trying to understand why humans keep engaging in ethnic cleansing or suicide bombing (on the latter, see Scott Atran). This is something I think anthropology is better at doing than a discipline that conceives its task as entirely transcultural.

In the 17th century, I hasten to add, understanding and even manipulating the elementary constituents of society would have been conceived as squarely part of philosophy: consider William Petty's 1672 plan to Anglicize the Irish population by inserting English women into Irish households. Petty has gone down in history as an economist and a demographer, yet he himself thought of his Anglicization scheme as a bit of natural philosophy, and conceptualized cultural change on the model of the transmutation of metals– which was also natural philosophy, or, more particularly, chemical philosophy. Try to fit all that into the disciplinary divisions du jour!


But to return to the prospect of a culturally informed science of mind: what ever became of it? Many anthropologists are still doing work that, in its piecemeal way, is contributing to the pile of empirical data that could be used for a culturally informed science of mind of the sort Lévi-Strauss imagined, but they do not seem to consider it as anything more than piecemeal work. Cognitive science continues to produce interesting results, but only a very small portion of it is ethnographically informed (and a smaller but very interesting portion of it, being produced, curiously, in Paris and mostly in English by Dan Sperber, Scott Atran, and their circles, is both consciously building on and improving the tradition of the anthropologists, while also communicating with people like Dennett). One result of this lack of interest in ethnographic comparison is that many of cogntive science's findings concerning the innate working of 'modules' for cognizing different domains of the human and natural environment run the risk, as Ian Hacking and many others have noted, of revealing only dispositions triggered by a particular cultural environment. The same point may certainly be made about what is now called 'experimental philosophy', in which social-psychological data about the intuitions of women and men in the street is gathered, in order to confirm or disconfirm philosophical views about which intuitions are right or not. But, obviously, a social psychologist's poll of modern Americans about their intutitions concerning, say, the moral valence of tearing off and devouring the arm of a victim of human sacrifice, is going to come up with different results than a similar poll taken in an anthropophagous society (and yes, there are anthropophagous societies; there might not be any openly doing what they've traditionally done, under the watchful eye of the 'world community', but what a proper human science would seek to understand is not the range of forms human society takes at present, but the range of forms human society can take).

These satellite research programs aside, in mainline philosophy an old and false dichotomy appears to linger, which dictates the range of possible approaches to the question of the unity and diversity of the human mind. One approach says all minds are very much the same. In fact they are all just like my mind, so all I need to do in order to learn what human minds are like is to think (see, for a very recent example, Galen Strawson's Selves). The other approach says that minds are very different, depending upon how they are situated in different kinds of bodies and given different biographies and different cultural worlds, so different in fact that if you don't have that body or belong to that world, you don't have any point of entry at all by which to understand. Understanding is replaced by Mitgefühl, and Mitgefühl comes to require Mitleben. This dichotomy seems to go back at least to a split between two different approaches to philosophy in the early modern period, which might be epitomized by the names 'Montaigne' and 'Descartes': Montaigne, who was fascinated by 'the cannibals' because he thought they revealed the absence of any unity between the world's various systems of meaning and value and conduct; and Descartes, who had nothing at all to say about the cannibals, or about anyone whose difference from himself might call into question his conception of exactly what is or is not among the things that are self-evident by the light of nature.

This split bears some ancestral relationship to the one currently marked in certain provinces by the labels 'continental' and 'analytic'. The former school bears particular responsibility for the abandonment of the Lévi-Straussian project, for at bottom the skeptical games of the post-structuralists are motivated by a rejection of the structuralist effort to comprehend otherness by comprehending the range of possible expressions of sameness. In at least this respect, if Galen Strawson is a philosophical descendant of Descartes (in methodology if not ontology), and Derrida of Montaigne, Lévi-Strauss is a Leibnizian, who believes that everything human beings come up with is an expression of the truth, not in the sense that every proposition assented to by humans is true, but in the sense that every attempt to make sense of the world is an expression of the same natural or rational order. This means, for Leibniz, that all learning is of use to the philosopher. In a stunning passage in the New Essays concerning Human Understanding of 1704, Leibniz writes:

When the Latins, Greeks, Hebrews and Arabs shall someday be exhausted, the Chinese, supplied also with ancient books, will enter the lists and furnish matter for the curiosity of our critics. Not to speaks of some old books of the Persians, Armenians, Copts and Brahmins, which will be unearthed in time so as not to neglect any light antiquity may give on doctrines by tradition and on facts by history. And if there were no longer any ancient book to examine, languages would take the place of books and they are the most ancient monuments of mankind. In time all the languages of the world will be recorded and placed in the dictionaries and grammars, and compared together; this will be of very great use both for the knowledge of things, since names often correspond to their properties (as is seen by the names of plants among different peoples), and for the knowledge of our mind and the wonderful varieties of its operations. Not to speak of the origin of nations, which is known by means of solid etymologies which the comparison of languages will best furnish. But of this I have already spoken. And all this shows the use and extent of criticism, little considered by some otherwise very clever philosophers who take the liberty to speak with contempt of Rabbinage, and in general of philology.

A big-tent conception of philosophy, to say the least. Or are Leibniz's New Essays not a part of the history of philosophy at all, but only a bit of cultural history?


I recall talking to an American Derrida scholar some years ago who insisted to me, in all honesty, that she did not know whether or not a chicken is a mammal. Such contempt for knowledge! Such shameful, unintentional self-parody! On the other side, I cannot count the number of times I've spoken with analytic philosophers who proudly enumerate all the areas of humanistic learning about which they know nothing (Stanley bemoans this too, but he thinks that members of the discipline need humanistic learning only in order to be more well-rounded people; I think you need it, along with scientific learning, in order to be a better philosopher). On both sides, there's a sort of compulsion to present oneself as pure: an enduring sense, one that has much deeper roots than the history of either analytic or 'continental' philosophy, and that probably takes a certain kind of religious persona as its model, that a philosophus cannot be a curiosus, that all the variety of the world presents an obstacle to the unity of thought. There have also been a few philosophers (or thinkers; whatever) in the past few hundred years, of whom I take Leibniz and Lévi-Strauss to be fine examples, whose respective life-projects might be characterized as a search for the unity that gives rise to, and endures behind, the wonderful variety of the world: philosophy through curiosity, and not just alongside it.

For an extensive archive of Justin Smith's writing, please visit www.jehsmith.com.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone