Too ontological, too rigid, too ahistorical but magnificent

This work is licensed under the Creative Commons | © Stephan Feuchtwang. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau4.3.025


Too ontological, too rigid, too ahistorical but magnificent

Stephan FEUCHTWANG, London School of Economics

Comment on Descola, Philippe. 2013. Beyond nature and culture. Translated by Janet Lloyd with a foreword by Marshall Sahlins. Chicago: University of Chicago Press.

Beyond nature and culture constantly disarms a reader, like me, who has engaged in a narrower comparison of universes, or cosmologies as I have preferred to call them. Its scope is so much bigger that whenever I find differences that seem so pro-found as to question Philippe Descola’s scheme of things, I come on the following pages to a reminder that each of the four families of universes contains huge differ-ences. But I will persist in my narrower comparison between cosmologies in China and in Europe and the questions they raise about the whole four-part schema.

The most basic starting point, which will also be my ending point, is the rhetori-cal—which is to say persuasive—role played by “ontology.” I understand it to be a superior substitute for both “culture” and “ideology.” Superior to culture because it goes beyond values, even the mightily expanded theorization of values in economic and moral anthropology, to include with values a kind, or mode, of knowledge of the world. Superior to ideology, for the same reason and because it includes not just institutions, such as state apparatuses and rituals of fertility, but also a way, or mode of experiencing the world that tests cognition of the world in practice. This is, indeed, a persuasive substitution and doubly so because it is set up to challenge the normative status of what, by this schematization, turns out to be only one of four ontological schemes, the nature/culture dichotomy. But like rhetorical devices, it teases the reader into the possibility that what would otherwise be described as “metaphysics” is an alternative reality, a universe among relative universes, and thus leaves open the question of knowledge of what is a world shared by us all.

The philosopher G. E. R. Lloyd, whose main comparators are the same as mine, tries in his book Being, humanity, and understanding (2012) to tackle this problem by pointing out that “science” must always include not just procedures of testing, proof, and argument, but also hermeneutics, and that it is always framed in relation to power holders, must therefore be persuasive to them, and its results can always be censured or constrained by other kinds of authority, let alone that of the scientific community or priesthood. But when he writes of “science” he refers only to the archives and texts of the Mesopotamian, Chinese, and Greek states and to the recorded astronomical observations, mathematics, and modes of argumentation among their literate elites. He allows for mutual checking among them and with “modern” scientific observations, on the premise of a shared world. But should he allow shamanic knowledge into “science”? Lloyd (2012: 68) has doubts because the archival accumulation of knowledge records is absent, and they provide an indis-pensable temporality of checking and testing.

This seems to distinguish between ontology and science, though all sciences have an integument of assumptions and exist in what Descola calls “collectives” whose members share ritual, religious, and other institutions, such as the Greek habitual placing of humans between gods and beasts, and the institutions of state that in-cluded the cults of the gods. So, do some ontologies, those of stateless societies, lack the conditions for “science”? In which case must their ontologies, animism and to-temism, be “metaphysics,” but not physics? However these questions are answered, they in turn raise the proposition that literacy and the formation of hierarchies of authority into states are the preconditions for two of the ontologies, analogism and naturalism, and conversely escape from states and literacy or simply being without them is the political condition of the two others, animism and totemism.

Before returning to that contention, let us move away from physics, broadly de-fined to include mathematics, astronomy, chemistry, medicine, and perhaps more, and take into account the important fact that Descola’s scheme is based on the dif-ferent ways in which “humans” are differentiated, which is to say whether or not they are distinguished from beasts and plants with other physical forms, and how these other human or totemic interiorities are differentiated from other objects in the universes of the animist and totemic collectivities. We could say that all four ontologies are anthropologies, including those of analogism and naturalism.

Descola, in his disarming way, admits that his whole bird’s-eye view of the on-tologies of humanity is of course, inevitably rooted in the assumptions of naturalism (2013: 303). Even while he is trying to create distances from it that the other ontologies provide, particularly those of the Amazonians, the Achuar, with whom he lived and hunted, he is still bound to consider human and cognitive sciences in the very establishment of his four-cell grid of ontological families of universes. These sciences establish to his and our satisfaction the universality of the distinction between interiority and physicality (121). Is there more to be found univer-sally, for instance in the very distinctiveness of both the physicality and interiority of humans, recognized as is this distinction in their own ways by all ontologies?

Right at the start of the book (Descola 2013: 4) where he quotes from his translation of one of his Achuar hunting companions, he has put an “as it were” qualification to their human identification of their prey: “they are, as it were, our relatives by marriage.” Does this qualification not imply a universal similarity of distinguishing humans from beasts but then extending to beasts (and plants) a quasi humanity in some ontologies but not in others? For the inhabitants of the animist or totemic on-tologies, Descola argues, the physical difference is superficial, the “as if ” essential. Yet it is still an “as if.” For Lloyd this is an example of different semantic stretches in different languages and their references, including metaphor and myth vital to rendering what is a multidimensional reality, consisting of qualities as well as of specifiable (chemical or otherwise reducible) contents. So, the hermeneutics of ref-erence is the key to the differentiating not only of languages and their usages but to ontologies themselves. Crucially, any one language can be translated into another and itself be questioned by the semantic stretches that each deploys, and it is this “understanding” to which the title of Lloyd’s book refers and uses helpfully as a way of understanding Descola. So, not “science” but understanding of understandings and some things held in common by all anthropologies would be a Lloyd version of Descola’s scheme.

Lloyd does, however, present a clear criticism, not of the scheme in general but of the historical depth of naturalism that will lead me to a contrast with Chinese cosmology and ultimately back to the whole scheme.

Greek philosopher-scientists in their habitual competitive disputations—Games of the mind to persuade the jury that their arguments fatally disposed of all other Masters—sought a nature behind human cultural appearances, an existence be-yond and behind, which many of them equated with the divine. This reduces their scheme to a two-fold distinction between divine nature and culture. For Lloyd, there is therefore a much longer history of naturalism than Descola’s fixation on Enlightenment scientific rationality allows. When humanity is itself conceived to be near the divine and the manifestation of God’s redemptive self-sacrifice in Christian theology, the divinity of nature is turned into a steeper hierarchy, with the human, in God’s image, near the top. This is a transformation of Greek science and philosophy, but nevertheless there is an evident continuity. So, why does Descola not see the continuity that Lloyd sees? Is it because he has already persuaded him-self that what preceded naturalism in Europe was analogism? Or perhaps he wants to reserve only for this ontology a history, and not the others—their historical con-ditionality upon literacy and state-like if not state authority? Do we need to rethink the cases of African nonstate analogism that he cites (Descola 2013: 221–26)? In particular, I would question whether the defining characteristic upon which Desc-ola dwells, namely the parsing of persons into several persons, or souls as they are also rendered, each with their own networks of mutual being with subjects both human and nonhuman in appearance, is confined to analogism. On earlier pages (117–20) where he is establishing the universality of interiority as opposed to phys-icality, he seem to admit a far wider spread of such divided personhood. To place it in analogism, he seems to equate it with hybridity, the joining of taxa from several observable beasts or plants into one hybrid body, such as the chimera, the dragon, or the sphinx, monsters of the late Bronze Age, early state formations. But this is surely quite distinct, a fabulous fabrication to manifest and communicate with nu-minous, or divine beings. It seems to me still arguable, therefore, that two of the on-tologies are those of the stateless or of the left-behind, or the escapees from states.

I turn finally to the comparison with which Lloyd and I are concerned, between civilization in Europe and in China, both of them placed in the grand family of universes that Descola names “analogism.” The defining problematic of analogism is the totalization of a world of singularities. Totalization is achieved through the proposition of correspondences among the singularities, ordering them and locating inherent disorder spatially, outside the centralized, hierarchized, and thus total-ized world, which in China is designated by Tianxia, All under Heaven, or in late medieval Europe by the Great Chain of Being reaching down to Hell, a hierarchical location of the makers of disorder.

Here I invoke what Lloyd points out, the glaring difference between a cosmol-ogy of elements, conceived as basic substances and the composition of things, in-cluding diseases, out of them in various balances (Greece and then Europe), and a cosmology composed by basic processes of change (China). These processes of change, deceptively named the Five Elements, though they should be translated as Five Phases—Water, Fire, Metal, Wood (or Plants), and Earth—interact in two cycles, one destructive the other constructive. In other words, the problem of de-struction, fragmentation, and disorder is not displaced to the outer regions of All under Heaven, it is inherent in the composition of a total universe. There is no premise of singularities. The universe and the correspondences that are based on these five processes are in two further orders. One is a cosmogonic order, starting from nothing and moving into a fertile chaos from which emerges the pivot and the complementary processes of yin and yang, out of which stem the five phases that produce the different things. The pivot introduces the other, more obviously hierarchical order of humans on earth with all other earthly things, between an underworld of water and an encompassing heaven, including stellar positions. Tian is the encompassing location of the principles of change, constant change, both secular and cyclical, including disorderly change. This is in sharp contrast to a uni-verse of beings, ontic entities, as posited by Greek and then Christian hierarchies of humanity between beasts and the divine, encompassed by an Unmoving Mover who through Christianity becomes the absolute other, whose existence is the sub-ject of hermeneutics and of course, of ontology. God is the ultimate singularity. Where Descola discusses production and the fabrication model or metaphor of the Biblical universe and wants to show how other universes, in this case a Chinese universe is quite differently conceived (2013: 322–23), he seems to acknowledge, disarmingly yet again, this huge difference within analogism. To posit the problem of the existence of singularity and its creatures, other singularities (even without exporting to China the hierarchy of approximation to the ultimate) as common to all analogical schemes and therefore to China as a founding premise is quite obvi-ously wrong. Can both be analogical, just because Chinese cosmology is a practice of divination, ritual, and speculation through correspondences? Are these corre-spondences of a type with the Great Chain of Being, or of a different type?

Let us accept they are of the same type. If they are of the same type, then the problematic in Christian ritual practice, as well as Chinese, of sacrificial offerings is surely just as much separation by communication, even more than joining.1 Ritu-als are representations, making visible in the world an unseeable encompasser who in Christian theology is an absent presence and a present absence, intrinsically and unbearably blinding, who must be kept separate and in Chinese cosmology incommensurably greater as a world-making embodiment of principles of change who must be treated as honored guests who become hosts through representations and are kept and keep themselves separate. So again, the premise of singularities is unwarranted. It seems to me to be a function of the structural rigidity of the logic of the four-cell grid. It is still possible that the cosmology of five phases in harmonious or in disordering interaction was preceded by a cosmology of singularities, but this is an empirical problem, whereas for Descola it is, I think, a problem of his grid, that is, of his structural logic.

My final question is whether it is even appropriate to call these families of uni-verses “ontologies,” when Chinese ritual and other practices, including philosophi-cal disputations and speculations, do not inquire into the existence of “being” or the thinking about thinking by which humans approximate the Unmoved Mover who thinks the world in Greek speculation. Philosophy endogenous to China has no preoccupation with being, no ontology. Cosmology in the changing civilization out of which this Chinese philosophy arises is based on the problem of preventing disorder under Heaven, ritually communicating with cosmogonic origins and with heavenly deities, whose most privileged places are in the border regions of All under Heaven. It is not a philosophy of joining singularities or communicating in a hierarchy of singularities and their classes of being.

I am not arguing for Chinese exceptionality. Instead I am using the cosmology that I find in the civilization of China to question the rigidity of the scheme that drives Descola’s magnificent comparisons of worlds of interiority and physicality and to question whether he needs to describe them, or we (including Lloyd, who apparently does accept) need to accept their description, as ontologies.


Descola, Philippe. 2013. Beyond nature and culture. Translated by Janet Lloyd with a fore-word by Marshall Sahlins. Chicago: University of Chicago Press.

Lloyd, G. E. R. 2012. Being, humanity, and understanding: Studies in ancient and modern societies. Oxford: Oxford University Press.



Stephan Feuchtwang
Department of Anthropology
London School of Economics
6th Floor, Old Building
Houghton Street
London WC2A 2AE, UK


1. Very many thanks to Tom Boylston for making this point.