The anthropology of art has produced some of the most stimulating and interesting writing of the past 35 years on the arts known to me, in particular the essays and last book of Alfred Gell in the 1990s, and the more recent two books by Carlo Severi. In 2021 the anthropologist and anthropological theorist Philippe Descola published a magisterial account of the world’s art, which last year was published in English as Forms of the Visible: An Anthropology of Figuration. This recent book is itself the development of a set of claims argued in his earlier masterwork of 2005, published in English in 2013 as Beyond Nature and Culture. In this and in a follow-up post I’ll try to give an initial presentation and critical assessment of Descola’s work as a major contribution to the study of the world’s visual art. This framing goes against the grain of Descola’s explicit statements, where he claims to be considering what he calls ‘figuration’ or visual imagery as given or embodied in two- and three-dimensional surfaces and objects. At the end of my consideration I’ll offer some reasons for thinking that Descola’s topic is best understood as primarily visual art and not primarily embodied visual imagery; but even if my counter-frame proves inapt, still everything that he does discuss undoubtedly contributes to our understanding of visual art.
I’ll begin with the briefest possible sketch of the central philosophical-anthropological claims of both books insofar as they provide the motivation and philosophical framing for the more recent book. One can generate a structure of possible general ways in which human beings make sense of the world, others, and themselves by starting from the thought of an individual or collective subject attempting to identify some indeterminate ‘other’ (thing, plant, animal, person, world) (Descola (2025), p. 26). The human subject attempts to identify and make sense through setting up a web of similarities and differences between and among different kinds of others in two dimensions, physicality and interiority, which all human beings keep distinct. This generates a grid of four conceptual possibilities: humans and ‘others’ have (a) similar interiorities and similar physicalities; (b) similar interiorities and dissimilar physicalities; (c) dissimilar interiorities and similar physicalities; and (d) dissimilar interiorities and dissimilar physicalities. Since these are the four conceptual possibilities of determining the nature of things generally, Descola calls them the four ‘ontologies’ (Descola (2013), p. 122 and throughout his work). He then identifies each possibility with familiar terms from anthropology and philosophy: (a) is ‘animism’ wherein humans and other beings (especially plants and animals) have similar interiorities though manifestly different bodies; (b) is ‘totemism’ where humans and (some) animals have both similar interiorities and (some) similar physicalities by virtue of both being derived from “an eponymous animal or plant”; (c) is the naturalism characteristic of the modern West where humans and others both have physical bodies, but only humans have a non-physical interiority; and (d) is ‘analogism’ wherein in each kind, both self and other, is dissimilar from every other in both body and interiority, and so sense is made by constructing a web of analogies between and among beings. For any human collective, one of these four ontologies predominates, but collectives are not typically ontologically monolithic, both because of partial or marginal use of other ontologies available through cultural contact, and even because each ontology draws upon conceptual possibility in principle available to anyone; and a collectivity might, depending upon circumstances and the problems it faces, mobilize resources from one of the non-dominant ontologies (Descola (2025), p. 201). As a piece of general anthropological theory Descola introduces the term ‘schema’ to indicate the abstract structures “that organize understanding and practical action” (Descola (2013), p. 102) and that are embodied in human psychology, practices, and institutions. Some schemas may well be human universals, but that “it is above all acquired schemas that are at the center of attention of those interested in the diversity of customs around the world” (ibid) and “constitute one of the principal means of constructing shared cultural meanings” (p. 103). There are a number of different forms of such schemas, including identifications (“by means of which I can establish differences and resemblances between myself and other existing entities by inferring analogies and contrasts between the appearance, behavior, and properties that I ascribe to myself and those that I ascribe to them” (p. 112)), relations that arise between entities once identifications are established (pp.112-13), ways in which entities are organized temporally and/or spatially, mediations whereby a conventional device (such as sacrifice, money, or writing) and its usage sets up relations, kinds of explicit categorizations within taxonomies, and our topic, “various systems of figuration, understood as the action by means of which beings and things are represented in two or three dimensions, using a material medium” (p. 114).
Bringing together the parts of Descola’s philosophical anthropology generates the research project resulting in Forms of the Visible: Descola hypothesizes that figuration as a general schematizing practice will primarily exhibit across human societies one or another of the four ontologies. As indicated in the first paragraph above, Descola characterizes figuration as “the act through which real or imaginary objects are represented in two or three dimensions using some sort of material medium” (Descola (2024, a book of interviews wherein Descola characterizes his career as an anthropologist and how he arrived at his distinctive concerns and claims), p. 126). Figuration is a universal impulse (Descola (2025), p. 5) to act in certain ways and with certain recurrent mechanisms (p. xiv). The product of figuration always (i) makes visible something otherwise invisible and (ii) substitutes (under some description) for that which is invisible or absent. As human action, figuration is not equivalent to some class of the world’s artifacts, in particular works of art, but rather constitutes a unified topic by virtue of (i) the kinds of acts from which (ii) embodied images arise, and which (iii) are perceived and received in certain ways. For Descola figuration is the production of images (pp. 5 and 7), and he equates images with Charles Sanders Peirce’s conception of ‘icons’ as a kind of sign characterized as bearing relations of resemblance. On Peirce’s massively influential account, there are two other kinds of signs, one of which he terms ‘symbols’, where the sign bears some conventional relation to its referent, and ‘indices’, where the sign refers by virtue of some causal connection or as of a result of spatial contiguity or temporal closeness (so smoke qua index refers to fire). Accordingly he often refers to the products of figuration as ‘iconic artifacts’, as opposed to seemingly aniconic artifacts such as some kinds of decoration or works of abstract art. However, in light of the image’s two functions of making visible the invisible and substituting for something invisible, the notion of an aniconic artifact dissolves under Descola’s analysis into what are “in reality complex signs that function simultaneously as indices--they are receptacles, parts of a whole, the trace of an action, thus quasi-persons--and as icons--they make visible a fundamental property of the object whose place they are taking: its existence.” (p. 12) So in the term ‘iconic artifact’ the term ‘iconic’ is explicative and not something picking out one type of artifactual image.
Drawing from a large and cross-disciplinary range of philosophy, anthropology, art history, and semiotics, Descola claims that iconic artifacts should be analyzed in what he calls three registers (p. 475): the ontological, the formal, and the pragmatic. The first two registers are straightforward: the ontological concerns what kinds of objects and relations are prominent in the social world under consideration (one world might have maize and snakes that communicate with gods, another might have celebrities and violent storms caused by global warming); and the formal concerns the structural features of iconic imagery generally (some implied point of view, and some spatial and temporal ordering of the contents of imagery). The pragmatic concerns not just the characteristic uses of a kind of imagery, but especially (here Descola follows a number of recent thinkers, particularly Alfred Gell in Art and Agency) the sense that an iconic is itself a quasi-agent that does things in the world and/or as an index renders visible the agency of its maker or intended recipient. So Descola’s primary project, one that occupies most of the book’s 600 pages, is to analyze successively the iconic artifacts distinctive of each of the four ontologies (again: animist, totemic, analogist, naturalist). A further and especially interesting aspect of the project is to consider at length in two chapters kinds of iconic artifacts that bear characteristics of two or more of the different ontologies. His first example of such hybrid imagery are the works of the Tsimshian, a group on the far northern western shore of Canada, and whose works juxtapose totemic and animist elements. His second example is contemporary visual art, wherein many artists work with multiple ontologies.
As I have reached my customary length for a blog post, I’ll stop here with the presentation of the skeleton of Descola’s book. In the next post (Part 1b) I’ll sketch his fascinating accounts of the four ontologies and the two hybrid modes.
References and Works Consulted:
Philippe Descola, Beyond Nature and Culture (2013)
--The Composition of Worlds: Interviews with Pierre Charbonnier (2024)
--Forms of the Visible: An Anthropology of Figuration (2025)
Vincent Descombes, The Barometer of Modern Reason (1993)
Alfred Gell, Art and Agency (1998)
Charles Sanders Peirce, On Signs (1991)