Navigation – Plan du site
Autumn symposium

Natural Forms and Symbolic Classifications

Philippe Descola
p. 49

Notes de la rédaction

Source: La lettre, no. 33 May 2012

Texte intégral

1The question of the relationship between the perceived forms of organisms and the way in which human beings identify them and detect certain characteristics in order to use them in a wide variety of classifications has captured the attention of anthropologists and psychologists for over a century.

2The categorization of plants and animals seems to demonstrate certain characteristics of its own, particularly the fact that it often appears in a taxonomic form. This means that it is presented with a hierarchy of inclusions on several ranks, which is much rarer in other domains such as emotions, artefacts or inorganic materials. It is therefore not absurd to think that the form of organisms, and the supposed adequacy between this form and some of the functions it permits, plays a part in the particular way in which human beings classify them. Although this is still controversial, certain authors thus argue that a specific cognitive device activates the perception and the taxonomic distribution of natural objects. This device corresponds to what developmental psychologists call a naive theory, meaning a presumably universal scheme of inference which structures the expectations of any individuals regarding the characteristics and behaviour of organisms. More generally, and insofar as most of human history has taken place in a context of intimate and permanent interaction with plants and animals, there are grounds to believe that our cognitive activity has been deeply shaped by this experience. For all these reasons, organisms constitute a particularly interesting domain for a better understanding of the mechanisms of categorization.

3Now, there are several conflicting theories on the matter. One emphasizes classification according to the similarity of attributes: it is because individuals demonstrate a finite set of similar prominent features that we recognize them as belon­ging to the same class. Another draws on causal inference, in other words the intuition that a class of organisms possesses a stable essence, either because of the transmission of identical characteristics through reproduction, or because each class represents a unique synthesis of the functions needed to live. Finally, a third theory highlights the prototypicality of organisms, meaning the principle that certain members of a taxon constitute its prototype inasmuch as they are perceived as being the most representative of their class. In one way or another, these theories all use the form of organisms as a classificatory index. Attribute-based classification establishes a relation of belonging to a class founded on a predicative judgment according to which the specific and visible characteristics recognized in any object are the necessary and sufficient condition for it to belong to a class. Classification through causal inference implies the recognition of a functional identity demonstrated by morphologically singularized organs. As for prototypical classification, it is founded on the recognition of a general typical form which acts as a focal model for the class and serves as the basis for denomination. The hypothesis I wish to defend here is that these three cate­gorization theories are complementary rather than ­mutually exclusive. In the domain of organisms, at the very least, completely different mechanisms of categorization and catego­rical reasoning are applied, depending on the classificatory contexts, the targeted levels of generality, and the intended uses of these classifications.

Haut de page

Pour citer cet article

Référence électronique

Philippe Descola, « Natural Forms and Symbolic Classifications », La lettre du Collège de France [En ligne], 7 | 2015, mis en ligne le 02 novembre 2015, consulté le 21 janvier 2019. URL :

Haut de page


Philippe Descola

Anthropology of Nature

Articles du même auteur

Haut de page

Droits d’auteur

© Collège de France

Haut de page