In one of their few co-authored articles, “On some primitive forms of classification,” Émile Durkheim and Marcel Mauss set out to establish that humans do not classify by an internal necessity of their understanding, nor by reproducing an observable order in nature, but by projecting onto objects a model of classification provided by society and its divisions. The reception of this article over time by anthropologists has been paradoxical: criticized for its ethnographic flaws and invalidated by later research in cognitive psychology on the categorization of natural objects, it nevertheless continues to arouse keen interest because of the hypothesis it proposes of a link between the classification of things and the classification of social groups.