Abstract

In societies with a non-elementary degree of complexity, we find institutions, social roles, promises, marriages, corporations, enterprises, and the large variety of what we can label “social objects”. On the one hand, we commonly speak and think of such entities as if they existed on a par with entities such as tables and persons. On the other hand, there is a clear link between what people think and how people behave and the social domain. We argue that the widespread “reductionist” approach in social ontology fails to account for both those aspects of social reality, and put forward a new approach. The main idea is that a particular kind of social objects, namely documents – and more generally records of social acts – are the ground of social reality. The fundamental difference from the reductionist approach is that the content of collective intentions will turn out to be not so important in accounting for the ontological variety of complex social realities such as ours.