Abstract

This article deals with the role of time in causal models in the social sciences, in particular in structural causal modeling, in contrast to time-free models. The aim is to underline the importance of time-sensitive causal models. For this purpose, it also refers to the important discussion on time and causality in the philosophy of science, and examines how time is taken into account in demography and in economics as examples of social sciences. Temporal information is useful to the extent that it is placed in a correct causal structure, and thus further corroborating the causal mechanism or generative process explaining the phenomenon under consideration. Despite the fact that the causal ordering of variables is more relevant for explanatory purposes than the temporal order, the former should nevertheless take into account the time-patterns of causes and effects, as these are often episodes rather than single events. For this reason in particular, it is time to put time at the core of our causal models.