What Is It That Develops

In seeking to understand the processes that are at the origin of the dynamics of cognitive development, during the first stage of our research (Doise, Mugny, & Perret-Clermont, 1975; Perret-Clermont, 1978/1980; Perret-Clermont & Schubauer-Leoni, 1981), we followed the genetic Piagetian model of children from 5 to 8 years of age. Focusing our attention on the role of social factors, we did not anticipate calling into question the very precise and repeatedly corroborated description of the acquisition of logical operations that characterizes passage to the operative stage in standard conservation tests. Likewise, we also considered as well established the Piagetian distinction between learning process in the strict sense of the term (i.e., the acquisition of local competences that do not modify the basic reasoning structure) and the broader use of the term to refer to the development of general cognitive tools. The purpose of this article is to show that, although this point of departure has effectively allowed us to identify the role of certain social processes-in particular, the interaction between subjects having different points of view-we have, nevertheless, found that this approach has not left unscathed our Piagetian understanding of those cognitive processes just mentioned. It transformed our epistemological understanding of the subject under study: Even when we confronted children with so-called "cognitive tasks," the children always displayed behaviors that had concurrent cognitive and social meanings. We have become accustomed to referring to certain behavioral modifications as "cognitive development" or "conceptual acquisitions" or the learning of "specific cognitive skills." But different observations have drawn our attention to the fact that it is not relevant, from a psychological point of view, to totally disassociate the supposedly "cognitive" elements from those elements that reveal the rules of social interaction or from the social marking of the contexts in which subjects' conduct is embedded. Effectively, as soon as we seek to evaluate the performance of a subject to assess its advancement level, we are confronted with a subject response that derives its meaning not from the "absolute" criteria of a