pp. 35-59

Whereas existing studies of how public problems are constructed emphasize markedly public episodes in the process and are likely to take a cognitive approach, this text seeks to move beyond those limits and show a possible path to renewing such study. It first undertakes to show that problem-defining processes are concomitantly cognitive and social and that they involve practices of confrontation and/or negotiation between different actor groups. It then suggests the usefulness of shifting the focus from clearly public arenas to the more « discreet », less visible spaces in which public policy is produced. The initial areas of research – collective risks and public health issues – are used here to open up avenues for studying all categories of public problems.