Today as in the past, most often crises take people by surprise. This fact has recently provoked strong criticism of the ability of an economic theory to predict crises, to understand their course and to establish solutions to mitigate their effects. History can thus serve as a reservoir of facts and experiences, and the use of a broad chronological perspective has been recently highlighted as essential to providing a wider, comparative knowledge of past crises. Recent economic historiography has highlighted the importance of studying financial and commercial crises alongside agrarian and demographic crises, as well as questioning specific aspects of these shocks. Another important dimension stressed by recent historical studies is the importance of recognising that crises in the past occurred against a background in which uncertainty was the norm. In societies that experienced various forms of ordinary uncertainty (linked for example to the ‘dead’ season in food or textile production), crises constitute peaks of exceptional uncertainty.