Increasingly scenarios are used as an important component of long-term planning. But not all scenarios are equally valid and equally useful for the decision-maker. Defining a scenario as a "synthetic process which stimulates step by step and in a plausible fashion a series of events which eventually lead a system to a new state", this study examines two kinds of scenarios: exploratory, where the inquiry proceeds from the present situation to a future one, and normative, where the search proceeds from a desirable future to the present. For each type of scenarios three sets of theoretical problems are examined: 1) the role of values, which must be explicitely recognized and used as such; 2) the concept of causality, which in a scenario has to be dealt with differently than in an "ordinary" scientific research, 3) the problem of time and the need to break the linear conception of the link existing between events. Finally, the study examines a number of practical tools and criterias (coherence, interaction, …) with which to build and to judge scenarios.
Veuillez télécharger l’article en PDF pour le lire.