No abstract

A good deal of strategic choice has been given back to organizations, which have become actors of their (only partial) compliance with institutional demands that they in turn contribute to shaping. The reported case of the successful modernization of the French cancer centers and their reinstatement as the leaders in their field contributes to a better understanding of the role of leadership in institutional change because it demonstrates a positional approach to institutional leadership. Cancer centers' reformers were both central, because they were placed at the intersection of several potentially interdependent organizational fields or institutional spheres, and marginal to most but not all of them. This particular position of the change-entrepreneurs, with its relational constraints and also its resources, enabled them to initiate a successful drive for the transformation of the field of cancer care and also greatly explains the particular form it took. Our analysis underscores the interactive nature of institutional change, where the motor of change simultaneously structures and is structured by the process it is driving and where the initiators of reform have to create their proper and specific combination of old and new in order to build an innovative dynamic.

Comment comprendre la vigueur des controverses qui a accompagné aux États-Unis la montée de la thématique « néoinstitutionnaliste » dans les sciences sociales américaines depuis une vingtaine d'années ? Réagissant à l'article de Peter Hall et Rosemary Taylor sur « La science politique et les trois néo-institutionnalismes » (numéro 3-4, juin-août 1997 de cette revue), l'auteur souligne que la réalité visée par cette « redécouverte » du rôle des institutions comme variable indépendante en sociologie et en science politique doit être reliée aux trois significa-tions différentes du terme « institutions ». En portant plus particulièrement son attention sur l'institutionnaitsme sociologique, il critique sa propension à surestimer la rationalité non instru-mentale et sa difficulté à reconnaître l'hétérogénéité irréductible du champ social.