[Uncertainties about the self and the issue of the proper theoretical model in immunology].

Theoretical immunology constitutes a critical basis of all medical discoveries. Immunology has been dominated since the 1940s by the self/nonself model. Here we try to shed light on the origins of this theoretical model and to show how and why this model has been called into question during the last thirty years. This paper has three aims. Firstly, we explore the sources of the immune self, going upstream from immunology to ecology-biology, psychology and eventually philosophy. Here the key questions : is the immune self really analogous with the philosophical and psychological selves in which it originates? What is the signification and adequacy of such a conceptual borrowing? We suggest that the << self >> vocabulary in immunology is not clear and precise. Secondly, we present the experimental inadequacies of the self/non-self model. We show then how both the vagueness of the term << self >> and these experimental flaws casted doubt on theories of immunology. Among the several models that have been proposed recently, none has attracted a consensus. Some immunologists have even suggested that immunology should rid itself of theorical concerns and concentrate on molecular aspects. This suggestion, however, is unacceptable; hence it is still necessary to find a theoretical framework for immunology. Finally, we try to suggest a way to escape this uncomfortable situation of doubt. The immune << self >> and the immune << system >> (<< network >>) are rooted in strong metaphysical conceptions of identity, the main characteristic of which is to consider the organism as an enclosed and self-constructing entity. By contrast, based on experimental data about immune tolerance and host-pathogen interactions, we propose to consider organisms as open entities. To what theory does this conception lead? What would be the consequences of such a theory with regard to medical aspects?