Une nouvelle interprétation des phénomènes de schizophrénie débutante est ici proposée sur la base d’un modèle mimétique et attributionnel du soi-agent qui permet de concevoir une construction du soi unitaire car sociale de bout en bout, y compris dans la dimension d’agence. Le vécu caractéristique du sujet qui en vient à se sentir au centre de l’attention de tous y apparaît comme la version pathologique d’une « centralité » normale du soi relative à ce qu’il est convenu d’appeler le « miroir social ». L’hypothèse présentée ici est qu’un déficit d’attributions causales de la part de l’entourage tend à susciter une abréaction du schème du soi qui est alors porté à « halluciner » des signes d’attributions dans les moindres attitudes, attentions ou messages de l’entourage. Signes venant alimenter le soi non pas sous le rapport de l’estime de soi mais sous le rapport du sentiment d’agence nécessaire pour se sentir exister comme être animé, comme être capable d’un agir sur le monde. Un des avantages de cette conception est qu’elle permet d’éclairer la fragilité des adolescents sous le rapport de la schizophrénie.

A new interpretation of the onset of schizophrenia is presented based on a mimetic and attributional model of the self as agent. Two separate independent perspectives on schizophrenia will be analyzed and criticized in relation to this model. The first one from the works of Henri Grivois, inspired by the general anthropology of René Girard is therefore oriented toward the mimetic dimension of interactions. The critique will focus on the gratuitous reduction of imitation to the domain of bodily motor functions when a much wider understanding of the concept is needed. The second perspective, proposed by Richard Bentall et al., deals with the attributional style of paranoids and is therefore mainly cognitive. The critique will focus on the reductionist and solipsistic vision of a self unduly equated with self-esteem. I intend to show that their respective mimetic and attributional foci are both needed in order to have a better understanding of the delusional subject, for these two dimensions define the inter-subjective space in which the self is built. As a matter of fact, the self is traditionally described as having two fundamental components, the me or object self and the I or agent self. But only the first one, the me, has been thoroughly investigated by the so-called “social mirror” research tradition which originates in Cooley’s “looking glass self” hypothesis. The “I” has constantly eluded scientific investigation and has therefore been virtually ignored by psychologists in the twentieth century. Many researchers have even proposed abandoning it to the realm of philosophy. However, as soon as the social mirror metaphor is understood as the imitation of others’ attributions to the object self, we may conceive of the agent self as the product of these imitations of attributions along the causal dimension. The “I”, characterized by the feeling of being the first cause of one’s own behavior, would therefore be of no less social origin than the “me”. Attributions of intention, responsibility or causality made by others to the self, when mimetically interiorized, will feed the causal self, the “I”. This unitary and psychological conception of self allows a new interpretation of some characteristic aspects of the onset of schizophrenia. The typical feeling of being a focus of attention of an entourage sometimes extended to the whole world can now appear as just the pathological version of the healthy “centrality” of the self in the « social mirror ». The question is: how does the bifurcation of the dynamics of self towards pathology happen? I hypothesize that if the causal attributions that normally “feed” the self along the agency dimension are too scarce, this will lead the self schema to an “abreaction” which consists of “hallucinating” personal causal attributions in the minute attentions, attitudes and messages provided by anonymous public social interactions. All these signs, when perceived and delusionally interpreted, will help the subject maintain the vital feeling of being a “cause” capable of “effects” which characterizes animated beings as opposed to pawns. Hence, the fact that paranoids are prone to perceive others as having malevolent intentions towards them, far from being explained by Bentall’s hypothesis – since a bad social self image doesn’t improve self-esteem – can be interpreted as their longing for causal attributions. In other words, accusations, malevolent attention and intentions are good for the feeling of agency. This conception might shed some new light on the specific susceptibility of adolescents regarding schizophrenia.