One of several problems concerning the possibility of mental causation is that the causal potential of a supervenient property seems to be absorbed by its supervenience base if that base and the supervenient property are not identical. If the causal powers of the supervenient property are a proper subset of the causal powers of the supervenience base then, according to the causal individuation of properties, the supervenience base seems to do all the causal work and the supervenient property appears to be futile. Against this consequence it is possible to argue, first, that the relevant properties of causes must be in some sense proportional to the relevant properties of their effects and, second, that the principle of causal closure serving as a premise in the supervenience argument is probably false. The constraint that the relevant properties of causes should be proportional to the relevant properties of their effects together with the falsity of the closure principle leads to a restoration of the causal efficacy of supervenient properties.

In the center of this paper is a phenomenological claim: we experience ourselves in our own doings and we experience others when we perceive them in their doings as active in the sense of being a cause of the corresponding physical event. These experiences are fundamental to the way we view ourselves and others. It is therefore desirable for any philosophical theory to be compatible with the content of these experiences and thus to avoid the attribution of radical and permanent error to human experience. A theory of ‘subject causation’ according to which the active subject continuously and simultaneously causes physical changes is sketched. This account is—according to the phenomenological claim defended—compatible with the content of our daily experiences in doing something and in observing others in their doings and it has a number of further more theoretical advantages: it does not touch the autonomy of neurophysiology and it is compatible with a thesis of supervenience of the mental on the physical. It does however require a weak version of subject-body dualism.

Cet article conteste l’opposition wittgensteinienne entre explication causale de l’action et explication psychologique par des raisons d’agir, ainsi que la postérité de ce distinguo, notamment chez Davidson, Putnam et Fodor. Ces derniers auteurs sont présentés à travers le paradigme qu’ils partagent (le physicalisme non réductionniste : PNR), dont on montre ici la parenté avec l’émergentisme britannique des années 1920. Je fais appel à plusieurs arguments de Kim, qui reproche au PNR de ne pas être cohérent avec son matérialisme de principe. Kim propose un modèle réductionniste d’explication de l’agir humain (la réduction fonctionnelle) qui me semble concilier identification des bases neurales et non-élimination du mental. Je signale cependant deux modalités psychologiques qui ne semblent pas devoir être traduites en termes de fonctions (caractérisées par leurs relations causales) : l’aspect qualitatif du vécu conscient, et le caractère normatif de nombreuses conduites.

A survient sur B si et seulement si il est possible que B existe, et qu'il est impossible que B existe sans que A n'existe. L'usage de la notion par Armstrong amène à se poser des questions sur l'ontologie des entités survenantes et sur la relation de survenance : les entités survenantes sont-elles réelles ou non ? Sont-elles quelque chose de plus que la base de survenance ? Quelle est la nature ontologique de la relation de survenance ? L'auteur examine les différents propos d'Armstrong sur le sujet et discute quelques options possibles. B.L.B.