Non-Cartesian Substance Dualism

ions from the overall psychology of a person. Thus, for instance, the notion of a will without an intellect, or of a language module in the absence of belief and desire, is just plain nonsense. Finally, it will not do to speak of the self ’s psychological states and processes themselves—its beliefs, intentions, experiences, and so forth—as being ‘parts,’ much less as being substantial parts, of it. For this would at best be at all appropriate only on a Humean constructivist view of the self—the so-called bundle theory—which I reject entirely as incoherent. I conclude, therefore, that if the self is a substance, then it must indeed be a simple substance, entirely lacking substantial parts.10 The simplicity of the self goes some way towards explaining its unity, including the unity of consciousness that characterizes its normal condition. Where this unity threatens to break down—as in various clinical conditions such as those of so-called multiple personality, schizophrenia, brain bisection, and so on—we are indeed inclined to speak of a plurality of selves, or of divided selves. But I think, in fact, that such talk should again not be taken literally, and that the psychological unity that most fundamentally characterizes the self is not merely to be located at the level of consciousness. A divided consciousness is, I think, in principle consistent with selfidentity: what is not consistent with this is a radical disunity of beliefs and values, manifested in a radical inconsistency of thought and action. Of course, we all display mild inconsistencies, but no one person could intelligibly be interpreted as possessing the incompatibilities of belief and value that typically characterize two different persons. Now, a complex entity can act in disunified ways because the various incompatible or conflicting activities can be referred to different parts of that entity. Thus a corporate entity such as a firm or a club can act inconsistently because its members may act in conflicting ways. But the actions of the self—those that are truly predicable of it, because they are genuinely intentional, and not merely of the body, such as so-called reflex actions—cannot in this way be ascribed to different elements or parts within the self. So we see that the simplicity and the unity of the self are indeed intimately related, even