Abstract
Norman Malcolm's recent argument against the conceivability of mechanism rests on the claim that purposive
explanations of behavior – that is, explanations of behavior in terms of desires or intentions – are incompatible with neurophysiological explanations of behavior. I admit that intentions or desires can be causes of behavior only if they are necessary for behavior, and, generally, that events can be causes only if they are necessary for their effects (except in cases of over-determination). What I wish to deny is the contention that if neurophysiological states are sufficient for behavior, then desires or intentions are not necessary for behavior. This latter contention seems to me
the crux of Malcolm's argument for the incompatibility of mechanism and purpose; if this contention can be shown to be un acceptable, Malcolm's conclusion that mechanism and purpose are incompatible will remain without support.