Self-Knowledge

The Introspective Model

According to the Cartesian conception, all mental states are like objects presented in one's own mental arena we are introspectively aware of. In particular, they are transparent to their subjects; that is, if a subject has them, he is immediately aware of them and in a position to judge that he has them. Moreover, a subject is authoritative with respect to them—if sincere, if he judges to be in a mental state M, he is.

The Cartesian model has been widely criticized. First, since Sigmund Freud's discovery of the unconscious, the transparency of mental states has been questioned. Second, it has been noted that animals and infants have mental states yet can't self-ascribe them. Third, the discovery of self-deception, whereby subjects self-ascribe mental states they don't actually have, makes authority founder. Fourth, the Cartesian model would introduce a cognitive faculty—namely, introspection—modeled after sight, which, however, appears difficult to characterize in relation to mental states that aren't physical entities. Finally, Ludwig Wittgenstein pointed out how the Cartesian model would entail the view that psychological language is private to each subject. On the one hand, one could know only one's own mental states and may merely surmise those of others. On the other, if only I can know my current mental states, the reference of my psychological vocabulary will be known to myself only. Hence, whenever it will seem correct to me to apply a given term t to a current mental state M, my use of t will be correct. According to Wittgenstein's argument against the possibility of a private language (either in speech or in thought), this would entail that the distinction between correct and incorrect uses of t would collapse and, with it, the very idea that t could mean anything at all.

David Armstrong has proposed a refined version of the introspective model. In this view, through the operation of an inner subpersonal mechanism, subjects would become immediately aware of their mental states. Working reliably, such a mechanism would also ensure that subjects be authoritative about them.

Armstrong's model has been criticized for considering both transparency and authority to be the result of the correct operation of the inner subpersonal mechanism. These characteristics should be subject to perfectly acceptable exceptions in case the mechanism broke down. Authority and transparency, however, are traditionally considered constitutive traits of self-knowledge and their failure either implies a lack of conceptual competence or rationality or else is a sign that the mental state one fails to have knowledge of is unconscious. Yet—to [Page 677]contrast this view with how we usually think of subpersonal mechanisms—we would never think of blind subjects, whose visual mechanism is impaired and can't therefore see objects presented to them, that they either lack the relevant concepts or are being irrational or else are unconsciously seeing the objects.

...