Dreaming

1. The ordinary account of dreams

The common-sense theory of dreams is that they are mental experiences; that most of them are delusive; and that they are distinct from other delusive experiences, like hallucinations, in virtue of the fact that they occur during sleep. Our conscious access to dreams is effected mainly via memory; upon waking, most dreams are recalled as having occurred to us, without any interaction between us and them. There are, however, reports of lucid dreams, in which subjects feel that they can deliberate about actions they take in dreams, or are conscious of dreaming as it occurs.

A natural classification would see dreaming as akin to hallucinating and imagining, but also to remembering; as a consequence, some of the philosophical problems posed by dreams will be inherited from the class of problems posed by these other types of experience. You dreamt for instance of a concert given by Rubinstein (a concert you did actually attend), and of a sudden metamorphosis of Rubinstein on stage into a huge rabbit. That this is a dream about Rubinstein and not about another pianist is likely to be explained in the same way in which one explains that your recollections of Rubinstein are recollections of him (see Memory §3). That this is specifically a fancy-like hallucinatory episode goes hand in hand with the fact that its content is similar to that of fancies, and, like the latter, might be disconfirmed by memory, inconsistent with other experiences, and contrary to the laws of the actual world.

Sometimes external circumstances influence a dream; but the lack of a systematic agreement between the contents of our dreams and states of the external world indicates that dreams are mostly generated from within the mind. There has to be something like a dream engine – independent of conscious control – responsible for the production of our dreams (internal movements of the soul, according to Plato, and, for Aristotle, a sort of inertia of sensible impressions, similar to that explaining the formation of after-images).

Common-sense evidence that dreams are datable episodes in the mental life of a subject is given by phenomena such as external stimuli, like the sound of an alarm-clock, entering into the content of the subject’s dreams. During the 1950s and 1960s, experiments by W. Dement (1958), D. Foulkes (1966), N. Kleitman, M. Jouvet and others, established a correlation between dreaming and phases of so-called ‘paradoxical sleep’ (PS), characterized by intense brain activity, muscular atony, and, in some animals, rapid eye movements (hence the term ‘REM sleep’, which is somewhat imprecise, for animals like the mole or the owl have PS but no REMs). PS is periodical – it recurs after phases of orthodox (or NREM) sleep – and takes about one quarter of the time of the entire period of sleep. Subjects woken during PS can easily and vividly recall a detailed, picture-like, ongoing dream.