Chapter
Fifteen
Dreaming and beyond
Just as his head came off, he awoke to find that the headboard had fallen on his
neck (Maury, 1861, pp. 133–134). He proposed that dreams do not happen in real
time but are entirely concocted in the moment of waking up. This theory became
popular, perhaps because so many people have the experience of dreaming
about a church bell ringing or a wolf howling, only to wake to the sound of their
alarm clock. It is also psychologically plausible in the sense that humans are very
good at constructing stories and quick at confabulating. But it is not true.
In the 1950s, people sleeping in the lab were asked to describe their dreams
and they gave longer descriptions the longer they had been in REM sleep. Other
experiments tried incorporating external stimuli into dreams. Sounds, taps on
the skin, flashes of light, and drips and sprays of water have all been used, and
when they don’t wake the sleeper they can sometimes influence dream content,
allowing dream events to be timed. These results, as well as the animal studies
described above, show that dreams take about the same time as would waking
events. All this suggests that dreams are not concocted in a flash on waking up,
but really do take time.
Other responses to these doubts are more subtle. Dennett provides a selection of
fanciful theories playing with the relationship between experience and memory.
On the ‘cassette theory of dreams’, the brain holds a store of potential dreams
recorded and ready for use. On waking from REM sleep, a ‘cassette’ is pulled out
of storage, to match the sound of the alarm clock if necessary, and hey presto,
we seem to have been dreaming. On this theory, there are no real dreams. There
are no events or images presented ‘in consciousness’, but only recollections of
dreams that were never actually experienced. ‘On the cassette theory it is not like
anything to dream, although it is like something to have dreamed. On the cassette
theory, dreams are not experiences we have during sleep’ (1976, p. 138).
The point of this theory is not that it might be literally true (even if we update the
cassette to an MP4), but that it provides the basis for another possibility – that the
equivalent of cassette dreams might be composed during the REM period prior
to waking. We can now compare the normal theory that dreams are conscious
experiences during sleep, and the new theory that dreams are composed uncon-
sciously during sleep and then ‘remembered’ on waking up. The question is this.
Could we ever tell which was right?
The answer seems to be no. It is no good asking dreamers whether their dreams
really occurred ‘in consciousness’, because all they have is their memories and they
will always say ‘yes’. And it is no good looking inside their brain because even if we
could see the neural events that correlate with imagining cups of coffee or trying
to walk, we still have no way of finding out whether those neural events were ‘in
consciousness’ or not. There is no special place in the brain where consciousness
happens, or, in terms of Dennett’s later theory (1991), there is no Cartesian the-
atre in which the dreams either were, or were not, displayed. We are left with two
theories that seem empirically indistinguishable, so this is again ‘a difference that
makes no difference’.
But the tendency to distinguish conscious from unconscious elements of dreaming
remains common – even amongst researchers who claim consistency with multiple
drafts theory. One account models the phenomenology of dreaming as a feedback
system involving memories, interpretive processes brought to bear on them, and
‘The conscious output
of the dream is what
will be recalled by the
dreamer’(Cicogna and Bosinelli, 2001,
p. 38)