THE NARRATIVE OF DREAM REPORTS
PART 1
A Thesis submitted for the degree of Doctor of Philosophy
by
Mark Thomas Blagrove
Department of Human Sciences, Brunel University
March 1989
BRUNEL UNIVERSITY, UXBRIDGE.
DEPARTMENT OF HUMAN SCIENCES
MARK THOMAS BLAGROVE
THE NARRATIVE OF DREAM REPORTS (1989)
Two questions are addressed:
1) whether a dream is meaningful as a whole, or whether the scenes are
separate and unconnected,
and 2) whether dream images
are an
epiphenomenon of a functional
physiologicaL process of REM sleep, or whether they are akin to waking thought.
Theories of REM sleep as a period of information-processing are
reviewed. This is Linked with work on the reLationship between dreaming and
creativity, and between memory and imagery. Because of the persuasive evidence
that REM sleep is implicated in the consolidation of memories there is a review of
recent work on neuraL associative network models of memory. Two theories of dreams
based on these models are described, and predictions with regard to the above two
questions are made. PsychoLogicaL evidence of relevance to the neural network
theories is extensively reviewed. These predictions are compared with those of the
recent application of structuralism to the study of dreams, which is an extension
from its usual field of mythology and anthropology.
The different theories are tested against four nights of dreams recorded
in a sleep Lab. The analysis shows that not only do dreams concretise waking
concerns as metaphors but that these concerns are depicted in oppositionaL terms,
such as, for example, inside/outside or revolving/static. These oppositions are
then permuted from one dream to the next until a resolution of the initial concern
is achieved at the end of the night. An account of the use of the single casestudy methodology in psychology is given, in addition to a replication of the
analysis of one night's dreams by five independent judges. There is an examination
of objections to the structuralist methodoLogy, and of objections to the paradigm
of multiple dream awakenings.
The conclusion is drawn that dreams involve the unconscious dialectical
step-by-step resolution of conflicts which to a great extent are consciously known
to the subject. The similarity of dreams to day-dreams is explored, with the
conclusion that the content of dreams is better explained by an account of
metaphors we use when awake and by our daily concerns, than by reference to the
physiology of REM sleep. It is emphasised that dreams can be meaningful even if
they do not have a function.
ACKNOWLEDGEMENTS
I am grateful for the help of my supervisors, Professor Adam
Kuper, Professor Liam Hudson, and Professor Michael Wright. I also
wish to thank Professor Alan Stone and Dr. Vernon Dobson for many
useful suggestions. This research was funded in part by a grant from
the Ann Murray Award Fund, which was greatly appreciated and very
helpful. The texts of the dreams of the subject KJ were kindly
supplied to Professor Kuper by Professor Rosalind Cartwright.
THE NARRATIVE OF DREAM REPORTS
Introduction - Two Questions about Dreams
p.1
SCIENTIFIC TRADITION
Chapter 1 - The Natural History of REM Sleep
p.5
Chapter 2 - The Human Experience of Dreaming
p.12
Chapter 3 - Older Physiological Theories of Dreaming
p.29
Chapter 4 - Newer Physiological Theories of Dreaming
p.33
Chapter 5 - Experiments on REM Sleep and Memory
p.43
Chapter 6 - Imagery, Creativity, and Dreams
p.57
Chapter 7 - The Neural Basis of Memory
p.66
Chapter 8 - Three Current Theories of the Function
of Dreaming
p.93
Chapter 9 - The Scientific Study of the Meaning of
Dreams
p.115
PSYCHOANALYTIC TRADITION
Chapter 10 - Freud and the Experience of Dreaming
p.137
Chapter 11 - Dreams and Metaphors
p.141
ANTHROPOLOGICAL TRADITION
Chapter 12 - Structuralism and Dreams
p.150
EXPERIMENTAL PREDICTIONS
Chapter 13 - A Comparison of the Phenomenological
Predictions of the Major Dream Theories
p.171
THE EVIDENCE
Chapter 14 - Night 1, Dreams and Analysis
p.222
Chapter 15 - Night 2, Dreams and Analysis
p.239
Chapter 16 - Night 3, Dreams and Analysis
p.236
Chapter 17 - Night 4, Dreams and Analysis
p.274
DISCUSSION
Chapter 18 - Commentary on the Methodology of
this Study
p.292
Chapter 19 - Conclusions
p.339
Postscript - Suggestions for Further Research
p.381
References
p.384
Appendix 1 - 'Unlearning' has a stabilizing effect in
collective memories. Hopfield, Feinstein
& Palmer (1983)
Appendix 2 - The Complete Dream Texts
p.426
p.428
INTRODUCTION
There are two questions that remain unanswered in current
dream research.
1)
Do individual dreams have meaning as a whole, with each
scene related to the other scenes, or are they collections of separate
scenes which do not influence each other and which the conscious mind
then sometimes ties together?
2)
Are dreams related to low level neural activity in the
brain, such as neurochemical or synaptic changes, or are they related
to higher brain activity and akin to waking thoughts, albeit in a
different language?
Each of the possible answers to the two questions is
coherent, and each has cogency because of its everyday counterparts.
So, the first question can be paraphrased as: is a dream (or a series
of dreams) like a play, which progresses towards a conclusion, or is
it like a revue, with separate unrelated snippets, incidental pieces
and extemporary elaborations? The second can be paraphrased as: is a
dream like gurglings in the stomach, which do have their causes and
meanings, or is it like a composed song?
I consider that these questions may be answerable by a
detailed study of the phenomenology of the experience of dreams, and
that such data may be used to decide between some theories of dreaming
which are presently postulated.
I relate in chapter 1 some facts about the physiology and
biology of dreaming which will help us in an initial evaluation of the
theories presented later. In chapter 2 I give evidence about the
experience of dreams for humans, and I suggest that an important
1
aspect to be explained is the narrative of the scenes. This choice of
a feature of dreams which is often ignored in the emphasis on their
bizarreness and creativity mirrors the study by Rechtschaffen (1978)
of 'single-mindedness' as an almost continuously present and equally
ignored facet of dreams. I examine in this chapter the physiological
concomitants of the images of dreams, and in chapter 3 review old
theories of where these images come from.
Chapter 4 shows the problems and strengths in certain
theories which hold dreams to be an epiphenomenon of Rapid Eye
Movement Sleep, and introduces the idea of dream sleep being a state
of information-processing, separate and different from the processing
that occurs during waking life. Chapter 5 provides evidence that our
time of unconsciousness during REM sleep has beneficial effects on
memory storage and examines common criticisms of this idea. Chapter 6
expands on the conclusions of the previous chapter with evidence
memories are creatively manipulated during REM sleep, and relates this
activity to the fact that dreams occur as images, that REM sleep
provides a state of mind different from that of waking life.
Chapter 7 describes theories of memory storage in
associative neural nets, which are composed of many interconnected
artificial neurones. Two types of such networks are reviewed, with the
conclusion that their efficiency is increased if they have periodic
changes of state, during which their activity follows different rules
than normal. This serves as an introduction to chapter 8, which gives
details of two theories of human and mammalian dreaming which are
based upon these two neural network theories. I show that the theories
as published are not completely applicable to biological dreaming, but
2
that they do explain much of the data reviewed in the first 6
chapters, and that they have a similarity with a third,
psychoanalytically based theory of dreams.
Chapter 9 argues that a solely psychological account of
dreams is needed as opposed to the psychophysiological work of the
previous chapters. A review is given of the use of case studies and
content-analysis in psychology. Chapter 10 explains the different
facets of dreams from the standpoint of classical psychoanalysis.
Chapter 11 is concerned with Rycroft's revision of such an account,
based on dreams as a source of enlightening metaphors rather than
obscuring disguises of the unconscious. This chapter introduces the
notion that, as Jung (1968) wrote, 'the dream does not conceal; we
simply do not understand its language'.
In Chapter 12 I explain the suggestion that one method of
analysing myths, totems, \iris 1riip,
aiu various other languages, t1nat ol
Structuralism, may be applicable to the study of dreams. I give
examples of the application of the method in its original field, that
of social anthropology, and give evidence that the vicissitudes in the
development of social anthropology over the last hundred years are
similar to the changes in the theories of dreams over the same period.
In chapter 13 I compare the seven major theories from
chapters 3-12 from the stance of their phenomenological predictions
about narrative and condensation.
Chapters 14-17 contain the reports of every dream of the
night for a subject who attended a dream lab for 4 separate nights.
(The full-length reports are in the second appendix.) I analyse this
3
data in the light of the different theories, measuring the evidence
against the predictions made.
However, there are problems with the whole project of
interpretation and the discovery of what thought and meaning is
'really there' in texts. In chapter 18
I compare the scepticism
towards exegesis that Structuralism and Psychoanalysis both encounter,
and then reply to criticisms that the findings of this piece of dream
lab research, and other work using multiple awakenings, are
artefactual to the experimental regimen used. An account is given of a
replication of the analysis of night 4 by five independant judges. The
differences between this experiment on reliability and proposed ones
on validity are explained.
In chapter 19 I show what theoretical conclusions can be
drawn from the results of the previous 4 empirical chapters. I
conclude with an account of the features of the mentation present
during dreams and show its similarities to day-dreaming. Experiments
are proposed to explore this similarity.
Appendix 1 is the full text of the paper by Hopfield,
Feinstein and Palmer (1983) examined in chapter 7.
Appendix 2 is the full texts of the dreams and interviews
used in chapters 14-17.
(Note on terminology: although images and thought can occur
during non-REM sleep, and the subject can be said in such instances to
be dreaming, throughout this thesis 'dreaming' and 'dreams' refer to
mentation that occurs during REM sleep alone.)
4
CHAPTER 1
THE NATURAL HISTORY OF DREAI1ING
The Discovery of REM Sleep
In 1951 Aserinsky and Kleitman discovered slow rolling eye
movements in sleeping infants, which occurred in association with
periodic body movements which had been studied by other workers.
Dement and Kleitman (1957 a&b) proceeded to find the same movements,
lasting about 3 to 4 seconds, in adults, but also noticed periods of
rapid jerky eye movements during which there would be short pauses
with the eyes still. They also found periodic changes in the sleeping
subjects' EECs, with an awake-like desynchronized pattern of brainwaves during rapid eye movement sleep alternating with large slow
'delta waves' occurring during non-rapid eye movement sleep.
This REM sleep is dependent on the phylogenetically older
areas of the brain, notably the rostral pons, and so a decorticated
cat shows only evidence of REM sleep, without NREM sleep (Jouvet,
1969). Furthermore, Velluti (1985) found that there is a phasic oxygen
consumption change during REM sleep in the reticular formation,
hypothalamus, amygdala and cerebellum, but not in the neocortex or
white matter, which are phylogenetically recent.
With the exception of two specie , all mammals so far
studied show alternating periods of REM and non-REM sleep. Nothing
resembling sleep is found in animals below the reptiles, except a
periodic decrease in movement. One of the more primitive surviving
reptiles, the tortoise, shows no evidence of it. Chameleon lizards
have REMs themselves during sleep but brain electrical activity is
5
unchanged. Reptiles do sleep yet have no REMs, while birds do have REM
sleep.
The Ontogeny of REM Sleep
Roffwarg, Muzio & Dement (1966) found that human neonates
spend around 16 hours of the day asleep, of which half is in stage
REM. However, there are problems defining REM sleep at this stage
because its constituent parts (that is, REMs, desynchronized EEC, lack
of K-sleep spindles in the EEC, and total loss of muscle tone) are
often dissociated, rather than occurring together. While the total
amount of sleep decreases throughout life the amount of REM sleep
decreases faster. Among infants duration of NREM sleep is quite
consistent, while the REM periods are more variable. This would fit in
with the hypothesised linking of NREM sleep with bodily needs and REM
sleep with the needs of the brain, which is as yet functioning
imperfectly. There is increased movement during neonates REMS with
sucking and even grimacing in evidence, the latter not yet being seen
in waking life. This finding is similar to that of the posturing shown
by cats in REMS that have had their motor system disinhibited
surgically (Jouvet, 1969).
The decrease in dream time found in the growing human (which
is also found in other mammals) must be due co one of the following
reasons:
1) REM sleep becomes more efficiently organised, with a more
orderly switching on and off, and a decrease in Intermediate sleep (a
period of atonia, desynchronization of EEC and rolling slow eye
movements). Valatx, Jouvet and Jouvet (1964, reported in Clemente,
6
Purpura & Mayer, 1972, chp. 13) state that the onset of slow wave
sleep is contemporaneous with the achievement of cortical maturation.
Exemplifying the immaturity of the sleep system is the presence of
Intermediate Sleep in some animals (which has only some of the
characteristics of REM sleep), the 'trace alternant' EEG burst
pattern, which is present in humans for the first 1-2 months, the
absence of a stable distribution of sleep and wakefulness for the
first 1-3 months, and the amplitude increase in EEC with age through
infancy, which then decreases throughout life. The sleep periods
consolidate and lengthen in the first months, another indication that
the brain is approaching an optimum state by maturation rather than
that the erratic sleep and high REM time are actually functional.
There is a shift in balance of inhibition of neurones to that of
excitation occurring in the first 4-5 weeks, when there is ontogenesis
of sleep spindle bursts. McGinty (ibid. pp.276-278) opposes this idea
of brain immaturity with the finding that both quiet sleep and active
sleep are reduced by stimulus deprivation in kittens, so there may be
a functional reason for the prevalence of REM sleep, but his results
may be due to other factors such as eating changes in the animals
used. Ornitz (Development of sleep patterns in autistic children, in
ibid.) notes that auditory evoked potential to clicks was high in REM
sleep in infancy, suggesting that phasic inhibition of the nervous
system is not fully developed in the first year. Phasic inhibition
thus develops at a time of great decrease in REM time. Ornitz cites
Pompeiano (l967a&b) as showing that REM phasic events (such as the
actual REMs which occur during the period of REM sleep) depend upon
the integrity of the vestibular system, and that this dependence
7
matures from birth. Also note that because in the newborn REMS periods
follow waking, and it is only later that NREM periods intervene, it
may be that young animals have so much REM sleep because they have so
little NREM sleep.
2) REM sleep has a function in the younger animal but later
on this is not so necessary. Dreyfus-Brisac (1967, in Clemente et al.
ibid. pp.202-203) found few REMs at 24-26 weeks gestation, sparse eye
movements at 28-30 weeks, with dense eye movements at 32 weeks. The
major decrease in REM sleep occurs in the first 3 months after birth.
Roffwarg, Muzio & Dement (1966) suggest that REM sleep for the fetus
is a time of stimulation of the visual system (all other senses in the
uterus will be stimulated by the environment), for there is
substantial myelinization of the visual system before birth which must
be explained. The problem remains, though, as to why REMS should
persist in later life; after all, various reflexes in the newborn
(such as Babinsky's, on the feet) do die away and never return.
Dreaming may thus simply be like all other instances of childhood
overactivity and playfulness, which also mellow with time, but the
problem is why it does not fade completely.
3) Its functions are taken over by thought, imagery and
fantasy. See Cartwright & Monroe (1968) in which waking fantasy was
found to reduce post-deprivation REM rebound more than mathematical
reasoning on waking did. Similar to this suggestion is the claim that
eidetic imagery is lost after childhood because of the growth of
propositional thinking.
8
The Purpose of REM Sleep
Some have suggested that the stimulation from the brain stem
that mammals receive during REM sleep (termed 'PGO' stimulation) may
not contain any useful information, but is instead random stimulation
used to avoid the brain becoming comatose during its long periods of
useful NREM sleep. It may be objected that the brain would have
evolved a system of alternating NREMS-awake-NREMS-awake-NREMS-awake
rather than waste 2 hours per day (in the case of humans) randomly
stimulating itself, or have very short periods of REMS, as birds do.
Alternatively, Berger (1969) provides evidence that REMS has the
function of tuning the eye muscles in stereoptic animals (although the
mole has a large REM time!). That NREM sleep is itself important is
shown by experiments in which animals are deprived of all sleep; they
subsequently have more REM sleep than usual, but have an even greater
increase in the amount of stage 4 NREM sleep, which is made up in
preference to REM sleep in this 'rebound' paradigm (Agnew, Webb &
Williams, 1964, 1967). Another possibility is that the stimulation
from the brain stem (pons) is analogous with that obtained during
waking life from the Ascending Reticular Activating System. When we
are awake it is not the stimuli from the environment which keep us
awake, they merely impinge upon a brain independently made receptive
by the ARAS stimulation. It may be that PGO bu sts maintain the cortex
in a state in which vivid imagery is possible; in the next chapter I
will provide further evidence for this position. The two systems are
connected via the hippocampus, which has desynchronized activity when
the subject is awake, and hippocampal theta wave activity when the
subject is in REM sleep. Grastyan (1959) has found that hippocampal
9
theta rhythm also inhibits the activity of ARAS. PGO bursts are found
to occur before REM bursts, and are detected by implanted electrodes
in either the pons, lateral geniculate nucleus, or visual occipital
cortex. (This method is not used for humans.)
Winson (1985) notes that hippocampal theta is sometimes
generated during waking life in mammals, but only during speciesspecific activities involved with biological survival. The same
neurones are involved in both cases, and he proceeds to construct the
theory that, as the hippocampus is involved with the processing of
refined information from the neocortex and back either to the
neocortex or to the limbic system, REM sleep is a period of 'off-line'
information-processing, concerned with the consolidation of memories.
As an incidental piece of evidence, he notes that the two extant
examples of monotremes (the most primitive mammals, which hatch their
young from eggs), the duck-billed platypus and the spiny anteater, do
not have REM sleep. He also notes that the latter has a very large
frontal-cortex, proportionately greater than that of humans (1985,
p.57). The former has an even larger neocortex, measured as a
percentage of brain volume, than does the anteater (reported by Tauber
and Glovinsky, 1987). He hypothesises that REM sleep evolved in order
to facilitate the integration and Inemorising of experience without the
need for a massive prefrontal cortex which wo id have been performing
this function during waking life at the same time as its function of
'executive' of the whole brain.
After this brief summary of some of the major points of the
natural history of REM sleep, and an introduction to the idea that
REMS is a time of information-processing, we now proceed to a
10
description of the experience that humans have mainly during REMS,
that of dreaming.
11
CHAPTER 2
THE HUMAN EXPERIENCE OF DREAMING
Experiments on the Phenomenology of Dreaming
Dement and Kleitman (1957 a&b) found that subjects awakened
during REMS were much more likely to report that they had been having
a dream than were subjects awakened during NREMS. Foulkes (1962)
showed that not only do dreams occur at all stages of sleep to a much
greater extent than had been realised in earlier work, implying their
occurrence without concomitant PGO stimulation, but also that the
contrast between REM and NREM sleep experiences does not concern
bizarreness, which is the stereotypical hallmark of dreams, but rather
the following formal differences:
REMS reports NREMS reports
% yes
% yes
Subject's emotion
50
32
Other's emotion
55
31
Visual
90
66
Clear visual imagery
80
62
Physical movement (self)
67
38
42
21
Locomotion (self)
45
25
Locomotion (others)
34
57
37
17
63
33
1
17
Only one other character
More than one part
Scene shifts
Memory processes
12
(Also, see McCarley & Hoffman, 1981, for further evidence
relating REMs to dream activity, and Duffy & Lombroso, 1968, about the
effect of saccadic eye movements in waking life on visual imagery.)
Rechtschaffen (1978) and Fiss (1986) state that the ego in
REM sleep is less critical and more passive than the ego in NREM
sleep, and can only follow one train of thought, while Hunt (1986)
suggests that what is distinctive about dreams is what they lack; they
show single-mindedness and loss of context (context only being
restored on waking). He writes that they can be as creative or as
pragmatic as can waking thought.
The REM reports were longer and contained less references to
the subject's work activities, less continuation of a prior dream
thought and much less evidence of memory processes. It is the NREM
reports that have a greater correspondence to waking life. (These
results contradict the earlier theory of Dement and Kleitman, 1957b,
that NREM reports were memories of preceding REM dreams).
In addition to the data of the above table note that Hunt
(1982) reports that sudden changes of scene are only found in 20% of
home recall dreams, and that condensation (a dream image containing a
mixture of two parts of waking life, such as the image of a character
with the face of one friend and the gait of another) is very rare,
occurring in less than 10% of the reports. We therefore see that the
PGO stimulation, if it is involved at all, is correlated with a
greater vividness, affect and complexity, rather than bizarreness.
As noted briefly in the last chapter the PGO input may be
especially needed by the occipital cortex to maintain its ability to
produce pictures, for the effect of occipital lesions is to remove
13
imagery but still allow dream-like thoughts. Of interest here are the
doubtful findings of some studies aimed at correlating dream images
with specific eye movements during REM sleep; the PCO bursts are thus
not indicative of any particular dream image, or of a single fast
movement or a scene discontinuity. Therefore, PCO stimulation is as
unrelated to particular experiences as is ARAS stimulation. Now,
sensory stimuli do not directly keep the cortex awake, the ARAS does
that, and by analogy the content of dream images may not be sufficient
to ensure that the image occurs; a general facilitation may be needed,
as illustrated next.
Physiological Phasic Events and the Dream Experience
REM SLEEP
AWAKE
endogenous
s ens ory.
dream
stimuli
stimuli
EYES- L.CENIC.N.
ARAS
I
PONS
As the case of comas in patients with no cortical damage
shows us, the brain must be maintained in a receptive state, as well
as receiving stimuli, in order to be active. Hersch, Antrobus, Arkin &
Singer (1970) found that an injection of noradrenaline during NREN
sleep increased salience (defined as vividness, emotionality, and
bizarreness) and Chase (1972) reports a significant correlation
14
between cardiac and respiratory irregularity and dream salience
(though this tells us nothing about the direction of causation).
However, autonomic nervous system arousal decreases from earlier to
later REM periods (Snyder, Hobson, Morrison & Goldfrank, 1964) so it
is not possible to say that the difference between REM sleep and NREM
sleep is simply one of arousal. A specific factor involving the
occipital cortex may be needed, a role that PGO stimulation may
perform. This is a different proposal from that of Brooks (in Webb,
1973), who says "... a stimulus of brainstem origin prepares
the
visual system for the arrival of an entirely new image, and the
process manifests itself as a PGO wave... As a result the net image
will be perceived as
a new one, rather than as an initially
blurred or displaced version of the preceding image... the neuronal
events associated with the imagery occur in the intervals between PGO
waves." Rechtschaffen (1973) also suggested that the PGO burst breaks
up the flow of information. If PGO stimulation has the function of
aiding the
vividness of dream imagery, this would indicate the
importance of dream imagery itself, and suggest that dream images are
purposeful rather than epiphenomena of the physiology of REMS.
Supportive of the view of PGO stimulation as a random jolt
is Cohen, Edelman, Bowen and Dement (1972) and Steiner and Eliman
(1972) who report that Intracranial (electrical) Self-Stimulation
reduces REM rebound after deprivation, while REM deprivation increases
ICSS activity. (This work also relates the random stimulation more to
overall neurochemistry than to individual memories, and hence is
contrary to information-processing theories.) The opposing work of
Hobson and McCarley (1977), who propose that PGO stimulation is
15
structured has not been proved. They propose that the bursts contain
basic motor and spatial information around which the dream is built
(with affect being laid on later). Support to some extent for this
view comes from Lerma & Garcia-Austt (1985), who found a relationship
between hippocampal theta rhythm and phasic events - (the former are
connected with movement in waking life). Also note that Gadea-Ciria
(1977) found that decorticate cats had a dramatic change in PGO
sequencing, becoming grouped in bursts separated by longer intervals,
which "suggests that the cerebral cortex is involved in the regulation
of patterns of PGO sequences." (ibid.) REM sequences themselves are
also structured, as shown by Weber, Muzet, Schieber & Lienhard (1983).
Other work on the pons shows its relationship to structured, nonrandom, information: Glickstein and Gibson (1976) showed it to be a
relay station for sensory information that guides motor activity. They
found that visual information, which reaches the occipital cortex via
the lateral geniculate bodies, is relayed to the pons. There is then a
sequence pons to cerebellar cortex to thalamus to motor cortex, and
there is a similar pathway for other modalities. Lerner (1967)
suggested that the PGO stimulation reactivates body image schemata,
thus giving an explanation for the predominance of REM sleep during
childhood. (Similarly, Breger, 1967, considers that dreams give us
images to aid the feeling of bodily integrit .) Still, however, all
this work does not prove that the PGO bursts contain any information,
and we remain with the proposition that PGO stimulation is a general
activator of the cortical areas involved in dream imagery.
Further dismissal of the relevance of PGO stimulation to
bizarreness comes from Pivik and Foulkes (1968), who noted qualitative
16
changes in NREM mentation over the night. Ideation became more
dreamlike and dreamlike reports were more likely on awakenings with a
shorter orientation time. Moreover, Fiss, Klein and Shollar (1974)
reported that the dreams of two subjects (volunteers with significant
psychopathology) were more intense, vivid, and more clearly narrated
(but not more bizarre) after a regimen of REM awakenings. There was
also an increase in eye movement density. They did not find the
increases in the control group who were awakened at the end of each
REM period, rather than at the start. Foulkes (1966) had found both
these results for REM sleep. However, Watson (1972) found that
regardless of the presence or absence of REMs, PIPs (periorbital
phasic integrated orbital potentials, a phasic event detected near the
eyes) were related to distortion and bizarreness in NREM sleep.
Rechtschaffen, Watson, Wincor, Molinari & Barta (1972) also found PIPs
occurring during NREMS and related them to the bizarreness of the
reports, and Arkin (1978) suggested that such REM intrusions are the
cause of night terrors.
It is important that PGO stimulation is reported to be more
related to the vividness, movement and complexity of dreams than to
dream bizarreness. Initially Dement and Wolpert (1958) and Roffwarg,
Dement, Muzio & Fisher (1962) proposed that the dreaming subject is
following the movement of the image. This wa disproved by Jouvet's
(1962) finding that decorticated cats also have eye movements, yet
presumably no images, and also by the presence of REMs in newborn
humans. Also, Moskowitz and Berger (1969) found that whether recorded
REMs were horizontal, vertical or oblique did not correspond with
concomitant dream images. Berger and Oswald (1962) found a significant
17
association between activity/passivity of the dream events and
frequency and size of the eye movements. An active dream was defined
as one which would have been accompanied by many shifts of gaze, if it
had occurred in real life. The frequency and size of the REMs were
given more weight if they occurred later in the period, in assessing
the final correlation. Fifty dreams were classed as active, and in 42
instances the corresponding REM record was classed as active, 39
reports were classed as passive and in 23 of these cases the eye
movements were independently classed as passive. Furthermore, Pivik
and Foulkes (1966) showed that dreams after a period of dream
deprivation (which causes an increase in REM density when the subject
is allowed to dream again) are more vivid. Similarly, Dement and
Wolpert (1958) report that 'the amount of observed eye movement was
related to the degree of participation of subjects in the events of
the dreams'. Note, though, that Firth and Oswald (1975) found a weaker
correlation between dream activity level and REM density than found
previously. Also, Broughton and Mamelek (1980) showed narcoleptics
have a mean REM density within the normal range, yet they do have
particularly vivid dreams.
The degree of bizarreness in the dream is dependent upon the
subject's waking personality - Schechter, Schmeidler & Staal (1965)
(+.38, p=.00l) between scores of
found significant correlations
waking creativity and dream imaginativeness (defined as "condensed
symbols, vivid and bizarre images, a development of character and
setting, and other evidence of the lightening of ego defences") for 87
recallers. The rating scale evaluated logic, realism and ego-control,
but it did not rate subjects' moving images.
18
Foulkes and Rechtschaffen (1964) found that a pre-sleep
showing of a violent film led to REMS awakening reports that were more
imaginative, more vivid, more emotional and longer than did a neutral
film, but, significantly, it did not lead to more distortion or to
regression in time reference. Therefore, dreams can alter in vividness
without altering in bizarreness. (Unfortunately, there was no record
of REM intensity in the study for us to see if it correlated with the
bizarreness.)
A dream can contain more activity and still be entirely
realistic. For example, Rechtschaffen and Buchigiani (1983) found that
vivid material was no different in bizarreness than less vivid
material, on average. Some lucid dreams fit into the opposite
category, the subject may just passively watch and wonder about some
incongruous aspect of the experience, and then actively take part.
Furthermore, Hartmann (1984), attempting to provide further evidence
for his theory that REM sleep is caused by noradrenaline and serotonin
depletion in the brain, coupled with acetylcholine excess (and that
nightmares have dopamine excess in addition to these factors), gave
small amounts of l-dopa to subjects. He found that measures of
"dreamlikeness", "vividness" and "detail" were all increased. He notes
(ibid. p.259) that propranolol does the same.
Antrobus (1986) claims that dream cognition and
phenomenology is explainable in terms of cortical activation and
heightened sensory thresholds alone, and he proceeds to cite Lehmann,
Duniermuth, Lange & Meier (1981) and Williamson, Galin & Mamelak
(1983) , who report an association between dreaming indices and
spectral power in the frequency bands <12 Hz., but he notes that they
19
used small sample sizes. Conversely, Antrobus, Ehrlichman, Weiner &
Woliman (1983) found no relation between EEC power in the band 2-12
Hz. and several indices of dreamlike mentation (N=21). He proceeds
(1986) to give persuasive evidence against the relation of phasic
events with dream breaks and bizarreness. Zimmermann (1970) proposed
that waking thought occurs at a lower level of brain activation than
does dreaming, and LaBerge, Levitan & Dement (1986) found that lucid
REM dreams have an even higher level of physiological activation than
do non-lucid dreams. So, PCO stimulation rate correlates with the
above-mentioned dream indices but it is not certain that EEC values do
so also.
I will now review the relevant work done on awakenings
following PCO stimulation (or, more precisely, phasic events in the
case of humans, because the PGO bursts cannot be detected without
intracranial electrodes).
Moruzzi (1963) noted that tonic aspects of REM sleep (such
as the low voltage EEC trace and EMG suppression) are interspersed
with phasic aspects, notably REMs, which are found to correlate with
PCO spikes (Pompeiano, 1967b), with pupillary dilation (Berlucchi,
Morruzi, Salvi & Strata, 1964), electrodermal responses (Broughton,
Poire & Tassinari, 1965 a&b), shallowing of respiration (Aserinsky,
1967) and with augmented EMC suppression (Pompeiano, 1967b and
Wolpert, 1960). Many papers show that dreams with profuse eye movement
are more active, vivid, emotional and anxiety provoking than those
with few eye movements (Dement and Wolpert, 1958; Hobson, Coldfrank &
Snyder, 1965; Karacan, Goodenough, Shapiro & Starker, 1966;
Goodenough, Witkin, Lewis, Koulack & Cohen, 1974). The dreams become
20
more intense with increased REMs, but not more bizarre. Bizarreness is
more related to time passing since last contact with waking reality,
and with the subject's personality. Note also Salzarulo's (1972)
finding that EM density is at a maximum in the middle of each REM
sleep phase, and that "over the first FS [REMS] phases of the night EM
density stays almost constant whereas it increases abruptly at the
beginning of the last FS phases".
Molinari and Foulkes (1969) obtained reports from REMQuiescent and from REM-Active parts of sleep. Reports were divided
into Primary Visual Experience and Secondary Cognitive Elaboration
categories, but partly with the knowledge of the accompanying eye
activity. (Aserinsky, 1967, had speculated that REM sleep episodes
with ocular movement and those with ocular quiescence may be
differentiated by their "levels of dreaming" and not just in relation
to the scanning hypothesis.) As with tests of the scanning hypothesis
Molinari and Foulkes had subjects report their very last preawakening
experience. One hundred and sixty awakenings occurred, with SCE
defined as:
1) Thinking, being aware, recognising, interpreting
or 2) Forming alternatives, comparisons or conceptual
relationships
or 3) Presence of verbalisation or expl nation.
Anything else was classed as PVE: SCE can include visual experience
but must include one of the above. The results were:
21
[sleep onset]
REM - M
REM-Q
NREM-Asc.
S0-2
SO-i
%age SCE
12
80
77
57
64
%age PVE
88
20
23
43
36
Note that SCE should not be equated with Freud's secondary
processes, or with waking thought, because it can have regressive
features and a lack of reality testing. "The distinguishing
characteristic of PVE is not the presence of visual experience, but
the absence of an active intellectual orientation toward such
experience..." (ibid.). Foulkes and Molinari go on to compare the PVESCE distinction to that between condensation and displacement.
However, Medoff (1972) found that 3 subjects could not discriminate
REM-phasic from REM-tonic awakenings in terms of PVE vs. SCE. The
subjects were specifically asked about the presence of conceptual
material. Similarly, Foulkes and Pope (1973) used 3 arousal categories
and found no significant differences in bizarreness, as well as the
following:
Awakening Category
REM-burst
Sawtooth burst/no REM No sawtooth
Subject's
discontinuity scale
.80 *
1.25 *
1.12
12% *
21% *
21%
100%
100% *
63% *
Nothing earlier
recalled
Spontaneous report
of PVE
(* means significant difference)
22
(A sawtooth burst is a phasic event measured by the EEC over
the top of the head.)
The PVE-SCE distinction was not significant for interview data. There
was a greater likelihood of a discontinuity only if sawtooth-burst/no
REM is compared to REM burst, which no doubt had an accompanying
sawtooth burst, so this significant correlation is best taken as
specious.
Holmes (1976) replicated the Molinari and Foulkes (1969)
results. The latter had used a written questionnaire which Holmes
considered an interference with recall, and he therefore used tapes.
The reports were assigned to one of the categories PVE or SCE, on the
basis of the final event, and two further judges graded the report for
manifest drive expression, that is, sexual and aggressive
interactions. The REMs were classified as REM-Q only if there were 30
seconds of REM sleep with no REMs preceding waking. His results were:
%age
%age
%age
N
SCE
PVE
Visual Reports
REM - M
56
13%
87%
100%
REM - Q
38
89%
11%
47%
Again the linkage of REM-M (and hence of PGO bursts) to the visual
system is shown, rather than a linkage to the production of bizarre
associations alone. These bizarre associations occur independently of
visualisation (as shown by studies of blind people who have bizarre
mentation but no visual imagery of it during REM sleep). Holmes states
that the visual imagery associated with REM-Q awakenings is "less
peremptory", less "captivating" and less vivid, but not less bizarre.
23
The rest of his results are concerned with correlating recall from
these two categories with the subjects' personality, classified as
"diverger" or "converger" (a measure of, respectively, creative,
artistic thinking, and narrow scientific thinking) - that correlations
were found provides indirect evidence for the phasic/tonic
distinction. For example, convergers showed 50% recall on REM-M
awakenings, divergers 95%, while with recall on REM-Q awakenings the
convergers were superior. (These results he related to the threatening
content of some REM-M material which he said was dealt with by
repression in the case of convergers, and by rationalisation in the
case of divergers.)
The reasons for the greater amount of time - up to 30
seconds between phasic bursts - that divergers spend in tonic REM
sleep may be different from those suggested by Holmes.
Because Foulkes (1962) found that 66% of NREM reports were
visual, which shows that phasic stimulation is not a necessary
precondition for visual dreams, I suggest that one possible purpose
for them is to increase any vividness which occurs upon turning away
from the outside world. I also note that REM sleep reports are more
complicated than NREM reports from the same part of the night, but
late NREM mentation is similar to early REM mentation (Kales,
Hoedemaker,
Jacobson,
Kales,
Paulson
&
Wilson,
1967).
Much of the work just reviewed is based upon the idea that PGO bursts
are a unique feature of REM sleep and are somehow necessary for REM
sleep to occur. However, recent work has complicated this account, and
will be reviewed in the next section.
24
The Function of PGO Activity
Evidence has been aduced for the correlation between the
density of phasic events and the amount of activity in the dream.
Although this can support the idea that PGO bursts somehow maintain
the brain in an excited condition, recent work by various authors can
be taken to show that the causality works in the other direction, that
is, that it is the dream activity that causes the phasic PGO bursts.
Individual P00 spikes are very similar to startle reflexes in awake
animals, an EEG reflex rarely seen due to the mundane conditions that
experimental animals are kept under. Bowker and Morrison (1976) found
that spikes 'identical in configuration, amplitude and duration to
spontaneous P00 spikes can be elicited in the lateral geniculate body
and visual cortex by auditory ... and tactile stimuli during quiet
wakefulness, drowsiness, synchronized sleep, and PS.' (ibid. p.23.)
They cite evidence that waking EEG recorded eye movement potentials
are similar to REMS P00 bursts, and that both type of wave are
simultaneously abolished by local cooling of the pontomesancephalic
isthmus. However, the two types of wave (termed PGOrem and P00w by
Brooks and Gershon, 1971) do differ in some respects. P00 rem occurs
prior to eye movements, whereas PGOw follows them; P00w recorded in
the LGN are 50% of the amplitude of P00 rem occurring spontaneously
during REMS. However, Bowker and Morrison clam that this difference
is due to the degree of alertness common in the laboratory animals.
Brooks (1976) notes that ascending pathways responsible for P00w are
similar to those subserving P00 rem.
This suggests that PG0 spikes during REMS are indicators of
hyperalertness, and that during REMS 'because the internal alerting
25
stimuli are generated in the reticular formation rather than via the
retina, the eye movements of PS tend to follow rather than precede
PGOrems' . Bowker & Morrison (1976). They do not comment on whether the
alerting PGO bursts are spontaneous during REMS, or whether they are
orienting responses provoked by the dream imagery. The result of
Molinari and Foulkes (1969) supports the latter position, as does the
result of Foulkes and Pope (1973), in which only imagery spontaneously
mentioned as vivid, rather than that reported in answer to a
questionnaire, was found to correlate with PGO activity. Home (1988,
p.288) relates these orienting responses to the short-lived changes in
many physiological systems which occur during REMS. He proceeds to
suggest that 'REM sleep is very similar to alert wakefulness. We might
even begin to query whether REM sleep really is a state of sleep, or a
peculiar form of wakefulness within sleep!' (ibid. p.289.)
Adrien (1976 & 1977) reports that in the cat lesions of the
anterior raphe nuclei are followed by a state of permanent PGO
discharge, and also that visual experience has no influence on the
ontogenesis of PGO activity in the visual sensory system. Cespuglio
and Valatx (1977) write that 'there are significant differences in the
frequency and distribution of eye movements between the two strains.'
This they take as evidence of a 'genetic component in the pattern of
PCO activity.' (ibid.) They note that phasic activity in the cat is
marked in the visual system but not in the whiskers, and is marked in
the whiskers of a rat but not in the LGN: this they relate to the
reliance of the rat on touch and smell. They suggest that 'PGO
activity may act to maintain functional the systems necessary for the
survival of a particular species of animal.' (ibid.)
Whether Bowker and Morrison (1976) are correct or not,
26
has no implications for the debate over the meaningfulness and
function of dreams. However, if they are correct then theories of REMS
that emphasise the importance of PCO stimulation as an essential and
unique part of REMS are misguided, although the the bursts could still
be interpreted as cortical stimulation. Conclusions reached about PGO
stimulation keeping the cortex in a receptive state are not affected
by these results, although there is the possibility that a third
variable causes both the dream vividness and the PGO bursts, which
would thus be causally independent of each other.
Other Characteristics of Dreams Through the Night
What explanations can be offered for the increase in dream
intensity and bizarreness as the night progresses, and also for the
lack of bizarreness in children's dreams (examples of which are given
in Kimmins, 1920), the time of life with ,reatest toi ottth. C
Foulkes (1982b) found that children's earliest REM dreams, at ages 35, do not involve kinematic imagery: movement and activity are seldom
reported. They lack not only narrative but also any consistent
presence of human characters and social interactions. He also found
that children's dreams are just a recreation of past waking
experience, or a totally plausible portrayal of real-life
contingencies.
Pivik and Foulkes (1968) suggest that the increase in
bizarre content in dreams as the night progresses is a function of
total sleep time before the awakening. However, contrary to the claim
that the subject must be asleep a long time in order to develop
bizarre imagery, Bertini, Lewis & Witkin (1969, in Altered States of
27
Consciousness, ed. Tart) found that at sleep onset there are dreamlike
transformations of elements from films shown just before sleep; for
example, one subject "dreamt" of the head of a worm sticking out of an
apple after seeing a film about a birth, and another subject saw a
suitcase being emptied. Other work shows that longer REM periods have
more vivid, emotional and active dreams, but that their plausibility
and sensibleness are unaffected by REM time length, again showing the
lack of association between dream intensity and dream bizarreness.
Finally, I wish to emphasise the narrative nature of dreams
by reference to the findings of Kerr and Foulkes (1981) and also Kerr,
Foulkes and Jurkovik (1978). In both papers a patient is found to have
narrative dreaming and yet no visual imagery, either when asleep or
when awake. (However, Hudson, 1971, describes patients with Turner's
syndrome, who were used in the latter experiment, and shows how it is
not certain that they are devoid of visual imagery; they have many
genetic disabilities which could be affecting the results. Also,
Humphrey and Zangwill, 1951, reported three patients with parietooccipital damage, defective visualisation and spatial judgement, who
all had a loss of dream recall.) I also wish to mention the possible
relation between narrative and the frontal lobes, following Jus,
Jus, Villeneuve, Pires, Lachance, Fortier & Villeneuve's (1973)
report that
frontal lobe damage
(in lobotomised chronic
schizophrenic patients) leads to a halt in dream recall, the relation
between the posterior portion of the left-hemisphere and the
generation of imagery in waking life (Farah, 1984), and Foulkes'
(l982a) idea of
the
dependence of
language parts of the left brain.
28
dream production on the
CHAPTER 3
OLDER PHYSIOLOGICAL THEORIES OF DREAI'IS
There have been speculations about the cause of dreams for
over two thousand years. In this chapter I will describe some of the
most influential theories dating from antiquity to the turn of this
century. I will provide more detail about Freud's work in chapter 9.
The oldest theories held the mind to be passive during
sleep, with dream images caused by the impingement of stimuli external
to the mind. Freud saw other accounts as dividing dreams into two
types - those concerned with the past and those with predicting the
future. There was also a division between those who thought dreams
intelligible, or at least amenable to interpretation, and those who
considered them meaningless. In chapter 8 we will see that these
standpoints are still with us.
The Source of Dream Images
Later philosophers divided on their opinions of the source
of these stimuli which caused the images of dreams. Some held that
some stimuli could be supernatural, others that they derived from the
body (Maury, who had noted the changes in his dreams during periods of
illness) or from the outside world (Jessen, for example). Others
emphasised either the presence of waking concerns in dreams (Cicero,
J.Mauss) or the presence of subliminal impressions from the day
before, or from the remote past. For example, H. de Saint-Denys, who
noted that it was often just good luck when the source of an image is
discovered. Freud was to emphasise the presence of episodes of the day
before in dreams, especially minor, incidental events. He warned:
29
'It is easy to see how the remarkable preference shown by
the memory in dreams for indifferent, and consequently unnoticed,
elements in waking experience is bound to lead people to overlook in
general the dependence of dreams upon waking life...' (1953, p.19)
Some philosophers before Freud, however, developed the idea
that dreams result from endogenous processes in the brain. Lavie and
Hobson (1986) claim that Freud didn't review these in 'The
Interpretation of Dreams' because of his concentration on French and
German writers. This tradition was based in Britain. Some were
influenced by Leibnitz's dogma that thought and psychological activity
had to be continuous, carrying on during sleep, with the resultant
dependence of dreaming on waking mentation. Locke countered this
school in positivist style with the observation that we are only
rarely aware of these putative sleeping thoughts. Hartley followed
Locke with the claim that the state of sleep is so different from that
of waking life that the associations during sleep are different from
the products of waking vigilance. He also attributed the ease of
forgetting dreams to the difference in mental state between sleep and
waking, which coincides with some modern views.
In Scotland Reid introduced the idea of dreaming being a
state 'intermediate' between sleep and waking, in which one was able
to attend to important stimulations present n the outside world.
Stewart, in the early nineteenth century, built on this to suggest
that dreams are like waking thought, except that the will is absent,
leaving only associations between thoughts.
Also in Scotland Hamilton was elaborating Leibnitz's idea of
the continuity of sleep and waking. He traced the continuity between
30
perception and thoughts in the two states, and performed experiments
in which he was abruptly awakened during the night in order to catch
himself dreaming. Despite believing in the continuity theory he did
realise that the type of thinking during sleep was different to that
of the day. Similarly, Wundt, the founder of experimental psychology,
stated that stimuli impinge on the sleeping mind under conditions
favourable to the formation of illusions.
The Purpose of Dreaming
By the middle of the nineteenth century physiology had
advanced such that dreaming could be related to nerve reflexes and to
changes in the circulation of the blood. Laycock, in 1851, wrote that
sleep serves a restorative function for the body, and C. Lewes replied
that sleep restores fatigued nerves. The former also held that dreams
occur during imperfect sleep, when the mind is still conscious of
internal and external stimuli. Nearer the end of the century arose
the claim that dreams are the differential activation of separate
parts of the brain. Freud applauded such a view of dreams which placed
them within normal physiology, rather than as a response to illness.
Freud produced a theory of dreams in 1900 which went far
beyond a basis of memory, physiology and dream-day stimuli. He wanted
to explain how dreams deal with this materi 1, and also what the
function of dreams is. Most previous theories provided accounts of the
mechanism of dreams, but not of their purpose. He claimed that for the
writers who came before him, 'dreams are a reaction to the disturbance
of sleep brought about by a stimulus - a reaction, incidentally, which
is quite superfluous' (ibid. p.78). He applied this criticism to
31
Schemer's notion that the mind in sleep plays with and symbolises the
stimuli that impinge upon it. He also complained that even some of the
proposed mechanisms were inadequate - stimulation theories could only
account for a small number of dream images. He countered that our
mental processes can account for dreams just as they can account for
neuroses and slips of the tongue - the product is strange, however,
because the thinking in them is done with images, which are believed
in because the person has 'turned away from the external world' (ibid.
passim).
With regard to mechanism, Freud's major contribution is in
the notion of unconscious thoughts gaining access to consciousness
through disguise and through their association with events of the
dream-day with the purpose of protecting the subject's sleep. He held
that dreams do this by incorporating disturbing thoughts (both
conscious and unconscious) and the occasional disturbing sensation
into the images of the dream.
32
CHAPTER 4
NEWER PHYSIOLOGICAL THEORIES OF REM SLEEP
The Catecholamine Depletion Hypothesis
Hartmann (1973) and Hartmann & Bridwell (1970) proposed
that, whatever dreams do, REM sleep restores depleted catecholamine
levels. This explained work on anti-depressants which are
independently known to increase catecholamine levels, but which also
severely reduce REM sleep time, and the observation that manic
subjects are believed to be high in catecholamines and tend to have
less REMS (Schildkraut, Schanberg, Breese & Kopin, 1967). Conversely,
alphamethylparatrypt-amine and 6-hydroxydopamine both interfere with
catecholamine synthesis and both increase REM time (Hartmann, Chung,
Draskoczy & Schildkraut, 1971), while Nakazawa, Tachibana, Kotorii &
Ogata (1973) found that L-Dopa administration reduces REM rebound;
this concurs with reports that some schizophrenics have reduced REM
rebound and that one proposed aetiology for schizophrenia is of an
excess of the catecholamine dopamine. Hartmann and Stern (1972)
reported that whereas four days of REM sleep deprivation in rats
resulted in a deficit in acquisition of an active avoidance task this
was reversed by administration of L-Dopa (a precursor of the
catecholamine dopamine). He had previously shown that the antidepressant MAO inhibitors and imipramine both reduce REMDep learning
deficits. (See Linden, Bern and Fishbein, 1975, for a study relating
REM sleep to learning, and my fifth chapter.) Iskander and Kaelbling
(1970) reduced catecholamine levels in cats and found an increase in
paradoxical sleep - that REM rebound cannot increase indefinitely with
33
deprivation led them to hypothesise that REM sleep is related to a
'physiological system having absolute values, such as catecholamine
stores'
Iskander and Kaelbling (ibid.) provide a reply to the most
embarrassing problem for the catecholamine hypothesis, namely that REM
deprivation has a therapeutic effect on endogenous depressives, who
are hypothesised to have too little catecholarnine. (See Vogel, McAbee,
Barker & Thurmond, 1977, who found that depression improvement
correlates with REM pressure after deprivation. This improvement is
embarrassing for many theories that propose functions for REM sleep in
terms of information- processing and aiding memory.) They suggest that
while "there is probably no single aetiological factor in depressive
illness, the evidence suggests that changes in delta sleep are more
significant than changes in paradoxical sleep in this clinical state."
They review the evidence that depressive patients have more awakenings
during the night and a gross loss of delta sleep, with a less
noticeable reduction of REM sleep, and that patients treated with ECT
have paradoxical sleep returning to normal levels "at about the time
of clinical improvement", and that delta sleep also returns to normal
levels. Similarly, MAO inhibitors cause a decrease in PS time and an
increase in Delta sleep; they also increase serotonin availability,
which is related to delta sleep. It is thus oossible that lack of
paradoxical sleep in these patients is due to lack of delta sleep, and
it is suggested that the therapeutic effect of REM deprivation works
by increasing delta sleep time. Cartwright (1983) found shortened REM
latencies in women undergoing divorce, REM latency returned to normal
at 1-2 year follow up except for those initially most depressed, and
34
Iskander and Kaelbling (1970) similarly report that at the time of
clinical improvement PS activity returns to normal. PS may thus not be
directly related to depression and its cure, which lessens the impact
of the depression data towards the catecholamine depletion hypothesis.
However, Zarcone and Benson (1983) do report that subjects who rated
themselves as depressed had greater REMS eye movement densities than
did controls.
The catecholamine system (specifically noradrenaline) that
we are concerned with has widespread terminations in the forebrain, an
area which is implicated in mammalian intelligent activity. Hartmann
is suggesting that the NA system is inactive, and is possibly being
repaired, during REM sleep. This is indicated by Schildkraut and
Hartmann's (1972) finding that NA metabolism is increased after 72
hours of REM sleep deprivation. The NA system originates in the locus
coeruleus, which is confirmed by single-cell recordings to be inactive
during REM sleep (Hobson, McCarley & Wyzinski, 1975).
The Continuity Between Waking and Dream Experience
This theory challenges the assumption of the continuity
between waking and dreaming brain conditions. Ever since Leibnitz
there has been hypothesised a continuity between waking life and dream
content. This hypothesis has its experimental justification in studies
comparing the dream life of creative and non-creative people; for
example, Adelson (1959), who reports dreams of more creative college
women as being more interesting and innovative than those of less
creative peers, Domino (1976), who used two high-school groups that
differed in creativity and found more "primary process thinking"
35
(defined as unusual or impossible events, inexplicable
transformations, condensations etc.) from the more creative group, and
Schechter, Schmeidler & Staal (1965), who found dream diary reports of
art students to be more imaginative than those of science or
engineering students. If critical judgement, task orientation,
reality testing, sense of identity and free-will are connected with
the catecholamine system (as hypothesised by Hartmann, 1973, and by
Stern and Morgane, 1974), then the lack of these, and concomitant
bizarreness, are solely the preserve of REM sleep (and possibly some
biochemically disordered individuals).
Note that this argument for a lack of continuity between
dreaming and waking mentation does not depend upon the truth of any
particular biochemical theory of dreaming, although I have here used
Hartmann's theory for illustration and because of the mass of evidence
in its favour. This theory is anyway opposed by evidence that
noradrenaline may act as the trigger for REM sleep, see Kanno and
Clarenbach (1985) and Jouvet (1969), and by another biochemically
based theory of dreams and hallucinations, that of Hernandez-Peon
(1966). It is the presence of a temporary neurochemical disorder that
leads to the discontinuity prediction: that REM sleep has a
physiological function would not be enough to predict waking/dreaming
discontinuity.
The Brain Over-Excitation Hypothesis
The above theory of dream function emphasises the effect of
REM sleep upon the brain as a whole. A similar theory holds that REM
36
sleep acts to make the whole brain less excitable, and that dream
deprivation will increase susceptibility to epileptic attacks. There
is evidence that after REM sleep deprivation there is a change in
threshold of electrical current able to produce epileptic fits.
Cohen and Dement (1965) hypothesised that REM deprivation causes a
generalised increase in neural excitability. This was based on the
increasingly severe convulsive spasms of face and body seen in the REM
deprived animal, and the increased availability of cells able to
respond to a second click in REM deprived cats (the accelerated
auditory recovery, found by Dewson, 1965). He found base thresholds
for electroconvulsion for 48 rats. 24 were REM deprived but obtained
SWS and 24 controls were allowed to sleep normally for the six days of
the experiment. The former group showed a fall in the mean threshold
of 23%, whereas the control group showed a slight increase. However,
in a third group used to control for the non-specific effects of sleep
deprivation only 2 non-REM sleep deprived animals were used, one of
which had a 4% drop in threshold. Bearing in mind that some REM
deprived animals had a threshold drop almost as small as 5% it is not
proven that REM deprivation is the relevant factor; more NREM sleep
deprived controls are needed. He states that REM deprivation may be
affecting plasma electrolytes, steroids or hormones, and hence these
may be the mediators of the effect. In favour of this interpretation
is the finding of Ukporiniwan and Dzoljic (1984) that enkephalinase
inhibition antagonises the increased susceptibility to handlinginduced seizure caused by REM sleep deprivation.
An increase in REM sleep time is found after daytime
electrical stimulation of the Ascending Reticular Activating System
37
(Frederickson & Hobson, 1969), which accords with the biochemical
recuperation theories mentioned above, and may provide a link between
the two theories.
Another experiment explained by gross neurochemical and
electrical changes is that of Cohen, Duncan and Dement (1967), who
found that electroconvulsive shock reduced REM rebound in REM deprived
cats. (This can be related to Jouvet's (1967 a&b) postulation of
catecholamines as the mediator of REM sleep, to Pujol, Mouret, Jouvet
& Clowinski's (1968) finding that there is increased noradrenaline
turnover during REM rebound, and to Kety, Evarts & Williams' [1967]
finding of an increased turnover of noradrenaline in the CNS after
repeated electroshocks.) This result suggests that REM sleep and ECS
both augment NA synthesis, and that changes of parameters for either
REM sleep or ECS may be related to NA on a neurophysiological level
rather than to the 'dry' level. The suppression of REM phase in the
cat after ECS was shown by Cohen and Dement (1966).
Strangely, in view of the above theory that REM sleep
prevents over-excitation of the cortex, seizures can substitute for
REM sleep (also shown by Kaelbling, Koski & Hartwig, 1968). REM
deprivation leads to an increase in Intracranial Self Stimulation
rates, and a decrease in current threshold (Steiner and Ellman, 1972),
but again this result can be globally and cherni ally explained in that
noradrenaline is implicated in ICSS (Stein, 1966). Also, Cohen and
D ment (1965) found that amphetamine lethality is actually reduced by
REM sleep deprivation, and postulated that this was due to a reduction
in brain catecholamine levels. The experiments cited so far in this
chapter indicate that dream sleep is implicated in neurochemical
38
changes in the whole brain; such theories may or may not be at
variance with those to be reviewed in the next chapter, that claim
that
REM
sleep aids the consolidation of individual memories
themselves.
In favour of some connection between
REM
sleep and a
protective function, though, is the report by Dement (1976, p.92) that
a narcoleptic patient had complete
REMS
suppression when given a
monoamine oxidase inhibitor. Upon withdrawal of the drug the patient
'entered an almost continuous
REM
sleep interrupted by waking
hallucinatory periods' . Similarly, cats whose synthesis of serotonin
was temporarily inhibited by the drug PCPA had reduced sleep time
(REMS
and
NREMS).
After a few days they were able to sleep more, but
PGO spikes began to emerge into the waking state. However, it must be
noted that although the barrier between
REM
sleep and waking life can
be breached in this way, This does not prove that
REM
sleep has the
function of keeping PGO activity away from waking life. An analogy is
that extremely hungry people may try to eat grass, but that does not
prove that we normally eat food in order to avoid eating grass!
Alternatively, to remove a valve from a radio may result in a highpitched noise, but the valve is not present in order to suppress that
sound. Furthermore, according to Purpura, Shofer & Scarff (1965), in
the newborn mammal (and at a time of much
REM
sleep) there is a
precocious development of synaptic inhibition, maybe as a brake
against overexcitation before the appropriate sensorimotor systems are
mature. The threshold of seizure induct on is actually greater at
earlier ages. They also state that spontaneous activity of neurones is
virtually absent.
39
I conclude that evidence that REM sleep is connected with
the staving off of electrical seizures is not conclusive. However,
Home (1988, p.71) notes that human studies show that 'sleep
deprivation may promote epilepsy in those people with a history of the
disorder'. He states that the sleep-deprived brain lives from moment
to moment, without pre-planned preparation, and that 'although
subjects can quickly respond to stimuli when these arrive, subjects
seem to react to each stimulus as a sudden surprise, rather than treat
it as an expected event.' (1988, p.75.)
Information-Processing Theories
Newman and Evans' (1965, also reported in Evans and Newman,
1964, and Evans, 1983) computer theory holds that REM sleep is a time
for practising programs and erasing wasteful sub-routines. This theory
was inspired by the function of the off-line state in computers.
During this state a computer is run without actually having any effect
on the outside world, so that programs may be explored on their own,
without providing any output to the tasks normally controlled by it.
The theory can provide many reasons to explain the decline in REM
sleep with age, and also the unconsciousness and isolation of the
subject during sleep.
One objection to the theory is the infant's need for reality
feedback to obtain such schemata as object per tanence, attachment and
sensorimotor skills; whereas later in life the brain can obtain
important results by turning to itself, why then is dreaming
concentrated in the early years? One answer is that at early ages so
much more is new, even though so much less has been learned in total.
40
In later years we are just learning more of the same thing, rather
than whole new schemata. The REM period is thus providing a playback
of what has happened recently, which fits in with the prevalence of
images from the previous few days. REMS time thus correlates with
amount of information being picked up (and its ease of assimilation)
rather than with the amount already in storage. (The latter is a
prediction of any computer theory holding REM sleep as a checker of
all programs.)
Possibly all mammals are preprogrammed to decrease REM time
with age because few new concepts are learned later on in life; this
is true for human culture where after picking up new concepts like
alphabet, etiquette, language and morality most additions then are
just more of the same thing (e.g. a second language), knowledge then
increases quantitatively. But maybe the system is not rigidly prewired though and a conceptual jolt, such as bereavement or religious
conversion, could lead to a greater time spent in REM sleep. (See
Fishbein and Gutwein, 1977, for a review of the findings of increased
REM time after learning, and Maho, 1977, concerning the critical
learning period preceding stabilisation of performance during which
REM time is maximised. Cohen (1979, p.93) summarises evidence for REM
time being related to amount of material still to be learnt in a
learning experiment.) Cartwright (1986) cites evidence for REM time
changes during divorce proceedings, covarying with whether the subject
was depressed or not.
For no increase in REM time to occur upon presentation
of a conceptual jolt would not falsify this suggestion of this
computer theory, for vividness could change instead. Cohen (1979)
41
states that, "it is my impression that to the degree that physical
activation determines dream content, it will influence the quality
more than the specific content of the dream".
A similar theory, also based upon an information-processing
point of view, is that of the Activation-Synthesis Hypothesis (Hobson
and McCarley, 1977, with further experimental evidence summarised in
McCarley and Hoffman, 1981). It is based upon the claim that PGO
stimulation from the brain-stem provides structured input about
movement, and hence it explains the finding that REM periods with
greater frequency of PGO bursts occur with more active dreams (Berger
and Oswald, 1962). It proposes that the PGO stimulation contains
sensorimotor information which is passed to the Lateral Geniculate
Nucleus and cortex, and that the amount of information about movement
will correlate with the rate of activity in the dream imagery. This
theory provides not just an explanation for dreaming, but also a
purpose for REM sleep in the infant fetus, in that visual neurones are
provided with excitation in order to aid their maturation. (The other
four senses will be obtaining stimulation from the uterine
environment.) Unlike the first two theories mentioned this one
provides an information-processing function for dreams - we will now
examine another such theory, which holds that REM sleep is involved
with the alteration and storage of memories, rather than just with the
preparation of neurones prior to learning (e idence for which also
comes from Roffwarg, Muzio & Dement, 1966).
42
CHAPTER 5
EXPERIMENTS ON REM SLEEP AND MEMORY
REM Sleep Aids Recall
The advantageous effect of REM sleep on memory is beyond
dispute, as shown by the following brief review of studies linking
dream sleep to learning. It should be noted that some of the studies
can be interpreted as simply showing that REM sleep halts memory
interference, whereas others claim that REM sleep has a positive and
actively beneficial effect on the laying down of memories. This effect
on memories is separate from the claims of the 'brain restoration and
repair' theories; the high level of neural activity during REM sleep
would, in fact, point away from the repair type of explanation.
1)
REM deprivation prior to training, or immediately
afterwards, impairs the formation of a permanent trace, even though
the new information has been registered (Fishbein, 1970; Segales and
Domino, 1973; Fishbein, 1972). Most of the studies that do find such
an effect use simple tasks.
2) Some results have been obtained indicating that REMS
deprivation affects initial acquisition of responses. However, in
favour of no effect is the result of Oniani (re orted 1987), who found
that REM sleep deprived rats performed a passive avoidance task as
well as controls, provided that the deprivation was effected through
'nonemotional' awaking, by hand, rather than by the water tank method.
43
3) Post-training prolonged REM deprivation affects memory
consolidation (putting the contents of STM into LTM) and stabilisation
(of LT memories themselves). Fishbein (1971) indicated that REMS may
function in the maintenance of LTM, as well as in its initial
consolidation. Fishbein, McGaugh & Swarz (1971) showed this directly
by training mice in an inhibitory avoidance task, they were then REM
deprived for 48 hours, and then given electroconvulsive shock (which
usually only causes amnesia if given immediately after training). This
group were found to be amnesic, the control sham-ECS group were not.
The dream deprivation did not destroy the memory traces, rather it
left them fragile and liable to disruption. Fishbein and Gutwein
(1977) similarly go beyond the evidence that REM sleep facilitates the
conversion of learned responses into long term memory, to show that
REM sleep actively maintains the stability of already consolidated
memory traces. Memory traces are thus made permanent and prevented
from decaying. (Note, in opposition to this, van Hulzen & Coenen's
[1979] finding that rats deprived of REM sleep after shuttle-box
avoidance learning did not perform worse than controls. However, they
did awaken the rats every time that REM sleep started, as the means of
REM deprivation, and so it is possible that consolidation occurred
during these waking periods, offsetting the effect of lack of
consolidation during REM sleep. Also, it is possible that the memory
was maintained in a labile state for these experimental group animals,
REM sleep having the effect of making the memories more stable, rather
than just consolidating them before forgetting occurs. They suggest
that the excited behaviour that REM deprivation causes may be the
reason behind learning changes in these experiments.)
44
There is also the finding of Zornetzer and Gold (1976) that
locus coeruleus lesions prolong the period during which an established
memory is susceptible to disruption. (The locus coeruleus is involved
in the production of REMS.) The only maladaptive feature of the memory
trace is its fragility.
4) The phenomenon of REM time augmentation after learning.
Lucero (1970) showed that rats trained in a labyrinth-like maze had an
increase in REM time but no change in total sleep time, whereas rats
which walked the same distances in the maze but without learning
imposed had no REM augmentation - this result argues against the use
of the notion of greater brain excitation rather than learning itself,
to explain some of the results reviewed here. (Similarly, Matsumoto,
Nishisho, Suto, Sadahiro & Miyoshi, 1968, found that even exercising
rats to the point of exhaustion has no effect on REM sleep
parameters.) Greater detail on this effect is presented by Smith,
Kitahama, Valatx & Jouvet (1974), who trained mice in a multiple-trial
discrimination task and found that REM sleep augmentation dissipates
as the learning curve approaches the point of maximum learning. Maho
(1977) found that there was a critical learning period for 7 cats
undergoing avoidance conditioning, characterised by a high rate of
respiratory, heart rate and arousal reactions, as well as an increase
in PS length. This period preceded the stabilisation of preferences.
In opposition to these results Meienberg (1977 found that a subject
on a language course had no change in baseline PS length. This may
indicate, though, that dreaming is only concerned with knowledge of
affective significance. However, Meienberg's result is contradicted by
de Koninck (1978), who found that English speaking Canadians who took
45
an intensive course in French showed marked elevation in REM sleep,
but only when their learning was effective. Further evidence for REM
augmentation is provided by Leconte, Hennevin & Bloch (1972) with
rats, and Solodkin, Cardona & Corsi-Cabrera (1985) in chicks after
imprinting.
5) There is indirect evidence linking REM sleep with
learning, such as the following: Gutwein and Fishbein (1980) showed
that environmentally deprived mice had less PS following learning of a
brightness discrimination task than did normally reared mice. Stern,
Morgane, Panksepp, Zolovick & Jaloviec (1972) found REMS to be
increased after protein synthesis was inhibited - such synthesis is
connected with the formation of memories. Jasper and Tessier (1970)
and Gadea-Ciria, Stadler, Lhoyd & Bartholini (1973) found that
desynchronized EEG is correlated with increased acetylcholine in the
neocortex. Bowers, Hartmann and Freedman (1966) showed that REM
deprivation is related to a reduction in brain ACh levels; we know
that ACh is related to learning because an enriched environment
increases the rate of release of acetylcholine, as measured by the
presence of acetyicholinesterase. Also, Skinner, Overstreet & Orbach
(1976) found that physostigmine (an anticholinesterase) prevented the
memory disruptive effects of REMS deprivation, suggesting that the
disruptive effect is cholinergically mediated.
Spreux (1982) submitted 8 subjects to a learning of a
calculation. There was an increase in the ratio of high- to lowfrequency eye movements (<lsec. as opposed to >2 sec.) compared to the
reference night, but sleep durations did not change. This was
interpreted as showing the information-processing nature of REM sleep.
46
Other work which finds a relationship between amount of REMs and
intelligence includes Tanguay, Ornitz, Forsythe & Ritvo (1976), who
found that autistic children show an immaturity in the organisation of
eye movements into discreet bursts, and Clausen, Sersen & Lindsky
(1977), who found a larger REM latency and lower EM density in Down's
syndrome subjects compared to controls.
In experiments to validate the catecholamine hypothesis
(which states that REM sleep restores catecholamine levels after
depletion during waking) it was found that reduced learning due to REM
deprivation can be reversed by L-Dopa or AMPT administration, compared
to control REMDep groups. L-Dopa and AMPT administration alone do not
aid learning (Hartmann and Stern, 1972), but do increase catecholamine
levels. That a pharmacological procedure can reverse REMDep effects on
learning argues against the importance accorded to FGO stimulation in
some theories of dreaming. In chapter 2 I showed that PGO stimulation
is instead concerned with the production of vivid dream images, rather
than with the content of the images.
Work on protein synthesis also shows the similarity between
REM sleep and normal learning. Lambrey-Sakai (1972) reported that REM
deprivation in rats impaired the incorporation of amino-acids into
brain proteins. Furthermore, Pegram (1973) inhibited protein synthesis
with anisomycin and found that this suppressed REM sleep. However, it
is necessary to note Home's (1988, p.281) point that REM sleep is
more fragile than NREM sleep, and is more likely to be lost due to the
stress of the drug study, as well as due to any reduction in protein
synthesis.
Drucker-Colin, Spanis, Cotman & McGaugh (1975) found that the
47
total protein content of cat midbrain perfusates obtained during REM
sleep is even higher than that obtained during wakefulness, indicating
the learning function of REM sleep.
However, Home (1988, p.281) points out that measurement of
the protein content of fluid from the brain does not tell us about
what happens inside neurones and neuroglia, which is where memory
proteins would be presumed to be laid down. Any change in protein
composition of cells could just be due to the general increase in
brain metabolism, and not due to specific memory proteins. He also
points out that there are stressful side-effects of the drugs given
and that such stress can itself affect memory (ibid.).
6) According to Scrima (1984 & 1982) there is, at least for
narcoleptics, a greater advantage in having REM sleep after learning
complex associative information than in having NREM sleep, which he
says merely prevents retroactive interference. He also found that
recall of complex associative information is better after NREMS than
after a period of wakefulness, and that in general 'studies
demonstrating that dream sleep deprivation produced recall
deficiencies used complex associative information, presumably the
basis of human memory' as opposed to minimally associative
information. Similarly, Tilley & Empson (1978) compared the effects on
story retention of REMS deprivation against stage 4 deprivation. For
the 20 undergraduate subjects, recall accuracy in the former group was
much poorer.
Scrima postulates that during REM sleep there is an
active consolidation of memories.
48
7) As a contrast to the finding that recall is better
following a period of REM sleep than a similar time spent awake,
Portnoff, Baekeland, Goodenough, Karacan & Shapiro (1966) found that
NREM sleep impedes the consolidation of memory traces: however, I
have doubts about the claim that the control group, who were kept
awake before sleeping again, were unlikely to rehearse the learning
material during this time, which would lead to a comparative deficit
of the NREM group's learning ability.
Dreams and Information-Processing
That some thought processes occur during sleep was shown by
Berger (1963), with the incorporation of spoken stimuli into dreams.
In addition, Hoelscher, Klinger and Barta (1981) found that 'concern
stimuli' were incorporated significantly more often than 'non-concern
stimuli' in stage REM, which shows (unlike the Berger result) 'that
sleeping subjects are capable of ... complex and subtle cognitive
discriminations, based on the waking value of the stimulus to the
subject.' They found a much smaller rate of incorporation into stage 2
sleep. Similarly, Cipolli, Fagioli, Maccolini & Saizarulo (1983) found
that sentences could be incorporated into dreams depending upon their
meaningfulness, another example of discriminative thinking during
sleep.
Evidence for the processing during dreaming of waking nonconscious perceptions is provided by Shevrin (1986), who shows that
unreported aspects of a briefly flashed (and presumably unconscious)
stimulus are recovered in dreams. Further evidence is provided in
Nisbett and Wilson (1977) and Zajonc (1980). This idea is associated
49
with Poetzl, who saw the visual apparatus as like a photographic
darkroom in which the developing of images takes place in different
stages. However, some of this work, over the decades since Poetzl's
publication in 1917, did not use controls, and so blank images are now
added. Previously, it was easy to confabulate any connections between
dream and stimulus image, and so this tendency had to be controlled
for.
Unconscious Adaptive Thinking During Dreams
However, it must be explained how dreams that are not
consciously remembered are still beneficial to the subject. As Rycroft
puts it, in a book unconcerned with the physiological and functional
implications of dreaming:
anyone maintaining that dreams have meanings has to
explain why the recipient of dream-messages does not even listen to
them, let alone take heed of them.' (1981, p.47)
I intend to show that the fact that dreams are for the most
part not remembered is no counter to their proposed learning function.
Dixon and Henley (1980) refer to eleven areas of psychology research
in which stimuli are found to exert a perceptual or behavioural effect
without ever entering consciousness. For example, if words are
initially associated to electric shock and then presented in the
unattended ear in a dichotic listening experiment a galvanic skin
response will be evoked (which is not evoked to control words). The
subject has no recall of what words were presented in the unattended
channel. Moreover, Forster and Govier (1978) showed that the response
50
even generalises to phonetically similar words, so a low level of
processing is occurring. Henley (1976) found that words which were not
consciously perceived could disambiguate other presented words. Work
on commissurotomized patients shows that the right hemisphere can
learn without the subject being aware of the material picked up.
Similarly, Nisbett and Wilson (1977) describe Wilson's study of
subjects having no conscious knowledge of tones presented in the empty
channel of a dichotic listening task, but when asked to rank their
guesses they did show some unconscious knowledge.
A sunimary of much
of the rest of such evidence is provided by LeDoux (1985). Note also
Kaser (1986) who played a sung message to subjects which was speeded
up sufficiently to be incomprehensible consciously. It was mixed with
a normal music recording, the latter being played alone to the control
subjects. All subjects made a drawing before hearing the tape, a
drawing afterwards, and a drawing of a dream from that night. Blind
ratings showed a significant difference between the experimental and
control groups with regards to their post-tape and dream drawings. The
conclusion was that the unconscious or preconscious could perceive the
message, and that, therefore, unconscious processing can occur in the
brain, and sometimes come to mind during dreaming.
Dixon (1971, p.315) notes that the effect of subliminal
stimulation is greatest when the recipient is relaxed and passive, and
when the stimulus is well below, rather than just below, the awareness
threshold. He proposes that the limbic-midbrain activating system
(see chapter 1) leads to less restraint on associative processes than
the (waking) Ascending Reticular Activating System. Similarly, Haber
and Erdelyi (1967) note that free-associational activity between a
51
tachistoscopic presentation and subsequent recovery of hitherto
unperceived material increases the number of items recovered. Another
example is given by Sperry (1968) in the paper 'Hemisphere
Disconnection and Unity in Conscious Awareness'. He reports flashing a
question mark to the left hemisphere and an erotic picture to the
right hemisphere of a split-brain patient. While the subject said he
saw a question mark he also showed signs of blushing and arousal which
he couldn't explain.
That actual learning can occur during sleep (rather than
just incorporation into ongoing mentation as shown above) is indicated
by Evans, Gustafson, O'Connell, Orne & Shor (1966) who found that an
objective bodily response to spoken material could be acquired during
stage REM, but not necessarily remembered during waking life. A
typical suggestion given during stage REM was "Whenever I say the word
'leg' your left leg will feel extremely cramped and uncomfortable
until you move it." The cue word was then spoken in order to test
acquisition. In the next stage REM period only the cue word was used,
the cues alone were also used on a second night. Eleven of 18 subjects
responded while in REM sleep to the appropriate cue words, 6 of these
responded in a subsequent REM period. Eight of these 11 were unable to
specify what had been said to them while asleep, the other 3 were
aware of the cue word but not of its significance, and these three had
shown evidence of arousal (alpha rhythm) during presentation.
(However, the published paper does imply that, at least for the first
retention test and evaluation of behaviour, the same experimenter was
used as gave the suggestion.) Unfortunately, the results do not
indicate whether the instruction was incorporated into the dream or
not.
52
Learning can thus occur during REM sleep, even to the extent
that subjects can learn to shorten their REM periods to 10 minutes in
order to avoid the trauma of being waken up at a fixed time after the
start of a REMS episode by a loud noise and then asked difficult IQ
questions (Fiss & Ellman, 1973). This effect was seen on the first
recovery night after the conditioning was ended but wore off on the
second night. Similar proof of instrumental learning during sleep is
given by Williams, Morlock & Morlock (1966).
Objections to Work in this Paradigm
Home (1988, pp.278-279) suggests that it is the increased
drive behaviour associated with REM sleep deprivation that may
distract the animal from its learning task. Such an increase in drives
may be directly due to the REM deprivation, or be due to the stress of
the procedure. He notes that 'if REM sleep deprivation is more
stressful than a control condition, then the animal may be more
concerned about reducing stress through producing stereotyped
behaviours than about anything else.' (ibid. p.279.) He writes that
the REM deprivation studies give more support to the theory that REM
sleep is involved with drive reduction and the reduction of the level
of excitement produced during wakefulness (1988, p.282). The latter
proposal would explain the increases in REM sleep observed in animals
kept in 'enriched environments'.
In addition, some of the human data above (of better recall
after REMS than after NREMS) is also interpretable as an effect of
high arousal during REMS compared to that during NREMS, which may as a
side-effect be maintaining the memory trace in a more retrievable
53
form. Note that such an interpretation does not depend upon whether
the subject was more alert after REMS than after NREMS, which these
experiments do control for. (Some more dubious experiments did not
control for the difference in initial brain activity between someone
woken from REM sleep and someone woken from NREM sleep.) Such an
interpretation (which has been ignored in Scrima's work) means that
the results of these human experiments do not tell us anything about
the actual function of REMS. However, Tilley (1981) retorts that
consolidation may be helped by REMS precisely by keeping the recent
memory traces in a state whereby they can be easily retrieved, and
Chernik (1972) indicates that REMS does not have the same advantage
over NREMS with regard to minimally associative information.
Furthermore, work by Cartwright (1977, p.95-97) on the advantageous
effect of REMS on memory for words, as opposed to a night in which
REMS was not allowed, is relevant here. Both groups had the same
number of awakenings. The material better remembered appeared to be
specifically related to 'active, striving, waking life', indicating
that REMS involves some selective information-processing, relevant to
the balance of outer and inner attention, and that REMS is not leading
to better recall of information in general, which would be predicted
if the result were due to greater activation of the brain during REMS
than during NREMS. I also note that this 'semi-active' interpretation
of the REMS data does not explain the observed progression of dream
content 'from predominantly elements of recent experiences toward the
beginning of sleep to older experiences during later sleep and back to
recent experiences again toward the end of a normal night of sleep'
(Scrima, 1984).
54
Evidence for this particular active information-processing
view of dreams comes from Roffwarg, Herman, Bowe-Anders & Tauber
(1978), in an experiment in which subjects wore coloured goggles
during their waking hours, and were awakened at various times during
sleep. On the first night goggle-coloured events appeared in some of
the scenes of the first dream of the night, and by the fourth or fifth
nights the goggle-colour was appearing in just under half of the
scenes of the later dreams of the night. The proportion of gogglecolour in the first dream increased progressively. When the goggles
were removed the colour disappeared almost completely from all of the
dreams of the following night, whereas it persisted for a few days in
sleep-onset imagery. Any imagery recorded during NREM sleep did not
show as much coloration as did REMS and sleep onset imagery.
Similarly, Battaglia, Cavallero & Cicogna (1987) found that sleep
onset awakenings have more day residues than REMS awakenings, while
Verdone (1965) showed that early REM periods have less memories from
the distant past than do late REM periods. Cartwright concludes that
such evidence means that 'dreams begin with the feelings and concerns
the person was experiencing just before sleep. Those that follow are
related to the first, but are older examples of situations in which
the same feelings are experienced' (1977, p.129). She claims that the
later scenes have contemplated solutions to the concerns.
Interestingly, an old account by the psychologist W. Rivers of a
series of three of his dreams from one night shows that 'the main
features of the three dreams were determined by the conflict arising
out of the prospect of a new [university] appointment, the manifest
content being determined by an incident of the previous day which had
directly stimulated this conflict.' (Rivers, 1923, p.87.) In addition,
55
whereas the first dream was based directly on the conflict, the second
brought in a memory of a trip to India, and the third a boyhood
interest in Cambridge.
Conclusion
The balance of the evidence is thus that REM sleep is
relevant to the organism's need to consolidate change, while
consolidation of ordinary memories can occur during wakefulness,
possibly needing just concentration. Consolidation can thus be
pictured as writing about an experience on a piece of paper and
putting it in the most obvious file, while dreaming is the crossreferencing and retrieval of various similar files. The insights of
free-association after dreaming would then be due to using the dream
elements as clues for which files to take out again when awake.
We will now proceed to study evidence that REM sleep is not
just associated with the storage of waking mentation, but is itself a
time of creative thought.
56
CHAPTER 6
IMAGERY, CREATIVITY AND DREA1'IS
In this chapter I will give evidence to show that REM sleep
is closely connected with divergent thinking and creativity, and
attempt to link this finding with the imaginal nature of dreams.
REM Sleep and Creativity
Cartwright and Monroe (1968) found a smaller REM rebound in
REM deprived subjects who reported their mental content when awoken
from REM sleep, than with subjects who reported a list of digits when
awoken. (However, in interpreting the result of a changed rebound
length we must remember that method of deprivation can affect the
rebound; for example, light exercise also decreases REM rebound.) They
also found that the amount of compensation was negatively related to
the amount of content judged to be dream-like. This implies that
dreaming can be partly substituted for by waking fantasy, and
complements work which shows greater creative ability following REM
sleep than after waking from NREM sleep. In that work, Fiss, Klein &
Bokert (1966) found that thematic apperception test stories (in which
a story is told in response to a pictured scene of interacting
figures) obtained immediately after REM periods were significantly
more dreamlike than TAT stories told after NREM awakenings.
Lewin and Glaubman (1975) suggested that mental activity
during dreaming is not integrative and consolidating (as much of the
last chapter showed), but is divergent and explorative. They found
that for twelve subjects REM deprivation led to poorer performance on
measures of divergent thinking, such as uses fluency and originality.
57
Glaubman (1978) set subjects a divergent thinking task before sleep
and found that subjects responses after NREM sleep deprivation were
numerically greater, more divergent and original than after REM
deprivation. This implies that REM sleep increases divergent thinking
during wakefulness, which is easy to believe if we bear in mind the
bizarre and innovative nature of dreams. Shevrin and Fisher (1967)
found that a waking subliminal stimulus is transformed differently in
REM sleep imagery as opposed to NREM sleep imagery, and Shevrin (1986)
gives evidence that dreams are necessary to recover primary process
transformations that have occurred in the day, or even to produce them
initially.
Hartmann (1984, p.128) gives evidence that 'there is some
relationship between nightmares and creativity'. Most of the people in
his survey of chronic nightmare sufferers were artists or other such
creative people. Also, Belicky and Belicky (1982), in a study of over
300 college students, showed that those individuals majoring in art
tended to have the most nightmares and vivid dreams, while students
majoring in physical education had the fewest. Anecdotally, Rycroft
(1981, p.132) writes that Hartmann
'mentions that he knows five
physicians who report a decrease in sleep requirements following
successful psychoanalytic treatment, and that he has not heard of any
cases of change in the opposite direction.'
Hartmann, Baekeland and Zwilling (1972) also related
subjects' personalities to amount of dreaming by comparing naturally
long and short sleepers. They found that all subjects had almost
identical lengths of NREII sleep, which accords with ideas of the
bodily, physically restorative, function of NREM sleep.
58
Short
sleepers were found to have significantly less mean REM density,
significantly less REM periods, and shorter mean period lengths (which
was not significant). Short sleepers were averaging 5.5 hours actually
asleep, long sleepers 8 hours. With the help of personality
inventories and interview impressions they concluded that short
sleepers were more efficient, ambitious, socially adept and
politically conformist. Long sleepers had a wider range of employment,
most had some neurotic problem, were less sure of themselves, and some
were artistic. Hartmann concludes that they are more in need of reprogramming than the "pre-programmed" short sleepers. The long
sleepers also had more primary process thinking in their dreams, and
remembered more dreams at home than the short sleepers. (With regards
to this latter point, Cohen, 1982, found that for male subjects, those
who usually sleep longer each night have greater recall of dreams.)
Upon REM deprivation long sleepers had a greater shortening of the REM
cycle than did short sleepers, indicating the greater need for
dreaming in the former group. However, note the figures provided by
Webb & Agnew (1970), who found that long sleepers had 53% more REM
sleep than control sleepers, and 36% more stage 2 than controls, while
short sleepers had as much REM sleep and stage 4 as controls,
obtaining their lack of sleep by moving through intermediary stages 2
and 3 quicker and more efficiently.
Problematic for the theories of the crea ivity of dreams is
the mundane nature of many of them (as documented by Hunt, 1982) and
also of many waking fantasies. Such fantasies are characterised not by
bizarre images but by a plausible narrative. Freud (1908, 'Creative
Writers and Day-Dreaming') writes that the first traces of adult
59
fantasy are present in childhood play, a creative manipulation of the
environment. He states that adult fantasies are composed of a present
wish, which 'harks back to a memory of an earlier experience (usually
an infantile one) in which this wish was fulfilled; and it now creates
a situation relating to the future which represents a fulfilment of
the wish '. Freud emphasised the narrative nature of the fantasy; it
is an imagined, plausible story with the aim of reaching some egoistic
or erotic satisfaction.
Creative Compensation During Dreams
Of relevance to this are the following studies of
compensation in dream sleep: Wood (1962) showed that social isolation
for the day led to an increase in "social intercourse" in the dreams
of the subjects. Similarly, Hauri (1968, 1970) employed 3 pre-sleep
conditions in an own-control design which had 6 hours of each of the
following: relaxation, physical exercise, and studying. He found less
physical activity in REM and NREM dreams combined in the exercise
case. Compensation is also evidenced in Bokert (1965), who found more
thirst-related imagery on nights in which the subject was waterdeprived, and a negative correlation between evidence of such content
and morning thirst. Newton (1970) found that newly paralysed
individuals reported more physical activity, and long-term paralysed
individuals less activity, than controls. Finally, e la Pena, Zarcone
& Dement (1973) found that information-processing, as quantified by
REM activity, compensates for acute excess or deficiency of
information-processing during the day.
60
Creativity
Contrary to Freud, Singer (1981) considers fantasy to be
healthy and present in even emotionally satisfied people. He writes
that regular daydreaming can make the real world more familiar, and
lead to less mistaking of internal imagery and thoughts for
hallucinations; he notes that in sensory deprivation experiments it is
the subjects with less experience of daydreaming who are likely to
report that they have hallucinations. (Note also that Wagner and
Stegman, 1964, report that adult schizophrenia is not linked with an
early history of daydreaming.) He notes that daydreams, like REM
dreams, decrease with age - and also that it is more prevalent in
upwardly mobile socioeconomic groups than in those whose status is
assured and certain; the members of the former, more unstable group
also described their fathers as being quite different from their own
self-ideal, and do therefore seem to be using fantasy as an
explorative tool. Similarly, Cartwright (1974a) studied the influence
of a conscious wish or desire to change a personally relevant trait,
this time in a dream paradigm. She discovered that the dream-world
deals with the problem by revealing different affective attitudes
toward the concern than those expressed in waking life.
The adaptive nature of daydreaming is evidenced by Symonds
and Jensen (1961) in their finding that major themes of adolescent
fantasies were still present in the subjects' ad it life and work.
According to Singer (ibid. p.148), children showing greater fantasy
are more curious and lively and also need to have the skill of
organised thinking in order to imagine plots and their endings. He
also claims that it is the less imaginative children who are more
61
aggressive (ibid. p.l36). However, lest we oversimplify this subject
in assuming that fantasy during waking hours and bizarreness during
sleep are not pathological, the finding of Starker (1974) must be
borne in mind. He found that more anxious and low self-esteem subjects
had more bizarreness in their dreams.
Bolton (1972, p.202) writes of such creativity as a distinct
ability of the brain, which can be evaluated numerically. 'The
correlation between intelligence and creativity, as defined by
performance on tests of divergent thinking, decreases with increase in
intelligence and decreasing test atmosphere. ... Different tests of
divergent thinking and different measures (i.e. fluency, flexibility
and originality) within the same test correlate highly among
themselves to form a unitary dimension, although verbal and non-verbal
components may be distinguished.' However, we are still left with the
question of what this activity consists of. Of relevance is work by
Polanyi (1967), who has distinguished between tacit and explicit
knowledge. We are focally aware of the latter, whereas the former is
like the ground from which the latter emerges. Such a
conceptualisation allows us to claim that one can rearrange what is
already vaguely known in order to produce new knowledge, which is then
explicit. This is in answer to Socrates' claim that the maieutic
method shows that what we discover by thought must have been already
known anyway. In Polanyi's sense, tacit knowledg is used to make
explicit knowledge. Of experimental relevance is the work of Mednick
(1962, 1968) on the Remote Associations Test. In this test three words
are given, and a fourth one which associates with them all must be
supplied by the subject. The major difference between high and low
62
scorers on the test is found to be connected with attention
deployment. When Mendelsohn and Griswold (1964) had subjects memorise
25 words while another 25 were played on a tape recorder, it was found
that high RAT scorers could utilise cues from both sets of lists when
later asked to solve anagrams which involved some of the words.
Furthermore, Laughlin, Doherty & Dunn (1968) found that RAT was a more
sensitive predictor of incidental concept learning, while intelligence
was a better predictor for intentional concept formation. The latter
seems to require focal awareness, the former subsidiary awareness, and
I suggest that this work may form another link between dreams, with
their loose connections and lack of analytical thinking (despite their
'single-mindedness'), and creativity. This work may help us to tie in
dreams with the notion of incubation, for dreams also may be seen as
the production of explicit knowledge from tacit knowledge, which is
aided by the predominance in dreams of elements which in waking life
are incidental and, quite literally, in the background. (Incidentally,
Paivio, 1971, suggested that the discovery phase of the creative
process is mediated by concrete imagery.)
Luria noted such a predominance of incidental elements in
the recall of his brain-damaged patient Zasetsky. He related it to the
normal condition of sleep, 'your thoughts are confused and you can
readily become disturbed by things that appear trifling during the
day. A cortical condition such as this, but patholo ically induced, is
what Pavlov termed a "damped" ... condition. In this state the cortex
functions far less precisely and is barely able to distinguish the
essential from the inessential; the dominant characteristics of
objects (which it normally would discern) cease to predominate, but
63
are "levelled off" with secondary, less essential attributes.' (1975,
p.9/.)
Connected with the activity of incubation is that of play,
for both require the relaxation of external constraints. Such is also
true of dreams, and, as some have claimed, of mythology. Note that it
is the external restrictions at the time of these activities that are
decreased: the restrictions of the environment before the activity
(that is, respectively for the above cases - the knowledge that is the
background to the incubation, the objects played with, the events of
the previous day, and the ethnographic background) are all rigidly
determined and complexly structured. Wallach and Kogan (1965)
similarly found that a creativity dimension of tests is only found
when the experiment is performed in a playful, non-evaluative context.
In other environments it was found that a measure of intelligence is
as good at predicting individuals' results in the creativity tests. On
a different note, but still connected with the notion of play in human
creativity, Campbell (1984, p.l54) writes of artistic epiphanies of
early humans which are irrelevant to strategies of adaptation - he
notes a 30,000 year old oval plaque from the Dordogne, carved from
antler or bone, which has an engraved serpentine chain looped and
folded back on itself. Similarly, human dreams may also have a large
deficiency of function (redundancy) compared to dreams of other
animals, with the intrusion of such playfulness and symbolism into
them. This redundancy may explain the lack of pathological
consequences in patients with complete suppression of REM sleep as a
consequence of taking anti-depressant drugs.
64
Imagery and Creativity
That REM dreams appear as an image, rather than, say,
chaotic verbal sounds and sentences, may be related to their proposed
active divergent thought function on the basis of the following work.
Hargreaves and Bolton (1972) found a relation between several
divergent thinking measures and performance on a pair-associated
learning task when subjects were told to form images linking each
pair. A greater ability to image each pair would lead to better
recall. Similarly, Durndell and Wetherick (1976) found that subjects
who could control visual imagery well performed two divergent tasks
better than a group with less control of imagery, while no
relationship was found between reported imagery ability and
performance on two conceptual tasks. Similarly, Shaw and DeMers (1986)
found that for 84 high IQ children 'imagery has an important place in
both the verbal and nonverbal dimensions of the creative process'.
They found that vividness and control of imagery were both correlated
with originality scores. They cited Shaw's conclusion that 'imagery as
involved here seems to be used not as a primitive (i.e. irreducible)
process for recall of passively stored information, but as a process
of active manipulation of the given information. Thus, imaging may be
able to account for individual differences in the transformation of
information in the incubation stage of the creative process.' They
conclude that 'originality' and 'flexibility' ar the aspects of
creative thinking that are most related to imagery. However, note the
problems with the claim that imagery is a separate means of
information-processing from that of the use of propositional knowledge
which have been explored by Pylyshyn, 1973.
65
CHAPTER 7
THE NEURAL BASIS OF MEMORY
The Theoretical Basis of Neural Networks
Numerous authors have modelled some of the functions of
brains using associative nets. This approach differs from traditional
Artificial Intelligence work in that the properties of the hardware
are important; high-level symbolic processing is not abstracted from
the hardware by which it is carried out. The hardware consists of
numerous units, the activation of each being dependent on the
activation of many others to which the neurone in question is directly
connected. These are different from digital computers although the
latter can simulate the operations that the former perform - the
computer just takes a long time doing so. Brains have a particular
style of computation, unlike digital computers having to weigh
conflicting and cooperating considerations very rapidly, which is
different from the style of digital computers. To mimic this nets are
built which store information in the connections between switches
which are either on or off. They store representations in terms of
their features, whereas computers store a piece of information at one
address, which results in computers being bad at generalising from
information already held, but being very accurate concerning what is
held. As an example of feature analysis, P.Winston f MIT developed a
computer program by which the machine distinguishes whether an
arrangement of bricks forms an arch or not, but the problem takes
hours to solve. Similarly humans take hours to solve the different
types of problem that are rapidly processed by a computer. The
66
difference is because of the serial operation of computers as against
the parallel operation of the brain, and also of associative nets. For
example, the visual system has many inputs, with cells at each level
depending on information from many cells of the previous level, and on
lateral inhibition from neighbours. Kohonen actually found that
without this lateral inhibition his model of learning to recognise
faces from only one presented half leads to a very blurred image on
recall.
New attempts to create intelligent systems are thus
mimicking the neural structure of the brain, rather than trying to
imitate human thinking by writing programmes which perform some of the
symbolic manipulations of human thought. An associative net is for the
most part unprograinmed, the solutions are found by the system settling
down into a stable pattern of neural firing.
An associative net can be two arrays of cells, for the input
and output, or one array with intermediary cells and the input and
output cells arbitrarily assigned.
OUTPUT
c
- '-\
PUT
INPUT
OUTPUT
Both can work as an autoassociative net in that a pattern
can be typed onto the first array which leads to an output which can
67
be fed back in as the input until the system is at equilibrium (that
is, either the input and output arrays are identical or, in other
types of net, none of the units switch any more). This final result is
a memory.
There are various ways in which the net can be programmed
with a memory, which each involve clamping the units and also the use
of a synapse correcting algorithm:
Decrementing net - if an output cell fires when an input cell fires
then the link between them is cut, otherwise it is left. Young (1978)
reports this effect in the memory of the octopus, and likens it to the
chipping away of stone to make a statue.
Incrementing net - if two connected cells fire together the link
between them is incremented.
Hopfield net - link gains are adjusted according to the product of
input and output cells' firing rates.
Similarities of Neural Networks with Biological Brains
There is now evidence that memories are stored by synapse
connection strengths (it is found that the activity of the enzyme
calpain increases if the synapse is used frequently), and nets so
built have many properties in common with brains.
1) Generalisation, for many of the same synapses are
used by different memories.
2) A complete representation can be constructed from a
partial input. Only the partial input is needed for the net to search
68
its memory for the complete memory. As Ackley, Hinton & Sejnowski
(1985) state:
"Many problems in vision, speech recognition,
associative recall, and motor control can be formulated as searches.
The similarity between different areas of cerebral cortex suggests
that the same kind of massively parallel searches may be performed in
many different cortical areas." The ability to reconstruct a memory
from a displaced input is also a property of holograms; Wilishaw,
Buneman & Longuet-Higgins (1969) expand on this to show that useful
information storage features of holography can be found in associative
nets.
3) "Graceful degradation" is possible, the assembly
does not fail completely if some of the elements are lost. Similarly,
brain cells are lost from birth onwards but forgetfulness only
gradually increases, and that only after middle age. (Amnesias due to
brain damage arise from limbic system rather than temporal lobe
damage.)
4)
Sutherland (1986) points out that the mistakes
made by the nets when learning have similarities to the mistakes made
by children, for example, in learning regular and irregular verbs.
Although it is the study of vision that is the most advanced
of all the studies of the neural basis of the senses, it is work on
smell which has provided new insights into the network theories of
memory and thinking. Vision is topographical, whereas smell only
identifies things, and so whereas research on the former could
concentrate on how the brain extracts a mental map from the retina,
69
research on the latter has had to meet the problem of the role that
memory plays in the senses. This is because smell is a matter of
association, each smell is significant because of the memory it
evokes. Lynch (reported in The Economist, 29/6/85, author unnamed)
noticed the similarities between the brain's olfactory memory circuits
and computer circuits designed to associate concepts, in that each
nerve cell has connections with many others, including itself via
feedback routes. Lynch says that smell is closely connected with
memory and association, for the nose sends nerves to the olfactory
bulbs and thence to the hippocampus, via pyriform cortex. The pyriform
cortex also sends connections to the frontal lobes. The importance of
the sense of smell to mammalian evolution is noted by Jerison (1976).
He claims that early marmnals were nocturnal and hence needed to
enlarge the areas of the brain dealing with smell and hearing,
because, unlike with vision, which has some retinal processing, there
was no other site for the improved processing to take place. Later,
with the extinction of the dinosaurs, the mammals invaded daytime
niches and had to encephalise their visual processing in order to
integrate it with auditory and olfactory information. He proceeds to
describe language as a sensory system, which aids in the creating of
labels and the modelling of the real world. Dreaming, I suggest, would
then be the use of this 'sense', without the participation of the
other five. It was only later, he claims, that lang age was used for
communication between people, and I note that at this point the theory
is similar to that of Jaynes (1976). Jaynes writes that people have
only been self-conscious for a few millennia. Before that the righthemisphere of the brain made commands, which were heard as voices,
70
about activities that were needed. Only recently have we evolved to
think about activities consciously, and the voices are thus now
redundant. He considers that old myths of hearing the gods speak are
an atavism from that time, as are the voices of schizophrenia now.
This notion of an internal language is very apposite to the study of
dreaming.
Hebb (1949) formulated a rule for how assemblies of
neurones form. "Any two cells or system of cells which are repeatedly
active at the same time will tend to become 'associated', so that
activity in one facilitates activity in the other." The first cell is
more likely to activate the second one in future. Hebb's ideas are
also used by Hinton and Sejnowski's Boltzmann machine in its efforts
to distinguish a figure from its background. In 1973 Stent added the
possibility that connections between an active and an inactive cell
are weakened. Malsburg then noted that fluctuations in activity must
be synchronised to a few thousandths of a second for mutual
strengthening to occur, otherwise there would be confusion between
overlapping assemblies that should not be connected. Singer used
Hebb's ideas to explain how cells in the visual cortex of kittens
develop preferences for information coming from one eye rather than
the other.
Experiments with the mollusc Aplysia have shown the
molecular changes that occur at synapses during learning. The snail is
given an electric shock which sensitises the animal to give a strong
defensive reaction even to a light touch. The response is of
withdrawing its gill. If no further shocks follow the effect wears off
after about half an hour. Inhibitors of protein synthesis prevent the
71
long-term sensitisation while leaving the short-term version
unaffected. Kandel showed that the inhibitors of protein synthesis act
at the same site as the learning changes, rather than affecting the
whole animal (as happened in earlier experiments). Inhibitors of RNA
synthesis also block the long-term effect. After learning, new
proteins are found by using two-dimensional gel-electrophoresis. These
proteins might be involved in altering the area of contact between
cells, changing the number of synapses, or influencing the production
of enzymes that regulate the transmission of electrical signals. One
group found that upon learning a protein in the synapse is
phosphorylated, thereby closing channels in the cell membrane that
allow potassium ions to flow out. Such an outflow is needed in order
to return the cell from an excited (depolarised) state to a resting
state. The cell stays excited for longer and strengthens the
connection. It is the convergence of several inputs simultaneously
onto one cell that is crucial. One input activates the cell, which is
then simultaneously active with the other input cell - the connection
is then strengthened, or, in neural net terms, the connection weight
is increased. In addition, Bliss and Lomo found that electrodes giving
high-frequency stimulation to the hippocampus strengthened the
connections between a pathway there and the area that pathway
connected with for several weeks. This they called long-term
potentiation and later experimenters found that its prevention affects
the learning of spatial tasks in the rat. Crick (1984) suggested a
mechanism whereby such synaptic changes can be permanent despite the
rapidity of protein turnover. A summary of work on the molecular basis
of memory is provided by Rose (1986). (Note, however, that Igor
72
Aleksander, a pioneer in neural network theories, is now postulating
that synapses may not change gradually, and that each neurone does not
sum its inputs but rather acts as a pattern recogniser in a network of
logic neurones. This removes the need for single neurones to have to
calculate changes in weights when learning.)
The network models have much In common with cortex: all or
nothing spikes, thresholds, refractory periods, delay time for signal
transmission, inhibitory and excitatory post-synaptic potentials and
stochastic neural activity. Biological systems and the Hopfield net do
not have rigid synchronicity of firing. The Hopfield net mainly has
local connections, similar to visual cortex which has a quasicrystalline layered structure with dense short range interactions
within a column and diffuse longer range axonally mediated connections
between columns. Some neurones have both inhibitory and excitatory
inputs. Also, dendro-dendritic synapses have been found which do not
use action-potentials, our models will have to do this also.
Moreover, it is the decrementing net that is popular for a
model of learning in the cerebellum (Albus, 1971) and is far simpler
than the Hopfield net which postulates that the synapses must have a
memory for their own previous firing rate and that of the connecting
synapse, and must be able to calculate the change required for
learning. The decrementing net needs no such precision, the link is
either there or not. Like brain cells (including soir that respond to
different degrees of profile), the units have rapid changes in their
rate of firing, they all have a threshold and information is conveyed
according to whether a cell is firing or not, and at what rate.
Hopfield's cells have a firing rate of + or -1, while the original
73
correlative nets had cells with a variety of firing rates (-1, -2,
+3...). Similarly, biological neurones have action-potentials at a
rate dependent upon the input current, and many have a graded response
rather than action-potentials. A Hopfield cell would sum the gains
from cells leading to it and then reset itself at either +1 or -1. A
later version of the net, though, (Hopfield and Tank, 1986) used nonlinear, graded response neurones instead of the 2-state type, and was
found to be able to solve various more difficult computational
problems. However, an unrealistic property of the net is that a
negatively firing unit coupled with a negative connection weight
causes the positive firing (- x - = +) of the output cell.
In practice it is found that the Hopfield net works best
when there are roughly equal numbers of +vely and -vely firing cells
in the whole domain, otherwise many more mistakes in learning are made
per pattern learned. (If many separate memories had many non-firing
cells then the memories would correlate; such correlations, although
necessary in brains, cause problems with the Hopfield net.) At best
the network starts to make errors after learning O.15N patterns, where
N is the number of cells in the network. It still does better than a
decrementing net though, as long as it doesn't have a low number of
+vely firing cells. (Note that the brain does have such a paucity of
firing cells at each point in time.) Unlike the other nets, but like
animal brains, the Hopfield and correlative models cn be disturbed by
inputting the opposite of a pattern already learned, by varying the
sequence in which patterns are put in, or by showing one pattern too
many times.
74
How the Networks Behave
The simplest sort of net has direct connections between all
input and output neurones. In order to teach the net a pattern of
neural states the neurones are clamped with the desired firing
pattern. The connecting weights are then adjusted by an amount equal
to the product of the neural firing rates, that is, if both are
positively firing then the connection weight increases by 1, if one is
positive and one negative then the connection weight is decremented by
1. If the two neurones are set at -1 and -1 then the connection weight
is incremented by 1. Once this algorithm has been completed the system
can be tested - if the input is not yet learnt then the algorithm is
repeated. Once the input pattern is learnt the network can be taught
another input pattern. However, often there is no suitable set of
weights for all the vectors together. Intermediate levels must be
introduced which extract a hierarchy of features from the input
vector, and also cross-talk is needed between members of the same
layer. An example of a simple task which requires intermediate links
is teaching a net to distinguish two sorts of inputs, the input being
two 8-bit arrays, one of which is a shifted version of the other. The
net has to respond according to whether the shift is one space to the
left or one to the right. The difficulty is that each input bit,
considered in isolation, provides no information about what the output
should be, and therefore combinations of bits musc be considered,
which needs intermediate units that extract information combinations.
A whole list of such problems is provided in "Learning Internal
Representations by Error Propagation" by Rumeihart, Hinton & Williams
(1985) ICS Report 8506. This includes the XOR problem, which has the
75
following input and output vectors:
INPUT
OUTPUT
00
0
01
1
10
1
11
0
Again the input and output patterns are very different. A hidden unit
is needed to give an internal representation of the input patterns.
TPUT
0.5
The firing threshoLd for the hidden
and output units are 2 and 0.5.
Such hidden units were first used in the perceptron. In the
Hopfield net, which has a vast number of homogeneous units, any of
which may be designated as input or output units 1 any unit not chosen
as an input or output member can be thought of as an intermediate
76
unit. Initially there was a history of failure in searching for
learning algorithms for complex nets, because when the network did the
wrong thing it appeared to be impossible to decide which of the
internal connection strengths was at fault. (With perceptrons the
convergence theorem just guaranteed the weights for a single layer.)
However, there are various ways in which the hidden units can obtain
the correct threshold and connection weights: the experimenter can
assign the values or the experimenter can hope that the system will
itself set the correct values as it does when there are no hidden
units. Minsky and Papert (1969) were pessimistic about learning in
these multilayered systems, and Rumeihart, Hinton & Williams (1986)
found that the learning algorithm sometimes led to an incorrect stable
set of threshold and connection values, which they pictured as a
minima rather than the desired minimum into which the system should
have fallen.
It is hard to analyse such nets with their cross-talk within
layers and feedback from later layers to earlier ones. Hopfield's
asynchronous net, however, can be analysed in terms of a state space
like a bumpy surface (Hopfield, 1982), as can the Boltzmann machine.
Diagrams of the state space are given in Hopfield and Tank (1986) and
in Tank and Hopfield (1987). (The former paper also analyses the
computational properties of such systems with regards to optimisation
problems.) Each point in the space corresponds to a ertain pattern of
active and inactive units. The energy of a pattern is defined as minus
the
SUI11
of all the weights between active units in the pattern. A very
stable pattern will have very low energy and be pictured as a trough
in the state space. The system drifts into the parts of lowest energy,
77
without oscillating. Hopfield and Tank (1986) note that a nonsymmetric net would oscillate, and provide trajectories with great
computational power - however, the mathematics of such complications
are not understood as yet. If two units have a big positive weight
between them then patterns in which they are both active will be low
in energy, the system will tend to settle into such a state - but it
will tend to avoid a state where two units with a large negative
weight fire together. Problems arise though because a given input
value can produce more than one output depending on the initial states
of the intermediate units and the order in which decisions get made.
We can picture the instantaneous state of the network as a
ball bearing on a bumpy surface, we want to ensure that the relative
probability of ending up in two different minima depends only upon
their relative depths, rather than on the contingencies just
mentioned. We want the ideal situation where the logarithm of the
probabilities of being in one of two states is just their energy
difference. We will then have control over the probabilities of each
output, a control which just depends upon what the net has been
taught. To escape from local minima and to find the point of lowest
energy Kirkpatrick, Gelatt & Vecchi (1983) suggested starting with a
lot of shaking of the surface, resulting in 'thermal equilibrium', in
which the ball bearing constantly moves. With a gradual reduction in
shaking the system would evolve toward the state wit least energy this process was called "simulated annealing". To do this in parallel
networks the decision rule is changed. Originally a unit would turn on
when the sum of its inputs was positive, off when they are negative this algorithm always leads to a decrease in energy. Now the decision
78
is taken probabilistically so that it becomes possible to increase in
energy, in order to jump out of a local minima. Each output vector at
equilibrium will then have a fixed probability that doesn't vary with
time. These are called Markov Random Fields, and the Boltzmann machine
is one of them. Such a machine learns to give outputs to certain
inputs with certain probabilities, and this learning procedure can
solve the 8-bit shift problem, for the machine can figure out what to
do with internal units whose required behaviour is not specified for
it from outside. Such a machine would thrive on the thermal noise
already present in transistors, for the logic of the system is made
less rigid by the presence of noise.
Ackley, Hinton and Sejnowski (1985) interpret a unit being
on or off as meaning the system currently accepts or rejects some
hypothesis about the environment. A positive link between two units
means that two hypotheses tend to support each other. Link weights are
thus symmetric. "The energy of a configuration can be interpreted as
the extent to which that combination of hypotheses violates the
constraints implicit in the problem domain, so in minimising energy
the system evolves towards 'interpretations' of that input that
increasingly satisfy the constraints of the problem domain."
A Hopfield net cannot code for sequences and neither can a
Boltzmann machine; this is because they only have symmetric links. For
memorising sequences, as opposed to just memorising single patterns,
asymmetric links must be present, as they are in the cortex. A
similarity of the Boltzmann machine to cortex is the probabilistic,
asynchronous decision making. This stochastic element in neurones is
based on fluctuations in amount of transmitter released by the
79
presynaptic terminals.
For the Boltzmann machine a variation in "temperature" (the
measure of randomness in neural firing) of 25% from unit to unit does
not affect the annealing and equilibrium solutions, but its effect on
learning is not yet known. Such a variation would be expected in
cortex.
How the Nets Misbehave
Hopfield uses +1 and -1 as the possible firing states for
each unit, rather than the states 0 and +1 used by cortex. For a
fuller description of the workings of the net see appendix 1. His net
is adversely affected by the sequence of inputs and by multiple
inputs, but then, this is seen in cortex. De Bono (1969) shows that
all "memory surfaces" will be affected by the sequence in which
learning took place. However, a strange property of their net
discovered by Hopfield, Feinstein & Palmer (1983) was that some
memories are easier to recall than others, and also that there is a
class of parasitic traces of the type below:
inputted memory 1
- ++-+-+- -
inputted memory 2
+-+-++
inputted memory 3
++- -++- - +- -++- -+
parasitic memory
±+++---- +--++--+
They state that this would have a human counterpart of:
80
inputted memory 1
Walter white
inputted memory 2
Walter black
inputted memory 3
Harold grey
parasitic memory
Walter grey
but note that the second parts of computer memories 1 and 2 cancel
each other out, which is a very contrived situation. Obviously simple
arrangements of + and - can be so contrived that exact opposites
result, but surely "black" and "white" will not just have precisely
opposite neural firing patterns. They have much in common via their
connections with colour and light, semantically they are antonyms but
neurally they could well be very similar, as evidenced by
their frequent connection in word association tests (Clark, 1970).
They will thus have
overlapping neural representations. Another
objection to such a contrived system is that in the above example the
net is being taught to learn two outputs for the same input, i.e.
so it is no wonder that it provides confused outputs.
With regard to this suggestion that the problems of
parasitic memories and unequal recall strengths of memories are
because the Hopfield net is contrived to produce them, I refer now to
Runielhart, Hinton & Williams's (1986) experience that:
"... the error surface may contain local minima ... we
have only encountered this undesirable behaviour in networks that have
just enough connections to perform the task. Adding a few more
connections creates extra dimensions in weight-space and these
dimensions provide paths around the barriers that create poor local
minima in the lower dimensional subspaces." Adding more neurones also
81
eliminates the problem.
The contrived nature of the Hopfield net, and hence its
inapplicability to the case of brain cortex and dreams is further
illustrated by the microchip version of the net. The chip is designed
to work with almost equal numbers of +1 and -1 units at any one time
(completely different from the case of cortex), and, related to this
(in that a preponderance of - ye or +ve units means that memories are
not orthogonal), "... any correlation between memories causes some
memories to be weighted more heavily than others. Statistical
orthogonality between memories minimises this effect" (Sivilotti,
Emmerling & Mead, 1985). Also, "... the 'attraction' of stable states
decays exponentially with the number of memories subsequently added to
the matrix" (ibid.), which shows gross inefficiency. It may be claimed
that the unlearning function is only needed because the net is so
inefficient in the first place.
In order to eliminate these difficulties Hopfield et al
(1983) propose the application to the net of an 'unlearning
algorithm'. This involves giving the network a random input and
allowing it to settle into the nearest minima or minimum. When this
state is achieved all excitatory connection weights between similarly
firing units (either both positive or both negative) are decremented
by a small fraction, and all inhibitory connection weights between
similarly firing units are made slightly more inhibit ry. In this way
parasitic responses which are too prone to be set off by random noise,
rather than by a structured input, are weakened. In doing this the
unlearning mechanism "separates the memories" so that they interfere
less with each other, and so that their accessibilities equalise.
82
(This algorithm is similar to that used by Hinton for the
process of learning. His learning procedure has two phases:
1) Clamp the input and output terminals as required.
Where both are active the connections are strengthened.
2) Clamp the input terminals and see what output is
obtained. At first an incorrect answer will probably occur, decrement
by the same amount as in the first phase all connections between
active output and active input neurones.
This algorithm is continued until the correct answer to the
input is obtained. The next input and output vectors are similarly
learned.)
However, although the overlapping of traces is obviously a
disadvantage for a simple storage system, it is essential for an
intelligent system. Gardner-Medwin (1976) states, with regard to his
own design of an associative net, 'The associations between elements
in human experience are by no means random and equiprobable in the way
assumed for the analysis in this paper'. Similarly, G.Hinton (1985)
states:
"If there is some neat regularity in the mapping from input
vectors to output vectors, we would like the [system] to 'capture'
this regularity.., to give the 'correct' output vectors for input
vectors it has never seen before." Surely unlearning, the
orthogonalizing of memories which in reality are correlated, will
decrease the possibility of this 'capturing'. An example of a
regularity is:
83
INPUT
- -+
-+-
OUTPUT
+- -+-
where the middle digit of the output is
always the same as that of the input.
now a random input, as required for the unlearning
mechanism, will similarly evoke -- or -H- as the middle digits, and
decorrelation will then be applied; this must have the effect of
interfering with the system's coding of regularity, and hence with its
ability to act intelligently.
Problems in Evaluating the Unlearning Algorithm
What distinguishes these putative bizarre traces from valid,
adaptive ones, and why are the former supposedly eradicated in
preference to the latter by the unlearning mechanism? It is not being
claimed by Hopfield et al that they are simply weaker, but it is
parsimonious to hold that their relative weakness is the explanation
for the result of Hopfield et al, 1983, shown overleaf.
84
S.
-o
U
IUU
UU
No. of' unlearning trials
Fig. 1 The traction of random starting states which leads to
particle memories (accessibility). The five dashed lines are the five
nominal memories. The solid line is the total accessibility of all
spurious memories. In these trials e was set at 0.01.
85
There is no mention of the experiment continuing after the
initial eradication of most of the parasitic traces in order to
discover whether the remaining parasitic traces would likewise
decrease, and whether they would decrease faster than the valid
traces. I suspect that their result was due to the weakening of all
memories in the net, resulting in some of the very many small
parasitic memories being eradicated.
Unlike in human memory the valid memory items in the
Hopfield net are all initially of the same strength. An important test
of the unlearning theory would thus be to introduce a weakly learned
valid trace and to see the effect of unlearning upon it. In a human
memory there are certain traces which are called up frequently, such
as concerning one's job or friends. If these are valid and important
what protection have they from the unlearning mechanism, which might
mistake them for obsessions? A human memory does not have equally
accessible states, the production of which is one of the aims of
unlearning. In reply, Hopfield et al can argue that equally strong
memories become unequally accessible, and unlearning acts to restore
them to their true strengths, but surely it is the presence of
correlations that results in this inequality, and correlations are the
basis of intelligence. Valid correlations and frequently accessed
thoughts are threatened by the unlearning algorithm. They list two
classes of traces which must be weakened, obsessions and parasitic
bizarre traces. They claim that the unlearning function would
preferentially decrease such traces, because the method has "a
feedback algorithm about the accessibility of particular states." But
surely in learning if an input leads to a wrong output it is a case
for learning again, a random input will not cause a weakly learned
86
connection to be strengthened.
Possibly such a simple forgetting algorithm would give much
the same result as Hopfield et al (1983). The usefulness of such a
forgetting function is described by Levy (1988). However, there are
differences between the forgetting and unlearning functions in that
the former would lead to a general weakening of all connection
weights, whereas for Hopfield's algorithm the memory state must be
accessed first so that the change in weight for that particular
connection can be calculated. According to the decrementing equation
used in Hopfield et al (1983), unlearning has the effect of making
inhibitory links even more negative; notably Clark, Rafelski & Winston
(1985) define the combination of decreased excitatory links between
positively firing cells and the increase of inhibition between such
cells as 'forgetting'. Of relevance to these functions is the finding
of Cooper, Liberman and Oja (1979) that immature cells between the
lateral geniculate and visual cortex become more specific in their
response to visual patterns over the first 50 days of life in the cat,
but that dark reared cats show a decrease in specific cells.
Specificity can be regained with a return of patterned input, whereas
a random input leads to a widening of the cortical cell tuning curve,
irrespective of how sharply tuned the curve was originally. They found
that the gain in specificity is much faster than its loss due to
random input - I conclude that the effect of Hopfield et al's proposed
random stimulation on valid memories would be quickly remedied when
learning has to take place again, and that this experiment does not
argue against the application of the unlearning theory to mammals.
Cooper, Liberman & Oja (1979) provide more information about the
87
acquisition and loss of neural specificity.
The Brainwashing Algorithm
The 'unlearning' algorithm has similarities to the
'brainwashing' mechanism of Clark, Rafelski and Winston, which is
described in "Brain without Mind" (1985). This plasticity algorithm
causes a change in the initial quasirandom connectivity of a newly
formed net, in that excitatory connections between positively firing
cells are decremented, while inhibitory connections between positively
firing cells are made less inhibitory. In the latter case, and in that
there is no change made to connections between units that are not both
firing positively, this algorithm differs slightly from that of
Hopfield et al (1983). In addition, note that the unlearning algorithm
is applied after learning has taken place, whereas the brainwashing
algorithm is applied in order to prepare a newly-formed net for
learning. This helps the net to avoid catastrophic behaviour such as
'epilepsy' or 'dying'. In contrast to Hopfield's net, which drifts
towards one state of lowest energy, the net described by Clark et al
drifts into a cyclic mode (which they relate to short term memory).
They describe four types of net, differentiated by whether the number
of inputs to a unit is fixed or variable, whether the number of
outputs per unit is fixed or variable, and whether the value of the
connection weight is set at 1 or can be varied (symboli ed by If
the weight is set at 1 the net learns by cutting, rather than
adjusting, connections. If the weight is variable the net is called
plastic rather than static; there are then sixteen possible
modification schemes which are guided by ongoing neural activity. This
88
figure is arrived at because there are four configurations of neural
pair states (^+, --, +-, -+), the link can be excitatory or
inhibitory, and the value of the weight can be incremented or
decremented. (Overall the total value of all excitatory and all
inhibitory connections is kept constant, so that the net doesn't die
or become epileptic.) Like changing the number of output or input
connections this change in the weights has the effect of making the
units more dissimilar, leading to longer periods of cyclic modes, more
"eligible" neurones (those that are not constantly either on or off)
and not too many firing at once, for then next they would all be
within their refractory period at the same time. In order to reduce
the number of neurones active at any one time (which is more like the
actual brain) and to increase the cycle mode period we increase the
number of inhibitory neurones.
There are four useful algorithms, each of which uses a
combination of the sixteen possible modification schemes that can be
applied to the connection weights:
1) Brainwashing; this punishes all active units, i.e. when
unit i fires at time T and unit j fires at time T+t then their
connection weight decreases, whether the connection is excitatory or
inhibitory. No external stimulus is present. Clark, Winston & Rafeiski
(1984) applied this to nets with initial quasi-random connectivity in
order to make the net have more complex cycle modes. Note that the
algorithm is applied before any memories are in the net, unlike in the
Hopfield model. It is also not intended to reduce the accessibility of
spurious states, rather it prevents catastrophic behaviour in the
reverberatory system. The net is made less dependent on its initial
89
conditions (such as initial activity, refractory time, threshold
values) and hence better able to learn; the initial weights
distribution also has less influence on the ultimate distribution in a
brainwashed net. Irregular wide swings of activity occur, unlike the
initial regular, high level activity. Clark
et al (1984) found
experimentally that a (fraction of active neurones) at t=0 was 0.6,
leading to a=0.7 for the rigid net and 0.2<a<0.45 for the brainwashed
net.
2) Attention; the net is made more receptive to incoming
stimuli than to its own stimuli. This is performed after brainwashing
and before learning.
3) Engramming; active channels with excitatory links have
weights increased, if inhibitory the weight is decreased. The net is
clamped to an external stimulus.
4) Forgetting; there is no external stimulus. Active,
excitatory weights are decreased and active inhibitory links are
increased in value. This differs from brainwashing in that inhibitory
links are increased, which has the effect of decreasing trace
strengths. In the brain it has the result of two neurones not firing
at all, or their firing not being correlated.
Thus, a net is brainwashed, given a stimulus and we then
wait for the cyclic mode, which is then engrammed. This could then
either be forgotten, or more memories added. They state that
brainwashing can be applied later on in order to "even out the
accessibility of engranimed memories" and mention that brainwashing
'or anti-learning' may reduce the interference between recently
consolidated traces. However, they do say that these two points are
speculative.
90
Conclusion
Hofstadter (1979, pp.299-300), writes about how computers
are programmed in a language which would be almost untranslatable in
terms of the lower levels of machine language, or the even lower
matrix of the ^ or - state of all the gates. I suggest that there
remains the possibility that associative network theorists have
exceeded the experimental evidence in trying to explain the high-level
symbolic activity of intelligent thinking in terms of the brain's
hardware, as if all cognitive rules will simply emerge from a simple
network. Simple networks may have some properties not present in
larger networks, for example, Minsky (1975) suggested that small
barriers may exist inside minima in order to differentiate concepts
within a class, an effect which would not be seen in small networks. I
draw support also from Hopfield and Tank's (1986) comment that:
'Hierarchy is necessary to keep the number of synaptic connections to
a reasonable level. To extend the present ideas from neural circuit to
neural system, such notions will be essential'. The hierarchical
organisation of nets, as in the brain, may also obviate the need for
the reverse learning algorithm. Similarly, Johnson-Laird (1987) draws
attention to the differences between mistakes in language learning by
children and in the learning of rules for the past tense by
associative nets. He warns that it may be that 'high-level principles
are paramount, and that parallel processes are merely he brain's lowlevel language into which all mental life must ultimately be
translated ... At the other extreme, another option is that the mind
contains no high-level principles: there is only parallel distributed
processing from which all behaviour emerges.' (As an example, the
91
'cocktail-party effect' in psychology works on a higher level than the
'attention' algorithm of Clark et al, 1985.) This caveat to the
paradigm must be held in mind as we now proceed to study two theories
of REM sleep which are based upon neural network theories, and which
provide a link with the psychological and psychoanalytic theory of
dreams proposed by Palombo.
92
CHAPTER 8
THREE CURRENT THEORIES OF THE PURPOSE OF REM SLEEP
In the last chapter it was shown that state changes have
been found useful in the working of associative nets, that for
efficiency the nets are made not to remain in the same learning and
recall state all the time. We will now study two attempts to
assimilate this knowledge of the behaviour of neural nets to the
psychology and physiology of dreams and REM sleep.
The Crick and Mitchison Theory of Dreams
In 1983 Robert's (1886) idea of dreams as the excretion of
pathological thoughts was revived. The new theory was based on the
observation that some dreams contain bizarre condensed images. It
attempted to explain this by claiming that the components of such
images are real memories which have become mixed together: it stated
that a time is set aside by manunals for calling up these 'parasitic
memories', which then have a process of 'reverse learning' or
'unlearning' applied to them, so that the real memories are
disconnected from each other, and again made separate. A neural model
of this was made by Hopfield, Feinstein & Palmer (1983, see last
chapter), who found that if too many memories are learnt by the
network of artificial neurones then the network recalls mixed memories
as well as the real ones, because the memories overlap due to being
distributed over the whole net. Hopfield et al applied the unlearning
procedure to their neural network and found that it eradicated some of
the mixed memories, leaving the system working more efficiently. Crick
and Mitchison claim that such a mechanism occurs during REM sleep, and
93
allows mammals to fit more memories into a smaller brain than if it
was not applied. They also claim that parasitic memories, if not
eradicated, could be a cause of obsessions, recurrent dreams, a
tendency towards far-fetched associations ('fantasy'), over-response
to inputs ('hallucinations'), and psychotic thoughts. Such a
connection between dreams and insanity has a long history, witness the
statement by Hartley (1801, p.389) that:
'The wilderness of our dreams seems to be of singular use to
us, for interrupting and breaking the course of our associations. For,
if we were always awake, some accidental associations would be so much
cemented by as that nothing could afterwards disjoin them; which would
be madness.'
Before unlearning is applied the memories have to be evoked.
'The major inputs and outputs of the system should be turned off, so
that the system is largely isolated. It should then be given
successive "random" activations, from internal sources, so that any
incipient parasitic modes would be excited, especially if the general
balance of excitation to inhibition had been temporarily tilted
towards excitation. Some mechanism is then needed to make changes so
that these potentially parasitic modes are damped down.' (1983, p.112)
They claim this evocation is caused by the intermittent PGO
stimulation from the brain-stem, which is one of tte physiological
indicators of REM sleep. Whichever memory traces react to this
stimulation are damped down, leaving a greater proportion of nonparasitic traces. The theory holds that this will make the memory more
efficient. Note that Docherty, Bradford & Jan-Yen Wu (1987) have found
94
that some nerve cells can change from activating a neighbouring cell
to inhibiting it at other times, showing that temporary changes in
rules of neural functioning between two specified cells can occur in
biological systems.
Hughlings Jackson (1931) wrote about "doubling of
consciousness", where a patient shows a dreamy state with remnants of
normal consciousness at the same time, and Penfield and Perot (1963)
produced the same by electrical stimulation of the temporal cortex.
These two examples, and that of lucid dreams, show that it is
possible, physiologically, to 'dream' and to be conscious at the same
time. Crick and Mitchison have arrived at a function of REM sleep
which requires that we be unconscious, in that the hallucination of
what is being unlearned should not be remembered, as that would
vitiate the whole process.
An experiment that can be read both in favour and against
the theory is Feinberg's (1968) finding that there is a positive
relation between the amount of eye movement during REM sleep and
estimates of intellectual ability in a group of retarded adults. The
assunied greater incidence of PGO stimulation could be because the
brighter ones have to eradicate more false-associations than the less
intelligent ones, due to having more memories, or alternatively,
against the unlearning theory, because during sleep the brain carries
out some other cognitive function. Feinberg notes that actual brain
damage could be the cause of the correlation.
With regard to its emphasis on the functional significance of
PGO stimulation Crick and Mitchison's theory has similarities with the
Activation-Synthesis theory of Hobson and McCarley (1977). In the A-S
theory the activated forebrain "synthesises the dream by comparing
95
information generated in specific brain-stem circuits with information
stored in memory" (ibid.), whereas in the unlearning theory the brainstem stimulation contains no information at all. Instead, information
(usually parasitic) is said to be evoked by it.
The Activation-
Synthesis theory states that the best fits to the incomplete data
provided by the primary stimuli (from the pons via the vestibular
system) are called up from memory. The primary sensorimotor stimuli
thus provide a frame into which ideational, volitional and emotional
content is projected to produce the final dream images. Such features
as scene shifts, time compression and condensations may thus be
dictated by the brain-stem. Similarly, the unlearning theory contends
that condensation will occur following a PGO burst. An objection to
the A-S theory is that PGO stimulation is not a sensory input at all.
However, note that it is far from evident why it should be the
putative parasitic traces that respond more often than valid traces to
this supposedly random input, as was noted in the last chapter.
Although the Crick and Mitchison theory seems to predict
that more primitive mammals will need less REM sleep (in fact, humans
have the same proportion as moles, and there is evidence that REM
proportion correlates better with degree of binocular vision than with
evolutionary complexity) the additional variable of brain size means
that no such simple prediction is really made, for a greater brain
size would supposedly separate memories, obviating the need for an
increase in REM sleep time.
They note, as does Winson (1985), that the Australian spiny
anteater is a mammal which does not have REM sleep, but instead has a
massive frontal cortex. Whereas Winson explains this by the claim that
96
the larger cortex is needed to connect memories together, which other
inanmials achieve in REM sleep, Crick and Mitchison claim that the
larger cortex is needed in order to separate memories which would
otherwise overlap.
Passive Evocation of Mixed Memories,
or Their Creative Formation?
In order to provide more information on what the experience
of dreaming involves, which will be useful for evaluating the three
theories described in this chapter, I will refer to Berger (1963), who
presented first names to subjects during stage 1 sleep. The
incorporation of the stimulus into the dream was judged by an
independent experimenter matching the list of dream contents with the
list of names. It was found that emotionally meaningful and neutral
names were incorporated to an equal extent, and there were no EEC
changes associated with incorporation, indicating that this thoughtlike activity was occurring during uninterrupted dreaming. In order of
decreasing frequency the incorporation was by:
1) Assonance (Similarly, Jung (1918) noted an increase in
clang-reactions during word-association experiments with a drowsy
subject, showing the importance of assonant processing in subjects who
are not fully awake.)
2)
Direct incorporation
3) Incorporation by a direct personal association appearing
in the dream
4)
Name represented by a symbol or action in the dream.
97
This result fits in with all theories which claim that
dreams are a kind of thinking, for example, Evans' (1983) computer
theory states that we think out future actions in our dreams, and that
we need to be asleep lest envirorunental stimuli are incorporated into
the practising programs, although his theory predicts a more rational
and less poetic type of thinking than here.
In waking psychology Voss (1972) categorised the various
ways in which stimuli are encoded and related to each other into four
dimensions:
1)
Formal dimension (how many letters in the word)
2)
Syntactic dimension (type of word)
3)
Associative dimension (what the word can be associated with)
4) Semantic dimension (what the word means, or can be taken as
meaning)
He found that distraction will eliminate the higher levels
of encoding, starting with number 4) above. I relate this to Berger's
(1963) result that incorporation of an external stimulus into a dream
occurs in order of decreasing frequency as 1) assonance 2) direct 3)
associative and least likely 4) symbolic (which obviously relies most
on the semantic dimension, as in my dream in which being emotionally
overwhelmed was tied to "Konstanz", a lake in Switzerland [see chapter
12]). Dreaming is not identical to an awake state of distraction with
respect to incorporation of external and internal stimuli, but does
show a hierarchy of levels of processing similar to the work of Voss
just cited. Voss similarly found that the use of a distractor at input
produced a higher number of clang errors upon recognition testing of
the words inputted.
98
The Crick and Mitchison theory may be depicted thus:
mem. 1—. mem. 2
mem.l
mem.2
mem.l
mem.2
already present
parasitic link
PCO stilation
and thought
activates the link
unlearning
Learning and thinking do not take place in this scheme, so why should
direct incorporation be occurring? Why is a distorted stimulus
incorporated, an element looking much like the "parasitic" or mixed-up
elements present in many dreams, and which is a distortion that was
not produced by PGO stimulation, but by the normal processes of
dreaming? The dream in this instance is more a creator of bizarre
images than a destroyer of them. The incorporated element was not
randomly stimulated out of the memory system, and yet it became part
of the dream. Berger (1963) reported also that, although meaningful
and neutral names were incorporated to an equal extent, the former
often appeared in a manner dependent on their personal meaning to the
dreamer. Incorporation was also "at a point in the dream appropriate
to the time of presentation of the stimulus", and herLe definitely not
the result of PGO stimulation. The activity of incorporation shows
that dreams are more than replayed memory traces; I suggest that a
dream narrative is a continual elaboration of the last image, and
hence more akin to thinking than to reminiscing. Evidence for this
emphasis on thinking as opposed to the reproduction of memories is
obtained if we compare the REM dream images to those present at sleep
99
onset: Cicogna, Cavallero & Bosinelli (1986), who found that sleep
onset imagery and daydreams have greater use of real memories than do
REM dreams.
Active Thought During Sleep
Going further than the work on immediate incorporation we
can cite Oswald (1960), who found more K-complexes in the EEG record
upon playing a meaningful name to sleeping subjects than after playing
the name backwards. That this reaction occurs shows that the sleeping
brain is performing some thinking activity, which is known from the
common experience that some sounds can wake people whereas others of
the same intensity don't. Examples of thought used during sleep are
also given by Tart (1963, 1966): some subjects can use the presence of
dreaming as a stimulus to awakening and others can use some internal
measure of how much time has elapsed in order to awaken.
Active thinking in the form of bizarre symbols can even be
seen before the subject is fully detached from the environment, as at
sleep onset. At sleep onset one can sometimes achieve the concrete
visualisation of an abstract thought, such as seeing descending white
stairs which symbolise going down with sleepiness (quoted in Tart ed.
"Altered States of Consciousness"). Such an image is a metaphor based
upon two valid traces, those of stairs and of dropping asleep. If
anything I consider that the association formed be ween these two
amplifies them both. It can still be retorted on behalf of the
unlearning theory, though, that it is precisely because bizarre images
crop up at these times that they must be systematically produced and
eradicated during REM sleep, when they are made more profuse.
100
Our efficient recall of supposedly 'unlearnt' dream images
is not an objection to the unlearning theory, even though our recall
of REM dreams appears quite effective if compared with our inability
to remember experiences of NREM sleep somnambulism. I borrow an idea
of Foulkes (1978) here that the long-term memory system may be
working abnormally (he suggests it is working in reverse, pushing
memories back to the sensory areas, as did Freud in 'The
Interpretation of Dreams') but that the rest of the brain can use
short-term memory to recall the images that have been coming up.
A problem with the claim that it is the memory areas that
are changed during dreaming is that PGO stimulation from the brainstem ends up in the occipital cortex, that the saw-tooth waves then
travel from the occipital cortex over the top of the head to the
frontal lobes, and that the theta wave then present in the hippocampus
is also present during motor activity in waking life. It appears that
it is most areas except the temporal lobes that change their activity
during dreaming (Yasoshima, Hayashi, lijima, Sugita, Teshima, Shimizu
& Hishikawa, 1984). This, of course, ties in more with theories that
ascribe to dreams a creative, recombinative function, which would
require the use of the frontal lobes and with memory working much as
in waking life.
The Second Theory - Clarke et al's "Positive Learning"
It may be more parsimonious to fit these findings into the
neural network theory of Clark, Rafelski & Winston (1985) to accept
that the bizarre imagery present in dreams has the same source as the
loose thinking images present in drowsy subjects, and not to propose a
101
new function for the PGO stimulation except as a facilitator of this.
Clark, Rafeiski & Winston (1985) relate their neural theory to sleep
by suggesting that during NREM sleep non-essential experiences are
eradicated and that during REM sleep important experiences are
associated and abstracted. This is certainly in line with the
difficulty in remembering mentation in the former case and the
bizarreness of mentation in the latter case. It is the former that
they associate with brainwashing (see last chapter), the latter with
'positive learning' with 'heightened noise'. Noise is increased by
having the units fire probabilistically, rather than just summing up
their inputs and firing accordingly. Crick and Mitchison (1986) state
that the bizarreness of REM sleep argues in favour of its being the
time for brainwashing; the argument against this is that it is the
'abstracted' learning, the finding of new connections, that is causing
the bizarreness, that we have here a loose thinking (which accords
with the experimental psychology data) and not the reverse of
learning.
This stochastic
firing may be, in fact, the neural analogue
to incubation and fantasy at the psychological level.
This would
also tie in with the claim of Schatzman (l983a, 1986) that
some dreams actively solve waking problems.
high beta
p r oh a
of fi
102
Beta is a measure of the stochastic element in the firing.
Clark et al (ibid.) are suggesting that during REM sleep beta
decreases. Note that this results in the formation of new
associations, associations that were not so likely with the higher
beta, and not the recall of associations already present, as claimed
by Crick and Mitchison. Crick and Mitchison (1986) suggest that as
well as the unlearning mechanism there may be a need for a bias
towards bizarre associations during REMS - a decrease of beta would
produce this but then the associations produced would not necessarily
have occurred with the high beta of waking life, in which case it is
better to stay awake with a high beta and never have the need for
unlearning. The usefulness of probabilistic firing is that it can
cause a net to jump from one cyclic mode to another, so that the
system doesn't get stuck in unwanted minima. It is also useful in
pattern recognition for finding the nearest fit. For the cerebellum,
Albus (1971) has already proposed such a concept of generalised
excitation resulting in increased sensitivity to inputs because inputs
do not have to exceed a certain threshold to effect output.
I also relate it to de Bono's concept of lateral thinking,
which involves looking at a problem in a new way, sometimes by having
a new starting point in assessing a problem. "A random input from
outside can serve to disrupt the old pattern and allow it to reform in
a new way." (1969, p.243.) He likens this new way of thinking to
poetry, where words are used in provocative and extraordinary ways.
This extraordinary way of using a word has no validity in itself, it
may be completely nonsensical, but it results in a new arrangement of
information. He gives the example of Leonardo da Vinci's diaries being
lost for centuries due to being mis-filed in a library and similarly
103
we need to shake up our stored information in order to avoid dogmatic
thinking. Hudson (1967, 1968) gives evidence for the pervasive
influence of stereotyping upon perception of the attributes of members
of other groups; the groups he used were artists and scientists. This
dogmatic thinking has its advantages sometimes though, it leads to
quickness of response and familiarity with the surroundings. Rational
thinking is applied to the results of the lateral thinking, for
otherwise we would remain disorganised. De Bono states that "Vertical
thinking is used to dig the same hole deeper, lateral thinking is used
to dig a hole in a different place" (1970, p.12.) and that "The
distinction between the two sorts of thinking is sharp. For instance
in lateral thinking one uses information not for its own sake but for
its effect. In lateral thinking one may have to be wrong at some stage
[a bizarre intrusion?] in order to achieve a correct solution; in
vertical thinking (logic or mathematics) this would be impossible. In
lateral thinking one may deliberately seek out irrelevant
information." (ibid. p.11.) He also notes the similarity between
insight and humour, for both can raise a smile and release tension, in
that both "involve the restructuring of patterns" (ibid. p.10).
Roberts (1985) makes the similar point that dreams are
concerned with updating our personal myths, with 'reconstructing' our
cognitive worlds. Similarly, Krippner (1986) hypothesises that dreams
are the synthesising of one's existing myths w th one's life
experiences, with the caveat that they can sometimes be conservative
rather than instigators of change. The neural mechanism of such
creative thinking must, however, involve more than the simple
generation of new associations. Bolton (1972, p.188) asserts that
104
sensitivity to the relevance of each new association is necessary when
they are produced - he notes that if thinking consists only of
associations then we would not be able to discriminate between those
that are valid and those that are not, for the act of discrimination
is itself a thought.
Dreaming and Constraint Limitation
The increase in stochastic firing, which has been proposed
as an integral part of learning and of forming more connections
between independent memories, can be related to the often noted
perceptual ability of dreams, their ability to pick up on small
impressions which are swamped in everyday life by more obvious and
harder knowledge. For example, Fromm (1957) reports a man dreaming of
an acquaintance being unkind "with a cruel mouth and a hard face", and
finding out much later in waking life that this was an accurate
picture. (Incidentally, would the unlearning theory assume that this
picture was to be erased? The proposed unlearning mechanism would then
be working against the useful faculty of intuition.) Fromm (1957, p.5)
also states that "another odd thing about our dreams is that we think
of events and persons we have not thought of for years, and whom, in
the waking state, we would never have remembered." However, it must be
noted that if one allowed oneself to think loosely while awake for two
hours of the day no doubt many old memories would appear, which the
exigencies of waking life would usually prevent.)
Gardner-Medwin (1976) notes that during waking life
"deductive processes may assist the associative processes in recall",
by generating correct details and inferences or by eliminating
105
spurious details (which I have suggested is a functional alternative
to the unlearning mechanism). He goes on to note that dreams have many
of the characteristics of recall in which the constraints of logic
have been removed, such as when we overlook bizarreness. (One link
here with Al is that 'constraint satisfaction' networks are now being
used to implement rules about what can and what cannot occur
simultaneously in the environment, and that a relaxation of these
rules results in bizarre outputs. Foulkes, 1982, p.178, makes the
similar point that 'in dreaming there often seems to be a dissociation
of inference markers from the propositions to which they are attached.
That is, there is an illogic in the conclusions the dreamer draws from
phenomenal facts'. He says that this occurs despite the opposite
tendency of 'blending of concepts'.) He says that the reasoning
processes and 'the stringent demands of waking life' may actually act
as a censor for some memories, and that dreaming is needed in order to
keep them strong - this is thus the opposite of the unlearning theory!
(This also resembles the psychodynamic theory that dreams may express
thoughts which are morally censored in the day.) Note that he is
suggesting that REM sleep reruns actual memories, whereas Clark et al
(ibid.) suggest REM sleep forms new associations, while Crick and
Mitchison propose that spurious memories are accessed and unlearned.
On the former theory dream images are loose thinking rather than
faulty memories (though obviously any sort of thinking will need to
involve memories) and it is possible that the neural basis of this is
a decrease in beta leading to a larger spread of activation than in
waking life ('spread of activation' is also evidenced in some work on
waking memory and perception). That this spread is not dependent upon
PCO stimulation is evidenced by the presence of spread of activation
106
at sleep onset mentation.
The Function of PCO Bursts
In order to explain the results given in chapters 5 and 6 it
may be claimed that it is the random PGO stimulation that "joggles"
the new memories into position, such "joggling" resulting in the novel
links found in many dreams as well as the anecdotal reports of
creative links occurring during dreams. (Such "joggling" and random
stimulation has similarities with de Bono's (1969, 1970) postulated
"P0" function.) The unlearning theory holds, on the other hand, that
the PGO stimulation illuminates the parasitic link in order to aid its
erasure; it appears that it is the unlearning theory that is
multiplying hypotheses, postulating the spontaneous formation of
parasitic links and also a process of unlearning, while others are
content to assume that dreaming is merely a more vivid version of
daydreaming. An extension of Clark et al's theory is thus that it is
the random PGO stimulation that forms the links, rather than helps in
breaking them, if it has any function at all. I assume that any
maladaptive links formed by such "joggling" will be used much less
than adaptive links and hence be more susceptible to forgetting and
lack of recall anyway, especially as they are likely to be illogical,
and the use of logic is often a part of recalling.
However, as we have seen in chapter 2, the PGO bursts, even
if they do have a purpose, correlate with complexity and vividness
rather than bizarreness of images, and the latter is not then
contingent on initial random stimulation. Therefore the connection
107
with Hopfield, Feinstein & Palmer (1983) is no longer present, for the
neural net models presume such random input. The Hopfield work is not
relevant to dreaming if bizarre images are caused by other factors
such as the drift from the previous image or by unconscious processes,
although it can be replied that REM sleep dreams (i.e. PGO bursts
present) are generally more bizarre than NREMS dreams, and that the
former's relative lack of correspondence with waking life accords with
the claim that more parasitic memories are evinced in this state.
Narrative in Dreams
The question remains of why are the dream elements enmeshed
in a whole narrative? A reply acceptable to the unlearning theory and
to Clark et al's theory is possible along the following lines. Under
behaviour therapy it is necessary for the patient to act out, or
imagine, the maladaptive behaviour; for example, getting nearer and
nearer to a feared high drop. The problem itself may be expressed in
only a few words, such as "fear of heights", but for curing it the
whole person must live through an episode of that fear, that is,
construct a narrative about it. Sometimes a whole story is needed in
order to define what the maladaptive behaviour actually is. Maybe
sleep onset intervals are not long enough for a run-through of
narrative to occur, hence dream states, lasting up to 45 minutes, are
needed. However, it is rare for such re-runs of act al memories to
occur during stage REM, they are more often acts that could happen, or
be imagined to happen, but which haven't. I shall explore in chapters
12-16 this theory that a dream has a narrative because consecutive
solutions and classifications are tried out to problems the subject
108
has; I already noted in chapter 5 that old memories may be being
explored and recreated in order to find an appropriate place for an
experience from the dream-day to be stored.
Much of a dream is composed of valid associations.
For
example, a chair in a room in a dream may be in exactly the same
position as in waking life. Are these associations unlearned or are
they, so to speak, dragged into the dream by the false links to which
they are said to be tied? Dreamers can wander around in a lucid dream
and be fooled that it is real, a testimony to the number of proper
associations present. (And valid associations are needed to define
what the parts of a condensation are, that is from where they are
taken.) Therefore, despite Crick and Mitchison's denial (1986,
p.232[lO2]), the theory does require and involve narrative, because
narrative defines the components of the imag,es; the question is
however, whether narrative has any other uses than this, and how long
these narratives are.
It is thus not the whole narrative that is unlearned but
only those parts occurring at the point of PGO stimulation. A quite
meaningful dream can thus be produced on the basis of the initial
false association, but note that other false associations occur by
virtue of the sleeping state, rather than caused by the PGO
stimulation.
Palombo's Theory And The Importance Of Imagery
It is the importance of encoding memories with respect to
each other, which is emphasised by Clark et al, that is the basis of
Palombo's (1976, 1978, 1980, 1984a&b) assertion that dreaming is the
109
simultaneous imaging of present and past memories, in order to learn
(rather than unlearn) an association between them. The strength of a
memory is thus increased by connecting other memories to it, no matter
how bizarrely. He suggests that dreams are the recall of propositional
information held in long-term memory and that they allow relevant
affect to be reconstituted and relived, which is needed for re-filing
new memories affectively rather than simply cognitively. I suggest
that this results in a deeper level of encoding, much as there are
many such levels possible in waking life. That animals show little
adverse effect after REMS deprivation treatment is explained by this
notion of dream sleep being a sort of 'icing on the cake', a helpful
but not absolutely necessary addition to the cognitive processes of
waking life.
With our knowledge that dreaming aids memory consolidation
it is now possible to make suggestions about the part that the visual
aspect of dreams plays in this. Palombo (1978, p.52) suggests that the
re-formation of an image of the body and surroundings acts to evoke
the emotions that were present at the time the original experience
occurred, which helps in the correct placing of new emotional
experiences in long-term storage. However, I would hold that if
cognitive information about an event is somehow held in a
propositional form in the brain, it is just as possible that affective
information about the event is likewise available. I therefore suggest
a much simpler explanation, based on the work of Morris and Stevens
(1974). They found that subjects who were told to form images that
linked together three objects physically had superior recall to those
told to image the objects separately, or who were just given
110
conventional free recall instructions. Richardson (1978) found the
same for the linked image condition, and that the separate images
condition had poorer memory for most of the list to be memorised than
did those with conventional recall instructions. Note also that Morris
(1978) cites 14 studies in which there is no benefit in the formation
of bizarre images, these being no better than a sensible interactive
image. Palombo does make an important point, however, in relating REM
sleep to the processing of emotions. It may be that REM sleep is only
concerned with the processing of memories that are emotionally
meaningful to the subject, as indicated by Greenberg et al (1983).
According to Morris and Hampson (1983, p.253) it is not the
imagery as such that aids performance, but rather "the processes which
underlie interactive imagery, namely the identification and
exploitation of potential relationships between the Items". DreamIng,
I suggest, is an excellent means for exploring such 'potential
relationships' . They proceed to give evidence that producing a
meaningful story about items is not quite as efficient as making an
image. An image may also have more information present than a related
set of propositions, as evidencei by t1nB tnhiiq'e \xs
to a-ns'-t t1n
question "how many window panes are there at the front of your house?"
Most people answer this by evoking the relevant image, for the
numerical answer is not stored itself.
However, these unconscious learning experiments may be said
to be a far cry from some of the proposed functions of dream sleep,
such as problem solving and re-filing memories on a semantic basis.
Palombo (1978, p.131) claims that such higher-order functions would be
better achieved in waking consciousness, and that his proposed model
111
of matching present events to past memories is of a much simpler
activity than the activities of waking life. I counter to this that
our dreams do show a great amount of sensible and deep thought, for
example, in one of my dreams I invented and performed an 'experiment'
to decide whether I was dreaming or under the influence of a
hallucinogenic drug, and obtained the correct answer! Emphasising this
'wakeful' aspect of dreaming led Hunt (1982) to claim that everything
in dreams has a counterpart in waking experience, if such pathologies
as fugue states and depersonalisation are included. The second,
empirical part of the thesis will aim to discover whether during
dreams we do more than superimpose present experiences and past
memories, that we in fact explore possibilities in either of these
domains, and whether in different dreams there are different amounts
of exploration and amounts of present or past knowledge used. (I note
Professor Alan Stone's (1986) suggestion that dreams may not all be of
the same kind, that they may be as varied as waking thought is.)
The latter two theories are suggesting that dream sleep is
not just able to aid consolidation of memories, but also causes their
recombination. That many animals can use this type of thought is shown
by work on cognitive maps and representation. It is mainly mammals and
birds which are found to be able to produce internal representations,
and hence possibly images, and these are precisely the groups of
animals which evidence REM sleep.
112
Experimental Evidence - The Change of REM Time with Age
The three theories described here must accord with the
evidence in chapter 1 that REM time is greater in the fetus than in
the newborn, and decreases thereafter.
It may be predicted that under the unlearning theory the
brain will be credited with a larger number of false connections and
more parasitic information as it grows older, which would mean a
greater need for REM sleep later in life. To avoid this problem they
postulate that REM sleep also removes catastrophic problems inherent
in the net before the problems of mixed traces and learning arise .. . during development, the semi-random process of making synaptic
connections is likely to produce parasitic modes. It is these which
must be "unlearned" in order to obtain a well-behaved system' (1983,
p.113). REM sleep thus has a dual function because of the random links
and resonating structures present in the fetus and neonate (for
myelinization of neurones occurs until age 2 years and interneural
connections are still being made). This use of the unlearning
algorithm would thus prepare neurones for future learning, and it may
be that dream images are merely an epiphenomenon of this anticipation.
Note that this suggestion conflates two separate areas of associative
net research; those using resonating circuits, and the Hopfield work
on static representations (see chapter 7).
There are two other options to explain the negative
correlation of REM time with age:
1) REM sleep aids the consolidation of new memory traces or, as a
variant of the unlearning theory, aids the eradication of new
113
parasitic traces. The change in REM sleep time would then have to be a
function of brain maturity, evidence for which was given in chapter 1.
2) Something different from waking processes is occurring, new
memories are already in long-term memory but are now transferred to a
better fit with old memories. However, more REM sleep is needed for
fitting in conceptually new information, because the brain is not yet
'dogmatic' enough to file information automatically, as it will do
when older. Maybe this process cannot occur while the senses are
supplying new images and information. This would accord with Clark et
al's theory, and others involving cognitive restructuring. Clark et al
(1985) proposed that brainwashing may occur in the fetus or newborn in
order to provide a competitive disadvantage to active links while
connections are still forming. However, according to Purpura, Shafer &
Scarff (1965) there is a precocious development of synaptic
inhibition, maybe as a brake against overexcitation before the
appropriate sensorimotor systems are mature. The threshold of seizure
induction is actually greater at earlier ages. They also state that
spontaneous activity is virtually absent. This opposes the idea that
the brainwashing or unlearning function used in computers (to make a
net ready for learning) might also have a use in immature biological
systems, as an inhibitor of over-active neurones.
Notwithstanding this finding, there is still the problem of
why the algorithm should persist in later life, if it is the basis of
REM sleep. Maybe dreams are so bizarre because the physiological basis
for them is such an atavism of earlier life, which is possibly even
maladaptive in later life. This is contrary to the unlearning theory,
which holds that REM sleep is adaptive in early and later life.
114
CHAPTER 9
THE SCIENTIFIC STUDY OF THE MEANING OF DREAMS
Psychology and Physiology
Having reviewed and evaluated much evidence from physiology
which concerns the nature of dreams it is now necessary to introduce
another approach to the subject.
That physiology is relevant to the form of cognition is
illustrated by the following work:
1) Luria (1973, p.312) writes of parieto-occipital lesions
which disrupt the ability to survey and perceive simultaneously the
logical scheme of a whole sentence, leaving only the ability to
understand simple sentences. That arithmetical and constructional
activity are also impaired shows that arithmetic and sentence decoding
use a similar mechanism of quasi-spatial analysis. Physiology has here
contributed to psychological theory in the discovery of an association
and commonality between the two activities.
2)
Shallice and Warrington (1970) showed by the
neuropsychological study of brain-damaged patients that there are
specific impairments of long and short-term memory processes. This
double dissociation gave information on the function of the memory
system.
Conversely, physiological theories have to accord with
psychological facts. For example, the three-colour receptor theory in
visual perception must account for our ability to perceive yellow.
Similarly, a psychologist may use drugs to analyse personality, and a
115
pharmacologist may use a subject's behaviour for information about a
drug's effects. Physiology may thus provide us with the necessary but
not sufficient conditions for dream content. (Likewise, teleology can
be dependent upon causal mechanisms, although teleological
explanations are only valuable when discovered independently of the
knowledge of causal laws.)
In other words, the content of dreams is a psychological
fact and must be explained as such. For example, Berger's (1963)
incorporation experiments provided a law of the relative likelihood of
the different ways in which an external stimulus may be incorporated;
it provided a law of how assonance is more common than direct
incorporation, which is more common than symbolic or free-associative
incorporation. Foulkes (1985, pp.160-161) writes that studies on odd
properties of dreaming may provide evidence for psychological theories
and models of memory. For example, 'dissociations (expected features
are missing in dream imagery), regressions (one character appears as
at an earlier age, while the rest are portrayed as they now are
known), faulty conclusions (... because it's Saturday I must hurry to
work), and reversals (a friend becomes an enemy, or vice versa...)'
may provide information about 'dimensions of featural analysis
the ways in which we know the "same" people over time..., the kinds of
inferences we associate with particular units of knowledge. Reversals
may prove explicable not from the old Freudian adage that "You really
hated that 'loved' person," but from the fact that part of what we
know about people, objects, and events is what they are not. We know
whales are not fish not so much by inference ... as by direct
representation.' (ibid.) He states that for dreams 'the ways in which
it joins together or does not join together knowledge representations
116
also may help us to map out the organization of memory.' (ibid.
p.161.)
It is therefore a valid to assert that laws can be
discovered, or should be looked for, on the psychological realm
independently of the search for physiological laws and bridging laws
between physiological and psychological events. D.O. Hebb emphasised
the necessity of devising independent psychological laws in stating
that:
specifying the physical embodiment does
not generally lead to an increase in
explanatory power
... neurologising is
merely a form of intellectual displacement
activity' (D.O. Hebb, 2949.)
A corresponding assertion is made by Gardner:
'Why gamble that an elementarist approach
will eventually yield large units, when one
has the option of beginning instead with
these
large
units,
which
seem
closer to the data and the experiences of
everyday life ?'
(1985, p.97.)
Gardner proceeds to note that neurologists ignorant of
linguistics can only have naive intuitions about langLrnge and language
disorders, and that 'one cannot have an adequate theory about anything
the brain does unless one also has an adequate theory about that
activity itself.., one can know every brain connection involved in
concept formation, but that won't help one bit in understanding what a
117
concept is.' (1985, pp.286-287.) New advances in science do not only
occur when a reductionist theory arises, new disciplines do come up.
Fodor (1981) uses the computer analogy of psychology being
concerned with software, and physiology with hardware. This
functionalist approach disconnects psychology from the need for
specific physiological backup. The merits of functionalism versus the
connectionist approach of Crick and Mitchison, which implies an
intimate relation between mental and physical function are an issue of
current controversy. It remains true, however, that the same
psychological functions can be carried out by physically distinct
physiological systems, as shown by work on the recovery of singing
ability in some birds (Gardner, 1985, p.281). Mehler, Morton and
Jasczyk (1984) write that the adequacy of psychological accounts does
not depend upon a knowledge of a specific biological model, and that
'a particular psychological process might be underpinned at different
times by a variety of different neuropsychological processes and yet
from the perspective of the neuropsychological level of functioning
there may be ... no reason for treating these processes as members of
the same natural kind. Indeed, exactly something like this happens in
the case of speech production, where, owing to differences in the
shapes of their oral cavities, different speakers may produce the same
speech sounds using very different vocal gestures.' (1984, p.87.)
Furthermore, the same physiological eve it may have
different psychological concomitants. For example, Schachter and
Singer (1962) showed that the same degree of physiological arousal, as
induced by adrenaline injection, was interpreted differently by
subjects according to their cognition of the emotional state of a
118
fellow pseudo-subject. Valins (1966) similarly showed the effect that
cognition has on the interpretation of emotion by giving false
feedback of their heartbeat to subjects looking at pornographic
pictures. Subjects claimed to find pictures more attractive if the
picture had been presented at the time of increased heart rate
'feedback' , even though their physiological state was constant
throughout. Mehler et al (ibid.) note that 'our fields seem to be
dominated by the belief that our understanding of a given mental
capacity increases in proportion to the extent that it can be
localised in the cortex.' (ibid. p.88.) If such a reduction is
possible, it will only occur following the discovery of laws and
descriptions which are particular to each field alone, with the laws
of one field being no more basic than those of the other.
It is possible to compare theories of dreaming which
emphasise psychological/physiological isoniorphism or comparison (e.g.
McCarley and Hoffman, 1981; Gackenbach, Snyder, Rokes & Sachau, 1986)
with some theories that stress the autonomy of the realm of
psychology. The autonomy of the realm of psychology in dreams is
suggested by any evidence that dreams provide recombinative thought,
rather than a repetition of waking thoughts, or even a lack of
relevance to waking thoughts. I will now review such evidence and will
then describe the methodology of content analysis of meaningful texts.
Meaning and Teleology in Dreams
Schwartz, Weinstein and Arkin (1978) review work on the
meaning of dreams, pointing out that independent judges in two studies
were able to correctly sort dreams in accordance with the subject who
119
produced them, and to correctly order them according to their
occurrence both within and across nights. Similarly, Dement and
Wolpert (1958) found that 7 out of a study of 38 whole dream sequences
were united by a common overall theme. Kramer, Roth and Cisco (1976)
provide similar comfirmation of the personal meaning present in
dreams, in that judges could sort dream series according to
circumstance of occurrence, although they could not do this for
individual dreams. Kramer, Hiasny, Jacobs and Roth (1976) found that
independent judges could sort dreams according to dreamer and
according to night of occurrence, although they could not sort them
into the sequential order they were dreamt in. This provides evidence
for Foulkes' (1980, pp.249-250) contention that 'psychophysiological
correlational research now appears to offer such a low rate of return
as not to be a wise place for dream psychology to continue to commit
its limited resources. ... Dreaming is a mental process and it must be
studied as we now study other mental processes. Whatever brain events
accompany dreaming, what the dream is is a mental act.'
Further evidence that dreams have to be studied as mental
acts and not just as the correlates of REM sleep physiology comes from
Fiss, Klein and Shollar (1974). Subjects had 6 baseline nights of
uninterrupted sleep, 4 nights in which every REMP was interrupted, 4
nights in which interruption was close to the end of the REMP, and one
recovery night. Subjects were not REM deprived and were awakened by
having their names called. Dreams recorded under the interruption
procedure were more vivid and emotional than those collected under the
completion procedure. They were also equal in length to the completion
night reports, even though they originated from shorter REMPs.
120
Furthermore, Fiss, Eliman and Klein (1969) showed that there is a need
to complete dreams such that the carry-over of REM-related mental
activity into waking life was strongest in a REMP interruption
condition than in a REMP completion condition,
or a complete REM
deprivation condition. Being prevented from completing a dream was
therefore more disruptive than not being allowed to dream at all.
Rechtschaffen (1964) similarly found that frequent interruptions of
REM periods resulted in a degree of thematic continuity rarely
observed in the dreams of a single night.
The dream is therefore meaningful itself, which means that
to discover the nature of dreams it is necessary to take account of:
as well as
'I'
and
Fiss (1969) states that 'the biological need for a
particular sleep stage needs to be distinguished from the
psychological need for a certain quality of sleep experience ... to
avoid the fallacy of mistaking synchronous for isomorphic events.'
The role of psychological causation is such that, unlike
physiological causation in general, we may need to take into account
the subjects' idiosyncrasies. An idiopathic approach is needed for
these cases which do not meet the S-R claim that there are standard Ss
and Rs which do not need interpretation by the subject. This can be
illustrated by the following two studies, which at first sight appear
to contradict each other. Fiss (1980) found that for 20 alcoholic
inpatients who had just completed a detoxification programme, 80% of
high cravers dreamed about drinking, whereas 30% of the low cravers
121
dreamed about drinking. However, Bokert (1968), who experimented on
subjects made thirsty before going to sleep, found that subjects who
dreamed about drinking and/or eating drank less water and rated
themselves as less thirsty the following morning than did subjects
whose dreams did not contain drinking or eating themes. Fiss (1986)
resolves the contradiction by noting that in the 1980 study 'the
dreams of the low alcohol cravers all contained themes of drive
gratification (e.g. having a good time while drinking), whereas the
dreams of the high alcohol cravers all contained defensive or
conflictual themes (e.g. loss of a love object as a consequence of
being caught drinking). ... the ... study underscores the importance
of looking at the quality and not just at the quantity of
incorporation.' (1986, p.173.) This is a justification for attempting
to devise hermeneutic theories of dreams.
The following chapters of the thesis will test competing
hermeneutic and physiological theories against the phenomenological
data of a series of nights' dreams. The former types of theories need
to be justified, given that I have already described the usefulness of
the positivist study of REM sleep. A hermeneutic account appears to be
necessary because of the above evidence of the meaningfulness and
goalÜdirectedness of the dream images. To illustrate the coherence of
taking both these paths, if only to test one against the other, take
the example of a cat purring. A physiological accoun of the purring
would not allow the claim that 'the cat is happy' because this would
entail unjustified statements about the inaccessible private world of
the cat. However, to simply describe the tone and volume of the purr
is not objective, because it misses out what may be the subjective
122
experience of the cat. Sometimes the attempt to reduce a phenomenon to
a more fundamental level of explanation only succeeds in destroying
the meaning essential to an understanding of it; for example, just
describing the muscle movements which occur when someone shakes hands
with someone else. In this case there are many levels of explanations
which can be taken as the more basic: a historical account of the
forerunner of handshaking; an anthropological account of its nearest
equivalents in other cultures; a physiological account of the muscles
used; or a motivational account of the interpersonal consequences.
Similarly, although it is necessary to use physiological methods to
describe the involvement of REM sleep with memory processing, or with
a changed susceptibility to epilepsy, or to discover if some people
do not have the neurological the expression of dreams, the meaning or
lack of meaning of any one person's dreams is another matter, as
removed as is linguistics from the neurophysiology of language. Such a
search for meaning is still an empirical matter, and mostly differs
from physiology only with regards its subject matter. There is a
correspondence between humanistic theories of voluntaristic activities
and dream theories which emphasise the element of choice in dreams,
that one scene is intended to be there rather than another, although
it is an empirical matter to discover the limits of this freedom,
which would be constrained not only by the type of mentation
characteristic of dreams but also by the mnemonic activation present.
I am not prejudging the issue by claiming that a
phenomenological approach to dreams is necessary. It is possible that
dreams are an epiphenomenon of a functional REM state,
and that
dreams will then not be the sleeping equivalent of the symbols of
waking life, which functionalists claim possess an independence from
123
the brain's hardware (Fodor, 1981). The point is that theories which
do not detach the dream into separate unrelated parts and which regard
meaning as central to the whole set of scenes produced must be
conjectured and then tested against reductionist theories.
As an example of these two types of theories, take two
possible explanations for the result of Fiss, Kremer and Litchman
(1977). They found that incorporating a presleep stimulus into dream
content facilitated the subsequent recall of the stimulus in the
waking state. Opposed to the explanation that the dream image had an
effect on the subjects' memories is the rejoinder that it is subjects
who were more likely to remember the stimulus who dreamt of it, and
that the images do not therefore have any information-processing
function. Haskell (1986a, p.156) writes that dream research should be
an interdisciplinary subject with its object of research being dreams
and dreaming. 'With this approach professional identity is attached
more to the object of research than to a given discipline or method
.' (ibid.) This would allow both molar and molecular research. He
thus disputes Foulkes' statement that 'we more than have our hands
full trying to deal with processing so we needn't take on the onerous
chore of characterizing precisely what it is that's being processed'
(cited in Haskell, ibid. p.146). In other words, the mechanics of
processing should not be given priority over content and meaning.
Haskell writes that these are different but equally ecessary levels
of analysis. Lawfulness of action doesn't preclude meaning. In fact,
to suggest otherwise, he states (1986, p.153), implies that meaning is
non-lawful. In order to obtain data about the possible meaning of
dreams a method of analysis of the dreams' contents is thus needed.
124
Content Analysis in Psychology
Experimental work has not yet shown us conclusively the
purpose of REM sleep, let alone the dreams that accompany REM sleep.
There are two extremes for the possible relation of REM sleep to
dreams, that the latter are epiphenomena, or that the imagery is an
integral part of the function of REM sleep, having a feedback and
functional role. In the former case the dream can be meaningful or
not. To provide more evidence on this question of meaning there is an
alternative to comparing many subjects' dream images with their
physiology, or to comparing the images with waking concerns, as
described above. Instead we can take the images and compare them with
each other. Thus instead of having many subjects and a blanket
description of types of images (e.g. active, passive, coloured) we can
take many individual and detailed images and compare them to each
other for only one subject. Obviously this procedure can then be
repeated for more individual subjects; the point is that in
experimental sleep lab psychology many subjects are compared over a
few dimensions, whereas in interpretative work many dimensions are
compared for the one subject. (The validity of an interpretation works
at the level of one person, whereas reliability needs many subjects,
and/or many judges in order to be tested.) Either method may be found
to be unproductive. Dream images may show no relation, no isomorphism,
to REM sleep physiology, or individual scenes may show no relation to
other scenes. This is all, however, an empirical matter, and both
methods must first be applied, once theories are devised which predict
inter-image relations.
125
Content analysis is such a study of the psychological
meaning and significance of documents and records. It can be nonreactive, although in the study of all dreams of a night the subject
is obviously affected by the awakening regimen. It has been used to
assess such variables as sex roles in children's readers, social mood
changes, and the motive to avoid success (Lewin, 1979, pp.252-271).
That the assessments can be complex is shown by Kohlberg's scoring
manual of moral development, which runs to over 100 pages (ibid.).
Aliport (1961, p.387) used a matching technique to relate
data from individuals such as life histories, photographs, specimens
of handwriting, artistic productions, gait, etc. He also got judges to
match invented words to abstract line drawings (ibid. p.491-492). This
'has the advantage of permitting complex productions to be ordered
with other complex productions ... The method permits the quantitative
study of qualitative patterns, though ... it does not tell us why a
judge perceives congruence between them.' (ibid. p.387.) By this
method he aims for the morphogenic prediction of the future of any
particular individual. He does differentiate this work on style and
manner, however, from projective techniques, which are aimed at
unconscious content (ibid. p.493). Baldwin performed a content
analysis of the traits and events mentioned in 'Letters from Jenny'
(Allport, 1965). Not only were separate counts of subject matter made,
for example, money, art, her son Ross, but also correlations between
them. 'For example, when she spoke of Ross, how frequently was he
mentioned in a context of money, of art, of women, of favor, of
disfavor? When she spoke of money how frequently was this topic
associated with Ross, with health, with jobs, with death?' (ibid.
p.197.) He used a variant of the chi-square test to determine the
126
significance of each association and obtained principal clusters of
ideas and feelings. Later, automated content analyses could be used,
due to advances in computer techniques. However, in order to produce a
computer readable set of categories the original rich discourse had
its connotations reduced. Factor analysis was then used to determine 8
'most prominent traits' which were found to overlap with a list of
traits derived from common-sense interpretation. Ailport concluded
that 'quantification of the structure of a single personality is
possible by means of statistical aids applied to content analysis.'
(ibid. p.199.). He gives the following account of why such a
methodology is necessary: 'That science likes universals and not
particulars is
. .. a fact. Yet personality itself is a universal
phenomenon though it is found only in individual forms. Since it is a
universal phenomenon science must study it; but it cannot study it
correctly unless it looks into the individuality of patterning! Such
is the dilemma.'
(Aliport, 1961, p.9.) He also notes that how our
qualities interact with each other is far more complex than the
comparison of single qualities across individuals. The two
methodologies are thus distinguished in that nomothetic work can be
reliable and yet irrelevanr, whereas the idiopathic work which has the
merits of significance and understanding can be undermined by the lack
of laws and predictability.
He explores the fact that humans do not just exhibit coping
behaviour, but also expressive behaviour, which is the manner of
performing adaptive acts. The latter is spontaneous and difficult to
alter. For example, what is written is part of coping, whereas the
style of handwriting is expressive. He does note, however, that 'every
127
activity, even if loaded heavily with expressivity, has its origin
somewhere. That is to say, some stimulus lies behing the scenes. Even
a "spontaneous" act follows some sequence of instigation.' (ibid.
p.466.) He complains that up to then psychology had concentrated on
testing coping abilities. Similarly, Rycroft emphasises not what a
symbol in a dream means, but why that particular symbol was chosen,
which, like style, is also somewhat idiosyncratic. It is therefore
evident that some studies of the meaning of dreams will have to use
single-case studies; the pedigree of these in psychology will now be
reviewed.
Single Case Studies in Psychology
Most early work in physiology was performed on individual
organisms, with the assumption of complete population generality.
Pavlov also used single organisms, generalising his case by repetition
with other species. In psychology Broca did much work with one
particular aphasic patient, and Fechner was concerned with the
variability of stimulus detection in individual subjects. Wundt used
highly trained individuals to give introspective data on the
assumption that the results were generalisable to the population in
general. Ebbinghaus administered some 2,000 lists of nonsense
syllables to himself. Also related to learning and experience was the
Kelloggs' (1933) project of raising one young chim anzee in their
home. In the area of motivation Cannon and Washburn (1912) showed by
the use of a swallowed balloon that Washburn's stomach contractions
coincided with his introspective reports of hunger pangs. (The last
two studies came to my attention in reading Hersen and Barlow, 1976.)
128
Watson and Rayner's study of Albert, and the sporadic
accounts of multiple personality, are case studies important in
abnormal psychology. In cognitive psychology an abnormally efficient
eidetic imagery ability, the like of which was never seen in any other
subject, was reported by Stromeyer and Psotka (1970). A pattern of 10,
000 dots was retained by their subject in eidetic imagery for 24
hours. In social psychology a detailed and seminal study of cognitive
dissonance was provided by the participant-observers of a group whose
members expected the end of the world to be imminent (Festinger,
Riecken & Schachter, 1956). Edelson (1975) uses an analysis of just
one short poem for his study of language and interpretation in
psychoanalysis.
Dukes (1965, p.76) says of work such as this:
'A few studies, each in impact like the single pebble
which starts an avalanche, have been the impetus for major
developments in research and theory. Others, more like missing
pieces from nearly finished jigsaw puzzles, have provided timely
data on various controversies.'
However, he goes on to warn:
'This historical recounting of "successful" cases is,
of course, not an exhortation for restricted subject samplings,
nor does it imply that their greatness is independent of
subsequent related work.'
(ibid.)
He does not, therefore, go so far as Hudson's (1966, p.12)
assertion that statistical studies have little to offer in the realm
of meaning and interpretation. Dukes gives three occasions for which a
study of N=l is appropriate. 'If uniquness is involved, a sample of
one exhausts the population. At the other extreme, an N of 1 is also
129
appropriate if complete population generality exists. ... In other
studies an N of 1 is adequate because of the dissonant character of
the findings. In contrast to its limited usefulness in establishing
generalizations from "positive" evidence, an N of 1 when the evidence
is " negative, " is as useful as an N of 1,000 in rejecting an asserted
or assumed universal relationship. ... Teska's (1947) case of a
congenital hydrocephalic, 6.5 years old, with an IQ of 113, is
sufficient evidence to discount the notion that prolonged congenital
hydrocephaly results in some degree of feeblemindedness.'
(ibid.
p.77.)
He also notes that there are practical reasons for
sometimes choosing a single case study. 'Situational complexity as
well as subject sparsity may limit the opportunity to observe.' (ibid.
p.78.)
He proceeds to advise that 'instead of being oriented either
toward the person (uniqueness) or toward a global theory
(universality), researchers may sometimes simply focus on a problem.
Problem-centered research on only one subject may, by clarifying
questions, defining variables, and indicating approaches, make
substantial contributions to the study of behavior. Besides answering
a specific question, it may (Ebbinghaus' work, 1885, being a classic
example) provide important groundwork for the theorists.' (ibid.)
Lazarus and Davison (1971) similarly state that case studies can
generate hypotheses, which later may be more rigor usly tested. A
similar point is made by Valentine, who also tempers it with a
warning.
'Sympathetic imagination is relevant to the origin but
not the validity of explanatory hypotheses. Empathic
130
identification may serve a heuristic function but it does not
guarantee knowledge. Conjecture is not fact nor plausibility
probability.' (Valentine, 1982, pp.176-177.)
The study of individual differences started in the early
years of this century, partly to investigate Quetelet's idea of the
'average man'. This became the basis for the comparison of groups of
individuals. 'Biometricka'
was then founded, which recorded
statistical tests and quantitative research. Cattell rated people by
mental tests, originally devised by Binet for targetting tuition to
needy individuals, and this concern with individual differences led to
an emphasis on groups and averages. Fisher worked on the problem of
generalisability of findings, and the problem of inference from the
sample to the population.
Hersen and Barlow (1976, p.8) write that 'by the l950s,
when investigators began to consider the possibility of doing serious
research in applied settings, the group comparison approach was so
entrenched that anyone studying single organisms was considered
something of a revolutionary ...' This was despite Skinner's use of
single cases in his study of operant conditioning. ' ... instead of
studying a thousand rats for one hour each, or a hundred rats for ten
hours each, the investigator is likely to study one rat for a thousand
hours.'
(Skinner, 1966, p.21, cited in Hersen and Barlow, ibid.
p.29.)
Meanwhile, psychotherapy was using single cases in the
production of hypotheses, and the case study method was the sole
methodology in the first half of this century. For example, Breuer's
(1895) treatment of Anna 0., in which he admitted that he didn't know
131
which part of the treatment caused the cure, but in which the effects
on the many behaviours were evident. By the 1950s the lack of publicly
observable data in such work led to its falling into disrepute; the
effects of treatment could not be evaluated, especially due to the
lack of recording equipment and reluctance to take detailed notes,
'... from the viewpoint of single case experimental designs, this
rejection of the careful observation of behavior change in a case
report had the effect of throwing out the baby with the bathwater.'
(Hersen and Barlow, 1976, p.10.) Aliport put it that 'we stop with our
wobbly laws of generality and seldom confront them with the concrete
person.' (1962, p.407).
Given that terms have been accurately defined there are
still problems with the group comparison design. The finding of
homogeneous groups is difficult, as is the finding of matched
controls. With regard to this myth of patient uniformity 'Bergin
consulted a prominent statistician about a therapy research project
who dissuaded him from employing the usual inferential statistics
applied to the group as a whole and suggested instead that individual
curves or descriptive analyses of small groups of highly homogeneous
patients might be more fruitful.' (ibid. p.16.) Hersen and Barlow
(ibid. p.13) thus argue that 'is psychotherapy effective ?' is the
wrong question, and that 'what specific treatment is effective with a
specific type of client under what circumstances ?' is more useful.
Similarly, some dreams may be meaningful whereas some may not be.
Not only is there the problem of generalising from the
sample to the population, the clinician also does not know how results
from one heterogeneous group are generalisable to one patient. More
132
random samples are better if comparison is desired with the whole
population, but make the results less generalisable to the individual.
In addition, as important as inter-subject studies of outcome is the
study of the clinical course of a specific patient. In the psychology
of dreaming this has its counterpart in the changes of dreams across
the night, as opposed to how many subjects get wish-fulfilling or
problem-solving dreams.
These difficulties led to an upsurge of naturalistic study,
in which there is no systematic manipulation of independent variables.
For example, patients can be assigned to different treatment and
control groups not at random but according to the clinical decisions
of what is the most desirable treatment for each individual patient.
This illustrates the scientist-practitioner split, described by Bergin
and Strupp (1972, p.440) as being '... a prevailing philosophy among
behavioral scientists which subordinates problems to methodology.
the selection of a problem is dictated by the experimental design.'
They proceed to warn that 'the kinds of effects we need to demonstrate
at this point in time should be significant enough so that they are
readily observable by inspection or descriptive statistics. If this
cannot be done, no fixation upon statistical and mathematical niceties
will generate fruitful insights, which obviously only come from the
researcher's understanding of the subject matter and the descriptive
data under scrutiny.' (ibid.)
Apart from naturalistic experimentation the other reaction
was in the direction of process research. In this the outcome of
therapy is ignored and emphasis is placed upon defining variables
concerned with the individual patient-therapist interaction. A
133
counterpart to this in dream research would be to concentrate upon
scene changes and the relationship of one scene or dream to the next
one, rather than to evaluate the effects of dream deprivation or the
overall occurrence of particular images.
Shapiro (1966) advocated such a scientific study of
individual phenomena. Unlike Ailport he did manage to construct a
methodology to put this idiographic approach into practice. He
manipulated independent variables for a single subject by means of an
A-B-A design, showing that independent variables can be manipulated in
the single case, although it can be difficult to return the
independent variable to its original state in the second A section.
(Because we can never be sure what would have happened without the
independent variable change in an A-B type paradigm, Campbell and
Stanley (1963) refer to the A-B strategy as a 'quasi-experimental
design'.) In the same way the waking concerns of dream lab subjects
can be used as the independent variables, which then change week by
week. The only problem here is that the design will then usually
become A-B or A-B-C, in that the original state of specific concerns
isn't returned to. Note that Holt (1962) considers that the
idiographic method as used by Allport is simply the application of
nomothetic methods to individual cases, and that idiography is more a
choice of subject matter than of method. Caramazza and McCloskey
(1988) argue against positions such as the latter by stating that, in
neuropsychology, a patient-by-patient analysis, informed by theory, is
always needed before even considering the grouping of patients'
performances (ibid. p.524).
An example of the use of a case study in dream research is
given by Salley (1988). This was a report of the use of hypnotherapy
134
on Frank, a patient with multiple personality, two of whose 13
personalities claimed a dream production function. As is usual in such
cases the host personality only experienced fugues and did not know of
the existence of the other personalities. Under hypnosis one of the
personalities was found, who identified himself as Self, and who knew
of the existence of the other personalities and of Frank. Self acted
as a protector of Frank and claimed to use dreams as his only line of
communication to Frank; he would use dreams to explain seizures that
Frank experienced. Self stated under hypnosis that the seizures
reulted from a conflict with Frank during which Self would attempt to
force Frank to regain consciousness. 'That night Frank dreamt that he
was standing on a pedestal and two voices were shouting at him; one
voice shouting "Yes!" and the other "No!" The vibrations from the
shouting were so intense that the pedestal began to shake and split
open, whereupon he fell to the ground shaking. Free association to the
elements of the dream led Frank to relate the shaking to his seizures
and the screaming to internal conflict and his resistance to regaining
consciousness after a blackout. In the two years since he had this
dream, he has experienced no recurrence of the hysterical seizures.'
(ibid. p.113.) Therapy then focused on getting to know the other
personalities one at a time, with dreams depicting the relations
between the other personalities and to which Frank would free
associate in order to uncover repressed memories. alley concludes
that some dream theories, such as Crick and Mitchison, and Hobson and
McCarley, would have difficulty accounting for the data from this one
case, and that 'dream theories relying on information-processing
models and linkages between past and present data' (ibid. p.115) fare
135
better.
This example of the successful use of an idiographic case
study to provide evidence about the nature of dreams shows that
studies of the content of dreams need to be performed. To this end
psychological theories of the content of dreams must be produced to
guide our observations. In the following three chapters we will
examine three of these theories.
136
CHAPTER 10
FREUD AND THE EXPERIENCE OF DREMIING
In chapter 3 it was noted that the early philosophical
examination of dreams resulted in two accounts of the relationship of
dreaming to waking experience. One was that dreaming is discontinuous
with waking life, the other that mentation can never halt and carries
over from waking life into our dreams, in the process aiding our
experience of continual personal identity. Freud used elements of both
these views; he held that dreams use material which they have obtained
from waking life, often from infantile memories, and yet work on this
material with unconscious thought processes which are entirely
different from those used in waking rationality.
According to Freud a dream is the hallucinatory fulfilment
of a wish, the wish usually being unconscious, often infantile. The
representation of the wish usually requires some disguise due to its
reprehensible nature. In the process of interpretation the dream
becomes the 'royal road to the unconscious', giving us evidence not
only of the content of the unconscious but also of the method of
thinking that that part of the mind uses, that is, the primary
processes. The unconscious is more evident during sleep than when one
is awake because the 'censor', which controls the exit of thoughts
from the unconscious, is more relaxed in the former state. Relaxation
is not total, however, and the unconscious thoughts
ill have to be
disguised by the process of the 'dream-work' in order to be allowed
into the pre-conscious part of the mind without shocking the sleeper
and waking him or her up.
Although Freud gave some physiological justification for
137
these ideas, the emphasis on 'wishes' and 'censorship' means that this
is a top-down theory of dreaming, one in which each image is
meaningful and in which there is an active process of translation of
the basic 'dream-thoughts' of the 'latent-content' into the overt
'manifest-content'.
According to Freud the experience of a dream is in three
senses a regressive one:
1) Topographical Regression. The dream is visual, which he
claims is a regression to an infantile and unrealistic hallucinatory
way of dealing with the world. Thoughts are usually represented as a
scene in dreams, with the odd exceptional case of dream-thoughts
remaining as simple thoughts or as pieces of knowledge. In waking life
what is experienced is processed by the brain to become thought and
memories, producing a sequence of excitation. Freud held that in
dreams the direction of excitation is reversed, partly because when
one is asleep there is no current of excitation flowing in the usual
direction. An initial problem with this formulation is the suggestion
that imagery is held to be either infantile or peripheral to
cognition, whereas in chapter 6 I have already reviewed evidence that
the use of imagery is central to much of creativity and cognition.
Freud does remark, though, that in waking life 'this backward movement
never extends beyond the mnemic images; it does not succeed in
producing a hallucinatory revival of the perceptual images' (1953,
p.543).
2) Temporal Regression. The dream is a substitute for an
infantile scene which is modified by being transformed onto a recent
experience. By being transferred onto what is usually a harmless
138
memory the infantile wish is better able to evade the censorship. This
is achieved partly because the infantile wishes attract the recent
thoughts and memories to them - they are unable to bring about their
own revival and hence return as a substitute, a dream. He gives
evidence that a colour seen in one of his dreams was 'a recent
impression, which attached itself to a number of earlier ones' (ibid.
p.5 Li7). In the 1920s he revised the theory to include the possibility
that conscious thought and intentions of the day before could cause a
dream, with the aid of reinforcement from repressed material. These
he called 'dreams from above', as opposed to the original 'dreams from
below'
3) Formal Regression. The means of representation and
expression of the dream-thoughts are regressed in that the methods
used are primitive, not following the rules of the rational secondary
processes. Use is made of symbolism (both idiosyncratic and
universal), condensation (by which elements of the latent content
combine to form a single item in the
manifest dream), and
displacement (which is the change in what appears important between
the two levels of the dream).
The dream experience, the manifest content, is thus the
result of the activity of the dream-work on the latent content, or
dream-thoughts. The dream can appear well-structured and coherent on
some occasions due to the activity of secondary revision, but any such
coherence is irrelevant to the actual meaning of the dream, which is
discovered by the process of free-association from the individual
elements of the manifest content, each taken separately. Freud's
139
interest in dreams was in translating the manifest content back similarly he held that the dream-work 'does not think, calculate or
judge in any way at all; it restricts itself to giving things a new
form' (ibid. p.507). This means that 'the dream content seems like a
transcript of the dream-thoughts into another mode of expression'
(ibid. p.277), just like a rebus.
Freud thus emphasises the irrelevance of the narrative of
the manifest content, claiming that the meaning of the dream does not
lie in the manifest sequence of images at all, and that any apparent
meaning in the dream as told when waking up is merely a surface gloss
over the real message.
140
CHAPTER 11
DREAXS AND METAPHORS
'Even if one accepts ... that dreams are an
involuntary kind of poetry, one has to add
that they are usually also an uncompleted
kind of poetry'
(Rycroft, 1981, p.165).
Problems with the Disguise Theory
In the last chapter we sa ho' Freud held that diea'ms
protect sleep by a homeostatic process which defuses stimuli (external
or internal) that could shock the subject back to wakefulness. In the
same way he held that neurotic symptoms allow the safer, symbolic
expression of unwanted unconscious processes. Central to this theory
of dreams is the disguising function of the censorship. However, a
paradox results if we understand dreams, parapraxes and neuroses as
analogous to each other. The paradox is that in the latter two states
parts of the unconscious are expressed in psychopathologies in
question; rather than being 'disguised' it is as if they are being
translated into a different language ('somatized' in the case of
conversion neuroses).
Charles Rycroft found other problems with the disguise
theory of dream contents. He noted that not only do soir people have a
great ability to interpret their own dreams, but that what is
'disguised' on one night may be shown openly on another - 'it is
difficult to imagine that one night a dreamer may wish to repudiate
his sexual wishes and another night may admit them freely' (1981,
141
p.65). In addition, what appears disguised is often perfectly well
known to consciousness anyway.
In answer to these problems he developed the notion of
dreams as messages expressing unacknowledged parts of the self. He
thought that these parts arise because 'changes within ourselves
instigated by maturation and ageing ensure that there are always
potential, emergent aspects of the Self which the personal I self has
yet to assimilate' (1981, p.68).
Dreams as Metaphors
He postulates that such assimilation depends upon metaphor.
'Since the novel and unfamiliar can only be described in terms of the
familiar, and the abstract can originally only be described in terms
of the concrete, vocabulary is largely built on metaphor, and we use
it whenever we speak or write, often without knowing that we are doing
so' (ibid. p.70). I suggest that a corollary of this is that dreams
often appear bizarre upon waking because of their metaphoric function,
but that while they are happening we take the metaphors literally, and
hence rarely recognise bizarreness while dreaming.
We now come to a problem with Rycroft's theory. He states,
of images in dreams, that 'since ... they often make no sense if
presumed to refer to the objects they themselves depict, they must
often be symbols referring to or standing for something other than
themselves' (ibid. p.73). This, of course, not only overlooks the
aforementioned evidence that dreams are predominantly mundane and not
devoid of 'sense', but also begs the question of why we can presume
that the images are symbolic. Although he nowhere provides any proof
142
for his analogising of dreams with metaphors (apart from the selfknowledge that such a procedure can bring), and of dream images with
symbols, experimental evidence is provided by the work of Silberer
(1951) and of Antrobus (1978) on the metaphorical transformation of
abstract waking thoughts at sleep onset and REM sleep respectively.
Rycroft notes that Freud formulated his theory of dreams
before art became representational and expressive, as with Picasso and
James Joyce, leaving him with the realist belief that images either
show reality, do not show reality, or disguise reality. Rycroft goes
on to show the difficult relationship between art and psychoanalysis,
which arises because of the fallacious assumption 'that the primary
processes and secondary processes are mutually antagonistic' (ibid.
p.158).
We have seen how in each night it is claimed that a subject
may concentrate on a limited set of concerns, the dreams are singleminded. I will now introduce Rycroft's idea of the presence of
metaphors in dreams, which are held to be useful for classifying and
thinking about a specific problem. Such a breaking up of the subject
into parts is compatible with Lacan's derision of the concept of a
unified ego, the welcome illusion of which, he says, starts at the
mirror phase. With this notion of metaphors comes the idea that the
signifiers in the dreams are not arbitrary, they depend on their
particular referents; Barthes (l972b) similarly claims that signifiers
in myths are not arbitrary. This opposes Lacan's claim that signifiers
are always arbitrary and changing, defined only in terms of other
signifiers (see the arguments in favour of this in Benvenuto &
Kennedy, 1986, and Bowie, 1969, and the arguments against in Archard,
143
1984)
Rycroft (1981, p.79) cites Marion Mimer's hypothesis that
there is an 'innate tendency to apprehend all objects that are not
one's self by likening them to organs and processes which are one's
self...' and that the outside world 'provides us with a stock of
images which we can liken to our bodies and its processes and to which
our bodies can be likened'. This last sentence shows how there is no
privileged point of reference, whether it be the body or the
environment, in this means of thinking. Piaget (1962) considers that
this type of thought is purely assimilation to the subject's internal
schemas, as opposed to accommodation to the outside environment.
Metaphors are thus enabling the dreamer to think about abstract
concepts in terms of the known concrete environment, and vice versa.
The crux of this position is that the work of dreams is not
a simple imagined resolution of problems, with the dream being seen to
be obviously functional and with no mystery pertaining to the reports.
Thinking in dreams is made more complicated by the introduction of
past memories, metaphors, symbols, and the symbolisation of the
environment (not to mention condensation and displacement). It is also
complicated by the taking of notions or objects from one realm into
that of another, which is the basis of Hudson's (1985) Bric-a-Brac
model of dreams. Furthermore, not only is an initial problem
symbolised but the symbol is then treated as real, as standing for
itself. We thus find that not only must the initial problem be solved,
but the metaphor which stands in its place must be solved also. For
example, in my dream (see chapter 12) of being in a submarine (which
represented the analyst) on Lake Konstanz I had to solve the problem
144
of escaping from the vessel, which I did by means of an escape hatch.
Rycroft writes that:
... when used as a figure of speech, metaphor is used with
consciousness of its nature as a substitute for direct statement, when
used while dreaming, metaphor is presented as though it meant itself,
and it requires the addition of waking, reflective consciousness to
ascertain to what it applies. Or, alternatively, imagery in dreams
fulfils a function the reverse of that fulfilled by metaphor in waking
speech. Metaphor in waking speech adds to or defines more precisely
and vividly a meaning already and consciously intended; imagery in
dreaming lacks as yet the meaning that will turn it into metaphor.'
(1981, p.71.)
Such taking metaphors literally has been shown to occur in
subjects with damaged right-hand sides of the brain (Gardner and
Winner, 1977). I contend that this may be related to the reported
increase in right-hand hemisphere activity during REM sleep
(Goldstein, Stoltzfus & Gardocki, 1972; failed replication - Antrobus,
Ehrichman & Weiner, 1978; additional support - Gordon, Froonian &
Lavie, 1982; Bertini, Violani, Zoccolotti, Antonelli & Di Stefano,
1982; sceptically reviewed by Antrobus, 1984). If the right hemisphere
is connected with the understanding and production of metaphors then
its greater activity at this time would lead to more far-fetched
metaphors. Upon waking one is in a state of fun tional righthemisphere 'damage' , or decrement, compared to the erstwhile REM
stage,
and the metaphors appear bizarre,
and possibly
incomprehensible
Rycroft's work has much similarity with that of Baylor and
145
Deslauriers (1986, 1986-7), who base their theory of dreams on
metaphors conveying abstractions through concrete images, and who
interpret dreams by giving their subjects extremely long periods of
time to discover the metaphors for waking-life experiences in each
dream image. Rycroft's theory is based on the idea that conceptual
thought is aided by concrete imagery and metaphor; the former, I
propose, may be present in the dreams of animals, whereas the latter
would be a human extension of such images, which could then even be
contributed to by language, contrary to Hunt's (1982) opposition to
theories of linguistic sources of human REM dreams. Aspects of culture
could then also be fitted into our mind during dreaming, except that
our culture provides more efficient mechanisms for such an ordering,
such as learning courses. The external culture could even affect the
way that our mind in dreams is formed, as will be elaborated on in
chapter 18. Of course, if the main use that adult humans have for
dreaming is to obtain a better fit in storage, an elaboration of
connections, then a lack of dreaming would not be deleterious, the
organism would merely fail to improve as much as it could have done but because of the importance of improving as much as possible it is
predicted that the brain would be programmed to make up for any REMS
deficit, as is observed. Theories that identify REMS as removing
something undesirable (toxin theories as well as the unlearning
theory) only explain the REM rebound effect, not the lack of
deleterious psychological effects upon REM sleep deprivation. (See
Dement's, 1960, finding of "anxiety, irritability, difficulty
concentrating" which has been much less replicable than findings of
learning and physical disability.) Vogel (1977) reports that
schizophrenics are not adversely affected, and that psychotic
146
depressives are improved by REM sleep deprivation. Note however that
Newman and Evans (1965) claimed that Kales, Hoedemaker, Jacobson &
Lichtenstein's (1964) finding of no deleterious effects due to sleep
deprivation was because nothing novel was introduced to the
environment during the experiment, and that the subjects were confined
to a single room for observation.
Barthes (1972b) claims that a dream is like a sign, which
includes both latent meaning (signified) and manifest content
(signifier). The signifier is not arbitrary, though, for myth, he
says, is a second-order semiological system, constructed from a
semiological chain which existed before it. Myth takes hold of
language to form its own system, and he thus calls it a metalanguage
(ibid. p.115). The original sign has many of its connotations missing
and ignored; as an example he gives a Latin phrase in a textbook which
becomes a sign of the correct use of language, a sign of privilege, of
hard work, rather than of the original participants in the scene that
the phrase describes. He states that the sign becomes 'impoverished'
in being used as a signifier. But this original meaning of the sign
interacts with its new signification. This interaction of concept and
symbol was noted by Hall (1966), who wrote that what is important
about a symbol is not that it just stands for something else, but
rather why that particular symbol was chosen, and what can be said
about that signifier, apart from what it stands for. For example, to
symbolise a father by a General has different connotations from
symbolising him as a priest. Similarly, Barthes states that: 'The
relation which unites the concept of the myth to its meaning is
essentially a relation of deformation.' (ibid. p.l22). It is such
147
deformation which may be a source of bizarreness in dreams. When we
are awake we note all the usual waking meanings of a sign and this
clashes with the way it was used as a signifier in the dream. Barthes
states that the term used in the myth is 'deprived' of its history,
'they are half-amputated, they are deprived of memory, not of
existence' (ibid. p.122), and we experience something similar in
dreams in their poverty of full memory and full connotations compared
to waking life. (Note, however, that some forgotten individual
memories may still reappear in dreams despite this.)
Rycroft states that the dreamer '...
is
likely to be pre-
occupied with what I have called earlier his biological destiny'
(1981, p.80). He goes on to say that this results in the use of body
symbolism in ways that would be disconcerting in waking life (ibid.
p.80). As I will elaborate on later, I consider this finding to be
more to do with what his patients knew about his Freudian theories,
and certainly KJ's dreams, in chapters 14-17, while completely
egocentric, are more concerned with professional standing than
biology, as we will see.
For Rycroft dreams concentrate on the biological and
egocentric facts of life; birth,se,
stat, css,
A dream produces metaphors which often stand halfway between the vast
area of culture, or the other vast area of na ure, and the
circumscribed world of our biology and egos; for example, a dream I
had compared worries about physical weakness to a broken motorcycle.
According to Rycroft dreams and poetry are produced because
'symbolisation is a basic human need, not a symptom produced by
148
conflict and repression, and ... human behaviour is a language, not a
set of mechanisms for discharging tensions' (ibid. p.13). This
synibolisation, however, does not result from accurate, definitive
translation of each referent, but also depends upon all the other
referents and symbols used, such that the message becomes synchronic,
rather than diachronic. Rycroft's theory is here similar to one method
of studying signs, that of Structuralism, which will now be reviewed.
These theories of dreams as a language will then be compared to the
non-linguistic theories of chapter 8, which hold that although parts
of dreams can be taken as meaningful signs, no message is intended.
149
CHAPTER 12
STRUCTURALISM AND DREAMS
I intend to sunimarise in this chapter two revolutions in the
theorising about human culture that have occurred this century.
(Details of the individuals involved can be obtained from Kuper,
l986b, chapters 3 and 7.)
Leach (1982, p.16) writes that early anthropology, 'the
study of primitive man', rested on the premise that 'all non-Europeans
are stupid, childish, barbarous and servile by their very nature'.
However, near the end of the nineteenth century sociological thinkers
were claiming that the difference between 'primitive' and 'civilised'
'lay in the type of society into which the individual happened to be
born' (ibid. p.22). This meant that the primitive thought was coherent
and valid, but that the basic assumed propositions that inferences
were drawn from were false.
From 1924 Malinowski started to teach at the LSE. He
believed that the social behaviours and institutions of the primitive
societies served the immediate practical needs of the individual, that
is, they were all functional. Another type of this 'functionalism',
that of Radcliffe-Brown, held that the institutions maintained the
society as a whole.
There was thus a movement from these institutions, such as
myths, being seen as irrational and childish, to being seen as solving
a problem for the society, and maintaining its equilibrium. Myths
would thus be used as an explanation and justification for the status
quo. Demythologising was applied to the texts, including the bible, in
150
an effort to separate what is functional and reasonable from what is
irrational. Evans-Pritchard noted that in their totemic system the
Nuer 'conceptualize the animal world on the model of the social
world' (Levi-Strauss, 1969, p.30), that is, as containing lineages,
and thus introduced the notion of metaphor.
The study of dreams underwent a similar progression, though
much later in the present century (although in the case of the Crick
and Mitchison theory dreams are still held to be irrational). From
being seen by Freud as a completely different way of thinking, as
primary process rather than secondary process, dreams became
thought of as having
a.
problem solving function (Schatzman, 1983a &
1986; Cartwright, 1986) or at least as insightful (Hall, 1966; Fromm,
1957; Shevrin, 1986 [on subliminal perception]) and metaphorical
(Rycroft, 1981).
Kuper (1979 has suggested that dreams may be seen in terms
of the next anthropological revolution, that of Levi-Strauss'
structuralism, and I intend now to provide some definitions of this
method, and some examples of its use. (Its application has been
widespread and has affected the study of history, literature,
linguistics and the social sciences.)
Structural Analysis in Anthropology
The initial premise of structural anthropology was that what
was most important for study in the interrelations of people in
primitive cultures was not what was communicated, such as women or
money, nor who made the communication, but rather the structure of the
151
system of communication as a whole, and the transformational
differences between each structure and the structure of other
cultures. For example, if caste systems are compared with totemic
systems a major difference is that the former are endogemous whereas
the latter are exogamous.
This search for structure and
transformation was first applied to kinship systems, but Levi-Strauss
later applied it mainly to myths, because of their greater lack of
constraint by the material surroundings. (The importance of this was
that he proposed that the structure of mind can be best ascertained by
studying thought when it is most independent of the outside world.)
Myths were held to be giving a message which necessitated
the myth being analysed as a whole, with regard to other myths, and
hence as a system of signs. The actual message was held to be
unconscious, below the manifest level of the myth. Similarly, P.Cohen
(1969) points out that to see myths as merely functional, or
justificatory of the status quo, or explanatory does not really
explain 'why the social function should be performed by myth and not
by some other device'. To explain this it is necessary to examine 'the
nature of mythical symbols' and 'the structure of mythical thought.'
(ibid.) The actual purpose of the myth, according to structuralism, is
the mediation of contradictions in a society's beliefs, which are
expressed as binary distinctions, X and not-X. Cohen (ibid.) gives an
example of the use of the method in analysing the stor of the Tower
of Babel. He suggests that there is the 'seeming paradox: if men are
descendants of common ancestors they are one people; on the other
hand, although all men are thought to be descended from common
ancestors, they are not one people. Secondly, there is the seeming
152
paradox that God provides men with consciousness but denies them
ultimate knowledge of his own being and whereabouts'. Although Cohen
does not state that A:B::C:D in this case, he does argue that 'ideas
and sentiments which are felt as irreconcilable are presented and that
the tension is resolved by an event - the destruction of the Tower and
the creation of permanent divisions between social groups'. Although
he has doubts about the falsifiability of the method, he appraises it
as directing attention to 'a central characteristic of myth, the
expression of conflicting principles and attitudes', and to 'the
subliminal intellect' (ibid.). However, we must remember Levi-Strauss'
contention that what looks like useless dichotomising and
differentiating of anything in the environment still serves a
function, that of defining parts of the environment.
Barthes (1972a) stated that 'It is probably in the serious
recourse to the lexicon of signification ... that we must finally see
the spoken sign of structuralism.' As well as this concentration upon
signs there is also the study of the position of signs one with
another, and also the study of the sign as a member of an opposition.
According to Sturrock structuralism 'studies relations between
mutually conditioned elements of a system and not between selfcontained essences. There is nothing essential or self-contained about
a given word; the word "rock" let us take. That occupies a certain
space, both phonetically and semantically. Phonetically i can only be
defined by establishing what the limits of that space are: where the
boundaries lie if it crosses which it changes from being the word
"rock" to being a different sign of the language - "ruck" for
instance, or "wreck", which abut on it acoustically. Semantically, we
153
can only delimit the meaning of the signifier "rock" by
differentiating it from other signs which abut on it semantically,
such as "stone", "boulder", "cliff". ... without difference there can
be no meaning.' (1979, p.10.) He expands on this by taking an analogy
used by Saussure: 'A piece in chess falls under the category of pawn
or bishop not by virtue of what it is independently of the game, which
is a carved bit of wood or ivory, but by virtue of the value invested
in it by the rules of chess and by the differentiation of pawns from
knights from queens and so on. We may not be conscious of this
differentiality every time we move a chess piece, but it is very clear
that such an "event" is wholly determined by the structure of the game
in which it occurs' (1986, p.21).
It is often remarked that Jakobson's study of phonology
provides proof for the structuralist notions of opposition and the
meaning for individual elements provided by the overall structure.
Jakobson divided phonemes into opposed pairs of 'distinctive
features', such as 'voiced'/'unvoiced' , and 'nasalized'/'not
nasalized', finding, for example, five such oppositions in French, and
three in Turkish. However, I find it doubtful that the discovery of
the use of oppositions in audition, which is a simple physiological
process, can tell us much about the use of oppositional and structural
thinking at the level of the mind. Evidence for the latter must come
from other sources.
Leach (1976, ch.13) gives examples of how differentiation is
used in human culture. For example, in our culture, ordinary people
engaging in lay activities wear clothes of any colour, but priests
engaged in lay activities wear black, and when engaged in religious
154
activities wear white: also, a bride wears white, whereas a widow
wears black. Similarly, 'the yellow robe and shaven head of the
Buddhist monk is designedly contrasted with the white robe and partly
shaven head of the Brahmin priest, and both are contrasted with the
ash covered body and long hair of the Indian spirit medium.
(ibid. p.58) He proceeds to give other examples of oppositions which
are used to distinguish parts of our lives, such as Noise/Silence
being related to Sacred/Profane. Levi-Strauss (1977, p.117) describes
the layout of some villages and the social organisation of their
inhabitants: 'A circular street runs around the storehouses, with the
huts of the married couples built at the outer edge. This Malinowski
called the "profane" part of the village. But not only are there
oppositions between central and peripheral and between sacred and
profane. There are other aspects too. In the storehouses of the inner
ring raw food is stored and cooking is not allowed. . . In the
Trobriands we see, therefore, a complex system of oppositions between
sacred and profane, raw and cooked, celibacy and marriage, male and
female, central and peripheral.' In the Bororo village the opposition
between the centre and the periphery of the village is also the
opposition between men (the owners of the central men's house) and
women (the owners of the encircling family huts). 'We are dealing here
with a concentric structure of which the natives are fully aware,
The central area containing the men's house and the dancing place
serves as a stage for the ceremonial life, while the periphery is
reserved for the domestic activities of the women...' (ibid. p.142)
Levi-Strauss analysed myths in terms of such oppositions,
and gave three rules for his method (1978a, p.65):
155
'1) A myth must never be interpreted on one level only. No
privileged explanation exists, for any myth consists in an
interrelation of several explanatory levels.' [So, while from the
analysis of dreams Rycroft writes that one uses the outside world's
oppositions to make sense of one's own life, Levi-Strauss would
eniphasise that one also uses oppositions concerned with oneself in
order to make sense of the outside world.]
'2)
A myth must never be interpreted individually, but in
its relationship to other myths which, taken together, constitute a
transformation group.' Levi-Strauss states that 'each myth taken
separately exists as the limited application of a pattern, which is
gradually revealed by the relations of reciprocal intelligibility
discerned between several myths.' (1986, p.13.) [Hall (1966) and Jung
(1968) made the same point for dreams, as did Freud himself when he
advocated the comparison of dreams from the same night, following work
on 'dreams in pairs' conducted by Franz Alexander (in Fliess, 1950,
pp.336-342).] Similarly, Leach (1969, p.22) writes that 'the novelty
of the analysis ... does not lie in the facts but in the procedure.
Instead of taking each myth as a thing in itself with a "meaning"
peculiar to itself it is assumed, from the start, that every myth is
one of a complex and that any pattern which occurs in one myth will
recur, in the same or other variations, in other parts of the complex.
The structure that is common to all variations becomes apparent when
different versions are "superimposed" one upon the other'. However, of
paramount importance are not the similarities between different texts,
but their differences, so we first tabulate the changes that occur
between the first long dream and the last one, because this can act as
156
a guide to the main oppositions involved.
'3) A group of myths must never be interpreted alone, but
by reference: a) to other groups of myths; and b) to the ethnography
of the societies in which they originate.' [The oppositions and their
permutations on one night may relate to those on other nights, and may
refer to the oppositions of one's waking life. According to Kuper
(1989) Levi-Strauss is himself now emphasising point a), as if myths
are a 'hail of mirrors', reflecting only each other.]
From Levi-Strauss' 'The Structural Analysis of Myth' (in
Structural Anthropology, 1977) we learn that, for the analysis,
invariant functions in a myth, or group of myths, are bundled
together, and then compared with other such bundles. A forerunner of
such work was Vladimir Propp, who classified parts of fairy tales by
their motifs, such as 'interdiction', 'flight', 'lack', 'villainy'. He
then found that all the tales followed the same pattern, although in
some a bundle may be left out or permuted. He also claimed that the
transformation of one type of tale to another type followed certain
laws. Levi-Strauss applied similar procedures to the myths of the
Americas, claiming that 'the proof of the analysis is in the
synthesis. If the synthesis is shown to be impossible, it is because
the analysis is incomplete' (1978a, p.134). However, he found that
each part of a narrative had many different attribute any of which
could be relevant, and it is therefore the overall context which
determines which part is bundled together with what. He gives the
example of eagle and owl appearing in one narrative, which would mean
that the relevant opposition is day and night. Each animal has many
157
attributes that can be chosen to be used in the code. Boon (1972,
pp.l57-l58) writes of frogs being popular in South American myths
because they are particularly 'good to think' due to being seasonal
(and hence temporally discontinuous), vocal, bridging land and water,
and extendable in size.
As well as this synchronic analysis it is also necessary to
follow how oppositions are permuted diachronically, both within and
between myths. An example of such a permutation is given by Kuper
(1986b, p.76) with regard to the naven ceremony in New Guinea. 'The
ceremony involved transvesticism and other dramatic reversals of
normal behaviour. For example, the mother's brother of the person
being honoured dressed in grotesque female attire, offered his
buttocks to his sister's son, and acted the female role in a fantastic
similitude of copulation with his wife.' Kuper notes that this led
Bateson to the notion of 'schismogenesis', the rule that 'oppositions
are continually and dialectically heightened once begun' (ibid.).
Levi-Strauss claims that this unconscious processing is constrained,
that '... individual human beings (at play, in their dreams, or in
moments of delirium), never create absolutely: all they can do is to
choose certain combinations from a repertory of ideas which it should
be possible to reconstitute.' (1964, p.60.)
Poole (1969, p.13) says that the actual analysis involves
reading '... the langue and the parole [i.e. the lang age system and
each speech act] in a given code, be it mythical, psychological,
economic, social, literary or political. It is to define the signified
and the signifying in a given context within these codes. It is to
carry out either a synchronic study or a diachronic study of these
158
codes, or both if we can or if we want to, but not to confuse our
modes of study. It is to learn to read off the opposition pairs in a
code. Finally it is to watch carefully for position and inversion in
these codes. We have thus to think and to analyse in terms of a total
field of communication in the given society, or myth, or work of art
we are studying. All this is possible because of the linguistic
model.'
Levi-Strauss (1978b, p.24) states that he finds internal
evidence for the structuralist account of myths from the recurrence of
the same events in different myths around the world, despite their
making up different messages ('...the same absurdity was found to
reappear over and over again, then this was something that was not
absolutely absurd.' ibid. p.11). He says that myths are thus a closed
system, unlike history. Dreams similarly have a very simple
repertoire, being very concerned, as Rycroft (1981) notes, with our
biological functions and survival, although the evidence of chapters
14-17 in this work is more in favour of dreams being egotistically,
rather than just biologically concerned. As with myths, dreams are
distinguishable by the idiosyncrasies of the subject, which are
introduced into what is a very stereotyped product. Levi-Strauss
(ibid. p.9) states that the structuralist approach is 'the quest for
the invariant, or for the invariant elements among superficial
differences'. He uses an alternative to reductionism (or at least at
the initial application of the method; eventually a relating to
neurology in the case of the structure of myths and kinship is
claimed) in that he looks for relationships between elements, and the
thesis of Kuper s work is that the same may be done for dreams,
despite the tradition this century of working out where each image
159
came from independently of the others. Opposing this is the
functionalist approach to dreams (Hunt, 1982; Schatzman, 1986, on
problem solving in sleep) which holds that the imagined environment
of the dream world is simply a model of the real world, in which we
can practise behaviour as if in the real world. An alternative view is
to see some objects and events of the dream world as symbolic, albeit
very concrete symbols, in which, as in myths, there is 'the respect
for and the use of the data of the senses' (ibid. p.13), as opposed to
conceptual and scientific thought. He proceeds to distinguish
scientific from primitive thought in that the latter only imagines it
has power over the universe; this is also true for dreams.
Examples of the Method of Analysis
As an example of the application of the method I will take
'The fate of Lot's wife' by Aycock (in Leach and Aycock, 1983). This
myth has one of the recurrent themes of mythology, that of
immobilisation, to explain which, he says, we must bring in the wider
symbolic structure of the myth in order to discover if the
iinmobilisation is mediating conflicting oppositions (as in The Story
of Asdiwal, Levi-Strauss, l978a, pp.146-l85). He notes the following
transformations in the story:
160
BEGINNING
END
Lot meets two male angels, who are He is seduced by two females, his
disguised, and are strangers
daughters, who are 'disguised' by
to him
his drunkenness
The impending destruction of Sodom The creation, due to the incest,
by God
of the Moabites and Ainmonites
Within the walls of Sodom, a city
A cave in the hills above Zoar
A society of homosexuals, who thus
Incestuous association, justified
cannot have children
by the need for children
Aycock claims that Lot's wife is a mediator for these
oppositions. She is turned into a pillar of salt between the two
geographical locations of Sodom and Zoar - and 'the salt into which
she is transformed has ... not only the connotation of sterility, but
also the connotation of purity, since she cannot, by definition,
participate in the abominations and impurity of either Sodom or the
caves in the hills' (ibid. p.117). He further claims that this method
can provide insight into other irnmobilisations in the bible, such as
'Noah in his ark upon Ararat, Isaac tied to the altar below Abraham's
knife and the ram caught in the bushes who substitutes for him, and
Jesus on the cross' (ibid. p.118). He also notes that the oppositions
are also standard themes: 'excess and restraint applied to food and
sexuality the city and the wilderness . ..' (ibid.). In case such a
method of exegesis appears contorted, Leach remarks that 'Prior to the
Reformation all Christians took this [method] for granted. Not only
was the whole of the bible true but it was all true in the same way.
161
It could be read as a synchronous story in which the different parts
were internally cross-referenced. It is only during the last 150 years
that a quite different attitude has come to dominate biblical
scholarship. Truth is now equated with "historical truth" and since it
has become apparent that large parts of the bible could not possibly
be "true as history" in a strict sense the task of the scholar has
been seen as that of sifting the true from the false' (ibid. pp.2-3).
With regard to Leach's comment above that the portrayal of
Jesus on a cross between heaven and earth is similar to other myths of
immobilisation, I note that the term 'mediator' is frequently used
with regard to what the theologians call his 'work' on the cross.
Opposition
Heaven
Earth
Mediated by:
Immobilisation between the two
Good god Evil mankind He is treated as if evil, and as if a
sacrifice, yet forgives his tormentors
Life
Death
He is killed yet resurrects - in the
meantime he is in a temporary liminal
place, some say on a journey to hell,
others that he was in paradise.
He is said to mediate between god and man, which is made
necessary by the danger of the meeting of what is sacred and what is
profane. That this danger requires some liminal area for the meeting
to take place is frequently observed in anthropology. The Roman
Catholic church has caused a further evolution of this liminality, in
that Mary and the saints are believed to intercede for the supplicant
162
to Jesus, who in turn mediates to god. Leach (1969, p.10) notes that
'this pattern is built into the structure of every mythical system,
the myth first discriminates between gods and men and then becomes
preoccupied with the relations and intermediaries which link men and
gods together'. He states that such mediation between two poles of an
opposition is performed by a third, 'abnormal' category, such as
virgin mothers or incarnate gods. It may be suggested that bizarre
dream elements will often be such mediators. They are bizarre because
often the opposition cannot be reconciled - for example, our
existential oppositions of death and life, freedom and security, or,
in the case of myths, that between the prohibition of incest and the
status of the original two humans. Leach (1969, p.22) claims that the
latter problem is met in Genesis by the recurrence of incestuous and
immoral relations, such as those of Lot's daughters and the men of
Sodom, which makes Abraham's incest with his half-sister Sarah less
onerous. Similarly, in 'The Legitimacy of Solomon' (in Leach, 1969) he
notes the unresolvable contradiction in Israelite society of endogamy
(to keep the community pure) versus exogamy (because of the many
foreigners). An attempt at a resolution is made by the tension between
stories of the downfall of kings after marrying foreign wives, versus
the presence of tribes that are neither Jew nor gentile, such as the
Samaritans (ibid. pp.54-55).
The Implications of Structuralism for Studying Dreams
According to Levi-Strauss' theory myths aim for the
resolution of conflicting ideas. This line of thought applied to
dreams predicts that a dream has a problematic beginning leading to a
163
final solution, but that the dream does not use the same thought
processes as the waking human does. It predicts that the successive
scenes are reached not by small positive or negative increments of
part of a previous scene, but by 'clear-cut relationships such as
contrariness, contradiction, inversion or symmetry' (Levi-Strauss,
1981)
Dreams, by analogy, would thus be concerned with the
resolution and exploration of tensions, not just with their
abreaction, hence their is predicted a dialectic in the narrative of
dreams: the tension is not merely depicted, as if it is essential,
necessary, and immutable, although some dreams will stop at this.
We therefore see that dreams are claimed to be doing more
than re-presenting waking life - like myths they aid classification,
and are hence constructive and re-constructive. On this issue of
classification:
'It might seem that the system of differences between animal
species is too powerful to match the much weaker differences between
human groups. Members of the same society look alike and live in
similar ways and conditions; social groups do not differ in the ways
natural species do. But the point precisely is that human groups are
trying through "totemic" institutions not to match two pre-existing
systems of differences, but rather to build one system with the help
of the other. They are trying not so much to e press social
differences as to create or strengthen them. In this respect the force
of the animal system is never excessive; whatever aspects of it can be
mapped on to the social system are welcome.' (Sperber, 1979, p.32.)
I now wish to introduce the work of Professors A.Kuper
164
(1979, 1983, 1986a) and A.Stone. Their joint paper (1982) analysed one
dream, the 'Irma dream' of Freud, and coupled with the other analyses
(of a set of dreams of a Plain's Indian and of a whole night's dreams
of a dream lab subject, 1(J) they provide a plausible and theoretically
grounded account of how one dream image changes into the next, and
therefore an account of the dream's narrative nature. They found two
progressions in the former dream; the sequence of diagnostic medical
rounds in a hospital (preliminary evaluation by Freud, note made of
Irma's 'pale and puffy' appearance, other colleagues perform more
extensive examinations, and finally the supply of a definitive
diagnosis and aetiology), and also the sequence of examination from
mouth to throat to upper body, lower body and finally intestines. They
also found 'a movement from superficial, false and less important
findings to deeper, truer, and more important findings' (Freud first
notices lesions in the throat, but the final diagnosis is of infection
and dysentery). They note also that the 'two sequences not only move
in a coherent direction, down and deeper, they present a typical
quasi-logical transformation'. An example of one such transformation
they give from that dream is 'the mouth that resists voluntarily
opening becomes an anus that involuntarily opens'.
In this theory the narrator is held to use the 'science of
the concrete', externalising conceptual thought by the use of whatever
comes to hand in the observed everyday universe. What comes to hand
are not tools or signs designed especially for solving the specific
problem, much as an engineer has, but rather "a collection of oddments
left over from human endeavours" (Levi-Strauss, 1966, p.19) which are
more akin to the tools used by a handy-man, a bodger, or, in LeviStrauss' culture, a 'bricoleur'. There is nothing else at his disposal
165
to use, and the "set of tools and materials ...
is
always finite"
(ibid. p.17). Examples of this finite set of tools are the physical
oppositions present in the environment, such as Inside/Outside or
Distance/Closeness, or Lower body/Upper body. I propose discovering
whether Levi-Strauss' claim about the properties of these tools found
in myths, that they are "permutable, ... only on the condition that
they always form a [closed] system in which an alteration which
affects one element automatically affects all the others" (ibid.
p.2O), is also applicable to any oppositions found in dreams. Note
that because many oppositions may change at once this form of thinking
is not a paragon of the scientific method, but still the system of
classification needed (in dreams as well as myths) is very precise,
often the equal of any modern taxonomy, and does obey rigorous logical
laws.
An essential element of the analyses is not just that
abstract thoughts are represented by concrete images, which Silberer
was indicating long before Rycroft's theories, but that both operate
in terms of oppositions. Boon (1985, p.169) puts it thus, '[LeviStrauss] has praised Durkheim's emphasis on contradictions that
sustain social and cultural divisions; here Durkheim foreshadowed
structuralism. Standard functionalist theories consider contradictions
in any system as potential obstacles to its proper functioning, which
must be corrected, repaired, purged, or cured. This i. a therapeutic
model in which contradiction is not so much integrated as released and
tensions felt by the actors thus eased. In contrast, structuralist
theories consider contradiction unavoidable; this much they share with
various schools of dialectic, including Hegelian and Marxist ones.
166
Systems, such as mythic variants, operate not despite contradiction
but by means of it...' Similarly, Lacan writes of how concepts or
attributes are defined as much by what they are not, as by what they
are; for example, a man considering his masculinity may proceed by the
contemplation of what it means to be feminine. In the same way, LeviStrauss writes that, 'like a myth, a mask denies as much as it
affirms. It is not made solely of what it says or thinks it is saying,
but of what it excludes.' (1982, p.144.)
By studying dreams themselves Haskell (1986b) came to the
similar conclusion that:
while specific imaginal [dream] content may be
important, the essential mechanism of resolution is probably
structural. For example, the cognitive operation of negation or
contradiction transformed in the dream to imaginal action forms is the
essential resolution mechanism.' He gives as an example 'turning
something upside down'. The resolution is in the structure and not the
content, and abstracts common mnemonic features of all content that is
related to the problem. So when one problem is solved, the same occurs
for many others. He suggested that 'the structure of logical relations
occurring on some levels of dream imaging takes the form of an actionlogic where objects, things and events are physically manipulated and
move about in imaginal (symbolic) space.' (ibid.) But he proceeded to
note that:
'Presumably the story presented in a dream is somehow
parallel to the real-life conflict situation, and when resolution
occurs in the dream story, it is thereby resolved in the psychological
reality. But this does not cognitively explain how the resolution is
accomplished.' (ibid.)
167
We must study the methods of this putative problemresolution in dreams in detail, in order to answer how, as opposed to
why, supposed resolutions occur in dreams. In the structuralist
account of dreams it is predicted that, as for myths, that resolutions
occur by the application of '... rules of transformation which enable
us to shift from one variant to another by means of operations similar
to those of algebra.' (levi-Strauss, 1977, p.235.)
Continuing the analogy, the background information that the
subjects give about themselves in the interviews thus correspond to
the knowledge that an anthropologist has of the background culture of
the mythmakers, or, put another way:
'Structural analysis can reveal unsuspected depths of
reference and inference meaning for any particular series of myths. In
order to squeeze this significant out, the anthropologist must apply
his prior knowledge of the culture to his analysis. He uses inference
the other way round, from the known culture to the interpretation of
the obscure myth. This is how he discerns the elements of structure.'
(Douglas, 1978, p.169.)
Part of the appeal of Structural Anthropology has be n that
it claims to provide insight about the structure of thought, and
further evidence for the validity of the project would be available if
the structural analysis of dreams were as fruitful as that of myths.
The aims of the project are summed up by Skinner, in 'The Return of
Grand Theory in the Human Sciences':
'The study of rationality has thus come to be
168
a major focus of research ...' (1985, p.16).
Theoretical Problems with the Analysis
The standard criticism of the structuralist account of
myths, which was taken up by van Veizen (1988) in his critique of
Kuper and Stone (1982), is that the sexual and aggressive side of
people is ignored. For example, Spiro (1979) reanalyses the Bororo
myth which had been studied by Levi-Strauss in 'The Raw and the
Cooked' . He complains that two incidents in the myth, the son's rape
of his mother and the father's subsequent assault of the son, are
reduced to the opposition Distance/Closeness, thus eliminating the
manifest content of aggression and sex. The myth continues with the
son killing his father, and the son taking revenge upon his mother and
stepmothers. Levi-Strauss' overall analysis, however, relates these
incidents to the origin of cooking fire and rain. Spiro remarks that
this analysis 'converts all acts of violence into metaphors for
nonviolent structural relationships' (ibid.). However, to claim that
the myths of primitive groups cannot be concerned with intellectual
puzzles and aetiological issues, as if they are stuck at an early part
of mankind's history, or as if they are more i touch with their
aggressive or sexual emotions than we are, is considered by many to be
a nineteenth-century view of these peoples. Similarly, to claim that
dreams are only concerned with sexual and aggressive wishes may be an
early twentieth-century atavism.
Levi-Strauss' concern with the intellect is exemplified in
his recent writing about Freud's 'The Interpretation of Dreams'. In
'The Jealous Potter' (1988) he contrasts his own relativist treatment
169
of mythic elements with Freud's frequent use of set translations for
symbols (ibid. p.188). But such a narrow concern is itself
intellectual, unlike how an interest in Freud's instinctual theory of
the unconscious as opposed to Levi-Strauss' cognitive view of the
unconscious would be. He has chosen as the point of interest in
Freud's favourite work a very minor and cerebral part of that work.
170
CHAPTER 13
A COMPARISON OF THE PHENOMENOLOGICAL PREDICTIONS
OF THE MAJOR THEORIES OF DREAMS
Structuralism, Psychoanalysis, and Dream Narrative
Both structuralist and Freudian analyses of dreams postulate
a submerged conflict which instigates the sequence of images. In the
former case the conflict is rather conceptual, for example myths of
the original single human may oppose common knowledge of generation
from male and female with belief in the autochthonous origin of man,
whereas in the latter case the conflict is more instinctual. The two
methods are similar in their activity of finding the conflicting
elements. All the elements of the text are viewed out of their
narrative order; for the structural analysis the syntagmatic chain
becomes a synchronous table:
A-B-C-D-E-F-G becomes
B
A
C D
E F
C
where elements in one column have some similarity to each other. In
his essay 'The Structural Study of Myth' (1977, pp.206-231) LeviStrauss studies the Oedipus myth, to illustrate a synchronic analysis.
He finds that the first column (the overrating of blood relations) is
to the second column (the underrating of blood relations) as the third
column (the denial of the autochthonous origin of man) is to the
fourth column (the persistence of the autochthonous origin of man).
171
The myth has the function of relating one unresolvable opposition to
another unresolvable opposition, by means of placing them in the same
syntagmatic chain. (Jakobson, 1965, stated that the failure to form
such chains may be the cause of certain aphasias.) This relating of
oppositions is synchronous, Levi-Strauss calls it 'paradigmatic', and
does not depend upon where in the chronology of the myth the elements
occur. Levi-Strauss' logical paradigmatic analysis is thus at the
opposite pole to the analysis of folk-tales carried out by Propp, who
found that many tales were similar in their means of progression, but
with different characters substituted, or other changes made to the
syntagmatic chain. Silverstone (1981) puts Greimas between the two
poles. Greimas reduced the number of functions that Propp perceived in
the tales, and paired them into oppositions. Some of these oppositions
are then ruptured at the start of the tale, and 'the task of the
narrative is to remedy the various dimensions of rupture which have
been established initially' (ibid. p.94). The narrative would then
mainly follow the familiar trail of the hero accepting a task,
confronting the villain and ending with a marriage.
In the psychoanalytic account of the dream-chain,
associations are made to each element which results in latent elements
being found, which can tie together two or more temporafly distinct
manifest elements. Freud stated that the parts of the manifest dream
are only tied together by these latent elements, for example: 'It is
natural that we should lose some of our interest in the manifest
dream. It is bound to be a matter of indifference to us whether it is
well put together, or is broken up into a series of disconnected
separate pictures... In general one must avoid seeking to explain one
172
part of the manifest dream by another, as though the dream had been
coherently conceived and was a logically arranged narrative.'
(Introductory Lectures, 1963, pp.181-182.) Both accounts therefore
postulate a disjointed collection of images, neither assume a smoothly
flowing narrative, they both assume a latent realm which binds
together these images. This latent realm is, by definition, timeless
in the Freudian account, and in the structural account can be a simple
conflation or comparison of two beliefs which is then translated into
a diachronous set of images. For both methods the dream is a whole
piece, the later parts are presaged by the earlier ones, while the
later images refer back to the earlier ones. We have a whole work with
interconnecting parts, rather than just a simple one-dimensional tale.
The basis of their difference is the significance to be
given to the sequence of the manifest dream. For the structuralist
account the manifest plot attempts to resolve some latent conflict,
which results in an extra determination for the images, in addition to
the synchronous determination above. However, resolutions do occur in
dreams looked at analytically, for example, a wish may be fulfilled
only after a struggle between contending and opposing wishes. This
presence of narrative in the Freudian account of dreams can be
overlooked because of Freud's emphasis on the irrational nature of the
unconscious. For example, he explained some of the gaps between dream
scenes as a result of the inability of the sleep g mind to use
conjunctions and logical relations: so, 'simultaneously in time' is
the concrete picture used to depict 'logical connection' (1953, vol.4,
p.314), whereas the relation 'just as' is depicted by similarity
between dream elements (ibid. p.320), and causality is represented by
either a transformation occurring before our eyes, or by havng a
173
preliminary dream with a longer sequel (ibid. p.316). He almost
ignores narrative, as shown by this elision of conjunctions, because
of his emphasis on the individual elements and symbolism of dreams.
(An effort from within the psychoanalytic movement to show the
significance of the manifest content 'as a form of adaptation, of
problem-solving' has been made by Spanjaard, 1969. In addition, from
the same tradition, Mahony, 1977, studies the interrelationships of
elements of the manifest content.)
However, we know that conjunctions and logical operations
can be present in the thoughts of a dream character, such that, for
example, in a nightmare a character may decide to climb somewhere in
order to escape a lion. Now, if logic ad xarrati
ca
within an action then they must be able to be present between actions,
and between scenes, and hence be part of the process of going from one
scene to another, especially in view of the way that scenes can refer
one to another. We have reached the question of the presence of cause
and effect in dreams and I now digress to the realm of philosophy.
Hunie opposed any use of the terms 'cause' or 'effect' on the
grounds that contiguity in space and time between two events, such as
the cue ball in snooker hitting a red ball and the latter then moving,
no matter how many times that contiguity has been noted, was not proof
that the same relation would obtain in the future. He claimed that we
only see separate events, and that we cannot see this supposed
relation of causality between the events. A major argunient against
this has been that we only know we have a cue ball because of the
events it can be involved in. If it is placed on the table it doesn't
sink into it and it doesn't move randomly by itself (like a jumping
174
bean does). To even talk about a lack of causality between objects
requires that we can place the objects by reference to what they do,
by being able to construct a narrative about them.
Similarly, for elements to be pictured in a dream we must
give them some narrative, some cause and effect, even if that is only
the dreamer looking from one part of the image to another, or just
knowing what the image is not. If narrative is present within an
image, in order to define an element of it, then it will surely affect
what the image does next, and we will then have a narrative for the
whole scene. Note though, that this cause and effect does not have to
always be identical to the pragmatic version seen in waking life, it
can be symbolic, or even irrational.
One dream I had went as follows: 'I was running down a
corridor away from my psychotherapist. I fumbled to find a key which I
used to lock a door between me and her. I kept the key.' Jung reports
one in which he descends by stairs below a house into deeper and
deeper realms. To define these realms, these levels, as well as to
define the key as something to keep a therapist away from me, reguires
narrative. The order in which the events occur does matter to the
meaning of the elements and the overall plot, despite Freud's
insistence to the contrary. Some structuralist analyses depend on
classifying elements in terms of what happened to the element
temporally, for example, the mother travelling downstream as opposed
to the daughter travelling upstream. To look for symbols and
associations, or to produce the above synchronic table, is merely the
first step of the analysis, but one that does involve having one eye
on the surrounding narrative of each element. In practice, both
175
structuralism and psychoanalysis then use these synchronic elements as
a key to decode the overall narrative. To concentrate on the key
alone ignores the way we give meaning to events, by constructing a
narrative about them.
Both the structuralist account, and that of Freud, find
resolutions in dreams. In the above dream of mine I resolved the
conflict of fear of therapy versus wanting it by locking the door and
yet keeping the key, so that I was the one who could choose whether to
go back. (Note that neither theory predicts that resolutions will
always be happy or wish-fulfilling at the manifest level. This is
evidenced by Cartwright's [1974 b] finding that periods of dream sleep
do not affect completion of crossword-puzzles or word-association
tests, but that they do lead to more 'negative' or pessimistic
solutions to a Thematic Apperception Test.) Wherever an opposition
must be superseded, whether for obtaining a wish-fulfilment or to show
its likeness to another opposition, narrative is necessary. Even if
complex logical operations cannot be expressed in dreams we can still
find time sequences, a dream may be illogical in parts but without
logic and narrative at other parts those parts would not be
comprehensible. For example, Freud (The Interpretation of Dreams,
1953, p.494) tells of a dream which included two simultaneous ideas,
that of a marriage, and that of an arrest: but in order to define
these two human activities a narrative is needed, so that the dreamer
knows what is being watched. In the same way, the strange lawless
behaviour of quantum particles is only revealed by the lawful and
predictable activity of the measuring apparatus, and art therapy is
only of value because of the surrounding structure which allows and
176
holds the period and medium of creativity and illogicality.
Freud considered that the 'dream-thoughts' produce a dream
by the processes of displacement and condensation, with an eye to the
representability of these thoughts in the eventual rebus. There is
then an optional process of secondary revision, which can turn this
initial product into what he called a 'well-constructed dream'. (This
revision adds to the dream, whereas the other functions of the
censorship are involved with omission.) He stated that (1953, On
Dreams, pp.666-6E7) 'Considerations of intelligibility are what lead
to this final revision of a dream. ... From the point of view of
analysis, however, a dream that resembles a disordered heap of
disconnected fragments is just as valuable as one that has been
beautifully polished and provided with a surface'. A dream therefore
appears to be logical due to this process of secondary revision; this
is summed up by his statement that 'I will not deny that critical
thought-activity which is not a mere repetition of material in the
dream-thoughts does have a share in the formation of dreams. . . .this
thought-activity is not produced by the dream-thoughts but by the
dream itself after it has already, in a certain sense, been
completed.' (ibid. p.313) However, the evidence is that this narrative
is not always produced after the 'real' dream, as if this dream-facade
consists of just conjunctions added to the disjointed images.
Sometimes, the narrative is an integral part of the images, and Freud
says as much in his short paper 'On Dreams' (p.667): '... in the
erection of a dream-facade use is not infrequently made of wishful
phantasies which are present in the dream thoughts in a preconstructed form, and are of the same character as the appropriately
named "day-dreams" familiar to us in waking life'. The form of the
177
day-dreams is, of course, a narrative, and so we have it on Freud's
authority that some dreams do contain an originally narrative aspect.
What, however, if this claim is wrong, and dreams do only
have a narrative element because of confabulation in the face of these
inchoate images? A waking model of this is the experiment of
Cazzaniga, LeDoux and Wilson (1977). Verbal commands were presented to
the left visual field of a splitbrain subject, P.S. This command was
thus only perceived by the right cerebral hemisphere. Following the
command he produced the response requested, and was then asked 'Why
did you do that?' In trial after trial his left hemisphere, which
controls speech and language, but had not seen the commands, responded
in ways consistent with the information available from his behaviour.
When the command was 'rub' he said 'itch', and when it was 'laugh' he
said that the experimenters were funny. These responses were not said
as guesses or jokes, but as facts to explain the previous behaviour:
the subject honestly believed the confabulation. Similarly, LeDoux,
Wilson and Gazzaniga (1979) presented two pictures simultaneously, one
to each visual field. Each hemisphere then had to choose from eight
pictures the one that went with the initial picture. In one case the
left field item was a snow scene, and the left hand pointed to a
shovel, while the right hand pointed to a chicken to match the chicken
claw presented to the right visual field. When asked why he chose
those items he said that he saw a chicken claw and so pointed to the
chicken, and that you have to have a shovel to clean te chicken shed.
This was not said in jest, the speaking left hemisphere had to account
for the behaviour of the left hand and so confabulated a solution, a
conjunction between the separate images. However, even if such
178
confabulation is the sole source of dream narrative the study of such
narrative remains valid and valuable, for we are still studying a
human ability, that of how scenes are made to lead one to another.
Whether narrative is present from the beginning, or whether it is
added when some raw material has already been produced, we are still
studying how humans construct narrative. Freud called the secondary
revision the first interpretation of a dream, we may thus be studying
not the pure product, but rather an interpretation of it; yet it is
still interesting to look for any permutations and progressions in
this overlaid narrative, for that also shows how our thoughts are
structured. As an analogy, I note that myths certainly take similar
pure products of the outside world and lay a narrative over them, and
the study of these myths similarly remains important.
Freud introduced the idea of secondary revision because he
denied the presence of intellectual ability in the initial stage of
dream creation. 'Everything that appears in dreams as the ostensible
activity of the function of judgement is to be regarded not as an
intellectual achievement of the dream-work but as belonging to the
material of the dream-thoughts and as having been lifted from them
into the manifest content of the dream as a ready-made structure.'
(1953, p.445.) He proceedeX to gt
inc
b'j tat'c
of dream activity which do appear to involve intellectual ability,
namely number calculations, speeches and judgements. He gives examples
where a number or speech also occurred in waking life and said that
'as a rule, the repetition is ill-applied and interpolated into an
inappropriate context, but occasionally, as in our last instances, it
is so neatly employed that to begin with it may give the impression of
independent intellectual ability in the dream' (ibid. p.459). He
179
treats numbers and speeches in dreams just as he treats objects, as
standing for something else, or as a day-residue. One such dream was
the following: "The dreamer was in a big courtyard in which some dead
bodies were being burnt. 'I'm off,' he said, 'I can't bear the sight
of it.' He then met two butcher's boys. 'Well,' he asked, 'did it
taste nice?' 'No,' one of them answered, 'not a bit nice' - as though
it had been human flesh." He notes that the last two speeches were
taken from waking life, from a visit to their neighbours who were
eating, and so were not intellectual judgements formed during the
dream (ibid. pp.420-421).
Now, speeches do sometimes occur out of context in dreams,
but here they fit together, and we must ask if this is simply
fortuitous, or the result of secondary elaboration, or the result of
thought processes at the moment the subject is woken. Freud had a rule
of disregarding apparent coherence in dreams - each element must be
traced back on its own account, by the use of free-association. After
the text of one of his dreams which involved many acts of logic
(concerning the criticism of the absurd idea that Goethe could have
made a literary attack on a young acquaintance of Freud's) he stated
'Every step in this set of logical conclusions, however alike in their
content and their form, could be explained in another way as having
been determined by the [latent] dream-thoughts'. (ibid. p.450). Now,
Freud has made us aware that a conclusion in the dream may also be
present in the dream-thoughts, and thus in waking life; but not
everything in a dream must stand for something else, for, in his own
words, 'a cigar is sometimes just a cigar'. An act of logic in a dream
could be just that, rather than an allusion to an event from waking
180
life. He is also leaving out his own discovery of overdetermination,
that an event in a dream can be the result of numerous separate trains
of thought. I am proposing that according to a Freud's overall account
a seemingly intellectual activity in a dream can be the result of a
waking event, and/or simultaneously of the subject thinking while in
REM sleep. His later revision of the theory to include 'dreams from
above' confirms this.
Freud was concerned with the source of individual dream
objects and actions taken on their own, which are strung together
(eventually) in a narrative. Similarly, the structuralist method
removes individual elements from the sequence and asks how they
arrived there, 'why is the eagle the totem for this tribe?' 'Why did
Connie show K.J. a record cover?'
(see chp.14.) Neither niethoti den?&ns
a smooth narrative, with no sudden changes of scene, and each can
subsume the possibility of the sequence being added to in order to be
made more intelligible. The presence of what Freud called added
'connecting thoughts' is compatible with both theories.
I consider that Freud's dogmatism with regard to the lack of
intellectual ability in dreams was due to iis emphasis on t'rie
difference between waking and sleeping life, and that he overstated
his case in this instance, while perfectly correct in observing that
'... the dream-work is not simply more careless, more irrational, more
forgetful and more incomplete than waking thought: it is completely
different from it qualitatively and for that reason not immediately
comparable with it'. (ibid. p.507). Dreams do have the strange
activities of condensation and displacement, but they also have
counterparts of waking thought. In fact, they are very similar to
waking life, which is one reason for why we are so rarely surprised in
181
them, so rarely are we cognizant of the fact that we are dreaming.
One example of Freud's contrary wish of not wanting to
completely deny the logical nature of dreaming is his criticism of
those who attributed the creation of the dream to the moment of
waking, especially after Silberer's experiments on hypnogogic imagery.
He obviously considered the dream needed time for its production.
Also, the first part of 'The Interpretation of Dreams' is critical of
those who claim that the dream is a meaningless jumble of images.
Silberer worked on the transformation of waking abstract
thoughts to concrete images while he was drowsy. For example, he wrote
'In the morning, at waking, while I was at a certain depth of sleep (a
twilight state) and reflecting over a previous dream and in a sort of
way continuing to dream it, I felt myself approaching nearer to waking
consciousness but wanted to remain in the twilight state.' He then saw
the following scene, 'I was stepping across a brook with one foot but
drew it back again at once with the intention of remaining on this
side.' (Quoted in Freud, ibid. p.504.)
Freud agreed that this
symbolisation in the dream state may occur, but stated that it is less
common than secondary revision, for this provided an example of
intellectual activity while dreaming which was not likely to be also a
day-residue, and hence did not fit his theory. He also stated that:
'What seems to occur more frequently are cases of overdetermination,
in which a part of a dream which has derived its material content from
the nexus of dream-thoughts is employed to represent in addition some
state of mental activity.' (ibid. p.505).
This statement allows me to claim that the psychoanalytic
view of dreams as irrational is not contradicted, but rather is
182
complemented by the structuralist view of their rational and cerebral
function. In addition, this statement shows that both study, at least
sometimes, texts in which aspects of the subject's world are
represented by objects in the subject's experience, which are put to
another use, in bricoleur fashion. It may be noted that Nietzsche, who
used many of the concepts attributed to Freud, was a master of such
concretisation of abstract thoughts, ('man is a bridge that must be
overcome', 'can you not hear the grave-diggers who are burying god?').
Freud concentrated on the source of the images in the subject's life,
and in the dream-day. The structuralist analysis of dreams is
confronted with much more mundane objects, and concentrates on their
relations one with another, and on the relations of relations, and on
the source of images in terms of the other images.
Whether Badcock's thesis of the derivation of Levi-Strauss'
work from Freud's is true or not (he states that the former is a
cybernetic version of psychoanalysis [ibid. p.112]), the Structuralist
method of dream analysis is seen to be a synthesis of the two. LeviStrauss and Freud are compatible also in that both aim to find the
hidden, unconscious determinism, which lies beneath the surface
events. Levi-Strauss likens them both to the science of geology in
this respect, and Freud himself drew a connection between the two
fields with his statement that:
'I have long been haunted by the idea
that our studies of the content of the
neuroses might be destined to solve the
riddle of the function of myths...'
(From the letter of S. Freud
to D. Oppenheim, October 1909.)
183
In conclusion, the relation between these two methods of
exegesis, and the recently proposed structuralist analysis of dreams
may be summarised thus:
FREUD
LEVI-STRAUSS
dreams
myths
transformation of
transformation of
instinctual latent
intellectual latent
content into independent
content into manifest
manifest elements, in
elements which themselves
order to satisfy instincts
transform, in order to
in the unconscious \
satisfy structural ucs.
SYNTHESIS
dreams
intellectual, language-based transformation
of egocentric, but often consciously known,
latent thoughts into manifest message which
progresses mytho-logically towards a solution.
Freud, Palonibo, and the Narrative of Dreams
Of relevance to the question of the narrative of dreams is
the work of the psychoanalyst S.Palombo. He holds (1978) that a
condensation is the meeting of two memories, one from the subject's
present waking life, and one from earlier in the subject's life.
Dreaming has the task of matching such memories in order to file
184
correctly the subject's latest experiences. The dream narrative is
thus solving the problem of which part of the 'memory tree' to fit the
day residue into. He states that anxiety dreams result when the new
experience cannot be easily reconciled during the dream with the
person's usual schemas. The dreamer wakes up due to this stimulus of
anxiety, and acknowledges the dream consciously.
'Awareness of the anxiety dream in his waking state gives
the dreamer an opportunity to bridge the gap with a new mediating
experience which enlarges the connection between the experiences of
past and present.' (ibid. p.17)
The memories form a tree-like structure, with the earlier
ones, which are of greater affect, at the trunk. The search starts as
close to the trunk of the tree as possible, and the path of least
resistance down a branch is then followed. He believes that dreams reaffect memories, which are stored without information about affect
(although no evidence for this is given).
This theory is fully in accordance with Freud's 'The
Interpretation of Dreams', from which Palombo derives not only
examples of what can be taken as the matching process but also the
physical metaphor of the Dalton lamp. This device was used in Freud's
time to compare the likenesses of members of one family by
superimposing lamp projections of the photographs of two members of
the family. Palombo considers that condensations in dreams are such
superimpositions of two memories, with the aim of discovering a
similarity between them. He claims that a greater similarity means
that the superimposition in question would be less bizarre. In that
case, the memory of the new event is filed in the memory system at the
185
place of the similar old memory. This theory accords only partly,
though, with the modern view of dreams as problem solving, although no
doubt in the search for solutions past actions will be reviewed.
Palombo writes that dreaming may be efficient at comparing memories,
but not necessarily at solving single problems. He proposes that such
skilled behaviour is best undertaken while one is awake, showing a
difference between his theory and the functionalist ones. The
following dream, though, taken from 'The Interpretation of Dreams'
(1953, pp.369-371), shows that Palombo's work is also in accordance
with the structuralist analysis of dreams. (This is important, because
I consider the Palombo work provides for dreams a link with the
evolutionary value of dreaming, as a part of the memory processes.)
'I was running down the staircase in pursuit of a little
girl who had done something to me, in order to punish her. At the foot
of the stairs someone (a grown-up woman) stopped the child for me. I
caught hold of her; but I don't know whether I hit her, for I suddenly
found myself in the middle of the staircase copulating with the child
(as it were in the air). It was not a real copulation; I was only
rubbing my genitals against her external genitals, and while I did so
I saw them extremely distinctly, as well as her head, which was turned
upwards and sideways. During the sexual act I saw hanging above me to
my left (also as it were in the air) two small paintings ... At the
bottom of the smaller of these, instead of the painter's signature, I
saw my own first name, as though it were intended as a birthday
present for me.. .'
Freud notes that the subject had, on the dream-day, seen
some paintings in a shop, and also heard of a servant girl who had
186
become pregnant on the stairs. From the structural point of view we
note the opposition in the dream between large and small, shown by the
two adults and the child, and also by the two paintings. However, the
dreamer also remarked that the stairs in the dream 'belonged to the
house where he had spent the greater part of his childhood and, in
particular, where he had first made conscious acquaintance with the
problems of sex'. We can thus conclude that, simultaneously with the
portrayal of present problems and their solutions, the dream depicts a
childhood association of the problems.
To obtain such a depiction requires a narrative, for the
individual memories are themselves temporal, and also because the
search itself may take time. Similarly, the depiction of resolutions
in a myth analysed structurally needs a time element, and hence also
needs narrative. Furthermore, definition needs a concept of what an
entity is not, as part of the description of the entity, as noted by
Structuralist writers, and particularly emphasised by the poststructuralist Derrida and by Lacan.
Structiiralism and Symbolism in Dreams
It is a prediction of the structuralist account, and that of
Rycroft, that for humans their dreams have a change in nature with
time, both phylogenetically and ontogenetically, in the same way that
an infant will change its cognitive functioning from sensorimotor to
concrete operations to formal operations. In dreams we can play with
whatever material we can use while awake, facial gestures when young,
simple concrete wishes when older, and puns (the similarity between
'Konstanz' and 'constant' is an example from later in this chapter)
187
and language when older still. For example, simple waking functional
actions performed by a young person, such as digging a hole, may have
ideational attachments if performed by someone older, attachments such
as burial, death, rebirth, and the earth. REM sleep may have evolved
to aid consolidation of memories of concrete actions and experiences
(the lack of proprioceptive and other feedback certainly argues
against it being a practice mechanism for the body's benefit), and so
could easily develop in the human individual to aid consolidation and
elaboration of conceptual attachments to these concrete aspects of
experience.
According to the structuralist account, the area of the
Symbolic is analysed in the same way as the Real, that is, by its
oppositional properties. But whereas all these properties are present
at once in waking life, and are subservient to the exigencies of that
life, the claim is that when dreaming we can concentrate on one
opposition at a time (which is simultaneously compared to other
oppositions, such as the differences between two tribes being likened
to that between, say, the bear and the eagle). Similarly, Barthes'
(1972b) essay on wrestling shows how the excessively acted movements
and expressions of wrestlers show the notions of good, bad, justice
and punishment in an unambiguous way. This makes wrestling so unlike
other sports, which are less of a ritual. As Sturrock (1979, p.62)
puts it:
'In daily-life the signs which the world presents to us are
wholly ambiguous, their meanings are many and uncertain; but for the
space of an hour or two, and within the confines of the wrestling
ring, the signs are utterly unequivocal.'
188
The same can be said for some dreams, although displacement
results in great ambiguity and an obfuscation of any message present,
and also it may not be apparent unless many scenes are compared what
the relevant oppositions are, according to the structuralist account.
The Predictions of the Crick and Mitchison Theory
At first sight the theory seems to imply that chronic dreamrememberers will be prone to instabilities, on average, more than nonrememberers because an association or event in the dream should have
been unlearned. Such a prediction accords with Crick's antipsychoanalytic bias, in that the remembering of dreams would no longer
be a virtue. In contrast to this, Singer and Schonbar (1961) found
that 'high and low daydreamers differ along a dimension which might be
termed self-awareness, or acceptance of inner experience', and there
is no study which shows a greater psychopathology among rememberers of
night dreams. Schatzman (1983b) stated that 'If their theory were
true, and their advice [of not attempting to remember dreams] valid,
much psychotherapeutic practice and many people's habits of recalling
dreams would have to change'. However, I consider this to be a false
objection to the theory. The theory certainly predicts that a nonrecalling dreamer will be more stable than someone who has not dreamt,
because the parasitic information remains in the latter case. But a
recaller is different from someone who has not dreamt at all, On
recall this person may indeed reinforce the parasitic link, undoing,
according to the theory, whatever unlearning took place, but will then
proceed to postulate links other than this one to try to explain the
production of the parasitic link: the single association is not left
189
isolated but all manner of reasons, some no doubt confabulations, are
put forward to provide a context, as well as the dream context, for
the conflation. Crick and Mitchison (1986, pp.239[lO9]-24O[llO])
rightly state that many of these reasons will be confabulations, but
they do add to the subject's knowledge of him/herself, which is
beneficial, rather than a cause of "instability". The unwanted link
has its existence prolonged but the overall effect is surely to cancel
out that instability.
pp
Hypothesised non-
Dreamer with damping
dreamer
Recailer with
new valid links
Instead, a real problem for the model is where bizarre links
from a dream are found to be immensely enlightening, to such an extent
that even if, as the theory's proponents say, the free-associations
produced on waking are later confabulations and not the real causes of
the dream, the cause of the enlightenment is not deserving of erasure.
Experimental data are, however, needed to illustrate this possibility,
and for this I repeat a dream I had while considering ending a period
of analytic psychotherapy.
In short, the dream started with myself on a river or lake
digging for lignite coal which I put on a barge. I then entered a
small submarine which was on the lake's surface and went down the
conning tower, but I was worried about water coming in, or of myself
being stuck, or the submarine sinking. Below there was a wooden door
and books on an old wooden bookcase. I got out by a large exit hatch
and realised that the lake was Lake Konstanz in Switzerland. I then
190
warned some children on the shore not to approach the submarine.
Later that morning I had a session of psychotherapy. The
therapist, Jennifer, had a wooden bookcase and the word 'lignite'
conjured up associations of gelignite and explosive submerged
material. A more major association in the dream was of Konstanz with
the submarine which I feared would let in water. The word 'Konstanz'
reminded me of a holiday at that lake I had had with K., a woman who
at that time made me feel smothered and drowned, it also associated to
'constant', the opposite to how I felt about the therapist, whom I
knew I could not keep seeing regularly. Before the dream I had never
associated the two women, but I was helped immensely by doing so
thereafter. Yet, according to the unlearning theory, at least one of
the links H, J, or C (in the following diagram) was 'parasitic' and
should have been damped down.
DIGGING FOR COAL
CHILDREN ON SHORE
LEAKY SUBMARINE
LAKE KONSTANZ
WOODEN DOOR.
J
F
B
C
C
Constant
Jennifer
WOODEN BOOKSHELF
D
A
K. dragging me down
191
(Note that the parts of the actual dream are shown as capitals, with
the associations to these in small letters.)
contend that links A-F could be
Crick and Mitchison
confabulations, and not the original reasons for the production of the
dream. However, whether or not they were the causes they are still
meaningful, so much so that an efficient brain would not put in danger
of erasure any such links as important to its world-view as:
Jennifer must show constancy
Lake Konstanz reminds me of K.
Jennifer is possibly dragging me down as K. did
Jennifer is like a leaky submarine
Furthermore, phonologically I will always recognise the
similarity between "Konstanz" and "constant", and my memory and visits
to the therapist associate a wooden bookcase with her.
Given that links A-D were already in my mind, even if they
didn't cause the dream, and that link E is a useful discovery, it
would be most unwise of the brain to set about erasing the links H, J,
L & M even if they did exist prior to the dream, because:
1) the whole structure would then be held together by fewer links,
2) H, J, L & M although not memories themselves, can hold other
memories together,
and 3) in the process of evoking these supposed parasitic links many
valid links are also brought up; if these are unlearned also then the
system seems to be loosing its memory ability overall.
If the links H, J, L & M did not exist prior to the dream,
in the same way that if they had occurred in a daydream we would not
192
ascribe to them prior existence, I fail to see which other link Crick
and Mitchison would claim my mind would be trying to erase. I am left
with two alternatives:
1) Links H, J, L & M were formed by the mind during dreaming
and their expression enabled me to link Jennifer and K. to my
advantage
2) All elements in the network existed before the dream, and
due to their proximities false associations result. The unlearning
theory holds that these are malign rather than benign.
The difference between these points is whether we can say
that the links H, J, L & M aid the retention and recall of the useful
links, or are a hindrance. In favour of the former option is a finding
in memory research that the greater the number of connections one can
imagine between elements to be remembered the greater the accuracy of
recall, and the most efficient method of remembering elements is to
construct a narrative involving them. (Some research suggests that to
produce a bizarre narrative or picture is even more helpful in
producing efficient recall, but I am tempted to believe that the extra
effort and thought needed to create a bizarre image is more likely to
be the cause of better recall in this case.)
Furthermore, although a Hopfield net may be assumed to hold
orthogonal memories (that is, the sum of any two memory matrices
multiplied together is zero) and mixed memories result when nonorthogonality is introduced, such correlativeness is the basis of
thinking and learning, and must be preserved. Animals do not just
193
learn many single patterns, but also the connections and correlations
between them, as shown in my dream above: a memory storage system is
not designed for this task, although it may be designed to produce
"best fit" solutions for single, independent inputs.
I conclude that for animal brains the non-orthogonality
(i.e. non-independence) of memories is actually helpful for the
efficiency of the overall memory, although the opposite is true for
the Hopfield net. Unlearning is thus a useful algorithm for the
Hopfield net, but would not be so for animal brains. On the basis of
this data Crick and Mitchison are therefore wrong to base a theory of
animal dreaming upon the Hopfield net analogy. In contrast, Palombo
and Clark et al state that REM sleep in humans and animals has the
function of assisting memory, by discovering and depicting
correlations between memories, which leads to enhanced intelligence.
The unlearning algorithm is useful for the efficient recall of
orthogonal memories in one type of storage system, and for the priming
of newly-formed reverberatory nets, but it has no counterpart in the
intelligent mammalian brain, except possibly for the priming during
REMS or NREMS of previously unused sets of neurones, ready for the
next day's learning. This off-hand suggestion by Crick and Mitchison
(1983) puts less emphasis on the functional evocation of bizarre
images during REMS than does the main part of their theory. The
possibility that bizarreness is an epiphenomenon of such a neural
preparation would be immune to all the objections made here against
the unlearning theory. Note that if such a preparatory function were
true for REMS this would fit in with the decrease of REMS time with
age (see chp. 1) only if it were also true that there were fewer and
194
fewer neurones available for new learning. Conversely, if the brain
has ample neurones for the experiences of a lifetime (which is much
more likely), or if previously used ones are recycled, then we would
expect the amount of such preparation to stay more constant with age,
as is found with NREMS time. This shows that it is possible to derive
more than one theory of dreams, and more than one set of experimental
predictions, from the single paradigm of neural networks. They both
meet in the suggestion by Hughlings Jackson (1835-1911) that sleep
fulfils the dual function of sweeping away unnecessary memories from
the previous day and of consolidating or maintaining more necessary
ones.
Empirical Evidence of Relevance to the Unlearning Theory
Crick and Mitchison (1983) claim that parasitic traces are
the cause of recurrent dreams and obsessions, with the possibility
also of causing the false associations which press upon the minds of
some schizophrenics. Brown and Donderi (1986) have, in fact, reported
that recurrent dreamers have a lower self-reported well-being than
non-recurrent dreamers and past-recurrent dreamers, which fits in with
the unlearning theory. However, much work shows that there is probably
very little in common, aetiologically, between dreaming and
schizophrenia (Vogel, 1968; Vogel, Barrowclough & Giesler, 1972). In
favour of the unlearning theory is the result of Benson and Zarcone
(1982), who found more middle-ear activity during REM sleep in
schizophrenics than in normals or schizoaffective patients. Middle ear
activity is a phasic event, occurring with PGO activity in dreaming
subjects. However, this paper does not show any intrusions of such
195
phasic events into NREMS or into waking life. Similarly, Benson and
Zarcone (1985) found no difference between schizophrenics and controls
with regard to REM intrusion from REMS into NREMS. In favour of the
theory, however, is the result of Greenberg, Pearlman, Fingar,
Kantrowitz & Kawliche (1970), in which dream deprived subjects showed
an increase in specific impulses or feelings as measured by a
Rorschach test. However, only four subjects were used, and 'changes in
form, movement, colour and regression yielded no consistent
results.' (ibid.)
A serious problem for the Hopfield model is how the system
would treat the mixed memories that result from tasks such as the
spreading activation memory experiments of Grossman and Eagle (1970).
It must be asked whether the unlearning mechanism would treat such
false memories as parasitic. If it were to do so then the system's
intelligent activity would be adversely affected, for although these
experiments on spreading activation and inattention lead to errors in
the lab, the system of interconnections between traces and different
degrees of encoding lies at the heart of our ability to think and
predict future stimuli. Collins and Quillan (1972) state that 'the
locating of paths between concepts... is basic to both comprehension
and recall', which implies that if one trace is activated it tends to
call up a few others - they are thus correlated, unlike the memories
in a Hopfield net. In the same volume Postman (1972) states
'Organizational processes were seen as critica . . .because of the
limited capacity of the memory system.' He notes how cuing can
counteract retroactive interference, which is one of the few examples
of traces being confused with each other.
196
At the centre of the unlearning theory is the concept of
'parasitic' information being held in the brain alongside valid
elements of knowledge. Using the analogy of an associative network the
latter elements correspond to data given to the memory by the
programmer, the former to spurious memories which crop up on recall,
often composed of a mixture of the valid memories.
The output in the computer model has its human counterpart
in, say, a verbal statement, but recall effects are not part of the
model. For example, slips of the tongue are not the result of mixed
representations at the level of storage (Motley, 1985), and cognitive
set contributes to human memory interference problems.
The interference of traces with each other was introduced as
an alternative to trace decay theory to explain Ebbinghaus' (1885,
'Memory') result that more trials are needed to relearn a list of
nonsense syllables after longer retention intervals. It suggests that
forgetting occurs because the trace is either masked or obliterated by
other traces of events,
as opposed to just decreasing as a function
of time. It is difficult to test the strong version of the theory,
that without interference no forgetting occurs, irrespective of the
time elapsed, because it is difficult to find a situation in which
intervening activity can be avoided and also because any means of
inhibiting activity is also likely to have other side-effects.
The similarity between Associative Interference Theory and
the Network ideas which form the basis of the associative network
paradigm is that the former has the following two assumptions:
1) that the essence of learning is the association of
stimuli and responses, and
2) that the phenomenon of transfer of past training is basic
197
to the processes of learning and forgetting (that is, that past
learning affects present learning).
Webb (1917) attributed RI (retroactive interference) to the
active disruption of the previously learned list A-B by the new list
A-C, and stated that when the A-B list was relearned there was a
tendency for the stimulus A to reinstate the competing response C. But
McGeoch (1932 & 1936) attempted to explain all forgetting as response
competition, in which A-B does not then weaken A-C, but simply
competes, and when A-C is stronger the C response is made rather than
B. Note that with the Hopfield model of learning there is, on the
contrary, direct interference between traces at the level of storage.
Similarly, Underwood (1948 a&b) assumed that in addition to response
competition there was also the unlearning of A-B associations during
A-C learning.
However, when natural language was studied differences were
found from those predicted in the lab (Keppel, 1968). This shows us
the gap between the artificial conditions from which the principles of
interference theory were derived and forgetting in everyday life.
According to Baddeley (1976, p.93), 'Interference theory has
difficulty demonstrating its relevance outside the rigid confines of
the verbal-learning laboratory and faces increasing competition from
more cognitive views of memory.' (Just as the Hopfield/Crick and
Mitchison model is in competition with cognitive and linguistic views
of obsessions and dreams.) He proceeds to explain Lhe major phenomena
of forgetting by a simple trace theory, in which a search process must
locate the trace followed by a process of discrimination between
traces/signals. RI effects can usually be avoided by cuing the subject
198
with the names of categories comprising the list to be recalled, when
the experiment is of free-recall. This indicates that the locating of
an item is the difficulty, for the item is as efficiently stored as
when it was first memorised, while it is the inter-memory links that
were made at that time which determine the efficiency of recall later
on. This suggests that work linking dreaming to memory should
concentrate on the assumption that the dream is providing new
connections between memories, rather than somehow strengthening each
memory independently. In paired associate learning the interference
seems to occur at the levels of both trace discrimination and trace
location, which suggests the idea of selective attention. Here a
signal remains discernible even if the interfering signal is much
stronger, as with the 'cocktail-party effect', whereas with the
Flopfield model the stronger memory had an overwhelmingly distorting
effect on the whole net. Baddeley (ibid. p.97) gives this graphical
interpretation of interference:
Trace
strength
rumination
memory
Time
Recall thus depends upon the relative strengths of traces.
Note that interference is not occurring at storage, the traces are
199
physically independent and hence do not need the "separation of
memories" that Hopfield's experiment demonstrates is useful in his
artificial system. The incredible accuracy and persistence of human
memory traces is shown in Sacks' (1985, chs.l5, 16 & 17), Penfield and
Perot (1963) and Luria (1968). The final result of all the work on
human interference is that the effects are more to do with the
subject's conceptual strategies than with storage interference, and it
is conceptual strategies that are missing from the work of Hopfield,
Feinstein & Palmer (1983). It is a change of set that results in a
change in the proposed signal detection and hence an alteration in the
manifest interference. With regard to this, I cite Lindsay and Norman
(1977, pp.351-354), who report an experiment in which a piece of prose
was given two titles, one concerning a celebration about achievement
in space, the other about a day in the park. One sentence in the whole
piece was comprehensible only with the former title and subjects given
that title were significantly more likely to recall that sentence
along with recall of the rest of the prose, than were the subjects
with the other title. The Hopfield net has neither this intelligent
recall activity nor a forgetting function - maybe these could
eradicate parasitic traces rather than the net having to unlearn them.
I conclude that interference problems in humans are at a higher level
than the physical storage, and also that problems of thought such as
obsessions and psychoses are also at a higher level of cognition than
the semantic memory traces used as subject matter. Further evidence on
the mixing of autobiographical memories at recall is given by Williams
(1976). As Hopfield et al (1983) conclude, though, unlearning may be
applicable to associative nets even if it is irrelevant to animals.
Crick and Mitchison claim that PGO stimulation brings to
200
mind the mixed memories. The PCO stimulation, however, which
correlates with the eye movements, is found to correlate with the
activity and vividness of the dream (see chapter 2), not with dream
bizarreness. Under the unlearning theory an increase in PGO
stimulation would be expected to correlate with an increase in
bizarreness, with more parasitic traces being accessed and unlearned.
The work on relating discontinuities, vividness and bizarreness to PGO
bursts does not accord with the unlearning theory. Even when Foulkes
and Pope (1973) contrasted 38 reports of REM bursts greater than 4
seconds with REM bursts of 3 seconds or less there was still no
difference in bizarreness. Watson (1972) did find a significant
difference between REM burst (and also PIP with no REM) and no REM/no
PIP mentation with regard to bizarreness. Similarly, Rechtschaffen,
Watson, Wincor, Molinari & Barta (1972) also reported an increase in
bizarreness during phasic events. However, Cohen (1979, p.l93) reports
Watson saying that subsequent research has not been as clear-cut.
Foulkes, Shepherd, Larson, Belvedere & Frost (1972) showed that
preadolescent Ss could discriminate REM-phasic from REM-tonic arousals
in terms of visual activities ascribed to subject as a dream character
but not in terms of overall presence of visual imagery or a PVE/SCE
distinction. The unlearning theory would allow visual experience and
cognitive mentation during a phasic event, as long as it tended to be
bizarre, but the theory has great difficulty explaining the lack of
increased bizarreness and discontinuities which re found, in most
studies, during phasic events.
The unlearning hypothesis must explain the dreaming that
occurs between PGO bursts, for the often suggested rationalisation
201
that occurs in this period seems to work against the postulated
unlearning that would have just taken place. Crick and Mitchison might
claim that these breaks are instances of the brain being inefficient.
The paradox is that if unlearning occurs all through the dream it will
affect the valid associations present (which some say are more in
evidence during the tonic periods), and if it is not then a connection
just unlearned will be subject to rationalisations, to incorporation,
which works
to
reinstate the connection in memory.
Molinari and Foulkes (1969) stated that the PVEs do not have
to be eruptions from the unconscious, merely a more active part of the
dream. Holmes' results indicate that the former is true. Yet in
neither case is bizarreness the distinguishing feature. The results of
Bowker and Morrison (1977) suggests that the PGO bursts are an
orienting response during REM sleep, much like those during waking
life, although of greater frequency. This makes the bursts much less
important than Crick and Mitchison suppose and opposes any idea that
they are a causal agent in dream production. If it can be assumed that
PCO bursts can result from dream images of a particularly vivid or
surprising kind, which is the opposite of Crick and Mitchison's
proposal, an explanation for the result of Foulkes and Pope (1973)
is produced. They found that dream vividness tested by a questionnaire
did not correlate with physiological phasic events, but that
scenes mentioned spontaneously as of great vividness did so correlate.
Work cited in chapter 6 showed that those people who have
greater REM sleep are more creative and artistic. However, this fact
cannot be used to judge the truth of the unlearning theory (that
dreams are bizarre in order to reduce fantasy in waking life) because
if the opposite correlation had been found, that daytime creativity is
202
associated with mundane REM dreams, Crick and Mitchison could reply
that individuals are creative due to lack of fantasy unlearning at
night! The theory is hence unfalsifiable at this point. Similarly,
Holmes (1976) found that 'divergers' [creative thinkers] spent more
time in REM sleep but had a lower density of REMs than did convergers
[scientific, narrow thinkers]. Convergers had a longer first period of
REM sleep than did divergers, and had a shorter REM sleep latency.
Unfortunately these results can be read either way for the unlearning
theory: convergers could be said to be so logical because of the
large amounts of unlearning they obtain each night, while on the other
hand it can be argued that the convergers do not need as much
unlearning as those more artistically inclined. The result of
Cartwright and Monroe (1968) (in which waking fantasy substitutes for
REM sleep) is the opposite of what the unlearning theory would
predict, Crick and Mitchison would have it that more compensation
(i.e. greater REM rebound) would follow a time of greater fantasy, and
that fantasy without PGO stimulation and the accompanying unlearning
function could never substitute for dream sleep, which, they claim,
has the function of unlearning.
The Meaningfulness of Human Dreams
A major empirical difference between top-down (Freudian,
structuralist, Palombo's, Evans', Rycroft's) and bottom-up (Clark et
al, Crick and Mitchison's) theories of dreaming occurs in the
attribution of meaning to dreams. Trivially, it is possible to claim
that any realistic images that a human produces may be made
meaningful, but the question here is whether some meaning is intrinsic
203
to the images, and to the way they are put together, whether or not
they are later interpreted. Evidence of relevance to this is seen in
Verdone (1965) and Foulkes (1966). They found that dreams from later
in the night, and later in each dream, referred to episodes earlier
in the dreamer's life. Clearly, some explanation must be found for
this pattern, which indicates that the dreams of the night have some
sort of progression in what they refer to, or mean.
In an effort to control for subjects' experiences prior to
dreaming Cartwright (1986) studied divorcees and related the different
types of dream seen to the emotional and healing state of the dreamer.
She found that dream sequences show problem-solving progress when
waking dysphoric affect is moderate, and poor quality dream-work when
affect levels were too high. The problem-solving dreams were
characterised by longer reports, a wider time perspective (i.e. not
just concentrating on the past), including the self in its marital
role (rather than ignoring that role), other people being present in
the dream, and by the needs of self-esteem and control being more
frequent in accounting for dream behaviour. However, some of these
results would be predicted by a passive theory of dreaming, which
would state that dreams are only meaningful in that they randomly
play-back images from a meaningful waking life. Her findings of the
temporal settings of dreams would then just reflect the temporal
concerns of the waking subject, that a married subject will think
about the present, a depressed divorcee about the past, and a non depressed divorcee about past, present and future. Similarly, the fact
that she found that 74% of the depressed divorcees had at least one
dream in which no other character appeared may be a reflection of
204
their actual waking lives, rather than of any psychopathology itself,
or of any active thinking during dreams. Better evidence that dreams
are not just a passive reflection and recall of waking life and
fantasies, and that dreaming is definitely an active response to
waking life is shown by the physiological work of Cartwright (1983).
She found that EM density was positively correlated (p=O.39) with
depression scores of divorcees, and that REM latency decreased with
depression scores. It was also found that the effect of divorce on REM
sleep was most severe and long-lasting 'in those for whom the life
event would be expected to induce the most affective upheaval and
whose traits required greater cognitive reorganisation' (1986,
p.4l8[288}), that is, in those women who scored higher on the scale of
Traditional/Liberated regarding female roles. A similar connection
between REMS and problem solving was found by Greenberg and Penman
(1975). They found an increase in REMS time in the night following a
challenging psychoanalytic session.
We must take care in the interpretation of this
correlational data, though, for
no evidence of direct causality
between life events and the functional change of dream images is
provided by these results.
Relation 1)
Change in environment causes depression
Relation 2)
Depression correlates with traditionality score
Relation 3)
Depression correlates with change in dream content
Relation 4)
Depression correlates with
change in dream sleep
physiology
Relation 5)
Change in dream
sleep physiology correlates with
change in dream content
205
Now, relation 5 is not one of causation, because REMS
parameters only affect dream activity, not content (see chapter 2),
and so we must find a third variable to cause them both, which is the
waking experience of depression. It therefore remains possible to
interpret these experiments as it being the neurophysiology of the
depressed state that alters the REMS parameters (see chapter 4), and
that dream content is still a mere passive reflection of waking
experience. Relation 4 would then be one of direct causality, and
relation 3 would be caused by a third variable, by waking life
experienced in a depressed way. It is therefore necessary, in order to
prove active thought is occurring during dreams, to show that some of
the events occurring in dreams are not taken or translated from waking
life, or that they use a different mode of thinking than waking life
usually does.
The alternatives for the causes of the changes in dream
images are thus (with an arrow indicating the link between cause and
effect):
Cl
__ I
il-4 I2—I3
CT
II I
Ii
Cl
12
13
dream images &
thoughts
C2 C3 waking concepts
& thoughts
dream thoughts
one image
Freud's account,
passively reflecting
causes the
that of the rebus
waking thoughts
next
206
A fourth possibility is that Waking Concept 1 causes the first image,
and that subsequent images are the result of bad and loose thinking.
This is a popular account of dreaming, and its possibility of being
true tells against Hall's (1966, p.88) dogmatic fourth rule of dream
interpretation, that the dream be taken as a whole rather than meaning
being ascribed to each part. That images may appear for no sensible
reason argues against his over-strong contention that 'nothing appears
in a dream which the dreamer does not put there himself' (ibid. p.87).
Such a contention also means that dreaming would then be the only
method of communication known that was not subject to 'noise'. (See
Campbell, 1984, for an introduction to this aspect of communication
theory.)
Evidence for the active vertical (i.e. waking concept to
sleeping image) transformational nature of dreams comes from Silberer
(1951), Antrobus (1978), and Luria (in Eysenck, 1986, pp.130-131) who
managed to show the logical transformation of waking conceptual
stimuli into dream images. Haskell (l986b) proposes that abstract
feature analysis is used in this metaphoric, symbolic thought. He
holds that it is
probably hard-mired into te brain aad is stwilar ta
the mathematical function of transformation of invariance. Any one
abstract meaning may thus be represented in many ways in a dream, and
he links this to the concept of overdetermination Whilst having
regard for Timpanaro's (1976) warnings about the validity of arguing
for a causal sequence of waking psychic thoughts to dream image (or
verbal slip) from the data of a seemingly meaningful image, I contend
that here we are on steadier ground than Freud was. Silberer's
transformations are simple and do not rely on notions of the
207
unconscious. Similarly, Cartwright (1986) finds meanings in successive
dreams which seem to have sufficient cause in known conscious thoughts
(though obviously the mechanism of transformation is unconscious). For
example, a dream of a violent cat trying to get out of a woman's arms
was likened to her husband; this I consider shows a passive
characteristic of dreams, viz, their ability to transform waking
thoughts.
A purpose for such concretization of thought is provided by
Atwood and Stolorow (1984, p.103, noted by D.P.Juda): 'When
configurations of experience of self and other find symbolisation in
concrete perceptual images and are thereby articulated with
hallucinatory vividness, the dreamer's feeling of conviction about the
validity and reality of these configurations receives a powerful
reinforcement. Perceiving, after all, is believing.'
The mere transformation and translation of waking concepts,
however, does not indicate the presence of active horizontal
(i.e. image to image) thinking in the narrative. As an analogy, an
event may be written about, and hence transformed into the printed
page, and yet the only thinking performed is in this transformation,
not on the page. We may contrast this to typing a problem into a
computer, which usually results in more than a restatement of the
physical facts in another language. Hall is thus begging the question
when he claims that 'we study dreams in order to find out what people
are thinking about during sleep' (1966, p.10) when the thesis of his
book is simply that the 'images of dreams' may be 'pictures of our
conceptions'. On his scheme it remains possible that the thinking
occurs during waking life, and that the dream is just a rebus of these
208
thoughts.
The active meaningfulness of dreams (as against passive
translation of waking life and thoughts) is indicated by Schatzman's
(1986) finding that successful problem-solving dreams often end with a
spontaneous awakening, which he takes to mean that the dream's work is
finished and ready for viewing. However, I consider it possible that
the waking may occur simply because the dream has alluded to a part of
the dreamer's conscious waking life, possibly by chance, which raises
questions for the sleeping mind about where the problem has come from.
Work on the difficulty of maintaining lucid dreaming has suggested
that such consciousness of one's mental state may be antithetical to
the continuance of the dream state, except for a very few people, and
so the subject wakes up. I therefore do not think that Schatzman's
finding is conclusive with regard to the presence of creative
horizontal thought in dreams.
An argument for active thinking during dreams is provided by
Hall's observation (1966, p.28) that our dreams do not reflect a
random sample of waking events; instead they select themes. For
example, 'conveyance and recreational settings have a higher incidence
in dreams than they do in waking life' (ibid.), and current affairs
and work are rarely mentioned. Our dream activities are rarely the
routine chores of life. Despite this, however, he proceeds without
proof to beg the question about the meaningfulness of dreams by
writing about 'one pretty reliable rule to follow, if the dream does
not make sense taken at its face value one should look for a symbol
which when appropriately translated will make the dream sensible'
(ibid. p.98). But even irrational or accidental actions in waking life
209
can be rationalised in this way, and made to appear motivated. As
already noted, however, Antrobus, Silberer and Luria's experiments do
show that such a translation is justified, at least sometimes.
Evidence for dreams as important in themselves (as opposed
to an epiphenomenon of physiological REM sleep) comes from Fiss,
Ellman & Klein (1969). They showed that the need to complete an
interrupted dream by mentation in waking life was quite independent of
any need to have REMS itself. They showed that dreams must be
completed, and not just substituted for by more REM time. Being
prevented from finishing a dream was found to be more disruptive, as
measured by projected anger and frustration, than if the subject was
not allowed to dream at all. (However, I note that the same may be
said for the physiologically necessary activities of eating and
urination!)
Crick and Mitchison and the Meaningfulness of Dreams
An essential part of the unlearning hypothesis of dreams is
that 'parasitic' memory traces are activated by random stimulation of
the cortex. This stimulation is random in the sense of containing no
information, the actual PGO bursts do not occur at random, being under
the control of the cotex (Gadea-Ciria, 1976 & 1977). The unlearning
theory proposes that the imagery during REM sleep is composed of
'parasitic' bizarre memory traces which may or may not be related to
other traces evoked in the same dream; Crick and Mitchison leave it
undecided whether the whole of the memory is stimulated at any one
time or just parts thereof. According to the theory a complete dream
210
can hence only tell us anything original about the real world if it
stimulates us, when awake, to evoke non-parasitic associations to the
dream elements.
On the face of it the theory thus predicts that a dream as a
whole piece of "work" will not have a meaning, and denies that some
meaningful part of the dream (such as a problem or wish) could have
been the stimulus for the rest of the piece. The whole dream is thus
more like gurglings in the stomach (or, at most, like a disconnected
revue) than a symphony; although individual sounds and associations
can be found in both gurglings and symphonies only in the latter case
are they exerting any formative influence to the other parts.
However, it is conceivable that a dream theory for humans
based on language and thought, a theory stressing the insightfulness
of dreams, could be consistent with the unlearning model if there is a
tonic phase of ratiocination following each phasic PGO jolt. However,
this ratiocination would be a top-down process, and to postulate it
would go against the simplicity of the bottom-up approach of Crick and
Mitchison. Such a ratiocination can be pictured as a random pulling
out of stored files in order to break false links followed by the
subject debating about why certain files have been taken out and where
to return them to. Yet some of these dream images appear so
overdetermined, as with Lake Konstanz in my dream mentioned earlier in
this chapter, that the subject may be incredulous at the claim that
the initial product was not only random but also maladaptive. However,
note Hunt's, 1982, point that "information" obtained from the IChing and Tarot cards also appears overdetermined in many
circumstances. Obviously Tarot cards do not fall into their order of
play because of any meaning ascribed to them, but once they are played
211
'sense' can be made of their order.
I conclude that the unlearning theory can allow for short
periods of narrative during dreams, in addition to the narratives
needed for the definition of items in the dream as already discussed.
It remains an empirical matter as to how long these narratives in
dreams actually are, but we must now consider what is involved in the
formation of dream narratives.
The Production of a Meaningful Narrative
The movements of a symphony may use different instruments
and a different tempo, so much so that the novice does not appreciate
them as coming from the same work. As each movement progresses though,
there are intrusions from the initial theme of the work, some go
unrecognised, some are played with different instruments, and all
through there are references to other parts of the work. The initial
theme is played with, distorted, repeated, dissected and combined with
other themes until the final crescendo. Such a piece can be produced
in two ways, corresponding to two types of dream theory:
1) The piece is written in its entirety before presentation
to the audience, so the work that has gone into it may be considerably
longer than the time it takes to play or be shown. D C.Dennett (1979)
postulates that a dream experience on waking is due to the sudden
playing of a pre-recorded dream 'cassette' in one go, it may even be
played backwards from how it was originally written. Similarly, the
Freudian position can allow for the primary processes to be working on
the material while we are awake, with a final polishing-off and then a
212
presentation during sleep.
2) The piece is improvised before the audience, resulting in
admiration from them at the skill and memory of the artist(s); the
thinking behind the piece is simultaneous with its transmission, and a
part can be repeated or harked back to, and used as a new stimulus.
This corresponds in dream theories to problem-solving and computer
models.
Such a harking back to earlier parts of the dream shows that
the narrative is not composed of individual random selections from our
memory. So, is the unlearning theory proposing that traces are chosen
not quite at random, that parasitic traces with something in common
are activated together? Now, surely our dreams are not composed of a
collection of successive random images alone, for example, with my
Konstanz dream (above), I did not see a succession of separate though
semantically related scenes:
lake - coal coal -barge
inside-drowning
lake - submarine
downstairs-stuck
submarine - enter
downstairs-bookcase
not only were the scenes on the whole continuous but some actions
necessitated knowledge of what had occurred previously in the dream
(and sometimes in my life). For example, when I was inside the
submarine I remembered that it was in a lake, after looking at the
bookshelf I still searched for an exit, and once on shore I remembered
how dangerous it had been and instructed children not to approach the
submarine. This is more like the stream of mental processes present in
waking life than separate images which are forced to fit with each
213
other to provide a confabulated meaning for the whole.
The stream of mental processes of waking life is partly
generated by external stimuli, with thought providing connections as
well as additional stimuli. In dreams the cortex is self-stimulating,
one image providing the stimulus for the next in some instances of
decision-making and recall of each scene, while in other instances an
image just seems to appear. This sudden appearance may be explained by
the new idea of PGO stimulation or by the older idea of unconscious
mental processes, either way some images lay the ground for others
leading us to conclude with one of the following options:
1) Some amount of thought produces some, or all, of the
subsequent images,
2) Images are produced independently of each other, the
brain ties together the result into a coherent plot and adds cause and
effect between images.
The second option does not fit with the actual dream
experience. This option claims much more than the idea of secondary
elaboration does over the aspect of the independence of each image.
The theory of secondary elaboration is similar but acceptable in that
its workings are manifested in waking life after the dream is recalled
and is thus susceptible to experimental manipulation. It claims that
links may be incorporated into a dream such that once one wakes up the
link appears to have occurred before the image it links did: de SaintDenys called this method of incorporation the 'retrospective illusion'
(1982, p.117) and it has recently been supported by Seligman and
Yellen (1987). Winson (1985) suggests that we may register sensations
at different times but consciously have an illusion of simultaneity.
214
This can be partly countered with the evidence of lucid dreams, which
are the most extreme example of images being produced by thought, as
opposed to narrative being an overlay for the images (see Green,
1968).
Now, the first option above is acceptable to the unlearning
theory in that the first image produced may be a parasitic one and yet
the succeeding ones can be non-parasitic. But surely this will reduce
the efficiency of the proposed mechanism because non-parasitic traces
will now be damped down. (Unless the unlearning mechanism is only held
to be working when PGO stimulation appears, and then the mentation
returns to normal learning until the next PGO burst, which leaves the
theory with the problem of what happens to bizarre traces produced
when PGO stimulation is not occurring.) I believe therefore that while
the putative mechanism of PGO bursts and unlearning could allow for a
self-stimulating cortex (as in 1 above) the mechanism would have to be
backed up by a block on the brain making sense of the images, a block
on the searching-for-meaning activity of the brain, which would
otherwise lead to the subsequent adaptive traces being unlearned. If
it has to be assumed that unlearning only takes place during a PGO
burst, not during the tonic interval between them, we then have a
similarity with the work of Molinari and Foulkes (1969).
Although this addition could square the unlearning theory
with the symphonic, rational and narrative properties f dreams I have
already mentioned in chapter 2 the doubts about there being any
differences in the content of dream mentation at times of PGO
stimulation and PGO quiescence. Furthermore, not only has an unproved
unlearning mechanism been postulated, it also seems necessary to
215
restrict its activity to the time of PGO events; thus unlearning would
have to be a sudden phenomenon, quickly turned on and off, in order to
avoid the unlearning of non-bizarre parts of the dream. It would be
easier for the brain's narrative and searching-for-meaning activity to
be believed to be switched off for the whole REM period (just as
muscle tone is) - but empirically that is not found. Our ability to be
curious and to puzzle during dreams is claimed by C.Green, ibid., to
be a major precipitating factor for lucid dreaming. An objection to
this point is to deny that the narrative ability of the brain can ever
be switched off, but there are such times as I shall now describe.
While drowsy, at sleep onset or just after waking, a
succession of images can be seen at play (for a description of which
see Foulkes & Vogel, 1965, and Tart, 1969, section 2). The scenes are
very simple, usually having just one object or worà w'hich suflers
continual transformations - it is harder to remember the preceding
images than with REM images of a similar duration. This state of mind,
composed of no extensive narrative, and simple images combining, is
closer to the predictions of the unlearning model than is the state of
mind at stage REM. For example, Cicogna, Cavallero & Bosinelli (1986)
note that the sleep-onset state, and daydreaming, refer to waking
memories much more than do REM dreams.
To sum up, we have seen that the idea of 'random
stimulation' is at the heart of the reverse learn ig model. 'The
normal inputs and outputs are then disconnected and the net is given a
random input. When the net has given a response, the synapses between
inputs and outputs are then adjusted to reduce this association
slightly.' (Crick and Mitchison, 1986). However, humans and other
216
animals are not simply stimulus driven. Although an input may be
random, such as white noise, it does not follow that outputs will be
random for complex, intelligent creatures. Obviously, a neural net is
stimulus driven, and we expect a random or nonsensical output to
answer a random or nonsensical input; but a mammal or bird has such a
complex neural apparatus in between sense and motor organs that there
is still intelligent activity in the brain even when it is sensorily
deprived or randomly stimulated.
It appears that the PGO bursts are not stimulating any
content of the brain, but instead allow such content to be vividly
imaged. The active nature of such a process indicates that the dream
images are themselves the purpose of REMS and PGO bursts, and that the
images are therefore meaningful. But on this evidence we can only
assert that it is pieces of dreams that are meaningful. On the
evidence just reviewed we cannot say anything more than that dreams
contain insightful, but separate, scenes, which are connected at most
via the subject's underlying concerns. There is only a little evidence
that any progression is involved in the content of one scene to the
next, or one dream to the next, except inasmuch as one scene provides
the starting material for the next one.
Alternative Accounts of Narrative
Seligman and Yellen (1987) claim that narra ive is used to
tie together separately produced dream images in order to impose a
meaningful pattern upon them. As an example they cite an experiment by
Barnard iKarmel, in which randomly flashing lights were shown to a
group of people. After a while some popular music was played and the
217
lights started to appear to be flashing in time to the music. The
experimenter concluded that such a misperception was due to the active
searching for meaning and sense which the brain was imposing upon the
incoming stimuli.
Their account of the subject actively restructuring an
experience is similar to Freud's statement that: 'There can be no
doubt that the censoring agency, whose influence we have so far only
recognised in limitation, and omissions in the dream-content, is also
responsible for interpolations and additions in it.' (1953, p.489)
Freud called them 'connecting thoughts', and Seligman and
Yellen also distinguish between the vivid parts of a dream and these
more cerebral additions. Such an embellishment is acceptable to the
dream theories of Palombo, Clark, Crick and Mitchison, structuralism,
Freud and Rycroft, that we are studying. In the case of the
structuralist account of dreams there is the lesson from anthropology
that myths are still analysable despite minor changes and
interpolations - Levi-Strauss states that this robustness makes myths
very different from poems. Only Crick and Mitchison's theory demands
any retrospective patching up of the dream, the others can have this
patching up occurring at the same time as the images are produced,
much as the constraint of finding a rhyning word will be relevant to a
poet as he or she evokes the next word in a poem. Such a constraint
would act much like that of Freud's 'representability' simultaneously
with the other creative abilities in dream formation.
It is found that subjects woken up quickly from sleep have
more bizarre dreams than those who are woken up slowly; subjects
restructure the whole dream to make it accord better with common
218
sense. However, such a finding of the effects of return to
consciousness tell us nothing about any editing which may have been
occurring while the subject was completely asleep, although it does
leave the possibility that whenever one wakes, and whatever the
abruptness of waking, some embellishment does occur.
It is necessary to ask when exactly this interpolative
activity is supposed to take place, if it is supposed to occur during
sleep. With the Karmel demonstration the events to be connected occur
contemporaneously, but with dreams the events are consecutive, often
following each other quite smoothly. Freud states that secondary
revision: 'fills in the gaps in the dream-structure with shreds and
patches. As a result of its efforts, the dream loses its appearance of
absurdity and diconnectedness . . .s (ibid. p.490.) Such revised dreams
'appear to have a meaning, but that meaning is as far removed as
possible from their true significance'. (ibid.)
We are still left with no proof that connecting thoughts are
added to some putative inchoate dream, which is somehow more basic
than these added links. Seligman and Yellen (ibid.) claim that the
difference in vividness between connecting thoughts and 'Primary
Visual Experiences' indicate that the later are somehow more original,
that they are the earlier parts of the final product, but do not
disprove the possibility, expanded upon in chapter 2, that all parts
of the dream are as basic as each other, that some are made more vivid
by their coincidence with PGO bursts, and that the only proof of
active creative changes in a dream script occurs upon slow waking,
that is, with the intrusion of consciousness into the actual
remembrance of the dream.
219
If it is true that some revision can occur, even in the
sleeping subject, there is still the interrelationship of various
images to study empirically and to explain. If much of that
interrelationship is explained by the revision, then that very mental
process must be studied. We can still study whether the later images
in a dream bear any relationship to the earlier images, even if that
relationship is warped or hidden somewhat by the postulated revision
process, and we can indeed further study whether parts of dreams are
related to parts of other dreams from the same night. As a preparation
for doing this I will summarise the predictions that the different
theories are making.
Phenomenological Predictions
The different theories predict different relationships
between the images:
Progression towards realistic resolution of ostensive waking
concerns; Evans' computer theory.
Progression towards resolution of metaphorical translations
of waking concerns; Rycroft, Kuper and Stone.
Presence of oppositions with their progressive permutation
in arriving at this solution; Kuper and Stone.
Comparison of waking concerns to past events, possibly with
a search for resolutions; Palombo.
No prediction about the presence of any progression from one
dream image to the next; Freud.
220
Denial that any progressive mentation is present in dreams,
although short narratives may be needed for the definition
of the contents of images; Crick and Mitchison; Clark,
Rafelski & Winston.
The theories also make different predictions about the
presence of condensation in dreams:
No condensation predicted, the dream world should be as
sensible as the waking world, although new adaptive programs
may be being explored; Evans.
Condensation expected, and predicted to often be
enlightening for the dreamer. Condensations will be between
recent and distant memories;
Palombo.
Condensation expected, sometimes enlightening for the
dreamer, with recent/recent, distant/recent, and
distant/distant memory combinations; Freud; Rycroft; Clark,
Rafelski & Winston.
Condensation expected, especially at the resolution or
attempted resolution of relevant oppositions; Kuper and
Stone.
Bizarre, confusing, unenlightening condensations predicted;
Crick and Mitchison.
Bearing these predictions in mind, we will now proceed to
the study of some recorded dreams.
221
Descola P.
Head-shrinkers versus shrinks: Jivaroan dream analysis.
Revised version of a paper presented at the 46th International
Congress of Americanists, Amsterdam, July 1988.
Levi-Strauss C.
The Jealous Potter. (Trans. B. Chorier)
University of Chicago Press.
(1988)
Spiro M. Whatever happened to the id?
(1979) ,81:5-13.
American Anthropologist
Van Veizen H.
Irma's rape: the hermeneutics of scructuralism and
psychoanalysis compared.
In: The Psychoanalytical Study of Society
(L. Boyer & Grolnick S. eds.) (1988) Analytic Press, Hilisdale NJ.
CHAPTER 14
THE NARRATIVES
OF THE DREA}IS OF FOUR NIGHTS
The first part of this thesis has shown the central place
that narrative holds in the experience of dreaming. That it is more
common than either bizarreness or condensation means that it is
usually overlooked as one of the formal attributes of dreams, or, as
in Crick and Mitchison (1986), that little attempt is made to explain
it once it is acknowledged. This attribute may enable us to discover
the function of dreaming. Maybe it is a particularly human ability to
weave the images together, maybe it is a result of our ability to
speak; but whether or not narrative is present in the REM sleep
activity of animals it tells us that, at least for humans, low level
accounts of dreaming such as the unlearning hypothesis or brainrelaxation or the eye-muscle activation theory must explain the amount
of narrative that is empirically found.
I wish to take up a suggestion
oi
Iou1'ies in the to'fio'wing
chapters, in an attempt to empirically test the predictions made by
the different theories delineated in the last chapter.
Foulkes (1982a) bids dream workers to concentrate on the
form of the dream and not its content. This he thinks will enable us
to discover the process by which a dream is put together, and should
be attempted before we aim to understand or predict the content of
dreams. An example of this method was the work of Foulkes and others
on REM-Active and REM-Quiescent imagery, which ained to discover if
there was more active imagery when PCO stimulation was occurring (see
Molinari & Foulkes, 1969; Pivik & Foulkes,1973; and chapter 2).
Haskell (l986h) also suggests that the logical structure of dreams is
222
more easily and more exactly manipulatable than is their content; with
respect to this he cites the work of Antrobus (1978) on the
transformation of incorporated conditioned visual stimuli. This work
involved conditioning the subjects to associate certain visual stimuli
with certain tones. When the tones were played to subjects when asleep
some aspects of the relevant visual stimuli were incorporated into the
dreams, but transformations had occurred. For example, a man cutting
the bark off a tree with a knife was transformed into the image of
cutting a pie with a kitchen knife (ibid. p.576).
NIGHT 1: DREAMS AND ANALYSIS
The following dreams were recorded at a sleep lab in
Chicago in 1973 as part of a project on changing levels of anxiety
during the day and anxiety levels in dreams. There were four subjects
in the study, chosen because of their exceptional recall ability. They
were inducted by Professor Cartwright, had their REM dreams recorded
by lab technicians, and knew the study was about daytime anxiety and
dream anxiety. The subjects had to rate their anxiety on Trait/State
scales at various times during the day, were asked about the content
of their concerns in pre-sleep interviews, and had to rate their
dreams for anxiety. The setting was that subjects would sleep in one
bedroom while one technician monitored their sleep. The standard
protocol for measuring sleep and REMs was used: 'Right/left EOG,
central EEG referred to the opposite ear, and chin EMG. The lab
protocol was to wait five minutes into the first REM before an
awakening was made, ten minutes into the second REM, 15 into the
third, 20 into the fourth and any subsequent REM.' (Cartwright, 1988.)
223
KJ's eye movements were monitored by the EOG, which can differentiate
waking, sleep onset, non-REM sleep and REM sleep from each other (and
even help distinguish stage 4 sleep from stage 2 in that there is a
small pick-up of the slow wave sleep waves by the EOG electrodes near
the eyes). Dr. Cartwright (1988) writes also that KJ was woken
specifically within bursts of REMs; this taken with the set questions
he was asked about the presence and reality of images show that in all
cases here he was woken during REM periods of a specified duration.
The EEC was used to differentiate waking from each sleep stage and
from stage REM, which has a distinct waveform somewhat like
wakefulness but with no alpha-waves and the presence of saw-tooth
waves. The EMG registered a small drop in neck muscle tone when the
subject fell asleep, and a complete loss of tone when the subject
entered REM sleep.
KJ slept well on all the nights. This was remarked to him
after the first night by the technician 'You really fell asleep very
fast ... and you did have lots of dreams'. KJ had four REM periods on
the first night, which is average, 6 on the second, 4 on the third,
and 5 on the fourth, and so the indications are that he slept well and
normally on those nights, especially as the REM periods were long and
did not result in spontaneous awakenings on KJ's part, which does
happen with worried subjects. His good first recall night indicates
that he had spent the traditional one adaptation night in the lab
beforehand; no mention is made in the protocols of such a night, which
is the usual procedure. The presence of so much lab related imagery
for the first night's recalls does accord with the lack of mention of
any previous experimental (i.e. awakening) nights spent in any lab.
224
When woken he was asked the standard 'What was running through your
mind just at the moment I called?' After the report he was asked about
the emotional tone of the dream, and what distortions were present.
Subjects would give associations to the material in the morning, as
well as state if any memories appeared in the dreams.
Six years later the series were given to Professor Kuper as
data for a theory of his which had only recently been devised.
A Summary of the Dreams, and Pre- and Post-Sleep Interviews.
KJ was one of a group who went once a month to a dream lab in
Chicago. He was a graduate student. Joan Smith is a fellow graduate
student; Chris is his major professor dissertation supervisor; he
lives with Connie; Dr. Ros Cartwrig\it is the department prol:essor; 're
lab, assistants are technicians who run the sleep lab at night,
operating the EEG, EOG and EMG equipment.
Night 1. (30-31/7/73)
Pre-sleep interview.
He has been reading some fiction and realises he is out of touch
with how he was a few years back, due to his concentrating on
psychology. 'I feel like I'm sacrificing all that to do what I am
doing right now.' He is not disappointed about specialising in
psychology, but realises how big the outside world is. 'I was just
realising that in terms of my life it is just such a heavy head think
all the time. Everything is psych. You know there is not someone
forcing me to do it, but I feel that my head is being shaped and that
is how I am looking at the world - as a clinical psychologist. Hearing
my own conversation, hearing the way I look at things, how I observe
225
them - I guess that is what is getting me down - everything is
analysed. I'm being perceived more and more as a psychologist and that
is something I never wanted to be on the outside. When I'm with people
I realised that all I talk about is psychology and I'm talking with
other friends who are not in psychology and I'm still talking about
psychology. We used to spend a lot of time with people, but we don't
do that much anymore. The last year I sometimes woke up wishing that I
would not be me, I'm just so tired of being me and dragging this mind
around, but that is not the case at all this year.' He is also worried
about not passing the prelims the second time, and now feels like he
is constantly being tested. He has to present a patient the next day
and is worried about his professor's reaction, as KJ has the
reputation of being 'oppositional'. 'She doesn't know me but she's
heard about me.' He has been using the rest of the students as a
support in public arguments with the department heads. He has been
wondering about going into therapy.
1st REM Awakening, 5 mm. into REM, 1.40 a.m.
A feeling of pursuit with no other people involved.
2nd REM Awakening, 10 mm. into REM, 3.05 a.m.
'I am in a sleep lab and kind of there's a small room in
between two rooms and I'm standing there just at the moment that you
called me waiting to unlock that door to go into another door inside.
There's nobody in the inside room right now which surprised me because
a few minutes ago there was a whole bunch of people in there and here
I'd come back and it's empty now so I'm standing there looking for a
226
key and ... maybe there's a window where I'm standing, an outlet to
the world, and I'm thinking about the ball game or the city or why it
is the door is locked. As I was standing there at the door trying to
unlock it as I was standing there fumbling for the key I was thinking
about the ball game. I had been in this room earlier with a lot of
people [who were watching the ball game]'. People had been delegated
responsibility as lab assistants, KJ had been delegated responsibility
too, which was connected with unlocking the door. He had come out of
one bigger room and was now in the small room in between that large
room and the one he wanted to get into. 'I think the person in charge
was Dr. Cartwright. I don't remember seeing her there but I know that
was who was in charge. In the dream itself there could very well have
been an awareness that this was a sleep lab...' The room he was in
'had the feeling of - kind of like the john - with that towel rack and
that continuous cloth towel.' He thought that those people who were
there earlier should be there still.
3rd. REM Awakening, 15 mm. into REM, 4.55 am
'This is another sleep lab dream. I got up from the sleep
lab to go to the kitchen to get some ice cream ... I'm standing at the
refrigerator with the freezer part open.' 'I took out of the freezer a
camera...' The scene becomes like his apartment, and then - Rosemary
walks in and then Roz walks in and 'asks me what I was taking pictures
of ... I wasn't taking pictures ... I was carrying the camera around
because of the light meter ... I didn't even know what I was doing
with it on. Then the whole sleep lab crew - two or three people - went
into the kitchen to make liverwurst sandwiches' and the crew have a
227
good time. The bed then becomes like his bedroom at home and Connie is
there. She is talking to IU and he is annoyed by it, because he wants
to sleep. He was lying in the bed with his feet at the end his head is
usually at. He asks her about two objects that had moved: one was like
a mirror. She placed the second object over to the right side of the
bed and over on the left side of the bed where the TV was she placed a
mirror; 'I didn't like the position'. And then she was concerned that
it would interfere with the TV aerial if she placed it too close to
the TV. 'At the very end she was asking me if we couldn't go to the
Italian Village [restaurant] one night ... asking me to make a
commitment for us for next weekend.'
He then added another part of the same dream:
He is in the underground on the el which is dirty and dark.
He gets off the train and waits for another. When a train arrives he
has to decide if it is his. The conductor says which it is and KJ goes
back to the waiting room. The room becomes like a lecture auditorium
and someone asks 'what are you doing here?' Although KJ was waiting
for a train he says he wants to buy an automobile, in order to be
sarcastic. A few minutes later many people arrive and the same
aggressive man starts to read out nationalities. They jump to their
feet when called and go to the bottom level. Finally he asks 'who is
Jewish?' KJ is annoyed and does not jump up, but just moves one step.
He then starts to think he is dreaming, and then reads the professor's
report on what he has just dreamt. He is confused at this because the
report is so objective and clinical.
4th. Awakening, 20 mm. into REM, 6.14 am
He is in bed in the lab talking to a male experimentee.
228
Another male brings in fruit juice, and other gifts are brought in,
including a magazine. A woman brings in a political cartoon, of Nixon
or Mighty Mouse, but he can't understand it and feels stupid. He feels
comfortable with his companion.
After getting up he goes down a long corridor. He looks out
in the street where there are people waiting at restaurants for
breakfast. Professor Cartwright comes up to him and says that Chris
wouldn't like him to be getting this external stimulation, but he
believes that that is really her opinion. Back in the lab he finds
that Chris has also slept there, as has another man, John. His room is
now like a sun-porch, and he wonders whether he should take all his
clothes with him or leave them there. In this sun room he is holding a
black light bulb and a can of air-freshener. Looking out of a window
he sees a school yard and students. Everyone he sees is coming out of
one door except for a small boy who climbs through a hole - 'but
that's what kids do'. A larger kid then manages to get through the
same hole.
Additional recall, 6.55am.
He is in a downstairs office kissing a colleague on the
cheek. Someone walks in but is pleased rather than surprised or
shocked. He has his own desk.
In the post-sleep interview he relates the dream events to
his present academic life as mentioned before going to sleep.
229
ANALYSIS
There are five obvious condensations in the above dreams.
1) The dream lab which was also the 'john'.
2) The sleep lab which was also his apartment.
3) The sleep lab bed which was also his bed at home.
4) The train station waiting-room which was also an auditorium.
5) The dream lab room which was also a sun-porch.
Each of these have some amount of narrative around them to
define them, as Crick and Mitchison may expect.
Kuper (1983) analysed the dreams of this and the following
night. He apparently found that the narrative of the first set of
dreams reflected the chronology of a night spent in the lab: from
leaving the lab workers to having a dream recorded, wondering whether
his performance is satisfactory, not knowing whether to take his
clothes with him the next morning, and finally meeting fellow workers
the next day. This is not surprising: subjects in a dream lab (as
opposed to being studied at home) not only sleep differently on the
first night but also tend to incorporate the external happenings
(worries about the experiment, procedures adopted, etc.) into their
dreams. A sceptic may decide that it is predictable that someone
spending a night in a dream lab for possibly the first time will dream
of such an experience, and in the correct order chronologically - this
may happen especially if the worried subject spends a great amount of
time falling asleep after each awakening, allowing time for rumination
over the attendant expected circumstances.
According to Kuper (ibid.) the subject's own conflicts, of
over-involvement in psychology, and of professional status, were
230
present in this progression and were resolved in the last scene with
KJ, a psychologist with an office, accepted by colleagues. Such a
wish-fulfilment would accord with the theories of Freud and Evans.
In lU's own words his concerns are that:
'I feel that my head is being shaped and that is how I am looking at
the world - as a clinical psychologist.'
'I'm being perceived more and more as a psychologist and that is
something that I never wanted to be on the outside.'
.because of the situation I have put myself into I will continue
to do it. I will continue to act this way [analysing everything] and
still be frustrated.'
This dichotomy between the inside world of the department
and the outside world of Chicago is depicted in the dreams of the
night, as Rycroft, and Kuper and Stone, would predict. In reality, he
is now mixing socially with few people except psychologists, although
that wasn't so in the past.
He mentions his insecurities towards the senior staff, but
that he will now confront them in private rather than in front of
other students. He continually feels 'like I'm tested', and has just
failed two exams.
He is wondering about whether to go into therapy. Note that
in the dreams of nights 1 & 2 there is often ambiguity about whether
he's the psychologist or the patient.
The first scene contains a reversal (a sinilar co-varying of
oppositions is found in myths, and structuralists are much concerned
with them in their analyses):
231
He was with others in the room who were watching an outside activity
is followed by
He is outside the now empty room, he is thinking of the outside
activity and looks outside, it is hard to use the key to get back in.
This scene connects with his pre-sleep statement that: 'I
don't even hang around with them [graduate students] very much - we
don't have very many friends. We used to spend a lot of time with
people, but we don't do that much anymore.' I expanded upon such
recapitulatory dreams earlier in the thesis (chapter 13); they are
dreams which symbolise the events of a long span of time in just a few
images. This idea is somewhat different from Palombo's, in that he
holds that dreams will contain real past and present memories, rather
than the symbolic encapsulations of whole eras which I claim is
present in this case. It does accord with Rycroft's notion of
metaphorical translations of waking concerns.
In the third awakening he is outside the lab but still has
on him observation equipment, much as he states that being a clinical
psychologist penetrates into his everyday life. In the dream he is
embarrassed when asked what he was taking pictures of. He says he's
using it for a light-meter; I take that to mean to see better, to
observe better, as psychology is claimed to help one to do. Meanwhile,
all this argument detracts from his enjoyment of the ice-cream.
In the next scenes of this awakening he further investigates
the activity of choosing, and relates it to authority. He doesn't
admit to the railwaynian that he's having to choose which train to
take, instead he says he's waiting to buy a car.
232
In the fourth awakening he is told by his professor that he
can read in bed but can't get stimulation from people outside. He had
slept on the sun-porch, a place half inside, half outside, which
accords with the structuralist notion of the search for mediators
between oppositions. Earlier, gifts were brought in for him and
another male subject by a male experimenter. A woman comes in and
shows him a cartoon in a magazine; he doesn't understand it and feels
self-conscious. However, he feels comfortable with the male companion.
The magazine, which he associates with Newsweek, is a product of the
outside, in that it reports news, but he associates it with doctors'
offices, the inside; it is what one is supposed to be reading. (Can
the cartoon be related to the children? Can we relate the adult
reading [Newsweek] to the child's reading [cartoon], which is of 'a
significant person, like maybe a fat characterisation of Nixon, but
maybe it was mighty mouse,...'? Such speculations are now
unanswerable, and some of the dream theories would bid us concentrate
on the overall form of the dream, rather than the meaning of
individual items of content, anyway.)
There is a greater amount of movement from room to room in
this night than in the others. There are three sorts of places in the
dream, the actual lab, experimenters' and psychologists' offices, and
Chicago. Structurally, these differ along the following three
dimensions: (As is usual in structuralist notation, a '+' indicates
the first item of the opposition in question, tnd '-' the second
item.)
233
+
+
inside/outside
psychologist/patient
+
equal/subordinate
+
The four
Chicago sun-porch
offices
lab
+
+
main locations are thus defined by three
oppositions. There are also many places which are in between the
outside and inside, such as the corridor.
Overall he progresses thus
inside patient superiors
outside patient superiors
1
inside patient colleagues
inside
'If
psychol. colleagues
So, despite his stated antipathy towards being a
psychologist, and the possible symbolisation in the 'john' scenes of
the elimination of this part of himself (Marks, 1988), he finishes the
night as a psychologist, in an office.
Such a change in dimensions from one scene to another was
also noted by Kuper (1979), in which four settings were analysed as
differing in terms of two sets of oppositions:
Green countryside versus Town
and Green countryside looking 'happy' versus
Green countryside looking 'sad'
versus
Green countryside finally looking not 'so bad'.
234
The problem is that for any change of scene it is always
possible to find some dimension on which they differ, otherwise there
would not be any recognisable change. That the dimension has some
aetiological significance for the change, and that it is a dimension
with some further meaning, is a claim some would think far-fetched.
After the first awakening there is depicted in each dream an
opening to the outside world. Awakening 2: in the small room outside
the lab there is "a window - a kind of outlet to the world again". He
then thinks of the ball game, and prior to this there were the
technicians watching the ball game on TV in the lab. Awakening 3: In
their bed-room Connie interferes with the TV aerial by means of a
mirror. There is then a sub-plot with KJ in an underground station.
Awakening 4: He walks through a corridor to get to a Chicago street.
He goes back to his bedroom which is a sun-porch. He looks out of a
window to see two kids climbing through a hole to get out of school.
He says the small room is like 'the john', a place which in
real life is half inside and half outside. Also, standing at the door
he describes as unpleasant (as on the fourth night he describes
himself as indecisive). He says in the morning (see appendix) that the
difficulty of using the key made him feel distant from the
experimenters. It is not clear from the report whether he is entering
the experimenters' room or the subject's room. He does say that it is
his responsibility to go into that room, however, which implies that
it is actually the lab.
235
Inside
Outside
lab
laundries in the
offices
street. Restaurants
kitchen
bedroom
auditorium
(window4le)
Liminal places
small room (john)
sun-porch
underground station
narrow corridor
I will now list the remaining reversals, which are part of
the movement from scene to scene, and which are claimed by the
structuralist account to show that a dream 'juggles the elements of an
issue which engages the dreamer - not in a random way, but rather by
patiently rearranging the elements of the issue, combining and
recombining them' (Kuper, l986a, pp.213-214) and that 'the mind defines
a system of differences - a set of binary oppositions which can be
used to specify and differentiate the items of information
the problem
of which
is composed ... the statement of the problem is then
subjected to a series of transformations, by negating, inverting,
reversing or substituting its elements.' (Kuper, 1986a, p.213.)
236
He walks into kitchen with camera for light measurement.
1)
followed by
He walks into sun-room with black light-bulb.
He doesn't speak and is criticised by Roz about the camera.
2)
followed by
He criticises talkative Connie for moving the TV and mirror.
3)
Connie complains about interference with the TV aerial.
followed by
Connie wants a link with the outside world (going to restaurant).
4)
Technicians are eating and he wants to sleep.
followed by
Connie wants to go out to eat and he wants to sleep.
With the last opposition we can see that there is a problem
of the falsifiability of this method of analysis. If one scene has a
similar opposition to another then it can be counted as a confirmation
of the importance of the factors involved, if it does not have a
similar opposition, then it can be counted as a transformation, a
permutation of the previous scene. The only answer to this is to
collect more examples of the reversals, in order to discover if such a
stepwise mytho-logic is being used. This approach finds support in
Mary Douglas' essay 'The Meaning of Myth', in wh ch she takes issue
with some of Levi-Strauss' purported reversals, but also claims (1978,
p.160) that the importance of reversals to meaning is undeniable in
some myths, such as the story of Asdiwal (in Levi-Strauss, 1978a),
237
where successive oppositions of height are finally resolved with the
hero immobilised half-way up a mountain, and where some quite bizarre
scenes are analysed in terms of being the opposite of an earlier, more
sensible, scene. Another reversal found by Levi-Strauss (1977)
involved tribes that showed familiarity between father and son yet
formal respect between uncle and nephew, which mirrored others in
which the father represented family authority while uncles were
treated with familiarity.
One task in this and the next three chapters is to perform a
structural analysis on the dreams of each night, so that Kuper and
Stone's theory can be compared to the others.
Finally, for this night, I note incidents which are similar
to those in night 3. Connie 'waits on' KJ in this night, whereas in
night 3 she appears as a 'maid' from a TV series. This night he waits
for a train and must decide whether to catch the A or B, whereas on
night 3 he has to decide about catching a bus. Although these last
two points may not be significant they illustrate the importance of
comparison between scenes, rather than the taking of associations to
individual elements.
238
CHAPTER 15
NIGHT TWO: DREAMS AND ANALYSIS
In the pre-sleep interview recorded four weeks after the
first night he told of having to complete an overdue paper for his
professor, that he had been reading about Transactional Analysis, and
that he had been reading of a baseball player who was in the wrong
league for his ability.
1st. Awakening.
He sees 'a wagon', it has 'four wheels' with one axle
connecting them like on a train, with a little cartoon man next to
each one.
2nd. Awakening.
He sees a sketchy scene from Treasure Island.
3rd. Awakening.
He is experimenting on a man to test the man's skills about
making judgements about his state of mind. He is then in the suburbs
where the man works.
4th. Awakening.
A review of the last dream. He was 'writing a theme, my
account of this person who was making the judgemnts', but it could
have been about someone else. He writes a paragraph on the aetiology
of his behaviour - asking was the guy retarded or just socially
deprived?
239
5th. Awakening.
He is telling a young person how to behave in a certain way,
'in order to achieve my or our ends' but 'he didn't know without me
telling him' . He is outside that person's house, the man's mother is
indoors. The second time they are there they go through the back door
but 'it was trickier this time' - KJ was inside having prepared a
script for the young man who was talking to his mother. KJ could see
the man's sister through another window. The mother questioned KJ
about 'a date' with the young man. She was pleased about it and asked
when he could visit them again. This made KJ flustered. KJ was trying
to get permission for the young man to be out of the house. There was
some advanced planning for both of these visits. On the second visit
it was raining.
Awakening.
He is going through a turnstile with his bookbag in
order to get inside a department store and there's one stand for
bathroom or kitchen utensils or appliances and he was looking for one
in particular. He gives the bookbag to the cashier or this guy sitting
in a booth, who is an older guy, maybe the owner of the store. He then
walks up a set of 'strange' steps but can't get further than the
eighth for then he would have to turn 180 degrees to continue going up
the other steps, which appears impossible. He stood there for a while
and there were people on the set of steps above him and there were
other people on his steps so he figured it was possible. The old man
then said that he can go up, and he then realised how to do it. He
found that it was 'just a matter of turning' and went up the steps. He
240
was still looking for that same kitchen or bathroom utensil or
appliance, with the help of the older man. He was looking for that
upstairs but didn't find it and then decided to look in the toy
section for something else, for a particular toy. He had a sense the
toy was round.
Next, he was involved with some 'heavy problem solving' on
the second or third floor of a house or apartment building. He was in
a room with two other males and was 'getting an image of a circle',
which was connected with 'evaluating or assessing' a patient. They
discuss the correctness of his interpretation of the patient.
Next, KJ is in a room cleaning up. Connie, who initially was
crying and had a distorted voice, conies walking in holding a record
cover. The picture on the record cover was in black and white and was
of a baby, or something to do with gentleness, or with families.
They discuss a man who lives down the street who has lied
about where he came from, and also about his job. They decide he is
schizophrenic.
[In the original text the following is written now:
'4th (?) Awakening, 6:50, about 10 mm. into REM.']
On interview he said that in the room on the third floor 'we
were sitting on the floor in kind of an old place and spinning
something - a gyroscope, ... spinning, ... there was a narrow metal
shaft, a vertical pin or something and I was sp nning something on
it.' The other men had also been spinning something around on this
little pinhead on the floor. 'It seems we were making up a story to go
along with this little toy or whatever it was, it seems as if we were
241
giving it human characteristics, describing a person that this little
thing was.' There was a difference of opinion and they were seeing
experimentally what this little thing could do. KJ describes it that:
'the first time we ran this thing I had it hooked up so there was a
little something that was on top, above the spinning part that smoked
and then some more people came in and I was demonstrating my point or
something. Now I put whatever it was that was smoking and it seems as
if it was a ... cigarette butt ... on the cylinder but below the
spinning thing and when it was done this way then that convinced me
that it was much more effective in demonstrating my point.' These
other people had been in another room and now they came in to see what
KJ had done. He says '.. .that spindle thing seems to be on a much
higher level than usual functioning', later he says that it is like a
propeller, with the ascending smoke driving it round.
In the morning interview he was asked whether the dreams
tied in with the pre-sleep interview. He replied that they did in
terms of his completing patient reports, which involve history,
symptoms and so forth. He stated that the first dream and the second
dream seem to relate a lot to hypnogogic experimentation which he was
also attempting on these nights.
242
ANALYSIS
I can only find two condensations in the course of this
night.
1) The second or third floor of a 'house or apartment building' which
was also part of the department store.
2) The record cover which was 'something to do with gentleness, or
with families'.
The former was embedded in much narrative, whereas the
second one wasn't.
KJ starts off as an observer of a man doing hypnogogic
experiments, KJ himself had previously been such a subject. Next comes
a dream in which he is recapitulating what occurred in the last dream
and he makes an initial diagnostic assessment of aetiology (whether
the patient is retarded or socially deprived). He writes a theme on
this patient. The next dream has a different patient but we note there
that KJ's relationship to him is of the next level of psychological
testing, in which the subject's behaviour is now manipulated. KJ is
giving him orders concerning what to do at the patient's mother's
apartment. There KJ is observed by the sister and questioned by the
mother. He becomes submissive and flustered at this, not wanting to
lie about how much he can do. Another visit is planned: he had arrived
in the rain. I consider that this dream shows some of the problems met
with in hospital out-patient work; arranging home visits and being
watched and questioned by relatives who want the best and most
frequent treatment to be given.
We can read the next awakening as showing the next step in
243
the formalities of patient care. He is involved with 'heavy problemsolving' on the second or third floor; he has an interpretation of
some patient and they all make up a story about a spinning toy, giving
it 'human characteristics' over which there is a difference of
opinion. This 'spindle thing' is on a 'higher level than usual
functioning'. With his girlfriend, he then diagnoses a man who lives
down the street. I interpret these last scenes as portraying the later
aspects of patient diagnosis, the case conferences which occur in
hospitals to give medical professionals the chance to theorise about a
patient in his or her absence, above the everyday level of the ward
both literally and also in terms of explanations for behaviour being
'higher' than what is observed. This new model, which is like a
gyroscope, was found with the help of an old, authoritative man, who
is the owner of the whole building. Finally, he and Connie see through
the lies of the paranoid schizophrenic and the case is finished. In
the dream he was clearing up a room when Connie arrived, which is an
apt picture. The happy ending and use of metaphors accord with Freud's
theory, and with Rycroft's.
Nothing in the texts has had to be distorted to find this
progression in the dream series, and no scenes have had to be
excluded. Such a movement through the night is reminiscent not only of
the progressive chronology found in the first night but also of Kuper
and Stone's (1982) interpretation of Freud's Irma dream, which showed
two progressions in that dream. It is also similar to Propp's analysis
of folk tales, which showed an initial confl Ct, followed by a
journey, aid given by another character, a battle or fight, and
finally a resolution in the form of the finding of treasure or a
marriage.
244
Onto the psychological assessment sequence, (the melodic line
of the dream series), is placed the dialectical argument concerning
his worth as a psychologist (the harmony), starting with the initial
opposition:
Good psychologist vs. Bad psychologist
He can improve vs. He is retarded rather than
by learning
environmentally deprived
Help given by vs. He looks for
ING
older colleagues
the model himself
He describes a theory, vs. Patient disagrees
others arrive to listen
Patient is lying, KJ is
a good psychologist.
Note that the initial thesis in each step is opposed by its
antithesis. The synthesis is the result, the compromise, between the
two. (Although Kuper and Stone's analysis, and mine here, both find a
dialectical argument as well as simple progressions it remains a
possibility that this is a consequence of the medical content of these
particular dreams, rather than an intrinsic style of all dreams. Freud
was dreaming of a medical examination, and KJ of treating patients,
both of which are logical, stepwise, argumentative activities.
However, if this is the explanation for the form of the dream then we
must assume a great dependence of dream mentation on waking styles of
245
thought, which I consider doubtful.)
The key used for the transformation of his mental concepts
into physical terms is:
Up:Down :: Model:Behaviour,
and Up:Down :: Psychologist:Patient.
Such a key to translation fits in with Rycroft's and with
the structuralist view of dreams.
With regard to this depiction of concepts in terms of the
physical environment, Leach (1982, p.147) notes that according to a
member of a primitive tribe cosmologies will control present social
life, 'the anthropologist must regard the ancestral cosmos as an
imaginary projection of present experience.. .' Similarly, KJ's
experiences are being externalised and thought about in terms of the
dream environment.
Note that in the first night a woman shows him a magazine
picture which he couldn't understand, and this made him feel stupid,
whereas in the second night Connie's picture is shown when he is a
successful psychologist. It is late on in the night when we see that
it is KJ who works on the model, he no longer needs help; those who
come in only do so 'to see what I had done'. The importance of the
opposition Up/Down to this resolution may explain the curious way that
KJ proves his point to the others - he places a roach underneath a
propeller so that the rising smoke causes it to rotate. This image and
that of a black and white photo of a mother or baby may also be
related to the very first brief awakening, in which KJ sees four
246
wheels with a little cartoon man next to each one. Whether it is KJ's
memory of that first awakening, or that that dream is unconsciously
transformed as if it was never conscious, we can see that the later
awakening does not merely include, translate, or show again, that
earlier scene. The scene is transformed progressively, as if the
dreamer had an aim in mind.
The first and second lines of the dialectical argument are
both stated in the dream as applying to the old patient, at first as
to whether he is competent enough to assess a mental state and next as
to whether his problem is due to deprivation or retardation. The next
dream involves a young patient; similarly, in the third dream he looks
firstly for an old appliance, and then for a (young) children's
object. However, unlike this permuting of Old/Young, the opposition
Male/Female is not important in this dream - neither sex has a
monopoly of any trait.
Of equal importance to the oppositions Old/Young and Up/Down
is that of Revolving/Static, which may be related to KJ's 'original
problem' (see Levi-Strauss 'The Structural Study of Myth', 1977,
p.216) of who is performing, who is evaluating, of whether he is the
patient or the psychologist. This is transformed into a symbolic
'derivative problem' of ascending vs. revolving. It is this derivative
problem that is resolved in the final scene by the presence of the
record cover, a stationary cover of that which rotates - the picture
on the cover also encapsulates the Transactional Analysis concept of
PAC, while simultaneously KJ resolves the origina problem by showing
himself as a successful evaluator and psychologist.
Throughout, there is an emphasis on stages, on floor levels,
which connects with:
247
1) his worries about which sleep stages he can maintain, REMS
vs. hypnogogic awakenings,
and 2) the levels of Parent, Adult, Child used in Transactional
Analysis. He behaves as a child when at ground level with the woman,
is next guided upstairs by an old man while looking for an adult
implement, and ends up on the third floor teaching old men. This
parent-like image is then followed by Connie showing him the 'family',
or 'child' picture,
and 3) his pre-sleep banter about baseball players being in
the wrong league for their ability.
Despite these changing activities and levels I suggest that
KJ is a psychologist throughout this series, although at one point
needing to receive instruction from an older man (colleague?) and
although reverting to a passive role with respect to the patient's
mother as a way of ignoring her demand (which is, in fact, one of his
own real life coping mechanisms).
So, the night progresses thus:
Cartoon men on the end of axles.
Original problem depicted (three awakenings).
Under instruction from an old man, he, a human, revolves and
starts to ascend.
Upstairs, an old man helps him to find a 'round' toy.
He makes smoke ascend, to instruct old men on how the
248
propeller (now given human characteristics), revolves.
Childlike, or inferior, Connie (she cries, then is a maid)
shows him a record cover with 'baby', 'gentle', 'family',
'mother'
characteristics.
In this scheme the structuralist claim would be that the
record cover is the resolution, it contains that which revolves; the
night ends with he and Connie summing up a person with psychiatric
illness, and simultaneously 'clearing up'. According to Crick and
Mitchison's theory the cover, as it is an overdetermined image, would
be a parasitic memory, rather than the product of a long series of
creative thought.
Note that the order in which revolving and ascending occurs
differentiates him from the model of the patient: he revolves and then
ascends, whereas the model revolves because of the ascending smoke. We
see that KJ is different from the patient because KJ evaluates and
ascends (in status); it is his job that makes him different from the
patients! IU evaluates the patient and thus performs and is given
status, whereas the patient must perform ('I was telling him how to
behave.') and is then evaluated. It is this discrimination, this
difference, which solves KJ's original problem of whether he is the
psychologist or the patient, and he then goes on to solve the problem
of whether he is a good or bad psychologist.
249
Downs tairs
Can't find adult
Revolves when
appliance
ordered by old
\
/man
Second floor
Looks for round
ASCENDS
Old man assists
toY\/
Third floor
Old men watch
With ascending smoke
gyroscope revolves\
His apartment.
KJ is parent & psychologist
The dominant theme throughout the night is his professional
competence, can he produce a viable model of the patient's behaviour?
There is also the bureaucracy of professional relations, that is,
dealing with family, seeing himself in what the patient does, and
dealing with colleagues. In the pre-sleep interview one of his main
concerns had, in fact, been finishing a paper for his teacher, and
reading T.A. and Gestalt theories. Getting work done and seniority
both crop up in the night. In the post-sleep interview he says the
dreams were concerned with patient evaluation, and the main opposition
in the dream series is thus good psychologist/bad psychologist. In
addition there are two physical oppositions by which he attempts to
help solve this difference, old/young, and up/down. It is the
ostensive importance of these oppositions that suggests that the
250
application of Evans' account to KJ's success in these dreams may be
ignoring some of the evidence, although KJ's activities do fit into
the 'practising programs' scheme. After the first two brief
awakenings, the importance of which will be shown later, the plot
concerns an old patient (most of lU's patients are old), then a young
patient, then the young psychologist has help from older colleagues
upstairs, then lastly he is upstairs proving his theory.
Up
Down
Young
Old
(Again, it can be remarked that if one scene takes place
downstairs then the next can only either be there again, or upstairs.
In the former case a structuralist would look for another opposition,
and in the latter case claim that one had been found! However, it may
be claimed that with a whole series of dreams such ad hoc analysis is
not possible.)
These two physical oppositions form the setting for the
scenes, pace Hunt who considers the setting to be only that, with no
symbolic function, although the environment can have such a function
in human waking life. (See chapter 8 in Structural Anthropology vol.1
about the dual organisation of tribal camps.)
The Up/Down opposition divides the characters into two
groups, patients and senior psychologists, leaving the one other
251
player,
KJ,
moving between scenes. Height is not mentioned in the
final scene, possibly that aspect has been resolved, for he is
successful, and yet with family rather than with colleagues. Age,
however, still figures in that a picture of a baby or mother is shown
to him. The structuralist claim is that we find abrupt oppositions in
the dream between two scenes because they are a means of solving the
problem of defining areas of
KJ's
life. Similarly, Leach notes that
'it has become increasingly apparent that neighbouring small-scale
communities, even when they are lumped together under the same
"tribal" label, are just as likely to be sharply contrasted as they
are to be very much the same. The contrast may be a significant
feature of the overall pattern' (1982, p.142). For example, in the
fifth awakening the following oppositions occur:
OUTSIDE
becomes
INSIDE
Parent
KJ
Mother
Adult
Mother
Sister
Child
Patient
KJ
& Patient
As was shown in chapter 12 in the analysis of the myth of
Sodom, we can most easily discover if there are any relevant
oppositions by comparing first to last scenes. If we do that for this
night of dreams we find there is a change from distorted people on the
end of axles to a distorted Connie (both literally, in the way he
reports her image to look, and in that she is crying) who brings him a
record cover. The record cover is the final mediator, it neither
ascends nor revolves but is connected intimately with each because of
252
its cover (depicting Parent Adult Child) and its contents, which can
revolve.
What is important to a structuralist analysis is not to
concentrate upon ascending itself, but rather ascending in relation to
remaining at ground level, and then this opposition in relation to the
opposition revolving/static. Similarly, Levi-Strauss (1977, p. 46)
states (about the avunculate):
'The error of traditional anthropology, like that of
traditional linguistics, was to consider the terms, and not the
relations between the terms.' Sperber puts it thus: 'If relationships
between items are considered, then shared or contrasted features stand
out as the basis for symbolic associations. The greater the number of
items related, the fewer the features which are likely to play a role:
one should study not symbols but symbolic systems.' (1979, p.30.)
Applicable to the oppositions that KJ appears to be using is
Levi-Strauss' statement that 'the elements of mythical thought ... lie
half-way between percepts and concepts' (l966,p.l8). They are
obviously concrete, but they "resemble concepts in their powers of
reference" (ibid.). Note, however, that there is a small difference
between lU's use of the oppositions up/down and revolving/static to
externalise his thinking, and his use of the opposition inside
Chicago/outside in countryside in the next night. The latter is a
literal depiction of his worries, he is wondering whether to leave
that city, whereas the first opposition is metaphorical, and the
second one is almost an arbitrary signifier. That the link between a
signifier and what it signifies is arbitrary was a crucial premise of
253
Saussure's theory of language. Rycroft would predict the utilisation
of just metaphorical symbols, whereas for the unlearning theory the
more bizarre the metaphor the more urgently must it be unlearned. By
contrast, Kuper (1987) states that a recent tendency in Levi-Strauss'
work is to claim that myths are not necessarily engaged with problems
from everyday life, but rather 'reflect each other', thus allowing for
this toying with almost ungrounded signifiers.
The use of the concrete opposition ascending/descending to
facilitate thinking about the abstract notion of status is not
entirely arbitrary, as some signs are. Systems can be envisaged in
which it is the physically lowest member who has the greatest
authority, for example, the devil in medieval Christian cosmology is
in hell while his minions are on earth, above, but elevation is still
important. Similarly, spinning/static is not entirely arbitrary in its
use as a representation of performing the science of psychology, for
the science is connected with the phrase and activity of 'spinning a
story' , and many would say that much of psychology (as with
sociobiology) is simply the production of just-so stories. It is the
patient whose ascending (i.e. Parent Adult Child) aspect is examined,
leading to the spinning of a story, the writing of an explanatory
report, while it is KJ's spinning of a story which results in his
increase in status and altitude. Each of these oppositions is thus in
Leach's (1976 p.l2) terminology an example of a 'natural index',
rather than a 'symbol', which is completely arbitrary. Leach states
that myths, totems and rituals use 'metaphoric condensations' of
elements from different contexts, such as comparing the difference
between human groups to the difference between two animal species. He
254
says this is needed because "to think clearly we must externalise the
concepts, akin to using pencil and paper." (ibid. p.17). Similarly,
Levi-Strauss considers that totems are chosen not because they are
'good to eat', but because they are 'good to think'.
So, I conclude this chapter by noting that in dreams we are
finding a whole environment full of messages, many of which are
symbolic, and are manipulated in terms of oppositions. Of relevance to
this is Levi-Strauss' (1966, p.267) comment that:
'The idea that the universe of primitives (or supposedly
such) consists principally in messages is not new. But until recently
a negative value was attributed to what was wron,Ly taken. to b a
distinctive characteristic, as though this difference between the
universe of the primitives and our own contained the explanation of
their mental and technological inferiority, when what it does is
rather to put them on a par with modern theorists of documentation.'
That the messages in dreams are so simple, and composed of
simple, abrupt oppositions, should come as no surprise, if we note
also the single-minded and naive way that the dream characters act.
255
CHAPTER 16
NIGHT THREE: DREA1'IS AND ANALYSIS
Pre-sleep interview: he complains that he has much
more work to do, and that he's 'still up in the air' about
arrangements for his practicum. He is in charge of a course but has
not officially been delegated the responsibility by Benton. His
clothes are all stained from the rain. He complains about a baseball
commentator whose game statistics KJ doesn't agree with.
Hypnogogic Awakening.
He sees 'a couple of jet planes', 'something to do
with international relations'; they are stationary, like a picture in
a book.
1st REM Awakening. l.45am, 5 mins. into REM.
He is outside a town getting a drink.
2nd REM Awakening. 3.45am, 10 mins. into 1i.
He is in a car with his mother, they pull up into a
cul-de-sac at a summer resort. It isn't a good place to park, 'we
didn't park within the lines'. As they get out there are people
around, at a sort of buffet, and he looks for an attractive woman
there, although they are all older. He wonders if it is appropriate to
'check out the scene' in this way. The buffet is happening in a lodge,
and the food is supplies for an ocean-going voyage, which are
somehow 'stolen' goods. The food is mainly caramel, with nuts. His
mother's friends are there next, and she asks him if he wants to talk
with them. He says 'no' but they still speak, telling him that he is
on the wrong course and at the wrong university. He starts to feel
256
unsure of himself when one couple say that with his course options he
may end up not wanting to go back to school. One of their sons used to
be in his class and then went to a different school.
He is then on a bench at the resort with his parents
and feels that he is in the wrong school, and that he should just walk
into Harvard or Yale. He thinks of Colleen, as if she was able to do
that.
Lastly, he is eating the caramel, with marshmallows
and walnuts.
3rd Awakening
With Connie he is travelling by car on back roads,
through familiar areas. They are in conflict about the route, 'whether
we could travel over water'. Connie was saying that it was physically
impossible for them to travel over the water. 'We sort of decided by
testing it out. I think that's why we went one direction, then the
other, then went back then continued on our way. ... But then once we
got underway ... we just buzzed along in the car.' They see a small
pond with grey, dead logs in it, all stripped of their bark. They
could see the bottom. The pond is at an angle with the logs at the
bottom like a dam. They are at the top of the pond. They are then with
others in the car, driving downhill. Somehow they turn back but carry
on the same way. As they travel downhill in the midst of the lakes
they make fun of him in a joky way, which he would prefer them not to
do. He is now a passenger. One of them is sad due to not visiting Otis
airforce base as usual, and KJ is amused as 'it was no big deal', or
maybe 'because she was so accustomed to going to Cape Cod [where the
257
base is] every summer'.
They are then flying through the air towards a town
that looked like Chicago. He was the leader as they were being 'sucked
through the air'. Like a roller-coaster they swooped down, landing at
a bank where they asked the president for $10 million. They walk down
the bank steps with the president and his employees and start to
laugh. He thinks of marketing coloured pancake batter with the money.
4th Awakening. 6.45am. 20mins. into REM.
KJ is alone on a motorbike, looking for a place to
park on a small city street at an intersection. The intersection is at
the bottom of a hill, he is driving down and sees no spaces on the
right side of the road. On the left side there are two spaces, he
takes one and five other cars try to dash into the other space. 'The
person who did get the space was not the person who was first to the
spot, the person who had come up from the direction I had come from,
from the direction behind me had tried to park and he was a young,
little bit longish haired person ... he parked his car ... not between
the two lines' . When he gets out of the car he sees that something is
wrong, gets back in, and parks within the lines. KJ starts to realise
that this is a dream and knows he will be woken up soon to relate it.
KJ thinks it strange that the meters allow so much space for each car.
He is talking to an old Wooster friend, Bob, about
his experiences in graduate school in the dream lab. Bob had dropped
out many years ago; he had a letter for KJ. They are then lying down
next to a fireplace and Bob changes into Craz. Although he hadn't seen
Craz in a year Craz just gives him a perfunctory 'hi' and starts to
transact business, about $10 for some dope. The receipt is scrawled on
258
a piece of paper, which is somehow also a cheque made out to someone
else, with Craz having the intention of turning the money over to KJ.
KJ is confused at this, and is also annoyed at Craz talking business
when they have only just met again. Craz says 'I'm really glad to see
you' in a cold way that wasn't convincing. They argue over the balance
of the money, because KJ will be left owing him money for the deal Craz says this is not possible to rearrange the deal because he has
not got the dope on him, which surprises KJ. At this point Craz
becomes Bob.
He is at Washington University and sees Connie's old
roommate with a Wooster friend of his. He runs after her and the scene
is like at Junior High school, and he remembers having to decide at
that time which bus to take home to save money. Down a corridor he
sees someone who looks like her £rom the ba, 1rit t s a 'man,
brother. He is told that she is at Mandolin hall, which is not part of
the university.
Dream told on waking up:
He is with a young mother and her five kids. They are all
retarded and he is talking to her about how to take care of the kids.
He is sad because he has been friendly with the kids. He walks onto a
porch and sees them, maybe all boys, looking over some railings. Their
backs were stained and KJ knows they have been like this for days. He
is sad because the mother had irresponsibly not bathed them.
He is then at an outdoor concert with Bob or Craz. The kids
were close by and two of them come up to say goodbye. 'We assured them
that the next time we were by we'd stop by ... I kind of shook hands'
with the little kid. They did not expect such an adult courtesy from
259
the kid, that he would come over alone to where they were standing. KJ
then explains to Craz or Bob who the kid was.
ANALYS IS
I have found six obvious condensations on this night.
1) The food at the buffet which is also supplies for a ship.
2) Bob, who changes into Craz.
3) The receipt for the dope, which is also a cheque.
4) Craz, who changes into Bob.
5) The scene at Washington University, which is also like
Junior High school.
6) Someone who looks like Connie's old roommate, but who is
actually the roommate's brother.
There are certainly some other bizarre happenings, such as
flying through the air, but these cannot be classed as 'condensations'
according to Freud's use of the term, nor as 'mixed memories', as
defined by Crick and Mitchison.
In order to pick out any relevant metaphors or progressions,
as required by Rycroft's theory, and that of Kuper and Stone, I will
now compare the first long awakening to the last awakening and
material from waking up.
260
4th Awakening and Waking up
2nd Awakening
Car parked above beach in
Motorcycle parked at bottom
cul - de - sac.
of hill.
His mother is present.
Kids' bad mother is there.
Disrespect from mother's friends
Respect from mother's child.
Old people present.
Children present.
They criticise his schooling
He gives psychological advice
to the mother, 'how to take
care of the kids'.
He admires fellow-student Colleen. He is respected by drop-out
Bob.
The dream between these two is also midway in certain
respects concerning its plot:
He is a passenger in a car that doesn't park;
it is
travelling a long way between familiar locations associated
with his schooling.
As the car descends a steep hill there is joky criticism
from his friends but he isn't angry. He then travels a long way in the
sky, then goes down, and then descends the bank steps.
The next awakening, the fourth, has him riding alone, again
downhill, but now he can park correctly, whereas the 'longish haired
person' does not do so at first. He is critical of his friend Craz for
261
impoliteness, whereas in the 2nd awakening he was the silent victim of
impoliteness; furthermore, the dream just before waking up has a kid
being polite to him. The last dream certainly shows a simple wishfulfilment, which would accord with both Freud and Evans.
We can now compare the second and fourth/last dreams, and
then the third and fourth/last dreams, to obtain the following code:
Criticised : Admired :: Above : Below
and Ribbed : Admired :: Descending : Low
Here, an internal comparison, personal to KJ, is related to
a physical comparison, or opposition, which is one of the objectives,
in Levi-Strauss' view, of myths. Similarly, in 'The Story of Asdiwal'
(l978a, p.162), Levi-Strauss notes that 'the hero goes from east to
west, then returns from west to east. This return journey is modulated
by another one, from the south to the north and then from the north to
the south'. He notes that this movement in the myth corresponds to
actual movements made by the tribe during the year, and is related to
the economic and social aspects of the tribe. In this case, for KJ,
being above is related to criticism, it is safer to be lower down,
just as KJ had complained in the pre-sleep interview of having too
much responsibility. This use of the metaphor of altitude accords with
Rycroft and with Kuper and Stone.
Comparing the first, albeit hypnogogic awakening to the last
scene there is a change from planes being above ('up in the air', as
he describes a prospective interview with a supervisor), with the land
below and a concern with international relations, to equal relations
with a child at ground level. But it is not simply a case of LU
262
progressively
changing level, there is also the question of the
position of those whom he is involved with: his superiors moan, he
cannot find a suitable woman among all the old women at the buffet, a
peer sells him dope, there is no great antagonism with other peers
and, finally, a child is friendly. But, while he is 'going down' in
the world, using a motorcycle rather than a car, and also attending a
concert, he is still a psychologist - it is the other actors in the
scenes who change in social status, and change in their ways of
treating him.
So, the argument running through the night can be tabulated
thus:
Initial Opposition
Second Opposition
Third Opposition
Responsibility of
college
Responsible for journey
& responsible in the
air, but enjoys it
Carefree,
with drug-addict
& drop-out
Inferior' s
pleasantries
Peers' joky ribbing
Superiors' complaints
HIGH
(above beach)
DESCENDING
LOW
(in air, on road, lake & steps) (ground level)
263
Similarly, in 'The Story of Asdiwal', Levi-Strauss notes the
presence of the oppositions Heaven/Earth
(east) ,
Earth
(west)/Subterranean world, and finally, Peak/Valley. The latter
opposition is resolved when the hero is trapped halfway up the
mountain because he forgot his snow-shoes (Structural Anthropology,
vol.2, p.163). See chapter 12 for other examples from mythology of
inimobilisation signifying the mediation of an insoluble opposition.
Under the structuralist account, the sloping lake could be a
mediator in the second opposition, for at the same time it is high and
low, symbolising the ability to be both responsible and less
responsible at the same time. Note his complaint in the pre-sleep
interview about the statisticians who were studying the performance of
a baseball star and finding him not quite up to the mark. This
complaint about those who, quite literally, remain on the side-lines
may be related to 1(3's
problem of resporitltt'
t
a'k c
it, and the criticism involved. We can thus concur with Clement, who
states:
'As always in myths there are two opposing dangers: the
danger of too much and the danger of not enough, the danger of excess
and the danger of lack.' (1983, p.135.)
There is thus the opposition Up/Down as a correlate of this.
The first awakening has the single image of planes in the air, with
the accompanying idea of "international relations" an area of great
responsibility. The sloping lake and its position next to hills
emphasises the transition from up to down, and then one of the people
264
they are travelling with mentions her holidays near the Cape Cod
airforce base.
This night of dreams gives us an example of the superiority
of a structural, as opposed to a solely functional (problem-solving)
evaluation of dreams, as put forward by Evans. What possible use to
KJ's waking life could be the scene of him flying through the air?
Such an action can solve no waking problems in a realistic way, but it
fits in well with the oppositions high:low and high:descending.
Similarly, the anthropologist Firth found that 'As far as the majority
of animal totem species is concerned the economic interest in them is
not of a pronounced type' (1930-31, p.297), and that animals and
vegetables which are important to the society are not always accorded
equal importance in the myths and rituals of the society. In the same
way, Levi-Strauss tabulates different foods of the Tikopia with regard
to the labour needed to grow them, the complexity of the ritual
intended to make them flourish, the complexity of the harvest rites,
and the religious importance of the clans which control each kind of
food. He concludes that the
'table does not correspond with the
totemic system, since the number of plants in it is greater; the yam,
which is controlled by the highest clan, and the ritual of which, both
for its cultivation and for its harvest, is also the most complex,
occupies the last place in importance as food and the second in labour
demanded. The "non-totemic" banana tree and sago palm are objects of
more important ritual, both to raise them and to gather their fruits,
than are the breadfruit tree and the coconut palm, both of which are
nevertheless "totemic"...' (1969, p.136). He notes (ibid. p.96) that
for the Tikopia it is the inedible fish, insects and reptiles that are
associated with supernatural beings; Firth suggests that this is
265
because 'creatures which are unfit for human consumption are not of
the normal order of nature...', and hence can be used in the
classification of supernatural elements. They are used not because of
any function they may have for the tribe's subsistence, but rather for
the function of classification.
In addition to the High/Low opposition there is the
distinction between Young/Old, which gives the following correlations:
Criticised:Admired :: Old:Young
and Ribbed:Admired :: Peers:Young
which is more in keeping with the flying scene (when KJ
is high, not criticised, and carefree) than the previous two equations
I produced. Overall, his environment, both physical and social,
changes thus during the night:
Up
Down
Old
Young
The notion of 'the wrong place' occurs early in the night.
He is in the wrong place to meet a young woman, he worries he is at
the wrong college, there isn't a good place to park in the cul-de-sac,
and he is advised not to travel along the dirt roads, nor across the
lake (by Connie, who thus shows concern and responsibility). Later, a
'long-haired man' who was not the first at a parking space is
nevertheless the first person to park there, yet then parked in the
266
wrong place, not between the lines. KJ himself is accused by old
people of possibly ending up not wanting to go back to college. (Note
that in the dreams of the first night he was in the wrong place in the
auditorium, and when not able to eat with the lab technicians.) The
idea of 'the wrong place' has been translated quite literally into
dream images. Professor Alan Stone (1986) has noted the incestuous
imagery connected with his mother and the cul-de-sac which is 'not a
good place to park', and Adam Kuper has commented that the very scene
is typical of American adolescent dating. There are also homosexual
images across the nights. All of this would be ignored in a
structuralist analysis which concentrates upon the formal oppositions
of the dreams, and would lack emphasis on such deep emotional meanings
behind the dream, which would need the subject's free-associations in
order to be explored (this is a common criticism of the structuralist
technique). The lack of associations with KJ's past also makes it
difficult to evaluate Palombo's hypothesis. Phenomenological methods
which would bear on this are discussed in the postscript to the
thesis.
As well as people changing place there are also the
transformations of Bob into Craz and of Nancy into a man which took
place in the second scene of the dream reported on the 4th awakening.
These are meaningless condensations according to Crick and Mitchison,
and so we need to discover if there is any importance attached to
these sex differences. The only people friendly to him are males,
although some are critical. Females are either a paragon of academic
excellence, critical (the wives, and Connie saying he can't travel
across the water) or, in the last scene, irresponsible. In all cases
267
they are extreme, unlike the homely men he meets. It may be claimed,
therefore, that there is some message in the dreams behind these sex
differences.
The initial opposition of responsibility and complaints may
be made more abstract by relating it to the opposition Giving/Taking,
which is certainly indicated as being of importance by the number of
transactions between people in the different scenes. For example, in
the penultimate scene KJ has an all too brief exchange of pleasantries
with Craz, and is then annoyed that they then only talk of money and a
cheque.
There is also the opposition of Inside town/Outside town;
while in the first night there was Inside lab/Outside in Chicago, and
we will see that on the last night he is in town all along, seemingly
working out the implications of the workplace's confinement rather
than wanting to be elsewhere.
According to the structuralist theory, other reversals which
permute these oppositions through the night are:
1) the change from old, overconcerned mother in the initial scenes to
young, unconcerned mother in the final scene.
2) on narrow roads and hills outside town he is in a car, whereas on
narrow roads and hills inside town he is on a motorcycle. This
juxtaposes outside/inside with constrained/free.
3) the change from responsible married people in the initial scenes to
less responsible younger people in later scenes, some of whom change
sex.
268
It is possible to categorise these actors in the scenes thus:
Overappropriate / Appropriate / Inappropriate
(In place)
(Out of place)
His mother
Colleen
Craz
Old people
Bob
Young mother
Kids
KJ
KJ (finally)
As in the fourth night there is the aspect of looking
for/out, and the opposition inside/outside. He's in the wrong place
because of what he wants to look for. According to the next night's
interview he had been considering leaving Chicago (in the dream
"Connie was saying 'no, it's a physical impossibility'"). Unlike the
last night this night has the aspect of not wanting to stay, whereas
in night 4 it's being inside that's the problem, although he resolves
that by judging it to be safer. On night 3 he sees the need for
responsibility, and on night 4 the need to be within the institution.
These changes are no doubt due to his waking thoughts, though, rather
than an effect of dream mentation.
We can summarise the remaining progressions as follows:
1) During the night there is a change from having a drink
outside townto cold buffet food, which is probably stolen
to investing in selling uncooked food
to buying dope.
269
This is a declining standard of food, progressing from
cultural to natural, an opposition that Levi-Strauss considers
important in myths. He considers the opposition so important that he
merits Rousseau as the first ethnologist, by virtue of his attempts to
find a basis in nature for culture (Structural Anthropology, vol. 2,
p.34.). This change in the nature of food is partly connected with
the opposition Give/Take, from stolen food he moves to investing in
selling food to buying illegal 'food'. Such an obviously intellectual
concern with the naturalness of 'food' is typically a feature of LeviStrauss' analyses. The dope is only a 'food' metaphorically speaking,
and yet attention is drawn to it as part of a possible sequence
because of the illicit nature of the previous real foods mentioned.
2) There is the following decrease in constraints as the
night goes on:
Respectable married
Carefree peers
couples (friends
Mother's children, he
mentions they are all
of mother)
retarded and dirty
(The dirty children are possibly a reference to KJ's comment in the
pre-sleep interview that due to the rain his clothes had stained and
the colours had run.)
Such a progression easily fits in with a simple Freudian
wish-fulfilment model.
3) The lack of constraint with regard to he position of the
cars during the end of the night compared to earlier:
270
Small parking space
'buzzed along
parking at intersection
at end of the
in the car'
with many other cars.
It is someone else who is
cul-de-sac
not between the lines.
This progression accords with Evans' notion of the
increasingly successful practising of programs.
But we can also see these progressions and permutations
occurring within scenes. For example, in the second REM awakening he
reports:
He is with mother, there is sexual symbolism (i.e. the culde-sac, and that cars are used for teenage courting.)
He doesn't want to know old people
He searches for a woman but there's only old people
He notes smart Colleen.
(I consider that these scenes fit wish-fulfilment accounts
as much as analysis in terms of structural oppositions.)
Similarly, in the third awakening, there are three scenes, each of
descending with other people:
Descending & being ribbed, reference to airforce
Descending as leader of flying group.
Descending steps of the bank.
(In the first two cases they are travelling over long
distances, without being quite sure of the route.)
271
So, we have found oppositions and progressions within
awakenings, as well as between them. This is important, for without
the former this would only be evidence of progressions between one
dream report to the next dream, rather than from scene to scene.
Remember that the subject has to be awakened to tell us of the dream,
and progressions from report to report could thus be an experimental
artefact. With the results of between awakening permutations, we may
have introduced a new day-residue, that is, the previous report, into
each subsequent dream, whereas we see now that within one awakening we
find a progression of scenes, so it may be possible to extrapolate
from this to claim that between non-awakened dreams there is also a
progression. Still, at the very least it may be claimed that this work
shows the permuting of oppositions from a dream report by the next
dream. That the permutations advance to a resolution, and that they
are based on the waking life of KJ is also found. However, on this
point, a complete sceptic about dream research may claim that KJ had
such problem-solving, waking-life reflecting dreams because the
experimental set-up (of pre-sleep interview and post-sleep study of
the connections between the dreams and waking-life) led him to believe
in this type of dream. In the same way, analysands having Freudian
analysis have Freudian dreams, and those with Jungian analysts dream
of archetypes. I answer to this that it is proven that dreams can
incorporate pieces of knowledge from waking-life, such as knowledge of
what the analyst wants to hear involving Freudian symbols, but that
this incorporation of conscious contents into the dream process is
quite different from the subject changing the form of the whole
night's dreams by deliberate and conscious intention. (Haskell, 1986
272
makes the similar point that 'expectations and demand characteristics
should not be an influence upon structure. It may be assumed that
logical structure is perhaps wired-in neurologically ... and/or that
it is relatively impervious to experience...') The evidence appears
that the form of dreams shown here, of progression of one dream to the
next (and one scene to the next) with the use of dialectical arguments
and logical permutations of oppositions, is caused by the unconscious
processing of the mind, although the prospective contents of the
dreams are known to a great extent by the conscious mind.
273
CHAPTER 17
NIGHT FOUR: DREAMS AND ANALYSIS
In his pre-sleep interview KJ complained about not having
done enough work for a class the next day, and that there were
conflicts with a colleague, Benton - 'I have to do exactly what he
wants and it's very hard to fit into his structure of things'. He was
also finding it hard to work at home each weekend. He has decided to
remain in Chicago, which Connie agrees with.
1st. Awakening.
No recall.
2nd. Awakening, 1.3Oam, 5 mm. into REM.
He had just ordered a pizza at a grocery store or
restaurant, and then went to a magazine rack to look at the new
Playboy magazine. He ordered that issue, and the man went off to wrap
it, but then the place became Like a çiizza arZar &ad the aar weat Cc
put it in the oven. While the pizza was cooking KJ realized that it
was an issue that he had already seen, and didn't want it. So he went
up to the man, who had gray hair and was old, but he was busy and
didn't help. KJ asked him to pull the pizza out of the oven, but the
man was playing with a doll's face that was made out of soft
styrofoam; he was cutting the face with a sharp point and KJ realized
that if he was not careful the little doll's head would be ripped.
Then he began taking care of someone else's pizza putting something
on it. In the meantime a young boy, maybe eight years old, was serving
someone else. KJ was still waiting for the older person and asked him
if he could take the pizza out of the oven. He said no, that it was
274
too late, that 'you have to have it done all the way, or not done at
all', for he can't 'do it half-way'. So KJ was disappointed and
resigned that he would have to eat it anyway, and would then wait
until next month, when the next issue of Playboy came out. He walked
over to tell Connie that he couldn't take out the pizza, and that they
would have a different kind next month.
Upon questioning he stated that there was a distortion in
'the magazine switching from Playboy to a pizza', and that the six
year old looked older than that, 'he looked like a ten year old, but
he said he was six years old'.
He then reported these other two segments:
The experimenter was coming into the room with some long
scissors to fix some of the wires, but KJ was curious and confused as
to what they would use scissors for. He played with them in his right
hand, but they were very hard to open and close.
He then added a segment which preceded the pizza sequence:
He was in an apartment house, living on the third floor. Although this
was the top floor somehow there was someone living above him. He
didn't like the man who lived upstairs, because somehow he was
annoying KJ. At one point KJ was out on the back porch, clearing off
all the snow into a hole in the floor to get rid of it all. He had a
sense of clearing it off and making it clean. Then, there was an
airplane flying just under a cover of clouds, and 'the cover of clouds
represented this other guy'. The airplane was trying to go as high as
it could, up to the bottom edge of these clouds, and KJ could hear
some hard scraping sounds of the top of the wing against the bottom of
275
the clouds. He felt good that the plane was taking off the bottom edge
of the clouds; Connie came out onto the back porch and was not pleased
with what was happening.
3rd. REM Awakening, 4.35am.
He was standing around outside on the grass playing catch
with a football with some others, some of whom were opponents. He had
just arrived on the scene, but after one or two balls was ready to go.
He then heard an ambulance siren in the background and was worried if
the police were coming. He did not want to take the risk of staying
and so started packing up his stuff into a bag in order to get on a
bike and go, but the straps were tangled up into the front wheel.
4th REM Awakening, 5.54am.
KJ is standing in an unfurnished apartment which has many
windows. He is looking out and is with his brother and mother. There
are several windows close together on the right, but there's only one
window over to the left. He points out to his mother that the trees in
the left hand window were now growing and somehow were different than
they had been, for they were alive and growing. His mother was
sceptical of that, so he watched those trees for awhile, for a few
minutes, and pointed out to her how in fact the tree was alive, and
changing, how it had gradually started moving and growing. She saw his
point. They then went down the steps and were outside on the grass,
where they were with his brother and Connie; there was an exchange of
gifts at this time. He gave his brother a sealed nvelope and KJ was
pleased to see him opening it. KJ had never seen him open a birthday
card before, and he calls it a pleasant scene. But his brother then
added up what he had received, and gave them a ten dollar bill in
276
exchange. KJ was returning to the apartment, and at the doorway that
was on the ground floor there was an envelope, and there was another
birthday card, and he opened it up. This gave him some kind of
reaffirmation of being a good person,
KJ is going to the third floor of his apartment but finds
that it's not his apartment, because it's laid out differently.
There's a woman there, who smiles at him very pleasantly, and says
that she's waiting for the lady inside to let her in, and yet the
doorway is open. KJ found it strange that she should be standing
there and waiting when the door is open. She explained to him that she
was going in to see this lady's beautiful furniture. He then decided
to look at the furniture also, but found that most of the place was
junk, with lots of mattresses. There were lots of people in there,
like a big party, but he doesn't know them. He was following around
the lady of the house and noticed a young 'weird' guy who has 'got a
lot of kids', and was mentioning the triple mattress. KJ leaves,
possibly with the 'weird guy', and sees a picture of him. Next, he was
on his motorcycle and the same fellow was walking down the street, he
thinks that this guy was responsible for the wires being off his ears
in the lab.
KJ then gets on his bike and does a U-turn. He is driving up
a hill, on a motorcycle, in fairly heavy traffic. His bike and a
'little red car' with a woman in it both overturn. He tries to fix the
bike himself. His friend Steve starts talking about how the guy was
weird, and then complains about how difficult it is to take photos of
KJ.
Lastly, KJ is in a V.A. cafeteria located down on a boat
277
dock, like a pier for fishing. He is with friends and they go into an
old, long, flat building and he feels uncomfortable. There is a
cafeteria with different frozen foods and cameras behind glass. Then
they obtain some strange food from a can. They walk all the way back
down to the end of this long cafeteria, where there are some
strangers. Then he sees a submarine and a ship at sea. The ship is
long, and the submarine is in it's hold, The scene changes to the sub
being way down in the water and the ship finally getting the hold door
open by opening up in half, and trying to take in the submarine. It
was as if it had a place for storing it. The ship missed the first
time, and so it had to do it again. The method of opening was like
that of a transport plane.
5th. REM Awakening, 6.5Oam.
He is outside in the sun with some young people.
In the next scene he sees a train which is running on the
ground level; it's not elevated but has tracks in the middle of the
road. 'There was this little automated vehicle running on the tracks,
that was very peculiar. It had a tiny two wheel thing in the front,
that didn't have any depth or height, it was as if a Ford with two
wheels. And that had attached to it kind of a trailing little car,
with a little compartment on it, it wasn't a car, it was some kind of
a wagon.' He pulls out a pear and cleans it off on his shirt. He is
then walking up the street, and sees some young Spanish kids. He is a
little alarmed at seeing them, uncertain of what they're going to do,
although they are very young kids. KJ stands up against the buildings,
in kind of a passageway, waiting for them to go by, and the
surroundings become a living room; he starts talking to people in
278
there. He doesn't manage to finish with them, but goes out to the
street, back out to the sidewalk, to see where these other kids were,
and they weren't in sight. He had expected to see them pass by, but
somehow they had disappeared. He then walks through an old long
hallway in an old industrial factory, he enters a long twisting
hallway which was somehow a short cut out to the alleyway where he
parked his motorcycle earlier. At the exit door the motorcycle is in a
different position from where they parked it. Some of the instruments
and the handle-bars are damaged, but he can fix them.
Next, he was sitting around with others in an apartment. His
professor was looking for an industrial job for him, about which he
was not pleased. A stranger was there who one person said lived in an
apartment in Chicago, while one person said he knew him across the
street, and that he's a student there. KJ then comforts a secretary
who is very emotional and who then becomes an older woman. Steve then
starts talking to KJ, who asks him not to interrupt. Steve then asks
KJ a question that puts him on the defensive. KJ finishes with the
secretary and for a time is centre of the conversation.
Morning Interview:
The experimenter says that he did have to enter the room
before KJ's dream about the scissors. KJ continues:
'As I was recounting that to you I was thinking about all
the Freudian implications. Cutting off my penis or castration or
something like that.'
About the dream with the snow and the airplane KJ said that
'it made sense to me that I didn't like him, but there were more
279
things that happened in the dream because I knew I felt okay about not
liking and it made sense to me because he did or he was something. I
was very grateful when that plane did his thing with the top wings.
There was a very large scratching sound like he was scraping off the
bottom layer of this guy that is represented by a cloud. It was kind
of like a narrow space in the sky. The wings were like on top of the
plane. I'm having problems dealing with Benton these days and so the
first thing that comes to my mind is that it could be him...'
When asked to relate the dreams to his pre-sleep interview
he stated that the anxious parts of the dreams reminded him of how he
functions best with some amount of anxiety in waking life. The
experimenter commented that he was resolving problems in the dreams
better as the night went on.
280
ANALYSIS
I have found three condensations during this night.
1) The grocery store that becomes a pizza parlor.
2) The Playboy that is a pizza.
3) The cloud that is somehow a man whom KJ does not like.
His concerns were 'finding all these things to do instead
[of his work]', deciding to remain in Chicago, and only having cleaned
up his desk when he had a paper to do instead. In the first scene
there is an allusion to the problem of whether he is assertive or not:
similarly, the pizza, constrained by the oven, is also a 'Playboy',
albeit one that is old and out of date. The sexual significance of
this is further underlined by the juxtaposition which then occurs:
Old Pizza-guy cuts a dolls face,
followed by
The technician coming in to cit the. ire. attache.d to J's xead.
KJ mentions at interview the worries about castration that
this latter incident caused him.
In the interview before the first night he commented that in
becoming a clinical psychologist 'I feel that my head is being
shaped'. Even participating in this experiment of dream recall must
have been for him a further example of being 'shaped', of getting
deeper into psychology. (Note that 'shaping' is a technical term in
psychology meaning to reinforce behaviours which progressively
approximate to the experimenter's desired goal.) Each of the first
three scenes involve a sharp object, with IU being passive; each of
281
the sharp objects is harming something - the plane's pointed wings, a
pointed object cutting a face which could result in the head being
ripped, and lastly the scissors. Simultaneously, there is confinement,
first caused by the clouds, then by the oven and the unhelpful owner,
and lastly by being on a bed in a room. Maybe the notion of
confinement can explain why the scissors are not easy to open, an
otherwise arbitrary and pointless inclusion. Connie complains about
him in both scenes 1 and 2, further underlining his lack of power in
the face of the confinements. The pizza scene fits in with Evans'
view, whereas the scene with the scraping plane wing accords with
Freud's wish-fulfilment theory and with Rycroft.
In the structuralist account, the initial scene contains the
first reversal in the dream series: he starts at the bottom of the
tenement clearing away a white substance (snow) downwards; next a
plane above the tenement is scraping at a white substance (clouds)
above it. The cloud cover is constraining the plane while earlier the
snow is constrained by the hole in the ground and by the broom. We
note also that the previous day he had been 'clearing away' work on
his desk, enabling him to 'find stuff I was looking for', whereas now
such clearing away more resembles confinement.
Scenes 4-7 depict him not wanting to stay, not fitting in.
Four reversals occur:
Outside, his brother opens a card,
is followed by
Inside a doorway KJ opens a card.
(note that the third night showed a preoccupation with giving and
receiving.)
282
KJ doesn't fit into the photos but fits in at the lady's flat
despite weird furnishings,
is followed by
A strange guy is in a photo but doesn't fit in at the flat (for he
is weird).
KJ wants to leave the field but is prevented physically,
is followed by
A lady is outside the room, wanting to enter, although the door is
open (i.e. she is not prevented physically)
He is among strangers in a room where furniture is being sold,
is followed by
He is with colleagues in a furnished apartment (final scene).
In scene 5 he sees a group of trees outside that are growing
and changing, he also has a pleasant time outside. We must bear this
in mind for when the outside world is represented as dangerous in the
penultimate scene.
Sexual overtones are present in the fourth awakening, scene
6. He first meets a pleasant woman outside the apartment, she wants to
look at old mattresses and cribs inside; next there is a 'weird' man
inside who has many children and talks of triple mattresses: however,
KJ wonders if he himself is the weird guy. His motorcycle then crashes
when he does a U-turn and a woman's car overturns. Somehow, LU says,
the weird guy was responsible for the electrode being off, which
implies that KJ is himself responsible for this, which he links with
his own castration.
283
As well as the apartment with lots of old furniture being
sold there is the apartment which is empty but that has windows,
through which he can see growing trees (although his mother is
initially sceptical of this). That there is then a family scene of the
giving of birthday cards shows the dichotomy in his life of home
versus work, the choice between which he mentioned in the first night
as a source of friction between him and Connie.
Scenes 8 and 9 show large containers, scene 10 has an
idyllic outside setting and with scenes 11 and 12 KJ is safe, moving
around but inside buildings. At first glance the change in his
fortunes over the night would fit in with any theory of dreams
involving problem-solving or wish-fulfilment, as shown by the
comparison of the initial scenes with the last ones:
LATER SCENES
EARLIER SCENES
Anger at man upstairs.
With colleagues in apartment
Confusion at what experimenter is
Deals with Steve and with the
doing and non-assertive towards
emotional secretary, and gets
pizza-seller.
a job.
In pizza-parlour, misjudges the
He knows how to avoid the kids.
age of the man's kid assistant.
He doesn't know what the scissors
He can fix the motorbike.
are for.
Certainly this successful change accords with Evans, Rycroft
and the structuralist theory, and with Freud in that there is an
undisguised wish-fulfilment.
284
This change is related to his statement in the post-sleep
interview that the dreams show him becoming more at peace with the
department, and not fighting the confinements there, which is just
like his own history at the place. However, he also says that the last
dream is 'negative' and has more anxious elements, and likewise there
is certainly no progression in the attributes of the vehicles in the
sense of a continual improvement:
Awakening 1) he condones what an independent plane is doing
2) no vehicles
3) his bicycle is stationary with straps tangled to it
4)
he has an accident on the motorbike, and the ship
has to attempt twice to recapture the submarine
5)
he knows how to mend the motorbike and sees an
automatic vehicle on tracks.
There is, however, a logical progression of the vehicles
when considered in terms of 'confinement', and this notion encompasses
most of the other aspects of the night's scenes, and also does not
leave us having to ignore the above six reversals as inexplicable or
chance events. The progression involves the permuting and final
resolution of physical oppositions during the night, as a concrete
picture of the resolution of his own personal conflict, that of hating
the confines of the department, as epitomised by his relations with
Benton. The physical oppositions are Up/Down, Inside/Outside, and,
less importantly, Sharp/Blunt and Left/Right.
285
plane
UP
apartment, 3rd or 4th floor
uphill on motorcycle
train at ground level
ship
submarine
DOWN
hole
strange apartment
one window
many windows
strange apartment
scissors held to ear
LEFT
RIGHT
Furthermore, during the night he produced a series of
confining structures:
Cloud cover - oven - hole - field - straps - photo - apartment
can for food - ship's hold - train's compartment - apartment
which are all relevant to the Inside/Outside opposition.
Some of the containers are themselves contained:
holds
holds
long ship --------submarine
long pie
long cafe ---------cans & cameras *
holds
holds
286
(* which hold food and photos, which each occur elsewhere in the
night.)
These containers are at various levels, the plane and the
submarine both attempt escape from an object at a different altitude,
but such distinctions of height are not present in the final scenes.
The penultimate scene resolves the oppositions Up/Down and
Inside/Outside by involving an automated vehicle 'that didn't have any
depth or height', with a 'trailing little car with a little
compartment on it' running on tracks. (Note that tracks confine that
which moves outside.) There is thus an increase in movement as the
scenes progress, much of it within containers or confining places,
until we end with the moving container.
plane that can't escape
container on tracks
DOWN
I
submarine that almost escapes
'V
TIME
There is evidence that there are two personal oppositions
that are being resolved simultaneously with the resolution of the
physical oppositions: these are Being contained/Freely moving, and
Fighting back/Accepting the situation. According to the structuralist
account the scenes can be seen to progress as an argument, a temporary
resolution being found at each step until we reach the final scene.
Just before the final scene he shows no desire to be outside, for in
287
order to escape the freely moving kids outside KJ enters a passageway
that becomes a living room, and the advantage of being inside is
recognised.
He hates confinement vs. Strength of containers
/
He is forced to fit in vs. He sees growth outside
He explores outside vs. Danger outside
& sees compartment on
tracks outside
Acceptance with colleagues in
apartment, offer of job.
We note that the keys to the transformation of abstract
concepts to the concrete environment depicted in this dialectical
argument are:
Above:Below :: Aggressive:Submissive
and
Inside:Outside :: Constriction:Growth
The concrete elements depict the abstract concepts (as
Rycroft emphasises) and change stepwise in the following way towards
restatements of the initial problem, and a final resolution (in
accordance with the claims of structuralism):
288
Initial Opposition Second 0. Third 0. Mediator Fourth 0. Mediat.
Confinement &
frus trat ion
Clouds
Ship
Safety
Train
Apartment
Danger
Sub
Plane
Freedom and growth
Both the concrete oppositions and the abstract oppositions
resolve themselves simultaneously. He is finally neither aggressive
like the plane nor submissive like the snow and, possibly, the
submarine. Pe is confinec 'with 1nis o1'ages a'n
s t\i
'nt
attention. He remarks in the post-sleep interview that he started off
feeling frustrated at the department when he arrived there but is more
'detached' and accepting of it now. KJ also says that this gradual
acceptance occurs daily in that, 'as the day goes on I rejoice about
the good things and forget the bad'. The night's progression thus
simultaneously mimics both his long- and short-term history at the
department, which fits in with Freud's concept of overdetermination.
Maybe this change in the perception of authority is depicted in the
change from sharp shaping instruments to the blunter ship and straps,
and to the final acceptance and being 'on the tracks'. Such a final
picture is reminiscent of the common phrase of being 'on the rails',
289
or 'on the right track'.
It is also necessary to note the changes in the opposition
Young/Old, because this has occurred previously on nights 2 and 3. KJ
misjudges the age of the kid assistant in the pizza parlour, and in
the last scene a secretary appears older when she becomes emotional.
This opposition is not as important here as in the previous nights,
though.
In the last scene there appears a man who either lives in an
apartment across the street or lives outside Chicago. (KJ mentions
that at one point '. . . the new guy came in, and mentioned the other
peoples' apartment'.) This is relevant to KJ's situation. In real
life they have decided to remain in Chicago, just as earlier in the
night KJ has to tell Connie that the pizza can't be taken out of the
oven. Note that in the first night Connie wanted him to go out (i.e.
away from work) to an Italian restaurant, and here he finds that the
pizza is stuck in the oven!
This use of concrete imagery simultaneously with the
abstract concepts may be compared to the following summary of part of
Wittgenstein's work 'On Certainty', cited by Hunt (1986), which also
illustrates Levi-Strauss' contention that in the comparison of
oppositions, there is no privileged perspective:
'Wittgenstein emphasises the immense difficulties and
relativities of real observation: it is never pure but always more or
less influenced by the various pictures we bring along with us. We
know something only in relation to something else. In ordinary
thinking and science we treat everything as fixed and finished when it
never is in order to use it as a perspective or view for the thing we
290
wish to question. Contrary to widespread assumption, the metaphoric
vehicle is not necessarily better known than its referent.'
I conclude that the texts of this night again show the
stepwise progression towards a wish-fulfilling solution, and also the
comparing of existential oppositions to physical oppositions as a
means of resolution. On the fourth night that KJ had his dreams
recorded in the lab a much greater variety of content occurred than in
the previous three nights. This diversity belied the fact that a
logical progression was present in the series, just as in the others,
resulting in a resolution of his main concern that night of 'having to
do exactly what [Benton] wants and it's very hard to fit into his
structure of things'. Similarly, de Saint-Denys noted that for his
dreams 'At first sight nothing could be more incoherent than this
series of images. Yet nothing could be more logically connected, once
one has grasped the link between the. tcLe.as .' (1'2, .0 .
291
CHAPTER 18
COMMENTARY ON THE METHODOLOGY OF THIS STUDY
In this chapter I wish to compare the epistemology of the
structuralist project to that of psychoanalysis, and also to justify
the application of the structuralist method to dream texts. Problems
with the data collection methodology are explored and then the results
of a replication of the analysis of the fourth night by five
independent judges are described.
The Falsifiability of Structuralist and Psychoanalytic Analyses
A comparison between these two theories is useful because of
the various warnings that a critical and mischievous history of
psychoanalysis can provide for us. For example, Timpanaro (1976, p.78)
writes:
'One patient told Groddeck that he had read a book in which
mention was made of the sea. Groddeck's comment is: "The patient has
an unresolved Oedipus complex, and he is a fool not to have realised
in fact that by means of the association Meer = la mere he was
speaking of himself" (Carteggio Freud-Groddeck, Milan, 1943, p.69).
One should note that the patient was German: his unconscious,
nonetheless, made its confession in French!'
Similarly, E. Jones (1958), in his biography of Freud,
relates a story about one of the earliest analysts, Stekel. Stekel
eventually left the movement having shown a marked lack of
intellectual conscience and honesty, after earlier being acclaimed for
his intuitive grasp of symbolism. Jones writes that:
292
'In a paper [Stekel] wrote on the psychological significance
people's surnames have for them, even in the choice of career and
other interests, he cited a huge number of patients whose names had
profoundly influenced their lives. When Freud asked him how he could
bring himself to publish the names of so many of his patients he
answered with a reassuring smile: "They are all made up", a fact which
somewhat detracted from the evidential value of the material.
(1958, vol.2, p.153.)
Wittgenstein made a criticism of psychoanalytic
interpretation along the same lines when he remarked that, using the
'logic' of free-association, one can start with any of a collection of
objects on a table and find that they are all connected in a pattern.
(In: Wollheim and Hopkins [eds.], 1982, p.8.) Freud himself mentioned
this problem in 'The Psychopathology of Everyday Life' (SE, vol.VI,
1953, p.250 n.2) , concerning the evidence of experiments with
associations from numbers. Schneider had found that he and his
subjects were able to produce associations to numbers that were chosen
at random; the associations gave the appearance of having determined
the choice of the presented number. Freud accepted that these results
could apply equally to word association, but concluded that
experiments on presented items give us no knowledge about the cause of
words which arise spontaneously.
Another philosopher, Ernest Geliner, notes that
psychoanalysis remains to a great extent untestable because not only
the unconscious processes but also the very cure are defined and
interpreted by the analyst (1985, p.176). Even the initial reasons for
entering analysis can be re-interpreted over time.
Freud claimed to have found a message in neurotic symptoms,
293
dreams, and parapraxes, but there is the problem of proving this in
that the external sign and the interpretation of the internal mental
state can be unfairly manipulated by the analyst as part of any
'proof'. S. Timpanaro (1976) has a different objection to the notion
of psychoanalytic interpretation: he writes that even Freudian
explanations for slips of the tongue (even for most critics an
acceptable part of the Freudian corpus) cannot be checked. He accepts
that pre-conscious thoughts and concerns may intrude into one's
speech, but uses his knowledge of the degeneracy of written texts
through the centuries to show other reasons for the elision of letters
and whole words, and for the mistaking of one word for another. Most
important among these effects is 'banalisation', that is, '...the
substitution of one word by another whose meaning is actually or
apparently the same, but whose usage is more familiar to the copyist'
(ibid. p.21).
As an example he gives the case of the omission of the word
'aliquis' (Latin - someone) in a Latin quotation by a young man whom
Freud met on a train.
Freud proceeded to gather free-associations
from the man (not so 'free' complains Timpanaro) which resulted in him
finding the 'cause' of the slip, that the man was worried about his
girlfriend's possible pregnancy. The story is written in full at the
beginning of Freud's 'Psychopathology of Everyday Life'. Timpanaro
notes, however, that not only is the meaning of the quotation barely
changed by the omission, but that the phrase is then more in accord
with Latin grammar. He then shows that all of the six words in the
phrase can be similarly connected with the notion of pregnancy, just
as 'aliquis' was associated to fluid and the abortion via the name of
294
a murdered child-saint. For example, 'exoriare' means 'arising' in the
passage, but can also mean 'birth' (ibid. p.44). In fact, Freud does
note that one word was emphasised in the quote ('exoriare'), and
derives two routes by which this is connected with the concern over
the pregnancy. Timpanaro comments that (ibid. p.47):
'The curious thing is that Freud sees this profusion of
competing explanations as confirmation of his method's validity,
without ever asking himself whether this superabundance, this
unlimited supply of explanations might not be an indication of the
weakness of his construction. ... a particular explanation that claims
to be scientific, must not be such as to elude all forms of controls.'
Jung (1968) makes the similar point that 'if you are riding
in a Hungarian or Russian train and look at the strange signs in the
strange language, you can associate all your complexes. You only have
to let yourself go and you naturally drift into your complexes.'
Similarly, Piaget (1971) points out that it is sometimes
difficult to know whether a structure is imposed on the content of a
text or is there already.
Timpanaro then likens the dubious connections that Freud
made between a known slip and a known mental concern to medieval
etymology (ibid. p.77). An example of this was 'a wood (lucus) is so
called because it gives no light (lux).' This free use of negation in
constructing a connection resolved any problems recalcitrant to the
method of subtracting or adding letters, and 'even in the eighteenth
century critics spoke ironically, and rightly so, of this omnipotence
(and thus immunity to proof or disproof) of etymology...' (ibid.).
Freud wrote an essay called 'Negation' in 1915 which left him with a
295
similar power.
His argument also applies to the work of Palombo in so far
as the latter deems some dream events to stand for real memories, and
hence needs free-association to retrieve these real memories. It is
the argument of this thesis that it is the structure of the dream,
rather than its content, that is usually determined by the
unconscious, and little use of KJ's associations is made, save for the
clarification and definition of the actual dream images and scenes.
Note, however, that we do need to know of KJ's general waking
concerns, just as the application of the method to ethnographic data
does require some knowledge of what the content items mean to the
society in question, for we must know what is the relevant opposition
for each item, for example, what does an eagle mean for this culture
in this context, that it is high or that it is not a seavengar?
Freudian interpretation of slips and associations is often
criticised for its unrepeatability, and hence its falling out of the
domain of science, On the contrary, structural analysis, and work on
symbolism, relies on 'bundles' of the particular relation in question.
As has been shown, the dominant representations of KJ's main concerns
for each night are repeated in dream after dream. Similarly, Hall
(1966) advocated interpreting one dream with the aid of others in
order to find why a particular symbol was used; Cartwright (1986)
found a similar repetition in the dreams of divorcees - one had
different aggressive men in various situations. Scientific rigour
would, however, allow an inference to be drawn from one instance of a
representation if we could be certain about the cause of the symbol,
that is, if we knew what the waking stimulus was with certainty. Such
exactness was cited by Eysenck (1986, p.131) from an experiment by
296
Luria. Luria read frightening scenes to hypnotised subjects who were
told to forget the scene upon waking. They later went to sleep but
were frequently found to dream of a similar scene, but with
alterations by means of symbolism, e.g. a rapist was changed to
someone who was snatching a handbag. (The question of whether the
interpretation of a dream gives us any clue as to its means of
production was tackled by this experiment, but is often ignored, for
example, Hermans (1987) relates the dreamer's waking life 'valuations'
to the 'valuations' of a dream, but only after the subject has chosen
the dream on the basis of its relevance anyway.)
Lagrand justifies the vagueness of psychoanalysis, and its
recalcitrance to falsification, as being signs of a science in its
youth. Timpanaro notes that this is an excuse for a gross
individualisation for all psychic occurrences, such that
generalisations cannot be made. He says this leads to a tendency 'to
self-conservation of the theories rather than their creative
advancement' (1976, p.216). Such problems are present to a much lesser
degree in my work because direct and repeated representations of
concepts are found, rather than individualised single events. (It is
true, however, that the dream as a whole is a single event, and that
it is not possible to draw general conclusions from one dream, or from
one subject. However, this individual level of analysis is the only
way to discover the reason for dreams, and needs to be repeated for
many dreams and many subjects, rather than altered.)
I suggest that the structuralist analysis of dreams does not
have these problem with falsification that psychoanalysis does because
both the initial problem and the state of the mind are both to some
297
extent conscious and public, in that both are common-sense texts
supplied by the subject.
Psycho analys is
Waking behaviour
symptoms,
Structuralism
dream text
dream reports
is related to
Structure of Thought
psychodynamics,
problems,
instincts
conceptual
oppositions
of the subject
Gellner (ibid.) mentions the vagueness in psychoanalytic
explanations: ' . . . in one very important sense, the point about
psychoanalytical depth inquiry into individual minds is not that it is
too rigid but, on the contrary, that in a special sense it is
extremely and indeed excessively open and corrigible. Roughly
speaking, psychoanalysis maintains that all surface data are suspect
and unreliable, and many or indeed most depth data (i.e. analytically
secured data) are also suspect.' Crews (1986) makes the similar point
about speculative Grand Theory in the human sciences, by stating 'Of
course such bogus experiments succeed every time. All they prove is
that any thematic stencil will make its own pattern stand out'.
Such cannot be said of the work on KJ's dreams, in that only
a couple of problems are noted each night, and so the question of how
many concrete oppositions exist in the dream is not 'excessively
open'. Any problem with falsification must therefore only arise in the
relating of day-problems to concrete oppositional images. I have
managed to relate KJ's most obvious problems of the day to the most
298
obvious oppositions in the dreams, even though finding the latter can
take weeks of work. The dynamics of the oppositions are considerably
more simple than the host of hydraulic activities that psychodynamics
is said to engage in, although this simplicity is picked upon by
Jenkins (1979) in his scorning of the idea of a structuralist
unconscious as being vague, contentless and trivial.
Apart from this simplicity of the structuralist project
there is the essential superiority that the dream text has over
symptoms, and even over the texts of myths. This consists in that the
latter two products change with each telling, for they are observed by
the conscious subject as they change over time (for example, the
progressions of the hysterical symptoms of Anna 0. after their initial
occurrence). The reports of dreams, however, change as a result of not
being told, as a function of speed of waking up. The dream is a whole,
a piece, before it is looked at and related. The relating is like a
mechanical translation of what had just been seen. Furthermore, less
scouring through the subject's life is needed in the search for a
connection between dream text oppositions and the problems of waking
life than is often the case for the interpretation of symptoms. This
area of research on the subject's psyche is thus more circumscribed
than that of psychoanalysis (although what may be written about the
rituals and prohibitions of a whole society is great in both
theoretical systems).
Applying Structuralist Analysis to Dreams
A myth changes in the telling, both geographically and
temporally, and Levi-Strauss aims to attribute these changes to the
299
proposed mytho-logic. But if we study one series of texts it is
heartening to know that the leeway or variation within each version of
a text can be ignored, that we can study the following
transformations:
dream 1
>dr.2
> dr.3
)dr.4
rather than confuse matters with the simultaneous variance of each
piece, as with the mythological data:
variation 1
myth 1
variation 2
)var.l
>var.l
>va .1
)var.2
>var.2
xj
)var.2
This greater simplicity of the dream data is a result of the
text being so close to the dream experience itself. In a way the
subject matter is purer because of the lack of intrusion from
consciousness (assuming the subject was woken abruptly), because each
dream is usually only related once, and because there are no external
constraints on the movement of the dream, save for the incorporated
day-events, unlike the case of variations of myths, which may be the
result of many impersonal exigencies, such as a change in the
society's means of production or sustenance. (Similarly, there are
many external influences on kinship systems, on which the
structuralist method is also applied.) Because of the importance of
this lack of consciousness in all these cases I would hesitate to
apply this method, or the psychoanalytic method, to lucid dreams.
I have noted above differences in the production of dreams
300
from that of myths, which in fact ease the application of the method
to the former. However, it may be said that such differences should
lead to the conclusion that the method is inapplicable to dreams.
Dreams, after all, are a private and unrepeated event, unlike myths
and rituals.
I disagree with this objection for many reasons. Like Freud,
Levi-Strauss is aiming to relate various cultural phenomena to
nature. In studying Australian totemic systems and primitive kinship
patterns he is studying groups that are in the midst of the natural
world. Kinship is intimately connected with sex and group survival,
and totemic identity is closely connected with the surrounding natural
events and classifications. (This is in contrast with life in Europe
where kinship is not always obligatorily connected with progeny, and
its relation to the survival of the nation can be ignored by
individuals. Identity is also connected with choices of consumption
and production, as much as with a correlation with natural
attributes.) I contend that dreaming, as an endogenous function of the
brain, is where we again find ourselves concerned with nature, with
emotional and biological existence, and also with the intellectual
problem of classification, which is the starting point for culture.
Badcock (1975, p.65) notes that kinship is a system of
communication of women between intermarrying groups of men, that
totemism is a language of one's social structure and its groups, and
that myth is a system of signs in which the mind finds its place in
nature, and communicates with the outside world. I add that ritual
appears to be a language of the hierarchies and structure within a
group (for example, with the priest at one end of the church and the
picture of Jesus or Mary smiling down from above the people). These
301
are all languages, in that reciprocal communication, be it of women,
status, group position, or cosmological concepts is taking place. I
suggest that dreaming is also a language, a representation of the
self, and its biological, social and emotional destiny, to the self.
This analysis fits in with Levi-Strauss' division of
language between code and message, langue and parole, which he
obtained from Jakobson. Badcock suggests that Levi-Strauss used the
former as an emphasis on the functional whole (as with Durkheim), and
the latter as an emphasis on the possibilities of changing that whole
(as with Marx). Similarly, I have found a code in the dreams I have
analysed; for example, in the second night that KJ spent in the sleep
lab after mentioning his worries about his worth as a psychologist his
physical position in the environment of the dream followed the simple
code:
up:down :: psychologist:patient
But there were also the events and changes of the dreams, changes
which used this basic code to construct a message, a message that
could eventually even alter the code, as shown by the absence of that
opposition in the last scene.
It may also be said that the results of the structuralist
approach to dreams are not as stunning as was the interpretation that
Levi-Strauss gave to the Oedipus myth. This is because of the
concentration on form which this analysis has undertaken, and also
because, in this study, the problems in K.J.'s mind are more mundane
than the features of the Oedipus myth which endear it to us.
302
The Reality of the Oppositions
The questions remain, however, whether these oppositions are
repeated because we have woken the subject, who then sleeps after
speaking of the opposition in question, and whether independent judges
would obtain the same analysis. The second point will be addressed
after the next section. With regard to the first point I note that the
idea of a dream taking up and permuting a waking opposition is just as
important theoretically as the ideal finding of a dream transforming
the oppositions of a previous undisturbed dream: in both cases dreams
are shown to be a type of thinking. I would not take the path of the
post-Structuralist writer Derrida in claiming that as the author
(dreamer) is not completely in charge of the text then the critic
(interpreter) has an equal role to play in the construction of its
meaning. I maintain that a 'transcendental signified' is present in
the dream experience itself, and that we must beware that, as Sturrock
states: 'Structuralism and post-Structuralism can both lead to
hyperactivity in the reader by encouraging him or her to be abnormally
attentive to each moment of a text, knowing as we now do that
everything in it means, and means to excess.' (1986, p.151).
Methodological Problems with Multiple Awakening Studies
There are three problems with the experimental method used.
1) The subject was woken at the end of each REM period, so
we are really studying only the first dream of a period of sleep. The
subject's report becomes a day residue for the next dream, which is
obviously unlike the situation of a normal night's sleep. At best we
303
must assume that the report is forgotten during the succeeding period
of NREM sleep, and the elusiveness of dreams even after reporting when
awake does give evidence for this. At worst, we must admit to studying
only the initial transformations of day residues which occur in one
dream; although this is itself significant information about the
differences and similarities between waking and dream cognition.
I will answer this methodological problem with the following
points:
a) We find we are not dealing with a whole series of what
turn out to be first REM periods. The dreams which K.J. gives us
become longer and more vivid all through the night, just as occurs if
a subject is allowed to sleep and dream normally.
b)
Verdone (1965) conducted an experiment to relate time of
awakening during the night to age regressive content of the dream. He
found that early dreams have recent memories and that dreams from
later in the night had more memories from early in life. This was
found despite his regimen being that the dreamer was awakened many
times each night. The temporal age of dream content carried on
progressing despite the awakenings.
c)
Dreams that are recalled are usually forgotten within
minutes unless actively repeated. It is quite possible that the
information recalled by K.J. is largely forgotten during the
intervening period of NREM sleep. However, the texts do indicate that,
with prompting, K.J. could recall many of the nights' dreams in the
morning.
d)
If the experimental paradigm is largely different from a
normal night's sleep and any progressions between dreams that we find
304
are an artefact of the regimen, we are still studying an important
phenomenon. Like Silberer, we are then studying the transformation of
waking thoughts into dream thoughts, and then the transformation of
thoughts within a dream, but, unfortunately, not the transformations
that would normally occur, unknown to the outside world, between
uninterrupted dreams. We have, however, been using dreams to study the
type of transformations which are also found in the unconstrained
activity of myth-making, because dreams are held to be similarly
unconstrained by the outside world. The regimen used in this
experiment is certainly aimed at examples of unconstrained thought,
though not at a normal night's sleep, and the finding of such
transformations and progressions here would bode well for the theory
behind the original application of structuralism to myths. After all,
myth-making is itself partly conscious, the myths are 'known even w'hen
not being told, and we must not decry finding the same mytho-logical
activity in a different paradigm, just because the paradigm is not
exactly the one we would like ideally.
2) In taking reports of the experience of dreaming, care
must be exercised in assessing artefacts of the procedure used. For
example, Cicogna, Cavallero & Bosinelli (1982) found awakening reports
to differ from morning, 3
day and 1 week reports, especially
concerning restructuring of the
material and a decrease in
'interactive' as opposed to 'associative' sentences. The former were
sentences of the report related to motivation, the 1 tter were related
to cognition. They also note that spontaneously recalled dreams may be
better structured than interrupted ones, which they attribute to
secondary revision.
However, Belicki and Bowers (1982) found that
305
pre-sleep instructions to pay attention to a particular dream content
were more efficacious than post-sleep instructions, which they
concluded were insufficient to account for dream report changes
following pre-sleep instructions. Foulkes (1966) showed that subjects
interrupted at the early stages of a dream gave more mundane reports
than those woken later on, and similarly whether the awakening was
abrupt or gradual is important, the former leading to the more bizarre
imagery (Goodenough, Lewis, Shapiro, Jaret & Sleser, 1965). Of greater
interest to this study are the findings of Rechtschaffen (1964) and
Fiss (1969) that repeated interruptions of REM sleep heightened the
degree of continuity between dream themes. This is an important point
in that, of necessity, research on series of dreams must involve
waking the subject during each dream. Note, however, that the subject
KJ in the empirical part of this thesis was only awakened once per REM
period, rather than repeatedly, and so the results obtained here can
still be claimed as intrinsic to the dream, rather than to the regimen
used.
Furthermore, that we are studying a subject in the lab
rather than at home will affect the content of dreams. Hartmann (1984)
reports that chronic nightmare sufferers have less frightening dreams
when in the lab. Early studies showed a difference between home and
lab dreams, although Weisz and Foulkes (1970) dispute them. Lastly,
Austin (1971) found that convergers (defined as someone who is
.substantially better at the intelligence test than ... at the
open-ended tests." Hudson, 1966, p.55) reported on 65% of REMS
awakenings as remembering dreaming whereas divergers reported some
recall on 95% of REM awakenings with the former giving shorter
reports. So, the subject's personality will affect the type of reports
306
we obtain.
3) There is the problem of the experimental set-up and
expectations for KJ in this instance. KJ did sleep well but also had
an emotional and professional involvement with the very department
that he was at that moment a subject in. In addition, although the
experiment was about anxiety levels, leading to such comments as 'so
you think there was an underlying tone of anxiety throughout all your
dreams' from the technician, there is the problem of experimenter bias
and subject expectations. For example, at one point the technician
does ask how the dreams tie in with the pre-sleep interview. This
introduces two issues:
a) how ideal the situation was for KJ, in that he does have
problems with the department (describing himself as 'oppositional')
and must hand in an overdue report to Professor Cartwright, who is in
overall charge of the sleep project, after the second night.
b) whether there were expectations for a particular result
by KJ or by the experimenters.
The first point would only lead to problems if KJ found it
difficult to sleep or dream normally, or if he censored his dreams to
make them more acceptable. The evidence about his ability to sleep has
been addressed earlier. Evidence against the second possibility is
provided by his reporting on the first night kissing a colleague, the
issue of anti-Semitism, and the issue of Ros's lost tape-recorder; on
the third night admitting worries about being in the wrong school; and
his honest talking about drugs and his aggressive fantasy about Benton
on the fourth night.
307
The presence of specific concerns, even though their source
lies with the university, does make the task of analysis easier. Much
work on problem-solving in dreams has used intellectual problems
(Schatzman, 1983; Lewin & Glaubman, 1975) or emotional problems of no
relevance to the dreamer (Cartwright, 1974): the stress of various
situations in KJ's life does increase the ecological validity of the
present study of resolutions in dreams.
Point (b) is the more important in that given KJ is sleeping
and dreaming normally the actual dreams he does have may be influenced
by his own or the experimenter's expectations. Although the
application of structural anthropology to the dreams was only
suggested many years later it is the case that the stbjects and
experimenters did expect there to be some relation between waking life
and the dreams. There are a number of points to be made with regard to
this this:
1) the major and quantitative part of the experiment was
about anxiety levels in waking life and in the dreams, not about
specific resolutions of problems.
2) KJ was asked about many things following his dreams.
Their emotional content; his associations to them; his degree of
participation in the scenes; and the presence of real memories, among
others. The variety of theories which, if interested, he could have
guessed that the experimenters were checking was thus vast. KJ was a
naive subject but did, no doubt, have greater k owledge of dream
theories than most people. This would, however, if anything, have
increased the difficulty in having any single expectation about the
experiment: does he have compensatory dreams for his anxiety level, or
308
wishes fulfilled in each dream, or memories cropping up in each dream,
etc.?
The prompts that the experimenter gave at the end of some of
the nights, to sum the dreams up and to provide an interpretation, did
not really go further than the common folk knowledge that dreams are
related to our everyday lives: it would be very difficult to find
anyone, especially someone with good dream recall, who does not
believe this or consider it a possibility. The work of Cartwright
(1986) and Schechter, Schmeider and Staal (1965) shows that any
variation in dream mentation style is an unconscious consequence of
waking cognitive style and emotional life, rather than any thoughts
about the form of dreams one should have, or temporary expectations.
Singer (1981, pp. 68-71) makes the corresponding point that the three
styles of day-dreaming (anxious-distractible, guilty-negative and
positive-vivid) correlate with long-standing personality traits. The
dismal lack of replication of reports of the solving of intellectual
problems in sleep, to be expanded upon in the next chapter, shows the
difficulty of producing short-term changes in the content of dreams,
let alone their form.
3) As is usual, KJ did not treat a dream report as a new day
residue input for the next REM period. The connections of metaphors
between different awakenings that I have found often took a long time
for the author and independent judges to uncover: as is usual in sleep
lab experimentation the subject does not have time to analyse each
dream before the next episode. The correspondence between each dream
and waking life does indicate that waking life is a greater influence
on each dream than is the preceding dream, and that the modifications
produced by the previous REMP's mentation are oblique and certainly
309
not obvious to the dreamer. There is almost no intrusion of the
manifest content of each dream into the next, and when he does compare
dreams it is on the level of the literal actions that he is
performing, rather than the diverse and recherche metaphors that I
have found.
4) It is known that long-standing expectations of subjects
can gradually affect the content of dreams; for example, the
introduction of Freudian or Jungian imagery by analysands. In a recent
replication of incorporation experiments Fonagy and Ennis (1988)
deliberately did not tell subjects that names would be played to them
during sleep in case the subjects then introduced lots of named
characters into the dreams. This evidence, however, is on a totally
different and much more manifest level than the characteristics of
dreams that I have found. Subjects in psychoanalysis do alter their
manifest dream content over time, but this is greatly different from a
sleep lab subject recently inducted into the lab changing the form of
his dreams, especially at the level of metaphors and their
manipulation. Griffin and Foulkes (1977) showed that for 29 subjects,
over 10 nights, it was not possible to consciously manipulate their
dreams by pre-sleep deliberation.
The Reliability And Validity of the Structural Analysis of Dreams
The structuralist analysis of the four nights of dreams was
performed by one author. Prior to this Kuper (1983) analysed the first
two nights' dreams. He emphasised the oppositions Subject vs.
Psychologist, and also Inferior vs. Superior, as I did, but also
spotted Male vs. Female, (ibid. p.173), which I considered
310
unimportant. This shows the necessity of addressing the issue of
reliability of the analyses.
Reliability
There are several claims being made by the structuralist
account of the phenomenology of dreams. There are physical and
abstract oppositions which differ from night to night, but which are
continuously invoked during each night; there are dialectical
progressions through the night along one or more dimensions, and there
are minor reversals during the scenes of each night.
A check of reliability aims to ascertain whether welltrained analysts can obtain the same results for all these findings.
Obviously the presence of the major oppositions and the accompanying
progression are more important than the plethora of minor reversals
(although Haskell, 1986b, stresses the latter) in that the former are
claimed to tie the whole night together in a meaningful way.
However, for the method to be deemed valuable there must
also be a test of the validity of the analysis. The use of a standard
dream dictionary, for example, may result in reliability of
interpretation, but does not guarantee that the interpretation is
valid. Similarly, if dreams are used for predicting the future we
would assess their interpretations as invalid, even if the method was
reliably used, or the products of the method reliably obtained. A
reliable method for the computer analysis of components and relations
in 135 Ge myths was developed by Maranda (1972), and yet the validity
of the resulting analysis of the predominant relations must still be
assessed. Reliability is tested by direct replication, in which the
311
variable remain the same. In order to test validity there must be a
systematic replication in which variables such as setting, subject,
experimenter, etc., are changed in order to determine generality, and
to discover exceptions.
Validity
This refers to the presence of a connection between the
agreed analysis and the waking concerns of the subject. Oppositions
and progressions found must be shown to relate to waking concerns.
To test validity would entail swapping the concerns of one night with
those of another, and attempting to make sense of the analysis of one
night in terms of the concerns of the other night. Experimentally this
is achieved either by giving udges one set of Oxeams an a oce of
concerns, or by giving them one concern and a choice of a few series
of dreams. The latter type of method was used by Kramer, Roth and
Cisco (1976) who found that although individual dreams could not be
connected with waking concerns 'the judges were able to match a
night's dreams to their presumed circumstances of occurrence at a
highly significant level' (ibid. p.316). The validity of the finding
of a progression can be tested for by analysing for the presence of a
progression in one series of dreams, and then altering the order of
the dreams and reanalysing. If a new progression, or even the same
one, is found, then this is evidence that the presence of a
progression is a common and unenlightening artifact of the
experimental procedure, and is in reality insignificant. Validity can
also be assessed by inserting a dream from another series into a whole
series from another night, in order to discover whether the same
312
oppositions can be 'found' in the inserted dream, which would again
show the results of the method to be largely artifactual.
Reliability is checked simply by giving the same texts to
independent judges and finding if they agree on the basic oppositions
and progressions. I propose concentrating on this aspect of assessment
of the work, rather than the question of validity, because:
1) without reliable analyses we are in no position to check
validity,
and 2) there is already a large body of work showing that
individual dreams (Kramer, Hlasny, Jacobs and Roth, 1976) and dream
series (Kramer, Whitman, Baldridge and Lansky, 1964; Kramer, Roth and
Cisco, 1977) can be reliably related to daytime concerns by
independent judges. The latter paper shows that judges could correctly
sort sets of night dreams according to their 'circumstances of
occurrence', although the same could not be done for single dreams.
They postulated that this difference between individual dreams and
series arose either because more useful information resulted from
having the many dreams, or because each dream could highlight what was
important in the other dreams, that it was partly 'the result of
reducing non-specific information and highlighting recognizable,
useful information.' This is precisely the claim of the structuralist
technique, that a comparison of one dream with another will provide
information about what is important in each dream and what parts can
lead to a valid comparison with waking concerns.
Before providing data about the application of the method to
dream texts by different judges mention must be made of the
independant use of the same methodology by the Jivaro Achuar of
313
Equador. Descola (1988) found that among this people there is a
specific method of dream analysis which is used for predicting the
future. He notes that:
the rule of interpretation mainly consists ... in an
inversion between the apparent content of the dream and the message it
portends, in terms of the nature/culture axis: the attributes and
behaviour of animals forbode specific human deeds, while human acts
announce the actions of particular animals. For example, a dream of
charging peccaries will be interpreted as the presage of a skirmish
with enemy warriors, while a man dreaming of sexual intercourse with a
woman is thus warned of a possible snakebite.' (1988, p.8.)
Other dreams are reduced to an eLementary form wlifch
'reduces the content of the dream to a single image that can be
submitted to a general formula of inversion or transposition: the
attributes of natural beings are translatable into human behaviour,
while cultural activities are the register in which relations to
animals are played out. The interpretation of ... dreams is thus
strictly metaphorical.' (1988, pp.11-12.)
However, such use of metaphor is not by the noting of simple
one-to-one correspondences.
'... they seem to differ markedly from the usual techniques
of symbolic correspondence apparently common to many archaic
societies. The predictive content of these dreams is revealed through
a double and probably simultaneous process: the selection of a short
sequence and a systematic inversion or transposition of its signified.
The Achuar, like several other amazonian societies, make use of a
simple principle of conversion that presupposes a correspondence
314
between fields of practice and sets of notions usually held to be
irreconcilable: humans and animals, up and down, aquatic and aerial,
male tasks and female tasks. However, the conversion process is
applied here less to the content of the symbols interconnected by the
sequence selected for gloss, than to the relation it expresses; being
of a purely logical character, this relation lends itself easily to
the operations of permutation by homology, inversion or symmetry from
which the augural message springs. These permutations seem to be
grounded in an elementary grammar with a probably finite set of
rules.' (1988, pp.13-14.)
He goes on to note (ibid. p.21) that this system of
interpretation is not only normative, but can also generate an
infinite number of statements. The method is thus more reliably
applied than are particular interpretations reliably obtained, for any
relation within the dream can be chosen as significant, and have the
method of exegesis applied to it. This is then an account of dream
interpretation rather than dream production.
Production must be investigated by:
dream 2 comparisons,
dream 1
). interpretation studies,
as opposed to dream
simply because any activity can lead to associations of affect or
memories, although the affects or memories were not necessarily
encoded into the activity when it was produced.
The way in which the interpretations described by Descola
occurs, though, does provide evidence that the type of thinking
described by Levi-Strauss (passim.) and Kuper (ibid.) is not peculiar
to those authors accounts.
'... instead of attributing a constant signification to
315
dream symbols, they emphasize the logical operations through which
symbols are connected; it is not the metaphorization of dream objects
that holds a divinatory value, but the metaphorization of their
relations.'
(Descola, 1988, p.25.)
Taking this evidence in addition to the work of Silberer it
is therefore postulated that metaphorisation of relations could occur
in going from waking concerns to dreams, and the extent of this must
be empirically discovered. A dream itself is then the interpretation
of a waking 'story' and contains an encoded memory much as speech
does, as opposed to the warning of Foulkes (1985, p.l92) that
'intentional meanings are not the only meanings events can have'. He
is warning that, for example, the 'meaning' of raised blood levels in
urine is not an intended sign, but can still be read as a sign:
similarly, Foulkes gives the example of the hidden meanings and
'unintentional behaviors' (ibid.) that can be present in a
politician's speech.
However, even if a dream does not have an intended message,
even if a dream is not like an 'unopened letter' (Fromni, 1957), the
way in which it is formed and the ways in which it is interpreted can
provide evidence about types of cognition. We will now test the
reliability of the application of the structuralist method of
analysis, a method which makes claims about both the formation and
interpretation of dreams.
Experiment to Test Reliability
Four judges were obtained from an undergraduate psychology methods
course. Only one judge had any familiarity with structural
316
anthropology. It was explained that they would spend approximately 20
hours on this project, which would entail learning the method and
applying it to night 4. They were given copies of Levi-Strauss' paper
'the Structural Analysis of Myth'; Kuper (1986); Kuper and Stone
(1982); my analysis of nights 2 and 3; and the complete text of KJ's
night 4 dreams, which meant that they had to devise their own precis
rather than be guided by mine. They were then given the following
instructions:
PROJECT ON THE STRUCTURALIST ANALYSIS OF DREAM SERIES
1) In this project you will apply the structuralist method
of analysis to one night of dreams, having read two illustrative
analyses of the previous series provided by the same subject, KJ, a
psychology student.
2) In a sense there is no single correct answer in t)iis
analysis. Whatever you end up with at the end of the project is valid,
it shows what an independant analysis arrives at in the time
available, even if it is less than, or different from, other analyses.
3)
It will be necessary to talk to me and to each other
about what the analysis entails, but you should not divulge what you
have found of the analysis of the last night's dreams.
4)
Levi-Strauss claims that in myths there is a central
concern of the mythmakers which can be expressed as an opposition,
such as over-rating of blood relations versus under-rating of bloodrelations, or endogamy versus exogamy. This can appear in the myth
317
directly, as well as in the form of a derived opposition, which can be
a physical metaphor for it, such as up versus down, old versus young.
The relevant oppositions can be found by comparing the beginning and
end of the myth/dream series. For example, on night 2 the opposition
up/down is important in many of the dreams of the night, as is
revolving/static. Such oppositions are synchronic (i.e. their temporal
position in the series is unimportant) and one needs to be imaginative
to spot what the relevant physical opposition is.
5) He also predicts that the diachronic (i.e. temporal)
movement of the text is dialectical. One scene (thesis) is followed by
another scene which on one or more dimensions is opposed to it
(antithesis), resulting in a further scene which is a combination and
resolution of these two (synthesis), and which then produces its own
antithesis. For example, in night two he is given help by an older
colleague in looking for an adult appliance. Next, he looks for a
child's object himself (antithesis). Next, he shows the workings of a
spinning toy to his colleagues (synthesis). The oppositions which are
relevant are young/old and acting on his own versus being helped.
6) Whether you come up with the same oppositions and
progressions in the series is a measure of the reliability of the
method. Whether this can then be related to KJ's concerns of the day
is a test of the method's validity, which is a different question
entirely. On the second night his major concerns w re his position as
a student vis-a-vis older colleagues, and the Parent-Adult-Child
theory of transactional analysis, which fits in with the physical
oppositions used in the dreams.
318
7) I should be in Brunel everyday of the week if help is
needed.
Judge 1
This judge, who was conversant with structuralist analysis,
found the following opposition:
passive : active :: old : new.
She derived it from such elements as the football and drugs
episode, which was an "old" scene; the old furniture; the rusty
scissors; the industrial part of town (old, she postulates); trees
'alive and growing'; Piper cub aeroplane (presumably new). These she
relates to KJ's increasing assertiveness across the night:
'For KJ the old is related to being passive, there is less
discomfort by staying with what is familiar. This could be seen in the
pre-sleep interview, when to do his work would mean to "stay with the
pain" whereas he could "rejoice about the good things" if he just
"killed time".
She states that the initial problem of:
KJ can assert himself
KJ cannot assert himself
is initially resolved by: 'The plane cuts the bottom edge off the
man/clouds, Connie tells him to stop what he is doing.'
319
She spots the following two parts of a dialectical
argument:
'KJ tries to do a U-turn but he and a woman in a car
overturn' versus 'KJ sees the trees alive and growing and convinces
his mother of the interrelationship with the trees'.
and 'KJ is afraid of the young kids and hides' versus 'KJ finds
his way through twisted hallways and takes what he wants'.
She notes that overall 'he cannot solve his problems with
Benton, with his work, or with Connie by getting someone or something
to do the work for him; by going back (U-turn); by passively following
a woman's lead; by rushing around them; by taking what he wants
without communicating what he is doing; by hiding; or
by
owi thezz
to organise his life. He has to stand up to Benton, his work and
Connie and "hassle it back and forth".' 'The dream's resolution is a
compromise: both active ("Stands up to Steve") and passive ("Listens
to Carol") play a positive part in the solution to KJ's waking-life
problems.'
This judge did mention another opposition, that of
male:female as being related to change:no change. Such an
interpretation was based on the opposition of Connie to his liking for
the plane's activity, and his mother's lack of recognition of the
trees, but it does ignore the obstreporous older male in the forst
awakening, and helpful Ros in the last one. This illustrates that it
is not possible to force any opposition to fit in with the dream data.
320
Judge 2
This judge found the following oppositions:
1.
A. Prospective
B. Active
2.
Immediate
Passive
versus
C. Aims
Needs
A. Intra
Inter
B. Internal
versus
External
C. Stationary
3.
Ignorance
Moving
Knowledge
versus
He relates 1A and lC and 2C to KJ's lack of ability to get
down to hard work, and also states:
'Active versus passive refer respectively (or appear to) to
the concern he feels that he is being dominated by his superior
Benton, and his ability to steer a relatively autonomous course in
terms of action.'
He later notes the following:
Active
'Passive
(1) listened to and
(1) unsuccessful at assertion
in Pizza Parlour/News store
expounding at the
political banquet
(2) angry with man upstairs
(2) phoned up
but passive
(3) talked to by Carol
(4) center of conversation
in apartment
(5) in escaping from the
children
321
(6) moving through warehouse
(7) afraid of the children
but active.'
He also wrote:
"2A. Intra versus Inter (defined by his movement in space relative to
particular environments) seem respectively to refer to a) a lack of
autonomy in his experience of control over outcomes (also pertinent to
the active versus passive opposition) and perhaps an awareness that
habituated programs of action are inhibiting change and b) (Inter)
where freedom of movement physically in the dream, indicates a mastery
over situations in real life. With the further implication that he
possesses the ability to cause changes that conform to his
requirements from life."
Following my instructions in the handout to compare the
first and last dreams he wrote:
"An analysis of the oppositions between the first and the
last dream
LAST:
FIRST:
Moving
Stationary
Remaining within the same
(1) Little automated vehicle
environment:
(2) KJ walking
(1) In bed
(3) Moves into passageway to
hide
(2) In pizza parlour/news store (4) Moving along haIlaj in old
industrial building
(3) In apartment
(5) In a rush to go somewhere
while talking to Connie
(6) The bike had moved
322
(7) The Latin children magically
disappear
Mostly intra environmental
Mostly interenvironmental
sequences
sequences."
This judge then derived a long dialectical argument in
which there is a continual swapping between external and internal
loci. He writes that outside has a 'sense of space and exposure', and
also the 'access to other environments', but which later has an '"out
on the streets" atmosphere suggested by threatening children', to
inside someone else's living room, to a final scene of 'reconciliation
of the idea of being inside as purposive but unproductive with the
idea of inside and productive.' [I had also written of the Fourth
opposition of the night being Safety versus Danger, with its mediation
being the apartment.]
As an example of another dialectical change that this judge
found, I will now transcribe his account of a 'movement between active
and passive'
B: ACTIVE:
'A: PASSIVE:
Apartment scene,
Makes effort, action taken but
direct action taken
attempts to change pizza order
scissors scene
and rejection of copy of Play-
action possible
boy ultimately fail to create
the kinds of changes he is
looking for
323
C. ACTIVE AND PASSIVE:
Active in field physically, passive socially
but engaged (socially). Possession of transport,
active in trying to get away but still nothing
created in accordance with overt wishes.'
He proceeds to write:
'These three scenes provide an escalation of activity. In
the first he takes no action, in the second he does but it is
unsuccessful, and in the third there is a sense in which both outcomes
(non-action, action without outcome) are accounted for while at the
same time augmented somewhat. There seems far more will at work in
scene C.'
This author then proceeds to give many examples of KJ's
increasing competence over the night.
'Ignorance
(1) over impossibility of changing order
(2) of what the scissors are for
(3) of how his floor is the top floor but there's someone above him
(4) and of what the sirens mean in the field scene
Knowledge
(1) In political banquet KJ is portrayed possessing information that
is of interest to others
(2) has knowledge of his environment during warehouse scene
324
(3) plays the role of wise interpreter in relationship with Carol
(4) KJ sees himself as wrongly represented by Steve and thus in some
sense is less knowledgable than KJ'
He goes on to note the following reversal:
'Age change
(2) Carol becomes an older
(1) Child is perceived as younger
woman'
than he says he is
He also notes KJ's increasing assertiveness with regards Connie.
Judge 3
This judge used a symbolic analysis to show the importance
of control/depletion and the difficulty of any change occurring
between them, that is, there is a problem with achieving control, for
it can't be obtained at one go. She hence shows that such control must
be progressively obtained through the night.
"KJ vents his anger with Benton and is depleted and unable
to go forward (incident in pizza parlour). Feelings of lack of control
and passivity (experimenter incident) and anxiety (playing catch with
two footballs and ambulance/police). Considering the implications of
changing his course/or job aim, having to go into strange areas of
study with new people, take in new information. IU trying to distance
himself from these possible choices. By allowing Mother to question
his perception and the incident of money/brother he is ruminating upon
what effect this might have with his peers/family and financial
circumstances Fear of failure (return to child status - latin
325
children) and how any changes might affect Connie (not givibng her
information she might want). His worked over bike, the twisted route
he takes both represented change, which might be difficult, but his
argument with Steve and his control of the outcome, allow him somwe
sense of choice, of continuity, of control."
She spotted the following two codes at work:
Confused : Resolute :: Directional : Turning
and
Apart : Involved :: Vast : Specific
(with people)
(spacially)
The latter correlation captures much of my finding that
involvement with other people is restricting him. The former
correlation does involve the passivity-inability : activity-competence
dimension (noticed by myself and by this judge in her symbolic
analysis) that KJ is complaining of with regards Benton. Objectively
the physical opposition of directional : turning that she uses does
not seem a useful description of the first scenes, but its similarity
to the notion of (spacially) free : constrained should be noted.
Judge 4
This judge concentrated on the aeroplane scene at first.
"Within this segment it seems as if there are clear
oppositions in terms of up/down. The Subject is on a lower level than
the person upstairs and lower than the clouds. Resolution and
satisfaction are both achieved by S managing to ascend to a higher
326
level, with perhaps S taking on the manifestation of the small
aeroplane.
mediator
dofl
Clouds
S
plane
(Benton)
It seems as if the resolution lies in the functions of the
plane. The plane is used to do something to the clouds which could be
said to represent Benton. The satisfaction which S feels with this
scraping may be an indication of reaching nearly the same level as
Benton and somehow exercising his control through the plane on the
output of the cloud (Benton's work instructions)."
This judge proceeds to note that the sweeping and the
scraping are both 'lateral movements'.
lateral movements
scraping
sweeping
result
snow goes down
possible change in snow (?)
(Benton's work/
(S's determination of type/
work gets done)
form of work)
(mundane)
(more interesting)
unhappy
satisfied
He notes an inversion in the Pizza scene on the same
awakening:
327
"The only other binary oppositions which appear within this
segment appear to be young and old: the old man ('grey hair') and the
young boy. However, there also seems to be an inversion of the
characteristics of the two - the old man playing with a (toy) doll and
the young boy serving at the counter (an 'adult' job)."
He finishes by noting that through the night:
"There may also be an intersting opposition of inside and
out with the boundaries between the two shifting continuously."
Judge 5
This judge found the following changes between the first
and last reports:
5th Awakening
"2nd Awakening
Man telling KJ he had to have
KJ telling Steve R. not to
pizza
interrupt
(2)
Young boy
Older woman
(3)
Doll with sharp edges
Chris' soft pants
(4)
Man playing with doll
KJ listening to emotional
(1)
talk from Carol
(5) KJ resigned to man's decision
re. pizza
KJ as centre of
conversation
(6) man living upstairs
woman living downstairs
(7) aeroplane above
train at ground level
(8) clearing snow
sitting out in sunshine."
328
She notes that there is a "passive/active opposition,
resolved when KJ talks with Carol in the last awakening..."
This judge writes that in the pizza scene KJ "accepts the
assistant's authority", that there is a "mid-way" dream of the
football scene in which he is "concerned with authority", and the last
apartment scene in which "KJ is himself exerting authority". She
procedes to write that "this progression is similarly expressed in
relation to the metaphor of altitude ... the cloud, representing the
man living upstairs whom KJ is annoyed with, is scraped underneath by
the top of the aeroplane's wings . . ." She notes that the night starts
off with KJ's annoyance at a man who lives above him, that he is later
on the same landing as the woman who owns the flat, 'quick to assert
to her that he is not the woman owner's husband', and that finally he
comforts Carol, 'the secretary from downstairs'.
Comparison with my analysis
I discovered the following concrete metaphors in the fourth
night:
Above : Below :: Aggressive : Submissive
and
Inside : Outside : : Constriction : Growth
Judge 1 found, amongst others, the oppositions Active (or
assertive) : Passive, and change, growth:no change, which was
related to New:Old.
Judge 2 found the abstract opposition Active : Passive, and also the
concrete oppositions 'Intra, Internal, Stationary' : 'Inter, External,
Moving'.
329
Judge 3 found the abstract oppositions Confused : Resolute, and
Control : Passivity; in addition there was the concrete oppsition of
(spacially) Vast : Specific.
Judge 4 noted the concrete oppositions Inside : Outside, and Up
Down.
Judge 5 found the correlation Up : Down :: Exerting authority : Under
authority. This judge noted that the resolution of this question of
authority occurred when KJ was comforting Carol. Similarly, I had
written 'Both the concrete oppositions and the abstract oppositions
resolve themselves simultaneously. He is finally neither aggressive
like the plane nor submissive like the snow, and, possibly, the
submarine. He is confined with his colleagues and is the centre of
attention.'
In the table on the next page there is a summary of the
work by this author and by the independent judges on the fourth night,
as well as the one analysis by Kuper and Stone; the analyses of KJ's
night's l&2 by Kuper; and the present author's initial analysis of
four dreams of subject LK, also from the Chicago sleep lab. Although
there is the confounding variable of four different subjects being
used, it is evident that there is smaller variability in oppositions
found across the judges used for the fourth night than there is (1)
across subjects and (2) across the four nights of the one subject KJ.
330
KJ: Niit 1
LK
Niit 2
Niit 3
Nit 4
Arlyst
(18)
Iride dept. :thicago wtsi±
PshoLogist:Patit
(Inferior)
1:Dcwi
Criticisi:ActhirJ
PsthoLogist:Pati,t
Spimir:Static
Abe:Be1c
Aixe:BeLci
This author
A9gressiw:Sbnissive
Old:Yc*r
Iride:Qjtside
Shcn Kird-iess:
Ccrtricticn:Grcth
Sufferir
Passi:Actie
Je1
U1iae, Grmith:Old, m
Passive:Pictive
Jue2
Staticnary:Mir
Inteimi :Externat
(s8tiaLly) Vast:Secific
Jue3
Apart: Irwol
Ccnfts&:Resolute
L:Dcw,
Jue 4
Irside:Ojtside
L:Dri
Jute 5
Act ive:Passiw
Iride lth. :thicago cutside Inferior:çerior
In (12)
Male:Fai1e
Ma1e:Faite
PshoLogist:S&bject
Pshotogist:&bject
Hea±A
Iatrogiic:Ca.sJ I' tiit
Physiology:Pshology
Plain's Irth, (1979)
Ccu,tryside:Tcw,
Magir others:Beim nnagl
Actors:Patients
331
Athil Ktçr
(& ALi Stcre)
Summary of evidence used by each analyst.
An asterisk indicates that a piece of evidence used by
myself was also used by the independent judge in the context of
the same opposition. Writing in the judges column indicates that
they used evidence that I had not spotted.
Above : Below
This author
plane (man upstairs)
Judge 4
Judge 5
up:down
up:down
*
*
*
apartment, 3rd or 4th floor
uphill on motorcycle
train at ground level
*
motorcycle on hill
*
ship
*
submarine
*
hole
strange apartment
secretary from downstairs
whom he comforts
332
Inside : Outside
This author
Judge 2
Internal
External
Judge 3
(spacially)
Vast: Specific
no evidence
In pizza parlour
*
Pizza in oven
Judge 4
In field with football
*
no evidence
Outside: brother opens card
Inside: KJ opens card
Inside lady's flat
*
KJ doesn't fit into photos
Lady outside wants to come in
KJ outside on rn/c
*
Trees outside
*
Idyllic outdoor political setting
*
Inside restaurant on peer
*
Train' s compartment
Outside there is danger
*
In the apartment
*
333
Constriction : Growth
This author
Judge 1
Judge 2
No change:Change
Stationary:
Old:New
Moving
Intra: Inter
Pizza kept in oven
Cloud; hole
Can't leave field
Field is an 'old' scene
In bed
Old scissors (I'd noted they *
were hard to open)
Lady outside wants to enter
Sees trees outside growing
Old furniture
*
Car does U-turn
Ship catches submarine
*
Idyllic outside setting
*
Moving train (& compartment)
Dangerous kids outside
Afraid of kids in 'old'
part of town
Moving inside buildings
*
*
gets through twisted hallways
Given job by people in
apartment
*
but he doesn't want the
industrial job
334
A9gressive: S&bnissiw
This aithor
Jue 1
Jute 2
Jue 3
tive:ssive
active:Fessive
*
*
*
CcnfisaJ xut scissors & exrirIiter *
*
*
IJrn-Q. prfiup tn ni,,
Pl
Jue 5
*
asserts itself Wt fails
(er at rim thz'e)
*
*
*
Ccrnie disprcs
In field, fails to esc
*
*
*
'sc
caxerr
with aithority'
he dees U-turn,
caTvirtes
car ectcxr6
sether
critical
*
Rides nyc cutside
S±zmrire vs. ship
active at
-t
nves thrc&i
rthc*.se
Afraid of bt cai aveid kide
*
Ca fix InDtorcle
*
Deals with Steve & secretary
*
*
ies short cut
*
deal ir with secretary
resolves the csitiai
335
*
*
Two of the judges attempted to find some parts of a dialectic
argument. In the following diagram, taken from the last chapter, the
numbers in parentheses indicate that the judge in question had noted
that part of the argument.
He hates confinement vs. Strength of containers
(1 notes apartment and bed
(2 notes KJ makes effort,
but not scissors) \ /
action is taken)
He is forced to fit in (1) vs. He sees growth outside (1&2)
\..
He explores outside (l&2) vs. Danger outside (l&2)
\/
Acceptance with colleagues in apartment,
offer of a job.
(l&2)
Judge 2 summarises this resolution as 'phone call
enquiring as to plans suggests external locus, as does presence of
many people hence reconciliation of the idea of being inside as
purposive but unproductive with the ideas of inside and productive.'
The following reversals were spotted by the judges:
Judge 2 noted that overall KJ interacts with single people
at the start, and groups near the end. He spotted 'child is perceived
as younger than he says he is' whereas 'Carol becomes an older woman'.
Also, 'airplane is up and in the air' whereas the 'trolley is on rails
and travels along the ground'.
Judge 3 mentions that he is angry with someone above
336
(Benton) and concerned with someone below (Carol), and confused in the
pizza parlour as opposed to being firm about the interruption from
Steve.
Judge 4 notes that early in the night he is
fearful/apprehensive about an old man, and fearful/apprehensive about
Latin kids later on. This judge also spotted 'the old man playing with
a (toy) doll and the young boy serving at the counter (an
'adult'job).'
Judge 5 spotted the change from 'man living upstairs' to
'woman (secretary] living downstairs', and suggested that KJ seeing
the trees grow outside may 'reflect KJ's concern about time-wasting
and not getting any work done here.'
I conclude that it was possible, in a short time, to teach
the method to subjects who had had, with one exception, almost no
knowledge of it. Not only did they derive oppositions similar or
identical to those found by myself, but the evidence they gave for the
oppositions overlapped greatly with mine. Many reversals that I had
noted were also found by the judges, as were large parts of the
overall dialectical argument.
It may be remarked that some of the oppositions are
sufficiently vague that they could be found in any text. However:
(1) The basic theme of the fourth night, noted by all judges, that
there is increased successfully assertive activity and movement across
the night is not so vague that it could be applied in reverse to the
night.
(2) The assertion:submission abstract opposition and its corresponding
physical counterpart of constriction:growth, are certainly
337
inapplicable to nights two and three, as well as to the Irma dream
analysis.
(3) There are a multitude of other oppositions which these can be
differentiated from, such as far:near; fast:slow; long:short. It is
not the case that there are a few vague oppositions which can be read
into any text.
Conclusion
I concluded at the start of this section that the
structuralist method is more falsifiable than the psychoanalytic
method. I then showed that its application to dreams is as justifiable
as its application to kinship networks, rituals and myths. From
chapter 13 I also conclude that the results of a progression within
the manifest dream are not in conflict with the work of Freud on the
origin of each separate dream image. From an examination of the
regimen used and the question of experimenter bias I also conclude
that the progressions found are not artefacts of the particular
regimen of awakenings used: that a regimen of multiple awakenings is
used, however, is unavoidable. The study of multiple awakenings is the
closest we can come to studying the interrelationship between dreams.
I then showed that not only can the structuralist
methodology be taught to naive subjects, but that its application by
independent judges to a dream series from one night led to results
very similar to mine.
338
CHAPTER 19
CONCLUSIONS
Summary and Discussion of Results
The initial description of seven theories of dreams showed
that much of the empirical evidence that we have about the REM state
(chapters 1, 2, 4, & 5) can be used to support almost any of them. For
example, the evidence from chapter 2 that the physiology of the REM
state has correlations with some factors of the experience of dreaming
does not prove or disprove whether that experience is epiphenomenal or
has itself some purpose. Similarly, while chapters 5 and 6 showed that
the REM state is functional with regards memory and creativity there
was only the possibility that dreams play an active part in memory and
creativity also. It was therefore necessary to obtain predictions
about the phenomenology of the experience which could be used to judge
the theories.
One such prediction concerns the degree to which dreaming is
discontinuous with waking concerns. As noted in chapter 3 philosophy
in centuries past divided into two extreme schools over this, with the
addition of a compromise school by Freud. All the theories stated,
such as those of Palombo, Clark, Evans, Rycroft, and Kuper and Stone,
as well as the physiological theories (described in chapter 4) and the
first theory derived from neural nets, that of Crick and Mitchison,
can support the notion of a continuity between waking thoughts and
that of dreams. The empirical research of the last four chapters has
supported the continuity theory, first mooted by Leibnitz, in finding
an interrelationship between waking and dreaming life with regard to
339
their content.
The theories make predictions about the type of thought
present, as described at the end of chapter 13. Evans predicts that
the type of thought will be much as in waking life; physiological
theories (e.g. the sentinel theory in chapter 1) can allow an
independent simultaneous dream world and hence make no prediction
about its form or content; the other theories have their own unique
predictions about how the form of dreams differs from that of waking
thought.
Palombo's theory could not be checked because information
was not taken about the presence of KJ's memories within the dreams.
With our data it is only possible to investigate how each scene is
related to other ones thematically, although for a full account of the
subject it will eventually be necessary to determine 'the ways in
which mnemonic activation and thematic planning interact in dream
formation. The implicit idea is that "What happens next?" in a dream
depends partly on what memories then are in a state of activation and
partly on the narrative requirements that what happens next must be
some sort of dramatic continuation of what happened last.' (Foulkes,
1985, p.157.) An account of why some memories are involved will be
needed in the future, and Palombo provides one speculation about this,
and yet the progressions that are seen in each of the four nights go
beyond the predictions of Palombo's theory. Certainly there are parts
of the dreams of KJ which KJ states derive from e periences from long
before, yet to state that this means that it is for informationprocessing purposes that these are juxtaposed with recent memories is
quite speculative. In addition, although some of the people involved
340
dated from his past he recognised very little of the dream features as
being real memories. The presence of so many metaphors in the dreams
also points to the dreams being creative acts rather than the running
together of actual memories. It may even be suggested that the high
figure reported by Palombo (l984a) for dreams containing actual
memories may be partly due to experimenter bias, because the subjects
were his psychoanalytic patients, and were instructed to search their
memories for sources of their dreams.
One strength of this theory, however, is that Palonibo (1978)
considers that such a rational activity as problem-solving is solely
part of waking life. The evidence for its occurrence in dreams is
often poor, which has led to papers that support it either being very
verbose with little empirical content [Roberts, 1985], or just having
the repetition of anecdotes from earlier studies [Schatzman, 1986].
One exception is the paper by Dave (1979), in which one subject after
rational-cognitive therapy solved a problem in his life, whereas nine
did so after having a relevant hypnotic dream.
It would not be fair to cite the presence of such long
plots in the four series of dreams as evidence against Palombo,
because to do so would rely on a naive view of how memories are made
or stored (it would rely on depicting them as quick snapshots rather
than as interconnected schemes), and also because as a psychoanalyst
he does allow for the secondary revision of the dream material. The
memory system would no doubt also require the use of narrative for its
exploration. The strength of Palombo's theory is triat it addresses the
issue of why there is such a mixture of memories within each dream. In
common with Freud's account of old wishes deriving aid from new events
for the aim of representation, this has the advantage over simplistic
341
theories of the chronological coding and storage of memories, such as
used by Verdone (1965). A weakness of the theory, however, is its
unfalsifiability: even if REM sleep is found to aid storage of
memories it is difficult to prove that the mixture of memories in
dreams has any input into this. This argument is similar to that
between Paivio's dual-coding hypothesis and propositional theories.
Given that dreams do provide a juxtaposition of temporally distinct
memories, the question is whether they exist in order to do that.
The theory of Clark, Winston & Rafelski (1984) and Clark,
Rafelski & Winston (1985) also uses the idea of memory-matching, but
here memories which are of similar ages can be compared. The first
weakness of the theory of 'positive learning', however, is whether any
explanation is being provided by their notion of an 'decreased beta',
that is, increased stochastic neural firing, to account for the
mnemonic activation present during dreams. Evidence for stochastic
neural firing in waking life is provided by Pritchard (1961) and by
Warren (1961), who played a word or phrase repeatedly and found that
the subjects would hear illusory changes, which are explained as due
to habituation occurring. It is possible that dreaming creates a
similar effect as with the Warren experiment, not by habituation of a
neural pattern but by a decrease in beta (a measure of the randomness
in neural firing) leading to that pattern being drifted away from,
towards a looser association. This is termed 'beta joggling' by Clark
et al (1985) and although it could be the neural tasis of the thought
joggling and lateral thinking mentioned in chapter 6, this remains a
circular argument unless physiological evidence of the changed
activation algorithms of the nerve cells is provided.
342
Again it must be noted that KJ did not remark upon an unduly
large number of real memories being present in his dreams. Also, how
is the scene of the aeroplane in night four to be described? As a
symbol standing for a real life experience? Once such questions are
asked and the symbolic or metaphorical aspects of dreams are given
their full weight, following the evidence of the repeated and
congruent metaphors shown here, it must be asked what type of thought
is happening? Such links between scenes were noted as long as a
century ago by de Saint-Denis (1982 edition), who termed them
'abstractions'.
The repeated metaphors found argue against the Evans
computer theory to an even greater extent. That theory's emphasis on
action and programs is congruent with the narrative aspect of dreams,
and with their frequent mundanity, and yet the repetition of the same
relations and referents in such physically dissimilar scenes produces
a problem. Furthermore, much of what KJ saw in the dreams was
inoccuous, and yet the Evans theory marks much of the contents out for
erasure. Very few realistic resolutions occurred, possibly because so
many of the scenes are so short. Ostensively each image did not use
the same scheme or programme as a previous image, but often did
provide a resolution at a metaphorical level for previous scenes.
These resolutions, however, were often not realistic; for example, the
record cover on night 2, or meeting the children at the end of night
3, or the train on night 4.
Despite Crick and Mitchison's emphasis on bizarreness and
erasure the mundanity of so many of the scenes stands out (e.g. the
343
giving of presents, the football game, the open-air political
meeting). The juxtaposition of many of the scenes looks incongruous,
and yet there was so little that was bizarre within any of the scenes.
One strange aspect was that so many of the scenes seemed to synibolise
the same referent as the adjoining scenes, or to refer to its
opposite; this was much more evident than the bizarre condensations
predicted by Crick and Mitchison.
Hopfield, Feinstein and Palmer (1983) have shown that one
model of memory needs an unlearning process to keep it free of false
information, but it does not follow, even if long-term memory is also
based on associative networks, that a similar method of cleaning is
needed; maybe thought can keep the memory clean or at least ensure
that the postulated false traces are never accessed anyway, for they
would need to fit in with the overall logical system in order to be
accessed. The evidence that the mixing of human memories does not
occur at the level of storage was reviewed in chapter 13.
In chapter 4 the work on the catecholamine hypothesis was
used to illustrate the proposition that some physiological theories of
REM sleep predict that dream mentation has a sharp dichotomy with
waking mentation. A corollary of this is that it is pointless to
eradicate or unlearn fantasies during REM sleep because with normal
levels of catecholamines the brain can adequately perform this task
during waking life anyway. The unlearning hypothesis is thus analogous
to punishing someone's behaviour when all that is really needed is to
remove them to another situation, where that behaviour would not then
occur. In addition, many experiments reported in chapter 5 can only be
interpreted as showing that REMS aids the consolidation of memory
344
traces, or makes ones that already exist (and show no interference
problems) less fragile. For example, Segal, Disterhoff & Olds (1972)
showed that hippocampal theta rhythm, which is an integral part of REM
sleep, is also associated with learning in adult life, thus showing a
similarity between the two states with regard to their physiology.
Work on protein synthesis also shows the similarity between REM sleep
and normal learning. Lambrey-Sakai (1972) reported that REM
deprivation in rats impaired the incorporation of amino-acids into
brain proteins, thus learning is itself disrupted by REM sleep
deprivation: under the unlearning theory it would be predicted that
amino-acids would not be taken up anyway during REM sleep, when
unlearning is supposed to be happening.
Ontogenetic data, as reviewed in chapter 1, opposes the
unlearning theory because at the same time in life as there is little
in memory storage there is the greatest amount of REM sleep, although
Crick and Mitchison (1983) do suggest that unlearning could be a
preparation for future learning in the fetus.
Dreaming may aid the storage of memories by increasing the
number of interconnections between memories - this obviously is an aid
to retrieval as well as to intelligent thinking. Work on associative
nets is not yet concerned with interconnections between many different
memories, it is concerned with inputting many separate memories onto
one network. The function of dreaming then may be working on a higher
level than simply aiding storage and consolidation. The importance of
interconnections between memories is that with each increase in
encoding used there are then more cues which will allow recall of that
memory. Memories are thus made stronger by having more
interconnections with other memories, whereas in the Hopfield net
345
strength is related to the neural connection weights of that one
memory. Animal memory is not akin to the Hopfield state space (see
ch.7) with its different sizes (strengths) of troughs (traces), rather
it is composed of troughs of indeterminate depths but with various
numbers of interconnections between them. The unlearning theory
ignores the adaptiveness of memories being correlated, and the
contribution this makes to intelligence and creativity.
The other main weakness of the Crick and Mitchison account
follows from the work reviewed in chapter 2. The PGO bursts are found
to correlate with vividness and not with bizarreness, and are possibly
not the cause, but a consequence of, vivid images.
As was shown earlier the presence of a plot is only a
problem for a naive application of the Hopfield net to human memory;
Foulkes (1982b) gives evidence that the ability to form dream plots is
only achieved with the maturation of concrete operations in the child,
and proposed that there is a basic mechanism for dreams with a
maturing narrative ability laid on top. However, the prediction from
the Hopfield work would surely be that plots would be short, only long
enough to provide definition for the content of the images. A
qualitative account of the length of the scenes has been given in
chapters 14-17; Rechtschaffen (1978) has noted that they are
considerably longer than day-dream scenes in waking life. Both the
Clark and Hopfield theories are contradicted by this finding. They are
in a similar position to Hobson, who states 'activation synthesis has
not yet attempted to account for the thematic-narrative constancies of
dreaming in any systematic way.' (1988, p.271.)
The great strength of the Hopfield/Crick and Mitchison
346
theory is that it accounts for the difficulty we have in ascribing
meaning to dreams. Dreams have to be interpreted because the meaning
is not apparent. Not only do these interpretations differ but once a
dream is interpreted we usually say 'this is what it must have meant',
rather than 'this is what it
did mean'. Even when an intention is
found it must often be the result of evidence picked up from all over
the dream, rather than from the end of the dream; dreams sua1ly
appear more meandering than interpretable sentences.
As shown earlier the psychoanalytic account of dreams does
allow for the presence of narrative amongst the disjointed scenes, and
its reliance on symbolism is obvious. The strength of the Rycroft
account is in its emphasis on the presence and importance of metaphors
in dreams. For Lakoff and Johnson metaphor 'unites reason and
imagination. Reason, at the very least, involves categorization,
entailment, and inference. Imagination, in one of its many aspects,
involves seeing one kind of thing in terms of another kind of thing what we have called metaphorical thought.' (1980, p.193.) However, the
data provide a problem for psychoanalysis, and also for Rycroft's
reformist account, in the presence of a continual exploration of
oppositions from one scene to the next. For example, the changes noted
from the first report on night three to the last report show a move
from 'older people criticise his schooling' to 'he gives schooling
advice to a mother'.
A progressive wish-fulfilment was shown for the four nights,
whereas for Freud a wish-fulfilment is not an end result but rather
one of the ubiquitous latent thoughts which would pervade many parts
of the dream. For Freud the parts of the dream were interconnected,
347
but by their relation to dream thoughts rather than by manifest
aspects of the dream itself, as evidenced by the dream data here.
In the light of the predictions made at the end of chapter
13 with regard to the relationships between the images and with regard
to the presence of bizarre or condensed images, it is necessary to
conclude that the structuralist theory of Kuper and Stone does explain
the features of the data which are present. Metaphors were found to a
great extent, contrary to the predictions of Clark et al, Crick and
Mitchison, and Evans, and not only were the metaphors grouped in terms
of oppositions, which goes beyond Freud and Rycroft's predictions,
they also progressed dialectically. With regard to the predictions
about bizarre condensations, only some of the theories could be
tested. Due to the lack of information about memories present in the
dreams Palombo's prediction of the mixture of recent and distant
memories could not be tested. In addition, KJ most often seemed
puzzled by the condensations. They seem to have no place in Evans'
theory, which is hence contradicted. That they could sometimes be
connected with foregoing events (e.g. the record cover in night 2)
contradicts Crick and Mitchison's account that dreams start with
condensations, but allows for no real differentiation between the
theories of Kuper and Stone, Clark et al, Freud or Rycroft, because of
the lack of data with regards memory combinations present.
One weakness of the structuralist account is often held to
be its unreliabilty and dependence upon the personality of the
interpreter. However, according to Boon, structuralism does make
specific and falsifiable claims. 'Another sort of criticism of LeviStrauss centers on his discovery procedures. While it has been
348
suggested that Levi-Strauss fails to make these procedures clear, I
would counter that he makes them so clear as to leave us incredulous
before their simplicity' (Boon, 1972, p.128). The method is certainly
clear enough to have been taught to 5 undergraduates in a matter of
hours. Boon shows how one structural analysis can indeed be better
than another one; in the case of an Omarakana village geographically
dichotomised by the opposition Male/Female he shows how the dichotomy
Free-sex/Regulated sex is more accurate (ibid. p.127). It does remain
a problem, however, that there were some differences between the
author's account and that of the independent judges, and yet we are
attempting to discover just some of the rules involved in the
production of very complex imagery which cannot be related to simple
contemporaneous inputs to the subject. The possible theoretical
importance of reversals, for example, has been noted by Haskell
(1986b) and by Foulkes (1985), and we should certainly be on the wrong
track if we did not experience difficulties and disagreements in
identifying them in practice.
Dreaming Nentat ion
The permutation of oppositions and their progression through
the night have been shown to be on the surface of the dream and can be
recognised. This has not been appreciated before because of the
necessity of taking reports of all dreams through the night, the
necessity of using reporters who can remember many scenes from one
dream, and the necessity of concentrating upon the form of the text
itself, rather than attempting to 'demythologise' it by making it as
sensible as possible, or taking numerous associations which lead us
349
away from the dream. Only a few associations of the subject have been
used in this study. I have taken literally the common belief that
dreams are a form of thinking, and so have used the surface text to
test this assumption. For if one wants to discover the presence of
directed thought in waking life one does look for syllogisms, for the
exploration of alternatives, of counterfactuals and extreme cases.
These are found in lU's dreams and betoken a sophisticated level of
thought - similar to the 'sophisticated use of language' which Archard
(1984) claims is so surprising when found in the supposedly pleasureseeking and primary-process using Freudian unconscious.
'I do not apply the method of free-association because my
goal is not to know the [patient'sJ complexes; I want to know what the
dream is.'
Jung (1968)
Such a use of associations can lead to convoluted arguments
which detract from the text, although it has its therapeutic uses. For
example, Hall (1966) reports a patient's dream in which he was
punished by his father for breaking windows - rather than interpret
this as an example of disobedience and rebelling against authority
Hall interpreted it as punishment for the unconscious wish to
penetrate the mother. Similarly, Foulkes (1978, passim) relies greatly
on associations, although I consider this to be due to the dream
example that he uses, Freud's 'Botanical Monograph', which has all the
hallmarks of a short NREM or hypnogogic event.
What I have found to be unconscious in dreams is not a
reified unconscious mind, but rather an unconscious process of 1)
transformation of concepts into concrete terms, which is as much
350
unconscious as much of the processes of mental arithmetic, creative
incubation, and the sudden recall of memories, and 2) the
transformation of these concepts and concrete terms by means of their
constituent oppositions. For example, on night 4 KJ compared the
freedom to develop which he would find outside the department and
outside Chicago to trees that he could see outside through a window,
but later depicted the dangers of being outside.
A dream is therefore a mostly unconscious processing of
mainly conscious and preconscious (ignored rather than repressed)
concepts. The most simple proof of this is in fact the phenomenon of
Freudian dreams itself, with their imagery of underground tunnels and
erectile constructions. The patient unconsciously forms these from
conscious knowledge of Freudian psychology, whereas a Jungian patient
forms archetypal images from the unconscious processing of what the
patient is rewarded and cued for in therapy sessions, and from general
knowledge of Jung. This exemplifies the malleability of dreams by
conscious knowledge. The dream is such a surprise because of the
unconscious processing of this knowledge, not because it draws from
memories and instincts which belong to a completely unknown part of
the mind.
It is because I have likened KJ's dreams to his waking
conscious thoughts, as he did, and shown that many of the details are
comprehensible in terms of these thoughts, that it is possible to link
dreams with incubation and creativity, as in chapter 6. Creative
thought is believed to require a period of accumulation of facts,
after which the issues are mulled over or, quite literally, slept on I wish to emphasise the pre-conscious or conscious nature of the
351
elements that are worked on, in contrast to the unconscious nature of
the actual processing. Similarly, a painter will be conscious of the
model, and of the model's environment, and of the paints and canvas,
but then creates a painting in a partly unconscious manner. Foulkes
provides evidence that, once started, there is a drive towards the
completion of the processing, such that 'on awakening midway during
REM periods, laboratory subjects often say things like "It wasn't
finished yet" about the dream they have reported, even though they
profess not to know the specific semantic intentions underlying what
they have dreamed already, or that would have guided the dream to its
completion' (1982a, p.180). It has already been noted that there is
more unease amongst subjects awakened from the middle of dreams than
from those not allowed to even start a dream. On this urge to carry
forward the thinking being performed Rechtschaffen (1978, p.102)
writes that 'there is a definite chronological march of theoretically
connected material, which probably proceeds without significant
detours for longer periods of time than most spontaneous waking
thoughts... Perhaps the single-mindedness of thematic coherence is
possible because attenuated reflectiveness and imagination prevents
interruption by competing thought streams'.
It must be stressed, however, that a solution may not be
obtained, there may be no diachronic progression but rather only a
synchronic restatement of the problem, such as with the first three
scenes of KJ's fourth night. Similarly, Kramer, Whitman, Baidridge &
Lansky (1964) claim that some nights show a thematic progression from
dream to dream, with alternation between disturbing and resolving
dreams, while others just have a repetitive restatement of the
original problem.
352
Evidence for the meaningfulness of dreams is seen in those
that 'recapitulate' symbolically a large part of the dreamers life. I
use this word because of its evocation of an old, but
now
discredited, theory of embryology, which held that each fetus quickly
goes through all the forms of its evolutionary ancestors. This type of
dream shows more than the symbolism of part of someone's life, but
rather gives an overview. I will give four examples of this special
kind of symbolism, which must surely be showing a greater amount of
thinking than the simple vertical symbolic transformations written
about by Silberer.
1)
In KJ's first night of dreams, written out in full in
chapter 14, he tells of a night in the sleep lab. Early on he is in a
room full of experimenters, he leaves the room, and when he comes back
it is empty. This is also how he describes his long term relationship
with people at the university, that he started there going out with
them for much of the time, and now rarely does so.
2) On the last night (chapter 17) the first dreams show
great frustration at constrictions, which are later accepted (in the
form of a moving train). This, in fact, recapitulates his long term
history at the department.
3)
In my 'Konstanz' dream (cited earlier in chapter 13) I
start digging up explosive material from a lake by myself. I then
enter a submarine but worry that it will descend in the lake and that
I will not be able to escape. I leave the submarine, stand on the
shore, advise others not to enter it, and finally ask a young boy to
353
help me retrieve some of the material. This dream clearly encompasses
over a year of my life, with its worries of whether to start, and
later finish, psychotherapy.
4) In my 'Natasha' dream I start in a karate class, with
all its attendant bodily activities, and then enter a library. I later
leave the library and the dream finishes with me kissing the neck of a
friend. This dream, as well as showing the internal dialectic of the
body versus the intellect, does review my life history of normal
adolescence turning to over-intellectual scholar, and returning to a
compromise between the two.
Palombo's theory (of the matching of past memories with
recent events) certainly accords with this idea of a symbolic
recapitulation. (It is also possible that dreams such as these will
still show, as much as any other dreams, the Structuralist claim of
'clear cut relationships such as contrariness, contradiction,
inversion or symmetry' rather than 'small positive or negative
increments' [Levi-Strauss 1981]. This is because only a few of the
supposed oppositions of the dreamer's life will be held to be
emphasised in any one recapitulatory dream.) This idea has
similarities with Kohut's (1977) notion of 'self-state dreams', which
portray in vivid, symbolic terms the current state of the dreamer's
self but without a time dimension involved.
With humans, the advent of the realm of the symbolic leads
to changes in waking and sleeping thought. Metaphor and metonymy can
now exist in waking and sleeping thought, with the disadvantage that
in the latter case they are taken to be real. This leads to no feeling
354
of incongruity while dreaming (except when it prompts a lucid dream),
but does on awakening. We may postulate that the dreams of animals are
never bizarre (unless a form of condensation occurs) because they do
not use symbolism and metaphors. Metaphor and metonymy become part of
waking memories and thought, and waking behaviour also becomes part of
this symbolic system. So, whereas a male stag will fight for
territory, the human male's territory acquires symbolic value,
extending way beyond the specific uses to which it is put. This
symbolic realm also invades sleeping thought, where it is taken as
real. The environment of a dream can thus change suddenly for it is
not just the background to the dream events but is simultaneously a
symbol which can be discarded. Thus, in the Konstanz dream 1 travelled
instantaneously from inside the submarine to the shore, without
noticing any journey, because I was actually exchanging one symbol for
another.
In this thesis the notion of the signifier being
permanently available for new uses has been illustrated. This relies
on the multitudinous properties of each object, such as, for example,
the submarine. KJ used the property of the submarine that it voyages
at a lower level than a ship - I used its properties of being a
container, and possibly sinking. This follows from Rycroft ridiculing
the idea of producing a list of sexual symbols because there is 'no
reason to suppose that there are any objects which could not be used
by someone somewhere to construct a sexual metaphor' (1981, p.79).
It has been shown that the properties of an object or
concept are intimately connected with the properties it does not have,
with what it symbolises a lack of. We do not just find a logic of
abstracted properties in dreams, but rather a binary logic of
355
abstracted properties. However, it is a peculiarity of the dream state
that alternatives cannot be held in mind at the same time, the
'single-mindedness of dreams' means that 'these alternatives seemed to
occur sequentially one at a time' (Rechtschaffen, 1978, p.102). (Lucid
dreams may be an exception to this rule that while dreaming one cannot
imagine alternative situations.) When one alternative is given the
others are still present, as shown by the high use of opposites in
word association tests. Conversely, when we focus on one attribute we
downplay others: 'every description will highlight, downplay, and hide
- for example: ... I've invited a renowned cellist to our dinner
party. ... I've invited a lesbian to our dinner party... Though the
same person may fit all of these descriptions, each description
highlights different aspects of the person.' (Lakoff & Johnson, 1980,
p.163.)
The discussion of the importance of what an item is not
brings us to the presence of dialectical arguments in dreams. A
connection between dialectics and metaphor has been noted by
Heilbroner, who relates the two to the unconscious and the
preconscious. ' . .. much "creative" thought hinges on the possibility
of discovering analogies, linkages, syntheses and the like between
hitherto separated entities. This mental act of "leaping" from one
thought to another has an obvious analogue to the idea of
relationship, so central to dialectical philosophy. ... dialectics is
at bottom an effort to systematise, or to translate into the realm of
manageable, communicable thought, certain unconscious or preconscious
modes [e.g. ambivalence] of apprehending reality, especially social
reality.' (1980, p.56.)
356
Of relevance to this finding of oppositions and dialectical
thought in cognitive psychology is the work on reversal theory and
personality in motivational psychology. Michael Apter has suggested
that humans are not always motivated by the desire to restore
homeostasis, but that we may show reversals between pairs of bistable
states, such as the reversal of the relaxation-anxiety continuum with
the boredom-excitement continuum (Apter and Rushton, 1981).
We now come to an empirical result concerning these
oppositions which could not have been known in advance. Whereas a
scientist ostensibly and consciously works by the changing of one
variable at a time, we see in this thesis, and also in Levi-Strauss'
results, that in dreams and myths two or more oppositions vary
simultaneously. In fact, Levi-Strauss claims that the variance of one
opposition leads to the simultaneous permutation of all oppositions in
the system. This breaks the rules of scientific reasoning by means of
formal operations.
The use of reversals of oppositions has also been noted by
Foulkes, who writes:
'Another seeming peculiarity of dream experience is the way
in which dream imagery seems to portray reversals of what we take to
be our standard waking conceptions. The traditional interpretation of
such reversals is that they are motivated, but there may be a less
complex and more empirically justifiable explanation. Semantic memory
research suggests that, in the featural and semantic analysis of
certain concepts, we may code directly what a concept's opposite is or
what a concept specifically is not. Thus, the time a human takes to
verify, in a reaction-time experiment, a proposition such as "a whale
357
is not a fish" may be less than the verification time for "a whale is
not a bird", precisely because one relationship has been taught, and
encoded, directly, while the other has been left to inference' (1982,
p.178).
In showing the presence of mytho-logic in human dreams I
have shown it not to be limited to the waking thoughts of primitive
peoples - for some reason it is used by each and every human, and has
time set aside for it. Work should now be done on providing evidence
from waking life for the use in creative thinking of dividing the
world into oppositions (as in de Bono's P0 function of aiming for the
opposite of what was intended), the comparison of oppositions in
metaphors, and the search for mediators and resolutions, as in many a
detective novel. I have arrived at this result because of the
comparison of dreams with myths. Kracke (1987, p.52) recommends the
same project because 'myths are obviously a highly developed art form,
an "oral literature" . . . which can attain great subtlety and
complexity of structure and thought. If dreams and myths partake of
the same fundamental kind of image-based thinking then the type of
thought process characteristic of dreams is capable of high
development, and of the expression and elaboration of complex and
subtle ideas. Dream thinking, or primary process, is not merely a
degenerate, regressive form of adult logical thought, but is rather a
distinct form of thinking, ... as valid as logical, categorical
thought, but appropriate for different kinds of p oblems.'
Haskell (l986b) makes the similar point that:
"The identification of an array or group of operational
structures could increase our current understanding of
358
logical and psycho-logical development and function. Such
research may yield several forms of cognition not readily
apparent in waking thought, and inform us about the basic
structure(s) of cognition."
It would be a mistake, however, to see this type of thought
as completely outside our usual waking thinking. Lakoff & Johnson
(1980) write that much of our thinking is not based upon the
dictionary definitions of concepts and objects, but upon metaphorical
thought.
... if you look in a dictionary under "love", you find
entities that mention affection, fondness, devotion, infatuation, and
even sexual desire, but there is no mention of the way in which we
comprehend love by means of metaphors like LOVE IS A JOURNEY, LOVE IS
WAR, etc. If we take expressions like "Look how far we've come" or
"Where are we now?" there would be no way to tell from a standard
dictionary or any other standard account of meaning that these
expressions are normal ways of talking about the experience of love in
our culture.' (ibid. p.115) 'This suggests that understanding takes
place in terms of entire domains of experience and not in terms of
isolated concepts.' (ibid. p.117.)
They note that the use of metaphor is so pervasive that it
is often overlooked, for example:
'Your reasons came through to us.
It's difficult to put my ideas into words.
Try to pack more thought into fewer words.
His words carry little meaning.
359
Your words seem hollow.'
(Lakoff and Johnson, 1980,
p.11.)
The particular metaphors that are used are based upon our
physical experience of seeing the world and having a body. For
example, 'What's coming up this week' uses the physical basis that
'Normally our eyes look in the direction in which we typically move
(ahead, forward). As an object approaches a person ... the object
appears larger. Since the ground is perceived as being fixed, the top
of the object appears to be moving upward in the person's field of
vision.'
(ibid. p.16.)
Other uses of the concretization 'up' occur in such
statements as 'things are looking up', which is based upon the
correlation:
Good : Bad :: Up : Down.
'He is High-minded' and 'That was a low trick.'
are based upon the correlation:
Virtue : Depravity :: Up : Down.
This latter one is part of the same network of metaphors as:
High status : Low status :: Up : Down,
as well as:
Rational : Emotional :: Up : Down.
The latter correlation is the basis for such statements as:
'He couldn't rise above his emotions' and '... I raised it back up to
the rational plane.' (Examples taken from ibid. pp.16-17.)
Even the use of rising or falling intonation in sentences
('Will you ever learn?') follows a pattern of use which does convey
added information about the speaker's knowledge and intent.
These examples show that it is difficult to differentiate
the physical from the cultural basis of a metaphor, and that many of
the metaphors are linked together within a culture. The metaphors used
360
may also change across cultures; it is easy to imagine that for some
cultures ACTIVE IS DOWN whereas PASSIVE IS UP, and Lakoff and Johnson
(ibid. p.161) note that the Hausus speak of the nearest part to us of,
say, a boulder, as its
back, rather
than its front, as we would say.
It must be emphasised, however, that there is an
experiential basis to these metaphors due to our experience with
physical objects, and that this enables us to view 'events,
activities, emotions, ideas, etc., as entities and substances' (ibid.
p.25) and hence quantify and reason about them. Asch (cited in
Billow's [1977] review of the psychological literature on metaphor)
has performed inter-cultural work on the basis of metaphors common to
different cultures. For example, the morpheme for straight means
honest, righteousness, and correct understanding in Chinese, Old
Testament Hebrew, and Homeric Greek (ibid. p.84).
The metaphors can become more complicated however, in that
there may be many attributes in common between the metaphor and
referent. Each may have similarities of structure, as in the metaphor
ARGUMENT IS WAR, although 'the structure is partial because only
selected elements of the concept WAR are used.' (Lakoff & Johnson,
1980, p.84.)
They note that enlightenment can result from the use of
some metaphors. For example, the phrase 'the solution of my problems'
(ibid. p.143) can lead to thinking about problems in terms of a
chemical which can dissolve them, but that they can also precipitate
out again. 'Much of cultural change arises from the introduction of
new metaphorical concepts and the loss of old ones. For example, the
Westernization of cultures throughout the world is partly a matter of
361
introducing the TIME IS MONEY metaphor into those cultures.' (ibid.
p.145.) Note, however, the cross-cultural work of Asch showing the use
of the same metaphors in some different cultures (Billow, 1977).
The Function of Dreaming
The features described in this chapter provide an answer to
the first question posed in the introduction to the effect that dreams
have meaning as whole pieces, rather than being collections of
separate scenes. The second question is answered with a functionalisttype discovery that dreams are the product of high level symbolic
brain activity, rather than the epiphenomenon of a functional
physiological process. However, a third question now results: given
that the dreams are meaningful, do they have a function as well?
In the four empirical chapters we have seen a multitude of
metaphorical images, coupled with the solving of minor problems in the
dreams. It would be possible to tie the production of metaphors to the
biological necessity of increased connections between similar
memories, and the active discovery of such similarities. This would be
important because Rycroft's idea of dreams as thinking that occurs at
night provides them with no biological function. The function would
then be that it is the
metaphor that will bridge two present
concerns, or a present concern and an earlier memory. For example, in
the above Konstanz dream (chp.l3) the psychotherapist (present) was
being connected to my mother (past) via the iinag of the submarine.
(Not only is the submarine a womb-like container but the waters of the
lake are reminiscent of birth.) Memories are present in dreams and can
be a fertile source of lateral and creative thinking, by putting past
362
experience at the service of present needs and problems.
This hypothesis of the phenomenology of dreams can then be
related to the natural history of dreaming. Experimentally we know
that during REM sleep animals practise behaviour programs, and
elaborate and consolidate memories (both within and between different
chronological sources). This needs some amount of narrative ability,
possibly commensurate with the animal's degree of independence from
fixed action responses. These animals thus live in the realms of the
Real and the Imaginary (to use Lacan's terminology) during waking
life, with many mammals using the hippocampus to produce cognitive
maps of the environment (O'Keefe & Nadel, 1978) and with primates
able, in addition to imaging the world, to put themselves on such
cognitive maps (Gallup, 1977). Humphrey (1980) suggests that dreams,
as well as play and parental manipulation, have the effect of
providing opportunities for introspection, and hence the ability to
form such dream images would provide an evolutionary and functional
advantage to higher species.
However, such a speculation seems unfalsifiable in that the
contribution of the dream images to our self-image is difficult to
measure, and just as difficult to separate from the contribution of
our self-image and introspection to our dreams. Given that dreams are
not only fleeting, usually unknown, and certainly present in an
altered state of mind (which works against recall in the waking
state), it could be argued that their actual contribution to our
information-processing is quite doubtful.
Therefore, that we have found them to be meaningful, as
well as showing active and recombinative thought, does not entail that
they have to be purposeful as well. For example, sugar in the blood
363
can be read as a sign of diabetes but is not caused by the need to
produce that sign. Similarly, that a metaphor is enlightening does not
mean that to produce enlightenment was its purpose. Any enlightenment
can result from a subsequent exploration of a metaphor, irrespective
of why or how it was produced initially. The reason that dreams have
been thought of as purposeful, as against the apparent
superfluouseness of day dreams, is that the physiological state
accompanying dreaming is pre-programmed, whereas boredom or automatic
activity, the usual accompanyments for day dreaming, do not show a
drive for regular occurrence.
Antrobus' (1977) paper 'The Dream as Metaphor: An
Information-Processing and Learning Model' may thus be divided into
two independent parts, for dreams may produce metaphors without having
any learning or information-processing function at all. Antrobus
writes, somewhat tautologously, that 'since the context of the
original event may be available less often during REM sleep than
during wakefulness or other sleep stages, metaphors should be
constructed more frequently in that sleep stage.' (ibid. p.333.)
However, this lack of context is not the only reason for the
prevalence of metaphors in dreams. Lakoff and Johnson (ibid. p.3)
write that 'metaphor is pervasive in everyday life, not just in
language but in thought and action.' It appears to be more prevalent
in dreams because, rather than being
spoken, the metaphors are seen.
This occurs because the scenes are hallucinatio s rather than real,
and so UP and DOWN, for example, can be seen as well as thought of.
Yet the amount of metaphor is possibly no greater than in waking life.
Antrobus is correct to see metaphor as the use of memories out of
364
context, but the striking attribute of dreams is the vividness of the
metaphors and not their prevalence in comparison with waking life.
There is a common theme between:
(1) this conclusion derived from combining the work of
Antrobus and Lakoff and Johnson,
and
(2) the empirical part of this thesis showing similarities
between waking and dreaming mentation contents,
arid (3) the empirical part of this thesis showing that REM
dreams have a logical structure, indicating active
goal - directed thought,
and
(4) the kinship between waking daydreams and night dreaming
on the level of form, as shown by the following two
studies.
Foulkes and Fleisher (1975) studied subjects in relaxed
wakefulness as a control for measurements of sleep mentation. Despite
taking steps to remove any expectation from the subjects' minds that
their experiences would be dream-like, they found that 19% of reports
were described as hallucinatory (that is, the images were thought to
be real when they were happening), and 25% were classed as regressive,
which included bizarre or fragmentary images. For 15% of reports there
was no reality content at all, 'the subject is not controlling his
thoughts, he is not aware he is in the laboratory, and his mentation
is hallucinatory' (ibid. p.69). Such mentation obviously did not last
nearly as long as it does during REM dreams. I must be emphasised
that day dreams should not immediately be cast as 'regressive': The
uses of the 'stream of consciousness' during waking life to summon up
memories and its role in working memory are both adaptive, although
365
the 90-minute bizarreness cycle of day dreaming (Kripke and
Sonnenschein, 1978) may not be.
Reinsel, Woilman and Antrobus (1986) postulated that 'the
long thematic sequences of REM sleep, in contrast to waking, are
achieved, in part, because of the high perceptual thresholds of that
state, which prevent the disruption of mentation sequences by external
stimuli.' (ibid. p.259.) They collected mentation reports from three
groups of subjects. The waking with [auditory] stimulation group had
many Topic Units and a high Total [word] Recall Count, a waking
without stimulation group had less of each, and that the REM sleep
condition had even less of each again, in an extrapolatory manner.
(The latter authors do beg the question, though, of whether noise is
the only disruptant of ongoing mentation, and, of course, the stages
of non-REM sleep also have high perceptual thresholds, although not
the same cotical activation as in the REM state.)
The continuity between dreams and waking cognition is
emphasised in both cases. Home has gone so far as to say:
'We might even begin to query whether REM sleep really
is a state of sleep, or a peculiar form of wakefulness
within sleep.'
(1988, p.289.)
The conclusion from all this is that dreams are not a
special type of thinking which has to have a special mechanism as an
explanation, be it unlearning or the comparison of memories. Dreams
appear to be further along the same continuum of use of concrete
metaphor and imagery as daydreams, evidence for this being provided by
the existence of the intermediate cases of hypnogogic imagery (Vogel,
366
Foulkes & Trosman, 1966), lucid dreaming, and non-REM dreaming. (In
the latter case, non-REM dream reports don't show any less moment-tomoment continuity than do REM dreams; Foulkes and Schmidt, 1983.)
Dreaming may simply be what the mind does when the brain is in REM
sleep, just as day-dreaming occurs when we are bored, and hypnogogic
imagery when we are drifting off to sleep. Dream images cause other
images to arise, even in the next dream period, but they may not have
any function or lasting effect, unless, like with day-dreams, we
consciously pay attention and remember a particular part of one. There
must be at least a short-term effect of having had one particular
dream or scene, however, in order for the next dream or scene to
follow on from it. The empirical evidence of this thesis cannot
differentiate between this proposal, that dream images have no lasting
effect unless consciously remembered, and the possibility described in
chapter 5 that unconscious information-processing with a long-term
effect is occurring.
Evidence provided by Singer illustrates this proposition
that individual images may be meaningful but themselves take no part
in information-processing: he shows (1981, pp.77 & 81) that daydreaming may keep us alert during monotonous tasks. If this is said to
be one function of day-dreaming it exemplifies the proposal that the
individual images may be functionless while the activity as a whole,
or its physiological counterpart, is functional. This is similar to a
conclusion about myths which is a recent extrapolation by Levi-Strauss
of his earlier theorising:
myths operate, as it were, in a hail of mirrors, but
they reflect only each other. This implies that myths - and dreams are not mechanisms for communicating messages in a secret language.
367
They are best understood as modes of thinking, or perhaps more
precisely, reflection.'
(Kuper, 1989, p.31.)
From this conclusion it is possible to provide some
solution to the tension between physiology and psychology which has
pervaded this thesis, the study of which first conferred
respectability on the subject of dream research. The elements of
dreams appear now to be based upon (1) each other, and (2) the
linguistic metaphors, live and dead, present in the dreamer's culture,
and so hopes of tying the images down to physiology look misguided.
Tying the images up to psychology and cultural presuppositions,
however, may be quite fruitful.
The Physiological Basis of Dream Cognition
This thesis started with an analysis of research connecting
the physiology of REM sleep to the phenomenology of the accompanying
dream. To be precise, it was the form of the dream experience that was
important; for example, number of scenes, number of characters,
temporal age of incorporated memories, vividness of imagery, amount of
physical movement, etc. Despite the hopes of Hobson and McCarley
(1977), who aimed to tie specific activities in dreams, such as
flying, with isomorphic physiological conditions, the physiological
work could only be related to overall properties of the imagery, and
not to specific thoughts and images themselves.
In order to provide more empirical evidence about the
activity of dreaming the content of four nights of dreams were then
analysed. This was found to bring us back to the original question of
form, because it was found that the content of the dreams was
368
following a concretizing and dialectical argument, which constrained
the actual content that was included in the dreams. Rather than
content being a random filler of the frames of the dream, it was shown
to be constrained not only by the content of the waking concerns of
the subject, but by a particular method of thinking.
Day
Dr e am
Dream
Content
Content
Concerns
Day
Concerns
Form 2
Form 1
Form 1
1'
Physiology
Physiology
(Arrows denote some degree of correlation, which must still
be determined.) Form 1 is the collection of dream properties stated in
chapters one and two, whereas form 2 is the collection of additional
dream properties found in the four dreams of KJ; that is,
concretization, progressive movement, and dialectical argumentation.
It is necessary to note, however, that for some subjects there may
sometimes, or even always, not be this effect of form on content. That
it has been shown in the case of KJ is rather an important
falsification of the theories that deny active thought during REM
sleep completely.
To find this active thinking some components of dream
cognition had to be analysed, just as Freud took apart the content of
dreams to find condensation and displacement. Similarly, concerning
the history of cognitive science, Luria writes that 'the acceptance of
thinking as a once-and-for-all indivisible act, which can be described
369
by subjective methods and which cannot be broken down into its
components, was a retrograde step compared with associationism,
closing the door to its scientific investigation.'
(1973, p.325.)
This introduction of a second type of form to the
characterisation of dreams is important because it leads us out of the
false content versus process dilemma proposed by Foulkes (1985,
pp.l96-2O4), which he uses to advocate the study of dreaming as
opposed to dream content. He starts by noting that the use of
associations to the images may or may not tell us of their sources
(ibid. p.196) and adds that our ignorance of the sources is further
added to in that 'little dream variation seems to be explained by the
manipulation or observation of waking variables' (ibid. p.198), and
that 'once we acknowledge that the sources of dreams are mnemonic
rather than behavioral ... the prediction of dream content becomes
much more complex.' (ibid.) He wittily sums this up as 'this is not
like predicting what you'll consciously think the first time you see
Paris; this is trying to predict what you'll consciously think,
period, when you're not intentionally regulating your own conscious
thought.' (ibid. p.200.)
He takes this as evidence that 'since it seems that the
activation of mnemonic elements during dreaming and their selection
for dream processing is random and arbitrary, it's not likely that the
particular content of our dreams - in and of themselves - serve any
adaptive function.' (ibid.) He then states that th visual form, (for
example, the presence of some elements in dreams as parallel to some
of waking life, and some as not parallel) means that there is some
predictability about the characteristics of all dreams, and that the
370
reasons for these formal properties must be discovered. He then
emphasises processes rather than content, dreaming rather than dreams.
However, this thesis has shown that another form, or processing, is
involved in dreaming, that is, the use of concrete oppositional
metaphors, with their progressive and dialectical argumentation.
Although we are then unable to predict content, we may be able to
predict the types of content to some extent.
Foulkes makes the point that from a knowledge of form (e.g.
the presence of long past and recent memories) we can construct
possible functions of dreaming (e.g. 'the integration of recent
evocative memories with older evocative memories', ibid. p.202), with
the emphasis that these are just possibilities. Similarly, it must be
emphasised that to discover concrete oppositional metaphorisation in
dreams does not prove that that activity is their fwwtion - that niay
rather be one way that the mind works when it is not functioning
purposefully. That this is functional also remains only a possibility.
Unconscious abilities to engage in concretization and to
derive dialectical arguments were found, but there is no reason to
expect that they have a physiological basis simply because they seem
to hold irrespective of content in this case. Although Luria (1973,
p.340) analyses the form of various mental activities (i.e.
operations, teleological programming, final checking) and connects
these types of activities with brain physiology, the functions he uses
are much grosser and broader than those of concretization and
dialectical argumentation.
This psychological activity cannot then answer Leach's
question 'in what way is the Culture of Homo sapiens inseparable from
the Nature of humanity' (1970, p.112. ) In other words, how are our
371
ways of thinking, our cultural methods of classification and
calculation, for example, which seem to deny our physical nature and
set us apart from the animals, actually a product of our brain's
physical nature. Leach proceeds to propose that, regarding the use of
binary features, 'I can see no reason to believe that they are human
universals'
(ibid. p.113).
Some parts of the form of dreams have known physiological
correlates, such as dream vividness, and a bridging law may be
proposed. There is no certainty for a physiological basis for, say,
Verdone's (1965) result, although some proposal would be within the
bounds of speculation. However, results such as that of concretization
are even further from the reach of physiology and of bridging laws.
Such a conclusion, that mental events are explicable in terms of other
mental events, just as physical events cause physical events, is in
line with the functionalist approach that 'all kinds of physically
different things could have human software' (Fodor, 1981, p.118).
Dream research is thus left in much the same position as studies on
the stream of consciousness when it comes to the inadequacy of current
studies to draw any parallel between physiology and mentation content.
This conclusion is further reinforced by evidence mentioned earlier
about the similarity of dreams to waking day-dreams. Even if REM sleep
is eventually explained by non-information-processing theories such as
that of catecholamine repletion, or brain excitiation reduction, as
described in chapter 4, or the sentinel hypothesis dreams could still
show a faithfu', if pointless, representation of waking life.
This conclusion does not then go as far as the denial by Bannister
(1968) of any connection between physiology and psychology, but is
372
similar to Skinner's (1950) claim that physiology is irrelevant to
psychology in the construction of psychological laws. Theories of
dreaming are in the same position of the search for relevant variables
as was learning theory in the 1950s.
'There seems to be no a priori reason why a complete
account [of animal learning] is not possible without appeal to
theoretical processes in other dimensional systems.' (ibid. p.2l5.)
Any explanations of the content of dreams will therefore
need an account of the use of signs, and a comparison with other sign
systems, to which we now proceed.
Dreams and Other Human Languages
Whatever the function of REM sleep in animals, for us dreams
are just one of many human representations of the world. They can be
differentiated from the other human representations along two
dimensions. Firstly, the linearity of the product, the presence of
time in its portrayal, whether it has a beginning and end or whether
it leads onto what is represented next. Secondly, the arbitrariness of
the link between sign and referent - is it a convention or is it
constrained by what is represented? (These links would be connotative
or denotative.) This latter distinction of sign and referent is well
known in philosophy, but should not be confused with de Saussure's
differentiation of signified and signifier, the first of which is
acoustic, the second semantic, while both of them are mental items.
373
Photos, paintings Films, plays Stream of Consciousness
decreased
Semantic memory
Dreams
Episodic memory
arbitrariness
Formulae
Literature,
Natural language
I of signs
myths, music
increased linearity of the representation
(This use of the dichotomies arbitrary/nonarbitrary and
diachronic/synchronic in fact illustrates a central structuralist
thesis, that it is useful, and possibly necessary, when considering a
concept, to also consider its opposite.)
The first column lists representations that are summaries,
extractions from a larger amount of experience.
Following the discovery that elements of a dream constantly
refer to, amplify, permute, repeat or presage other elements, as
occurs in poems, symphonies and plays, dreams can be put in the second
column, which lists representations that are whole pieces. Although
related in time they are constrained by beginning and end.
The third column consists of the bases of human culture,
those which make us different from animals: yet these are resident
within the person, as opposed to being a tool or artefact, like books
or photographs.
The horizontal ranks are differentiated in that the bottom
rank contains what Peirce called 'symbols', which have no natural link
with their content (even in the case of onomatopoeic words), the
374
middle rank has 'indices', which have natural motivation but are open
to interpretation, and the top rank has 'icons', which are each found
to refer 'to the object that it denotes merely by virtue of characters
of its own...'
(Peirce, 1978, p.143).
All members of the above chart are language systems in the
sense of being representations of the outside, and personal, worlds.
Even memory and consciousness are to some extent users of symbolism,
of signs, despite their felt immanence. Formulae have components which
represent measurable factors in the world, with some valid symbols
being absent from any particular formula, and such symbols also
appearing in other formulae which then elaborate their definition and
attributes. Episodic memory similarly uses signs, which are stylised
representations of events. The actual event no longer exists but
leaves its mark in the brain. Each component of episodic memory,
however, is not an uninterpreted snapshot. Usually they must be
reconstructed, partly with the use of deduction. Episodic memory
leaves out much of what was present at the time of the referred-toevent, we can imagine putting new elements in (think of your breakfast
table this morning - add five milk bottles to it), and it has elements
which are interpreted beforehand and which can be summoned up. The
analogy with spoken sentences, and hence with language, is apparent.
Structuralism is the study of signs in elation to other
signs, and in relation to what each sign does not refer to, and it is
these signs, and the rules by which they are put together, that
constitutes a language. We note that, in the above chart, dreams are
found between four separate pairs of languages. This positioning forms
the basis for four
theories of
dreaming:
375
1)
Semantic memory - Dreaming - Episodic memory
Evans' (1983) computer theory proposed that at night we
practise day-time routines, eliminating unnecessary parts and making
the routine more automatic. Such a practising is entailed in the
transition from episodic to semantic memory, to make the former more
general. Dreams can also take a semantic fact (e.g. this is what my
house looks like) and build an episode around it. Dreams show us
episodes, or events, which are based on semantic, general knowledge,
and on events of the dream-day.
This activity can be related to another Structuralist
distinction, that of langue and parole, that is, language as a whole
system and each individual speech act. Each of these cannot exist
without the other - language cannot be studied apart from individual
speech episodes, and individual speech acts cannot occur without
knowledge of part of the community's whole language system. Now,
individuals in the language community are each dependent on the
language as a whole, but do have the effect collectively by means of
many speeches of slowly changing the language over time. The former
are instances of, and dependent upon, the latter, but do have the
effect of incremental changes over time. This interrelation and
dependence of parole and langue is paralleled by theories of dreams
which claim that the single event or episode of having a dream affects
the subject's personality and more long-term schemes. Much work has
been done on the consolidation of memories during REM sleep (Li &
Shao, 1981; Pearlman, 1983; Palombo, 1976) and on the possible forming
of new connections between memories, resulting in deeper encoding.
(See chapters 5 & 8.) Such hypotheses of dreams having some
376
information-processing effect on the dreamer do provide an
evolutionary rationale for them, whether the hypotheses are true or
not, but the problem of evaluating these theories is glaring. The crux
of the problem has been elaborated earlier in this chapter, and in
chapter 13, with respect to Cartwright's (1986) paper showing a
correlation between types of dreams and waking life personality and
personality changes. It may be impossible ever to evaluate whether it
is the waking personality and events that cause the dream, or whether
the dream can have any reciprocal effect on waking life.
2)
Films, plays - Dreams - Poetry, symphonies
Hudson's (1985) description of dreams shows their
resemblance to these contrasting examples of artistic pieces. He sees
the dream as a meaningful whole which is played to an audience of one.
'As a metaphor, the cinema has deep attractions. It enables us to
distinguish clearly between the film studio, the world of preparation
and construction, and what happens in the spectator's mind once the
film in its finished state has hit the screen - the screen itself
constituting the boundary or interface between one world and the
other. On this analogy, Freud's dream theory is seen as a director's
theory; a question of scenarios, shooting scripts, and footage left on
the cutting-room floor. ..' (ibid. p.154)
3)
Photos, painting - Dreams - Natural language
Finding itself between these two gives the dream its
features of puns and condensation. Dreams are a mixture of direct
377
representations and corrupted arbitrary representations. For example,
the use of the pun 'Lake Konstanz' - 'constant' in my dream mentioned
in chapter 12. An event can be translated into language which is then
manipulated, the result is then translated back and replayed as a
picture of events. Freud catalogued many examples of such
manipulation, for example the botanical monograph dream. Note that
photographs and paintings are examples of one-way communication, while
natural language is mostly a two-way communication; dreams are somehow
in between, from oneself to oneself.
4)
Formulae - Dreaming - Stream of Consciousness
That whole areas of experience can be related to other
areas, that this relation can be succinctly expressed, and that this
expression may be reduced to an algebraic relation (such as A/not A
B/not B) was studied by Levi-Strauss in his application of
Structuralism to anthropology and mythology. He claimed that elements
of the world cannot be thought about except in terms of their
opposites (e.g. plenitude thought of in terms of hunger), and also
that each opposition can be thought about in terms of other
oppositions, whether geographical, natural, cultural, or whatever.
Whereas in the case of formulae it is obvious that signification is
being used, it is less obvious in some dream , and is rarely
recognised in the apparently immanent experience of consciousness. The
post-structuralist writer Derrida has written of the illusion of
believing that presence precedes signification, which he says has
dominated Western philosophy for two thousand years. He says that it
is writing that first gives us an inkling of the mistake of believing
378
that spoken sequences of words, or sequences of thoughts, are capable
of abstraction from time - for it is writing that can be more
permanent than, and detached from, its author. Sturrock writes of this
view: 'We cannot be fully "present" even to ourselves in so far as we
must of necessity commune with ourselves in a system of signs that is
not ours alone but a social institution. "Hearing ourselves speak" is
the illusory model of intimacy and immediacy which Derrida suggests
has enabled us conveniently to ignore the true nature of signs. It is
a model which shortens the circuit of communication so much that it
may no longer appear as a circuit...' (1986, p.143).
The eight language systems around dreaming in the diagram
are each mainly constrained by either nature or culture, perception or
thought.
Photos - Films
Cs.nesst
- - - i
constrained by nature,
S.Memory I Dreams
E.Memory1
4... - -
Formulae
& perception
Literature N.Language
constrained by
culture, thought
All nine are language systems in that they contain a
selection of the total number of possible sig s, and use this
selection to represent real or imagined events of waking life. For all
four of the above pairs, where the dream is between two language
systems, the dream is a meeting place for culture and nature, and for
thought and perception. It achieves this by having its feet in both
camps, by being at the same time perception and thought. In dreaming
379
we think what we see, and see our thoughts, both at the same time.
Similarly, de Saint-Denys (1982 edition, p.34) noted that: 'the image
connected with each idea appears as soon as the idea arises. ... The
image of the dream is to the idea which calls it forth exactly what
the image of the magic lantern is to the lighted glass-plate which
produces it.'
This positioning of dreams with respect to other languages
says nothing about why this is so, or about the truth or falsity of
the above four types of theory which can be related to this
positioning. I am placing it amongst other examples of narrative and
representation and emphasising that dreaming has a narrative aspect,
the study of which, as well as the application of Levi-Strauss' theory
of mythic narratives, may then tell us why dreams have a narrative.
380
POSTSCRIPT - SUGGESTIONS FOR FUTURE RESEARCH
As well as the attempted replication of the present contentanalytic study with different subjects I suggest finding further
evidence for the transformational, narrative and functional aspects of
dreams by the following quantitative work.
After each dream period subjects will be woken and a
complete report on the dream will be obtained. In the morning the
subject should be asked to associate to each thematic unit (Cicogna,
Cavallero & Bosinelli, 1986) with instructions to pay attention to any
memories that may have been incorporated, either into the foreground
or the background of the scenes. The theoretical background of the
methodology of using post-dream associations in order to discover predream sources of the dream is described by Cavallero and Cicogna
(1984). The search for memories provoked by single incidents is
described by Rubin, Wetzler & Nebes (1986, pp.202-203) in work on
recall of autobiographical memory across the lifespan. The subject
will rate the date of the episodic-memories associations and also
group the associations according to present concerns that they may be
relevant to. Within each concern group the associations will be rated
by being placed in order of decreasing relevance to the particular
concern.
The concern-relevance order will then be compared by ANOVA
to the chronological order of the thematic units which provoked the
associations. The information-processing paradigm would predict either
a preponderance of concern-related associations at the start or at the
end of the night's dreaming. The associations can be divided into
strict episodes, abstract self-references, and semantic traces (after
381
Cicogna, Cavallero & Bosinelli, 1986) in order to discover if episodic
or semantic associations are produced by dreams from a particular time
of night, with the assumption that this will provide evidence as to
whether particular dream periods have differential access to semantic
as opposed to episodic memories, or vice versa.
There has been found a progressive resolution of problems in
dreams (for example, Verdone, 1965; work cited by Cartwright, 1977;
this thesis) through the night and I hence predict either an increase
of relevant memories through the dreams of the night, or the inclusion
of progressively earlier memories - either of which will indicate the
information-processing ability of dream images, as opposed to
information-processing being afforded by the REM state alone.
With the associations for each dream listed together the subject
concerned will then rate which associations are similar to other
associations. If similar associations are produced more often by
temporally close thematic units and scenes than by distant ones this
will be evidence that memory sources used during dreams are not
activated at random. Obviously this hypothesis relies completely on
the assumption of Cavallero and Cicogna (1984) that dream associations
provide information about the sources of a dream.
When analysed closely the result of Verdone (1965) relies
greatly on the difference in memory dates between the contents of the
first and second REMPs, with little change between the second and
subsequent ones. Furthermore, the regimen of waking subjects 5 or 12
minutes after the start of the REMPs could have caused REM
deprivation, and hence enhancement of dream vividness, which itself
382
could have given rise to more remote memories being accessed. For this
experiment it will therefore be necessary to wake subjects as near to
the end of a REM period as possible.
In addition to the above comparison of age of memories
across dreams I suggest comparing the memories present within REM
dreams with those present in day-dreams. The methodology for the
collection of such reports is given in Foulkes and Fleisher (1975). As
reviewed in the last chapter, the topographical and formal regressions
found in dream images are also present in day-dreams. The next task is
to find any similarity or difference between temporal regression of
REM and day-dreams. The work of Battaglia, Cavallero & Cicogna (1987)
on the presence of day-residues, recent-residues and remote-residues
in sleep onset mentation and REM dreams can be summarised thus:
Frequent
Less frequent
By extrapolation I
REM
SO
Occurrence
Day-, Recent-,
Remote-,
Recent-,
Remote-, Day-.
predict that for day-dreatus the freguency
of occurrence of the three types of temporally referring associations
will be:
Day-residues > Recent-residues > Remote-residues.
383
REFERENCES
Ackley D., Hinton G. & Sejnowski T.
A learning algorithm for
Boltzmann machines. Cognitive Science (1985),9:147-170.
Creativity and the dream.
Adelson J.
Merrill-Palmer Quarterly
(l959),6: 91-97.
Adey W., Bores E. & Porter R. EEC sleep patterns after high cervical
lesions in man. Archives of Neurology (l968),19:377-383.
Adrien J. Lesion of the anterior raphe nuclei in the newborn kitten,
and the effects on sleep. Brain Res. (1976), 103:579-583.
Adrien J.
factors.
Ontogenesis of PGO activity in the kitten: epigenetic
(1977)
Sleep 1976. 3rd European Congr. Sleep Res.,
Montpellier 1976, pp.3-7. Karger, Basel.
Agnew H., Webb W. & Williams R. The effects of stage 4 deprivation.
Electroencephalog. Clinical Neurophysiology (1964),17:68-70.
Agnew H., Webb W. & Williams R. Comparison of stage 4 and 1-REM sleep
deprivation. Perceptual and Motor Skills (l967),24:851-858.
Albus J. A theory of cerebellar function. Mathematical Biosciences
(1971) ,lO:25-61.
Alexander F.
Dreams in pairs.
In: Fliess R.
The Psychoanalytic
Reader. (1950) pp. 336-342. Hogarth Press, London.
Allen S., Oswald I. & Tagney J. The effects of distorted visual input
on sleep.
Report to the 1st. international congress of the
Association for the Psychophysiological Study of Sleep, June 1971,
Bruges, Belgium.
Allen S., Oswald I., Lewis S. & Tagney J. The effects of distorted
visual input on sleep. Psychophysiology (1972),9:498-504.
Ailport G.W.
Pattern and Growth in Personality.
384
(1961) New York:
Holt, Rinehart and Winston.
Aliport G. The general and the unique in psychological science. J.
of Personality (1962), 30:405-422.
Aliport C. Letters from Jenny.
(1965, interpreted and edited by C.
Ailport.) Harvest/HBJ SanDiego.
Antrobus J.
The dream as metaphor: an information processing and
learning model. J. of Mental Iniagery (l977),2:327-338.
Antrobus J. Dreaming for cognition. In A.Arkin, J.Antrobus, & S.Ellman
(eds.), The mind in sleep: Psychology and psychophysiology (1978)
(pp.569-581). Hilisdale, NJ: Lawrence Erlbaum.
Antrobus J.
Neurological approach - the hemisphere-dreaming
controversy: resolution of a paradox.
(1984) In: Psychology of
Dreaming, M. Bosinelli & P. Cicogna (eds.), CLUEB, Bologna, Italy.
Antrobus J.
Dreaming: cortical activation & perceptual thresholds.
J.of Mind & Behavior(1986),7(2&3):l93[63]-2l2[82].
Antrobus J.
Cortical hemisphere asymmetry and sleep mentation.
Psychological Review (1987) ,94:359-368.
Antrobus J., Ehrlichman H. & Weiner M.
NREM: failure to replicate.
EEG asymmetry during REM and
Sleep Res. (1978), 7:24.
Antrobus J., Ehrlichman H., Weiner M. & Wollman M.
The REM report
and the EEC: cognitive processes associated with cerebral hemispheres:
their ratios and sums, cross trial contrasts and EE window duration.
Sleep (1983),82:49-5l.
Apter M. & Rushton C. (eds.) Reversal Theory and Personality. (1981)
South West Inter-Clinic Conference, Wiltshire.
Archard D. Consciousness and the unconscious. (1984) Hutchinson.
Arkin A. Night-terrors as anomalous REM sleep component manifestation
385
in slow wave sleep. Waking and Sleeping (1978),2:143-147.
Arkin A. , Hastey J. & Reiser M. Post-hypnotically stimulated sleep
talking. J.of Nervous and Mental Disease (1966),l42:293-309.
Aserinsky E.
Physiological activity associated with segments of the
rapid eye movement period.
(eds.)
In: Kety S., Evarts E. & Williams H.
Sleep and altered states of consciousness.
(1967, pp.333-
350.) Williams & Williams, Baltimore.
Atwood C. & Stolorow R. Structures of subjectivity: explorations in
psychoanalytic phenomenology. (1984) Hillsdale N.J., Atlantic Press.
Austin M. Dream recall and the bias of intellectual ability. Nature
(1971) ,23l:59-60.
Badcock C.R.
Levi-Strauss; structuralism and sociological theory.
(1975) Hutchinson & Co., London.
Baddeley A.
The psychology of memory.
(1976) Basic books, NY.
Baekeland F., Koulack D. & Lasky R. Effects of a stressful presleep
experience on electroencepholograph-recorded sleep.
Psychophysiology
(1968) ,4:436-443.
Bannister D.
The myth of physiological psychology.
Psychol. Soc.
(1968),21:229-231.
Barthes R.
Critical Essays.
1972 translation, Northwestern
University Press. (Originally published 1964 in French.)
Barthes R. Mythologies.
Bull. Br.
(a)
1972 translation, Jonathen Cape, London.
(Originally published 1957 in French.)
(b)
Battaglia D., Cavallero C. & Cicogna P. Temporal reference of the
mnemonic sources of dreams.
Perceptual & Motor Skills (1987),64:979-
983.
Baylor G. & Deslauriers D. Understanding dreams: method, maps and
metaphor. Dreamworks (1986) ,5:46-57.
386
Baylor C. & Deslauriers D.
background and theory
Dreams as problem solving: part 1:
Imagination, Cognition and Personality (1986-
87) ,6:l05-ll8.
Beck S.
The science of personality: nomothetic or idiographic.
Psychological Review (1953), 60:353-359.
Belicky D. & Belicky K. Nightmares in a university population. Sleep
Research (1982) ,l1:116.
Belicky K. & Bowers P. The role of demand characteristics and
hypnotic ability in dream change following a presleep instruction.
J. of Abnormal Psychology (1982),91:426-432.
Benson K. & Zarcone V.
Middle ear muscle activity during REM sleep
in scizophrenic, schizoaffective, and depressed patients. Am. J. of
Psychiatry (1982) ,l39:1474-l476.
Benson K. & Zarcone V. Testing the REM sleep phasic event intrusion
hypothesis of schizophrenia. Psychiatry Research (1985),15:163-173.
Benvenuto R. & Kennedy R. The works of Jacques Lacan. (1986) Free
Association Books, London.
Berger R. Experimental modification of dream content by meaningful
verbal stimuli. Brit.J.Psychiatry (1963),109:722-740.
Berger R.
Oculomotor control: A possible function of REM sleep.
Psychological Review (1969) ,76:l44-164.
Berger R. & Oswald I. Eye movements during active a d passive dreams.
Science (1962) ,137:60l.
Bergin A. & Strupp H.
Changing frontiers in the science of
psychotherapy. (1972) New York: Aldine-Atherton.
Berlucchi C., Morruzi C., Salvi C. & Strata P. Pupil behavior and
ocular movements during synchronized and desynchronized sleep.
387
Archives Italiennes de Biologie (1964),102:230-244.
Bertini M., Lewis H. & Witkin H. Some preliminary observations with
an experimental procedure for the study of hypnagogic and related
phenomena. In: Tart C. (ed.) Altered States of Consciousness. (1969)
John Wiley and Sons. N.Y.
Bertini M., Violani C., Zoccolotti P., Antonelli A. & Di Stefano L.
Performance on a unilateral tactile test during waking and upon
awakening from REM and NREM sleep. pp.383-385 In: Koella W. (ed.)
Sleep 1982, Basel: Kager (1983).
Billow R.
Metaphor: a review of the psychological literature.
Psychological Bulletin (1977) ,84:81-92.
Black A.
The natural history of obsessional neurosis.
In:
Obsessional States. (ed. H.Beech) 1974, Methuen, London.
Blagrove M. A further application of structural analysis to a dream
series. (1988) Unpublished.
Bokert E. The effects of thirst and related auditory stimulation on
dream reports.
Paper presented to the Association for the
Psychophysiological Study of Sleep, Washington D.C. (1965).
Bokert E.
The effects of thirst and a related verbal stimulus on
dream reports. Dissertation Abstracts (1968), 28:122-131.
Bolton N.
The psychology of thinking.
(1972) Methuen, London.
Boon J. From Symbolism to Structuralism. (1972) Blackwell, Oxford.
Boon J. Chapter on Claude Levi-Strauss in The Return of Grand Theory
in the Human Sciences. ed.Skinner Q. (1985) Cambridge University
Press.
Bowers M., Hartmann E. & Freednian D.
Sleep deprivation and brain
acetylcholine. Science (1966),l53:1416-1417.
Bowie M. Chapter on Jacques Lacan. In: Structuralism and Since. Ed.
388
Sturrock J. (1979) Oxford U.P.
Bowker R. & Morrison A.
hyperalertness.
The P00 spike: an indicator of
(1977) In: Koella W. & Levin P. (eds.) Sleep 1976.
Karger, Basel. pp.23-27.
Breger L.
Functions of dreams.
J.Abnormal Psychology (1967),
Monograph no.641.
Breuer J.
Anna 0.
(1895) In: Standard Edition of the Complete
Psychological Works of Sigmund Freud, (1955) Hogarth Press, London.
Briggs G.
Acquisition, extinction & recovery functions in
retroactive inhibition. J.of Experimental Psychology (l954),47:285293.
Brooks D. The P00 phenomenon. Sleep 1976, 3rd. European Cong. Sleep
Res., Montpellier 1976, pp.15-18. (Karger, Basel 1977).
Brooks D. & Gershon M. Eye movement potentials in the oculomotor and
visual systems of the cat: a comparison of reserpine-induced waves
with those present during wakefulness and rapid eye movement sleep.
Brain Res. (1971), 27:223-239.
Broughton R. & Mamalek M.
The effects of nocturnal gamma-
hydroxybutyrate on sleep/waking patterns in narcolepsy/catalepsy.
Can. J. of Neuroscience (l980),7:23-3l.
Broughton R., Poire R. & Tassinari C. The electrodermogram (Tarchanoff
effect) during sleep. Electroenceph. & Clinica Neurophysiology
(1965) ,l8:691-708. (a)
Broughton R., Poire R. & Tassinari C.
effect) during sleep.
Electrodermogram (Tarchanoff
Perceptual & Motor Skills (1965),20:l81-l82.
(b)
Brown R. & Donderi D.
Dream content and self-reported well-being
389
among recurrent dreamers, past-recurrent dreamers, and nonrecurrent
dreamers. J.Personality & Social Psychology (1986),50(3):612-623.
Campbell D. & Stanley S. Experimental and quasi-experimental designs
for research.
(1963) Chicago: Rand McNally.
Campbell J. Grammatical man. (1984) Pelican, London.
Canon W. & Washburn A. An explanation of hunger. American Journal of
Physiology (1912), 29:441-454.
The case for single-patient studies.
Caramazza A. & McClosky M.
Cognitive Neuropsychology (1988), 5:517-528.
Cartwright R.
The influence of a conscious wish on dreams: a
methodological study of dream meaning and pre-sleep stimuli.
Abnormal Psychology (1974),83:387-393.
(a)
Problem solving: waking and dreaming.
Cartwright R.
J. of Abnormal
(b)
Psychology (l974),83:45l-455.
Cartwright R.
J.
Night Life, Explorations in Dreaming.
(1977),
Englewood Cliffs N.J., Prentice-Hall.
Cartwright R. REM sleep characteristics during and following mooddisturbing events. Arch. of General Psychiatry (1983),40:197-201.
Cartwright R.
point of view.
Affect and dream work from an information processing
J.of Mind and Behavior (1986),7(2&3):4ll[281]-
428[298]
Cartwright R. (1988) Personal communication.
Cartwright R. & Monroe L. Relation of dreaming and REM sleep: The
effects of REM deprivation under two conditions. J.Personality and
Social Psychology (1968),1:69-74.
Cavallero C. & Cicogna P.
research.
Models and strategies of sleep mentation
In: Psychology of Dreaming. Bosinelli M. & Cicogna P.
(eds.) (1984) CLUEB, Bologna.
390
Cespuglio R. & Valatx J. Genetic factors of the phasic phenomena of
paradoxical sleep. (1977) Sleep 1976. 3rd European Congr. Sleep Res.,
Montpellier 1976, pp.7-iO. Karger, Basel.
Chase M. (ed.) The sleeping brain. Brain Information Service/Brain
Research Institute. Los Angeles (1972).
Chernik D.
humans.
Effect of REM sleep deprivation on learning and recall by
Perceptual & Motor Skills (1972),34:283-294.
Cicogna P., Cavallero C. & Bosinelli M. Analyzing modifications across
dream reports. Perceptual and Motor Skills (l982),55:27-44.
Cicogna P., Cavallero C. & Bosinelli M. Differential access to memory
traces in the production of mental experience. mt. J. of
Psychophysiology (1986) ,4:209-216.
Cipolli C., Fagioli I., Maccolini S. & Saizarulo P. Associative
relationships between pre-sleep sentence stimuli and reports of mental
sleep experience.
Perceptual & Motor Skills (1983),56:223-234.
Clark H. Word associations and linguistic theory. In: New Horizons in
Linguistics, ed. Lyons J. (1970) Baltimore, Penguin Books.
Clark J., & Winston J. & Rafelski J.
networks.
Self-organization of neural
Physics Letters (1984),lO2A(4):207-2ll.
Clark J., Rafeiski J. & Winston J.
Brain without mind: computer
simulation of neural networks with modifiable neuronalinteractions.
Physics Reports (l985),l23:216-273.
Clausen J., Sersen E. & Lindsky A.
Sleep patterns in mental
retardation: Down's syndrome. Electroencephalography & Clinical
Neurophysiology(1977),43(2):187-l91.
Clemente C., Purpura D. & Mayer F. (eds.)
nervous system. (1972) Academic Press, N.Y.
391
Sleep and the maturing
Perceptual and Motor
Cohen D. Dream recall and total sleep time.
Skills (1972) ,34:456-458.
Cohen D. Sleep and Dreaming. (1979) Pergamon, N.Y.
Cohen H. Sleep and self-stimulation in the rat. In: M.Chase
(ed.) Sleep Research vol.4, (1972) Brain Information Service/Brain
Research Institute, Los Angeles.
Cohen H. & Dement W. Sleep: changes in threshold to ECS in rats after
deprivation of PS. Science (1965),l50:1318-13l9.
Cohen H. & Dement W. Sleep: suppression of REM phase in the cat after
ECS. Science (l966),l54:396-398.
Cohen H., Duncan R. & Dement W. The effect of electroconvulsive shock
in cats deprived of REM sleep. Science (l967),156:1646-1648.
Cohen P. Theories of myth. Man (1969),4:337-354.
Collins A. & Quillan M. How to make a language user. In: E.Tulving &
W.Donaldson (eds.) Organization of memory.
(1972) Academic press,
New York.
Cooper L., Liberman F. & Oja E. A theory for the acquisition and loss
of neuron specificity in visual cortex. Biol.Cybernetics (l979),33:9-
28.
Craik F. & Lockhart R.
research.
Levels of processing; A framework for memory
J. of Verbal Learning and Verbal Behavior (1972),ll:67l-
684.
Crews F. In the big house of theory. [Critical review of 'The Return
of Grand Theory in the Social Sciences, ed. Q.Skinner.] The New York
Review 29/5/86 pp.36-42.
Crick F. Memory and molecular turnover. Nature (l984),312:1O1.
Crick F. & Mitchison G.
The function of dream sleep.
(1983) ,3O4:lll-ll4.
392
Nature
Crick F. & Mitchison C.
REM sleep and neural nets. J. of Mind &
Behavior (1986),7(2&3):229[99]-250[120].
Dave R.
Effects of hypnotically induced dreams on creative problem
solving. J. of Abnormal Psychology (1979),88:293-302.
de Bono E. The mechanism of mind. (1969), Pelican Books, London.
de Bono E. Lateral thinking. (1970), Pelican Books, London.
de Koninck J.
Intensive language learning & REM sleep.
Sleep
Research (1978) ,7:146.
de la Pena A., Zarcone V. & Dement W. Correlation between the measure
of rapid eye movements of wakefulness and sleep. Psychophysiology
(1973) ,lO:488-500.
Dement W. The effect of dream deprivation. Science (1960),l31:17051707.
Dement W. Some must watch while some must sleep. (1976) W.W.Norton,
N .Y.
Dement W. & Kleitman N. Cyclic variations in EEC during sleep and
their relation to eye movements, body motility and dreaming.
Electroencephalography & Clinical Neurophysiology (1957) ,9: 673-690. (a)
Dement W. & Kleitman N. The relation of eye movements during sleep to
dream activity: An objective method for the study of dreaming.
J.Experimental Psychology (1957) p 53:339-346.
(b)
Dement W. & Wolpert E. The relation of eye movem Lt5, body motility
and external stimuli to dream content. J.Experimental Psychology
(1958),55: 543-553. (a)
Dement W. & Wolpert E. Interrelations in the manifest content of
dreams occurring on the same night. Journal of Nervous and Mental
Disease (1958), 126:568-578. (b)
393
Dennett D.
Brainstorms: philosophical essays on mind and psychology.
(1979) Harvester Press, N.Y.
de Saint-Denys H.
Dreams and how to guide them.
(ed. M.Schatzman)
(1982, originally published in French 1867.) Duckworth, London.
Dewan E.
The programming (P) hypothesis for REMs. (1969) Physical
Science Research Papers, no.388. Air Force Cambridge Research
Laboratories, Project 5628.
Dewson J.
A central neural change coincident with REM deprivation in
the cat. Paper presented at the 5th annual meeting of the
Association for the Psychophysiological Study of Sleep, Washington
D.C., March 1965.
Dixon N. & Henley S.
Without Awareness. (1980) In M.Jeeves (ed)
Psychology Survey No.3, London, Allen & Unwin.
Docherty M., Bradford H. & Jan-Yen Wu. Co-release of glutamate and
aspartate from cholinergic and CABAergic synaptosomes.
Nature
(1987) ,330:64-66.
Domino C.
Primary process thinking in dream reports as related to
creative achievement.
Douglas M.
J.Con. & Clin. Psychology (1976) ,44:929-932.
Implicit meanings - essays in anthroplogy.
(1978)
Routledge & Kegan Paul.
Drucker-Colin R., Spanis C., Cotman C. & McGaugh J.
Proteins and REM
sleep. (1975) In: P.Levin & W.Koella (eds.) Sleep 1974: Instinct,
Neurophysiology, Endocrinology, Episodes, Dreams, Epilepsy and
Intracranial Pathology. (1975) S.Karger, Basel.
Duffy F. & Lombroso C.
Electrophysiological evidence for visual
suppression prior to the onset of a voluntary saccadic eye movement.
Nature (1968) ,2l8:1074-l075.
Dukes W. Nl
Psychological Bulletin (1965), 64:74-79.
394
Durndell A. & Wetherick N.
The relation of reported imagery to
cognitive performance. Br.J.Psychology (1976),67:501-506.
Eagle M. & Ortof E. The effect of level of attention upon 'phonetic'
recognition errors. J.Verbal Learning and Verbal Behaviour
(1967) ,6:226-231.
Language and dreaming: the interpretation of dreams
Edelson M.
revisited. Psychoanalytic Study of the Child (1972),27:203-282.
Edelson M. Language and Interpretation in Psychoanalysis. (1975) New
Haven: Yale Univ. Press.
Evans C. & Newman E.
Dreaming: an analogy from computers.
New
Scientist vol.24, no.419, pp.577-579, 26/11/64.
Evans C.
Landscapes of the night.
(1983) N.Y. Viking Press.
Evans F., Gustafson L., O'Connell D., Orne M. & Shor R.
Response
during sleep with intervening waking amnesia. Science (1966),152:666667.
Eysenck H. Decline and fall of the Freudian empire. (1986) Penguin.
Farah M.
The neurological basis of mental imagery; a componentical
analysis.
Cognition (1984) ,18:243-261.
Feinberg I.
Eye movement activity during sleep and intellectual
function in mental retardation. Science (1968),159:1256.
Ferry C.
Networks on the brain.
New Scientist (1987), vol.115,
no.1569, pp.54-58.
Festinger L., Riecken H. & Schachter S.
When Prophecy Fails - a
social and psychological study of a modern group that predicted the
destruction of the world. (1956) Harper & Row, N.Y.
Firth H. & Oswald I.
Eye movements and visually active dreams.
Psychophysiology (1975) ,12:602-606.
395
Firth R. Totemism in polynesia.
Oceania (1930-31), vol.1.
Fishbein W. Interference with conversion of memory from short-term to
Commun. Behav. Biol.
long-term storage by partial sleep deprivation.
(1970) ,5:l71-175.
Fishbein W. Disruptive effects of REM sleep deprivation on long-term
memory. Physiology and Behavior (1971),6:279-282.
Fishbein W.
Alterations of the sleep states: sleep deprivation:
total, partial, and selective. In: M.Chase (ed.) The Sleeping Brain
(1972) pp.347-355. UCLA Brain Information Service.
Fishbein W. & Gutwein B.
Paradoxical sleep and memory storage
processes. Behavioral Biology (l977),19:425-464.
Fishbein W., Mccaugh J. & Swarz J. Retrograde amnesia: ECS effects
after termination of REM sleep deprivation. Science(l971),172:80-82.
Fisher C. Dreams, images, and perception: a study of unconscious and
preconscious relationships. J. of the American Psychoanalytic
Association (1956) ,4:5-48.
Fjss I-I.
The need to complete one's dreams.
(1969). In J.Fisher &
L.Breger (Eds.) The meaning of dreams: some insights from the
laboratory. California Mental Health Research Symposium Monograph
No.3, pp.38-63.
Fiss H. Dream content and response to withdrawal from alcohol. Sleep
Research (1980), 9:152.
Fiss H.
An empirical foundation for a self psychology of dreaming.
J.of Mind and Behavior (1986),7(2&3):161[3l]-192[62].
Fiss H. & Ellman S. REM sleep interruption: experimental shortening
of REM period duration. Psychophysiology (1973),lO:510-516.
Fiss H., Elinian S. & Klein G. Waking fantasies following interrupted
and completed REM periods.
Archives of General Psychiatry (1969),
396
21:230-239.
Fiss H., Klein G. & Bokert E. Waking fantasies following interruption
of two types of sleep. Arch.Gen.Psych. (1966), 14:543-551.
Fiss H., Klein G. & Shollar S. 'Dream intensification' as a function
of prolonged REM period interruption.
In Psychoanalysis &
Contemporary Science, (1974) NY, IIJP. pp.399-424.
Fiss H., Krenier E. & Litchman J. The mnemonic function of dreaming.
(1977) Paper presented at the meeting of the Association for the
Psychophysiological Study of Sleep, Houston, Texas.
Fliess R.
The Psychoanalytic Reader. (1950) Hogarth Press, London.
Fodor J. The mind-body problem. Scientific American (1981), 244:124132.
Fonagy P. & Ennis M.
Dreaming and environmental stimulation.
(1988) Paper presented at the Conference for the European Association
for the Study of Dreams, London.
Forster P. & Govier E. Discrimination without awareness ? Q.J.Exp.
Psychology (1978) ,30:282-295.
Foulkes D. Dream reports from different sleep stages. J.Abnormal and
Social Psychology (1962) ,65:l4-25.
Foulkes D. The psychology of sleep. (1966) Scribner's: New York.
Foulkes D. A grammar of dreams (1978). New York: Basic Books.
Foulkes D. Dreams and dream research.
In: W, K ella (ed.) Sleep
(1980) pp.246-257. Basel: S. Kerger.
Foulkes D. A cognitive-psychological model of REM dream production.
Sleep (l982),5(2):l69-l87.
Foulkes D.
Wiley.
(a)
Children's dreams: longitudinal studies.
(b)
397
(1982) John
Foulkes D.
Dreaming: a cognitive-psychological analysis.
(1985)
Lawrence Eribauni Assoc. Hilisdale, N.J.
Foulkes D. & Fleisher S.
Mental activity in relaxed wakefulness.
Journal of Abnormal Psychology (1975), 84:66-75.
Foulkes D. & Pope R.
Primary visual experience and secondary
cognitive elaboration in stage REM; a modest confirmation and an
extension. Perceptual and Motor Skills (l973),37:l07-118.
Foulkes D. & Rechtschaffen A. Presleep determinants of dream content:
effects of two films. Perceptual and Motor Skills (1964),19:983-1005.
Foulkes D. & Vogel C. Mental activity at sleep onset. J.Abnormal
Psychology (1965) ,70(4) :231-243.
Foulkes D., Shepherd J., Larson J., Belvedere E. & Frost S. Effects of
awakenings in phasic and tonic stage REM on children's dreams reports.
Sleep research (1972),l:104.
Frederickson C. & Hobson J. The sleep of cats following prolonged
electrical stimulation of the brain stem reticular activating system.
Psychophysiology (1969) ,6:27l.
Freud S. The interpretation of dreams. (Orig. 1900) (1953 S.E. vol.5)
Hogarth Press, London.
Freud S. On Dreams.
(Orig. 1901) (1953 S.E. vol.5) Hogarth Press,
London.
Freud S. Creative Writers and Day-dreaming Cong. published in 1908).
(1957 S.E. vol.9 pp.141-153.) Hogarth Press, London.
Freud S. Introductory Lectures on Psychoanalysis. (1953-57) Hogarth
Press, London.
Freud S. & Oppertheim D.
Dreams in folklore.
(1958) International
Universities Press, New York.
Fronim E. The forgotten language. (1957) New York: Grove Press.
398
Cackenbach J., Snyder T., Rokes L. & Sachau D.
Lucid dreaming
frequency in relation to vestibular sensitivity as measured by caloric
stimulation.
J. of Mind and Behavior (1986), 7:277-298.
Gadea-Ciria M.
Plasticity of ponto-geniculo-occipital waves during
paradoxical sleep after frontal lobe lesion in the cat. Experimental
Neurology (1976),53:328-338.
Cadea-Ciria M. Sequential discharges of phasic waves in the pontine
oculomotor system during paradoxical sleep in chronic mesencephalic
and decorticate cats.
EEC & Clinical Neurophysiology (1977),42:709-
712.
Cadea-Ciria M. , Stadler H., Lhoyd K. & Bartholini C.
Acetylcholine
release within the cat striatum during the sleep-wakefulness cycle.
Nature (1973) ,243:518-519.
Callup C. Self-recognition in primates: a comparative approach to the
bi-directjonal properties of consciousness.
American Psychologist
(1977) ,32:329-38.
Cardner H. The Mind's New Science. (1985) Basic Books, New York.
Cardner H. Epilogue to the 1987 edition of the Mind's New Science,
Basic Books, New York.
Gardner H. & Winner E. The comprehension of metaphor in brain-damaged
patients. Brain (1977) ,lOO:717-729.
Cardner-Medwin A.
The recall of events through the learning of
associations between their parts.
Proc.R.Soc.Lond.B.(1976),l94:375-
402.
Carskof B. & Sandak J.
Unlearning in recognition memory.
Psychonomic Science (1964) ,1:l97-l98.
Cazzaniga M., LeDoux J.& Wilson D. Language, praxis and the right
399
hemisphere: clues to some mechanisms of consciousness.
Neurology
(1977),27: 1144-7.
Geilner E. The psychoanalytic movement. (1985) Paladin, London.
Glaubman H. REM deprivation and divergent thinking. Psychophysiology
(1978) ,15(l) :75-79.
Glickstein M. & Gibson A. Visual cells in the pons of the brain.
Scientific American (1976),235:90-98.
Goldstein L., Stoltzfus N. & Gardocki J.
Changes in interhemispheric
amplitude relationships in the EEC during sleep.
Physiol.Behav.
(1972) ,8:81l-815.
Goodenough D., Lewis H., Shapiro A., Jaret L. & Sleser I.
Dream
reporting following abrupt and gradual awakenings from different types
of sleep. J.Personality and Social Psychology (1965),2(2):170-179.
Goodenough D., Witkin H., Lewis H., Koulack D. & Cohen H. Repression,
interference, and field dependence as factors in dream forgetting. J.
of Abnormal Psychology (1974),83:32-44.
Grastyan E.
The hippocampus and higher nervous activity.
In:
M.Brazier (ed.) The Central Nervous System and Behaviour. Pp. 119-205.
Josiah Macy, Jn, Foundation, NY (1959).
Green C.
Lucid dreams.
(1968) Oxford.
Greenberg R. & Dewan E.
Aphasia and REM sleep.
Nature
(1969) ,223:183-184.
Greenberg R. & Penman C. REM sleep and the analytic process: a
psychophysiologic bridge. Psychoanalytic Quarterly (1975) ,44:392-403.
Greenberg R. & Perinian C. If Freud only knew: a reconsideration of
mt. Review of Psychoanalysis
psychoanalytic dream theory.
(1978) ,5:71-75.
Greenberg R., Peaniman C., Fingar R., Kantrowitz J. & Kawliche S. The
400
effects of dream deprivation: implications for a theory of the
psychological function of dreaming. British J. of Medical Psychology
(1970) ,43:l-ll.
Greenberg R., Pearlman C., Schwartz W. & Grossman H. Memory, emotion,
and REM sleep. J.Abnormal Psychology (1983),92(3):378-381.
Greenspoon J. & Ranyard R.
Stimulus conditions and retroactive
inhibition. J. of Experimental Psychology (1957),53:55-59.
Griffin M. & Foulkes D. Deliberate presleep control of dream content:
an experimental study. Perceptual and Motor Skills (l977),45:660-662.
Grossman L. & Eagle M. Synonimity, antonymity & association in false
recognition responses.
J. of Experimental Psychology (l970),83:244-
248.
Gutwein B. & Fishbein W.
Paradoxical sleep and memory. Selective
alterations following enriched and impoverished environmental rearing.
Brain. Res. Bulletin (l980),5:9-12.
Hall C.S.
The meaning of dreams. (1966) McGraw-Hill, N.Y.
Hargreaves D. & Bolton N.
research.
Hartley D.
Selecting creativity tests for use in
Br. J. Psychology (1972),63:451-462.
Observations on man, his frame, his duty, and his
expectations. (1801) London: Johnson. [Cited in Lavie & Hobson, 1986.]
research. British J. of Psychology (1972),63:451-462.
Hartmann E. The function of sleep. Yale University Press (1973).
Hartmann E. The Nightmare: the psychology and biology of terrifying
dreams. (1984) Basic Books, NY.
Hartmann E. & Bridwell T. Effects of AMPT, 1-DOPA, and 1-tryptophane
on sleep in the rat. Psychopharmacologia (1970),7:3l3.
Hartmann E. & Stern W.
Desynchronized sleep deprivation: Learning
401
deficit and its reversal by increased catecholamines. Physiology and
Behavior (1972) ,8:585-587.
Hartmann E., Chung R., Draskoczy P. & Schildkraut J.
Effects of 6-
hydroxydopamine on sleep in the rat. Nature (1971),233:425-427.
Hartmann E., Baekeland F. & Zwilling C.
between long and short sleepers.
Psychological differences
Archives of General Psychiatry
(1972) ,26:463-468.
Haskell R.
Cognitive psychology and dream research: historical,
conceptual, and epistemological considerations.
J. of Mind and
Behavior (1986), 7:131-160. (a)
1-laskell R.
Logical structure and the cognitive psychology of
dreaming. J. of Mind & Behavior (1986),7(2&3):345[215]-378[248J. (b)
Hauri P.
Effects of evening activity on early night sleep.
Psychophysiology (1968) ,4:267.
1-Jauri P.
Evening activity, sleep mentation, and subjective sleep
quality. J. of Abnormal Psychology (1970),76:270-275.
Hebb D. The organization of behaviour: a neuropsychological theory.
(1949). Wiley, NY.
Heilbroner R.
Marxism: for and against. (1980) W.W. Norton, N.Y.
Henley S. Responses to homophones as a function of cue words on the
unattended channel. B.J.Psychology (1976),67:559-567.
Hermans H.
The dream in the process of valuation: a method of
interpretation. J.of Personality & Soc. Psychology (1987),53:163-175.
Hernandez-Peon R.
Neurophysiologic model of dreams and
hallucinations. J.Nervous and Mental Disease (1966),14l:623-650.
Hersch R., Antrobus J., Arkin A. & Singer J.
Dreaming as a function
of sympathetic arousal. Psychophysiology (l970),7:329-330.
Hersen M. & Barlow D.
Single case experimental design. (1976)
402
Pergamon, Oxford.
Hilgard E. Divided consciousness: multiple controls in human thought
(1977) NY: Academic Press.
and action.
Hinton C.
Learning in parallel networks.
Hobson J. A. The Dreaming Brain.
BYTE, April 1985.
(1988) Basic Books, N.Y.
Hobson J. A., Coldfrank F. & Snyder F. Respiration and mental activity
in sleep.
J. of Psychiatric Researeh (1965),3:79-90.
Hobson J. A., McCarley R. & Wyzinski P.
Sleep cycle oscillation:
reciprocal discharge by two brainstem neuronal groups.
Science
(1975) ,189:55-58.
Hobson J. A. & McCarley R. The brain as a dream state generator: An
activation-synthesis
hypothesis
of
the
dream
process.
Am.J.Psychiatry (1977),134(l2):1335-l348.
1-loelscher T., Klinger E. & Barta S. Incorporation of concern- and
nonconcern-related verbal stimuli into dream content. J.Abnormal
Psychology (1981), 90(1):88-91.
Hofstadter D. Goedel, Escher, Bach. An eternal golden braid. (1979)
Penguin Books, London.
Holmes M.
REM sleep patterning and dream recall in convergers and
divergers.
Ph.D. thesis, Edinburgh University (1976).
Holt R.
Individuality and generalization in the psychology of
personality. J. of Personality (1962), 30:377-404.
Hopfield J.
Neural networks and physical systems with emergent
collective computational abilities.
Proc.Natl.Acad.Sci.USA (1982),
79:2554-2558.
Hopfield J., Feinstein D. & Palmer R. 'Unlearning' has a stabilizing
effect in collective memories. Nature (l983),304:158-159.
403
Hopfield J. & Tank W.
Computing with neural circuits: a model.
Science (1986) ,233:625-633.
Hudson L. Contrary Imaginations (1966) Penguin Books, London.
Hudson L.
Arts and sciences: the influence of stereotypes on
language. Nature (1967),214:968-969.
Hudson L. Frames of mind. (1968) Methuen & Co. Ltd. London.
Hudson L.
Intelligence, race, and the selection of data.
Race (1971),12: 283-292.
Hudson L. Nightlife - the interpretation of dreams. (1985) Weidenfeld
& Nicolson, London.
Humphrey M. & Zangwill 0.
Cessation of dreaming after brain injury.
J.Neurol.Neurosurg.Psychiat. (1951),14:322-325.
Humphrey N.K.
Nature's Psychologists. (1980) In B.Josephson &
V.Ramachandran (eds.) Consciousness and the physical world. Oxford,
Pergamon.
Hunt H. Some relations between the cognitive psychology of dreams and
dream phenomenology. J.of Mind & Behavior(1986),7(2/3):213[83]228 [ 98]
Hunt H. Forms of dreaming.
Perceptual & Motor Skills (1982),54: 559-
633.
Iskander T. & Kaelbling R. Catecholamines, a dream sleep model, and
depression. Amer.J.Psychiatry (197O),127:83-90.
Jackson J. Hughlings.
In: J. Taylor,
Selected writings of J.
Hughlings Jackson. (1931) London.
Jakobson R.
Towards a linguistic typology of aphasic impairment. In:
Disorders of Language. Ed. de Reuck A. & O'Connor M. (1965) London,
Churchill.
Jasper H. & Tessier J. Acetylcholine liberation from cerebral cortex
404
during paradoxical (REM) sleep. Science (l970),172:60l-602.
Jaynes J.
The origin of consciousness in the breakdown of the
bicameral mind.
Jenkins A.
(1976) Boston: Houghton Mifflin.
The social theory of Claude Levi-Strauss.
(1979)
MacMillan Press, London.
Jerison H.
Palaeoneurology and the evolution of mind. Scientific
American (1976) ,234:90-101.
Johnson-Laird P. Connections and controversy. Nature (1987),330:12-l3.
Jones E.
Sigmund Freud: Life and Work vol.2. (1958) Hogarth Press,
London.
Jouvet M.
[Research on the nervous structures and the mechanisms
responsible for the different phases of the physiology of sleep.]
Arch. Ital. Biol. (1962),lOO:125-206.
Jouvet M. The states of sleep.
72.
Scientific American (l967),216:62-
(a)
Jouvet M. Neurophysiology of the states of sleep.
(1967),47:ll7.
Physiol. Rev.
(b)
Jouvet M. Biogenic amines and the states of sleep. Science
(1969) ,163:32-41.
Jung C.
Studies in word association.
(1918)
Translated by
M.D.Eder, London.
Jung C. An analysis of a patient's dreams. In: An lytic Psychology:
its Theory and Practice.
(1968) Pantheon Books and Routledge &
Kegan Paul, London.
Jus A., Jus K., Villeneuve A., Pires A., Lachance R., Fortier J. &
Villeneuve R.
Studies on dream recall in chronic schizophrenic
patients after prefrontal lobotomy.
405
Biological Psychiatry
(1973) ,6:275-293.
Kaelbling R., Koski E. & Hartwig C.
Reduction of REM sleep after
electroconvulsions - an experiment in cats on the mode of action of
electroconvulsive treatment. J.Psychiat.Res. (1968) ,6:153-157.
Kales A. , Hoedemaker F. , Jacobson A. & Lichtenstein E.
deprivation: an experimental reappraisal.
Dream
Nature (1964),204:1337-
1338.
Kales A., Hoedemaker F., Jacobson A., Kales J., Paulson M. & Wilson T.
Mentation during sleep: REM and NREM recall reports. Perceptual and
Motor Skills (1967),24:555-560.
Kanno 0. & Clarenbach P.
Effect of Clonidine and Yohimbine on sleep
in man: polygraphic study & EEC analysis by normalized slope
descriptors. EEC & Clinical Neurophysiology (l985),60:478-484.
Karacan I., Goodenough D., Shapiro A. & Starker S.
during sleep in relation to dream anxiety.
Erection cycle
Archives of General
Psychiatry (1966) ,15:183-189.
Kaser V.
The effects of an auditory subliminal message upon the
production of images and dreams.
J.Nervous & Mental Disease
(1986),l74(7) :397-407.
Kellogg W. & Kellogg L.
The ape and the child.
(1933) New York:
McGraw-Hill.
Keppel C.
Retroactive and proactive inhibition.
In:T.Dixon &
D.Horton (eds.) Verbal Behavior and General Behavior Theory. (l96)
Englewood Cliffs N.J. :Prentice-Hall.
Kerr N., Foulkes D. & Jurkovic G.
Reported absence of visual dream
imagery in a normally sighted person with Turner's syndrome. J.Mental
Imagery (1978),2:247-264.
Kerr N. & Foulkes D.
Right hemisphere mediation of dream
406
visualisation: a case study.
Cortex (1981),17:603-609.
Kety S., Evarts E. & Williams H. (eds.)
Sleep and altered states of
consciousness. Proceedings of the Association for Research in Nervous
and Mental Disease (1967) vol. 45, Williams & Williams, Baltimore.
Kininiins C.
Children's dreams.
(1920)
Kirkpatrick S., Gelatt C. & Vecchi M.
Longmans, London.
Optimization by simulated
annealing. Science (1983) ,220:671-680.
Kohut H.
The restoration of the self.
(1977)
International
Universities Press, NY.
Kramer M., Whitman R., Baldridge B. & Lansky L. Patterns of dreaming:
the interrelationship of dreams of a night.
J. of Nervous and Mental
Disease (1964), 139:426.
Kramer M., Hiasny R., Jacobs G. & Roth T. Do dreams have meaning ? An
empirical enquiry. American Journal of Psychiatry (1976), 133:778-781.
Kramer M., Roth T. & Cisco J. The meaningfulness of dreams.
Sleep
1976, pp. 314-316. Karger, Basel (1977).
Kripke D. & Sonnenschein D.
In: Pope K. & Singer J.
A biologic rhythm in waking fantasy.
The Stream of Consciousness.
(1978) John
Wiley and Sons, N.Y.
Krippner S.
Dreams and the development of a personal mythology.
J.
of Mind and Behavior (1986) ,7(2&3) :449[319] -462[332]
Kuper A. A structural approach to dreams. Man (1979 ,14(4):645-662.
Kuper A. The structure of dream sequences.
Culture, Medicine and
Psychiatry (1983) ,7:l53-175.
Kuper A. Structural anthropology and the psychology of dreams. J.of
Mind & Behavior (l986),7(2&3):333[203]-344[2l4] .
Kuper A.
(a)
Anthropology and Anthropologists - the modern British
407
school. (1986) Routledge & Kegan Paul, London.
(b)
Kuper A. Symbols in myths and dreams. Encounter (1989, March, pp.2631.)
Kuper A. and Stone A. The dream of Irma's injection: a structural
analysis. Am. J. Psychiatry (l982),l39(lO):l225-l234.
LaBerge S., Levitan L., & Dement W.
Lucid dreaming: physiological
correlates of consciousness during REM sleep. J.of Mind & Behavior
(1986),7(2/3):251[l211-258[128]
Lakoff G. & Johnson M. Metaphors We Live By.
(1980) University of
Chicago Press.
Lambrey-Sakai F.
(1972) Cited from the French by W.Fishbein &
B.Gutwein, Paradoxical sleep and memory storage processes. Behavioral
Biology (1977) ,l9:425-464.
Lashley K.
Mass action in cerebral function. Science (1931),73:245-
254.
Lashley K.
In search of the engram. Symposium of the Society of
Experimental Biology (1950) ,4:454-482.
Laughlin P., Doherty M. & Dunn R.
Intentional and incidental concept
formation as a function of motivation, creativity, intelligence and
sex. J.Pers.Soc.Psychology (1968),8:401-409.
Lavie P. & Hobson A. J.
Origin of dreams: anticipation of modern
theories in the philosophy and physiology of the eighteenth and
nineteenth centuries. Psychological Bulletin (1986),lOO:229-240.
Lavie P., Tzinchinsky 0., Metania Y. & Epstein R.
Hemispheric
asymmetry, REM sleep and dreaming. In: Bosinelli M. & Cicogna P.
(eds.) The Psychology of Dreaming. (1984) CLUEB, Bologna.
Lazarus A. & Davison C. Clinical innovation in research and practice.
In: Bergin A. & Garfield S. (eds.) Handbook of psychotherapy and
408
behavior change: an empirical analysis. (1971) New York: Wiley.
Leach E. Genesis as myth. (1969) Cape, London.
Leach E. Levi-Strauss. (1970) Fontana, London.
Leach E. Culture and communication, the logic by which symbols are
connected. (1976) Cambridge University Press.
Leach E. Social anthropology. (1982) Fontana, London.
Leach E. & Aycock D.A.
Structuralist interpretations of biblical
myth. (1983) Cambridge University Press.
Leconte P., Hennevin H. & Bloch V.
Increase in paradoxical sleep
following learning in the rat: correlation with level of conditioning.
Brain Research (1972),42:552-53.
LeDoux J. Brain, mind and language. (1985) In Brain and mind (ed.
D.Oakley) Methuen, London.
LeDoux J., Wilson D. & Gazzaniga M.
consciousness.
Beyond commissurotomy: clues to
In M.Gazzaniga (ed.) Handbook of behavioral
neurobiology, vol.2:543-54. (1979) Plenum Press, New York.
Lehmann D., Dumermuth C., Lange B. & Meier C. Dream recall related to
EEC spectral power during REM periods. Sleep (l981),lO:151.
Hippocampal theta rhythm during
Lerma J. & Carcia-Austt E.
paradoxical sleep. Effects of afferent stimuli and phase relationships
with phasic events. EEG & Clinical Neurophysiology (l985),60:46-54.
Lerner B.
Dream function reconsidered.
J. of Abiormal Psychology
(1967), 72:85-100.
Levi-Strauss C. Tristes Tropiques. (1964)
Athenium.
Trans. by J.Russell. N.Y.
(Orig. 1955.)
Levi-Strauss C.
The savage mind.
London.
409
(1966).
Weidenfeld & Nicolson,
Levi-Strauss C. Totemism. (1962) (Pelican edition 1969, London.)
Levi-Strauss C. Structural Anthropology volume 1. (1958) (Peregrine
Books 1977 edition, London.)
Levi-Strauss C. Structural Anthropology volume 2. (1973). (Peregrine
edition 1978, London.)
Levi-Strauss C.
(a)
Myth and meaning. (1978) Routledge & Kegan Paul,
London. (b)
Levi-Strauss C.
The way of the masks.
(1982)
Trans. by Sylvia
Modeiski. Seattle: Univ. of Washington Press. (Orig. 1975.)
Levi-Strauss C. The Raw and the Cooked.
(1964) Penguin Books (1986
edition, London.)
The effect of REM deprivation: is it
Lewin I. & Claubman H.
detrimental, beneficial or neutral? Psychophysiology (1975),12:349353.
Lewin M.
Understanding Psychological Research. (1979) John Wiley and
Sons, New York.
Li D. & Shao D. Effect of REM deprivation on consolidation of memory
in rats.
(1981).
Abstracted in Psychological Abstracts (1982),
vol.68, published by the American Psychological Association.
Linden E., Bern D. & Fishbein W.
Retrograde amnesia: prolonging the
fixation phase of memory consolidation by paradoxical sleep
deprivation.
Physiology and Behavior (1975),14:409-412.
Lindsay P. & Norman D.
to psychology.
Human Information Processing an introduction
(1977) Academic Press, N.Y.
Lucero M. Lengthening of REM sleep duration consecutive to learning
in the rat. Brain Research (l970),20:3l9-322.
Luria A. The Mind of a Mnemonist. (1968) Harvard Univ. Press., Ma.
Luria A. The Working Brain.
(trans. B. Haigh) (1973) Allen Lane,
410
London.
Luria A. The man with the shattered world. (1975) Penguin, London.
Maho C. Physiological concomitants of learning and PS in the cat.
Physiology & Behavior (1977),18(3):431-438. (Abstract.)
Mahony P. Towards a formalist approach to dreams.
mt. Review of
Psychoanalysis (1977) ,4:83-98.
Maranda P. Qualitative and quantitative analysis of myths by
computer.
In: Maranda P. (ed.) Mythology. (1972) Penguin, London.
Marks D. Personal communication.
(1988)
Matsumoto J., Nishisho T., Suto T., Sadahiro T. & Miyoshi M. Influence
of fatigue on sleep.
Nature (1968), 218:177-178.
McCarley R. & Hoffman E.
REM sleep dreams and the activation-
synthesis hypothesis. Am.J.Psychiatry (1981),138:904-9l2.
McGeoch J. Forgetting and the law of disuse. Psychological Review
(1932) ,39:352-370.
McCeoch J.
Studies in retroactive inhibition VIII Retroactive
inhibition as a function of the length and frequency of the
interpolated lists. J. of Experimental Psychology (1936),19:674-693.
McGeoch J. & MacDonald W.
Meaningful relation and retroactive
inhibition. Am.J. of Psychology (193l),43:579-588.
Medoff L.
Psychological correlates of the EEG sawtooth wave.
Unpublished Master's Thesis, Univ. of Wyoming (1972)
Mednick S. The associative basis of the creative process.
Psychol.
Rev. (1962),69:220-232.
Mednick S.
The remote associations test.
J.Creat.Behav.
(1968) ,2:213-214.
On reducing language to biology.
Mehier J., Morton. & Jusczyk P.
411
Cognitive Neuropsychology (1984) ,1:83-116.
Meienberg P. The tonic aspects of human REM sleep during long term
intensive verbal learning.
Physiological Psychology (1976),5(2):250-
256.
Mendelsohn C. & Griswold B.
Differential use of incidental imagery
in problem solving as a function of creativity.
J.of Abnormal &
Social Psychology (1964) ,68:43l-436.
Minami H. & Dallenback K. The effect of activity upon learning and
retention in the cockroach.
Am.J. of Psychology (1946),59:l-58.
Minsky M. & Papert S. Perceptrons: an introduction to computational
geometry. (1969) MIT Press.
Minsky M.
The psychology of computer vision. (1975) McGraw-Hill,
N .Y.
Tonic and phasic events during sleep:
Molinari S. & Foulkes D.
psychological correlates and implications.
Perceptual and Motor
Skills (1969) ,29:343-368.
Morris P. Sense and nonsense in traditional mnemonics. In Gruneberg
M. (ed.) Practical aspects of memory. (1978) Academic Press, N.Y.
Morris P. & Hampson P.
Imagery and consciousness.
(1983) Academic
Press (London).
Morris P. & Stevens R.
Linking images and free recall.
J.Verbal
Learning & Verbal Behaviour (l974),13:310-315.
Morrison A. & Bowker R.
PGO spikes: a sign of startle reflex
activation. In: Chase M., Stern W. & Walter P. (eds.) Sleep research
vol.4 p.36. Brain Information Service/Brain Research Institute. Los
Angeles (1975)
Moruzzi G.
Active processes in the brain stem during sleep. The
Harvey Lectures (1963),58:297.
412
Moskowitz E. & Berger R.
they related?
Rapid eye movements and dream imagery. Are
Nature (1969),224:613-614.
Motley M. Slips of the tongue. Scientific American (1985),253:114119.
Nakazawa Y., Tachibana H., Kotorii M. & Ogata M. Effects of L-Dopa on
natural night sleep and on rebound of REM sleep.
Folia Psychiatrica
et Neurologica Japonica (1973), 7:223-230.
Newman E. & Evans C. Human dream processes as analogous to computer
programme clearance. Nature (1965) ,206:534.
Newton P. Recalled dream content and the maintenance of body image.
J. Abnormal Psychology (1970),76(1):134-139.
Nisbett R. & Wilson T.
Telling more than we can know: verbal reports
on mental processes. Psychological Review (l977),84:231-259.
Oakley D. & Eames L.
The plurality of consciousness.
In: Brain and
Mind (1985, ch.8) (Ed.D.Oakley) Methuen, London.
O'Keefe J. & Nadel L.
(1978) The Hippocampus as a cognitive map.
Oxford, Clarendon Press.
Oniani T. Does paradoxical sleep deprivation disturb memory trace
consolidation?
English Abstract in Psychological Abstracts
(1987) ,74:55-56.
Oswald I. Discriminative responses to stimulation during human sleep.
Brain (1960),83:440-453.
Paivio A.
Imagery and verbal processes. (1971) N.Y. Holt, Rinehart
& Winston.
Palombo S.
The dream and the memory cycle.
mt. Rev, of
Psychoanalysis (l976),3:65-83.
Palombo S.
Dreaming and Memory: A new information-processing model.
413
(1978) Basic books, New York.
Palombo S.
The Cognitive act in dream construction.
J. of the
American Academy of Psychoanalysis (l980),8:185-20l.
Palombo S. Recovery of early memories associated with reported dream
imagery. American J. of Psychiatry (1984),14l:l508-l5ll.
Palombo S.
Deconstructing the manifest dream.
Psychoanalytic Association (l984),32:377-393.
Peariman C.
(a)
J. of the American
(b)
Impairment of environmental effects on brain weight by
adrenergic drugs in rats.
Physiology & Behavior (1983),30:l6l-163.
Pegram V. Effects of protein synthesis inhibition on sleep in mice.
Behav. Biol. (1973),9:377-382.
Peirce C.S. Collected Papers, vol.2. (1938) Hartshorne C. & Weiss P.
(eds.) Harvard University Press.
Penfield W.
The interpretive cortex.
Penfield W. & Perot P.
Science (1959),129:1719-1725.
The brain's record of visual and auditory
experience: a final summary and discussion. Brain (1963),86:595-696.
Piaget J. Plays, Dreams and Imitation in Childhood. (1962) New York,
Norton.
Piaget J.
Structuralism.
(1968)
Routledge & Kegan Paul (1971
edition, London.)
Pivik T. & Foulkes D. 'Dream deprivation': effects on dream content.
Science (1966) ,153:1282-1284.
Pivik T. & Foulkes D.
NREM mentation: relation to personality,
orientation time, and time of night.
J.Consulting and Clinical
Psychology (1968),32(2):144-151.
Polanyi M.
The tacit dimension.
(1967). Routledge & Kegan Paul,
London.
Pompeiano 0.
Sensory inhibition during motor activity in sleep. In:
414
Neurophysiological basis of normal and abnormal motor activities.
M.Yahr & D.Purpura (eds.) (1967) pp.323-372. Raven Press, N.Y.
The neurophysiological mechanisms of the postural and
Pompeiano 0.
motor events during desynchronised sleep.
In: Sleep & Altered States
of Consciousness. Kety S., Evarts E. & Williams H. (eds.)
(1967)
(b)
pp.351-423. Williams & Wilkins, Baltimore, Maryland.
Poole R.
(a)
Introduction to 'Totemism' by C.Levi-Strauss.
(1969)
Penguin, London.
Portnoff G., Baekeland F., Goodenough D., Karacan I. & Shapiro A.
Retention of verbal materials perceived immediately prior to onset of
non-REM sleep. Perceptual and Motor Skills (1966),22:75l-758.
Postman L. A pragmatic view of organization theory. In:E.Tulving &
W.Donaldson (eds.)
Organization of Memory
(1972) Academic press,
N .Y.
Postman L., Stark K. & Fraser J. Temporal changes in interference.
J. of Verbal Learning & Verbal Behavior (1968),7:672-694.
Pritchard R.
Stabilized images on the retina.
Scientific American
(1961) ,204:72-78.
Prytulak L.
Natural language mediation. Cognitive Psychology
(1971) ,2:l-56.
Pujol J., Mouret J., Jouvet M. & Glowinski J. Increased turnover of
cerebral norepiniphrine during rebound of paradox! al sleep in the
rat. Science (1968),l59:112-1l4.
Purpura D., Shofer R. & Scarff T. Properties of synaptic activities
and spike potentials of neurons in immature neocortex.
J. of
Neurophysiology (1965) ,28:925-942.
Pylyshyn Z.
What the mind's eye tells the mind's brain: a critique
415
of mental imagery.
Psychological Bulletin (1973),8:l-14.
Rechtschaffen A. Interrelatedness of mental activity during sleep.
Archives of General Psychiatry (1963),9:536-547.
Rechtschaffen A.
Discussion of W. Dement's "Experimental Dream
Studies" in Masserman J. (ed.) Science and Psychoanalysis (1964),
7:168-179. N.Y. Grune and Stratton.
Rechtschaffen A.
The psychophysiology of mental activity during
sleep. In: J.McGuigan & R. Schoonover (eds.) The Psychophysiology of
Thinking (1973) Academic Press, NY.
Rechtschaffen A. & Buchigiani C. Visual dimensions and correlates of
dream images.
Sleep Res. (1983),12:189.
Rechtschaffen A., Watson R., Wincor M., Molinari S. & Barta S. The
relationship of phasic and tonic periorbital EMG activity to NREM
mentation. Sleep Research (1972),1:114.
Reinsel R., Wollman M. & Antrobus J. Effects of environmental context
and cortical activation on thought.
Journal of Mind and Behavior
(1986), 7:259-276.
Richardson J. Mental imagery and the distinction between primary and
secondary memory. Q.J.Exp.Psychology (1978),30:471-485.
Rivers W.H.R.
Robert W.
Conflict and Dreams.
(1923) Kegan Paul, London.
[The dream explained as waste-disposal.] (1886) Hamburg.
Roberts K.R.
Worlds and world reconstruction
Advances in
Descriptive Psychology (1985) ,4:17-53.
Roffwarg H., Dement W., Muzio J. & Fisher C. Dream imagery:
Relationship to rapid eye movements of sleep.
Archives of General
Psychiatry(1962) ,7:235-258.
Roffwarg H., Muzio J. & Dement W. The ontogenetic development of the
human sleep dream cycle. Science (1966),l52:604-618.
416
Roffwarg H., Herman J., Bowe-Anders C. & Tauber E.
The effects of
sustained alterations of waking visual input on dream content.
Chp.9
in The Mind in Sleep, ed. A.M. Arkin, Antrobus J. & Eliman S. (1978)
Hilisdale N.J., Lawrence Eribaum Associates.
Rose S.
Memories and Molecules.
New Scientist (27/11/1986),
no.1536:40-44.
Rosenzweig S.
Idiodynamics in personality theory with special
reference to projective methods. Psychological Review (1951),58:213223.
Rubin D., Wetzler S. & Nebes R.
Autobiographical memory across the
pp.202-211 In: Autobiographical Memory. (ed. D. Rubin)
lifespan.
(1986) Cambridge University Press.
Rurneihart D., Hinton G. & Williams R.
Learning representations by
back-propogating errors. Nature (l986),323:533-536.
Symbolism and its relationship to the primary and
Rycroft C.
secondary processes. mt. J. Psychoanalysis (1956),37:137-146.
Rycroft C. The innocence of dreams. (1981) Oxford U.P.
Sacks 0.
The man who mistook his wife for a hat. (1985) Duckworth,
London.
Sagales T. & Domino E. Effects of stress and REM sleep deprivation on
the patterns of avoidance learning and brain acetylcholine in the
mouse. Psychopharmacologia (Berlin) (1973),29:307-3 D.
Salley R. Subpersonalities with dreaming functions in a patient with
multiple-personalities.
Journal of Nervous and Mental Disease
(1988) ,176:l12-l15.
Salzarulo P.
Variations with time of the quantity of eye movements
during fast sleep in man.
Electroenceph. Clin. Neurophysiol. (1972),
417
32:409-416.
Schachter S. & Singer J.
Cognitive, social and psychological
determinants of emotional state. Psychological Review (1962), 69:379399.
Schatzman M.
Solve your problems in your sleep.
(a)
(1983) ,vol.98,no.1361,pp.692-693.
Schatzman M.
New Scientist
Such stuff as REM sleep is made of. New Scientist
(1983) ,vol.99,no.l375,pp.796-797.
(b)
Schatzman M. The meaning of dreaming. New Scientist (1986) ,vol.101,
no.1540/1541, pp.36-39.
Schechter N., Schmeidler C. & Staal M.
Dream reports and creative
tendencies in students of the arts, sciences, and engineering.
J.Consulting Psychology (1965),29:415-421.
Schildkraut J. & Hartmann E.
Turnover and metabolism of
norepiniphrine in rat brain after 72 hours on a D-deprivation island.
Psychopharmacologia (1972) ,27:17-27.
Schildkraut J., Schanberg S., Breese C. & Kopin I.
Norepinephrine
metabolism and drugs used in the affective disorders: a possible
mechanism of action. American J. of Psychiatry (1967), 124:600.
Schwartz D., Weinstein L. & Arkin A.
Qualitative aspects of sleep
mentation. Chp.6 in: The Mind in Sleep. Eds. A. Arkin, J. Antrobus &
S. Ellman.
(1978) Lawrence Erlbaum Assoc. N.J.
Scrima L.
Isolated REM sleep facilitates recall of complex
associative information. Psychophysiology (1982),19(3):252-259.
Scrima L.
Dream sleep and memory: New findings with diverse
implications. Integrative Psychiatry (1984),2:211-216.
Segal M., Disterhoff J. & Olds J.
Hippocampal unit activity during
classical aversive and appetitive conditioning.
418
Science
(1972) ,175:792-794.
Seligman M. & Yellen A.
What is a dream?
Behav. Res. Ther.
(1987) ,25:1-24.
Shapiro M. The single case in clinical psychological research. J. of
General Psychology (1966), 74:3-23.
Shaw C. & DeMers S.
The relationship of imagery to originality,
flexibility and fluency in creative thinking.
J. of Mental Imagery
(1986) ,lO:65-74.
Shevrin H. Subliminal perception and dreaming. J. of Mind & Behavior
(1986) ,7(2&3) :379[249) -396[2661
Shevrin H. & Fisher C. Changes in the effects of a waking subliminal
stimulus as a function of dreaming and non-dreaming sleep.
J. of
Abnormal Psychology (1967) ,72:362-368.
Silberer H.
Report of a method of eliciting and observing certain
symbolic hallucinatory phenomena. In O.Rapaport (ed.) Organization &
pathology of thought. NY, Columbia Univ. Press. pp.195-207.
Silverstone R.
The message of television. Myth and narrative in
contemporary culture. (1981) Heinemann Educational Books, London.
Singer J.
Daydreaming and fantasy. (1981) Oxford University Press.
Singer J. & Schonbar R.
Correlates of daydreaming: a dimension of
self-awareness. J. of Consulting Psychology (l961),25:l-6.
Sivilotti M., Emmerling M. & Mead C.
A novel
associative memory
implemented using collective computation. 1985 Chapel Hill Conference
on VLSI.
Skinner B. F.
Are theories of learning necessary?
Psychological
Review (1950), 57:193-216.
Skinner D., Overstreet D. & Orbach J.
419
Reversal of the memory
disruptive effects of REM sleep deprivation by physostigmine.
Behavioral Biology (1976), 18(2):189-190.
Skinner Q. (ed.) The return of grand theory in the human sciences.
(1985) Cambridge University Press.
Skynner R. & Cleese J.
Families and how to survive them. (1983)
Methuen, London.
Slamecka N.
Retroactive inhibition of connected discourse.
J. of
Experimental Psychology (1961) ,62:295-30l.
Smith C., Kitahama K., Valatx J. & Jouvet M.
Increased paradoxical
sleep in mice during acquisition of a shock avoidance task.
Brain
Res. (1974),77:221-230.
Snyder F. The Phenomonology of Dreaming. In L.Madow & L.Snow (eds.)
The Psychodynamic Implications of the Physiological Studies on Dreams.
(1970) Springfield Ill. Thomas.
Snyder F., Hobson J., Morrison D. & Goldfrank F.
Changes in
respiration, heart rate, and systolic blood pressure in human sleep.
J. of Applied Physiology (1964),19:417-422.
Solodkin M. , Cardona A. , Corsi-Cabrera M.
Paradoxical sleep
augmentation after imprinting in the domestic chick. Physiology and
Behavior (1985),35(3):343-348.
Spanjaard J. The manifest dream content and its significance for the
interpretation of dreams.
Int.J. of Psychoanalysis (1969),50:221-
234.
Sperber D.
Chapter on Levi-Strauss. In: Structuralism and since,
Ed.J.Sturrock. (1979), Oxford U.P.
Sperling C.
Information in a brief visual presentation. Unpublished
doctoral dissertation. Harvard University (1959).
Sperry R.
Hemisphere deconnection and unity in conscious awareness.
420
American Psychologist (1968),23:723-733.
Spreux F. Paradoxical sleep modification after a learning in man.
Cahiers de psychologie cognitive (l982),2:327-334.
Starker S. Daydreaming styles and nocturnal dreaming. J.Abnormal
Psychology (1974) ,83:52-55.
Stein L.
Paper presented at the International Symposium on Anti-
depressant Drugs, Milan, April 1966.
Steiner S. & Eliman S. Relation between REM sleep and intracranial
self-stimulation. Science (l972),l77:l122-1l24.
Stern W. & Morgane P.
Theoretical view of REM sleep function:
maintenance of catecholamine systems in the central nervous system.
Behavioral Biology (1974) ,ll:l-32.
Stern W., Morgane P., Panksepp J., Zolovick A. & Jaloviec J.
Elevation of REM sleep following inhibition of protein synthesis.
Brain Research (1972) ,47:254-258.
Stone A. Personal communication (1986).
Stromeyer C. & Psotka J.
The detailed texture of eidetic images.
Nature (1970), 225:346-349.
Sturrock J. Chapter on Roland Barthes. In: Structuralism and since,
Ed. J. Sturrock. (1979) Oxford U.P.
Sturrock J. Structuralism.
Sutherland S.
(1986) Paladin, London.
Parallel distributed processing.
Nature (1986),
Symonds P. & Jensen A. From adolescent to adult.
(1961) Columbia
323:486.
University Press.
Tanguay P., Ornitz E. , Forsythe A. & Ritvo E. REM activity in normal
and autistic ch ildren during REM sleep. J. Autism and Ch.
421
Schizophrenia (1976) ,6(3) :275-288.
Tank D. & Hopfield J.
circuits.
Collective computation in neuronlike
Scientific American (1987),257(no.6):62-70.
Tart C.
Effects of posthypnotic suggestion on the process of
dreaming.
Unpublished doctoral dissertation. (1963) University of
North Carolina.
Tart C.
dreaming.
Some effects of posthypnotic suggestion on the process of
mt. J. Clin. Exp.Hyp. (1966),14:30-46.
Tart C. (ed.) Altered states of consciousness. (1969) John Wiley &
Sons, N.Y.
Tauber E. & Glovinsky P.
New views on the function of REM sleep in
the evolution of mammals. Contemporary Psychoanalysis (1987), 23:438445.
Tilley A.
Retention over a period of REM sleep or non-REM sleep.
Br. J. of Psychology (1981),72:241-248.
Tilley A. & Empson J.
REM sleep and memory consolidation..
Biological Psychology (1978),6:293-300.
Timpanaro S.
The Freudian slip, Psychoanalysis and textual
criticism. (1976) New Left Books, London.
Tulving E. & Donaldson L. Organization of Memory. (1972) Acadeaic
Press, N.Y.
Ukponmwan 0. & Dzoljic M. Enkephalinase inhibitic antagonizes the
increased susceptibility to seizure induced by REM sleep deprivation.
Psychopharmocology (1984) ,83:229-232.
Underwood B. Retroactive and proactive inhibition after five and
forty-eight hours. J. of Experimental Psychology (1948),38:29-38. (a
Underwood B. 'Spontaneous recovery' of verbal associations.. J. of
Experimental Psychology (1948),38:429-439. (b)
422
(1964) Cited by Himwich W.
Valatx J., Jouvet D. & Jouvet M.
Developmental changes in neurochemistry during the maturation of sleep
behavior, pp. 125-140 In Clemente C., Purpura D. & Mayer F. (eds.)
(1972).
Valentine E. Conceptual Issues in Psychology. (1982) George Allen &
Unwin, London.
Valins S.
Cognitive effects of false heart-rate feedback.
J. of
Personality and Social Psychology (1966) ,4:400-408.
Van Hulzen Z. & Coenen A.
Selective deprivation of sleep and
consolidation of shuttle-box avoidance.
Physiology and Behavior
(1979) ,23:821-826.
Velluti R.
An electrochemical approach to sleep metabolism: a p02
paradoxical sleep system.
Physiology & Behavior (1985),34:355-358.
Verdone P. Temporal reference of manifest dream content. Perceptual
& Motor Skills (1965),20:1253-68.
Effets de la desafferentiation
Vital-Durand F. & Michel F.
peripherique sur le cycle veille-sommeil chez le chat. [Effects on the
sleep cycle of deafferented cats.] Arch.Ital.Biol. (l97l),l09:166-186.
Vogel G.
REM deprivation:III Dreaming and psychosis.
Archives of
General Psychiatry (1968) ,18:312-329.
Vogel G., Barrowclough B. & Giesler D. Limited discriminability of
REM and sleep onset reports and its psychiatric implications.
Archives of General Psychiatry (1972),26:449-455.
Vogel G., Foulkes D. & Trosman H. Ego functions and dreaming during
sleep onset. Archives of General Psychiatry (1966), 14:238-248.
Vogel C., McAbee R., Barker K. & Thurmond A. Endogenous depression
improvement and REM pressure.
Archives of General Psychiatry
423
(1977) ,34:96-97.
On the relationship of associative and organizational
Voss J.
processes. In: E.Tulving & W.Donaldson (eds.) Organization of memory.
(1972) Academic Press, N.Y.
Wagner N. & Stegman E. Does the schizoid child develop into an adult
schizophrenic? (1964) Unpublished report cited in Singer J. (op.cit.)
Wallach M. & Kogan N. Modes of thinking in young children. (1965).
Holt, Rinehart & Winston, New York.
Warren R. Illusory changes of distinct speech upon repetition - the
verbal transformation effect. Brit.J.Psychology (1961) ,52:249-258.
Watson R.
Mental correlates of periorbital phasic integrated
potentials during REM sleep. Ph.D. thesis, Univ. Chicago (1972), and
also in Sleep Research (1972),1:116.
Webb L. Transfer of training and retroaction: a comparative study.
Psychological Monographs (1917) vol.24. no.104.
Webb W.
Sleep: an active process; research and commentary. (1973)
Scott, Foresman: Glenview, Ill.
Webb W. & Agnew H.
Sleep stage characteristics of long and short
sleepers. Science (1970) ,168:146-147.
Weber L., Muzet A., Schieber J. & Lienhard P.
rapid eye movement production during
REM
Characteristics of
sleep in man: organization
and rhythmicity. EEG & Clinical Neurophysiology (l983),55:l5l-l55.
Weisz R. & Foulkes D.
Home and laboratory dreams collected under
uniform sampling conditions. Psychophysiology (1970) ,6:588-596.
Williams H., Morlock H. & Morlock J.
sleep.
Instruniental behavior during
Psychophysiology (l966),2:208-216.
Williamson P., Galin H. & Mamelak M.
Spectral EEG correlates of
mentation during sleep. Paper presented at the International Meeting
424
of the Sleep Society, (1983), Bologna, Italy.
Wilishaw D., Buneman 0. & Longuet-Higgins H.
Non-hollographic
associative memory. Nature (1969), 222:960-962.
Brain and Psyche - The biology of the unconscious. (1985)
Winson J.
Anchor Press/Doubleday New York.
Wittgenstein L. Conversations on Freud. In: Philosophical Essays on
Freud. Wollheim R. & Hopkins J. (eds.) (1982) Cambridge U.P.
Wolpert E.
Studies in psychophysiology of dreams: II An
electromyographic study of dreaming. Archives of General Psychiatry
(1960) ,2:231-241.
Wood P.
Dreaming and social isolation. (1962) Unpublished doctoral
dissertation. Univ. of South Carolina.
Yasoshima A., Hayashi H., lijima S., Sugita Y., Teshima Y., Shimizu T.
& Hishikawa Y. Potential distribution of vertex sharp wave and sawEEC & Clinical Neurophysiology
toothed wave on the scalp.
(1984) ,58:73-76.
Young J.
(1978) Oxford U.P.
Programs of the brain.
Zajonc R. Feeling and thinking. American Psychologist (1980),35:151175.
Zarcone V. & Benson K.
rated depression.
Increased REM eye movement density in self-
Psychiatry Research (1983>,S:65-31.
Zimmerman J., Stoyva J. & Metcalf D.
Distorted visu 1 experience and
REM sleep. Report to the 9th Annual Meeting of the Association for the
Psychophysiological Study of Sleep, (1969), Boston, Mass.
Zimmerman W.
Sleep mentation & auditory awakening thresholds.
Psychophysiology (1970) ,6:540-549.
Zornetzer S. & Gold M.
The locus coeruleus: its possible role in
memory consolidation. Physiol. Behav. (l976),16:33l-336.
425
[ETTtTNA'flJl
NATURE VOL 04 54 JULY tI)
APPNDZX I
NATljJijj (1983), vol.304 PP.158-159
'Unlearning' has a stabilizing effect
in collective memories
J. J. llopfleldt, I). I. Feinstein & R. G. Palmert
California Institute of Technology. Pasa4en.
Calthrnia 91125. USA
Dui.r University, Durham. North Carolina 27706. USA
Bell Laboratories, Murray Hill. New Jersey 07974. USA
Crick and litchison' have presented • hypothesis for t?'e luL>
tional role of dream sleep Involving an unlearning' process.
We have Indepen Jently carried out mathematical md computer
modelling of le..rising and 'unlearning' n a collective neural
network of 30-1,000 neurones. The model network has a
content.pddresuble memory or 'associative memory' which
allows U to learn and store many memories. A particulz
memory can be evoked In Its entirety when the netwok is
stimulated by any adequate-sized subpart of lb- Informatloa
of that memory1. But different memories of the same size are
not equally easy to recall. Also, when memories are learned,
spurious memories are also created and can also be evoked.
Applying an 'unlearning' process, similar to the learnIng processes but with a reversed sign and starting from a noIse Input,
enhances the performance of the network In accessing resI
memories and In minimizing spurious one,. Although oisr model
was aot motivated by higher nervous function, our system
displays behaviours which are strikingly parallel to those needed
for the hypothesized role of 'unlearning' In rapId eye movement
(REM) sleep.
In the most symmctric form of collective memory in our
dynamic neural network', each neurone, j, has two states, and
is described by a variable -i1. The instantaneous state of
lhc system of N neurones can be thought of as an N-dimensional
vector having components s, of size 1. The neuronea arc interconnected by a network of synapses, with a synaptic strength
T from neurons j to neurone i. The instantaneous input to
neurons i is
input to I - E T,sj
No. of unlearning trials
Fig. I The fraction of random starting states which leads to
particle memories (acceaibility). The five dashed Iine are tt'.e five
no,rinal memories. The solid line is the total accesaihi!ty of all
spurious memories. In these irish a was set at 0.01.
wh_os r, is the present state (*1) of neurons j. The neural
state of the system changes in time under the following
algorithm. Each neurorse i interrogates itself at random in time,
but at a mean rate W, and readjusts it state, setting M. ±1
according to whether the input to at that mom-nt is greater
or less than zero. The neurotics act asynchronousl,.
This algorithm defines the time evolution of the state of the
system. For any symmetric connection matrix, therc are stable
states of the network of neuronea. in which each neurons is
either 'on and has an input 0 or 'off' and has an Input <0.
These stable States will not change in time. Starting from any
arbitriy initial state, the system reaches a stable state and
ceises to evolve in a time of —3/W.
The stable states of the system can be arbitrrilv assigccd by
an appropriate choice of T. Supeose it different N-dimensional
state vectors
s 1 tOn 0.25N
are to be stable states of system. If these state vectors are
su.iiciently different, and if the synaptic connection matrix Tu
is givel by
then the states e' will be stable states of the system.
This network now functions as an associative memory. If
started from an initial State which resembles somewhat state
e' and which resembles other p'(a ^ t) very little, the. sta:e will
evolve to the state ', The states i' arc evokable memories,
ind the system correctly rccorist,'ucls an entire memory from
any initial partial information, as long as the partial information
was sufficient to identify a single memory. Detailed properties
of the collective operation of this network have been
described previously2.
The form of the 7',, matrix can be described as an incremental
learning rule. To learn a new memory gr", increment 7',, by
learnM" T""
S.ili..m)s/4)/21u1 5S.-02501 00
C Iwi M..
426
La
NATURE VOL 3414 II JULY
LETTERSTONA11JIt
In biology or in circuits, this would be done by placing the
system in slate ,,&—for example, driven by external inputs .—
and enabling a learning process that allows all T 1 to increment.
The information needed by each synapse is local—the mcccrflent for synapse ij does not depend on the global structure of
the new state or past memorics, but only on "' and ".
Undet this algorithm, when random starting states arc
chosen, some stored memories are much more accessible than
others, that is, considerably larger numbers of randomly chosen
initial states lead to some memories than o others. This is a
vagary of the particular set of memories which ha'.e been
learned. It occurred to us that it would be poss.ble to reduce
this unevenness of access (which can be intuitively described
as the "50% of all stimuli remind mc of sex" problem) by
'unlearning'.
Specific unlearning was implemented by choosing starting
states at random; when a final equilibrium state s' was reached
it was weakly unlearned by the incremental change
unlearnT-es,O<ec 1
Figure 1 illustrates the effect of unlearning on the accessibility
of five stored memories in a set of 32 neurones. Accessibility
is quantita,ively defined as the fraction of random initial states
leading to a particular final stable state or group of states. The
unevenness of the lines is due in part to statistical nc'e in the
simulation. The accessibility of the nominal assigned memories
initially ranges over a factor of 3, but converges with unlearning
to a spread of only a factor of 1.4. Thus the accessibility is
much more uniform (or in Crick-Mitchison terms, the relative
stability of the modes made more uniform) alter specific
unlearning, and the system wiU have functionally improved
recall.
In our model the storage of a set of assigned memories in
7',, also produces a set of spurious stable states which were not
innerted as memory states. One of the strong effects of unremembering is to reduce the total accessibility of spurious stateS,
as shown by the solid line in Fig. 1.
The qualitative reason (or the success of unlearning comes
from the behaviour of the 'energy' E, defined for any State a as
E-I* I
the example
Memory I
Memory 2
Memory 3
Spurious memory
where grey is taken as a category equally resembling black and
white. This spurious state is more stable when 'Harold' and
'Walter' have a significant correlation—perhaps Harold' and
'Harry'. These particular spurious states are not simply transitive logical associations of the form Ai-.B, B.-.C; -.A.-.C.
They are truly spurious 'illogical', associations, but perhaps
'plausible' as they come from correlations in the structure of
memories.
In our simple system, unlearning improves memory function
both by the equalization of accessibility and the suppression of
spurious memories. We asked whether other simple algorithmic
changes such as clipping the T matrix or a threshold effect
produce an equivalent improvement in memory performance.
These two do not, presumably because they lack the essential
element of the present scheme, that is, the feedback via the
algorithm of informatirn about the accessibility of particular
states. We believe the results found will be inscnJve to
whether the state comp"ncnt values are taken as 0 and I or ± 1.
The REM sleep hypothesis of Crick and Mitchison' refers
to higher level processing. Our example illustrates that from a
mathematical viewpoint the general idea could work as they
described, If the Crick-Mitchison hypothesis is correct, one
might ask about correlations between the structure of the
spurious linkages in modelling and the strange associations
present in dreams.
We thank F. Crick and I). Wilishaw for discussions. 'This
work was supported in part by NSF grant DMR-8 107494 and
by the System Development Foundation.
T14,SM1
The change of neural State with time according to the asynchronous algorithm enonotonically decreases E until a final
stable state is reached—either a stored memory or a spurious
memory. Any stable state has, for a given T,1 , an energy
E. There is a strong tende4scy for the states having the deepest
energy valleys to collect from the largest number of random
starting states, that is, deep valleys are also broad. When a final
stale e' is unlearned, its energy E' is specifically raised and its
valley of collectio;, diminished relative to other states. While
this argument indicates why accessibility of stored memories
should be made more nearly even by unlearning, only a detailed
analysis shows why the spuri'us states should be so sensitive
to it. Too much unlearning will ultimately destroy the stqred
memories.
We have identified a class of spurious States, which in their
most elementary form have their origin in triples. As an example
on 16 neurones
Memoryl
Memoryl
Memnory3
Spuriousmemory
Walter, white
Walter, black
Harold, grey
Walter, grey
++++---- ++—+—+-+—+—++
++--++-- +--++--+
++++---- +--+^--+
The stability of the spurious memory is enhanced if the first
half of memory is weakly correlated with memories 1 and 2..
Mathematical analysis of the statistical stability of such spurious
ilales shows that they arc typically less stable than the assigned
memories, and that the stability will also depend on correlations
with other memories. The nature of these spurious states can
be described by analogy in terms of higher level function by
427
R.,ó.d 3* D.b., *912;
pd IS Mw *983.
I. C.&F.C*M.Id..G.N...3S4,lII-ll4(19tt,.
2. 14op&M. J. 1. P.. --- A.4 &I. U.S.A. 7, 2*34-2354 119121
NDPEDIX 2
THE HIL TEXTS OF K.J'S FOR NIGHTS' DREAMS
KJ r am of a grap tio rmnt axe a ntnth to a drean lth in thicago. He as a gra±ate strxmt.
Joai nith is a feLlcw gra±ate stithrt; thris is his riajor professor dissertaticn sr.çervisor; he lives with
Carrie; Ras is the crient professor.
NIGHT 1. (30-31/7/73)
pre-sleep interview.
E. I w*.jld Like to talk a Little bit xut rkrat kird of thca,ts are goir thrct, ytu' mird r
jwt before ytu are to go to sleep.
S. ItratanI thir*ira.it?
-
E. Yes.
S. I'm thirkirG aixut cLinical ythotogy a-ti rkiether.... I ruld really like to get cut of. There's
this scaery by Tan Nold ad stuff that I ras jist thirkir9 dxnit hcw I got so ttnred m by readir thxut
Tm ld. Reaiir9 that I've heei cut of tcsdi with sirte a few years ago. IJien I get into saie ficticz, I
really aijc' it. I ckrr't take the tine to cb that at all. I ras jLst kird of thirkir@ that there are ras to
cb it a-ti I krtw that I rjxitd enjoy it. I feel like I'm srificir- all that to d rkiat I an ckir9 riit
riw. The La-9er I stay in it the harder it is to get cut of it. I we alL of the tential that is there.
That's rkrat I've bee, thirkir9 aixut.
E. Are yui feelir kirti of sad or disxrintad?
S. Not recessari ly, jist kird of Lockir ar erythir a-ti seeir-9 hcw big the r-izrrld is ad how
maLl the cut off is. It's kird of a reccrmideraticn, ai ucertainty, bit rot ressarily a disxrintnuit.
Not regret or that I na a mistake, jist that there is all that to cb cut there, ad I have to keep this in
mird.
E. Soyur,ttofirdapLaceforitairgatltheotherthirssothatitcbesn'tgetLeet.
S. Cl,
, Lecase I rover decic to give it p. Do yai i.Eit to krr râ,at else I ras thirkir of ?
E. Yes.
S. I'm a little ca-cerr thxut talon-ow. In cLass I to present [sic] my Fetieit ad that shctild he
mi interestir scae with my professor there altha, I ckn't have that rnxh avdety aixut it. My professor,
she's reaLly flakey a-ti I thi't r,t to cb aiythir to çear that I an none cçpsiticral thai I an. She'll
pick ri,t p cii that. She ckresn't ki-o.r rim bit she's heard aixut rim. Everythir I cb kirti of fr.rnels ip the
d-ra-reL. So I have already elimir-ateel scum of the thirGs that I ras goir to ck realisir that she's goir
to read than this ray, even thct, I cbi't believe in the the ray is effective at aLl.
E. Do yai feeL that she is goir to be cverly critical a-ti that ycu ckn't have the freecbn to
p-esait the Itia,t as ycu ri*.jld like to.
S. No, I las jist realisir that in terlal of my life it is jist sich a heavy head thir* all the
tine. Evei-ythir is syth. Ycu ki-u.i there is rot saueam forcir rim to cb it, bit I feel that my head is
irg sliced a-ti that is hcw I an lockirg at the werld - as a cLinical sythotogist. Hearir9 my aii
ccmersaticn, hearir the ray I look at thirGs, how I cerve then - I g.ess that is wat is gettir rim thai
- erythirg is malyzed. I cpreciate thirs, I see the hearty of it, bit the, I a-alyze it. I'm rot
satisfied with jist cpneciatir9 the bea.ity. I'm inpir all this stricture ai myself. Ar-ti saietines I kild
of get cut of myself ad look heck in a-ti say fine ytu ca-i have that, bit rather thai 95% Let's cut it heck
to afxut 50% off the jcb - cii the job you can be 95% a-ti be excellent, bit let's cut it ... became ... rot
rer raited to
be
ai the outside. I ±i-i't rait to
be
the role in aLL social reLaticrshi as the
thologist. IJia, I'm with prcple, I reaLized that all I talk aixut is
droLogy a-ti I'm talkir with
other- friar ri,o are rot in ology a-ti I'm still tatkirG thxut syohology. I saletirims thirk, gee
saletirras I ctg,t to talk aixut saiethir other thai s'd,ology, bit that's i.hat atweys calms to mird
particuLarly all I haig arard with is grthte stu,ts in mholc9y. I thr't e.i hwg arctrd with then
ry inch - we th,'t have very rialy frie-t. No wed to spend a Lot of tine with Fecpte, bit we thi't cb
that mxli aimore. The last year I saretirmes w*e ip wishir that I rculd rot be ire, I'm jwt so tired of
428
ire ad draggir this mird arcu4 hit that is ret the case at all, this year.
E. Well, have yw bear thirkir aixut ha. yeu car nify yctr behavior ?
S. Yes, I jwt i.art to kird of keep it in mird.
E. Are these jist pessirg thcaits or are they serias?
S. It ney ret he sericte erxi to ck alythir9 rix,..it, hecase of the sitwticn I have pit myself
into I swlxct that I will ccritirae to c it. I will caitin.E to ect this wey ad stilt be fn.strated. I
thirk I look forwerd to a tine- sirte reedir Tan Weld I've been rrir that if that tine is er goir
to care, it's goirg to be lcrger thai I expected. It seeim Like the Laer I stay in it the sore ttrret
revisicn that is goirg to care ad I'm rover goir to get cut of it. The q,estiort is iat an I goir to th
aiait it W,ich is pretty nuth tharacteristic of ire ad ret nmkirg decisicre. Thirkir ip the cptiam, bit
rever actwlly decidir rkiith wey to go, so I rarmin frtstrated I experierte a tot of fn.stratim here
tecase I decic I ited the eper, bit I ckrr't t to pit ip with alL the crep.
The other thirk (sic] that really isn't goirg thra, my mird riit ru is the i.i,ole sitImtim
within the depertht ad ret pessir the preline the seccrd tine ad that realty affected my self-ccrtept
ad really aere of ccrstaitty beir tested. I thir* the tw. big pecple .kro are head of clinical are ret too
fcrd of so ad it really has affected ire. Ei I &i't krxw in a lot of respects to their I thi't ithitify
with than as clinicia's. The fact that they are in ixver ad that they are caroUr my it's reaUy
teen a big decisia, for ire to neke as to Eiether I an goir9 to d i,at they ,t ad stay here ad get the
degree or an I goir to break a.my ad take a rest. It .ea really hard ad sort at çofu(. to g
with it. So I necie the decisim to stcp actir9 cut so nixh ad rot to ccrrfrart pecple in p.blic ad to
casult with than in private. Q-e of my au irseoxities t to se the rest of the str.xrts as a sl.ç4xrt so
that I felt that I had sore stra-th by ccnfrcntirg the professor in class ad to get the stuits agairst
the professor irsteal of goirg to his office privately ad talkirg to him aixut my feelir i.hith prc*thty
wxitd have prcri,ced sore chae, bit I thirk it wes the grcp sqçxx't I wes lo*irg for. After I rra my
isia, I felt nuh better rixut acceptir the stuff. I an gettirg nãrere the other wey. The peinfut pert
is feelirg like I'm tested ad jr.st this kird of peraoia - uxrir iiat dees the professor thirk or has
he tu'ral agairet ire - especially in these clinical ireetirgs ãiere these tw bigges are in,vl. I've tern
thirkir that there ast be sarethir ira with ire sirte I didi't çess die çxel ire the sxd tree. ('iii as
gcod as e'er4xdy else ad jist a kird of fear of hecanirg assoriated with failu's. Mi I goir to go thrc*,
aother year ad fail it again in the splr'g or be failed ar it. The starde are nude hirer ad hiier
eac*i tine I go thrcti it. So I've bear sr.çer depressed this week. I've Lear qestia,ir, is Life in
clinical peyehology goir to be this wey, will I really have to thae my Ld,avior in order to get alcrg
of partiwlar paple here ? It's bear
really kird of fri,tenirg. Thirkir here yeu've bee, goirg alcrg so well. In the fall yw were jist feelir
like the hiiest yw have ever bee, in yar life. Ycti krew jr.st feelirg aper self-ca,fithit ad mi sirce
the winter ad all this werk in my heed ad rot keepir9 r.p with it ad fallir heuird. I've beer iaxrirg
with pecple, or is it jist this particular sitratiai ad a ccqule
if there is saiethirg wrorg with ire ad have very seriasly been thiokirg within the last week aixut goir
into ther' ad kird of Lookirg at it ad firdir cut. I ised to really accept my a..ri evalwtiai of thir
ad rt*i I'm q.estiaiir melf ad my evallEticra i.i,ith is really scary. But sirte last thu-sdey ad fridey
I've been feelirg nixh better. But nucy, tteschy ad we&escby I wes realty depressed ad I'm rot Laelly
that wey at all. O-e dey I mi,t be depressed bit that's alL. But that's ret army mird taii,t. FiLlir cut
that weekLy thiok nude ire realize that nury I an depressed. Tiesdey I dep'essed. Wey I wes
ckpressed. I tha,t Ken this has been three chys in a rai that yw have i clepras. d. Ycu've rover bee,
depressed this lcrg.
E. Is that aixut it mi ?
S. I car talk forever.
E. Ckay, the, I'LL say gocx±uiit r.
1st REM Awekenirg, 5 mm. into REM, 1.40 a.m.
S.
kird of p.rsuit, sore ji.st kird of the feelir of PXSuit.
I%xn q,estia,irg KJ said that ro other pecscre were invvlud.
429
2rd REM A,,kenir, 10 mm. into REM, 3.05 a.m.
S AgainI±n'thave'thiruverycLear,hithatIhaveiskirdofenmnmgeofaperscn-ci,cJi,
aother sLeq Lth drean. The part I reiether r is: I an in a slew Ith axl kir1 of there's a sietl roan
irtetween twe roai axi I'm staxlir there j.st at the ntnEnt that ycu cat Ll ne waitim to u,Lcdc that cixir
to go into a-other ±or iride. There's n±xxty in the iri roan rit rt ãiith sn.rpris&1 ire txica.ise a few
mirutes ago there was a iJrole bnth of pecçiLe in there a-nJ here I'd cane hack a-nJ it's aipty run so I'm
stadir there Lookir for a key a-nJ thirkir9 thir Like nmybe there's a wirthn iJiere I'm sta-dir a-nJ I'm
thirkirg thxtit the helL gane or the city or i.iny it is the ckor is Locki or sarethir Like that. There's a
Little bit more I thir. us I was staxiir9 there at the cbor tryir9 to unlock it - ycu krmn, lookir for the
riit key, I thirk there was a wirthn - kird of am cutlet to the werld again. I thirk as I was sta-dir9
there funblir9 for the key I was thirkirg thxut the helL gane. Ncw a Little bit earlier en I thir-k I had
bean in the roan I'm tryirg to get into with a n.tnole lot of pecpLe, I thirk ever.1xdy watthir the helL
game or sane pecpLe were watthir the helL gale, samethir Like that. I had been in this roan earlier with a
Lot of pecpLe a-nJ - I'm ji.st an the tip of rane±erir sane more stuff with that. I thi't runaxir the
saxe, hit I krun that there was a feelir9 of ctegatad resjxi-amibiLity associatad with there heir9 l
sista-nts in the dream - kird of Like a direct tie to n4nat's h enir here.
E. Try a-nJ ccrcentrate jest cn n.kmat hçaed in the dream a-nJ save the associatias for the nocnir,
S. This was in the drean too, pecpte had been clegatad resxrsibility as lth assistaits.
E. Were ycni c*-e of these pecpte ?
S. Not I, same of the other xcple in the dream. No, dich't reccgiize any of the other pecple. I was
samehew a perticiFa-nt, hut I knew that I wasn't a clele... di, wait, I thirk I was a persai nIio was
c1egatad respensibility too .es, stax1ir there at the cborway, tryir to uitrxk it was a se-se of rot
jist beir-g a Farticiait, hit I had same resjxrsibility. W-nat had hpa-ed earlier, I was thirkir9 axut
niat had hed in the roan earlier a-nJ ha it was different fran riit there run shere like I'm all atere
Like in the niiole sitn.tiai there canirg cut of a-e bigger roan ad then beir in this kird of mnell roan
irtweai n.knere I'm sta-dir a-nJ then goirg into this other roan that I'm tryir to go into. So there was a
same that I was cn-e of those pecple with resp]-sibility. It was very mnxh like the slew lth resprsibility
eva, in the dream hut I cki-n't rema±er it specifically beir - I thirk the persa-n in tharge yeli - was Dr.
Cartwi-iit. I th-n't reierher seeir9 her there bit I krtw that was n.kno was in tharge. In the dream itself
there ccnld very weLl have been an aare-ess that this was a sLew l, I'm pretty sure there was hut I
ccuLcki't thotutely say that there was.
l.xti further q.estia-nirg he statad that as he was stadirg there tryirg to get into the ckor the
awira-nElt was iliite, 'it was a solid color', 'it also had the feeLirg of kinJ of Like the jthn - with
that tcwat rack a-nJ that ca-ntirw.s cloth tc*.neL. In a se-se that was n.iiat the key - I thirk the key was kimxl
of like the key a-n those tcseL thirgs. The roan I was staxlirg in kird of ha-I that ceLity Like the jd,n. As
I was sta-dirg there either I was wa-crir nJiy I was havirg difficulty uilockirg the ±or or n.iiy everycre
else wasn't there - I miit have bee-n eadirg a rote that was an a wall, cu the ±orway that I was tryirg to
get into. That sn.gestad to ire that those pecpte 1io were there earlier shcuild be there still.' He goes ci
to say that 'there were other pecpte in the early Fart of the drean hit those are the other pecple I ckn't
reccgiize as beir famiLiar.' Fir-ally, he remarks that 'sta-thrg there at the cixir, was u-pleasait.'
3rd. REM Awakmir-g, 15 mm. into REM, 4.55 arm
S. This is a-other slew Lth dream. I got i fran the sleq Lth to go to the ki dien to get sane ice
cream becase my mmcnith was real dry a-nJ I'm staiJirg at the refrigerator with the freezer Fart i ad I go
thram a cc*.ple of ice cream bars a-ri then I go thrc*, a fnxesicte ad all the tine the refrigerator
cbor's open. Then also, I gess I also tork cut of the freezer a calera a-nJ I have a camera arcud my reck.
ThanI goci this is Like inmyertiasit - a-nJ I go into cre of the other roais axl I'm watkirg
arcu-d there a-ri I'm rot sure nJiat I'm dDir-g there. I waLk arc*rd there for a few mirutes a-ri thai Ramary waLks in a-nJ then Roz walks in a-nJ asks ire nknat I was tekir9 pictures of a-nJ I said sanethirg like
I wasn't takirg pictures that I was carryirg the camera ar-aid becaise of the Liit meter - or nunbl.erl same
kird of thirg that I dicki't eva-n u-dersta-d - becase I dic*i't even krun nJ-rat I was ckirg with it ci. Then
the nhole sLeep Lth crew - twa or three pecpLe - wait into the kitchen to sake liverwjrst saxi.iches ad
ycu' re alL havir-9 a real good tine nakir-9 I iverwrst saxiches ad nJiai ycu giys all came heck I wait hack
into the bad nJiith was Like my Ircan at hare a-ri Ccrnie (my girLfria-d) is there run ad she was with me
430
i4e, i wes '.p talkir to ne afxut salethir9 ad I wes aroyJ E' it, I dich't at to talk I Jist ited to
go heck to sleep. The, I it heck into the Iroan. Lets see there wes a - there were t cbjects I wes
ccrcerrul with hair9 nzwei I wes Lyir in the i with my feet at the ad my heal is is&lly at - in my
at haie. I asked her - I wes lakir for a place to pit these t cbjects. Qe • like a mirror ad the
other - I'm rot sare, a samll dresser or a herrel or na,4x a TV or saiethirg, at aiy rate, she placed the
cbject that I ca,'t ramiber over to my riit side of the bed ad over ai the Left side of the bed shere the
lv es she placed a mirror ad I dicti't like the piticn ad she wes talkirg to ne ±*xit this ad that ad
I dick,'t it to talk to her or athirg, I ji.t ted her to go wt - I i.ited to go k to steep ad she
pit the mirror in ore place ad I said shy th,'t j ji.st leai it ip againet the well over there ad thai
she wes ccrcerr about it wouLd interfere with the TV if she placed it too clam to the TV - it uLd
interfere with the aerials - the e,tame cii the TV aid so she stirk it over cii the left side ad asked ne I ckn't reisaiber shat it wes, she ited saiethir fran cm ad I di±i't ,t to hassle with her, bit at the
very ad she wes askirg ne if we cc*jt&c't go to the Italia, Village ore niit, she wes kird of lookirg
forwerd to that like I wesn't talkir with her or did,'t ,t to sperd aiy tine with her so she wes kird of
Like askirg ne to nmke a cannitnit for cs for next weekerd, next satirday ni,t or samthir to spa-cl tine
with her Like goirg to the Italian Village for dirrer or saiethirg.
&,ai asked if aiy distorticis were presait he replied 'Ccnnie was nakir arraa1its for goirg to the
Itatia, VilLage, that seeied Like a distorticn, evei in the dreen, as soai as she said it it did-i't scu-d
riit, it was as if she was kird of schciijlir a, pintnent shic*n didn't seen riit at alL - kird of ne
fittirg her into nay sche±ile - that dich't nmke sa-6e. The other distorticn - just with rot kroiir shat I
es thir9 with the canera. I felt sanehcw goilty or ca-spicw.E or saiethir like that shai everytniy cane
pxrir c.it of the roan ad I was staxlir9 there with a canera ad I didn't krcw shy I had it cii ad saieae
askir9 ne shat were *i takir pictcres of a-cl I .asn't takirg pictcres of a'thir - I dick,'t kri shat I
wes cbir - kird of felt like I was cait with the goo bit I didn't kru shat I had the.
Ijie, asked xiUt his affect dxir9 the drean he stated that it was 'Real pleasait eatir9 the ice crean
bit after everbxty cane act ad started nmkirg roise ad stuff then it was irpleasait ad I c.a,ted to go
Leck in there a-cl fall asleep.'
He the, a±J ax,ther pert of the sane dream: S. This is nuth nore weird. I'm in the
udergrc&rd cm the el ad the el canes trtckirg in a-cl it's like ai oval she.l track - pretty large, a-cl
I'd just cane fran the cçpite side of the oval cii a train axi I get off the train ad it's real shitty
Like the el is a-cl dark ad dirty a-cl ily, I get off that train ad I'm goirg to La staiJir there waitir
for the next train to cane - tu kro., I th,'t walk thrc*n the tu-rel or athirg, I just staid there in
the sane sjece waitirg for the next train ad shile I'm stadir there a-a train canes bit it's rot my train
- sane pecpLe get cm, sane pecple get off - rothir9 of ccrseqme it cbesn't seen ad -- there's reaLly a
Lot missin-g fran this dream, I've cnLy got pieces of it -- ad I go p sane stqs into like a waitir9 roan
Like a train &pDt aid I'm waitin-9 for the next train. Ch, before I did that I was tryirG to thirk of i,ith
trainIwaswaitirfor--wasita,AtrainorwasitaBtrain?Ihadtothirkforamiruteshithitwas
that I baited ad ad that other train, she, it cane alcr was ai A train a-d the e-ireer or ccn±ctor
caLled it c*it ad I reneitererl havirg to decide hn, cb I nt the A train or B train ad so the A train
cane bit I was waitirg for the B train, ad the, I wa-nt beck 14) to this waitir9 roan te thir ad I wait
adsatth.,, ... di, di, theplace that I satck.w-ncnwes tiers of ccnrete or sanethir it wasn't stairs,
it wes like a sta-Jiuii or sanethir-g, bit no backs cii the stairs, just ccctrete, ad I was sittir an ore of
the tcp tiers aid sane giy cane alcrg sho acted as if he werked there ad he was ur- ad rTmyLe wearirg a
CIA u,ifornn or saiethir ad stood ckwi there helc*n ne , the rawestad or new rack [sic] or saiethir ad
he, in a real rasty kird of tore, I g.ess kird of ai accisatory tore said 'Wiat the hell are u cbirg here
?' I feLt very, an the a-a had, ccnfused as to shy he shcul&n't kr shat I was cbirg there becase
cbia.sly I was waitir9 for a train bit I decic to he wry facetias in respxdirG to him ad I said 'I'm
aitir - I cane here to hey a, aitan±ile' in kird of a sarcastic or cb-oxias way to tell him I'm weitirg
for a train - like shat the helL cb .tu thin-k I'm cknin-? This is shere you wait for a train ad that's ii,at
I'm cbin-9 here a-cl he was sanecknat take, aback by that a-cl thai I thin-k realised that of carse that's shat
I was dair-g was waitir for a train. Thai a few mirtites later there was a shoLe ness [sic] of prcple sittirg
cn these tiers with no - I ckrn't kr if tiers is the ri,t wan-cl, large sts that you'd sit a-n - all these
FeCPLe are sittin-g there waitin- for the el r ad this guy was still there a-nJ he addressed a gssticn -
431
he stiLL kird of rmsty or very ccnfrcntir I thirk - ad I s stiLL sittirg ai Like the hiest step rot the hiiest bt the rrext to the Last rcw ad eer4xdy eLse iøs in frcnt of ire so I cc*.iLd see evertre
else ad I th,'t reisnber the cmtext, .kiat preceded it, b.it he said - be started Like caLLir- wt
reticrrelities of prcple to see tio eertrre s ad askir then to care focrd - Like tu'd juip to cufeet ad care rirt thri to the bttan LeveL there as he caLLed it wt ad it s reaLLy kird of ca,fra,tir
Like 'ALL riit, i.kro's PoLish?' - 'CX, Jio's ItaLia,?' axi he t thrctr Like three of those ad dick-r't
get ire in my of those categories ad like ever1xdy else bed Im caL Led ad thm he said 'CX, rho's
Jish?' ad that s no ad I Jist kird of barced dewi cre step ad rareir seated. Again, I i.as kird of
reaLly a-9ry ad realLy and at his i.krole attitu a-cl his lkrole proach in a essive-aggressive - ku-cl
of canir-9 c1rr a-re step, b.it I ild-r't carpLy with him fuLLy. I i.as canirg &w, to -- gettir9 ip ad goirg
ckw-r me step to ac1ax*Leee that I i.as a Jew a-cl I ±n't krow iEry he as cbir it or i.kiat the caitext as
ad so I sLid cbn that me step. As I recaLL, I as su-prised that he picked no ip as lir9 a Jew becase I
tas the arlycre that Larkeel Like I i.as a Jew, rthxfy eLse did tone ad it seers he bed kird of a sin
mniLe ien caLLirg a.rt the naticraLities,
j kr, Like gettirg eve4:xxly, there 'ore nmkre tm pecpLe ad
I as sirpriscri at that ad it seei Like he bed a reaL sin., kird of carpLacmt sidLe - lcnir9 that I ras
Jeish or saiethirg. Except, after that haqxred, at as goirg thrc*, my mini .as, gee, here I an
&eenirg that I havm't bee, ioken to give this rejxx-t ad thm ikrat I tha4it
of i-ext isis that here isis
Roz again a-cl I sew her sheet ad a, her sheet ware Like - in te - isis iat bed ji.st occured betwee, ire
ad this train giy in the train staticn. I reed the List ad it was jwt kird of a netter-of-fact, cbjective
rortirg of ikiat hpmei a-cl that the other goy caLLed the rares a-cl prcple stoed a-cl a-cl he caLLed for
the Jew ad Km sLid th.n into the i-ext Lc.rer thirg there. It dich't have my of my kird of arotaticn or
stuff like that, that I'm wed to seeirg m my drean records, it was jwt kird of cbjective, ikrat she
thctit as the &server haç-ied. Nci I'm reaLLy caifwed as to itiether I bed this drean - this dreari
marred before the other Lcr9 me I jwt toLd yw xut with the sleep Lth - bit I'm rot su-e i.kiether I
twLLy ke ip a-cl thc*g-it th, fin, I jwt bed a ck-ean ad they did-r't wake ire or ihether, ycu krow, I
&eant the ii,ole thirg. I thirk that's nostLy ikrat I i-ureter.
He a±i rrext:
S. 'This me started cut realLy reitral a-cl thei got reaLLy eroticreL - I was
reaLly ary at this gjj, - very aroticrsi(.
'The arty prrsai I reccgrized was Dr. Cartwrirt. After the eara-te of this giy - after I was
ref Lectir-9 m my exria-ce of havirg the drean ad rot burg aiakened.
'At the Foint i.tie, Dr. Cartwrit was there - ckn't ki-rM if I was asleep
uthrstaxl ikry that was Lecase I tha,t it was definitely REM.'
E.
is there aiy nore ycu cm recaLL?
or
a-ake. I ccuLch't
S. There was jwt a fLash of birg a, mi el. I ±n't kru if it was before I got to the staticn
if I got bock m me Later. I thirk I isis goirg hare - I'm rot sire.
or
4th. Awakmirg, 20 mm. into REM, 6.14 an
S. IwasgettirgLpfrantheLthinthenomirg-IiasptajrgwjthabLackLm,tbiLbatthe
mrsnt ycu caLled ire a-cl I bed jwt kj-irxkeei over a spray mist refresher cnto the fLoor. lkrat the dr-ean's aLL
afxut is -- I thirk I bad ii*m ip fran the dreari a-cl I was in a sLeep Lratory aLthcti the settirg was
differurt ad I weke ip fran the drean a-cl I wmt ±wi to the u-cl of a mrrcw corricbr wearirg my ejaias or
Lth cart or saiethirg a-cl bavirg aLL the warks a my heed ni-cl stoixi there at the ae of the ckorway a-cl it
Like Clark St. ip m the rorth side a-cl there was a La.rdry at the u-cl of this haLLway ad across the
street there was a Large Laxdry a-cl dewi the street to the Left there ware FecpLe Li ad ip at ar aitckror
kird of Like restaraits for breakfast. It was strr' over there ad the, ad the, to the rorth, waLkirg ckw,
the sidewalk ware Lots of pecpLe a, their iay to wark aid I was dispinted that it was nori,irg aLready a-cl
lsiS
there I isis ku-cl of sta-dirg there, rot rureterirg havirg gotte-r cut of , jwt rereterirg bavirg waLked
dew, this kirci of haLLway ad the, Dr. Cartwrirt again, she's jwt eeryi,ere, cares aLcrg ad kird of
ishers ire Lock to the sLeep Lth sayirg 'Chris n't Like yixJ if ycu de this' ad I dickr't uthrstad ihy she
'as telLirg no that Chris izuLch't Like it - I figored that she did-,'t Like it - she searad kird of armyeel
that I bad ga-re cut there - it was a stinuLatia, rt of the thirg - so I asked her - I shcuLct,'t be gettirg
a.itside stinuLaticn, it wasn't aLLc - I asked her if there was a differu-ire Letweer readirg before I iit
to Il ad my sta-dirg there Ltxkirg at the pecpLe ad she said there was. So there I was beck in the L
bit it was ±ysicaLLy differe-rt. There was Chris tatkirg amy with a b.jth of ycu sLeep Lth cpLe warkirg
or him a-cl thm - Like he'd sLept in the Lth that nirt - axi the, I waLked ' the haLLway to a-other i-can
L32
ad there t,ms Jthn with his cbor rtLy cça, ad he was gettirg i. Thai I go into the roan iJiere I was in
ad again it's kird of Like my erth-it - Like there's a dinirg roan a-d a sin parch ad that's iJiere I had
slept. Herb was there ad he was gettirg dressed to go cut - he was ji.st wakirg ip ad gatherirg ip his
stuff ad he had cxi his ,ts ad jaia tcçs ad I was kird of waLkirg arctrd the roan aLcne ad Lookirg
cut the win± ad there were Lots of pecpLe waLkirg ±w, the street ad everxdy had a winter coat cxi ad
it wasn't reaL coLd winter, bit faL L. As I Lookeci c*.Jt ad sa that I tha.it to myseLf that I'd better wear
a jacket to schooL to be warm ao.ii a, the notorcte. Thai I was ±cidirg hc to get myseLf ready ad it
seaiad at the tine that I was thirkirg that I wasn't satisfied with hcw I had th-e it the Last tine I had
been in the Lth so I was ccidirg âiether I shc*ild gather i.p aLL my stuff ad then get dressed, or get
dressed first ad ad ip havirg to walk beck to the Lth severaL tines. I watchir Herb to see i,at he
had ckne ad so I g..ss I gathered ip my stuff or saiethirg - I pit cxi aLL my cLothes ad then after I had
t dressed caipLetely I reaLized that na I didi't t to have my shirt ai becase they'd get the
collcdicn aLL over it ad e.erythirg else ad that I'd prc±thLy have to take off my shirt again. Thai I
balked into the sin roan ad was hoLdirg this very imlL, shed Li,t buLb that had been for a black
liit or saiethirg ad I was nrir if it iLd wark for my taiL tiit a, my netorc1e iJiich was broken
ad I looked at it a-id realized that it wajLct,'t. Then I waLked into the front roan hotdirg this ad krxxked
over a cai of air freshener. I was aLso Lookir-9 cut this wiri± - there was a kind of ai oLd schooL nd ad
a old sd,coL bi ldir- ad there were tw3 sets of th.ble dxx's that p.sh cut fran the irsic ad as I was
atthir9 saie stits - the thble dxr cii the rit was cicsecJ a-id there was a i.ãiote a foot ad a half
sqare cxi the other haLf of the ri1t had set of cbors ad everylxdy was caliirg cut of the Left had cr
except as I was sta-dir9 there watchir-9 a mielL child cLiidrs thr*xi this cbor. It strtck ne as reaL weird
that sarecre iajld go thrtxi alL the effort of clinbirg ip that hii irstead of goir9 cut the other chor,
bit I jiEt fixed weLL, that's shat kick ck, d saiethirg jLst becase it's hard to c or for the reLty
or saiethir. This ' a reaL mmLL kid a-id I was sirpriseci at ha., weLL he had aainiered that a-id then
there was a ni.ri, Larger kid, this tine Like he alrizst waLked thra.i, like he'd jLst stepeci ar the icd
thrai the hoLe a-id the hoLe was qiite a tot bigger r so this kid didi't have to sq..Eeze thrc*, or
athirg, jLt stç cxi thrai. Thai I said Cl,, that's hcw the other kid got thra, becaEe the hale was
so nu± bigger - there's a distortia, there.
EarLier in the drean, I thir-k before this shoLe seqte bega, that I'm teLlirg yw thcut - I was
in a sleep Lth en the talkirg to my carçaiicn, iJ,oever that was - that is, it was a-other neLe i.io was
in the sleep lth bit I ckn't recogiize it as beirg Jthi or Herb a-id I had jist caiplair thcut beirg
thirsty again ad sarea-e had bra.it in fniit juice aid I ha±t't dru-k it bit I was talkirg to Wioever my
caw-ticn was ad Lo ad oLd in walks a-other a-e of the ecple fran the Lth and this tine it was a rraLe
J,o walked in with a-e of those Large ca-s of, I thirk, i-icot ora-e juice. He had overheard tie thra.,
the micrcçttcre and I feLt kird of bed beca.se I had aLready had this other caa of juice saieore had bra,t
tie earlier. I expressed my preciaticn for that and he ,t beck cut ad then it seai Like there was a
iJiole series of gifts i.ãiere saiea-e fran the sleep lth wuLd waLk in and give ire a negazire - ae nazire,
I gess it was Ne.sweek. Ard thai a-other tine ae m fran the Lth, ho seaiexJ fani Liar, walked in and
cçened the nazire to a e with a fuLL-size cartcxzi drair -- a draiirg of a sigiificait rscn, Like
n4e a fat characterizaticn of Nixcn, bit nm4e it was Miity Mace, I ckn't krt, jist a faniLiar face
tirgipa l,ole age. There was no ctim. I was thirkirg in... I ckn't get it... neybe this shots I'm
rot aLL that iert. I miit hate fireLLy gotten it bit I ckn't kr,. Piyay, I didi't thirk it was very
ftrri a-id I dic*i't krow tiiy it was beir- pinteci cut to tie. The negazire dic*i't have a rare cxi the cover, it
tea jtst Like a weekly nets nazire or saiethirg so I assured it was Newsweek, I thi krn tkiy - I ±n't
read Newsweek, or aiy other weekLy news nagazine. The cartoon nede ire feeL reaLLy self-ccrscicw ad seLfdeprecatirg... I ccuLch't eai see the pint of thy the rrscn shcx it to tie. The roan was hot and my
nuith &y; my caraiicn agreed ad I feLt canfortthLe with him.
lxn cestiauirU he raTer-ks that Clark Street lcekecl wry different, that the episcx with Dr.
Cartii-iit was irpLeasa,t a-id that the rest of the dreaii was 'Less irpLeasent, prctthly eti neutral.'
Ackiitict-aL recaLL to this cb-ean, 6.55 an
S. Amther rt of the sare drean jist care beck to tie. I'm in my office ±w'stairs and ioai 9nith
wes there ad I went ip to her ad mthraced her ad kissed her cii the cheek. She was sittirg at a-other
iy's desk i-Audi is jist exactLy cqxsite mine in the office so that it i-ild be Like the desk I got ip fran
a-id her desk were La-gthwise at cçxsite axis of the roan. It seats that after I athracecl her saiecre waLked
433
in ad said saiethir - Like thxut this display of affecticn beir very curias or saiethir like that. The
rsa-i i1io watk&1 in wasn't Lçset or shockeJ or athir like that - jLst kird of - like sr.rprisrd, hit also
a se-se of pleasr.re at seeir9 it.
E. Did )#tu feel this?
S. No, the rsa sho walked in -- well, ,, I felt that way toa, very hy axl I iited to
expess my affecticn - I felt real gocd ad Joe, felt real goixi hit the rsai wio walked in was kird of
su-p'ised hit it wasn't a rejectir or shocked kird of thir9. It was pleasir9 for then as well. It seers as
tha it was a wim as cçpised to a rim, I'm rot sure ikiether it was sarea olckr or xer - like a
stithit or faculty kird of thirg. Really rot sure i.ho it was.
Ki's Interp-etatiai of his Dri&.
1st. Awakerir: u-certain, rerhs sare rcesaitaticn
a-d pesait ca-cerrs.
of
hecticriess, tesicn
of
presmt Life style
2rd. Akmir: In this &ean I seen to be lorir for the faniliar exçerie-ce of the lth, fran the
winter sle cai-se. Perhçs Lookir for the faniliar faces irside the little roan, of the çecple fran the
class; they were there earlier, I rqxx-t, refers to Last tine in the sle 1th. Also ra.i in the &ean I an
m the cutside, as if imtir to be nore in ccntrol, have my ha-c in the experine-iters' roan, çxittir
elactrcdes a others, watthir the machines. As &eaTet- I an ctricts as to .iiat is goir ar irside that
little roan r4-iere the experinelters are; mtir to be the sibject, hit also mtir to be the experinmter,
or else feet cler to than. Resp-sibility issie semis to r-ese-rt my lack of resçxrsibitity ase I an
cnly the sibject, the ca,ftsiar stalE fran my str-or se-se of resxrsibility ad my corcern over beir thle
to erforrn - Ferforin via my &eais. My rot havirg the rit key, or rot beir9 thte to get the key to fit the
cixir also reflects a se-se of dista-ce, beir kt cut of the experinmters' roan, fornerly havir the key
to the Ith cirir the winter, beir9 cicser to R.C. thai rom. 9rellress
of roan I an stadir in nay be the
sthologically arell roan I an sleir-g in in the Lth versr.s the large size (tholcgicalLy) of the
mqerinmters' roan.
3rd. Asakmir: (nore Locse strirgs ad u-certainty in this interpretaticn.) I nay be nakir the slew
1± none cammfortthle L' rrakir it Like my erthmt. All I nake of the ice creen scere is my thirst that
ni-it in the lth. Leavir-9 the refrigerator deor cç, is salethim Ccrrrie ckres a lot kriLe p-qarir ford ad
that aroys ire. I nay ha.'e held it cça, in the c)-ean to coal off, as I r also hot in the lth. Camera nay
Le same s±oliain of inagery, penra-elt record of visian, sixh as a record of a &ean - the te. Also te
recorder/carera were associated in theft fran car of Ccrnie's carera ad R.C. 's te recorder. My goi It
xut havir9 the camera in the ci-ean - my ca-cern ±cut R.C. 's reacticn to te-recorthr theft? The la,ter
ad gigglir of the lth crew Leevir-g their roan ad enterirg the kitche,; emterirG my pivate cknain of my
&eams; my cacern over their beirg stra-ers, rot feelir9 terribly canfortthle, a-cl in the &ean as in the
lth, aroyrd at the roise of the lth crew, gilir, occasia-at latter early in the erir, accalFauied
byn, an perfonia-ce audety a-cl difficulty faLlir- asleep. Liverw.zst saxtithes - saiethir I have never
eatal? The conflict with Ca-ide seers to be r-esautative of the difficulties in air relatici-ship, very
nixh the case in the &ean of her .autir to deal it cut, disass it, ad ire beir tired, rot r.ertir to
talk, or deal with it. Ca-nie is askir ire in the c)-ean,
as she is askir9 ire in ar relaticsship, to camiit
tine strictly to her, so that she's rot alwa Last. The l is reversed, perhçs stgests the dista-ce I
feel caçered to before. Ca-cern with thjects ain seers to be the relatia-ship be e-i is, vis-a-vis
thjects I place in the way, or her continal efforts to please ire, irmpm'e relaticu-ship, help tie with school
by takirg care of aimaller cbjects (mirror ad other uiithitifird cbject). I ckn't see how all varias
elmmmts relate into a ctheraut &ean.
3rd. Awakmir9, seccrd pert:
S. krith route cb I iaut to take in life, in school, seams to be the there here. Ha.i in the dreen I
air caifraitir the prcblen of waitirg; my directicn (my train) is rot here yet, so I have to wait for the
rimt train to care in. In a wry real se-se I an waitirg for sigrificait chwges in school to ocar in the
rear future. 'Ikrat the helL are .tu deir- here?' Perhs a pert of myself askirg ire the qesticn; tie
becanir defa-sive, th-cudas, refisir to deal with myself to resolve a-swer ad feel gocd axut it, so I
refise to deal with it, actir irdig-aut. Or yux me nay be anirnxs, gtcI inage of cartnmt, i.kiith I
434
ro 1crer see as old as I is to. The ckrtn,t askir ire kat I'm cbir here, perhaçs my attrib.iticn that
they ±n't see my view at all, that they ckn't reccgiize that I an iitir for a train; perhs my
akwl&erasit in my drean that they thi't ccrmicr my perfoniaxe satisfactory, axi see ire actir as if I
an in a hotdir9 ttem, reither en the A train or the B train. I ,tei the nxr nai to be taken ±eck L'
my reply, bit I ckn't thirk he realty ros, prt±ebly jist wrote no off; again my eectaticn that others see
see ikrat I'm dDir axi accept it. Also wartir to be left alcne ' the train nEster, if that's
iho he is, as I it to be left alcre L' nasters of the deFertmlt. Ci the tcp step I an thxw the yc*r
my psiticn,
as I often feel thove the nasters here at Circle, en a different lel bit still in their train
staticn. I an the nest distait fran the train nester ewr after other cp1e, riro rn.st be other stlxts,
are seat
en the uanfortthle ccexrete tiers (a Ia Circle). The other stits juip ii,en the nester calls;
fliLe I resprd slc*ly, ccidir hat to th, firelly caxeiirG to fall into lire. I in't care ckwi alt the
'ay, I n't leave carpletely my step, bit I will care a little oy. I'll bitterly offer sare caipranise to
the certrent bit I try my best to sake my trip rk. kklirg of other ecple to the tiers say be my givi
neLf the &çxrt of other stithits, perhs enly to fird alt that they are will irg to step forerd axi
carply fully with the nester. My surprise that he reccgrises ire as a Jew, scuds terribLy nixh Like a recent
açeriece I hai with aro of the 'nesters' iho surprised ire by seeir thirs in ire that I thctit i.ere
hickn. Futherrrore, .s beth beir Jets, perhs I felt I shculd have received better treatnit fran aother
Ja. Firally I turn to see R.C. iotthir cwer ne, &servir tkrat's goirg en, bit enly recocdir it
bthaviorally, my sr.rprise to see jist a strairt acca.rtt of the interacticn. Sitiotien with R.C. semis to
ref Lect my catern with R.C. as rotective fis-e, notherly or otherwise, a-d my seeir her less interest&1
(in my] relaticiship with nester, bit FErhs nore as research sthject.
4th. Ainkenir S. This drean sears to strike
en
tto Levels. The first sears to
be my
alticiFetial of the Irtmir9 in the sleep Lth, iokir p early, beirg dissatisfieti with my niit's sleep,
with my Ferforixe in the Ith. Qi aother level it nay be again tpir the acanic ccnftict of t1,ich toy
to go, caceni for i.1tat's geir9 cii irsidr, wrsr.s that's goir en cutsi±; cutsic the cartnit, cutsic
in the city of Chicago. My iolkirg cutsic to Clark St. becase of my caiplaints of rot beirg thle to fall
asleep fits wry clesely with my real exFeriace in the l, havir breakfast, but in the drean the caern
for gettir9 dressed e,çJiently - very nuh a caxem for ire that nornir became of ai early ointrrmt.
The efforts by the lth crew to accarmxte ne in the dreaii, nnxh as I preciated exFeriniters' efforts in
the lth with the cp of toter, carefully pJttir en electrcdss, gaeral nice treatrrastt a-d attenticn. Alcrg
the other level of aalysis, R.C. axi Chris are both ray clest &lvisors. R.C. semis to be sakirg for
herself a-cl Chris in atxrragirg ire to attexi a little nore to hat's goir9 en irmi the c rtranit; speixi
Less tine for the present cut ar the streets a-cl sore tine irsic atterdir9 to school. The prcple Wtsic
aarirg coats nay reflect my cacern for the Fall, my heavy bed in the certnit, a lot to carry.
FLrtherriore I an in ray real Life piticn of sta-dir at the wirn±w of ax erias,t lockirg cut at pecpte,
Wiile I take breaks fran my rk. The to bmys 4io clinb thrai the wirthi are chcsirg different rwtes to
get out of the school. They, as I an tryirg, are rot chocsirg the faniliar, easy route. The first b' has a
difficuLt tine, snngestir9 the first cviait has a difficult tine, bit the cxnir withs for the rext txy.
If the boys are ire, t.kiith I presule they are, the sr.gestien is that if I stick with as route,
Fersevere, it becaros easier to get cut that other dxr. Or, that
as
or
if I
I took at the prcbleir a little lcr9er,
the sqere of the cçiir, (the sqerm-ess of the cortirait, the limitedsss) becaros less limited, with
fore ram to Fer-init cviaits to have access. To extexi my arbialete in interp-eta • en, I cculd also
stest that the children exitir the school in the drean are the stuits gr&thtir, that nest care cut
cre toy, bit still there are a few that d rot fit the ronn. In that se-me I nay be willirG to reirein here
at Circle as lcr
as
I cai lccl to the future a-cl see myself rot becunir m-gulferl,
or
broken, bit that I
cai at Least crat cut or clint cut the toy I iait. To no Netstaek is scnethir that ore is sqçxDs&l to
real, that is often very interestir to reed, bit tiiidi I read crily in dxtor's offices, etc. The experiece
with the cartani in the negazine s&gests iay feelirs of irmecurity, inferiority in regards to my prthlars
Fessirg prelins. My hid, expectatial of myself, yet my discre,t Ferfornu-ce - I fijre I should be thle to
urcsta-d the cartoen, I figure I shca.ild be thle to ess the prelim, yet I fail in both tasks. This failu-e
in the drean a-cl in life is of trenas-cb.s cacer-n to no as I challe-ge my self-ccrcept,
Calpetece, tx1er if I have itiat it takes.
Sitatien with Joai 9nith is cisias; I ckn't krv..i at that Feint it occrxraJ in the
I kithutz occasiarelLy, rrekir9 sses at each other. ReLaticrmhip to drean is uclear.
435
my same
of
She a-cl
lkxn further interviewir he said that all the ck-eaas 'seen to relate to my cwi a,xiety thxut
erfortnirg ... The fact that I cculch,'t sleep. I idce ip pralEturely ad jt start wa1kir arc*.rd ad
stuff Like all that leaJir to nDthir - ji.t lyir ake b.siress is rot for ne. Also my caxem with all
the mise, all the gigglir ad I was LirG sereitive to it becase I wasn't fal1ir asleep.'
E. Ycu were acti.Lty hearir the gigglirg or yw were eauirg?
S. Jim I was a.ke I heard the gigglir ad i,en I was dreanir, I also heard the gigglir. t,a,
ertre pxrir cut of the roan i,ith ever it was into the kitchen ad gettir their stuff Like that
nekirg liverwjrst satiithes ad stuff like that.
E. So that was also a drean, ycu' rot beir thle to faLL asLeep. The ice crean was relat to jt
that ycu weren't thle to go to sleep.
S. I was thirsty. I an rot sure shy I was staxiirg there with the refrigerator door cçel...
E. I thirk rit after yc*.l ha the ice crean ycu took a carera cut of the freezer.
S. I ck*i't recall takiog it cut of the freezer, bit I raasiter havir it arcud my neck i.ien all of
u care cut ad I felt kird of goilty haviog it bit I krew that I ha*,'t chre a,ythirG ad t I still felt
giiLty xut havir9 it ad I did-i't take aiy pictures. ... Each tine I wake ip I pAl&J the bla*et further
ad further dcw,. It seared to get hotter ad stuffier as the nit it an. Ccniie very often leaves the
refrigerator door cç in the rthmt shen she's preçeriog ford ad she doesn't like to have to keep
cçBlirg it...
S. It seare that yai were all goiog into the kitchen to neke liwrwirst sadiches ad I uiteJ to go
heck to sleep or scnethiog. I was eatiog ice crean for qiite a lcr tine. I haJ three different ice creaTe.
ThefirsttinElhada- thelastthir9IhaJwasafuecyclethatwasc*lastickthathaJthepqerall
shitty an it ad soft ad hard to haxile. There were twa kir of ice creari before that, bit I chr't
rsrerber shich kird they were, like ice crean an a stick kird of thiog. It seared like f ecycles except
that the-i I got to the freezer I knew this
different fran the others. They were all real gocd. ... Then
thai I went into the -ormn Ccrnie either foltc*1 ne in or she was in there already or we weit in there
tcgether. I was lyir9 &wi. The Led was at the other erd of the roan ad I was at is&lLy is the foot [sic]
of the I-J ad there was the sare sace that we have an the wall side of cu' i, bit there was sore a, the
riit side ad for sare reasai the way these twa cbjects were ptac&i was sareshat irrportent thict, she •
tryiog to do this for ne. I was lyiog ±w, ad she was in a se-se waitiog an ne. The first doject I thc*4rt
beir achieval, bit then I thc&it it was rot. t..hen she got to the mirror I stgestI that thy doesn't
she, the teLevisict, was agairst the wall, ji.st leen it ip agairet the televisicri ad she was ca-rerred that
it uLd interfere with the recepticn or the enterras or saiethiog. I wauld have agreed earlier thxut it,
ca.j krcs, bit tho's watchir- IV? I dicWt ad I jlst said okay ad she stcxxl then ip ciceer. Say I was lyiog
here ad she pit than here to the left of my heed.
E. The TV was a different cbject. She lied alreazty placed sarethir-g alreedy over this side of the
thict, I tha,t was the TV rxitil I realised it was the TV over here. Logically, it mirt have baa, aother
TV.kdthelasttoert of thed-eanwasheraskir9rre inakirdof wayas if shedid,'tseeneforaweek
'do I get to see yrxi next week? How xut Le goiog cut to eat together to the I tal iai Village?' ...
rever did go to eat there, it's too expe-sive. It's jLst struk so surprisir9ly in the drean, askir9 em in
that kird of way. ... it certainly is rot rqxesaltative of hcw we relate. It was in a se-se foniel, like
she was alticiatirg with excite1-it thich ties in. It is rot that umtural. This suirier I have been
stayir9 at hare a lot. I werted to get work chre, bit I ha't gotten av work done so we haven't gore cut
to have a gocxi tine ad t I stayed hare ad did-Vt get the work ckrre. So in that se-se s e does welt ire to
get my slit together better so that we ca-i decide thai we will go cut ... But the way she said it was very
stra-ge.
E. Do ycu have ay idea of that is irrportalt about nrvirg the cbjects or her doirg it? Yw said it
as irlpc-ta-rt to leave than there they were.
S. Yeth. I was thirkiog after I we-it back, the mirror that scux real sigiificait particularly
sirioe I arrrmrcissistic . . ycu krs it wasn't Like I was there lookiog at myself or eve-i in a peiticn so
that I cculd look at nielf ... It was facirg this way, bit if I LOCked into the mirror it would be into
that corner of the roan
S. The two hrcgogic thiogs. I felt a little goilty thxut askir-g Roz so that I ca-ide that ad thai
rot haviog a-vythirg. I felt giilty ad then kird of a failure. The failure beirg the first real drearr.
E.
Ji.st a pDint of referexe. Ycu really fell asLeep wry fast ... ad ycu did have lots of
436
S. Yeth, I rea1iz that I shculd Lock at myself a-d rot at others. I get ry hii exctaticrs of
se1f.
S. (The giy at the stat icn] s really arroga,t. The type of rsa that nokes ne really ary.
i felt self-ccrscicw as a Jew
Ycu kr,.i (aiti-senitimid ms my projecticn of my cw ccrem. ... Ke jLst
starta cat Lirg cut rmticnatities ad I rever excti him to get mire.
437
NIGHT 2
E. lat are the thats goirg thrc&, stir mird as yc&r thx*jt to fall asle?
S. Dr. cartwriit's pr - gettir9 it thie this week fran a ccu-se in the winter crter.
ss that I'm pIeasi that I'm nore relaxat Last nrnth I s really ccrcerrl with
I
i.iiether I as goir to be thle to fall asle ad &ean. ... I thirkir9 &it a beck I wes
Jist rea:1ir. Trasacticrml Armlysis ad Gestalt, a thick xut my ccrwicticr. ... This bxik is really
excitir. ... I wes thirkir xut Herry Aarai i,o is cii the b3ttan of the Naticrml Leage in terms of hare
E. He's in the wra-g Leage. ... C4cay. Gocd niit.
1st. Awakrnir9.
S. I'm rot sire, I have a vage inmge of, nm/ce a wegai, hit pr&thly nore like fcu- i.ãieels,
na4 havir ae axle carmctirO than Like a, a train, nke a little cartoai nm rmxt to each am. I'm rot
sire if that's ii,at there hit i.ha, I thirk thxut it that's irat I get.
. Awekrnir.
S. Miit have bea, a sketchy sca-e in Treasure Isla-d or saiethir bit I really ±*,t reie±er
goir9 m there, jist a brief scam.
3rd. Aweka,ir.
S. I r.ns thirkir ... it wes as if I were ai exeriniter or clinicia, ad in the sittmtia, I'm
±scribir9 I wes, i kro.i, a clinicia, for this nm ad the mm - it's really hard to describe â,at I imt
to say - it ke goir9 ay - in the cl-em I'm thirkir xut this rim ad ha caitait his skilLs are in
nekir9 jIe1mts xut - di, pr&dly the state mcple were in Wia, they said sarmthir - there wes like a
viscnL irie in b' mird - I wes &ivir like in the sr.b.rbe b' a, cça, gare ad it wes as if that wes
ere this giy wei-keJ. Ard I as thir*ir of iâ,ether or rot this g.iy .o.jLd urlersta-d i.irat am of this
thirgs like uxrsta-dir a thc*.it - as canim fran the prcp1ate nmtal state of the rscn. In my mird
I • tryir to describe, it wes alrnst testir this gjy - is this uy CI( to ccntirr.s deir this, Ca,
discrimirete? This wes Like a sarple in a wey to see if he cauld ck the task that wes assigi.
he
Ixn interview, K.J said that he wes 'rreitally ai tive articiFmt, bit rot visr.lLy', that the
sm as 'very immtiaml', axi that 'there wes a-other rsai I wes exaninirg this gy with'.
4th. Awekrnir9.
S. J,atweswesareviewofthelastdr-ean. Iwesgoirthrc*,, kird of categorically, as if I
ies writir9 a thare, my accant of this Fersa, i,1-io wes nEkir the juGTmts -
or it mi,t have been a-other
rscn, ad decidir ha inxi sase it node. mcifically that srxh ad suh a rsc*i wes - the nwory keeFs
fLciir9 in ad cut - I wes writir a mrr1i a, the etiolcgy of his Lavior - the sçecifics of it - it
escss ne r. s this g.iy in fact retar or wes he ... jist sccial crhmticn or sai thirg, w krm,
Jist fai Lire to learn the sial (7) skills. There's norm to the dr-em, earlier, bit I ckn't kru. if I cai
reesr it.
0, interview
he
said he wes 'reviewir mmtal descriptics, of this g.iy, rot writirg - ro vistel
iffagay... utioraL'.
5th. Asekrnir9.
S. I wes sta-thr a, a sidewelk ad I wes tellir this tall slader goy, yrxr rscn, ad this
wes ireide, I wes tellir him ha.i to behave. I ckn't reimiber W,y I wes
es Like his base ad his mother
teLLirghimbc*ito,ave, hit I kr I wes tellirg him to,ave in a certain', ycu krcs, in order to
iiem my or ar ade. There wes a sase that we weren't werkir9 so nuth as a ir becase I ha to tell
him, he dith't krrc* withcut ire tellir9 him. Jist as ycu weke me .hat hai occurreJ, later cii
438
we
ha gas beck
into the ertn,t, his nother's mrthB,t, a seccrd tine e hai gcne into the beck ±or ad again I hal
expLair to him at I hal ,tal him to d. It was trickier this tine, he r the beLL or saiethir
an
he s st&dir there taLkir to his nether a iA,ite he was ±ir that I was kird of back ay a the sts
or beck prdi or saiethir axi off to the Left, Like iraic the rthit. I ccuLd see his sister or saieae
else thro q, arother wirch, kird of watthir iiiat was geir ai axi I ert of the script that wa hal
peçered - it uLd irv.vLve the nether qestimir tie so at the çccp1ate tine I care p ard atswerai
q.esticze axl the nether askalne - I g.ess it kird of Like a cte or sate kird of arra-eiits - I thir* it
as mre Like a cte - it seat Like a haineoL kird of thirg t.tiere she askal no - she expressal her
ptess&re that I was takir her sa wt or saiethir axi - Like I was takirg him alt riit ro..i, rit at this
naimt axi then she askal no xut the text tine, as if weLL, I g.ess u'LL be canirg by again very scai a-ri that kird of threw a C?) catfi.sal no. Ard rmi I was to Lcrer Like the strategist or pLaTter, I becare a
sthnissiw, kird of flisteral LittLe lx, stadirg there resjxzxlirg to her ad it was very kird of infaitiLe
in mswerirg the cstias ati jt.st kird of stuthLirg a-d tryirg to care p with sate kird of a-swer that
o.iLth't be a Lie- .i know, Like I'd be k ttescy, or we&eay or rext satu-r±ty or iiatever but .kiich
wuld ccntirie to create the inpressicn that I w-tteI t.i,ith sarahcs seei to get this other giy cLearaxe
or sarethirg for rot havirg to be arcu-si or sarethir9. I'm rot cLear exactly i.ttat I'm tryirg to get. The
seccrd casicn that we went to his nether's hase strikes no that it's thip outside, I'm stardirG a, sate
brick fLoor ad there's a serse of Li,t rain or &izzLe or jist thrpiess. Again, there was earLier pla-nirg
for the first a-ri the secaxi tines, bit I ckn't rit r rarerter i1at it
.
Awakenirg.
Cl,, I hal ji.st said this rtiaiLar rsai i.es raoid sthizcçlirenic - let's see, W-ro was it? I
th-r't krs. Let's see, in this siti.aticsi there's twa ecpLe in it. Cl,, I knos. I'm in a ram cLea-rirg (p or
satethir, I'm rot are i.ktat a-ri Carrie cares waLkirg in hoLdirg a record caer. She was cLea-iim p the
Livir rain ad hal this record cover. The picttre a, the record caer was in bLack a-ri iiiite ad was of a
ey,t1e, I thi,* it was a a the record caer, I thick i.es iiat it is or, if rot a heLy, sarothirg
that feels wry fartiLy bit I'm rot are if a nether or rot. Except then she ro Lcrger is cryirg, rcw she is
ikcL
a-cL t'vc txgec c
ctr9,
Pm no o&ser'irg a-ri she says,
krcw xut so a-ri so iJio n
in tha, the street, welL he said he's fran sareere in New
Harshire, welL he's rot a-ri i bw that he said he warkal at so ad so, welL he ckresn't ad he says that
di,
le ar-a here hho are alt to get him or sarethir - wall irayte I still an ire ase I say - this is kird
of a fani Ly jcke, a joke between no a-ri Ccrrrie, ra-oid schizcçfrenic a-ri I'LL tel L y* W,at that nears.
That's at - h,irg sha, vj caLLal no. I sal then it r becase nore of the &ean canir9
Lack to no bit it's fLa. ay again. Sate tine before iktat I ±scribaJ to ytu I'm irrvLvaJ in sate kird of
heavy p-cbLan soLvir9 thir-g ad it searei to be tokirg pLace ai the seca-d or third floor of a hc&se or
artne,t biiLdirg - it seers that at aie pint we Left the seca-d fLoor a-ri wa-it p to the third a-ri I'm in
this rain with twa other naLes a-ri I keep gettir a-i inage of a circLe. I'm rot are if that's at the
&earr , I thick I'm rot gettir-g i,at I wa-it - the ine is the feetirg of a circLe a-re again, it scux
Like ire evaLrstir (a, my] art, either evaftmtir- or assessirg, again, it seats to be a stient or else I'm
thscribir or disctssirg with these other- twa giys the correctress of my intet-p-etaticti of sate Fetient Or
sarethir at the a-ri a-cl stiLL a-other Fert of the &ean - I'm walkirg thra, a-i area - sort of waLkir9
thra, a store jist to toke a short cut or to care to the other side of it. This is a Large kird of
sdLkycarthit store a-cl I'm carryir9
my green Lxokbmg a-ri first I'm outside the area, t*i have to go
thrct4, a LittLe tnt-stiLe or salethir9 to get into the store. 0, the riit jr.st aljacait to it there's ore
stad or a m or a bethrccin or kitchen uta-siLs or pLia-es a-ri I'm Lookirg for ore rtiwLar ate, I
±i,'t krm J,at it was. I Lcal a Lcr tine for it too a-cl I ±n't ratErber. I decide to go in the store so
I give the cashier or this giy sittirg in a booth, ad he's ai oLder giy, older tyo rsa, - at-er of the
store - give him my bxkbag a-ri he's got that arcu-ri his shouLder sittirg there ha-gir9 to that a-ri I walk
ip this set of stq nJ,ith were very strage st bit I waLk p Like ei,t stairs, bit then i.i,at u'd have
to di iciLd be ti.rn 180 degrees to cmtirte goir9 np the other sts bit it was as if there was ro aviLthLe
ey of rmkir-, it was as if my top step was FeraL LeL to the text step, yet my step went p south a-ri a-ri the
text step wait p rorth a-ri there was ro Laxiirg a, nJ,ith to tnrn to reverse you- actia, C?) ad I stocd
there for a i.hi Le a-ri there were ecpLe ai the set of steFs ab,e ire a-ri there were other prcpLe ai my sts
so I figral it was possibLe, axi the g' said jist go p those stçs a-ri I'm Lockir9 at him, how the hell
ca-i I get np these sts a-ri then I stocri there a-ri stocxJ there a-ri fireL Ly I realize host to cb it. It seats
ae step açpeeral, nmte a La-dir-9 a-re step wide or I jist tu-nal the coiner, at airy rate, it becare easier
439
ad it was a rretter of tumirg axi I went p the steçs ad walked arcuxi a LittLe bit. I was Lookir
for - I ne stiLL Lookir for that sate kitchen or ethroan utersiL or pLierte. I have the ica that it
as se kird of thir that sticks ai the waLL, bit it's ret Like a so dish or toiLet pqr hoLcr that
sticks ai the waLl hecase I sa arm alunirun rack a-d said that wasn't shat I ,ted. I have the serse of a
bi.ah, bit I th,'t thirk it was a bruih either. At army rate I was Lookir for that stairs axi it gets a
Little fr.zy shat haçparmed next bit at a' rate I finish Lookir9 for that ad dicki't fird it axi then
ccic iJiy thm't I Look in the t' secticn for saiethir eLse, Like there was a FerticuLar tc' there I was
lcc1dr for. I have a serse the tc' was rorri, I have the inege of, earLier irrage of - it seers as if I Fred,
imem I was in the tc' secticn, or at me ste in this sthlocky deçerthent store, that erether nweger tpe,
oLcr rscn was assistir9 ire firdir shat I was Lockir for. The Fert that's missir, I thm't relelter
cariirg ckw, the steps, I thirk I reither, re I ±n't rareiher hcs I r Left that pLace, ne4 I never
did, ria1 that was the l4tairs of the other pert that I feLt. The pert that is missiog is 4mat the
futtim was of shat was goir9 ar I.stairs in that roan with the serse of evaLwtim or asseseisnt of ijiat I
thciit aireedy or shat I'd seen in a sitwtia, or saiethirg. Ma*e I'LL get that Later, that's aLL for r.
C), interview he said that in that roaii 'we were sittirg m the fLoor in kird of am oLd place axi
spimirg sarethirg - a gyroscc, re a ream.. re, th , spimir, Let's see there i a ruarra netat
shaft, a verticaL pin or sarethiog axi I was spimiog sarethirg am it - ãmat hç.emed, I thir* I waLked into
this roan &d there were these other twa g.s sittiog am the floor I thir with this LittLe pin or saisthiog
with a verticaL axLe
or
saiethirg. They were spimir9 saTethir arcuxi am this Little pirlmeed
it sears we were nakiog p a story to
go
aLcr9 with this LittLe t'
or
am
the fLoor.
.imatever it was, it seers as if
we
here givirg it hui'ar, characteristics, cscribir9 a persam that this LittLe thir9 was. There was a differate
we were kirti of Like seeir experinitaLLy krat this LittLe thir9 ccuLd ±. So he said it was
aethirgadl saidro it was, di gosh, Ithm't *i,at I said it was. At army rate the first tine we ram
this thirg I hed it hooked p so there was a LittLe sarethiog that was am tcp, de the spimirg pert that
ercked ad then sare nore FecpLe care in a-d I was thrtrstratirg rib' pint or saiethirg. Ncs I pit i,atewr
it that was mkirg ad it seers as if it was a yw krr cigarette bitt or roach or sarethirg am the
of cpinicn aJ
cyLircr bit heLa the spimiog thir artS she, it was ckxe this way then that ccrwirtai ne that it was rruth
osre effective in thrustratiog my pDint. These other prcle sho care in, I thm't reccgrize amy
they hal hem in this other roan eraged in sale kird
of
activity
ad
of
than bit
they had been there iJien I had arri
ad ro they care in to see iJiat I had the rather thai warkiog am it. ... artS iJiat were the
characteristics I was attrib.itiog to this thiog? I thm't krmi. I thirk this pert jist precai the pert
krere Ccrnie care in with the record cover artS ri present ne [sic] with stiLL et aether petient this gy sho Li thai the street, I thirk that was the seme. There seers to he a distortiam Wien
Ccrnie care in - in her voice, a-d rm4e Like aLL of her becaie Like a nervcas...' '...that spirdle thirg
sears to he am a nu± himer LeveL thai sr.eL frxcticnirg ... ltairs with spirdle thirg was him1y
arotiaeL the dertnømt store was miLdLy aroticrL artS the sitwtiam with the record cover was mildly
aloticreL. PLeasant, wry pLeasant.'
4th. C?) Awakeniog, 6:50, thxut 10 mm. into REM
No recaLL or ine.
In the ntrnir interview he was asked iJmether the cl-ears tied in with the pre-sle interview. He
rlied:
S. Yeth, in teimis of the ers that I have been caipLetir have teen petient rxx-ts artS the
rer that I an dir for Roz is a thermy rqxrrt aixut a petiemt the thermy r Less so than the ers
that I have been warkiog am iJmith have te pieciog together, yii kro.i, Like history, syrptars artS so forth.
The first cl-ear atS the secaxi cl-earn seen to reLate a Lot to the h.ptgogic attarpts. The first cl-earn iJmen I
wke p seared o he exactLy iJiat I was dDir with the hngogic stuff.
E. The first me aixut the wagam?
S. No, ro, rot the ccsmtemt of the hrogogic, jist ire tryiog to rererber hptgc9ic artS - cam
u tell ire iJmat the first cl-earn was?
E. Yes, the first cl-earn was 'I was thir*iog ... that he was assig-ed.'
S. Yech, that's rimt. It rmkes sese to ire. Me tryir to figre wt - welt, here I ckzi't
rarerter heir waken p cI.riog the hypegcgic artS is it becase we thm't have the ste rimt. Ycu krt*, an
I in the ri1t ste to he in the hyxegogic. Yedi, I was am exerinmter or cLinician in the sase of
aqerirrentir to see if I ccaLd thveLc4 the hypogogic ad ire rot heir9 thle to tell how carçetent his
440
skiLls are ad h&s nakir9 a jua1it Axut ikat state cple are in a-d na kirg ip ad rot krr*ir
thether - the discrimirmticn ias extraiely difficult in the hrgogic to differentiate between alt the
inEges that were goir thrcai my mird i,ith I presuiuJ
Mi I wes stilt aweke bit i4,id, precerierl in
fat the inage that wes there. Ycu krcw, wes that iJiat I
cir iien j cat tern no or wes that at I
thirkir9 before. 1 ckn't kra. Is that clear ruJ?
seccrd drean tu start cut s'iro it wes a revia.i of the Last drean. Yw
E. Yes. With
also said cu were writir9 a there akxtit tu' accctrnt of the jixanB1t of this perscn.
S. I said I wes writir a there?
E. Yes.
S. th,itwesacttottyinmymirdthatIweswritirit....Itseai6LikeiienIfirstdeLp
that it sel to no to be a ccrltirLaticn of my evaLwticn of my evatwticn of my evakticn. After I gave
that first drean ad then Mi I wes goir beck to sLe I wes thirkir aLxut ijiat that neent ad i.iiether it
as
in fact jist in sbet ic form of
my exFerierce
with the h,pxgogic... It
tiait itiith it very nuh scux like, bit it seaes like it
the sale
E. Does the y xar to be fanitiar to ciJ?
S. No, it wes Like a-i olcr persai. The aily rscn that uLd
There is ro articutar atiait, it's ji.st that all
my
nay be
werkir thrcuji a
again.
be
faniliar uLd
be my
Ietient.
tients are otcr ocpLe.
uthiuhtotakethis (infaitile] rote [withthenother]?
E.
S. Sanetines I will bthae s&bnissively rather thai nake a canfrcntaticn ad I'll be very
abnissive ... Do yu km Traactia-el kelyis? I'll kird of drcp cut of the ailt ad nave into the
thiLd's role ad pit the other rsai into the rait ad relate that i so that eerythir is rceiver1 as
rot in fact - I ckn't
a chiad fran nother. ... It wes a chvise to get him cut of the hc*.se ... ad I
thirk I
goirg to be chir sanothir with him or ±irg a-iythirg ad et she wes very enth.siastic like
'Ch, hcw rthrful i wilt be tir him cut ai tiosthy ad i±cschy ad I'll be seeim vi satu'chy',
ad no beir evasive becase that wesn't the case ... earlier in the drean ... I kri i.kiy I i.ms cbirg iiat I
as
chir9 ad
we
were in it together. He n't
my
victim
or
aiythir. Thraiwt the drean this wes the
strategy to se ad I wes thuirait in the relaticrahip.
S. (In the third drean) I wes Lçstairs in this roan ad there wes cre roan with a lot of iys
or cbir sane kird of task ad then I it into a-other
cbirg sanethir. Like pLayir9 sane kird of gane
tJiere there wes cre giy
or
tw other g.iys ad I thirk there were twe kir of tasks that
we
were werkir ai
ad ae of than wes this little natal spindLe kird of gyroeccçe,... bit rot a great big raid thir like a
grcsc. It haJ the feelir of a-other etimt evalwticri iknere this wes a Fatient ad we alt got
different ichas as to ijiat is goir9 an with him ad I an thinxstratir this ao prcblan. I thn't thirk I hxl
saTethir9 canir in, I Leerr by beir there. I wes eçerinaitir with it ad I sayir 'this is iJiat he
is aLL xiJt' ad thai saioae cane in ad they weren't sire that it 1 it ad then it wes in a serae like
a rceri that wes ai this pin that n. aidriog ... i4en sane of the other cpLe cane in fran the other roan
ad I ns chlu'stratir9 to than ad I thc*.it that it wit I really werk if I pit the roach Lrrnaath so that
the stoke canir9 ip will case - sort of like a prqtter - it wilt sake the prcçelter werk. It werkerl.
E. So pretty nixii alt of
*.r
ci-eals are ±aLirg with Fetient evallEtiCti.
S. It seats that wey.
44:1.
Niit 3 (24/9/73)
E. I iLd Like *i to talk a Little thxiut the thirs that are ai
mird roi.
S. The stuff that is goirg to he canirg ip rext qarter. I'm goirg to he b.sier this qarter
sore so thai aiy other q.arter. I said that Last sçrirg, bit this qarter is ei sore so. I'm jist kird of
reiieair the thirgs that are canirg p. Tatorra. I have ai interview with a ther' strvisor for my
cticun this falL ad that's still .p in the air. I'm thirkirg thxut in,' cc&rses ad the varic*.s thirgs
ir,vl with my Masters. I stilL have aother iooaiplete that I did-i't d this sumer. Ccrnie ad I are in
a rent siitch. Tcthy she staJ haie ad I
I have reaLly hem thirkir9
xut
it to school. Ustally I'm haie shm she goes wt to ork. ,at
a Lot is that my disaEsim secticne take a Lot of resço-eibility ad that
I ms in charge - rot heirg thlegatrd the respsibility, bit I ass Be-itcn jest exlxctrd ne to th it. Very
armyir, especially sire three sectici-s .ere rot caierei tcxy. I jist take a Lot of prith in iEiat I d in
aysecticfs ad I thir* u have a res4xribiLity to the stix,ts ad I resmt the fact that that kirri of
screwed ne ad that's 4,at I thir Mi I ms callirg - I ms tryirg to get tcgether pecple ad to fird
cut iJ,ith T.A.s were goirg to take iiid, hctrs ad I thirk this shoold have boa, takm care of Lcr9 ago. I
as thirirg xut in,' clothes .i,idi are all stairul with the rain. Tmi,t I as wearirg my Leather jacket,
a tai aEnter, than a shite ni.stin sjxrt shirt ad thai a blie p4lcer - the layeral Look. I tock off in,'
ccat ad my sweater as all grem a-d b-c.n ad I assui it as fran my leather coat. Thai I took that off
ad my jute shirt as grea ad Ixcwu. So than I thc*.,t it as a grea, jersey that I had m that hasn't
Lea, ashed yet ad I thcs.,t that it as dyirg it iratad of my Leather jacket. My çalts will stilL
p'±thLy he soakad ad I have class aLl thy taiorra.
E. If they are sti IL wet taiorrcM ne,te ya ca ie the &ier to finish ckyirg
than
in the
inxiuirg.
S. Ckay, that is a
grxd
i. Than I as thirkirg xut jiat I
that I as here (xut the Ferfonlace
E.
ckay,
of
as
thirkirg xut the last tine
eheL I pLayers ad their prrfonate statistics].
good ni,t.
1st. Hyogogic Awekmirg, 12.02 an.
A cciple
of yet planes - I th't krtw - ad saiethirg to th with interraticral relatiaa.
ccracic*.s of thirkirg thwt soiethirg interratia-aL, saiethirg to th with Vietran or
S.E. Asia, ad that as jit hef ore ad I wes a.are of that, of havirg that kird of inagery. ibm the rext
thirg as kird of yet ptaes. I th-'t have a pictue of than flyir9 aver the sky or athirg, jLst kird of
S.
st Lefore that I
as
staticrary, kird of - I th,'t kru, like a picttre m a e or sarethirg, ytu km, they kird
saie asprct of interraticral reLatias or saiethirg. That's all.
of
resa1t
1st. N Aca,irg, 1.45 an, 5 mm. into RBl.
S. Let's see. I'm rot are, r4e heirg cutsith of tcw to get a cir*
cutsi of tcw.
or
scnethirg. Beirg
d. REII Awekmirg, 3.45 an, 10 mm. into REM.
S. I thirk I as plLirg saTe mtruts cut of a - tcçpir-9. There wes this big piece
of
stry
heckadforthadlwespillirgrutsa-dcaruielwtofit.
Ia-cw, Like ai a big sheet ad as I asit aLcr the piece, yw krs, fran left ad there ild he a
scriptia, of - ritten - of hat as heirg a± at each Feint. Here as mi interestir9 bit with saie
batruts, cas ad cars-el ad there as also a-other ae. This ae had narsl-imllcn.s ad cherries ad
Soiethir, i,ater ad it as this thscriptim that as goir9 ai this lcr type thirg ad yw kru4 I'd boon
YQJ
442
irg that for qJite saie tine ad ... ai this thirg ... it as of a, this Festry, or i.iiatewr it s, s
for a fonier occasicn that I had caie ai ad the rest of the c-eeai I g-ess s thiit this fonier cccasiai.
Let'ssee.IS...thiSPLaCeiSareSOrt...1WasSittir...IthirkasUllerresOrtPlaCethatIs1Xt
at aLL faniliar ithysicaLLy. Let's see iiat's goir a, here. Let ne try to retrae ... I ca,'t ... Let ne
sort of rk... Ji.t prior to this thir9 with this cameL estry sheet I s sittir at a tthle, kird of a
piaic Ixrch deaL with my rents, irLuiirg
my
father ad Let's see, I had jist nw saie kird of
di my gcd,
treie-dxs iriit in regarde to school. I s sittirg ai that horü, ad s thirkir to melf -
I hcçe I ckn't rqress this ... I thirk it had saiethir to dD with hoim, for saie reesai, ±cidir@ I s
thsoluteLy in the wra place ..., or nth I isn't in school ad ... the ir6iit s that my gcx:I, neyte I
shc*ild lk into a school Like Harvard or YaLe ad jiEt lk in ad talk to than in or±r to get in, hot
it's nore caTptex thai that. Asscciatad with this is Like aLso at the tthLe I thirk - it seals to ho
Collesi. I'LL teLL u Later
ColLee, is ad it s as if she s dxut to de this ad I thct,t if she
caild ck it then nm'ke I, then nmybe it
the nest, it
the slmrtest iy for ne to ck it. This isn't
extLy ri,t ixit I'm gettirg there. The gereral, aLri,t the ga-eraL gist
ji.st saie kird of ircr&JibLe
irsiit I got, that seais to ho a distorticn, aLthc*., I cai't reaLty grasp ihat it s. I ies sittir there
L
silentLy, after I'd caie L.çxn this ... it r Like siLete at a tthle, ad it iøs kird
felt aiu-ge toscrean, jist toscreanuit, hit I dich'tth it. th, Let ne
go
of
grcsir9 ad I, I
hack in the &ean a bit rm.
kw. This i again, j kn, a sumer resort t place ad I had pL lad ip to this place, I thirk with
nother ad gotten cut
of
the car, aLL rit, re, Let ne thir.
my
pL Lad ip ad i.e i.ere Lcekir for a
thir, ad of the ±iwr.ey ad then i.e
We
Farkir se ad it i.s a mmLL, circular
fJ
kird of
hai to reverse direction in a miaLL circLe there at the ad, ad there r-sn't a gocd pLace to park di&i't park within the Limes, i.e jist pLLaJ the car to a stcp ad bafore i, I g.ess, i,es a big cthin, big
lcxe, I gess, or that kird of thir9, I g.ess a Lce ad th.n beLcs i.here i.e cculd,'t see it i.as, i.as Like
the Lead, area. We got cut of the car ad i.ere kird of su-wying the area. Ch, far cut, I've got it mw. Ard
r' nether aL L riit, i.e got cut of the car ad there i.es jiEt this, u krm, alL these pacpLe arcud,
ad there i.ere Lots ad Lots of
my
nether's friath ad
my nether
askai tie.., there's Mr. ad Mrs. Talbart,
wuld i Like to taLk to... i.uLd ii Like tie to intrcdxe tu to Mr.
ad Mrs. TaLbart? Aid I said, tt
krcw, I said 'me' ad these, Lath of the pacpte neitia- are pacpLe I kro.i, are friax of my nether's
ad that - ad, u krc*, I had re desire i.hatsoever to taLk to than ad then I talkad to the GLars, my
iether askeJ if I rit&J to taLk to the GLrs: me. Aid the, Mr. Taltert alL of a su±ien is stading there
cçprsite tie so I shake, I kird of hesita,tLy shake hax with him. It iesn't at aLL canfortthLe, it did,'t
feeL 9xdat aLL. &it the gist
of
the thing r - i.i,y had I decithi to
go
into-- to go to a - a ti.a rs
testers p-cgran at Weshir9tai lh-iiv. rather thai.. - ad then planing ai stcçpir9 after that, thc&i I wuld
thle Latter a,, shctiLd I decide to, to retun for a Ph.D. rather thai going straiit into a-other
ptgran, kird of Like at Harvard or saiething. Mate at Harvard, I thi't kr i.here else - ad going
hair9, it i., I i.s having to explain it to e'er4xdy there,
stmaiit thra., to get the Ph.D.. Aid I •
that i. kimd of the si ttim ad it very utanfortthle ad I did,' t ialt to have to de that
he
rtiaLarLy Hr.GLazier, iJ,o - i.iicse sal isad to ho a fried of mime ad i.it to t.,iversity of Rcd,ester.
Rcd,ester, No? Wiere the hell did La C?)
go
to sdxaL? I th,'t kru.i. Rcxthester is tiad in there, his wife
gces there rn, I ckn't krui if he iseJ to
go there, I ckn't krm. At any rate, she, I firelly ra, into Mrs.
Glazier ad she asks ne with jist sid, sirprise ad, realLy, cçxsir my decision, yw krt, ii,y did yw
to de this p-ogran ad I expLair to her hcu that I i.a-lt to stcp after tw years -- ad go cut ad
try it or sarething ad ha'e the cptia, of going back ad she kird of, her paint of via is that if, if I
Stcp the, I'm mever going to i.,t to go back ad stading there kird of feeling, u kim, I n ret actiLly
CCide
s.re of m.seLf ad my decision ad rir if I realLy will go back to school. I thirk that's i.hy I didi't
i.ait to have to explain it to pecple. Also it's - part of this ihole scae here - me, ne€e a little bit
earLier, neybe i.M, i.e had jist parkad ad gotten ait of the car, I i.as kird of sizing ip the sittatia,
here, jist seeing if it i.es prcpriate salehcM ad it seais xtpriate in a sase ase, Lockir9 for a
-- I i.as Lcx*ing for a attractive i.nu, or saiething -- for i.hat? It i.ms alL older i.ani, so there
erm't av tu prcpLe ad it i.s as if it i.es a party
or
sarething ad I i.as kird of checking cut the
sce-e there with my nether. I raineber, u kro.i, met firding axdy there, yw kri., that I fctrd
attractive
or saiethirg particularLy to start cut there ae i.niai i.ho, yw kre, for sare reason, her face
stcad wt ad she i.sn't at aLL attractive ad had kird of short gr'ish hair ad it i-as ji.st kird of plain
ad u-attractive La*ir9 ad I - her face is very clear to ne. It's ret a faniliar face. Then it seess i.e
i-are kird of i.alking arou-d, irside of this Lcxe, or i-hatever, ad yw krew, I i-as tryir to decide if it
443
as prcp'iate or rot ord - th, Let's see if I ca, see - there's a reaL serse of bizarreress here - a
distorticn of - kind of, I ckn't krw shere this is - laid wt vis&eIly, m a chart or diagran or saiethirg,
es like five c's of - of Jiat - there were sare - there were sq:plies, i,ith iro1u fcxxl I g.ess, krith
had Iea, stolen, or taken a.y fran the pecple i,are it wes ad it wes kind of, tu krir, visrIly, it es
kird of, say we're larkir at
en 8 1/2 tines 11 and we're Lookir at it so the Ia-g ey is vertical and
there'd be a six eragr±s, or six series of se-ita-ces, yw krts, neke ru±ered, in i4,ith there es ro
etthoraticn other then jist en iten -- a 1istir, itanizaticn of the, yw kni, these itais - the gox sale of then were focd, others were - I'm rot su-e exactLy, they had the se-se of Leir9 sçplies, yw
it Like a, en oceen-goir ship, it wes, yw kro, en old ae, I thirk in the 18(X)s it wes crossir9 the oceen
-- these se-se of s&çplies were irclx ad it wes like a real bed thir like these, it had been taken. I'm
rot su-e .á,o they were taken fran or by i.tian they were taken or exactly hcs they were laid, there's a se-se
of there - of it - of there havir9 been a biffet or saiethirg -- I'm rot sixe b.it I wes jlst really shockad
ed very sad to see alL these thir9s had been gae or taken, shatever. I thirk it wes here, at this pint,
that I es jist she, I first begen scribir9 the &ean of seeir like a cockie sheet with this essir C?),
essirg beck ad forth, kird of cararel ad bkratever it wes stry -- it ms very flat, it iesn't thick like
a ck*.irut, it os pretty, it s pretty flat ad it wesn't all caniel, eva, thar I'm cscribir- it as if
it iere aid camel, the scripticn that I read alcz-9 of it s a little camel here, the i.hole thir loc&nd
Like camel, b.it there were cnly sjxcial ores that had said camel, let ne go into a cscriptia, of this
thir9 nore eLrateLy ... as I said, it -
flat, besically camel cc'ernd a-ri the a-ily reascn I krew there
here ruts in that thir r.es beca.Ee there
a littLe rk coloraticn ad the cscripticn said there iøs a
rut utrneath ad as I read alcrg, I read alcr qiite a lcrg tine, in fact I thirk I as readir it
kwer, I thirk I startad at the bettan of the sheet ad r.es goir9 neyte frau riut to left irstead of
fran Left to ri,t, neybe I did-Vt krc* shich wey I wes dir-g it, bit rot u,til I had got to the tcp S cxve
or atewr did I, I taste it and reach, yw kru., readuir in with my left ha-ri, my firer a-ri my thuib a-ri
it ms very difficuLt to pill it wt beca.ee it wes really, kind of, very very gooey, cri.z-thy camel or
inatever. I ckn't... I ckn't rereiter actr.LLy gettir into it -- e.'er actually gettir9 it into my ecuth.
For sare reasci-r, there's a TV g.Jic gist to this or a flax.r to this, as if the scripticr, that I as
readir9 wes a TV g.Jic ±scriptia, of this piece of fond Like, like the lv .iic has a cscriptia,
reçer, of a TV shc, it s kind of a TV gJic-ress se-se to it.
(.xn fu-ther q.astiir
he
of
a
said:
S. Yes, I ranerter welkir9 arcuxi, it seers like I i.aLknd ara.rd irsic the lc'e ad this, I
thirk this es the b.iffet se-se, it seers like there wes fcxxi set Ip ai the varicts tthles - yw krn., like
pia,ic tthLes in size - I thirk this r.ns shat wes there. There had bee, a social gatherir thir c&itsic bit
I thirk I es also irsic checkir9 cut this place. It seers I as theckir9 it cut for ire, rot for my nether
I es theckir it cut for ne. Yeh, the visuml thirg, it sears at the a-ri of the drearu, at the a-ri of the
d-ean it es hard to cscribe,
we were kind of sittir9 a, a pialic be-dn - we were, my nether es there a-nd
I thirk my father as there, I'm rot su-e sho else wes there a-nd we were...
I thirk I've told yw all the distorticrs. Ych, the distorticn thxut that I wes at shir9tar
Ihivecsity rot here a-ri that my nether wes goir to intrcd.ce ire to these prcple that I aLready kro.i.
3rd. kkenir
S. were piLir9 cut of the -- we had jist arriJ, ire a-nd several frien, we had jLst arri1
by fLyir thrcti the air, at a bEr*, nnyte in Detroit, or saiekrere cz the east coast to thicago in this
kind of fentasy d-ean, we were just sucked thrc*, the air ad I as beir carried by saieae else - I
kind of the lead rsa in cu- - I g.ess - ycu cen't call it a sq..adra,, we were all kind of flyirg in a-e
gra.p bit I r.es kind of like burg in a se-se, yaj kr-ow - frct,t shirt - a-nd it wes terrifically excitirg ad
frintenirg too. As we were swxpirg ckw,, to lord in frcnt of this be* presith,t's C?), in frcnt of the
Eak,ycukrow, Likearollercoasterkirdof thirg itwesforneedwejiste,t irsickrthe -- wesa4the
be-k president a-ri we -ut irsic a-nd told him we ,ted to hirrc*. $10 millie, ad we're sta-dirg there for
444
a little i.kiile waitir, I gss, for him to get together ad we start&1 welkiog ±*n the st, as this Ik
presi±nt, as this bak presic1t ad than far other La-k officials ad as we're welkiog chn the stqs I'm
kird of axerir9 i4y it is that we're berrcir9 $10 millicn, ad ti,ile we're stadiog there I gss I
laad - began to Lai a little bit, I jist that this iioLe thiog wes so aheurd. As we're welkir ckwi
the steFs, I ckn't krw iJere to the baseient or i.kiatever fran the Lwk, I startad thirkir r iat are we
- well the first thiog that esse thra
goiog to d with $10 mill.icn ad that I thc&t of wes that I
my mird wes nerketiog i&ntical icake better to, I th,'t krn, krit Jenina or saieae, ad jist bririrg
ithitical stuff as that ad jist ckagirg it in a differant - in a differant bix ad ckir9 a trmterã*.
praioticrol caiigi ad ±ckd we'd have to cb saiethirg to nEke it look or taste a little bit differant I
tha,t that we t..ild pit saie kird of focd colorir into it to nake it a little differant in color. Nc*
that haççerel before this thole thiog - this trip - I thirk we're leaviog fran the East Coast to haie ad
we're in a car ad in the car are Berrett, Ccrnie a-ti ne ad I g.ess twe other friert, I'm rot sure rit
rn tho they are bit we're in a reaL gay inxd, I g.ess real h' to be goir a-ti we took off - kirti of via
ter, my hcne, ad went thra. a lot of faniliar areas a-ti we go ip
ore of - I ss we startad fran
Like - place to place travel
thra, these dirt roa ad thirgs a-ti it was very u,realistic, it
very brief ad we did-Vt go thra, a thole lot of thiogs, there were a lot of siits that in a drean strwk
ne as beir wry faiiiLiar, of my chilctocd, like places that I knew very welL bit I thirk of then, it seais
like they are faniliar. We uld açproach ore sort of dirt road off to the sick ad Ccn,ie uld ask ne if
the road thrcL. Paxtcri, a dirt road i.p thrai here wasn't a scanic rwte that ild be, xi kr, a1cr
ax cstirmticn, a'd I *ild say to, it's jist kirti of a dirt r that goes ip into the ad that
n.ild be kird of strage, yw krnsi, that Ca-nie nild ask ne xut the kird of miall, dirt road, iiich
did,'t seen at all, yw krvzw, ai route you kro., we werm't goir9 by hi's or aythirg we were kird of
cbirg the beck r arard Weoater. Let's see, there were we ai ax way to? I g.ess, Derier, was saTqlace
were an our way to (?) ad, th, let's see rt.i, earlier in the drean it seals like it was ji.st ne ad
we didn't take that tu'n-off to Paxtai then before LS was a kirti of a
pxd that was jLEt like it pearal, in the a-can aiway, it was ithitical as it was, as I thc*.nt it had
bean in my childnocd, excqt for the fact that there were nwiy dead logs disrsal thra, this xrd in
we
Ccrnie. Let's see, after that
kird of at an
n.knith I could (see] strait thra, to the bettan of this p:iti. In fact, the prrd
it wasn't flat, it was at ai a-gle ad the logs were, you krtsn, alcr the hettan of this eall prti - lir
ip sort of, we were kird of at the tcp of the p:rd ad it was goirg cks.rtnill a.ay fran is - the logs were
kird of horizantal, were horizontal to ax persective, ax ale, a-ti they were servirg a thnnirg kird of a
fu-cticn, they were kird of thimir-g the Ford - they were all very stri1 of theIr berk, they were very
gray. Than we were, I thirk after that we n.,t ±n a hill ad I ss there were like five of i.s in a car
at this tine - Be-nett a-ti Ca-nie a-d I ad twa others a-ti we &oe again thra, saie, again, saie kird of
minI I tan fani Ly kird of a place a, ax journey to the west a-ti than we tu-rI riit beck arcuti ad want
beck to there we cain fran bit the, saiehc*n cantirned alcr ax way. We were - at ore Faint I kro* we were
goirg &bn a very steq hill a-d I was, let's see, I esserger ad there were all these lakes, these
nice Lakes ala-g the way bit my thole psitim was very Faculiar in the grap, let tie fire out that my
psitiai was. Kird of Like I was the eqse of jckirg ad kiftzitzir, I was the cbjact of saie kicthrg.
tnat was it xut exactly? I had, saTethirg that I had ckre was the reascn that I, that they were kird of
-- I g.ess I iuld prefer kirti of rot beirg the
kithitziog tie ad nnkir tie feel -- I wasn't agry bit I
cbject, or - of the jokir ad thirgs. I thn't reieter exactly that we were jckirg rixut. Let's see, di
i, at ore Foint we were goirg dew, a lcrg stem hiLl ad this pire ad it was wocted all aran-ti is ad
greet big lake a-ti as we aj:prcached this Lake me of the people in the car says that they were really sal
becase they hadn't been to Otis airforce bese all this suiner lcrg - v.i krow, as if this was saiethir
they did every sunier, bookir-g forward to every sumer ad it as if hive liit was that she refer -Ithirkitwasashe--wasreferriogtoasOtisairforcebese,thiChiS,UknY.i,anairfOrCebeseinCE
Cod ad I thcn,t that was very, I kird of laiel to myself a-ti thirkir9 in the drean that she nnkirg nore
of it thai it was warth or it was to big deal that she didn't get to go, becase see we were goir rm ad - or the fact that she lisi in N.Y. a-ti this was in Crpe Cod ad was really far ay... I'm rot sire if I
was Laiir9 becase of -- became she was so accistaiui to goirg to Ce Cod e'ery sumer or saiethir, or
goir-g there often -- rot exactly sire, I ckn't rela±er a lot of details that hçe-nad alct-g this jcur-rey
bit I ess earlier, the Fart that was really kird of seekiog out the route -- I reieiber discissir, kird
of, with Ca-nie thich way we ild go. There was saie, yw krc, kird of nnjor kird of canflict thich was has saiethiog to a, with ax tam route. I ckn't krts the exact retire of the confLict - of thidi way we
445
uLd go, i,ether we cc*.ild go - .krether we could travel over water, or shether we hal to go by Lati or it
sears Like there's sare kird of faitasy cpaLity to it. kd I gess we sort of ccic by kini of testir it
wt. 1 thirk that's kry we weit as directicn, the, the other, then weit huck then ccrrtiri.1 en ctr way, it
was kird of a testirg iIiith route t.aiLd be best. But then crce we got utlerway then I ±i,'t krrosi, we jLst
bzz aLcr in the car. But then at the erd of this thir here .here I began thscribir the drean, tu
krrM, therewewerefLyirgintheaira-ditwasifwecaiei natcw,thatatfirstgLa-roeLook Like
Chicago, rticuLarLy seeirg the Hatock ta.er hut then as I Look again, kird of oçei ray es, it wasn't a
very cLear visicn, it was, foggy - I couLck,'t see too cLearly - I realiz1 that -- I was thirkirG gee, we
haer't gas far eai to be at Chicago t, this rust be Detroit or sarethirg ad I LOOkerJ again ad sa.i
it reaLly wasn't the Ha-roock BuiLdir, it was auth rruLLer - lirt coLor b.iiLdir that reserbL&1 the
architecture of the Haxock ad then it n sase that we were that far ad that's ii we jist real Ly kird
of swacped riit th.n into the street to berrc*4 -- to go to the k to berrow the $10 miL Lien.
His a-swers to qJesticrs irdicated that there were: 'sale other pLaces en route very earLy,
isIa-o, xn, wacy tys places arcud .kxster ... the Logs in the pad, they were very gray..., the FxltJ
was carpLeteLy tra-sçsrent - we fLew, r.iiat Looked Like Chicago wasn't Chicago, I g.sss we drove acrs the
water, ..., kird of cbirg thirs with the car that you ca-r't reaLLy ck. ...
we were ccidirg if we
couLd sake it, ytu krrow, Ccrnie sayir, ro we can't go that way - u krroiw, a physicaL irrpjssibiLity ad yet we saicw did. ... I was a rticipa-rt, hut in a ssiw role, Like shen I was beir carried - I
±n't rererter the other twa peopLe in the car - I thirk as • a girL, it could have cha-ged. There were a
Lot of peopLe stadir en the be* ikren we arrived. Care was a p3ticarm, or a * .erd sta-dir-g there,
rear the -k president. ... PLeasant, exct for the wry ad - as we swacçel ±&.n that was very
irpLeasant, very scary.
4th
kenir, 20 mm. into REM, 6.45 an.
S. I was, I thirk, at shirten Liriwrsity or sarethirg. I thirk I was at shirgten L,iversity
again ad I was sta-dirg - Let's see ... EarLier in the -- let's see... riit then shen you caLLed ne I as
sta-dir irsic of a fcr or sarethir of the strxient u-ricn huiLdir9 of shirten U,iwrsity ad I sa.
Ca-nie's old rocinmte sta-diog there ad I was talkim with a falTer
ter fried of mire ad I caLLed out
rl.ri,iog
becase she was reaL far away
to Na-roy ad she we-it out the cbor so I took off after her ad I
ad as I was r-u-nir off after her to say heLLo ad stuff I was pessir by aLL these u- people en the
waLkway ad that was very nW, like the feeLir-g of juiior hir, sare people were rr*irg kird of
strroptitiasLy axf - or Like college, yeb, Like coL Lege, ad ii,i Le I was rtrrrir -- see here they were
goir9 to get ar the limes for rericeJ fare lis ticket that hed been in juiior hir C???) ad as I was
rin,ir- after Na-roy thirkirg how 1, I g.sss hcs I ised to ccic iErether it was warth.kri Le to take the us
hare - saw the au-my ad take the hs hare or take a regular b.s ad get hare otherwise (?) -- saiethirg
Like this ad, i krro Like, I was sure Na-roy hai ga-s ±wi this nEin roed, this rein waLkway excopt I
ccul&r't fird her ad then I sa, op theed I sa a taLL girL, Na-roy is tall, ad it Looked Like Na-roy fran
the back - with short hair, her hair was out shorter i,ei I sa her in the Lou-e there for a nuirt ad she
was waLkirg hoLdir harts with a gy ad as I n.n op ad get cLe to her ad sayir9 hi, DelL C?) i.i,ich is
her Last rie, il-rat we caLL her she trxrm arctrd ad it's a am ad he starts aswerir ne ad it's as if I
irm1iateLy krm it isn't that DeLL, it's her brother. I ckn't krIoM her brother, she ±esn't haw a brother,
this iy di&r't Look aythirg Like her ad I was reaL su'prise:l ad asked her ihere Na-roy was ad he said
she pr&thLy we-it over to Maxblin HaLL ad I was real disxDinted became 1 figa-ed rro I wzn't be obLe to
fird her became I di&r't krrow ihere reaLLy to Lock ad I krsw rirere Ma-r±)lin Hall was even thci that's
rot reaLLy a liii Ldir- at Washir9ten l,iwrsity. -- But that's kird of how the di-ean errJ rit there. But
earLier in the drean before I sa Na-roy, oh, ray gcrod-sss, alL ri,t, we'lL get a-other segrelt ±ne. I was
taLkir-g to an oLd Larter fried of mire raied Bob ad I hah't seen him in severaL years ad oh, Crass (?)
was in this drean too. I was teLtir- Bob I gsss xut my experie-roes in grairute schooL or saisthir in the
Last caple of years... teLLir him xut ray experia-ces in the drean Lob. That was prctthLy ikrat we were
talkir alxut. He was back in schooL again -- rø..i it's takirg place in
ter ad he was in l4xster again
en thristnss vacaticrr ad it was kird of strae, ytu kru.i, the fact that we were in Ilxmter again ad the
446
fact that he's been cut of schooL for iraly, nary years. He drccçol cut Lcr ago ad rsi he was hare ai
thristims break. He hod a Letter with him that he haJ salt to ire ad kird of it hod jLst arri to ire, or
he hod nEiLed, ad yet, yeu krn, he hod it in his had ad he gave no the Letter ad I reed the Letter ad
tu krtw, was jiot surprised that he hod written to ire ad I ckn't kr k'iat the letter said or aiythir.
But it was r4iile I was taLkir with him -- it seaio like it was in a rozin i.ã,ere we were kird of Iyir9 ck.n
very caswLLy ad nme there was a firLace goir9 or sarethir klai I sa. Narcy. I thirk Bcb chaiged to
aether friaxi of mire, Craz C?) ad again it sears Like I was in aother segTait of the ckean that I'LL get
to in a mirute. Craz cares aLcz-g ad he gives ire a very erfr.z-ctory hi thc*i I haai't sear him in a year
ad innliateLy, Iike,starts trasactir bsiness with ire ad takes cut this LittLe foLr or saiethir ad
cn a Little crurbLed ip piece of
er he's ha-dire no a receipt a receipt, I gess, for $10 or saiethir
ad ar it he's got scra4ed in reaL
LL haritirg that's wry difficult to read, scratched cut, it's
was ir8 cut to saioaie else ad thai at the bettan of this check, i krn, he hod written hcs he was the
authorized agent of - or r.irate'er else ad he ne tumir over these ftr to ire, Ken Jacd:s, irsteod of the
perscn rared a, the check. It was reaL ccrrfr.ir, ycu ki, ai the littLe rote it said to see xre, or
âratewr ad there was rothir9 thwe - there was ro expla-atory rote
e Like he hod irdicated. So I'm
lookir9 at this thir9 gettir real caifrsed, sort of. Thai,
ess the trst th'ir I said to 'him was kru., kird of facetiasly, ycu krr, hi. Ycu krtw, I was really, ycu krru, stressir-, kird of real
a-nace, ycu kj-n, I hach,'t seer him in a year ad it's realLy good to see ycu ad here ya jist care in
ad jist give ire a wry brief hi ad thai start taLkir siross netter we have. Cli, the b.siness flatter
for ckpe. This $10 was irat I hod given than or sarethirg to Wy sare ckçie ad he says to ire, ycu kr*i, yeh,
I'm reaL gLad to see ycu too, exct he said it, ycu kru, in sh a cold way that it wasn't at aLL
cawirtir ad I c*raLLer9ed him a, it ad he assured ire that it really was good to see ire. Yet in his
behavior, his Liavior was wry discreart fran that. So then, )1J kru, I stested, iJ,y cbesn't - ikry
th-r't I jist give yar the rarairthr of the irrrey - the belae - rather thai ycu givir ire beck this riasy
ad then ire stilL cir yc*.J the irore irtrey ad he said becase he dicti't have the stuff yet, r4rich nadir
sase at the tine althct, it surprised ire becase I thait, ycu kmi, he wuld have it with him. I thirk
it 1xut this tine that he becaie Bcb ad was ro lc*er Craz. All riit. ALL ri,t, beck to ar earlier
segle-rt. I thirk wry early in this dr-iran I was ai my riotorcycle alcre ad I'm Lookirg for a place to perk
a, a groll kird of city street that was at the intersectia, - the street that I was actrally a, was kird of
- was a hill ad I was at the bettan of the hill irith is iiere the intersecticrr was ad I was drivirg ck*.rr
this rood tcsard the intersectic*i ad sa ro sçaces a the rit sick of the rood. Thai a, the Left sick of
the road, of the street, there were twa sçues Iry a arkir rioter ad I p.L Led over into, ya krxM, the
first of the twa Leavir the secaxi sce ci theari of ire ad as 2 çrae n there were like f5e other
cars alL jockeyir psiticrs for this sçece thead of no ad that was very distorted in the serse they
weren't alt lirod ip ad they weren't all canir fran the sale directia, ad in fact, ycu krui, ft di&r't
look as if they were tryir to get into this sce uitil they actr.aLly did it. Thai, ycu kr, ewryixdy
tried to chsh into this sçace ad the rsa, r.tro actralLy did get the sçce was rot the rscn iio was first
to the spat, the prrscn i,iio hod care rp fran the directia, I hod care fran, fran the directia, Idiird no hod
tried to perk ad he was a yrxr, Little bit lcrish haired erscn ad he Ferked his car ad got cut ad
I,ere he parked his car, again cares a distortia,, rot betwear the twa lines, by the neter, bet way ip
thread, close to aether car iio hod left, really a treasxkr.s arc&rrt of sce so that, yw krc*, aiy big car
cculd have easily have fit in ad I thcr.rt that was stra-9e that the, the rioters uld Is set ip in srxh a
way that ycu kr there wzuLd be so rnxh sçece between cars. Ard krai he got cut of the car ad Looked hcs
he had rked he thc*it there was sarethir wra ad got back in ad backed ip into the yw kr, in
between the lires irore, ad then I stood there ad Locked ip my bike ad as I was Lcckir r.p my bike I
thcr4it to myself, Ken this is a &ean. It certainly feels as hudred rcart realistic. Yw kr here ycu
are Lockir ya.r bike p. Aid ycu kr I Looked arcud ad ewrythir was rfectly clear, prrfectLy
realistic. There were ro distortias visrally ad I cci wait ad see ycu'll be *en ip scai ad ycu'IL
be taLkir thxut this &ean ad reteLLir-g it as a drean. Aid I jrst kird of filed that aay, even thcrjr I
cculd feel as thc&i it was a drean, my mird jist said, 'Hey, this is a drain'.
Remirr
of
drean told after gettir ip:
447
S. IwaswithafaiiLy,akirdofargnotheradnayteherfiekideadtikethekidewere
or saiethir to varyirq degrees ad it was a pretty tcrg segi,t with than tbct I ckn't
aLL retar
receher aLL of it - aLL the detaiLs of it. It was kird of ne ta1kir to the nether or nn',4e there a
father there too, ro I dcn't thir there was a father ycu krcw, xut hcs the kid is ad hcM to take care of
the kide ad thirs Like that I was real sa:i axut the sitwticn hecaBe I had got to he real friedty with
the kide ad aL L ad I thirk Ccrnie was there at this tine r ad kird of Like at the ad I t
cnto kird of en cç hack pxch ad there en the hack pxth were aLL the kide stadir9 they were Little kide
I thirk fran age three to age tweLve or salethir9 ad I thirk they were aLL Ixi,s? Mayhe they were alL boys
ad they were stadir there a, this mLL LittLe hack prd, Lockir9 over the raiL ir ad aLL of 'an were
aLL discoLorad, kird of as if their entire hacks were green, grass stair ad brc.n grass stair ad I was
reaLLy pzzLad, I thick, jt.st reaLLy iet at the fact that, ycu kr*, for sate reasai I krei that they hal
boa, this way for deys, ycu krrM, their nether hach't hathaJ than at aLL ad was reaL irrespsibLe ad jIst
reeL sad thxut that. I thick that was the Last scee
rot takir9 care of the kide, hathirg the kide ad I
with the kide ad then Later i,en I taLkirg with Bth or Craz the, the - it kird of Like, sort of
Like, sittir at a, cut±or caxert ad these prcple with their kide were saietâ,at cicsa by ad the eial Ler
kide caie ip to the twa of is a-ti kird of, ycu krm, fcrdty,said gcxxkye to is ad we assu-ed then that the
text tine we were by we'd stcp by ad see 'an again. I kird of shook hat with 'en, Let's see, it was a
Little kid - there was saiethir9 fur' xut the way we shock hart - ycu krtw scnethir weird. Both - we
we rover exprcted him -- the iLt kird of
were both very taxhrd that the kid cale i.p to say gcxxt
cairtesy to say gocxt.e to saieae - to cats over aLas to i.here we were sta-dirg to say gooci to is ad
that kird of interrapted Is taLkir with Bth or Craz oever it . That, I thick I briefLy expLairI to
ktoewer it was, Bcb or Craz, ycu krxM, i,o the kid was a-ti .kat ha-I jiet gas en with then. I thick that's
it.
44
NIGHT 4
Pre-Sleep Interview:
E. t.kat are the thir9s goirg m in tur mird r-?
S. Really thirkir txut my 101 class taiorrc. I ckn't have that prerai t, the reairg a-d
all the other stuff that I have to d. I nest certainty wilt drean xJt that tmit. At niit, I often
ck-ean xut my 101 secticr. A lot nore ccnflicts have arisen in terns of that in rkirg with Bentcn. I
thc&it I s goir to be thle to cb my ai thir. I have to ck, exactly i.iat he baits a-d it's very hard to
fit into his strttLre of thir9s ard ark with neterial that really ckesn't interest no too nuh. So I'm
really stper frtstrat&1. A lot of stuff has been goirg m this st week that has been associated with that.
Let's see, .iiat else. kd there is atweys the rk thir. Niit after niit - I'm even gettir nore huig ip
m It r y.,i - rot beir thle to - it's really a bettle.
E. Are tu jist rot cbir it or...
S. Yeth, I jest stay at hare ckn't go wt ad I ji.st fird other thirs to ck a-d there are the
bxks sittirg cç,. If it s reaJirg that I had to
de
I cctitd ha-die that, b.it it's the writirg - p1ttir
it cbii m er a-ti caimittirg myself. I'm kirti of dewtcpir a pticbia by rot dDir it - I jiEt fird all
these thirs to cb instead. I ckri't even irderstad myself i.iy I c it. Jrst in teniB of my life - ar
arall picture - I thirk I've decithi to stay here for sure - foltcir this rhole weekerd exoriace, the
Ith. All the thirs Rcn, Chris a-ti I dealt with - that's rot thsolutety for sure, bit I'll have to see ha4
it's goir9 to be. It's goir to be pretty gocd. I jest sp3ke to Rcn tcniit for t.o a-ti a half hcurs xut
it - deir 0.D.
E. t.krat is 0.D.?
S. Orgarizaticrat develcpTent. I myself miit get into that - so I nmy be takirg a practicun - ar
0.D. practicun ad a ther' practicun. tks. Let's see, that's thxut all. I gess havirg talked that over
with Ca-nie the other niit, I kird of got that straiita-I cut i.irere we are. She's beer - as nixh as I
rt to go I've beer very aibivatent becase there are liore resarces here in Chicago. tirereas for her there
is rothirg at all, except no. So it's okay with her if I i.art to leave, beca.se she has ro real desires to
stay here, bit iren I told her iiere thirgs were at - she agreed that it wtild rrske a lot of se-se to stay.
E. So it felt very gcxxl to talk it cut.
E. How have tu been feelirg for rnt of the cay tcdey?
S. Cl,, extrerely giilt ricklm. Wxlerately depressed. Tcdey has beer the werst cby of the week. It
i.srlly is. Mc*-tby, isr.ally is becase I cb a little bit of werk drirg the week a-ri the, say, well, I'll ±
it ciw-irg the week-er-ti a-ti I cb roe cirirg the week-ed. I kill a lot of tine sr.nby gettirg reaiy to cb my
rk a-ti then firotty arcu-ti ten o'clock I'll sit th.n to cb my werk ati Leralty wird ip sleepim cnly three
fc&r hours sr.rday. Last nirt I na-aged rot to cb aiy werk at at I. I dic*,'t Leave the artnE-rt fr-ca alt dey. I looked at alt the thirgs that I had to cb, clea-i L my desk ii,ich ha±r't been clearI in a la
tine a-ti fcu-d stuff that I wes lookirg for. The noin thirg that I had to cb wes a eper that I dich't cb
ati I st ip iritil 1.() an rot ±ir9 it. So I felt - as well as beir tired - as a piece of shit. I'm
or
goir-g to have to discaor hy I an thLe to naintain it urtil I forget thxut it. I lam that I'm in a pretty
goal nrxd. c, my regretful thir-g - I'm rot a-are of beir rticularly regretful at this manit. I krrw this
nornirg Like - r - very regretful. As the cby goes m I rejoice xut the goal thirgs ad forget the Led
- the, I get cks.n a-ti I thirk .I,at the hell u,tit saiothirg else hçe-s ati I bxrce ri,t beck. I ckn't
let it stay with no. If I ejtd stay with the rein a-ti be depressed for a week thai I utd cla the werk
Lecase it *jtd be so reinfut. I ckn't disciplire myself. I werk strictly 1,en I feet like it. That's all.
449
E. Okay-gocdnit.
1st. Awakaiir.
No recall.
2rd. AwakmirG, 1.3Oan, 5 mm. into REM.
E. tjiat was nj-nir thrc*i ycu- mird ji.st at the mirnt I callad?
S. I had jLst oi-der&J a pizza at a, (eLse) I was in a kird of a grocery stoc-e-restara-it deal,
ad, th, I -t to a nagazirm rack, th ... I wait to the nmgazire rack to look at the raw Pla.4x'. kd I was
tellirg Ccrnie, ad, th, the net-s clerk caie over, ad I kir like orderm-1 that am, that issue. kd so he
took it, ad like ta-it off like to wr, it, or th, or saiethir, b.it thai it Lecare like a pizza arlor, ad
it was - a-ti he wait off ad stixk it in the cn. Ah, ... tkii le, ad tiii le in a sm-se it was cookir, this
pizza was cookir-, th, I becaie aware that it was ai isst.e that I had already seai, ad didi't wait it - it
wasn't the new issue. So I go ip to the -- next to the urn, ad he was kird of a, older urn, with gray hair,
a-ti I was sta-dirG there, Like waitir for him to help rim -- b.sy with iJiat he's d3ir -- ask him if he
caVt pill the pizza cut of the ova,, I ±n't wait it. Ard I sta-d there for qiite a .tiile, th, I ±n't n.sl,
L ad telL him riit ay. Un, I'm lockir ad I see him playirg with a...un... this kird of deli - dell's
face, that's nmde cut of soft styrofoan, that's cut art of, th ... ad the styrofoan's cut cut with
sarethir with ku-ti of sharp little eces, little pints. Ar-ti if ycri miss ip am little pDint, yar also blar
it, the little dell's head is kirde ripped. It was pretty cute, I that. Ard thai he begar, ycu krxM,
takir-g care of saima-e else's pizza, p.Ittir- salethir9 c*i it. Ard also in the neaitine, th, there was a
yc*r9 Ixy there, a fellc*r, nakm eiit years old or tm-i years old, saiethir like that, a b' also Ldiird
this 1cM ccuiter, a-ti a-other cLstaler a-i the other side. He said -- ycu krtw, like waitirg to order a pizza
-- a-ti he said, 'Well I g.ess I'm ...', sarethim to the effect that he was here before -- this is the
astaier -- t,ai he was here before dich't he get fran this w- fellcM? fran this, this six year old. Ard
he nrntici saiethirg like that, this six year old. Ar-ti the six year old qJesticr said '... Un, ycu jr.st
orderad ycur drirks fran rim last tine, iu order-ad - ad ycu orderad the pizza fran the other gay'. So the
other iy, ycu krcw, agreej, 'Ycu're rit'. A little e ad thai wa-it ad did his order. But it sears as
if rtw, ycu kro.r, the nui, this older prrsa-r that I'm waitim for, was mw nakir9 the pizza, pittir9
salethir9 a, it. But after all this had occu-raJ, thai fit-ally I stcq warkir for a ntnit, ad I said, I
askad him if he cca.ild take mine cut of the ovai. tin, I Faid try dellar already already, bit I, ycu krur, told
him I did-r't wait it rur, askad him if he cculd take it cut. Ar-ti he said ro, that it was too late, ycti have
to have it th-e all the way, or rot ckrre at all, he cai't, ycii kr-ui, de it half-way. So I was disp3intad,
a-cf resirad that th ... that I was ga-ra eat it, eat it aay, if it was -- pr&thty have a-other, a-ti
that we, yci krui, tuild have am pizza ad we wa.ilcti't pit off havirG am uiti I Like next nnth, thai the
next ntrith's issue of Playboy caim art. (Pase) I g.mss I had jist -- Aniray, I was walkir- over to tell
Ccrnie that he cc*.ilch't take it cut of the i ru-i, a-ti after JLSt like goir thrai the ratiral process,
sta-dirg there ala-re a-ti ±cidir-g thiether, ycu krur, wel II gess we'd have to eat this pizza, a-ti thai i-ext
nrnth we waild have a-other am of -- It was as if, ycu kr-cM, this was like a smcial cathiraticz-i of pizza
a-ti althc*i we likm-1 this, we preferr-ad the - this other ku-ti of pizza, bit if we ha*i't or-dereci that we're
waitir9 for next ritnth, well thai we' IL have this sFecial carbiraticii pizza alysay, a-ti thai, ad t-hio krmrs,
un, shat will hpen? We'lL have the other kird next ntnth, a-ti rimyte it'll be a tkrile alyt-sy. But if it'll
be a-hi le, we'll be ready for aother pizza Ly thai alysay.
ixn cp.msticniir he statad that there was a distorticn in 'the nagazirm switchir fran Playfx' to
a pizza', that the six year old lookad older thai that, 'he lookad like a ten year old, bit he said he was
six years old'. The oily pecple in the drean were KJ, Ccr-nie, the older rsai, the little b', ad a-other
ctstaTer.
He thai r3rtad these other twa segirnts:
S. It tertJ to be a different ku-ti of thir, Like, th, yw were canirg into the roan here, to
fix sare of my wires, ad ycu bra.it a scissors with yrxJ, ad th, a lcr pointy scissors, ad ycu care in
cntheritsideofmy,adycuwerewar-kir9mthethir--a)thetiiateveritwasthere, ad!, xi
450
uld ise a
kru., s curic*.s, becase 1 as ccnfied as to shy tu had brc&,t a scissors in, shat
to play with it in my ri1t had, ad it •
very
scissors for. Ah, ad I picked it i.p, ad I, cci
diff... it s wry hard to c ad close. And then I looked cn to see if, .t*.i krm, if there s a screw
that held it, ad that rede it tiit. Aid I thc&it there s, I sa it ca.ild be lcxaied, ad it should be
loosened. It wes real stiff. Ph... aid I wes thir*ir, hew strae, ycu krca, hew cculd ytu ise the scissors
to fix shatever ycu were goir to fix. Ycu cai't ji.st cut a wire, that's rot ... ad I thirk that's all the
se
I rber thxut.
He then a±led:
S. All rit. This segiit, this seg1it seais to have preceded the - the pizza, ... I s like
in a ter,t, in ai rtiient hcuse. Aid I kind of livir cn the third floor ad like this s the tcp
floor, t saiew there ns saieae Livir ithe ne. Aid for saie reascn I dicli't like this this g' sho
lived içstairs. Un, or he as aroyir to ne, or kin saiethir. I'm rot too clear m the details, bat like
at ore point in the seglent, it sea like I s cut m
my Lack porch, th, nay clearir off all the sr,
like i kr, as if there .as a hole in my floor or saithirg, cn my back porch, ad I ias p.hir all the
srø,a tc*arde - ycu krt.i, kirtb a sinple jcb of jist gettirg rid of all the sroi, or if it saie other
thir9s, ycu kru4, a serse of clearirg off ad nekirg it cleai. Un, that kind of a ser ge of - excqt I wes
a-cry at this g..iy lçstairs. Ycu kix, there's - it's I've forgotten a lot
of
the details fran it, bat I
...asifiti.ascutofmy
view, it as flyir jwt uxier a cover of cicix, aid the cover
of
cicu rnpresented this other g'. Aid in
flyirg, that airplae as tryirg to go as hid, as it cculd, i to the ece, ip to the bDttan ece of these
cicu, ad as it kt flyir9, ycu krc.i, this short distace, yw krru, it scr... I caild hear saie hard
scrir scu-ds of the tcp of the wiry agairst, ycu kroi, the bDttan of the clout. Aid hearir9 it , ycu
krcsi, a gocd feelir, beca.se it, ycu kr,, it nea,t that the pla-e as flyirg close ad as like takir off
the battaii ece of the cLcu, ad that
very satisfactory, that's shat I waited it to d, that
5 FT
intmtiai. Aid then - saiüiew take off that battan &e... Aid, Ccrnie cale cut, I gss, cnto the back
porch, Un, to kind of like find cut shy - shat I ckir, ad shy. Aid I'm rot sire of the interacticn
here, b.it it semis as if she as ccnfi.sed as to shy I wes cbir9 it, aid ias, th, kind of s*.n_gestir that I
rot. Yeth, I thirk that's all.
3rd. REM Awakenir, 4.35an.
The scm-e as xut ... it semis to be oi-gaiuized arard, centered arcuxL.. I as stadirG arc&rd
cutside a-i the grass ad we're playirg catch with a football. Ru- yiir arcixd, deir that. Aid also -- I also
sa other Fecple there, fran th - there jist semied to be a lot of than arcu-d, that sort of pecple. Ph, it
semis Like there also were cçpsir9 ... Un, ad I had jist arrived ai the scm-e, recently, ad I played
catch with than for a littLe shile, ad then, after a-a or twe balls, I wes ready to go cii, ad waited to
go. Un, yai kr that &ean experimxe, in the last few namnts, few mirutes, I exerieroe- Un, yw
krcsi, eqeriatir hearirg ai arbAae I as hearir9 a siren in the Lackgroird, aid ne Leir9 ccrcerr
i4iether they were, shether the police were cauiir, ad I udersta-d that I shwld - yc*.i krow, that I
shculch't see that, ad rot take the risk of stayir there, hair9 arcud. Like I'm pockir9 ip my stuff
into my tote bag, to get ai my bike ad go. Aid there were twe footballs there, aid it kirb set ne back
bacase they were aily possirG arcud a-a, saishcw. Aid then my yel lew - shat de ycu call than? strs,
ycu kim, bat ycu wait to str thirs to the seat with, ad they were all ta-sled ip into the frcnt sheel,
aid saiethir else they were in kirth of a pottern. Jt.st as I weke ip, I ias still playirg with those
thirGs, jist as I i-as tryir - tryir to cb it real fast, ad gettir rarvas iAxut it...
lkxn qsticnir-9 he said that the eçerie-ce wes mildly mioticiel, pleasait.
4th REM Awakenir, 5.54an.
S. Iwasdriviripahill,atthe . cnanotorcycLe, in fairlyheavy traffic. Aid ... th
to neet a train, Un, kind of a hi.ay, ad kirth fell off that in ckiir so. I kind of cut off saieshere in
turnir arcu-d qjickly cii a u-divic hiay. Aid ah, a little red car with a iw in it, ad she ... bat
looked peeved shen I in, sha, I did it,
ad
kirtki in the process, th... I ckn't kru.i, it semis bath of is
ad it misfired, ad sort of
overturrJ saieshere. As I s tryir9 to accelerate ip this hill, cii the gas
451
a fried of mire, Steve, i4io is a otogrçher, ad he's tryirg to neke the point that ah, that he's cirg
the -- that he had a Lot of experiere with ne i.iiere ±otogrAiicaLLy I hal -- %.hen he was ci, the caiera,
ad for a t1otogrier, I had saie thir9s that were -- that nede it difficult for him, or irccrwenient for
him, I &n't krtsi. I'm rot too sure at it is, bit sac*i he was the picture-taker at my ccnfirnEtic*,,
ad, saie other kird of thirgs i,ith, u kr, didVt hpm all the tine that I knew him. 1iat hçe:l
earlier, ji.t before I was pla-nir en canirg here, th (si, ․) I'm returnir9 saiere, goir ip to the
third floor of my artnnit, ad .4,en I get to the third floor, I fird that it - it's rot my ertnait, cnly
it's Laid cut differently, cut there in the hallway, ad that there's a ini there, ad she mniles at ne
wry pLeasatly, ad s's that she's waitir9 for the lady irEide to Let her in, or saTethir9. Ard the
Ard she exp1airl to ne that she was goir in to
shculd be stadir there ad waitir i1al the dDor is
see this lady's beaitiful furniture, or saiethirg. So I tell her that I - that it in the wraorthent, that I'm rot really her hibad or 14,atever, that I ±n't live there. Ard - ad I start ±wi the
stairs. Ard then I decide, th wait a miri.jte, this porsai is goir in there to look at the furniture, ad I
really enjoy Lookirg at pecples' old fuTliture ad old style, i,y th,'t I thirk of askir these pecple to
take a look too? So, the m's en the stairs, ad she ad a yxrg porscn, ad there's this ae short, real
hea aitiqe tthle in the frcnt as ycu walk in. But the rest of the place is like aLL juk, ad there's
lots a-cl lots of pecple in there... Oily like a big porty. Aid I'm jiEt follewir arcu-d the lady of the cut en the La-dirg, the frent ladir9, the frcnt hallway, I thirk. Aid
of the hci.se, ad this in, .ho
ad,, I fard cut it was alL jirk ftrnitu-e, crfts a-cl bthies' crils, ycu kro.i, for taller- ad ckthle
nettresses, ad triple nEttresses, it i a terrible disintnt, lockirg at all that ft.rnittre like
that. So we're welkire arctrd the ha.se, ad there's a ruiter of pocple there, bit fran the look of thires
she shci. than each roan, frau the roan we were sta-dire in it jist locked like a he porty, it looked
like a sattr±y ni,t porty, with everytxdy the .aTm krtis there. Ar-cl, a-cl I was kirt sta-dirg in the
kitchen, ad lookire into the other side of the roan, ad seeir9 lots ad lots a-cl lots of ecple. Lhs...
krtsi, I gess the porsa, sho was interested in biyire saie of these thirs, ycU kris, like friax, ad so
forth, was ro Lcrer the im iio i in the frent hall, bit was r this - this giy, a giy, he seais
kirth weird, jLst like the stuff he's interested - ycxi krui, the Da,ish stuff, a-cl I knew that he's got a
lot of kith, ad he's nn,tienire the triple nmttress, ad so forth, ii,ith nakes ne thirk he's kirt weird.
(Faze) So I'm leavir9 there... I'm leevirg there... (a.se) So I - I leave there, I thick, with this guy?
with this stra-e g/? ... un ... I &n't krts. There's a picture there of this strae guy, bit I tad to
be nore of a-i thserwr. Cwld there be suiecrie else i.j,o it with this stra-e giy, ad then ha-I prcblelE
with him also... Hc*ever, I'm rot - I'm rot sire i.ho it is, E-iether it's ne, or I'm &servir saiecre else,
bit ad,, I'm leavir9 there, ad ... (case) th ... I ckn't - I - I gess - there's a g in my drean, bit
then jist - jist before the scere that I begai with en - en the notorcle, in, I'm beck en the notorcycle,
ad ad,, ad then the saie fel lew was welkirg ckn.n the street, ad sanehcw it - it, yw krsi, it node plain
the neenire of ad,, yc*.i- ri,t, this guy is weird. Un, in a se-se the evithe for it was the fact that my
riit ear electrcde was of f again. Aid ycu kr-ow sanehci this guy was reslxrsible for it. That was jist
before I got en the bike ad polled cut ad did that U-turn I told ycu adxut before. He had a bike, ad was
-- or I was tryire to fix my bike myself, hen I - there was a green wire that iait to - ,ird my rit
eer,axdIwastryitofixthat--fiXtheelasticarttogetitbacken...Un...kdthat-that'5
kia,, ycu krow, it seaiod Like 4ien Stew bega, taLkire to rim, either my veice or actimlly talkire, he
starts cut with talkirg with this guy, hew he we- hew Stew was ri,t adxut this guy beire weird, ad then
he was talkire adxut hew I did all kirth of thirGs hhen he was takirg pictures that nae it harder for him
to take pictures. I'm rot sure. ALL rit. Then a capLe other thirgs fran earlier in the cb-ean. Ard cre of
the... thelast... IthirkitwassqpsedtobeaV.A.cafeteria,adit'slocatedck$ncnsalekirdOf a
beat dcck, Like for shiipire, like a pier. Aid it's a stra-e hour of the thy, a-cl I s&çxse we're ckw,
there. I it thai there with Lots of ecple. Sale friax, I'm rot sire ho. Un, nmybe Cliff, nmyte sarecre
else. Ah ... a-cl it's strae beir9 there, th.n ' the shipoir pier. Aid we go into this old, Lcrg, flat
biiLdire ad I'm feelire rnanfortthLe deir it. Aid way cksn at the erd, there was scam kircl of cafeteria
there. Aid, ad,, we it thai there, ad in the shcwase, yw krow, the different fcxxis like ,ird the glass
were all these little sort of frozen fotx displf -- were all these canera Leses. ... sane fictiticas
brad, sane caiera that I'd rever seen. Ah, ad they were -- ad goi in I had saie kird of ai expectaticn
of it beirg qiite ..., enly I heard that they were real cheap there, or sanethire. I go in there, ad there
are all these stra-e - stra-e braxis of caieras. So I'm Lookire, ad pleased with that. Then we get ar
452
focd, or sarethiog, ad ah, I ckn't krcw shat that was -- it was strae focd, or it wasn't real focd, or
shatever. Fu1V, it was sate kird of a cai - or was a
(te rue cut).
Un, saie kird of fur' lookir9 focd a, a - a, aie piece of rard, rye - or satethiog that was,
a, ae piece of bread. I'm rot sure. The food was strae to ne. Un, bit I g.ess we reid for
it, I'm rot sure. Then th, we lay the chaige a, the canter, n.sh arcuxl the cashier, ad then we walk all
the way heck ckn to the ad of this lcrQ sort of cafeteria, ad then a caple of strage- stragers were
tu krrM, cnly
there. N-d - let's see - j,o were they? I'm rot sure. Ch, i.iat they're tiaJ into, it seels, is a-other iJiole
thiog that got than all together, bit un ... there's kird of a sibmrire thiog in here, ad I'm &serviog
saie ship at sea, ad ... I ckn't krut ... tin, bit a-y, I cth't kmw hc* it cate axut, bit this ship
kird of lcrg, ad the stbmrire that it had kird of like in it's hold, ad the i.hole - t.i,ole piliog way
wayhck,th,kirdoflikethegiyintheYella.iSiblErire,th,th, Iforgethismte,Yul?--th, Itha,t
it was YuL. O.K. But the, tin, the Lcrety Heart's CUb tim, that old giy. But at ay rate, I tad to he a - a
dista,t clserver of - of this, tu kr, the sthrarire haviog torn away fran the nmin ship, or - or sate god
thin thiog. Nd I kr like in the distaxe, tu krtw, it's kird of, or tiiatever the ship's ±irg, ytxi kn,
retainiog the mioller s&.b in. The scae of, ui krw, the st.b heirs way thai, in the water, or saiethirg. I
ckn't krs, still is missirg. But then the ship firmlly got the door ci satehat. It seats like kind of
this ship kir cjxnod i.p in half, ad triad to take in the sthtmrirte. Ycu krtw, it sei like it had a
place for storirg. But rather thai the sibmrire see this ship split in half to kirtb take in the s*blErine,
kird of like th the frei,t, or jet tra-sxx-t, that cçers t.p, yai kri, the ckors ad - a hie d3or, ad
ta-ks a-id that drive into it, that kit-id of thiog. kid it the - the ship missod the first tine, ad it dich't
ckt it prcçrly the first tine, ad, so it had to ck it again. So. Mae this follcMirg that thir, that we
wait ck*.n to that cafeteria, sateway in the drean -- bit I was ji.st the cerver of the st.bmrire a-id the
ship sca-e, shereas I was ai active erticiait in the cafeteria episc*. Arey. All riit. Nd then there
was a-other shole seganit saie,here else, shich seats to have precadad the th the erthmt scae, yai krs,
i.kere I was goirg in the artnent. Ard that is, th, t.1,ere I reimiter the hegimirg, I'm stadir-g in a
rtnait, prc±thly the third floor, or the fc*.rth floor, salethir9 like that, ai erthmt that, I ckn't
thirk it's ftrnished, ad it's got a lot of wirtbs, a-id I'm sta-diog there lockirg cut. Nd I gss I'm
there with my brother ad my nether. Nd there's severaL wirihs close together ai the rit, bit there's
atty ae wiri±* over to the left. Arid I kir pint cut to my nether, at first, I g.ess, that the tree? or
the trees, the green tree? ro, that isn't I dich't rotice the green leaves ad stuff a, the trees, that
the trees in the left had wirn±.P wash shat? havirg sate, or were rmi grcwirg? or saiehc were different
then they had Lea,, ad were, ycu kmt, alive ad grcsirg. Nd my nether was skq,tical of that, so I watchal
those trees for aiti le, for a few mirutes, a-id p3intal cut to her hc in fact the tree was alive, ad
cha-iog, t.I,atewr. Un, ycu kr, it had grazthlly startal rrtwirg, startal grcitirg, ad then am begat, I
helie,e, th, grcsiir in the wirxkM, ad that was prctthly the tine t.iiere I was goim to - i.here I jt.st th,
j,ere it saithc* certifial my thir, ad she sew my paint. Nd see that, at the tine, ycu kr that iazuld
irvolve satethirg that was a mystery, grciieg a, a, I thirk, sate kit-id of point, I was illt.stratirg it, to
the tree grcsiirg - to the interrelaticrahip with the trees. Un, bit it seats that thei we ,t thai the
were wtsi a, the grass, ad r, it- it semis that there was far of is, I thirk far was
ad my nether, ad Ccrnie ad I, ad, there was at exchaige of gifts at this tine. I had given
my brother, ycu krm, a sealed erivelcpe renirdir9 him of his birthc' perty, or satethirg like that, ad he
steFs, ad
we
my brother,
was in the process of cçmir that at the tine, ad that was gratifyirg to ne, saishow. At the tine of the
drean, I was thirkir9, ii, he's cpe-tir9 it ritt here ad rmi, aitsicle. That was a strae feeliog, for ne,
heca.se I've rover semi him cçm a birthcy card hef ore, a-d it - it was a very Iniqm a-id pleasa-tt scate,
we were a, the grass, aid cçiirg my Letter, ad that. Un, a-id it seats as if I had all - as if I had
received previasly - nm-,ke rot previc*.sly, bit satetitie, ycu kir.w saistitte previctaty, a birthcy card or
satethiog fran him. I &n't krtw. The,, it seats like Ca-nie had also given him satethiog? sate gift,
already. Nd thai it was as if - the stra-ge pert was, all rit, he had gotten sate gift, in a Letter, it
seats, ad he had sent - he had tiD sate kind of, satehow, a± i.p shat he had gotten. Arid the,, he jist
gave its? or ,tm.4je Ccz-nie gave him the,? or he gave my nether, or satethim, I th,'t krew ii,o, ten chllars, a
ten chllar bill t.hid, was sateha, ycu krtw, in exchage for pert of i.hat he got. He wasn't giviog it all
back, he gave nest of it beck, aid the ten d3l lars was for twa people. I th,'t krow. Saielxdy a±ai it wrcrg
scuethiog. Al,, so I was Leavir9 this scam - that was a very brief wtcbor scam - ad I was retu-nir to
the ertrrent, as I returr to the ertnmt a, the cborway that was a, the grcu-d floor - there was at
or
453
erLcpe, ad there was a-other birthdey card, ad I cçied it i, ad it seaied to
watchir Steve - I'm rot sure - bjt yai krti., cç1ir it - as if
we
be a card that I'd heai
ware beth faniliar with the drean. It
Looked Like saie kird of faniliar birthdey card. Aid I exeriened that as saie kird of reaffirnmticn of
my
beir a gocxi perscn, saiethir aixut givir him a birthdey card, or saiethir m that order. Aid r* I want
ip the stairs, ad this is the erthEnt iere that other story took place, that, tu krow, iiere it was rot
exactLy my rtnent, i4iere my artnmt was Like cre deor, kri, rorth, ad i.iioewr was actwLLy west,
west of this ore, ad ... of that arUrent, ad the iJiole story of iã,at haçpmed there.
(.xn beir asked k-ether there was a-iy distorticn in the way faniliar cpLe
represented,
he
or
cbjects were
said:
S. Ah, th. I thirk it was actwt Ly iiai they occurred. Let ne jiEt kid of ru, thra, than.
Distorticre, they were the stiLls. The i1otogriier, he was riekir the pint thxut hai I hed ckrte â-iatever
it is that I was deir rm before, axi he ixinted cxt a pattern, ad yet I - it's stra-e iiere he cculd
have care fran becaEe I was &ivirg a notorcLe at the tine, ad mt he's a ssaer.
(.xn beir asked if he was ai active mrtici,t in the scams,
he
replied:
S. ALL except for the sitrmticn that had the sthiarire in it, the ship. Other Tha, that, I was ai
active articiFa-1t.
5th. REM Aikenir, 6.SOan.
S. A brth of ecple were sittirg arcu-d, in rèiat seai Like, I was sittirg arc*.rd in ai
artnmt. I j.st then asked... Cartwriit 1= Lookir for ai irdEtrial jcb for no? Aid I was askir than
J,y. I i.ms askir... thris? ad Ren? 1-ry Cartwri,t was boc1ir for ai irãmtriaL jcb for rim. I gess
that's- before that Diam aiswered the ha-e? Aid it was Rcn, ad she was tryirg to fird wt if my schaile
for i j,e, I was goir to be seeir, xi kri, for this irdEtrial client. I di&r't udersta-d that becase it
were jtst sittir arcu-d in that did-Vt even said plasible. I ckn't krrsj, it was ccnfrsir. (ra.se)
in the livir roan scam, ad, we did that for a4ri Le, I ... there was a stra-ger here .tio- did he
kr-o. other cple here or rot? Arrimy, I did-Vt krn him, ad he said hello. Ce rsen told no he dicki't
live in a-i ertnmnt here in thicago, ad am mrscn said he knew him acrcms the street, he's a stuit
here, ad t other pecLe were in the bei Ldirg over here en the other side. It's real stra-9e. thris was
saTeciiere, a-d Dave a-cl ... said that he had ha in a qer en ... for sthool, ad he cr-uipled ip the
rai draft of that çqmr, ... I jist get a feelir that a 4,ole hxe-pxje of thirs that hçeied in this
roan here, ... Aid thris is chair his 1aits, pittir9 these hhite, soft -its
en - cha-ir-g his pa-its.
EarLier in this mrtnit I was talkir aixa.it saiethir to Carol, the secretary fran ±wstairs, ad I
ha±i't seen her in ari le, she'd been a.oy, ad she care in, it searm I was sittirg at a t±Le acrcms fran
her, ad she was talkir, ad she was into sate real heavy aTotiaml stuff. I ckn't raim±er i.hat it was,
bt she was beirg real aiotic*-ol. But then she becaie ai oLder nai, I gess. She becare ai older rm.
Al,, ad saream else walked Lb', ad kirr looked in en r.i,at was goir en. But she centirt, ad she was
really wry eroticral ly irvLved, ad cryir, iien she talked aixut rJiat it was. Aid I had been, I gess I'd
bee, taLkir to Steve Reise before I begar talkir with Carol, ad he rt into a-other roan, ad started to
talk to saimam else, ad then
he
calm heck, ad i.âiile I was still talkim to Carol, Listmir to her, he
started taLkir to no. So I turred to him, ad asked him to hold ento it, that I was taLkir to Carol, ckn't
intern.pt no. Aid he said sarethir aixut that we were, in fact, talkir before.
E. Do tij rereitrer...
S. No I dich,'t. I thirkheia,talrioto. Ardwekirgotip, adhassled itbecka-dforth. Ah,
I thirk saiethir abxit rceptien checkir. Aid he asked no a qostien that put no en the defe-mive shen he
acttmlLy asked no the q.Jesticn, to show that, Un, the ' he asked no the qiestien ccntriLijted to no feelir
defa-miw in a-swerirg it. It kirn seared Like he was &ewir to the qmstien in aswerirg no. It was like
centrib.itir saro blare to ire for really misu-derstadir, for rot rcwir ad stuff, all the blare... So,
454
then he went back to i,at he was ckiir, ard gss I finishi with Carol. Aid that seen to brirg ne i.p to
the ioLe scene hççnir9 in the i.i,ole Livir roan, with all these pecple -- sarethir9 was goir a,. For a
tine I was kir the center for the ccmersaticn. th! I did-i't iroLi saie scenes -- well jwt before the
scene i.k,e, the new uy caie in, ad nenticr1 the other pecpLes' artheit, ad so forth. Earlier in the
drean I was outside, ad interactir with sale pecple, I'm rot s&re iiio, bit..., I'm tatkirg with saie
pecpte, ad we're tatkir at a titicat - saie kird of a pDliticat barqiet. I ckn't krxsi if they were
- tu- pecpte fran Atterretives, or fran â,at exactly. But I was interactir with, th, outside, in
the sui, ad I was sittir at a tthte, ta1kir, with saie pecple. Aid then there's arother scam, Like a
street scene, by a, kirda like &w, by Wabash, except ro, becase it's - the train is nxnir a, the grc*rrJ
Level, it's rot the eLevat, it's all tracks in the micklle of the road, ad there's ro ... for children.
Ar-ti I was walkir9 aLcr here, with saiea-e a-ti I have n' lu-titeg with ne, a-ti I have a bid, of pears in
it, a-ti sare ples, a-ti I p.lL alt a pear, cLean it off a, ray shirt, a-ti the-i I get distractad. Ariway, I
pit back ray pear, I ckn't kr-tsr krith pear it is, the pear I'm holdirg is very soft, ad saimhcsr I didi't
wa-it that am. So ev-itielly I fir-ti a rear, bat the arty ar that seem... Aid then walkir ip the street,
as I walk amy, I see a iJiole bid, of, th, I g.ess yc&r Saiish, Latin kide. I'm a Little aLaniad at
seeir then. I'm kir-da afraid, u-certain of iat they're goir- to de, aLthct, they are wry t*x kide. Ar-ti
I stocd ip agairt the biildirs, in kird of a pessageway, waitir- for then to go by, ad it seei like I it jwt see like it bacaim a livir roan, Ixcase, you km.,, I was talkir9 to pecple in there. Yet I
never kirda finistrad with than, a-ti then wa-it out to the street, back out to the sidewalk, to see i.i,ere
these other kide were, a-ti they weren't in sirt. I expectad to see then kirda pess by. I never saw than
pess by shen I was in this roan. So .i,en I got out, I exxctad to see then riit arcu-ti there, I dichi't kr-tsr
k-,ere they were. It was all kirt stra-e. They kirth disearad. Then, at saie pint, I was wa1kir
thra9, a, old hallway in a -- I was goir thrct, a Lcrg -- it was in this kirtb old factory, mistrial
secticn, a-ti I was goir9 thrct, a lcr9 hat tway that twistad in sare -- saimbow in the drean it was very
fani Liar, it was like I hal been there before. Aid shortly after I enteral the deor, to this Lcr hat Lr.my
wait L.p this Lcr hat lway, ard the hallway seail twistad a Little bit, kirda a, the way cut. ActimLLy, I'm
kirda takir a short cut cut to the aLleyey r.bere I perkad my notorcle earlier. Aid as I go ip to the
exit cbor, fran this kirt shortcut, there are a bid, of, r.tratever they calL then, by the ckorway. Ar-ti
they're jist sittir there, kirda in a mimlI pile. Ard walkir by that, I ji.st pickad ae of those. Ar-cl the
way I did it was by walkir alt the way back to the new baitdir-, r.4,ere I hal care fran, ad went out the
fra,t ckor. Aid the, I was ca,ferrir9 with Ccn,ie abut, uli, ne goir back a-ti gettir- than, except there
was a qerret became she waitad to have sore inforrrmtia, ad I wasn't tell irg her, I was in a nsh, I was
sayirg, jist wait twa mirutes, I'LL be rirt back. I g.ess I di&r't wait to tell her I was goir to take
sare? or? sore infornotict, that she wait&l that I wasn't givir9 her? Ar-ti that create-I a lot of 1a1 fee1ir
between the twa of is. So if - she-i I go alt to get the notorcLe, it nm1 severaL feet fran i.here it was
in the alley shen we perkad it. Aid, a Lot of the irstnjiuts ad all that stuff a, the bike - the ha-tilebars were way off, Like tu-nei mr, ad the mirror was turned over, a-ti the fr-a-it rotor perts were turned
er backward. ... we could fix it by pillirg the ha-die-bars cb.n, ... Aid we reached the ccxtlLsia, that
the bike nLst have been warked over shi Le it was perked there, or sarethir. The, I thir* - I th,'t thirk we
actielly axied ip drivir off... I'm in the livir roan scam that I descritI, the roan with Ra, ad Chris
a-ti the other stulents. Ar-ti then Diare beir there, a-ti the - takir- a nessage or the Fhcrle. Kirda like en
acadenic, sear Like ire a-ti the O.D. issis I was taikir to Reise xut before.
S. Cli, ore other Little thir. tJ,ile I was a, the street shere the tra- t.here the railroad
tracks, the trolley tracks, i.i-rere there was a sr.b.ay tracks wait by, sta-dir9 there agairst the wall a-ti
watchir, I saw this Little aitcueted vthicle ru-nir a, the tracks, that was very peculiar. It was in a
se-se of a, I thir-k jist a littLe, tiny twa t.kieel thir in the frait, that dich't have ay cDth or hei,t,
it was kir-da like as if a Ford with twa sheets. Aid that had attached to it - there was sare pipirG to the
thir - kirtiof a traiLir littLe - Little car, with a Little caqrthent a, it, it was - wasn't a car, it
was a, well, saie kird of a wagcn, or I ckn't kr-tsr, sarethir
Mornir Interview:
S. I was hçç' to see the new issrm of Playx' out,
455
so
that I could check that out. That's how the
.iioLe thiog started. Then after I kird of ordered it, he took it a.my to cook it. He took it a.y, I ckri't
km ii,y. I dich't ji.st y for it ad take it. Then I Looked at the cor ad said 'ha, this looks Like ai
old cver', this is Last nath's isse ad I hed to int that yeth it was. But I thciit it was this
ntnth's issi. -- bit ne,te it cha-iged. it's sore Likely that it chaed - I xesuie it did becaEe
thirs char,e a Lot - to ai older iss. 1at stade cut to ne is my lack of aggressiveness in telliog him as scm as I fard cut - that it wasn't iiat I ited. I neen so that I wulcli't have to have the pizza that
it waild he a lcr9 tine before the rext iss is cut
I recessarily dich't ,t. I ratimalized it that ne
ted the kird we ordered, b.it this me was alrit. In
so nke we n't pizza Like that a-y. We
a.kmir I felt a-d aroyed that I ha't hem nore aggressive. That's saiethiog that I cb all the
tine. I nem I'm iswlly very aggressive, bit in me sitwtim that I'm rot aggressive is Like with sates
pecple. Ycu km, Like shen I it cha-ige for a dellar or saiethirig or cashing in bettles at the Jewel I'll
wait in a Lcrg lire ireteed of going aheai of the ecple. Saietines I'm really rot aggressive that way a-d
it really a-ns Ccz-nie ad she is besical ly wry uoggressive, bit in those sitticrs she is very
aggressive. ... in the drean ahere I waited ad ward ip having to Wy the pizza becase it hed alreaJy been
shen he said that it had already been cooked.
cooked. I felt
The experinenter the says that
scissors. KJ cmtirs:
he
did have to enter the rin before KJ's drean xut the
S. As I was recantirg that to u I was thirkirg dxut all the FrarJiai inplicatias. aitting
off my iis or castratim or saTethirg like that. I get that my sase of yttir caning in with than was both ytu brc&,t me tool in - curing the &ean I had the fmtasy that fran the reading yw Icui that
scnethirg was wra-g ad ytu ad Lirde got together to decide klat toots to Ixing in there with yw to
reir, a-d the me tool yw brait in was the scissors ad this is my fa-itasy in the di-ean, yw decicI
that, bit hcsi was the scissors going to wark. Yw ca-i't jist cut a wire - ycu reir wires rot cut wires. So
it was a se-se of ccnftsim. You dich't have to ise it for athirg. You did I.klatewr yw did ad I was ji.st
sitting there playing with the scissors ad they were hard to cçel. There was a cc*.ple spits of r'st m the
scissors.
E. The segient that preceded that me was yt&r living in a terenent hase ad 'there was a giy
living thwe no ad I dich't like him'. IJiat kird of feelings d ytti have aixut that?
S. I &n't relEiber athirg Like that. During the drean it nede se-se to no that I did-i't like
him, bit there were sore things that hmed in the dreaii hecaLse I knew I felt okay aixut rot liking a-d
it node se-se to no becase he did or he was saiething. I was very grateful shen that plale did his thing
with the tcp wings. There was a very large scratdiing said like he was scring off the bettan Lar of
this giy that is r-esmted by a clad. It was kird of like a mrra spece in the sky. The wings were like
en tcp of the pla-e. I'm having prthteis dealing with Bentcn these deys ad so the first thirg that cabs to
my mird is that it could be him even tha, I ckn't have my recollectim of the drean that it was him or
my of the perticular things that have haFcmed between no ad Benten recently. In my cascic*.s state ri
that,, he is a source of ca-ten for no.
E. Ckay, let's go en to the third REM. 'I was stading arctrd playing catch with saio giys a-d a
footLelL...' Ycu said that yw figored that yw shc*jLdi't be there ad ycu were pecking L ycl-r stuff.
S. Yeah. That was - it wasn't as if I weit there to play footbell. I ckn't krui shat it was that
I went there for. I have feelings aixut aTbivaLaxe, bit it wasn't aixut thrc*.iirg a foothel I arard. I was
fore ccnerr aixut then mioking hash. It wasn't Like a football gale, jiEt sta-ding arc&rd ad a Lot of
trees. No notorcyctes, jIst bicycles ad stuff scattered en the grcud. Qte in a shiLe way off in the
distaxe I wxild hear a siren very far ay. It wasn't as if they were closing in en is - they were far
aey. That nDJe no thir*, 'gee, I wnler if that has a-iythirg to de with i.s here'. It wasn't you- typical
&pe scene pera-oia. In telling the ck-ean I was surprised I had a th-eaii like that becase I mioke so Little
these ckiys ad I'm rot at all ca-cerr about b.sts or aiything Like that.
E.
cth't
yw
try ad ut all these things together ad see if
456
you
cm caie ip with saie kird
of general thaie for the ni,t ad see hc ft cculd fit in with ijiat we were taLkir xut fore ytu It
to steep.
S. The dreais were pleasat, bit they all had aixias eteasits in than ... I'm ised to
futicnir with a certain air&rit of aixiety at alt tirts so that it ckiesn't irrcitate ne. I g.ess I
roticed in every sirte form that I filIal Cut, I'm at Least rnxrateLy avias except ai rare acasicrs.
Thirs weren't peachy aid rosy, bit they were pesitive, I was eijc1yir Life in my dreais, bit there were
always p-thLeis to evercaie, sa4rces of discanfort. Let's see hai did I deaL with these in the &eals. The
thir with Steve Reise, yeth, I was very Lset Jt that becase I take pride in my camuiicaticn skills. I
ins pretty clear that I - I gess I felt challer&J by iiat he said. The eleas,t with the siras canir, I
was ccrcernel. I had a little difficulty in leir9 becase
my
str ins arcud the iiieel ad aLL tal&J
Lp.
E. Ycu dich-i't resolve the pizza thirg very well.
S. No, I was real passive in there.
E. It seals Like ycu were resolvir thirGs as ya.i were goir thra, the ni,t. Ycu started alt
with pizza ikiere ycu weren't thle to resolve aythir ad were Left feelir kid of fn.strat&l &cut the
iiiole thir.
S. Aid did that have the thir iiiere ij caie in?
E. Yes.
S. Yes, that's also very passive.
E. Yes, xj never asked ne itiat I was goir to cb with then, ad in the next dream u realal en
airpLaie to cane ad take care of the giy, yai ccuLch't cb it yctrself. But then ycu started resolvir
thirs - with the fosthalL ycu were tryir to get cut of the siti.aticn - ycu started becanir active.
S. I ckn't krc i.kny the eLaiits of otogriiy cane into that dream twice with Stew. I an also
gettir into iotcgr±iy a Little more ro..
E. So yci..i thir* there was
en
inrLyirg tore of aixiety thrc*,c*it all yctr dreans?
No, I ckn't thir that was the p-aJanirit thir. The last dream shifts more tcsnrd the negative.
There are more aixic*s elennits in the segrents. I fird nnelf processir9 it in terms of hcw I'm dir with
life irsteal of lookir9 at the dreans. I feel like I have a better ccz,trol ever my Life r, nuh
nore cptimistic xut my life here, I feel nuh more in cantrol. I'm nuth more at peace with my ewircrmnnnt
my
here. I was fitir9 it. I feel nuth better xut it. That seans to be in there. That cnlity that there
stilL has to
be
prthlais to
be
overcane ad stuff that still gets in the way, bit they ±n't cilitate me,
I still say the drean is pleasant despite havir those aspects in than becase I ±n't experiee than as
terribly as before Like with the dream at the railrcal staticti [nit 1]. That dream blew me wart iiiereas
here there was jLst sarethir9 to be overcane. A Little uainfortthle bit I gess there is more of a sese of
detacfiient thai frtstraticz-s.
457