lth

l
V: i

l
i

ll

W
I

I
i

I

l
I

H

I
(DOOM

 

LENGTH OF DREAM REPORTS AS A
FUNC'NON QF POSITION EN A DREAM SERIES

Thesis {:01- “so Dogma of M. A.
MICHIGAN STATE UNWERSITY
Audrey Recht
1963

TH 5315

LIBRARY

Michigan State
University

 

LENGTH OF DREAM REPORTS AS A FUNCTION
OF POSITION IN A DREAM SERIES

by

v.
Audrey Recht

A THESIS

Submitted.to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology
1963

, . Q
F.
2' ,t t: J

‘. _\;\

ABSTRACT
Length of dream reports as a function
of position in a dream series

by Audrey Recht

The object of this study was to determine in what
way the length of dream reports vary with their posi-
tion in a dream series. Since it seems a priori that
content varies directly with dream report length, then,
the amount of content may vary with the position in the
dream series. This is an important consideration with
dream selection procedure for investigations of types
and frequency of content in dreams.

It was hypothesized that the dream report length
would, over group data, either decrease, as the result
of defensive repression, or increase due to a learning
effect, as the dream series progressed.

Dream reports in the form of a dream series were
collected from 339 students (12h males and 215 females)
as term paper assignments for a psychology course.

A word count was used to determine the dream re-
port length, and the mean length at each position was
calculated for males and females separately. Because
of the possibility that those individuals who continued

to certain positions were a homogeneous group with par-

Audrey Recht

ticular characteristics of their own, the male and fe-
male drewm series were divided into subgroups and the
mean length of dream reports was determined for each
separate subgroup.

The females showed a trend to report more dreams.
From the statistical procedure of rank correlation on
the separate subgroups, results showed a trend to an

increase in length of dream report as the dream series

progressed.

’HflunflAQ CQWng kiuawav/

ACKNOWLEDGMENTS

The author wishes to express her thanks to Dr. I.
Kremen for his interest and instruction in connection
with this study. In addition, I am especially grateful
to my statistical advisor, Dr. B. Karon, for the very
helpful assistance and time he has rendered me in the
analysis of the data. Also, I wish to express my
thanks to Dr. A. Rabin for the time and consideration

he has made available to this study.

Introduction
Procedure
Subjects
Method
Results
Tables
Graphs
Discussion
Summary
References

Appendix

TABLE OF CONTENTS

Page

11
17
20
35
38
39 .
ho

 

INTRODUCTION

Dreams have been and still are of great interest
to man. Dreams play an important role in the work of
the psychoanalysts, and the phenomenon of dreaming is
now being subjected to empirical research. Calvin S.
Hall (2,h) initiated studies using the dream series
as a research tool. A dream series is a collection
of recalled dreams of an individual over a period of
time.

Within the last several years, Hall has been ex-
tending dream series studies attempting to categorize
and to quantify the manifest content of the dream re-
ports. The establishment of normative data of dreams
necessitates content analysis and frequency counts of
the manifest dream content. He developed an empirical
classification system for the content analysis by list-
ing content mentioned in a large sample of dream.reports
in dream series form and then by devising classes, such
as settings and objects, and subclasses, such as build-
ings and furniture, for the content. Hall (3) first
presented data from the analyses of dream reports in
terms of percent of occurrence. In his more recent man-
uals (5), he presents rates per 1000 dreams for each of

his subclasses of content and rates within each major

-1-

subclass. For example, in his classification of settings,
Hall states that buildings were the setting at the rate
of hob times per 1000 dreams and that office buildings
were the setting at the rate of 11 per hob times. These
rates were based on the analyses of 3032 dream.narratives
collected from male and female college students. This
same procedure and presentation was followed with each
rate.

However, concerning the sampling technique in se-
lection of dream narratives, there are three major, re-
lated questions to consider. First, one must ask whether
the length of the dream report varies systematically
with its position in the dream series. Dream.producti-
vity may be greatest at the start of a dream series
rather than at the middle or at the end, yielding the
longer dream narratives at the beginning. Secondly, it
seems a priori that content should vary directly with
the length of the dream report. It is only reasonable
to expect that the longer the dream.report the more cp-
portunity there is for more and varied content. The
third consideration logically follows from the first
two, that being whether the amount of content varies
with.the dream report's position in the dream series.

If the majority of dream reports used for a frequency

-2-

 

count are the more productive dream reports, the possi-
bility of bias, such that the longer dream reports con-
tribute more material to the frequency counts, is in-
troduced if one does not control for position in the
series.

If this then is the case, that dream report length
varies systematically with its position in the dream
series, there is reason to be skeptical about Hall's
base rate data for one cannot determine whether he con-
ducted the counts with every aVailable dream, with ran-
domly selected dreams at different positions in the
dream series, or with a control for dream reports of
approximately equal length.

With respect to the question of dream report length
varying with the position in the dream series, Kremen*
reports the impressionistic observation that, in collect-
ing a series of dream diaries, many students, after hav-
ing started their dream diaries, stated that they no lon-
ger could recall their dreams. Due to increased attention
to their dreams and attempts at interpretation (or misin-
terpretation), conflicts and/or socially unacceptable
emotions may have been brought to the surface. It would

then be necessary to utilize defensive measures and so

 

a personal communication

this phenomenon could be hypothesized as a consequence
of defensive repression. If this is indeed the cause
of the students' inability to remember their dreams,
then, over group data, one would expect a decrease in
length of dream reports as the dream series progresses.

On the other hand, it could be that as one becomes
concerned with and is attentive to his dreams, recall
would be greater. This raises the question as to whe-
ther a practice and learning effect applies to dream
series collection. In this case, one would eXpect, over
group data, an increase in length of the dream report as
the dream series progresses.

The objective of this study is to determine, there-
fore, in what way length of dream reports vary as a

function of position in a dream series collection.

PROCEDURE

Subjects

All students enrolled in an introductory psych-
ology of personality course at Michigan State Univer-
sity over four terms, from September, 1961 through
December, 1962, were required to undertake a term pa-
per. These 1121 students had three options for their
term paper topic; a fourth Option was added for the
third and fourth semesters. The first option was auto-
biographical in nature including such possibilities as
discussion of one's family background or the evolution
of one's values and opinions. The second Option was
the writing of one's private personality theory. The
third was keeping a dream diary over a nine week per-
iod. The fourth was the construction of a fake dream
diary. The specific instructions for the latter two
assignments are presented in the Appendix. Of these
1121 students, #16 choose to do a dream diary.

Those dream.series that contained less than ten
dream reports were excluded, for so small a number
seamed inadequate for obtaining a good perspective of
the relationship of dream.report length to position in
the dream series. 0f the collected 416 dream series a

total of 339 (12h male and 215 female) dream.series

-5-

contained at least ten or more dream reports. Table I
(see page 17) presents a breakdown of the total number
and according to semester of the number of dream series
collected, the number and percent of the dream series
containing less than ten dreams (and therefore not used),
and the number and percent of dream series used in this
study. Discussion of Table I will be reserved for the

Result section (see page 11).

Method
The dependent variable in this study being dream
report length, the words in each dream report were
counted by use of the following seven point counting
procedure:
1. When two or more dream reports
are given for the same night
enter (a) the total number of
words combined for the two or
more dreams and (b) the number
of words for each dream, is.
100 (AS and 55).
It was decided to regard two or more dream reports a
night as one position in the series. It was possible
that a subject would consider, upon awakening, one
dream of two portions, as a change in setting, as two

separate dreams. Thus it was decided to consider the

entire night's dream.productivity as being one position.

-6-

2. Begin counting the number of
words at the point where the
subject states "I dreamed..."
and end the count where the
subject states "I awoke."

("I dreamed" and "I awoke" are
included in the count.)

Since most dream reports begin with "I dreamed" and end
with "I awoke," this will be more or less a constant
factor throughout.

3. Include in the word count
such statements as "some-
thing else happened, but I
can't remember it."

h. All statements preceding,
included in, and following
the dream.report that
identify people or places
are to be counted as part
of the dream report. Count
the exact words used by the
subject; do not rewrite.

5. Count as two words contrac-
tions, is. didn't.

6. Count as one word: cities
(Grand Rapids), numbers
(one hundred and three),
schools (MSU), holidays
(New Year's), and time
(four o‘clock). All
hyphenated words, is.
broken-hearted, are to be
counted as one word. The
words "Lake Superior" would
be given a two word count.

7. Do not count any extra
explanatory statements
nor any associations that
are used in addition to
what is necessary for a
statement of the dream.

This point allows subjective judgment to enter the
counting procedure, and, therefore, it was decided

to determine reliability of word count. An assistant
counted the dream reports of thirteen dream.series.
Five dream.reparts from each of these thirteen dream
series were selected at random (65 dream reports) and
counted again by the writer. The product-moment cor-
relation was .99976, indeed a high reliability.

After the number of words in each dream report
of the 339 dream series was counted, with each of the
four semesters, the dream series were combined for the
males and for the females and a mean number of words
for each position in the dream series was plotted.
Each graph was viewed for any great deviations (ie. a
plot extending sharply upward with others extending
sharply downward) according to semester (especially
important because of the variation of three or four
options of term paper tOpics in procedure). No large
deviations were observed, and the mean number of words
was then calculated combining all the dream series of
the four semesters into the two main groups of males
and females.

It was noted that those few writing long dream

diaries reported shorter dream.reports than those

-8-

 

writing shorter dream diaries. A decrease toward the
end then could be due to those few writing consistently
shorter dream reports. Therefore because of this pos-
sibility that those who continued to certain positions
were a homogeneous group with particular characteris-
tics of their own contributing unduly to the graph, the
dream series' of the male and those of the female were
separated into subgroups at certain positions in terms
of the number of subjects reporting a specific number

of dreams .

Data Analysis

 

Rank correlations were made for each subgroup and
t-tests of significance were made for each correlation
coefficient in order to determine if there was any sys-
tematic relationship between dream report length and
the dream's position in the dream series.. For the rank
correlations, the rank of one was assigned to the low-
est mean length of dream reports and the ranking of
position followed position number.

The length of dream reports from forty subjects
selected at random from 108 (29 males and 79 females)
possible subjects (with the exception that twenty had
to be males and twenty had to be females) were come

pared (the males and females separately) over a span

of twenty positions. The method of allowances (6) and
Friedman's two-way analysis of variance by ranks (8)
were calculated in order to determine the significance,
if any, of any differences in length of dream report

from.one position to any other position in the dream.

series.

-10-

 

RESULTS

In order to determine whether males or females
wrote dream diaries more often, the following chi

square test was performed:

 

 

Table A.

Males Females Total
Number doing
dream diaries 173 2A3 416
Number not
doing dream
diaries 319 326 205
Total 552 569 1121

 

'X2 = 15.65%, ldf, p_<.01

M

It can then be concluded that females wrote dream
diaries more often than males. A second chi square
test was done to determine who reported fewer dreams
of those who did dream diaries-.3 From Table I it can
be seen that a total of 77 dream series contained
less than ten dream reports, and, of these 77, h9
male and 28 female dream series contained less than
ten dream.reports. The following chi square test was

performed:

-11-

Tabfi Bo

 

 

 

 

m M n
Males Females Total

Number with less

than 10 reports N9 28 77

Number with more

than 10 reports 124 215 332

Total 173 243 hlb

 

ET?
I = 18.9625, 1 df, 24.01

 

It can then be concluded that females wrote signifi-
cantly more dream diaries containing more than ten
dream.reports.

A further chi square analysis was performed to
determine if the above finding held with those male
and female subjects reporting thirty or more dreams

to those reporting less than thirty dreams.

Table C e

 

 

 

Males Females Total

Number with 30

or more dream

reports 9 18 27
Number with

less than 30

dream reports 115 191 312
Total 12h 215 339

 

2
X = .1Lm35, 1 df, 11.80

 

 

 

 

 

The results are not significant in this analysis.

-12-

The male dream.series extends to h? positions,
as can be seen in Figure I, with three-fourths of
the males (N 93) extending to position 13, approxi-
mately one-half (N 58) to position 17, and about one-
fourth (N 29) to position 20; less than 10 N extend
beyond position 30.. The female dream series ex-
tends to 50 positions, as can be seen in Figure II,
with about three-fourths of the females (N 163) to
position 15, one-half (N 107) to position 18, and
one-fourth (N 5h) to position 23; less than 10 N ex-
tend‘beyond position 3h. There appears to be a trend
for the females to report more dreams. This is in
agreament with the above finding that females handed
in a significantly greater number of dream series con-
taining more than ten dream reports.

The male and female subgroups are represented by
Table II and by Figures III through XV. The subgroups
were determined by the number of subjects continuing
to certain positions. Thus, for the males, 12h sub-
jects went to position 10, 105 subjects to position
12, 58 subjects to position 17, and so on. The mean
length of dream.report for each position was deter-
mined for each separate subgroup. This means, for
example, that for Figure IV, the mean length of dream
reports was determined from those 105 subjects going

-13-

to position 12, while for Figure V, it was determined
from those 58 subjects going to 17, and so on.

The results of the rank correlations between po-
sition number and shortest to longest mean dream re-
port length on each subgroup and the t-tests of sig-
nificance for each correlation coefficient are pre-
sented in Table III.’ The correlation coefficient for
the male subgroup 1, that group that went to position
10, and that for the female subgroup 8, that group
that went to position 15, are significant at the 1%
level. The correlation coefficient for the male sub-
group 6, that group that went to position 29, and that
for the female subgroup 9, that group that went to po-
sition 18, are significant at the 5% level. For the
females as a whole, there appears to be a trend (sta-
tistically significant for subgroups 8 and 9) for the
length of dream reports (in terms of the mean number
of words) to increase somewhat as the dream series pro-
gresses. For the males, there appears to be a trend
(statistically significant for subgroup l) for the
length of dream reports to increase as the dream series
progresses. However, it is to be noted that with each
successive male subgroup, extending to more positions,
this tendency progressively decreases until for sub-

groups 5 and 6 (statistically significant for subgroup

-1u-

6) the situation is reversed-~the tendency is now for
the length of dream reports to decrease as the dream
series progresses.

The results of the method of allowances and the
Friedman two-way analysis of variance by ranks were
not significant. It is then suggested that dream re-
port length is independent of the position at which
it occurs, that is, that short and long dream reports
appear at different positions in the series.

From Figures III to XV there appears to be an
oscillation phenomenon, that is if the number of words
goes up on one dream, a tendency to go down is observed
on the next, and vice versa. However, it seems that
such an apparent oscillation will occur by chance given
random sampling from a fixed distribution., To assess
whether there is a real oscillation effect, one would
have to derive the exact probabilities of oscillation
by chance, and compare our empirical findings. This
is somewhat complicated, and it does not seem likely to
yield significance (Karen*).

The results show a trend for females to report
more dreams, a trend for dream reports to increase in
length as the dream series progresses (with the ex-

ception of the two male subgroups), and a trend for

 

* personal communication

-15-

dream.report lengths to be independent of the particular

positions at which they occur.

-16-

 

m.ww méa mam mm gum onth

 

 

 

 

 

 

 

 

 

N..:. m .wN emfi om m: ode-2.
m .Hw mama omm 2.. 01v #308
ohm HQ; do 0H H5 @3th
v.2. m.w~ mm mH be 3m:
mom: Zorn
m .ww N. .3 we a E. @3th
o.~m 0.3 3‘ m cm 032
$3 mafia-e.
«Km-w oda om Ni co mamas-oh
«inc bdm Tm ma we 332
N23 young?
quo m .0 pm N am 03th
méo >.wm 2 NH an 2.3-4
1:: 3mm
gnome-m 0H
oomb tomb uoz tomb Eman- mmod oouoofioo
uGeuaonH «sebum-m abacus-Z no?» Hon—5.92 Hongdz Hen-0H.

 

 

womb one oouoodoU moaaom Edema Ho :30??on .H 3an

17

Table 11. Male and Female Subgroups.

 

 

 

Subgroup Number of
Number Positions N
. *Male 1 10 124
2 12 105
3 17 51
4 20 29
5 24 17
6 29 10
M 7 10 215
8 15 163
9 18 107
10 23 54
11 27 29
12 31 16
13 33 11

 

18

Table III. Rank Correlation and t-Test Results for the Male and
Female Subgroups.

 

 

 

Subgroup
Number Correlation t-Test
34313 1 .35758- 4.21944**
2 .11888 1.21258
3 .09804 .73530
4 .07369 .38319
5 -.24435 -.95297
6 -.65419 -2.42050*
Female 7 . 12727 1. 87087
8 .24286 3.18147**
9 .24871 2.61146*
10 .08498 .61611
11 .03175 .16510
12 .15202 .57768
13 .22995 .71285

 

:1:
Significant at the 5% level.

>101:
Significant at the 1% level.

19

com-“mom
N om

k
1

he

4mm
fi"
nip—O
V‘
M
O
m
cur-Ln

.iLm
‘P
J-m
l-‘U
duo
p1
dpm

-3...

L

.2:

 

 

imm~

.. 0mg

.lmbH

Gama n 73 momaom EmoaQ 0132 ..H oudwfim

 

.rmoa

spzom ;o .quuan ueew

1U
r-|.l')

 

 

1b

 

mm

q-

season
om mN cm
H

AmHN n 73, moans-m Emman— 03th

mg

.= flew-m

v2

aim

L

1—
If}
[s
.-I

1m»

Iced

21

ImNH

noma

spxoM 3:0 .19qu neew

 

-mod

Gowfimom

OH
.

«pm

 

1| mNH

1r omH

 

1..th

3.2 n 7: - .oz asoembsm e112

echo?
mo
hunch-Z
Geog

A: baa-mam

22

mom-“mom

 

 

m- o-, m
n - bl
‘ a q
-emm-
mono?
HO
menu-H32
one:
..om-
[T mp H

32 u E N .02 asohms-am 33a .2 chem-m

23

c033 om

Cd

in
i1

.1“,

1:2:

ImNH

 

Liam-

Awm n 73 m .02 ASOHwnu-m 3.32 .> 0.3th

3:03
mo

non-b.92-

smog

24

_ 0
PM

cos-mom

ma OH
.

HP

an“)

an. u E e .02 9-833 33>-

160A

J

fiNH

 

16mg

.H> eta-mam

max—"93
HO
Hun—652
“adv:

25

em

sea-Eon

fl . S.

 

-1--

a o

P h
- .

 

A: u 73 m .02 Adoamnsw 2.32

.:> 93m:-

L

l

.mh

.2:

1mm.”

 

romH

mono?
mo
gong-oz
smog

26

eon-tom

 

 

 

mm
. MN WM mm OH m
1 _ . 1 1 1
1.0m
Jump
Loo-
mono?
- mo
{mug awash-Z
4- use:
llomfl
8- uzv .e 9on _ . :2.-
b Z now 3.34 H; enema
tmz

 

27

cowuwmom

0H m

mono?
iromd no

use;

 

L

.mhd

33 u E 1. .oz aseembsm Beach .x- 939m

Hench-z

28

don-30m
ma 3 m

‘-

 

mono?
1.. mmd mo

Geog

 

om.“

AM: n 73 w .02 adonmndm 0?th .K 0.3-th

Hench-Z.

Z9

cos-eon

mg mg A: m

P _ .
a . J

-b

. mono?
+£- so
Hooch-Z

use:

 

rom-

:o- u 7: e 62 defies-em flea-eh .Hx eta-mam

30

MN

cos-eon

 

 

om ma 2 m
a i . 1 .
.. on:
mono?
.: mNH mo
Henna-Z
use:
Some

Gem n 73 o“ .02 assume-W madam-h

.mx charm

31

more ad

 

 

 

eN mm em m- o- m
T .. i u u “
-.oe-
--m-
.
-rem-
am u E 2 .oz macaw-Em ewe-e...- .Ex obs-m

mono?
mo

32

Renata-Z.

use:

-bl-fi

N
ail-o
N

cos-mom

ma
-

 

 

7: u 7: NH .02 adoamndm odmfioh

.Zx charm

van

.2.

.2:

ImNA

 

com.“

33

mono?
mo
none-32
use;

eo-aeea

ON m- o-

u q

 

 

 

 

 

 

:- u E 2. .oz aeoemosm Sac-o...- .>x «Sm-m

.140

L

cod

Pmm-

.oma

 

SS

34

mouo>P

nonnnsa
adeeA

DISCUSSION

Authenticity is given to the dreams by the follow-
ing factors. First of all, each student had the Oppor-
tunity to choose a topic. It was emphasized that those
dreams must be real. The subjects were informed that
code numbers, and not their names, would.be used for,
identification purposes. It was felt that in this way
the subjects would offer less resistance in relating
all remembered elements of the dreamy—no matter how so-
cially unacceptable they may be. For the Spring, 1962
and Fall, 1962 semesters, each student if, in choosing
a dream tOpic, had the further choice of writing a fake
or real dream series. Naturally, the real dream.pre-
sented in the dream series is the recollection and not
the complete reproduction of the actual dream.

Ramsey (7), in reporting a study by Manaceine, who
found that the number of dream reports is less for those
subjects with lower intellectual capacity, suggests that
individuals with greater intellectual capacity would re-
flect this in the richness and variety of the content in
their dream reports. Bearing this in mind, the range of
intellectual functioning of the subjects in this present
study was narrowed to approximately an above average

range; this range control was achieved by the selection

-35-

of college students.

An important factor concerning the dream report
is that of the limitations involved in translating the
predominant visual imagery to written statements. In
short, some subjects may be loquacious and give detailed
descriptions whereas others may give only vague and gen-
eral descriptions. This difficulty of transferring vis-
ual imagery to written statements, however, was present
'in all cases and would seem more or less a constant
throughout an individual's dream reports.

In accordance with Ramsey's (7) statement that
studies have demonstrated the trend for women to report
a higher frequency of dreaming than men, the present
study revealed a trend for females to report a higher
frequency of dream reports than males. Any attempt to
explain the reasons for this sex difference would be
conjecture.

It can be stated in general that the tendency (not
always statistically significant) was for the dream re-
ports to increase in length as the dream series pro—
gressed. This finding lends support to the learning
and practice hypothesis, and does not support a defen-
sive repression hypothesis. The two exceptions to this

finding (one only being statistically significant) are

-36-

from the male subgroups writing the longer dream series.
These two exceptions may be explained, perhaps, by

boredom or other outside interests and work.

-37-

SUMMARY

The object of this study was to determine in what
way dream.report length varies with its position in a
dream.series. It was hypothesized that the length
would either increase, due to a learning effect, or
decrease, as the result of defensive repression.

Dream reports in the form of a dream series were
collected from 339 students (12h males and 215 females).

The females showed a trend to report more dreams,
and there was a trend for an increase in length of

dream.report as the dream series progressed.

-38-

1.

2.

3.

6.

7.

REFERENCES

Edwards, A. L. Statistical Methods for the Behav-
ioral Sciences. New York: Holt, Rinehart,
and Winston,l96l.

Hall, C. S. Diagnosing personality by the analysis
25 dreams. J. abnorm. soc. Psychol., 19u7,‘gg,
-79,

Hall, C. S. What people dream about. Scient. Amer-
ican, 1951, 18k, 60-63.

Hall, C. S. Current trends in research on dreams.
In Brower, D. and Abt, L. E. (Eds.) Progress in
Clinical Psychology. New York: Grune and
Stratton, 1956, g, 239-257.

Hall, C. S. Institute of Dream Research. Technical
Manuals No. 1, 2, 3, h, 5, 1962.

Mosteller, F. and Bush, R. R. Selected quantitative
techniques. In Lindzey, G. (Ed.) Handbook of
Social Psycholggy. Reading, Massachusetts:
Addison-Wesley Publishing-Company, Inc., 19Sh,

1. 3011-307.

Ramsey, G. V. Studies of dreaming. Psychol.ABull.,
1953: 521 1432.1155-

Siegel, S. Nonparametric statistics for the Behav-
ioral SEIences. New YorE: McGraw-Hill Book
Company, Inc., 1956, 166-173. '

 

-39-

AP
PENDIX

-uo-

Dream.Diary: All you need do for this assignment
is keep a record of the dreams that you have or the '
dreams that you recall during this term. You will be
expected each time you awaken from sleep, either during
the night or in the morning, to make a reasonable effort
to recall what you might have been dreaming beforehand.

 

Several observers have reported that one can more
often recall a dream or more numerous details of a dream
if, when, waking up, one lies perfectly still, not mov-
ing and keeping one's eyes closed. That would seem a
good procedure to follow each time you awaken from sleep.
Try first to recall whether you were or were not dream-
ing; and then, if it is your impression that you were
dreaming, attempt to capture the details of the dream.
When you have done this, write out your impressions as
fully and completely as possible. It will help to keep
pad and pencil close by your bed. In writing out a dream,
do not modify the details, do not censor, do not strug-
gle after prose perfection. Write out your impressions
as they occur to you, whether the sequence be logical or
not. You can transcribe these later in the day but here
too do not modify your original script. Faulty syntax
will not be met with scorn. Also, should you suddenly
recall a dream some time during the day, record this as
welle '

Please number each dream sequentially and include
the date of the night on which the dream occurred. Since
this assignment requires little active work, everyone who
undertakes it will be expected to write out all the dreams
they remember, beginning this night and ending two nights
before the due date.

Fake "Dream? Diagy: This is a somewhat different
term paper option. Those who choose this may simply
fake a dream diary. In making up such a series of fic-
titious dreams, only one consideration is of importance:
attempt to make your series resemble as nearly as pos-
sible what you think a "real" dream diary would contain--
but do not consult any of your classmates who are actually
doing a "real" dream diary. Construct your fake diary on
your own, guiding yourself throughout by what you think
happens in dreams.

 

In handing in your paper, follow the same format as
that for a "real" dream diary, numbering each fake dream

~L11-

report sequentially and including a date (preferably
when written). It would be best by far not to leave
this task for the last night: one cannot do a good
job in one sitting. Do it gradually over the term.
The expected number of fake dream reports per term
paper will be announced later.

-ug-

fr’nfi"!

. I! 'I' I 'V
«D‘QWV

1 835 e

:11sz a:

'1 '23-‘25

r “‘5': we

 

HIIIHHINIH

||I|
l
l
Ill-III
I'll!
Ill
I
ll

1
3