Domhoff, G. William (1996). Finding Meaning In Dreams: A Quantitative Approach. New York: Plenum.   Use these links to quickly navigate through the book:
 • Introduction
 • Chapter 1: The Scientific Study of Dream Content
 • Chapter 2: The Hall/Van De Castle System
 • Chapter 3: The Quality of the Data
 • Chapter 4: Normative Findings
 • Chapter 5: Age Differences in Dream Reports
 • Chapter 6: Cross-Cultural Studies of Dream Content
 • Chapter 7: Consistency and Change in Long Dream Series
 • Chapter 8: The Continuity Between Dreams and Waking Life
 • Chapter 9: The Repetition Dimension
 • References

Chapter 8: The Continuity Between Dreams and Waking Life in Individuals and Groups


In this chapter we will attempt to show there is meaning in dreams by exploring the relationship between our dream content categories and waking thought and behavior. This is what people usually have in mind when they ask if dreams have meaning. They want to know if dreams reveal anything about a person's psychological make-up, especially anything not previously known or understood. As should be clear by now, we do think dreams have meaning in this sense of the term. We think dreams reveal people's conceptions, concerns and interests. It is important to understand people's conceptions because they are one basis for their actions in the world. It is important to understand people's preoccupations because they determine where people will expend most of their energies (Hall, 1947, 1953a).

We believe the findings to be presented in this chapter demonstrate a continuity between dreams and waking life: the concerns people express in their dreams are the concerns they have in waking life. What they dream about is also what they think about or do when they are awake. We call this claim the "continuity hypothesis." Just as the "consistency hypothesis" characterizes our major findings on people's dream lives over the space of time, so too the "continuity" hypothesis summarizes what we have learned about the relationship between the dreaming and waking minds. Taken together, the consistency and continuity hypotheses are our major evidence for the overall "meaning" of dreams in the sense of (1) regularity and (2) correspondence with other psychological variables.

Although the continuity hypothesis is a straightforward and simple one, the actual nature of the continuity is somewhat complicated. The continuity usually is with both thought and behavior, but sometimes it is only with waking thought. For example, people who have highly aggressive dreams are not always aggressive people in waking life, but they admit to many aggressive thoughts and fantasies during the day. Similarly, people who have frequent sex dreams are not always sexually active in reality, but they entertain the same thoughts in waking life and sometimes practice frequent masturbation to the accompaniment of their sexual fantasies. These two examples are not happenstance ones. It is our hypothesis, based on a limited number of cases, that sexual and aggressive preoccupations in dreams are the ones most frequently manifested only in waking thought. Our best case example of this point will be presented later in the chapter.

Most of our evidence for the continuity hypothesis comes from "testimony" by the dreamer or people who know the dreamer. Sometimes this testimony is augmented by what might be considered more "objective" or "physical" evidence, as when a child molester is jailed for his behavior or mental patients are hospitalized. For the most part, though, our evidence comes down to testimony, and for some of our most satisfying analyses we have relied exclusively upon the testimony of the dreamers.

On a few questions, however, the testimony of the dreamer, or at least the initial testimony, has not supported our inferences. What are we to make of such instances? If the person has corroborated our inferences, say, 36 out of 40 times, on what basis do we make rejections of contrary testimony? As we all know, testimony is not always accurate, but that does not resolve our problem. We do not propose to deal with this issue until the conclusion of the chapter, but we did want to alert readers to it before it arises.

The search for relationships between dream content and waking behavior has been sidetracked by a repeated--and unexpected--finding: there is little or no relationship between dream content and standard projective techniques like the Thematic Apperception Test (see Hall, 1956, for a review of most of these studies). For example, Mary Osterberg (1951) coded dream series and TAT stories obtained from ten males and ten females for frequency, intensity, and object of aggression using precursors of the Hall/Van de Castle scales. There was no relationship between dream reports and stories for any of these categories. Gordon's (1953) comparison of dream reports and TAT stories found dream reports contained more aggression, tension, fear, passivity, and inadequacy. Similar negative findings with the TAT are reported by Leman (1967), Ben-Horin (1967), and Zepelin (1980-81, 1981). The results with the Rorschach are not any better (e.g., Bolgar, 1954; Eagle, 1964).

Nor is there much evidence for a strong correlation between dream content and objective tests like the MMPI, California Psychological Inventory (CPI), and Edwards Personal Preference Schedule (e.g., Winget and Kramer, 1979, for a review of such studies through the early 1970s). We will focus primarily on those few studies reporting some positive results, all of which support the continuity hypothesis.

Rychlak and Brams (1963) collected two or more dreams from 15 males and 15 females in a summer school session ranging in age from 19 to 52. They scored the dreams on four thematic dimensions and several of the content categories in their scoring system, which overlaps greatly with the Hall/Van de Castle system, as shown briefly in chapter 2. They then compared their findings with many dimensions of the MMPI and the Edwards Personal Preference Schedule for the same subjects. The scores on the dream dimensions and the personality tests were dichotomized at the median, which created a series of 2 x 2 tables that were tested for statistical significance with chi square. Results at the .01, .05, and .10 levels of significance were reported. Given the number of comparisons they made, several significance differences would be expected by chance alone. No measure of effect size is reported.

There were very few significant relationships between their dream content categories and the tests. However, the findings on the thematic scoring were generally supportive of continuity between dream content and personality dimensions. For example, people with positive interactions as a main theme of their dreams tended to show more social confidence and social control on the personality tests, and people who were high on "reward" themes involving striving and recognition tended to score higher on scales showing concern with social status. Themes of tension (anxiety, frustration, and hostility) were more often present for those who scored higher on hostility and psychopathology scales. Those who had the most unpleasant dream reports tended to score higher on MMPI scales for dominance and aggression (Rychlak and Brams, 1963:229-231). However, it is difficult to make very much out of this study due to the small number of subjects and dream reports, the subjectivity of thematic scoring, the large number of statistical tests that were run, and the lack of information on effect sizes.

Foulkes and Rechtschaffen (1964), as one part of a larger study of the effects of different films on the laboratory dream reports of 13 male and 11 female subjects, studied the correlations between 22 MMPI scales and the subjects' own ratings of their dreams for such dimensions as emotionality, unpleasantness, violence, dramatic quality, and degree of distortion. Once again, several significant differences could be expected by chance in such a large matrix. Perhaps the most interesting finding for our purposes, and one consistent with those of Rychlak and Brams (1963), was a positive correlation between various hostility scales derived from the MMPI and unpleasantness in dream content (Foulkes and Rechtschaffen, 1964:994). In a later study of dream reports from 14 boys ages 13-15 who each spent two nights in the sleep lab, Foulkes et al. (1969) found that the degree of active participation in the dream action correlated positively with dominance on the CPI. They also found a positive correlation between CPI aggression scores and physical aggression in dream reports. Once again, these findings are compatible with our continuity hypothesis.

Further tentative support for the continuity hypothesis comes from the findings on two nights of laboratory dream reports from 24 subjects divided into high and low expressors of impulses in waking life, as determined by an MMPI subscale. Those who were high on impulse expression in waking life also tended to express more sexuality and hostility in dream reports (Ben-Horin, 1967). Van de Castle (1968; with Holloway, 1970) reports positive correlations between some Hall/Van de Castle categories and the MMPI profiles of several types of psychiatric patients, but the studies do not involve large numbers of subjects and no details are given. On the other hand, Fletcher (1970), in a study of 529 dream reports, found no relationship between measures of aggression in dream content and various MMPI scales.

Looking at the overall findings from this handful of studies with objective tests, they seem to hold out some possibility for the use of such tests in studies of dream content. However, the results have to be treated with caution due to relatively small sample sizes and the lack of replication studies. Indeed, the fact that we could find no published studies in the Psychological Abstracts between 1973 and 1992 attempting to link specific categories of dream content to dimensions of objective personality tests may be a telling comment on how little potential investigators see in such studies after reviewing the literature.

The negative findings with projective techniques and the tentative findings in the few studies with objective personality tests leave personality researchers with no established ways to link dreams with waking life. This is yet another reason why there is so little study of dream content in scientific psychology. So, if there is going to be any advance in finding connections between dream content and waking life, new directions will have to be taken. That is why we have relied on testimony in our own efforts. It is also why we think analysts of dream content may have to create their own objective tests, based on their findings within dream reports, to find the linkages to waking life. The old research strategy started with the existing tests and tried to determine if their dimensions had any connections to dreams. This strategy in effect tested the assumption that the various dimensions of waking personality are reflected in dream content and found little evidence for it.

Similar conclusions can be drawn from a study comparing themes in daydreams to several personality measures (Gold and Reilly, 1985/86). Thirty male and 32 female college students took an objective personality test, listed their most important current concerns, and then kept a diary of their daydreams. No correlations were found between any of the personality dimensions and daydream themes, but a little over half of the daydreams related to the students' five most important current concerns, clear evidence for the continuity hypothesis.

Due to the meager findings with personality tests for both dream content and daydream themes, we have adopted a new working assumption: dream content may not be about "personality" in the usual sense of the term. Instead, dream content may provide us with different information about people than most personality tests do. Since we have found dream content reveals conceptions and concerns, that should be our starting point in developing or selecting objective tests for the study of the correspondence between dream content and waking behavior.

Each of the individual case studies in this chapter is a blind analysis in the sense that we did not have any knowledge or expectations about the dreamer. We did have some general impressions of our two most famous subjects, to be introduced shortly, but we had no idea of what inferences might arise from a quantitative content analysis of their dream reports. The individual cases we have brought forward in this chapter are the best we have out of an only slightly larger pool of possibilities. They are the ones on which the most quantitative work has been done and the most information is available. We regret, for example, that we could not report more about the lives of Jason, Dorothea, and Marie, for their long dream series and consistent patterns would have made them excellent subjects for this chapter.

This chapter consists of eight case studies and a discussion of findings with various kinds of clinical patients ranging from anxiety neurotics to schizophrenics. Although there are some promising leads in the group studies of patient populations, for now we put our greatest stock in the individual cases. We think they are the most fruitful direction for new research in the next several years.

We have decided there is no more appropriate place to start the chapter than with our case study comparing the dream reports of the two most famous dream theorists of the 20th century. Their dream series are not as long as we would like them to be, and we do not have detailed information on all aspects of their personal lives, but we can think of no greater opportunity to test the power of our approach than the dreams of Sigmund Freud and Carl Jung. To find patterns in the dreams of Freud and Jung that no one has noticed before would be a strong recommendation for quantitative content analysis.

The Dreams of Freud and Jung

The material for this section consists of 28 dreams reported by Freud in two of his books, The Interpretation of Dreams (1900) and On Dreams (1901), and 31 dreams reported by Jung in his autobiographical Memories, Dreams, Reflections (1963). The two series present us with a certain handicap because they are so short, although we have found as few as 20 dream reports may on occasion reveal significant aspects of a dreamer's personality (Hall, 1947).

Moreover, some differential biases may have operated in the selection of the dreams reported by the two men. For example, Freud may have selected dreams favorable to his theory, and Jung may have selected dreams favorable to his theory. It does appear evident that the reasons for relating their dreams in the first place were quite different. Freud used his dream reports to illustrate various aspects of his dream theory; Jung's purpose was the more personal one of illuminating the nature of his inner life and development. This difference in purpose is evidenced by the books in which they appeared. Freud's dreams are published in scientific treatises; Jung's dreams are reported in his autobiography. There is also the possibility that these dreams are highly memorable or vivid ones not typical for either dreamer, although our findings cast doubt on this concern.

In spite of the problems of brevity and possible selection biases, we expected our method to reveal differences between the two dream series congruent with the lives of the two theorists. Each dream was typed on a 5 x 8 card. Freud's 28 dreams and Jung's 31 dreams were shuffled together before Hall coded them. Freud and Jung were then compared with each other and American male dreamers.


Freud and Jung are like each other and the norms in most of the basic character categories, as can be seen in Table 8.1. None of the differences is statistically significant. However, there are some differences that relate to what we know about the two men. Freud has more characters in his dream reports than Jung does, 85 versus 70, although Jung reports more and longer dreams. The number of lines per character is 3.4 for Freud and 6.5 for Jung. This "density coefficient" for people in dream reports is much higher for Freud than it is for Jung. Jung's dream narratives are filled with descriptions of scenery, architecture, and objects rather than with people. This difference appears to be compatible with what is known about the two men. Freud was a highly sociable person. He had many close friends and disciples with whom he interacted on a very personal basis. Jung was more solitary and kept would-be disciples at a distance. He spent much time in scholarly pursuits poring over old manuscripts, and he was a lover of nature. Jung said of himself, "Today as then [meaning in childhood] I am a solitary" (1963:41-42; cf. 32, 356).

Table 8.1 goes here

This difference in sociability between the two men is congruent with other evidence from their dreams. Jung's interest in nature is compatible with the fact that he had a higher animal percent than Freud. He writes in Memories, Dreams, Reflections that "I loved all warm-blooded animals.... Animals were dear and faithful, unchanging and trustworthy. People I now distrusted more than ever" (1963:67). More mystical, fictional, and historical figures turn up in Jung's dreams than in Freud's. This suggests Jung lived more in a world of the past whereas Freud lived more in a world of the present. Indeed, Jung wrote that for years he felt more closely attuned to the past, especially to the Middle Ages and the 18th century, than to the present (1963:34-35, 87).

Jung dreams more about members of his family whereas Freud dreams more about friends and acquaintances. This implies that Jung's sociability expressed itself within his immediate family, and Freud's social life was centered more outside of the family. There is evidence in Ernest Jones's (1953-55-57) three-volume biography of Freud that he expended an enormous amount of time and energy in keeping up a large network of friendships and in trying to maintain cohesion within the psychoanalytic movement.


We turn next to the findings on objects because they have some connections with the inferences we made about Freud and Jung's interests based on the character analysis. There are many more objects in Jung's dream reports than in Freud's, 297 versus 196, which suggests again that Jung was more object-oriented and Freud was more person-oriented. Further support is provided for this statement by the kinds of objects each dreamed about. Jung dreamed more about houses, buildings, and architectural details, especially windows, doors, and walls, and more about nature and landscape than either Freud or the norm group. Freud, on the other hand, dreamed much more about parts of the body, particularly those located in the head, than either Jung or the norm group.

Jung's dreams contain no references to money, whereas Freud's incidence of dreams of money is the same as the norm group. Both men are low in the implements category, especially references to weapons and recreational equipment, probably not surprising considering that both of them were intellectuals and scholars.

Aggression and Friendliness

The findings so far have been interesting, but hardly earthshaking. They take on more significance from this point on. Freud and Jung have some very revealing differences between each other and the norms in their patterns of aggression and friendliness. However, these differences do not manifest themselves in our most general indicators, the A/C and F/C indexes. There are 16 aggressive and 16 friendly interactions in Freud's dream series; Jung's has 14 aggressions and 11 friendly interactions. Their A/C and F/C codes are much the same, and they are in close accord with our findings for American males between the ages of 30 and 80 (Hall and Domhoff, 1963b, 1964). Nor do they differ from each other or the norms on their degree of involvement in aggressive and friendly interactions, or in the percentage of time they are victims of aggression.

With regard to the role of befriender and befriended, however, the two men are poles apart. In every instance where Jung is involved in a friendly interaction, he initiates the friendliness. Freud, on the other hand, more often plays the role of being the recipient of friendliness (8 out of 11 times). The norms fall midway between the figures for Freud and for Jung. This finding suggests that Freud would expect people to come to him and be sensitive to rejection if they didn't. We do not have a great deal of behavioral evidence on this point, but we do know that Freud was deeply hurt because Jung did not make an effort to visit him when Freud made a trip to Switzerland in 1912, shortly before their close friendship of six years dissolved (Jones, 1955:144).

By far the most striking and informative finding in comparing the two series concerns the large differences in the patterns of Freud and Jung's aggressive and friendly interactions with male and female characters. As we have said so often, the typical male has more aggressive interactions with males than with females, and more friendly interactions with females than with males. Jung's aggressive and friendly encounters with males and females are fairly typical. He has an aggressive interaction with about one in every four male characters in his dreams, and none at all with females. Jung has about an equal number of friendly encounters with males and females, which deviates slightly from the norms and makes Jung more friendly with everyone. In Freud's dreams, the typical pattern is reversed. He has an aggressive encounter with one out of every four female characters, and almost none with males. On the other hand, he has many more friendly interactions with males than with females. His pattern is similar to what we reported for Jason's long dream series in chapter 7.

There is evidence that the waking preferences of the two men are compatible with these findings. Jung (1963:48-55, 73, 93-96) felt much closer to his mother than his father, and the evidence for his conflicts with his father is abundant in the early chapters of his autobiography. Jung is known to have had many affairs with women (Wehr, 1985). He did not spend much time in male social groups. On the other hand, many people have concluded after reading Freud that he was hostile toward women (a summary of the evidence can be found in Chodorow, 1978: chap. 9). Jones (1955:421) says Freud found women "enigmatic." He also writes that Freud was "quite peculiarly monogamous" and that "the more passionate side of married life subsided with him earlier than it does with many men" (Jones, 1955:386, 421). Jones says "it might perhaps be fair to describe his view of the female sex as having as their main function to be ministering angels to the needs and comforts of men" (1955:421).

As regards Freud's feelings for men, we know he had a very intense relationship with his fellow physician, Wilhelm Fliess, in the 1890s. Freud spoke of overcoming his homosexuality and said alternations of love and hate affected his relationships with men (Jones, 1955:420). Freud wrote to his friend and colleague, Max Eitingon, that "the affection of a group of courageous young men is the most precious gift that psychoanalysis has bestowed upon me" (Jones, 1955:419).

We will return to these findings when we comment on the ending of Freud and Jung's friendship.

Success and Failure

The experience of success and failure in their dreams also distinguishes Freud and Jung. Most males, as shown in chapter 4 have an equal amount of success and failure in their dreams. So does Jung, but Freud has much more success than failure. In fact, he only fails once and succeeds six times. This suggests that Freud was more strongly motivated to succeed than Jung was. Jones' (1955:415) judgment that fame meant very little to Freud does not square with the pretty obvious fact that Freud aspired to greatness. Jung, on the other hand, although he may have had the same aspiration, did not do many of those things that would help him to achieve fame. Unlike Freud, he did not found an international organization with its own journals and publishing house. He did not establish a chain of institutes throughout the world to promote his ideas. He did not encourage disciples. He preferred his stone tower to the bustle of the scientific marketplace. He did not seek worldly success, although he did not refuse it when it knocked on his door.

Good Fortune and Misfortune

Good fortune in dreams, as we know, is a rare quantity; misfortune is commonplace. Freud and Jung, true to this cross-cultural pattern, have more misfortune than good fortune in their dreams. In fact, Freud has no good fortune at all, whereas Jung has more good fortune than is to be expected. Freud's large amount of success relative to failure, coupled with a lack of good fortune, suggests he pictured himself as succeeding through his own efforts and not as a result of luck. Jung was more likely to view the world, at least in his dreams, as a cornucopia showering benefits upon a person. Jung's autobiography suggests he was more fatalistic than Freud. He was inclined to let things happen to him, to let his life be lived rather than to live it. He wrote (1963:358) that almost everything in his life "developed naturally and by destiny." Jung (1963:356-357) felt his life was ruled by forces over which he had no control, and did not completely understand, although he spent much of his adult life trying to understand them. Freud was more rationalistic. By exercising reason, he felt that people could master the world.

Summary and Speculations

We think the several differences between the dream series of Freud and Jung are congruent with their behavior in waking life. Freud was a striving person who liked the company of other men and held to a theory that has a negative view of women in general. Jung, on the other hand, was more solitary and interested in nature, and in his passionate relationships with women.

We also think the dream evidence may shed light on the break-up of their close friendship. Freud sought intense relationships with other men. He wanted to bring men together into a successful psychoanalytic movement. Given Jung's more solitary orientation, his preference for nature and architecture, and his enjoyment of the company of women companions, it is not difficult to imagine that he would be put off by what he may have seen as Freud's adhesiveness. Moreover, Jung did not share Freud's driving ambition.

True enough, Freud and Jung came to have strong theoretical differences, but such differences do not necessarily mean the end of friendship, or of at least cordial relationships. By way of contrast, two other Swiss intellectuals who were attracted to psychoanalysis remained warm friends with Freud even when they developed intellectual differences. One, Oscar Pfister, was a Protestant minister, an unlikely friend for a secular member of the Jewish community (Meng and Freud, 1963, recount Pfister's friendship with Freud). The other, Ludwig Binswanger, who broke with Freud to introduce existentialism and phenomenology into psychiatry, corresponded with Freud for the rest of Freud's life. Binswanger subtitled his book on their relationship Reminiscences of a Friendship (1957).

We are not saying theoretical differences are unimportant. We also recognize that Freud was deeply disappointed by Jung's theoretical disagreements because he thought of Jung as the "son" he had "anointed" to take over the helm of the psychoanalytic movement (Jung, 1963:361). Jung also was a far stronger personality than Pfister or Binswanger. Nevertheless, we think the patterns we have found in the dreams of Freud and Jung are of more than theoretical interest. These patterns seem to relate well to their waking concerns, and that is why we have speculated on their possible usefulness in understanding the relationship between the two men.

More generally, we think this first case analysis shows quantitative content analysis can extract "meaning" from dream reports. The fact that there were relatively few dream reports and a lack of deep information on the two men makes the case less conclusive than it might be, but perhaps more impressive for the same reasons.

The Dream Life and Literature of Franz Kafka

Franz Kafka (1883-1924), one of the greatest writers of the 20th century, wrote down 37 of his dreams in diaries and letters between 1910 and 1923. Hall and Richard Lind (1970) did a quantitative content analysis of those dreams and then made predictions about his concerns and interests before reading other material in the diaries or any of the biographies and reminiscences on Kafka. The dream reports can be found in Appendix B of Hall and Lind (1970:97-124).

It is not known why Kafka wrote down the dreams appearing in his diaries. He rarely commented on them and never attempted any interpretations. It is known, however, that he was critical of psychology in general and psychoanalysis in particular. In the case of the six dreams he included in letters to Milena Jesenska, a woman with whom he was in love for many years, it seems clear he was using the dreams as a means of communicating his mixed feelings about their relationship.

Despite the fact that there are only 37 dream reports, the quantitative content analysis yields a very good portrait of Kafka's personality because there is so much consistency from category to category. This case is also interesting because it shows how useful mundane categories such as "activities" and "objects" can be.


There were 135 characters in Kafka's 37 dream reports. This is a mean of 3.6 per dream report, significantly higher than the mean of 2.4 characters for the normative group. This finding suggests a first inference about Kafka: perhaps he is more preoccupied with people than the typical person. The characters were divided into three usual categories: individuals, groups, and animals. There were no differences from the norms. Ninety-five of the 135 characters in Kafka's dream reports could be identified as individual males and groups of males or individual females and groups of females. Of these 95, 63 were male and 32 female, yielding a male/female percent identical with the male norms. The 132 human characters were divided into those who were familiar to Kafka (family, relatives, friends, famous people) and those who were unfamiliar (strangers). Once again, the findings were very similar to the norms.

Aggression and Friendliness

Kafka's aggressive and friendly interactions are also fairly typical, but there are also some suggestive individual uniquenesses. At least one aggression occurred in 15 of Kafka's 37 dream reports. This percentage of 41 is not significantly different from the percentage of 47 for the normative sample. However, when the 24 separate acts of aggression are divided by the number of characters, the resulting A/C score of .18 is significantly different from the normative proportion of 34 at less than the .001 level of confidence (h = .37). The physical aggression percent is 38 for Kafka and 50 for the normative group, but that difference is not statistically significant. Kafka is less likely to be involved in the aggressions in his dream reports than the normative group (.62 vs. .80, h = .40).

However, when Kafka did take part in an aggression, he was far more likely to be the aggressor than the victim, opposite of the normative group. His A/C index with males is significantly lower than the norms, but with females it is similar to the normative group. Generally speaking, then, there is less aggression in Kafka's dream reports due to an unusually low number of aggressive encounters with other males. There is also a strong tendency to witness aggression rather than participate in it.

There are 24 friendly interactions in Kafka's dreams, exactly equal to the number of aggressive interactions. Although aggressive interactions slightly outnumber friendly interactions in the normative sample, Kafka's aggression/friendliness percent is not significantly different. Similarly, when friendly interactions are divided by the number of characters, the F/C score of .18 is not significantly different from the finding of .21 for the normative group. Nor were there any differences in witnessed vs. involved friendliness or in dreamer as befriender vs. dreamer as befriended. His distribution of friendly interactions with males and females yielded no statistically significant differences.

When we look at Kafka's patterns of friendliness and aggression with males and females, which was so revealing in the case of Freud, another type of interesting pattern emerges. Unlike the typical male, he has equal amounts of friendliness and aggression with both genders due to less aggression with males and less friendliness with females (see Table 8.2).

Table 8.2 goes here

Activities and Objects

The activities in Kafka's dream reports are similar to the norms in all categories. The objects categories in Kafka's dream reports are comparable to the norms except in one category: body parts. Kafka is three times more likely than the average American male dreamer to mention one or another part of the body, which is an h of .52 when the two percentages are compared. We will return to this interesting finding on body parts after summarizing the findings for other coding categories.

Other Coding Categories

Kafka proves to be very typical in most other coding categories. There are no differences on misfortunes, good fortunes, food and eating elements, or emotions. Successes and failures occur with about equal frequency. The only difference in any of these categories is that Kafka witnessed others having success more often than the norms (.44 for Kafka, .11 for the normative group, a difference significant at the .01 level of confidence, h = .78). This witnessing of successes fits with his pattern for witnessing aggression.

A Portrait of Kafka

What kind of picture of Kafka emerges from this analysis? First, he is unusually concerned with people. Second, he is not as aggressive as other males. Third, he seems unusually preoccupied with his body. Finally, his lower level of friendliness with female characters suggests his feelings toward women are more mixed than is typical for most men.

This portrait of Kafka accords well with what emerges from biographies. He was a quiet, passive person who preferred watching to doing. He worked as a clerk in a quasi-governmental insurance association and never moved out of his parents' home. He never finished his three novels. He studied people intently and his favorite activities were reading, attending the theater, sightseeing, and walking (Hall and Lind, 1970:46-48). He was not involved in any active sports or hobbies. He had deep concern about his body and the bodies of others. Although he was handsome and six feet tall, he thought of himself as skinny and compared himself unfavorably with his more muscular father. There are constant negative comments about his body in his diaries. He was repelled by bodily imperfections in others. He was interested in the nudist movement and in natural health practices (Hall and Lind, 1970:39-43).

Kafka had several close relationships with women as a young adult, but they were characterized by hesitation and indecision. He never married. He was engaged to one woman, then the engagement was broken. They became engaged again, and once again decided not to marry. He had two other similar relations before developing a satisfying association with a woman toward the end of his life, when he was dying of tuberculosis (Hall and Lind, 1970:51, 75-76).

Hall and Lind also applied the Hall/Van de Castle coding system to Kafka's three novels (Amerika, The Trial, and The Castle). They found few or no connections to his dream reports. They said it was "plainly evident that the interactions between the heroes and other characters in the three novels are much more complex than Kafka's social interactions in his dreams" (Hall and Lind, 1970:67). Kafka's understanding of the human condition was not limited to his own personal preoccupations. His interest in observing others served him well.

We conclude that there is a connection between Kafka's dreams and his waking life, but not between his dreams and his literature. More generally, we suspect that the alleged close connection between dreams and artistic productions is a cultural myth. The "dreams" referred to in such claims are more likely daydreams, reveries, or hallucinations in most instances.

The Dreams of a Child Molester

Between September 15, 1963, and February 8, 1967, a convicted child molester whom we will call Norman wrote down 1,368 dream reports. He initiated the project for his own private reasons. He was in a mental institution 80% of the time that he was keeping his dream diary, often writing dreams on paper bags or laundry lists (Bell and Hall, 1971:v, 3). Norman was 34 years of age when he undertook this project and 38 when he ended it.

Alan Bell, a clinical psychologist doing an internship at the hospital where Norman was held, interviewed Norman approximately 20 times between October, 1966, and April, 1967. During these interviews he learned of Norman's ongoing dream record. Bell later wrote to Hall asking him if he would be interested in analyzing Norman's dream series. Hall undertook the task without knowing anything except the age and sex of the dreamer. Although Hall knew nothing about Norman when he began the analysis, it did not take long to figure out he probably was a child molester because of frequent mentions of looking at little girls' genitals or having sexual thoughts when in the presence of children. Hall had previously analyzed the dream series of a 50-year-old transvestite who cross-dressed in his dreams as well as the dream series of male homosexuals, so he knew the continuity hypothesis had been supported for sexual behavior in at least some cases.

In addition, it also was clear just from reading so many dream reports that Norman lived at home with his mother and sister when he was not institutionalized, had never married, enjoyed reading and swimming, and worked as a helper in printing shops. Although such information is evidence for the continuity hypothesis, it is not the kind of information sought by those who study dream content. In this particular case, the interesting question became a search for why Norman developed into a child molester.

Once the dream reports were coded, Hall wrote a character sketch of the dreamer. Bell then compared this sketch with his clinical information, test scores, and reports furnished by five different institutions where Norman had been held for seven years between the ages of 20 and 34 for child molestation (Bell and Hall, 1971:6-9).


There are many unusual features in the character patterns in Norman's dream reports. His mother appears four times more often than would be expected from the norms. His sister, who is three years younger, appears ten times more often than expected. There is not a single appearance of his father in the 1,368 dream narratives. Given these family patterns, it is not surprising that his male/female percent is extremely low at 50/50 vs. 67/33 for the norms (h = .35). Beyond his female family members, Norman dreams primarily of unknown males and unknown females. In particular, there is an extremely low incidence of female friends and acquaintances. As for the males who are known to him, they are usually his fellow inmates, not friends of long standing. There is also a large incidence of minors (characters under age 18) in Norman's dream reports. His friends percent (known characters divided by all human characters) is only 9. This figure is far below our normative figure of 31 for males (h = .57). Combined with the very low male/ female percent, it suggests that Norman is a very atypical dreamer in terms of characters.

Aggression and Friendliness

There is nothing unusual in Norman's pattern of aggression and friendliness except for the important point that he is a little less aggressive than other males (Table 8.3). Norman is also well below the norms on his physical aggression percent (27 vs. 50, h = .48). He shows a certain passivity in that he is more likely to be the victim of an aggression and the recipient of a friendly act (the befriended). He has many friendly and few aggressive interactions with minors. If there is a saving grace in Norman's dream reports, it is in the fact that he is not a very aggressive person.

Table 8.3 goes here

Sexual Activity

The typical male dreamer has 12 "sex dreams" per 100 dream reports. Norman has 13 per 100, not a statistically significant difference. What distinguishes Norman from most other males is the variety of characters with whom he is erotically involved. Whereas heterosexual adult males have erotic encounters almost exclusively with peer females, Norman is involved in or witnesses sexuality in his dreams with males and females, and adults and minors, as can be seen in Table 8.4.

Table 8.4 goes here

Moreover, Norman is unusual in the range of sexual acts in which he engages in his dreams. He is both active and passive in his sexual encounters, as well as oral, anal and genital. Most of all, he has sexual fantasies more frequently and sexual intercourse less frequently in his dream reports than the male norms, as shown in Table 8.5.

Table 8.5 goes here

Preoccupation with the Body

The frequency of Norman's references to body parts in his dream reports does not differ from the norms, but the parts of the body he dreams about are different than the typical male. In the normative group, references to the head and extremities exceed references to the torso, anatomy, and sexual organs. That is, the "torso/anatomy percent" mentioned in chapter 2 as a possible indicator of psychopathology is expected to be 31, but in Norman's case it is 54 (h = .47). This finding fits with his sexual and gender preoccupations.

Hall's Portrait of Norman

Hall thought the most extraordinary feature of Norman's personality was his pervasive emotional immaturity. He is dependent on his mother and sister. He has no friends. He prefers to be around children. His sexual desires are unfocused, perverse, and mostly in the realm of fantasy. He suffers from gender confusion. Hall believed the absence of the father from the dream series might be its most tell-tale sign. He speculated that the father was either absent or else had used Norman sexually (Bell and Hall, 1971:84). The evidence further suggested that Norman was ambivalent toward female genitals, but had a strong feminine identification. The emphasis on fantasy in his sexual activities in dreams suggested that Norman was a compulsive masturbator. The frequent urination and defecation in the dream reports, something we have not mentioned until now, suggested Norman might still be a chronic bedwetter even as an adult.

Norman's Life History

Norman reported he was a victim of childhood sexual abuse. This turns out to be a fairly frequent pattern for child molesters. Beginning at around age 4, Norman was forced to perform fellatio on his father for three or four years. Norman said he enjoyed the experience at the time, but came to have severe guilt feelings about it later on. This testimony confirms Hall's main conjecture: Norman's problems involved his relationship with his father (Bell and Hall, 1971:21). When asked who were the most positive influences on his life, Norman named his mother, sister, aunt, and a male college professor who had helped him recently; three of the four are women. When asked who were the most negative influences on his life, he named four males.

Norman was a frail, unimposing, stooped-over person who lacked energy or aggression in waking life. He seemed to be making an enormous effort at self-control. Norman thought of himself as primarily heterosexual, but had never had sex with a woman. He said his masturbation fantasies were exclusively heterosexual, unlike his sexual fantasies in dreams. He had had sex with other males and at least once with an animal (Bell and Hall, 1971:25). Norman's main sexual outlet, however, was the same compulsive voyeurism present in his dream reports. He wrote that for "as far back as I can recall, I have had a morbid yet fascinating curiosity about the female genitals" (Bell and Hall, 1971:26). This is what led to his child molestation, which consisted primarily of voyeurism for several years, but included exposing himself at least once and fondling children several times. As time went on, he knew he was becoming more aggressive with the children, and it worried him (Bell and Hall, 1971:28).

Clinical observations and tests supported Hall's view of Norman as an immature and dependent person. They also supported the idea of gender confusion and showed his sexual life was primarily at the level of fantasy (Bell and Hall, 1971:88-90). However, there were some disagreements between Hall's predictions and the clinical/test findings. Norman was not a bedwetter and he successfully resisted masturbation for weeks at a time. He thought masturbation was wrong and often felt depressed afterwards (Bell and Hall, 1971:25, 94). He was able to resist his preoccupation with his body; he was very interested in spirituality. As Bell and Hall (1971:96) put it, "an analysis of the dream content reveals very little about his defensive maneuvers." For us, this is a very important conclusion.

Norman was released from the hospital in late spring, 1966, and was never hospitalized between that year and 1970, when the book on him went to press. He earned an A.A. degree at a local community college. He spent much of his time reading in his bedroom at home or in the school library. There is even a happy ending to the story as of 1992. Norman continued to stay out of institutions. He married a woman slightly older than he is in the mid-1970s. In 1992 he was in his mid-60s and living in a sunbelt state with his wife. According to Bell, Norman in his old age is a kind and gentle person (Alan Bell, personal communication, July 22, 1992).

Norman's long and unusual dream series is a graphic demonstration of the continuity between dreams and waking life. It also demonstrates the strengths and weaknesses of using dream content in personality research. Norman's dreams reveal his basic conceptions, concerns, interests, and preoccupations in dramatic fashion, but not all of his behaviors. It seems to us an important lesson that defenses may not be revealed by dreams. We are reinforced in our view that dreams have unique and useful things to tell us, but also in our view that we may not be able to find everything we want to know about personality in dreams. It is no wonder that the attempt to correlate dream content with standard personality tests has not been very successful.

The Dreams of a Neurotic Patient

In November, 1967, clinical psychologist C. Scott Moss sent Hall 58 dream reports from a 28-year-old married man who was in psychotherapy with Moss. Moss (1970: chap. 3) has provided an account of how the patient analyzed four of these dream narratives while under hypnosis. Knowing only that the dreamer was a male and in therapy, Hall did a quantitative content analysis of the dreams. Hall then rated the patient on 42 brief statements concerning his interests, preoccupations, and attitudes. These ratings were then compared with independent ratings by Moss on the same 42 items.

The first unusual feature of this man's dreams is similar to what was found for Norman in the previous section, a low number of friends and acquaintances. The characters in his dreams are either family members or strangers. We are therefore reinforced in our hypothesis that the friends percent may be a useful indicator in predicting psychopathology.

The dreamer is high in aggression with all classes of characters, including females, and low in friendliness with all classes of characters, again including females. He therefore has an atypical A/F square in several ways, but low friendliness with females may be significant in distinguishing males with mental health problems, as will be seen later in the chapter in a study of hospitalized male psychiatric patients. Table 8.6 presents these results.

Table 8.6 goes here

The problems shown in Table 8.14 are also reflected in his aggression/friendliness percent (all aggressions divided by aggressions plus friendliness): he is 81, the norms are 62 (h = .43). However, there are three pieces of better news in the areas of aggression and friendliness:

  1. He is not high on physical aggression percent; thus, he is an angry man, but probably not a physically violent one.
  2. His victimization percent is low, which means there is an assertive component to his anger.
  3. He has a normal rate of befriending others in the few instances of friendly interactions.

The dreamer has his share of misfortunes, but he is not above the norms on misfortune percent; this is a positive sign for him. The dreamer fails in nine of 15 strivings, which is slightly above the norms, but with an N too small for statistical significance (60% vs. 50%). Here the counterbalancing feature is that he is striving so often. Finally, this man's negative emotions percent is 97, as opposed to the normative percentage of 80 (h = .58). There is only one instance of happiness in 58 dream reports.

Hall interpreted these numbers to mean that the patient sees himself as an aggressive, angry person in a hostile and unfriendly world. He sees himself as ineffectual. He is anxious and unhappy. His strengths are that he is not a victim of aggression, and not merely a recipient of friendliness. He fails, but at least he tries. Here we would add one other finding not mentioned as yet: he is high on active activities in his dream reports.

In order to put these and other quantitative findings into a form that could be compared with Moss's clinical observations, Hall then developed 42 brief statements and rated the dreamer as high, medium, or low on each of them. He then asked Moss to make the same ratings. Hall and Moss agreed exactly in 28 of their 42 ratings. They disagreed by one step in nine cases, and they disagreed completely in five instances. The odds of this level of agreement occurring by chance are less than one in 1000 (.0007). From Hall's point of view, the most important items were the five where they disagreed completely. He thus wrote as follows to Moss for further information:

In his dreams, he had many references to games and sports so I gave him a high score. You said his interest in this area was low. I said he had a low interest in travel because there were few travel references in his dreams, and you said his interest was high. You said his relations with his mother were good, and I said they were poor. Actually, he only mentions his mother once in the 58 dreams, and that is in connection with her apartment. He said "my brother and I were trapped in my mother's apartment with a dozen women and were preparing to fight off an Indian attack." I figured that anyone who had such poor relations with women, including his wife, must also have poor relations with his mother. Could you enlighten me on this point?

I said he was low on expressions of sorrow because he felt this emotion only once in his dreams. You said he was high. Finally, you said he was a very passive person and I said he wasn't. I judged lack of passivity (or high activity) on the basis of his activities in his dreams which were often strenuous and physical. Is this a case of compensation? In general, I don't feel that dreams are compensatory. An active person has active dreams, a passive person has passive dreams.

Moss replied on each of the issues Hall raised. His replies will be discussed in the following paragraphs. They show that Hall actually was right on three issues--interest in sports, lack of interest in travel, and high level of activity. Hall may have been wrong on the emotion of sorrow. Hall very likely was wrong on the man's attitude toward his mother for reasons we will discuss shortly. With this overview in mind, we turn to Moss's replies. They are important because they show that theory-based clinical "interpretations" are one thing and personal concerns are another on several issues.

Interest in sports

Moss agrees that the man was interested in sports, but Moss did not interpret this as an "interest" because the patient was an "abysmal participant" and because he "alluded to sports or 'games of chance' as a means of reflecting how he felt about therapy or his dealings with people."


Moss agreed the patient "had no real interest in travel per se." Moss said he rated the patient's interest in travel as high because "he did make periodic references to returning to California where he had been before he was married and used this as an escape mechanism from reality."


Moss conceded he was on "tenuous ground" in saying the patient was very passive. He agreed the patient was an active person. Moss thought of him as passive in a therapeutic sense, as the following paragraph explains:

Finally, I did say that he was a very passive person, but here I feel I am on more tenuous ground. The patient was an active (acting out) person all his life, but he came into therapy because his coping mechanisms had failed. He was caught up in a series of self-defeating, hostile, acting out behavior and he reacted to it with an attitude of despair. I thought of him as being "passive" in the sense that he was analyzing this behavior, attempting to work toward something more advantageous. In this sense, he had despaired of trying to work through his problems using his old methods. But it is correct that he was an active person, having active dreams which for the most part reflected the reality of his life, and so I believe that we could agree that he was a high activity person.


Hall inferred that the patient was low on sorrow because it appeared only once in the 58 dream reports. Moss said the depth of the patient's sorrow only showed up in his reactions to the dreams, by which Moss means "guilt and remorse." Here we may see the difficulties of agreeing on what a word means in the realm of emotions. Moss writes:

The patient's dreams almost inevitably were filled with hostile reactions, but in his interpretation and subsequent evaluation of the dreams, he was often filled with guilt and remorse, particularly in areas involving his wife. The patient throughout most of therapy developed psychosomatic symptoms which I again relate to his conflict and sorrow.


Hall made his comments on the patient's conception of his mother based on only one dream, a dream in which the patient and his brother were trapped in their mother's apartment. There is no interaction with the mother in the dream, nor any indication of her presence, but only a mention of her. Hall in effect overgeneralizes when he says, "I figured that anyone who had such poor relations with women, including his wife, must also have poor relations with his mother." Moss's reply is very convincing on this point:

In regard to his attitude toward his mother, the patient's attitude toward both men and women left much to be desired, certainly the attitude towards his wife was fraught with hostility and ambivalence. However, the patient held a rather positive attitude toward the mother throughout therapy as far as I could determine. Early in his life, he held a somewhat fictive goal of rescuing his mother from the tyranny of his father, and throughout therapy he always spoke of the mother in protective, understanding terms.

The Patient's Case History

Moss provides some useful information on the patient's motivations and symptoms in the chapter on the hypnotic interpretation of his dreams. He came to therapy because of marital tension, general nervousness, and physical symptoms such as stomach aches. He worked as an inventory clerk. He had an IQ of 121. Moss says the patient seemed psychologically naive and "exhibited minimal insight into the basis of his difficulties and a proneness to attribute them exclusively to an organic etiology" (1970:33). However, the patient "appeared genuinely motivated and was able to interact effectively despite his anxiety" (Moss, 1970:33). The patient improved considerably in the course of 48 sessions. A follow-up inquiry ten months after therapy ended found that he was "reasonably satisfied with his marital and work situations and he had not felt the need for additional psychiatric assistance in the interim" (Moss, 1970:56).

The patient had been hospitalized 11 years earlier when he made a "suicidal gesture" while in the military overseas (Moss, 1970:32). He became depressed and received a medical discharge. In an autobiographical sketch he wrote shortly thereafter he said his family upset him because of constant bickering. He said nothing he did ever pleased them. He characterized his general attitude toward people in a way that fits exactly with Hall's inferences: "I try to stay away from them, they make me sick" (Moss, 1970:32).

Moss characterizes the patient's father as a "strict disciplinarian" who was self-aggrandizing, impatient, and intolerant, and his mother as "meek, hardworking, and long-suffering" (Moss, 1970:38-39). The patient had a younger brother who was more physically robust. The brother had violent quarrels with the father and bullied the patient. The hypnotic dream sessions suggested that the patient was deeply angry with his father, brother, and wife, which certainly fits with the dream findings on aggression. In the course of the treatment he came to a better understanding of his relationship to his wife and "recalled with remorse the occasions when his wife had tried to communicate with him" (Moss, 1970:38). He came to realize that she couldn't express her feelings any better than he could.

In all, Moss's account of the patient's tensions and insecurities accord extremely well with Hall's quantitative findings and ratings. As with Freud, Jung, Kafka, and the child molester, there is considerable continuity between dreams and waking life. Beyond the general confirmation of the continuity hypothesis, there are three important points that emerge from this collaboration between a quantitative content analyst and a psychotherapist:

First, Hall's incorrect inference on the patient's attitude toward his mother shows no inferences should be made from one character class to another. Specifically, hostility toward a general category, such as women, should not be presumed to include specific significant others who are included in that category unless there is hostility toward them as well. Conversely, it would be risky to assume from hostile interactions with a mother or sister that the dreamer would have hostile interactions with all women in waking life.

Second, quantitative content analysts have to make very clear what kind of information they want from the psychotherapists after the analysis is completed. Psychotherapists will have a natural inclination to utilize their theories and to provide their interpretations if questions are not very focused, as is evidenced by Moss's reaction to questions about the dreamer's passivity and interest in sports and travel.

Third, extreme deviations from the norms on our key indicators of possible psychopathology were very accurate in this instance, and we will see further support for them in some of the studies of patient groups later in the chapter.

Sam and Tony

The next two cases concern males who were neither famous nor in psychotherapy. They each separately happened to pick up Hall's The Meaning of Dreams (1953c) and then wrote him to offer their dream series for study. They are included here because everything Hall knew about them was learned through their responses to questions he formulated based on a study of each dream series. This gives us an opportunity to show we do not have to rely on published biographical sources or the judgments of psychotherapists for our information on waking life, as we did in the previous four cases. These examples show that there are very simple and objective ways to gain useful information from the dreamers themselves. By this point there is no need to present the usual quantitative findings from the dream series. We will concentrate on the predictions that were made and confirmed.

Sam was 20 years old when he offered his series of 50 dream reports. From the quality of his letter and his willingness to answer a series of objective questions, Hall decided that he would be a useful subject even though there was nothing very unusual about his dream reports. Based on both high and low frequencies, Hall asked him to respond to what were in effect either/or or yes/no questions. The dreamer, of course, did not know the findings from the analysis of the dream series. The responses are grouped by coding categories.


Unlike the male norms, there were many more indoor than outdoor settings. It was therefore inferred by Hall that the dreamer preferred indoor activities to outdoor ones. The dreamer was asked whether he preferred one type over the other. He replied he was almost entirely interested in indoor activities.


There were five atypical frequencies in the objects categories. Sam rarely dreamed about tools, household articles, nature, or clothes. It was inferred he would have little or no interest in craft activities, household chores, nature, or clothes. Sam replied he had no interest in any of them. On the other hand, money appeared more often than expected in the dream reports. Sam said he was very interested in money.


It was predicted that Sam would say he feels very close to his family because many members of his family appear in his dream reports and that he does not have many close friends because relatively few friends are mentioned in the dream narratives. Both predictions were confirmed.

Social Interactions

Sam had more aggressive interactions with males than with females. He was not above the norms, but he was asked if he felt males are more hostile than females. He said he felt males were more hostile. Sam was a victim more often than he was an aggressor in his dream reports. He says he feels in waking life that he is more apt to be the victim of an attack than an initiator. There was more aggression than friendliness in his dream reports, although not unusually so, but he was asked whether or not people were usually friendly and helpful. He replied that he did not think people are usually friendly and helpful, which fits with his aggression/friendliness percent.

One prediction was not confirmed. Hall thought Sam would be interested in guns and other weapons in waking life because there was an above-average reference to them in the dream reports. He replied that he "hated" guns and weapons, and had nothing to do with them. Here is one of those places where it would have been useful to be able to probe further or find other sources of information. Overall, however, the findings from Sam's 50 dream reports allowed Hall to develop a fairly complete and very accurate portrait of the dreamer.

Tony, a schoolteacher in his middle 30s, kept a record of his dreams for a year. His preoccupations, interests, and behavior patterns are reflected extremely well in his dream reports. It is obvious that he is a schoolteacher because he is teaching school frequently in his dreams, and that he is worried about growing old, for he dreams of his hair turning gray or falling out, and of physical incapacities and aging. Tony's dream reports show him to be very conscious of time and money, with an interest in doing household repairs, all of which he acknowledged. There are few attempts at success and no failures, and Tony reports that he is not ambitious and has no fear of failure. There are few aggressive interactions in his dream reports and Tony says he is not an aggressive person.

Tony is atypical in his dream interactions in that he has more friendliness with males than females, which squares with his waking feelings. Tony has sexual relations with both males and females in his dream reports, and many of these sexual contacts are with adolescents. He has dreams where he is rejected by women in favor of other men, and where he feels inadequate in sexual relations with women. When combined with the fact that there were slightly more sexual interactions with males than with females, the negative interactions with women suggest that he prefers male sexual partners.

Tony reports the following sexual history. His first sexual experience was with a boy when he was 16. His first heterosexual relation occurred a few years later when he was in the military. Then there were some casual affairs with young women and a four-year affair with an adolescent boy. At the time he was corresponding with Hall he was seriously involved with a woman and thought his homosexual impulses were waning. In short, there is great continuity between Tony's sexual patterns in dreams and in waking life.

In summary of this section, Sam and Tony's dream series are fairly typical in showing continuity with both waking thought and behavior. But not all cases are so straightforward, as our discussion of Karl will now document.

Karl: A Difficult Case

Karl was in his early 30s when he began keeping a record of his dreams. He provided Hall with a little over 1,000 dream reports in a three-year period. We are discussing the findings from his dream series because they illustrate what appears to be a lack of continuity between dreams and waking life. The large number of dream reports also makes this a useful case because they allowed Hall to focus on dream reports that repeat key content preoccupations.

Karl's parents were divorced when he was age 13. He continued to live in the family home with his mother and two brothers. He was a talented football player in high school and played one year in college on an athletic scholarship. He married when he was in his early 20s. He and his wife separated toward the end of the three-year period when he was keeping his dream journal and later divorced. They had two children. Karl works as an engineer. He is a jogger and weightlifter. He is 6'2" tall and weighs over 200 pounds. We will concentrate on three areas where there seemed to be discontinuity between Karl's dream reports and waking life:

  1. His interest in football.
  2. Violent aggression.
  3. The frequency of sexual activity.

Karl had 65 dream reports that included athletic activity. Most of the time he was playing football again. Hall made the simple prediction that the dreamer was very interested in football. At that point, of course, Hall knew nothing about Karl's high school football successes and his college football scholarship. Karl wrote back that, yes, he used to be interested in football, but had taken no interest in the game since he quit it after his first year in college. He said he never watched games after that either.

So, here was a major disagreement on a simple issue, but it turned out that Karl never really gave up his interest in playing football, which is what he continued to do in his dreams. Later he wrote with some embarrassment that several years after he was out of college, just five years before he started the dream journal, he tried out for a professional football team but was not successful. He also now wrote that even at that point he would like to work out with a professional team if he had the opportunity. "Who knows," he wrote, "maybe I'm kidding myself about not liking football. Perhaps I do, and don't want to admit it. I don't really know, but there's some sort of hang-up rooted in it."

If Karl had responded only on a yes-no objective test, Hall's inference would have been scored as incorrect. The lesson is we should not always take an immediate "no" for an answer, but then the question becomes how to probe further without risking the possibility of suggesting or forcing the desired reply. Here the usefulness of information from others becomes apparent, but that often raises issues of confidentiality and privacy unless clear understandings have been reached with subjects.

Karl has an above-average number of violent and sadistic dream reports. In one dream, for example, he kills his father with a shot from a high-caliber rifle. In another he sees unknown men starving and tormenting a woman they have chained to a chair. When Hall made the prediction that Karl became involved in fights occasionally and harbored angry feelings, he received a denial. Karl reported that he never engaged in fights. Moreover, he regarded himself as being a friendly, warmhearted, peaceful person.

There are two things going on here. First, Karl is actually not all that friendly to some people. Second, the inference perhaps was too general. It should have been focused on those people with whom Karl had aggressive interactions in his dream reports. Karl readily acknowledges anger toward his mother, father, and wife. His letters to Hall are full of outright hatred for his father and resentment toward his mother and wife. He quit football in his freshman year at the university partially to settle a score with his father, who derived pleasure from his son's successes on the gridiron. Karl wrote as follows about this episode:

The sorry bastard was all set to soak up the old glory at my expense. No sooner was he all situated [his father became active in the quarterback club and was elected president of the alumni association] than I slipped out of the situation that May. It really left him high and dry, really frosted his ass. It was great, a real pleasure.

So, Karl agrees with his dream reports that he dislikes his father and resents his mother and wife, but it does not follow for him--and perhaps does not follow in reality--that he is an angry, violent person. This case may demonstrate the opposite type of mistake to the one made with Moss's patient. Hall used the patient's hostile interactions with various women to infer that he disliked his mother, and was probably wrong. With Karl he used hostility toward mother, father, and wife to infer aggressive interactions with a wider range of people, and perhaps was wrong. In other words, we have findings that support those theorists who argue that at least some aspects of "personality" are situation specific, only here we would call what we have seen "person-specific," so to speak. In any case, the danger is one of overgeneralization from specific dream content.

Finally, Karl has many passionate sex dreams that are described in vivid and uninhibited detail. Over the three-year period he had sexual relations in his dream reports with 38 different female characters, many of them repeatedly, most of them women he knew and to whom he felt attracted in waking life. He also had sexual relations with two male characters. Hall therefore inferred that Karl had an active sexual life that may have included some occasional homosexuality.

Wrong again, but this time the problem is not a disagreement with Karl. We include this failed prediction because it shows clearly that sometimes the continuity is with waking thought and fantasy, not with behavior. In actuality, Karl had a miserable, constricted sexual history adding up to frustrated fantasies and frequent masturbation. Karl did not have sexual intercourse until he was married because he considered premarital intercourse "immoral and unintelligent." From the beginning of his marriage he did not find sexual satisfaction with his wife and became resentful toward her. He claims she never "warmly initiated intercourse" and did not like foreplay. He thinks this "coldness," as he puts it, was the reason he was sometimes impotent when they did have intercourse. Karl did not have extramarital affairs, and he had only one sexual encounter with a woman in the first year or so after he and his wife separated. He has never engaged in a homosexual act. Here, then, is a person whose sexual behavior was not predicted from the sexual activity in his dream reports. The continuity is with his waking sexual fantasies. His one sexual behavior is masturbation, once or twice each day.

Most of the lessons we learn from Karl's case are too obvious to summarize. There was far more continuity than seemed apparent at first, but it was masked by his blind spots, or by poor questions. Most of all, though, the findings on aggression and sexuality show the continuity can be with waking thought only.

This finding on sexuality and aggression with Karl was not unique among the several similar blind analyses that Hall carried out by correspondence with individuals who sent him dream series. Occasionally there would be a lack of behavioral continuity for more mundane content categories, such as a woman who dreamed frequently of foreign travel, but only daydreamed about it in waking life. For the most part, however, such behavioral discontinuities as there were occurred in the areas of aggression and sexuality. In future studies, one major task should be to determine if there are elements in dream reports that will allow us to predict whether the continuity between dream content and wakefulness will be only in the area of thought and fantasy.

Person One

Person One is a 21-year-old male college student who wrote down 40 of his dreams the summer after he graduated from high school, then another 20 during a six-week period in his junior year in college. The series is a little shorter than we now prefer, but it has two striking features that make it useful. It was coded for us by Adam Schneider.

As shown in the h-profile in Table 8.7, the most unusual finding for Person One is his very low male/female percent of 38/62 (h=.59), something we rarely see in individual male series. He reports that his father died when he was age 2, and he did not have a stepfather until age 10. The vast majority of his friends are women and his only close family members are his mother and grandmother. He does not like men as much as women, and does not form close friendships with them.

Person One is also unusual in his low level of aggressions. The rest of his percentages and indexes are close to the norms, although he is high on familiar characters. These findings typify Person One. He is an unaggressive person who spends most of his time with a few close (female) friends. He is a low-key, normal person. What makes Person One worth discussing here, of course, is the way in which the coding system picked up the two issues on which he is an atypical male with only 60 dream reports--his preference for women as friends and his low level of aggression.

Table 8.7 goes here

Dream Content in Patient Populations

There have been many studies of dream content in various psychiatric populations over the past 40 years, but most of them are anecdotal in nature, use untested coding systems, or include only small numbers of subjects. The earlier studies are summarized by Kramer (1969, 1970) and Kramer and Roth (1979b), who conclude that there are only a few consistent findings, such as more family members in the dreams of depressives and more hostility, emotionality, and bizarreness in the dreams of schizophrenics. Some studies using the Hall/Van de Castle coding system since the Kramer and Roth (1979b) review was completed show more promise, but the results are inconsistent.

There are several likely reasons for why the findings on psychiatric patients have been so meager and inconsistent. There may be variation from hospital to hospital in how patients are diagnosed and classified. Patients within the same diagnostic categories may have been in different phases of their illnesses. The possible effects of medication or hospitalization on dream content are not controlled. It also may be that there are not large-scale group differences from what "normal" people dream about, as will be suggested when we consider the first psychiatric dream content study using the Hall/Van de Castle system (Hall, 1966b).

However, we think at least part of the problem lies in the inadequacy of the measures utilized in previous studies, which is why we proposed a number of possible psychopathology indicators in chapter 2. That is, there may be more differences than the older measures have been able to detect. We will first of all summarize several suggestive findings from studies using the Hall/Van de Castle coding system, and then suggest that the "h-profile" might be of use in future studies.

Rather than reporting on the relevant studies in a one-by-one fashion, or grouping the studies in terms of the type of patient studied, we are going to organize the findings in terms of our psychopathology indicators, showing which psychiatric groups have been found to be atypical on a given indicator in one or more studies. Since some of the findings are contradicted in other studies, we are only claiming that the "positive" results are "suggestive."

Lack of friends and friendliness

There are several studies suggesting a smaller number of friends and friendly acts in the dream reports of some types of patients. In fact, a low F/C index with female characters was the major finding in Hall's (1966b) study of 211 dream reports collected from 50 male patients who were grouped into four diagnostic categories: five patients who were both schizophrenic and alcoholic, 20 patients who were schizophrenic, 15 patients who were alcoholics, and ten patients with a variety of other diagnoses. The dream reports of the four groups were compared with each other and with the male norms.

The dream reports were coded for characters, social interactions, success and failure, misfortune and good fortune, and food and eating. The differences among the patient groups were few and unrevealing: dream reports from schizophrenics were shorter, and there were more food and eating elements (actually, drinking) and fewer sexual interactions in the alcoholics' dream narratives. Moreover, there were only a few differences between the patients as a group and the Hall/Van de Castle norms. For example, there was a lower male/female percent (58/42 vs. 67/33, h = .19), which may be worth a look in future psychopathology studies of males. But the biggest and most interesting difference--in fact, the only other difference--concerned friendly interactions, especially with females. This is seen first of all in the atypical aggression/friendliness percents displayed in Table 8.8. The same finding is revealed in the familiar 2 x 2 table of A/C and F/C scores with male and female characters.

Table 8.8 goes here

Table 8.9 goes here

Low friendliness also was a striking finding in a study of female patients in Paris, France. Fifteen were schizophrenic, 12 were other types of psychotics (Schnetzler and Carbonnel, 1976). While these authors did not calculate an aggression/friendliness percent, we can use the figures in their Table 3 to determine that it is 62 in the schizophrenic group and 78 in the other psychotic group, as compared to 47 for their control group of 15 normal female subjects and 52 for the Hall/Van de Castle female norms (Schnetzler and Carbonnel, 1976:373).

A similar finding is reported in a study of female outpatients in London, England, who suffered from high anxiety states (Gentil and Lader, 1978). Twenty patients each mailed in five dream reports in stamped envelopes they were provided by the researchers. These dream narratives were compared with those collected from 25 female volunteers. According to the authors (1978:301), one of the "most significant" findings was the lack of friendly interactions, but the statistic used does not make it possible for us to make comparisons with our norms.

Findings in a comparison of dream reports from depressed and schizophrenic patients studied in the laboratory point to a low level of friendliness as well as a lack of friends in patient dreams (Kramer and Roth, 1973). Ten depressed patients, five male and five female, contributed 91 dream reports, and 13 schizophrenics, 11 male and two female, contributed 217. There were more strangers in the schizophrenics' dream reports (71%) and more family members in the depressed patients' reports, but the striking thing to us is that both groups had an extremely low friends percent, 18 for the schizophrenics and 22 for the depressives. We conjecture that a similarly low friends percent would be found in other studies reporting more relatives in depressives' dreams (e.g., Bollea et al., 1978).

Aggression and victimization

Several studies suggest there is more aggression in the dream reports of various patient groups than in control groups or the Hall/Van de Castle norms. For example, Firth et al. (1986) found more of what they called "destructive violence" in the dream reports of violent patients. Often this finding is independent of any findings for low friendliness; that is, it is not part of an aggression/friendliness percent. Some of these studies report that patients are more often victims in their dream reports. In our terms, the patients have a very high victimization percent. In this brief review we will focus on the studies reporting high victimization percents.

The Paris study by Schnetzler and Carbonnel (1976), just discussed in terms of low friendliness, found psychotic patients are more likely to be victims in aggressive interactions. So did a study of acute depressive patients studied in a laboratory in Italy (Bollea et al., 1978). The study of highly anxious women outpatients in London also found higher victimization in the patient group than the control group (Gentil and Lader, 1978). There is also a Canadian study reporting higher victimization for male and female asthmatics than neurotics or hypertension patients (Levitan and Winkler, 1985). This study did not use Hall/Van de Castle scales, so the several patient groups are not comparable with our norms. Stretching things a bit, perhaps we can include here Hauri's (1976) finding of more "masochism" in the dream reports of people who had come out of a reactive depression.

Patricia Carrington's (1972) excellent study of 30 female schizophrenics in a New York City hospital provides impressive evidence for a high degree of aggression and victimization in a patient population. Carrington's subjects were mostly first-admission patients in an acute phase of their illness. None had received electroshock within the previous three months. They ranged in age from 15 to 39, with a median age of 19. Compared to Carrington's control group, a comparable number of college women in the same age range, the schizophrenics' reports contained more aggression, especially physical aggression against the dreamers. In our terms, they were high on victimization percent and physical aggression percent.

Anger and victimization were found as important elements in dream reports from anorexic and bulemic women. Brink and Allan (1992) compared dream reports from 12 women with eating disorders to those of 11 normal women on a 91-item rating scale. The anorexic and bulemic dreamers had more instances of being attacked, being ineffective, hating themselves, and feeling anger.

We do not want to imply that all studies of patient groups show a higher rate of victimization. Riemann et al. (1990) did not find this to be the case in their study of depressive patients in Germany. Similarly, Barrett and Loeffler (1992) compared 20 college females classifiable as depressed by the Beck Depression Inventory with 21 non-depressed college females and found differences suggesting that there is not much of anything in depressives' dreams: fewer dreams are reported, their length is shorter, there are fewer characters, and there is less anger. In fact, the generally mixed findings on depressives are one reason why we say that the positive studies we are reporting are merely "suggestive."

Negative emotions

There is some suggestion in the literature of a higher negative emotions percent in patient populations. This was the case, for example, in Bollea et al.'s (1978) study of acute depressives in Italy. There is also some evidence of a decline in negative emotions in the dreams of depressed patients after four weeks of treatment with trimipramine (Riemann et al., 1990:97); however, these findings are self-ratings of dream emotions by the patients themselves, so they may reflect the general improvement in their condition, not changes in the dreams themselves.

Miscellaneous findings

There are a few other suggestive findings in the literature relating to our hypothesized psychopathological indicators. The anxious patients in the Gentil and Lader (1978) study had more mentions of body parts and the depressives in the Kramer and Roth (1973) study had a very low success percent. Carrington's (1972) study of 30 hospitalized female schizophrenics found more misfortunes, including misfortunes to the dreamer's body, than in the control group.

New directions

The general findings on low friendliness, high victimization, and high negative emotions in several different patient groups make us hopeful that more differences could be found in studies using our psychopathology indicators relating to these issues: friends percent, F/C index, aggression percent, and A/C index. As for our other indicators, they are basically untested in the literature except in the individual case studies earlier in this chapter. They thus provide us with the possibility of entirely new directions. Indexes like the physical aggression percent, the A/F square, the torso/anatomy percent, and the bodily misfortunes percent seem especially promising to us. To aid in the use of these indicators, we have put the male and female normative expectations for each of them into Table E.1 in Appendix E.

Given that none of these indicators is likely to be specific to any particular diagnostic category, we think a further step needs to be taken. We think there is potential in using the h-profile to see if there are patterns of deviations from the norms indicative of different types of psychopathology. For example, the patterning of the victimization percent and the physical aggression percent might differ in physically violent people and people who were physically abused as children. Both groups might have a high physical aggression percent, but violent people might have a low victimization percent and abused people a very high one.

The indicators to be included on the h-profile in psychopathology studies could be determined in much the same way the MMPI was created, that is, by seeing which indicators actually differentiate "normals," neurotics, and various types of psychotics. In the short run, however, we think the h-profile is likely to be most useful in studying lengthy dream series from individuals, rather than groups of patients. This is because of the difficulties in securing a good, complete set of dream reports from any psychiatric population.


In all, we think the findings presented in this chapter are good evidence for our claim that the conceptions and concerns found in dream content appear to be the same ones operating in waking life. The waking mind and the dreaming mind seem to be basically one and the same, a strong argument for the idea that there is meaning in dreams.

However, we recognize that much more research needs to be done before this conclusion can be stated with confidence. We need to demonstrate continuities in representative samples of subjects who have been shown through testing or observation to vary along behavioral dimensions corresponding to our findings on dream content. The degree of aggressiveness is the most obvious candidate here, but so is assertiveness, which can be measured with a dominance scale. We also need a wider range of case studies with long dream series where more in-depth personal information can be obtained from the subjects. In this regard it would be ideal to work with people who have written down their dreams as part of a Jungian analysis so personal information could be obtained from their analysts as well. We also need to see if the h-profile can help us identify various types of psychopathologies, which would be all the more interesting if some patterns on the h-profiles corresponded with various psychopathology profiles on the MMPI.

Now that we have presented our findings, we can return to the issue of predictions not corroborated by the dreamer. How should we deal with Sam's denial of an interest in guns or Karl's claim that he is not an aggressive person? We think there are two main answers. First, we need to compile a catalogue of failed predictions to see if any pattern emerges in them. Are there common issues where dreamers do not support our claims? Secondly, we need to develop more case studies where we can obtain corroborating testimony. For example, if a man's women friends or his Jungian analyst said he was hostile to women, thereby agreeing with the content finding, we would be less inclined to accept his denial of our inference.

The hypothetical example at the end of the previous paragraph leads to another point. We need to elaborate Hall's (1953b) claim that some people are more open to their conceptions than others. We also need to extend it to say that people may be open to some of their conceptions and blind to others. It is not enough, however, to put forth such an hypothesis. We must have psychological tests permitting us to predict in advance the kinds of conceptions the dreamer is likely to deny.

We think this chapter shows that the quantitative content analysis of dream series and dream sets can tell us something new and interesting about some aspects of people's motivation and personality. We do not claim we have said anything new about character traits or defenses in this chapter, but we do think we have shown that dreams are a good window into people's conceptions, concerns and interests. As Foulkes (1985:192) puts it from a cognitive point of view, "because they are so little responsive to anything but what's on our mind, dreams may be one of the purest reflections we have of the structure and processes of our mind." In Hall's (1947, 1953a) terms, dreams bypass cultural and personal defenses to reveal underlying conceptions and concerns with an uncluttered clarity.

We now have reached the end of our quantitative road, at least for the time being. We hope we have shown readers how the quantitative study of dream content can be rigorous, systematic, and revealing, and that further studies using our methods and statistical conventions would be a valuable contribution to the literatures on both dreams and personality. We hope we have shown there is meaning in dreams.

Although we have reached the end of our quantitative road, we have not quite reached the end of the book. In the final chapter we want to make use of our quantitative findings for two general purposes: as an anchor point for a slightly new way of thinking about dreams in general, and as a link to recent work on the content of daydreaming, reveries, thought intrusions, and similar types of relaxed waking cognition.

Continue on to Chapter 9

Go back to the Dream Library index.

dreamresearch.net home page dreamresearch.net contact info