Figure 3, based on the Hall and Van de Castle findings with young college men and women, shows that the content indicators control for these correlations. Comparing dreams with less than 175 words to those with more than 175 words in three different samples that range from 50 to 300 words, there are no systematic differences between shorter and longer dreams on any of the percentages and ratios when the two samples are compared. The physical aggression percent for women is unexpectedly lower in the longer dreams, and the familiar setting percent lower in the men's dreams, but the same differences are not found in the other samples, so it is likely they are an anomaly. And, as we will see, the bodily misfortunes percent is lower for Barb Sanders in the longer dreams, but the same difference does not show up with the other two samples. On the other hand, and as expected, the longer dreams are usually higher on the "at least one" indicators, demonstrating that these indicators -- and, by inference, the rating scales used in most dream studies -- are extremely sensitive to dream length.
Note: The differences in this figure are expressed in terms of the units of the effect size statistic "h," which are about twice as large as a percentage point. An h of .20 or less is considered small. The h statistic is explained in the statistics section of DreamResearch.net.
But the bigger test is with the Barb Sanders series because it is the largest study of the issue to date. It includes dreams that range in length from 2 to 500 words. This range encompasses just over 95 percent of her dream reports. The starting point was a random sample of 250 dream reports that ranged from 50 to 250 words in length. These dreams were coded for characters, social interactions, misfortunes, and emotions. Then random samples of 200 shorter and 200 longer dream reports were added. These samples were only coded for social interactions, the most important and sensitive coding categories, because of limited resources.
The study reveals that the "at least one" indicators rise consistently as word length increases until reaching a plateau at 400 words. This finding shows that longer reports have more social interactions, as most researchers might expect. It is also very likely that most other coding categories would show the same sensitivity to report length.
On the other hand, none of the rates and percentages is affected by dream length once a minimum of 50 words is reached. This means that the indicators do an excellent job of controlling for dream length up to at least 500 words. A selection of these results is displayed in Figure 4. At the same time, the results in Figure 4 also make clear that the content indicators do not do a good job of correcting for the distortions that appear in dream reports of 25 to 49 words. The findings for dreams under 25 words in length, which are not included in Figure 4, are even more distorted. This reinforces the earlier decision by Hall and Van de Castle to exclude dream reports shorter than 50 words. It seems unlikely that repeatable and scientifically useful results can be obtained with short reports.
Note: The differences in this table are expressed using the effect size statistic "h."
Our more detailed analysis of the Barb Sanders series began with a Hall/Van de Castle content analysis of a random sample of 250 dream reports (the "baseline 250") ranging in length from 50 to 300 words. As noted a moment ago, this baseline sample was coded for characters, social interactions, misfortunes, successes and failures, and emotions. The work was done by Sarah Dunn, Melissa Bowen, and Heidi Block, who worked in different pairs on different categories. Several reliability checks yielded the high percentages of agreement shown in Table 2. All differences of opinion were resolved by discussion among the coders to provide a uniform result for entry into the computer.
The first important finding is a methodological one: the codings from any 125 of the 250 dreams in the random sample replicate the overall results almost exactly, as determined by the approximate randomization program available to researchers on request through DreamResearch.net. However, there are many deviations from the overall results with subsamples of 100 dream reports, and the drop-off is large at 75 and 50, demonstrating once again how risky it is to accept or reject hypotheses based on inadequate samples sizes. For example, the A/C index of .33 is usually between .30 and .35 with randomization samples with 125 dream reports, but it ranges from .27 to .39 with samples of 75.
The findings on sample size, when combined with the findings on dream length, lead to two basic rules for future dream research:
These two rules are deeply upsetting to traditional dream researchers. Relying on what they learned in basic statistics courses, they think that sample sizes of 30 dream reports are sufficient. They forget that when statisticians say that two samples of 30 each can be compared using the t test, they are assuming there are 30 observations in each sample. But with dreams, the elements being studied are not present in every dream report, so there are many less than 30 "observations" in 30 dream reports. Only 50% of dreams at best have at least one aggression, for example, and about 40% have a friendly interaction, and so on down the line from there, with sexual interactions logging in at about 10%.
Traditional researchers also don't like the 50-word minimum. They know how much trouble it is to convince a large number of people to spend several minutes writing down a full dream report. We suspect from inspecting the dream reports collected by others -- which they often gave us only reluctantly, and sometimes only after we said we would go to the American Psychological Association's committee on ethics, which insists that data be shared if it has been published and if participants are not named -- that many studies are based on inadequate reports.
Methodological cautions to one side, what are the findings?
The General Substantive Findings
The first substantive analysis concerns the overall consistency in the dream series. A comparison of the first 125 dreams in the baseline sample with the second 125 (Table 3) shows that the dreamer is consistent within 5 or 6 percentage points in all but two or three categories. Her male/female percent fell from 58/42 to 49/51, placing her very close to the female norms of 48/52. It is also interesting that she dreamed less of friends and more of family members in the second half of the series, but the familiarity percent -- which relies on these two categories, plus prominent characters -- remained more consistent. These findings on consistency support several previous studies demonstrating consistency in dream content over decades. This result is now so well established that it is deviations from it that are of interest in developing a better theory or in understanding a specific person. It is a finding that can be used as one litmus test for determining the adequacy of any dream theory.
These findings on consistency with the Hall/Van de Castle coding system can be supplemented and reinforced by a study we did of the Barb Sanders series using word and phrase searches on DreamBank.net. For this analysis, we compared her first 3,116 dreams, which she wrote down before we met her, and which we used for all the Hall/Van de Castle codings, with the 1,138 she has given us since that time. At least one of the 13 main people in her life (parents, ex-husband, three siblings, three children, granddaughter, and three best women friends) appear in 33.6% of the dreams in the first set and 35.1 percent of the second set. Her interest in theatrical productions, as a writer, actor, and producer, is reflected in the fact that 4.9% of the dreams in the first set contained one of several terms related to this activity, as compared to 5.2% for the second set.
We also compared the first and second sets using long word strings for each of the five emotions that are coded for in the Hall and Van de Castle coding system: anger, apprehension, sadness, confusion, and happiness. The percentages are very similar, with the biggest difference, 4.6 percentage points, on happiness. For the total number of dreams with at least one emotion, the figures are virtually identical, as seen in Table 4 below:
These overall results on consistency do not deny that there is day to day fluctuation in dream content, or that striking external events or some unusual physical condition, such as an upset stomach, may sometimes affect dream content. However, they do suggest that most dream content must be drawn from a reasonably circumscribed set of schemata and scripts. Such a result, we might add, is completely unanticipated and unexpected by most tradition dream theories.
Characters and Social Interactions
Characters and social interactions are usually the most psychologically revealing aspect of a detailed content analysis, and the Sanders series is no exception. The random sample contains 884 human characters, 3.4 per dream report. Since the female normative figure is 2.7, this finding suggests that Sanders is more intensely involved with other people in her dreams than the typical woman. This presumption is supported when the 679 social interactions are analyzed in terms of rates per character, that is, the A/C, F/C, and S/C ratios. As shown in Table 5, the dreamer is 19 to 24 h points higher that the normative figures for these three ratios. These are small but statistically significant effect sizes. What makes them noteworthy is that all three indicators are elevated.
Although Barb Sanders has a higher rate of social interactions than the normative sample, she is very typical in her A/F percent, 49, which means that she is equally involved in friendly and aggressive social interactions when all dream characters are considered. However, she does differ in that she is far more likely to be the aggressor, 50 to 33, which is 36 h points above the norms, a moderate effect size. Most of these aggressions are angry thoughts toward a person, critical comments, or rejections. She is about average on befriender percent,
To provide an indication of her patterns of interaction with key people in the dreamer's life, my research assistant Ryan Harvey coded all of the friendly and aggressive interactions in the entire dream series with her parents, favorite brother, daughters, and two close women friends. The findings, which are summarized in Table 6, correspond with how the dreamer and the four interviewees describe her relationships with these people in waking life.
For example, her mother is the most important and difficult people in Barb Sanders' life. Sanders provides the following portrayal of her mother:
Her mother appears in 239 dream reports, or 7.7 percent of the total, which is more than any other familiar character. The A/C ratio with her is .70, well above Sanders' average in the baseline 250 for all characters, .32. The A/F percent between them is 72, well above the dreamer's normative figure of 49 (h=.48). This A/F percent is consistent over the entire series, as shown when the dreams are divided into thirds.
On the other hand, Sanders says she has a more positive attitude toward her father, a claim that is corroborated by her friends. He appears in 213 dream reports, second only to her mother. The A/C ratio with him is .36, not far above her normative figure with all characters, and the F/C is .37, once again slightly above her normative figure. The A/F percent with him is 50. While there is nothing striking about her relationship with him, it is dramatically different from her relationship with her mother.
Sanders' middle daughter is almost as problematic for her as her mother. This daughter was 4.5 years old at the time of the divorce, and was most upset by it. At age 14 she ran away from her father's home and came to live with Barb Sanders. She did poorly in school, could not hold on to a job, and suffers from severe psychological problems. She had a daughter when she was a teenager and soon left her to Barb Sanders to raise. She still returns to live with Barb Sanders from time to time. Sanders worries about her constantly and there is great tension between them.
This daughter appears in 165 dream reports. The A/C ratio is .92, even higher than with the mother, and the F/C ratio is also very high at .52, well beyond Sanders average for all characters. Taken together, these two ratios show that there is a high rate of interaction between them. The A/F percent is 64, with Sanders initiating 79 percent of the aggressive interactions and 70 percent of the friendly interactions, far above her averages for all characters. These indicators provide an accurate summary of how Sanders conceives of their relationship.
By contrast, Sanders dreams only half as often of her oldest and youngest daughters, who adjusted to the divorce better, went to school in their father's home state, saw their mother primarily during summer vacations, and live normal adult lives. The A/F percents, 44 with the older daughter and 31 with the younger daughter, show that she has more friendly than aggressive interactions with both of them, which reflects her more positive relationship with them. As with the middle daughter, Sanders is more likely to be the initiator of both aggressive and friendly interactions.
The dream reports also capture her positive relationships with the favorite people in her life. For example, Sanders has great affection for the brother closest to her in age, who appears in 97 dream reports, which is one more than the total for her other two siblings combined. The A/F percent with him is 25, almost the mirror opposite of her interaction pattern with her mother. Positive patterns are also apparent with two women friends, but there are differences on some indicators, which reflect her different pattern of interaction with each of them. Sanders met her closest friend of long standing, Ginny, when she returned to college for her M.A. After college Ginny married, moved to a city 100 or more miles away, and raised a family, but she and Barb Sanders remained in close touch. Her husband and children are also friendly with Barb Sanders, who has visited at their home frequently over the years. Ginny appears in 96 dream reports and has an A/F percent of 23, the most positive balance with any known character. The comradely nature of their relationship is seen in the fact that they are equally likely to initiate friendly or aggressive interactions.
Sanders met another close friend, Lucy, when Lucy was a student at the community college where Sanders worked. Lucy, who is several years younger than Sanders, is outgoing and dramatic, and she and Sanders soon ended up working together in musicals and theatrical productions. Sanders is the big sister in this relationship, giving Lucy instructions, helping her, and becoming annoyed when Lucy is late or resisting direction. This pattern is reflected in the fact that Sanders initiates 78 percent of the many friendly interactions between them as well as 78 percent of the relatively few aggressive interactions.
These analyses of Sanders' dreamed social interactions with several of the significant people in her life are only a starting point in terms of understanding her waking conceptions of these people. Highly detailed studies with any of these characters would be likely to yield rich findings because of the large number of dream reports. To demonstrate this claim, somewhat more detailed analyses are next presented on two very different subsets of dream reports. One subset concerns her ex-husband, the other a man she was infatuated with for nearly a year in the middle 1990s, before she accepted the fact that he did not care for her romantically.
The Ex-Husband Dreams
Barb Sanders first met Howard, her future husband, when they were in high school, where they had nearby lockers. They sometimes danced or flirted, but he had a steady girlfriend, and in her senior year she fell in love with Darryl, the person she still considers the true love of her life. However, that relationship did not work out for reasons that she explained in the interview:
At this point, Howard was attending a major university far from their hometown, and was no longer seeing his high school girlfriend. He and Sanders then reconnected when he came home for the summer, and they were married after a year and a half courtship. Although it is risky to rely on memories of 35 year earlier, especially after a painful divorce, Sanders recalls that she harbored some doubts about the relationship even then. Perhaps this is not surprising in the context of the highly mixed feelings she harbored about Darryl and the end of the relationship with him. She also remembers Howard as a person who was insensitive to her need for tenderness and expression of feeling, as a person who just wanted sex. She felt that their sexual interactions sometimes felt more like rape than seduction.
Howard appears in 133 dream reports. He is surpassed only by her mother, father, and middle daughter in the frequency of his appearances. Dividing the dream series into thirds, there are 48 Howard dreams in the first third, 40 in the second third, and 45 in the final third, suggesting her marriage and divorce are a lasting concern. These dreams have a fairly regular structure. They usually start with Sanders noticing that Howard is back, causing her considerable apprehension or annoyance. At the outset he is often seeking reconciliation, although on occasion she is the one who is thinking about the possibility of getting back together. As the dream unfolds, Howard usually tries to initiate a sexual interaction, through a touch or a kiss, but Sanders is either hesitant or repulsed. Sometimes she is tempted, but then changes her mind. For example:
In 1996 and 1997, the Howard dreams seem to become somewhat more benign. They appear to contain more reflection and regret, and more discussion between the two of them. In one or two dreams, she even entertains the idea of reconciliation, an idea for which there was no basis in her or his waking reality. They lived far apart, never saw each other, and rarely communicated. In addition, Howard had been remarried for many years. This impression of a change in the tone of the Howard dream reports is borne out by the large decline in the A/F percent when they are divided into four chronological segments; there are proportionately more friendly interaction in the later years. This result is shown in Table 7.
Sanders' reflections on her feelings about Howard in March, 2000, parallel the main themes in the dreams, as well as the changes, as shown in these excerpts from the interview with her:
Then, in the midst of this apparent softening of attitude toward Howard, he died unexpectedly of a heart attack in April, 1997, without any history of serious illness. At this point the dreams about him seem even more reflective, and sometimes include the awareness that he is dead, even though she is interacting with him in the dream:
Despite this trend toward a more positive resolution of her feelings about Howard and the divorce, the more negative dreams still appeared on occasion. On November 13, 1997, for example, about six months after his death, she has a dream that is very similar to the five from 1976 to 1991 excerpted earlier:
Sixteen months later, on February 13, 1999, Sanders has a dream in which Howard suddenly appears and asks for sex. She thinks about it, but then "I pull back and say 'no'." He then "gets angry and grabs my arm and forces me over the wall of a house and tries to rape me violently." She hits him and knocks him unconscious. When he starts to regain consciousness, she escapes in a car.
In all, the dreams about Howard are a classic example of the operation of the repetition principle over the space of nearly 25 years. There is some decline in the negative aspects in these dreams, and there are changes in her waking reactions to them, but the issues they reflect remain essentially unchanged. It is as if these themes are embedded in her vigilance/fear system and are subject to reactivation under circumstances that cannot be determined with the information that is available for this study.
Moreover, these repetitive patterns seem to persist even though Sanders feels she has gained greater perspective on her failed marriage in waking life. Her belief is supported in the interviews with her friends, who note that she talks about him less and expresses less anger than in the past. But the dreams do not change quite as much as her thoughts about him. This fits with the cognitive view that dreams reflect mental encodings of past waking experience, not the experiences themselves, so are not likely to be changed by further reflections. It also fits with the idea that the updated aspects of an overall conception are no more likely to appear in dreams than the older one due to the lack of reality constraints during sleep. Taken together, these ideas lead to the hypothesis that waking feelings may change more, or more quickly, than dream feelings.
The dreams about Howard could be said to contain a "wish," in the Freudian sense, in that they clearly show a regret that the marriage was not successful, and they sometimes express a hope for reconciliation as well. There is also a wish for sexual gratification in several of the dream reports. At the same time, there is a traumatic quality to these dreams that overwhelms the wishful dimension. The fears and negative emotions persist. And yet, periodically there are dreams that clearly seem to "resolve" the issue and signal that the dreamer is ready to move on. But they are eventually followed by another negative dream. This finding should be given consideration before accepting claims in the clinical literature about the frequency and importance of "resolution" dreams. Such claims often suffer from the lack of longitudinal follow-up.
Dreams Of a Failed Infatuation
In late September/early October, 1994, several years after she had dated anyone regularly, Sanders met a man at a party whom she found very attractive. They had a common circle of friends and a mutual interest in the theater and dreams. They struck up a friendship, and a few months later they were in a small playwriting group together. Later they were in the same dream-sharing group as well. The man is 12 years younger than Sanders and did not seem interested in more than a friendship, which Sanders basically understood. Nonetheless, she became very infatuated with him and entertained the hope of a romantic relationship. Her friends were sure that nothing would come of the relationship, and worried about her.
However, two felt it would have positive aspects because it would add new zest to her life however it ended up. The friendship blossomed over the space of a year. Perhaps Derek gave some indications that he did care for her romantically more than she realized at first, mostly through heartfelt conversations. Whatever the signals, Sanders came to feel betrayed when the relationship did not go further. She became very upset when he showed affection towards another woman in her presence. When he came to a meeting of the dream group with a date, Sanders expressed her annoyance to him, to the great surprise of his date, and in effect ended the friendship in early April, 1996. She saw him only once or twice in passing after that.
Derek appears in 43 dream reports during the time period covered by the systematic analysis, then in another 4 dreams in the portion of the dream journal written after this study began.3 The first dream occurred on October 7, 1994, just a few days after she met him. Thereafter, the frequency and content of the dream reports reflects the rise and fall of her hopes about him.
Overall, the dreams have a very high rate of friendly and sexual interactions, and a low rate of aggressions, especially physical aggressions, as shown in the h-profile in Figure 5. This h-profile compares dreams concerning Derek to dreams with Howard in them and the baseline sample.
Thirteen of the first 16 dream reports in which Derek appears contain sexual or intimate physical interactions with him, such as warm hugs or resting her head on his lap. These early sensual dreams are in general very positive and full of anticipation, but they also express her fear that he is not does not care about her, and in one dream he even chases her after he has an orgasm:
As Sanders comes to accept that the relationship is not going to develop in the way she hoped, her sexual interactions with Derek become less frequent. She also has dreams in which she is angry with him or jealous that he is having sex with someone else. He is more peripheral in the dream reports in which he does appear, and some characters are now described not as "Derek," but as "like Derek." On May 2nd and 3rd, 1996, about a month after she broke off their friendship, she has dreams in which she is upset because he is with another woman:
Then, after a period of 5 months in which she records 72 dreams, but none of Derek, she has one that she calls "the kiss of forgiveness:"
A week later, she has a dream that implies they are still friends, even though she has not seen him in months:
After this pair of dreams, Derek appears in only 4 out of next 461 dream reports. By contrast, he appeared in 43 of the previous 334 dream reports. Two of the dreams after the apparent resolution dreams are relatively benign, but in one someone like Derek is betraying her, and in the other he is tormenting her, so he is not entirely gone from her mind:
The Derek dreams are striking for the fact of sexual intercourse and other sexual intimacies that have no correspondence to her waking reality. They are clearly wishful dreams. They are continuous with her waking hopes, but not with her waking life. They also directly reflect her fears about him and the relationship. Unlike the Howard subseries, they do decline greatly in frequency once the wishes are gone. This leads to a testable hypothesis for future studies: it is only possible for a significant person to disappear from dream life if there have been no real-life intimate interactions involved.
There are several reasons why the Derek subseries is deserving of much more thorough study than has been possible here. It is long enough to make detailed analyses possible, but not so long as to be overwhelming. It occurs in a relatively circumscribed period of time and is mixed with 334 other dream reports between October, 1994, and October, 1996. It has been coded for several Hall/Van de Castle categories, the results of which are available in the SearchCodings section of DreamResearch.net. In addition, it contains much material that might lend itself to a blind metaphoric analysis that could be compared to the coding results. Finally, the interviews with Sanders and all four friends have highly detailed commentary on the relationship. Thus, it is possible that highly focused analyses of subseries such as this one may be an excellent way to study dream meaning in great detail.
For now, though, enough has been done with the Howard and Derek subseries, and the other character studies presented in this section, to show that Barb Sanders' dream reports present an accurate portrayal of her relationships with the important people in her life. These findings support a hypothesis that derives from earlier studies of many other dream journals: people's dreams about family and friends enact their conceptions and concerns in regard to them. Such dreams are like dramas that reflect waking relationships; they are miniature soap operas.
Although this is an important and useful conclusion, it does not follow that all the other elements in these dreams are equally accurate and informative. There may be nonsense mixed in with the coherent aspects in dreams about family members and close friends. Nor does it follow from these findings that the many dreams that do not include these known characters are consistent with waking conceptions. Thus, there remains a great deal of dream content that remains to be explained.
It appears from reading through Barb Sander's dreams that she has a strong interest in theatrical performances, whether as an actor, singer, or director. When a word string containing all relevant theater-related terms is entered into DreamBank.net, it retrieves 169 dream reports from the overall series, 77 of which actually involve Sanders auditioning for, taking part in, or directing a theatrical production. This large number of theater dreams is in fact continuous with Sanders' waking interests. She acts and sings in productions, some of which she writes herself. She also very much enjoys directing theatrical performances.
In half of the theater dreams where she is involved as a performer or director, Sanders sees herself as giving an excellent performance. Sometimes she comes out of the audience to give an unexpected performance that is met with great approval. In others she is "discovered" at the last minute to be worthy of a starring role. It is clear from these positive theater dreams that Sanders has, or once had, very high hopes and ambitions. She would very much like to be in the public limelight as an esteemed artistic figure. This is shown most dramatically in a dream report where she, her middle daughter and, another woman find themselves singing on stage to great acclaim. Sanders reports that she loves the applause, but wants to do a solo, and ushers her daughter and the other woman aside:
This inference of great ambition is supported in the interview with her friend Lucy, who often performs with her:
But an equal number of the theater dreams contain rejections and misfortunes. She does not win the part, or people leave as she is about to perform. She misses a rehearsal, can't find the theater, or nearly falls off the stage. In seven instances, she forgets her lines or is afraid she will forget them, and in two others she does not have a script. These negative events are consistent with two of her waking concerns in regard to her public performances. First, she is indeed afraid she will forget her lines, as attested to by both her and her friends. Second, she does feel she is often ignored or unappreciated, a point that is stated most frankly by two of her friends.
So, just as Sanders' dreams featuring significant others enact the nature of her relationships with them, the theater dreams seem to be variations on a few of her major concerns about artistic performances. She wants to be noticed and thinks she deserves far more attention than she receives, but she worries that she will forget her lines and that people will ignore her. These concerns are the "themes" that are acted out to varying degrees in each dream relating to the theater and performances. That is, each individual dream can be seen as a specific instance from which generic information can be extracted. This instantiation of generic information may be based on resemblance metaphors, that is, figurative expressions based on same shared common property or properties with the underlying concern. If it could be shown that resemblance metaphors are operative in generating at least certain types of dreams, such as ones that are variations on a theme, it might account for a significant minority of dream reports. Since there are almost limitless possible resemblances between various elements, however, any metaphorical interpretations have to be "constrained" by information in other dreams in the series. It would be a mistake to impose Freudian or Jungian or Gestalt views on the dreams since they can make any dream fit their theories. The trick would be to show that the resemblance metaphors fit with other metaphoric expressions in the same series and with the non-metaphoric elements in the dreams, such as what is found with a Hall/Van de Castle coding for social interactions.
Dream Elements That Are Not Continuous
Although the several analyses presented so far support the continuity principle, there are some elements in Sanders' dream reports that are not continuous with her waking life. They provide an interesting exception to the continuity principle that could lead to a better understanding of dream meaning, or perhaps they may reveal the limits of the conceptual systems available to the neural network for dreaming. For example, she has several dreams about cats and kittens, but especially kittens, that are neglected, deformed, or starving. The appearance of cats fits with her interest in cats in waking life, but contrary to the continuity principle, she does not worry about the health of cats in waking life. Nor does she fear that they might starve or be neglected.
An example of how the contradictory findings on neglected cats might be approached using DreamBank can be seen through the mention of "stray kittens." Five of the 8 uses of the term "stray" occur in conjunction with kittens, a contingency with a p value of .000. Four of these instances -- in January, 1981, October, 1982, and twice in October, 1986 -- are used in reference to men who are lost souls who do not amount to much in her eyes. The fifth, which occurred in December, 1981, concerns two actual stray kittens. The dream report begins as follows, then moves to unrelated topics having nothing to do with kittens or inadequate men:
In waking life, the equation of "stray kittens" and "lost men" is understandable to most people, based on their general knowledge of the world, as a conceptual blend. They see the connection due to their instantaneous comprehension of the characteristics of stray kittens that can be applied to at least some men -- e.g., helplessness, lack of attachment to an important source -- while ignoring the kittens' irrelevant properties -- e.g., needing to be breast fed, furriness, claws, mewing. In terms of Sanders' dreams, the research question is: does her understanding of rootless men as "stray kittens" underlie her dream of actual "stray kittens?" This example merely poses the question, and in no way begins to answer it.
Unlike kittens, horses are portrayed in a very positive light in Sanders' dream reports. She has 24 dreams in the overall series where she is riding a horse. In all of these dreams, she portrays herself as an excellent rider. In one dream she rides like the wind, in another she learns quickly and becomes an excellent rider, and in another her father praises her for her riding skills. Taken together, these dreams give the impression that she learned to ride as a child and likes horses, but this is not the case. The following excerpts from the interview with her give the flavor of her reactions to inferences about her conceptions and waking experiences concerning horses:
Although the dreamer begins by saying she "loved" the few occasions where she rode a horse as a child, the general thrust of the interview contradicts the expectations based on the continuity principle. She later says that horse dreams are "no reality." The point could be stretched to say that the dreams are continuous with her wishes, but the problem is that the continuity principle would predict that her negative experiences with dreams should be reflected in the dreams as well. Later in the interview she recalled more about horses as follows:
So, beautiful horses are part of her pleasant memories, and a plausible basis for her positive dreams of riding horses. Thus, the horse dreams seem to be based on her early thoughts about horses, with little regard for her few and often negative waking experiences. Thus, the research issue is how to tell these possible wish dreams from more realistic dreams.
There is also a lack of continuity in relation to her use of guns and rifles in her dreams. There are 33 dream reports in the overall series in which she is holding or shooting a gun, always with confidence. She fixes and reloads guns, kills dangerous animals, and fights off human attackers. In one dream she captures seven men at gunpoint. In another she grabs a woman's gun and kills her. Based on the continuity principle, these positive dream actions led to the inference that she might have learned to shoot guns as a child and still enjoys doing so. However, the inference is incorrect, as seen in the following interview material:
The unexpected answers to questions about horses and guns suggest that the wrong questions may have been asked. They were not cognitively focused. If dreams stem from how a person organizes her or his knowledge, rather than actual events, then it would have been more useful if open-ended questions had been asked about her thoughts and fantasies concerning horses and guns, not about her experiences with them. The role that they play in Sanders' thinking would have to be probed more fully before moving to the level of general conceptual metaphors in an attempt to understand these dreams.
Unusual Elements in Dream Reports
The unusual elements in dreams -- distortions in familiar settings, impossible acts like flying under one's own power, and metamorphoses -- immediately come to mind when most people in Western cultures think of dreams. Although such elements are less frequent in representative samples of dream reports than popular stereotypes suggest, they do happen in 10 to 35 percent of dreams in various studies, and they once again raise the possibility that there are nonsensical aspects to dreams due to the limited capabilities of the conceptual systems available during sleep.
It also may be that some unusual elements are nonsensical and others are metaphorical, which means that distinctions among various types of unusual elements might be useful. For example, Sanders has a dream in which she and her friend Ginny are both married to Ginny's husband. This situation is obviously impossible in a cultural sense, but not in a physical sense, as when animals talk or people fly under their own power. Given Sanders' great affection for Ginny, her husband, and their children, as shown by the social interactions in the other 90 dreams she has about them, it is plausible that such a dream expresses a desire to be part of a familial relationship with Ginny and a sensual relationship with her husband. That is, it is within the context of other dreams from the same series that studies attempting to separate nonsensical from figurative dreams may be possible.
Composite characters based on two different people sometimes appear in the Sanders dream reports, usually designated by a slash ("/"), as in "Dwight/Howard," a composite of her favorite brother and ex-husband. These composite characters lend themselves to study as possible conceptual blends. In the 43 dreams in which Derek appears, for example, 8 of those appearances characterize him as "like" someone else or as a combination of himself and another person. Most strikingly, he appears as "Derek/Darryl" in three dream reports, which yokes him with the true love that she rejected for unfaithfulness at age 18. In this case, there are two pieces of information that suggest that this composite character may represent a conceptual blend. First, she said in the interview that these were the two men she loved the most. Second, they are also similar in that she never had sexual intercourse with either of them.
Interestingly, 5 of Darryl's other 31 appearances in the dream series involve comparisons or composites in addition to the 3 composites with Derek. The network of likenesses and composites centering around Derek and Darryl is presented in Figure 5. It is another example of how the study of unusual elements in dreams might proceed.
Character metamorphoses are relatively rare in dreams, occurring only 12 times in the 1000 dream reports comprising the Hall/Van de Castle normative samples. However, they often strike dreamers as remarkable and mysterious when they occur. Although it is common in waking life to say that a person is "like" someone, or for an object or event to "remind" someone of some other object or event, metamorphoses seem far removed from waking thought patterns even though they are often seen in movies and videos. They therefore provide an interesting challenge and opportunity for those who hope to find meaning in all aspects of dream content.
There are only 4 instances of character metamorphoses in the Sanders baseline 250, not enough for a systematic study:
To find a sample of metamorphoses in the entire Sanders series, the terms "changes into," "turns into," "becomes," and "is now" were entered into DreamBank.net. It is unlikely that these four terms capture all the metamorphoses in the series, but they do provide a large sample that is probably representative of the overall population of metamorphic changes in the series. The sample has the added advantage of including metamorphoses of objects, which are not coded for in the Hall/Van de Castle system. After eliminating phrases such as "the argument turns into a fight," "he becomes angry," and "the food is now ready to serve," the initial yield of 132 dream reports boiled down to 50 instances in 49 dream reports that qualified as metamorphoses.
Thirty of the 50 metamorphoses include a human or animal character at the beginning or the end of the transformation. Thirteen of these changes are human to human transformations, but in seven dream reports people change into animals, creatures, or objects, and in another seven animals or objects turn into people. These are also two occasions where one animal turns into another, and one where a male puppet turns into a female puppet. Thus, there is no one pattern involving character transformations.
A few of the changes seem to be similar to the cases of composite characters discussed in the previous section, such as when her ex-husband is now one of her brothers, or a man is now like Derek, or a woman turns into Faye Dunaway. It is also interesting that 9 cases involve babies or young children, who are obvious instances of relatively rapid transformation. However, as the following list shows, some of the metamorphoses involving babies or children seem to have more potential to be meaningful than do others:
There are also transformations of characters that might make sense in terms of adult slang:
However, there are several changes that don't seem to have any immediate plausibility, even when the context of the dream is added. For example, a yellow wooden horse turns into an artistic man, a horse becomes a cat, a very large pig becomes a piglet, and a spider becomes a miniature man, who in turn becomes a light bulb.
In addition to the 30 transformations that include animal characters, there are 20 cases where one object turns into another. Fourteen of these changes involve objects that are linked to travel or movement, which may suggest a possible underlying pattern connected to one of several conceptual metaphors. For example, "Change is Motion," as in "his mood jumped from morose to ecstatic in less than a minute," and "Processes are Movements," as in "the meeting goes for two hours." Both of these metaphors are pervasive in everyday thinking, and compatible with a metaphor closely related to issues of self, "Life is a Journey," as in "she's come to a fork in the road" and "they have one more mountain to climb."
In terms of the specific object-to-object metamorphoses in the Sander series, several of the transformations are fairly straightforward, such as when a small car becomes a big flashy car, a car becomes a house trailer, or a car becomes a hearse. Other changes involving vehicles seem less straightforward, as when a bus becomes a series of kiddie cars, a vehicle becomes 5 different wheelchairs hooked together in a row, a table becomes a circus train, or a pick-up truck becomes a tape recorder that has to be pushed up a hill There are also interesting changes that occur in passageways that may relate to "Change is Motion" or "Life is a Journey": a bridge becomes a stairway, another bridge becomes a boat, a road becomes a hallway, another road becomes a staircase, a waterfall becomes a freeway, and a hallway is suddenly the inside of a diesel truck.
Five of the remaining 6 object-to-object metamorphoses concerned household objects or wearing apparel. They involve small changes and do not immediately suggest any relation to motion metaphors: a cup becomes a bowl, a bowl of milk becomes a doily that the milk soaks through, a mirror turns into a space heater, a table becomes a room, and a scarf becomes a wet folded newspaper. The final object-to-object transformation is a familiar one from slang: a banana turns into a penis, and the dreamer has sexual intercourse with a man.
While these 50 cases suggest some potential metaphoric meanings, it is not possible to provide convincing evidence that any of these metamorphoses relate to the systematic findings presented earlier in this chapter, despite the context provided by the overall series. It will take further investigations, and perhaps new approaches, if a case is to be made for "meaning" in these elements.
This case study has only scratched the surface of the Barb Sanders series. It is a demonstration of possibilities, not a definitive analysis. However, enough has been said to illustrate that the main characters, social interactions, and activities in the dreams reveal her waking conceptions and concerns in relation to the significant people and interests in her life. On the other hand, there are elements in the dreams that do not immediately seem continuous with her waking conceptions, such as her excellent riding and shooting. These elements may reveal the limits of cognitive capabilities during sleep, or they may be the products of figurative thinking. Similarly, the unusual elements in the dream reports, ranging from composite characters to metamorphoses, may define a dimension that goes from the metaphorical to the nonsensical. The resolution of these seeming anomalies will require many further studies.
We hope you can find a way to contribute to our understanding of the Barb Sanders series. Until we all start staring at the same data and build on each other's results, there's not much hope for progress on understanding meaning in dreams.
Go back to the "Interesting Findings" index.