Printer Friendly



Most empirical research on dreams has focused on content and structure, while linguistic features have received far less attention. The present study investigated dream language in a critical developmental stage: early adolescence. Narratives of the dreams of 145 early adolescents were tape-recorded and transcribed, and the frequencies of various grammar forms and common words were calculated. The most common nouns for the entire sample were house and mother. The most frequent verbs were go and do. Males' dream narratives contained a greater number of such words as animal, long, enter, and kill. Females more often used intransitive verbs and such words as teacher, horse, and put. Several features differentiated older from younger early adolescents' dreams. The results indicate that linguistic features of dream narratives are affected by age and sex, displaying interesting parallels with clinical theories on dreams and early adolescence.

Manifest dreams have been given increasingly more consideration, both as representations (not only as the deformation of latent contents) and as meaningful stories in themselves, beyond dreamers' associations (Greenberg & Pearlman, 1978; Pulver, 1987). Analyzing dreams can produce materials related to day residues and significant childhood memories (Palombo, 1984). The Freudian dream model, with its relation between superficial and deep structure, has been compared to that of transformational linguistics (Heynick, 1981). Linguistic competence in dreams is much higher than Freud thought, since he denied that dreams could produce new propositions if not through the editing of secondary elaboration (Heynick, 1981). Nevertheless, psychoanalysts continue to place special importance on the single dream and its interpretation (Haesler, 1994). Studies on typical dreams are sparse, with the exception of dreams about falling (Saul & Curtis, 1967) and embarrassment (Saul, 1966).

Apart from psychoanalysis, research has been conducted on the manifest content of dreams (see Webb & Cartwright, 1978), and scales for measuring dream reports have been elaborated (Foulkes, 1971; Hauri, Sawyer, & Rechtschaffen, 1967). Several studies have revealed links between dreams and gender (Hall & Domhoff, 1963; Winget, Kramer, & Withman, 1972; Urbina & Grey, 1975), personality traits (Spadafora & Hunt, 1990), mental disturbances (Kramer, Withman, Baldridge, & Ornstein, 1970; Beck & Ward, 1961), and sociocultural conditions (LeVine, 1966; Roll, Hinton, & Glazer, 1974).

Dream content has been investigated from a developmental perspective (Hall & Van de Castle, 1966; Foulkes, 1971, 1982; Turpin, 1976). Children's dreams have been found to be realistic representations of their lives--their dreams being a follow-up of their waking hours and not a discontinuous experience (Webb & Cartwright, 1978). In psychoanalysis, it is acknowledged that dreams change according to age. Children's dreams reflect the interaction of drives and ego development. They represent the most pressing concerns and tasks for children at the different stages of growth. In dreams, children give symbolic or metaphoric expression to their developmental struggles (Ablon & Mack, 1980; Mahon, 1992; Greenberg & Pearlman, 1978).

As objects of analysis, dreams are tales, that is, one of the most direct and universal expressions of the human capacity to produce stories. Studying a dream as a narrative shifts attention from interpretation to analysis: (1) interpretation makes the meaning of a dream clear to the dreamer, who is trying to grasp its references to both external and internal worlds; (2) analysis identifies the tale's form, characters, theme, plot, and structure. A dream cannot be interpreted without the associations that provide references, but it still can be analyzed as a text, and its morphology and narrative structure studied.

The analysis of dream narratives can involve several facets (e.g., content, structure, language). The present study analyzed the frequency of grammar forms and words used in dream telling. Since this kind of research does not require a synthesis based on content categories, it allows generalizations that are independent of any theoretically based interpretative model.

Emphasis has been placed on the uniqueness of dreams rather than on their universality, on their referentiality rather than their textuality. In general, they have been interpreted as expressions of infantile desires or considered elaborations of the problems of waking hours. If, however, typical content is emphasized, dreams can be seen as a way of giving shape to conflicts occurring during the life cycle. The typical problems of life form the common ground activating the emotional symbolization of age-specific anxieties or situations. Therefore, the dreams of those at the same developmental stage can be studied collectively (rather than the wishes of a single dreamer), and the elements common to a subgroup of individuals sorted out.

In early adolescence, the body changes, personal autonomy grows, peer pressure is particularly important, while the relationship with parents is reconsidered and their authority reduced. In view of this, analyzing dreams during this period of life should make it easier to find meaningful age- and sex-specific linguistic elements. This research attempted to answer the following questions: (1) Which are the most common words in early adolescents' dream language? (2) Do early adolescents of different sex and age describe their dreams using different language? (3) Are there any parallels between the main features of early adolescents' dream narratives and the developmental tasks in this life stage?


Dreams were collected through interviews of early adolescents from two junior high schools. A school sample was chosen for two reasons. First, in Italy junior high school is compulsory. Therefore, students came from different social, economic, and family backgrounds, all the more so since one school was located in the center of town and the other in the suburbs. Second, since the recording of dreams was carried out during regular school hours, participation entailed minimal inconvenience. The interviews took place in an isolated classroom, sheltered from external noises and unwanted listeners.

Participants were asked to describe one or more dreams. This procedure was different from that of Hall and Van de Castle (1966) and Roccioletti, Zulu, and Bertini (1983), who collected ten dreams for each subject and randomly sampled five of them for subsequent analyses. They did so on the assumption that a subject who was requested to describe a single dream would select a particularly unusual or emotionally relevant one, and considered this to be an undesirable outcome. In the present study, however, it was felt that this phenomenon would increase the chances of identifying themes related to stage-specific developmental conflicts. In fact, clinicians agree that patients' selection of dreams is important. The first dream reported at the beginning of psychotherapy is thought to be predictive of the entire course of therapy. An interesting study on panic attacks has recently used single dream collections as a useful source of information (Free, Winget, & Withman, 1993).

The tape recording of dreams was chosen because it lets unconscious communicative qualities prevail (e.g., slips of tongue, momentary lapses of memory, unnecessary repetitions). Writing them down, on the other hand, requires linguistic competence and skills that might allow secondary processes to take over.

Data Collection

The first of two meetings with subjects took place in their classrooms, where research methods and goals were explained and questions answered. It was emphasized that participation was voluntary, and confidentiality and anonymity were assured. They were advised that dream telling would be tape-recorded, and only interviewer and interviewee would be present. It was stressed that teachers were not members of the research team and that the collected material would not be available to them. Students were told that dreams should be recent and described in detail. A single dream would be enough, but others could be included. A week later, interviews were begun among those who agreed to participate in the study.


One hundred forty-five early adolescents (74 males and 71 females) participated. They were divided into three age groups by sex. There were 57 (28 females and 29 males) 11-year-olds, 45 (22 females and 23 males) 12-year-olds, and 43 (21 females and 22 males) 13-year-olds.

In order to make sure that dream language differences among the six groups were not due to social or demographic background, educational and professional achievements of parents, social adjustment, peer relations, and school performance and adjustment were analyzed. No statistically significant differences were detected.

Analysis of Data

The tape recordings of the dream narratives were transcribed following the rules suggested by Mergenthaler and Stinson (1992) in order to obtain texts suitable for computer-assisted analysis. The frequency of each word in the dream narratives was counted by means of a software program. Since the program identifies words according to their spelling and not their semantic value, different rules for each part of speech had to be defined before starting the analysis of data.

All noun forms (singular, plural, feminine, masculine, which in Italian have different endings) were grouped together. The same criteria were applied to possessive adjectives and pronouns, which in Italian end differently according to the sex and number of the object/person they refer to.

All verbs were reduced to the corresponding infinitive form, including the past participle, even though in some instances this tense form could be used as an adjective or a noun. This choice was not likely to cause any bias in the data, considering that sometimes only the presence of an agent makes the passive voice of a verb different from a participle working as an adjective. However, whenever the same verbal ending might have implied different semantic values, the related words were grouped according to their semantic meaning. For instance, partita, the female past participle of the verb partire (leave) was not grouped with the noun partita (match), with which there is no semantic relation. Of course, auxiliary verbs were not taken into account.

Whenever the same word could be used as a different part of speech (e.g., verb, adjective, noun, adverb), its context determined assignment. For instance, the Italian word solo was put with the group of adverbs, adjectives, or conjunctions according to its different meanings: only, alone, or yet. Indefinite and qualifying pronouns and adjectives, as well as articles, adverbs, and conjunctions, were not considered.

Data were analyzed from two perspectives: prevalence of the various grammar forms--transitive, intransitive, and reflexive verb forms, along with the total frequency of verbs, nouns, and adjectives, and person and number of pronouns; and prevalence of particular words--the ten most frequent verbs, nouns, and adjectives in each group of early adolescents. To avoid inclusion of idiosyncratic terms, words used by less than three subjects were not included. Seventy-nine words were selected for further analysis: 33 nouns, 28 verbs, and 18 adjectives.

This study focused on the effects of sex and age on the language in early adolescents' dream narratives. Sex differences were investigated by dividing the dreams of males and females into two large texts. These texts were then analyzed by age group. The frequencies of each word in the six subgroups were examined using a common chi-square test. For significant results, an additional chi-square test was used to assess sex differences, and age differences were assessed by means of the Brandt test.


Grammar Forms

Analysis of the number of personal and possessive pronouns (see Table 1) indicated that males' dream narratives included a higher number of the first person singular than did females' narratives. Thirteen-year-old males did not use the third person singular and first person plural as much as younger males did. For females, the frequency of the third person singular was higher for 11-year-olds than for older females.

The prevalence of transitive, intransitive, and reflexive verbs are reported in Table 2. Male students did not use as many intransitive verbs as did females. Further, for males, the frequency of transitive verbs decreased with age. Females' dreams showed the opposite trend.

Males' dream narratives included more verbs than did females' narratives, especially for 13-year-olds (See Table 3.) For both sexes, the total number of verbs increased with age.

Most Frequent Words

The ten most frequent nouns and verbs for the entire sample are listed in Tables 4 and 5. Tables 6, 7 and 8 show the differences in the prevalence of selected nouns, verbs, and adjective between males and females and different age groups.

Nouns. The noun teacher was significantly more frequent for females than for males, who hardly ever mentioned teachers. The words cousin and horse were also more frequent in females' dream narratives. The noun animal was more frequent in males' dream narratives.

A number of nouns varied by age group. Thirteen-year-old males used animal, window and foot less frequently than did 12-year-olds. The opposite was true for time and people. The nouns person and uncle showed a peak prevalence for 12- and 11-year-old males, respectively.

Female 11-year-olds used the words father, foot and horse more often than did 12- and 13-year-olds. Staircase, person and people and people were more frequently mentioned by older females, while use of the world cousin peaked for 12-year-olds.

Verbs. The following verbs were more frequently used by males than by females: wake up, feel I hear, enter, and kill. The verbs must and put were more often used in females' narratives.

The frequency of the verbs put, go, want, take and do decreased with age in males' narratives. Three verbs (wake up, feel / hear and seem) showed the opposite trend, increasing for 12- and 13-year-old males.

The verbs go, must and take were more frequent for 12-year-old females than for 13-year-olds. Do and turn were significantly more frequent for 11-year-old females than for 12- and 13-year-olds. The verbs feel / hear and look for were more frequently used by 13-year-old females. Eleven-year-old females infrequently said that they dreamed they were coming from somewhere.

Adjectives. The analysis of adjectives revealed few significant difference. Long was more often used by males than by females, and the frequency of fast increased with age for males.


The most frequently used nouns can be divided into groups: family members and friends (mother, father, brother / sister, friend), general terms (point, person, time), and objects (house, school, car). Verbs can be divided as follows: verbs having a semantic relation with sleep (dream, wake up), verbs of action (go, do, stay), verbs describing competence and will (know, want, be able to), and descriptive verbs (see, say). It would be interesting to compare the frequency of these words in dream narratives with their frequency in other narratives. The high number of references to family members, as well as the significant presence of the word friend--relationships that are particularly important for early adolescents --is notable.

It is significant that the word mother was used more often than father. This is consistent with findings from psychosocial research, which have suggested that the mother represents the central relational object for both males and females, even though she is also an identification object for females. In addition, the three age groups were homogeneous; no significant variations were found in the frequency of mother. Another frequent noun was house, which underscores the importance of the home as a context for dream events.

With the exception of dream and wake up, verbs that are usually important in a narrative were frequent. It would be interesting to compare the frequency of the verb see, here an expression of the way to perceive dreams, with its frequency in other narratives.

The most frequent words in the present sample and contents indicated by Freud as typical in dreams had the presence of parental relationships in common, but not sexual relationships. There were references to death, but not to birth.

It should be emphasized that the groups differed significantly in the frequency of several words and grammar forms. The larger number of first person singular pronouns in males' narratives seems to indicate poor narcissistic development. A lower frequency of intransitive verbs for males seems to indicate a poorer relation to objects: males use "we do" rather than "I am related to someone."

The choice of words by males also suggests impulsiveness and aggression. Verbs such as kill and enter imply dreams with aggressive content, which frequently leads to wakening.

For females, more verbs expressed a sense of duty and there were more references to school and interpersonal relations (e.g., teacher). In females' dream narratives, the word cousin was noted to increase in parallel with a decrease in the word father. This might express a widening of the relational circle: a cousin is in fact a figure in between family and extrafamily circles, a possible forerunner of the sexual object.

With increasing age, use of action verbs decreased (for males: put, go, want, kill, take, do; for females: go, take, do, turn). Verbs describing feelings and perceptions increased (males: wake up, feel/hear, seem; females: feel/hear, look for).

Females used words representing typical Freudian symbols, such as horse and staircase. This may suggest a symbolic prefiguration of a proprioceptive sexual experience, an expression of a precognition of sexual relationships. When dreams do not find any reference to experience, either in the past or in the present, they produce symbols. It is difficult to assess the value of adjective frequencies, but perhaps the word long, used more often by males, has symbolic meaning.

In sum, the present research confirms the usefulness of a linguistic analysis of dream narratives. Results indicate that the language used by early adolescents to describe their dreams differs significantly by sex and age. This may reflect differences in underlying focal conflicts.

Although the findings are consistent with widely accepted theories of adolescent development, other factors may have been at play. Differences in linguistic background, or in expressive or cognitive skills, may have influenced the results. Different rates of dream-related physiological development may have also been a factor.

There were two additional limitations of the present research. First was the lack of a control group of adolescents or of latency-stage children. Statistical analyses referred only to sex and age differences in the dream language of early adolescents. Whether the findings are specific to early adolescence or if they apply to other developmental stages cannot be stated.

Further, the study was conducted in a specific social and cultural context: a large metropolitan area in northern Italy. Only similar studies carried out in different geographical areas could make it clear whether the results are generalizable.

Additional investigation into the multiple dimensions of dreams, such as emotions and object relations, are necessary. Nevertheless, a quantitative analysis of the language used to describe dreams provides useful information through the linguistic choices made by the dreamer/narrator.

Paolo Azzone, M.D., Salvatore Freni, M.D., and Daniele Vigano, M.D., Institute of Psychiatry, University of Milan.

Alfio Maggiolini, Ph.D., and Katia Provantini, Ph.D., Institute of Psychology, University of Milan.

Reprint requests to Professor Salvatore Freni, M.D., Istituto di Clinica Psichiatrica, Universita degli Studi di Milano, Via Francesco Sforza 35, 20122, Milano, Italy.


Ablon, S. L., & Mack, J. E. (1980). Children's dreams reconsidered. Psychoanalytic Study of the Children, 35, 179-216.

Beck, A. T., & Ward, C. H. (1961). Dreams of depressed patients: Characteristic themes in manifest content. Archives of General Psychiatry, 5, 462-467.

Foulkes, D. (1971). Longitudinal studies of dreams in children. In J. Masserman (Ed.), Science and psychoanalysis. New York: Grune & Stratton.

Foulkes, D. (1982). Children dreams: Longitudinal studies. New York: Wiley-Interscience.

Free, N. K., Winget, C. N., & Withman, R. M. (1993). Separation anxiety in panic disorders. American Journal of Psychiatry, 150, 595-599.

Greenberg, R., & Pearlman, C. (1978). If Freud only knew: A reconsideration of psychoanalytic dream theory. International Review of Psycho-Analysis, 5, 71-75.

Haesler, L. (1994). Le role du reve dans la pratique psychanalytique aujourd'hui. The current role of dream in psychoanalytic practice. Psychanalyse en Europe, 43, 6-31.

Hall, C. S., & Domhoff, B. (1963). A ubiquitous sex difference in dreams. Journal of Abnormal and Social Psychology, 66, 278-280.

Hall, C. S., & Van de Castle, R. L. (1966). The content analysis of dreams. New York: Appleton-Century-Crofts.

Hauri, P., Sawyer, J., & Rechtschaffen, A. (1967). Dimensions of dreaming: A factored scale for rating dream reports. Journal of Abnormal Psychology, 72, 16-22.

Heynick, F. (1981). Linguistic aspects of Freud's dream model. International Review of Psycho-Analysis, 8, 299-314.

Kramer, M., Withman, R. M., Baldridge, B., & Ornstein, P. H. (1970). Dream content in male schizophrenic patients. Diseases of the Nervous System, 31, 51-58.

LeVine, R. (1966). Dreams and deeds: Achievement motivation in Nigeria. Chicago: University of Chicago Press.

Mahon, E. J. (1992). Dreams: A developmental and longitudinal perspective. Psychoanalytic Study of the Child, 47, 49-65.

Mergenthaler, E., & Stinson, C. H. (1992). Psychotherapy transcriptions standards. Psychotherapy Research, 2, 125-142.

Palombo, S. R. (1984). Deconstructing the manifest dream. Journal of the American Psychoanalytic Association, 32, 405-420.

Pulver, S. E. (1987). The manifest dream in psychoanalysis: A clarification. Journal of the American Psychoanalytic Association, 10, 99-117.

Roccioletti, G., Zulli, A. M., & Bertini, M. (1983). Il sogno nell'eta evolutiva: Una ricerca sul contenuto manifesto. The dream in developmental age: Research on manifest content. Rome: Bulzoni.

Roll, S., Hinton, R., & Glazer, M. (1974). Dreams and death: Mexican-Americans vs. Anglo-Americans. Interamerican Journal of Psychology, 8, 111-115.

Saul, L. J. (1966). Embarrassment dreams of nakedness. International Journal of Psycho-Analysis, 48, 552-558.

Saul, L. J., & Curtis, G. C. (1967). Dream form and strength of impulse in dreams of falling and other dreams of descent. International Journal of Psycho-Analysis, 48, 281-287.

Spadafora, A., & Hunt, H. T. (1990). The multiplicity of dreams: Cognitive-affective correlates of lucid, archetypal and nightmare dreaming. Perception and Motor Skills, 71(2), 627-644.

Turpin, E. (1976). Correlates of ego-level and agency-communion in stage REM dreams of 11- to 13-year-old children. Journal of Child Psychology and Psychiatry, 17, 169-180.

Urbina, S., & Grey, A. (1975). Cultural and sex differences in the sex distribution of dream characters. Journal of Cross-Cultural Psychology, 6, 358-364.

Ward, C. H., Beck, A. T., & Rascoe, B. (1961). Typical dreams. Archives of General Psychiatry, 5, 116-125.

Webb, W. B., & Cartwright, R. D. (1978). Sleep and dreams. Annual Review of Psychology, 29, 223-252.

Winget, C., Kramer, M., & Withman, R. (1972). Dreams and demography. Canadian Psychiatric Association Journal, 17, 203-208.
COPYRIGHT 1998 Libra Publishers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Azzone, Paolo; Freni, Salvatore; Maggiolini, Alfio; Provantini, Katia; Vigano, Daniele
Geographic Code:1USA
Date:Mar 22, 1998
Next Article:Child Welfare Research Review (Vol. 1).

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters