Domhoff, G.W. (1996). Finding meaning in dreams. A quantitative approach. r.2 The Hall-Van de Castle System.pdf

Finding Meaning in Dreams: Chapter 2

Strona 1 z 22

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 2: The Hall/Van De Castle System of Content Analysis

Introduction

The first task for a dream content coding system, as for any content analysis, is the formulation of

a set of categories encompassing the relevant aspects of dream reports in a reliable and useful way.

Abstractly speaking, the categories in a system of classification can be either theoretical or empirical.

A theoretical category is one derived from a theory of personality and applied to a dream report. For

example, a category derived from Jung's theory of archetypes or Freud's theory of the castration

complex is a theoretical category. An empirical category, on the other hand, is one developed in a

trial-and-error fashion from a reading of numerous dream reports with no theoretical intentions in

mind. "Friends," "aggressive interactions," "physical activities," "misfortunes," and "successes" are

examples of empirical categories in the Hall/Van de Castle system.

In practice, however, some categories or groups of categories are both theoretical and empirical.

In the Hall/Van de Castle system, for example, aggressions are categorized in terms of aggressor and

victim, and friendly interactions are classified in terms of who initiated or received the friendly act

(befriender and befriended). These are strictly empirical categories. However, a score combining the

percentage of initiated aggressive and friendly encounters might be designated as a measure of

extroversion, a theoretical category (Van de Castle, 1969:189).

In this book only empirical categories will be utilized because there are more of them and they

have proved to be more interesting and useful. Moreover, theoretical categories are far more difficult

to develop, not only because the investigator must be steeped in the theory he or she wishes to

utilize, but also due to the fact that even proponents of the theory may disagree about which

empirical elements in the dream report are embodiments of the theorist's claims. In other words, both

theory interpretation and dream interpretation are difficult enterprises.

The problems with theoretical scales can be seen in the short history of the castration anxiety,

castration wish, and penis envy scales Hall and Van de Castle attempted to develop on the basis of

psychoanalytic theory. They defined castration anxiety as (1) an injury or threat to the dreamer's

body; (2) an actual or threatened injury or loss to any animal or object belonging to the dreamer; (3)

the inability of the dreamer to use a gun, piece of machinery, or vehicle; or (4) a male dreamer

changing into a woman, acquiring a woman's secondary sexual characteristics, or wearing women's

clothes (Hall and Van de Castle, 1966: 126-130). Castration wish was operationalized in terms of the

file://E:\prace\artykuþy\sny\ksiazki\finding_meaning_in_dreams\Finding Meaning in ...

2009-02-11

Finding Meaning in Dreams: Chapter 2

Strona 2 z 22

same four criteria, except they happen to another person in the dream report. Penis envy was

indexed by (1) acquisition of an object with "phallic" characteristics; (2) admiration of a man's

physical characteristics or possessions; or (3) a female dreamer turning into a man or acquiring male

secondary characteristics (Hall and Van de Castle, 1966:131-133).

In an initial study, Hall and Van de Castle (1965) found in accord with psychoanalytic theory that

significantly more males than females displayed castration anxiety, whereas significantly more

females expressed castration wish and penis envy. However, when they tried to replicate the study

with a larger sample as part of their normative study, they found more penis envy in men's dreams

(25 dreams in 500 dreams) than in women's dreams (13 dreams in 500 dreams). The small number of

occurrences for both genders is as damning for the usefulness of the categories as the unexpected

findings. The differences on the Castration Wish Scale also were reversed in the normative study,

with 45 such dreams for males and 39 for females. Only the difference on Castration Anxiety was

replicated.

To make matters worse, the only other use of one of these scales suggested it was measuring

power disparities between males and females, not the castration complex. Women living in

traditional preindustrial societies with high male dominance scored higher on the alleged penis envy

scale than women in more egalitarian traditional societies (Nathan, 1981), not a finding expected

from psychoanalytic theory. This study, combined with Hall and Van de Castle's inability to

replicate their earlier study, leads to the conclusion that there is no future for these theoretical scales.

The only positive note coming out of the effort is the reassurance that the coding system was

objective enough to lead to results contrary to the expectations Hall and Van de Castle developed

from their original study.

The empirical categories in this book were formulated after reading and studying a large number

of dream reports. No class of items appearing fairly frequently was omitted. On the other hand, some

classes with low frequencies were included because of their potential psychological significance or

to give completion to a larger set of classes of which the small class was one part.

The first step in creating good and useful categories is ensuring that their scope and limits are

defined clearly so there can be a high degree of agreement between coders as to what elements in a

dream report should and should not be included in a given category. No matter how psychologically

significant a category may be potentially, it is useless and has to be discarded if there is no intercoder

reliability.

Developing reliable categories, as noted earlier, is not an easy task. It only can be accomplished

by painstaking trial and error, and then examples must be developed to aid new investigators in using

the system. Even with all this effort, there are always gray areas at the boundaries and new elements

that do not fit the coding system. Difficult coding decisions and elements that cannot be coded are

often frustrating, especially for newcomers and those who do not realize that there is an arbitrary

element in any classificatory system. Difficult coding decisions, combined with the considerable

amount of time it takes to use the system, are among the reasons why some investigators prefer to

use quicker and more subjective global rating systems that assign a single score to a dream report.

Such problems cause others to drop quantitative content analysis altogether in favor of qualitative

analyses based upon traditional dream theories or metaphoric understandings common to all of us.

There are no established rules or guidelines for constructing empirical categories. That is, there are

many different ways a set of elements can be organized. Hall and Van de Castle tried to create

categories that seemed to fit with everyday conceptions and at the same time would prove to have

psychological significance. Whether their judgment was good or bad should be decided by the

results obtained using the system, and by comparing their system with those created by other

investigators.

file://E:\prace\artykuþy\sny\ksiazki\finding_meaning_in_dreams\Finding Meaning in ...

2009-02-11

Finding Meaning in Dreams: Chapter 2

Strona 3 z 22

Although the system to be presented in this chapter is in a certain sense arbitrary, we believe

anyone starting out to devise a set of empirical categories based on reading thousands of dream

reports would end up with a very similar system. In fact, we will present evidence to this effect later

in the chapter when we compare the Hall/Van de Castle system with coding schemes developed by

other investigators. This similarity of coding systems is due to the fact that most dream reports have

a setting or series of settings, and one or more characters who think, act, interact, and express

emotions. Dreams, in other words, have some resemblance to plays.

In principle, categories can be either "nominal," standing by themselves, or "hierarchical,"

meaning each category is a point along a general continuum. Hierarchical categories involve ratings,

rankings, or the assignment of "weights." They represent different degrees of a phenomenon: more

or less, greater or smaller, stronger or weaker. They assume one category is more or less than any

other category. A hierarchical scale is called a "rank scale" or "ordinal" scale if the categories are

only arranged from highest to lowest, with no implication that they are equal distances from each

other. If all categories are equally distant from each other, the scale is called an "equal interval"

scale. If a scale has an exact zero point, such as weight does, as well as equal intervals, it is called a

"ratio" scale.

There are serious difficulties with hierarchical scales in studying dream reports. For example,

when rating scales are based on subjective estimates concerning "vividness," "confusion," or some

similar dimension, then they are often less reliable because coders have to make difficult

comparative judgments. When "weights" are assigned to different elements arranged along a scale,

there are often implicit assumptions difficult to justify in psychological terms. In one hostility scale,

for example, a death is coded a "4," an injury is coded a "2," and discomfort is coded a

"1" (Sheppard, 1964). By this reckoning, two injuries or four discomforts have the same "weight" as

a death. In one dependency scale, one point is given if partial reliance on another individual is

mentioned, but six points are given if there is a specific reference to food or eating (Whitman et al.,

1961). This assumes that mentioning a sandwich shows six times as much dependency as accepting a

helping hand from another (cf. Van de Castle, 1969:193, for an elaboration of this critique).

To avoid these problems, all the scales in the Hall/Van de Castle system are nominal ones. They

assign no ranks or weights. Instead, they compare various discrete categories as equals and aggregate

discrete categories for overall scores. The eight categories of the Hall/Van de Castle "scale" for

aggressions provide an example of this point. Each aggression in a dream report is tabulated into

categories ranging from (1) covert feelings of hostility to (2) verbal criticism to (3) rejection or

coercion to (4) verbal threat of harm to (5) theft or destruction of a person's possessions to (6)

chasing, capturing, or confining to (7) attempts to do physical harm to (8) murder. The frequencies

for each category can be compared with norms. Categories one through four, the nonphysical types

of aggression, are summed, as are categories five through eight, the physical aggressions. Finally, all

categories of aggressions can be totaled for an overall aggression score. There is no information lost

in this system, and there are no assumptions made about how much "stronger" or "heavier" or

"weightier" one aggression is compared to another.

In closing this introduction to the Hall/Van de Castle system of content analysis, it cannot be

stressed enough that empirical content scales such as are about to be presented yield "findings," but

not explanations for those findings. That is, it is essential to make a distinction between "findings"

and "interpretations." For example, as will be shown in chapter 4, there are some gender differences

in the dream reports of American college students. However, there is nothing in these findings that

tells us whether the differences are cultural, psychosocial, biological, or some complex combination

of these factors. Unlike the situation with good theoretical scales, where correct predictions are in

effect "interpreted" by the theory, it is necessary to search further for interpretations of empirical

findings. Sometimes plausible interpretations can be found by looking at results with other empirical

categories; sometimes they can be found in comparisons with other individuals or groups. We will be

making very few interpretations in this book, and none of the interpretations we do make is tied to

file://E:\prace\artykuþy\sny\ksiazki\finding_meaning_in_dreams\Finding Meaning in ...

2009-02-11

Finding Meaning in Dreams: Chapter 2

Strona 4 z 22

any of the classical dream theories derived from clinical practice. Our approach may be useful to

those researchers who want to test hypotheses derived from clinical theories, but this book is not

based upon or beholden to any of them.

We now turn to an overview of the coding system so readers can gain a general sense of its

breadth and depth. The emphasis will be on definitions, rationales, and highlights because the system

can appear overwhelming in its intricacy and detail when it is first encountered. A complete

presentation of the system and how to use it accurately can be found in Appendix A, where each

coding rule is followed by examples of codeable and noncodeable items. As a further aid to those

who want to use the coding system for their own research, Appendix B presents Hall and Van de

Castle's original coding of every codeable element in ten dream reports from a young adult male.

The Hall/Van de Castle system consists of ten general categories, most of which are divided into

two or more subcategories. The ten general categories are:

Characters

Social Interactions

Activities

Striving: Success and Failure

Misfortunes and Good Fortunes

Emotions

Physical Surroundings: Settings and Objects

Descriptive Elements

Food and Eating

Elements from the Past

We begin our exposition of the system with the characters category.

The Classification and Scoring of Characters

The characters category consists of people, mythical figures, and animals. These three general

types of characters are then broken down in a number of different ways. First, all three general types

can appear as individuals or groups. Thus, "a woman," a "Greek god," or "my favorite kitten" is one

character, but so is "a group of women," "three goddesses," or "a litter of kittens." If characters were

not differentiated into individuals and groups, problems would arise about estimating group sizes,

and groups consisting of what the dreamer thought to be dozens or hundreds of people could

introduce serious distortions into a simple frequency count of individual characters. Still, we can

make differentiations from within a group that allow for individuality. If the dreamer reports that he

or she was "talking with a group of women, one of whom was my mother," we code for both a group

(the women) and the individual character who was singled out (the mother).

For human and mythical characters, there are three additional general categorizations: gender,

identity, and age. In the gender category, individual characters can be male, female, or "indefinite,"

with indefinite meaning gender is not identified in the dream report. Groups of characters can be

male, female, indefinite, or of both genders (that is, identified in the dream report as a group

containing both men and women).

The identity classification of characters concerns such factors as relationship to the dreamer,

occupation, ethnicity, and the prominence or celebrity of characters. For example, characters can

file://E:\prace\artykuþy\sny\ksiazki\finding_meaning_in_dreams\Finding Meaning in ...

2009-02-11

Finding Meaning in Dreams: Chapter 2

Strona 5 z 22

have the following relationships to the dreamer: immediate family, other relatives, known

characters (friends, acquaintances, classmates), and strangers (unknown to the dreamer).

The list of occupational identifications is of course very large. So is the list of ethnicities.

"Prominent" characters is a category for famous people who are known to the dreamer by general

reputation, but not personally (e.g., leading political figures, famous entertainers, and sports stars).

Fictional, dramatic, imaginary and supernatural characters are also included as "prominent"

characters because they are familiar to the dreamer by their reputation (e.g., cartoon characters,

mythical figures, characters in TV serials). Finally, there is a residual identity coding category called

"uncertain" for those characters whose identities cannot be established with certainty from the dream

report.

Age is the fourth and final general category after individual/group, gender, and identity. The four

age categories are adult, teenagers, child (ages 1-12, or referred to as a child in the dream report),

and baby (under age one or referred to as a baby or infant).

There is one major coding issue with regard to the character categories: deciding when to code for

the presence of a character. After trying several alternatives, Hall and Van de Castle found they

could achieve the highest reliability and make the maximum use of all the information in the dream

reports if there is a coding for characters even if they are only mentioned ("I told him my dad was a

good person."), referred to (e.g., "I told him about my sister's new car."), or thought about ("I

wondered how my brother was doing."). Conversely, there are conventions precluding the coding of

characters mentioned in a generic sense (e.g., "anyone can sing," "dogs are friendly animals") or

referred to in order to establish that they were not in the dream (e.g., "it was not my older son, but

my younger one"; here the "older son" is not coded).

As Appendix A explains, there is a detailed notational system for describing each character in a

dream report. "A man I did not know" is a 1MSA, that is, an individual character (1), a male (M), a

stranger (S), and an adult (A). "A little baby girl that I knew" would be a 1FKB, that is, an individual

character (1), a female (F), a person known to the dreamer (K), and a baby (B). In other words, a

coding is assigned to each human and mythical character for each of the four general categories in

the order the categories were introduced in previous paragraphs.

Each immediate family member has her or his specific coding symbol, but we will leave most of

the details for the appendix. However, we note here that 1MFA is a "father" and 1FMA a "mother"

to show how important sequence is. That is, an M in the second position in the sequence means

"male," but an M in the third position means "mother;" an F in the second position means "female,"

but an F in the third position means "father." Such little details of the coding system are slowly

learned as one works with it. Once learned, the efficiency and usefulness of the system became

apparent. Animals, incidentally, have one generic coding symbol for all of them, which is "ANI."

They are coded only for individual or group. Thus, two spiders, three bears, or a herd of buffalo all

would be coded 2ANI, and nothing more.

In all, the entire character system is only slightly more detailed than the summary we have

provided here. The complications caused by dead characters, imaginary characters, and characters

who change into another character in the dream report are discussed in the appendix.

The Classification and Scoring of Social Interactions

With the cast of characters introduced, attention turns to the unfolding of the play. Lines are

spoken and the plot develops. The relative emphasis given to dialogue as contrasted with actions will

depend upon the author of the dream and the message he or she wishes to express. A character's

remarks may serve to insult, flatter, or "proposition" another character, or a character may act by

file://E:\prace\artykuþy\sny\ksiazki\finding_meaning_in_dreams\Finding Meaning in ...

2009-02-11

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: