Title Page
 Table of Contents
 Review of the literature
 Summary and discussion
 Appendix A: Biographical material...
 Appendix B: Paired subjects with...
 Appendix C: Informed consent...
 Biographical sketch

Group Title: relationship between selected personality factors and cancer
Title: The relationship between selected personality factors and cancer
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00099499/00001
 Material Information
Title: The relationship between selected personality factors and cancer
Physical Description: vii, 61 leaves : ; 28 cm.
Language: English
Creator: Rivera, Diana L. B ( Diana Luise Blanchard ), 1944-
Publication Date: 1983
Copyright Date: 1983
Subject: Cancer   ( lcsh )
Personality   ( lcsh )
Counselor Education thesis Ph. D
Dissertations, Academic -- Counselor Education -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Thesis: Thesis (Ph. D.)--University of Florida, 1983.
Bibliography: Bibliography: leaves 57-60.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Diana L.B. Rivera.
 Record Information
Bibliographic ID: UF00099499
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000450206
oclc - 11437315
notis - ACL1874


This item has the following downloads:

PDF ( 2 MBs ) ( PDF )

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
    Review of the literature
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
    Summary and discussion
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
    Appendix A: Biographical material on David Bakan
        Page 50
    Appendix B: Paired subjects with criteria used for pairing
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
    Appendix C: Informed consent form
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
    Biographical sketch
        Page 61
        Page 62
        Page 63
        Page 64
Full Text







Copyright 1983

Diana L. B. Rivera

To Lisa


Grateful acknowledgement is given for the contributions

of my supervisory committee: to Dr. Ted Landsman, for his

careful and caring criticism of my work and for his leader-

ship in this project; to Dr. Janet Larsen, for her valuable

suggestions and for encouraging me with her belief in the

value of this undertaking; to Dr. James Joiner, for his

accessibility and support; to Dr. John Saxon, for his advice

on research design and methodology; and to Dr. Roy Weiner,

who made it easy, for the warmth with which he received a

visitor to his field of knowledge.

Thanks go also to the University of Florida Medical

Oncology faculty, to Joann McGhin for making space avail-

able, and to Loretta Robinson, who typed the manuscript.

Finally, with special gratitude, I thank those who

participated as subjects.



ACKNOWLEDGEMENTS ........................................... iv

ABSTRACT ............... .....................................vi

ONE INTRODUCTION ....................................... 1

Agency ................................................ 2
Communion ...................................... .....3
Development of the Agentic Personality.............4
The Freudian Theory of Cancer ...................... 6

TWO REVIEW OF THE LITERATURE. ..........................10

Sexual Maladjustment.............................. 11
Inhibition of Aggression ..........................13
Loss of a Significant Person ..................... 15
Lack of Social Involvement ........................16
Inhibition of Maternality. .........................18
Repression......................... .... .............19
The Structure of Personality ......................19

THREE METHOD ............................................. 21

Subjects .......................................... 22
Using the PRF to Measure Agency ....................25
Using the PRF to Measure Communion................28
Validity and Reliability of the PRF................31
Procedure .............................................35

FOUR RESULTS........................................... 37

FIVE SUMMARY AND DISCUSSION ............................ 44

Conclusions....................................... 44
Impact of the Study............................... 46
Limitations of the Study. ............ ..............47
Directions for Further Research....................48




C INFORMED CONSENT FORM ............................. 56

REFERENCES ........................ .. .... ..................57

BIOGRAPHICAL SKETCH ..................... ...................61


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Diana L. B. Rivera

August, 1983

Chairman: Ted Landsman, Ph.D.
Major Department: Counselor Education

This study investigated a relationship between person-

ality and cancer based on Bakan's concepts of Agency and

Communion. Agency refers to that aspect of personality

motivated by the need for individuation and separation;

Communion refers to that aspect of personality motivated by

the need for unity and belonging. Thirty cancer subjects

were compared with a matched group of 30 cancer-free sub-

jects on performance on selected scales of the Personality

Research Form. Agency/Communion ratios were constructed

for each subject by dividing total score on scales fitting

the definition of agency by total score on scales fitting

the definition of Communion. The cancer group was found

to score significantly higher on Agency/Communion ratio than

the non-cancer group. Significant differences between the

two groups were also found on three of the individual scales.

These findings support Bakan's theory of an association

between Agency and cancer. They further help to clarify

the concept of a possible cancer-related personality type.


This is a study in personality theory. Its purpose is

to provide empirical evidence for or against a hypothesis

that the development of cancer is partly a function of

personality. Results of this study will impact the field of

personality, the continued research of cancer as a psycho-

physiological disorder, and the area of psychological inter-

ventions for the prevention and treatment of cancer.

Conflict theory, the view of personality and behavior

as products of global opposing forces, is the foundation for

the hypothesis. Freudian psychoanalytic theory is a psycho-

social version of the conflict position. David Bakan

(Appendix A), who hypothesizes a relationship between

personality and cancer, takes the intrapsychic conflict

position (Maddi, 1968). This position holds that the failure

to achieve and maintain a dynamic balance between the two

major life tendencies--essentially the tendency toward in-

dividuation and the tendency toward integration--leaves the

organism in a state of instability which, depending upon

direction and degree, causes or allows to occur every kind

of malfunction from neuroticism to physical illness. Bakan

contends that cancer is one manifestation of imbalance

between the two forces, which he calls Agency and Communion.



An agent is a power that acts; agency, the facility or

state of acting or exerting power. These words derive from

the Latin agere: to drive, lead, act, do (Webster's New

Collegiate Dictionary, and The American College Dictionary.)

David Bakan (1966) uses Agency, and his own adjective

form, agentic, to refer to the intrinsic force toward au-

tonomy and individuation hypothesized by himself and other

personality theorists such as Freud (1920/1950), Rank

(1929), and Angyal (1941). The connection between the

dictionary definition of agency and this drive in the di-

rection of self and separation is perhaps clarified by the

Rankian view that every independent action is both a sepa-

ration and a self-assertion. Choosing to act--to be an

agent--is itself a differentiation of the individual from

other people and from the physical environment. Bakan's

selection of Agency rather than a more obvious word, such as

Angyal's autonomy, seems to be partly for the purpose of

designating a principle more general than most of us are

used to thinking about. Agency is not a human prerogative.

Agency is a basic characteristic of living matter that mani-

fests itself on all levels of organization: in the politics

of societies, the behavior of organisms, the functioning of

tissues, the separation and differentiation of cells.

Agency characterizes existence as an individual, yet

at the same time points to the futility, indeed the non-

viability, of the individual in and for itself. The parts

of an organism must cooperate or the organism will die. A

person's life must merge with the lives of others or suc-

cumb to meaninglessness and absurdity. Bakan quotes Hillel:

"If I am not for myself, who will be for me?" This speaks

of Agency. "But if I am only for myself, what am I?"

(Bakan, 1966, p. 14.)


The other basic directionality of life Bakan calls

Communion. Communion comes from the Latin communion, mean-

ing mutual participation (Webster's New Collegiate Diction-

ary.) It signifies the participation of the individual with

a greater whole, a union with others and the world requir-

ing loss of self and self-consciousness. While this aspect

has its negative possibilities, as indicated by Otto Rank's

appellation "fear of life," Bakan stresses its positive

nature. Nowhere does he indicate the notion that the in-

dividual might be so overtaken by the communal force as to

be totally subsumed. One must assume that he views

Communion as unable by definition to be a domineering force.

Communion is that which mitigates against domination. Its

function seems to be that of bringing the individual into

harmony with its context, thereby creating a higher and more

meaningful level of organization.

The two forces, Agency and Communion, are in one sense

inexorably antagonistic. In another sense, they are compli-

mentary. Through a developmental process of differentiation

and integration we become living organisms, develop person-

alities, form social systems and ecosystems. We resolve the

conflict of our dual inheritance only to have it reappear

with increasing complexity -- and thus we grow.

Development of the Agentic Personality

Bakan is not specific about the development of a person-

ality dominated by Agency. Among theorists who share Bakan's

intrapsychic conflict position, the clearest statement of

personality development is provided by Otto Rank (1929,

1945). Both theorists postulate a personality which ex-

presses individuation (Agency) at the expense of unity

(Communion). Rank regards this problem as a developmental

"handicap" (Maddi, 1968).

According to Rank the exclusion of unity by individua-

tion begins in early childhood, during a critical period in

the development of what he calls will. Will, like some

aspects of Freud's ego, refers to that sense of self which

simultaneously directs psychological growth as an individual

and integration with the environment. The earliest mani-

festation of will is counterwill (Rank, 1945). Around the

age of two, a child's most salient sense of self derives

from his ability to say no. In resisting environmental

pressures, his parents, even his own impulses, the child

affirms himself as a separate person with a volition dis-

tinct from other forces. When parents prize their child's

emerging personhood while helping the child adapt to the

realities of the environment, counterwill normally develops

into mature will. But if the child at this stage experi-

ences a hostile environment, either because the parents

teach the child that he is "bad," or because he is generally

unsuccessful in learning to adapt to the limitations of

reality, the result may be a defensive fixation on counter-

will. Such an individual in adulthood will tend toward a

sense of separateness that is accompanied by hostility and

moralistic guilt. The personality is well differentiated,

but not well integrated. Lacking a mature will, the in-

dividual is disadvantaged in his ability to balance separa-

tion from the herd with constructive integration with the

world (Maddi, 1968).

Bakan believes that the danger of losing one's balance

in personality development lies in the tendency of Agency to

repress Communion. Prevailing life-strategies for several

hundred years have been agentic. Agency arises as self-

protection, self-assertion, self-expansion; as the formation

of separations; as isolation, alienation, aloneness; in the

urge to master; in repression of thought, feeling, and

impulse. Agentic strategy has enabled us to make great

strides toward conquering the physical environment and

establishing technological societies. But at what cost?

Notable among the ills that Bakan associates with unmitigat-

ed Agency is cancer. In what he calls the Freudian theory

of cancer, Bakan explores the rationale for a connection

between cancer and agentic features of personality and

indicates lines of empirical investigation leading to this

possible conclusion (1966, Chapter V).

The Freudian Theory of Cancer

Freud spoke of the instincts as mental representations

of somatic processes. He identified conversion hysteria,

in which the inhibition of a physical function accompanies a

psychological repression. It was Freud's intention that

psychoanalytic theory address the whole person. He further

believed that the intrapsychic mechanisms discovered through

psychoanalysis had relevance to the study of cellular


In Beyond the Pleasure Principle (1920/1950) Freud

proposed a death instinct which resides both in the ego and

in the individual cell. The normal purpose of this death

instinct is to promote the interests of the individual, as

opposed to the collective, leading to death from forces

"immanent in the organism itself" (p. 51). It may be said

that the death instinct preserves and protects the in-

dividual from external forces so that it can die in its own

time, of its own volition. Motives associated with this

process are self-preservation, self-assertion, mastery,

separation of the organism from the outside world, and

repression (Freud, 1920/1950; Bakan, 1966). Bakan quotes

Freud: "A strong egoism is a protection against falling

ill, but in the last resort we must begin to love in order

not to fall ill, and we are bound to fall ill if, in con-

sequence of frustration, we are unable to love" (Bakan,

1966, p. 171).

The very defenses which function to protect the

organism from destruction from without, if unchecked, will

destroy the organism from within. An over-defended ego is

cut off from the healing eros, the life instinct, operative

in loving and relating. Freud viewed such an imbalance

between the internal narcissistic needs, and the external

interpersonal needs, as incompatible with life. An in-

dividual thus insulated from its kind cannot maintain

health. The organism's efforts toward self-preservation

in this over-defended position result instead in a vicious

circle of greater insulation and greater imbalance.

Bakan noticed a parallel between Freud's reasoning and

the experimental findings of Hans Selye (Bakan, 1968, p. 26

ff). Selye (1956) described diseases of adaptation, in

which automatic physiological reactions associated with

self-preservation are self-injurious. Freud and Selye,

says Bakan, are looking at the same process: the individual

becoming ill from immanent forces elicited under conditions

of actual or potential threat, "which run off automatically

outside the dominion of conscious control" (1968, p. 44).

These reactions, because of their automaticity, bear no

direct relationship to the duration or degree of external

danger and can only be mitigated by the mechanism of

surrender (Selye, 1956), so that the body will stop defend-

ing itself. Freud treated psychological defenses in this

same way in the beginning stages of psychoanalysis.

The parallel with Selye may be useful in clarifying

the somatic process represented by Freud's death instinct.

Additional clarification is provided by the allusions to

cancer in Freud's writings. Bakan believes it was no co-

incidence for Freud to begin theorizing on the death in-

stinct during the time that he was developing a cancerous

growth in his jaw and palate. He revealed the discovery of

his tumor, calling it a tissue rebellion, about two years

after the first publication of Beyond the Pleasure Principle

(Jones, 1957). Bakan finds this book to be self-analytical,

as was characteristic of some of Freud's work. In Beyond

the Pleasure Principle, there is a specific reference to

"malignant neoplasms" (p. 68), where Freud states that

malignant cells have become narcissistic in the same way

that the personality becomes narcissistic without healthy

interpersonal relationships. This follows an attempt to

establish a relationship between the death instinct and

asexual reproduction (pp. 62 ff.).

In context of Freud's holistic orientation, particu-

larly his linking of what takes place between organism and


environment with what takes place on the cellular level,

Bakan finds Freud to be using the idea of the death in-

stinct to get at the psychological component of the state

in which cancer occurs. The terms Agency and Communion are

analogous in a broad sense to death instinct and life in-

stinct. More specifically, they are equivalent to the

strong egoism of which Freud spoke, and the love (Freud) or

surrender (Selye) which mitigates against self-destruction.

Where Freud suggested that the behavior of cancer cells was

a manifestation of narcissism, Bakan chooses to say that

cancer is an agentic disease.


Bakan cites a number of studies relating psychological

and interpersonal phenomena to the development of cancer

(1966, p. 17 ff.). He points out that the body of research

is sparse, that studies have largely been conducted outside

of Freudian contexts, and that each study taken by itself

allows alternative explanation of the data. It is their

collective impact, he says, which lends validity to the

Freudian theory and suggests an association between un-

mitigated Agency and cancer. Much the same can be said of

the research that has been conducted since Bakan's 1966

publication. More recent studies of personality and cancer

have been improved by the more frequent use of control

groups. The body of research is still sparse, however, and

there is no study which attempts to focus on the Agency/

Communion aspect of personality.

Categories that appear in the literature, and which

Bakan believed were indicative of unmitigated Agency as a

factor in the etiology and course of neoplastic disease,

are sexual maladjustment, inhibition of aggression, loss

of a significant person, lack of social involvement, and

inhibition of maternality. Repression is a sixth category



which appeared in the literature after Bakan's publication,

which also lends support to his views.

Sexual Maladjustment

Sexual relationships are one channel for expression of

Communion through intimacy with another person. In the case

of sexual maladjustment, Communion is impaired in its

ability to mitigate against domination of the personality

by Agency. Studies on the sexual functioning of women with

breast cancer or cervical cancer indicate a tendency toward

sexual maladjustment. One study found evidence of sexual

maladjustment in men with prostate cancer.

Tarlau and Smallheiser (1951) and Wheeler and Caldwell

(1955) found rejection of the feminine role and negative

feelings toward sexual relations in women with cancer of the

breast and women with cancer of the cervix. Bacon, Renneker

and Cutler (1952) found sexual inhibition and frustration at

a higher level in the personalities and histories of women

with breast cancer than had been observed in neurotic women.

These investigators also found in an intensive study of five

of their subjects that they chose mates who were cold,

sadistic, alcoholic, seclusive, impotent, uninterested,

opposed to having children, or monumentally narcissistic

(Renneker, 1963).

In some breast cancer subjects, Renneker (1963)

found a tendency to act out by becoming sexually pro-

miscuous after the disruption of a relationship. Cobb

(1953) observed a "sexual preoccupation leading to multiple

marriages" in her study of men with cancer of the prostate

gland (p. 52).

Reznikoff (1955) studied TAT responses of women with

breast cancer, benign breast tumors, and no tumors, and

found the cancer group to exceed both the benign tumor group

and the no tumor group in responses of alienation toward

masculine figures. He also found the benign tumor group

to exceed the no tumor group in such responses, suggesting

an ascending order in the correlation between feelings of

alienation toward men and degree of breast pathology. In

the same study, the cancer group tended to have married

later in life and reported a greater frequency of unhappy

marriages. Lerer (1975) compared histories of breast cancer

subjects with histories of benign tumor subjects. Those

with malignant tumors had been less assertive, had developed

fewer interpersonal relationships, had used less successful

and more archaic defense mechanisms, and had exhibited a

higher degree of frigidity, than did subjects with benign


Several of these studies have methodological problems.

Tarlau and Smallheiser (1951) and Wheeler and Caldwell (1955)

did not design their studies to determine whether the subjects'

negative feelings were a premorbid condition, or whether they

might result from lowered sexual self-esteem accompanying

perception of their disease. A similar criticism can be made

about Reznikoff's (1955) alienation toward masculine figures


on the TAT. Cobb (1953) did not use a comparison group.

These problems are avoided by Bacon et al. (1951), Renneker

(1963), and Lerer (1975) who do use comparison groups, and

who focus on long-term life style and historical data. Their

studies are, however, open to the criticism that they failed

to control for known medical risk factors.

Inhibition of Aggression

According to Freud, aggression is a way of directing

the death instinct outward. "It is extremely telling,"

says Bakan, "that one should find cancer patients to be

people who are remarkably unable to express aggression."

(1966, p. 198). He supports this with the findings of six

studies. Cobb (1953) observed that prostate cancer patients

do not express aggression, are compliant and superficially

cooperative. A majority of Bacon, Renneker and Cutler's

breast cancer subjects had no technique for discharging

anger, and in fact denied ever having been angry (Bacon,

et al., 1952; Renneker, 1963). Life histories of cancer

patients indicated that twice as many showed inhibition

of aggression as compared with a cancer-free group (LeShan

& Worthington, 1956b). Rate of growth of malignant tumors

of the same type has been observed to differ among in-

dividuals, with the extremely polite and acquiescent tending

to have more rapidly progressing diseases than the more ex-

pressive personalities, and those who might be described

as too mean to die (West, 1954; Blumberg, West & Ellis, 1954).

Bakan notes one study (LeShan & Gassman, 1958) in

which a patient achieved measurable though temporary

shrinkage of visible metastatic growths, following some 45

hours of intensive psychotherapy resulting in her ability to

ventilate and accept hostility toward her children. This

study can only suggest the possibility of a cause-effect

relationship and cannot be taken to confirm that suppressed

hostility directly affects tumor growth. A later study

(Greer & Morris, 1975), however, confirms the conclusion of

LeShan and Worthington, that cancer patients show inhibition

of aggression as a trait of personality. Greer and Morris

compared women with malignant breast tumors and women with

benign breast tumors, during the period of time between re-

port of symptoms and diagnosis. They found that the cancer

subjects showed a pattern of abnormal release of anger,

usually extreme suppression, which was both contemporary and

existing throughout previous adult life. Their innovative

design controls for the effect on emotional response which

might result from the subjects' awareness of having cancer.

In this case, all subjects were awaiting diagnosis and did

not know whether their conditions were benign or malignant.

It would have been wiser, however, for the investigators

to follow the benign tumor group to see whether any of

them developed malignant tumors within a ten year period

after the first biopsy, since there is a higher incidence

of subsequent cancer among women who have had biopsies.


A related study by Kissen and Eysenck (1962) provides

evidence which may be conflicting, depending upon how it is

interpreted. Male cancer patients were found to rate high-

er on extraversionn" and lower on "neuroticism" than their

controls. Since these investigators equated low neuroticism

with diminished outlet for emotional discharge, their con-

clusions agree with the others. Kissen (1963) conducted

another study on emotional discharge in males with cancer

and confirmed that this group does tend to bottle up and

conceal emotional difficulties. The Kissen and the Kissen

and Eysenck studies are both disputed by Huggan (1968a),

who disagrees with their interpretation and with their

method of collecting data, which was by self-report. Huggan

believed subjects had distorted their self-ratings in order

to be perceived more positively, but allows that "faking

good" can itself be evidence of inability to discharge nega-

tive affect. Interestingly, another investigator found

elevated MMPI lie scale scores in cancer patients

(Schonfield, 1975). These subjects were all women. Hagnell

(1966) reported a significant association between person-

ality traits resembling extraversion and cancer in women,

but not in men. He was unable to explain his findings.

Loss of a Significant Person

Losing one's primary source of love and belonging deals

a nearly mortal blow to the communal side of the personality.

Might not the void created by such an absence leave the


individual temporarily at the mercy of agentic processes?

Loss of a significant person as a precursor is one of the

oldest observations to be made about cancer. Some physicians

began hypothesizing in the eighteenth century that the

death of a near relative, or the grief associated with the

death, might somehow be related to the subsequent develop-

ment of a malignancy, because it seemed so frequently the

case that their patients had experienced such a loss shortly

before the onset of cancer (Bakan, 1966). Results of

psychoanalyses of 100 cancer patients in 1926 led to the

conclusion that a precipitating factor was object loss

(Evans, 1926). Controlled studies, for the most part, have

confirmed recent loss of a significant person as occurring

more often in the histories of cancer patients than in

control groups (Peller, 1940; Bacon et al., 1952; Greene,

1954; LeShan & Worthington, 1956a; Green & Miller, 1958;

Neumann, 1959; Muslin & Pieper, 1962; Muslin et al., 1966).

The most recent study comparing life histories of cancer

patients with controls fails to confirm this theory, find-

ing no significant difference in loss of a significant

person between the cancer group and the cancer-free group

(Schonfield, 1975).

Lack of Social Involvement

One indicator of a healthy balance between Agency and

Communion is the presence of satisfying interpersonal re-

lationships. Bakan points to two previously cited studies

showing lack of social involvement as characterizing the

personalities of cancer patients. Cobb (1953) found her

subjects not only to avoid emotional involvement, but also

to have had "difficulty in making their way into a world of

adequate social relationships" (p. 254). Reznikoff's (1955)

subjects had fewer interests outside the home, as well as

husbands who spent less time with them, than did the control

group of healthy women. Later studies agreed with these

findings. Booth (1960) found secretiveness and autistic

orientation toward others in cancer patients, and classified

them as anal types. Brown, Katz, and Kaufman (1961) inter-

preted House, Tree, Person drawings made by cancer patients

as showing egocentricity and defective interpersonal re-


The argument should be made that it is difficult to

tell whether we are measuring precipitating psychological

events, or measuring psychological reactions to disease.

One approach to this issue has been to study breast cancer

and benign tumor subjects who have not yet been biopsied,

such as the study that was conducted by Greer and Morris in

1975, where subjects found to suppress anger were sub-

sequently diagnosed as malignant. Another approach is to

gather data from the subjects' premorbid histories, as did

Lerer (1975), who found breast cancer patients to have been

less assertive and to have had fewer interpersonal relation-

ships than those whose tumors were benign.

Abse, Wilkens, and Vande Castle (1974) studied lung


cancer subjects and compared them to their healthy age mates.

He found that older subjects were more like their age mates,

exhibiting no significant differences between the two groups

above age 56. Since cancer tends to correlate more highly

with age than with any other known variable, it may be that

earlier-onset cancer would be more likely to involve a psycho-

logical predisposing factor. This was the case with Abse's

subjects. Below age 56, the cancer group did differ signifi-

cantly from their counterparts. Younger cancer patients were

characteristically more overly conscientious in work at-

tendance, less active socially and sexually, markedly more

constricted in interpersonal relationships, and more troubled

in handling dependency needs than their healthy age mates.

Inhibition of Maternality

The normal mother-child relationship strikingly ex-

emplifies Communion. Maternal feelings represent a loss of

self through union with another being, which transcends the

separateness of individuals. Two studies of maternality in

female cancer patients are cited by Bakan (1966) both in-

dicating its inhibition or absence. No others appear in the

literature. Bacon et al. (1952), out of 40 female cancer

patients, found only one who had a desire for children.

They speculated that a psychological block against pregnancy

might somehow create a disturbance in hormonal balance which

contributed to the development of breast cancer. Reznikoff

(1955) found that cancer subjects, as compared with benign

tumor and no tumor groups, were more ambivalent toward


taking on the responsibility of children and "distinctly more

fearful and threatened by the birth process" (p. 101).


There are three studies in the literature which do not

fall into the research categories mentioned by Bakan, but

which can be interpreted as supporting the theory of a re-

lationship between cancer and the agentic feature. These

studies indicate repression (one manifestation of Agency) as

a personality characteristic of individuals who develop

cancer. Bahnson and Bahnson, in 1967 and again in 1969,

found strong repression and denial as life-long, continually

utilized defense mechanisms in persons who develop cancer.

Huggan (1968a) confirmed this, finding in particular re-

pression of anxiety.

Concerning the psychosomatic view of cancer, Grinker

(1966) points out that its proponents are not searching for

or claiming a psychological cause. Causes are presumed to

be simultaneously psychological and somatic. The task is to

become able to identify the psychological variables which

interact with physiological variables to affect etiology,

development, therapy, and outcome for the patient.

The Structure of Personality

As with the perennial issue of nature versus nurture,

we cannot differentiate between characteristics of the pre-

morbid personality and possible effects on a subject's

contemporary personality of the disease process. It must be

acknowledged that this somewhat cloudy area exists in all

attempts to study personality. It should also be noted that

personality is a highly stable configuration of psycho-

logical traits which ordinarily are not subject to dramatic

change and which do not fluctuate as a function of the

social and biological pressures of the moment.

According to Freud, the formative period for person-

ality development is early childhood. Those processes which

Freud referred to as id are present at birth. Those pro-

cesses which he called ego begin to differentiate from id

at birth and are massively impacted by the early environ-

ment, so that the learning which determines an individual's

traits and characteristics is a product of heredity and the

particular environment available during the formative

period. Those processes which Freud called superego are

introjected from the parents when ego is developed suf-

ficiently that the child is capable of self-perception and

parental identification, beginning about age two or three.

Superego's values further modify the ego, and by about the

age of five the basic personality is formed (Freud, 1953).

Not all personality theorists, even those identified with

Freud, adopt so radical a position on the early closure of

personality. The consensus, however, is that if person-

ality changes at all, it changes very slowly (Maddi, 1968).


The intention of this study is to compare two groups

-- a cancer group and a non-cancer group -- on degree of

dominance of the personality by agentic features as described

by David Bakan. Theoretically, all motives can be classified

as either agentic of communal. For the sake of accuracy,

only those traits and conditions named by Bakan were measur-

ed. They are

Agency Communion

self-protection sense of oneness with

self-assertion other organisms


separation lack of separation

isolation contact

alienation openess

aloneness union

mastery noncontractual cooperation

repression lack and removal of


It was hypothesized that the cancer group would exhibit

a significantly higher ratio of Agency over Communion than

the non-cancer group.



The cancer group was composed of adult outpatients at

the University of Florida Shands Teaching Hosptial. They

were selected on the basis of current diagnosis of systemic

cancer and homogeneity with respect to prognosis. Selected

subjects had known their diagnoses for at least one month

prior to testing, in order to minimize any possible post-

traumatic effect on their test-taking behavior, and were

expected to live for a year or more.

It should be noted that patients under treatment for

cancer at Shands do not represent a random sample of people

with cancer. These patients are engaged in seeking the help

and support of others. In contrast to portions of the

population who receive their diagnoses and go home to die,

or pursue some form of self-treatment, patients at Shands

are mong the most communal of the cancer population. While

this pre-selection factor may have made it more unlikely to

find a significantly higher Agency/Communion ratio in the

cancer group, the risk of Type I error was reduced.

The non-cancer group was composed of 30 subjects who

had no history or symptoms of cancer. They were drawn from

a larger pool of volunteers solicited from college students,

university employees, hospital visitors, and friends and

relatives of the cancer subjects. The 30 used to make up

the comparison group were chosen by matching with the cancer

group for known medical risk factors in the disease popula-

tion. Paired subjects with criteria for pairing are describ-

ed in Appendix B.

Risk factors associated with specific diseases repre-

sented in the cancer group were obtained from etiological

and epidemiological data compiled by Fraumeni (1975) and

Schottenfeld and Fraumeni (1982). Age tends to correlate

with cancer in general; sex tends to correlate with some

types of cancer, as well as with some personality variables

(Jackson, 1974). Age and sex were therefore used as pairing -

criteria in every instance. Occupational exposures, such as

to ionizing radiation, certain dyes, and certain chemicals,

would have been matched had they been found in the cancer

group. National origin and race were the same for all sub-

jects. A two- to four-fold increase of risk exists for re-

latives of persons with cancer of the stomach, large in-

testine, lung, uterus, and breast; so subjects in these

categories were paired with comparison subjects having

similar family histories. Lung cancer was also matched for

smoking behavior, uterine cancer for age at menarche within

one year and age at first pregnancy within two years, and

breast cancer for age at menarche within one year, age at

first pregnancy within two years and nursing of children.

All other cancers were matched only for age within two

years and sex. Pairing was blind for all information other

than matching criteria.



Since no instrument was available that had been design-

ed for the specific purpose of measuring the agentic and

communal aspects of personality, one had to be found which

measured variables closely approximating Bakan's de-

scriptions of Agency and Communion. The Personality Research

Form, or PRF (Jackson, 1974) was selected to measure Agency

and Communion because of the close correspondence of its

scale descriptions to Bakan's descriptions of agentic and

communal characteristics. The PRF is a personality in-

ventory originally derived from Murray's needs, the theo-

retical constructs defining internal components of person-

ality and behavior in Henry Murray's theory of personality

(Murray, 1938).

Murray, like Bakan, is basically a psychoanalytic

thinker. He was trained by members of Freud's inner circle,

and his theory of personality makes use of the Freudian con-

flict model (Maddi, 1968). Also like Bakan, Murray modifies

Freudian theory with the position that instincts exist which

are compatible with communal life (Murray & Kluckhorn, 1956).

Where Bakan does not elaborate on the concrete aspects of

personality, Murray (1938) offers extensive classification

of personality characteristics, most notably a list of 40

needs including but not limited to the characteristics

that Bakan uses to define Agency and Communion.


The PRF has reduced Murray's list of needs into 20 in-

dependent personality variables. Form E, used in this

study, has 352 items in 20 personality scales and two in-

dicators of test-taking attitudes and validity. It re-

quires about an hour to complete. Reading level is fifth

to sixth grade as calculated by the Dale-Chall formula

(Dale & Chall, 1948). Twelve of the scales use definitions

which either duplicate or very closely approximate the defi-

nitional terms employed by Bakan, so these twelve were used

to measure Agency and Communion. Agentic scales are

Achievement, Autonomy, Cognitive Structure, Defendence, and

Harmavoidance. Communal scales are Affiliation, Impulsivity,

Nurturance, Play, Sentience, and Succorance.

Using the PRF to Measure Agency

Measuring self-assertion

Self-assertion is measured by the scales Autonomy and

Dominance. The definition of Autonomy uses the trait ad-

jectives "uncompliant" and "undominated" and says that the

high scorer breaks away from restrictions and may rebel

against restraints. The definition of Dominance uses the

trait adjective "assertive" and describes the high scorer

as expressing opinions forcefully, and enjoying the role of


Measuring self-expansion

Self-expansion is the agentic growth and development of

the individual. It is measured by Achievement. The defi-

nition of Achievement uses the trait adjectives "striving,"

"attaining," and "self-improving" and says that the high

scorer is willing to put forth effort to attain excellence.

Measuring separation

Agency is manifested in separations, whether of the

individual from other individuals or from the environment,

or separation as differentiation of parts. Autonomy

measures separation of the individual from others, while

Cognitive Structure measures the need for psychological

division of the world into its separate parts. The defi-

nition of Autonomy uses the trait adjectives "free," "in-

dependent," "individualistic" and describes the high scorer

as not tied to people, places, or obligations. The defi-

nition of Cognitive Structure uses the trait adjectives

"defining," "explicit," "clarifying," and "precise," all of

which refer to carefully distinguishing individual things

from other things. The high scorer does not like ambiguity.

Measuring isolation

In addition to implying separation and aloneness,

treated above and below, isolation implies restraint from

interaction in order to reduce the risk of harm. This

aspect of isolation is measured by Harmavoidance. The

definition of Harmavoidance uses the phrases "withdraws from

danger" and "stays out of harm's way" and describes the

high scorer as not enjoying exciting activity and avoiding


Measuring alienation

Alienation is measured by Defendence. The definition

of Defendence uses the trait adjectives "chip-on-the-should-

er," "touchy," "defensive," and "self-condoning" and says

that a high scorer suspects that people mean him harm.

Measuring aloneness

Aloneness is measured by Autonomy. The definition of

Autonomy uses the term "lone-wolf." The high scorer enjoys

being unattached.

Measuring mastery

Mastery as accomplishment is measured by Achievement.

The definition of Achievement includes the trait adjective

"accomplishing" and describes the high scorer as aspiring

to accomplish difficult task. Cognitive Structure measures

a form of psychological mastery of one's experience, as in

the trait adjectives "exacting," "perfectionist" and "seeks

certainty." The high scorer wants all questions answered

completely and wants to make decisions based upon definite

knowledge. Mastery as ascendency and control is measured by

Dominance. The definition of Dominance uses the trait ad-

jectives "ascendent" and "controlling." The high scorer

attempts to control his environment and to direct other


Measuring repression

Repression refers both to the psychological defense

mechanism which prevents certain material from entering

consciousness and to the state of being repressed in the

common sense of being restrained from expression. The PRF

does not attempt to measure unconscious processes as such.

It does, however, measure the traits of rigidity in think-

ing and defensiveness in behaving which are concomitant with

psychological repression. Cognitive Structure is defined in

part as rigidity, and Defendence is defined in part as de-

fensiveness. These scales are as close as the instrument

will come to measuring this aspect of repression. As for

repression in the common sense, Defendence uses other de-

fining trait adjectives which indicate restraint of ex-

pression: "secretive,'.' "wary," "rationalizing," "guarded."

Low scores on the Impulsivity and Play scales, discussed

in the next section, would also be expected of a repessed


Using the PRF to Measure Communion

Measuring the sense of oneness

A sense of oneness with others is measured by Affilia-

tion. The definition of Affiliation uses the trait adjec-

tives "gregarious," "warm," "neighborly" and says that the

high scorer enjoys being with and maintaining association

with people.

Measuring lack of separation

One indicator of lack of separation would be low score

on a scale measuring separation. Other indicators are the

scales that measure traits of togetherness. These are

Affiliation, Nurturance, Play, and Succorance, which are

described under other headings in this section as they per-

tain to sense of oneness and contact.

Measuring contact

Contact as direct experience through the senses is

measured by Sentience. The definition of Sentience uses the

terms "sensuous," "sensitive,""enjoys physical sensations,"

"notices environment" and describes the high scorer as

noticing smells, sounds, tastes, and the way things feel.

Contact as close association with others is measured by

Affiliation, described above, and by Play, Nurturance, and

Succorance. The definition of Play uses the trait adjec-

tives "playful," "jovial," "prankish," "sportive" and says

that the high scorer spends a good deal of time participat-

ing in games, sports, and social activities. The definition

of Nurturance uses the trait adjectives "aiding," "minister-

ing," "assisting" and says that the high scorer gives com-

fort, assists others, and is interested in taking care of

others. The definition of Succorance uses the traits "help


seeking" and "craves affection." The high scorer seeks con-

tact in the form of love, protection, and advisement.

Measuring openness

Openness as the state of being natural, unabashed, and

unconcealed is measured by Impulsivity and Succorance. The

definition of Impulsivity includes the traits of spontaneity

and incautiousness, and Succorance includes "confiding."

Also measured by Succorance is "defenselessness," which

implies openness in the sense of lacking protection. The

high scorer on Impulsivity speaks freely; the high scorer

on Succorance confides readily.

Measuring union

Union is measured by Affiliation, Nurturance, and

Succorance, described above. All are ways of joining with

others. Nurturance includes maternal/paternal feelings,

which would seem to create a deeper sense of union than

that required by friendship or dependence.

Measuring non-contractual cooperation

Non-contractual cooperation is measured by Affiliation.

A defining trait adjective of Affiliation is "cooperative."

Nurturance also indicates non-contractual cooperation, in

that it measures helpfulness and assisting, and the high

scorer offers a helping hand.

Measuring lack and removal of repression

Low scores on scales measuring repression would be one

indicator of lack of repression. Scales that directly in-

dicate lack or removal of repression are Impulsivity and

Play. The definition of Impulsivity uses the defining trait

adjectives "uninhibited," "irrepressible," "impulsive" and

says that the high scorer gives vent to feelings and wishes,

and expresses emotions. Play indicates lack or removal of

repression in the sense of being restrained or subdued. The

defining trait adjectives for Play include "laughter-loving,"

"jolly," "prankish," "gleeful" and the high scorer is light-

hearted and does many things for fun.

Validity and Reliability of the PRF

Concurrent validity

Jackson and Guthrie (1968) studied correlations between

PRF scale scores, behavior ratings by peers, and self-ratings,

in college students. They found correlations of .32 with

peer ratings and .45 with self-ratings. A correlation of

.18 was significant at the .01 level.

Kusyszyn (1968) studied correlations between a subset

of PRF scales and trait ratings by peers in college students

and found a median correlation of .40. When the raters

shared living quarters with the assessees, the correlation

was .47. A correlation of .29 was significant at the .01

level for the total sample, and a correlation of .42 was

significant at the .01 level for the group living together.

In the total sample, the PRF scale Autonomy correlated

significantly with peer ratings at the .05 level, while all

others in the subset correlated significantly at the .01

level. In the group living together, Aggression and

Nurturance correlated significantly with peer ratings at the

.05 level, while all others in the subset correlated sig-

nificantly at the .01 level.

Jackson (1974) reported median correlations, using all

20 PRF content scales, of .56 with trait ratings by peers

and .52 with behavior ratings, in college students. A

correlation of .28 was significant at the .05 level, and of

.36 at the .01 level. The Abasement scale did not correlate

significantly with either behavior ratings or trait ratings

in this particular study, although it did in others.

Sentience did not correlate significantly with behavior rat-

ings, but with trait ratings it did correlate significantly

at the .01 level. Among the other 36 correlations obtained

in the study, 29 were significant at the .01 level, and

seven were significant at the .05 level.

From a replication of this study, Jackson (1974) re-

ported median correlations of .49 with behavior ratings

and .54 with trait ratings. A correlation of .31 was sig-

nificant at the .05 level, and .40 was significant at the

.01 level. Again, Sentience did not correlate significant-

ly with behavior ratings, but did correlate significantly

at the .01 level with trait ratings. The same was true for

Understanding, which in the original study correlated sig-

nificantly with both behavior and trait ratings. Also dif-

fering from the original study was Abasement, which did

correlate significantly this time with both other measures.

Of the 34 remaining correlations, 30 were significant at

the .01 level, and four were significant at the .05 level.

These studies demonstrate the convergent validity of the

PRF, with the reported correlation coefficients exceeding

those usually found for personality inventories.

Discriminant validity

In order to measure discriminant validity as well as

convergent validity, Jackson (1966) developed a procedure

which he called multimethod factor analysis. This procedure,

used to correlate a number of traits measured by a number of

methods, results in a matrix which shows only heteromethod

validity coefficients, so that common factors cannot be in-

fluenced by monomethod variance.

Jackson andGuthrie (1968) used multimethod factor

analysis to measure the PRF scales by three other methods:

self ratings, peer ratings, and behavioral descriptions.

They reported exceptional regularity in the loading of the

appropriate factors with PRF scales. The 20 scales loaded

on 18 factors, with two factors being defined by two PRF

scales and criterion measures. One of these two factors

was loaded positively by three Dominance measures and

negatively by three Abasement measures. The other was

positively loaded by three Aggression measures and nega-

tively by three Impulsivity measures. All other factors

were defined by single sets of PRF scales and criterion

measures. Jackson concluded that each scale provided a

unique contribution to the assessment of personality.


Item analysis of the PRF shows unusually good homogenity

for personality scales (Jackson, 1974). For the 20 content

scales, Jackson reported .91 as the median of lower bound

Kuder-Richardson formula 20 values, with a range of .80 to


In a study of the stability of PRF scores, Bentler

(1964) found test-retest reliabilities over one week's time

ranging from .69 to .90 for the content scales, with a

mean reliability of .80. These were considered lower bound

estimates, since the testing conditions were not identical.

Odd-even reliabilities over two weeks' time were re-

ported by Jackson (1974). Reliabilities ranged from .72 to

.92, with a mean reliability of .85 for content scales.

PRF-E, the form used in this study, reduces the number

of items per scale to the best 16 out of the 40 original

items per scale. Since reliability is influenced by test

length, Form E compromises somewhat between higher relia-

bility and shorter testing time. Two samples of PRF-E

odd-even reliabilities are reported by Jackson (1974). In

a sample of psychiatric patients, reliabilities for con-

tent scales range from .29 (Cognitive Structure) to .84,

with a mean reliability of .73. In a sample of college

students, reliabilities range from .50 to .91, with a mean

reliability of .68.


Selected subjects were asked to read and sign an in-

formed consent form (Appendix C) and to provide personal

data for matching purposes, as outlined in the section on

Subjects. Each subject was given the self-administered

Personality Research Form in an area suitably free from



The PRF was scored according to standard procedure for

the instrument, resulting in a score for each scale rang-

ing from 0 to 16. The six agentic scale scores were added

together, and the six communal scale scores were added

together. These were expressed as an Agency/Communion ratio

for each subject and converted to decimal form for ease in


Analysis of the data

The hypothesis that the Agency/Communion ratio would

be larger for the cancer group than for the non-cancer group

was tested for statistical significance by a comparison of

difference scores. Because of the directionality of the

hypothesis, a one-tailed test was used. The non-randomness

of the samples and the matched group design indicated the

use of Sandler's A-statistic, a derivative of Student's T

which is preferred over T for ease of computation (Sandler,

1955). An alpha level of .05 was selected.

A comparison was made between the agentic variables

only, and between the communal variables only, to determine

whether the source of the difference between the two groups

lay more in one of these dimensions than in the other, or

whether the difference lay exclusively in the ratio of

Agency over Communion. A scale by scale comparison was then

made in order to note more specifically the degree of con-

tribution by each scale to the overall difference between

the groups.


Results of the study are summarized in Tables 1 through

5. Computation of Sandler's A showed the Agency/Communion

ratios for the cancer group to be significantly higher than

the Agency/Communion ratios for the non-cancer group. The

computed A value was 0.12. The table value of A for a

one-tailed test with 29 degrees of freedom is 0.368 or less

for significance at the 0.05 level, 0.193 or less for sig-

nificance at the .01 level, and 0.161 or less for signifi-

cance at the .005 level. Therefore the difference between

the two groups was significant at the .05 level, as hypo-

thesized, as well as at the .01 level and the .005 level.

A comparison of the agentic scales only showed the

cancer group to be significantly higher than the non-cancer

group in Agency independent of Communion. The computed A

value was 0.29, which was significant at the .05 level.

Also, a comparison of the communal scales only showed the

cancer group to be significantly lower than the non-cancer

group in Communion independent of Agency. The computed A

value was 0.13, which was significant at the .01 level.

A scale by scale comparison of differences between the

two groups indicated the following:

Of the agentic scales, the greatest difference was in

Cognitive Structure, followed by Harmavoidance, Defendence,

Autonomy, and Dominance. Achievement had no bearing on the

overall difference between the two groups, showing a slight

negative difference approaching zero. Of the communal scales,

the greatest difference was in Play, followed by Succorance,

Nurturance, Sentience, Impulsivity, and Affiliation. Three

of the 12 scales showed differences of a magnitude reaching

significance independent of any other scale score.

Cognitive Structure was higher in the cancer group than in

the non-cancer group, with significance at the .05 level.

Play and Succorance were lower in the cancer group than in

the non-cancer group, Play being significant at the .0005

level, and Succorance at the .01 level. When the groups

were broken into subgroups by sex, comparison of Agency/

Communion ratios showed that both the female and the male

cancer subgroups scored significantly higher than their non-

cancer counterparts. The females-only difference was

significant at the .01 level, and the males-only difference

was significant at the .05 level.

There was no significant difference between the two

groups on either the Infrequency scale (measuring random

responding, lack of understanding, or deciption) or on the

Desirability scale.


Agency/Communion Ratios and Difference Scores
of Paired Subjects

Pair # Cancer group Non-cancer group Difference Difference2

1 0.82 1.17 -0.35 0.12
2 1.41 0.85 0.56 0.31
3 1.42 0.87 0.55 0.30
4 1.71 0.81 0.90 0.81
5 1.38 1.12 0.26 0.07
6 0.87 0.42 0.45 0.20
7 0.78 0.52 0.26 0.07
8 1.14 0.84 0.30 0.09
9 1.16 1.93 -0.77 0.59
10 1.14 1.11 0.03 0.00
11 2.03 1.41 0.62 0.38
12 1.00 0.45 0.55 0.30
13 1.28 1.15 0.13 0.02
14 1.07 0.58 0.49 0.24
15 1.00 1.21 -0.21 0.04
16 1.08 0.89 0.19 0.04
17 0.73 0.68 0.05 0.00
18 1.04 0.49 0.55 0.30
19 1.04 0.93 0.11 0.01
20 1.09 1.40 -0.31 0.10
21 1.04 1.04 0.00 0.00
22 0.76 1.42 -0.66 0.44
23 1.22 0.65 0.57 0.32
24 0.98 1.40 -0.42 0.18
25 2.31 0.63 1.68 0.28
26 0.87 0.81 0.06 0.00
27 1.28 1.18 0.10 0.01
28 1.46 0.74 0.72 0.52
29 0.85 0.75 0.10 0.01
30 1.74 1.17 0.57 0.32
ZD=7.08 s D2=5.83


Computations of Sandler's A for Agency/Communion
Ratios, Agency only, and Communion only

A = D2/(gD)2 df = 29

A = 5.83/50.13
A = 0.12
A(29) = 0.12, p<.005

A = 6339/22201
A = 0.29
A(29) = 0.29, p<.05

A = 8229/62001
A = 0.13
A(29) = 0.13, p<.01


Values of A for Individual Scales in Descending Order of
Contribution to Overall Difference

Agentic Scales:
Cognitive Structure

Communal Scales:



*significant at .05 level
**did not contribute to overall difference
***significant at the .0005 level
****significant at the .01 level


Agency/Communion Ratios and Difference Scores
of Female Subjects Only, and
Computation of Sandlers A

Cancer group Non-cancer group Difference Difference2

1.41 0.85 0.56 0.31
1.71 0.81 0.90 0.81
1.00 0.45 0.55 0.30
1.28 1.15 0.13 0.02
1.07 0.58 0.49 0.24
1.00 1.21 -0.21 0.04
1.08 0.89 0.19 0.04
0.73 0.68 0.05 0.00
1.04 0.49 0.55 0.30
1.04 0.93 0.11 0.01
1.09 0.40 -0.31 0.10
1.04 0.04 0.00 0.00
0.76 1.42 -0.66 0.44
1.22 0.65 0.57 0.32
0.98 1.40 -0.42 0.18
2.31 0.63 1.68 0.28
0.85 0.75 0.10 0.01
SD-=4.28 0D2=3.4

A =-D2/(5:D)2 df = 16

A = 3.4/18.32

A = 0.19

A(16) = 0.19, p<.01


Agency/Communion Ratios and Difference Scores
of Male Subjects Only, and
Computation of Sandler's A

Cancer group Non-cancer group Difference Difference2

0.82 1.17 -0.35 0.12
1.42 0.87 0.55 0.30
1.38 1.12 0.26 0.07
0.87 0.42 0.45 0.20
0.78 0.52 0.26 0.07
1.14 0.84 0.30 0.09
1.16 1.93 -0.77 0.59
1.14 1.11 0.03 0.00
2.03 1.41 0.62 0.38
0.87 0.81 0.06 0.00
1.28 1.18 0.10 0.01
1.46 0.74 0.72 0.52
1.74 1.17 0.57 0.32
ZD=2.8 D2=2.68

A = D2/(gD)2 df = 12

A = 2.68/7.84

A = 0.34

A(12) = 0.34, p<.05



The primary objective of this study was to determine

whether certain characteristics of personality approximat-

ing Bakan's definitions of Agency and Communion are associ-

ated with cancer. Results indicate that such an association

does exist. Following are the four findings presented in

the previous chapter, discussed according to their contri-

butions to an understanding of this association, as well as

their indications for Bakan's theory.

1. The Agency/Communion ratio was significantly higher

for subjects in the cancer group than for subjects in the

non-cancer group. This finding strengthens Bakan's theo-

retical position by supporting his assertion that cancer is

associated with an agentic personality. Bakan has made the

suggestion that the relationship between personality and

cancer is possibly etiological. Alternative interpretations

are also possible; for example, agentic characteristics of

personality may constitute secondary symptoms of cancerous

conditions. Based on current knowledge of the stability of

personality and the reliability of its measurement, the

symptomatic explanation is unlikely. Within the present


research design however, such alternative interpretations

cannot be ruled out.

2. The cancer group scored both significantly higher

on the agentic scales, and significantly lower on the

communal scales, than did the non-cancer group. This

strengthens Bakan's theoretical position by its consistency

with his view of the interaction between Agency and

Communion: that Agency tends to suppress Communion, and

Communion tends to mitigate against domination by Agency.

This view would predict that an abundance of Agency would

keep Communion at a relatively low level, or similarly, that

a low level of Communion would permit the development of a

relatively high degree of Agency by failing to mitigate

against the tendency of Agency to dominate.

3. Comparison of the cancer group and the non-cancer

group on single scale scores shows considerable variation.

A values range from 0.08, a significant difference at the

.0005 level, to 112.50, indicating nearly identical per-

formance of the two groups. The most dramatic differences

between the groups are found in Play, Succorance, and

Cognitive Structure. The cancer group showed a relative

depression on the Play and Succorance scales, and a relative

elevation on the Cognitive Structure scale. On the basis of

these data, typical personality traits of a person who has

cancer might likely be high need for certainty, definition,

and precision; perfectionist tendencies; rigidity and self-


control; combined with lack of playfulness, and avoidance

of or resistance to the help, protection, and support of


This finding is compatible with Bakan's position. But

Bakan does not speculate on the fine points of what might be

the typical personality profile associated with cancer--only

that the personality characteristics associated with cancer

would indicate domination by Agency. The tentative assertion

of a type, then, constitutes a refinement of Bakan's theory.

4. Performance on the Infrequency and Desirability

scales shows no significant difference between the two

groups. This indicates that the performance of both groups

is equally credible. It is further worthy of mention be-

cause it disagrees with the conclusions of Huggan (1968a),

cited in Chapter Two, that cancer subjects distorted their

self-ratings more than control subjects and is inconsistent

with the findings of Schonfield (1975) that cancer patients

had elevated lie scales on the MMPI, also cited in Chapter


Impact of the Study

Theoretical Impact. One of the functions of theory is

to generate testable hypotheses, which in turn can generate

data. The data can then be fed back into the theory to

modify, strengthen or refine it, and to generate new hypo-

theses. A major thrust of this study was to bring the data


to bear on David Bakan's theory of personality, particular-

ly his Freudian theory of cancer. It can now be said that

Bakan's position bears up under empirical scrutiny. Two

directions are conceivable as emerging from the data:

development of a typology of personality based on Bakan's

theory, and discovery of a specific personality type which

may be associated with cancer.

Practical Impact. Efforts toward psychological inter-

vention with people who have cancer should be based on a

sound understanding of the psychological correlates of cancer.

This is equally true whether the interventions focus on

personal adjustment, rehabilitation, coping with grief, or

the use of psychotherapy as a treatment modality along with

medical interventions. It has been said that knowing what

kind of person has a disease is as important as knowing

what kind of disease a person has. Results of this study

may increase our understanding of the appropriateness and

effectiveness of psychological interventions. The as-yet

unexplored territory of cancer prevention through promotion

of communal climates for the formative stages of personality

development is a possible long-range implication of these

findings, should further study substantiate the idea that

a greater susceptibility to cancer exists in individuals

with certain personality traits.

Limitations of the Study

Freudian concepts cannot be thoroughly explored without

delving into unconscious processes. It is a limitation of


this or any quantitative study involving Freudian concepts

that exploration of the unconscious dimension of personality

is sacrificed in favor of the greater accuracy and precision

with which we are able to explore the conscious dimension.

Similarly, in the conversion of Agency and Communion from

abstractions to measurable concrete characteristics, the

chance exists that something may be lost. It is hoped that

what is gained in empirical clarity and practical applic-

ability is the greater part, and that the integrity if not

the entirety of Bakan's theory has been preserved.

Another limitation of the study is simply that it is

not longitudinal: the future status of the comparison sub-

jects is unknown. Presentation of the data at this time

precludes the opportunity of eliminating subjects from the

comparison group should they develop cancer at a later time.

While matching for age is a control for the effect of ageing

on risk of cancer, additional validity could be gained by

following the non-cancer subjects.

Directions for Further Research

Further research is needed to assess the etiological

significance of the results of this study. Longitudinal

studies are necessary to determine whether or not a greater

probability of developing cancer exists in presently healthy

persons whose personalities are highly agentic, as well as

to follow comparison subjects to see whether they remain

cancer-free. Studies of the ways in which personality
mediates a possible stress and cancer relationship might also


provide insight into the etiological role, if any, played by

personality factors.

Another research area suggested by the results of this

study is the question of whether the course and outcome of

a disease can be influenced by alleviating repression and

developing latent communal aspects of personality. Compara-

tive studies of subjects who had accomplished these changes

versus subjects showing no psychological change in this

direction would address this question. Finally, replica-

tions of the present study in other contexts are important

to assess the generalizability of the results.


David Bakan was born in New York on April 23, 1921. He

received his PhD in psychology in 1948 from Ohio State

University. Until 1961 he was a member of the faculty of

the University of Missouri. He then joined the faculty of

the University of Chicago, where he stayed until 1968.

Currently he is Professor of Psychology at York University

in Toronto, Ontario. Bakan is recognized for his research

and writing on Sigmund Freud, on the psychology of religion,

and on the scientific method. His own theory of personality

emphasizing Agency and Communion was introduced in 1966 in

his book The Duality of Human Existence.

Historically, Bakan might be classified as a Freudian.

His postulate of ubiquitous opponent forces as the basis of

personality and of human culture has its roots in psycho-

analytic theory. Yet in one respect Bakan moves sharply

away from Freudian thinking: in a word, Freud described

psychological functioning as defensive, whereas Bakan con-

fines his treatment of defensiveness to the negative aspects

of personality and behavior. Bakan's view of healthy

functioning leans toward the tendency for positive growth

which characterizes the self-actualization theories of


The hypothesis of a relationship between unmitigated

Agency and cancer is David Bakan's unique contribution to the

field of personality theory.



Pair Numbers

1 Cancer subject: white male, age 42, with melanoma.

Non-cancer subject: white male, age 41.

2 Cancer subject: white female, age 40, with cancer

of the breast, age at menarche-ll, age at first

pregnancy-20, did not nurse, no significant family


Non-cancer subject: white female, age 39, age at

menarche-10, age at first pregnancy-19, did not

nurse, no significant family history.

3 Cancer subject: white male, age 24, with melanoma.

Non-cancer subject: white male, age 24.

4 Cancer subject: white female, age 30, with uterine

cancer, age at menarche-12, no pregnancies, no

significant family history.

Non-cancer subject: white female, age 29, age at

menarche-12, no pregnancies, no significant family


5 Cancer subject: white male, age 36, with Ewing's


Non-cancer subject: white male, age 37.

6 Cancer subject: white male, age 22, with Hodgkin's


Non-cancer subject: white male, age 22

7 Cancer subject: white male, age 19, with

malignant lymphoma.

Non-cancer subject: white male,

8 Cancer subject: white male, age


Non-cancer subject: white male,

9 Cancer subject: white male, age


Non-cancer subject: white male,

10 Cancer subject: white male, age


Non-cancer subject: white male,

11 Cancer subject: white male, age


Non-cancer subject: white male,

age 20.

24, with Hodgkin's

age 24.

28, with Hodgkin's

age 28.

56, with Hodgkin's

age 58.

66, with carcinoid

age 65.

12 Cancer subject: white female, age 33, with cancer

of the breast, age at menarche-13, no pregnancies,

no significant family history.

Non-cancer subject: white female, age 32, no

pregnancies, no significant family history.

13 Cancer subject: white female, age 48, with cancer

of the thyroid.

Non-cancer subject: white female, age 49.

14 Cancer subject: white female, age 27, with cancer

of the breast, age at menarche-10, no pregnancies,

no significant family history.

Non-cancer subject: white female, age 27, age at
menarche-ll, no pregnancies, no singificant family


15 Cancer subject: white female, age 50, with

metastic colon, no significant family history.

Non-cancer subject: white female, age 49, no

significant family history.

16 Cancer subject: white female, age 26, with

breast cancer, age at menarche-12, no pregnancies,

no significant family history.

Non-cancer subject: white female, age 26, age at

menarche-12, no pregnancies.

17 Cancer subject: white female, age 54, with uterine

cancer, age at menarche-13, age at first pregnancy

-30, no significant family history.

Non-cancer subject: white female, age 54, age

at menarche-13, age at first pregnancy-31, no

significant family history.

18 Cancer subject: white female, age 31, with acute

myelogenous leukemia.

Non-cancer subject: white female, age 31.

19 Cancer subject: white female, age 24, with breast

cancer, age at menarche-14, no pregnancies, no

significant family history.

Non-cancer subject: white female, age 23, age at

menarche-14, no pregnancies, no significant family


20 Cancer subject: white female, age 65, with cancer

of the colon, no significant family history.

Non-cancer subject: white female, age 65, no

significant family history.

21 Cancer subject: white female, age 60, with cancer

of the breast, age at menarche-14, age at first

pregnancy-25, did not nurse, no significant family


Non-cancer subject: white female, age 59, age at

menarche-14, age at first pregnancy-27, did not

nurse, no significant family history.

22 Cancer subject: white female, age 53, with acute

lymphocytic leukemia.

Non-cancer subject: white female, age 52.

23 Cancer subject: white female, age 25, with acute

myelogenous leukemia.

Non-cancer subject: white female, age 25.

24 Cancer subject: white female, age 65, with acute

myelogenous leukemia.

Non-cancer subject: white female, age 67.

25 Cancer subject: white female, age 35, with breast

cancer, age at menarche-12, age at first pregnancy

-24, did nurse, no significant family history.

Non-cancer subject: white female, age 35, age at

menarche-12, age at first pregnancy-22, did nurse,

no significant family history.

26 Cancer subject: white male, age 23, with acute

lymphocytic leukemia.

Non-cancer subject: white male, age 24.

27 Cancer subject: white male, age 25, with ganglio-

neuroblastoma (brain tumor).

Non-cancer subject: white male, age 26.

28 Cancer subject: white male, age 18, with Hodgkin's


Non-cancer subject: white male, age 19.

29 Cancer-subject: white female, age 60, with lung

cancer, has never smoked, no significant family


Non-cancer subject: white female, age 59, has

never smoked, no significant family history.

30 Cancer subject: white male, age 52, with osteogenic


Non-cancer subject: white male, age 51.



This is a study investigating a relationship between

personality and cancer. Subjects are selected for participa-

tion on the basis of current diagnosis and/or treatment of

cancer by a University of Florida Health Center physician.

There is no monetary compensation for participation.

I understand that I will be given a questionnaire on

which I will answer whether a series of statements are true

or false as applied to me. My responses are to be kept

confidential, and are to be used only for research purposes.

I understand that no hazards to my health will be in-

volved and that no compensation for injury can be received.

I understand that my participation is voluntary and

that I may withdraw at any time.




Signature Date


Abse, D., Wilkins, M., van de Castle, R., Buxton, W.,
Demars, J., Brown, R. & Kirschner, L. Personality and
behavioral characteristics of lung cancer patients.
Journal of Psychosomatic Research, 1974, 18(2), 101-113.

Angyal, A. Foundations for a Science of Personality.
New York: Commonwealth Fund, 1941.

Bacon, C., Renneker, R. & Cutler, M. A psychological
survey of cancer of the breast. Psychosomatic Medicine,
1952, 14(6), 455.

Bahnson, C. & Bahnson, M. Denial and repression of
primitive impulses and of disturbing emotions in patients
with malignant neoplasms. Annals of the New York Academy
of Science, 1966, 125, 827.

Bakan, D. The Duality of Human Existence. Boston:
Beacon Press, 1968.

Bakan, D. Disease, Pain, and Sacrifice. Boston:
Beacon Press, 1968.

Blumberg, E., West, P. & Ellis, F. A possible relation-
ship between psychological factors and human cancer. Psycho-
somatic Medicine, 1954, 16(4), 227.

Booth, G. Lung cancer and personality. International
Psychosomatic Cancer Study Group, August 10-13, 1960.

Brown, F., Katz, H., & Kaufmann, M. The patient under
study for cancer, a personality evaluation. Psychosomatic
Medicine, 1961, 23, 166.

Cobb, A. A social psychological study of the cancer
patient. Unpublished doctoral dissertation, University of
Texas, 1953.

Dale, E. & Chall, J. A formula for predicting read-
ability. Educational Research Bulletin, 1948, 28, 37-54.

Evans, E. A Psychological Study of Cancer. New York:
Dodd, Mead and Company, 1926.

Fraumeni, J. Persons at High Risk of Cancer. New York:
Academic Press, 1975.

Freud, S. Beyond the Pleasure Principle. New York:
Liveright Publishing Corporation, 1950. (Originally
published, 1920.)


Freud, S. The Standard Edition of the Complete Psycho-
logical Works. J. Strechey (Ed.). London: Hogarth Press,

Greer, S. & Morris, T. Psychological attributes of
women who develop breast cancer: A controlled study.
Journal of Psychosomatic Research, 1975, 19(2), 147-153.

Greer, S. & Morris, T. The study of psychological
factors in breast cancer: Problems of method. Social
Science and Medicine, 1978, 12(3-A), 129-134.

Greene, W. Psychological factors and reticuloendo-
thelial disease: I. Preliminary observations on a group
of males with lymphomas and leukemias. Psychosomatic
Medicine, 1954, 16(3), 220-230.

Greene, W. & Miller, G. Psychological factors and
reticuloendothelial disease: IV. Observations on a group
of children and adolescents with leukemia: An interpretation
of disease development in terms of the mother-child unit.
Psychosomatic Medicine, 1958, 20, 124-144.

Grinker, R. Psychosomatic aspects of the cancer
problem. Annals of the New York Academy of Science, 1966,
125(3), 876-881.

Grissom, J., Weiner, B. & Weiner, E. Psychological
correlations of cancer. Journal of Consulting and Clinical
Psychology, 1978, 43(1), 113.

Hagnell, O. The premorbid personality of persons who
develop cancer in a total population investigated in 1947
and 1957. Annals of the New York Academy of Science, 1966,
125(3), 846-855.

Huggan, R. Neuroticism and anxiety among women with
cancer. Journal of Psychosomatic Research, 1968, 12(3),
215-221. (a)

Huggan, R. Neuroticism, distortion, and objective
manifestation of anxiety in males with malignant disease.
British Journal of Social and Clinical Psychology, 1966, 7(4),

Jackson, D. Multimethod factor analysis of convergent
and discriminant validity. Paper read at the meeting of the
Society of Multivariate Experimental Psychology, Atlanta,
Georgia, 1966. Reprinted in Psychological Bulletin, 1969,
72, 30-49.

Jackson, D. Personality Research Form. Goshen, New
York: Research Psychologists Press, 1974.

Jackson, D. & Guthrie, G. A multitrait-multimethod
evaluation of the Personality Research Form. Proceedings
of the 76th Annual Convention of the American Psychological
Association, 1968, 177-178.

Jones, E. The Life and Work of Sigmund Freud, Volume
III, The Last Phase, 1919-1939. New York: Basic Books,

Kissen, D. Personality characteristics in males con-
ductive to lung cancer. British Journal of Medical Psychology,
1963, 36 1.

Kissen, D. & Eysenck, H. Personality in male lung
cancer patients. Journal of Psychosomatic Research, 1962,
6, 123.

Kusyszyn, I. A comparison of judgmental methods with
endorsements in the assessment of personality traits. Journal
of Applied Psychology, 1968, 52, 227-233.

Lerer, S. A comparative study of psychological factors
in illness of the mammary gland -- cancer and cystic fibrosis.
Revista Colombiana de Psiquiatria, 1975, 4(2), 127-144.

LeShan, L. & Gassman, M. Some observations on psycho-
therapy with patients suffering from neoplastic disease.
American Journal of Psychotherapy, 1958, 12, 730.

LeShan, L. & Worthington, R. Loss of cathexis as a
common psychodynamic characteristic of cancer patients: an
attempt at statistical validation of a clinical hypotheses.
Psychological Reprints, 1956, 2, 183. (a)

LeShan, L. & Worthington, R. Some recurrent life history
patterns observed in patients with malignant disease. Journal
of Nervous and Mental Disorders, 1956, 124, 461. (b)

Maddi, S. Personality Theories, A Comparative Analysis.
Homewood, Illinois: The Dorsey Press, 1968.

Murray, H. Explorations in Personality: A Clinical and
Experimental Study of Fifty Men of College Age. New York:
Oxford, 1938.

Murray, H. & Kluckhorn, C. Outline of a conception of
personality. In C. Kluckhorn, H. Murray, and M. Schneider,
Eds. Personality in Nature, Society, and Culture (2nd ed.)
New York: Knopf, 1956.

Muslin, H. & Pieper, W. Separation experiences and
cancer of the breast. Psychosomatics, 162, 3, 230.

Neumann, C. Psychische besonderheiten bei kredspatientinnen
Zeitschrift fur Psychosomatische Medizin, 1959, 5 91-101.

Peller, S. Cancer and its relations to pregnancy,
delivery, marital and social status: I. Cancer of the
breast and genital organs, II. Cancer of organs other than
reproductive: total cancer mortality. Surgery, Gynecology,
and Obstetrics, 1940, 71, 1-8.

Rank, O. The Trauma of Birth. New York: Harcord,
Brace, 1929.

Rank, O. Will Therapy and Truth and Reality. New York:
Knopf, 1945.

Renneker, R. Psychoanalytic explorations of emotional
correlates of cancer of the breast. Psychosomatic Medicine,
1963, 25, 11.

Reznikoff, M. Psychological factors in breast cancer:
A preliminary study of some personality trends in patients
with cancer of the breast. Psychosomatic Medicine, 1955,
17(2), 100.

Sandler, J. A test of the significance of the difference
between the means of correlated measures, based on a simpli-
fication of Student's t. British Journal of Psychology,
1955, 46, 255-226.

Schonfield, J. Psychological and life-experience dif-
ferences between Israeli women with benign and cancerous
breast lesions. Journal of Psychosomatic Research, 1975,
19(4), 229-234.

Schottenfeld, D. & Fraumeni, J. Cancer Epidemiology
and Prevention. Philadelphia: W. B. Saunders Company,

Selye, H. The Stress of Life. New York: Liveright
Publishing Corporation, 1956.

Tarlau, M. & Smallheiser, I. Personality patterns in
patients with malignant tumors of the breast and cervix:
An exploratory study. Psychosomatic Medicine, 1951, 13(2),

West, P. Origin and development of the psychological
approach to the cancer problem, in J. Gengerelli & F. Kirkner,
The Psychological Variables in Human Cancer: A Symposium.
Berkeley: University of California Press, 1954, p. 24.

Wheeler, J. & Caldwell, B. Psychological evaluation of
women with cancer of the breast and cervix. Psychosomatic
Medicine, 1955, 17(4), 264.


Diana Rivera was born Diana Luise Blanchard on May

5, 1944, in Crown Point, Indiana. She graduated from Crown

Point High School in 1962. She majored in psychology at

Valparaiso University, receiving the Bachelor of Arts degree

in 1972 and the Master of Arts in Liberal Studies in 1974.

From 1974 to 1977 Ms. Rivera was a member of the psychology

faculty at Valparaiso University. In 1977 she came to

Gainesville, Florida, with her daughter, Lisa, to pursue

graduate study in rehabilitation counseling at the Uni-

versity of Florida. She was awarded the Master of Health

Sciences degree in 1978 and entered the doctoral program

in counselor education, specializing in theories of person-

ality and counseling. She was married in 1982 to James H.

Pitts. Currently residing in Gainesville, Ms. Rivera is a

Florida Licensed Mental Health Counselor and a Certified

Rehabilitation Counselor. She expects to receive the degree

of Doctor of Philosophy in August of 1983.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

I / /

Ted Landsman
Professor of Psychology and
Counselor Education

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Jane ly. Lara.
Prgogessor or Counselor

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

ames G. Joit5 r
Associate Professor of
Rehabilitation Counseling

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
r, /

John P. Saxon /
Associate Professor of
V Rehabilitation Counseling

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Roy/S. Weiner
Professor of Immunology and
Medical Microbiology

This dissertation was submitted to the Graduate Faculty of
the Department of Counselor Education in the College of
Education and to the Graduate School, and was accepted in
partial fulfillment of the requirements of the degree of
Doctor of Philosophy.

August, 1983
Dean for Graduate Studies
and Research

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs