Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Referential Dissociation And Response To Stress
(USC Thesis Other)
Referential Dissociation And Response To Stress
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
REFERENTIAL DISSOCIATION
AND RESPONSE TO STRESS
by
J. Morgan Thomas
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Sociology)
June 1970
71^16,440
THOMAS, J. Morgan, 19 35-
RE FE RENT IAL DISSOCIATION AND RESPONSE TO
STRESS.
University of Southern California, Ph.D., 1970
Sociology, general
University Microfilms, A XEROX C om pany , Ann Arbor, Michigan
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED
UNIVERSITY O F S O U T H E R N CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES. CALIFORNIA 9 0 0 0 7
This dissertation, written by
......J.:„.Mor5[an_Thomas.................
under the direction of h.X£... Dissertation C om
mittee, and approved by all its members, has
been presented to and accepted by The G radu
ate School, in partial fulfillment of require
ments of the degree of
D O C T O R O F P H I L O S O P H Y
' D ean
D a te 055®„.L?.7 . 9.
DISSERTATION COMMITTEE
latrtnart
ACKNOWLEDGMENTS
I wish to thank my wife, Connie Thomas, for making
this dissertation possible and Herman Turk for his
careful supervision and guidance during the writing.
I am grateful to the Drinking Driving and Traffic
Safety Project and, in particular, Oksana Didenko, who
made available the data runs used for analysis.
Graduate training and the initial draft of the
dissertation was completed under a National Institute
of Mental Health funded Ph. D. training program in the
Demography of Social Disorganization, Grant No. MH-10243.
The literature review was accomplished in part
under Contract No. FH-11-7099, U.S. Department of
Transportation, Drinking Driving and Traffic Safety
Project, University of Southern California.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ................................ ii
LIST OF TABLES................................ V
Chapter
I. REVIEW OF THE LITERATURE ON DRINKING
DRIVING, ACCIDENTS, AND STRESS........ 1
Introduction 1
Accident Proneness 2
Alcohol as a Causal Factor 4
Alcoholism 7
Life Style - "A Man Drives as He Lives" 9
Variables Linked with Accidents and
Drunk Driving 11
Background Variables 11
Demographic Factors 15
Immediate Stressful Events 18
Stress and Response 19
Cause of Stress 2 4
Summary 27
II. RESEARCH METHODOLOGY AND DESIGN .... 28
Introduction 28
Data Collection 28
Representativeness of the Sample 31
Problem of Using Convicted Drunk
Drivers as Representative 34
Design 35
Summary 3 6
III. DATA ANALYSIS.......................... 3 7
Introduction 37
Data Analysis 40
Possible Interpretations 42
Interpretation of Results 46
iii
Chapter Page
IV. DEVELOPMENT OF REFERENTIAL DISSOCIATION
THEORY.................................... 6 8
Introduction 68
Referential Dissociation 72
Types of Dissociation 7 3
Concept of the Social Self 79
Control and Identity 81
Concept of Stress 85
Summary 89
V. RESPONSE TO THE STRESS OF REFERENTIAL
DISSOCIATION ............................ 91
Introduction 91
Generalized Response Categories 92
Adaptive vs Maladaptive Responses 94
Response to Rejective Dissociation 96
Suicide as the Result of Loss
of Control 97
Frustration-Aggression 102
Mental Illness 104
Summary 10 6
VI. RECAPITULATION............................ 107
Introduction 107
The Link Between the Literature Review,
the Data, and the Theory 108
Alternative Explanations 112
Proposed Design 113
Relationship between "Deviant Responses" 117
VII. SUMMARY.....................................119
VIII. SUGGESTIONS FOR FURTHER RESEARCH .... 122
Stress and Adaptation 122
Variables in Future Research 125
REFERENCES.........................................127
iv
LIST OP TABLES
Text Table
Page
A Proportions of Driver License Applicants
and Convicted Drunk Drivers for Court
Sample and for Los Angeles County, 196 8,
by Race-Ethnicity....................... 33
B Proportions of Driver License Applicants
and Convicted Drunk Drivers in Sample
and Los Angeles County, 1968, by Sex. . . 33
Table
1. Percentage of Population within Driver
License Applicant and Offender Categories
for Seven Indices of Separation, Los
Angeles County, 196 8....................... 49
2. Z Scores Between Driver License Applicants
and Offender Categories for all
Separation Indices......................... 50
3. Percentage of Population within each Offender
Category for Six Indices of Separation by
Age .....................................51
4. Percentage of Population within each Offender
Category for Six indices of Separation by
S e x........................................... 52
5. Percentage of Population within each Offender
Category for Six Indices of Separation by
Race........................................... 53
6. Marital Status, by Length of Time Living
Alone, Driver License Applicants and
Offender Categories, Los Angeles County,
1968......................................... 54
v
Table Page
7. Separation Index - Time Living Alone,
Driver License Applicants and Offender
Categories, Los Angeles, 1968 ........... 55
8. Separation Index - Presently Employed,
Driver License Applicants and Offender
Categories, Los Angeles County, 19 68. . . 56
9. Measures of Association between Indices
of Separation and Offender Categories . . 57
10. Separation Index - Person Lived with until
Age 16, Driver License Applicants and
Offender Categories, Los Angeles
County, 1968.............................. 58
11. Number of Kinds of Separation Experienced
by Driver License Applicants and Offender
Categories who Reported Problems with
Alcohol, Los Angeles County, 19 60 . . . . 59
12. Percentage of Population within each
Offender Category for Six Indices of
Separation by Alcoholism.............. 61
13. Separation Index - Not Working - Number of
Jobs in Five Years, Driver License
Applicants and Offender Categories,
Los Angeles County, 196 8............... 62
14. Separation Index - Employed Number of Jobs
in Five Years, Driver License Applicants
and Offender Categories, Los Angeles,
1968 .................................... 63
15. Number of Times Married, Driver License
Applicants and Offender Categories,
Los Angeles County, 1968 ............... 64
16. Total Minor Violations (Traffic), Driver
License Applicants and Offender
Categories, Los Angeles County, 1968 . . 65
Table Page
17. Total Major Violations, Driver
License Applicants and Offender
Categories, Los Angeles County, 1968 . . 66
[
k
18. Percentage of Population within each j
Category, Driver License Applicants j
and Convicted Drivers, for Seven
Indices of Separation by Occupational
Status.................................. 67 ;
i
CHAPTER I
REVIEW OF THE LITERATURE ON DRINKING I
DRIVING, ACCIDENTS, AND STRESS j
i
Introduction j
This dissertation has developed out of an attempt toi
i
find a unifying concept which will explain the drinking |
i
driving and accident phenomenon. Three major directions J
will be explored - a literature review of drinking driv- j
i
ing and accidents, an analysis of data available on j
i
i
convicted drunk drivers, and a theoretical exploration
of the major points arising from both the literature and
j
the data analysis. j
i
The review of literature will concentrate on two
types of events - drunk driving and automobile accidents.j
Both these events are linked in the search for a unifying!
I
principle because past research has labelled them
responses to various forms of stress. The purpose of
the review will be explored further using data from a |
large drunk driving population in a metropolitan area. I
I
i
On the basis of the review of literature and suggestions !
!
emanating from a consideration of drinking driving data, j
a theory is developed which utilizes stress arising from j
perceived loss of control of the environment. |
2
Accident Proneness
In the search for a common element in the two
responses of drunk driving and accidents, the following
theoretical approaches will be explored: accident
proneness, alcohol as a causal factor, alcoholism, and
life style. Also explored are those variables which
have been linked with accidents and/or drunk driving.
In the past, one of the most prominent ideas of
accident causation was the accident proneness concept
suggested by Greenwood in 1919 and 19 20. Briefly, this
concept relates to the supposed propensity of some
individuals to have more accidents due to specific psy
chological and response attributes. Tillman in 1949
conducted a study which indicated that a small proportion
of the bus drivers in his sample contributed dispropor
tionately to the total number of accidents. Similar
studies attempted to identify specific social and psycho
logical characteristics common to the higher accident
group.
Subsequently, several studies (Thorndike, 1951;
Webb, 1955) have been done which tend to indicate that
the accident proneness concept was not tenable statis
tically when attributed to a total population. A
Poisson distribution applied to accidents among large
groups of drivers does not support the notion of accident
proneness since "some individuals will have more acci
dents on the basis of chance alone." Similarly, it is
not the same individuals who have accidents year after
year (Thorndike, 1951). McFarland and Moore (1955) have
reviewed the different attribute and ability studies and
have indicated the general lack of association between
tests and accidents.
McFarland and Moore (1955) in reviewing the
accident-proneness literature indicated the relative
lack of significant differences between accident and
control populations on sensory and psychomotor abilities.
However, they had noted minor differences between indi
viduals with high and low accident rates on certain
psychosocial and temperamental attributes.
Generally, there appears to be insufficient evi
dence for a strict concept of accident proneness. If
there are individuals who tend to have more accidents,
the environmental factor, precipitating events, social
supports, and availability of particular defense mechan
isms, in conjunction with accident potential situations
(i.e., availability of car) must all occur in associa
tion with each other to produce the accident. This may
not occur often enough to the same individual so that
he can be identified by frequency of accidents. It
4
would appear more reasonable that several different
personality characteristics and pathologies would tend
to heighten the chance of accidents when specific
stresses are present along with low adaptation ability
and contributing environmental factors. McGuire (1970)
has suggested that there are two major forms of accident
proneness, shortr-term proneness involving external
factors and long-term proneness involving internal
factors and character traits. While this neither con
firms nor disaffirms the presence of stress as a major
variable, it leaves the way open for further refinement
and interpretation. Given specific situations, certain
individuals are more apt to respond inadequately. What
are these situations, and what are the characteristics
of the individual? These questions will be dealt with
later.
Alcohol as a Causal Factor
Perhaps more research has been done on this partic
ular idea than any other in traffic accident research.
The effects of alcohol on the physiology of the individ
ual have been well documented. To name just a few
studies and documentation in this area, Carpenter (1962)
has reviewed the effects of alcohol on reaction time,
motor skills, and positional nystagmus. Gruner, Ludwig,
and Domer (1964), Talland (1966), and Moskowitz and
5
Depry (1968) have documented the effects of alcohol on
auditory vigilance and divided attention tasks; Goldberg
(1943) studied the ability to concentrate and comprehend;
and Loomis and West (1958) recorded the ability to per
form a simulated driving task. In each case, the main
effect was a slowing down of the processes of integra
tion, adaptation, and reaction time. Accident research
has come up with some interesting findings: McCarroll
and Haddon (1962) found that 41 percent of the accident
responsible group had a marked degree of drunkenness
(Blood Alcohol Content [BAC] level of 0.25 percent and
greater). This degree of drunkenness was not found in
the control groups. "It appears highly probable that the
use of alcohol was a casual factor in the deaths of one-
half or more of the drivers killed in the single vehicle
accidents... studied." Similarly, Gerber (1962) found
alcohol present in 34 percent of the drivers in the
fatal accident population studied in Ohio.
Schmidt and Smart (1959) in their research paper
remarked that the alcoholic drivers had more accidents
when drunk but no more than the general population when
not drinking. They indicate that the difference is
related to the frequency of consumption and to the phy
siological effects of alcohol instead of specific
personality factors.
6
Smith and Popham (1951) found a significant excess
of accidents occurring to individuals with levels as low
as 0.03 - 0.05 BAC over those with no BAC level.
Similarly, Lucas, et ad. (1955) found that individ
uals with BAC levels over 0.15 percent were 8 times
more likely to be in an accident as in a nonaccident
group. Holcomb in 1938 found 33.2 percent of the acci
dent drivers as compared with 6.3 percent of the controls
with BAC levels over 0.05 percent. Alcohol levels of
0.05 percent plus were found in 22.4 percent of accident
involved drivers versus 8.5 percent for controls matched
for time and place. Other studies document these find
ings (Barmack & Payne, 1961; Haddon, 19 62; Borkenstein,
et al., 1964). Further, Hyman (196 0) hypothesized that
about 80 percent of all men between the ages of 20 and
60 drink at least occasionally and about 90 percent
drive at least occasionally. Therefore, most adult men-
will probably on some occasion drive while under the
influence of alcohol. The amount of hidden "delinquency"
here is tremendous. But why, then, do so few of these
millions end up in accidents? Mainly because drinking
does not always lead to accidents. But still, while
50 percent of the fatal accident population can be
linked to alcohol sufficient to cause deterioration of
driving skill, we still must account for the other 50
percent with relatively low BAC levels or with alcohol
involvement.
The main import of these results is that while
alcohol is present in a large portion of the accident
population, it is certainly not the only factor involved
and leads to the suspicion that there may be some other
factor which causes some individuals to become involved
in accidents whether or not intoxicated.
Alcoholism
Alcoholism appears to be related to drunk driving
and to accidents. However, in most cases, the felony
driving records of drunk driving accident individuals
are brief or nonexistent indicating lack of serious or
fatal accidents.
Selzer (1966) showed that a population of 50 alco
holic drivers had twice the moving violations and acci
dents as a non-alcoholic group. However, severity of
these accidents was not rated. Schmidt and Smart (1959)
made the observation that the accidents of 98 alcoholic
clinic patients were almost always preceeded by drink
ing, but when not drinking, the accident rate of this
population was no higher than the rate of the general
population. This finding has been replicated by Selzer,
Payne, Quinn, and Westervelt (1965). They further noted
that nearly all of the accidents in this group were not
8
of the multiple vehicle collision type (i.e., they
generally ran off the road or into a fixed object).
A study by Payne indicated that 49 percent of the
alcoholics sampled said they drive more recklessly when
drinking while only 16 percent of the non-alcoholic
population indicated the same inclination to drive more
recklessly, and 28 percent indicated that they never
drove when drinking. However, the entire population of
alcoholics admitted driving after drinking. While
alcoholics probably drive more often after drinking than
the general population, it is possible that the popula
tion is less likely to have fatal type accidents. This
does not contradict the fact that fatal accident popula
tions generally have a high proportion of alcohol mis
users in their ranks.
According to Waller (1968) "... about half of (the)
persons killed or severely injured in crashes have been
drinking, usually in quantities that substantially exceed
average social drinking patterns." This suggests that
problem drinking and alcoholism may be present.
Selzer has documented this further by showing that
37 percent of the fatality group in his sample were
alcoholics.
The high percentage of alcoholics in the accident
population tends to suggest that alcoholism is a strong
9
factor in accident etiology. It is the single predom
inant factor which has been identified, although it
cannot account for the total number and severity of
accidents on our highways. How does it do this? It
is possible that the personality characteristics of
certain types of alcoholics increase the possibility of
accidents when repressed aggression {self or other
directed) is released, and although alcohol may lower
the ability to react quickly, this fact does not provide
i
the complete explanation we seek nor answer our questionsj
concerning the over-all cause of the serious accident.
Thus, alcohol or alcoholism alone does not provide a J
sufficient answer to the problem. j
Life Style - ”A Man Drives as He Lives"
i
One of the more interesting theses that have been !
I
developed in accident research is the idea that the I
individual drives as he lives. The accident population
appears to contain a high proportion of individuals who
lead an erratic life, who have many social problems, and j
]
who have experienced and reflect primary group disorgan- I
I
ization in childhood and as an adult. I
!
Tillman and Hobbs (1949) using 96 accident fatali- I
i
ties in a Canadian study, found that 66 percent of the i
I
repeaters were known to one or more public agencies as j
I 10 i
; against 9 percent of the accident-free drivers. Similarly!
j McFarland (1954) in a study of truck drivers found
8 0 percent predictability using past public agency
records. Waller (1967) in a public agency check found |
; that 87 percent of the drunken drivers, 76 percent of
| the drinking drivers with accidents, and 39 percent of
i the nondrinking drivers with accidents were known to
| community agencies. Further, number of arrests per
person for drinking accident drivers was 3.9 in the
: Waller study and 1.6 for nondrinking accident drivers.
i
: 1
; Further, the percent of persons with two or more alcohol
; i
: 1
; arrests in the drinking or hit-and-run accident sample j
i was 50 percent. The percent of persons in the drinking j
I
or hit-and-run sample with non-traffic offenses on their j
j
, records was 66.7 percent and for the accident without j
I drinking 33.3 percent. j
i
; . 1
Barmack and Payne found that 87.5 percent and [
! !
: 73.5 percent of the drinking accident and nondrinking j
; accident population had previous traffic infractions as j
against 50.0 percent of the control population. j
! i
While such knowledge aids in predicting accidents, j
; i
i it does not lead to an explanation of the accident j
i I
! phenomenon or why the accident occurred at a particular i
; i
; time. It does lend support to the major thesis of this
■ paper that there may be a variable present which appears
i
■ to lie behind the problem drinker, the alcoholic, and I
: 11 ’
| the individual who continues to recidivate that may |
| account for a large amount of the predicitve variance.
j
| The presence of past records with public agencies may
I indicate that the individual has not been able to suc-
j cessfully handle his environment. Extensive criminal
!
[
records may indicate an unwillingness or an inability
j
! to handle life in a normative way.
j E
Variables Linked with Accidents I
; and Drunk Driving j
In the past, several variables had been found to I
I
i
be related to accidents or drunken driving such as j
i
j ground variables (i.e., childhood trauma, family dis- j
organization, history of truancy), demographic variables j
(i.e., age, sex, marital status), and specific psychol
ogical variables related to stressful events preceding
the accident or drunk driving offense.
i
; Background Variables
; Tillman and Hobbs (194 9) have shown that the high-
i accident driver as against the low-accident driver has
had greater disorganization in childhood (aggression
i
; and phobias, truancy, and disciplinary problems) and
i
! relatively poor adult adjustment (frequent AWOL's,
j
; sexual promiscuity, admitted bootlegging, job history
I
of being fired, and five or more previous jobs).
12
Barmack and Payne (1961) attempted to account for
early trauma (separation of one or both parents from
the child before age 16) and for the incidence of prob
lem drinking among parents of the subject population.
The largest differences in incidence of trauma and
parental problem drinking appeared between the drinking
and the nondrinking accident populations rather than
between the accident and control populations with a
greater proportion of the drinkers having had childhood
trauma and one or both parents as problem drinkers.
Unfortunately, no other major studies have been
done on background or childhood variables in relation
to the drinking accident problem. The Tillman-Hobbs
and Barmack-Payne findings must be taken as suggestive
only.
Recently, there have been a number of studies on
personality factors as related to accident and non
accident populations, as well as alcoholics and non
alcoholics within an accident population. The major
assumption appears to be that certain psychological
patterns lead to increased accidents.
Selzer1s study (1964) identifies several general
psychopathological characteristics of drunk driving
and accident populations.
Selzer (1963) indicated that 67 percent of a drunk
driving population suffered from one or more classfiable
psychiatric illnessess exclusive of alcoholism. It was
further estimated that "... no less than 78 [percent] of
the arrested drivers had pathological drinking problems.
These included passive-aggressive personalities (most
numerous), alcoholism, schizophrenia, psychoneurosis,
depression, sociopathic personalities, and paranoia.
Selzer found 41 percent of a population of auto fatali
ties had one or more symptoms of psychological disorder
including paranoid thinking, suicidal thoughts or acts,
depression, and violence. This is contrasted with 17
percent disorders within the control group. When the
fatality population is subdivided into alcoholic-non
alcoholic the percentage of personality disorders rises
to 69 percent for the alcoholic and 25 percent for the
non-alcoholic populations.
One of Selzer1s conclusions indicates that auto
mobile accidents might be purposely caused by the
alcoholic to express regressed anger. Backing up this
assertion is much alcoholic research which has shown
the chronically hostile state of the alcoholic and the
inability to release hostility due to a dependence need
and a fear of alienating those upon whom the alcoholic
is dependent.
14
At an earlier date, Conger (1959) identified
similar personality characteristics in a population of
10 airmen and 10 controls — although the sample was
small, the findings appear to be confirmed by Selzer's
study and by previous accident research in general. For
the accident repeater, the variables hostility with low
control, aggressive behavior, egocentricity, separation
anxiety, fantasy preoccupation, and stimulus-bound low
tension tolerance appeared more frequently than for a
non-accident control population. There was consensus,
however, on the findings that the accident population
did not fall into a single psychopathological category.
The most recent study coming to our attention is
the Baylor University of Medicine study of 25 fatal
accident drivers. Seventy-six percent of the accident
fatal population had personality disorders as against
8 percent of the control group. The largest single
category was alcoholism (60 percent) with antisocial
personality (sociopath) falling next (24 percent). Other
disorders included paranoid (4 percent), obsessive-
compulsive (8 percent), hysterical (8 percent), passive-
dependent (4 percent), passive-aggressive (4 percent),
and schizoid (4 percent).
Several conclusions can be drawn from the above
studies: (1) Alcoholism as a clinically diagnosed
1
15 !
disorder appears to be a consistently present factor in |
the fatal accident population; (2) Various clinically j
evident personality disorders including alcoholism are j
generally found in the accident population, particularly ]
among those classified as had-been-drinking; {3) The
proportion of the accident population which had-been-
drinking range from approximately 50-76 percent. This
depends heavily on the sampling methods used or whether
i
i
any sampling methods were used at all; (4) The excessive
intake of alcohol for a majority of the had-been-drinking
group indicates more than merely social drinking. Since
a fair percentage of the high BAC group do not have fatty
changes in their liver (indicating alcoholism), the
heavy drinking may have been related to a specific stressJ
Further, alcoholism appears to be combined with other
!
I
clinically observable personality disorders. It may be j
|
that these disorders in conjunction with excess alcohol I
i
intake, when combined with an environmental stress j
j
precipitate the fatal "accident." j
!
Demographic Factors
There are several environmental factors which
appear to have a strong correlation with traffic acci-
i
dents. These are age and marital status. We will briefljj
1
discuss the major findings related to these variables. !
16
The majority of studies have indicated that age is
a dominating factor (see for example, Borkenstein, 1964;
Haddon, 1962). Youth and the elderly appear to be over
represented in the accident population. Impulsiveness
and the problems of transition to adulthood along with
untrained driving are cited as responsible for accidents
to youth while the declining abilities of the aged are
cited for accidents at the other end of the age spectrum.
For the younger ages, there is the transition from child
to adult with the rise of new sets of expectations, new
responsibilities, and marriage.
Pelz and Schuman (1967) in their study of young,
dangerous drivers found a high proportion of youth who
were finding the transition from child to adult diffi
cult and were working rather than continuing schooling.
But how do we account for the slight rise in number
of accidents for men around age 35-50 (Lauer, 1952;
Brown, 1968)? Certainly, the increased accident rate
is not due to the same factors found in the younger and
older age groups. For the middle aged, changes of jobs,
careers, realization of nonaccomplishment of youthful
ambition may play a large role, and for the aged there
is retirement, loss of loved ones through death, and
the changed status of being old. However, if there
is poor adaptive ability at any age then rapid changes
17
can be extremely stressful.
The other variable which appears to discriminate
in all accident populations is marital status. The
married-living-apart often appears to be over-represented
in the accident group. This higher rate relative to the
control population has not been adequately explained and
may be due to varying degrees of stress related to a
perception of not belonging, loss of a sense of identity,
or even a lessening of the restraint placed on the
individual by the society. Of interest here is the
percentage of accidents attributed to the separated and
not married categories. The separated have the highest
accident rate relative to the control population fol
lowed by the divorced and then the single and the
widowed {Brown, 196 8). Barmack and Payne found that
the proportion of men in the accident group who were not
living with their wives lent support to the idea "that
the drinking accidents may be selective of individuals
with current marital adjustment problem(s)." Brown
(1968) substantiated the findings relative to the adjust
ment problem by showing that 80 percent of the fatality
population had experienced one or more stresses within
the 24-hour period preceding the accident as compared
with 12 percent for the controls. The major category
was interpersonal conflict.
18
Immediate Stressful Events
The recent emphasis on stress in accident research
has focused on the delineation of major "stressful”
events in the individual's life prior to his accident.
Several studies have concentrated on obtaining informa
tion on stress and have indicated the highly predictive
nature of the stressful events. Selzer (1967) found
through an investigation of personal conflict/ tragedy,
and vocational and financial stress among the accident
population that the presence of one or more stresses
was evident in the accident population in 52 percent
of the cases as against 18 percent for the control
population. A subdivision of the accident population
had one or more stresses in 72 percent of the cases and
the non-alcoholic in 42 percent of the cases.
Tabachnik, et a^. (1967) in their attempt to
differentiate a suicide population from an automobile
"accident population" found that the only variable
found consistently in the "accident" population was a
recent change in the individual's life in relation to
a job or to his immediate family. In most cases, these
changes appeared to be beneficial increases in respon
sibility (promotion or a new baby, etc.). However,
these changes may have created stress due to a temporary
inability to cope with the change thereby resulting in
19
conflict or intrapersonal stress which could not be
readily resolved. Tabachnik indicated that the individ
ual might be facing an identity crisis and was attempting
to attain a psychological moritorium.
The above studies can only be suggestive of the
part played by stress in the etiology of accidents.
However, there are ample implications from other studies
that immediate stress and the inability of the individual
to cope with and adapt to environmental changes genera
ting stress is an important factor in much pathology.
Stress and Response
The major factor which has consistently appeared
in past research is nonnormative response to the environ
ment. The idea of stress as the precipitator of much
deviant behavior has been suggested in several of the
theories. The individual is continually forced to
react to changes in the environment. Such changes can
pose problems for the individual and cause stress as a
preadaptation state. It is, however, the individual
perception of the stress and the adaptive response made
or attempted which is of importance in determining the
behavioral outcome. It is the behavioral outcome with
which we are concerned here. If different interpreta
tions of stimuli and adaptive responses are indicative
of different degrees of ability to successfully cope
20
with changes in an individual's environment, then those
individuals who cannot successfully adapt to a particular
event are more likely to give an inappropriate response.
Such responses might include drunk driving, accidents,
suicide, psychosomatic disease, and mental illness. It
is our purpose to look at the findings of past research
from the stress-adaptive-response viewpoint to see if
such a perspective unifies the findings of accident
research to a greater extent than previous single factor
theories.
The preceding suggests that the accident proneness
concept may still be tenable. Proneness may not be
manifested consistently since individuals with low
adaptive ability when faced with rapid change tend to
insulate themselves against such change. However, when
change does occur and other defenses are not readily
available to protect the individual, then there is a
greater possibility of inappropriate responses. Certain
maladaptive responses might lead to greater accident
potential. Such responses to stress — as inattention,
uncontrolled aggression, obsessive rumination, fantasy,
proneness to sleep, drunkenness — would all be danger
ous when engaged in by the driver of a car. Thus, those
who have obsessive, compulsive, or highly impulsive
reactions would be more likely to end up in the accident
population when faced by a severe stress.
21
It is possible, given the unequal distribution of
major stresses in the lives of most persons and the
adequacy of minimal defenses for low level stress, that
major accidents occur so infrequently to poor adaptors
as not to show up more than once in accident populations
selected for study (thus, the lack of significant
findings as to accident proneness when large populations
are tested over time).
Looking at alcohol as a cause of accidents gives
further evidence of the part stress and differential
adaptive response might play in accidents and in drunk
driving. Thus, there may be a specific set of person
ality characteristics in some subsection of the alcoholic
population which makes them more susceptible to accidents
when intoxicated or when .there is a particularly high
stress to which the individual finds it hard to adapt.
Of particular interest are the accidents which are not
alcohol involved. Differentiating betxveen the had-been-
drinking accident population and the had-not-been-
drinking accident population may lead to the findings that
different coping mechanisms-are attempted by individuals
in the two populations. It is possible that the drunk
driver with many small accidents is alleviating stress
through chronic drinking while the fatal accident driver
finds drinking does not alleviate stress, but only makes
22
it worse. It has been shown in at least one study
(Waller, 1968) that the fatal accident driver does not
generally have long records of alcohol-related offenses
or other accidents which indicates that there may be a
rather short-term period of stress for which the
individual does not have adequate coping mechanisms.
One other possibility is a lower visability in terms of
police arrest segment of the population for this segment.
Also, there is the possibility that these are younger
individuals who have not had time to accumulate long
records. This possibility needs to be tested.
Selzer's work has illuminated the part that imme
diate stress plays in the etiology of accidents. This
stress combined with alcohol as an intervening variable,
particularly at high BAC levels, can create the neces
sary lack of judgment or the fatal burst of aggression,
whether self or other directed, to cause a major acci
dent. The percentage of psychopathology in the accident
population attests to the inadequacy of response already
in operation in the lives of the fatal accident drivers.
Poor adaptability when united with a major stressful
event may lead to self-destruction.
Alcoholism alone cannot account for even a majority
of the fatal accidents. As was indicated earlier, the
23
alcoholic generally makes up only 25 to 33 percent of
the fatal accident population. Certainly/ alcoholism
unrelated to other factors (i.e., personality variables)
cannot be called the single major cause of accidents
although a significant portion of every accident popula
tion contain response tendencies exist in the alcoholic
which may lead to alcoholism and to accidents (self
destruction).
The inadequacy of response to the ordinary stress
of life is clearly illustrated by the "drives as lives"
hypothesis. The fact that many public social agencies
possess records on the drunk driver is a testimony to
the drunk drivers continual search for help in adjusting
to and continuing to exist in society. Maladaptive
responses might tend to accumulate leading either to
increased minor accident potential or to desperation and
a suicidal tendency.
Waller's finding that the drinking accident group
and the drunk driving group contained the largest
percentage of persons known to the welfare and probation
departments supports the "inappropriate response"
hypothesis. Also of significance is the finding that
36 percent of the drinking accident population and
27 percent of the drunk driving population have arrest
records for violence. Could this indicate an inability
to contain major stress?
24
Certainly, the inability to respond adequately to
society and to the stresses experienced needs to be
taken into account in developing research models for
identifying the drunk driver and the fatal accident
driver.
Cause of Stress
Selzer's work on deceased drivers and Brown's work
on drunken drivers indicated that a majority of these
populations had personal and social conflicts within
the recent past (prior to accident or drunk driving ar
rests) . Tabachnich's work (1967) on suicide indicates
that there is one variable which consistently appears
in the histories of a "near fatal" accident population -
a recent or contemplated "favorable" event such as
fatherhood, job promotion, etc....... The hypothesis
advanced is that these individuals are facing an increase
of responsibility and are going through an identity
crisis and are looking for a psychological moritorium.
However, a different interpretation can be placed on
these findings. Since the basic personality of the
auto-death group included such characteristics as dis
tantly integrated with significant others, active and
exhibitionistic, and upset when slighted or criticized,
it can be hypothesized that these individuals needed
greater reference- group support for self-valuation and
25
had fewer inner resources to draw on for self-
identification. Thus, any change which caused a
perceived withdrawal of previously supportive reference
groups; e.g., the withdrawal of a wife as attention is
shifted to a new baby and the withdrawal of previous peer
groups through a change of status, has disastrous effects
with a resultant loss of ego support and a loss of a
definition of the individual as meaningful.
The work of McCord and McCord on alcoholism and the
data gathered from the Sommerville Youth Study indicated
that those youth who later became alcoholic were des
cribed as aggressive, apparently self-sufficient, and
critical of their families indicating a possible attempt
to overcome dependency needs through rejection and self-
reliance .
The McCord's interpretation was that as these boys
approached manhood the dependency needs became more
insistent and the possibility of relations to new
reference groups shifted and withdrew and new ones were
not easily found; therefore, escape was found through
alcohol.
Studies of accident repeaters highlights the fact
that those who are separated from others are more often
found in the accident repeater groups. Conger (1959)
found that accident repeaters had less hostility control,
26
greater aggression, extremes of egocentric-sociocentric
attitudes, and high separation anxiety among other
traits. Each of these traits should indicate both actual
and perceived separation from others.
Tillman and Hobbs in 1949 indicated that accident
repeaters had army, job, and marriage instability as well
as instable backgrounds (i.e., parental-marital trouble,
excess childhood phobia, and school strain and adjust
ment problems). These portray a disjuncture between the
individual and his major reference points. Supporting
these findings, Barmack and Payne (1961) compiled data
which indicated that drinking-accident-involved drivers
as against a control group have a more extensive back
ground of childhood separation from parents before the
age of 13 or were separated and raised by others for at
least six months while both parents were still alive.
They further found that individuals in the married
- living apart category were over-represented in their
sample. While there appeared to be no significant
difference within the accident group, there was a sig
nificant difference within the drinking-accident and
the control groups (22.5 percent vs 8.0 percent).
Further, the percentage of drivers who were married but
currently living alone and who reported broken childhood
homes was higher among the drinking accident drivers
27
(35 percent) than among the nondrinking accident group
(14.3 percent).
Summary
The studies reviewed suggest that one of the
principle factors in accidents and drunk driving might
be separation from others and the resulting stress lead
ing to attempts to adapt to the environment albeit non-
normatively and often inadequately. In order to subs
tantiate this suggestion, data collected on 764 convicted
drunk drivers and driver1s license applicants were
analyzed. The results of the analysis led to the devel
opment of a unifying theory of behavior, stress, and
maladaption.
CHAPTER IX
RESEARCH METHODOLOGY AND DESIGN
Introduction
This chapter describes the selection of the drinking
driver sample, the collection of the data, and the
selection of the control group. Also included is a
discussion of the problems encountered in using convicted
drinking driving offenders as representative of a deviant
population.
The primary data was gathered as part of the
Drinking Driver and Traffic Safety Project, University
of Southern California, funded initially under a contract
with the State of California and subsequently taken
over by the U.S. Department of Transportation under
a three-year contract. The data has been collected
as part of an effort to construct a prediction model
to identify potential drinking drivers and to predict
recidivist drinking drivers.
Data Collection
Interviews and questionnaires were completed during
a five-month period by convicted drunk drivers in four
courts in Los Angeles County which together handle
approximately 60% of the total drinking and driving
29
cases in Los Angeles. These courts included Division
35 and Division 51 of Los Angeles Central Municipal
Courts, the Van Nuys Municipal Court, and the East
Los Angeles Municipal Court. Financial and practical
considerations did not permit procuring data from the
entire volume of cases passing through these courts.
As a result, a random sample was selected. This was
done by the interviewer obtaining each day's arraign
ment list. The data base is made up of 1,320 persons
selected at the Department of Motor Vehicles and 7 04
court respondents. The court respondents are broken
down as follows: 358 first offenders, 171 second offen
ders, 132 three-plus offenders, and 103 with reduced
convictions of reckless driving. The reckless driving
category is not used for analysis except in two tables,
"Marital Status by Length of Time Living Alone" (Table 6)
and "Number of Kinds of Separation Experienced"(Table 11),
where they are lumped together with the first offender
drunk drivers due to similar population characteristics.
Utilizing first and two or more offenses as categories,
a table of random numbers was used to select one-third
of the first offenders (N=358) and three-fourths of the
two or more offenders (N=303). Again using a table of
random numbers, five or six of those initially selected
30
were chosen to be interviewed; the remainder were given
questionnaires.
A comparison group was selected from Department
of Motor Vehicle (DMV) license applicants. In order
to obtain a representative population, race/ethnic
and socioeconomic proportions were looked up for Los
Angeles County from Census data. The DMV offices from
which subjects were to be drawn were picked-on the basis
of these statistics. The control sample (N=1320) was
drawn from the Van Nuys, Montebello, Torrance, Long Beach,
and Compton offices.
The sampling procedure at the Department of Motor
Vehicles consisted of interviewing the first available
license applicant each morning and the next available
applicant as soon as the previous interview was completed.
In this way, 5 or 6 applicants were interviewed per
day. The interviewer attempted to select two persons
at the same time. The first person was given a question
naire to fill out with instructions to return it to
the interviewer when finished and the second person
was interviewed. In order to avoid overselection of
females in the sample due to the relatively small number
of female drinking drivers, a modified quota sampling
technique was used. Only males were selected four days
of the week with both males and females selected one day
31
giving a total of no more than 10-20 women per week.
A questionnaire and an interview were utilized to obtain
demographic data. The questionnaire was pretested on
a small court sample and a DMV sample and changes and
additions made for the final instrument.
Data from public records were also collected on
each individual in both the DMV and the court samples.
The public records reviewed were Criminal investigation
and Identification (CII) and DMV offense records. Other
possible sources were checked during the pilot study,
but the relative paucity of data precluded their use.
These sources included credit systems, alcoholic rehab
ilitation clinics, private welfare organizations, the
County Hospital, and the Welfare Board.
Representativeness of the Sample
The final sample was not wholly representative
of the Department of Motor Vehicles license applicant
population in Los Angeles County for the following
reasons: (1) a late start was made in several DMV
area offices thereby cutting down the proportion of
persons obtained from various cultural and racial
sub-populations highly represented in the different
area offices, and (2) the impracticability of using
the Los Angeles Central Office (which handles a high
|
32
proportion of Negro and Spanish-American applicants)
due to administrative snags. As a result, the
Spanish speaking and Black populations were under
represented. The attempt was made to insure the DMV
applicants were representative of the general or
driving population. However, the age structure was
slightly different for the two populations, and
socioeconomic status was not matched due to the unavail
ability of court data, thus giving a somewhat biased
population.
It was felt that if the DMV and court samples
were first matched on race/ethnic, sex, and socioeconomic
variables then other variables which were deemed signi
ficant in differentiating drunk drivers from non-drunk
driving populations could be analyzed. Otherwise, there
would be a question as to the effect of socioeconomic
status on drinking and driving habits. To insure that
the population was representative, interviewers were
sent to each municipal court in the Los Angeles County
area to look up files of drunk driving for a one-month
period. This was done while results were being analyzed
from the pilot run. Unfortunately, when interviewers
searched court records for the race/ethnic, education,
and income information on all convicted drunk drivers
in L.A. County, only the two variables, race and sex,
were available. The comparison between the populations
on these two variables is indicated below.
TEXT TABLE A
PROPORTIONS OF DRIVER LICENSE APPLICANTS AND
CONVICTED DRUNK DRIVERS FOR COURT SAMPLE
AND FOR LOS ANGELES COUNTY, 1968,
BY RACE—ETHNICITY
Convicted Drunk Drivers
License L.A.
Applicants County
Sample_________Sample_______ Total
Caucasian 85. 0 37.1 53.0
Black 4.0 38.1 27.1
Mexican-
American 7.2 21. 0 17.2
Others 3.8 3.8 2.7
100. 0 100.0 100.0
TEXT TABLE B
PROPORTIONS OF DRIVER LICENSE APPLICANTS AND
CONVICTED DRUNK DRIVERS IN SAMPLE AND
LOS ANGELES COUNTY, 1968, BY SEX
License
Applicants
Sample
Convicted
Sample
Drunk Drivers
L.A.
County
Total
Male 81.5 96.5 94.5
Female 18.5 3.5 5.5
100.0 100.0 100.0
34
Due to the non-representativeness of the data according
to race and sex, the sample cannot be extrapolated to
the population. In order to control for the major bias,
that of socioeconomic class, the data analysis will use
occupational status as a control.
Problem of Using Convicted Drunk Drivers as
Repres entative
In the past, almost all studies dealing with the
drinking drivers have used official court records as
data sources. This procedure leads to a major difficulty
in interpreting research findings with regard to the
total drinking driving population. Those who are arrested
for a particular illegal act may be quite different
from those who commit the same act but are never arrested.
It is well known that many people commit prohibited
acts for which they are never arrested or convicted.
The differences may lie in variables unrelated to the
actual offense; i.e., driving newer but less flashy
cars or manifesting a more responsible attitude when
stopped by police or driving in areas with fewer police,
and so forth. The point is that differential rates
of arrest between two populations may be due to factors
unrelated to the actual frequency or seriousness of
the offense. As a result, the findings of the present
35
research cannot be extrapolated beyond the convicted
drunk driving population.
Design
The purpose of the current research is to substan
tiate the association between the variables selected
as indicative of separation from others and drunk driving.
Several methods will be used to assess the existence
of such a relationship. Each variable will be tabulated
against seriousness of the drunk driving record. Secondly,
alcoholism will be used as a control variable so that
the existence of separation, apart from alcoholism, can
be assessed. Alcoholism is determined by a self-report
item related to drinking in the morning which was highly
correlated with other self-report indices of alcoholism
and with number of alcohol-related offenses. Third, each
of the offender populations and the driver license popu
lation will be compared to one another on the basis of
having or not having experienced one or more forms of
separation from others. Fourth, occupational status of
the respondent will be used to indicate the effect of
socioeconomic status on the degree of drinking, offender-
non-offender category, and separation.
-1
36
The indices of separation (those variables
indicating a loss, rejection, separation from others)
used in this research are as follows:
(1) maritial status: separated and divorced
(2) persons living with: living alone
(3) length of time living alone: short length of
time living alone (two years or less)
(4) employment status - number of jobs:
presently unemployed - 3+ jobs in last five-
year period
(5) employment status - number of jobs:
presently employed - 3+ jobs in last five-
year period
(6) persons lived with until age 16: loss of
one or both parents in childhood
Summary
The difficulty of collecting a representative sample
of the court populations and the problems of population
bias due to factors possibly centered around arrest
patterns means that the sample cannot be extrapolated
to the population of convicted drunk drivers nor to a
population of non-canvicted drinking drivers. However,
by controlling for race, age, sex, and socioeconomic
status, it is possible to give an indication of the
variables operating in the drunk driving population in
an attempt to find a major index associated with drunk
driving.
CHAPTER III
DATA ANALYSIS
Introduction
The review of the drinking driving and accident
literature has shown a recent emphasis on the concept
of stress and the idea of separation from previously
assumed roles (high divorce rates, job changes, and
living alone). The present data can be used to generate
further hypotheses related to the causes of drinking
driving and to substantiate past findings.
In order to substantiate the presence of an associ
ation between separation from others and drunk driving,
all variables even vaguely related to what might be
hypothesized as a separation between the individual
and others important in his life were used. However,
in utilizing these variables certain limitations and
problems in interpretation are apparent:
(1) Indices of separation in the individual’s life
give no indication of time of occurrence.
Thus, a divorce eight years before the drunk
driving incident might have little to do
with the present incident.
(2) Time priority questions are difficult to
assess in the case of multiple offenses since
the drunk driving offense could either be the
cause or the result of a separation from
others, be unrelated, or drunk driving and
separation could come from a common cause.
37
38
(3) The use of demographic variables as indices of
separation do not indicate whether or not there
was stress present which led to the response of
drunk driving. Stress is a construct only in this
research and is assumed to arise when there is a
separation from others. Stress is considered a
preadaptation state and will lead to adaptive or
maladaptive responses in an attempt to reduce the
psychological or physiological need state of the
individual to a manageable level. Stress is
assumed to be greatest immediately after a separa
tion from important others when the individual
feels rejected or left alone without the "neces
sary” support or need fulfillment.
(4) Several interpretations other than stress due to
separation can readily be given to the data.
Alcoholism leading to both separation and drunk
driving, a general maladaptive response pattern,
and a subcultural response leading to traffic
law violation will be evaluated as alternatives
to the stress hypothesis due to separation.
One of the problems with a straight forward cause and
effect interpretation between indices of separation and
drunk driving (i.e., separation causes stress which leads
to drunk driving) is the fact that time priority cannot
be established for the two-plus drunk drivers. Therefore,
for the two-plus offenders it is possible that separation
took place after, or as a result of, prior drunk driving
offenses.
In the case of the first offender population, there
is clear priority between the indices of separation and
the offense of drunk driving, since all information on
previous separations was collected at the time of the
offense. For this reason, the analysis will concentrate
its argument on the first offender group.
i
The control variables used in eliminating alterna
tive explanations for the data were alcoholism and occu
pational status. Offense category data were divided j
j
according to the dichotomy alcoholic and non-alcoholic j
i
as determined from self-report data on frequency of morning
I
drinking. Controlling for occupational status was accom- j
plished through the division of offense category data |
J
into four subdivisions - managerial/professional, clericalj
skilled, and unskilled. Further, due to the difference
between the samples with regard to age,, sex, and race-
i
i
ethnicity, the data were controlled on these three j
variables. A test of significance (difference of propor- |
tion test [Z score] was used. The magnitude of associ- |
F
ation between the variables has been computed using j
gamma (see Goodman and Kruskal, 1954), which in the 2x2 |
table is equal to Yule's Q. ;
40
! Data Analysis
r
i The major hypothesis of an association between
z
: separation variables and drunk driving is generally
i supported by the data (significant at p ^ .05 using a
i one tailed test) (Tables 1,2). This relationship
remains when age, sex, and race-ethnicity are controlled
(Tables 3,4,5). The only categories where the associa-
■ tion is greater than the .05 level, using a difference
! of proportion test, are "living alone less than one
: year" (see Table 7 for full table) and "not presently
; employed" (see Table 8). The lack of a significant
: association in the category "living alone less than one
i year" may be due to chance or to a younger age group in
the driver license applicant population who have never
been married. This offsets the larger population of
recent separation and divorce cases in the drunk driving
i population. Table 6 suggests this interpretation since
; approximately 50 percent of those living alone have never
i been married in the driver license sample as against
14 percent for the convicted drunk driver sample.
The higher proportion in the driver license appli
cant population and the nonsignificance of the associa-
f
tion between the offense categories and the separation
j
; category "not presently employed" may be accounted for by
' the younger driver license applicant population with more
individuals below labor force age. However, when age i
i
is controlled (Table 3) the trend toward higher present j
!
I
employment for the offender population remains. When I
|
a measure of association is computed, time living |
alone shows a low magnitude of association and employ- j
i
ment has a high negative magnitude association (-.695).
There is one variable which is clearly an index
of separation from others prior to any drunk driving j
!
offense. This index is "persons lived with until age !
J
16." Table 10 clearly indicates that decreasing familial j
|
relationships is positively associated with drunk driving, j
Table 11 (in two parts, based on self-report data ;
i
regarding alcohol problems) indicates that when combining j
all separation variables, a greater proportion of the
convicted drunk driver population than the license appli
cant population have had one or more dissociating experi
ences. Correspondingly, within each alcohol problem j
category, the greater the number of previous arrests, I
i
the greater the proportion with a separation experience j
i
as defined in this research.
42
| i
I ;
\ Possible Interpretations I
I |
Rather than blindly accept the straight line inter- J
i j
pretation between separation experience and drunk driving, i
I [
j other possible interpretations will be discussed. Since :
! drunk driving is related to the ingestion of alcohol, the !
! possibility of alcoholism as an explanation of both
■ drunk driving and separation from others must be assessed.!
Further examination of Table 11 indicates that ;
1 i
; alcoholism may have little to do with dissociation \
j
since — for each of the two kinds of offenders — those i
i
with admitted two-plus problems with alcohol are similar I
: in proportion of separation experiences to those \*ith no |
i
; problem with alcohol, while the highest proportion of j
i the population are those with one alcohol-related problem, j
; i
: The alternative explanation for the observed relation-j
ship that drinking causes both arrests for drunk driving ]
' f
i i
and also leads to separation of the individual from j
1 others is not tenable from a review of Table 12 due to :
: the minimal shifts in proportions when alcoholism is j
! i
controlled except in two categories. !
I If alcoholism is a causal factor, controlling for al- j
i 1
I i
! coholism should remove the observed relationship between
43
separation variables and drunk driving. However, the
previous relationship and substantive significance is
sustained for most of the variables. Where there is not
a significant difference between the driver’s license
applicants and the first offenders, there is one between
the first and two-plus offenders. Alcoholism (self-
reported items related to drinking) does not appear to
be a prior causal factor in both drinking driving arrests
and in separation. The relationship between drinking
driving and variables indicating separation does not
partial out on alcoholism except for the variables
"living alone," and "employed - number of jobs in five
years." In the case of "employed - number of jobs in
five years," the lack of a substantial difference is
in the alcoholic population where a difference would be
expected if alcoholism was causal.
Yet another explanation is a generally maladaptive
life style. Thus, the individual may generally respond
to life in a manner which leads to separation from
others with drunk driving merely a manifestation of this
maladaptive response. Assessment of the number of job
changes in a five-year period (Table 13 & 14) and the
number of marriages (Table 15) of the drunk driving popu
lation contrasted with the license applicant population
indicates the feasibility of this interpretation.
44
For those not working (Table 13), 44 percent of
the drunk driver populations compared to 23 percent of
the license applicants have had 3 or more jobs in a
five-year period. Similarly/ for those employed
(Table 14) the proportion of drunk driver population who
have had five or more jobs in five years is about twice
the proportion of license applicants. Further, Table 15
indicates a significantly greater number of marriages for
the drunk driver populations over the license applicant
population. If the maladaptive model fits the present
data, the drunk driving populations would have substantial
minor and major crimes. Tables 16 & 17 substantiate this
hypothesis.
Yet another explanation involves a subcultural or
lower socioeconomic class response which deviates from
the general cultural norms. Thus, the drunk driving
population generally has a higher number of traffic
violations and other crimes, has a higher divorce rate,
greater number of job changes, and is more highly unem
ployed. However, by controlling for socioeconomic status
(occupation), it can be seen that a subcultural explana
tion is not wholly tenable (Table 18).
Table 18, offense categories by occupational status,
indicates that when controlled for socioeconomic status,
the magnitude of association between drunk driving and
45
the separation variables is generally upheld. The
variable "living alone less than one year" does not
sustain the relationship between the offense categories
in the managerial/professional status, but does sustain
the association in the other three occupational categor
ies. The reason for this drop is not readily apparent
although indicating that those living alone in the drunk
driving category have generally been living alone for
longer lengths of time than has the general population.
Secondly, the drunk driving population tends to be
presently employed to a greater extent than the control
population although the difference is nonsignificant.
Present employment status does not appear to have a
significant effect on drunk driving. Overall, occupa
tional status does not explain the relationship observed
between the separation variables and the offense categor
ies although evidencing an effect on the degree of drunk
driving within categories. Five of the six variables
retain the magnitude of association seen earlier when
occupational status is controlled.
In summary, several alternative explanations have
been suggested for the basic associations found between
drunk driving offenses and variables initially selected
as representing separation. These alternative explan
ations are:
46
! (1) generalized stress of separation leading
j to response of drunk driving,
(2) alcoholism leading to both separation from
others and drunk driving,
{3) an individual drives as he lives - maladaptive
response to life situations in general,
! (4) a subcultural response pattern leading both
j to arrests for drunk driving, nontraffic-
| related crimes, and constant job and marriage
changes.
| Interpretation of Results
i
Five of the seven variables used as indices of
separation from others significantly differentiate the
drunk driving population from the driver license
; applicant population. Controlling for alcoholism (self-
| reported on several questions related to alcohol) does
i not affect this association substantively although the
!
| effect of alcohol problems on drunk driving increases as
| the number of separation variables decreases. This
suggests that alcohol and separation may both combine to
cause drunk driving.
The finding that the drunk driving population has
imore job and marriage shifts than the license applicant
ipopulation suggests that the drunk driver population
i experiences extensive separation from others. That it
! has extensive criminal records relative to the driver
' license population indicates a non-normative life
pattern as well. Utilizing only first offenders, it
47 j
i
i has been possible to demonstrate that this separation
i
variables preceded the drunk driving charge but not
necessarily drunk driving. While it has not been pos-
|
I sible to reject the maladaptive response model as an
i explanation which overrides stress as a necessary
1 factor/ we have not deleted the stress construct. Stress
1 may be an ever.present factor in the maladaptive |
responses. We would, therefore, be concerned with j
5 i
i long-term stress leading to an inability to handle even !
; «
! i
| minor stresses successfully (success being defined as j
[ i
: that response which does not lead to further separation). i
! 1
I Thus, the minor separation experience as well as the majors
: !
j
: one which is visible demographically (i.e., divorce,
change of job, etc.) would create the maladaptive response.j
The subcultural or status hypothesis is inadequately j
■ supported by the data and indicates that across all status j
i I
‘ categories, there is an association between indices of j
■ separation and drunk driving. !
i i
j It is the purpose of this paper to unify these j
findings into a consistent theory which can then be \
tested using more specific data. While the preceding j
; analysis cannot be used conclusively to prove a direct
| causal connection between separation, stress, and non- I
i
; normative responses such as drunk driving, it has cast
I
i
j
I
48
doubt upon several other explanations. This leaves the
question of the connection between drunk driving, sepa-
ration from others, and maladaptive responses open to
further and more specific research.
The following chapter will show the development of
a theory of separation, stress, and maladaptive response,
indicating how various types of maladaptive response lead
to non-normative actions.
TABLE 1
PERCENTAGE OF POPULATION WITHIN DRIVER LICENSE APPLICANT AND
OFFENDER CATEGORIES FOR SEVEN INDICES OF SEPARATION
LOS ANGELES COUNTY, 1968*
Separation No Offense Convicted Drunk Drivers
Indices License Applicants First Offenders Two-Plus Offenders
Percent % Base Percent % Base Percent % Base
Separated/
Divorced 9.3 (1276) 33.1 (345) 39.2 (298)
Living Alone 8.3 (1305) 21.4 (347) 27.6 (298)
Living Alone less
than one Year 28.7 (163) 33. 8 (83) 34.7 (93)
Presently Unemployed 23.5 (998) 17.2 (297) 20.0 (240)
Employed - 3+ Jobs
in Five Years 20.4 (996) 31.8 (296) 30.0 . (239)
Unemployed - 3+ Jobs
in Five Years 22.7 (220) 44. 8 (47) 42.3 (52)
Lived with Other than
Parents til Age 16 19.9 (1310) 37.8 (357) 37.4 (302)
* Total category Ns in each instance are different due to non-response items.
TABLE 2
Z SCORES BETWEEN DRIVER LICENSE APPLICANTS AND OFFENDER
CATEGORIES FOR ALL SEPARATION INDICES
Dissociation
Variables
Z Score Between
License Applicants/
First Offenders
Z Scores Between
First Offenders/
Two-Plus Offenders
Z Score Significance Z Scores Significance
Separated/Divorced 9.96 “= ■ . 001 1.49 NS
Living Alone 6.97 ^ .001 1.84 .05
Living Alone less
than one year -.76 NS -.12 NS
Presently Unemployed
i
to
to
CO
.05 . 85 NS
Employed - 3+ jobs
in five years 3.82 .001 -.41 NS
Not Working - 3+
jobs in five years 3.07 .001 -.17 NS
Lived with Other than
both parents til
age 16 5.34 .001' 1.23 NS
ui
o
TABLE 3
PERCENTAGE OF POPULATION WITHIN EACH OFFENDER CATEGORY
FOR SIX INDICES OF SEPARATION BY AGE
16; - 21 22-34 35 - 64 65 +
Separation
Variables DLA
1st
Off.
2+
Off. DLA
1st
Off.
2+
Off. DLA
1st
Off. '
2+
Off. DLA
1st
Off.
2+
Off.
Separated/
Divorced/
Widowed 1.0 8.3 0 5.8 21.1 23.6 8.4 34.7 37.4 24.1 20.0 50.0
Alone 4.7 16.7 0 10.1 20.0 21.5 32.3 23. 0 30.8 21.3 20.0 20.0
Alone less
than 1 Year 57.1 33.3 0 41.9 41.4 70.0 21.0 29.8 35.5 7.7 0 0
Unemployed 36.4 28.6 50. 0 20.7 16.5 9.0 18.5 17.7 22.1 66.1 0 60.0
Employed 3+
Jobs 50.9 62.5 0 29.9 44.1 35.0 11.9 21.1 27.6 5.9 0 0
Lived with
Others til 16 18.7 28. 6 50.0 18.9 33.8 38.0 19.5 32.3 36.3 26.1 40.0 50.0
TABLE 4
PERCENTAGE OF
FOR
POPULATION
SIX INDICES
WITHIN EACH OFFENDER
OF SEPARATION BY SEX
CATEGORY
MALE FEMALE
Separation
Variables
License
Applicant
1st.
Off.
2+
Off.
License
Applicant
1st.
Off.
2+
Off.
Separated/
Divorced/
Widowed 7.2 27.7 35.3 9.6 38.9 100.0
Living Alone 9.3 21.8 27. 8 7.4 15.8 50.0
Alone One Year 31.7 33.8 34.1 11.1 33.3 100.0
Unemployed 16.6 15.4 19.5 53.1 50.0 0
Employed 3+
Jobs 8.3 31. 8 30.8 13.5 25.0 0
Lived with
til 16 19.7 32.2 37.3 18.8 40.0 0
Ul
NJ
TABLE 5
PERCENTAGE OF POPULATION WITHIN EACH OFFENDER
FOR SIX INDICES OF SEPARATION BY RACE
CATEGORY
CAUCASIAN BLACK MEXICAN-
AMERICAN
Separation
Variables DLA
1st
Off.
24-
Off. DLA
1st
Off.
2+
Off. DLA
1st
Off.
2+
Off.
Separated/
Divorced/
Widowed 7.6 35.3 45.3 16.7 25.4 34.0 4.9 20.3 25.0
Alone 9.4 29.0 33.3 12.0 22. 7 28.4 3.5 10.4 18.2
Alone less
than 1 year 47.2 47.6 53.3 33.3 37.0 28.6 50.0 77.8 78.6
Unemployed 22.6 17.6 24.4 21.6 13.9 20.8 30.1 21.3 14.3
Employed 3+
Jobs 27.1 33.3 32.6 25.0 23.3 27.7 30.5 43.4 29.3
Lived with
Other til 16 18. 9 27.5 27.6 33.3 39.6 43.7 25.0 31.2 32.1
Ln
u >
TABLE 6
MARITAL STATUS, BY LENGTH OF TIME LIVING ALONE
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Length of Time Living Alone
Less Than One Year More Than One Year
Convicted & Convicted &
Marital License Reckless Drunk License Reckless Drunk
Status_________Applicant__________Drivers__________ Applicant__________Drivers____
Percent Number Percent Number Percent Number Percent Number
Married,
but alone/
Divorced/
Separated 53.6 (22) 86.1 (56) 45.2 (38) 75.0 (96)
Never
Married 46.4
(19)
13.9 (9) 54.8 (46) 25.0 (32)
Total 100.0 (41) 100.0 (65) 100.0 (84) 100.0 (128)
TABLE 7
SEPARATION INDEX - TIME LIVING ALONE
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES, 1968
Time Living
Alone
No Offense Convicted Drunk Drivers
First Offenders Two-Plus Offenders
Percent Number Percent Number Percent Number
Less than
6 months 18.9 (23) 20.0 (16) 20.6 (19)
6-11 months 9.8 (12) 13.8
(U)
14.1 (13)
1-2 years 17.2 (21) 16.2 (13) 18.5 (17)
3+ years 54.1 (66) 50.0 (40) 46 .7 (.43)
Total 100.0 (122) 100.0 (80) 100.0 (92)
cn
(J!
TABLE 8
SEPARATION INDEX - PRESENTLY EMPLOYED
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Presently
Employed
No Offense
License Applicants
Convicted
First Offenders
Drunk Drivers
Two-Plus Offenders
Percent Number Percent Number Percent Number
Yes 76.4 (1001)
r ' -
■
(N
G O
(292) 80.0 (240)
No 23.5 (308) 17.2
(16)
20.0 . (60)
Total 100.0 (1309) 100.0 (308) 100.0 (300)
( ji
c r >
TABLE 9
MEASURES OF ASSOCIATION BETWEEN INDICES
OF SEPARATION AND OFFENDER CATEGORIES
Dissociation Gamma between Driver License-
Variables______________Applicants and First Offenders
Separated/
Divorced/
Widowed .657
Living Alone .5 09
Living Alone
less than one year .118
Presently Employed -.695
Employed - 3+ Jobs .291
Unemployed - 3+ Jobs .469
Lived with Others
til 16 .420
TABLE 10
SEPARATION INDEX - PERSON LIVED WITH UNTIL AGE 16
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY/ 1968
Person lived with
until age 16
No Offense
License Applicants
Convicted
First Offenders
Drunk Drivers
Two-Plus Offenders
Percent Number Percent Number Percent Number
Both parents 80.5 (1054) 67.2 (240) 62.6 (189)
One parent 14.6 (192) 21.9 (78) 28.1 (85)
Relatives 3.4 (44) 8.1 (29) 7.6 (23)
Other 1.5 (19) 7.8 (10) 1.7 (5)
Total 100.0 (1309) 100.0 (357) 100.0 (302)
Cn
00
TABLE 11
NUMBER OF KINDS OF SEPARATION EXPERIENCED BY DRIVER LICENSE
APPLICANTS AND OFFENDER CATEGORIES WHO REPORTED
PROBLEMS WITH ALCOHOL
LOS ANGELES COUNTY, 1960
Number of
Separation
Variables
Driver
License
Applicants
Convicted Reckless and Drunk Drivers
who Report no Problem with Alcohol
Reckless
& First
Drivers
Offenders Two-Plus Offenders
Percent Number Percent Number Percent Number
None 48.9 (429) 37.2 (89) 27.9 (26)
One or
More 51.1 (448) 62.8 (150) 72.1 (67)
Total 100.0 (877) 100.0 (239) 100 .0 (93)
cn
vo
TABLE 11
(continued)
NUMBER OF KINDS OF SEPARATION EXPERIENCED BY DRIVER LICENSE
APPLICANTS AND OFFENDER CATEGORIES WHO REPORTED
PROBLEMS WITH ALCOHOL
LOS ANGELES COUNTY, 1960
Number of
Convicted Reckless and
Drivers who Reported One
with Alcohol
Drunk
Problem
Convicted Reckless and Drunk
Drivers who Reported Two-Plus
Problems with Alcohol
Separation
Variables
Reckless
& First Two--Plus
Reckless
& First Two-Plus
Percent Number Percent Number Percent Number Percent Number
None 26.0 (19) 16.7 (11)
38.5 (47) 30.3 (40)
One or
More 74.0 (54) 83.3 (55) 61.5 (75) 69.7 (92)
Total 100.0 (73) 100.0 (66) 100.0 (122) 100.0 (132)
C T l
o
TABLE 12
PERCENTAGE OF POPULATION WITHIN EACH OFFENDER CATEGORY
FOR SIX INDICES OF SEPARATION BY ALCOHOLISM
License Applicants 1st Offender 24- Offender
Separation
Variables Total
Alco
holic
Non-
Alco. Total
Alco
holic
Non-
Alco. Total
Alco
holic
Non-
Alco.
Separated/
Divorced/
Widowed 9.3 28. 6 7.4 33.1 37.5 27.0 39.2 45.8 32.9
Living Alone 8.3 14.3 8.9 21.4 37.5 19.6 27.6 31.7 27.2
Alone less
than 1 year 28.7 100.0 28.6 33.8 33.3 31.2 34.7 36.8 34.8
Unemployed 23.5 16. 0 23.3 17.2 19.2 16.6 20.0 29.8 16.8
Employed 34-
Jobs 20.4 20.0 20.0 31.8 45.0 30.9 30.0 31.4 28.8
Lived with
Other til 16 19.9 16.7 19.7 37.8 38.5 32.0 37.4 30.0 38.9
< 3 \
TABLE 13
SEPARATION INDEX - NOT WORKING - NUMBER OF JOBS IN FIVE YEARS
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Not Working:
Number of jobs
No Offense
License Applicants
Convicted
First Offenders
Drunk Drivers
Two-Plus Offenders
Percent Number Percent Number Percent Number
Never 4.5 (10) 2.0
(1)
0 (0)
1 - 2 72.8
(161)
54.0 (27) 57.7 (30)
3 - 4 14.5 (32) 26.0 (13) 30.8 (16)
5+ 8.2 (18) 18.0 (9) 11.5 (6)
Total 100.0 (221) 100.0 (50) 100.0 (52)
O'!
to
TABLE 14
SEPARATION INDEX - EMPLOYED NUMBER OF JOBS IN FIVE YEARS
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Employed: Number
of Jobs in five
years
No Offense
License Applicants
Convicted
First Offenders
Drunk Drivers
Two-Plus Offenders
>
Percent Number Percent Number Percent Number
Never 79.6 (692) 68.2 (176) 70.0 (140)
1 - 2 15.4 (134) 20.5 (53) 21.0 (42)
5 - 6 2.8 (24) 7.0 (18) 5.5
(11)
7+ 2.2
(19)
4.3
(11)
3.5 (7)
Total 100.0 (869) 100.0 (258) 100.0 (200)
O'I
w
TABLE 15
NUMBER OF TIMES MARRIED
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Number Times
Married
No Offense
License Applicants
Convicted
First Offenders
Drunk Drivers
Two-Plus Offenders
Percent Number Percent Number Percent Number
1 79.0 (747) 68.8 (190) 66.8 (173)
2 18.5 (175) 25.4 (70) 25.1 (65)
3+ 2.5 (24) 5.8 (16) 8.1 (21)
Total 100.0 (946) 100.0 (276) 100.0 (259)
o\
TABLE 16
TOTAL MINOR VIOLATIONS (TRAFFIC)
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Total Minor No Offense Convicted Drunk Drivers
Violations License Applicants First Offenders Two-Plus Offenders
Percent Number Percent Number Percent Number
0 48.2 (641) 29.1 (104) 15.5 (47)
1 22.7 (312) 17.6 (63) 15.8 (48)
2+ 29.1 (386) 53.3
(191)
68.7 (208)
Total 100..0 (1329) 100.0 (358) 100.0 (303)
O'!
U1
TABLE 17
TOTAL MAJOR VIOLATIONS
DRIVER LICENSE APPLICANTS AND OFFENDER CATEGORIES
LOS ANGELES COUNTY, 1968
Total Major
Violations
No Offense
License Applicants
Convicted
First Offenders
Drunk Drivers
Two-Plus Offenders
Percent Number Percent Number Percent Number
0 99.0 (1312) 74.6 (267) 58.8 (171)
1 .7 (10) 15.4 (55) 19.3 (58)
2 .0 (0) 5.3
(19)
7.6 (23)
3 .0 (0) 4.7 (17) 14.3 (43)
Total 100.0 (1322) 100.0 (358) 100.0 (301)
ci
c r >
TABLE 18
PERCENTAGE OF POPULATION WITHIN EACH OFFENDER CATEGORY, DRIVER
LICENSE APPLICANTS AND CONVICTED DRIVERS, FOR SEVEN
INDICES OF SEPARATION BY OCCUPATIONAL STATUS
Managerial/
Professional Clerical Skilled Unskilled
Separation
Indices DLA
1st
Off.
24-
Off. DLA
1st
Off.
24-
Off . DLA
1st
Off.
24-
Off . DLA
1st
Off.
24-
Off .
Separated/
Divorced/
Widowed 22. 6 42.5 50.0 21.2 46.2 50.7 28. 8 41.8 23-6 41.2 42.9 53.2
Living Alone 9.9 26.3 40.5 7.9 25.5 32. 4 12.5 24. 6 14.3 6.9 14.7 30.8
Living Alone
less than 1
year 47.3 42.9 43.7 37.5 56.0 59.1 43.7 50.0 53.8 36.4 53.3 57.1
Presently
Employed 19.4 12.7 9.5 17.4 15.2 21.9 30.1 15.5 23.4 21.5 22.9 17.7
Employed, #
Jobs 5 Years 14.6 31.1 25.0 22.9 28. 6 26.2 28. 2 9.3 42.4 33.9 37.1 30.3
Unemployed, #
Jobs 5 Years 21.4 28. 6 33.3 22.2 53.8 28.6 29.0 62.5 50.0 36.0 35.0 40.0
Lived with Other
than Parents
Until Age 16 19.1 26.6 28.6 22.2 26.6 37.0 22.0 35.2 33.8 21.0 39.2 41.7
CHAPTER IV
DEVELOPMENT OF REFERENTIAL
DISSOCIATION THEORY*
Introduction
In order to develop a unified theory on which data
can be tested further, it is necessary to set forth
various assumptions about man's basic nature and his
response patterns. One of the current assumptions
regarding man has been the idea of consistency as a motive
force toward change.
In sociology and social psychology, there has been
a rapid growth of theory and research on status inconsis
tency, balance theory, congruity theory, and action
dependence theory. All of these trends refer to a consis
tency in the interrelations and juxatpositions of elements
which constitute the thinking, feeling, and perceiving
individual — the psychological individual. Further,
stress or tension has become a general matter of concern
insofar as the addition, loss, or modification of one
or more of these elements has consequences for certain
or all of the other elements or for the psychological
individual as a whole.
Much of the work done in these areas has concentrated
on inconsistency between psychological elements referred
68
69
to as expectations and cognitions in the perception
of the individual and posits the need of the individual
to attain consistencyi The fact that such a consistent
state seldom occurs does not seem to deter the use of
this assumption as a theoretical basis for research.
Accordingly, the main purpose of the present writing
is to provide a concept which replaces the assumption
of a consistent or balanced state in man. Change in
the individual from one state to another is not the
result of inconsistency between conflicting cognitions
so much as it is a search for identity. This is obtained
through control over the external environment - an adapt
ation to the environment in order to fulfill the basic
psychological and physiological needs. Thus, if control
through adaptation demands inconsistency, then such incon
sistency will be sustained by the individual. The most
vivid form of such inconsistency is shown in the schizo
phrenic who, in an attempt to deal with conflicts in
the environment, will develop conflicting personality
types which are used to gain identity and control through
interaction with the environment. Other examples of
inconsistency are readily evidenced in a social context
by the submerging of feelings of hate in order to obtain
other needed "goods’ ' for sustaining the physical and
70
psychological states of the individual.
Identity through control develops through interaction
with the environment. Put another way, interaction makes
up the social identity of the individual.
Thus, when interaction is diminished or when there
is a loss of important sources of need fulfilment, the
control exercised by the individual and, hence, his social
identity is temporarily diminished. Attempts will be
made by the individual to regain the withdrawn interac
tants or to protect the self from further loss through
various coping mechanisms such as diminution of need for
specific types of support, devaluation of self, searching
for alternatives, etc.
Thus, coping mechanisms are all those processes
which the organism may use in attempting to cope with
its environment and protect itself from that stress which
is potentially destructive of identity. One coping mecha
nism is the use of alternative roles. Alternative role
sets, as specific prearranged response patterns, are
easier to fulfill in specific circumstances and, thus,
make easier the control of social objects forming part
of the identity. The role of the drunk is far easier to
play than the role of husband, and environmental control
is regained by redefining the contextual situation
71
in one's own terms {i.e., not responsible for actions).
He knows what to expect from others and can react
in a very structured way without incurring immediate
or long-range obligations while fulfilling preset
expectations relating to lack of ability to perform
certain roles while drunk. New roles are adopted
in order to:
(1) permit the fulfilling of a particular self-
concept,
(2) retreat from an unstructured role to a more
structured one (expectations are delimited)
(3) retreat to a less demanding role - one which has
fewer expectations (alcoholic has fewer expecta
tions than does father).
These mechanisms enable the individual to control
his environment, but what happens when this control
is lost.
Since withdrawal or perceived loss of important
sources of need fulfilment leads to a diminution
of social identity, it is important to understand
the causes and effects of such loss. Little work
has been done concerning effects on the individual
when he perceives a withdrawal of social objects once
controlled. Such withdrawal and its perception will
be called referential dissociation. It is hypothe
sized that referential dissociation will lead to
72
partial loss of identity and a high degree of stress
due to loss of control over the environment. This
stress will be manifested in several ways depending
on the adaptive response of the individual. The concepts
referential dissociation, identity, stress, and adaptation
will be explained before in-depth discussion of
the theory is presented.
Referential Dissociation
Referential disssociation is a perceived withdrawal
(from associational ties) of social objects representing
the social identity of an individual. Thus, when
an individual experiences the withdrawal of groups
or persons which represent parts of his social identity,
stress (a pre-adaptation state) occurs; and the individual
reacts by (1) attempting to regain the association,
(2) failing this, he turns to various previously
learned coping mechanisms, (3) if these do not provide
a satisfactory solution to the problem, the individual
may attempt new coping mechanisms, and (4) if these
fail, the individual may seek to end his life if
the associational ties had ultimate identity meaning
for him. An example of referential dissociation
is the perceived withdrawal of peer esteem when
a man is passed over for promotion or when interaction
73
with him is decreased after he has made a blunder
in his job. More permanent dissociation comes through
divorce or the loss of a loved one through death.
Several basic assumptions underlie the concept
of referential dissociation. First, dissociation
must be perceived. Secondly, perception of a withdrawal
may be caused either by an actual withdrawal by
an individual or by a misinterpretation of an innocent
act which results in a reaction causing dissociation.
Third, regardless of the reality of the dissociation,
the sense of loss of identity and resulting diminution
of worth takes place to the degree to which the referent
is important to the subject's self identity. Fourth,
various coping mechanisms are set in operation by
the individual. The specific coping mechanism used
depends on many factors including those mechanisms
found to be successful in the past, the possibility
of using old mechanisms in the present environment,
the latitude of the environment in permitting certain
responses, and the initial success of the mechanism.
Types of Dissociation
Referential dissociation can be subdivided into
four types: The first type is atomistic dissociation
as a condition of a society increasing in specialization
74
or division of labor accompanied by, in Weber's
terms, increasing rationalization. Thus, with the
proliferation and specialization of occupations,
the individual becomes increasingly lost as a significant
and meaningful part of the social network. As personali
ties are molded in increasingly uniform fashion,
adequate competence becomes the major quality demanded
in a job, and the individual becomes less meaningful
as a personality and, consequently, instantly replaceable,
thereby less able to control the broader aspects
of the system as it affects his life. As individuals
become less necessary and more numerous, they become
more mobile. Mobility itself increases the atomistic
nature of man. Sorokin makes this point strongly
when he states that "mobility diminishes intimacy
and increases psychosocial isolation and loneliness
of individuals." {Sorokin, 1927) This is similar
to the lack of restraint by society (Durkheim's
anomic condition) (Durkehim, 1951). The major emphasis
here is not the lack of restraint or the lack of
collective activity, but the perceived withdrawal
of the social referents and the resulting loss of
personal identity through loss of control. Without
reflection from some object, social man can have
no meaning. Similarly, as a man is vertically mobile,
75
he loses identity with previously meaningful or
supportive referents and faces a period of separation
through withdrawal from and by old identity groups
and, alternatively, a testing period by new identity
groups. Thus, individuals with a strong sense of
personal worth generated from strongly supportive
identity groups in the past and those with supportive
groups which do not reject the vertically mobile
individual are able to sustain the transition period
with little or no stress. However, it appears endemic
that as society becomes more fragmented and atomized
through emphasis on specialization, rapid replacement
of personnel and public controlled socialization,
the stability of identity groups is diminished. This
leads to greater isolation and a greater need for
continuous referential support.
The second principal type of dissociation is
punitive or rejective dissociation. Here an individual
is perceived as deviant and is rejected by others
either as a means of punishment, or for self protection,
or through lack of desire for further affiliation
and responsibility, or because of the need of a scapegoat.
A phenomenon which often appears here is the deviant
reaction of the rejected individual which causes
him to be further rejected by other groups, thus
76
creating a rejection syndrome. The concept of a rejec
tion syndrome developing from initial rejection by one
part of the society and the consequent reaction of other
parts of the society needs to be studied. Even if the
deviant does not react in a deviant manner, knowledge
of his rejection by another group tends to cause further
rejection because of the pervasiveness of the normative
culture. Rejection by one group due to deviance probably
means the individual is deviant with regard to the norms
of other sectors of the society, and the individual
may find himself blocked on several fronts, although
the different sectors may never communicate.
The theoretical work on the functions of the deviant
for group cohesion emphasizes the positive aspects of
the deviant for group growth. However, the effect on
the deviant may be totally destructive. The graveyard
of social rejects is found in the suicides, the ranks
of the alcoholics, the skid rows of every strata of
society, the mentally ill, and, perhaps, the problem
drinking driver who "accidently" kills himself.
Conflictive dissociation is the third type of
dissociation. This type of dissociation arises
through the coming together of two or more societal
normative patterns giving conflicting expectations
for action. This is similar to the concept of status
77
inconsistency, except the result of such a convergence
is the inability to fulfill all of the expectations im
posed adequately, thus leading to a partial withdrawal
or rejection of the multi-role incumbent.
Such a convergence of conflicting role demands is
a consequence of an organic society in the Durkheimian
sense and closely akin to the effects of atomistic dis
sociation, except that the dissociation is a result
of inability to fulfill expectations of major reference
groups and individuals rather than a disjuncture between
the major social institutions and the individual. Due
to the demands of the social structure, the individual
feels a discrepancy between (1) his ability to fulfill
conflicting roles, (2) his desire to be accepted by
other role participants, and (3) his control over the
perception and reactions of others.
The types of dissociation above cover the major
causes of rejection and withdrawal. The only type of
dissociation left arises from the perceiver and is due
to a misperception of reality - imaginary dissociation.
Such misperceptions can cause a reaction resulting
in a withdrawal by others (a self-fulfilling
prophecy) and a compounding of the initial perceived
78
dissociation. The paranoid state is an example of
such misperception. The individual symbolically
organizes a threatening pseudocommunity out of the
fragments of the social acts of others. Cameron
has indicated that such a perception leads the individual
into conflict with the real community which in turn
rejects him. Lamert has argued that the paranoid
individual does not react against a pseudo-community
but a real community which acts differentially toward
the paranoid in an often overtly organized and conspira-
tional way (Lamert, 1967).
In summary, the four types of dissociation are:
atomistic, a disjuncture (alientation) between the
institutions of society and the individual due to
a high degree of specialization and low level of
participation by the individual; rejective, in which
the individual is rejected by those groups with which
he is identified; conflictive, in which the individual
is caught between conflicting role expectations and
finds himself dissociated from full participation
and acceptance and, thus, control over his environment,
and imaginary dissociation in which no dissociation
has initially occurred but an incorrect perception
forces the individual into a deviant reaction resulting
79
in a self-fulfilling prophecy. The present writing
will focus on the rejective form of dissociation.
Concept of the Social Self
Mead characterizes the physiological organism
without a self as existing in the world of experience
without placing the objects encountered in relation
to the self - a bare "thereness of the world." (Mead,
1956) But in organizing experiences, the self is
used as the experiential center and meaning is defined
in terms of the self. Further, Mead indicates that
we tend to organize all experiences into the self
and organize our memories "upon the string of our
self," and "the self, as that which can be an object
to itself, is essentially a social structure, and
it arises in social experience." (Mead, 1956)
This conception lays the foundation for the
dissociation hypothesis. The separation of objects
related to the self (those objects which the individual
controls and, therefore, are part of the self, causes
a disruption of the self-identity. There is a necessity
to reorganize one's identity. At the same time, if
that part of the self which is separated makes up
a large part of one's identity, than not only is there
a disruption of the identity but a general feeling
80
of loss of control and a corresponding sense of meaning
lessness. As Cooley has stated, the development
of the self is associated chiefly with the "exercise
of power, of being a cause, ideas that emphasize the
antithesis between the mind and the rest of the world."
{Cooley, 1922) It is through control of objects
and of the self that man draws things to him, or
put another way, holds on to those things that have
become meaningful. On the one hand, it is made up
of drawing objects to the self (thus, control
exercised over them) and on the other, the unified
substance of those things which have been controlled
and held and have become identified with the person.
Socially, the sum total of these things and processes
are the self. When the processes are thwarted, a
breakdown occurs. Thus, mental disease might be
characterized as attempts to cope with and control
the environment in such a way that the control attained
gives meaning and value to the self. Any loss or
withdrawal of those elements which have become identified
with the self in such a way as to indicate lack of
ability to control is a disruption of the self and
leads to stress. Release of one segment of this
self through the voluntary acceptance of an alternative
81
set of ideas creates change in the social self, but such
a change can be experienced without corresponding stress
due to the continance of control. The individual rejects
the old in order to accept the new, but should part of
an identity taken away from the individual non-voluntarily
then stress is immediately experienced. An example
of voluntary change is when an individual emotionally
outgrows a certain type of role model and selects a new
one. An example of non-voluntary change may be the
loss of an ideal image as when a revered model suddenly
displays the fact that he has clay feet. This is a dis
sociation, a withdrawal, a loss of control over a part
of the social self.
Control and Identity
Interaction with others has defined the boundaries
of acceptance-rejection of the self by others. We learn
just how far we can go without losing control of the
situation.
Interaction is a process of continually testing
these boundaries as well as fulfilling more basic
needs. But due to fear of rejection, violation of
the boundaries is never too great. Since many situa
tions tend to be novel and the boundaries are not
always preset, situations tend to be pre-structured
:
82
to facilitate operations within an unknown environment
without immediately facing the ambiguity of the
acceptance-rejection problem. The other person is
forced to act within a prearranged structure thereby
putting him off guard and in an ambivalent position.
In this context, then, loss of control means a loss
as to the exact parameters within which acceptance-
rejection operates and, therefore, loss of a sense
of identity {one feels belittled by the situation,
not at ease and unsure of himself).
The loss of control creates stress and causes
ambiguity as to identity. Depending on the stability
of the individual’s past reenforcement, such a
situation can cause a destruction of the previous
identity and a devaluation of the self or cause an
attempt to regain control by reemphasizing his previous
identity. In the latter case, greater productivity .
usually results. Alternatively, a mental breakdown
due to stress attests to the previous ambiguity of
the individual's identity. Those individuals who
are uncertain of their identity tend to be extremely
cue conscious and aware of their environment and
of the acceptance-rejection boundaries.
The element of control enters through the principle
of reduction of ambiguity {an artifact of conservation
83
of energy in adaptive responses). Since it is not
possible for an organism to construct totally novel
responses to each new situation (conservation of
energy), control of the environment is necessary to
some extent. Thus, it is known that individuals
tend to react consistently in their autonomic patterns
of reaction to different types of situations, and
from one time to the next (Lazarus, p. 172). The
identity of the individual, to have some constancy
over time, must involve some parameters related to
environmental situations and must be rewarded when
elicited in response to similar previous circumstances.
If rewards are not forthcoming then the act will
be extinguished. (see Skinner, 1953; Homans, 1961)
The individual must then be able to exercise some
control over the environment since total ambiguity
leads ultimately to- the extermination of all behavior.
When an individual loses control over situations
which have great identity value, he can no longer
fulfill his identity by re-enacting previous response
or role patterns. This happens when a role partner
withdraws and refuses to be restricted by the
previously drawn parameters (i.e., the situation
is no longer controlled unless the individual can
change roles. This may involve tremendous sacrifice
84
of particular self-conceptions). Thus, various so-
called coping mechanisms or mechanisms of "giving
up" - psychosomatic disease, insanity, or mental
breakdown, and suicide are attempts to redefine the
situation and re-establish control - the adoption
of a new role in order to re-establish control. By
becoming sick, one adopts a new role - one that is
structured and defined such that the individual
regains control of his environment, even though it
may place the person in a dependent position.
The alcoholic does this continually in order
to redefine the situation. The role of drunk
is an easy one to fulfill and becomes a highly
dependent condition. But, as in other instances
of dependent role adoption, the major reason is control
over the environment even if the control is one far
less than the individual usually exercises.
Thus, loss of control when such loss means rede
finition of a crucial role is tantamount to loss of
identity and, therefore, loss of previous identity
meaning unless alternatives are readily available.
When control is no longer perceived as possible,
when the loss of identity has become complete, than
there is a true giving up. This giving up culminates
in a total acceptance of fate, a loss of self identity,
and a total submission to the environment.
85
Concept of Stress
The concept of stress is central to the study
of man in society. Man's development, his interaction
with the social and physical environment, and social
change at all levels of aggregation depend on change
within the individual participants. Such change
is brought about through reaction to the external
environment and to the internal environment as it
is shaped by the external. Such reaction can be
labelled stress. Stress has been defined as "the
state manifested by a specific syndrome which consists
of all the nonspecific induced changes within a biologic
system" (Selye, 195 0) or as defined by Basowitz,
et al., "we should not consider stress as imposed
upon the organism, but as its response to internal
or external processes which reach those threshold
levels that strain its physiological and psychological
integrative capacities close to or beyond their limits."
(Basowitz, 1955) This definition defines stress
as a pathological condition and thereby narrows the
broader conceptualization used by Selye. It is
this narrower definition which will be used in the
present theory since we are dealing with that syndrome
which takes the organism beyond its integrative capacities
86
to the point where control of the environment is
problematic.
The above definition can be readily adapted to
a sociological framework by the assumption that the
specific changes we are talking about are a combination
of immediate responses to social stimuli or responses
to physical objects which are integrated in a social
context. Such responses have a resultant social meaning.
The purely innate responses to physical stimuli; i.e.,
fear of extinction of body tissue, basil regulation
of body temperature and blood pressure, and reaction
to degrees of heat which cause pain are reactions which
are sometimes given social meaning (i.e., withstanding
severe pain denotes manliness), but stress as a nervous
system response will not be used in this discussion.
The above definitions indicate states of stress
but do not relate them to resulting behavior. Stress
is a preadaptation state - a state which forces the
organism to respond. The response important here is
a social response - an attempt to change a physiological
or psychological state through adaptation.
In 1964, Schmale outlined several general constructs
which had come out of stress research. These constructs
are as follows: (1) "stress on a psychological level
87
involves a perception which threatens the individual
with a loss of some significant aspect of his relation
ship with his environment, (2) the threat requires a
change in functioning in order to overcome or adapt
to the actual or anticipated change in environment,
(3) what is perceived as psychological stress and how
such stress is reacted to is a highly personal matter,
(4) the personal factors involve genetic equipment,
early experience during important psychologically
critical maturational periods and the subsequent adap
tation to the earlier derived intrapsychic representation
of personal and cultural demands." When this is placed
into a social context; i.e., "the loss of a significant
aspect of his relationship with his environment," then
the above definition of stress and its causes and results
for the individual come close to the definition of refer
ential dissociation and its concomitants as set forth
in this research.
Stress arises through dissociation and perceived
loss of control. Adjustment can be made to partial
loss of control if the role is not rigidly prescribed,
if the individual does not depend heavily on his ability
to totally control the situation, or where the loss
of control appears temporary and, thus, does not require
88
a re-arrangement of the social self- However, if total
control is jeopardized then that part of the self-identity
tied to the role is lost and to the extent that this
forms the total identity, more or less stress is experi
enced. Studies done of the Chinese indoctrination
programs during the Korean War (Schein, 1956) are illus
trative of what happens when the environment becomes
ambiguous and self identity is questioned through the
use of various techniques of persuasion, threats of
punishment, dissociation, manipulation of information.
Since control is not possible except on the terms of
the captors {not real self control), the prisoner is
faced with total loss of self or acceptance of pseudo
control over his own identity through the mechanism
of cooperation. The Asch Study on conformity represents
the attempt to control one's environment and maintain
identity. Those whose identity was weakest in the situ
ation being the ones to most readily conform.
Thus, for the purpose of this writing, stress can
be summarized as the organismic preadaptation state
arising in response to environmental change. Put another
way, it is a psychophysiological state of an individual
created by any environmental change which necessitates
an increased adaptive response. The response elicited
89
may be appropriate or inappropriate depending on the
individual's coping abilities.
Summary
This Chapter has presented the referential dis
sociation concept and has related it to stress, a pre
adaptation state, due to the perception of withdrawal
of major referents. Since social identity is made up
of the controlled environment, the perceived withdrawal
of major referents leads to a loss of identity and an
attempt to regain or form a new social identity. Specific
responses to the stress of dissociation will be delineated
in the next Chapter.
90
FOOTNOTE
*The Dictionary of Psychological and Psychoanalytical
terms defines dissociation as 1 1 ...a process whereby (or
condition in which) a group of psychological activities
possessing a certain unity among themselves lose most
of their relationships with the rest of personality and
function more or less independently." "...in dissoc
iation the segments subsequently maintain or even
increase their internal organization." (English and
English, 1961)
When this concept is placed in a social context
(i.e., a referential context), it can be said that
there is a general process of withdrawal of specific
relationships with social others and a necessity of
functioning independently. The process of withdrawal
of specific social objects is labelled referential
dissociation.
CHAPTER V
RESPONSE TO THE STRESS OF REFERENTIAL DISSOCIATION
Introduction
Response patterns are determined by an attempt to
retain unity of identity-control, presumably on a new
level with each change that takes place in the environ
ment. What happens when unity cannot be achieved? What
happens when there is a lack of adaptation? As with
all integrated systems which are self-maintaining, there
are various mechanisms and a finite number of permutations
in which the mechanisms operate together to insure unity
and control. Thus, when control once attained is upset,
varying attempts are made by the individual to retain
previously specified parameters. Such operations are
not functionally attached to any one goal state except
insofar as the innate capacities, capabilities, and needs
of the individual to operate and maintain identity-control:
sets parameters for the operation of the individual within
his environment.
These parameters are fairly rigid. For instance,
the type of response is limited by the level of knowledge,
the learned responses of the individual, and the para
meters set by his basic personality traits. If the basis
91
92
for learning a complex set of operations had not been
previously set, the problem of understanding the more
complex level is thwarted. Similarly, an individual who
has learned a particular set of responses cannot be
expected to respond in a totally different way from these
response sets. This is particularly visible in the
"mentally deranged" where "adaptive" learned responses
are not appropriate to the given stimulus as defined
by the social context.
What happens, then, when an action is precipi
tated which cannot be readily incorporated within the
learned response patterns of the individual? What com
pensations are made, where does the breakdown occur,
and how is the breakdown repaired?
Generalized Response Categories
The mechanisms used to establish or re-establish
control depend on the type of loss which is experienced,
the demands of the environment, and the general reaction
mode of the individual. When the loss of a spouse occurs,
consolation in the arms of another woman while making
up for a certain sense of loss does not aid in the
control over one's environment and cannot be indulged
in with impunity. However, accepting the condolences
93
of friends and relatives and going through a period
of celebacy retains the requisite control over the envir
onment through the assumption of an interim identity -
that of mourner and widower.
On the other hand, should a loss occur through
social mechanisms, such that certain types of recog
nized control are difficult to obtain {inability to
attain a high status position), then supplementary
controls are set up. Such supplementary controls can
be the establishment of an identity (the hippy and beat)
which is set against the normative forms of identity
and control in the society. This is negative control
in that it is utilized to obtain recognition for certain
qualities meaningful to the larger group while reviling
the outward manifestations of status as recognized by
that group. Thus, certain virtues (peace, freedom,
love) become mechanisms of control while outward appear
ance becomes a symbol of differentiation and separateness,
thereby establishing a new social identity.
The type of change is determined by the reaction
mode of which there are three major types:
1) adaptive-versatile mode
2) fixed mode
3) changeable mode
i In the adaptive-versatile mode, the major reaction
| to loss will be a realignment of the environment so
; that the loss is covered - a new identity is assumed
which takes the place of the lost identity (i.e., the
! student who is rejected from graduate school turns
i
■ hippie).
|
! In the fixed personality mode, the individual
I attempts to regain the lost identity by making various
| concessions, by attempting to improve his "value" or
: worth in the eyes of the lost referent through various
"reward" or "punishment" states in order to reestablish
; the prior identity (i.e., the wife who is rejected
j becomes a model housekeeper). Identity change is less
: and any permanence in the loss has a more severe conse-
| quence on the emotional structure of the individual.
For the changeable mode, there is a general shifting,
between extreme attempts to regain the lost referent,
i to attempts to adopt a new identity. The shifting will
I occur suddenly depending on the perception of control
; gained through a particular stance at any one time.
| Adaptive vs Maladaptive Responses
i
! Under the stress of dissociation, the effect of
| the response to the environment is either adaptive
95 !
i
or maladaptive. An adaptive response limits the possi
bility of further dissociation while giving the maximum
possibility of reestablishing the individual's control
over his environment. A maladaptive response enhances j
I
the probability of further dissociation and further j
loss of control. Adaptive responses may even include
those responses which might be maladaptive from the
society's viewpoint (i.e. drunk driving) if in fact
it does reestablish the individual's control over his
environment and does not lead to further dissociation.
While a non-normative response may be adaptive for the
individual it is by definition not socially adaptive.
We are, thus, working with a response which is adaptive,
and the individual is attempting to use the difference
to his advantage (i.e., by contravening the norms of
the society he retains control over the withdrawn
referent individual). However, since the normative
structure is the basis for most evaluation of the res
ponses of others, the non-normative response tends to j
|
cause further dissociation and, thus, ultimately becomes j
I
1
maladaptive for the individual. A maladaptive response i
on the societal level is maladaptive for the individual
because it may cause further dissociation and loss of
control. A socially maladaptive response which was at
96
one time adaptive for the individual may become mala
daptive due to changes in the environment. This will
be particularly stressful since the response has been
built into the reflex pattern of the individual and
is difficult to change.
All responses are seen as attempts to reestablish
control over the environment as that control is defined.
For the passive dependent individual, the reestablishing
of control may be forcing another to play the dominant
role and take care of him as one would a child, while
for the aggressive independent individual, reestablishing
control may be the attainment of submission by another.
Thus, attempts to make the other who has withdrawn
feel sorry for his withdrawal because it has "caused"
an arrest for drunk driving, may be as adaptive as
bringing the other person flowers in an attempt to
repair a ruptured relationship.
Response to Rejective Dissociation
This section will describe specific reactions to
rejective dissociation which lead to rejection by other
parts of the society. As an example of rejective dissoc
iation, I will use an individual who cannot fulfill a
specific role placed upon him by others. Other parts
of the system are placed under strain and counter
97
pressures are put into operation to help relieve the
individual of certain responsibilities, or the individual
is bypassed in order to maintain system functioning
without the individual's participation, in response,
the stress of rejection and loss of control may be
experienced, and depending on past defense mechanisms
and adaptive learned responses the individual may either
recoup his losses or find that he is responding inade
quately to the loss and is, thus, losing complete control
of the changes taking place. He finds himself even
more isolated. If the individual perceives no way out
of the isolation or rejection through re-entry into
former roles or through alternative roles, he may be
forced to alleviate the stress experienced through various
non-normative responses. Several types of non-normative
responses will be discussed here as they relate to
attempts to regain identity-control (over the indivi
dual's environment). These responses in order of their
presentation are suicide, aggression, and mental illness.
Suicide as the Result of Loss of Control
While Durkheim does not posit stress per se as
a cause of suicide, his egoistic and anomic suicide
point to qualitatively different forms of dissociation.
98
In egoistic suicide, there is a lack of societal inte
gration of the individual, "... [since] the individual
ego asserts itself to excess in the face of the social
ego and at its expense, we may call eogistic the special
type of suicide from excessive individualism.1 1 {Durkheim,
p. 20 9, 1951) The individual is thrown onto his own
resources without the collective support of strong group
norms against suicide or the moral support for growth
of the individual as a part of the whole. According
to Durkehim's study, this form of suicide is found in
a society where secularization and individualism is
important and where loss of cohesion characterizes
the social bonds.. Thus, Durkheim sets up and supports
three propositions such that suicide varies inversely
with the degree of integration of religious society,
of domestic society, and of political society. (Durkhiem,
1951)
The other form of suicide is labelled anomic by
Durkheim. In such a state, a man is set adrift with
unattainable goals, he is unregulated by society in his
goals, and he suffers the torment of unquenched ambition,
"...it is not true, then, that human activity can be re
leased from all restraint." (Durkheim, p. 252, 1951)
Such conditions exist in severe economic crises (economic
99
disaster or suddenly increased power and wealth). The
sudden reduction of aspirations and the necessity for
unaccustomed restraint or, on the other hand, the release
of all restraint and the upsetting of the norms of
social structure create a heightened degree of anomie.
Such a state can be regarded as a dissociation from
previously sustained norms and restraints. Durkheim
equates the two forms of suicide when he states,-
"Certainly, this and egoistic suicide have kindred ties.
Both spring from society's insufficient presence in
individuals." (Durkheim, p. 258) He then goes on
to indicate how they differ, with egoistic suicide due
to a deficiency in the collective activity of the society
and anomic suicide due to a deficiency of societal
restraint in the passions of the individual. The dissoc
iation is evidenced even more clearly in the following
passage. "Two factors of suicide especially, have a
peculiar affinity for one another: namely, egoism
and anomie. We know that they are usually merely two
different aspects of one social state; thus, it is
not surprising that they should be found in the same
individual. It is, indeed, almost inevitable that the
egoist should have some tendency to non-regulation, for
since he is detached from society, it has not sufficient
hold upon him to regulate him." (Durkheim, p. 288,1951)
100
Durkheim's anomic and egoistic suicide can be
seen as the result of dissociation. The main impact
of his analysis is that suicide is due to the loss of
control or restraint by society, but it is this loss
of control by society which tends to leave most reactions
unbridled and man no longer has the same sense of control
over his environment. His reactions no longer elicit
the same responses as previously, and he loses his
identity. It is this dissociation due to loss of control
and identity which leads to increased suicides.
Henry and Short in their book Suicide and Homicide
take Durkheim's approach in delimiting their theory
of suicide. They postulate that "(1) ... suicide rates
in high status categories are higher than suicide rates
in low status categories because high status categories
are subject to fewer external restraints than low status
categories ... (and) (2) ... the probability of suicide
varies inversely with the strength of the relational
system ..." (Henry and Short, 1954)
The restraint of the relational system can be
seen in the lower suicide rate in the ranks of the
"married, among residents of rural areas, and of well-
organized urban centers.
101
Henry and Short are hard pressed, however, to
explain the increase of suicide with age and the fact
that the non-white female rate is higher than the non
white male. The strength of the relational system
is used to explain the age phenomenon but only from
the point of view of external restraint. As far as
the non-white female rate being higher than the non
white male, they use the explanation that the higher
status and occupational position of the Negro female
in the family system makes her more susceptible to
business cycles.
However, in both these cases, utilizing the theory
of referential dissociation, the relational system itself
may have the effect of cutting off close contacts with
others regardless of the extent of restraint by the
system. This causes dissociation and stress leading
to defense mechanisms and possibly suicide as an alter
native measure to a sense of hopelessness or an attempt
to exercise control, albeit the last act of control
to make the other feel sorry, to feel quilty, etc.
While restraint may be one intervening variable, certainly
the resulting dissociation is the major process leading
to suicide.
102
Thus, disintegration of the social bonds is related
to the concept of dissociation and loss of control.
Where this dissociation is greatest due either to the
mechanization of society, excessive individualization,
or the lack of restraint due to disorganization, there
will be a high suicide rate.
Frustrati on-Aggr e s s ion
In this brief exploration of certain reaction
formations and maladaptive responses, the frustration-
aggression hypothesis should be explored for its value
in explaining reaction to perceived dissociation. This
hypothesis posits that frustration or blockage of primary
(innate) or secondary (learned) drive states in the
individual causes a violent reaction which is either
directed inward against the self or outward against
others. Put in terms of the referential dissociation
hypothesis, there is an attempt to reduce the stress
generated by the inability to control the environment.
Such reactions may be directed against the object per
ceived as causing the frustration (maladaptive in that
it can cause further dissociation) or against a less
threatening substitute (adaptive, if it does not cause
dissociation of important referents). These two responses
103
with regard to objects are called aggression and
displacement respectively. The frustration-aggression
hypothesis put forth by Dollard, et al. (1961) cannot
be accepted uncritically. There are many different
reactions to frustration which can neither be labelled
aggression nor displacement. Both aggression and
displacement, whether self or other-directed, implies the
will to unity or to self assertion. However, there
is the reaction of helplessness and hopelessness —
a general deletion of the self which follows severe
frustration and an inability to control one's environment.
Mental breakdowns and psychosomatic diseases may be
the result of a general giving up — a release of the
basic desire to maintain the unity of self stemming
from a feeling of hopelessness. Many studies can be
cited to back up this hypothesis. Principally relevant
is a psychosomatic medical study done by Schmale (1964)
which indicated a positive correlation between object
loss of a significant other, attitudes of giving up,
and the onset of disease. Similar conclusions were
reached by Hinkle and Wolff (1957). The major finding
of research on stress by Schmale (1964) was that stress
and the onset of disease was a product of neither high
physical or environmental stress but the result of
experiencing a change in life to which there appeared
to be no solution. Therefore, there appears to be
a total giving up due to inability to regain control
and, therefore, the loss of identity. The impulsive
act which leads to fatal automobile accidents may be
the result of a situation when the individual is over
whelmed by his loss. Yet another form of giving up
is the process of self indulgence (self-pity) and attempts
to increase the reason for self pity and pity by others.
This is perhaps one of the more common reactions to
loss of control and, thus, identity. Drinking enhances
the state of self pity and the individual may consciously
or subconsciously cause injurious things to happen to
him in order to increase the state of self indulgence
and indulgence by others. This is labelled as a "giving
up" response. A distinction is made between other-
oriented giving up and self-oriented giving up — one
being a sense of helplessness and the other hopelessness.
Mental Illness
Mental illness, on the other hand, may not be
a giving up process or a mechanism which is directed
toward homeostasis of the internal system but a coping
mechanism which gives meaning to the individual, provides
him with an alternative identity through which control
can be established based on a "different view of reality,"
one which cannot be obtained through previous identities.
| This shift to meaning and identity-control as the mechan-
! isms leading to response and change relative to the
j
external environment may overcome some of the basic
’ discrepancies found in using homeostatic or consistent
|
! models in the explanation of human action. Menninger's
j thesis on homeostasis, heterostasis, and entrophy postu-
I late a general return to a balanced position through
varying levels of coping processes which eventually
! put the system into balance (Menninger, 19 63). Just
■ what this balance is or what the psychological equivalent
; to physiological changes are which tend to return the
j organism to a steady state are not explicitly stated.
■ The fact of constant tension, of willingly increasing
1
tension (studies on the earthworm which indicate "boredom,
i
; and a movement toward stimulation and change), of desires
|
j for breakdown of the body and ultimately death of the
| organism through psychosomatic disease and suicide are
|
: poorly accounted for by a homeostasis model. On the
; other hand, the attempt to regain identity through control
J
1
over the environment leads to an alternative explanation
; which is consistent with empirical fact.
106
Summary
Various responses such as suicide, psychosomatic
illness, and mental illness have been outlined as
attempts to re-establish control over the environment
when such control has been perceived as lost.
CHAPTER VI
RECAPITULATION
Introduction
This chapter will accomplish four tasks:
(1) provide a summary of the link between the litera
ture review on drinking driving and accidents, data
on drunk drivers, and the theory of referential
dissociation; (2) indicate the possible alternative
explanations to the drunk driving and accident phe-
nomemon; (3) indicate how the referential dissociation
theory could be tested in order to eliminate the alter
native explanations; (4) provide an overall summary
of the relationship between drunk driving and accidents,
and other types of "deviant" responses.
The theory of referential dissociation: (1)
examines the basic assumptions about man as a respond
ing organism, (2) discusses a rationale for particular
types of response, (3) indicates a response mechanism,
and (4) shows that this response mechanism can have
alternative manifestations.
The basic as sumptions of the theory are as follows:
a) man responds in order to maintain control over his
environment for purposes of fulfilling his psycholog
ical and physiological needs; (b) such control is not
107
108
exercised to create a consistent state, since man
sustains much inconsistency in order to obtain fulfill
ment of his needs; (c) man is a socially interacting
organism; and (d) his interaction provides him with a
social identity.
The rationale presented for response is a percep
tion of loss of control of those elements which are
necessary for fulfilling psychological and physiological
needs. Such perception occurs when a part of an
individual's social identity is seen as withdrawing.
This withdrawal leads to a loss of self identity and
a sense of lost control.
The response mechanism occurs in an attempt to
regain control by reincorporating the withdrawn element
or by attempting to find a substitute identity.
The manifestations of the response mechanism may
be adaptive or maladaptive, normative or deviant.
The link between the literature review, the data,
and the theory
The literature review and the data analysis give
substantiation for different parts of the above theory.
The basic assumptions about man are not dealt with by
the literature review or the data but underlie the
theoretical framework.
109
The major theory states that stress arises due to
perceived loss of environmental control associated with
the loss of, or withdrawal of, identity referents. The
attempt to regain control may lead to adaptive responses
or to various nonnormative or maladaptive responses of
which drunk driving, accidents, suicide, etc., are
manifestations.
Several studies on drinking driving and accidents
are supportive of this interpretation. In particular,
the studies of substantial childhood trauma in the lives
of high accident drivers (Tillman and Hobbs, 19 49) and
of drinking accident drivers as against nondrinking
accident drivers (Barmack and Payne, 1961). Several
studies directly confirm the dissociation idea by
indicating that a higher proportion of the accident
population than of the nonaccident population were
married-living apart (Barmack and Payne, 1961) and were
separated to a greater extent in the accident over the
nonaccident population (Brown, 196 8).
. Related directly to the idea of immediate stress
as an antecedent to accidents are several studies which
indicate a recent separation. Selzer (1967) found that
personal conflict, tragedy, as well as vocational and
financial stress were found in 52 percent of an accident
population as against 18 percent for a control population.
110
Tabachnick's study (1967)/ indicating a major change in
responsibility for those involved in automobile accidents
can be interpreted as a separation from previous roles
and previous role encumbents.
Of related importance is the finding that a large
percentage of each accident group exhibited a personality
disorder. Several studies have indicated the extent of
hostility with low control/ separation anxiety, aggres
sive behavior (Conger/ 1959), antisocial personality,
alcoholism (Brown, 1967), and passive aggressive, alco
holism, schizophrenia, paranoid thinking, suicidal acts
or thoughts, depression, and violence (Selzer, 1963) in
the drinking accident population. Many of these
disorders would tend toward separation from others or
the withdrawal of others.
Several studies, however, uphold the alternative
hypothesis related to a generally disorganized life
style which may lead both to separation from others and
to drunk driving. Thus, the Tillman and Hobbs study
(1949) and the McFarland review (1954) indicate that
the drinking accident driver was known to public
agencies for social deviance and/or problem and that
the number of arrests is much higher for the drinking
accident driver than for other groups (Waller, 19 67;
Barmack and Payne, 1961). In several of the studies,
Ill
other factors were present which suggests that some
form of stress was also attendent on the accident or
was implied by the past history of the driver.
The data analyzed on drunk drivers further support
the idea of stress due to separation by indicating a
significant relationship between a major proportion of
the separation indices utilized and drunk driving.
Two indices of separation, utilized "length of
time living alone" and "presently employed" first
appeared to suggest that the stress hypothesis is not
correct. However, the relationship was found to be
in the right direction for the length of time living
alone, suggesting that the less the time living alone
(recent separation) the greater the number of drunk
driving arrests. The presently employed category may
not stand against the theory. Since it is possible
that being a slightly older population than the control
group a greater proportion would have jobs.
The association between drunk driving and separa
tion indices is evidenced even when alcoholism is
controlled. The only major problem is one of time
priority: did the separation lead to drunk driving or
did prior drunk driving lead to separation? The only
other question of importance relates to the cause and
effect of the association found. Thus, a generally
112
disorganized life might lead to both drunk driving and
separation. Since data cannot satisfactorily answer
this question the alternative explanation is left open.
Alternative Explanations
The data analysis and background research tend to
substantiate the association between separation and
drunk driving. There is still the possibility, however,
that the relationship between the two variables is not
causal — or, if causal, not an immediate cause and
effect.
The first alternative is general disorganization
in the individual's life - he "drives as he lives" and
generally gets arrested for drunk driving or other
deviant acts because of a particular style of life
which is impulsive and nongoal directed. To carry the
argument further, this s£yle of life would also account
for the separations experienced.
A second alternative explanation relates to what
can be labelled as long-term dissociation. The
individual feels a general separation from others
because of early dissociation in life; this separation
becomes particularly relevant at certain times in later
life although not necessarily connected with immediate
stress due to separation from others.
With the initial explanation relating to referen
tial dissociation we have three alternative approaches
(1) drunk driving as the result of the immediate
stress of dissociation/
(2) drunk driving as the result of a particular
style of life which also produces the
independent stress of dissociation,
(3) drunk driving as the result of the stress of
dissociation which may have occurred in the
past but has particular relevance in the
present.
The data analysis has several limitations in
distinguishing one possibility from the other.
(1) it does not distinguish between acute and
chronic stress,
(2) controls are lacking for antecedent variables
which might account for both separation and
the responses studied.
Proposed Design
This leads to the third purpose of this chapter,
an explication of the design necessary to determine
which of the above hypothesis can be rejected. The
first problem is one of delineating a population which
can be tested. This can be done by isolation of a
specific response.
The use of a specific type of response is easier
when the response is a particular "deviant" response
since it is usually well defined and visible (i.e.,
drunk driving, suicide, mental illness). The
114
major problem with lumping all of an apparently similar
response type together is the different reasons for the
response. Thus, for a part of the drunk driving popula
tion, the individual may be "accidently" apprehended on
his way home from a party where the "deviance" of the
response was not necessarily a response to the stress
of a major withdrawal or part of a particular life
pattern. Further, the drunk driving population may be
a severely biased population due to police arrest
patterns. In this case, there may be more indices of
separation in the arrested population than in the control
population because separation is part of a particular
life style. In order to eliminate these two facts,
it would be necessary to select several populations
which are not as heterogeneous regarding external
factors. Thus, a fatal accident population, controlled
for responsibility for the accident, might be a more
homogeneous population in that there is no police bias,
and fatal accidents are not a way of life. However,
there is difficulty in collecting information. This
can be avoided by utilizing near fatal accident cases.
This could readily be accomplished through use of
hospital records, followup surveys, and interviews with
those persons responsible for near fatal automobile
accidents.
1X5
One way in which to assess the effects of dis
sociation on different individuals is to select a dis
sociating event which effects many persons in a similar
way, such as a large-scale job lay-off or the death of
a spouse in an automobile accident. However, several
factors which might bias the results must be taken
into account in selecting the population for study;
(1) the degree to which the dissociation can be
rationalized and handled through denial
(2) the existence of readily available alterna
tives
(3) the degree of identify formation between the
individual and the withdrawn referent
(4) the degree to which the loss is perceived
as permanent
(5) the degree of emotional preparation for
the event
(6) the degree to which the individual feels
rejected
The lay-off would have to be discounted on several
of the above grounds because of the insulating factors
present in a majority of cases. On the other hand, an
interview of auto accident widows might involve a
population with a greater sense of dissociation but
it would be necessary to take the different types of
learned response into account. Since the major theory
is one of response to stress, it is difficult to select
one population of individuals undergoing a particular
116
"stressful" event and predict the response of individuals
within that population because of the insulating
variables and the different types of learned response.
It is necessary to select a population which will have
as few insulating factors as possible.
With either type of population, whether specific
event or specific response, several types of data must
be collected.
(1) data on stress immediately preceding the
incident,
(2) data on perception of separation,
(3) data on separations any time during the
preceding lifetime.
In order to test data to distinguish between the
"drives as lives" hypothesis and the stress hypothesis,
it would be necessary to show that stress was not an
important determinant of behavior - specifically, the
stress arising from perception of separation. In order
to accomplish this, it would be necessary to correlate
(1) degree of generalized separation from others (an
individualized perception of social distance from others,
(2) perception of actual withdrawal of others, and
(3) behavioral indices of separation with several forms
117
of response using indices of impulsive behavior,
alcoholism, and disorganized life style patterns such
as numerous offenses, numerous separations (four-plus
jobs in 5 years} as controls.
Relationship between "Deviant Responses"
Finally, it is important to delineate the relation
ship between drunk driving, accidents, and other deviant
responses. What is the unifying concept? In each case,
the deviant response is a response to perceived loss of
control.
Underlying all types of response is an attempt to
gain physiological and psychological needs and the
attempt to order the environment to assure the provision
of these needs. When specific needs which form a part
of the individual's identity are perceived as withdrawing,
a response is elicited which is calculated to regain
control. Often this response is maladaptive or if not
maladaptive it may be deviant relative to the generally
accepted response patterns. Any response which causes
greater dissociation is labelled maladaptive. Suicide
may be a response calculated to gain control over
another (make the other person feel sorry for what he
has done), just as the drunk driver may be attempting
to redefine the situation in order to reestablish control.
118
While it has been possible to lend some support
to the referential dissociation theory through the
literature review and the present data analysis, more
substantive research is needed.
CHAPTER VII
SUMMARY
The review of the literature suggested that separa
tion from others produced stress which led to drinking
driving and to accidents. Drinking was not considered
a causal factor but only an intervening variable to
observable behavior leading to arrest or to accidents.
The data analysis utilizing samples of the drunk driver
population drawn from Los Angeles County courts indicated
that when controlled for alcoholism, and occupational,
status, separation from others appeared to be more
evident in the convicted drunk driving group than in
the driver's license applicant group. Based on this
substantiation of the hypothesis developed from the
literature review, a theory of separation from others
was developed. A summary of the basic points developed
follows:
( . 1 ) Man's basic drive is a sustaining of identity
through control over the environment in such a way as
to fulfill basic physiological and psychological needs.
(a) Any response to the environment (i.e.,
another person) is an attempt to sustain a particular
type or level of control over the- degree of stimulation
necessary to fulfill phyiological and psychological
needs. Thus, given individual differences in capaci-
119
120
ties, capabilities, and innovative response patterns,
certain responses are utilized repeatedly and become
identified with the individual.
(b) These response patterns tend to become
set and through time external objects are accepted
insofar as they respond to the individual's basic
response patterns and provide his basic needs.
(c) The external objects responded to
overtime become identified with the basic response
patterns and, in sum, make up the social individual in
terms of his interaction with the external environment.
(d) Any shift in the response of these social
objects is countered by attempts toward realignment
(i.e., reassumption of control).
(2) Interaction with other social selves or social
objects defines the boundaries of the self by setting
the parameters of acceptance-rejection.
( . 3 ) Interaction is a process of continually testing
these boundaries and, thus, confirming control; but due
to fear of rejection and thus loss, the violation of
boundaries is seldom too great.
(4) When there is a perception of loss of control,
an attempt is made to regain control through retension,
realignment, development of an imaginary world, substi
tution , removal, etc.
120
ties, capabilities, and innovative response patterns,
certain responses are utilized repeatedly and become
identified with the individual.
(b) These response patterns tend to become
set and through time external objects are accepted
insofar as they respond to the individual1s basic
response patterns and provide his basic needs.
(c) The external objects responded to
overtime become identified with the basic response
patterns and, in sum, make up the social individual in
terms of his interaction with the external environment.
(d) Any shift in the response of these social
objects is countered by attempts toward realignment
(i.e., reassumption of control).
(2) Interaction with other social selves or social
objects defines the boundaries of the self by setting
the parameters of acceptance-rejection.
( . 3 ) Interaction is a process of continually testing
these boundaries and, thus, confirming control; but due
to fear of rejection and thus loss, the violation of
boundaries is seldom too great.
(4) When there is a perception of loss of control,
an attempt is made to regain control through retension,
realignment, development of an imaginary world, substi
tution, removal, etc.
121
(5) There are several levels of control loss -
called here referential dissociation.
(a) loss of control over the mechanisms of
larger social structure (atomistic).
(b) loss of control over individuals important
to fulfillment of basic needs (rejective).
(c) loss due to inability to fulfill conflicting
demands of various role positions (conflictive).
(d) perceived loss where none in fact has
occurred (imaginary).
(6) Loss of control gives rise to stress. Stress
is a preadaptation state in which the organism is taken
beyond immediate integrative capacities to the point
where control of the environment becomes problematic.
( . 7 ) The initial loss of control leads to further
loss due to what can be called a rejection syndrome if
the response is maladaptive. Rejection builds up as
a chain reaction from other sectors of the society.
(8) Suicide as a response to control loss is an
attempt to regain control of the environment by forcing
others to react in a prescribed way - guilt, grief, "if
I had only done this and this, he wouldn't have killed
himself."
CHAPTER VIII
SUGGESTIONS FOR FURTHER RESEARCH
Stress and Adaptation
No definite conclusions can be advanced regarding
the causes of drunk driving, accidents, suicide, mental
illness, but it appears that several new dimensions
might be profitably explored:
(1) The relationship of stress to change in the
human condition - as antecedent to pathology
and deviant behavior - needs to be studied
in greater depth. It has been suggested
that all non-normative behavior is motivated
by the stress of dissociation and the attempt
to regain control and, thus, establish
identity.
(2) The type of adaptation possible and the
previous pattern of adaptation used by the
individual must be determined in order to
successfully interpret reaction to stressful
experiences.
C3) The part that dissociation plays in the human
being has been observed throughout the history
of man. What is its meaning and what reactions
does it create? This paper has suggested one
possible avenue of approach - the loss of
control and identity - and has attempted to
show that deviant reactions are associated
with attempts to regain control, thereby
sustaining the identity.
Of interest in future research may be the part
played by the socialization process in different social
classes, creating different learned forms of release
for aggression. Is it possible that the lower class
122
123
atmosphere is conducive to drinking and to release of
hostility and aggression? Does this act as a safeguard
for the individual against major accidents? Does the
socialization process through which the middle class
individual goes, with its attendant control of emotions
and the drive for status goals, create a greater pos
sibility for explosive releases of repressed anxiety at
particularly stressful periods?
The trend toward social psychological research
designs in attempting to understand stress and reaction
has generated many questions for which we do not
yet have answers: (1) Is stress a major factor in
the etiology of deviance? (2) How does the psychological
state of an individual lead to greater or less deviance?
(3) What governs the threshold of resistance to
stress? (4) What part does control play in the determina
tion of the type of stress release mechanism used?
(5) How do primary and secondary groups effect the
mechanisms used to reduce stress? (6) What part do
childhood experiences (trauma) play in increased
susceptibility to various reactions? As yet we do not
have large enough sample populations to control for
type of stress and type of precipitating event.
Nor has any work been done on the ways in which different
individuals react to perceived stress given specific
124
personality disorders and various background variables.
Given certain background variables (i.e., childhood
trauma), specific personality types (adaptive states),
indices of ability to control stress, and precipitating
stresses as well as latitudes for action within the
environment, it may be possible to predict different
types of reaction to stress with surprisingly high
accuracy.
The concept of. the life cycle and the change
which occurs in terms of the expectations of self
and others and the changes in role as prescribed
by the society may effect some individuals more than
others. Thus, those with more rigid coping abilities
may not be able to adapt as easily to normal life
cycle changes. In this context, the idea of self
concept may have high predictive power. The individual
with a poor self concept may be affected to a greater
extent by changes in his environment since he may
not have sufficiently successful adaptive response
patterns.
Still another area as yet unexplored is the
effect of changes in society on the response of
individuals. If the environment determines to some
extent the responses made by an individual, then
changes in the way the individual interacts with
125
other parts of the society will change as the environment
changes. Thus, suicide has been correlated with
economic fluctuations and with the growing number
of inconsistent demands made by an individual's primary
and secondary groups. Mental illness has been correlated
with the increasing division and break-up of the primary
groups in a society and with the attendant strains
taking place in upward status mobility.
An area needing further study is the
extent to which accidents, drunk driving, suicides,
mental illness, and psychosomatic diseases are all
correlated with changes in societal values and environ
ments . Are these merely different mechanisms for
coping with changes and/or different forms of giving
up when faced with inability to control or adapt
to rapid change?
Variables in Future Research
In future research, the following variables must
be accounted for in order to gain predictive ability:
(1) Environmental stimulation and interaction
{physical and social)
(2) Need or drive state at time of stress
(3) Previous learned responses
(4) Previous associative meanings of environmental
stimulations
(5) Environmental prescriptions
126
(6) Innate response mechanisms
(7) Available response alternatives
(8) Previous environmental adaptive level
(9) Cyclic changes in the internal environment
(aging, arousal levels, etc.)
(10) Definition of the situation
(11) Attitude sets (helplessness, hopelessness,
low self esteem)
(12) Environmental supports
That this appears to be a list too long and
detailed to be taken into consideration in present
research indicates how superficially we have delved
into the area of stress in an attempt to understand
man.
REFERENCES
Alexander, F., "Accident-Prone Individual," Public
Health Reports 64, (1949), 347-362.
Arbous, A .G. and J.E. Derrich, "Accident Statistics
and the Concept of Accident Proneness," Biometrics 7,
(1951), 340-432.
Asch, S.E., "Effects of Group Pressure on the Modification
and distortion of Judgements," in H. Geutzkow (ed.) ,
Groups, Leadership and Men, (Pittsburgh. Carnegie,
1951).
Barmack, Joseph E. and Donald E. Payne, "Injury-Producing
Private Motor Vehicle Accidents Among Airmen,"
Bulletin 285, (Highway Research Board, 19 61).
Barmack, Joseph E. and Donald E. Payne, "The Lackland
Accident Countermeasure Experiment," Highway Research
Board Proceedings 40, 1961, 513-522.
Basowitz, H., H. Persley, S.J. Norchin, and R.R. Grinkein,
Anxiety and Stress (New York: McGraw Hill, 1955),
p. 289.
Bjerver, K.B. and L. Goldberg, "Effect of Alcohol
Ingestion on Driving Ability," Quarterly Journal of
Studies on Alcohol 11, (March, 195Oj , 1-30.
Bjerver, K.B., L. Goldberg, and P. Linda, "Blood Alcohol
Levels in Hospitalized Victims of Traffic Accidents,"
Proceedings of 2nd International Conference on Alcohol
and Road Traffic, (Garden City Press Co-Op, Toronto,
1955), 92-102.
Borkenstein, R.F., R.F. Crowther, R.P. Shumate, W.B. Ziel,
and R. Zylman, The Role of the Drinking Driver in
Traffic Accidents^ (Department of Police Administration,
Indiana University, 1964).
Carpenter, John A., "Effects of Alcohol on some Psycho
logical Processes," Quarterly Journal of Studies on
Alcohol 23, (June 1962), 274-314.
127
128
Carpenter, John A., "Contributions from Psychology to
the Study of Drinking and Driving," Quarterly Journal
of Studies on Alcohol, Supplement 4, (May 1968).
Cisin, Ira H., "Social Psychological Factors in Drinking-
Driving," in Fox and Fox {eds.}/ Alcohol and Traffic
Safety, (Department of Health, Education, and Welfare,
1963) .
Coldwell, B.B., D.W. Penner, H.W. Smith, G.H.W. Lucus,
R.F. Rodgers, and F. Darroch, "Effect of Ingestion
of Distilled Spirits on Automobile Driving Skill,"
Quarterly Journal of Studies on Alcohol 19, (1958),
p. 590.
Conger, J.J., H.S. Gaskill, D.D. Glad, L, Hassel,
R.V. Rainey, and W.L. Sawyer, "Psychological and
Psychophysiological Factors in Motor Vehicle Accidents,"
Journal of the American Medical Association 169, (1959),
1581-1587.
Cooley, Charles H., Human Nature and the Social Order,
(New York: Charles Scribner's Sons, 1922), p. 77”
Cosper, R. and K. Mozersky, "Social Correlates of Drinking
and Driving," Quarterly Journal of Studies on Alcohol,
Supplement 4, (May 1968).
Dollard, John, et a!L. , Frustration and Agression, (New
Haven: Yale University Press, 1961).
Dunbar, H.F., "Susceptability to Accidents," Medical
Clinic of North America 28, (1944), 653-662.
Durkheim, Emile, Suicide, trans., John A. Spaulding and
George Simpson, (Glencoe, 111., The Free Press, 1951),
pp. 209,252,257,258,288.
Efron, Vera and Mark Keller, Selected Statistical Tables
on Consumption of Alcohol and on Alcoholism, (Rutgers
Summer School of Studies on Alcohol, July 1963).
English, Horace B. and Ara C. English, A Comprehensive
Dictionary of Psychological and Psychoanalytical Terms,
(New York: Longman, Green, & Co., 1961).
129
Gerber, S.R., "Vehicular Fatalities in Cuyahoga County,
Ohio, USA: Twenty Years Experience (1941-1960)," in
Proceedings of Third International Conference on Alcohol
and Traffic Safety, (London, 1962).
Gibbs, Jack P. and Walter T. Martin, Status Integration
and Suicide, (Eugene, Oregon: University of Oregon
Press, 1962).
Gold, David, "Statistical Tests and Substantive Signifi
cance," The American Sociologist, (February 1969),
42-46.
Goldberg, L., "Quantitative Studies on Alcohol Tolerance
in Man. The influence of ethyl alcohol on sensory,
motor, and psychological functions referred to blood
alcohol in normal and habituated individuals," Acta
Psysiological Scandanavia 5, (1943), 1-128.
Goodman, Leo and William Kruskal, "Measures of Association
for Cross Classifications, 1," Journal of the American
Statistical Association 49, (December 1954), p. 740.
Greenwood, M. and H.M. Woods, "The Incidence of Industrial
Accidents upon Individuals with Special Reference to
Multiple Accidents," Rep. Indust. Res. Bd. No. 4,
(London, 1919).
Greenwood, M. and G. Yule, "An Inquiry into the Nature
of Frequency Distributions representative to Multiple
Happenings with Particular Reference to the Occurrence
of Multiple Attacks of Disease or of Repeated
Accidents," J. Roy. Statis. Society, (1920), p. 83.
Gruner, 0., 0. Ludwig, and H. Domer, "Zur Abhangigkeit
aldoholbedingter Aufverksamkeitsstorungen vom
Blutalkoholwert bei niedrigen konzentrationen,"
Blutalkohol 2, (1964), 445-452.
Haddon, W., "Alcohol and Highway Accidents," in
Proceedings Third International Conference on Alcohol
and Traffic Safety, (London, 1962).
Haddon and Bradess, "Alcohol in the Single Vehicle Fatal
Accident," Journal of the American Medical Association
19, (1959), p. 1587.
130
Hayman, Max, "The Myth of Social Drinking," The American
Journal of Psychiatry 124, (November 1967), 39-48.
Henry, Andrew F. and James F. Short, Jr., Suicide and
Homicide, (New York: The Free Press, 1954), p. 75.
Hinkle, L.E. and H.B. Wolff, "Health and the Social
Environment: Experimental Investigations," in
Leighton, A.H., J.A. Clausen, and R.N. Wilson (eds.),
Explorations in Social Psychiatry, (New York: Basic
Books, 1957) .
Homans, George Caspar, Social Behavior, Its Elementary
Forms, (New York: Harcourt, Brace, and World, Inc.,
1961)“.
Hyman, Morton M., "Accident Vulnerability and Blood
Alcohol Concentrations of Drivers by Demographic
Characteristics," Quarterly Journal of Studies on
Alcohol, Supplement No. 4, (May 1968).
__________ , "The Social Characteristics of Persons
Arrested for Driving While Intoxicated," Quarterly
Journal of Studies on Alcohol, Supplement No. 4,
(May 196 8) .
Jellinck, E.M., The Disease Concept of Alcoholism,
(New Haven: College and University Press, I960).
Jones, Edward E. and Harold B. Gerard, Foundations of
Social Psychiatry, (New York: John Wiley and Sons,
Inc., 1967) .
Kendall, Maurice G., Rank Correlation Methods, (New York:
Hafner, 1955) .
Lazains, Richard S., "Stress Theory and Psychophysio-
logical Research in Emotional Stress," Lennart Levi
(ed.), Proceedings of an International Symposium,
Applied Medical Defense Research, Stockholm, (New York:
American Elsevier Publishing Co., Inc., 1967), 42-177.
e
Lemert, Edwin M., "Paranoia and the Four Dynamics of
Exclusion,1 1 Human Deviance, Social Problems, and
Social Control, (New Jersey, Prentice-Hall, Inc., 1967),
197-211.
131
LOomis, T.A. and I.C. West, "The Influence of Alcohol on
Automobile Driving Ability," Quarterly Journal of
Studies on Alcohol, Supplement 4, (May 1968).
____________________________, "The Influence of Alcohol on
Automobile Driving Ability/An Experimental Study for
the Evaluation of Certain Medicolegal Aspects,"
Quarterly Journal of Studies on Alcohol 19, (1958),
30-46.
Lucas, G.W.H., W. Kalow, J.D. McColl, B.A. Griffith, and
H. Smith, "Quantitative Studies of the Relationship
between Alcohol Levels and Motor Vehicle Accidents,"
Proceedings of the 2nd International Conference on
Alcohol and Road Traffic, (Toronto, 1955).
McCarroll, J.R. and W. Hadden, "A Controlled Study of
Fatal Automobile Accidents in New York City," Journal
of Chronic Diseases 15, (1962), 811-826.
McDougall, W. and M. Smith, "The Effects of Alcohol and |
some other Drugs during Normal and Fatigued Conditions,"!
Great Britain Medical Research Council, Special Report
Series No. 56, (London: Her Majesty's Statistics
Office, 1920).
McGuire, Frederick L., "A Typology of Accident Proneness,"
Behavioral Research in Highway Safety, (January 1970).
McFarland, R.A., "Alcohol and Highway Accidents: A
Summary of Present Knowledge," Traffic Digest and
Review 5, (1964), 30-32.
McFarland, R.A. and Roland C. Moore, with B. Warren,
"Human Variables in Motor Vehicle Accidents," Harvard
School Public Health, (Boston, 1955).
Mead, George Herbert, George Herbert Mead on Social
Psychology, Anselm Strauss (ed.), (Chicago: University
of Chicago Press, 1956), pp. 200,204.
Moore, Wilbert E., Foundations of Modern Sociology Series,
(New Jersey: Prentice-Hall, Inc., 1965).
Moskowitz, H. and D. Depry, "Differential Effect of
Alcohol on Auditory Vigilance and Divided-Attention
Tasks," Quarterly Journal of Studies on Alcohol 29,
(March 19 68), 54-63.
132
Mulford, H., "Iowa's Drinking Problem Driver 1961: With
a Method of Identifying Drinking Drivers in a Survey
Sample," Social Problems 12, (1964), 196-211.
Mulford, H., "Drinking and Deviate Drinking," Quarterly
Journal of Studies on Alcohol, (December 1964), 634-65 0.
Neilson, Royal A., Alcohol Involvement in Fatal Motor
Vehicle Accidents in 41 California Counties in 19^6,
(San Francisco: California Traffic Safety Foundation,
July 1967).
Parsons, Talcott, Societies: Evolutionary and Comparative
Perspectives, (New Jersey: Prentice Hall, 1966).
Phillips, Derek L., "Rejection: A Possible Consequence
of Seeking Help for Mental Disorders," American
Sociological Review 28, (December 1963), 963-972.
Pittman, D.J., "Alcoholism: An Interdisciplinary
Approach," Proceedings of a Conference on Community
Health, (St. Louis: Washington University Press,
1959).
Popham, R.W., "Alcoholism and Traffic Accidents: A
Preliminary Study," Quarterly Journal of Studies on
Alcohol 17, (1956), 225-232.
Schein, Edgar H., "The Chinese Indoctrination Program
for Prisoners of War: A Study of Attempted 'Brain
washing'," Psychiatry 19, (1956), 149-172.
Schmale, Arthur H., Jr., "Object Loss, 'Giving Up,' and
Disease Onset," in Symposium on Medical Aspects of
Stress in the Military Climate, (Washington, D.C.:
Walter Reed Army Institute of Research, 1964).
Schmidt, W. and R. G. Smart, "Alcoholics, Drinking, and
Traffic Accidents," Quarterly Journal of Studies on
Alcohol 20, (1958), 631-644.
Schmidt, W., R.G. Smart, and R. Popham, "The Role of
Alcoholism in Motor Vehicle Accidents," Traffic Safety
Research Review 6, (1962), 21-27.
133
Schuman, S.H., D.C. Pelz, N.J. Ehrlich, and M.L. Slezer,
"Young Male Drivers, Impulse, Expression, Accidentism,
and Violations," Journal of the American Medical
Association 200, (196V), 1026-1030.
Selye, H., The Stress of Life, (New York: McGraw Hill
Book Co., 1956).
_________ , "Personality versus Intoxication as Critical
Factor in Accidents Caused by Alcoholic Drivers,"
Journal of Nervous and Mental Diseases 132, (1961),
p. 298.
Selzer, M.L., Charles E. Payne, Jean D. Gifford,
William L. Kelly, "Alcoholism, Mental Illness, and the
Drunk Driver," American Journal of Psychiatry 12 0,
(October 1963), 326-331.
Selzer, M.L., "Drunk Driving and Fatal Automobile
Accidents," Traffic Digest & Review, (June 1964), p. 31.
Selzer, M.L., C.E. Payne, F.H. Westervelt, and J. Quinn,
"Automobile Accidents as an Expression of Psycho
pathology in an Alcoholic Population," Quarterly
Journal of Studies on Alcohol 28, (September 1967),
505-516.
Selzer, M.L. and J.E. Rodgers, "Fatal Accidents: The
Role of Psychopathology, Social Stress, and Acute
Disturbance," American Journal of Psychiatry 124,
(February 1968) , p. 8 " !
Selzer, M.L. and N.J. Ehrlich, "A Controlled Study of
Alcoholism, Psychopathology, and Stress in 96 Drivers
Causing Fatal Accidents," (Highway Safety Research
Institute, University of Michigan) (to be published).
Shibutani, Tamotsu, Society and Personality, (New Jersey:
Prentice-Hall, Inc~ l"9"62) .
Skinner, B.F., Science and Human Behavior, (New York:
McMillian and Co., 1953).
Smith, H. and R. Popham, "Blood Alcohol Levels in Relation
to Driving," Canadian Medical Association Journal 65,
(1951), 325-328.
134
Sorokin, Pitirim A., Social Mobility, (New York: Harper
and Bros., 1927), 522-525.
Syme, L., "Personality Characteristics of the Alcoholic,"
Quarterly Journal of Studies on Alcohol 18, (1957),
288-302.
Tabachnik, N., "The Psychology of Fatal Accidents," in
Edwin Shneidman (ed.), Essays in Self Destruction,
(New York: Science House, Inc., 1967).
Talland, G.A., "Effects of Alcohol on Performance in
Continuous Attention Tasks," Psychosomatic Medicine 28,
(1966), 596-604.
Thorndide, R.L., The Human Factor in Accidents with
Special Reference to Aircraft Accidents. FinaT
Report Project 21-30-001. (Randolph Field, Texas, Air
Force School of Aviation Medicine, 1951).
Tillman, W. and G. Hobbs, "The Accident-Prone Automobile
Driver; A Study of Psychiatric and Social Background,"
American Journal of Psychiatry 106, (1949), 321-331.
Waller, J.A. , "Identification of Problem Drinking among
Drunken Drivers," Journal of American Medical Associ-
tion 200, (April 1967), 114-120.
_____________, "Accident Patterns Associated with Drinking
and Violation Patterns among Drivers with Alcoholism,
Other Medical Conditions, and No Medical Condition,"
Quarterly Journal of Studies on Alcohol, Supplement 4,
(1968), 118-137.
_____________, "Chronic Medical Conditions and Traffic
Safety," New England Journal of Medicine 273, (December
1965), 1413-1420.
_____________, "Fact and Fiction about Accidental Injury,"
Northwest Medicine 67, (1968), 451-457.
Waller, J.A., "Patterns of Traffic Accidents and
Violations related to Drinking to some Medical
Conditions," Quarterly Journal of Studies on Alcohol,
Supplement 4, (May 1968).
Webb, W.B., "Illusive Phenomena in Accident Proneness,"
Public Health Reports 70, (1955), 951-956.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Status Consistency Among The Clergy
PDF
Social Components Of Housing Cost In The Western Metropolis
PDF
Structural Factors Affecting Fertility In Large United States Cities
PDF
S.T.P.: A Simulation Of Treatment Processes
PDF
The Construction And Empirical Test Of A Theory Based On Selected Variables In Small-Group Interaction
PDF
Differential Fertility Behavior And Values In Rural And Semi-Urban Costa Rica
PDF
An Empirical Examination Of The Relationship Of Vertical Occupational Mobility And Horizontal Residential Mobility
PDF
A Sociological Approach To The Etiology Of Female Homosexuality And The Lesbian Social Scene
PDF
Economic Differentiation And Social Organization Of Standard Metropolitanareas In The United States: 1950
PDF
A Study Of Factors Related To Police Diversion Of Juveniles: Departmentalpolicy And Structure, Community Attachment, And Professionalization Of Police
PDF
Transformational Processing Of Sentences Containing Adjectival Modifiers
PDF
Factors In The Conviction Of Law Violators: The Drinking Driver
PDF
Ecology Of Negro Communities In Los Angeles County: 1940-1959
PDF
Familism, Suburbanization, And Residential Mobility In A Metropolis
PDF
Determinants Of Intercounty Migration: California, 1970-1973
PDF
Consensus Of Role Perceptions In A Welfare Planning Council
PDF
Employment Distribution, Income, And City Size: A Statistical Analysis
PDF
Some Behavioral Consequences Of Career Success: A Synthesis Of Reward Andbalance Approaches
PDF
Resource allocation among organizations within urban communities
PDF
Heart Rate Response To Stress: A Mathematical Model
Asset Metadata
Creator
Thomas, J. Morgan
(author)
Core Title
Referential Dissociation And Response To Stress
Degree
Doctor of Philosophy
Degree Program
Sociology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Sociology, general
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Turk, Herman (
committee chair
), Henry, James P. (
committee member
), Labovitz, Sanford I. (
committee member
), Van Arsdol, Maurice D., Jr. (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-470175
Unique identifier
UC11363187
Identifier
7116440.pdf (filename),usctheses-c18-470175 (legacy record id)
Legacy Identifier
7116440
Dmrecord
470175
Document Type
Dissertation
Rights
Thomas, J. Morgan
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA