toys in the attic: Values and Value-Orientations 465 4 The Theory of Action and Its Application 4.4 Social Behavior and Personality
Development1 - ROBERT R. SEARS In
recent years, a number of useful methods have been devised for measuring social
behavior and the individual personality.
The opinion poll, small group observational procedures, and attitude
scales have contributed notably to the precision with which the action
of groups can be measured and their future behavior predicted. Similarly, in the field of personality and
motivation, such devices as the thematic apperception test, doll play, behavior
unit or time sample observations, and standardized interviews have become more
and more effective for providing objective and quantified statements about
significant variables. From a
practical or applied standpoint, some of these methods have been enormously
valuable. Market surveys, studies of
morale in the military services, diagnostic analyses of disturbed children, and
comparative studies of techniques of teaching have yielded findings that have
much improved the quality of human output.
In effect, the past decade has put in the hands of any competent
technician procedures which permit the empirical discovery of facts and
principles that hitherto had been the province of so-called men of wisdom. For many areas of human action intuitively
skillful lucky guessing has given way to precise and replicable
investigation. The result is a
vigorously expanding body of empirical knowledge about the behavior of both individuals
and groups. One
might feel encouraged, indeed, about this progress but for one thing - there is
no systematic theoretical structure to integrate the empirical findings. By a theory,
I mean a set of variables and the propositions that relate them to one another
as antecedents and consequents. This involves such logical impedimenta as definitions, postulates, and theorems; the
definitions of variables must be mutually exclusive; intervening variables must
ultimately be reducible to operations; the reference events specified as the
consequents in theorems must be measured independently of the antecedents from
which they are derived, and so on. The
general procedure of theory construction is sufficiently standard that it needs
no explication here. 1
This is a modification, with some additions, of the author's Presidential
Address delivered before the American Psychological Association at Chicago,
September 3, 1951, and published in the American Psychologist, VI
(September 1951). 466 The Theory and Its Application The findings to be integrated are those that
describe consistent relationships between behavior (or its products) and some
other events. Essentially, these are
the descriptive behavioral relationships that comprise the disciplines of
individual and social psychology, sociology, and anthropology. Individual and
group behavior are so inextricably intertwined, both as to cause and effect,
that an adequate behavior theory must combine both in a single internally
congruent system. The
chief advantages of a theory are two.
First, it is economical in the sense that it permits many
observed relationships to be subsumed under a single systematic
proposition. For example, it has been
found that severely punishing children's aggression at home reduces the amount
of aggression, that reproving aggressive doll-play acts reduces these, that
societies having severe negative attitudes toward children's aggression contain
little in-group aggression, and that play groups with severely anti-aggressive
leadership exhibit little quarreling. All these observed relationships can be
summarized by the two propositions that punishment creates aggression-anxiety
and that aggression-anxiety reduces aggression.2 Another way of stating the economy point is
that observed relationships have greater generality if the variables involved
are part of a larger theory. An appallingly small number of the relationships that
have been discovered in social psychology can be generalized beyond the
immediate situations in which the studies were made. With respect to attitude measurement, for example, one might well
ask whether any general principles of an antecedent-consequent nature have been
found. In personality study, descriptions of qualities are usually specific to
the particular person examined. And an
increasing number of investigations are casting grave doubts on the predictive value - the validity - of these
descriptions. The other
virtue of a good theory is that it permits the use of multiple variables and
their relating principles, in combination, for the prediction of events. For instance, it is common knowledge among
teachers that a child who has been punished at breakfast is likely to be
aggressive and uncooperative in school.
Frustration breeds aggression. But if one adds to this the principle of
stimulus generalization, one can predict that children who are very severely
punished at home will be nonaggressive in school. Social Behavior and Personality Development 467 2
These particnlar findings could he conceptualized without the intervening
variable of "aggression-anxiety," but this concept is needed for an
adequate incorporation of several other findings into a theory of
aggression. There
has been little opportunity in the behavior sciences as yet to gain the
advantages of such compounding of propositions. Some efforts have been made to discover the personality
characteristics of persons who behave in specifiable ways in groups. The results have been minimal, probably because there is no systematic connection between personality variables and those describing
social actions. Clinically
one can often get a satisfying feeling that tie "understands" a
particular person's behavior, but post hoc
understanding with ad hoc
principles is no substitute for an internally coherent system of
predictive laws. The lack of such a system is
not for want of hard work on the problem. Nor is there lack of brilliant achievement along the way. Neither social psychology nor personality
study is new, and through the last half century there have appeared several
reasonably elaborate theoretical formulations to systematize some of the facts
in both fields. In social psychology there are those of William
McDougall [8] and Floyd Allport [1]; in personality,
those of G. V. Hamilton [4], Kurt Lewin [7], Gordon Allport [2], H. A. Murray
[91, and the successive refiners of psychoanalytic
theory. However, in the main,
these systematizations have dealt with either individual
or social behavior but
not with both. What is
needed at present is a single behavior science, with a theoretical structure that will account for
the actions and the changes of potentiality for actions both of individuals and of groups. Every theory must have a subject matter. It must be a theory about something,
obviously. A certain class of events must be selected for
explication. These are the reference
events, the consequents for which antecedents are discovered. The basic events
to which behavior theory must have reference are actions. This follows from the very nature of our
interest in man. It is his behavior,
the things he does, the ends he accomplishes that concern us. From a
logical standpoint, a theory is of value to the
extent that it orders a set of observations. There are many kinds of observations that
can be and have been made of social and individual behavior. Some of these have involved inferred traits
or needs; others have related to perceptions or to states of
consciousness. By the criterion of logic,
a theory that takes any of these phenomena as its basic reference events is
acceptable. But
there is another criterion to be considered, the practical
one. It is reasonable to ask what kind of events are important to us. On this score, action is clearly more
significant than perception or traits. The clinician must make judgments about personality that
will permit predictions of behavior. Will the patient attempt suicide? Will his performance at intellectual tasks continue to
deteriorate? Will his level of social problem-solving improve under an
anxiety-reduction therapy? Likewise, the teacher and the parent undertake methods of
rearing a child with expectations that his actions will change in a particular
direction. They want him to add more
accurately, or paint more freely, or cry less violently when he is
disappointed; even those changes commonly interpreted as perceptual, such as
art or music appreciation, are evidenced in the form of choices as to where to go, what to look at, what to listen
to. 468 The Theory and Its Application The
situation is even clearer with respect to social behavior. The social engineer
is concerned with such questions as whether a certain parent-child relationship
will establish habitually dependent behavior in the child, whether the eventual
marriage of a courting couple will terminate in divorce or in the social facilitation of the labors of the two people, whether citizens will buy
bonds or vote for a Congressman, whether a group will be shattered or solidified
by external opposition that is, whether there will be an increase
or decrease in cooperative efforts and in-group aggression. Aside
from the fact that a behavior science rather than a need or perceptual
science is of the greatest use to us, there
is an evident practical advantage.
Human beings deal with one another in terms of actions. The teacher has direct observation of the
performance of her pupils. The parent
or the husband or the foreman or the congressman can have only inferrential knowledge
of the ideas or desires of those with whom he interacts. But he can
describe the conditions that impinge on people and he can take note of
the behavioral consequences.
To put the argument briefly: actions are the events of most importance,
and actions are most available to observation and measurement. This is
not to say that needs or motives, perceptions, traits, and other such
internalized structures or processes are irrelevant. Any scientific system must contain both operational and intervening
variables that are independent of the reference events forming the subject
matter of the system. But the choosing
of such variables must depend on their contribution to a theory that will
predict actions. There is no virtue in a descriptive statement that a person or
a class of persons possesses such-and-such a trait or need unless that
statement is part of a larger one that concludes with a specification of a kind
of action to be performed. To describe
a person as having "high emotionality" or "low sensitivity"
or "diffuse anxiety" is systematically acceptable only if other
variables are added that will, together with these internal personal
properties, specify what kind of
behavior can be expected for him under some specific circumstances. Social Behavior and Personality Development 469 MONADIC AND DYADIC UNITS Reference
has already been made to the necessity of combining
individual and social behavior into a single theoretical system. The reasons are obvious. In any social interaction, the interests,
motives, habits, or other psychological properties of the acting individuals
determine to some degree the kind of interaction that will occur. The shy youngster is likely to have less stimulating learning
experiences with his teacher than is a bolder one; the traveler in a foreign
land who knows the language forms different kinds of friendships than the
traveler who uses an interpreter. Conversely, the social
milieu, the interpersonal relationships, within which a person acts
determine his psychological properties.
A man in a subordinate role cannot act as a leader;
a child reared as the younger of two develops differently from one reared as
the elder of two.
Whether the group's behavior is dealt with as antecedent and the
individual's as consequent, or vice versa, the two kinds of events are
interdependent. To
demand a combining theoretical framework is one thing, but to get it from a psychologist is quite
another. In spite of their long
prepossession with social influences on the individual, psychologists think monadically. That is, they choose the behavior of one person as their
scientific subject matter. For them, the
universe is composed of individuals.
These individuals are acted upon by external events, to be sure,
and in turn the external world is modified by the individuals'
behaviors. But the universal laws sought by the
psychologist almost always relate to a single
body. They are monadic laws, and they are stated with reference to a
monadic unit of behavior. The
main variables that compose such systems have been presented diagrammatically
in many ways. Some are so well known
that they represent, virtually, signatures for the theorists who devised them. There are Tolman's schematic sow-bug, Hull's
behavior sequence, Lewin's field structure, and Miller and Dollard's learning
paradigm. These diagrams differ
considerably in the kinds of variables they incorporate. Some emphasize reward and reinforcement;
others do not. Some are time-oriented; others are descriptive of a
nontemporal force field. All specify antecedent stimulus conditions and consequent
actions, but in very different ways and with quite different systematic
constructs. But there is one thing in
common among them - they are all monadic. But if
personality and social behavior are to be included in a single theory, the
basic monadic unit of behavior must be expandable into a dyadic one. A dyadic unit
is one that describes the combined actions of two or more
persons.3 A dyadic
unit is essential if there is to be any conceptualization of the relationships between people, as in the
parent-child, teacher-pupil, husband-wife, or leader-follower instances. To have a science of interactive events, one
must have variables and units of action that refer to such events. While it is possible to systematize some
observations about individuals by using monadic units, the fact is that a large proportion of the properties of a person that
compose his personality are originally formed in dyadic situations
and are measurable only by reference to
dyadic situations or symbolic representations of them. Thus, even a monadic description of a
person's action makes use of dyadic variables in the form of social stimuli. 3 Although the
prefix means "two," the term is used here simply as the minimal
instance of multiplicity. Similar
principles would hold whether the interactors were two or more. 470 The Theory and Its Application This is
exemplified in Fig. 1, a
diagram of a monadic behavior sequence that, as will be seen, can be expanded into a dyadic sequence. One aspect of this figure deserves comment,
the "environmental event."
This concept refers to the changes produced in the environment by the
instrumental activity; these are the changes necessary for the occurrence of the
goal response. The teacher trying to
increase participatory activity in a class of children, for example, gets her
reward when the youngsters spontaneously start a team game at recess. She makes her goal response - she has
achieved her aim – when the environment
changes, that is, when the children play a team game. Or a boy is seeking approbation from his
father; he hits a three-bagger; his father grins with satisfaction. The grin is the boy's environmental event in
his monadically conceived action sequence.
Fig. 1. THE MONADIC INSTIGATION-ACTION SEQUENCE Social Behavior and Personality Development 471 This
concept achieves importance in the present context, because it is the necessary
connecting link between a monadic and dyadic systematization of behavior. The framework for such a description is
shown in Fig. 2. For convenience the two persons are labeled Alpha and Beta. A dyadic situation exists whenever the actions of Beta are, or produce, the environmental
events for Alpha, and vice versa.
The behavior of each person is essential to the other's
successful completion of his goal-directed sequence of action. The drives of each are satisfied only when
the motivated actions of the other are carried through to completion. The nurturant mother is satisfied by the
fully loved child's expression of satiety, and the child is satisfied by the
expressions of nurturance given by his mother.
Fig. 2. THE DYADIC SEQUENCE It must
be made clear in this connection that "environmental
events" are only those changes
in environment produced by the behavior of the person under consideration. The stroke of lightning that splits a log
for the tired woodcutter is not in this
category, nor is the food given the newborn
infant by his mother, nor the empty taxi
that providentially appears when the rain is hardest. These are certainly characteristics of the environment, manipulanda that govern in some ways the
future behavior of Alpha, but they are not environmental events. They were not
induced by any action of Alpha. Fig. 3. THE DYADIC SEQUENCE WITH ANTICIPATORY
RESPONSES O THE ENVIRONMENTAL EVENT This is
an important distinction. Unless the
interaction of Alpha and beta is based on something other than the fortuitously
useful conjunction of their individual actions, there is no interdependence of
each on the other. There is, in effect,
no dyadic
system, only a piling up of parallel monadic sequences. The
factor responsible for maintaining stability of the dyadic unit is exhibited in
Fig. 3.
It is the expectancy of
the environmental event, diagrammed in a notation similar to that used by Hull
for the anticipatory goal response [6]. The existence of such anticipatory responses can be derived from
the monadic principles of learning. Alpha's actions, whether
instrumental or goal, that involve manipulation of the environmental events
produced by Beta's behavior move forward in Alpha's sequence in the form of reduced or symbolic responses which,
in turn, instigate response-produced cues. These are the expectancies
of Beta's supportive behavior, and this is
the mechanism by which a dyadic behavior unit can be derived from the combining
of two or more monadic units. The
development of this part of the behavior theory is perhaps a task for Sociologists.
Cottrell [3] and Parsons [10] have given attention to the matter, and the next
step appears to be the selection of appropriate variables. 472 The Theory and Its Application The
assertion has been made that any useful theory must be a theory of action. By definition, then, it will be dynamic that is, having to do with force or energy
in motion. The term dynamic
has been so abused by psychologists
during the last half century, however, that its meaning is no longer clear. Perhaps it never was. But with successive "dynamic
psychologies" - those of Freud, Morton Prince, Woodworth, Lewin, and a
host of contemporary theorists - the meaning has been more obfuscated than
ever. Sometimes it refers to a
motivational approach, sometimes to a developmental, sometimes to an emphasis
on unconscious processes. Mostly, I
suspect, it merely means that the theorist is revolting against what seem to
him the stultifying structuralistic unhuman inadequacies of his
predecessors. It boils down to a
self-attributed accolade for virtue, a promise to deal with important
characteristics of real live people rather than dry and dusty processes. This
is a waste of a good word. By no means
all modern psychological systems are dynamic; some are trait-based and some are
need-based. No one would deny that
combinations of habit structures do exist and do provide a kind of integrated
consistency in a person's behavior.
Likewise, no one would attempt to order the events of human action without
variables that relate to motivation, including those kinds that cannot be
verbally reported by the person himself.
But there is more to dynamics than motivation. There is change. Changes
in behavior are of two kinds.
For a theory to be dynamic, both must be systematized, separately
but congruently. One is on-going action
and the other is learning. In Fig. 1 the sequence of events
beginning with the instigators Sext, Sp,
and Scog,
and ending with the goal response Rg, is a
single behavioral event. In other words, both
the external factors and the internal ones
(the potentialities of the person) that initiate action are indicated. The
diagram describes such a predictive statement as this: that everything else
being equal (i.e., nothing else contributing to the variance), a hungry man who
sees a refrigerator, knows there is food in it, and knows how to get at it,
will eat if the refrigerator door opens when he manipulates it. Principles that
relate antecedent motivational factors to subsequent behavior are
dealing with on-going action; they are statements about the
resolution of field forces.
"Frustration produces aggression" is an example, albeit one
which is sometimes hard to swallow because in real life there are always so
many other variables besides frustration that contribute to response
variance. Social Behavior and Personality Development 473 Obviously,
however, no predictive statement can be made about on-going action unless
certain things are known about the person's potentialities
/or action. Action does
not take place with an organism containing a psychological vacuum. The person has certain properties that
determine what kind of be- havior he will produce under any given set of circumstances.
His motivation is weak or strong, he is frustrated or not in various
goal-directed sequences, he has expectancies of the consequences of his
behavior. Unless these various
properties of the person are known, it is impossible to have any
systematization of on-going action. And
unless the changes in potentialities
for action are systematically ordered, there is no possibility of constructing
an on-going action theory that will enable one to predict beyond the
termination of any single sequence of behavior. In Fig. 1, the various potentialities for action are
specified by Sp
(motivation) and Seog (cognitive structures). In large part these characteristics are a
product of learning. The successful
completion of a behavior sequence is a reinforcement, and this modifies the
drives and habit structures of the person in certain lawful ways, these laws
being part of the body of the laws of learning. In other words, there is a
change in the person's potentialities for action. It is to be noted, therefore,
that although Fig. 1 describes a single behavior sequence, there are two ways
of ordering the events that compose it.
Both refer to changes, to energy in motion. To be dynamic, a theory of behavior must encompass both.4 PERSONALITY AS ANTECEDENT In
this framework, personality is a description of those properties of a person
that specify his potentialities for action.
Such a description must include reference to motivation, expectations,
habit structure, the nature of the instigators that activate instrumental behavior,
and the kinds of environmental events that such actions will produce. Furthermore, all these factors must be
described in terms of the dyadic aspects of the behavior that occurs. That is, the kinds of Betas who can serve as
instigators for particular responses must be specified, and the environmental
events that Beta creates for Alpha must be described not only as they fit into
Alpha's activity but also as they fit into the whole motivational sequence of
Beta. 4
The most elaborate theory of on-going action is that of Kurt Lewin [7], but
his field theory has never been developed to care adequately for problems of
personality development (learning).
Similarly, the developmental theory of G. V. Hamilton [4] gave an
excellent account of the changes in potentiality for response but did not cover
so effectively the problems of on-going action. 474 The Theory and Its Application This
will give an adequate description
of a personality, but it is not sufficient for a theory
of personality. For this all these
factors must be treated as part of an antecedent-consequent proposition. Sheer description of the properties of an
object is of little value, either scientifically or practically, since the
ultimate aim of any theory is to provide lawful predictions of those events
that form the subject matter of the theory.
This can be done only when "if x, then y" principles are added
to description. Personality theory is
adequate only if it predicts behavior. In
behavior science, personality must be treated as both antecedent and
consequent. As antecedent, it is part of the total matrix that must be
known in order to account for either individual or dyadic action. In recent years various approaches to
personality have too much depended on assumptions of fixed traits and fixed
needs. This has led to measurement
procedures that do not include reference to the social stimulus conditions
under which the traits or needs will be expressed. As Sanford has said, in connection with a study of leadership, there
is no trait independent of the conditions that elicit it. Leadership is a quality in a person's behavior only if there are followers who react to
him as a leader. Most behavior with
which the personality psychologist is concerned is either directly dyadic or is
in response to symbolic representation of the dyad. Therefore, any conceptualization of the person's properties
must be done with consideration of the properties of the various Betas
with whom Alpha is interactive. A
simple example of the measurement problem created by these considerations arose
in connection with some data on aggressive behavior analyzed in the Harvard
Laboratory of Human Development [5, 12].
Forty preschool children were the subjects. Two main measures of aggressiveness were secured. One was overt and socially directed
aggression. This measure was obtained
both by teachers' rating scales and by direct observation. The other was projective or fantasy
aggression revealed in doll play. By a
fixed trait or need assumption, one would expect these two measures to
correspond somewhat. They did -
somewhat! The correlation was
+0.13! But
further analysis makes the meaning of this relationship clear. These children's mothers were interviewed
concerning their methods of handling the youngsters' aggression at home. On the basis of this information it was
possible to divide the children into three subgroups which had had
different degrees of severity of punishment for aggression. Social Behavior and Personality Development 475 In
Fig. 4 the frequency of both overt and fantasy aggression are shown for these
three subgroups. It is to be noted that
while the "mild" and "moderate" groups show a mean
correspondence in amount of aggressive behavior of the two kinds, there is a
radical disagreement in the "severe" punishment subgroup. These latter children, on the average,
behaved rather nonaggressively in preschool, but in their doll-play fantasies
there was an abundance of aggression.
One could ask whether these children are very aggressive or very
nonaggressive. Do they have strong need for aggression or weak? Interpersonal aggression is measured by frequency of
aggressive acts occurring during four hours of observation in preschool;
fantasy aggression is measured by mean frequency of aggressive acts occurring
during two twenty-minute doll-ploy sessions.
Punitiveness of mother is based on ratings of interview material
concerning severity of mother', punishment of child's aggressive acts at home.
Charted values are medians of the three groups, sizes of which are: Low = 7,
Medium = 23, High = 10. 4. RELATION OF
INTERPERSONAL AND FANTASY AGGRESSION TO MATERNAL PUNITIVENESS Even
if these questions could be sensibly answered, which they cannot, the answers
would be of little help in predicting the future aggressive behavior of these
children. To accomplish the latter, which is our aim, there must be an analysis
of the social stimulus conditions under which the future behavior is to be
observed. The
minimum specification would concern whether the behavior would be observed in a
nursery school or in a permissive doll play experiment. With a conceptualization of the dyadic
variables involved, however, it is possible to make a statement that goes
beyond the narrow confines of these two measuring situations. In this
instance, the more general statements can be made that, first, the
amount of aggression will be a negative function of severity of punishment; and
second, with severity of punishment held constant, the amount of
aggression will vary positively with dissimilarity of the dyadic situation to
the original punishment situation in the home.
476 The Theory and Its Application PERSONALITY DEVELOPMENT The
systematization of personality development requires a different approach. When personality
factors are considered as antecedents to individual or group
behavior, the laws of on-going action
are involved. But when personality development is the matter for
study, the laws of learning are the bases. What is needed in this case is a set of principles that
will describe the way in which the child's potentialities for action - that is,
his drives, habits, cognitive structures, and expectancies - are changed by the
experiences he has throughout his life.
This
is a difficult problem, both logically and empirically. Personality is partly the product of a
lifetime of dyadic action which has modified the individual's potentiality for
further action. The changed
potentiality is therefore partly a product of his own actions. For example, in the data concerning child aggression and severity of maternal punishment for
aggression, the mother's actions in punishing the child were doubtless
influenced in part by the amount and kind of aggression exhibited toward her by
the child. Thus, the dyadic behavior
that served as an antecedent to the differential display of overt and fantasy
aggression by the child was contributed to by the child himself. Logically,
and practically, a good theory requires that antecedents and consequents be
entirely independent of one another. It would be most satisfactory if the
child did not influence the mother's behavior, and if we
could then say something about the effect of severity of punishment on later
behavior. One solution to this problem appears to be a careful
measurement of the child's contribution to the dyadic relationship and a
partialling out of that influence in the comparison of antecedent mother
behavior with consequent child behavior.
If this
procedure does not prove feasible, as it may very well not, a developmental
theory can still be constructed in which the
antecedent variables are specified changes in the mother's contributions
to the dyadic mother-child interaction.
Such a theory would be more defensible logically, for it would be taking
formal account of the dyadic nature of the learning situation. Empirically, however, it would be
considerably more difficult. For the
partialling-out method, naturalistic data are appropriate; natural variation in
child-rearing methods, as this is found in any group of mothers, can be used as
the antecedent. But if we are forced to
use specified changes in maternal behavior as the antecedent, the research task
will be complicated not only by the necessity of securing families in which
such changes can be made, but by the long wait
from early life, when the changes begin to be introduced, to later childhood,
when the personality consequents are to be measured. Social Behavior and Personality Development 477 In any
case, it is clear that an effective approach to the problems of the development
of personality and of the influence of personality on the behavior of groups
requires a theory that has the following properties: its basic reference events
must be actions: it must combine congruently both dyadic and monadic events; it must account for
both on-going action and learning; it must provide a description of
personality couched in terms of potentiality for
action; and it must provide principles of personality
development in terms of changes in potentiality for action. To
spell out in detail the specific variables
that must be defined for use in this theory is beyond the scope of the present
paper. There are two general
bodies of concepts and their relating principles, however, that appear
promising. One of these is the
set of definitions and postulates that compose the laws of learning.
Whether the particular formulations used by Tolman, Hull, Guthrie, or
Skinner are selected seems of little importance at the moment. Those of Hull and Tolman have certain a
priori advantages, but the main point is the use of whatever laws of learning
will best serve to account for changes in potentiality for action. The theoretical formulation of the research
in our own laboratory stems from Hull through Miller and Dollard. The second
set of defined variables contains conceptualizations of those secondary
motivational systems that arise universallyy as a product of the dyadic
relationship between mother and child [11].
These include aggression, dependency, self-reliance, anxieties,
competition, and status-seeking, as well as the various consequences of the training inherent in the
socialization of the primary drives of hunger, sex, and elimination. The exact forms of behavior potentiality
created in each of these motivational areas are different from child to child
and from culture to culture. But the biological nature of man, coupled with his
universal gregariousness, gives rise
to various learning experiences that every child endures in one fashion or another. BIBLIOGRAPHY 1. Allport,
F. H. Social Psychology. Boston: Houghton, Mifflin, 1924. 2. AlIport,
G. W. Personality: A Psychological Interpretation. New York: Holt,
1937. 3. Cottrell,
L. S. "The Analysis of Situational Fields in Social Psychology," Amer. Sociol. Rev., VII (1942), 37~382. 4. Hamilton,
G. V. Objective Psychopathology. St. Louis: C. V. Mosby, 1925. 5. Hollenberg,
Eleanor, and Sperry, Margaret. "Some Antecedents of Ag. gression and Effects of Frustration in Doll Play,"
Personality, I (1951), 32A3. 478 The
Theory and Its Application 6. Hull, C.
L. "Goal Attraction and Directing Ideas Conceived as Habit Phenomena," Psychol. Rev., XXXVIII (1931), 487-506. 7. Lewin,
Kurt. A Dynamic Theory ol Personality. New York: McGraw-Hill, 1935. 8. McDougall,
William. An Introduction to Social Psychology. London: Methuen, 1908. 9. Murray,
H. A. Explorations in Personahty. New York: Oxford University Press, 1938. 10. Parsons,
Talcott. The Social System. Glencoe, Illinois: The Free Press, 1951. 11. Sears, R.
R. "Personality Development in Contemporary Culture," Proc. Amer. Philos. Soc., XCII (1948), 363-370. 12. Sears, H.
H. "Relation of Fantasy Aggression to Interpersonal Aggres. sion," Child Development, XXI (1950) 54.
ideological furnishings for the homeless
mind
daurril
library: talcott parsons
ACTION
DYNAMICS