150.19 H5M 63-02157
Henle
Documents of Gestalt. psychology
DOT 2 7 1978
JOL"2
$
1
DOCUMENTS
OF
GESTALT
PSYCHOLOGY
DOCUMENTS
OF
GESTALT
PSYCHOLOGY
EDITED BY
Mary Henle
UNIVERSITY OF CALIFORNIA PRESS
Berkeley and Los Angeles : 1961
University of California Press
Berkeley and Los Angeles
California
Cambridge University Press
London, England
<g> 1961 by
The Regents of the University of California
Library of Congress Catalog Card Number: 61-14554
To Wolfgang Kohler
PREFACE
At the 1959 convention of the American Psychological Association,
Wolfgang Kohler delivered a presidential address whose text serves as
an introduction to the present volume. In his brief survey of "Gestalt
Psychology Today," Professor Kohler, one of the founders of this move-
ment, indicated certain differences in outlook and assumptions between
Gestalt psychology and some other trends in American psychology. He
also expressed confidence in the possibility of fruitful collaboration by
representatives of the various approaches. Given this situation, it seems
opportune to make available in a single volume a number of recent
papers in which the principles and methods of Gestalt psychology are
clearly expressed by a group of authors for whom this approach has
been central.
The present volume is intended to supplement the fundamental texts
of Gestalt psychology, for example, those by Kohler, Wertheimer, and
Koffka. A forerunner of the present collection is A Source Book of
Gestalt Psychology, edited by Willis D. Ellis, which contains abbrevi-
ated translations into English of a number of the early theoretical and
experimental publications of Gestalt psychologists. The present book
brings together more recent contributions by the founders of Gestalt
psychology and by some of their immediate collaborators and students.
With the exception of the essays by Wertheimer, which are not else-
where easily accessible to the psychologist, most of these papers have
been published since 1950. They thus illustrate the recent thinking and
viii : Preface
findings of the authors. They include work in the fields of social psy-
chology, motivation, and art, in addition to new developments in the
exploration of the cognitive processes, for which Gestalt psychology
used to be best known.
The concepts of Gestalt psychology have been central for some work-
ers in the field, have markedly influenced others, and have slightly
tinged the work of hundreds more. Limitations of space have dictated
the omission from the present volume of authors whose inclusion would
have given a more rounded picture of the scope and influence of con-
temporary Gestalt psychology. Perhaps the most conspicuous omission
is that of Kurt Lewin, whose work developed out of Gestalt psychology
and whose approach and assumptions remained in many respects similar
to those of Gestalt psychologists working in other fields. Lewin's recent
papers are excluded because they have been collected elsewhere and are
readily available in book form. For the same reason the work of Fritz
Heider and of Lewin's students is not included.
In a sense this has become a volume of work from the New School for
Social Research. As the reader will notice, all the senior authors whose
work is included have had some connection with the New School,
whose graduate psychology department was founded by Wertheimer in
1933 as part of the University in Exile. It must be mentioned, however,
that the New School is presently the principal academic connection of
only one of the authors, and that this volume by no means represents
the varied points of view of all the psychologists at this institution. Still,
the New School has in some sense provided a center for Gestalt psy-
chology in America. This is, of course, not the first instance in the his-
tory of psychology in which one or a few universities have provided a
home for a particular approach to our science.
This book was edited while the editor was a Fellow of the John
Simon Guggenheim Memorial Foundation. It is a pleasure to call
attention to an additional debt this book owes to the Foundation.
Four of the studies in this collection were made while their authors were
Guggenheim Fellows.
MARY HENLE
CONTENTS
GESTALT PSYCHOLOGY TODAY
Wolfgang Kohler
PART I. ESSAYS BY MAX WERTHEIMER
ON TRUTH 19
Max Wertheimer
SOME PROBLEMS IN THE THEORY OF ETHICS 29
Max Wertheimer
ON THE CONCEPT OF DEMOCRACY 42
Max Wertheimer
A STORY OF THREE DAYS 52
Max Wertheimer
x : Contents
PART II. GENERAL THEORY
PSYCHOLOGY AND EVOLUTION 67
Wolfgang Kohler
SOME PROBLEMS OF ECLECTICISM 76
Mary Henle
GESTALTEN YESTERDAY AND TODAY 90
Rudolf Arriheim
THE PRESENT SITUATION IN BRAIN PHYSIOLOGY 97
Wolfgang Kohler
PART III. COGNITIVE PROCESSES
BRIGHTNESS CONSTANCY AND THE NATURE OF
ACHROMATIC COLORS 109
Hans Wallach
THE KINETIC DEPTH EFFECT 126
Hans Wallach and D. N. O'Connell
PERCEPTION AND ATTENTION 146
Wolfgang Kohler and Pauline Austin Adams
SOME CONSIDERATIONS CONCERNING THE RELATION
BETWEEN PERCEPTION AND COGNITION 164
Hans Wallach
SOME EFFECTS OF MOTIVATIONAL PROCESSES
ON COGNITION 172
Mary Henle
Contents : xi
PART IV. SOCIAL PSYCHOLOGY AND MOTIVATION
PSYCHOLOGICAL REMARKS ON SOME QUESTIONS OF
ANTHROPOLOGY 203
Wolfgang Kohler
EFFECTS OF GROUP PRESSURE UPON THE
MODIFICATION AND DISTORTION OF JUDGMENTS 222
Solomon E. Asch
FORMING IMPRESSIONS OF PERSONALITY 237
Solomon E. Asch
ON FIELD FORCES 286
Mary Henle
PART V. PSYCHOLOGY OF EXPRESSION AND ART
THE GESTALT THEORY OF EXPRESSION 301
Rudolf Arnheim
THE METAPHOR: A PSYCHOLOGICAL INQUIRY 324
Solomon E. Asch
EMOTION AND FEELING IN PSYCHOLOGY AND ART 334
Rudolf Arnheim
Wolfgang Kohler
DARTMOUTH COLLEGE
GESTALT
PSYCHOLOGY
TODAY
In 1949, the late Herbert Langfeld gave a lecture in Europe in which
he described what appeared to him to be the major trends in American
psychology. He also mentioned Gestalt psychology; but he added that
the main observations, questions, and principles characteristic of this
school had become part of every American psychologist's mental equip-
ment. I was not so optimistic. And, in fact, the very next year attempts
were made to explain the molar units in perception by processes which
gradually connect neural elements. Soon afterwards, a theory of con-
ditioning was developed, according to which more and more com-
ponents of a stimulus object are gradually conditioned, and the course
of the whole process can be explained in this fashion. Such theories may
prove to be very useful, but one can hardly say that, at the time, thek
authors were greatly influenced by Gestalt psychology. It is for this and
similar reasons that a new discussion of old questions seems to me
indicated.
Reprinted with permission from the American Psychologist, Vol. 14, No.
12, December, 1959. This selection was the Address of the President of the
American Psychological Association at its sixty-seventh annual convention in
Cincinnati, Ohio, September 6, 1959.
2 : Wolfgang Kohler
I should like to begin with a few remarks about the history of Gestalt
psychology because not all chapters of this history are generally
known. In the eighties of the past century, psychologists in Europe were
greatly disturbed by von Ehrenfels' claim that thousands of percepts
have characteristics which cannot be derived from the characteristics of
their ultimate components, the so-called sensations. Chords and melo-
dies in hearing, the shape characteristics of visual objects, the rough-
ness or the smoothness of tactual impressions, and so forth, were used
as examples. All these "Gestalt qualities" have one thing in common.
When the physical stimuli in question are considerably changed, while
their relations are kept constant, the Gestalt qualities remain about the
same. But, at the time, it was generally assumed that the sensations in-
volvecf are individually determined by their individual stimuli and must
therefore change when these are greatly changed. How, then, could any
characteristics of the perceptual situation remain constant under these
conditions? Where did the Gestalt qualities come from? Ehrenfels'
qualities are not fancy ingredients of this or that particular situation
which we might safely ignore. Both positive and negative aesthetic
characteristics of the world around us, not only of ornaments, paintings,
sculptures, tunes, and so forth, but also of trees, landscapes, houses,
cars and other persons belong to this class. That relations between
the sexes largely depend on specimens of the same class needs hardly
be emphasized. It is, therefore, not safe to deal with problems of psy-
chology as though there were no such qualities. And yet, beginning
with Ehrenfels himself, psychologists have not been able to explain
their nature.
This holds also for the men who were later called Gestalt psycholo-
gists, including the present speaker. Wertheimer's ideas and investiga-
tions developed in a different direction. His thinking was also more radi-
cal than that of Ehrenfels. He did not ask: How are Gestalt qualities
possible when, basically, the perceptual scene consists of separate ele-
ments? Rather, he objected to this premise, the thesis that the psycholo-
gist's thinking must begin with a consideration of such elements. From
a subjective point of view, he felt, it may be tempting to assume that all
perceptual situations consist of independent, very small components.
For, on this assumption, we obtain a maximally clear picture of what
lies behind the observed facts. But, how do we know that a subjective
clarity of this kind agrees with the nature of what we have before us?
Perhaps we pay for the subjective clearness of the customary picture
Gestalt Psychology Today : 3
operated before there is a perceptual scene and which thus influence the
characteristics of this scene. Are we allowed to impose on perception
an extreme simplicity which, objectively, it may not possess?
Wertheimer, we remember, began 'to reason in this fashion when ex-
perimenting not with perceptual situations which were stationary, and
therefore comparatively silent, but with visual objects in motion when
corresponding stimuli did not move. Such "apparent movements," we
would now say, occur when several visual objects appear or disappear
in certain temporal relations. Again in our present language, under
these circumstances an interaction takes place which, for instance,
makes a second object appear too near, or coincident with, a first ob-
ject which is just disappearing, so that only when the first object, and
therefore the interaction, really fades, the second object can move to-
ward its normal position. If this is interaction, it does not, as such, occur
on the perceptual scene. On this scene, we merely observe a movement.
That movements of this kind do not correspond to real movements of
the stimulus objects and must therefore be brought about by the se-
quence of the two objects, we can discover only by examining the physi-
cal situation. It follows that, if the seen movement is the perceptual
result of an interaction, this interaction itself takes place outside the
perceptual field. Thus, the apparent movement confirmed Wertheimer's
more general suspicion: we cannot assume that the perceptual scene is
an aggregate of unrelated elements because underlying processes are
already functionally interrelated when that scene emerges and now
exhibits corresponding effects.
Wertheimer did not offer a more specific physiological explanation.
At the time, this would have been impossible. He next turned to the
problem of whether the characteristics of stationary perceptual fields
are also influenced by interactions. I need not repeat how he investigated
the formation of molar perceptual units, and more particularly of
groups of such objects. Patterns which he used for this purpose are now
reproduced in many textbooks. They clearly demonstrate that it is
relations among visual objects which decide what objects become group
members, and what others do not, and where, therefore, one group
separates itself from another. This fact strongly suggests that perceptual
groups are established by interactions; and, since a naive observer is
merely aware of the result, the perceived groups, but not of their de-
pendence upon particular relations, such interactions would again occur
among the underlying processes rather than within the perceptual field.
Let me add a further remark about this early stage of the develop-
4 : Wolfgang Kohler
ment. Surely, in those years, Gestalt psychologists were not satisfied with
a quiet consideration of available facts. It seems that no major new
trend in a science ever is. We were excited by what we found, and even
more by the prospect of finding further revealing facts. Moreover, it
was not only the stimulating newness of our enterprise which inspired
us. There was also a great wave of relief as though we were escaping
from a prison. The prison was psychology as taught at the universities
when we still were students. At the time, we had been shocked by the
thesis that all psychological facts (not only those in perception) consist
of unrelated inert atoms and that almost the only factors which com-
bine these atoms and thus introduce action are associations formed
under the influence of mere contiguity. What had disturbed us was the
utter senselessness of this picture, and the implication that human life,
apparently so colorful and so intensely dynamic, is actually a frightful
bore. This was not true of our new picture, and we felt that further dis-
coveries were bound to destroy what was left of the old picture. Soon
further investigations, not all of them done by Gestalt psychologists,
reinforced the new trend. Rubin called attention to the difference be-
tween figure and ground. David Katz found ample evidence for the role
of Gestalt factors in the field of touch as well as in color vision, and so
forth. Why so much interest just in perception? Simply because in no
other part of psychology are facts so readily accessible to observation.
It was the hope of everybody that, once some major functional princi-
ples had been revealed in this part of psychology, similar principles
would prove to be relevant to other parts, such as memory, learning,
thinking and motivation. In fact, Wertheimer and I undertook our early
studies of intellectual processes precisely from this point of view; some-
what later, Kurt Lewin began his investigations of motivation which, in
part, followed the same line; and we also applied the concept of Ges-
taltung or organization to memory, to learning, and to recall. With
developments in America, Wertheimer's further analysis of thinking,
Asch's and Heider's investigations in social psychology, our work on
figural aftereffects, and eventually on currents of the brain, we are
probably all familiar.
In the meantime, unexpected support had come from natural science.
To mention only one point: Parts of molar perceptual units often have
characteristics which they do not exhibit when separated from those
units. Within a larger visual entity, a part may, for instance, be a corner
of this entity, another part its contour or boundary, and so on. It now
seems obvious; but nobody in psychology had seen it before: the same
Gestalt Psychology Today : 5
happens in any physical system that is pervaded by interactions. These
interactions affect the parts of the system until, eventually, in a steady
state, the characteristics of all parts are such that remaining interactions
balance one another. Hence, if processes in the central nervous system
follow the same rule, the dependence of local perceptual facts on con-
ditions in larger entities could no longer be regarded as puzzling. Com-
parisons of this kind greatly encouraged the Gestalt psychologists.
In America, it may seem surprising that enthusiastic people such as
the Gestalt psychologists were intensely interested in physics. Physics
is generally assumed to be a particularly sober discipline. And yet, this
happened to us most naturally. To be sure, our reasoning in physics in-
volved no changes in the laws of physics, and no new assumptions in
this field. Nevertheless, when we compared our psychological findings
with the behavior of certain physical systems, some parts of natural
science began to look different. When reading the formulae of the physi-
cist, one may emphasize this or that aspect of their content. The particu-
lar aspect of the formulae in which the Gestalt psychologists became
interested had, for decades, been given little attention. No mistake had
ever been made in applications of the formulae, because what now
fascinated us had all the time been present in their mathematical form.
Hence, all calculations in physics had come out right. But it does make
a difference whether you make explicit what a formula implies or merely
use it as a reliable tool. We had, therefore, good reasons for being sur-
prised by what we found; and we naturally felt elated when the new
reading of the formulae told us that organization is as obvious in some
parts of physics as it is in psychology.
Incidentally, others were no less interested in this "new reading" than
we were. These other people were eminent physicists. Max Planck once
told me that he expected our approach to clarify a difficult issue which
had just arisen in quantum physics if not the concept of the quantum
itself. Several years later, Max Born, the great physicist who gave
quantum mechanics its present form, made almost the same statement
in one of his papers. And, only a few weeks ago, I read a paper in which
Bridgman of Harvard interprets Heisenberg's famous principle in such
terms that I am tempted to call him, Bridgman, a Gestalt physicist.
We will now return to psychology. More particularly, we will inspect
the situation in which American psychology finds itself today. The
spirit which we find here differs considerably from the one which charac-
terized young Gestalt psychology. Let me try to formulate what mem-
bers of this audience may have been thinking while I described that
6 : Wolfgang Kohler
European enterprise. "Enthusiasm?" they probably thought. "Feelings
of relief when certain assumptions were found less dreary than those of
earlier psychologists in Europe? But this is an admission that emotional
factors and extrascientific values played a part in Gestalt psychology.
We know about the often pernicious effects of the emotions in ordinary
life. How, then, could emotions be permitted to influence scientific judg-
ments and thus to disturb the objectivity of research? As we see it, the
true spirit of science is a critical spirit. Our main obligation as scientists
is that of avoiding mistakes. Hence our emphasis on strict method in
experimentation and on equally strict procedures in the evaluation of
results. The Gestalt psychologists seem to have been guilty of wishful
thinking. Under the circumstances, were not some of their findings un-
reliable and some of their concepts vague?"
I will at once admit two facts. Almost from its beginning, American
psychology has given more attention to questions of method and strict
proof than Gestalt psychology did in those years. In this respect, Ameri-
can psychology was clearly superior. Secondly, sometimes the Gestalt
psychologists did make mistakes. Not in all cases was the reliability of
their findings up to American standards, and some concepts which they
used were not immediately quite clear. I myself once used a certain con-
cept in a somewhat misleading fashion. I had better explain this.
What is insight? In its strict sense, the term refers to the fact that,
when we are aware of a relation, of any relation, this relation is not
experienced as a fact by itself, but rather as something that follows from
the characteristics of the objects under consideration. Now, when pri-
mates try to solve a problem, their behavior often shows that they are
aware of a certain important relation. But when they now make use of
this "insight," and thus solve their problem, should this achievement be
called a solution by insight? No it is by no means clear that it was also
insight which made that particular relation emerge. In a given situation,
we or a monkey may become aware of a great many relations. If, at a
certain moment, we or a monkey attend to the right one, this may hap-
pen for several reasons, some entirely unrelated to insight. Conse-
quently, it is misleading to call the whole process a "solution by insight."
This will be particularly obvious when the solution of the problem is
arbitrarily chosen by the experimenter. Take Harlow's excellent experi-
ments in which primates are expected to choose the odd item in a group
of objects. "Oddity" is a particular relational fact. Once a monkey at-
tends to it, he will perceive it with insight. But why should he do so
during his first trials? His first choices will be determined by one factor
Gestalt Psychology Today : 7
or another, until he happens to attend, once or repeatedly, to the oddity
relation just when he chooses (or does not choose) the right object.
Gradually, he will now attend to this particular relation in all trials; and
he may do so even when entirely new objects are shown. Surely, such a
process should not simply be called "learning by insight." If Harlow
were to say that, under the circumstances, it is learning of one kind or
another which gives the right relation and corresponding insight their
chance to operate, I should at once agree. What, I believe, the monkeys
do not learn is insight into which object in a given group is the odd one;
but they must learn to pay attention to the oddity factor in the first
place. I hope that this will clarify matters. They have not always been
so clear to me.
When the solution of a problem is not arbitrarily chosen by the ex-
perimenter, but more directly related to the nature of the given situation,
insight may play a more important role. But, even under these circum-
stances, it is not insight alone which brings about the solution. The
mere fact that solutions often emerge to the subjects' own surprise is
clear proof that it cannot be insight alone which is responsible for their
origin.
But I intended to discuss some trends in American psychology. May
I confess that I do not fully approve of all these trends?
First, I doubt whether it is advisable to regard caution and a critical
spirit as the virtues of a scientist, as though little else counted. They are
necessary in research, just as the brakes in our cars must be kept in
order and their windshields clean. But it is not because of the brakes or
of the windshields that we drive. Similarly, caution and a critical spirit
are like tools. They ought to be kept ready during a scientific enterprise;
however, the main business of a science is gaining more and more new
knowledge. I wonder why great men in physics do not call caution and
a critical spirit the most important characteristics of their behavior.
They seem to regard the testing of brakes and the cleaning of wind-
shields as mere precautions, but to look forward to the next trip as the
business for which they have cars. Why is it only in psychology that we
hear the slightly discouraging story of mere caution over and over
again? Why are just psychologists so inclined to greet the announcement
of a new fact (or a new working hypothesis) almost with scorn? This
is caution that has gone sour and has almost become negativism
which, of course, is no less an emotional attitude than is enthusiasm.
The enthusiasm of the early Gestalt psychologists was a virtue, because
8 : Wolfgang Kohler
it led to new observations. But virtues, it has been said, tend to breed
little accompanying vices. In their enthusiasm, the Gestalt psychologists
were not always sufficiently careful.
In American psychology, it is rightly regarded as a virtue if a man
feels great respect for method and for caution. But, if this virtue be-
comes too strong, it may bring forth a spirit of skepticism and thus pre-
vent new work. Too many young psychologists, it seems to me, either
work only against something done by others or merely vary slightly
what others have done before; in other words, preoccupation with
method may tend to limit the range of our research. We are, of course,
after clear evidence. But not in all parts of psychology can evidence im-
mediately be clear. In some, we cannot yet use our most exact methods.
Where this happens, we hesitate to proceed. Experimentalists in particu-
lar tend to avoid work on new materials resistant to approved methods
and to the immediate application of perfectly clear concepts. But con-
cepts in a new field can only be clarified by work in this field. Should
we limit our studies to areas already familiar from previous research?
Obviously, this would mean a kind of conservatism in psychology.
When I was his student, Max Planck repeated this warning over and
over again in his lectures.
Our wish to use only perfect methods and clear concepts has led to
methodological behaviorism. Human experience in the phenomenologi-
cal sense cannot yet be treated with our most reliable methods; and,
when dealing with it, we may be forced to form new concepts which, at
first, will often be a bit vague. Most experimentalists, therefore, refrain
from observing, or even from referring to, the phenomenal scene. And
yet, this is the scene on which, so far as the actors are concerned, the
drama of ordinary human living is being played all the time. If we never
study this scene, but insist on methods and concepts developed in re-
search "from the outside," our results are likely to look strange to those
who intensely live "inside."
To be sure, in many respects, the graphs and tables obtained "from
the outside" constitute a most satisfactory material; and, in animal psy-
chology, we have no other material. But this material as such contains
no direct evidence as to the processes by which it is brought about. In
this respect it is a slightly defective, I am tempted to say, a meager,
material. For it owes its particular clearness to the fact that the data
from which the graphs and tables are derived are severely selected data.
When subjects are told to say no more than "louder," "softer," and per-
haps "equal" in certain experiments, or when we merely count how
Gestalt Psychology Today : 9
many items they recall in others, then we can surely apply precise sta-
tistical techniques to what they do. But, as a less attractive consequence,
we never hear under these circumstances how they do the comparing in
the first case, and what happens when they try to recall in the second
case.
Are such questions now to be ignored? After all, not all phenomenal
experiences are entirely vague; this Scheerer has rightly emphasized.
And, if many are not yet accessible to quantitative procedures, what
of it? One of the most fascinating disciplines, developmental physiology,
the science investigating the growth of an organism from one cell,
seldom uses quantitative techniques. And yet, nobody can deny that its
merely qualitative description of morphogenesis has extraordinary sci-
entific value. In new fields, not only quantitative data are relevant. As to
the initial vagueness of concepts in a new field, I should like to add an
historical remark. When the concept of energy was first introduced in
physics, it was far from being a clear concept. For decades, its meaning
could not be sharply distinguished from that of the term "force." And
what did the physicists do? They worked and worked on it, until at last
it did become perfectly clear. There is no other way of dealing with new,
and therefore not yet perfect, concepts. Hence, if we refuse to study the
phenomenal scene, because, here, few concepts are so far entirely clear,
we thereby decide that this scene will never be investigated at least
not by us, the psychologists.
Now, I had better return to Gestalt psychology. Let me try to show
you how Gestalt psychology tends to work today by discussing a more
specific issue, an issue on which scores of American psychologists have
worked for years. We shall thus be enabled to compare the way in which
they approach this issue with the Gestalt psychologists' approach.
The issue in question refers to the concepts of conditioning and mo-
tivation. One school seems to regard conditioning as almost the process
with which the psychologist has to deal. In a famous book with the
general title Principles of Behavior, the late Clark Hull, then the most
influential member of the school, actually dealt with little else al-
though he often used other terms. He felt that even such facts as think-
ing, insight, intentions, striving, and value would eventually be ex-
plained by a consistent investigation of the various forms of condition-
ing. We are all familiar with the basic concepts of his theory. Hence I
will say only a few words about it. When conditions in an animal's tissue
deviate from an optimal level, a state of need is said to exist in this
tissue. Such needs produce, or simply are, drives which means that
10 : Wolfgang Kohler
they tend to cause actions in the nervous system, some more or less pre-
scribed by inherited neural connections, others of a more random na-
ture. Drives are also called motivations. None of these terms is to be
understood in a phenomenological sense. They always refer to assumed
states of the tissue. The main point is that, for biological reasons, states
of need must, if possible, be reduced and that this may be achieved by
certain responses of the organism to the given situation. In case first
responses are of a random character, learning or conditioning will often
select such responses as do reduce the needs in question. In a simple
formulation, the well-known rule which governs such developments is
as follows: when a response has repeatedly occurred in temporal con-
tiguity with the neural effects of a certain stimulus, then this stimulus
will tend to evoke the same response in the future provided the re-
sponse has caused a reduction of the need. I will not define such further
concepts as habit strength, reaction potential, afferent stimulus inter-
action, reactive inhibition, and so forth, because they will play no role
in my discussion.
But one term seems to me particularly important. Many recent, and
important, investigations are concerned with so-called "learned drives,"
an expression which has, of course, this meaning: if a neutral stimulus
is repeatedly followed by conditions which cause a primary state of
drive such as pain, and the corresponding fear, then the fear with its
usual effects on behavior will gradually become connected with that
neutral stimulus, so that the stimulus alone now evokes the fear and
its overt consequences. Certain drives are therefore said to be "learn-
able" in the sense that they can be attached to facts which, as such, are
not related to the drive and hence would originally not evoke corre-
sponding responses.
Some experiments in the field of conditioning in general are most
interesting. I will only discuss the concepts used in the interpretation of
this work and the conclusions which it is said to justify.
To begin with these conclusions: They refer to certain human ex-
periences which, if the conclusions were justified, would have to be re-
garded as strange delusions. I mean our cognitive experiences. Suppose
somebody discovers by accident that, every time he subtracts the square
of a given integer from the square of the next integer in the series, the
result is an odd number. A more learned friend now explains to him
why this is a necessary rule, undoubtedly valid beyond any tests ever
done by a person. The explanation refers to simple relations and to rela-
tions among relations all readily understandable and the final out-
Gestalt Psychology Today : 11
come is convincing. Now, is the understanding of the relations involved
to be explained in terms of conditioning? Nothing in conditioning seems
to give us access to the psychological fact which I just called under-
standing; and, since an understanding of relations is essential to all
cognitive achievements, the same applies to the whole field.
Explanation of our intellectual life in terms of conditioning would
simply mean: its reduction to the operations of an often most practical,
but intrinsically blind, connection of mere facts. Promises that such an
explanation will nevertheless be achieved cause in the present speaker a
mild, incredulous horror. It is not the business of science to destroy evi-
dence. Behaviorists would perhaps answer that arguments which refer
to human thinking as an experience are irrelevant, because science is
only concerned with facts observable from the outside, and therefore
objective. This answer would hardly be acceptable. The behaviorist's
own objective observations are invariably observations of facts in his
perceptual field. No other form of objective observation has ever been
discovered. Consequently, the behaviorist cannot, without giving more
particular reasons, reject reference to other individual experiences
merely because they are such experiences.
Thus we are justified in considering a further example of human
experience. A need or drive, we are sometimes told, is a motivation. I
do not entirely agree with this statement for the following reasons. A
need or drive, we remember, is supposed to be a particular state in the
tissue. There is no indication in Hull's writings that such a state "points
beyond itself" toward any objects although it may, of course, cause
movements, or actions of glands. Now it is true that the same holds for
certain needs as human experiences; because, when a need is felt, it does
not always point toward an object, attainment of which would satisfy
the need. At the time, no such object may be in sight; in fact, no such
object may yet be known. But when the proper object appears, or be-
comes known, then the situation changes. For, now the subject feels
attracted or (in certain instances) repelled by this object. In other
words, an object may have characteristics which establish a dynamic
relation between the subject and that object. According to common
experience, it is this dynamic relation which makes the subject move
toward, or away from, the object. We ought to use different terms for a
mere need per se and the situation in which a subject is attracted or re-
pelled by an object. Otherwise, the dynamic aspect of the latter situation
might easily be ignored. I suggest that we reserve the term "motivation"
for this dynamic situation. Here we are, of course, on familiar ground.
12 : Wolfgang Kohler
Motivation as just described was Kurt Lewin's main concern in psy-
chology. He clearly recognized the part which certain characteristics of
an object play in establishing the dynamic relation between this object
and the subject. He called such characteristics of objects Aufforde-
rungscharaktere, a term which then became "valences" in English.
So far as I know, there are no valences in objects, no attractions and
no repulsions between objects and subjects in the behaviorist's vocabu-
lary. I am afraid that, in this fashion, he misses a point not only im-
portant in human experience but also relevant to what he regards as
true science.
How would a Gestalt psychologist handle motivation in the present
sense? He would begin with the following psychological facts. I do not
know up to what point Lewin would have accepted what I am now going
to say. My facts are these: (a) In human experience, motivation is a
dynamic vector, that is, a fact which has a direction and tends to cause
a displacement in this direction, (b) Unless there are obstacles in the
way, this direction coincides with an imaginary straight line drawn from
the object to the subject, (c) The direction of the experienced vector is
either that toward the object or away from it. In the first case, the vector
tends to reduce the distance in question; in the second, to increase it.
(d) The strength of both the need present in the subject and of the va-
lence exhibited by the object can vary. Both in man and in animals it
has been observed that, when the strength of the valence is low, this
reduction can be compensated for by an increase of the need in the
subject; and, conversely, that, when the need is lowered, an increase of
the strength of the valence may compensate for this change.
When considering these simple statements, anybody familiar with
the elements of physics will be reminded of the behavior of forces, (a)
In physics, forces are dynamic vectors which tend to change the distance
between one thing (or event) and another, (b) Unless there are obsta-
cles in the way, a force operates along a straight line drawn from the
first object (or event) to the other, (c) The direction in which a force
operates is either that of an attraction or of a repulsion, of a reduction
or of an increase of the given distance, (d) The formula by which the
intensity of a force between two objects is given contains two terms
which refer to the sizes of a decisive property (for instance, an electric
charge) in one object and in the other. It is always the product of these
two terms on which, according to the formula, the intensity of the
force depends. Consequently, a reduction of the crucial term on one
side can be compensated for by an increase of the term on the other side.
Gestalt Psychology Today : 13
We have just seen that the behavior of vectors in motivational situa-
tions is the same as the behavior of forces in nature. Gestalt psycholo-
gists are, therefore, inclined to interpret motivation in terms of such
forces or, rather, of forces which operate between certain perceptual
processes and processes in another part of the brain, where a need may
be physiologically represented. We have no time to discuss the question
how cortical fields or forces would cause overt movements of the organ-
ism in the direction of these forces.
Now, not everybody likes the term "force." Its meaning, it has been
said, has anthropomorphic connotations. But, in human psychology,
we simply must use terms which if I may use this expression "sound
human." If we refused to do so, we would not do justice to our subject
matter which (to a high degree) is human experience. To be sure, in
physics, Heinrich Hertz once tried to do without the concept "force."
He actually wrote a treatise on mechanics in which he avoided this
term. And what happened? He had to populate the physical world with
unobservable masses, introduced only in order to make their hidden
presence substitute for the much simpler action of forces. Ever since
that time, physicists have happily returned to the old concept "force,"
and nobody has ever been harmed by the fact.
The present reasoning leads to a conclusion which distinguishes this
reasoning from the treatment of motivation in the behaviorisf s system.
Clark Hull was a great admirer of science; but, to my knowledge, he
hardly ever used the concepts characteristic of field physics. The funda-
mental distinction between physical facts which are scalars (that is,
facts which have a magnitude but no direction) and vectors (which have
both an intensity and a direction) played no decisive part in his theoriz-
ing. His main concepts were obviously meant to be scalars. There is no
particular spatial direction in a habit strength, none in a reaction poten-
tial, and none even in what he called a drive state. Hence, the core of
modern physics as developed by Faraday and Maxwell had no influence
on his system. For this reason, and also because he refused to consider
motivation as an experienced vector, he could not discover that the
operations of motivation appear to be isomorphic with those of fields
or forces in the brain.
But, if motivation is to be interpreted in this fashion, certain assump-
tions often made by behaviorists may no longer be acceptable. Take
the concept of learned drives. As I understand this term, it means that
learning can attach a drive state to a great variety of stimuli which, as
such, are neutral facts. Now, so long as a drive is not regarded as a
14 : Wolfgang Kohler
vector, this seems indeed quite possible. But, if the drive in Hull's sense
is replaced by a motivational force which operates between a subject
and some perceptual fact, no arbitrary connections of this kind can be
established. For, now motivation becomes the experienced counterpart
of a force in the brain, and this force depends entirely upon the relation
between conditions in the subject and the characteristics of the per-
ceived object. There can be no such force if the object is, and remains,
a neutral object. Forces only operate between objects which have the
right properties. Any example of a force in nature illustrates this fact.
How, then, are the observations to be explained which are now in-
terpreted as a learning of drives? After all, some learning must be in-
volved when an originally neutral object gradually begins to attract or
repel a subject. From the present point of view, only one explanation
is possible. Supposing that the subject's need does not vary, learning
must change the characteristics of the object, and thus transform it
into an adequate motivational object. One instance would be what
Tolman calls a sign Gestalt; in other words, the neutral object would
become the signal for the appearance of something else which is a
proper motivational object. This expected object would now be the
object of the motivation. Or also, when a neutral object is often ac-
companied by facts which are natural motivational objects, the char-
acteristics of such facts may gradually "creep into" the very appearance
of the formerly neutral object and thus make it a proper motivational
object. Years ago, comparative psychologists in England stressed the
importance of such processes, to which they gave the name "assimila-
tion." They regarded assimilation as a particularly effective form of
an association. And is it not true that, as a consequence of learning, a
coffin looks forbidding or sinister? I also know somebody to whom a
bottle covered with dust and just brought up from the cellar looks most
attractive. As a further and particularly simple possibility, the subject
might just learn more about the characteristics of the given object itself
than he knew in the beginning; and the characteristics revealed by this
learning might be such that now the same object fits a need. It seems
to me that all these possibilities ought to be considered before we accept
the thesis that motivations in the present sense can be attached to actu-
ally neutral objects. Incidentally, similar changes of objects may also
be responsible for the developments which Gordon Allport once re-
garded as evidence of "functional autonomy."
You will ask me whether my suggestions lead to any consequences
in actual research. Most surely, they do. But, since I have lived so long
Gestalt Psychology Today : 15
in America, and have therefore gradually become a most cautious
scientist, I am now preparing myself for the study of motivation by
investigating, first of all, the action of dynamic vectors in simpler fields,
such as cognition and perception. It is a most interesting occupation
to compare motivational action with dynamic events in those other
parts of psychology. When you do so, everything looks different, not
only in perception but also in certain forms of learning. Specific work?
There is, and will be, more of it than I alone can possibly manage.
Consequently, I need help. And where do I expect to find this help?
I will tell you where.
The behaviorist's premises, we remember, lead to certain expecta-
tions and experiments. What I have just said invites us to proceed in
another direction. I suggest that, in this situation, we forget about
schools. The behaviorist is convinced that his functional concepts are
those which we all ought to use. The Gestalt psychologist, who deals
with a greater variety of both phenomenal and physical concepts, ex-
pects more from work based on such premises. Both parties feel that
their procedures are scientifically sound. Why should we fight? Many
experiments done by behaviorists seem to me to be very good experi-
ments. May I now ask the behaviorists to regard the use of some phe-
nomenal facts, and also of field physics, as perfectly permissible? If we
were to agree on these points, we could, I am sure, do excellent work
together. It would be an extraordinary experience and good for psy-
chology.
PART I
ESSAYS BY MAX WERTHEIMER
Max Wertheimer
LATE OF THE GRADUATE FACULTY,
NEW SCHOOL FOR SOCIAL RESEARCH
ON TRUTH
Science is rooted in the will to truth. With the will to truth it stands or
falls. Lower the standard even slightly and science becomes diseased
at the core. Not only science, but man. The will to truth, pure and un-
adulterated, is among the essential conditions of his existence; if the
standard is compromised he easily becomes a kind of tragic caricature
of himself.
The scientific situation with reference to the theory of truth is com-
plicated at present. In the last decades logicians and epistemologists
have worked intensively on its problems; many complications have
emerged; some of them seemed to menace the whole inquiry. New ap-
proaches have been made, much positive work has been done. In this
paper I shall discuss only one aspect of the problems, and in the simplest
way. What follows has to do with things that the natural man feels as
self-evident; but the theory must envisage these things. Examples will
be drawn from everyday life.
I begin with the classical definition of truth, a point of departure that
lends itself to the simplest exposition of the subject, although what I
have to say applies equally to many other approaches. According to
the traditional definition it is propositions that are true and false. A
statement or proposition is true which corresponds with its object, and
Reprinted with permission from Social Research, Vol. 1, No. 2, May, 1934.
20 : Max Wertheimer
vice versa a proposition is false if it fails to correspond with its object.
Truth, so conceived, is a general quality of propositions. It is a ques-
tion of the relationship between the proposition on the one hand and
the object on the other.
Many new problems emerged in the criticism of this old formulation:
for example, what, strictly, did the term "correspondence" mean, what
could it mean? What is meant when we say a proposition should "cor-
respond" with its object? Or when the question is raised as to whether
reference to the object does not involve an illegitimate transition? There
are many other fundamental questions that have led to new formula-
tions.
But regardless of the necessity for changing the definition of truth
in the sense of these objections and regardless of the great importance
of these changes in some respects, the claim contained in the simple
form of the old definitions is in itself important enough. Statistics, pub-
lications of one kind or another, do contain figures that are plainly
false; in this respect there is lack of conscientiousness and worse. The
standard is often lowered, especially in general statements in which
whole realms of thought are disposed of on the basis of personal pre-
dispositions or of a few individual facts. What the old definition requires
is straightforward fidelity to facts in the sense in which a proposition
corresponds or fails to correspond with the facts.
But truth demands more.
If we consider the function of truth in life, in living thought and being,
then the old definition is unsatisfactory. I will touch here only on one
relatively simple point.
An example: a man hires another to steal something out of a desk;
the theft is discovered; it has been established that the second man was
seen near the house; the judge, who does not know their connection,
asks the first man whether he took the article from the desk; the first man
answers "No," gives his alibi, and is discharged. He did not take the
article from the desk. His statement that he did not is true according
to the definition. Nevertheless, he lied.
The difficulty need not necessarily invalidate the classical definition
of truth. One could say that the difficulty was due to the way in which
the suspect was questioned. The investigator should have asked, not
"Did you take the article from the desk?" but "Are you guilty of stealing
these valuables?" Why was the judge so stupid? The judge's stupidity
is irrelevant. It has nothing to do with the fact that the man would have
made an untrue statement, had he said that he did take it. In the old
On Truth ; 21
terminology, this is known as the false conclusion based on many ques-
tions. Or another explanation may be offered. One may point out that
the word "taking" means not only the physical fact of taking but like-
wise its cause. There are two meanings of the word "take." For each of
these meanings, the old definition of truth clearly establishes what is
true and what is false.
But solutions of this kind tend to eliminate the problem instead of
solving it. The issue is waived. The way to go about solving the problem,
it would seem to me, is to attack it directly. The investigator's question
to the suspect is not an isolated fact in a vacuum. It is an integral part of
a well-defined situation in which the investigator, the suspect, and the
theft form a characteristic whole. The detail that the suspect did not
himself remove the valuables from the desk stands in an important and
characteristic relation in the whole situation. Were the suspect really
not guilty, had he nothing to do with the whole thing, then this detail
would in the new whole have an entirely different function, a different
significance, a different role.
The proposition, the statement, "I did not take it from the desk,"
corresponds with reality; but with a piecemeal reality, torn from its
context, seen as a piece, blind as to its connections as a part in a related
whole; or in another related whole in which the suspect appears inno-
cent. The real truth must take account of any statement, and equally of
its corresponding object, as parts of related wholes. A thing may be true
in the piecemeal sense, and false, indeed a lie, as a part in its whole. We
must distinguish the object as piece ja|, the object as part of its whole
O Q
, , the object as part of another whole
For the time being let us indicate piecemeal truth and falsity by t and
f ; and by T and F what we have called the real truth, in which the state-
ment and its object are considered as parts in their related wholes. The
example that we have been discussing is of the form tF. For the function
of truth and falsity, it is not a question only of the statement in itself
(an sich), but of the statement S and the object O as parts of their
wholes; S and O, in their roles, in their functions as parts of their wholes.
The truth here does not consist only in the correspondence between the
proposition and its crude, isolated object. If the object turns out to be a
part in the pattern of a definite situation, then the proposition is really
true when it corresponds not only with the part as such but with the role
that it plays in the whole.
If each item in the situation were isolated from every other, if our
22 .- Max Wertheimer
world were in every instance nothing more than the sum of isolated
facts, if we had to do only with "atomic facts," then the old definition of
truth, the t and f, would be adequate. If, however, these facts or data
exist, not in isolation but as parts of a whole, determined by their func-
tion in this whole, then it leads to blind and false conclusions to consider
them as pieces. The whole plan of traditional logic, in all its rules and its
general laws, is set up to deal with piecemeal content in a summation
relation. And for this it is adequate.
Logistic (Logistik), the study of relational networks and of implicit
definitions, provides the possibility, indeed the necessity, for seeing con-
tent as a part of its whole, but in a limited sense. It defines content by
its place in the relational network, but the network is built up as a sum-
mation. (Cf. in logistic the new approaches to a structural theory of
truth. ) Logistic has failed hitherto to study the relationship between
content as piece and content as part. In reality we frequently have the
possibility of considering as pieces what are really parts of a system.
Science for the most part indeed finds itself in this situation, at least in
the early stages of a problem.
Before we proceed, let us consider briefly some other examples of
this form. A newspaper writes: "We know now just what to expect from
Minister X. Such are the steps he intends to take! At the banquet held
on the twenty-seventh of this month, he declared himself against . . ."
Indignant denial: "The truth is that Minister X was not even present at
the banquet."
Any number of examples may be found in statistical material and in
balance sheets. In the course of the last decades a whole technique of
the form tF has been developed, the technique of doctoring balances,
etc. This does not refer necessarily only to deliberately misleading fig-
ures. Simultaneous newspaper reports on the same facts in journals be-
longing to different political parties afford the psychologist and the
logician a veritable treasure house of the various forms of tF.
Moreover it is not at all necessary to conceal any of the relevant de-
tails in order to arrive at blind or false statements in the sense of tF. All
the data may be laid before the reader or hearer but in such a way as to
deceive him by the technique of shifting the emphasis, displacing the
center of gravity (Umcentrierung) . An example of the simplest tech-
nique: during the war newspapers in some countries were compelled to
give out reports in the exact words of the general staff. They achieved
what they wanted by the use of heavy type for some parts, and it often
happened that in this way entirely opposite impressions were produced.
On Truth : 23
I remember cases in which I received so strong an impression that the
reports were different, that I could not believe they were the same until
I had compared them word for word. (Cf. the old false conclusion by
inflection.)
These were cases of the form tF. Instead of the two truth values t and
f , we have schematically the four combinations tF, tT, fF, fT. That there
are cases of tT and fF is clear. Are there also cases of f T, that is, false as
pieces, true with respect to the whole, to reality? Yes: for example, an
excellent caricature. It may be wrong in practically every detail and yet
be a truer representation of its object than a photograph which is accu-
rate in every detail. Other cases are anecdotes and stories about men,
events and ages which are known to be inventions yet nevertheless hit
the nail on the head. Se non e vero, 6 ben trovato. But of course it must
be ben trovato.
Propositions of this type present an inherent danger, however. The
danger is due to the fact that it is usually easier to prove the truth of an
item in isolation than in its role in the whole. Indeed, many propositions
for which the claim is made that they are true in relation to the whole,
to the essence, are really false both in themselves and in relation to the
whole. (Cf. the history of some scientific schemata.)
If, then, we are really to find out what is true and what is false, we
must direct our attention to the role of any particular item in the whole
of which it is a part.
This necessity has generated a number of problems that are as inter-
esting as they are difficult. The basic problem here is the function of a
part as a part of its whole. It is a central problem for the Gestalt theory.
We may formulate it as follows. Facts occur in these ways: first, in
isolation, as such, or as units in a sum; secondly, as parts of their wholes
and a part may figure in two or more different wholes. Often there is
the possibility too of cutting off as a piece what should really figure as a
part. What are the differences in these cases?
One may ask why we speak of two kinds of truth and not merely of
two kinds of data. The answer is that the two kinds of truth may deal
with the same data, but that one (T,F) goes to the heart of the matter
while the other (t,f ) may remain external namely, when the set of
facts is not merely a sum.
The approach that we have used, the formulation of a truth function
T,F in relation to the old t,f, is not the only possible theoretical ap-
proach. The attempt can be made to do justice to objects by other
24 : Max Wertheimer
approaches. The problems in question can only be pointed out, they
cannot be dealt with even briefly here. But what is necessary is that the
theory clearly envisage these objects. From whatever angle we approach
the theory, concrete problems of research are generated in any serious
attempt at a theoretical solution.
Remote as it may appear from logic as customarily defined and
treated, we need a logic of objects. If there is any objection to calling it
logic, the name is a matter of indifference. I believe that these questions
rightly belong to logic; they are not questions merely of psychological
fact but they raise issues in terms of correct, incorrect, adequate, true,
false, logical, illogical.
Attempting to find a different approach, one may say that the truth
function is not touched in all this, that in the burglary example, for
instance, the difficulty is a matter simply of variously defined concepts.
We need only to define carefully and to agree on definitions. Can any
one suppose that all the difficulties consist solely in the failure to define
clearly? We can readily see that the real problem is not solved; the
question then is, in a given situation, in a given context, which definition
is justified, correct, significant. In a context it is not a matter of in-
difference, of purely subjective appraisal which definition is to be
chosen, which things are to be discussed. There are instances of mis-
leading, inappropriate, inadequate definitions. And in the question here
raised, of the logical formulation of definitions, we find again, if we look
more closely, the problem of piece as opposed to part. Structurally this
is entirely analogous to what was discussed above. (Cf. examples, pp.
26-27 below.)
Similar considerations hold for the problem of abstraction. There are
logically bad, incorrect, misleading abstractions. All this is germane to
the logical problems of the formulation of concepts.
The problem, to be sure, in a certain sense disappears if logic is to be
nothing but the formal theory of deduction. In that case the question of
an adequate formation of concepts that does justice to objects falls
away, and the coordination of the model with reality becomes purely a
matter of chance. There are similar problems, however, in a purely for-
mal logic of implicit networks.
In traditional logic and in logistic the truth function t,f is funda-
mental. In some of the newest developments, logicians define all logical
constants such as not, and, or, following, etc., by means of tables of
truth functions. If we follow the method of multiplying truth functions,
if we take T,F into consideration, very radical consequences follow.
On Truth : 25
The cases of tF and fT prove to be very complicated special cases. We
soon see that the crux of the problem is the problem of the objective,
o
content relationship between |a| and , . When, how, in what are they
different? Solutions prove to be too easy where the content of a part is
held to be unaltered in cases of abstraction, or of the recurrence of simi-
lar pieces or sections in different wholes. Whether there is a difference
and what that difference is difference in role, in function, in dynamics,
or actually in content itself this is a task for research which leads to
very concrete and formal problems.
If we put |a| and |b| in a real relationship as parts of a whole, then we
have at all events not merely these two, each by itself as before, but we
must consider the question of their alteration: |a| + |b|-> , + K*
The same holds of the abstraction: in a real or a blind "abstraction,"
lib"'" bl
K
the result a , |a|. This is already the case
in relational networks: if a part is really taken from a system, then the
remainder, implicitly defined, is substantially changed. And the same
3. 3.
question is involved if we consider the relationship of , and
The general science of logic demands the same rigorous exactitude
as classical logic but classical logic represents a special case, like the
Euclidean special case in geometry. For the most part it is adequate in
those instances in which the inner functional content approaches zero.
A fundamental difficulty might be discerned in what follows. One
should be distin-
o
might ask whether the truth values |a| or and
guished, whether we should not likewise take into consideration the
a
ab
differences
abc
and
abed
etc. Would not this lead to radical uncer-
tainty? Until we had exhausted all the data, would we not be in perma-
nent danger of a tT becoming transformed into a tF? And to exhaust
all data is of course impossible.
There are actually instances of changes in truth values. If we look
more closely we soon see that this is not a question of an objection but
of a field of research. In the first place it by no means follows that the
mere addition of indifferently assorted facts continues to affect the truth
value at all points. All things are not interconnected in the same way;
there are fields, there are questions, that are relatively self -enclosed. If
to some object situations we continue to add new facts, we see that at a
26 : Max Wertheimer
particular point the truth value actually no longer changes. And this
has its own inner reasons. A Gestalt, a pattern, limits itself. Whereas in
general the variation of a part demands the simultaneous variation of
the other parts, this no longer follows, for example, when the variation
takes place outside the boundaries and the field of the Gestalt. The
question of establishing the boundaries of a Gestalt, as of a practical
situation, is not a matter of choice but is subject to examination on
grounds of being correct or incorrect, logical or illogical.
Moreover, if there are situations in which what we have here dis-
cussed is questionable, we should not permit our vision to be obscured
for that reason. The cases of tF that actually occur are often entirely
unequivocal.
I mention a special case. It happens that in a fight between two
parties, the way out is found in this form: for each party his own point
of view is immediately relevant; it seems impossible to bridge the differ-
ences in outlook; we have at first
ab
in the one case and
the parties are starkly opposed. We find the solution
ac
abc
in the other;
the common
solution that takes into account the viewpoints of both instead of doing
violence to one by the other.
The form of this mode of solution touches questions of domination
by the individual social and cultural milieu in the latest discussions on
the theory of truth.
Here I should like to include a few more simple examples illustrative
of the principal theoretical points.
1 . Note these two musical motifs:
Im
r _._
-f
-)
Corresponding to the motif at the left, assume the continuation in D flat
major; at the right the continuation in D minor. Such a continuation is
not absolutely necessary, however.
Someone may say that the first two tones and the first interval of the
second motif are the same as in the first motif. That is true, for the piano.
The musician replies that you have completely failed to understand. In
the motif at the right it is a diminished fourth, the motif at the left a
major third; the second tone at the left is the tonic, at the right the
On Truth : 27
passing-note. And it would be false, illogical, if the second tone in the
motif at the right were to be written as D flat. If we fail to hear the tones
as different in these different combinations then we have understood
absolutely nothing of what is involved in these motifs. 1 And actually the
singer, the violinist, often intones quite differently in these two instances.
Here we have our example if we confine ourselves to the motif on
the right and the listener takes the first two tones in isolation, really ab-
stracts them, blindly unconscious as to the role, the function, they have
in their whole (tF).
2. Take several members in a mathematical series. A formula is
thought out; and it becomes evident then how the formula and the serial
structure and the variability are entirely different if the further develop-
ment of the series is taken into consideration. Or we have a mathemati-
cal series. We characterize it by a particular formula, blind as to which
member it is of a family of curves.
3. Or the engineer discovers in his measurements that he has a
straight line. Measurements taken in other subsections show that he is
dealing not with a straight line but with the asymptote of a hyperbola.
Another step is indicated if we consider now in all its relevances the
earlier remark following the tF examples. All the "data" may be given
in sum, all the tones of a melody, all the members of a series nothing
omitted, accordingly, from the sum of the data. And yet nothing may
have been grasped. What is really involved may not have been seen at
all in these cases in which the sum of the data is present as the sum of
individual pieces or fragments; and this in spite of the fact that all the
single relations may also have been given in sum. In addition to the tF
and tT, we have the t (?) .
Someone makes a serious request, "Tell me the truth about this."
The answer may contain many facts that are individually valid and can
leave the questioner entirely ignorant. I have failed to understand, to
grasp, if I know everything as a sum, if I have not grasped the inner
connections of the whole, the inner determining principles, if the indi-
vidual data are not present as parts, transparent, determined as parts.
Here too we have to deal with concrete problems, with the problem of
grasping what is really involved in the transition to what is understood
from what is not understood. The same kind of understanding is neces-
sary in order to make a reasonable prediction.
iCf. examples in the Z. PsychoL, 1933, 129, 353 if. [Reprinted in M.
Wertheimer, Productive Thinking, enlarged edition (New York: Harper, 1959),
pp. 260-265. Ed.]
28 : Max Wertheimer
I should like briefly to indicate a second point here. The classical
view, according to which truth and falsehood are qualities of proposi-
tions or assertions, seems to limit the meaning of truth and falsehood
artificially. It is not only and primarily a question of what anyone says or
states. Truth and falsity, indeed understanding, is not necessarily some-
thing purely intellectual, remote from feelings and attitudes. In many
of the previous examples the most important thing is not the statement
but the whole position, a man's attitude toward the thing itself. It is in
the total conduct of men rather than in their statements that truth or
falsehood lives, more in what a man does, in his real reaction to other
men and to things, in his will to do them justice, to live at one with them.
Here lies the inner connection between truth and justice. In the realm
of behavior and action, the problem recurs as to the difference between
piece and part.
Science is rooted in the will to truth.
If it is clear that science demands not only scrupulous fidelity to facts,
however indispensable, but that it demands more what we have here
indicated by T; if decisions on this basis are reached with greater diffi-
culty; if in certain scientific fields definite decisions are at present im-
possible for technical or for fundamental theoretical reasons; all these
considerations cannot weaken the claims of truth. Indeed the responsi-
bility of the scientist increases with his difficulties, the demand grows
for unswerving determination in the search for truth, for that relation-
ship to the things themselves that means to do them justice, and wills
the truth untainted by wishes or commands.
Such is the real origin of the claims of science to free discussion: in
the interests of safeguarding scrupulous fidelity to facts, in the attitude
of every scientific worker toward his work, in his readiness for the most
exacting proof of his theses. A conception is too easily influenced by
will or by wish, too easily becomes instead of the object of research an
idol, an end in itself. And even if we suppose that science should func-
tion in the service of life, we must beware lest the instrument corrupt it-
self. It is of no little importance to preserve in its purity the instrument
that is the search for truth.
But the stake is not the instrument; it is man himself.
Max Wertheimer
LATE OF THE GRADUATE FACULTY,
NEW SCHOOL FOR SOCIAL RESEARCH
SOME PROBLEMS
IN THE THEORY
OF ETHICS
The old conception of homo sapiens implies that some faculties, some
abilities are important for man (although not always actually realized) .
Among these are the ability and tendency to understand, to gain in-
sight; a feeling for truth, for justice, for good and evil, for sincerity. Con'
nected with the realization of these abilities is the old conception of
human dignity as an inner task of man. Only then is one a true man,
if . . . , only then do we have a truly human society, if ...
This conception of homo sapiens does not overlook human blind-
nesses and weaknesses, the differences between men and between situa-
tions. It does not ignore the fact that these faculties are often concealed
and overgrown, that their development is sometimes hampered; that
not every man is always a homo sapiens, that he may at times be his
own tragic caricature; that there are often conditions and circumstances
which interfere with their development or realization; that there are
forces which often conceal these human qualities or counteract their
development.
Reprinted with permission from Social Research, Vol. 2, No. 3, August,
1935,
30 : Max Wertheimer
Today we often encounter a thesis radically opposed to the concept
of homo sapiens. This thesis maintains that it is simply wrong to speak
of "man" in such a way, that "man" is only a pale abstraction, a fiction.
It affirms that men, races, cultures are fundamentally different from
one another in their conceptions and evaluations. To speak of the ethics
is, according to this thesis, meaningless; we have only a variety of dif-
ferent systems of evaluations and these are mere historical facts chang-
ing with history. All values are fundamentally relative, changing with
place and time. If we construct the systems of axioms for the ethics of
various peoples and times, we have as a result simply a sum of different
systems. Each system is only a fact like any other system. Many use this
doctrine of the relativity of ethics today as a self-evident statement.
Manifold factors gave rise to this thesis. 1 will mention here only two
relevant factors in the development of science. In recent decades, fortu-
nately, there has been a wide development of special scientific investiga-
tions of the facts involved, investigations seeking to discover the numer-
ous facts concerning our problems in ethnology, sociology and cultural
history. With the assembling of the results, material for comparison is
to some degree at hand. There have been reported very different moral
concepts and evaluations. The conclusion was drawn by some that di-
versity as a fundamental principle has been established and that the
old idea of unity is an empty fiction. Accordingly, this old idea is to be
considered only as an historical fact determined by the conditions of
time and place under which it was developed.
Formally, certain aspects of modern logic favored this development.
Axioms now often appear only as propositions which are so chosen that
the single data in a system may be deduced from them. Some formulate
the problem as follows: The task of science can only be to state the
single facts in a field and to seek, formally, general propositions which
represent these facts. With the exception of logic and mathematics axi-
oms have an arbitrary character. Opposites of one kind of axioms are
always conceivable; differences in axioms cannot be settled except by
arbitrary decision. Thus a system of evaluations of a certain social group
may be represented by one system of axioms, the evaluations of another
group by other axioms. "There is no way to prove, either by logical rea-
soning or by scientific demonstration of facts, which of two conflicting
norms is more right than the other. The choice between two such norms
must be an arbitrary one; all norms are facts with an absolutely equal
claim to recognition." Thus we have different systems of ethical axioms
representing simply different sets of facts, and that is all.
Thus one may formulate: Different axioms of evaluations, even if
Some Problems in the Theory of Ethics : 31
contradictory from one system to another, are equally sensible. You
can get another ethics, equally sensible, if you reverse the evaluation
signs in the axioms of one ethics, substituting plus for minus and minus
for plus in the norms.
Certainly, this last statement is not true. Are there not common con-
ditions in human societies, both on the technical and on the human side,
limiting the possibilities? Are there not identities of conditions in human
societies and in men?
But first of all, we want here to ask a formal question: is the direct
inference from the statement of different (contradictory) facts of evalu-
ations to the heterogeneity in axioms logically valid?
It is not. It would in a simple way be valid if the objective were here
nothing more than a mere cataloguing of naked facts; if there were no
differences in the meaning of contents, in the causation and determina-
tion of attitudes touching our problem; if all the facts were of equal
theoretical rank; or if I consider the facts without further theoretical
consideration, psychological or sociological, if I see them torn from
their determining nexus and compare them as such. But facts of evalua-
tions cannot be dealt with offhand as if they were of the same rank.
There are, for example, critical psychological differences in the deter-
mination. I need only to mention suggestion, inertia, temporary blind-
ness.
Consider a relatively simple example. It often happens that men in a
psychological crowd situation take part ardently, and a short time later
are unable to understand how they could have behaved so blindly. They
did it under the blinding psychological conditions of the crowd situation.
Their behavior was possible only because of a tremendous narrowing
of the field of consciousness or a narrowing of the field of actual deter-
mining factors.
Is the transition from the one state to the other correctly described
simply as a transition from one system of ethics to another? And have
we then simply two contradictory systems of axioms of equal claim and
rank? Is our problem satisfactorily settled with such a statement? Is the
task of science here only to state the bare facts and to formulate generali-
zations, to state the different axioms? Do we not have to study how these
different modes of behavior have been caused? And is it not possible in
such cases that different conditions have led to different behavior and
to different evaluations, although the principles of evaluation are identi-
cal? Temporary blindness to some of the factors involved can change the
factual behavior under the identical ethical principles.
Such conditions as blindness are only special cases. Another type of
32 : Max Wertheimer
case comes readily to mind. Because of a fallacy, a fallacious suggestion,
because of a current theory, of a certain superstition, because of a cus-
tom or tradition the behavior or evaluation toward an A is formed in a
certain way, for the reason that this A means a B for the person under
these conditions. The result is an apparent contradiction between the
evaluations of A in different groups, but does not mean necessarily a
difference in the fundamental causal and determining evaluation of B.
We have then a diversity, a relativity in the meaning of A but perhaps an
identity in the determining principle B. And in a similar way different
biological or sociological conditions often require, because of an identi-
cal basic need, different factual evaluations of an apparent content A,
an A which is the same only when taken as an atomic fact but different
in the role it plays in the whole situation.
The logical form may be illustrated in the simplest, baldest case. In
one group there may be a positive evaluation, a positive response to the
object A, in another group there is none. Now, I have to ask, what is
the cause, what is the reason for this positive evaluation of the A in the
first group? It may be, in the simplest case, a conditioned response. The
A is thus evaluated because it is conditioned, connected with B, which
is positively evaluated. 1 It may then be that we find in both groups the
same fundamental positive evaluation of the B.
Group I Group II
First step:
Factual evaluations
(conditioned responses) A+
Conditioning BA+
Sources (axioms, reflex) B+ B+
Second step:
Factual evaluations A+ A
Conditioning BA+ DA
Sources B+, D D , B+
Third step:
Factual evaluations A+, C C+, A
Conditioning BA+, DC BC+, DA
Sources B+, D B+, D
1 The response to A may be the "conditioned reflex," the response to B a
"reflex."
Some Problems in the Theory of Ethics : 33
The next step is that, in different whole situations, an A changes nec-
essarily as part, in its role, in its function. This step transcends the purely
summative constellations. 2
If we try to look over the available factual material concerning the
various evaluations so far as we know them on the basis of the present
state of research, with consideration of the theoretical problems in-
volved, there appears a mixture of different things. On the one hand, we
have evaluations formed by varying meanings, varying traditional cus-
toms, usage, current theories, differences in the situation, etc., and in
these there is a marked variety. These facts do not give directly the
sources and principles of the evaluations in behavior. If we search for
real principles, axioms, on the other hand, we find that the matter in
most cases has not been sufficiently clarified; if we look for contradictory
norms which would directly, clearly, unmistakably form a basis for con-
tradictory axioms, there is little material at hand; if we look for contra-
dictory differences in simple human relationships, in cases uncompli-
cated by misunderstandings, etc., in situations clearly concerned with
open, friendly trustfulness, or scheming, cunning, selfish, petty deceit-
fulness, with wise understanding, sincere integrity, or reckless, egoistic^
callous brutality or obvious, brutal injustice; then the facts seem to tend
more strongly in the direction of fundamental identities.
Perhaps we can already trace the beginnings of a third stage in scien-
tific development. In an earlier period, the first stage, ethnology used
European concepts simply as universal and self-evident criteria. Then
comparative research among different societies brought to light a wealth
of diverse evaluations leading to the second stage: there seemed to be
simply diversity, different ethical systems. Today the situation in eth-
nology seems to be advancing to a third stage. On the one hand, we have
learned that in the determination of evaluations we must take into ac-
count so many biological and sociological, religious and historical com-
plications, that we see that the inference from the single facts of evalua-
tion to diversity of principles and axioms was too hasty; the apparent
differences are not direct evidence for differences in the axioms. On the
other hand, in clear cases of human relationships uncomplicated by
"pejorisations" concerning truth and falsehood, justice and injustice,
the facts tend more in the direction of fundamental identity than in the
direction of contradictory norms.
2 Cf. Max Wertheimer, On truth, Soc. Res., 1934, 1. [This is the previous
article in the present volume. Ed.]
34 : Max Wertheimer
The problems of scientific ethics are very complicated. If the prob-
lems are to be studied scientifically we must try to distinguish the various
problems involved and modestly proceed step by step in clearing up the
field. We must study the differences in the causation, in the determina-
tion of factual values. We must also study differences in the logical
structures in the facts of evaluation. I wish here to raise only a single
problem of the many in this rich field, a point frequently overlooked in
the present theories.
It is customary to speak of values indiscriminately as if they fitted
into only one scheme. The usual scheme is the following: we have on
the one hand an object, a content, and on the other hand the subjective
feeling, the subjective judgment of value. To the object is only added
an evaluation; this may be positive or negative; it is subjective in deter-
mination. Logically speaking, evaluations are determined externally to
the object and are arbitrary with respect to the object.
For this scheme the following example may serve.
I call a food good because to me it tastes good, or because it is in
fashion, because others call it good, because it is now considered aristo-
cratic to prefer this food to other foods, etc.
This evaluation may be true for me, for another person not true. The
evaluation depends on the relation to the subject, his subjective feeling
which is entirely external to and arbitrary with respect to the object.
The logical structure is: the object A and the added plus or minus
evaluation.
We may compare with this example the following:
A judge convicts an obviously innocent person because the guilty
person has bribed him, or because he hopes to achieve by this con-
viction a certain profit for himself or for his group. Or, a poor, hungry
child has been given a piece of bread to eat; an adolescent rowdy comes
along and grabs it to use it as a football.
If we try to understand such cases with the same formulation as we
used in the earlier example, in terms of the arbitrary addition of a sub-
jective evaluation to an object, then there is a lack of clarity, something
does not jibe; the matter demands clarification, there seem to be various
problems involved. Is the logical structure the same as in the case of
tastes?
We have here an objective situation and a behavior, an action, and it
is important first to sense, to see, to comprehend how this action actually
stands in the situation. The most important question is not that of a
subjective evaluation standing outside of an object, but the relations
Some Problems in the Theory of Ethics : 35
within the happening itself, how the action meets the requirements of
the situation, the zueinander of the two, the relation between the situa-
tion and the action. The factual behavior may be determined without
regard to the situation, blind to what it accomplishes in reality for the
object. The behavior may do violence to the situation. The behavior may
be determined so as to be appropriate to the structure of the situation, to
accomplish what the situation requires. We have here various objective
qualities, qualities concerned with the relationships in the situation.
Whether a particular instance of real behavior is a case of the one or the
other of these three classes is a matter for which my seeing or failing to
see, my subjective evaluation, is irrelevant. Whether I evaluate it posi-
tively or negatively does not change by an iota the issue which quality
of the three is present in the situation, what role the behavior really
plays, what kind of part it is in the situation. This point must be under-
stood first of all. If someone in a real situation fails to see it, and perhaps
acts accordingly, then he has not a different ethics but he is blind to the
main issue.
It is difficult for many theorists to see clearly this difference in the
logical structure and to realize its relevance for the theory.
What is involved here in principle? In logic or mathematics there are
decisive principles, such as, for example, the principle of contradiction.
This is not an arbitrary subjective matter, it concerns the material and
its relations.
But are there comparable principles for ethical questions, something
which will naturally be entirely different in this different field? Many
theorists deny it. We do not have such principles, they assert. Moreover,
there seems no theoretical possibility for such principles because of the
basic subjectivity of all values.
Perhaps the formulation of the problem was too narrow. Is it not
possible that for certain problems of values there are principles which
are identical with those in other fields, so that some principles in logical
thinking would only represent a special case of a common principle? I
will try to indicate an approach which may open the way to theoretical
studies. It involves some radical changes in theoretical concepts but,
on the other hand, it seems to formulate more directly something which
has always been intended. I can sketch here only one side of the matter.
Someone in adding makes seven plus seven equal fifteen. This hap-
pens; such additions are real facts. And he says, I call it good because I
love the number fifteen, or because I have set up the principle that in
the addition of two whole numbers, the sum shall always be divisible by
36 : Max Wertheimer
five, or, as may happen occasionally in the addition of a bill by a waiter,
because I have a personal interest in the fifteen.
The determination of the fifteen is blind to the objects or in violation
of that which is demanded by the structure of the objective situation. If
I prefer the fifteen in this case, if I evaluate it positively, this is irrelevant
to the fact that the fifteen is wrong.
What is the structure? The situation, "seven plus seven equals . . ."
is a system with a lacuna, a gap (eine Leerstelle). It is possible to fill
the gap in various ways. The one completion, fourteen, corresponds to
the situation, fits in the gap, is what is structurally demanded in this sys-
tem in this place, with its function in the whole. It does justice to the
situation. Other completions, such as fifteen, do not fit. They are not the
right ones. They are determined by caprice, in blindness or in violation
of the function this gap has in the structure.
We have here the concepts of "system," of the "gap," of different
kinds of "completion," of the demands of the situation, the "required-
ness." 3
The case is similar if a good mathematical curve has a gap, a place
in which something is lacking. For the filling in of the gap, there are
often, from the structure of the curve, determinations which indicate
that the one completion is appropriate to the structure, is sensible, the
right one; other completions are not. This is connected with the old
concept of "inner necessity." And not only logical operations, conclu-
sions, etc., but also happenings, doings, being, can be, in this sense,
sensible or senseless, logical or illogical. 4
We may formulate: given a situation, a system with a Leerstelle,
whether a given completion (Lueckenfullung) does justice to the struc-
ture, is the "right" one, is often determined by the structure of the sys-
tem, of the situation. There are requirements, structurally determined;
there are possible in pure cases unambiguous decisions as to which com-
pletion does justice to the situation, which does not, which violates the
requirements and the situation. 5
3 These are quite complicated concepts; I may mention only that these
concepts are connected with recent developments in mathematics, science, and
Gestalt theory.
4 This transcends in some respects the boundaries of the usual theory of
traditional logic.
5 Some fundamental principles of logical thinking are special cases of this
principle. Thus you must not introduce a proposition blind to the conditions
of the situation; it is not right to act blind to the structure or to other given
data; in the law of contradiction, the filling in of a non-A is simply not per-
missible logically, if in the system a plus-A is already included. This is but the
baldest special case of the general principle.
Some Problems in the Theory of Ethics : 37
Now the case of the bribed judge is structurally similar to our first
mathematical example. We have here again a system with a gap. If the
judge convicts the obviously innocent man for reasons of personal in-
terest or interests of his group, he may prefer this behavior subjectively
but this does not alter the fact that his decision is unjust.
This case is very close to the traditional logic; but we can see the
principle in other cases also. Here sits a hungry child; yonder a man,
who is building a small house, and lacks a single brick. I have in one
hand a piece of bread and in the other a brick. I give the hungry child
the brick and take the soft bread to the man. Here we have two situa-
tions, two systems. The allotment is blind to the functions of the gap
filling. 6
These cases may serve to illustrate the principle in a preliminary fash-
ion. Many further steps are naturally required (and with a single princi-
ple the task is not done). First a clarification of the terms "determina-
tion," "requiredness," etc., is necessary. 7
A further step is involved when such a system with its gap is to be
considered as only a part of a more inclusive system. If a gangster needs
a revolver at a critical moment in order to carry out his particular task,
then the gangster with his need is logically a part of the larger system of
human society, in which he with his wishes is to be considered as a func-
tional part. If there were always an infinite regress to ever larger wholes
with subsequent changes in the determination, then the principle would
be endangered in its application. But this is a point which cannot be
settled with a simple generality but which presents a matter for research.
It seems that in many cases there are not unlimited changes with enlarge-
ment of the whole, that often enlargement no longer changes the deter-
mination. And there are many cases in which a larger and more gen-
erally embracing field is common to conflicting fields.
However the matter may be, whatever the formulation of the further
steps in the theoretical construction, and what place and rank the princi-
ple mentioned will have among other principles, it seems to be necessary
to realize what is here involved.
I may mention that it has been customary to look upon logical think-
ing as an entirely intellectual field necessarily separated from feelings,
attitudes, tendencies. The principle mentioned is not only static, to be
6 Or it may be caused by scorn. There may be men who evaluate scorn
positively, but this does not change the fact that the allotment does not do
justice to the situation.
7 1 cannot deal with this here. I may only mention that the usual simple
dichotomy of to be and ought to be has to be revised. "Determinations," "re-
quirements" of such an order are objective qualities.
38 : Max Wertheimer
taken merely as a list of assertions about determinations, but touches
on dynamics. Logical operations, logical proceedings have a great deal
to do with feelings, attitudes, real behavior; they include them. Logical
operations, as in the case of the thinking scientist, contain and demand
such things as sincerity in doing, the will to do justice to the material.
We may add this further example. A young, idealistic party member
is passionate in the negative evaluation of members of a certain race. It
is not sufficient in such cases to give the formulation: in one system of
evaluations, members of this race are positively evaluated, in another
negatively. This young man perhaps behaves thus only because he has
been brought to this state through suggestion, propaganda, through the
wanton slander that this race is a poisonous snake. He does not really
behave with respect to the A (members of this race) but to a B which
he has been taught to identify with this race. The real problem here lies
not only in the behavior of the young man, but in the enforcement of
the blind identification. This involves a corollary to the principle men-
tioned. To take away by artifice the possibility of seeing the true situa-
tion, through the enforcement of blind judgments, of improper narrow-
ing of the mental field, induction of blind centering, deprives man of
the prerequisites for our problems.
The principle mentioned differs radically from many usual definitions
of justice. Many traditional definitions use criteria which are strange,
external to the essential centering of justice as viewed by this principle. 8
I repeat, the formulated principle is no patent solution. It is rather a
starting thesis for a field of research. This field demands not a simple
solution but research and it seems necessary, as in each proper science,
to start with the study of the most concrete, transparent examples. We
have to start from "pure cases" in the scientific sense of this word. It
happens that relativists argue against such examples; that they are not
evidence against relativism "because they are too obvious." But this is a
contradiction. In reality, the most obvious are the most needed at the
outset of scientific studies. To be sure, most cases are much more com-
plicated. The contents involved are very often not so easy to define as
the seven in our mathematical example, 9 but that should not obscure our
main issue.
There are many other problems I have not spoken of here. For in-
stance, there are cases in which the situation at first does not contain the
8 Compare such a statement as that the state creates law, with the statement
that the state is an attempt to realize, to discover, to guarantee justice.
9 Wertheimer, op. cit.
Some Problems in the Theory of Ethics : 39
fact of a gap, where the gap is generated by an idea of man, by a new
goal, etc. In such cases the goal is theoretically a part of the material of
the situation and comes under the requirements of being the right one.
To be sure, this brief remark does not solve this problem; another princi-
ple is involved here. I have also not dealt with the so-called cases of
conflict, which for many theorists stand curiously enough in the fore-
ground. They present complications. Some may believe that all applica-
tions to concrete human situations are so complicated that there is no
possibility at all of using the principle mentioned. But there can be no
doubt that at least there are, negatively, serious and brutal infringements
which are perfectly unambiguous; and on the positive side there re-
mains, in cases too complicated for an exhaustive answer, the possibility
of proceeding according to the best knowledge (nach bestem Wissen
und Gewisseri) ; here the corollary mentioned plays an important role.
A difficulty can be seen in the fact that the problems of truth in logic
itself are complicated. Whatever the answer may be, that modest sense
of truth which declares fourteen to be the right answer in our mathemati-
cal example may suffice. It would be wonderful if all propagated or
factual norms of values might be as adequate as this.
A few words about the psychological side of the problem. Are these
"requirements," these "qualities" of justice and injustice, etc., psycho-
logical realities for man?
We have had very little real research work in this direction, but this
much is clear, one cannot assume that men always feel or see these re-
quirements vividly, and that they act accordingly. Sometimes one may
have the impression that the real quality of man is not homo sapiens
but rather insipiens. Whether these requirements are vivid and effective
for a man depends on many factors, among them historical factors, con-
ditions of his Standort, the state of his glands, etc. But we should not,
as often happens, confuse such factors with what is meant by the require-
ments and confound conditions of evaluation from both fields. In the
latest development of psychology some results seem to indicate that
there are real psychological vectors tending in the direction of such re-
quirements. In some experimental psychological investigations the re-
sults have indicated that the old principle is wrong which asserts that all
acts of man are centered by the "ego" (striving for one-sided satisfaction
of ego interests) . Let us illustrate this by an analogy from the psychol-
ogy of perception. It is not true that in perception the world is always
centered around the individual. If I am standing in a special place in a
room, then the walls are not out of line but I am out of line with the
40 : Max Wertheimer
walls. The ego does not always determine the system of coordinates for
the objects, but the ego often feels itself as a part oriented with respect
to the system of coordinates of the objects. Similar facts were found in
experiments on thinking and action. The vectors often arise in actual
situations from the requirements of the situation, not from egocentric
interests.
Many other psychological issues might be considered in this con-
nection. I will mention only the following:
Important is the study of the "transitions" from, for example, the psy-
chological state in a crowd situation to the "awakening" which occurs
when these special conditions fall away. And there are also for most
men moments in their lives when in a concrete situation they feel awak-
ened, when they feel how narrow, blind, crooked they have been and
acted, when their eyes are opened and they feel that their former be-
havior was possible only because something of their best, their finest,
their most worthy was missing, that they had been robbed of it.
Further, there are experiences like the following. One knows a man
who is an outstanding example of a certain caste, whose entire behavior
expresses very definitely the evaluations of his caste. And in a serious
moment the outer shell falls away, and from behind this exterior there
now comes out a simple, good, somewhat immature man, for whom the
seemingly serious attitudes which he had exhibited are in fact like
strange, superficial clothing. There seem to be layers in men, and it is a
question of fact what the inner layers of men really are. Concerning our
problem there are opposing theses. I would believe that the optimistic
thesis is the right one, however difficult, indeed however impossible it
may be at times to penetrate to this layer.
Experience seems to indicate that the conditions often are not so diffi-
cult, if the situation is clear, transparent, simple and actual, and if there
are not conditions of blindness, etc. Often it is surprising how intense
are the reactions of men in simple human situations when faced with
clear, actual injustice.
In any case the questions here cannot be decided simply by means of
the statistics of the naked actions of men. There is required a study of
the deeper connections and determinations. But even if the optimistic
psychological thesis mentioned above should prove false, if it should be
demonstrated that most men are blind or inimical in their inner nature
to these requirements, would this mean a disproof of the principle? It
would not. There would then be two races, the one for whom these re-
quirements are vivid, the other which is blind or inimical to them. For
Some Problems in the Theory of Ethics : 41
there are good men, even if there be but few. But I believe that the
pessimistic thesis is false.
We have dealt here with only a few issues from a rich and broad field.
We may formulate as a result:
"Relativity" has an important place in the rich field of different
"meanings," different biological, economic, cultural, sociological, his-
torical conditions, etc., and it is important to study these facts as mod-
ern sociology, ethnology, and cultural history do, in order to achieve an
understanding. But in order to get real axioms and to study the central
ethical problems, it does not suffice to formulate axioms simply by
generalization from the factual evaluations. There are needed studies
of the causation, the determination, the structural conditions of the
evaluations; there are needed psychological studies of the causation,
the determination, the genetic development; there are needed studies
of the logical structures in forms of evaluations.
I think that the modern methods of ethnology, sociology, and cul-
tural history which have proved to possess great merit lead in themselves
beyond a hasty, superficial relativism concerning the real ethical prob-
lems, in that they involve the study not only of the factual evaluations
but also of their structural roles, thek inner causation, their inner deter-
mination.
Max Wertheimer
LATE OF THE GRADUATE FACULTY,
NEW SCHOOL FOR SOCIAL RESEARCH
ON THE CONCEPT
OF DEMOCRACY
If one tries to understand scientifically what democracy is, one is con-
fronted with this situation. On the one hand there seems to be simple
agreement as to what is meant by democracy, as expressed in the usual
definitions, including such characteristics as "government by the peo-
ple," "majority rule/ 9 "freedom of speech," etc., etc. On the other hand
if we look more closely, if we follow the way in which different men deal
concretely with special problems of democracy, the trend, the direction
of attitudes and arguments in actual political situations, in juridical de-
cisions, and in scientific discussions, there seem to be big differences
which often touch the very heart of the matter. Sometimes there are
open contradictions: both parties to an argument insist they are advanc-
ing the real democratic claim; judges contradict each other as to what
a certain democratic principle demands; criticisms of a scientific book
on democracy assert that "the book is scientific, to be sure, but you see,
what he is speaking of is not true democracy at all, he has not the right
idea of democracy." Various factors are involved, factors that differ in
Reprinted from Political and Economic Democracy, edited by Max Ascoli
and Fritz Lehmann. By permission of W. W. Norton & Company, Inc. Copy-
right 1937 by W. W. Norton & Company, Inc.
On the Concept of Democracy : 43
kind. Among them are certain logical-methodological problems, and it
is with these latter that this discussion is concerned.
The usual method of stating what democracy is, is the old traditional
one. Compare the object with other objects of the same class, compare
democracy with other forms of government, find out the similarities and
the differences, formulate them and you have characteristics, a number,
an "and-sum" of items which differentiate it genus proximum, dif-
ferentiae specificae. This method of isolating special items in subtractive
abstraction has its merits, seems often indeed to be the exact method,
but it has its dangers. Such an item is likely to be used blindly, with no
reference to the role it plays in the hierarchical logical structure; the
view is one-sided, the mental horizon artificially narrowed. Such an
enumeration of items is not enough. There is another method, which is
to investigate the structural function of the items and the hierarchical
structure of the whole idea. This method is certainly much more diffi-
cult. It is not so easy to decide such questions as it is to decide questions
in subtractive abstraction, but it is necessary.
In order to illustrate the formal problems that must be met, let us
consider some examples. These examples are given solely as illustrations
for the formal problems; the truth or falsity of their material content
will not concern us in this paper.
Comparing democracy with other forms of government we find, for
example, that in democracy the people shall vote and that the will of
the majority shall decide. This, we are told, is the democratic procedure.
But if we look more closely we may see that if we take "majority rule"
as an item for itself without regard to the function it has in democracy,
we are blinding ourselves, and sometimes others, to what it means.
Various things must be taken into account, first that the vote is meant
as a free vote, not a vote by threat or intimidation. This is not an acci-
dental addition but is logically determining in the structure. Secondly,
the voter must have access to as full and true information as possible,
and there must be free and open discussion. Again, this is not simply an
accidental addition. Thirdly, "Great issues cannot be resolved by count-
ing noses, but only by an appeal to what is right and what is wrong," is
a democratic statement. Majority without this tendency is not at all
democratic in the sense of the "good old democrats."
These requirements involve others. Full information for the voter is
one of the roots of the demand for free expression, which in this context
has the function of a technical means to the centering end. As a means
it is considered only the best available, not 100 per cent effective, which
44 : Max Wertheimer
is interesting from the logical standpoint. Moreover, in genuine democ-
racy free voting by personal decision is not meant simply as a means
for the voter to protect his private interests, the profit interests of him-
self or his group. It is a means for preserving his rights but his rights,
not his private interests. The voter is envisaged as responsible for a de-
cision that is right for the community, not simply for himself. There are
times in which rights and private interests may coincide, but the idea
as well as the reality of majority decisions between two profit-interest
groups is a caricature of the idea and of the attitude of old democracy.
Majority rule in a democracy also implies a characteristic attitude to-
ward minorities. Simply to deny or to violate the rights of minorities, to
blind oneself to their needs and their claims is not democratic. "We have
the majority, so what?" is not a democratic attitude. An unjust decision
against a minority because of majority-will is not democracy. Finally, a
defeated minority has not simply to bow, to recognize a decision as right
"because it is the will of the majority the majority having spoken, the
matter is settled." It is settled for the time being. But the minority has
not simply to give in. If the minority is convinced that the decision was
wrong or unjust, it becomes its duty to continue trying to clear the mat-
ter up, to find better arguments, a clearer presentation to help those of
the majority to see the truth.
Against all this one may argue that the democratic way to decide is
by majority, that of necessity the majority must eliminate aims and argu-
ments of the minority, voting against minority needs, tendencies, and
convictions. Since matters have to be settled and decisions have to be
reached, decision by will of the majority is the democratic way. Here
we reach a logically very interesting problem. Of course this is the demo-
cratic procedure. But this does not imply that the content of the ma-
jority principle is, taken in itself, a democratic principle. Some facts
seem to show that it is nothing more than a technical means toward
what is really wanted. Confronted with the necessity for a decision, the
wise democrat will not feel very happy if the proper aims of a minority
are brutally overridden by majority vote. In spite of seeing no better
practical way, he does not like it. It is no real solution for him. In itself
majority rule is by no means a democratic goal but only a technical
means, only technically a solution, the best that is available but far from
perfect. From the standpoint of logic it is not the content of the majority
principle which is truly democratic, but only its function as the technical
means to the real goal of more just decisions. It is not the will of the
majority that is wanted but the better decision. (The vote has another
On the Concept of Democracy : 45
source as well: the principle that man should not be subject to a ruler,
but himself responsible.) In order to understand such an item as the
majority principle we must not be satisfied with stating it by itself. We
must go on to the role it plays in the hierarchical structure of the whole.
Without this we fail to understand it at all.
One might add that the method of conciliation, in which one or more
representatives of two conflicting parties try to reach a just decision in
concrete discussions, although it does not count the noses of interested
party groups, is a democratic idea.
What has been said has consequences in actual real proceedings.
Men often act in the belief and the emphatic conviction that their way
is the democratic way, but it is only superficially democratic because
they are blind to functional meanings. In the meetings of democratic
bodies a member may not uncommonly be heard to say: "Why should
we discuss this matter at all? You are only a small minority. What the
majority wishes is clear. Let us proceed in the democratic way, let us
vote. And if you want to oppose calling for a vote on the measure now,
we will follow the rules and vote first on closing discussion. If you want
this formality, you can have it, but what's the use? You know we have
the majority in both cases." Often this is blind or contemptuous misuse
of what is really intended in democracy. Under the influence of a strong
democratic wave, to take another example, some people demand quick
introduction of the vote and speedy formation of a parliament. Others
emphasize the prior need of real democratic preparation for the vote,
the necessity for free propaganda to open the eyes of a people blinded
by a mighty, one-sided press. The first ones have their way and empha-
size that it is the true, democratic way. Finally during recent years some
men have been blinded by the idea that as true democrats they had
simply to bow to the will of the (alleged) majority.
Let us consider a second item which is interrelated with the first, as
nearly all items are in the conception of democracy. There axe discus-
sions, arguments, claims, court decisions in which the principle of free-
dom of the press is used as an item in itself, or in a one-sided connection.
There are instances in which the principle of freedom of the press is
used simply as a special case of freedom of business enterprise, of the
right of an individual to make profits. Combine it with the principle of
free speech, free self-expression, and, if only these two are taken into
account, the result may easily be emphatic assertion of the right not to
be bothered, not to be interfered with, not to be blamed for building
up a mighty chain of newspapers which by its business methods excludes
46 : Max Wertheimer
the possibility of nearly all other information, which is a tool for arbi-
trary "self-expression," one-sided information, and one-sided influence,
that may just happen to coincide with the selfish interests of oneself or
one's group.
Certainly the principle of freedom of the press is very important in
democracy, but its meaning is not meant to be restricted to conse-
quences for the individual and individual rights. It also has the social
function of providing the public with better information, if for no other
reason than because information is indispensable to voting in the true
sense. The underlying idea is that many will use self-expression because
all have the right to it, that the result will be better information because
numbers will help against bias, etc. Democracy requires that the people
shall know what is important in order to vote justly. This is a vital func-
tion of freedom of the press, but it is very different from the meaning
determined only by the two above mentioned principles of freedom of
business enterprise combined with freedom of expression. In this con-
text freedom is a means to an end.
(Historically the idea of freedom of the press is connected with lib-
eralism's optimistic view of laissez faire. It may be that dissatisfaction
with means for public information will bring about new policies directed
toward providing avenues of self-expression for those popular groups
that have hitherto been deprived of organs of their own.)
Many of the assertions that have been made may be questionable.
Their truth or falsity is a matter for historical and other investigation.
Here they are used simply as examples for the logical-methodological
problem. Each example shows, I think, the big difference between tak-
ing a single item as an item in itself or in one-sided determination, and
trying to understand it in its function in the whole structure. These
questions must be asked: How are such items interrelated, how deter-
mined, how centered? What is their role and function, their functional
position? Which items are central, which peripheral is their content
determined by other items, by which items, and how? We have to envis-
age them in their place in the hierarchical structure of the whole.
If we look at the different items by which democracy is usually de-
fined as a sum, or better, if we look at the manifold of items included
in democracy, there are at first two logical possibilities, viz., that all
these items or some of them are in an "and-sum" with some interrela-
tions, or that we have to deal with a hierarchical structure in which the
items are to be conceived as parts in their relational place, in their func-
tion in this whole. This question is different from the question of the
On the Concept of Democracy : 47
number and variety of historical sources, causes, etc. Certainly these
are important, but to look for the various items that may be at work is
a different problem from asking, "Is there an 'and-sum' or a structure?"
One might put as a possible question: "What is the heart of the mat-
ter? Is there a structural center?" Various hypotheses are possible. Sci-
entifically they have to be chosen not arbitrarily or one-sidedly, but
viewing the whole material in order to do justice to the manifold inter-
relations, in order to discover the structural center. Different hypothet-
ical centers are to be studied and followed through the relational net-
work, then compared for their merits. I will confine myself here to
sketching only some of the steps in connection with one hypothesis to
illustrate the structural problem.
If we look at the real beginnings of the great democracies take for
example the happenings in the United States and in France the main
point seems to be not opposition to the kingdom, to the king as king,
but opposition to injustice, a wish to avoid injustices done by the king,
both for oneself and the community, a wish not to be subjected to arbi-
trary, unjust commands of the ruler. As a hypothesis we may try to
conceive as the center the wish to create and to assure a more just
procedure, to get decisions and rulings that are not arbitrary but di-
rected by reason and justice; which means that rule by oppression,
violence, and trickery is opposed, that an open, honest way of procedure
is sought. It appears then that the will to change the form of government
is not at all primary.
It would be a logically secondary step toward better realization of
the primary aim, inasmuch as kingdoms have the greater danger of
arbitrary, unjust dealing, of dealing not determined by reason, justice,
the common good. Moreover, in order that men unjustly dealt with, for
example, may have the right to be heard, the possibility to appeal, to
participate in decisions, the concept of government by the people, of
parliament, is born. The principle of justice and reason as opposed to
arbitrary dealing is not only chronologically earlier than the people-
principle against the king-principle, it is likewise structurally primary,
central in this hypothesis. In this structure items like government by
the people, voting, majority rule, etc., etc., are to be conceived then as
secondary items determined by the center and their content must be
understood in the light of the centering idea the vote, for example, as
the enlightened vote of the people. This kind of vote, by reason of the
central idea, requires again as means to the end, furnishing open infor-
mation, permitting free speech, etc. Similarly with other items; the idea
48 : Max Wertheimer
of division of powers is, for example, likewise logically determined by
this central idea. Working this out in all the ramifications it becomes
clear that most of the characteristics, most of the institutional ideas,
of democracy are consistent if viewed in the light of the central idea of
justice and reason, supplemented by faith in the people, the idea of
homo sapiens, etc.
The problem of centering gives rise to such questions as these: What
is the structural place of the decidedly important item of autonomy of
the individual, participating in decisions, voting freely out of his con-
victions, the individual of the control social? It is possible to conceive
this as another main point coordinate with the first we mentioned, as the
centering idea instead of the first, or as secondary to it. They are some-
what different, at least in the emphasis. If we take as the center the
"inborn rights of the individual" in the directions indicated we get a
slightly different picture. I will say only briefly that studies of the inter-
relations of the two principles seem to show the principle of autonomy
as structurally secondary, an outcome from the first principle in a special
direction, just as the principle of equality, equality before the law, etc.,
seems a special outcome from the first principle.
Another problem arises as to the content of "liberty," of "noninter-
ference" as understood by liberalism in its connection with the harmony
theory of laissez fake. This idea has similar formal significance both in
political and economic respects (cf. the Boltzmann principle in phys-
ics). Perhaps we really have an "and-sum" here. These ideas certainly
include some features that are logically strange to the first principle,
"arbitrary liberty," for example, noninterference with "arbitrary" free-
dom of enterprise, etc., but certainly there are likewise inner connec-
tions, features that come very near to the first principle of justice and
reason. We cannot deal here explicitly with this complicated problem. I
shall mention only one point, that the will to courageous truth, to
objective reason, to just decisions, requires freedom of the man and of
his mind, but this does not involve the harmony theory of laissez faire
in enterprise, etc.
Irrespective of these and other features there appears to be a logical
structure of democracy with a hierarchy of parts. It is striking to see
how the different special items, the different points in the picture get
their meaning in their place, in their role, in their function as parts in
the picture along with the ethical and educational aims of democracy,
the will to truth, to open-mindedness, to fair play, to honesty, etc.
Viewed in this way the real essence of democracy seems to be not a form
On the Concept of Democracy : 49
of government, a sum of institutions, etc., but a certain real attitude in
life, behavior of a certain kind, not only in state matters but generally
in relations between men. This attitude has some characteristic similar-
ities to the role of the judge or the juror, rather than to the fighting of
interests. The state is viewed not as a governing body, but as the guar-
antor of justice and reason which has not to create law but to fulfill it,
to realize it by making tha rules.
It seems necessary to work out such schemes clearly for the different
part-items, their interrelations and their determinations as parts in the
scheme, to try to test the conclusions as to the structural function of the
items. There are methods for testing such hypotheses and for comparing
different structural hypotheses. Structural theses, structural centers,
are often blindly established. We often encounter statements and argu-
ments determined simply by artificially narrowing the mental field,
viewing parts as if they were in themselves the important thing. Cer-
tainly some court decisions, some political attitudes and arguments
would run otherwise if they were not determined by viewing items one-
sidedly, severed from their function in the whole.
Some theorists may ask the reasons for such investigations into the
logical structure of these things. In their view they are all secondary
things, "ideology," "rationalizations," "we should look for the real
forces behind them." I will not discuss this view here. I will merely
remark that they are certainly not only ideas; they live in the real atti-
tudes and actions of men. Moreover in studying the "real forces behind"
the ideas it seems equally necessary to see clearly what the ideas are and
how they are structurally related. Finally, similar formal problems recur
in the study of the forces themselves.
Another point must be added. The main point in our deliberation
has been the difference between an item seen in itself or in one-sided
determination and an item envisaged as part of the hierarchical struc-
ture, democracy. We have dealt with items within political democracy.
The question is repeated when we envisage, as we must, the structure
of democracy as a part functioning in the broader structure of the social
field. Stt ucturally democracy looks somewhat different as a part in var-
ious broader fields.
Suppose that our thesis is right in its main lines. Suppose that de-
mocracy is a hierarchical structure and that it has been realized to a
certain extent. Now this political democracy is to be viewed not as a
structure in itself but as a part in the social field, in the larger whole.
Let me give some hints on this next step, simplified in order to show
JO : Max Wertheimer
only the structural problem. Think of the frontier period. We may envis-
age it as a kind of social field in which the conditions of production, of
economic life, of the possibilities for the individual, etc., resulted on the
whole in rather good functioning, in mutuality, in a kind of equilibrium,
if for no other reason than because of the wide-open possibilities for
all men. Logically and structurally the conditions of life fitted well with
the attitudes, aims, regulations of democracy. What we may call the
"part-system" democracy, including laissez f aire, fitted in well, supplied
answers to the questions which arose, dealt with them in a satisfying
way. Now think of an important change in the social field in which
democracy is functioning as a part. Think of big changes in the eco-
nomic "part-system," the development of big capitalistic forms, indus-
trialization, masses of workers, depression, masses of unemployed, etc.
Strains and stresses are born. The system democracy has to face new
problems. Disequilibrium, strain, stresses in one part of the field, e.g.,
the economic, are not simply irrelevant for the "part-system" democ-
racy. Problems of the whole dynamics of the whole system arise.
In the frontier period the Boltzmann principle functioned easily for
the most part. Not only did the individual have his opportunities, the
other fellow had his and it was consistent to feel that if you satisfied your
private economic interests you were at the same time performing the
best service for the common good. But when big changes take place in
important parts of the field, it is no longer simple for the system of de-
mocracy to function as well as a part-system in the broader field. Ten-
sions arise with regard to this "part-system." There are new problems.
The old idea combined with the laissez f aire principle cannot so easily
overcome the difficulties and fit into the new field. Comparatively it
was much easier to function unquestioned in the former field of the
frontier period. Here the point is to envisage the functioning of a system
as a part in a broader system, to study the systematic consequences that
follow for the "part-system" from a changing of parts in the broader
field.
After the change in the broad field, democracy continues to function,
but under strain. The items in it get a somewhat different meaning
structurally. Freedom of enterprise, freedom of contract, for example,
mean something quite different under the new relations between em-
ployer and working masses. To put the extreme case, it is one thing to
have freedom of enterprise and freedom of contract work in real
mutual freedom, it is another to have them work as one-sided tools. The
principle of noninterference if it does not result in real mutuality ac-
On the Concept of Democracy : 51
quires a new meaning. A difference arises between a purely political
democracy and a democracy which includes the now changed realities
of life. Logically the content of some of the old items was not concerned
with certain items of reality because these were functioning well, were
giving mutual equilibrium. Now these old items are confronted with
new realities which must be taken into account in order to re-establish
the democratic principle. Holding rigidly to these items in their old
meaning without facing the new conditions, without asking what is
demanded in the light of the very principles of democracy, implies some
blindness. There are dynamic systemic demands in the new situation
which must be met so that democracy may again become a functioning
part in the whole system of the social field.
With regard to this problem there are different attitudes which mean
structurally different things. The first is: Wait, we have nothing to
change; we have to hold to the old meanings rigidly; the weather will
change, conditions will soon improve, the old meanings will again
function with ease. The second is: We have to improve various items
in order that the system may again work consistently with the demands
of the inner meaning of democracy; "real democracy needs more than
the old political and individual freedom," which is now only superfi-
cially freedom. The idea of social consciousness arises from a new mean-
ing of the items within democracy. The third attitude is: Democracy
is denied as not adequate. Some mean to suspend it temporarily in order
to make possible real democracy after a time. Others deny the right of
democracy to exist at all and set up utterly different principles. It would
be better to have the proponents of these different attitudes look at the
structural problem in a logically consistent and logically honest way
instead of arbitrarily and artificially recentering and narrowing the
logical field.
To summarize, stating and discussing items in subtractive abstrac-
tion is not enough. We have to consider their structural function. The
methodological approach is not only to compare different forms of
government by comparing items taken in subtractive abstraction, but
to study the inner structure of the object and to view the system struc-
turally in its functioning as a part in its field.
Max Wertheimer
LATE OF THE GRADUATE FACULTY,
NEW SCHOOL FOR SOCIAL RESEARCH
A STORY
OF THREE DAYS
I shall report what happened in the course of three days to a good man
who, facing the world situation, longed for a clarification of the funda-
mentals of freedom.
He saw: ideological devaluation of freedom had spread; freedom
in the humane meaning of the word was proclaimed false, outworn, use-
less; and the radiance of the old idea was often exploited for other ends.
Some men seemed to have lost sight of it entirely, without realizing what
they had lost. Confused by the complexity of actual situations many
became uncertain, basically unclear with regard to the very concept
of freedom, its meaning, value, actuality. Even men who loved freedom
deeply often felt helpless in the face of actual arguments. So it was
with our man; not that he felt uncertain in many or most of the concrete
issues; but he felt impelled to reach a fundamental clarification. What
at bottom is freedom? What does it require? Why is it so dear to me?
He was a humble empiric, open-minded, thirsting for information.
Of course, those three days of his search were only a beginning for
From Freedom: Its Meaning, edited by Ruth Nanda Anshen, copyright,
1940, by Harcourt, Brace and Company, Inc., and reprinted with their per-
mission.
A Story of Three Days : 53
him. He touched only some of the issues involved, for it was by chance
that he met just those men and read just those books. They represented
only certain points of view and the discussions were by no means ex-
haustive, yet I think that what he experienced was in many respects
characteristic, typical of some fundamental trends in actual thought.
Those were dramatic days for him, in which he became more and
more bewildered, but at the end of those three days he felt that he had
gained some clarification, that he now saw more clearly something that
only his heart had told him before.
He sought out a sociologist who was immersed in studies of this very
problem, and he asked his question. The sociologist was very kind. He
told him about the investigations of modern sociology, about the history
of societies, how ideas of freedom had developed in them and what
freedom had meant to them; he told him how different were the ideas
of freedom and the ways of realizing them, etc. Our man was fascinated
by the richness of what he heard. He felt that here were men with an
honest, sincere approach; these were serious studies, and he became
more and more hopeful. "You are the right man," he said. "I am sure
you feel as I do in the actual world situation," and he told him how he
felt.
"I share your feelings," said the sociologist. "I too hold proudly and
passionately to our traditional values."
"But why?" he was asked. "What is it that makes freedom so dear
to you also, and what is freedom essentially?"
"I am at one with the traditions of our people," said the sociologist.
"But if you ask me about the fundamentals, I must answer: It became
more and more clear in our studies that the standards, the evaluations,
the goals, that an individual has are shaped, determined, by the social
group, the society of which he is a part. Different periods in history,
different societies, different nations, have different views. Ethical stand-
ards are relative."
There was a long pause. After a time our man asked in a low tone:
"Is that all? Should what these others assert be true? Are our ideas of
freedom merely the historical standards of a certain time, now perhaps
outworn? Are there no fundamental standards; are the requirements of
freedom a fairy tale?"
"No fairy tale," said the sociologist, "but developed in and character-
istic for certain historical, cultural, and social settings."
"And nothing more?" asked our man. "Is no decision possible among
54 : Max Wertheimer
various systems? Are there no features that are basic in men with regard
to questions of freedom, no requirements for men, as men should be?
No features that are desirable, required in human society?"
"Here you are touching upon very difficult things," said the sociol-
ogist. "Some of my friends would say that the fight for freedom was al-
ways a fight against certain concrete restraints or compulsions and
meant, necessarily, different things in different times. Society in its
rules and institutions necessarily permits freedom, imposes restraints
that with time change in different directions. There are no axioms which
would allow me to speak of fundamental standards. To speak of 'the
man' or 'the society' is only a pale abstraction."
Our man became more and more bewildered. "Was this," he said,
"what your friends wrote and taught? And was this not one of the factors
in the developments we now face, one of the factors that paved the way
for political leaders proclaiming new and other national or racial ethics,
willfully and efficiently?"
"Do not overestimate the role of the opinions of sociologists," said
the sociologist quietly. "I told you that this is the position that most of
my friends take. And certainly they were sincerely driven to these con-
clusions by their findings, which contain great factors in their favor. We
cannot lightly dismiss them. I myself would not dismiss your questions
with their answer. I feel that these are genuine questions; that as sociol-
ogists we must face not only diversities in various cultures but also must
seek for fundamentals, for identities in the requirements of man and in
the dynamics of society in a doctrine of man and in a doctrine of
society. There have been approaches in this direction. I feel as you do
that in this context the problem of freedom will play a genuine role. But
these are scientific Utopias, my friend; we are far from any real insight,
far from even a real method of approach. There are some young soci-
ologists who are groping in this direction and grappling with the prob-
lem.
"But if you ask for a definition of freedom, not in terms of the full
reality of a specific society, which, of course, I should prefer to give you,
then my answer would have to be: absence of restraints, of compulsions,
of external hindrances from doing what one desires to do, and maybe
I should add absence of imposed internal inhibitions. . . . Though I
might say that such a definition certainly lacks concreteness."
Our man thanked the sociologist. He felt sad, puzzled, bewildered.
He came home, sat down, and reached for one of the books he had
ordered for his search. It was a novel by a famous writer of 1936. He
A Story of Three Days : 55
was too disturbed to read thoroughly. A certain page caught his eye,
He read, more and more excitedly, these sentences:
Anthony . . . turned over the pages of his latest notebook . . . he began
to read.
"Acton wanted to write the History of Man in terms of a History of the
Idea of Freedom. But you cannot write a History of the Idea of Freedom
without at the same time writing a History of the Fact of Slavery. . . .
"Or rather of Slaveries. For, in his successive attempts to realize the Idea of
Freedom, man is constantly changing one form of slavery for another. . . .
"Abolish slavery to nature. Another form of slavery instantly arises.
Slavery to institutions. . . .
"All modern history is a History of the Idea of Freedom from Institutions.
It is also the History of the Fact of Slavery to Institutions. . . .
"Institutions are changed in an attempt to realize the Idea of Freedom.
To appreciate the fact of the new slavery takes a certain time. . . .
"The honeymoon may last for as much as twenty or thirty years. Then
... it is perceived . . . that the new institutions are just as enslaving as
the old. What is to be done? Change the new institutions for yet newer ones.
. . . And so on indefinitely, no doubt.
"In any given society the fact of freedom exists only for a very small num-
ber of individuals. . . . For them, institutions exist as a kind of solid frame-
work on which they can perform whatever gymnastics they please. . . ."
Anthony shut his book, feeling that he couldn't read even one line more.
Not that his words seemed any less true now than they had when he wrote
them. In their own way and on their particular level they were true. Why
then did it all seem utterly false and wrong?
"Utterly false and wrong," our man said passionately. How was it
possible at all, he asked himself, for a man to formulate such asser-
tions! What he had read seemed unbelievable. At the same time he felt
strangely reminded of remarks he had encountered in the last years on
one or another occasion, for which these unbelievable formulations
seemed somehow fundamental. Now his longing for clarification
changed into a passionate drive. I must, I must see through all this.
Somehow it is a strange distortion to view the facts in this way seems
to press them into a blind and wrong direction. What is it that is wrong
in the fundamentals of this picture?
He took up the next book. It was a book from the year 1928 by a
famous psychoanalyst and dealt with culture. He read it through from
beginning to end. Again and again he turned back to some basic formu-
lations in it. There were some remarks of another character added here
and there, but in the main those formulations seemed to him nakedly
56 : Max Wertheimer
to express basic assumptions which led straight to those bewildering
passages he had encountered in the novel.
. . . every culture must be built up on coercion and instinctual renunciation.
. . . abandoning coercion and [abandoning] the suppression of the instincts
. . . would be the golden age, but it is questionable if such a state of affairs
can ever be realized. . . . the psychical sphere of culture . . . frustration
. . . prohibition . . . privation . . . the instinctual wishes that suffer un-
der them are born anew with every child.
. . . Such instinctual wishes are those of incest, of cannibalism, and of
murder.
... It is in accordance with the course of our development that external
compulsion is gradually internalized.
. . . Every child presents to us the model of this transformation; it is
only by that means that it becomes a moral and social being.
. . . Those people in whom it [the internalization of external compulsion]
has taken place, from being foes of culture, become its supporters.
. . . [but] a majority of men obey the cultural prohibitions in question
only under the pressure of external force, in fact only where the latter can
assert itself and for as long as it is an object of fear. This also holds good for
those so-called moral cultural demands.
. . . We have spoken of the hostility to culture, produced by the pressure
it exercises and the instinctual renunciations that it demands. If one imagined
its prohibitions removed, then one could choose any woman who took one's
fancy as one's sexual object, one could kill without hesitation one's rival or
whoever interfered with one in any other way, and one could seize what one
wanted of another man's goods without asking his leave: how splendid, what
a succession of delights life would be!
. . . [but] only one single person can be made unrestrictedly happy by
abolishing thus the restrictions of culture, and that is a tyrant or dictator who
has monopolized all the means of power. . . .
"Could this be true?" our man exclaimed. "Is this Man? Society?
Freedom? Is freedom lack of restraint of 'instinctual impulsions/ ex-
ternal or internal? Is Man essentially so determined, impelled by fear
of punishment or by habits, by internalized rules imposed on him by
compulsion?"
"I must see a philosopher!"
He went next day to see a philosopher and asked, "Will you tell me
please what freedom is, philosophically?"
The philosopher smiled. "This," he said, "is an old and famous topic
of philosophy down through the centuries. If you like, I can give you
A Story of Three Days : 57
the names of a great number of books which you can study are you
interested in the history of philosophy? There are a number of philos-
ophers who still deal with these questions, but if you like, I can try to
tell you briefly how the problem lies in modern philosophy as I see it,
and, I may say, as it has been well established in modern philosophy.
"The concept of freedom, of free will, of free choice, played an im-
portant role in various religions and in various philosophies. It was wish-
thinking. Modern developments in science and philosophy have shown
that there are no free acts. Causality governs them or, as we formulate
it, all actions take place under the principle of determination, are de-
termined by their causes; there is no such thing as an action leaping
into existence uncaused, and so what is going to happen, happens by
necessity. It is mere blindness if men believe that they are free to act or
to make decisions without realizing that their actions are the necessary
outcome of forces which determine their choice.
"You might look into the modern textbooks of psychology. In most
of them you will not even find mentioned such terms as free will, free
decision, etc.
"There have been discussions about this principle of determinism.
Some tried to save the old, outworn ideas by trying to defend a kind of
psychological indeterminism. But there are few who would still hold
these views to be defensible. There are some philosophers nowadays
who believe that the newest developments in physics, viz., the uncer-
tainty principle and statistics of probability, are again giving a foothold
to indeterminism. But one should not misunderstand the meaning and
role of these concepts in modern physics: they may make for some un-
certainty or chance happenings but they give no basis for the existence
of free will."
Our man lapsed into deep thought. "I think," he said, "I realize that
important consequences are involved in this philosophical discovery
of determinism. In looking, for example, at a man who has committed
a crime, we should not forget to look for the causes which made him
commit it. And we may find that his deed was due to factors which were
beyond his control. We must try to understand his deed from the factors
of causal necessity."
"Yes," answered the philosopher, "but don't forget that it is not only
in cases in which you may discover an external force that compelled
him, but also in cases in which it would have been said in olden times
that he acted of his own free will, on his own decision, with nothing ex-
ternal to compel him. Such a description is utterly superficial. A man is
55 : Max Wertheimer
determined even in these cases by the set of causal forces within him,
by his desires, instincts, acquired habits."
"Is there not this important factor," asked our man humbly, "that
man, after all, in a situation which calls for decision does not know of
the forces that will determine him and, therefore, practically will have
to choose, to decide? That everything is in fact determined may be of
value to someone looking into the past, after the decision has been
made, after the deed is done, but not before? And so the principle of
determinism does not perhaps do away with the questions of free deci-
sion."
"There are some," said the philosopher, "who try to make use of
this factor of past and future for our problem, again in connection with
new developments in modern physics. But don't you see, this does not
help indeed this may be the very reason why man is deceived about
himself, why he may appear to himself as free, which is nothing other
than that he does not know how in his seemingly free decisions he is
lawfully and by necessity determined by causes."
Our man felt uneasy about this answer, but, unable to clarify the
issue, he proceeded with another question. "Aren't those ideas of de-
terminism somewhat dangerous?" he asked. "I should guess that a man
who really comes to believe in determinism and to act sincerely in ac-
cordance with this belief would not only change his philosophical opin-
ions, but his very actions. He would become a fatalist, relieved of all
troubles in facing a situation that calls for a decision ... it will hap-
pen anyhow. . . ."
"True," said the philosopher with a sly twinkle. "But fortunately men
believe in their will, and even if they are philosophically convinced of
determinism, they will not make use of it in actual situations. On the
other hand, you may see in your remark a profound confirmation of the
very principle of determinism: even your belief or disbelief in the prin-
ciple may be a determining factor."
Suddenly our man jumped up from his chair. "Now," he said ex-
citedly, "permit me another question. If we state that all is determined,
does this change anything in regard to the real problems of freedom
(with the only possible exception of this problem of the realization of
fatalism)? Suppose we attach to every deed, to every action, to every
attitude, the quality, It is determined, 5 would not all real concrete prob-
lems of freedom remain just the same? The discussions between deter-
minism and indeterminism do not touch the real problem, in fact they
obscure it. Should the essence of free action be that it is in no way deter-
A Story of Three Days : 59
mined? Or if all actions are determined, that there are no free men?"
"Let us not mix up such practical problems with the philosophical
issue," said the philosopher. Here from the fullness of his heart our man
told the philosopher about his troubles, facing the world situation, about
his meeting with the sociologist and about the formulas in the books
he had read.
Said the philosopher, "Like you I am a lover of political freedom.
Certainly there is the very important problem of how much the State
should or should not restrain the freedom of individuals. These are
questions with which the sociologists and men of political science may
properly deal; but don't you see that the very foundation of all that you
have told me about the sociologist and the formulations in the books
is the modern discovery of determinism, of realizing it as basic in all
these questions?"
Our man realized this and was more bewildered than before.
The next day he said to himself, This is what I have learned:
1 . There is no freedom because all is determined, is the consequence
of causes. Or,
2. Freedom is absence of external restraints, of compulsion, free-
dom to pursue whatever wish may come to one's mind. Or,
3. Because such wishes may be due to whatever standards may have
been internalized on the basis of compulsion, freedom means to be able
to follow those instinctual impulsions without inhibitions.
Suddenly all he had heard in this context seemed to him utterly
strange, narrow, inadequate; superficial, oversimplified, wrongly di-
rected, blind to all the real problems of freedom, appropriate neither
to the nature of man nor society, out of focus on both. He felt the desire
to get away from all these terms and definitions, he wanted to face again
the real situation, to restate the problem in full view of life.
He first thought of what the sociologist had told him and soon felt
lost in the manifold features of history, its complexities, its diversities.
"First let me realize," he said passionately, "what I have seen with my
own eyes. Have I not seen in my experience strong and indeed very char-
acteristic cases of men, of children, who were free, who were unfree?
What were the essentials? My experiences, of course, are no sufficient
basis for statistical generalizations; nor do I wish to make any now.
What I want is to grasp, to realize, what I have seen."
60 : Max Wertheimer
He recalled a number of cases. Then he said, "Sometimes one sees
a man, and by the way he goes through life, by his attitudes, by his
behavior in dealing with life situations one feels: this is a free man, he
lives in an atmosphere of freedom. And so in observing children.
"On the other hand, one sees men or children, and feels strongly:
in their behavior there is no freedom there is no air of freedom in
their world."
It is, he thought, not easy to put into words what one faces so vividly
in these extreme cases. Let me think what were these cases concretely?
The free man, he recalled, frank, open-minded, sincerely going
ahead, facing the situation freely, looking for the right thing to do and
so finding where to go.
The opposite he first thought of children he had often seen inhib-
ited, pushed, or driven, acting by command or intimidation, one-track-
minded, chained to certain ways of acting and of thinking, even in
viewing situations the very curves of their actions, of their movements,
often showed these features, especially in meeting new situations. They
often looked like sorry products of external influences or like slaves of
any desire that might have come to their minds. Often they looked like
robots, somehow crippled, robbed of essential abilities, narrow-minded,
stiff, rigid, mechanical, their movements and postures often had the
effect of puppets on strings. And grownups still more so. (Even slave
drivers he had seen such in our times were they free? No, they be-
longed here.) Of course, many thus enslaved did not overtly behave
timidly at all just the opposite, brutal and overproud. But one sensed
the same unfreedom, sometimes one saw what happened when they had
to face a new situation in which their coat of armor was futile. . . .
And what experiences he had had in observing transitions!
If a child, if a man, having lived as that kind of slave, came to live in
another social field in which there was the real air of freedom, what
marvelous happenings had he not observed in such cases! Very similar
indeed to regaining health after a long illness.
Suddenly the whole problem appeared to him to be no longer a prob-
lem of philosophical schools of whatever standards or evaluations, but
a problem of hygiene it seemed to require the biologist studying health
conditions. This is a task of scientific investigation, he thought. But
not in terms of those previous theses. What conditions, what institutions,
make for the free? What for the unfree? And what price is paid in the
change?
This, it was now clear to him, was not to be viewed piecemeal, in
A Story of Three Days : 61
terms of a choice, of a wish, of an "instinctual impulsion," etc. One's
whole attitude towards the world, towards the other fellow, towards
one's group, towards one's own momentary wishes was involved. And
suddenly those theses dealing only negatively with freedom appeared
to him like saying that growth, that maturing, is absence of impedi-
ments to growth; that beauty is absence of ugliness; that good think-
ing is absence of mistakes; that genuine achievement is due to ab-
sence of inhibitions; that kindliness, or friendship, is nothing but
absence of hostility; that justice is any legal rule imposed arbitrarily.
"What we face," he said, "is not a problem to be dealt with in such a
piecemeal, negative way."
After a while he found himself thinking of his experiences in certain
specific situations in which there was clearly the one or the other kind
of behavior. He recalled discussions. What differences! In the way a man
faces a counterargument, faces new facts! There are men who face them
freely, open-mindedly, frankly, dealing honestly with them, taking them
duly into account. Others are not able to do so at all: they somehow
remain blind, rigid; they stick to their axioms, unable to face the argu-
ments, the facts; or, if they do, it is to avoid or to get rid of them by some
means they are incapable of looking them squarely in the face. They
cannot deal with them as free men; they are narrowed and enslaved by
their position.
For a moment he himself objected, Why are you connecting the issues
of freedom with all these features? With questions of being blind or
narrow-minded in contrast to facing situations with open eyes and deal-
ing with them honestly? Yes, he decided, I must; these things are most
closely and intimately correlated with the meaning and the facts of
freedom.
How was it in history, in the times when people honestly fought for
freedom? Those men fought against the arbitrary, willful acts of their
governments, they fought for fair and honest dealing. To those men
freedom was envisaged and endeared in these terms. Freedom was
sought and longed for not in terms of being able to do whatever might
come to one's mind, to act in as one-sided and as blind a way as one
might wish, to be free to brutalize the other fellow willfully. Were not
those praisers and lovers of freedom those very men who demanded
enlightenment for everyone, who fought for just dealing in courts, and
just laws?
Thinking of the three theses he had written down earlier, he felt as
if the scales had fallen from his eyes. The real question was, what kind
62 : Max Wertheimer
of attitude, what rules, what institutions make for the free, what for the
unfree? The real problem is not as in thesis (1), which seemed to say
that all determination, all causes and influences, are factors against
freedom; the problem is which ones are? This is a matter of causes and
consequences; some make for freedom in men, some for unfreedom!
"What nonsense!" he said. "If a man is blind, or sees things in a
distorted way and you open his eyes, give him knowledge, make him
see, you may thereby strongly influence him, change him, determine
him, but are you thereby limiting his freedom?
"And do not men have a healthy desire not to be blind or blinded, at
least in the long run? Are there not, thank God, some tendencies of this
kind in men? And in the dynamics of society?"
Thinking of theses (2) and (3), he said to himself, There is some-
thing in formulating freedom as absence of restraint, of compulsion; a
price is paid when spontaneity, genuineness, are impaired or destroyed.
Yet the very term compulsion means willful, arbitrary force. And spon-
taneity, genuineness, are certainly not adequately viewed in terms of
"whatever wish may come to one's mind" or in those "instinctual im-
pulsions." What he had read about happiness was not happiness, was a
crude caricature of happiness.
The assertions that "cultural institutions by necessity restrain, limit,
freedom" now appeared to him astoundingly superficial. Is limiting
freedom the essence of institutions for true education? of the roads that
society constructs? Likewise of the development of law and of courts
if understood not in terms of any arbitrarily imposed law, but of making
possible some degree of confidence in fair, just dealing? Is it not sheer
piecemeal thinking to say "restraint is restraint," if a kidnapper re-
strains, imprisons, a child in order to extort ransom, and if another
restrains the gangster from doing it in order to help the child? Is there
not in the very birth of cruelty, of brutality, the factor of being blind,
of being narrowed down?
And are there not tendencies in men and in children to be kind, to
deal sincerely, justly with the other fellow? Are these nothing but "in-
ternalized rules on the basis of compulsion and of fear"? He thought
of children whom he had seen grow how little did this blind sweeping
generalization apply to their kindness, to their desire for real grasp, to
their horror in the face of an act of brute injustice.
"What is needed," he said, "is a sincere study of the tendencies, the
vectors, their development in children, in men, in the dynamics of soci-
ety, but not in terms of such rash definitions, or of those Instinctual
A Story of Three Days : 63
impulsions,' assumed in blind generalization. These are tasks for em-
pirical study in the same way that problems of philosophy have become
problems of modern science. Old theses, dependent on the philosophical
school to which one adhered, should now be studied, discussed in sci-
entific investigations. To be sure, superficial statistics will not help; these
are deeper questions, involving the dynamics of men, of society. And if
these fine tendencies are often weak, if their awakening, their growth,
are often endangered, or if they are wholly overcome by other forces,
does this justify constructing substitutes on the basis of their very op-
posites, or overlooking them, denying them entirely? There was some
positive development in this direction. It needs help."
Marvelous tasks for investigations! he thought.
Then again he found himself thinking of the actual world situation.
In full view of it, of the actual happenings, he reread the three theses
he had written that morning. The whole line of approach appeared to
him cruelly to miss the issue by focusing on "whatever wish may come
to one's mind," and on those "instinctual impulsions." Was this the
issue? (Probably it is just blind restraint that breeds and feeds such im-
pulsions.)
Here are the basic issues, he felt, instead of in those three theses:
That human beings are exposed to injustice, to willfulness, to brutal-
ity; robbed of any hope of being treated with fairness, with kindness;
that institutions are destroyed which had slowly developed, guarantee-
ing some justice, some fair dealing.
That men are forced to keep silent in the face of acts of injustice, with
no possibility of helping the victims; forced even to help in performing
those acts against their will and better knowledge.
Still more, that men, even children, by willfully distorted information
become narrowed down, poisoned in their very souls, robbed of the
preconditions of free judgment through being blinded, robbed of what
in man and society is humane.
Now he felt more clearly why freedom was so dear to his heart.
What he had reached, he felt, was only a start. He saw that there are
other problems to be faced; problems of the physical, economic con-
straints of men by hunger, dire lack of means of subsistence; problems
of real cooperation (oh, what he had gone through were not prob-
lems of piecemeal individualism) ; problems of mutual justice between
groups; problems of the individual called as a member of his group not
only to cooperate in performing, but in facing and judging the very
goals; etc. But in all these as in other urgent problems what he had
64 : Max Wertheimer
gained did not seem useless. The task he felt was to face these problems
also with the attitude of the free man, productively, sincerely; real help
he felt would come only this way.
He was eagerly looking forward to the further steps.
Then he took his notebook and wrote down after the three theses:
"Logical remark. This is what I have gone through, logically: In these
three theses freedom is viewed in a piecemeal way and defined as a
thing in itself, cut off from its living role and function, basically merely
negative. Freedom is ( 1 ) a condition in the social field, and a terribly
important one. In viewing such a condition we should not view it as a
thing in itself and so define it, but we should view it in its role, in its func-
tion, in its interactions, in its consequences for men and for society.
Freedom is logically (2) not just a condition; what matters is how men
are and how they develop, how society is and how it develops. Freedom
is a Gestalt quality of attitude, of behavior, of a man's thinking, of his
actions. (Think of the difference between the free and the unfree, the
description of which was of course only a first approach to viewing the
essentials.) Now logically freedom as condition (1) and freedom as
Gestalt quality (2) must be viewed not as two pieces, but in their inti-
mate interrelation. Freedom as condition is only one factor, but a very
important one with regard to freedom as character quality. To put a
man (or even a dog) in chains has consequences. Some men to be sure
remain free in their hearts, even in chains, waiting for the moment to
throw them off. But there are men whom chains enslave to the core. And
here in the interaction between freedom as condition and freedom as
character quality, one understands the real meaning of brute restraint
and compulsion the consequences for the victim and for the oppres-
sor."
What matters is not a rash and elegant definition, but really facing the
issues.
PART II
GENERAL THEORY
Wolfgang Kohler
DARTMOUTH COLLEGE
PSYCHOLOGY
AND EVOLUTION
In the psychology of human perception it is customary to distinguish
between Empiristic and Nativistic interpretations of given facts. What
do these terms mean? The answer seems obvious: an Empirist favors
explanations of perceptual facts in terms of learning, while a Nativist
prefers explanations in terms of inherited mechanisms which are located
in the nervous system. 1 If there are any major difficulties in the Em-
pirisf s way of theorizing, such difficulties will not be discussed in these
pages. Rather, I propose to analyze what is commonly supposed to be
the Nativist's program; because the characterization of this program
which I have just mentioned, and which would probably be regarded as
Reprinted with permission from Acta Psychologies Vol. 7, Nos. 2-4, 1950.
lr The term "Empirist" may sound unusual to English and American psy-
chologists. But we need a special name for theorists who tend to give bewilder-
ing facts in perception an interpretation in terms of learning. The word "Em-
piricist" does not refer to such theorists. A philosopher of the Empiricist
school might actually prefer Nativistic to Empiristic theories of perception;
his principles do not compel him to prefer the latter. Again, philosophers who
are Rationalists rather than Empiricists often reason in the Empiristic fashion
when they deal with problems of perception. We should therefore confuse two
different issues if we were to use the name "Empiricist" both when we mean
a particular school in epistemology and when we talk of what I prefer to call
the Empiristic trend in the explanation of perceptual facts.
68 : Wolfgang Kohler
adequate by many psychologists, seems to me to be most seriously mis-
leading.
The main point is this. When referring to inherited particularities of
the nervous system, we clearly mean histological facts, even if in a given
case we may be unable to indicate precisely what these facts are. Thus,
when it appears that a certain phenomenon in perception is not brought
about by learning, we conclude that it is such histological conditions
which are responsible for its occurrence. I should like to show that this
inference is by no means generally justified. A phenomenon in per-
ception which is unlearned need not, for this reason, depend upon the
existence of special histological factors. Nativism is a dangerous term;
it covers several theoretical possibilities. Unfortunately, we are for the
most part aware only of the one toward which the expression "inherited"
points. It is important that we know what other factors may be involved
in unlearned function. For our issue is by no means restricted to a form
of theorizing in perception. Even the intellectual life of man will easily
be given a wrong interpretation if we do not realize that Nativistic
theories tend to include a far too special premise.
At the present time, all biologists agree that, if a nervous system ex-
hibits certain histological conditions, the cells of the species (and, in
particulars, those of the individual) are ultimately responsible for the
fact. It is the chromosomic equipment of the cells which has forced
morphogenetic processes to establish those structures. The chromo-
somic equipment of a species, on the other hand, is assumed to be a
product of evolution. Hence, if a theorist gives us a Nativistic explana-
tion of certain phenomena, and if his theory is of the kind to which the
name "Nativistic" commonly refers, he always assumes that, in the last
analysis, those phenomena are made possible by particular achieve-
ments of evolution.
It seems to me that this procedure is incompatible with the very mean-
ing of evolution. For the principle of evolution implies that all biologi-
cal events, including the functions of the brain, have some characteris-
tics on which evolution has never had any influence whatsoever.
In its most consistent form, the postulate of evolution maintains that
once the behavior of the inanimate world is sufficiently known it must
be possible to derive all biological facts from principles which hold for
nonliving systems. At the present time, nobody can be forced to accept
this radical postulate. I will nevertheless adopt it for my present purpose,
because in this fashion my task will be greatly simplified. It follows, of
course, that I cannot try to convince Vitalists or those who believe in
Emergent Evolution. In a sense, this does not affect my argument. The
Psychology and Evolution : 69
error which I propose to discuss is often made by authors who regard
the postulate as a necessary part of modern scientific thinking.
Evolution is commonly regarded as a principle of change or develop-
ment, and this is also the natural meaning of the term. Nonetheless, the
postulate which I have just mentioned is a postulate of invariance. It
does not indicate what actually happens when organisms develop from
inanimate systems, or when new species arise. Rather, it states that,
however such changes may be brought about, the same principles as
hold in physics and chemistry apply also to these transformations, and
to the forms of life which thus originate. Of course, those who agree
with this thesis will readily admit that present formulations of those
principles may have to be modified when attempts are made to apply
them to living systems. General formulations which excellently serve
the physicist's purposes may not explicitly refer to situations which play
a particularly important role in biology. In biological theory, principles
of science may therefore have to be stated in a way to which the physicist
is not accustomed. It goes without saying that reformulations of this
kind are entirely compatible with the postulate of invariance in evo-
lution.
When referring to principles, I have in mind such general proposi-
tions as the First and the Second Laws of Thermodynamics, but also the
Law of Dynamic Direction, a law which is implied rather than explicitly
formulated in physics. 2 But the postulate of invariance applies not only
to general principles. It also demands that no forces and elementary
processes occur in organisms which do not also occur in physics and
chemistry. By forces I mean such vectors as electric and gravitational
fields. Examples of elementary processes are electric currents and cur-
rents of diffusion. It does not, of course, follow from the postulate that
all forces and elementary processes which are known in physics must
also play a part in living systems. What is meant is merely that such
forces and processes as are actually encountered in organisms invariably
have counterparts in the inanimate world. It will be realized that all
concepts to which the postulate of invariance refers, namely, general
principles, forces, and elementary processes, are concerned with action.
Obviously, as I am now using this word, it applies not only to events
which involve changes but also to steady states. 3
2 Cf. W. Kohler, The Place of Value in a World of Facts. New York:
Liveright, 1938, pp. 306 ff.
3 I am aware of the fact that, in physics, the term action has also a much
more technical meaning. In the present connection, we are not concerned with
action in this sense.
70 : Wolfgang Kohler
If so much is supposed to have remained invariant in evolution, what
can have varied while evolution took place? There must be factors in
nature which can change irrespective of the fact that the general princi-
ples, the forces, and the elementary processes of all action remain the
same. Any textbook of physics can tell us what these factors are. The
same principles apply, the same forces operate, and the same processes
occur under conditions which vary widely from one system to another.
Take mechanics, the discipline which deals with the movements of ob-
jects. It is a form of action that objects in the neighborhood of our planet
tend to approach its surface. But objects may either be free to follow
the direction of the gravitational vector, or given conditions may restrict
this freedom. When placed on an oblique plane which is rigid and solid,
an object still approaches the earth, but it does so in the direction of
the plane and more slowly, because the resistance of the plane eliminates
the component of gravitation which would operate at right angles to
the plane, and only the component parallel to the plane accelerates the
object. Given conditions which exclude certain possibilities of action are
called constraints. The mechanics of solid objects is not the only part
of physics in which constraits modify action. If a gas is surrounded by
the firm walls of a container, these walls are constraints. Many processes
can occur in the gas, but all those are prevented from taking place which
would involve a displacement of the walls, and thus the gas cannot ex-
pand as it would otherwise do. In hydrodynamics, a rigid tube in which
a liquid is enclosed is obviously a constraint; in contact with the inner
surface of the tube, the liquid can move only in the direction of the sur-
face. It is perhaps not customary to use the same term in the case of elec-
tric phenomena; actually, however, when a nonconducting substance
surrounds a material in which electric currents spread, this substance
plays the part of a constraint.
No constraint in the sense in which we have just used the concept
makes a positive contribution to the action upon which it is imposed. In
this sense (although not in others) the role of such constraints is nega-
tive. They serve to exclude certain actions which would be possible if
the constraints were not present. But while in this fashion some com-
ponents of forces and of elementary processes are eliminated, the re-
maining components do not change their behavior. The laws which hold
for forces and processes are formulated in general terms so that, when
certain possibilities of action can no longer be realized, the same laws
still apply to such actions as are not prevented by constraints. On an
oblique plane, for instance, the component of gravitation which oper-
Psychology and Evolution : 71
ates in the direction of the plane accelerates an object in precisely the
same way as it would if the constraint were absent. The general principle
which is here involved is the principle of the conservation of energy. On
the oblique plane, increments of kinetic energy and losses of potential
energy are smaller for a given period than they would be in the absence
of the plane. Since both changes are of the same size, the principle holds
in this situation just as it does in the case of free fall. 4 Similar considera-
tions apply to the other instances of action under constraint which have
been mentioned in the preceding paragraph.
We can now return to our discussion of evolution. While the general
postulate of invariance in evolution claims that no essentially new kind
of action appears in living systems, it imposes no limits upon the con-
straints which may develop when certain inanimate systems assume the
characteristics of organisms, and when the various species acquire their
distinguishing traits. In this respect, the postulate demands only that
such constraints be established in a way which is compatible with the
laws of physics and chemistry. In organisms, many different forms of
action are, of course, combined (and mutually interrelated) which sel-
dom occur together in the simpler systems commonly studied by physi-
cists and chemists. Apart from this peculiarity of life, it can, according
to our postulate, be only specific constraints by which the living world
has been made possible. No examples of such constraints will here be
needed, since most histological structures may be considered from this
point of view. But, although the world of living creatures would not
exist if evolution had not introduced these structures, action in the
organisms can never be explained solely by the constraints to which it
is subjected. Constraints alone, I repeat, never cause any action; they
merely serve to modify actions which, as such, owe nothing to con-
straints. Thus, if our general postulate is accepted, any action in any
organism involves the operation of factors which are entirely inde-
pendent of evolution. We have seen that these factors are the forces and
the elementary processes of nature, and such more general facts as are
formulated in the principles of science.
Just as to any other biological processes, our reasoning must be ap-
plied to the cortical events on which the characteristics of mental facts
depend. Generally speaking, cortical action is also modified by con-
straints, and to this extent evolution is partly responsible for the way in
4 I am, of course, assuming that the influence of friction can be ignored. If
this influence is not negligible, the energy balance of the system becomes more
complicated, but the principle of the conservation of energy still remains valid.
72 : Wolfgang Kohler
which this action occurs. But, qua action, it can never be understood
only in such terms. For all action is also a matter of processes which
evolution has not affected, and which are now not affected by its prod-
ucts, the histological conditions found in nervous systems.
Since this argument may be too abstract to carry full conviction, I
will give a simple example. It has recently been suggested that the proc-
esses underlying organized perception are steady electric currents which
spread in the brain as a continuous medium. If this should prove to be
true, the distribution of such currents would partly be determined by
histological circumstances which evolution has established in the tissue.
Quite irrespective of such special conditions, however, the currents
would also follow the general laws which hold for any electric flow in
any resisting medium, and cannot have been altered by evolution. If
actually some other process plays the part which has just been ascribed
to electric currents, our argument must be applied to this other action.
It will now be apparent why we cannot accept the statement that the
explanation of all unlearned perceptual facts has to be given in terms of
histological conditions. The statement cannot be entirely correct for
any perceptual fact. It is only another form of the same mistake if all
unlearned functions are attributed to achievements of evolution. For,
quite apart from such achievements, unlearned functions are bound to
exhibit certain characteristics which they share with actions in the inani-
mate world.
Although our argument is so simple that it may almost appear as
banal, few discussions in which Nativistic explanations are being con-
sidered take account of the fact that when we deal with unlearned func-
tions we must always distinguish between action and its constraints.
All authors refer to histological conditions upon which such functions
depend, and thus, indirectly, to evolution; but few seem to realize that
any brain function whatsoever is also an example of actions which do
not, as such, depend upon such conditions. One cannot play Hamlet
without the Prince of Denmark; on our stage, however, we are con-
sistently trying to do so.
The present issue is now becoming particularly important because
the belief in the omnipotence of learning, which characterized an earlier
period in the development of psychology, is rapidly beginning to weaken
under the impact of evidence which points in the opposite direction. A
few years ago, a Symposium on Heredity and Environment showed the
change of trend in a most impressive fashion. 5 It seems, however, that
6 Symposium on Heredity and Environment. PsychoL Rev., 1947, 54, 297-
352.
Psychology and Evolution : 73
the factors involved in unlearned function which owe nothing to evolu-
tion, and can therefore hardly be called "inherited," were not explicitly
discussed at this conference. We do tend to ignore these factors. In an
excellent article on certain phenomena in human perception, a psy-
chologist recently explained that these facts can be interpreted either
in the Empiristic fashion or as consequences of the chromosomic equip-
ment of man. Obviously, the chromosomic equipment of man cannot
be made responsible for characteristics which action exhibits in inani-
mate systems as well as in organisms. I do not believe that the author is
a Vitalist; most probably, he would accept our general postulate of in-
variance. Nevertheless, action and the principles which it follows were
not mentioned in his paper. It is quite true that in the determination of
all events which we study in human perception some inherited condi-
tions and ultimately the constitution of our chromosomes must play a
certain role. But they can do so only by influencing processes, the nature
of which has not been altered in the least while evolution took its course,
and while human chromosomes originated. Consequently, all facts in
human perception have certain characteristics which are unrelated to
the specific make-up of human cells. In this respect, only one point is
open to debate. To what degree are constraints imposed upon physical
actions when these actions occur in human brains?
Our distinction has to be made quite generally, not only when prob-
lems in perception are being considered. In fact, if the distinction is
ignored in certain other parts of psychology, the consequences are actu-
ally much more important. In some cases, they may affect our very
concept of man, and may thus have repercussions even in philosophy.
In the early years of this century, a Naturalistic conception of man was
strongly recommended by some philosophers. One topic to which they
applied their program was human thinking. More specifically, they
maintained that knowledge and the attempts to widen its scope are not
primarily concerned with objective cognition. Rather thinking was to
be regarded as a particular mechanism which evolution has developed
in man, and which helps his species to survive. Now, human thinking
which is objectively adequate can hardly fail to have useful conse-
quences; but it does not follow that the usefulness of thinking in a par-
ticular environment is its most fundamental characteristic, and that its
adequacy, when it is adequate, must be defined in terms of its useful-
ness. For again, although human thought may to a degree be influenced
by special conditions which evolution has established, it is, first of all,
a form of action. Consequently, the notion that only such conditions,
i.e., histological factors, are responsible for the characteristics of think-
74 : Wolfgang Kohler
ing cannot possibly be defended. Thinking no less than perception must
also follow principles which are unrelated to the particular circum-
stances of human life, to evolution, and to histological devices. I some-
times wonder what those philosophers meant by nature when they de-
manded that man be understood in Naturalistic terms. Surely, they
ignored the most important of these terms.
In the meantime, their views have had an immense influence upon the
intellectual and emotional climate of our historical period. Generally
speaking, there is an optimistic trend in evolutionary reasoning. The
changes which have occurred since life first appeared on this planet are
commonly regarded as improvements. From this point of view, there
is, of course, a great temptation to regard human thinking as the very
greatest among all evolutionary achievements, and on this basis to feel
more optimistic than ever. The Naturalists did not make this mistake.
They realized, and sometimes apparently with a certain satisfaction, that
in the evolutionary explanation of human thinking as a useful tool this
thinking is actually devalued. There is general agreement among the
biologists that of all imaginable devices which might have arisen in evo-
lution those have become stable characteristics of a species which serve
to make this species better adapted to its environment. Now, although in
a way all adaptation implies improvement, it also has its less attractive
phase. To the extent to which the various functions in animals and man
are thus usefully conditioned, their value must be regarded as relative
to the particular environment in which the adaptation has occurred.
This is precisely what the Naturalists actually meant, or what their fol-
lowers understood them to mean, when they said that human thinking
is a product of evolution. When we now refer to "human nature," we
use the expression with an unmistakable accent on the adjective. We
seem to mean a quite particular part of nature, the one which is merely
human, or human in a restrictive sense. This is the point at which evo-
lutionary optimism has turned sour; at which evolution has become a
powerful source of the relativistic defeatism from which our intellectual
culture is suffering. And yet, if we follow the principle of invariance in
evolution, there is no cause for this particular form of our ailment. When
man is thinking, he invariably follows, at least in part, some principles
of action which hold everywhere, and can therefore not be suspected of
being merely relative to his particular environment. To be sure, he
would not exist at all, if evolution had not occurred. At present, he
would have little to think about if evolution had not given him sense
organs, properly conducting nerve fibers, and many other particular
Psychology and Evolution : 75
devices. But even a modest nerve impulse which travels along one of
those fibers obeys some general principles of action no less than the
constraints to which it is subjected. Similarly, man's thinking must have
some characteristics which are not in any sense determined by evolution
and, for this reason, significant only in a relative sense.
In some instances, human thinking may be strongly affected by in-
herited conditions; in others, it may follow mainly general principles of
action rather than such conditions. I have a suspicion that the latter
alternative is often realized when human beings grasp relations between
objects, and when they derive further relations from those which are
given. If this were true, there would still remain unanswered questions
as to the cognitive significance of such events. But whatever this signifi-
cance might be, it would not be limited by the fact that it happens to
be human beings in whom the events occur. Moreover, such processes
would probably show a certain affinity to the facts which man observes
in nature; because, as to certain fundamentals, he would find in such
observations what he can also find in himself.
How would a human being be impressed by principles of action if
he became aware of them as principles of his thinking? Factual gener-
ality, even if it is absolute, need not be related to such concepts as being
valid or evident; but it might be so related in the present case. Man
might find those principles necessary in the sense in which certain for-
mal principles actually appear to us necessary when we think. Naturally,
he would not feel that the recognition of such principles presupposes
observation of any particular facts. Rather, their necessity would seem
to him to have an a priori character. And yet, no subjectivism would
follow, because, as I said before, for excellent reasons an a priori of this
kind would tend to fit empirical evidence.
Possibilities such as these fairly obtrude themselves once the postu-
late of invariance in evolution has been understood and accepted. Even
so, it remains to be seen whether principles of action in nature can really
be recognized in the way in which human thinking proceeds. Obviously,
if attempts in this direction should end in failure, the postulate of in-
variance could no longer be accepted in its radical form, and a Dualistic
view of the world would become unavoidable.
Mary Henle
GRADUATE FACULTY, NEW SCHOOL FOR SOCIAL RESEARCH
SOME PROBLEMS
OF ECLECTICISM
Some ten years ago Woodworth, commenting on the situation in psy-
chology as a whole, wrote: "Some may lean toward one school and
some toward another, but on the whole the psychologists of the present
time are proceeding on their way in the middle of the road . . ." (21,
p. 254) . He suggested that "If we could assemble all these psychologists
[all the psychologists in the world] in a convention hall and ask the
members of each school to stand and show themselves, a very large
proportion of the entire group would remain seated" (21, pp. 254-
255).
A similar position was taken by Boring at about the same time:
During the 1930's the isms pretty well dropped out of psychology. . . . The
only reason for mentioning these four schools in this book is that the student
hears about behaviorism and Gestalt psychology and has a right to be told
what they are and that they are no longer important as schools. What was
good in all the schools is now simply part of psychology (4, p. 11).
The eclectics, rising above the conflict of schools, hold that psycholo-
gists today are in happy agreement. It is their position that no real
issues exist among the various points of view in psychology. For it is
Reprinted with permission from the Psychological Review, Vol. 64, No. 5,
September, 1957.
Some Problems of Eclecticism ; 77
only on the basis of such a belief that one is able to select from each
approach, combining the theoretical contributions of all. As Wood-
worth presents the middle-of-the-road position:
Every school is good, though no one is good enough. . . . One points to
one alluring prospect, another to another. . . . Their negative pronounce-
ments we can discount while we accept thek positive contributions to psy-
chology as a whole (21, p. 255) .
If eclecticism is as prevalent in contemporary psychology as the above
statements suggest, it becomes important to examine its consequences
for theory. It is the hypothesis of the present paper that the eclectics
have, to a large extent, succeeded in reconciling differences only by
obscuring theoretical issues. An alternative to this kind of eclecticism
will be proposed.
We may begin with an examination of specific instances of eclectic
reconciliations of differences. Examples of two kinds of eclecticism
will be discussed, one having to do with reconciliations of positions
which refer to the entire field of psychology the attempt to resolve
conflicts among "schools" or general points of view and a more cir-
cumscribed eclecticism relating to particular psychological problems,
As an instance of the first kind, Woodworth x writes:
A broadly defined functional psychology starts with the question "What
man does" and proceeds to the questions "How?" and "Why?" ... So
broadly defined . . . functional psychology scarcely deserves the name of
a school because it would include so many psychologists who have not pro-
fessed themselves. Now the question is whether our middle-of-the-roaders
are not after all members of this broadly conceived functional school. . . .
But if the middle-of-the-roaders are really functionalists, the question is
then whether the same would not be true of all the schools. Are they not all
functionalists at heart? (21, p. 255).
Commenting on such a f unctionalism, Boring wrote in 1 950 : "Wood-
worth believed that psychologists were more in agreement than thek
quarrels indicated, and he sought a system to which all could subscribe.
He very nearly succeeded" (3, p. 565) .
It is not likely to be denied that psychology today has a functionalist
flavor. The interest in the adaptive value of psychological processes is
everywhere apparent. Nor can there be much disagreement with a f unc-
X A number of the examples to be considered will be taken from Wood-
worth's writings because the present author regards him as one of the clearest
of the eclectics and one of those whose theories are to be taken most seriously.
78 : Mary Henle
tionalism defined in terms of these three questions of Woodworth's. But
it tells us very little about a psychologist to say that he is a functionalist
in this sense. What we need to know are the kinds of answers which a
particular psychology gives to these questions. These are the issues in
contemporary psychology, and here it is that disagreements arise. What,
for example, is the relation of re-inf orcement, or of repetition, to learn-
ing? How does the learning process proceed? What is the nature of the
fundamental human motives? How does the group exert its influence
on the individual? It is the answers to such questions as these that divide
psychologists. It would seem that Woodworth has succeeded in bringing
all together only by obscuring such issues. If all are functionalists today,
we still have the problem of examining the differences among the several
varieties of functionalism.
The point may be made more specifically in connection with the same
author's theory of the conditions of transfer of training, the doctrine of
identical components (19), reformulated by Woodworth and Schlos-
berg (23) as a theory of "common factors." It is put forward to re-
solve the differences between those who hold that transfer is a function
of the identical elements in two learning tasks, and those who maintain
that it is a matter of the application of common principles or other whole
properties to the two activities. Woodworth's view 2 is that anything
concrete can be transferred thus both identical elements and princi-
ples since "any idea that can be recalled, or any attitude that can be
reinstated is concrete enough to qualify. Perhaps anything that can be
learned can be transferred" (19, p. 207). Again, "what is successfully
transferred is usually something you can put your finger on a princi-
ple, a good emotional attitude, a technique" (22, p. 582) .
The controversy about the conditions of transfer is settled, in other
words, by saying that something is carried over from one activity to the
other. It is true that this is a formulation which covers most of the cases.
But it lacks an advantage of both of the theories it displaces, namely the
attempt to state the specific conditions of transfer. It would seem that
the differences are resolved only at the expense of any specific theory in
the area of the controversy. The theory covers all the cases only by
telling us nothing specific about any of them.
To return to the more general kind of eclecticism, there exist today
a number of efforts to reconcile the various significant theoretical posi-
2 Since the earlier formulation (19) is the more explicit, and since the later
(23) seems not to differ from it in principle, the former will be drawn upon
here.
Some Problems of Eclecticism : 79
tions in psychology. A number of authors have attempted to resolve the
differences between behavior theory and psychoanalysis, Gestalt psy-
chology and psychoanalysis, behavior theory and Gestalt theory. 3
Several examples of this trend in contemporary psychology will be
examined.
The problems seen above in Woodworth's formulations exist also in
Abt's statements about basic agreements between psychoanalysis and
Gestalt psychology with respect to the structure and development of
the personality:
Freud's multiple-structured self is not essentially different conceptually from
Lewin's division of the person into regions. The dynamic and economic inter-
changes that are postulated as occurring with respect to the id, ego and
superego in psychoanalysis find parallel expression in Lewin's system of
barriers and lie classes of movements across them (1, pp. 38 39) .
If Abt means that Lewin's metatheory is compatible with Freud's
theory of personality structure, a case could be (but has not been) made
for this position. But to equate Lewin's division of the personality into
regions with Freud's topographical analysis of the person is to lose all
the specific psychological insights of the latter and much of the meta-
theoretical contribution of the former. Lewin's inner-personal regions,
if translated into Freudian terms, would undoubtedly fail to distinguish
between id, ego, and superego; his motor-perceptual region includes
some but not all of the functions of the Freudian ego. 4 Lewin is, indeed,
largely unconcerned with the specific content of the personality, with
the distinction between conscious and unconscious motivation, and
with the historical development of the person; it is impossible to discuss
Freud's topographical divisions apart from such considerations. Again,
Lewin has not discussed the content and nature of the forces responsible
for behavior; thus the parallel with Freud's statements about dynamics
can be maintained only if one turns one's attention away from the spe-
cifics of Freud's successive instinct theories. It is indeed possible to find
3 Not aH of these discussions are eclectic, at least in the meaning used here.
Some attempt to understand the contributions of one psychology in terms of
the theory of another, rather than to resolve differences between them. Such
work, for example Personality and Psychotherapy by John Bollard and Neal
E. Miller, will not be considered here. It presents interesting problems of its
own which deserve separate treatment.
4 For example: "It is to a certain degree arbitrary where one draws the
boundary between the motor-perceptual system and the inner regions, whether
for instance one considers the understanding of speech as an event within the
boundary zone or within the inner-personal systems" (16, p. 178).
80 : Mary Henle
parallels between Lewin's statements about the tendency of systems
under tension to seek discharge and Freud's formulations about the
pleasure principle; but to equate the two is to lose the specific character
of both the pleasure and reality principles and to neglect Freud's theo-
ries about behavior which is independent of the pleasure principle (8).
A further illustration will be given of the tendency prevailing in con-
temporary psychology to reconcile the ideas of Freud and Lewin. An-
other author writes: "It is the thesis of this paper that a synthesis of the
ideas of Lewin and Freud provides a basis for the beginnings of an
integrated system of psychological theory . . ." (5, p. 206). More
specifically, "We have already noted the structural parallelism between
Freud's divisions of the personality and Lewin's psychical systems" (5,
p. 222) . This writer, it is true, limits the parallelism, remarking, "Freud
has provided the living clay for the Lewinian scaffolding" (5, p. 228 ) . 5
Still, closer examination suggests that essential differences between the
"living clay" and the "scaffolding" have been neglected. For example,
"one of these [defense mechanisms], projection, becomes the equivalent
for Lewin's unreality" (5, p. 222) . While there are, of course, important
components of unreality in projections, the two concepts are by no
means equivalent, if only because not all events on a level of unreality
(for example, dreams, fantasies, vague hopes and wishes) can be de-
scribed as projections, at least as Freud uses the term. Again, to put the
Freudian unconscious "directly into Lewinian language" by saying that
"there are sub-systems within the region of the self which are not in
communication with each other" (5, p. 225) is to slip over the specific
nature of unconscious processes in Freudian theory. Furthermore, if
this is meant as an equation, it fails to do justice to the very important
communications which do exist between conscious and unconscious
systems. The point may be illustrated by the dream, which draws upon
the person's waking experiences (e.g., experiences of the "dream day")
and which is recalled by the waking individual.
Another instance of a premature reconciliation of Gestalt psychology
and psychoanalysis may be taken from the work of Witkin et al. (18) . 6
These authors point out:
5 In this connection Bronfenbrenner comments on Lewin's neglect of the
content of psychical systems: "This is indeed an unfortunate oversight'* (5, p.
214). The thesis will be developed elsewhere that this is no oversight, but that
Lewin undertook a different task.
6 These remarks in no way detract from the excellence or the significance
of these authors' study.
Some Problems of Eclecticism : 81
Although psychoanalytic theory, in its conception of primary and secondary
processes, recognizes the relation between intellectual functioning and per-
sonality, it has not really been concerned with the nature of secondary proc-
esses. . . . Gestalt psychology, in contrast, has offered a well-developed
theory of cognition, in which the role of the nature of reality in determining
perceptual and thought processes has been emphasized. But . . . Gestalt
theory has on the other hand neglected the role of personal factors in per-
ception. By showing that a perceptual act cannot be understood without
reference to both personal factors and the nature of reality, studies such as
ours help to bridge the gap between Gestalt and psychoanalytic theory, and
provide a basis for bringing together the main aspects of both into a single
comprehensive theory of human psychological functioning ( 1 8, p. 48 1 ) .
If, as I believe, the authors are correct in saying that "there is still
lacking in psychoanalytic theory any specific account of cognition or of
the nature of secondary process" (18, p. 481), and that Gestalt psy-
chology has neglected personal factors in perception these are pre-
cisely the reasons why studies such as theirs do not help to bridge the
gap between the two theories. The gap can be bridged only by a true
reconciliation of existing differences; and since these two approaches
have been concerned, as the authors point out, with such different areas
of psychology, it is difficult even to know where the essential differences
lie. A systematic analysis of the assumptions of both psychologies, one
concerned with implicit as well as explicit assumptions, would undoubt-
edly reveal both important differences and surprising compatibilities of
the theories. But in the absence of such an analysis we cannot, without
glossing over real differences, say that a particular finding helps to bring
the two theories together. To do so, it would need to be shown (a) that
the results demand a theory which reconciles actual differences between
the two approaches; or (b) if the two psychologies are in agreement
in the area in question (which would require demonstration), that the
findings can be handled in terms which are compatible with both.
To say, in other words, that both personal factors and reality factors
determine a perceptual effect is to pose a problem. It is to point out that
we need a human psychology which will include both kinds of factors;
but it is not to say that such a psychology will be compatible with Gestalt
psychology or with psychoanalysis or both. To the present writer it
seems more likely that a finding which cannot be handled adequately
within the framework of either of these existing systems will demand,
not a reconciliation of the two admittedly incomplete theories, but
rather a new theory. This point will be discussed below.
82 : Mary Henle
A final example, which will show again how eclectic reconciliation
of differences may be achieved at the expense of a specific theory in the
area of the controversy, may be taken from Welch (17). This author
has offered some fundamental propositions which he believes should
be acceptable both to Gestalt psychology and to contemporary be-
haviorism. For example, "perceiving is the result of a stimulus com-
pound producing effects upon the sense organs which establish brain
traces similar to or in otherwise related to brain traces formerly estab-
lished" (17, p. 181). Surely everyone will agree that present percepts
are related to traces of past ones; where dispute exists it concerns the
specific effects of past experience on perception. Welch has succeeded
in reconciling the differences by omitting the specific area of contro-
versy. Again, we are told:
In interacting with its environment, the organism changes in many ways.
. . . [Among other changes] it may learn. Learning is the effect of a stimulus
compound or stimulus compounds upon the nervous system of the organism
and the responses which these evoke, that makes possible the establishment
of new responses, as a result of such experiences (17, p. 187),
Many psychologists may, indeed, accept this as a rough definition of
learning. But when one leaves this level of generality and raises the
question of how this process is to be envisaged, this happy harmony
disappears. Hilgard points out: "There are no laws of learning which
can be taught with confidence" ( 12, p. 457) . Likewise no one questions
the fact that "behavior of any type is the result of the interaction of the
organism and its environment" (17, p. 176) . But what is the nature of
this interaction? What are the roles of organism and environment? Here
are questions on which different writers have taken divergent positions.
(Cf. 11.) While no one will disagree that memorizing and generalizing
( 17, pp. 181, 182) occur, and while Welch's definitions might provoke
little controversy as rough identifications of the phenomena in question,
different theories exist about the nature of these processes.
It will be clear from the above discussion that the existence of facts
which all psychologists accept is irrelevant to the problem of eclecticism.
Likewise the circumstance that some developments in psychology have
called attention to facts ignored by others has no bearing on the issue.
The important questions are: How are these facts understood? What is
their place in the over-all theoretical system? Even where agreement
exists as to the facts, differences are current with respect to these ques-
tions.
Some Problems of Eclecticism ; 83
In all the examples considered here, it would appear that differences
have been reconciled and controversy eliminated at the price of ob-
scuring the issues with which research is concerned in contemporary
psychology.
Boring, years ago, pointed out the productive role of controversy in
scientific research (2). Not only does the eclectic lose prematurely the
advantages of controversy, he may to some extent give up the advan-
tages of theory as well. The above discussion contains the suggestion
that the eclectic at times renounces specific theory in the area of a con-
troversy in order to reconcile differences. This statement will be quali-
fied below. But now attention must be drawn to a consequence of the
intimate relation between fact and theory.
There is a certain amount of fact that can be discovered in the absence
of any theory. For example, time errors forced themselves to the atten-
tion of psychologists who were concerned with quite different problems.
For the most part, however, problems for investigation arise out of the
theories one holds. New facts are discovered in the course of research
designed to test one's hypotheses. To the extent, therefore, that the
eclectic gives up specific theory in the area of a controversy, he is handi-
capped in the discovery of new facts.
Closer examination will, however, often show implicit theories which
may contradict the eclectic's avowed intention by placing him in a posi-
tion on one side or the other of the (now only implicit) controversy.
Woodworth, for example, deals with transfer in terms of carrying over
something from one learning situation to another rather than in terms
of application of what has been learned to the training tasks as well as
to the new ones. That is, transfer is seen as occurring because knowledge
acquired in the original training is carried over to the new activity; the
new tasks, to the extent to which they are similar to the learned ones,
are considered already partially learned. The alternative is ignored that
what is learned is not tasks but principles or other whole properties;
thus the training activity may merely provide examples of the use of the
principle which can be applied equally to the new situation. Wood-
worth's theory is thus close to a theory of identical elements in this re-
spect, opposing one derived from the study of learning by understanding
(13, Chap. 5) and, indeed, unable to deal with many cases of such
learning. 7 This consequence is particularly impressive since, as will be
7 If a theory of common factors were correct, there should never be more
than 100 per cent transfer, since two activities cannot have more than 100
per cent of their factors in common. Yet, as Katona has shown (13), cases
84 : Mary Henle
illustrated immediately below, Woodworth is by no means opposed to
learning by understanding; it is another instance of the confusions which
eclecticism breeds.
In another place Woodworth calls attention to the following contro-
versy:
Among present-day theories of learning those which emphasize re-enforce-
ment or the law of effect minimize the perceptual factor, often, stigmatizing
it as "mentalistic" and impossible to conceive in physical terms, while those
which emphasize perceptual learning are apt to deny any direct importance
to the factor of re-enforcement (20, p. 119) .
In attempting to show that there is "no obvious incompatibility" of
these two factors, he makes (explicit) assumptions about learning as a
cognitive process 8 which would be likely to be unacceptable to many
S-R theorists, and (both implicit and explicit) empiristic assumptions
about perception 9 which many cognitive theorists might find equally
unacceptable (assumptions, incidentally, which are not necessarily con-
sistent with those about the learning process) . 10
exist in which performance on the test activity is superior to that on the train-
ing task.
It is of interest to note also that Woodworth's theory, while it succeeds in
reconciling the differences, appears to lump together cases that do not belong
together. There is evidence that transfer of specific data is different, in process
as well as in the magnitude of the effect, from the application of principles
derived from one set of data to new material. (Cf. 13.)
8 For example: "As to connections, several may be established before the
conditioning is complete, but the primary one connects the conditioned stimu-
lus with the meaningful character it acquires as the first event in a regular
sequence" (20, pp. 121-122). Also "In experiments that offer alternatives and
demand a choice, what has to be learned is a distinction between stimulus-
objects and not between motor responses. . . . What has to be learned is the
difference between the two alleys" (20, p. 122).
* "When a new percept is in the making when an obscure stimulus-com-
plex is being deciphered, or when the meaning of a cue or sign is being dis-
covered an elementary two-phase process is observable. It is a trial-and-
check, trial-and-check process. The trial phase is a tentative reading of the
sign, a tentative decipherment of the puzzle, a tentative characterization of the
object; and the check phase is an acceptance or rejection, a positive or nega-
tive reenforcement of the tentative perception" (20, p. 124).
Among the implicit assumptions seems to be the view that organization is
not primary in perception, nor prior to the effects of learning; as well as the
idea that there is no fruitful distinction to be made between perception and
interpretation.
10 As a final illustration, Welch states: "This distinction between elementary
and higher forms of learning involves the distinction between a situation where
the new elements are simple in nature, or simple in character and are simply
Some Problems of Eclecticism : 85
Several questions suggest themselves with respect to the theory im-
plicit in eclectic solutions.
(a) A question worth examining is whether there is a tendency for
such implicit theory to be too heavily weighted in the direction of tradi-
tional theory. As the above examples show, this need not always be the
case; but it seems plausible to think that when theory is not explicit, and
thus not examined, it draws upon doctrines prevailing both in psychol-
ogy and in the culture in general rather than upon the newer and less
widely accepted theoretical currents. In a similar connection Kohler has
pointed to a certain conservatism in eclecticism:
. . . it has been said with approval that psychology now tends to be eclectic.
Again, we have been told that in psychology we had better stay in the middle
of the road. I cannot agree with these prescriptions because, if they were
followed, psychologists would have to look first of all backward. In an eclectic
attitude, they would be too much concerned with ideas which are already
available; and, in attempting to find the middle of the road in psychology,
they would have to give too much attention to the tracks along which others
have moved before them. Such attitudes could perhaps be recommended if,
in research, security were an important issue. Actually there is no place for
it in this field. In research, we have to look forward, and to take risks (15,
p. 136).
(b) Another question which arises in connection with the theory
underlying eclectic solutions is the following: since such theory is often
implicit, and thus unexpressed and unexamined, is it adequate to lead
to the discovery of new facts? For example, since the idea of "carrying
over" (i.e., as opposed to that of "application") is only implicit in
Woodworth's theory of identical components, it seems unlikely that it
would be subjected to test. Or again, the implicit elementarism in
Welch's statement about learning (cf. footnote 10) is unlikely to be
tested, since the author's main focus is on other aspects of the statement.
Also worth looking into in connection with the theory implicit in
eclectic solutions are the questions of its adequacy for ordering the
facts and its susceptibility to proof or disproof. For example, to say that
"something" is transferred is too unspecific a statement of the conditions
of transfer to test empirically. Any finding of transfer seems to confirm
integrated, and a situation where the new elements are complex and integrated
in a complex manner" (17, p. 188). This statement implies an elementaristic
view of the learning process learning being envisaged as the integration of
elements which would be far from acceptable to aU the psychologists Welch
is trying to reconcile.
86 : Mary Henle
it, and there is no result which could disprove it. Again, it has been sug-
gested above that the theory implicit in a given eclecticism is not always
internally consistent. This is a question which deserves examination in
connection with particular eclectic psychologies.
We may summarize the discussion so far by saying that eclectics have
to a large extent succeeded in resolving conflicts in psychology by ig-
noring differences and obscuring the issues. Some reasons for dissatis-
faction with such solutions have been indicated. 11 Is there no alterna-
tive? It seems to the present writer that reconciliations can be reached
in psychology only by focusing on the existing differences, examining
them, and carrying on research to settle issues. If this is eclecticism, it
is eclecticism after the fact rather than the prevailing eclecticism before
the fact. And it is clear that it will not be a matter of reconciling existing
theories. Since competing theories on any particular issue in psychology
today or competing psychological systems each tend to be plausible
and to be supported by evidence, it is unlikely that any one will win a
clear victory over the others. Yet none can offer a fully satisfactory ex-
planation or else the controversy would not exist. Controversies do
not exist in science with regard to processes which are fully understood.
Thus the task seems to be one of arriving at new, more comprehensive
theories of the processes in question.
An example should make this clear. It seems safe to say that theories
of forgetting arising out of experimental psychology have found no ade-
quate place for the facts of repression. Nor have the psychoanalysts suc-
ceeded (or tried) to bring these facts into relation with a general theory
11 It is of interest to note that eclecticism seems to have presented similar
problems in other fields of knowledge in their comparative youth. I quote an
observation on the medical science of a century ago: "And as the rules de-
rived from fundamental truths seemed to come into unsolvable contradiction
with the experiences and the sanctioned standards of practice, there sprang up
under the name 'eclectic* the representatives of sober elucidation, of the juste
milieu, of the medium of the extremes. The breach between theory and prac-
tice, which they feared, was avoided or postponed if theory gave up the pre-
tension to penetrate into particulars and if practice agreed that, because of its
youthful immaturity, it should be excluded from counsel, and progress in
silence and in hope. The conflict was settled and peace was achieved, not by
the reconciliation of the parties, but by separating them. The so-called im-
partial examination of the facts should lead only to a middle road between
them. [The eclectics] thought they had principles and avoided their applica-
tion; they proclaimed themselves free and in practice clung to the consequences
of old dogmas. They practiced tolerance not because they included the truth
of each dogma, but because a chasm existed between theory and life, beyond
which theory didn't matter" (9, p. 9).
Some Problems of Eclecticism : 87
of memory and forgetting. Can the two kinds of theories be brought
together? It seems to me that the most fruitful starting point is not the
attempt to reconcile existing theories. Actually, useful theories of re-
pression do not exist. (Cf. 6 for a similar point, more generally stated.)
It is hardly sufficient to say:
Repression proceeds from the ego, which possibly at the command of the
superego, does not wish to be a party to an instinct cathexis originating in the
id. Through repression the ego accomplishes the exclusion from conscious-
ness of the idea which was the carrier of the unwelcome impulse (7, p. 19).
This statement contains no hypothesis about the processes involved,
about how repression can possibly be brought about. Thus there seems
to be no point to attempt to reconcile the theories of experimental psy-
chology and of psychoanalysis on repression; neither has an effective
theory in this area. What we need is to look into the processes them-
selves, in the light of what we know about forgetting in general ( cf . 1 ) .
Can affective processes act, for example, to produce a failure of the
Hoffding function, i.e., that selective interaction between present proc-
ess and memory trace which is the basis of recognition and the first step
in the process of recall? (Cf . 14, pp. 126 ff. ) Under what conditions can
emotional and motivational processes introduce interferences? An-
swers to such questions might lead not only to a hypothesis about the
nature of repression, but might also introduce considerable modifica-
tion into our present theories of the nature of forgetting in general.
The eclectics are, of course, right in maintaining that where a genuine
controversy exists in psychology, and where evidence seems to support
both sides, there is likely to be some truth to both positions. But they
solve their problem too soon. Existing theories cannot be made more
comprehensive by adding divergent ones together. They can be broad-
ened to include all the relevant evidence only by looking more deeply
into the phenomena with which they are concerned; and this means
arriving at new theories.
At this point the parallel between productive solutions of theoretical
problems and of personal problems becomes striking. In connection
with the reconciliation of opposites within the personality, C. G. Jung
points out that conflicts are never resolved on their own level. They are
outgrown. Only on a higher level can you see both sides.
88 : Mary Henle
SUMMARY
Examples have been presented to show that eclectics tend to resolve
conflicts in psychology by glossing over real differences and obscuring
the issues. Such solutions achieve harmony at the price of specific theory
in the area of the controversy, and thus sacrifice fruitfulness in the dis-
covery of new fact. Closer examination often reveals implicit theories
underlying such solutions, but unexpressed and unexamined theory can
hardly be expected to equal explicit hypotheses either in fruitfulness or
in adequacy in dealing with known facts,
It is here suggested that differences need to be resolved in psychology
not by denying them and attempting to combine existing theories, but
by focusing on the differences and using them to get a better view of
the relevant phenomena. We will achieve more comprehensive theories
not by combining existing ones but by understanding better the proc-
esses in question.
REFERENCES
1. Abt, L. E. A theory of projective psychology. In L. E. Abt and L. Bellak
(Eds.), Projective psychology. New York: Knopf, 1950.
2. Boring, E. G. The psychology of controversy. Psychol. Rev., 1929, 36,
97-121.
3. Boring, E. G. A history of experimental psychology (2nd ed.). New
York: Appleton-Century-Crofts, 1950.
4. Boring, E. G. The nature of psychology. In E. G. Boring, H. S. Langfeld,
and H. P. Weld (Eds.), Foundations of psychology. New York: Wiley,
1948.
5. Bronfenbrenner, U. Toward an integrated theory of personality. In R. R.
Blake and G. V. Ramsey (Eds.), Perception, an approach to per-
sonality. New York: Ronald Press, 1951.
6. Bruner, J. S. Freud and the image of man. Amer. Psychologist, 1956, 11,
463-466.
7. Freud, S. The problem of anxiety. New York: Norton, 1936.
8. Freud, S. Beyond the pleasure principle. London: Hogarth Press and the
Institute of Psycho-Analysis, 1950.
9. Henle, J. Handbuch der rationellen Pathologie. Erster Band (2nd ed.).
Braunschweig: F. Vieweg u. Sohn, 1846.
10. Henle, Mary. Some effects of motivational processes on cognition.
Psychol Rev., 1955, 62, 423-432. [See pp. 172-186 of the present
book.]
Some Problems of Eclecticism : 89
11. Henle, Mary. On field forces. /. PsychoL, 1957, 43, 239-249. [See pp.
286-297 of the present book.]
12. Hilgard, E. R. Theories of learning (2nd ed.). New York: Appleton-
Century-Crofts, 1956.
13. Katona, G. Organizing and memorizing. New York: Columbia Uni-
versity Press, 1940.
14. Kohler, W. Dynamics in psychology. New York: Liveright, 1940.
15. Kohler, W. The scientists and their new environment. In W. R. Crawford
(Ed.), The cultural migration. Philadelphia: University of Pennsyl-
vania Press, 1953.
16. Lewin, K. Principles of topological psychology. New York: McGraw-
Hill, 1936.
17. Welch, L. An integration of some fundamental principles of modern
behaviorism and Gestalt psychology. /. gen. PsychoL, 1948, 39, 175-
190.
18. Witkin, H. A., Lewis, H. B., Hertzman, M., Machover, K., Meissner,
P. B., and Wapner, S. Personality through perception. New York:
Harper, 1954.
19. Woodworth, R. S. Experimental psychology. New York: Holt, 1938.
20. Woodworth, R. S. Reenforcement of perception. Amer. /. PsychoL,
1947, 60, 119-124.
21. Woodworth, R. S. Contemporary schools of psychology (Rev. ed.).
New York: Ronald Press, 1948.
22. Woodworth, R. S., and Marquis, D. G. Psychology (5th ed.). New
York: Holt, 1947.
23. Woodworth, R. S., and Schlosberg, H. Experimental psychology (Rev.
ed.). New York: Holt, 1954.
Rudolf Arnheim
SARAH LAWRENCE COLLEGE
GESTALTEN
YESTERDAY AND TODAY
Written on the occasion of the 100th anniversary
of the birth of Christian von Ehrenfels (June 20, 1959)
The remarkable essay "On Gestalt Qualities" (6), published in 1890
by the Austrian philosopher Christian von Ehrenfels when he was in
his early thirties, has fallen on fertile ground. It has given a name to one
of the most characteristic schools of scientific thought in our time and
thereby established its existence formally. It is responsible for the word
"Gestalt" having acquired citizenship in the English language; and even
the term "Ehrenfels qualities" is encountered here and there in the
American psychological literature as a term designating Gestalt proper-
ties.
But Ehrenfels did more than baptize the new movement. With great
discernment he immediately posed, or at least alluded to, many of the
questions that the Gestalt problem raises in the minds of thoughtful
theorists even today.
We are accustomed to considering Ehrenfels' approach as transi-
Reprinted with permission from Gestalthaftes Sehen, edited by Ferdinand
Weinhandl. Darmstadt: Wissenschaftliche Buchgesellschaft, 1960. Translated
by the author.
Gestalten Yesterday and Today : 91
tional: boldly and accurately he showed that the traditional "atomistic"
method of science was unsatisfactory; but in order to go beyond the
mere summation of elements he added a further, hypothetical element,
intended as an explanatory device. Described in this fashion, the theory
'sounds quaint (9, p. 10). It seems to put new wine into an old bottle.
We have learned in the meantime that the Gestalt springs from the or-
ganization of the parts themselves, not from a quality added to the parts.
We do not say: the whole is "more" than the sum of the parts; we prefer
to assert that the whole is "something else" than the sum of its partsT
But does this not mean distorting the true situation in our own way?
Ehrenfels' Gestalt qualities were not simply a "gimmick," introduced
because an explanation was needed. He was an observer rather than a
sophist. When he examined an integrated whole, for example, a paint-
ing, he found indeed an over-all composition, but also a collection of
parts, which could be perceived one by one and were each distinguished
by an independent character of their own. To be sure, when he isolated
a patch of color or a shape it changed its character. However, seen in
context, a part was not simply "something else"; rather, more often than
not, it clearly resembled the isolated piece. The element did not simply
disappear in the whole. It underwent changes, lawfully dependent on
the interaction of whole-structure and part-structure. We are used to
comparing various Gestalt contexts with each other. But isolatipn^jpo,
is a Gestalt context after all, and indeed a particularl^rirumlnating one.
And therefore, although to assert that the whole^"adds" something new
to the parts is to speak imprecisely and uacautiously, such a formulation
points to a negle ( ctd aspect of, the Gestalt problem^
**Tt is true that Ehrenfels, inevitabljTp^lpr^ought of visual units
as the direct counterparts of the retinal stimuli, thus arriving at a set of
hypothetical elements which he confused with the actually observed
parts of a perceptual whole. /What he called the "foundation" of his
Gestalt qualities was the sum of the "sensations."/Today it is evident
that the often quite self-contained parts of a perceptual whole are not
unaffected by that whole but, on the contrary, created by it or at least
tolerated as sub-wholes, which are thoroughly different from the raw
material of elements. In this sense, progress has been made. If, however,
we define Ehrenfels' "foundation" in our own way as the sum of the
actually perceived, genuine parts and if consequently we ask the ques-
tion of how the whole is constituted of these parts, we can derive from
his concept a fruitful approach even today.
The so-called constancy hypothesis (4, p. 86) led psychologists
92 : Rudolf Arnheim
astray, not so much because it assumed a point-by-point relationship
between each act of consciousness and the corresponding process in the
nervous system, but because the physiological egmvalenj-s selected for
the purpose w^^JhejadditiYe^liiniiU.in .thosense organ rather thajLthe
field_processes-hat organize the, stimulus material in^the projection cen-
to this coupling of the "sensations" with the ac-
tivities of the peripheral sensory organ, Ehrenfels and his contempo-
raries assumed without question that one could speak of perception only
as long as there was a direct response to a stimulus, whereas memory
and the forecasting imagination had to take care of what preceded and
followed the immediate present. This meant that the experience of
listening to a melody or viewing an olSject in motion was assumed to be
cdnfpbsed "oFdiss'imilar 'elements, namely, direct percepts $nd indirect
"representations.^' Such a notion surely contradicts introspective evi-
dence, which suggests no such distinction. Do we see a dancer "arrive
out of the future and jump through the present into the past?" (2, p.
306). Even in this matter, however, Ehrenfels was farsighted enough
to remark that when Ernst Mach, unconcerned with such theoretical
scruples, talked about the direct perception of a sound-Gestalt, "he
obviously was speaking of perception in a sense that differed from the
ordinary."
In the course of time, theory has caught up with observation by recog-
nizing the projection centers in the brain as the probable physiological
equivalent of the perceptual fieldfin the brain field, the temporal suc-
cession of past, present, and future is translated into spatial relation-
ships, i.e., some sort of chain of traces, corresponding to sounds or
phases of perceived motion. To such a chain of traces further "links"
attach themselves seamlessly as soon as new percepts are transmitted to
the brain. To what extent these links fuse in unified wholes does not
depend on whether they derive from perception directly or only indi-
rectly, but depends as Ehrenfels recognized intuitively upon the
Gestalt structure of the sequence. "In the case of perceiving a walking
person," he says,
we can indicate, with some precision, th^range within which we clearly ap-
prehend the Gestalt quality of the percept, in this case, the visually perceived
motion. We are convinced that we see the most recent phase of the action
the latest step, in the case of medium-fast walking whereas the earlier steps
convey the quite different experience of being merely remembered.
"illusion," as he calls it, "is likely to come about because memory
preserves the full continuum of all the positions of the leg only for the
Gestalten Yesterday and Today : 93
latest step so that the corresponding temporal Gestalt quality can be
fully visualized, whereas there remains of the preceding one nothing
but an incomplete impression" (6, pp. 270-271). (in our present-day
language this means that the range of the psychological present is deter-
mined by the strength and boundary of the Gestalt context! A passage
of music, for example, which is conceived by the listener as an indi-
visible unity, appears to be "in the present."
Strictly speaking, anybody who assumes that the sensory experience
of a sequence of sounds is based on perception only for the short mo-
ment of the actual present but otherwise on memory, ought to make
the same assertion for the visual field, since objects are perceived by
successive scanning if they are at all extensive. Ehrenfels, however,
maintains that for spatial Gestalten the situation is much simpler be-
cause "all the parts of the complex on which the Gestalt is based are
given simultaneously" (6, p. 253). Evidently he has in mind the phe-
nomenal difference between the succession of what is heard and the
simultaneity or better, timelessness of what is seen. The question of
how this difference comes about would seem to us not at all easy to
answer since both kinds of process derive physiologically from temporal
successions of stimulations. Quite in general, psychologists are coming
to realize more and more that perception is not primarily an effect of
sustained sensory stimulation. On the contrary, perception seems to
record essentially the changes in stimulation. Even in the peripheral
sense organ any persistence of a stimulus seems to cause a weakening
of the response. In fact, Grey Walter in his studies of brain waves has
recently suggested that the terminal phase of the sensory process might
be the kind of scanning motion we know from cathode rays and that
the a-rhythm of the EEG may reflect such a "searching for pattern" (7,
pp. 108 ff. ) . The fact that our eyesight presents the things of our world
in timeless immobility, in spite of the dynamic origin of vision, may find
its explanation in the nature of the stimulus material. Also the cortical
center of vision uses its spatial dimension for the purpose of discerning
the locations of stimuli, whereas in hearing the same dimension is used
to distinguish pitch.
Essentially ... Ehrenfels thoughtjof a Gestalt as JL synthesis brought
about when physiological processes mae*llirect contact with each
" ' f*.ww*n *** ..*?'.,. ,. .. ,jr,- , * ,*,yjt,, >,Mt -n,-***,w *.(*.-.-*.'-*
other. He held that, for reasons not yefknown," spatial cohabitation in
tfie brain produced a "new" psychical entfty^iii^ still to be
determined. Psychology was unable to move beyond this step, decisive
though it was, until a Gestalt was understood to be the result of a
concretely describable whole-structure, which determines the place and
94 : Rudolf Arnheim
function of the parts constituting it. Only this further insight mobilized
experimental research. It showed, for example, that a Gestalt, in order
to come about, does not require the presence of all parts but only that of
the skeleton of factors which determine the total structure. The "effort"
often needed to produce a perceptual Gestalt had been observed by
Ehrenfels and explained by the need to supply missing elements of the
"foundation"; nowadays we are more inclined to believe that the "effort"
is needed when a conglomeration of parts is slow to resolve itself in a
comprehensive whole (2, p. 49, Fig. 44) .
Nevertheless, Ehrenfels recognized the fundamental problem of
structural organization when he raised the question of "why there
emerge from that infinite crowd of elements the particular Gestalten
that seem primordial and, as it were, privileged" (6, p. 288). Wert-
heimer has treated this problem thoroughly in what he used to call his
"dot paper" (8), which starts out with an answer to the question: What
are the rules according to which elements combine? Intuitively Ehren-
fels anticipated a principle that might be called complementary to
Wertheimer's "factor of similarity" (8, p. 309) by pointing out that a
white square on a dark ground is discerned because it "detaches itself
from its surroundings by different coloration" a factor of dissimilarity,
in other words, which appears as a useful supplement to the concept of
"grouping," particularly if we remember William James's briskly formu-
lated rule: "The law is that all things fuse that can fuse, and nothing
separates except what must" (3, Vol. I, p. 488).
Our philosophical predecessor is at his most modern when he ex-
plains that Gestalt qualities submit to abstraction. What he has in mind
is not only the strict identity of structure, which can serve as a common
denominator for all transpositions. He also points out that, for instance,
a common rhythm may be the basis of comparison for Gestalt patterns
otherwise different from each other. Most important, he suggests that
abstractions occur not only in the traditional way on the basis of com-
mon elements but because of a similarity in the total phenomenon, the
style or habitude "which often resists tenaciously any attempt at an
analysis based on the identity of single components" (6, p. 279). He
does not tell us how to go about such classification according to struc-
tural types; but then, up to this day, nobody else has dealt concretely
with this problem of abstraction based on structural similarities, funda-
mental though it is for the Gestalt approach to scientific method.
Incidentally, Ehrenfels extends his observations on structural simi-
larities even to the question "whether Gestalt qualities belonging to
Gestalten Yesterday and Today : 95
different and apparently incommensurable realms of perception for
example, a crescendo or the growing stronger of the light at dawn or
the mounting of suspense might not possess a direct similarity, which
would go beyond the mere identity of common traits (such as in this
case the time) and yet be inherent in the phenomena themselves" (6,
pp. 279-280). He describes here the principle of isomorphism, i.e.,
the structural similarity of processes in different media. Today, the
Gestalt psychology of expression and of the body-mind problem is
founded on this principle (5, p. 61; 4, pp. 56 fL; 2, pp. 363 ff.).
By pointing out that Gestalt qualities are subject to abstraction Eh-
renfels provided, at the same time and apparently without realizing it,
the "counterweight to the individualistic tendencies" of which he speaks
in the last paragraph of his paper. More recently, Gestalt theory has
indeed been claimed as an ally for the idiographic (geisteswissenschaft-
lich) view, according to which the uniqueness of the individual whole
excludes any kind of lawful generalization (1, p. 16) a misunder-
standing that, we discover, can be refuted by reference to the very
"prime source."
Ehrenfels touches the most sensitive spot of Gestalt theory when he
discusses a question likely to be raised in response to his approach. If
there is to be assumed an infinitely large number of mutual influences
among the parts or combinations of parts, would not "an unending
complication of our mental life" be the consequence? His most striking
counterargument is: But if we look at the facts, we do find that each
part is clearly given to us as to its own nature and function in the whole
(6, pp. 256-257)! This observation is certainly correct. It means to
us that the nervous system fulfills indeed the seemingly superhuman task
of having all the innumerable interactions among parts or complexes of
parts take place in such a way as to make the result appear with pre-
cision in each part and at each structural level of the whole. The miracle
is made possible by the fact that it takes place in perception. This means
it is accomplished by a physiological field process in which the forces
constituting the field organize themselves automatically in the direction
of a minimum of over-all tension.
Well and good; but what happens when the reasoning intellect under-
takes to trace such accomplishment? It seems safe to assume that con-
ceptual thinking is not a field process but consists in the tying of rela-
tions among items. Therefore if we attempted even in the simplest
example to describe the multitude of interactions and to explain why a
given part appears at its place in the whole in that particular fashion
96 : Rudolf Arnheim
rather than another, our interpretation, like meshes thrown over a
statue, could offer a relational network of any desired density but never
the Gestalt process itself.
If this is true we may have to think of the Gestalt as a limiting case, a
scientific objective which conceptual thinking approaches asymptoti-
cally but which because of its own nature it can never reach. Per-
haps we are faced here with one of the ineluctable differences between
perceiving (Anschaulichkeit) and conceptual reasoning (Begrifflich-
keit) . Granted that in the practice of research it makes all the difference
whether we describe the object of an investigation as the sum of inde-
pendent parts and relations or whether, with our mind intent upon the
whole, we select for our description the relations closest to the total
structure we perceive. In principle, however, we should have to realize
that conceptual reasoning, given its phenomenal as well as its physio-
logical character, is doomed to atomicity.
Once this is admitted, must we say that this limitation means the
tragic foundering of science? Or shall we remember with some serenity
that it is precisely in the nature of science to trace some of the significant
lines of force in what we call reality, not however to duplicate this reality
itself in all its complexity? This is a question which the psychologist may
have to hand back to the philosopher for further treatment.
REFERENCES
1. Allport, G. W. Personality. New York: Holt, 1937.
2. Arnheim, R. Art and visual perception: a psychology of the creative eye.
Berkeley and Los Angeles: University of California Press, 1957.
3. James, W. The principles of psychology. New York: Dover, 1950.
4. Koffka, K. Principles of Gestalt psychology. New York: Harcourt, Brace,
1935.
5. Kohler, W. Gestalt psychology. New York: Liveright, 1947.
6. Von Ehrenfels, C. Ueber 'Gestaltqualitaten.' Vierteljahrsschr. f. -wiss.
Philos., 1890, 14, 249-292. [Reprinted in Ferdinand Weinhandl
(Ed.), Gestalthaftes Sehen. Darmstadt: Wissensch. Buchges., I960.]
7. Walter, W. G. The living brain. New York: Norton, 1953.
8. Wertheimer, M. Untersuchungen zur Lehre von der Gestalt. II. Psych.
Forsch., 1923, 4, 301-350.
9. Wertheimer, M. Ueber Gestalttheorie. Sonderdrucke des Symposion,
Heft 1. Erlangen: Philos. Akad., 1925.
Wolfgang Kohler
DARTMOUTH COLLEGE
THE PRESENT
SITUATION IN
BRAIN PHYSIOLOGY
During the past ten years, I have worked in physiological psychology.
This term refers to a rapidly growing number of problems and pro-
cedures. Most firmly established is the interest in questions of localiza-
tion. Excellent investigations have shown how loss of function in one
part of the brain or another affects behavioral achievements. Equally
interesting are the results of tests in which the brain is either locally
stimulated or, as a whole, subjected to electric shocks. We are especially
impressed by the discovery that stimulation of certain parts of the brain
seems to be equivalent to an award and that of other parts to a punish-
ment. Attempts to influence behavior by chemical action have so far
not had equally satisfactory results. For, while the changes in question
are often strong, it is still difficult to describe them in precise terms. But
so great is the interest which the medical profession takes in such experi-
ments that present shortcomings of the chemical approach will probably
be overcome within the next few decades. Efforts to relate the chemical
characteristics of individual brains to behavior are just beginning to
yield promising findings.
Reprinted with permission from the American Psychologist, Vol. 13, No. 4,
April, 1958.
98 : Wolfgang Kohler
While there can be no objection to these endeavors as such, physio-
logical psychology should also be concerned with problems to which
such investigations rarely refer. First of all, what exactly happens in
the brain when certain psychological events take place? Even if we
knew in detail what areas of the tissue are involved in particular achieve-
ments, we would still not be able to answer that question. Let me men-
tion a simple example. To my knowledge, neither birds nor fishes have
a visual cortex; and yet, the pattern vision of certain birds and fishes
resembles that of the primates who have a visual cortex. Consequently,
a given kind of process may occur in different locations; and, if this is
true, we have to deal with a functional problem which studies of locali-
zation alone cannot solve. My own work in physiological psychology
refers to functional questions in this sense.
I do not believe that this part of our task can be entirely left to the
neurophysiologists. For, much as I admire their actual achievements,
we disagree on one major issue. At the present time, no evidence as to
the nature of brain function can compare with our own, that is, the psy-
chologists' evidence, and most of this the physiologists quietly ignore.
They obviously do so in the conviction that the main principles of pe-
ripheral nervous function are also those of central function. Hence,
since they know the former processes extremely well, they do not seem
to expect that, when studying the brain, they might sooner or later need
our help.
But the most important process of the peripheral nervous system is
the nerve impulse. If brain function, too, consisted mainly of such im-
pulses, then hosts of psychological facts could never be understood in
physiological terms. Nerve impulses are particularly lonely events.
While they travel along their fibers, each is almost, if not entirely, un-
related to the others. The principal characteristic of most psychological
situations, on the other hand, is interrelation of their parts. Some such
relations may be established by impulses which travel in transverse
neurons from one place in the brain to another. But in many instances
this explanation does not work because the interactions in question
depend upon the specific nature of the various interacting facts. Nerve
impulses which seem to be always events of the same kind can hardly
represent the specific characteristics of one local fact at the place of
another. Psychologists will also hesitate to admit that such different
phenomena as perception with all its modes, memory, learning, attend-
ing to, feeling, thinking, and planning can all be derived from such a
monotonous principle of action.
The Present Situation in Brain Physiology : 99
I do not, of course, deny that impulses travel in axons of the brain
just as they do in peripheral fibers. I merely suspect that, in addition,
there are other forms of brain action and that, from the point of view of
psychology, such other actions are no less important than the impulses.
Hypotheses about these further processes can be developed only by
those who are familiar with certain parts of physics, with neurophysiol-
ogy, and with the basic facts of psychology. I do not suggest that we
return to what was once called brain speculation.
In recent years, my collaborators and I have formulated such an hy-
pothesis and have then tried to verify it in physiological experiments
(10-18). The psychological fact from which we started was the segre-
gation of molar units or things in visual fields. It seemed to us a plausible
assumption that, when the brightness of a certain area differs from that
of its environment, this area will be pervaded and surrounded by an
electric current. The current inside and the current outside would flow
in opposite directions. Functionally, therefore, a sharp discontinuity
would be established at the boundary of the area.
A few years later, Wallach and I began to study the distorting effects
which prolonged inspection of a visual object has upon this object itself
and upon test objects afterwards shown in about the same region ( 17) .
I need not describe how we soon found ourselves in the good company
of Gibson, who had discovered similar effects in other instances (6).
Our own experiments showed that the observed distortions follow a
general rule. Test objects recede from the regions in which the previ-
ously inspected objects, and particularly their contours, have been seen
as though an obstruction had been established in these regions. Not
only our own observations but also those of Gibson could be derived
from this rule.
It was, of course, our next question whether such phenomena can be
explained in terms of known physiological facts. Does any process cause
obstructions in the nervous system? There is such a process. When a
current flows through cells of the nervous system, it affects their sur-
faces. Where it enters the cells, it raises the local impedance and thus
blocks its own way. Where it leaves the cells, the opposite happens. But
since, after a short while, the former effect becomes stronger, the final
result is an obstruction. Thus the current is weakened, and its distribu-
tion in the tissue changes. When it is interrupted, the affected tissue re-
mains in its altered condition for considerable periods. Test currents
which are now conducted through this region will, therefore, also devi-
ate from their normal course. They will be weakened where the im-
100 : Wolfgang Kohler
pedance has been raised, and their intensity will grow in less affected
parts. Hence, the current as a whole will recede from the affected region
just as, in a figural aftereffect, a test object recedes from the area previ-
ously occupied by an inspection object.
Our assumption that activated parts of the visual cortex are sources
of electric currents can, therefore, also explain the figural aftereffects.
Are we right in saying that the displacement of test objects in figural
aftereffects and the deflection of test currents in the neighborhood of
obstructed tissue are remarkably similar facts? I once had an opportun-
ity to demonstrate aftereffects to a well-known physiologist who was
not yet acquainted with these phenomena. After several observations,
he turned to me and said with a smile: "Nice demonstrations of electrot-
onus, aren't they?" Electrotonus is an old name for the obstructions
which currents establish in the nervous system.
When electrotonus was first discovered, it was caused by currents
conducted through the tissue from>the outside. Since, in our theory, this
condition is supposed to be established by currents which issue from
active parts of the tissue itself, we now had to demonstrate that there
are such currents. This proved to be a difficult enterprise, but eventually
we succeeded ( 1 2-1 8 ). At the present time, numerous records of visual
and auditory currents are available. They were taken from intact human
heads and from the exposed projection areas of cats. Occasionally,
visual currents have also been registered from the brains of monkeys.
I wish to thank Held, O'Connell, Neff, Wegener, Pribram, and Rosen-
blith for their invaluable collaboration.
Our findings agree with known facts of cortical localization and may
therefore be regarded as reliable. The electrotonic action of the regis-
tered currents is also obvious. Continuation of the flow in given parts of
the tissue soon reduces its intensity. So far, therefore, our factual evi-
dence agrees with our expectations, and our theory of figural aftereffects
seems now to rest on fairly firm ground.
Figural aftereffects are merely deviations from what we call normal
perception. An explanation of these effects in physiological terms must
therefore include statements about the processes underlying perception
in general. Actually, our suggestion that cortical currents play an im-
portant part in normal visual perception preceded our work on figural
aftereffects. This thesis, however, has not been generally accepted,
partly because it seems to be at odds with certain experimental findings.
Lashley and his collaborators (19) laid strips of gold foil on the
visual cortex of one monkey and thrust thin wedges of the same material
The Present Situation in Brain Physiology : 101
into the occipital lobes of another. It was assumed that these measures
would greatly distort the distribution of visual currents. When post-
operational tests showed no disturbance of the animals' vision, the
authors concluded that perceptual organization cannot be related to
such currents.
I do not believe that this conclusion is justified. Strips of gold foil
attached to the surface of the brain can cause no major disturbance of
cortical currents because the electromotive forces involved operate at
right angles to those metallic conductors. Under these circumstances,
only an exceedingly small part of the flow will be deflected into the very
thin strips and, since this part, too, must return to the other side of the
source, it cannot follow them for more than a very short stretch. Wedges
of gold foil thrust into the cortex will at first cause a stronger disturb-
ance; but the very fact that now a highly concentrated flow must pass
through these wedges will establish abnormally high electrotonic ob-
structions in the tissue around their tips, and soon the conductors may
practically be sealed up,
Sperry, Miner, and Myers (22), who worked with cats, took several
steps to distort the cortical currents. In one experiment, they inserted
tantalum wires into the visual cortex. In some animals, the tissue was
sliced by vertical cuts in various directions; in others, small plates of an
insulating material were pushed into the visual region. Extraordinarily
difficult discriminations between visual patterns which the cats had
previously learned to perform were, on the whole, not very greatly dis-*
turbed by these radical measures.
I have difficulties in understanding this finding. It seems to me in-
compatible not only with my own views but also with generally accepted
neurological evidence. For, autopsies revealed that, in several instances,
the visual projection system including the geniculate bodies had been
very severely damaged. Even during the initial training, the discrimina-
tions which the cats had to perform must have taxed their abilities to
the utmost. How could their achievements remain similarly precise when
much of the tissue necessary for vision had been destroyed? It is to be
hoped that these experiments will soon be repeated. The authors realize
that any material which is frequently used in animal experiments will
gradually acquire particular characteristics which may serve as extrane-
ous cues; but their statements in this respect do not answer one ques-
tion: With very few exceptions, they always used the same pattern, an
equilateral triangle, as the positive object, while up to 40 modifications
of this triangle served as negative objects. Were correspondingly nu-
102 : Wolfgang Kohler
merous copies of the positive object presented during the training and
the final tests?
No comments on our work have so far been made by the neurophys-
iologists. I suspect that they do not like its general trend. Cortical cur-
rents or fields differ from the more popular nerve impulses in three re-
spects. In the first place, they do not follow the all-or-none principle
which holds for the impulses. The currents are graded processes; the
degree to which their cortical sources are activated, and therefore also
their own intensity, varies from one instance to another. Physiologists
will hardly raise objections on this particular ground because they are
quite familiar with graded events in the nervous system. Such events
are known to occur at the surface of axons and also at synapses.
Secondly, unlike the nerve impulses and the electric rhythms of
cortical cells, the currents shown in our records have not the form of
short-lived waves; rather, they are quasi-steady states. But again, such
states occur quite regularly at synapses and have been recorded in nu-
merous physiological investigations. I will presently return to these
studies.
A third difference between cortical currents and nerve impulses refers
less to certain facts than to the interpretation of these facts. While the
currents spread freely in the tissue as a continuous medium, nerve im-
pulses follow prescribed linear conductors, the nerve fibers. But no
physiologist will deny that many electric states of cells and fibers have
fields which also spread beyond their sources. The field of nerve im-
pulses has been measured at considerable distances from the active
fibers (20). The alpha-rhythm of cortical cells can be registered from
intact human heads only because this rhythm establishes correspond-
ing currents not only in the cortex but also in the skull and the scalp.
Or take the evoked potentials which announce the first arrival of aj>
ferent impulses in the projection areas of the brain. It has been shown
that these potentials, too, can be recorded from the intact head. Phys-
iologists accept these facts, but they do not seem to regard them as very
important. The reason is obvious and has once been clearly formulated
by Adrian (2). To the extent to which neural processes are located in
or on histological elements, their behavior in the tissue is prescribed by
the arrangement of these elements; and, since this arrangement appears
to be most orderly and practical, the distribution of corresponding
functions will be equally precise and practical. On the other hand, the
currents which spread around the active elements are not subjected to
such restrictions. Hence, it is often assumed that, if they do not cause
The Present Situation in Brain Physiology : 103
confusion and disorder, their influence on the tissue and on other proc-
esses must be negligible.
Every now and then, a physiologist discovers facts which cannot be
reconciled with this interpretation of neural action. The fields of acti-
vated elements do influence processes in other elements. For instance,
under certain conditions the cat's spinal cord exhibits a regular rhythm.
Bremer (4) has demonstrated that this activity is synchronized from one
end of the cord to the other. The author himself remarks that the rhythm
spreads with a speed far beyond the speed of ordinary neural transmis-
sion and that the interaction involved must therefore be mediated by
electric fields of the active cells. To prove that this is true, he even made
a sharp cut through the cord which left its two parts unconnected by
histological elements, that is, connected only by physical contact. After-
wards, the rhythms on one side and the other of the cut were just as
nicely synchronized as they had been before. Similar observations made
by Gerard and Libet (5 ) in their work on the frog's brain are, of course,
generally known. It follows that neural interactions can be maintained
quite independently of messages which travel along fibers.
Even more important is another function of neural fields. In 1938,
Barron and Matthews (3) published a most important paper on the
steady potentials which impulses establish at dorsal and ventral synapses
in the spinal cords of frogs and cats. They realized that such potentials
must be sources of equally steady currents, and they also discovered
that these currents have a remarkable effect upon the cells through
which they pass. As soon as their intensity has reached a certain thresh-
old value, they throw these cells into rhythmic action, which is then,
in the form of impulses, propagated along the axons of the cells. To my
knowledge, all records of ventral synaptic potentials and their currents
show this effect. Consequently, while occasionally transmission at a
synapse may be a direct effect of the impulses which arrive in presyn-
aptic fibers, this transmission is for the most part mediated by synaptic
currents. Under the circumstances, one can no longer maintain that the
currents of the nervous system are mere epiphenomena.
Steady potentials of neural structures have been demonstrated by
several physiologists. Adrian (1) has shown that the electric response
of a beetle's optic ganglion is a steady state or current. Quite recently,
Goldring and O'Leary (7, 8) have registered steady currents from the
visual cortex of the rabbit and have thus confirmed our own findings.
There seems to be little doubt as to the origin of cortical currents. The
steady spinal currents discovered by Barron and Matthews issue from
104 : Wolfgang Kohler
activated spinal synapses. Most probably, the sources of perceptual cur-
rents are activated synapses in the cortex.
Let me add a few remarks about problems in my field which ought
to be solved next. First of all, figural aftereffects have been demon-
strated not only in the frontal plane but also in the third dimension.
Years ago, Gibson (6) discovered such effects in kinesthesis, and we
studied further examples in this modality. In addition, figural after-
effects occur in passive touch, and most probably also in hearing. Even
certain inhibitions which tend to impede learning may partly be caused
by what we have called "satiation." It has not yet been possible to apply
our physiological assumptions to all these phenomena. The reason is
simply that nobody knows the physiological facts which underlie visual
depth, kinesthesis, and learning. Naturally, we cannot tell how electrot-
onus affects these processes until their nature as such has been revealed.
Thus, we have three new tasks which follow from the study of figural
aftereffects.
As though this were not enough, there is another problem to be
solved in this field. Ever since Gibson's first discovery, it has been known
that, under certain conditions, figural aftereffects are not prevented by
eye movements of the subject. Again, when many objects of a given
kind slowly move across the visual field while the subject fixates a mark,
the moving pattern still causes a figural aftereffect, and this effect cor-
responds to the shape of the moving objects. During Ivo Kohler's well-
known experiments (9), the subjects could also move their eyes as they
wished; and yet, the resulting specific adaptations had much in common
with figural aftereffects. Now, i$ the obstructions established under such
conditions were all localized at exactly the same neural level, the statis-
tical result would be a homogeneous layer of satiation which could
cause no specific aftereffects. It seems to follow that the location of ob-
structions which arise at different times cannot be precisely the same
in other words, that time is spatially represented in the brain just as it
is in the geological strata on the surface of the earth. This may seem to
be a bold hypothesis; but it is the only assumption which can explain
such remarkable facts. Incidentally, there is also no other explanation
for certain facts in memory. Memory traces, too, remain individual
entities; although, statistically speaking, they must be established every-
where and would therefore form a homogeneous layer of affected tissue
if their location in detail did not depend upon the time of their origin.
As a further task, we must therefore try to discover how the temporal
dimension is represented in the tissue.
The Present Situation in Brain Physiology : 105
These are fairly formidable problems. Another task which is again
related to the figural aftereffects does not look quite so forbidding. I
mentioned that the term electrotonus is commonly used as a name for
the obstructions caused by currents. Actually, as we have seen, currents
affect the surfaces of cells in two ways: they establish an obstruction
only where they enter cells, and they lower the local impedance where
they leave them. The more precise name for the former change is ane-
lectrotonus, and the latter is called catelectrotonus. Strictly speaking,
therefore, our theory of figural aftereffects is mainly concerned with
anelectrotonic action.
But are we allowed to make use of this particular action alone and
simply to ignore that cortical currents must also establish catelectrotonic
changes? Under the conditions which give rise to figural aftereffects we
can safely do so, because under these conditions the anelectrotonic
changes are particularly strong. There are, however, situations in which
the catelectrotonic effects ought to be stronger, and where this happens
the flow of currents will be facilitated rather than impeded. As a con-
sequence, tissue in the catelectrotonic state must attract test currents
and, if these currents are those of perceptual objects, these objects must
appear displaced toward the tissue in question.
Physiological evidence permits us to predict under what circum-
stances currents and objects will be attracted by certain places in their
neighborhood. Best conditions will be those in which perceptual objects
just appear or disappear, and also those in which one object appears
while the other is just disappearing. On the other hand, when percep-
tion begins to be stable, catelectrotonic attraction will soon be over-
come by anelectrotonic repulsion. We all know perceptual facts which
only arise under the conditions favorable to catelectrotonic effects. I
am referring to the various forms of apparent movement. But certain
achievements of memory may also be related to catelectrotonic facilita-
tion. At the present time, our studies are mainly concerned with these
two issues.
REFERENCES
1. Adrian, E. D. Synchronized reactions in the optic ganglion of Dytiscus.
J. PhysioL, 1937, 91, 66-89.
2. Adrian, E. D. General principles of nervous activity. Brain, 1947, 70,
1-17.
3. Barren, D. H., and Matthews, B. H. C. The interpretation of potential
changes in the spinal cord. /. PhysioL, 1938, 92, 276-321.
106 : Wolfgang Kohler
4. Bremer, F. Some problems in neurophysiology. London: Athlone Press,
1953.
5. Gerard, R. W., and Libet, B. The control of normal and "convulsive"
brain potentials. Amer. J. Psychiat., 1940, 96, 1125-1153.
6. Gibson, J. J. Adaptation, after-effect and contrast in the perception of
curved lines. J.exp.PsychoL, 1933, 16, 1-31.
7. Goldring, S., and O'Leary, J. L. Experimentally derived correlates be-
tween EGG and steady cortical potential. /. Neurophysiol, 1951, 14,
275-288.
8. Goldring, S., and O'Leary, J. L. Summation of certain enduring sequelae
of cortical activation in the rabbit. EEG din. NeurophysioL, 1951, 3,
329-340.
9. Kohler, Ivo. Uber Aufbau und Wandlungen der Wahrnehmungswelt.
Vienna: Rohrer, 1951.
10. Kohler, W. The place of value in a world of facts. New York: Liveright,
1938.
11. Kohler, W. Dynamics in psychology. New York: Liveright, 1940.
12. Kohler, W. Relational determination in perception. In L. A. Jeffress
(Ed.), Cerebral mechanisms in behavior. New York: Wiley, 1951. Pp.
200-243.
13. Kohler, W., and Held, R. The cortical correlate of pattern vision. Sci-
ence, 1949, 110,414-419.
14. Kohler, W., Held, R., and O'Connell, D. N. An investigation of cortical
currents. Proc. Amer. phil. Soc., 1952, 96, 290-330.
15. Kohler, W., Neff, W. D., and Wegener, J. Currents of the auditory cortex
in the cat. /. cell. comp. Physiol, 1955, 45, Suppl. 1, 1-24.
16. Kohler, W., and O'Connell, D. N. Currents of the visual cortex in the
cat. /. cell comp. Physiol, 1957, 49, Suppl. 2, 1-43.
17. Kohler, W., and Wallach, H. Figural after-effects: An investigation of
visual processes. Proc. Amer. phil Soc., 1944, 88, 269-357.
18. Kohler, W., and Wegener, J. Currents of the human auditory cortex.
/. cell comp. Physiol, 1955, 45, Suppl. 1, 25-54.
19. Lashley, K. S., Chow, K. L., and Semmes, J. An examination of the
electric field theory of cerebral integration. Psychol Rev., 1951, 58,
123-136.
20. Lorente de No, R. A study of nerve physiology. New York: Rockefeller
Institute, 1947.
21. Sperry, R. W., and Miner, N. Pattern perception following insertion of
mica plates into visual cortex. /. comp. physiol Psychol, 1955, 48,
463-469.
22. Sperry, R. W., Miner, N., and Myers, R. E. Visual pattern perception
following subpial slicing and tantalum wire implantations in the visual
cortex. /. comp. physiol Psychol, 1955, 48, 50-58.
PART III
COGNITIVE PROCESSES
Hans W attach
SWARTHMORE COLLEGE
BRIGHTNESS CONSTANCY
AND THE NATURE
OF ACHROMATIC COLORS
PART I
The problem of brightness constancy arises through the following cir-
cumstances. The amount of light which is reflected by an opaque object
and which stimulates the eye depends not only upon the color of the
object but just as much upon the amount of light which falls on the ob-
ject, that is upon the illumination in which the object is seen. When in
spite of this, the seen colors are in agreement with the object colors,
when a given object appears to have the same color in various illumina-
tions, we speak of brightness constancy.
The majority of investigators who aim at all at functional explana-
tions understand this problem to mean: How is illumination registered
and in what way is it taken into account so that the experienced colors
remain constant when the illumination is varied? In this version the
problem is a difficult one at the outset, for illumination is never directly
or independently given but is represented in stimulation only inasmuch
Reprinted with permission from the Journal of Experimental Psychology, Vol.
38, No. 3, June, 1948.
110 : Hans Wallach
as it affects the amount of light which is reflected by the objects. To be
sure, we perceive illumination as well as surface color; a spot of light
here, a shadow there, a brightly lighted region near the window or the
dim light of dusk on everything. But the fact remains that both vari-
ables, object color and objective illumination, affect the eye through
the same medium, the varying amount of reflected light. If the seen
illumination were found to be in agreement with the objective illumina-
tion, in principle the same problem would arise which we face regard-
ing the surface colors. There is only one stimulus variable to represent
two objective variables each of which seems to have its counterpart in
experience. Under these circumstances investigation has largely con-
sisted in the study of factors by which illumination could be recognized
and in the demonstration of their effectiveness in bringing about con-
stancy.
The following observations suggested a radically different approach
to the writer. They concern some variations of an experiment by A.
Gelb which demonstrated brightness constancy in a most impressive
way. Gelb's experiment 1 is most conveniently performed by opening
the door of a dimly lighted room and by suspending in the frame a piece
of black paper. This paper is illuminated by a strong projection lantern
which stands on the floor or on a low table and is tilted upwards so that
the part of its beam which is not intercepted by the black paper passes
through the open door onto the ceiling of the adjacent room where it is
invisible to the observer. In the light of the strong lantern the paper may
look white instead of black. When a white piece of paper is held up in
front of the black paper so that it too reflects the strong light of the
lantern, the black paper assumes a black color. According to the usual
interpretation it looks first white because no cues for the special strong
illumination are available when this illumination affects only one visi-
ble surface. With the introduction of the white paper into the beam a
special brilliant illumination becomes visible and constancy is restored:
the two papers are perceived with their real color.
The arrangement of Gelb's experiment lends itself to a still more im-
pressive demonstration. When the black paper is presented alone, re-
ducing the intensity of the lantern light by small steps to zero causes the
perceived color of the paper to vary all the way from white through gray
to black. Every change in illumination is accompanied by a correspond-
ing change in the perceived color. However, when a larger white paper
1 Described in W. D. Ellis, A source book of Gestalt psychology. New
York: Harcourt, Brace, 1939, p. 207.
Brightness Constancy and Nature of Achromatic Colors : 111
is fastened behind the black paper so that the latter is seen surrounded
by white, the same changes in illumination do not at all affect the seen
colors, which remain white and black throughout. Paired in this way
the colors are immune to changes in illumination and remain "con-
stant." It is rather a change in the perceived illumination which now
accompanies the change in the objective illumination.
The question arises: what determines the color with which the black
paper is seen at a given intensity of the lantern light when the paper is
presented alone? Do we deal in this situation with an absolute relation
between the intensity of the light which stimulates a portion of the
retina and the resulting perceived color? In considering this question
we have to remember that there is another variable in the situation, the
dim general illumination of the room. When this is varied it becomes
immediately clear that this general illumination also affects the color
of the black paper. When, with a high intensity of the lantern light, the
general illumination is raised, the color of the black paper changes from
white to gray, and this in spite of the fact that the paper too now reflects
light of a somewhat higher intensity than before. Only relatively, that
is in relation to the light which comes from other surfaces, has the light
reflected by the black paper become less intense.
Such dependence of the perceived color on the relative intensity of
the perceived light should be demonstrable in a much simpler form, and
this is the case.
In a dark room a white screen is illuminated by the light of two slide
projectors. In one of the projectors an opaque card with a circular hole
of % in. diameter is inserted, and the bright image of the hole is focused
on the screen. The slide for the other projector consists of a blank glass
covered with an opaque card with a circular hole of 1 in. diameter and
with a % in. cardboard disk which is pasted concentrically into the
hole. Focused on the screen this slide produces a bright ring. The two
projectors are so adjusted that this ring surrounds the image of the % in.
hole so that the edge of the latter coincides with the inner edge of the
ring. The light intensity of the projectors can be changed by running
them on variable transformers or by letting their beams pass through
episcotisters.
We have then on the screen a circular region (disk) and surrounding
it a ring-shaped region which reflect light intensities that can be sep-
arately controlled. When the intensity of the disk is kept constant and
that of the ring is widely varied, the color of the disk may change all the
way from white to dark gray. The disk looks dark gray when the light
112 : Hans Wallach
reflected from the ring is of high intensity, and it becomes white when
the brightness of the ring is greatly lowered. When the light intensity
of the disk is varied and that of the ring is kept constant, the color of
the disk, of course, undergoes similar changes. Again it is quite clear
that the color which appears in one region, namely in that of the disk,
depends on the relation of the light intensity of this region to that of its
surroundings. This is true also of the ring. It can be shown in corre-
sponding fashion that its color depends on the relation of the intensity
of the ring to that of the disk.
When the ring is altogether omitted so that the disk is seen in com-
pletely dark surroundings, it ceases to look white or gray and assumes
instead a luminous appearance similar to that of the moon at dusk.
Lowering the intensity of the disk greatly does not change this mode of
appearance, provided the rest of the room is really dark; the disk looks
merely dimmer. The same observation can be made with the ring when
it is presented without the disk, or with both the ring and the disk when
they are placed far from each other on the screen. Opaque colors which
deserve to be called white or gray, in other words "surface colors," will
make their appearance only when two regions of different light intensity
are in contact with each other, for instance when the ring surrounds the
disk or when two oblongs have the longer edges as their common border.
The importance of a close contact for the emergence of surface colors
becomes strikingly clear in the following observation. The intensity of
the disk is adjusted to be one-quarter that of the ring, which makes the
color of the disk a medium gray. An opaque object is moved from the
side into the beam of the lantern which projects the ring so that part
of it is blotted out by the shadow of that object. When this happens the
gray color disappears almost simultaneously from that part of the disk
which is adjacent to the shadow. It looks as if the dense gray there were
dissolving, leaving the screen transparent to let a light behind it shine
through. Brought about in this fashion, the change from surface color
to a luminous appearance is quite impressive. That side of the disk
which is still well surrounded by a brighter ring continues to show the
gray color, and between it and the luminous side the disk shows a steady
gradient in the density of the gray.
These observations make it clear that, at least under these conditions,
surface colors occur in our experience when regions of different light
intensity are in contact with each other and that the particular surface
colors which come about depend on the relation of these light intensities.
They are apparently the product of nervous processes of limited scope,
Brightness Constancy and Nature of Achromatic Colors : 113
for close spatial contact between the regions of different light intensity
is required for their emergence. Moreover, the degree to which surface
color is present in a certain region depends on the intimacy of the con-
tact between this region and its partner. This is easily demonstrated
by the following observations.
No matter what the brightness relation between ring and disk be,
the ring will always show a less dense surface color and have more of a
luminous appearance than the disk. This becomes quite clear when two
pairs of such regions are presented for comparison which are so chosen
that the intensity of the ring in one pair equals that of the disk in the
other one, and vice versa. Even the region of lower light intensity in
each pair, which is perceived as a gray, has a more luminous appearance
where it occurs in the ring than where it occurs in the disk. The most
obvious explanation for this difference in the mode of appearance is
that the disk is more under the influence of the ring than vice versa,
inasmuch as the disk is completely surrounded by the ring, whereas the
ring is in contact with the disk only on one side. This explanation agrees
well with the observation reported earlier that the elimination of part
of the ring rendered that part of the disk more luminous which was then
no longer enclosed by a region of different light intensity.
This influence under which surface colors emerge is clearly a mutual
one. Though less so, the ring does display surface color. There is a
great difference in the mode of appearance between a ring which sur-
rounds, for instance, an area of higher intensity and an equal ring
presented in an otherwise dark field. Whereas the latter looks merely
luminous, the former shows in addition to some luminosity a distinct
gray.
The mutual influence on which the emergence of surface colors de-
pends must also account for the fact that the particular colors which
come about depend on the relation of the stimulating light intensities.
It is probably best conceived of as some kind of interaction which takes
place as part of the nervous process which underlies color perception.
It will be remembered that the dependence of the perceived colors
on the relative intensities of the stimulating light was also evident in the
variations of Gelb's experiment which were first reported. It remains to
be added that the transition from surface color to a luminous mode of
appearance can be demonstrated with Gelb's setup in the following way.
At first the special illumination of the black paper and the general illu-
mination of the room are so adjusted that the black paper looks white.
When now the general illumination is further reduced, the paper be-
114 : Hans Wallach
comes more and more luminous, and it ceases altogether to look white
when the rest of the room is completely dark. Luminosity of the paper
can also be produced by excluding the general illumination from its
immediate neighborhood. By such measures a rather luminous gray,
not unlike that appearing in the ring, may also be achieved. Thus it is
not only in projected rings and disks that luminosity appears as an
alternative to surface colors when adequate differences in intensity are
lacking or when the contact between those regions is diminished. Clearly
discernible segregated objects, as for instance a suspended piece of
black paper, function in the same fashion.
PART H
So far, we have become acquainted with the way in which surface colors
come into existence and with the manner in which they depend on the
stimulus situation. They depend on the relation of stimulus intensities on
the retina which are so located with regard to each other that the sub-
sequent nervous processes interact. Now the question arises what bear-
ing this has on the problem of brightness constancy.
In order to answer this question, some clarification of the nature of
brightness constancy is needed. One may say that brightness constancy
prevails when a perceived color is in agreement with the corresponding
object color. Object color is a persistent physical characteristic of a
surface, the property to reflect a certain proportion of the light which
falls on that surface. For instance, a surface which looks black under
constancy conditions reflects about 4 per cent of the illuminating light,
and a white one about 80 per cent. This property, called reflectance, is
not conveyed to the eye as such. It is rather represented to the eye by
light of a given intensity. This fact constitutes the problem of brightness
constancy, for the intensity of the reflected light depends to the same
degree on the color of the reflecting surface as on the strength of the
illumination. If in our environment illumination were always and every-
where the same, the fact that our visual sense is not directly affected by
reflectances but only by the reflected light intensities would not raise a
problem in perception, for the reflected light could represent the object
colors unequivocally. But illumination varies widely, even between dif-
ferent parts of the same visual field, and often very different light inten-
sities come to represent the same reflectance to the eye and, in con-
stancy, produce the same color in the observer's experience. When, for
instance, a medium gray which reflects 20 per cent of the illuminating
Brightness Constancy and Nature of Achromatic Colors : 115
light is presented once in an illumination of an intensity 100 and again
under light of an intensity 300, the intensities of the reflected light are
20 and 60, respectively; if complete constancy prevails, both stimulus
intensities lead to perception of the same medium gray. Similarly the
white background on which the gray samples are shown will reflect light
of the intensity 80 in the weaker illumination and of the intensity 240
where it is in the stronger illumination, and the two differently illumi-
nated parts of the background will probably both be judged as white. At
first glance no orderly connection between stimulus intensity and per-
ceived color seems to exist.
There is, however, one feature in the stimulus situation which re-
mains the same when the illumination is varied. The intensity of the
light reflected by the gray in the weaker illumination (20) stands in a
ratio of 1 : 4 to that reflected by the white in the weaker illumination
(80), and the same ratio exists between the intensities reflected by the
gray and the white in the stronger illumination (60 and 240) . It is easy
to see that in the case of any given set of object colors the ratios of the
intensities of the reflected light remain the same for any change in illu-
mination which affects all of them. 2 Thus, if the perceived colors were
to depend on the ratios of the intensities of the reflected lights, they
would remain unchanged when a given set of object colors were pre-
sented in changed illumination, and constancy would be assured. A
medium gray may serve again as an example. Although it affects the
eye with different light intensities when the illumination is changed, it
would be perceived as the same color because the ratio of the intensity
that it reflects to the intensity of the light reflected by the surrounding
white would remain the same, for a change in illumination affects the
latter in the same proportion.
At this point we have to consider the observations reported in Part I.
They suggested that the perceived surface colors depend on the relation,
not yet quantitatively defined, of the light intensities in interacting re-
gions. But we now find that constancy would result, if our visual percep-
tion functioned in such a fashion that the perceived colors depended on
the ratios of the intensities of the reflected light.
Thus, we merely have to make the assumption that the relation on
which surface colors depend is one of simple proportionality to give the
observations of Part I a direct bearing on the problem of brightness con-
stancy. If this assumption were correct brightness constancy would find
2 This is a simple consequence of the fact already mentioned that object
colors reflect a constant fraction of the illumination.
116 : HansWallach
its explanation in the very process by which surface colors come about.
This assumption can be tested by simple experiments. If it is correct,
the particular colors which are perceived in a pair of ring and disk
should depend on the ratio of the intensities of the two regions, and
only on that ratio. In other words, no matter what the absolute inten-
sities of ring and disk may be, the same colors should be seen in the case
of any pair of intensities which happen to stand in the same ratio to each
other. This is, in close approximation, the case, as the following report
of quantitative experiments 3 shows.
Two pairs of ring and disk were used, in order to permit simultaneous
comparison. The intensity of each of these four regions could be varied
independently.
Four identical projection lanterns equipped with 500-watt bulbs
were used for this purpose. They were arranged in two groups, and each
group produced on the screen a pair of ring and disk as described in
Part L They were all so adjusted that they gave their respective regions
the same light intensity. This was done in the following way. First a pair
of ring and disk was formed with one lantern from group I and one from
group II, and the intensity of one of them was varied until the contour
between the ring and the disk disappeared because of brightness equal-
ity. Then these two lanterns were restored to their respective groups and
similar adjustments were made within each group by varying the light
intensities of the not yet equated lanterns.
The intensity variations required by the experiments were brought
about with the help of episcotisters through which the lantern beams
had to pass before reaching the screen. This technique has the advan-
tage that the episcotister apertures are a direct measure of the relative
intensities in the various regions.
Measurements were made by the method of limits. Ring and disk of
one pair and the ring of the other pair were kept at constant intensities,
and the intensity of the remaining disk was varied in suitable steps until
the S judged the colors of the two disks as equal.
In the first experiment one of the rings was given the full illumination
of its lantern and the disk inside it received half of the intensity, for its
light beam passed through an episcotister of 180 degrees aperture. The
light for the ring of the other pair was cut down to one-eighth of full
intensity by passing it through an episcotister of 45 degrees aperture.
3 These experiments were performed by the students of various seminars in
perception and classes in experimental psychology at Swarthmore College
under the author's supervision.
Brightness Constancy and Nature of Achromatic Colors : 117
The aperture for the disk of the latter pair was varied in steps of two
degrees. The following are the means of one upper and one lower limit
for each of five Ss: 24, 26, 24, 23, 24 degrees with a total mean of 24.2
degrees. This result means that, on the average, a light intensity in a
disk corresponding to an episcotister aperture of 24.2 degrees when it
is surrounded by a ring of an intensity of 45 degrees aperture brings
about in the S's experience the same gray as does a disk of an intensity
of 180 degrees aperture inside a ring of an intensity of 360 degrees
aperture. There is only a small deviation from the value of 22.5 degrees
which with 45 degrees forms the same ratio as does 180 degrees with
360 degrees. Comparing the grays in the two disks was not difficult for
the Ss. The great difference in absolute intensity between the two pairs
of ring and disk (8:1) made the less intense pair look much dimmer,
but that did not affect the distinctness of the disks' color. However, it
made the rings look very different; though both were white, the more
intense one was by far more luminous. This latter observation, which
was also made in most of the following experiments, seems to be impor-
tant, for it corresponds to a fact which can be observed in real constancy
situations. When identical sets of object colors are placed in different
illuminations and appear approximately the same, the set in the stronger
objective illumination is often also seen to be more strongly illuminated.
Perceived illumination and the different degree of luminous appearance
which was frequently observed in our experiments seem, functionally
speaking, to be closely related experiences.
In another experiment a disk of 90 degrees intensity was shown in a
ring of 360 degrees intensity. This combination, which forms an inten-
sity ratio of 4 : 1, brings about a much darker gray in the disk. In the
other pair, the disk whose intensity was varied was surrounded by a ring
of 180 degrees intensity. The proportionate value for the disk is here
45 degrees. The averages of two upper and two lower limits for each of
four Ss were 46, 52, 45, 44 degrees with a mean of 47 degrees.
In the following experiment the disk of the brighter pair was varied
and a ratio of 3 : 1 between ring and disk was used. In the darker pair,
the ring had an intensity of 1 80 degrees and the disk one of 60 degrees,
and the variable disk was surrounded by a ring of 3 60 degrees intensity.
Five upper and five lower limits were determined for each of three Ss.
The means were 1 13, 1 15, 121 degrees. The proportionate value is here
120 degrees.
It will be noted that so far all deviations from the proportionate values
were in one direction. They all imply that, where they occur, a disk of
118 : Hans Wallach
proportionate intensity in the dimmer pair looks darker than the disk
in the pair of higher intensity; viz., in the first two experiments the disk
in the less intense pair had to be given a slightly higher than propor-
tionate intensity to give a color match and in the last experiment the
disk in the more intense pair had to be made objectively darker. Thus,
although these deviations are small, they deserve our attention. Experi-
ments with an improved technique were made to find out how significant
they are.
To facilitate measuring a variable episcotister 4 was used for the de-
termination of the limits. This device permits changing the aperture
by definite amounts while it is spinning. Only when the Ss had given a
judgment of equality was the episcotister stopped and its angle measured
with a protractor.
It has been described above how the intensities of the four lanterns
were equated at the outset of the experiments. These equations are
likely to contain subliminal errors which could affect our measurements.
In the experiments which follow the episcotisters were interchanged
between the groups of lanterns after half the number of limits had been
determined for a given S, so that the group which during the first half
of an experiment produced the brighter pair of ring and disk were made
to produce the dimmer pair during the second half, and vice versa. Thus
any error in the original lantern adjustment which would affect the
measurements during the first half of the experiment in one direction
would in the second half affect it in the opposite direction. In this man-
ner such an error will appear in the scatter of the limit values but will
not affect their mean.
The first experiment ( 1 ) done with this improved technique was one
with a small difference between the brighter and the dimmer pair. The
former had a ring of 360 degrees intensity and a disk of 180 degrees,
and the other pair had a variable disk in a ring of 180 degrees. Four
Ss took part in the experiment. For each one four upper and four lower
limits were determined. Table 1 presents the means of these limits. The
proportionate value is here 90 degrees. It will be noted that the small
deviations from this value are in a direction opposite to those previously
reported, for they would imply that a disk of proportionate intensity in
the dimmer pair is perceived as a slightly lighter gray than the disk in
the more intense pair.
1 Designed and built by R. Gerbrands, Emerson Hall, Cambridge, Mass.
Brightness Constancy and Nature of Achromatic Colors : 119
TABLE 1
EPISCOTISTER SETTINGS IN DEGREES FOR DISK WITHIN RING OF
180 DEGREES IN COMPARISON WITH DISK OF 180 DEGREES
WITHIN RING OF 360 DEGREES
Subjects Ad. McN. Ba. Cl.
Upper limit
88
86
85.5
90
Lower limit
84
84
79.5
84.5
Mean
86
85
82
86
Grand mean: 85
TABLE 2
EPISCOTISTER SETTINGS IN DEGREES FOR DISK WITHIN RING OF
90 DEGREES IN COMPARISON WITH DISK OF 240 DEGREES
WITHIN RING OF 360 DEGREES
Subjects Mo. Cr. Ke. Cy.
Upper limit
61
62
73
74
Lower limit
62
64
68
67
Mean
61.5
63
70.5
70.5
Grand mean: 66.4
This is not so with the results of the following experiment (2), in
which a still lighter gray was produced and in which the intensity of the
dimmer ring was only one-quarter of that of the brighter one. In the
dimmer pair the ring had an intensity of 90 degrees and the disk was
variable, while in the brighter pair the ring had 360 degrees and the
disk 240 degrees of light. The results are given in Table 2. With the
Ss Mo. and Cr., 10 upper and 10 lower limits were determined, with Ke.
and Cy. only six. Individual differences are larger in this experiment.
For two of the Ss there was a marked deviation from the proportionate
value of 60 degrees, which implied that for them a disk of 60 degrees
intensity in the dimmer pair showed a slightly darker gray than the disk
in the brighter pair.
Ten Ss were employed in an experiment (3) in which the variable
disk was surrounded by a ring of 360 degrees of light and the dimmer
pair consisted of a ring of 90 degrees and a disk of 30 degrees intensity-
Six upper and lower limits were determined for each S, except for Ss Mo.
and Cr., who again supplied 10 pairs of limits each. The average of the
120 : Hans Wallach
TABLE 3
EPISCOTISTER SETTINGS IN DEGREES FOR DISK WITHIN RING OF
360 DEGREES IN COMPARISON WITH DISK OF 30 DEGREES
WITHIN RING OF 90 DEGREES
Subjects Ca. Ga. Hs. Ht. Lu. Ro. Mo. Cr. Ke. Cy.
Upper limit 104.5 91 117.5 116.5 113 130 128 107.5 105 113
Lower limit 92.5 91 99.5 98.5 103 112 100 103 97 95
Mean 98.5 91 108.5 107.5 108 121 114 105 101 104 Grand mean: 106
individual means as shown in Table 3 was 106 degrees, a clear deviation
from the proportionate value of 120 degrees. It implies that the gray
in the disk of low intensity looks somewhat darker than a disk of pro-
portionate value in the brighter pair.
The direction of the deviations from proportionate values encoun-
tered in the last two experiments was such that they could be regarded
as the effect of a slight influence of the absolute stimulus intensities on
the color process which otherwise could be conceived as functioning
according to a proportional law. The question arose whether these devi-
ations reflected intrinsic properties of the color process or whether they
were introduced by incidental experimental conditions. An answer can-
not yet be given and must be left to further detailed investigation. How-
ever, an experiment which was performed with this question in mind
will be reported below, because it will add the data of still another com-
bination of intensities.
It was suspected that the presence of the brighter pair of ring and
disk in the visual field when the gray in the disk of the dimmer pair de-
veloped was responsible for the fact that this gray looked a trifle too
dark. If the high intensities of the brighter pair had an influence across
the spatial interval on the colors which emerged in the dimmer pair,
this is what should have happened. Such an influence can be avoided
by presenting the pairs successively. This was done in the following
experiment (4) . The intensities in the brighter pair were 360 and 180
degrees, the ring in the dimmer pair was 90 degrees and the disk was
varied. Table 4 shows for four Ss the means of four upper and four
lower limits. Ordinarily, with an intensity ratio of 4 : 1 between the
rings the deviation under discussion was to be expected. It did not ap-
pear. The slight deviation from the proportionate value of 45 degrees
was in the opposite direction.
In another experiment, however, successive presentation failed to
Brightness Constancy and Nature of Achromatic Colors : 121
eliminate completely the deviation under discussion. Experiment 3 was
repeated with three further Ss who did the experiment twice, once with
successive and once with simultaneous presentation. The limits listed
in Table 5 are the averages of four determinations each. Although suc-
cessive presentation reduces the deviation from the proportionate value
of 120 degrees, it does not eliminate it.
TABLE 4
EPISCOTISTER SETTINGS IN DEGREES FOR DISK WITHIN RING OF
90 DEGREES IN COMPARISON WITH DISK OF 180 DEGREES
WITHIN RING OF 360 DEGREES
Subjects
Ad.
McN.
Ba.
CL
Upper limit
Lower limit
43
41
42
40
41
42
44
44
Mean
42
41
41.5
44
Grand mean: 42
TABLE 5
EPISCOTISTER SETTINGS IN DEGREES FOR DISK WITHIN RING OF
360 DEGREES IN COMPARISON WITH DISK OF 30 DEGREES
WITHIN RING OF 90 DEGREES
Subjects
CL
He.
Be.
Presen-
tation
Simul-
taneous
Succes-
sive
Simul-
taneous
Succes-
sive
Simul-
taneous
Succes-
sive
Upper limit
Lower limit
Mean
110.5
99.5
105
121
99
110
99
97.5
98
108.5
99
104
104
94
99
112
104
108
These deviations from proportionate values appear rather insignifi-
cant when one compares them with the remaining effect of the propor-
tional law. For example, in experiment 3, which showed the largest
deviation, a disk of an intensity of 30 degrees aperture had on the
average the same color as one of an intensity of 106 degrees aperture,
that is, an intensity 3.5 times as high. The deviation from the propor-
tionate value of 120 degrees amounts only to 12 per cent.
It should be mentioned at this point that such experiments can also
be done with a less elaborate setup. Two color mixers and one pro-
jection lantern suffice for a crude demonstration of the proportional
122 : Hans Wallach
law. With the help of a large color wheel of black and white disks and a
small one fastened on top of it to the same mixer one can obtain a ring-
shaped and a circular region in which the intensities of the reflected
light can be varied independently. On one mixer, e.g., the large wheel
can be set to show a sector of 90 degrees white and the small one a
sector of 45 degrees white. To the other mixer are fastened a small
wheel with a white sector of 180 degrees and a large wheel of 360
degrees white. When the mixers spin in general room illumination, one
sees a dark gray disk surrounded by a medium gray ring on one mixer
and a light gray disk in a white ring on the other one. However, when
the mixers are placed in separate, strictly local illumination they look
quite different. That illumination can be provided by a lantern equipped
with an opaque slide which has two circular holes a good distance apart.
It projects two narrow beams of light of equal intensity. When the mix-
ers are placed each in one of the beams at such a distance from the pro-
jector that their wheels are covered by the light almost to the outer rim
and the rest of the room is entirely dark, both color mixers show a white
ring and a light gray disk much alike in color. The reason for this change
is easy to understand. Under local illumination the two color mixers
provide exactly the same pattern of stimulus intensities as the setup in
experiment 4, and thus the same colors develop as in that experiment.
In general illumination, on the other hand, the pairs of ring and disk
are surrounded by regions of other intensity, e.g., the light reflected by
the wall of the room, which cooperate in determining the colors which
come about in the pairs. If, for instance, light reflected by a white wall
forms the stimulus intensity of the surrounding region, that intensity
stands to the intensity of the dimmer ring in a ratio of 4 : 1, and in this
relation the ring should assume a medium gray color, as indeed it did.
It was explained above how the assumption that the achromatic
colors depend on the ratios of the pertinent stimulus intensities accounts
for brightness constancy. On that occasion complete constancy was
shown to follow from this assumption. However, complete constancy
has hardly ever been demonstrated experimentally. An object color pre-
sented in reduced illumination usually looks somewhat darker than
another sample of that color in full illumination, though not as much
darker as the difference of the reflected light intensities would warrant
if there were no constancy. Yet complete constancy would follow from
a direct application of the proportional law. Deviations from propor-
tionality which occurred in our experiments are by far too small to
account for the usual lag in constancy. The difficulty resolves itself when
Brightness Constancy and Nature of Achromatic Colors : 123
it is realized that the proportional law cannot be applied so simply to
this situation. Here the two pairs of regions, the sample and its back-
ground in full illumination and the other sample with background in
reduced illumination, are not as completely separated from each other
as the corresponding regions in our experiments, for the regions of
different illumination are in contact with each other, and the brighter
one can have an influence on the dimmer one. In other words, we have
here a case where three or more regions of different intensity interact.
Such processes have not yet been sufficiently investigated, and no report
can be made at this time. It seems, however, quite likely that a full in-
vestigation will furnish the rules for the prediction of the lag in con-
stancy in individual experiments.
This report may so far have given the impression that, apart from
the small deviations discussed, the proportional law permits prediction
of color equations if the pertinent stimulus intensities are known. How-
ever, this is so only with important qualifications. To a certain extent
also the geometrical arrangement of the regions of different intensity
has an influence on what colors come about in these regions. Some brief
experiments which permit a first appraisal of the importance of these
conditions will be reported below.
In the measuring experiments so far reported the width of the ring
was % of the diameter of the disk so that the area of the ring was four
times as large as the area of the disk. A reduction of the width of the
ring to % of the diameter of the disk so that its area was about the same
as that of the disk did not affect the color in the disk as the following ex-
periment shows, in which the colors in two disks were compared which
were surrounded by rings of different width. Both rings were given the
same intensity of 120 degrees aperture; the disk in the narrow ring had
an intensity of 15 degrees and appeared as a very dark gray; the disk in
the ring of standard width was variable. The mean of two upper and two
lower limits for a single S was also 15 degrees. A number of other ob-
servers were satisfied with that equation.
The width of the narrow ring was further reduced so that it amounted
to only M.6 of the diameter of the disk. The same constant intensities as
in the last experiment were used. The averages of two upper and two
lower limits for each of two Ss were 37 and 37 degrees. This result
means that a disk of 15 degrees intensity inside the very narrow ring
looked as light as a disk of 37 degrees intensity inside a ring of standard
width. The outcome of this experiment was so striking that we repeated
it with another combination of intensities. The intensity of the two rings
124 : Hans Wallach
remained the same, but the disk in the very narrow ring had an intensity
of 60 degrees. Again a higher intensity was needed for an equation in
the disk inside the standard ring. The averages of two upper and two
lower limits for the same two Ss were 87 and 86 degrees. However,
with this intensity ratio of 120 : 60, which produces a light gray, the
effect of making the ring very narrow was not so great. It amounted only
to 45 per cent, whereas in the case of a ratio of 120 : 15 which normally
produces a very dark gray the disk in the standard ring had to be made
145 per cent more intense. On the whole it looks as if the very narrow
ring which has only one-quarter of the area of the disk cannot make the
disk color as dark as does a ring of sufficient width.
As just reported, no difference in the effect of a ring which has about
the same area as the disk and of one which has four times the area of
the disk has been found. Two further measurements were made with a
much wider ring. Its width was 1 .5 the diameter of the disk and its area
15 times that of the disk. In one experiment the intensity ratio between
the wide ring and its disk was again 120 : 1 5. In the disk of the standard
pair the averages of four upper and four lower limits for the two Ss were
17 and 16 degrees. When a ratio of 120 : 60 was used, averages for
two pairs of limits were 66 and 63 degrees. The deviations from 15 and
60 degrees respectively are probably incidental. At any rate, they are
not in the direction which would indicate an enhancement in the effec-
tiveness of the ring with increased width. It seems that, once the ring
has an area equal to that of the disk, any further increase in its width
does not affect the resulting color of the disk.
It was reported in Part I that a ring looks more luminous than a disk
of the same intensity in another pair in which the intensities of ring and
disk are the same as in the first pair but interchanged. The question
arises whether such a reversal of intensities also causes a color difference
in the regions of equal intensity. Two pairs of disk and ring in which
the area of the ring was the same as that of the disk were presented
and lights were so arranged that in one pair the lower intensity was in
the ring and in the other pair in the disk. The two higher intensities
in the two pairs amounted both to 360 degrees, the ring of lower in-
tensity was kept at 45 degrees, and the disk of lower intensity was varia-
ble. Measurements were made with four Ss. The means of three upper
and three lower limits were 54, 71, 83, 86 degrees. These figures indi-
cate that for the same intensity ratio the lower intensity appears as a
lighter gray when it is given in the ring than when it is given in the disk.
A rather dark gray results from a ratio of 360 : 45 degrees. In the case
Brightness Constancy and Nature of Achromatic Colors : 125
of smaller ratios which give rise to lighter grays the differences in color
which result when the intensities of ring and disk are interchanged are
very much smaller. For an intensity ratio of 2 : 1 only a difference in
luminosity can be discerned.
SUMMARY
It was found that opaque achromatic surface colors are perceived when
light of different intensity stimulates adjacent areas on the retina. The
achromatic color which is seen in a particular region must be regarded
as the result of stimulation received from that region and of stimulation
from neighboring regions. Although these colors are qualities which
are perceived in a given region, they are products of an interaction
process, which depends on difference in stimulation in at least two
areas. In the absence of a suitable difference in stimulation a color of an
entirely different mode of appearance is seen. A single bright region in
an otherwise dark field, for instance, looks luminous instead of white,
and reducing the light intensity in that region fails to make it look gray;
it continues to appear luminous and merely becomes dimmer.
The first steps were taken to investigate quantitatively the rules of
this dependence in the simplest case, that of two regions of different
intensities of stimulation where one region surrounds the other. The
colors which come about under these circumstances depend in close
approximation on the ratios of the intensities involved and seem inde-
pendent of the absolute intensity of local stimulation. The region of
higher intensity will assume the color white and that of lower intensity
will show a gray (or a black) which depends on the intensity ratio of
the two regions. The greater the difference in intensity the darker will
be the gray which appears in the region of the lower intensity.
It can be shown that a dependence of perceived colors on the ratios
of stimulus intensities accounts for the constancy of achromatic colors
under varying illumination. Complete constancy would follow from
this rule of interaction of two intensities. The fact that measurements of
brightness constancy rarely give results which denote complete con-
stancy presents no difficulty for this explanation. These experiments
involve interaction between more than two regions of different stimulus
intensity.
Hans Wallach and D. N. O'Connell
SWARTHMORE COLLEGE
THE KINETIC
DEPTH EFFECT
The problem of how three-dimensional form is perceived in spite of the
fact that pertinent stimulation consists only in two-dimensional retinal
images has been only partly solved. 1 Much is known about the im-
pressive effectiveness of binocular disparity. However, the excellent
perception of three-dimensional form in monocular vision has remained
essentially unexplained.
It has been proposed that some patterns of stimulation on the retina
give rise to three-dimensional experiences, because visual processes
differ in the spontaneous organization that results from certain proper-
ties of the retinal pattern. Rules of organization are supposed to exist
according to which most retinal projections of three-dimensional forms
happen to produce three-dimensional percepts and most retinal images
of flat forms lead to flat forms in experience also. This view has been
held mainly by Gestalt psychologists.
Another approach to this problem maintains that the projected stimu-
lus patterns are interpreted on the basis of previous experience, either
Reprinted with permission from the Journal of Experimental Psychology, Vol.
45, No. 4, April, 1953.
1 Most of the work reported in this paper was done while the senior author
was holder of a John Simon Guggenheim Memorial Fellowship.
The Kinetic Depth Effect : 127
visual or kinesthetic. However, such empiricistic assumptions do not
explain much, unless it is made clear how those previous experiences
come about whose influence is supposed to account for the current per-
ception. Kinesthetic form perception itself presents far more complex
problems than does three-dimensionality in vision, and recourse to
kinesthesis appears at present quite futile. Retinal disparity can of
course account for that previous visual experience insofar as Ss with
binocular vision are concerned. Whether in the absence of binocular
vision, e.g., in congenitally monocular Ss, head movement parallex can
fully play the role of binocular parallex appears doubtful. It seems that
all we have left to account for an original experience of three-dimen-
sional form in monocular Ss is the assumption that certain patterns of
retinal stimulation will naturally produce experience of solid form.
However, once it is assumed that perception of three-dimensional form
can follow directly from retinal stimulation because of spontaneous
organization of visual processes alone, there is no point in postulating an
influence of past experience.
Unfortunately it appears that no one has succeeded in formulating
rules of spontaneous organization adequate to predict which pattern
of retinal stimulation will lead to perceived flat figures and which one
will produce three-dimensional forms. We have made a vain attempt
of our own and have become convinced that the three-dimensional
forms perceived in perspective drawings, photographs, etc., are indeed
a matter of previous experience. 2 In this situation the search for a visual
process which can account for an original perception of three-dimen-
sional form in monocular vision becomes imperative. Such a process
will be described in this paper.
When one moves about, the retinal image of solid objects lying to
the side of one's path not only expands but also distorts, because the
objects are seen successively from different directions. More specifically,
a retinal image distorts under these conditions as if the corresponding
object were rotated through a certain angle in front of the eye of a
stationary observer. -
This fact suggests a simple technique for the investigation of the per-
ceptual results of these distortions. An object is placed between a puncti-
f orm light source and a translucent screen and is rotated or turned back
2 Evidence supporting this point of view will be reported in a subsequent
paper. [Hans Wallach, D. N. O'Connell, and Ulric Neisser, The memory effect
of visual perception of three-dimensional form. /. exp. PsychoL, 1953, 45,
360-368. Ed.]
128 : Hans Wallach and D. N. O'Connell
and forth. Its shadow is observed from the other side of the screen. The
shadow-casting object is placed as close to the screen as possible,
whereas the distance between the light source and the object is made
large. Owing to this arrangement isometric projection is closely ap-
proximated. The shadows of a great number of three-dimensional forms,
solid or wire-edged, will be perceived as three-dimensional under these
circumstances. The shadows of some forms will look three-dimensional
only in such a moving presentation; that is, in none of the positions
through which such a form passes during rotation will it cast a stationary
shadow which looks three-dimensional. With such forms one can study
this effect in isolation. It will be referred to by the term "kinetic depth
effect." It is important because it answers our problem: It appears that
the kinetic depth effect can cause a genuine perception of three-dimen-
sional form in a monocular S whenever he moves and keeps looking at
an object which does not lie directly in his path.
Similar setups have been used before by Miles (3) and Metzger (2) .
Miles presented to Ss the shadow of a two-bladed fan wheel in rotation,
the shaft of which was parallel to the screen on which the shadow was
formed. His Ss reported a large number of different motion patterns,
most of them involving depth, but no attempt was made to find out
whether the object that was seen in motion was a good representation of
the fan wheel whose shadow was presented. Metzger also was not pri-
marily concerned with the perception of three-dimensional form which
such a setup can yield. In fact, he investigated with great thoroughness
the effects of a very special kind of arrangement which is not favorable
to the emergence of stable three-dimensional forms. He had arrange-
ments of vertical rods rotate about a vertical axis and showed their
shadows in a low oblong aperture which hid the ends of the rod shadows
from view. As in Miles's experiment the motion patterns seen by a given
S during an extended period of inspection were very changeable. More-
over, naive Ss mostly differ among each other as to whether the first
movement process which they perceive is in three dimensions or takes
place in a plane. Suggestion has a strong influence, in the course of
longer observation as well as initially. We shall see below that the pat-
terns of line shadows which Metzger presented to his Ss do not contain
the condition essential for the kinetic depth effect.
The depth observed in Lissajous figures (1,4) is the result of com-
plex effects and will be discussed in a later paper.
The Kinetic Depth Effect : 129
EXPERIMENTS TO DEFINE CONDITIONS OF KINETIC
DEPTH PERCEPTION
Experiment 1. Rotation of solids. One of the figures which we pre-
sented to many Ss was a solid block in the shape of a roof with sloping
gables, as shown in Fig. 1A. This was rotated continually about its
longest axis. Figures IB and 1C show two of the forms of the shadow
ABC
FIG. 1. Solid form (A) used in experiment 1, and two samples (B, C)
of its shadows.
during rotation. When any one of the three figures is shown stationary to
a naive S, it is described as a two-dimensional figure. When the solid
is slowly rotated so that the shadow undergoes continuous deformation,
S sees a three-dimensional solid in rotation. The direction of the seen
rotation may or may not coincide with that of the object behind the
screen, and this is in agreement with the conditions of stimulation which
give no clue of the object's real direction of rotation. Which one of the
two directions of rotation is seen seems entirely a matter of chance, and
with prolonged observation S usually experiences a number of spon-
taneous reversals of the direction of rotation. Occasionally an S sees for
long periods reversals after each rotation of 180 degrees.
In experiments like this the impression of three-dimensional form
is so natural that many Ss who are not psychologists are not astonished
by their observations. They correctly assume that behind the screen is
just such an object as they see. Only after reversals of the direction of
rotation have occurred do they begin to wonder.
Where the kinetic depth effect takes place, a rigid three-dimensional
form in rotation is seen instead of the distorted two-dimensional figure
given on the shadow screen. The distorting two-dimensional shape may
occasionally be seen after prolonged exposure of the same kinetic pres-
130 : Hans Wallach and D. N. O'Connell
entation, or when one looks from a sharply oblique direction at the
screen. The S's experience of a continuously flowing form seems to him
abnormal or unusual, although this is exactly what is given on the retina.
The essential difference between this two-dimensional experience and
the three-dimensional form which is usually seen seems to be that the
latter is unchanging and rigid instead of ever changing and flowing. The
changes in the shape of the retinal image are accounted for by a per-
ceived rotation of the three-dimensional object, whereas in the two-
dimensional process the seen movement distorts the form itself. As far
as we can see, the distortions of the perceived two-dimensional form
agree closely with the changes of the retinal image. But the perceived
three-dimensional form is not determined merely by what is presented
on the retina at a given moment. A single one of the projections of the
shadow-casting object which make up the changing shapes of (he
shadow does not look three-dimensional. Only the sequence of chang-
ing shapes gives rise to the seen three-dimensional form. In other words,
a single projection causes a three-dimensional form to be seen only be-
cause it was preceded by a number of other projections. By itself, it
simply does not convey enough data about the three-dimensional form
which it represents. The seen three-dimensional form is richer in struc-
ture than any single projection and it is built up in a temporally extended
process. The individual retinal image determines only which aspect of
the turning three-dimensional form appears to be given at the moment.
Thus, perceptual experience far surpasses what one should expect of a
process determined by momentary stimulation and seems to a higher
degree a product of the immediate past than of stimulation occurring
at a given moment.
Experiment 2. Partial rotation of wire figures. Experiments of the
kind just reported differ in two ways from the realistic situations in
which the kinetic depth effect occurs. In the first place, a shadow-casting
object is put through a complete rotation and produces a pattern of dis-
tortion on the screen which corresponds to that obtained when under
natural viewing conditions an S moves completely around an object,
which is hardly ever done. In the second place, in the shadow of a solid
object an edge of the three-dimensional form is visible only for a com-
paratively short period, namely as long as it forms a contour of the
shadow, whereas in the realistic situation it usually can also be seen
when it passes across the front of the figure. We therefore performed
quantitative experiments with wire figures which of course show all
The Kinetic Depth Effect : 131
edges continuously, and had them turn back and forth through an angle
of only 42 degrees.
FIG. 2. The wire "parallelo- FIG. 3. The wire "helix" used in experiment
gram" used in experiment 2. 2 (A) and its appearance from the top
Figures 2 and 3 each show projections of two of the wire figures used.
The figure represented in Fig. 2 can best be described as a parallelogram
containing one diagonal, which was bent along this diagonal so that
the planes of the upper and the lower half formed an angle of 1 10 de-
grees with each other. The other figure (Fig. 3 A) consisted of a piece
of wire twice bent to form part of what might be described as a triangular
helix. Figure 35 shows a view from the top. These figures were turned
back and forth through an angle of 42 degrees at a rate of one cycle per
1.5 sec. and their deforming shadows were shown to individual Ss one
at a time for as many periods of 10 sec. as were necessary to obtain a
clear report. After each 10-sec. exposure, S was asked for a report.
In the case of the parallelogram (Fig. 2) all of 50 Ss sooner or later
during the exposure periods reported seeing a three-dimensional figure
turning back and forth apparently very much like the wire figure behind
the screen, except that it sometimes appeared to S in its inverted form.
In the case of the "helix" 48 of the same 50 Ss reported a three-dimen-
sional form like the wire figure and 2 Ss saw a flat zigzag line which
distorted before their eyes, a process which corresponds to the changing
pattern of stimulation. Evidence that the three-dimensional forms which
were perceived were due to the kinetic depth effect and that none of the
projections represented in the deforming shadow by itself would have
been seen as three-dimensional was obtained in the following way. To a
new group of 22 Ss five such projections of each figure were presented
individually in random order. They were the projections of the two
extreme positions at which the figures stopped and started to turn the
132 : Hans Wallach and D. N. O'Connell
other way and of three intervening positions chosen to be at about 10
degrees of rotation apart. None of these stationary projections looked
three-dimensional to any one of the 22 Ss. Positions intermediate be-
tween the ones presented resembled them so closely that it seems im-
possible that they would lead to a radically different perceived form.
Often the turning wire figures were seen three-dimensionally im-
mediately upon presentation. Of 16 Ss for whom time records were
taken, 1 1 reported the "parallelogram" in the correct three-dimensional
shape after the first 10-sec. exposure period. In the case of the "helix"
only 4 of 16 Ss reported its shape correctly after the first exposure. The
reason for this difference is yet to be investigated.
Once it has occurred, the three-dimensional impression is so strong
that one cannot voluntarily see the two-dimensional figure which is
given on the screen. It has been mentioned that the perceived three-
dimensional figure may resemble the inverted form of the wire figure
rather than that figure itself. This is to be expected, inasmuch as a three-
dimensional figure and its inverted form have identical projections,
such that a given projection is always a representation of both, a figure
and its inverted form. The added fact that with prolonged observation
of the distorting shadow the perceived three-dimensional figure may
spontaneously invert in Necker cubelike fashion is also a consequence.
Experiment 3. Rotation of a truncated cylinder. A systematic in-
vestigation of the kinetic depth effect resulting from rotation of a solid
was attempted in the following manner.
A cylinder which is rotated about its axis casts a shadow which does
not change in any way, but any cut that is taken off the cylinder and
which is not at right angles to its axis produces a characteristic deforma-
tion of its shadow. A large number of wooden cylinders, about as high
as wide, were made and cuts were varied systematically. The solid forms
which were produced in this fashion were shown in complete rotation
to Ss who had no previous experience with our studies.
The results can be summarized in the following way: (a) Shadows
whose only deformation consists in an expansion and contraction in one
dimension will look flat; a dark figure is seen which periodically be-
comes wider and narrower. An example is the shadow of a rectangular
block which is rotated about an axis parallel to a set of edges, (b) Shad-
ows which display contour lines that change their direction and their
length will appear as turning solid forms. The roof-shaped figure de-
scribed above (experiment 1) is an example; the shadow contour pro-
duced by an edge of a gable tilts and changes its length simultaneously
The Kinetic Depth Effect : 133
(compare Fig. IS and 1C), (c) Curved contours which are deformed
without displaying a form feature which identifies a specific point along
the curve are seen as distorting, often even if for some reason the shadow
is seen as a three-dimensional form. This peculiarity is in disagreement
with our description of the kinetic depth effect and has delayed our work
for years. It is now clear that the perceived distortions of deforming
curved contours are not related to the kinetic depth effect at all. They
will be dealt with in a later paper. 3
Experiment 4. Rotation of straight rods. When it became clearly
recognized that shadows which display contour lines that change their
direction and length will appear as turning solid forms, we checked
whether this would also apply to single lines. Will a detached line which
changes its direction on the screen and its length at the same time display
a kinetic depth effect, i.e., will it appear to tilt into depth in such a way
as to account for its shortening? The following experiments show that
this is the case.
When a rod is fastened to the end of a vertical shaft at an angle of,
say, 45 degrees and is rotated about this shaft, the shadow which it pro-
duces tips from side to side. From a tilt of 45 degrees toward the left,
the shadow rights itself and goes through the vertical into a tilt of 45
degrees toward the right and back through the same positions. At the
same time it shortens and lengthens periodically. Its end points move
on horizontal lines across the screen. This process is invariably seen as a
motion in three dimensions, as nearly as one can tell exactly like the
real movement of the rod behind the screen, a rotation describing a
conical surface about a vertical axis.
This setup lends itself to a variation which is interesting because the
three-dimensional motion which is perceived is a different one from
that which the rod behind the screen undergoes. The shaft on which the
rod turns is tilted toward or away from the screen by 15 or 20 degrees.
This changes the pattern of expansion and contraction of the rod
shadow completely; its end points now move on elliptic paths on the
screen. Only rarely does the rod motion seem to describe a circular cone
like the real movement of the rod behind the screen. In most cases a
movement is seen in which the movement component of tilling toward
or away from S is much smaller than the lateral movement component
such that the surface described is that of an elliptic cone. Whereas the
real rod goes in one rotation from a tilt toward S into a tilt away from
3 Hans Wallach, Alexander Weisz, and Pauline Austin Adams, Circles and
derived figures in rotation. Amer. J. PsychoL, 1956, 69, 48-59. Ed.
134 : Hans Wallach and D. N. O'Connell
him, the perceived motion of the rod is restricted either to a changing
forward tilt or to a changing tilt away from S. Either one of these two
different movements can of course be perceived because, like a three-
dimensional figure and its inverted form, they would if they were ob-
jectively given produce the same projection.
It seems that what distinguishes in these experiments perceived move-
ment from the two-dimensional process which is given on the screen and
therefore on the retina is the fact that the perceived rod is seen with a
constant length. Tilling into depth is seen instead of shortening.
Inasmuch as the rod as a whole must anyway be seen to move, this
tilting motion is only a modification of a necessary process and not an
added change, which is what a perceived stretching and shrinking would
amount to. Thus, a tendency to see one motion instead of the two simul-
taneous movements of the two-dimensional process (a tipping from
side to side, and a stretching and shrinking) may be held responsible
for the depth effect. Another possibility would be a selective principle
according to which a line of constant length is seen rather than a chang-
ing one. No decision between these two possibilities can be made on the
basis of our present results.
Experiment 5. Rotation of T and A figures. It is important to realize
that a shadow line must undergo both a displacement and a lengthening
or shortening in order to produce a kinetic depth effect. Both these
changes must be given together. A change in length alone is not suffi-
cient to produce a reliable kinetic depth effect. This is shown in the
following experiment.
A piece of Ke-in. wire 4 in. long was fastened at right angles to a
vertical shaft of the same diameter so that the two formed a figure of
T shape. When this form was turned back and forth about the vertical
shaft through an angle of 42 degrees, the shadow of the horizontal wire
formed a line of 105-mm. length which periodically contracted to a
length of 75 mm. and expanded again. We presented it with a rate of
one period per L5 sec. for 10 sec. to 24 Ss.
Eighteen of 24 Ss saw a line in the plane of the screen expanding and
contracting and only 6 Ss perceived the line turning in the third di-
mension. Had the horizontal wire been in an oblique position with re-
spect to the axis of rotation so that its shadow had shown a displacement
in addition to the change in length, the kinetic depth effect would have
been obtained with the majority of the Ss, as the experiment with the
oblique rod would predict.
The Kinetic Depth Effect : 135
Because of its importance we had the experiment with the T figure
repeated by another E and obtained the same result: Of 40 Ss, 30 saw
the horizontal line expand and contract in the plane of the screen and 1
saw it turn.
These results must be compared with those of a different wire figure
which does produce a kinetic depth effect under identical conditions.
Comparable data come from a shadow presentation of an equilateral
triangle. Its sides consisted of wires 5 in. in length, and one of them was
tilted by 15 degrees against the horizontal. When this figure was turned
back and forth, the shadow of one of its sides changed from a slope of
45 degrees to one of 57 degrees and thus presented conditions favorable
to the kinetic depth effect. Of 20 Ss who observed the triangle for 10
sec., 17 reported a turning in depth and only 3 expansion and contrac-
tion. (With such a plane figure the kinetic depth effect consists, of
course, in the perception of a plane figure that turns into depth.) This
score is to be compared with that for the T figure where only one-third
of the Ss saw a turning. The difference is reliable at the .001 level of
confidence.
Still, the 6 and 10 Ss who saw the T figure turning in depth need be
accounted for, if our claim is correct that they are not the outcome of a
genuine kinetic depth effect. Here the following result is significant.
When the T figure was presented following the presentation of the
triangle, 15 out of the 17 Ss who saw the triangle turn saw the T figure
turn also. (Three Ss who saw the triangle expand and contract reported
the same for the T figure.) In other words, when a figure that Ss saw
turning preceded the presentation of the T figure, a large majority of
Ss saw it turning also. This shows a strong influence of a previous per-
ception on the manner in which the T figure is seen, for only one-third
of the Ss saw the T figure turning when it was given as a first presenta-
tion. This difference is significant at the .05 level of confidence. We
suggest that in the case of these latter Ss some such influence of a
previous experience, though a more remote one, has been at work also.
We are inclined to conclude that a line which changes in length but
is not displaced at the same time does not give rise to the kinetic depth
effect. This agrees well with the already reported finding that a solid
shadow which expands and contracts only in one dimension will not
show the kinetic depth effect either,
136 : Hans Wallach and D. N. O'Connell
THE ACCURACY OF KINETIC DEPTH PERCEPTION
When we described the kinetic depth effect in complex figures, we
stated that the change in the retinal image is accounted for in per-
ception by a rotation of a three-dimensional form. This implies, of
course, that a real form which is like the perceived one would in rotation
produce a sequence of retinal images very similar to that actually given,
or, in other words, that the perceived form resembles closely the
shadow-casting object. That this is the case has been confirmed by many
Ss to whom the shadow-casting object was directly shown immediately
following the kinetic presentation.
For more stringent confirmation several methods were used: In the
case of the "helix" (Fig. 3 ) , Ss were asked to bend a piece of wire into
the shape of the figure which they saw turning on the screen. Only Ss
who reported seeing a three-dimensional form turning back and forth
were given this test. Of 29 Ss, 13 made good reproductions, 12 fair ones,
and only 4 Ss made poor reproductions. Eleven of the 12 Ss whose re-
productions were fair were later asked to make another reproduction
while they looked directly at the turning wire figure. Of these only 7
were able to make good reproductions and 4 made only fair ones under
these conditions. Altogether there were 8 Ss who could make only fair
reproductions when they viewed the figure directly.
In the case of the parallelogram (Fig. 2), the accuracy of perception
by virtue of the kinetic depth effect was checked by showing S four
similar wire figures and asking him to pick out the one that matched
best the form he saw turning on the screen. As mentioned earlier, the
bend in the shadow-casting figure amounted to an angle of 1 10 degrees
between the planes of the two triangles which made up this figure. This
angle is, of course, characteristic of its three-dimensional form. If this
angle were 180 degrees, the wire figure would be plane. In our four
models the bend amounted to angles of 95, 1 10, 125, and 140 degrees,
respectively. The one with the 110-degree angle was an exact copy of
the shadow-casting figure; the other three models had the same height
as the standard and were made to produce projections on the frontal
plane which were identical with the projection of the standard. The four
models were inserted in a wooden block and handed to S. The choices
of 30 Ss who had previously reported seeing the three-dimensional figure
were 8, 17, 4, and 1 for the 95, 110, 125, and 140-degree figure, re-
spectively. Only one S found it impossible to make a choice. It is un-
The Kinetic Depth Effect : 137
fortunate that we did not include one more model with a still sharper
angle, but even so the data give a rough idea of the accuracy of the per-
ception of three-dimensional form which is based on the kinetic depth
effect. 4
Experiment 6. Rotation of luminous rod. Another attempt to obtain
a measure of the accuracy with which depth perception functions by
virtue of the kinetic depth effect was made employing a straight line
which rotated in an oblique plane. When a line is turned in a frontal-
parallel plane about its midpoint, the end points of its retinal projection
move on a circle. However, when it rotates in an oblique plane, its reti-
nal image changes in length as it turns, and its end points move on an
elliptic path. Therefore, when in the dark a luminous line is rotated in a
frontal-parallel plane about its midpoint and is exposed from behind
an elliptic aperture, its retinal projection will be the same as that of a
line which turns in an oblique plane, for the aperture causes the line to
be visible with the same changes in length. If the motion of the line that
rotates in an oblique plane can be correctly perceived with the help of
the kinetic depth effect alone, then the line rotating behind an elliptic
aperture should also appear to rotate in an oblique plane, because the
two lines produce identical stimulation so far as the kinetic depth effect
is concerned. Other cues for depth perception as, for instance, retinal
disparity, would give rise to experienced rotation in an oblique plane
only in the first case. Thus when the rotating line behind the aperture
is presented to a naive S and he perceives it turning in a properly oblique
plane, one can be sure that this is due to the kinetic depth effect. By
this procedure, as by the use of the shadow screen, the effect can be
studied in isolation; other cues for depth perception would tend only to
prevent the line from turning in an oblique plane.
A %-in. lucite rod, approximately 23 in. long, served as light source
for the luminous line. Its ends were flat and finely ground to admit a
maximum of light They were inserted in metal caps which contained
hidden flashlight bulbs and also served as mountings. The light from
these bulbs made the lucite rod appear to glow evenly over its whole
length. The rod was inserted in a U-shaped sheet-metal trough of proper
length and attached by the mountings. A bushing was fastened to the
back of the trough at its midpoint and attached to the horizontal slow
shaft of a reduction gear motor, so that the lucite rod could be turned
4 It should be noted that a turning by 42 degrees produces only a moderate
distortion of the shadow. Its width, which is most strongly affected, suffers a
reduction of only 27 per cent.
138 : Hans Wallach and D. N. O'Connell
in a vertical plane. The trough and rod were covered by a long strip of
cardboard into which was cut an aperture 22 in. long and % in. wide.
Through this aperture part of the rod's surface was visible. In front of
this apparatus was a frame to which cardboards with different elliptic
apertures could be attached parallel to the plane of rotation of the rod.
Three different elliptic apertures were used. They were 40 cm. long and
35 cm. wide, 39.5 cm. long and 28.9 cm. wide, and 40 cm. long and
20.6 cm. wide, respectively. They produced projections of the turning
luminous rod identical with the projections produced by a luminous line
that turns in a plane forming an angle of 29, 46, and 59 degrees, re-
spectively, with the plane of the aperture. They were attached with the
large axis of the elliptic opening in vertical position.
The S was seated in front of this apparatus at a distance of 9 ft. He
had before him a small table covered with a light gray cardboard. A
metal rod was joined at right angles to a shaft which was fastened verti-
cally to the table, so that the rod could be swung around in a plane
parallel to the table top about an inch above it. A degree scale was
marked out on the cardboard by which the position of the rod could be
read. With the help of the rod, S could indicate the position of the plane
in which the luminous line seemed to turn. A darkroom amber bulb
illuminated this arrangement in such a way that S could see the rod but
not the scale markings and that the remainder of the room was com-
pletely dark. After each setting of the rod E asked S to close his eyes
and then read the scale with a flashlight.
During testing E asked S to look at the luminous line before him and
close one eye, and that eye was covered. The luminous line which, when
presented in the appropriate elliptic aperture, produced the projection
of a 46-degree tilt was set into clockwise rotation at a rate of one revolu-
tion in 10 sec. When S reported that it turned in an oblique plane, he
was asked to turn the measuring rod before him into a position parallel
with that of the luminous line in rotation. This was repeated with the
59-degree and the 29-degree apertures, and thereafter the three aper-
tures were used twice more in random order for the purpose of practice.
Neither in this practice period nor later during the experimental trials
was S told whether his settings were correct or not correct. No time
limit was set on the presentation of the revolving line and S made his
setting when he felt ready. After a rest period of 3 or 4 min. the experi-
mental series began. It consisted of nine presentations; that is, each one
of the three apertures was presented three times in random order.
The means and SD's of the 15 means of the three measurements for
The Kinetic Depth Effect ; 139
each of the three degrees of tilt were 14.9 (SD = 6.6), 44.0 (SD =
3.4), and 59.8 (SD = 4.4) for the 29, 46, and 59-degree tilt, re-
spectively. There was no overlap between the means of the settings by
individual Ss from one tilt to another, and there was no overlap between
the individual settings which a given S made for the three tilts.
Where projections of tilts of 46 degrees and of 59 degrees were pre-
sented, the averages for all Ss came close to the expected values. How-
ever, in the case of the 29-degree tilt, the average of 14.9 deviates sig-
nificantly from this value; only one individual setting out of 45 is as
high as, or exceeds 29 degrees.
Why this is so is not clear. However, it should be pointed out that
the change in length which a line turning with a 29-degree tilt under-
goes amounts only to 12,5 per cent, and one of 14.9-degree tilt (the
value of the average) only to 3.4 per cent. In other words, a tilt of 14.9
degrees produces a change in length that is very likely below the limen.
Yet, that does not necessarily mean that those Ss who gave settings of
15 degrees or lower did not receive effective stimulation for a tilt.
Whereas the small angles of tilt (due to the negligible change of the
cosine function in this range of values) probably do not lead to a change
in length sufficient to produce a kinetic depth effect, settings of such
low values do not indicate that no tilt was perceived when these settings
were made. In experience, a tilt of 15 degrees is of distinct significance,
and objectively conditions of stimulation of 29 degrees were given.
It would have been important to find out whether these results can
be improved by making the luminous line wider. To make it wider
would have the advantage that the contraction of the line would be
given by a change in its proportions, that is, in figural terms, rather than
by a change in its absolute length. Improved results would indicate that
change in proportions is effective in producing the kinetic depth effect.
Unfortunately, this variation could not be done with the present setup,
because a wider line would have shown up the intersections with the
elliptic aperture by their changing obliqueness. A more expensive man-
ner of presentation would be needed.
OTHER FACTORS IN KINETIC DEPTH PERCEPTION
Experiment 7. The effect of angle constancy. Such a variation of the
experiment might have contributed to the solution of the following prob-
lem. It has been reported that shadows of solid blocks will produce the
kinetic depth effect only if they have contours which are displaced and
140 : Hans Wallach and D. N. O'Connell
change their length simultaneously. Should we assume then that the
presence of such contours solely accounts for the kinetic depth effect in
complex forms, imparting their depth to the whole figure, or do com-
plex forms produce such an effect in their own right? Just as we have
considered the possibility that a line is seen to move into the third di-
mension to account for the given change of its retinal image while it is
perceived with constant length, we might assume a tendency to see in
general rigid, unchanging forms instead of the given distorting shapes.
For the present this question must remain unanswered.
However, a question which can be considered a part of the question
just raised was actually put to a test. Most shadows of turning figures
do not only display contours which are displaced and change their
length; their angles also change. Is there a separate tendency for angles
to remain constant which produces kinetic depth effects?
To answer this question we used a figure which consisted of three
rods all meeting in one point under angles of 1 10 degrees and forming
a wire-edged representation of an obtuse corner (Fig. 4). When this
figure was turned back and forth through an angle of 42 degrees and its
shadow was shown to 56 Ss who before had seen its stationary shadow
as two-dimensional, 53 Ss reported seeing a rigid obtuse corner. In this
presentation two of the three dark lines forming the shadow not only
were displaced but also underwent considerable changes in length.
FIG. 4. The figure used in experiment 7.
Entirely different results were obtained when the length of the
shadow lines was made indefinite. To achieve this, the shadow screen
was covered with a cardboard with a circular aperture where the shadow
of the corner figure fell on the screen. The size of the aperture was so
chosen that the ends of the shadow lines were always hidden from S.
Thus, only movement of the corner point, angular displacements of each
of the three lines, and changes of the angles which they formed with
each other were visible. As the corner point shifted sideways, one of the
lines seemed to move farther under the aperture edge and another one
The Kinetic Depth Effect : 141
seemed to pull out from under it, and the length of all three of them
seemed indefinite.
All 22 Ss employed in this experiment reported seeing a flat figure
which distorted. Had the kinetic depth effect occurred, the Ss would
have seen instead a rigid three-dimensional form with constant angles.
However, no such effect was observed where changes in the length of
the lines which constituted the figure were not given, because that length
was indefinite. We may conclude that a displacement of lines which is
linked only with a change of angles does not give rise to the kinetic depth
effect; displacement of lines and change in their length is needed. The
question remains unanswered whether length must be understood only
in absolute terms, or whether change in proportion has an effect of its
own.
Experiment 8. Variation of distances between objects. Not only are
the retinal images of solid objects deformed when one moves about; the
same is true to various degrees of the projection of the whole environ-
ment. That the objects which make up the environment are seen ar-
ranged in three-dimensional space and with unchanging distances be-
tween each other may also result from a kinetic depth effect. Just as
some of the contours of solid objects produce appropriately changing
retinal projections when the objects are seen from different angles, the
projections of many of the intervals between objects change their length
and their direction when one moves about. That this has the effect of
producing visual depth can be demonstrated with the shadows of an
arrangement of several objects on a rotating platform.
We used spheres supported by thin vertical rods because the shadow
of a sphere in rotation does not change its shape and will therefore not
produce a kinetic depth effect of its own. Four spheres of 1%6-in.
diameter were arranged at the corners of a square concentric with the
platform. The four rods were all of different height so that all the inter-
vals between the shadows of the spheres were periodically oblique when
the platform turned, and changed length and direction simultaneously.
The lower part of the screen was covered so that the shadow of the
turntable itself was hidden. The arrangement was turned at a rate of
one revolution in 5 sec. and was exposed to individual Ss for 20 sec.
Under these circumstances all 30 Ss who took part reported the
spheres to move in three dimensions. Twenty-four Ss saw the spheres
in a rigid spatial arrangement which turned about its center just like
the actual arrangement behind the screen. The others saw them move
in open single file in snakelike fashion into depth. In this latter motion
142 : Hans Wallach and D. N. O'Connell
only the shorter intervals between spheres are rigidly maintained, and
each sphere changes direction of rotation with each excursion; but there
can be no doubt that this is an incomplete form of a kinetic depth effect.
It should be mentioned that 15 of the 30 Ss had no other experience
with these shadow experiments except for another experiment with the
spheres which will be reported below. It is apparent that the kinetic
depth effect will readily yield a perception of a rigid spatial arrangement
of unconnected objects.
Experiment 9. Variation of distances between objects. The arrange-
ment of experiment 8 offers still another opportunity to check on our
finding that a line must change both in length and in direction in order
to produce a kinetic depth effect. From our experiments with solid
blocks and with the T figure we concluded that a shadow which merely
expands and contracts in one dimension will not give rise to the effect.
Here we set out to show that the same is true for intervals between ob-
jects; we modified the experiment with the spheres to correspond to
these experiments.
All the rods were given the same height so that the spheres were
aligned on a horizontal line and the intervals between the shadows
changed in length only. This arrangement was shown to the same 30 Ss
and under exactly the same conditions that prevailed in experiment 8,
but was presented prior to it. The fact that the T figure proved so sus-
ceptible to the influence of previous perceptions made this sequence
advisable. The experiment with the luminous rod which has been re-
ported at length was done with the same Ss in between the two experi-
ments with spheres.
As in the case of the T figure, only a minority of the Ss now saw
movement in three dimensions, namely 10 out of 30. The difference
between this result and that of experiment 8 is reliable at the .03 level
of confidence. Of the 15 naive Ss, 12 saw the shadows move back and
forth in the plane of the screen, 1 S reported movement in three di-
mensions, and 2 saw in the beginning of the observation period the
plane and later the three-dimensional version. For the other group of
15 Ss who had observed the shadows of some wire figures before, the
numbers were: 8 flat, 2 three-dimensional, and 5 first flat and later
three-dimensional. When these results are compared with those of the
previous experiment where imaginary lines connecting sphere shadows
changed both in length and in direction, it becomes again apparent
that these are essential conditions for the kinetic depth effect and that a
mere expanding and contracting of retinal distances is insufficient.
The Kinetic Depth Effect : 143
This is the reason for our view that Metzger's work (2) is not di-
rectly concerned with the kinetic depth effect. He exposed shadows of
arrangements of vertical rods whose changing patterns presented Ss
with rectangular intervals which changed in width only. No other de-
formations were visible, because no marks were distinguishable along
the shadow lines, and the latter ended only at the edges of the aperture
in which they were given. As with the aligned spheres, no reliable depth
effects are produced spontaneously in naive Ss with such an arrange-
ment. Whether Metzger's work contributes to an understanding of the
kinetic depth effect remains to be seen when more of the nature of the
effect is known.
Experiment 10. Effect of set. In our experiments, the kinetic depth
effect results in two perceptual characteristics: (a) a turning in the
depth dimension, and (b) three-dimensionality of form. However,
when the effect is observed under realistic conditions, namely when by
moving about one obtains a changing retinal projection for a stationary
solid object, no turning is perceived. The reason for this is that the ob-
ject remains in unaltered relation to its environment with respect to
which S perceives himself moving. Thus, only three-dimensionality and
rigidity of form, seen instead of the deforming two-dimensional pat-
tern which is given on the retina, are here the overt manifestations of
the kinetic depth effect.
The fact that under realistic conditions the object remains in un-
altered relation to its environment needs some consideration, because
this is not so in our experiments. There the deforming shadow of the
object denotes a turning while the environment, that is, the screen on
which the shadow is shown, remains stationary. Under realistic con-
ditions, on the other hand, both the object and its environment are
given retinally with deformations which denote a turning in relation
to S. The kinetic depth effect transforms the deforming retinal pro-
jection of the environment into a three-dimensional structure, and once
this has happened, the perception of the object as a three-dimensional
form is probably facilitated. That such a facilitation is likely to take
place is indicated by experimental results which show a strong in-
fluence of preceding exposures on the readiness with which the kinetic
depth effect occurs.
The following results may serve as an example. When the "helix"
(Fig. 3) was shown following the presentation of one or two figures
which readily show the kinetic depth effect, all of 18 Ss saw it as three-
dimensional during the first 10-sec. exposure. This is to be compared
144 : Hans Wallach and D. N. O'Connell
with results in experiment 2 according to which only 4 out of 16 naive
Ss gave a clear report of three-dimensional form for this figure after
the first 10-sec. exposure.
The question of how this influence is exerted must remain open. It
is conceivable that it consists merely in a set to see a turning in the
depth dimension. However that may be, the influence is a strong one.
If such an influence is effective between succeeding exposures of
different figures as shown, it should be expected to work also within a
given visual field. Under realistic conditions, once the environment of
a given object is perceived as a rigid spatial structure which changes its
orientation with respect to the moving S, such an influence should
facilitate a kinetic depth effect for the object. There are several reasons
why the environment should easily be seen in this fashion. To mention
only one: the environment will usually contain familiar features which
would cause the facilitating influence of previous experience with simi-
lar situations to operate. Thus we have good reason to believe that the
kinetic depth effect takes place more readily under realistic conditions
than it does in our shadow-screen experiments.
SUMMARY
When a three-dimensional form, solid or wire-edged, is turned behind
a translucent screen and its shadow on the screen is observed, the
shadow will appear as a rule as a three-dimensional rigid object which
turns, quite similar to the physical object behind the screen. This hap-
pens notwithstanding the fact that S actually looks at a plane figure
which is being deformed.
One condition seems to be essential for the occurrence of this effect:
the shadow must display contours or lines which change their length
and their direction simultaneously. If this condition is not fulfilled, a
plane distorting figure like the one on the screen is perceived unless an
influence of previous perception operates*
This effect is believed to operate widely under ordinary circum-
stances. When one moves about, objects near one's path are succes-
sively seen from different angles, and this change in orientation of the
object to S is the same as occurs when the object is turned by an equiva-
lent angle. Thus, the object's retinal projection undergoes the same
deformation as do shadows in our experiments, and the same per-
ceptual processes should result.
The Kinetic Depth Effect : 145
REFERENCES
1. Fisichelli, V. R. Effect of rotational axis and dimensional variations on
the reversals of apparent movement in Lissajous figures. Amer. J.
PsychoL, 1946, 59, 669-675.
2. Metzger, W. Tiefenerscheinungen in optischen Bewegungsfeldern. Psy-
chol Forsch., 1934, 20, 195-260.
3. Miles, W. R. Movement interpretation of the silhouette of a revolving fan.
Amer. J. PsychoL, 1931, 43, 392-405.
4. Philip, B. R., and Fisichelli, V. R. Effect of speed of rotation and com-
plexity of pattern on the reversals of apparent movement in Lissajous
figures. Amer. /. PsychoL, 1945, 58, 530-539.
Wolfgang Kohler and Pauline Austin Adams
DARTMOUTH COLLEGE
PERCEPTION
AND ATTENTION
ARTICULATION AND ATTENTION
In a recent paper on perceptual organization and learning, the senior
author was led to the following conclusions: (1) In many situations,
learning is at least as much a matter of perceptual articulation as of
large-scale organization. (2) Large-scale organizations exert a strong
pressure upon their material, which operates against articulation of
this material. The larger organizations are likely to prevail, and to
prevent learning, unless objective conditions for articulation are par-
ticularly favorable. To a degree, the pressure of a large-scale organiza-
tion can, however, be overcome by Os who intend to learn and, for
this purpose, introduce or support the required articulation. (3) It
follows from (2) that the threshold for articulation within larger or-
ganizations must be high. The threshold is probably lower, but still
fairly high, when Os inspect the material in an attitude which facili-
tates the articulation. 1
These conclusions refer to the perceptual phase of learning. It should
Reprinted from the American Journal of Psychology, Vol. 71, No. 3, Sep-
tember, 1958.
1 Wolfgang Kohler, Perceptual organization and learning. Amer. L PsychoL
1958,71,311-315.
Perception and Attention : 147
therefore be possible to verify them in perceptual situations in which
learning plays hardly any part. To our knowledge, there have so far
been only two investigations which are concerned with the present
issue, an older study by Krechevsky 2 and more recent experiments
done by Krech and Calvin. 3 In both cases, results have been striking,
and in line with the present conclusions.
Since Wertheimer's work on visual grouping, patterns such as that
shown in Fig. 1 have often been used for demonstrating perceptual
organization. 4 Actually, the fact that in such patterns certain rows of
dots form horizontal (or vertical) lines is a demonstration of articula-
tion; for the complete patterns are seen as large units which have the
shapes of squares, and, when the horizontal (or vertical) lines are
formed, they remain parts of the larger organization. This has already
been emphasized by Krech and Calvin, who used such patterns in
their experiments. These investigators did not measure the threshold
for articulation in Wertheimer's patterns; nevertheless, their results
clearly prove that this threshold must be very high indeed; for when
distances between the dots in one direction were very much greater
than in the other, not a few Os remained entirely unaware of the fact,
and did not see the corresponding articulation even when the patterns
were shown many times in succession.
FIG. 1. An example of articulation.
2 1. Krechevsky, An experimental investigation of the principle of proximity
in the visual perception of the rat J. exp. PsychoL, 1938, 22, 497-523.
3 David Krech and A. D. Calvin, Levels of perceptual organization and
cognition. J. abnorm. soc. PsychoL, 1953, 48, 394-400.
4 Max Wertheimer, Untersuchungen zur Lehre von der Gestalt. PsychoL
Forsch., 1923, 4, 301-350.
148 : Wolfgang Kohler and Pauline Austin Adams
One might be inclined to object that this fact was only caused by the
short time during which the patterns were seen; namely, 0.06 sec. This
objection would not be justified. The threshold for articulation in such
patterns is also very high when exposure times are much longer. We
have recently shown this in very simple experiments in which the pat-
terns were seen for one or several seconds. To obtain an estimate of
the threshold for articulation, we varied the ratio of the distances in
the vertical and horizontal dimensions. Some observations were made
when Os were concentrating on a "cover task" which was meant to
deflect attention from the patterns; in others, they were expressly asked
to attend to the patterns as such. In our first tests, we used the former
procedure.
The patterns consisted of black circles with the diameter Ys in.,
which were attached to squares of white cardboard (7 in. square). In
one direction (either vertical or horizontal) the distance between the
circles remained constant; namely, 1 %e in. In the other direction, it
was reduced in small steps from 1 % 6 to %e in. To protect the patterns
against the effects of repeated handling, they were covered with cello-
phane. The contrast between the circles and their background was not
appreciably affected by this procedure.
The patterns were used as backgrounds for cardboard figures of
much smaller size, and Os were instructed to indicate whether they
liked, or disliked, each figure by removing it in one direction or the
opposite. The figures did not seriously disturb perception of the pat-
terns. Moreover, the patterns alone remained before the Os between
the removal of one figure and the presentation of the next.
During the choices of a given O, of which there were six in suc-
cession, the background pattern was not changed. Only four of the
available patterns had to be used in the present tests. Each served as
background in tests with 5 Os; no O saw more than one pattern.
Immediately after O's six judgments, E removed the pattern, and
asked him to describe the background on which the figures had been
seen. The statements obtained under these conditions varied little.
Sometimes the circles were said to have been evenly spaced; more
often, to have been arranged in vertical and horizontal rows. Clearly
distinguished from such descriptions were those which referred only
to horizontal (or only to vertical) rows. Since actual articulation in
one dimension goes with suppression of corresponding relations in the
other dimension, only statements of the latter kind can be regarded as
evidence that articulation has occurred.
Perception and Attention : 149
In one-half of our tests, distances in the vertical direction were
varied from one group of Os to another; in the other half, the varia-
tions occurred in the horizontal direction. Students of Bryn Mawr and
Swarthmore Colleges served as Os in the former case; and of Stanford
University in the latter.
In our tests with the cover task, results from 20 Os were as follows:
With the distance in the horizontal (or vertical) direction constant
( I2 /i6 in.), and with the distance in the vertical (or horizontal) di-
rection reduced to
7 /ie %6 %6 %e in-,
articulation in the vertical direction occurred
012 4 times
and in the horizontal direction
001 4 times
Since every O was tested with only one ratio of the distances, these
results are independent of any particular "set." The threshold for ar-
ticulation which the present numbers indicate can, of course, be re-
garded as a threshold only when differences among the individuals in
the various groups are ignored. On the other hand, the agreement be-
tween the two sequences of numbers suggests that, under the conditions
of our tests and with comparable Os, most individual thresholds would
lie between %e and %e in. The probable threshold in this sense is ex-
tremely high. Suppose that, on the average, the threshold lies at % 2 in.
The difference between this distance and the constant distance in the
other direction amounts to about 60 per cent of this greater distance.
In the next group of tests, Os were shown patterns of the same type,
with the instruction to describe each immediately after its presentation.
There was now no cover task. Under these conditions, perception was
obviously "attentional." We have not yet tried to introduce more spe-
cific attitudes than mere attention to the patterns. Obviously, during
intentional learning, the operations of Os often go far beyond this
simplest condition.
The Os (10 in number) were students of Bryn Mawr and Swarth-
more. All saw the complete series of patterns in its regular sequence,
beginning with the situation in which distances in the two directions
were the same ( 1 %$ in.) . Patterns in which distances in the vertical or
in the horizontal direction were shorter were shown in alternation. All
150 : Wolfgang Kohler and Pauline Austin Adams
patterns were exposed for 1 sec. Since articulation in the two direc-
tions seemed to occur at the same ratio of the distances, results for both
situations are here combined. Out of 10 cases, they are as follows:
Distance (in inches): i% 6 10 /ie 9 /ie 8 /i6 %6 6 /ie %6
Number of articulations: 1 2 8 9 10
These numbers are cumulative: once O had seen a pattern as articulated,
he also reported articulation in the following patterns, in which condi-
tions for it were further improved. For the group as a whole, the thresh-
old may now be said to lie between % 6 and % 6 in., that is, at a difference
between the two distances which is distinctly smaller than it was when
Os were occupied with a cover task. On the assumption that it lies at
*%2 in., it amounts to approximately 37 per cent of the larger distance.
Hence, the attentional attitude has considerably lowered the threshold,
although the threshold still remains very high. Articulation does not
often occur until distances in one direction are reduced by more than
one-third.
Since it seemed possible that a 1-sec. exposure was too short for
articulation to occur easily, the experiments were repeated at Stanford
University under the same conditions, excepting that now the patterns
were shown for 3 sec. There were again 10 Os. The results were as
follows:
Distance (in inches): i 2 / 16 IO /IG %6 8 /ie %e 6 /ie 5 /ie
Number of articulations: 2 8 9 10
There is hardly any difference between these results and those obtained
with the shorter exposure times. We may therefore trust the results of
both tests. They confirm the results of Krech and Calvin. In such pat-
terns the threshold for articulation is far greater than more familiar
spatial thresholds.
Krech and Calvin have demonstrated that the ease with which
articulation occurs in different persons is not related to their visual
acuity. The size of the threshold for our groups may be regarded as
proof of the same fact. Not one out of our 20 Os showed evidence of
articulation when the distances in the two directions were x %e and
1 %6 > that is, when the difference amounted to %, or 20 per cent of the
smaller distance. Most people can compare visual distances as such
far more accurately. In this respect, we find a very simple observation
particularly impressive. When confronted with one of our patterns
Perception and Attention : 151
which, in spite of a considerable difference between the two distances,
does not appear to us articulated, we have no difficulty in singling out
three circles which together form a right triangle, and in ignoring the
surrounding circles. 5 The great difference between the vertical and
horizontal sides of the triangle is perfectly obvious under these condi-
tions; and yet, when the triangle is allowed to disappear again in the
total pattern, the same objective difference is once more quite unable
to bring about articulation.
The conclusion that the suppression of articulation is caused by
organizational rather than by more elementary visual factors is also
supported by the following fact. In the sequence in which the various
patterns were shown, a next reduction of the distance-ratio occurred
just as often first in the horizontal as in the vertical dimension. Results
did not indicate that articulation occurred more easily in the horizontal
dimension, although the vertical-horizontal illusion might have been
expected to operate in this direction. Apparently, it had no appreciable
effect on the outcome of our tests. The illusion is, however, present
when a rigHt triangle is singled out. When this is done with a pattern
in which objective distances are the same in both dimensions, the
vertical distance is clearly greater.
While, under the conditions of our observations, articulation is sup-
pressed within surprisingly wide limits, it is, of course, no longer sup-
pressed under more extreme conditions. 6 Even when the patterns are
seen only incidentally, as they were in our first tests, articulated organ-
ization becomes perfectly natural with very great differences between
the distances in the two dimensions. After all, in the first tests, 8 out
of 10 Os did report articulation when distances in one direction were
three times as great as those in the other. In the same connection, it
must be remembered that, if in such patterns vertical (or horizontal)
lines are formed only against strong resistance, the same patterns have
smaller parts which are seen as such all the time; namely, the dots or
circles. Conditions for their survival as separate entities are, of course,
so favorable that they are not appreciably affected by their inclusion
in the total patterns or, under conditions of articulation, in vertical
or horizontal lines. 7
5 This is a much easier task than that of articulating the whole pattern to
form vertical (or horizontal) lines.
6 Krech and Calvin, op. ciL, 399.
7 Even the articulation with which our tests are concerned seems to occur
752 ; Wolfgang Kohler and Pauline Austin Adams
The sequence in which the patterns were shown under conditions
of attentional perception may have caused a certain set in favor of non-
articulated organization. The threshold found in these tests may, there-
fore, be higher than it would have been under other conditions. If this
is true, our results do not show the articulating effect of attention in its
full size, but we doubt whether, in the present situation, the threshold
is greatly increased in this fashion. When we inspect a pattern which
only a few of our Os see as articulated, it generally does not appear as
articulated to ourselves, and then a strong effort to introduce the "right"
articulation seldom produces more than temporary articulation in re-
stricted regions of the pattern. 8
We agree with Krech and Calvin on all facts, but not with some of
their remarks about Gestalt psychology. In the first place, their own
observations and ours can hardly be said to mean that, under certain
conditions, the principles of Gestalt psychology do not hold. Surely,
Gestalt psychologists have never maintained that differential proximity
always wins, even when organizational factors which operate in an-
other direction are extremely powerful. Is it an objection to the law of
gravitation that airplanes may rise from the ground, and stay in the
air for hours? Secondly, on the basis of their findings, the authors ob-
ject to statements that the effects of proximity are "immediately" per-
ceived. When, many years ago, Gestalt psychologists made statements
of this kind, they used the term "immediate" in protest against the
view that organization is merely a matter of learning which has grad-
ually transformed so-called sensations into objects and groups. The
term was not meant to deny that it takes organization a certain (very
short) time to complete its work. On the contrary, certain phenomena
such as the y-movement were always regarded as evidence for precisely
more easily under certain conditions. Occasionally, we have observed that a
pattern, which is not seen as articulated when shown in large size, begins to
show articulation when, under otherwise identical conditions, its size is greatly
reduced.
8 Articulation of larger patterns can be produced by distinctive similarity of
some parts as well as by differential proximity. Consequently, the questions
which have arisen in the latter case must also be asked in the former: (1) What
is the threshold for articulation when differential similarity is the decisive factor?
(2) Does attention affect this threshold just as it affects articulation by proximity?
So far we have done only preliminary tests when attention was deflected by a
cover task. On a middle-gray background, alternating rows of dark-gray and
white circles were shown when distances in both directions were the same. Only
2 out of 15 Os gave evidence of having seen the pattern as articulated when
describing it after completion of the cover task. In future tests, the brightness
difference between the circles will have to be varied.
Perception and Attention : 153
this fact. In the same connection, the authors state more specifically
that even simple perceptual structures are products of a rapid develop-
ment which proceeds from the more homogeneous to the more dif-
ferentiated. This again is a view with which Gestalt psychologists are
quite familiar. In fact, in an article in which he discussed perceptual
grouping, Wertheimer expressed exactly the same view a long time ago.
The tendency of larger organizations to prevent the emergence of
more particular structures is not restricted to patterns of dots. In
further tests, we used patterns of another kind which have also been
introduced by Wertheimer. Letters and words tend to disappear as
particular entities with their familiar characteristics when their mirror
images are added, as in Fig. 2. Few people who see this pattern are
spontaneously aware of the fact that its upper half is the word men. In
camouflage of this type, it is mainly the tendency of organization to
form closed entities which so changes some characteristics of parts that
they are not spontaneously recognized. For instance, when seen in the
pattern as a whole, the lines of the letter m and those of its mirror image
become mere boundaries of area between them which now assume the
figural character described by Rubin. When shown alone, the lines of
the letter have no such function; they are then merely linear phenomena.
Within the larger pattern, however, the word and its letters tend to lose
their familiar characteristics, and are therefore seldom spontaneously
recognized. The principle which is here involved differs from the one
which operates in Wertheimer's dot-patterns; and yet, as to the rela-
tion between larger organizations and particular parts of them, the two
situations resemble each other so much that we expected analogous
tests to give similar results also with the present pattern.
FIG. 2. A word and its mirror image.
As used in our observations, the pattern had a horizontal length of
1 8 cm., and a height of 3 cm. We varied the distance between the upper
and the lower halves of the pattern from 0-3.5 cm. in steps of 0.5 cm.
9 Dr. Mary Henle once did experiments of this kind which have not been
published. She found that the larger organization tends to prevail even when the
word and its mirror image were clearly separated in space.
154 : Wolfgang Kohler and Pauline Austin Adams
Preliminary tests made us suspect that, when Os are first shown the
pattern without separation, and then the various versions with grad-
ually growing separations, the effect of set tends to prevent awareness
and recognition of the word, even when separations are very great.
Tests for the various distances were, therefore, made with separate Os,
5 for each separation. Our Os were again students of Bryn Mawr
and Swarthmore Colleges.
In a first procedure, perception of the patterns was incidental. Hori-
zontal lines of approximately equal length were shown, one 1.8 cm.
above and the other the same distance below the pattern and, as a cover
task, Os had to compare their lengths. The time of exposure was always
1.5 sec. After six comparisons, the pattern was removed, and O was
asked what he had seen between the lines. In the reports, certain terms
such as "leaves," "hearts," and "wavy lines" occurred very often. The
word "men" was mentioned only when the patterns with very great
separations were shown, and even then only by a few Os.
The results from 40 Os, 5 at every separation, were as follows:
Separation (in cm.) : 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Times "men" observed: 00001021
Only 4 of 40 Os discovered that a well-known word was shown to
them, although in several groups the separation between this word and
its mirror image was large even at separations of 3.0 and 3.5 cm.
only 3 of 10 Os perceived the word.
We next performed tests under conditions of attentional perception.
Objective conditions remained the same, including the time of presen-
tation. The horizontal lines were also left in their places. Again, 5 Os
were tested with a given separation. Before the pattern was shown, the
Os were instructed to describe its appearance as soon as it had been
removed by E. The horizontal lines were not mentioned in the instruc-
tions.
The results with 40 Os, 5 at every separation, were:
Separation (in cm.) : 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Times "men" reported: 01234455
It is remarkable that, even when Os attend to the patterns before them,
separations such as 0.5 and 1 cm. do not regularly permit recognition
of the word. No more than 3 of 10 Os are aware of its presence under
these conditions, and only when the distance grows to 3 cm. is the word
seen by all. Since no set operated against its emergence, we can only
conclude that, with some Os, the word is still strongly affected by the
Perception and Attention : 155
larger organization when the separating distance amounts to about 2
cm. On the other hand, the difference between the present results and
those found under conditions of incidental perception is also very
great. When confronted with the same patterns, 24 (rather than 4) Os
altogether have now recognized the word.
There is no doubt that our Os were aware of the distance between
the two halves of the pattern. We often asked them to make drawings
in connection with their descriptions. For the most part, these drawings
consisted of two separate figures, even when the Os had not recognized
the word; but, just under these circumstances, the figures showed little
resemblance to the word or its mirror image. The drawings were gen-
erally too regular.
SATIATION AND ATTENTION
The conclusions drawn from certain learning experiments have been
confirmed in the preceding tests. We now turn to the question why at-
tention operates against the tendency of larger organizations to sup-
press more articulated perception. In trying to answer this question, we
must briefly refer to certain principles of natural science. More partic-
ularly, we must consider the fact that, when open systems receive
energy from the outside, processes in their interiors are intensified, and
tend to become more differentiated. It is our suggestion that the effect
of attention on visual patterns is a special instance of this fact.
Suppose that, in a closed system, inert velocities are immediately
destroyed by friction. Such a system pays for any transformations in
its interior by a reduction of the energy available for such work. When
this energy has been reduced to a minimum, the system is either in a
state of equilibrium or in a steady state. Several physicists have called
attention to the fact that, in such states, the distribution of materials
and processes within the system is maximally homogeneous or sym-
metrical. The reason for this fact is, of course, that the energies which
the system spends in transforming itself are closely related to differ-
ences, inhomogeneities, and asymmetries of materials and processes.
The very simplest example is that of a closed system in which only one
particular transformation is possible to begin with, which is brought
about by one difference of potential (in the more general sense of this
term). In this special case, the difference of potential is gradually
lowered to a minimum, and the corresponding process of transforma-
tion weakened.
156 : Wolfgang Kohler and Pauline Austin Adams
The present rule applies only to closed systems as wholes. Very
often, developments in parts of such systems occur in the opposite, the
"upward," direction. This is entirely in line with the general rule so
long as the local increase of energy is accompanied by greater decreases
elsewhere in the system, as it actually always is. Under such circum-
stances, the part in which developments occur in the direction of grow-
ing energy is a special instance of an "open" system. In such a part of
the whole (or closed) system, energy and the intensity of processes can
grow because other parts of the system play the role of sources from
which that particular section can borrow energy. When the energy
within an open system of this kind grows, existing differences and in-
homogeneities in its interior will be enhanced, and may develop to
maximal size. In extremely simple cases, an open system in this sense
will again be capable of only one transformation. This transformation
(and the supporting difference of potential) will now be intensified
until a highest value is reached. 10
The most obvious examples of open systems are the organisms. They
maintain (or even increase) the energy in their interior by absorbing
energy from their environment. If the parts of the environment which
thus serve as sources are included, we have, of course, again closed
systems, of which the organisms are open parts. In organisms, too, the
absorption of energy from the outside has the effect of giving processes
the upward direction, or of keeping them at high levels of energy.
What has just been said about organisms applies also to some of their
parts, for instance, to the visual projection areas of the human brain.
Quite apart from the energy inherent in afferent impulses, the visual
cortex may, or may not, receive energy from other parts of the brain.
The former condition seems to be realized when a person takes interest
in his visual field. For, under these circumstances, the segregation of
objects in the field is sharpened, differences between various parts of the
field are enhanced, and so forth. These changes have again the well-
known direction of developments within open systems to which energy
is transferred from the outside.
10 Some recent statements about open systems give the impression that the
meaning of this term is not always clearly understood. Authors who believe that
open systems are mysterious entities to which the laws of physics do not apply
are making a serious mistake. To be sure, the behavior of some open systems is
fascinating; but there is no evidence that this behavior follows principles which
are at odds with those of the physicists. It is, however, true that physics has so
far given attention only to a few examples of open systems, and that, as a con-
sequence, the relation between the behavior of some such systems and the princi-
ples of physics has never been explicitly formulated.
Perception and Attention : 157
It need not be the visual field as a whole which is so affected. What
we call "visual attention" may be directed toward particular parts of
the visual scene. It thus relates other (as yet unknown) parts of the
nervous system to such particular visual objects; and, in doing so, it
also seems to increase the energy level of the processes in question; for
local differences are again intensified, and the clearness of local struc-
tures is enhanced.
Our tests with Wertheimef s dot patterns may serve as examples of
such effects of attention inasmuch as, under conditions of incidental
perception, these patterns tended to appear as homogeneous distribu-
tions while, under the influence of attention, the differences between
the distances in the two directions became much more effective, and
articulation appeared where none had been seen before. Attention had
a similar effect upon the combination of a word with its mirror image.
As long as perception of this pattern was incidental, the word as a
visual entity remained submerged in the total structure in spite of
considerable distances between the two parts. Attention changed this
situation in the direction of a more specific organization, in which the
word could show its known characteristics, and thus be recognized.
Hence, attention operates in the direction of differentiation, and is in
this respect comparable to events in physics which transfer energy to
an open system from the outside.
If we are right in assuming that, under conditions of attentional per-
ception, the energy of the visual processes in question is increased, it
should be possible to demonstrate this fact in psychological experi-
ments. When complicated situations are involved, such a demonstra-
tion may be difficult. But in some open physical systems only one
process will be intensified under such circumstances. It therefore
seemed advisable to perform the necessary psychological tests under
similarly simple conditions.
According to a theory developed by Kohler and Wallach, the cortical
counterpart of a segregated visual object is pervaded and surrounded
by an electric current. 11 This current will be weaker or stronger de-
pending upon the difference between the brightness inside and outside;
but soon the flow will always be weakened by its own electrotonic (or
satiating) effect upon the tissue, whatever its initial intensity may have
been.
We have just assumed that, when attention is concentrated on an
11 Wolfgang Kohler and Hans Wallach, Figural after-effects. Proc. Amer,
philos. Soc., 1944, 88, 269-357.
158 : Wolfgang Kohler and Pauline Austin Adams
object, the underlying cortical process (which, according to Kohler
and Wallach, is the current of the object) will be intensified, and that,
as a consequence, the object will stand out more clearly; but this in-
tensification of the current will also accelerate its electrotonic action.
As a result, satiation in the crucial area will soon be stronger, and cor-
responding figural aftereffects will be greater than those caused by a
comparison object to which O does not attend. Kohler and Emery have
made observations which demonstrate this fact. 12 An object gradually
shrinks (by "self-satiation") when it occupies a given place for some
time. Now, when two equal objects are shown on the left and the right
side of a fixation point, attention may be concentrated on one or the
other. When this is done, the object to which O attends soon appears
smaller. If now attention is shifted to the other object, this object will,
after a while, appear smaller, and so forth. Since this phenomenon con-
firms the present reasoning, we decided to do the following tests. 13
Our first tests referred to the influence of visual attention upon a
well-known figural aftereffect in the third dimension of visual space.
When an object on one side of a fixation point has been seen for some
time, a test (T-) object shown in front of the former object is displaced
toward O; it appears nearer than a comparison object which is pre-
sented on the other side of the fixation mark in a symmetrical location
and in the same objective plane. If attention really intensifies satiation,
its influence should be demonstrable in a situation in which two equal
and symmetrically placed satiation (I-) objects are shown first, and
then two equal T-objects which have strictly corresponding locations
in front of the two I-objects. In such an objectively balanced arrange-
ment, the concentration of attention on one or the other I-object during
12 Wolfgang Kohler and D. A. Emery, Figural after-effects in the third dimen-
sion of visual space. Arner. 7. PsychoL, 1947, 60, 159-201.
13 It is by no means easy to make observations in which attention must be
shifted from one object to another while O fixates a mark between them. Visual
attention has sometimes been compared with the use of a flashlight which can
be turned one way or the other as its owner wishes. This comparison is seriously
misleading. Once attention is concentrated on one more or less peripheral object,
this vector cannot be transferred at will to an object in another peripheral loca-
tion. For a considerable period, it refuses to move from its first target to the
new one, even when O makes every effort to bring about the change. Visual at-
tention seems to be a far more earthy fact than we have realized in the past
From a practical point of view, it follows that, when the effects of attention on
visual objects are to be studied, Os cannot be expected to shift the vector from
one object to another whenever they are asked to do so. In our experiments, no
such instruction was ever given.
Perception and Attention : 159
the satiation period should have the effect of displacing the correspond-
ing T-object farther toward O.
The experimental arrangement which we used has been described by
Kohler and Emery. 14 The I-objects were light gray squares (1.5 in. on
a side) . They were shown at a distance of 10.5 ft. from O, who fixated
a mark 3 in. in front of their plane for 45 sec. Two T-squares ( 2 in. sq. )
then appeared 3 in. in front of the fixation mark, and O now compared
their locations in the third dimension. Some Os were instructed to con-
centrate on the left I-object and others on the right I-object during the
satiation period.
Since the distance between the squares and the fixation mark was
small (about 1.2 in.), concentration of attention only on one square
proved practically impossible. It seems that, in such a situation, the
width of the region to which O attends cannot be reduced ad libitum.
We therefore instructed our Os to concentrate on the fixation mark and
one of the squares as a pair of objects. This is a much easier task. The
Os also were asked not to judge immediately, because many figural
aftereffects tend to grow to some extent while the T-objects are per-
ceived. In the present experiments, this phenomenon was particularly
striking. Our Os were mostly students of Swarthmore College who did
not know the purpose of the tests. Of 19 altogether, 7 were instructed
to concentrate on the right object, and the remaining 12 on the left I-
object. All were tested twice. Before the tests, depth discrimination
when attention was concentrated on the fixation mark alone had been
examined in all cases. Accuracy had always been found so high that
actual measurement of thresholds would have been difficult with our
apparatus.
The results of our tests are shown in Tablel, in which plus refers to
judgments in the expected direction; zero, to the absence of any asym-
metry in the location of the T-objects; and minus, to judgments in the
opposite direction. "Right" and "left" mean the two directions of at-
tention, and the numbers indicate how many Os gave judgments belong-
ing to the three categories. The last column gives the p-values found in
a sign test, ignoring the zero-cases. Altogether, 31 judgments out of 38
agree with expectation, 4 disagree, and in 3 instances no difference has
been observed. The number of Os tested with attention concentrated
on the left side is greater, because it proved advisable to increase their
number when the observed effect seemed weaker on this side. The data
14 Kohler and Emery, op. cit. f 180.
160 : Wolfgang Kohler and Pauline Austin Adams
TABLE 1
THE EFFECT OF ATTENTION ON SATIATION: VISUAL TESTS
Direction Test + - p
Right
first
6
1
0.016
second
7
0.008
Left
first
9
1
2
0.032
second
9
1
2
0.032
of Table 1 suggest such an asymmetry. This phenomenon may be re-
lated to observations mentioned by Hebb. 15 A more specific relation
may exist to asymmetries in visual perception which Gaffron has de-
scribed. 16 Our tests have not been repeated in a situation in which T-
objects are displaced backward. The present form of the tests must be
regarded as more conclusive. 17
We felt that the present results were in need of confirmation. It also
seemed advisable to study the same problem in a different situation,
for instance, when the aftereffect which attention might, or might not,
intensify occurs in another sense modality. Further tests were, there-
fore, done with the figural aftereffect in kinesthesis, in which satiation
changes the width of a T-object. Is this effect also enhanced on one side
15 D. O. Hebb, Organization of behavior. New York: Wiley, 1949, p. 49.
16 Mercedes Gaffron, Right and left in pictures. Art Quart, 1950, 311-331.
17 Kohler and Emery, op. cit., 179. Dorothy Dinnerstein and the senior author
once did experiments in which the effect of self-satiation upon objects shown
before or behind the plane of the fixation mark was tested. A square was presented
either behind or in front of the fixation mark, and either on the right or the left
side of this mark. After a satiation period, the location of the satiated square was
compared with that of another square which now appeared on the other side of
the fixation mark, and in the same objective plane. In all observations of this
kind, the affected square was found to appear nearer the plane of the fixation
mark, irrespective of its objective location behind or in front of this plane. We
have recently repeated these tests with 6 Os. Without exception, their judgments
confirmed this rule. The rule suggests that, under conditions of self-satiation, the
depth of three-dimensional visual structures is gradually reduced. One can easily
demonstrate that this is true by presenting in a stereoscope two slightly disparate
projections of a three-dimensional object to the two eyes, which are given the
proper directions by means of fixation marks on one side of the two projections.
After a satiation period, two further disparate projections, identical with the
former, are added on the other side of the marks so that the three-dimensional
appearance of the old and the new visual objects can be directly compared. It is
then found that in the satiated object the third dimension is strikingly reduced;
it looks much flatter than the new object.
Perception and Attention : 161
(in one hand) when, in an otherwise symmetrical situation, O's atten-
tion is concentrated on this side? It is not known whether attending to
kinesthetic facts is basically the same process as concentrating on a
visual object; but, on the assumption that it is, we performed the fol-
lowing tests.
The experimental arrangement which we used has been described
by Kohler and Dinnerstein. 18 In a first experiment, two equal I-objects,
2.5 in. wide, were presented to the two hands for a period of 45 sec.
The T-object which O then felt with one hand was 1.5 in. wide. With
the other hand, he had to find a place on a scale, which appeared to
have the same width as the T-object. In a second experiment, the width
of the I-object was 1 in., that of the T-object again 1.5 in. Thus, the
aftereffects established in the two experiments, and therefore also cor-
responding differential effects produced by one-sided concentration of
attention, would lie in opposite directions. Moreover, in both experi-
ments, concentration on the side where later the T-object is presented
would have one, and concentration on the other side, the opposite effect
on the measurements.
The scale was always placed on O's right side. O was blindfolded.
Since subjective equality need not coincide with objective equality, 1 *
the first two measures were made without preceding satiation. In one
measurement, O's right hand was placed on the scale three steps below
the point of objective equality (POE) ; in the other, three steps above
this point. The actual sequence was one for one-half of the Os and the
opposite for the other half. The mean point of subjective equality
(PSE) for all 42 Os was +0.93 of one step on the scale. This value
differs significantly from the POE (p < 0.01 ) . The width touched with
the right hand was underestimated by 28 Os, overestimated by 9, and
judged equal by 5. Our results in this respect agree with those of other
investigators. 20 Only a few of our Os were left-handed. The average
PSE for the Os who concentrated, in the following tests, on the left or
the right side did not differ significantly from each other. The differen-
18 Wolfgang Kohler and Dorothy Dinnerstein, Figural after-effects in kines-
thesis. Miscellanea Psychologica Albert Michotte, Louvain: Institut Superieur de
Philosophic, 1947, 199.
19 Michael Wertheimer, Constant errors in the measurement of figural after-
effects. Amer. J. PsychoL, 1954, 67, 543-546.
20 G. S. Klein and David Krech, Cortical conductivity in the brain injured.
J. Personal, 1952, 21, 118-148; A. McPherson and Samuel Renfrew, Asymmetry
of perception of size between the right and left hands in normal subjects. Quart.
J. exp. PsychoL, 1953, 5, 66-74; Michael Wertheimer, op. cit., 543-546.
762 : Wolfgang Kohler and Pauline Austin Adams
tial aftereffects measured after satiation were, for each O, related to
his PSE as a zero.
Before satiation began, the Os were instructed to concentrate on the
impressions in one hand. As a pretext they were told that their results
would be compared with those of other Os who, during the same tests,
would learn some material by heart.
Table 2 gives the numbers of Os whose judgments had the expected
direction (+), of those who showed no differential effect (0), and of
those whose judgments were at odds with our expectation ( ). Out
TABLE 2
THE EFFECT OF ATTENTION ON SATIATION: KINESTHETIC TESTS
Experiment
Direction
+
p
1
right
10
1
0.001
left
10
1
0.001
2
right
8
2
0.004
left
7
3
0.008
of 42 cases altogether, 35 judgments agree with our hypothesis, no
judgment had the opposite direction, and in the remaining 7 instances
no direction was indicated. The p-values are derived from a sign test in
which zero-cases were not included. There seems to be no doubt that
our assumption is verified. The decision is less satisfactory in experi-
ment 2 than in experiment 1. Such a difference would have to be ex-
pected if, in experiment 2 the relation between the widths of the I- and
the T-object was less favorably chosen.
Since, actually, the judgments of our Os had the form of quantitative
deviations from their PSEs, we can add a further table (Table 3 ) which
TABLE 3
THE EFFECT OF ATTENTION ON SATIATION: FURTHER
KINESTHETIC DATA
Experiment
Direction
Mean
SB,,
P
1
2
right
left
right
left
+1.27
1.14
1.10
+1.15
.22
.21
.39
.39
<0.01
<0.01
<0.01
<0.01
Perception and Attention : 163
refers to such quantities, the means of the results of the individual Os in
each group, rather than to mere directions of judgment. Under SE m ,
the standard errors of the means are given. The unit in which results
are here measured is one step on the scale (% 2 in.)-
Table 3 confirms the results shown in Table 2. When, in such experi-
ments, attention is concentrated on the kinesthetic impression in one
hand, it intensifies the corresponding satiation.
SUMMARY
(1) In certain patterns introduced by Wertheimer, approximate
thresholds for articulation have been determined under conditions of
incidental and attentional perception. The thresholds are always very
high, but clearly lower when perception is attentional. The same result
was found when a word was shown at varying distances from its mirror
image. Under conditions of attentional perception, the word as such
is seen and recognized at much shorter distances from its mirror image.
(2) Attention intensifies the process which underlies the perception
of an object. Under such circumstances, satiation is accelerated, and
corresponding figural aftereffects are enhanced. Figural aftereffects
both in the third dimension of visual space and in kinesthesis show this
influence of attention.
Hans W attach
SWARTHMORE COLLEGE
SOME CONSIDERATIONS
CONCERNING THE RELATION
BETWEEN PERCEPTION
AND COGNITION
I look around in a familiar environment and find that nearly every
object has meaning. 1 This meaning I experience as an objective fact,
and I perceive it out there in the thing. A hammer looks like something
with which to drive a nail into a wall or something with which to smash
a vase. I have the impression that I perceive these meanings in the ob-
ject even while I realize that they do not come to me through my eyes
at the moment of perceiving them but must be furnished by a memory
function, for they were given by previous experience with the object.
I suspect that this discrepancy between experience and function is
responsible for some of the vagueness of conception in this part of
psychology. The remedy seems to be to acknowledge it explicitly: The
meaning which the hammer has for me is, functionally speaking, the
effect of past commerce with that object, but it is experienced as being
Reprinted with permission from the Journal of Personality, Vol. 18, No. 1,
September, 1949.
1 This paper was written while the author was a John Simon Guggenheim
Memorial Fellow.
The Relation Between Perception and Cognition : 165
seen in the object of equal status with color and form. We have here the
effect of a recall process in which recall is not experienced as such. In-
stead the contribution of memory appears as part of a percept.
There is no reason why this recall process should be essentially differ-
ent from other recall processes as, for instance, recognition and recall
by association. As Kohler, reviving an old argument of Hoffding's, has
pointed out (5; 3, pp. 126-144), any recall which is occasioned by a
perceptual experience involves a process in which such an experience
brings into function a memory trace of a similar experience of the
past. This is true also where the subject recalls an associated content.
When I pass a man on the street whom I have met before, I may, for
instance, recall his name. This recall presupposes an association be-
tween a visual trace, say of the man's face, and a trace of his name. But
the association alone does not explain recall of the name on this oc-
casion. Face and name left associated traces when I was introduced to
the man in the past. When I meet the man again, he is, of course, on
the whole the same physical object, and I may immediately experience
him as the same person I met before; but functionally speaking there
are two separate psychological events involved, the perception of the
face on the occasion of the first meeting and a present perceptual proc-
ess of the "same" face. The former is now represented by a memory
trace, and it is this trace which provides the access to the trace of the
name. Yet it is one of many traces of faces which I might recognize, and
so the question arises how the appropriate one is brought into function.
As Kohler has argued, the only possible answer is that the similarity
between the original process of seeing the face, now represented by its
trace, and the present process is responsible for the proper selection of
this trace. Recall by association, then, consists of two steps: A process
of recall by similarity by which the present perceptual process makes
contact with the trace of a similar process of the past, and secondly,
recall of a content associated with this trace.
The necessity for such an assumption is not avoided by thinking in
terms of responses instead of associated contents. One speaks of the
same stimulus when the same combination of physical processes affects
the sense organs of the subject or, more loosely, when these combina-
tions appear to the experimenter as the same. Yet, when the same
stimulus occurs repeatedly to produce a particular response, we deal,
of course, with different, initially unconnected psychological events,
and they must in some fashion manifest their identity in the nervous
system of the subject before that particular response can be aroused.
166 : Hans Wallach
In a relatively small number of cases the assumption might possibly be
made that afferent processes produced by different stimuli will arrive
at different places in the nervous system and manifest the identity of
the respective stimuli by their locale. But where the stimulus is charac-
terized only by its spatially or temporally extended pattern as, for in-
stance, in the case of visual forms, of speech sounds, and of noises, this
assumption cannot be made. Similarity between the processes produced
by different occurrences of the "same" stimulus must account for its
identification.
Recall by similarity may occur without recall of an associated con-
tent, namely, in pure recognition where it merely produces a feeling of
familiarity.
The same process of recall by similarity must be involved when
meaning is perceived, unless a set is operating. No matter what the na-
ture of meaning may be, as long as a meaning content was acquired in
the past it is necessary to postulate the same sequence of recall processes
in order to explain how meaning comes to appear in perceptual ex-
perience: first, a step of recall by similarity between the sensorily deter-
mined perceptual process and a trace complex with which the meaning
content is connected, and secondly, the coming into function of the
connected meaning. Since in a familiar environment most perceptual
objects appear meaningful, such recall processes must readily take
place all tie time, and because they do, their mediation usually goes un-
recognized.
I have just used the term trace complex, because often what appears
as a familiar or meaningful object has occurred in the past not only
once but at several different times. Frequently a number of these oc-
currences have contributed in different ways to the meaning content.
When recall takes place, the total meaning seems to be given. This indi-
cates that the recall process established contact, not with a trace of any
one of these occurrences, but with a trace complex which represents
features of a number of them. In short, the various occurrences must
have previously become connected so that they could contribute to
the total meaning content and later, in a recall process, function to-
gether. These connections are themselves the effects of recall processes
which took place in the past, when at the second occurrence of the
object contact was made with the trace of the first one, and later, when
the third occurrence brought into function the trace complex of the
first two, and so on. In this fashion, recall by similarity accounts for a
The Relation Between Perception and Cognition : 167
cognitive product of temporally separated events and, in many cases,
makes learning by repetition possible.
Often the function of a memory trace which participates in a per-
ceptual process is not merely to add its content to the sensorily deter-
mined qualities of color and shape. It may change the organization of
the primary process by imposing a different internal grouping as in the
case of the Street figures. 2 Or new perceptual qualities may emerge
after a trace reference has been established, as, for instance, physiog-
nomic qualities. They are seen only when the primary form pattern
has been recognized as a face. But when that has happened, the seen
physiognomic qualities depend mainly on the sensory conditions.
Yet no matter how intimate and varied in function the interplay of a
primary process and an aroused trace may be, it must be preceded by a
process of trace selection. Where no set is operating, this selection
process consists of recall by similarity and is therefore initiated by the
primary perceptual process and highly dependent on its characteristics.
The selection process must be distinguished from the interplay between
trace and primary process, for it is prerequisite for this interplay.
The intervening of this selection process before a pertinent memory
trace can influence a perceptual process introduces an element of for-
tuity into experiments on the effect of central factors on perception, or,
rather, what amounts to fortuity from the viewpoint of such studies. We
are all aware of the difficulties which the incomplete knowledge of the
subject's past means for experimentation and its evaluation. But even
if we had a complete inventory of the subject's previous experience,
we would not be much better off because it would still be to a degree
fortuitous whether a psychologically pertinent trace content actually
comes to participate in the present process. Kohler and Restorff have
demonstrated that whether or not an experience of the past through
recall by similarity gains an influence on a present process depends on
the content and structure of the time interval between that experience
and the present (1; 5; 3, pp. 126-144). Frequency and recency of
the mental event which the pertinent trace represents undoubtedly also
play a role in whether or not contact is made with the trace (1, 5).
Therefore, where trace selection is left to spontaneous recall, the in-
fluence of a need or of another central factor on perception is not alto-
gether a dynamic matter. This must be borne in mind when quantitative
2 See R. R Street, A Gestalt completion test. Teachers Coll Contrib. to Educ.,
1931, No. 481. Ed.
168 : Hans Wallach
results are interpreted. They represent not only the strength of the
central factor in the perceptual process but also the probability of the
occurrence of the contact with the pertinent trace.
This last consideration presupposes, of course, that, as in the case of
meaning, central factors gain whatever influence they may have on
perception through mediation of a more or less specific memory trace.
This is certainly true of values which with few exceptions are them-
selves mediated by meaning. This is also true of needs. On the whole,
particularly in a mentally healthy individual, a need becomes operative
only in specific situations. A given situation must be recognized as be-
longing to a certain kind which is specific for the operation of a particu-
lar need. It is even true of drives when they are aroused by their drive
objects. Except for the relatively rare cases where cathexis of a novel
drive object occurs in the perceptual situation, the capacity of an ob-
ject or situation to arouse a drive was acquired in the past, and this
product of previous cathexis is brought into function through a recog-
nition process.
That the arousal of a pertinent memory trace must precede the
influence of central factors on perception is particularly important
where these factors are supposed to affect recognition, as in the case
of tachistoscopic studies (6). It seems very likely that the arousal of a
pertinent trace amounts to recognition. But the assumption that values
and needs affect recognition seems to contradict this simple conception.
For these central factors which supposedly have an influence on
whether or not recognition occurs come into play through the arousal
of pertinent traces. This makes no sense if recognition is equivalent to
trace arousal, A solution to this dilemma will have to be found in a
closer analysis of the psychological processes which go on in such ex-
periments.
Up to the present the main interest of the studies on the influence of
central factors on perception has been in demonstrating such effects.
Such demonstrations show to what extent the organization of per-
ceptual processes can be influenced by contents and conditions of cen-
tral origin. But this is not the only way in which studies of this kind
may contribute to the investigation of perceptual functions. They may
be helpful in answering the following question: To what stage of or-
ganization must a visual process develop before it can arouse a memory
trace? Since, in the absence of a "set," central factors gain an influence
only after pertinent traces have been aroused, properly designed experi-
The Relation Between Perception and Cognition : 169
ments which attempt to produce such an influence may be a way to
answer this question.
I should like to clarify this question briefly: A number of perceptual
functions besides the strictly sensory processes have to take effect to
bring about visual percepts as we experience them. They are, to name
only those which are of interest in this context, the interaction process
on which the formation of dense surface colors depends, organization
due to grouping factors, and the formation of "figure and ground." The
question is whether they must also take effect before a pertinent mem-
ory trace can be aroused.
The answer to this question has certainly to be affirmative insofar as
the colors of the achromatic scale are concerned. The gray color seen
in a certain region of the visual field depends on the relation of the
intensity of stimulation received from this region and the stimulus
intensity of the surrounding region (9). An interaction between the
local processes must be responsible for the gray color. A light gray can
be changed into a dark gray either by lowering the stimulus intensity
in the region in which the gray is seen or by raising the intensity in the
surroundings. The end results of these changes cannot be told apart,
although in one case stimulation in the gray region changes and in the
other it does not. This shows that no sensory process which corre-
sponds to the intensity of local stimulation has trace representation;
only the product of the interaction process plays a role in recognition.
The evidence concerning grouping pertains only to line figures.
Demonstrations of Kohler's (4) and the work of Gottschaldt (2) have
made it clear that figures with which the subject is well acquainted will
not be spontaneously recognized when lines are added which prevent
these figures from appearing as separate units. I think it is easy to rule
out the possibility that the comprehensive figure which comes about
through such addition prevents recognition of the "hidden" figure by
initiating its own recall process in competition with the "hidden" figure.
Rather, "hidden" figures fail to be recognized because they do not ap-
pear as separate units and are therefore unable to make trace contacts.
Since in the case of line figures the formation of units is ruled by group-
ing principles, organization due to grouping factors should precede
trace arousal. There is certainly need for direct evidence to support this
reasoning.
The evidence concerning "figure and ground" is at present contra-
dictory. The essence of the distinction of figure and ground is that
170 : Hans Wallach
only the area designated as figure has form. Rubin has demonstrated
that an area will be recognized by its shape only if it is seen as figure
(7). This means that the sensory pattern must first be organized in
terms of figure and ground, before the figure areas establish contact
with specific memory traces.
This seems to be irreconcilable with the result of the brilliant experi-
ment of Shafer and Murphy (8). It demonstrated an influence of
previous reward and punishment on the perception of figure and
ground in an ambiguous design. An outline circle was divided by an
irregular vertical line of such a shape that each half of the circle could
be seen as a profile of a face of half-moon shape. Prior to the test in
which two such ambiguous figures were used each half -moon shape
was presented singly, and each presentation was always accompanied
by a reward or a punishment. In the test, the previously rewarded half-
moon shapes were predominantly recognized, that is, seen as figure.
There can be no doubt that in this experiment the aftereffect of
previous reward and punishment was connected with the traces of the
forms of the training series and could become effective only through
the arousal of these traces. Thus, figure-ground distribution is in-
fluenced by factors which can only come into play after the pertinent
trace has been aroused. This seems to contradict the conclusion drawn
from Rubin's result that formation of figure and ground must precede
trace arousal.
This contradiction disappears only when one takes into account that
Rubin worked with solid figures, that is, pattern where the contour
between figure and ground is formed by the borderline between differ-
ent colors, whereas Shafer and Murphy used outline figures. The black
line on white which here forms the contour between figure and ground
also has form in its own right and can therefore make trace contact. A
repetition of Shafer and Murphy's experiment with solid figures would
really go to the heart of the matter. My hunch is that it would fail
to demonstrate any influence of reward and punishment on figure-
ground distribution.
I have recently become impressed with the extent to which memory
traces participate in simple perceptual processes. They not only im-
part meaning or impose grouping as previously mentioned; they also
seem to be responsible for the perception of three-dimensional form
where perceptual conditions for primary organization in depth are ab-
sent or weak.
In this situation the full answer to the question, at what stage of
The Relation Between Perception and Cognition : 171
perceptual organization pertinent traces can be aroused, seems to be
very much needed. As matters stand I would say that the interaction
process on which the formation of dense surfaces depends, the forma-
tion of figure and ground, and in some cases, organization due to the
grouping factors must develop before traces can be aroused through
similarity of form. That such processes must occur before traces can be
aroused may possibly be one of the reasons for the great stability of
simple perceptual processes and the high degree of independence from
other psychological functions which they exhibit. One may say that,
up to a certain point, the development of percepts is protected against
interference by central factors, because the developing percepts are
inaccessible before trace contact has been made.
REFERENCES
1. Bartel, H. Ueber die Abhangigkeit spontaner Reproduktionen von Feld-
bedingungen. PsychoL Forsch., 1937, 22, 1-25.
2. Gottschaldt, K. Ueber den Einfluss der Erfahrung auf die Wahrnehmung
von Figuren. I. PsychoL Forsch., 1926, 8, 261-317.
3. Kohler, W. Dynamics in psychology. New York: Liveright, 1940.
4. Kohler, W. Gestalt psychology. (Rev. ed.) New York: Liveright, 1947.
Chap. VI.
5. Kohler, W., and von Restorff, H. Zur Theorie der Reproduktion. PsychoL
Forsch., 1935, 21, 56-112.
6. Postman, L., Bruner, J. S., and McGinnies, E. Personal values as selective
factors in perception. /. abnorm. soc. PsychoL, 1948, 43, 142-154.
7. Rubin, E. Visuell wahrgenommene Figuren. Copenhagen: Gyldendal,
1921.
8. Shafer, R., and Murphy, G. The role of autism in a visual figure-ground
relationship. /. exp. PsychoL, 1943, 32, 335-343.
9. Wallach, H. Brightness constancy and the nature of achromatic colors.
/. exp. PsychoL, 1948, 38, 310-324. [See pp. 109-125 of the present
book.]
Mary Henle
GRADUATE FACULTY, NEW SCHOOL FOR SOCIAL RESEARCH
SOME EFFECTS OF
MOTIVATIONAL PROCESSES
ON COGNITION
Work on the influence of needs and attitudes on perception and other
cognitive processes has proceeded with insufficient analysis of the
problem. It is the purpose of the present paper to outline a number of
possible effects of motivational processes on cognitive ones. 1 The in-
fluences to be described are regarded neither as established nor as
exhaustive, but are offered as hypotheses for future research.
Frequently it seems to be tacitly assumed that the mere presence of a
need or attitude is sufficient to account for an observed effect. The task
of research has become, therefore, one of relating the presence or ab-
sence of particular motivational conditions to quantitative variations
in performance on cognitive tasks. The point of view to be presented
here is, rather, that the finding of such a correlation only opens up a
problem: that of understanding how motivation influences cognition
(cf. 44; also 6, p. 139).
Reprinted with permission from the Psychological Review, Vol. 62, No. 6,
November, 1955.
1 This paper is part of a project undertaken when the writer was a Fellow of
the John Simon Guggenheim Memorial Foundation.
Some Effects of Motivational Processes on Cognition : 173
The point may be illustrated by reference to the question of the
effect of attitudes on recall of controversial material. There is some
evidence to indicate that we remember material we agree with better
than apparently equivalent material with which we disagree (e.g., 14,
32). That we agree with certain points is no explanation of their
superiority in recall. But if we take this finding as a point of departure,
it is indeed possible to find plausible reasons why material we agree
with should be favored in memory. These are, of course, only hy-
potheses which require independent testing, (a) It may be suggested
that an attitude functions as context for related material (cf. 5, p. 582) .
Presented data are understood in relation to the subject's existing atti-
tudinal structure. Items which are in harmony with the attitude find
their place in the structure in a simpler and more direct manner than
does opposing material. There seems to be little doubt although it
needs to be demonstrated for material comparable to that used in the
studies on recall of controversial material that structured material is
better recalled than unrelated items (24) . The superior recall of items
we agree with might follow, then, from the advantage of structured
over unstructured data in memory, (b) Material we agree with may be
better recalled because it is better understood than material which
opposes our own attitude. Material which is understood is known to
fare better in memory than that which we do not understand (24). A
point we agree with is not, of course, better understood just because
we agree with it, but because we have thought about it, have placed it
in context, etc. Nor are data that confirm our attitudes always better
understood; an insult may be very well understood, (c) Facts and argu-
ments we agree with may have an advantage in memory over opposing
ones when they are, to start with, more familiar- (d) Material we agree
with may be received in a more friendly manner, while that with which
we disagree may be rejected at the outset, regarded as nonsense, etc.
The former condition is, very likely, the more favorable for recall.
Other things being equal, material to which we have given attention is
favored in memory, (e) Intention to recall, a factor known to be im-
portant for memory, may operate to favor items with which we agree. 2
We may try to remember certain points because we wish to refer to
2 We are not concerned here with the problem of how intention operates in
memory. It should be pointed out, however, that the intention does not impose
itself on recall, but seems to act by altering other known conditions of recall.
The problem of the influence of intentions on cognitive processes may be of the
same order as the problem here under consideration.
174 : Mary Herile
them again, to use them in argument, because they are flattering to us,
etc.
Repression will not be discussed in the present connection because,
even if it operates in the experiments under consideration, there would
seem to be no way of knowing that it does. It would be necessary to
have detailed knowledge of the repressions of individual subjects in
order to guess what new data might be repressed by association with
them. It is, in addition, highly unlikely that the experiments which
deal with the recall of controversial material have set up the conditions
necessary to produce repression.
While there is evidence that makes the above hypotheses seem plausi-
ble, it is also becoming recognized that there are conditions under
which material which opposes our attitudes is favored in recall (cf. 2,
4) . We may begin to define these conditions too. (a) We may remem-
ber something that disturbs us precisely because it does not fit into our
schema. It might be that when an attitudinal structure is ripe for
change, such disturbing, contradictory material is favored, while the
organizing effects of attitudes enhance recall of confirming material at
other times, (b) Points we disagree with may stand out, in contrast to
repeated evidence for something we believe. It is known that such out-
standingness can favor recall. The fact that outstanding items within a
system are favored does not, of course, contradict the statement made
earlier that structured data are more likely than unorganized ones to
be well recalled, (c) We may remember something in order to refute
it, to accuse somebody of it, in some way to cope with it, etc. Intention
to recall, in the service of some other need, may also favor the opposing
material.
Finally, certain attitudes may operate to produce about equal recall
of material we agree with and of that with which we disagree for
example, an objective attitude, a desire to be fair, to hear both sides
of the case, a wish to recall as much as possible so as to do well in the
recall test, etc. (2).
It is clear from the variety of hypotheses presented above that it is
necessary to have detailed information about a particular subject's
understanding of, and attitude toward, the presented material if we
are to attempt any predictions about the kind of material which will
be favored in recall. It will also be clear that the really interesting prob-
lems about motivational factors in recall are not confined to the favor-
ing of some particular kind of material, but concern also the qualitative
dimensions of the individual subject's pattern of recall, involving the
kinds of processes suggested here. It may even be that the individual's
Some Effects of Motivational Processes on Cognition : 175
pattern of recall of personally relevant material is so distinctive that, in
the hands of a skilled clinician, it may yield projective data which agree
well with personal information derived from interview and projective
methods. 3
The above hypotheses, it will be noted, seek the effects of attitudes on
remembering in terms of factors known to favor or inhibit recall. They
introduce no new determinants, but are largely statements of ways in
which attitudes might alter the structural conditions of recall.
By contrast, much of the interest of current research in this field
has been to find distortions of cognitive processes by needs and atti-
tudes. Such distortions undoubtedly occur, although their frequency
under normal conditions may be questioned. But it is here suggested
that the influence of motivational processes on cognitive ones is not
limited to distortions, and that such effects are not necessarily the ones
most likely to give us an understanding of the processes involved.
Rather, as has been pointed out elsewhere (22), a possibly more fruit-
ful starting point for research would seem to be to look for changes by
motivational processes in accordance with the nature of the material on
which they act. 4 The present paper will seek to describe ways in which
needs and attitudes may alter cognitive processes in a manner that does
not violate the presented structure.
As is implied in the above remarks, it is maintained here that moti-
vational processes may influence, but do not produce, cognitive organi-
zation. A vector (e.g., a need, expectation, intention, or attitude) or a
trace system can influence a perceived form or another memory trace
only if the latter already exists. The problem, as Wallach (52) has
pointed out, is to determine the stage of organization to which a visual
process must develop before central factors can influence it. 5
To see motivational processes as operating through the presented
3 I am indebted to Florence R. Miale for a preliminary demonstration of this
kind.
Alper and Korchin (4, p. 35) make a similar suggestion: ". . . selective re-
call may well function here as it does in the so-called projective tests: the subject
reacts to the material selectively in terms of his needs and tension-systems, the
products of his recall being themselves projections of these needs and tension-
systems." These authors fail, however, to follow up their suggestion with detailed
comparisons of recall and personality data of individual subjects.
4 The theoretical consequences of taking distortions as the paradigm of the
influence of needs and attitudes on cognition have been shown elsewhere (22).
5 The present writer fails to see how the assumption of unbewusster Schluss
(e.g., 9) solves this problem. In order to initiate processes of inference or recall,
perceptual data must already be organized. We cannot make inferences about
an object until we can perceive it; the inferences thus do not account for, but
presuppose, the perception.
776 : Mary Henle
structure is not, however, to deny the influence of needs and attitudes
on cognition. In a recent paper, Postman (38) seems to reduce such
influences to others which are not themselves motivational, viz., set and
past experience. Certainly the latter processes have important influ-
ences on cognitive ones, and motivation may act on cognition through
them. Furthermore, they seem to cut across a number of the other
influences of needs and attitudes; but it will be maintained in the para-
graphs that follow that they by no means fully account for the effects in
question.
We are not concerned here with imagination, wishing, etc., which
for present purposes may be considered to be very largely determined
by motivational factors. 6 Rather, the present discussion is limited to
some of the interactions which occur between motivational and struc-
tural determinants of cognition.
The following hypotheses are offered as possible modes of influence
of needs and attitudes on cognitive processes. They may guide, but
they do not, of course, take the place of concrete analysis of particular
problems. Some of these hypotheses are already to be found in the
literature. Others seem to derive some support from existing studies,
even though these were not specifically designed to test them. In the
case of still other hypotheses, the task of testing them remains for the
future.
L A need or attitude may operate as a vector, pointing in one di-
rection rather than another. It is sometimes possible, under the in-
fluence of a need or attitude, to find an item which would otherwise be
unnoticed in the perceptual field. This is easy to demonstrate with
camouflaged items. Likewise a recall vector may bring things to mind
that do not occur spontaneously. (Of course the search refers to mem-
ory traces established in the past. The vector, to be effective, must have
something to point to; aroused traces, supplying this, enable the vector
to be effective. Here, as in other effects to be discussed below, the
attitude or need operates in cooperation with the individual's past
experience.)
It is likely that effects of pointing may be demonstrated with other
vectors besides simple search vectors in perception and memory. It
6 This is probably an oversimplification. It would be worth knowing to what
extent even these processes are limited by our knowledge, experience, and assump-
tions both conscious and unconscious about ourselves and about reality.
Some Effects of Motivational Processes on Cognition : 177
seems that, under the influence of a need or attitude, we are attuned to
events to which we would not otherwise be sensitive. Under the pressure
of an unresolved need, we find things to be angry or worried or hurt
about.
This hypothesis seems to be essentially the same as earlier views of
Bruner and Postman and others (e.g., 9, 42) on selective sensitization
to valued or needed aspects of the environment. In later writings these
authors seem to regard this process as a function of expectancies or
"hypotheses" only. Postman, for example, states that "There is little
evidence for direct sensitizing effects of motivational conditions on per-
ception" (38, p. 99). It is here suggested that this function needs to be
re-examined with respect to motivational states.
Some experimental work on the relation of recognition thresholds
to motivational states seems to be relevant. If these studies have indeed
demonstrated a lowering of threshold for need-related material that
cannot be accounted for by differential frequency or the operation of
specific expectancies, a process of pointing may be operating. For
example, Postman and Brown (39) have shown that experiences of
success attune the individual to tachistoscopically presented goal words
(e.g., "succeed") , while after failure the subject is relatively more sensi-
tive to deprivation words (e.g., "obstacle," "failure"). A possible in-
terpretation of this finding is that the persisting mood attunes the
individual to material congruent with it. McClelland and Liberman
(34) report comparable results for individuals differing in the strength
of their need for achievement. For example, subjects (Ss) with a strong
need for achievement, recognized goal and instrumental words relating
to achievement (e.g., "success," "achieve") faster than those whose
need was weak. Two studies on the influence of hunger on perceptual
sensitivity to need-related material seem also to be relevant. Lazarus,
Yousem, and Arenberg (31) report that recognition thresholds for
photographs of food objects declined with increasing hours of food
deprivation (within limits set by the habitual eating cycle), and Wispe
and Drambarean (54) found that need-related words were recognized
more rapidly than neutral ones under conditions of food and water
deprivation, but not when Ss were satisfied.
These experiments need to be repeated under conditions where pos-
sible sets for need-related items are not allowed to develop during the
experimental series. For example, the relative sensitivity to the first
need-related item in the series should be studied (cf . 28 ) . If the findings
should be confirmed under these conditions, they would suggest the
178 : Mary Henle
operation of the kind of sensitization to need-related material here
described as pointing.
2. Closely related to pointing may be the organizing effects of needs
and attitudes. Within certain limits we can voluntarily influence organi-
zations in the perceptual field, grouping together items which would
not spontaneously go together. Likewise in recall, interaction between
a process and a trace which would not occur spontaneously may take
place under the influence of a vector (27). In the case of other cog-
nitive processes, the same organizing effects of needs suggest them-
selves. It is a fact of common observation that when a strong need or
interest is aroused, the facts of experience organize themselves around
it. When I am working on a scientific problem, for example, everything
I read appears to bear on the issues with which I am concerned. It
seems that under the influence of an aroused need we perceive similari-
ties not otherwise noticeable. These vectors, I repeat, operate in con-
junction with the individual's knowledge and previous experience.
But here, as in the cases that follow, it is insufficient to refer to past
experience alone.
3. The perception of other relations is likewise influenced by needs
and attitudes. It has been suggested that under the influence of an
aroused need, the psychological field may be restructured so that learn-
ing occurs ( 1 ) . For example, an object previously seen as unrelated to
a goal may come to be perceived as the means to attaining it, one event
may acquire the meaning of a signal for another, etc.
4. An aroused need or attitude may act on cognition by selection
among the various possibilities presented. This hypothesis has fre-
quently appeared in the literature. It has not, however, customarily
been separated from sensitization or pointing, as discussed above (e.g.,
9,42).
The following example, among the many possible, illustrates the
selective effect of an attitude in cognition. If two individuals who hold
opposed attitudes are presented with a given fact, they do not neces-
sarily perceive the same fact, but each may select for it a different
meaning out of several possibilities it presents. As Asch puts it (5, p.
584) : "One can observe much adroitness in the manipulation of mean-
ings in the interests of an undisturbed outlook." Selective effects of
needs and attitudes are most familiar from the projective methods. In
the Rorschach, for instance, a variety of interpretations may be given
to a similarly perceived portion of a blot; consider, for example, the
variety of actions attributed by different individuals to the animals in
Some Effects of Motivational Processes on Cognition : 179
the side details in Card VIII. Examples of this kind could be multiplied.
The experimental literature provides comparable examples. Sanford
(46, 47) and Levine, Chein, and Murphy (33) found that hungry sub-
jects gave more food responses in the interpretation of incomplete or
ambiguous pictures than did satisfied ones, and that the effect increased
(within limits) as hunger increased. (The effect of the need cannot,
however, be separated from possible selective effects of a food set, in
the experiment of Levine, Chein, and Murphy.) There appears to be a
selection of need-relevant interpretations from among the many mean-
ings the ambiguous material could be given.
5. The need or attitude may supply context. Since the context may
influence decisively the manner in which an item is experienced, it fol-
lows that a given item may be differently viewed in accordance with
the need or attitude aroused. This effect of attitudes was illustrated
above in the case of a problem of memory. Needs may be expected to
have comparable effects, since they function typically in need-object
organizations or "sentiments."
6. One particular effect of the fact that needs and attitudes function
as temporally extended organizations deserves special mention: mem-
ory traces relevant to these organizations may be aroused. Past experi-
ence has important effects on cognitive processes, of which the more
relevant in the present connection may be such effects as the con-
tributing of meanings, the establishing of norms or adaptation levels
(20), and the rendering commonplace of some item of experience so
that it is overlooked or its significance lessened in the cognitive field.
Past experience likewise facilitates perception under conditions of re-
duced stimulation (37, 21, 23, 49, 50; but cf. also 17, 29). It probably
acts also as a selective factor, favoring certain possibilities among
those which are structurally given (e.g., 13, 53).
These effects of past experience have been demonstrated in cognitive
situations having little bearing on subjects' needs and attitudes. The
work on the role of past experience needs to be extended to other cog-
nitive situations which have motivational relevance.
7. A need may arouse an expectation, which is known to have cer-
tain effects on cognitive processes. For example, Titchener long ago
formulated a principle of prior entry. "The stimulus for which we are
predisposed requires less time than a like stimulus, for which we are
unprepared, to produce its full conscious effect" (51, p. 251). The
sensitizing, organizing, and selective effects of sets have been men-
tioned above. Bruner has suggested that the function of expectancies
180 : Mary Henle
"is to re-order the availability of traces" (7, p. 307; cf. previous hy-
pothesis). A number of recent experiments have dealt with the effects
of sets on cognitive processes and the conditions under which they
operate (e.g., 41, 10, 1 1 ). It is clearly important to distinguish between
direct effects of motivational processes on cognition and thek indirect
effects through the arousal of expectations, which frequently have simi-
lar consequences ( cf . 3 8 ) .
Postman and Crutchfield (43) have recently discussed the arousal
of expectations by a state of need. These authors presented incomplete
words for completion, varying the intensity of hunger of thek Ss and
the degree of selective set for food responses. They found the effects
of set in determining food responses to be larger than those of differing
intensities of hunger; the relationship between the latter and frequency
of food responses depended largely on S's expectation for such re-
sponses. They conclude (p. 217): "Intensity of need is one of the
variables which modify the operation of such general principles of
cognition as selective 'set within limits defined by the characteristics of
the stimulus-materials' 3
8. Needs or attitudes may make us unwilling to ask certain ques-
tions, discourage the deske to understand, keep us from considering
relevant evidence or from seeing the relevance of presented evidence.
They may narrow the mental field (1,5), with significant consequences
for cognition. Much current work has been concerned with such in-
fluences (e.g., the effects of prejudice, the clouding of judgment by
strong emotion). These effects are not, at the outset, to be dismissed
as entirely automatic evasions. Reasons are frequently used in these
instances to give them at least the appearance of sense. The cognitive
processes involved need to be understood.
9. Needs or attitudes, in a manner closely related to the last-men-
tioned effect, may cause us to overlook differences, to fail to make
distinctions, just as in the perceptual field a great contrast may make
us fail to see a lesser one. For example, several years ago the writer
heard a debate on a perceptual problem held before a group which was
bitterly opposed to the use of the experimental method in psychology.
The audience seemed not to know that a debate was going on, but
attacked both speakers alike for dissecting thek human subject matter.
They saw no important difference between the two positions presented.
Common experience suggests similar phenomena in the realm of social
and political attitudes. To a conservative person everything left of center
Some Effects of Motivational Processes on Cognition : 181
may seem to be radical, while the more radical individual may regard
everything right of center as reactionary. The range over which fine
distinctions can be made appears to be shortened.
10. A possible effect of needs and attitudes is a specific disturbance
of recognition and recall. Some evidence suggests that individuals who
are shown, by independent methods, to differ in the extent to which
given experimental material is disturbing to them, show corresponding
differences in their recognition thresholds for such material (15, 16,
30). If these findings are confirmed, and if alternative explanations
(e.g., differential readiness to report) can be eliminated, a disturbance
of die recognition of certain kinds of threatening material suggests it-
self. Again, the facts of repression, insofar as this mechanism involves
forgetting, raise a similar problem.
How might such a disturbance of recognition and recall operate?
The phenomena of repression suggest that it is not a matter of destruc-
tion of memory traces; for repressed material to express itself in dreams,
symptoms, and other forms requires that the corresponding traces
be intact. The disturbance seems, rather, to concern that interaction
between process and trace, based on their similarity, which underlies
recognition and is the first step in the process of recall by association.
(For a discussion of the selective influence of similarity in recognition
and recall, cf. Kohler, 26, pp. 126 ff.)
It is very tentatively suggested, then, that a vector may operate to
prevent that interaction between a present process and a memory trace
which is necessary for recognition and recall.
11. Needs and attitudes may act on the physiognomic properties of
experience. Consider, for example, the change which can sometimes
be noted in a person's appearance for us with the change from acquaint-
ance to friendship. That the change is not a matter of familiarity alone
is suggested by the fact that such physiognomic changes seem to be
much less pronounced when increasing acquaintance is not accom-
panied by friendship. It is true that the friend looks upon us with a more
kindly eye than the mere acquaintance, may be happier and more re-
laxed with us, etc., so that some of the perceived change is accounted
for by actual changes in his appearance. It is worth considering, how-
ever, whether changes in the person's physiognomic properties
changes which transcend differences in mood may not be, in part, a
function of the attitudes and sentiments we have for him.
It might be that, if Murray (35) has demonstrated a genuine in-
182 : Mary Henle
fluence of fear on judgments of maliciousness, it is of the nature of an
effect on physiognomic properties. (Other interpretations, however,
are possible.)
12. Needs or attitudes may animate, enliven, activate, or give out-
standing position to relevant parts of the cognitive field. These effects,
in turn, have certain consequences for cognition. 7 In a very real sense
the significant person stands out in a group. The effects of outstanding
position in perception, memory, and thinking are well known. As
another example, the possibility is suggested that Zeigarnik's main re-
sult the favoring of incomplete over completed tasks in recall (55 )
may be accounted for if it can be assumed that need tension may in
some way make a trace more lively or more active and thus increase its
availability. Gilchrist and Nesberg (18) have reported an experiment
which suggests an interpretation in these terms. Hungry and thirsty
subjects were asked to match the illuminance of just previously pro-
jected pictures of need-relevant objects. Increasing need gave rise to
increasingly bright matches. In the case of thirsty subjects, the error
dropped to its starting level immediately after drinking. If such a result
should be confirmed, it could be viewed as an instance of the enlivening
effect of need in immediate memory.
13. A strong need or interest leads us to exert efforts in its service.
These may show themselves in increased efficiency in cognitive tasks,
while a relaxation of effort reduces efficiency. In the case of excessive
effort there may be a disruption of performance. A number of studies
seem to permit interpretation in these terms rather than in terms of
actual perceptual change.
Bruner and Postman (8) found that when Ss were given electric
shock during a task, they judged the size of a disk which figured in the
task as accurately as control Ss; but when the shock was removed, their
size judgments increased. Since magnification in size is here equivalent
to increased inaccuracy, it is suggested that this effect is not a matter of
"post-tension expansion" in perception, but of relaxation of efforts in
a judging task. In another study (40) the same authors found that Ss
7 This hypothesis does not depend upon a principle of "prior entry" (9, p. 96) .
It sounds like some of Bruner and Postman's statements about perceptual "ac-
centuation" (e.g., 9, p. 100). In practice, however, these authors have used ac-
centuation only to mean increase in perceived size. As such, it comes under the
heading of perceptual distortions, which are not being discussed here. In the
present writer's opinion, accentuation of perceived size has, in any case, not
been demonstrated as a function of value, with autochthonous factors properly
controlled (cf. 2, 12, 19, 25, 36).
Some Effects of Motivational Processes on Cognition : 183
who had been harassed and badgered during an impossible perceptual
task showed higher recognition thresholds in another perceptual situa-
tion than control Ss, as well as premature and frequently nonsensical
prerecognition hypotheses. It seems plausible to regard these findings
as indications of disruption of test performance as a result of the actu-
ally reported excessive efforts of the experimental Ss. Rosen (45), on
the other hand, found that Ss who were able to avoid an electric shock
by correct perception had lower recognition thresholds for nonsense
syllables than members of a control group who did not receive shock.
Here it would seem that added, but not excessive, effort increased effi-
ciency of test performance.
Other reports suggest comparable effects of attitudes in perception
and memory. Allport and Kramer (3) found that anti-Semitic indi-
viduals were able to identify photographs of Jewish and non- Jewish
faces more accurately than Ss free of prejudice. The result is attributed
to the greater importance of racial identity to prejudiced persons. Simi-
larly Seeleman (48) found that different attitudes led to different de-
grees of effort and attention to an exposure series, with consequent dif-
ferences of performance in a recognition test. Thus individuals with
favorable attitudes toward the Negro correctly recognized more Negro
photographs than did anti-Negro Ss.
Such differences in performance are, of course, to be distinguished
from actual cognitive change under the influence of a need or attitude. 8
SUMMARY
This paper has been concerned with the question of how needs and
attitudes influence cognitive processes. The attempt has been made to
describe ways in which motivational processes alter cognitive ones in
accordance with the nature of the material on which they act. It was
suggested that needs and attitudes may act by pointing or sensitizing,
organizing and reorganizing, selecting, supplying context, arousing
relevant memory traces, arousing expectations, discouraging the desire
to understand, obscuring differences, disturbing the recognition proc-
ess, altering the physiognomic properties of experience, and animating
or enlivening aspects of experience. In addition, strength of motivation
8 Postman (38) has discussed other motivational influences on performance
in perceptual experiments, including the matter of selective reporting, which is
a serious problem for all attempts to study possible differences in the perception
of emotional and neutral material.
184 : Mary Henle
may influence performance on perceptual or other cognitive tasks
without producing actual cognitive change.
REFERENCES
1. Adams, D. K, A restatement of the problem of learning. Brit. J. Psychol.,
1931,22, 150-178.
2. Adelman, C, et al. An investigation of the influence of needs and attitudes
on perception and memory. Unpublished research, New School for
Social Research, 1951.
3. Allport, G. W. 5 and Kramer, B. M. Some roots of prejudice. J. Psychol.,
1946, 22, 9-39.
4. Alper, Thelma G., and Korchin, S. J. Memory for socially relevant ma-
terial. /. abnorm. soc. Psychol., 1952, 47, 25-37.
5. Asch, S. E. Social psychology. New York: Prentice-Hall, 1952.
6. Bruner, J. S. Personality dynamics and the process of perceiving. In R. R.
Blake and G. V. Ramsey (Eds.), Perception: an approach to personal-
ity. New York: Ronald, 1951.
7. Bruner, J. S. One kind of perception: a reply to Professor Luchins.
Psychol Rev., 1951, 58, 306-312.
8. Bruner, J. S., and Postman, L. Tension and tension release as organizing
factors in perception. /. Pers., 1947, 15, 300-308.
9. Bruner, J. S., and Postman, L. An approach to social perception. In W.
Dennis (Ed.), Current trends in social psychology. Pittsburgh: Univer-
sity of Pittsburgh Press, 1948.
10. Bruner, J. S., and Postman, L. On the perception of incongruity: a
paradigm. /. Pers., 1949, 18, 206-223.
11. Bruner, J. S., Postman, L., and Rodrigues, J. Expectation and the per-
ception of color. Amer. J. PsychoL, 1951, 64, 216-227.
12. Carter, L., and Schooler, K. Value, need, and other factors in perception.
PsychoL Rev., 1949, 56, 200-207.
13. Djang, S. The role of past experience in the visual apprehension of
masked forms. J. exp. PsychoL, 1937, 20, 29-59.
14. Edwards, A. L. Political frames of reference as a factor influencing
recognition. /. abnorm. soc. Psychol., 1941, 36, 34-50.
15. Eriksen, C. W. Perceptual defense as a function of unacceptable needs.
/. abnorm. soc. PsychoL, 1951, 46, 557-564.
16. Eriksen, C. W. Defense against ego-threat in memory and perception.
J. abnorm. soc. PsychoL, 1952, 47, 230-235.
17. Eriksen, C. W. The case for perceptual defense. PsychoL Rev., 1954,
61> 175-182.
18. Gilchrist, J. C., and Nesberg, L. S. Need and perceptual change in need-
related objects. /. exp. PsychoL, 1952, 44, 369-376.
19. Golden, S. An experimental investigation into some factors that deter-
mine the perception of size. Unpublished master's thesis, New School
for Social Research, 1950.
Some Effects of Motivational Processes on Cognition : 185
20. Helson, H. Adaptation-level as a basis for a quantitative theory of
frames of reference. PsychoL Rev., 1948, 55, 297-313.
21. Henle, Mary. An experimental investigation of past experience as a
determinant of visual form perception. I. exp. PsychoL, 1942, 30, 1-
22.
22. Henle, Mary, and Michael, M. The influence of attitudes on syllogistic
reasoning. /. soc. PsychoL, 1956, 44, 115-127.
23. Howes, D. H., and Solomon, R. L. Visual duration threshold as a func-
tion of word-probability. /. exp. PsychoL, 1951, 41, 401-410.
24. Katona, G. Organizing and memorizing. New York: Columbia Univer-
sity Press, 1940.
25. Klein, G., Schlesinger, H., and Meister, D. The effect of personal values
on perception: an experimental critique. PsychoL Rev., 1951, 58,
96-112.
26. Kohler, W. Dynamics in psychology. New York: Liveright, 1940.
27. Kohler, W., and von Restorff, H. Zur Theorie der Reproduktion.
PsychoL Forsch., 1935, 21, 56-112.
28. Lacy, O. W., Lewinger, N., and Adamson, J. F. Foreknowledge as a
factor affecting perceptual defense and alertness. /. exp. PsychoL, 1953,
45, 169-174.
29. Lazarus, R. S. Is there a mechanism of perceptual defense? A reply to
Postman, Bronson, and Cropper. /. abnorm. soc. PsychoL, 1954, 49,
396-398.
30. Lazarus, R. S., Eriksen, C. W., and Fonda, C. P. Personality dynamics
and auditory perceptual recognition. /. Pers., 1951, 19, 471-482.
31. Lazarus, R. S., Yousem, H., and Arenberg, D. Hunger and perception.
J.Pers., 1953,21,312-328.
32. Levine, J. M., and Murphy, G. The learning and forgetting of contro-
versial material. J. abnorm. soc. PsychoL, 1943, 38, 507-517.
33. Levine, R., Chein, L, and Murphy, G. The relation of the intensity of
a need to the amount of perceptual distortion: a preliminary report.
/. PsychoL, 1942, 13, 283-293.
34. McClelland, D., and Liberman, A. The effect of need for achievement
on recognition of need-related words. J. Pers., 1949, 18, 236-251.
35. Murray, H. A. The effect of fear upon estimates of the maliciousness of
other personalities. /. soc. PsychoL, 1933, 4, 310-329.
36. Norton, E. N. Is symbolic value an organizing factor in perception?
Unpublished master's thesis, New School for Social Research, 1950.
37. Ortner, A. Nachweis der Retentionsstorung beim Erkennen. PsychoL
Forsch., 1937, 22, 59-88.
38. Postman, L. The experimental analysis of motivational factors in per-
ception. In Current theory and research in motivation. Lincoln: Univer-
sity of Nebraska Press, 1953.
39. Postman, L., and Brown, D. Perceptual consequences of success and
failure. /. abnorm. soc. PsychoL, 1952, 47, 213-221.
40. Postman, L., and Bruner, J. S. Perception under stress. PsychoL Rev.,
1948,55,314-323.
186 : Mary Henle
41. Postman, L., and Brunei, J. S. Multiplicity of set as a determinant of
perceptual behavior. 7. exp. PsychoL, 1949, 39, 369-377.
42. Postman, L., Bruner, J. S., and McGinnies, E. Personal values as selec-
tive factors in perception. 7. abnorm. soc. PsychoL, 1948, 43, 142-154.
43. Postman, L., and Crutchfield, R. S. The interaction of need, set and
stimulus-structure in a cognitive task. Amer. 7. PsychoL, 1952, 65,
196-217.
44. Prentice, W. C. H. "Functionalism" in perception. PsychoL Rev.,
1956, 63, 29-38.
45. Rosen, A. C. Change in perceptual threshold as a protective function
of the organism. 7. Pers., 1954, 23, 182-194.
46. Sanford, R. N. The effects of abstinence from food upon imaginal proc-
esses: a preliminary experiment. 7. PsychoL, 1936, 2, 129-136.
47. Sanford, R. N. The effects of abstinence from food upon imaginal proc-
esses: a further experiment. 7. PsychoL, 1937, 3, 145159.
48. Seeleman, V. The influence of attitude upon the remembering of pic-
torial material. Arch. PsychoL, 1940, No. 258.
49. Solomon, R. L., and Howes, D. H. Word frequency, personal values,
and visual duration thresholds. PsychoL Rev., 1951, 58, 256-270.
50. Solomon, R. L., and Postman, L. Frequency of usage as a determinant
of recognition thresholds for words. 7. exp. PsychoL, 1952, 43, 195
201.
51. Titchener, E. B. Lectures on the elementary psychology of feeling and
attention. New York: Macmillan, 1908.
52. Wallach, H. Some considerations concerning the relation between per-
ception and cognition. 7. Pers., 1949, 18, 6-13. [See pp. 164-171 of
the present book.]
53. Wallach, H., O'Connell, D. N., and Neisser, U. The memory effect of
visual perception of three-dimensional form. 7. exp. PsychoL, 1953,
45, 360-368.
54. Wispe, L. G., and Drambarean, N. C. Physiological need, word fre-
quency, and visual duration thresholds. 7. exp. PsychoL, 1953, 46,
25-31.
55. Zeigarnik, B. Das Behalten erledigter und unerledigter Handlungen.
PsychoL Forsch., 1927, 9, 1-85.
Solomon E. Asch
SWARTHMORE COLLEGE
PERCEPTUAL
CONDITIONS
OF ASSOCIATION
The study of the formation of associations has followed a highly uni-
form pattern whose special character has escaped notice. The pre-
vailing mode of investigation possesses a one-sidedness that obscures a
fundamental question concerning the nature of associations.
The observation from which we begin is that the study of associations
has been restricted to associations between units. The terms entering
into association have been, virtually without exception, distinct and
independent units. Of the various relations that may obtain between
terms, only one has been selected for analysis; the consequence is that
our knowledge of associative processes rests almost entirely on a lim-
ited, and perhaps not representative, set of conditions. We propose
to extend the study of associations to certain hitherto neglected rela-
tions, and in particular to inquire into the role of the unit relation in
association.
There is an important theoretical reason for this partiality of in-
vestigation, which may serve to introduce the present discussion. The
Based upon S. E. Asch, J. Ceraso, and W. Heimer, "Perceptual Conditions of
Association," Psychological Monographs, Vol. 74, 1960.
188 : Solomon E. Asch
problem of association came into prominence during the modern pe-
riod as part of a more comprehensive doctrine, whose account of psy-
chological functioning started with a theory of sense data. The British
empirical philosophers, who were most responsible for this develop-
ment, described the original contents of mental life as discrete, irreduci-
ble sensations, each aroused by a given physical energy acting on a
particular sense organ. Given this starting point, a process was needed
to bind or join the elements, in order to account for the formation of
sequences in experience corresponding to the regularities of objective
events. Further, theorists in this tradition proposed to derive all psy-
chological functions other than sensory from perception to thinking
and imagination from associative combinations. Thus association
was made responsible for order in mental life. (Behaviorism adopted
these assumptions without substantial change. The conditioning of
responses was the formal counterpart to the "association of ideas.")
This starting point prescribed the identification of associations and
the interpretation of the underlying process. If the units of mental life
are discrete and irreducible, the achievement of association is to join
what is initially entirely unrelated. Thus virtually by definition, the
formation of associations early became synonymous with the associa-
tion of one unit with another discrete unit. The experimental movement
has adhered to this orientation with remarkable strictness up to the
present. Although investigation has only rarely worked with elemen-
tary stimuli, it has not deviated from the paradigm that pairs hetero-
geneous, and as far as possible, unrelated units. The paradigm has
served as the basic, general case, to which other conditions and effects
must be reduced. This is, we believe, the reason that investigation has
concentrated on "empty," rote associations, limited to the relation of
contiguity. 1
There are other relations equally deserving of study. The most chal-
lenging are those that obtain between terms belonging to the same unit.
As a rule, units are not homogeneous, but possess distinct parts or
aspects. When one perceives or recalls such a unit, one is perceiving or
recalling a definite order and relation among the parts or aspects. If so,
all questions concerning associations between units have their formal
1 The reason generally advanced for the study of rote associations is a technical
one, namely that they exclude unobservable effects of past experience. This argu-
ment conceals the decisive point. The need to control past experience restricts
the content of terms, but it does not necessarily prescribe the relations between
them.
Perceptual Conditions of Association : 189
counterpart when one turns to the internal relations of a unit. The
latter have not been the object of investigation; students have made
the customary assumption that the constitution of units is precisely the
problem of association, that units are the products of associations
between units. We considered that the similarities and differences
between these important conditions is a problem of consequence. Ac-
cordingly, we proceeded to compare the forming of connections be-
tween terms when the relations between them were systematically
varied, and in particular when they differed with respect to unity. We
asked, first, whether association follows the same course in these in-
stances, and second, whether they are the same or distinct in process.
The general procedure was that of the classical association experi-
ment. The stimuli were visual forms. The terms to be associated were,
as far as possible, constant in all experimental variations; the latter
differed mainly in the relations between the terms. All the data to be
described were obtained with one-trial learning.
THE CONSTITUTIVE RELATION
For the first step we selected stimuli bound by the constitutive relation
(see Fig. 1, column 1 ) . Each stimulus of this series is a particular visual
form whose contour is delineated by another, smaller form that is
identically repeated. We will refer to the over-all stimulus as the form,
and to the constituents of the contour as modes. In view of their
phenomenal character, we will call the percepts of this series unitary.
To anticipate the procedure briefly, the stimuli will be shown to sub-
jects who will subsequently be tested for recall of coherence between
form and mode, the latter corresponding to the a and b terms of the
standard association experiment.
For purposes of comparison we need a parallel condition corre-
sponding to the conventional association experiment. This series (see
Fig. 1, columns 2 and 3) consisted of pairs of terms a figure in con-
tinuous contour, and to the right of it a linear array of identical forms.
The left and right terms are identical in content with the forms and
modes of the constitutive or unitary series. Since the terms are clearly
heterogeneous, we will call this series non-unitary. In this manner we
constructed two sets of stimuli composed of the same contents but
differing in relation. We are concerned to compare the mastery of the
two series under the same conditions of learning and recall.
Let us briefly describe the differences and similarities between the
790 : Solomon E. Asch
series. In the non-unitary series the members of each pair are discrete
and heterogeneous; each is a perceptual unit, and they stand in the
relation of contiguity. The unitary series has the Mowing properties:
( 1 ) Form and mode are in a relation of dependence; the modes are the
constituents of the contour, (2) The spatial distribution of the modes
gives the over-all form. The modes contribute to the form by virtue of
their properties as points, not through their own form-character. (3)
Form and mode are identical in locus. Finally, the paired terms are
equally heterogeneous or dual in both series.
UNITARY
NON- UNITARY
UNITARY
NON-UNITARY
4 aaaaaaaer
FIG. L Unitary and nonunitary series.
Except where otherwise noted, the general procedure employed in
the experiments to be described was as follows. The stimuli of each
series were shown, to comparable groups of subjects, singly and in
succession. The conditions of learning and recall were identical in both
groups. Experimentation was done individually. Each member of the
series was exposed for 4 seconds. The instructions were to inspect each
stimulus (or pair) so as to be able subsequently to reproduce as many
as possible. Thus learning was intentional, and under "free recall" in-
structions. A test of recall came after 3 minutes. The subjects were men
and women undergraduate students.
Perceptual Conditions of Association : 191
Experiment 1. Following the learning procedure just described,
there came a test of free recall. The subjects were provided with blank
booklets, and were requested to reproduce each stimulus (or pair), one
to a page, as completely as they could, but to reproduce in part when
necessary. These reproductions were scored first for recall of content
alone, or of the forms and modes regardless of whether they were cor-
rectly joined. Second, we scored the frequencies of correct and incor-
rect pairings, and of recalls of one term alone. (For a detailed account
of the procedure and scoring, see reference 1, p. 6.)
The principal finding (see Table 1, experiment 1) is that the two
conditions were substantially alike in recall of content, but differed
markedly and significantly in joint recall. The unitary series produced
a higher level of accurate pairing and a lower level of inaccurate pairing
than the non-unitary series.
Experiments 2 and 3. These results were fully confirmed under con-
ditions of aided recall, when at the time of test the subject was pro-
vided with either the form (in continuous contour) or the modes (in
linear array) and requested to reproduce the other. Again recall of
content was much the same in the two series, but the unitary series was
distinctly superior in accuracy of paired recall (although the stimuli
initiating recall differed perceptibly from those of the unitary series but
were identical with those of the non-unitary series). See Table 1,
experiments 2 and 3.
Experiment 4. A further variation was done in which the forms and
modes of experiment 1 were paired randomly in a new way. The con-
ditions of learning and recall were those of experiment 1. The results
were substantially those of experiment 1 .
These effects persisted in full force when the conditions of learning
and recall were varied widely. The results were substantially the same
when the conditions of learning were incidental, and when the final
test was that of matching or of recognition. The latter findings establish
that the advantage of unity persisted when the burden of recall was re-
duced, and tell us that the experimental effect refers to the formation
of functional connections.
THE FIGURE-GROUND RELATION
Experiment 5. We proceeded next to the study of another stimulus
relation between terms that of figure and ground. This fundamental
CO
I
I
u
I
2 ^
f~* r*j
g
s:
B^
a
r
I
j-j +S
ON ^J* o ON
ON
ON
^ CO O 00
t-4 ON r-H CO*
CO O O CO
OJ r*; IO rH
ON O od od
Tfr rfr ^f CO
CN co ON r>
o *o" **f ON
vq
<<t
ON
ON
CO
C<J "^ 00
ci 10* co"
C* 01 rH
oo vo 10
i> Tf to
r^ c- vo
CM -* r-J
VO
10
o
CN
CS
CO
CO
CO
to
CO
o"
CO
vo
cs
s
CN
o
vo
o\
vq
ON
VD
CO
(S
Relation of
Inclusion:
Experiment 6
03 .2 r^
4-* ^ J-i
Perceptual Conditions of Association : 193
perceptual relation has not been brought into connection with the
associative problem, mainly because most students have divorced the
issues in the two regions.
FIG. 2. A sample stimulus for the figure-ground condition.
Figure 2 illustrates the relation. Each member of the series consisted
of a form in continuous contour, located on a ground of smaller forms
identical with the modes of the earlier series. The modes did not overlap
with the contour. All other conditions of learning and recall were those
of experiment 1.
Recall of content was somewhat but not significantly lower than in
experiment 1, but coherence of recall was significantly poorer than in
the unitary series, and insignificantly different from the non-unitary
series (see Table 1, experiment 5). The result is in accordance with
general evidence that the figure-ground relation is a segregating con-
dition.
Considered as a perceptual situation, the traditional association
experiment (here represented by the non-unitary series of experiment
1 ) consists of paired heterogeneous figures on a single ground. The
associative similarity between the latter and the figure-ground condi-
tion indicates that their segregating properties are approximately equal.
We consider this an illustration of the relevance of perceptual data to
issues in the psychology of association.
THE RELATION OF INCLUSION
Experiment 6. A further relation we studied was that of inclusion.
The series consisted of forms drawn in continuous contour, the inner
area being filled with the previously paired modes (see Fig. 3). Again
194 : Solomon E. Asch
the modes were drawn so as not to touch the contour. Learning was
intentional and the test that of free recall.
FIG. 3. Series for the included condition.
Recall of content was similar to the preceding conditions, but recall
of the connections between the terms was almost exactly intermediate
between the levels of the unitary and non-unitary series. Further exami-
nation showed, however, that this was not an intermediate condition.
Despite the standard procedure followed in constructing it, the series
was not uniform: some stimuli produced high levels of pairing, in con-
trast to others that were far more resistant to pairing. Inspection sug-
gested that the inner forms of some stimuli cohered with the contour to
establish a surface, but that the inner forms of other stimuli remained
simply an aggregation of small units enclosed by a contour line. To
check on this possibility, six judges who had no knowledge of the prob-
lem were asked to rank the stimuli in terms of "how well" the contour
and the included material went together. There was substantial agree-
ment among the judges; on the basis of the mean rankings the series
Perceptual Conditions of Association : 195
was divided into two halves, which we will call "unitary" and "non-
unitary."
The two categories of stimuli produced strikingly different results:
those judged "unitary" were far superior in pairing, and also in absolute
recall. Further, accuracy of pairing of the "unitary" subset was similar
to that of the unitary series of experiment 1, while the "non-unitary"
subset resembled the non-unitary series of experiment 1 (see Table 1,
experiments 1 and 6).
It is instructive to apply this analysis to the figure-ground series,
since the latter too contained forms that enclosed other forms. The
results fail to reveal the slightest trace of a difference between the sub-
sets, when the division is based on the judgments obtained in the in-
clusion series. When the modes extended beyond the contour into the
surrounding area, they lost their specific relation to the contour. The
figure-ground relation functioned throughout as a segregating condi-
tion, unlike the relation of inclusion.
Marked and consistent differences of coherence were found when
the perceptual relations between terms were varied. The relations in
question are describable geometrical properties. In addition, the results
grouped the several sets of relations in two categories, which differed
with respect to unity. The sources of the property of unity are diverse,
but the effects upon coherence were quite stable. Unitary percepts were
consistently superior in coherence; in this investigation unity was repre-
sented by the constitutive relation, and by the relation of inclusion
when it permitted surface formation. Non-unitary percepts were con-
siderably more resistant to joining; they were represented here by the
relation of contiguity between heterogeneous terms, the relation of
figure and ground, and the relation of inclusion when it did not permit
surface formation. The criterion of unity is phenomenal; it is a reliable
criterion and one obtained independently of the results. The conclusion
we reach is that coherence between terms is a function of the unit rela-
tion. This conclusion is based on data from only one modality, the
visual, and on a few properties within that modality; it will therefore
need to be verified over a wider range. It seems justified, however, to
propose that if we can specify the relative unity of two percepts, we can
predict their relative coherence.
196 : Solomon E. Asch
ALTERNATIVE INTERPRETATIONS
Before one can consider this conclusion as secure, it is necessary to
examine alternative interpretations and certain possible sources of
error.
The rationale of this study required that the several experimental
conditions should be identical except for the property that was varied.
This was the reason for the attempt to maintain the constancy of terms
while varying the relation between them. But a change of relation might
have introduced unintended modifications that could be responsible
for the observed effects.
1. The preceding comparisons require that the respective stimulus
conditions should be equally perceptible or discriminate, or that
differences in this respect should not favor the conditions that are
superior in coherence. Accordingly, we repeated experiment 1 (with
the unitary and non-unitary series, respectively) without change of
conditions except for the final test. The latter called for the recognition
of the individual forms and modes, when these were interspersed with
others that had not been part of the learning series. Thus we took
recognition as an index of perceptibility. The forms of both series were
equally recognizable, but the non-unitary series was superior in recog-
nition of modes. Since the non-unitary series was far poorer in coher-
ence, the main finding cannot be attributed to recognition differences.
2. The non-unitary series may have been more homogeneous and
therefore more susceptible to intraserial interference than the appar-
ently more differentiated unitary series. To check on this possibility
experiment 1 was again repeated, but was now preceded by familiariza-
tion with the individual forms and modes. We reasoned that familiari-
zation should reduce and equalize subsequent internal interference in
the two series, and thus reduce the experimental effect. Although the
procedure of familiarization was highly effective, the difference of co-
herence was not reduced; indeed, it became more substantial.
3. We need also ask whether the unitary and non-unitary conditions
differed in amount of information to be mastered. The relevant answer
is that the respective stimulus conditions were remarkably similar in
recall of content, differing only in coherence. The procedure of fa-
miliarization described above points to the same conclusion.
4. The relations here studied differed with respect to the spatial
proximity between terms; examination shows that proximity cannot
Perceptual Conditions of Association : 197
account for the findings. Identical relations of proximity (in experiment
6) produced the most different effects, while conditions differing in
proximity (experiments 1 and 5) were quite similar associatively. In
this investigation proximity was effective only as it mediated the prop-
erty of unity.
These alternatives, and certain others not here discussed, fail to ac-
count for the findings. We conclude that coherence is dependent on
unity, not on conditions only secondarily dependent on unity.
PERCEPTION AND ASSOCIATION
Traditionally there has existed a sharp separation between perceptual
and associative investigation. When the principal function assigned to
association was the joining of unrelated contents, it followed that the
observation of data was one kind of event, and their association an-
other. This was the reason that an association was initially conceptual-
ized as the establishment of a pathway between excitations, which was
independent of the contents joined. This divorce has persisted sub-
stantially to the present day. Students do not question that perceptual
conditions decide what will be observed and associated, but they insist
that the associative operations have their own exclusive character. This
is the cornerstone of what we may call the associationistic theory of
associations: it denies a direct effect of perceptual conditions on
association.
It is precisely this account that Gestalt theory challenges. The classic
formulation of Kohler (1929, 1941) proposes that association is the
aftereffect of organization in immediate experience, not an event
added to it. Traces preserve the organization of earlier events; and the
principles pertaining to each are intimately related, often the same.
The principal consequence of this position is that conditions favoring
organization in experience favor trace organization, and therefore
coherence of recall.
This investigation provides convincing evidence that perceptual re-
lations directly determine the formation of associations. It thus sup-
ports the theory of Kohler and raises a question about a postulate of
associationistic theory. At the same time, neither of the principal
theories of association have concerned themselves closely with the par-
ticular problem under discussion, with the difference between coher-
ence within a unit and the coherence of one unit with another. A con-
cern with this problem raises a new question to which we now turn.
198 : Solomon E. Asch
UNITARY AND NON-UNITARY COHERENCE
It would be entirely consistent with the evidence here reported to con-
clude that there is one process of coherence, or that the unitary and
non-unitary conditions differ in associativeness, but the conclusion
does not follow with necessity. To be sure, the obtained differences
were throughout quantitative, and this may appear to support a con-
tinuum interpretation; but the argument is not logically decisive, in
addition to which the presence of quantitative differences alone may
reflect limitations of procedure. We ask: Is the coherence established
within units and between units referable to a single process, or are they
distinct in process?
An answer cannot be found within the frame of current thinking and
evidence; to put the question is to realize how pragmatic is the pre-
vailing characterization of an association. We need to ask in what
respects processes of coherence might differ. The following comments
are necessarily tentative.
One hint toward a solution comes from the following finding re-
ported in this investigation. The occurrence of associative errors or
the mispairing of terms was remarkably low in unitary series (see
Table 1 ) . The relevant data are those that do not force choices upon
subjects; therefore free recall results are the most pertinent. Under free
recall test conditions the majority of subjects did not mispair at all,
and there were only rare instances of others mispairing more than once.
(There are also reasons for questioning whether the latter were au-
thentic associative errors; see reference 1, pp. 37-39.) A further find-
ing goes in the same direction. When subjects separately reproduced,
in free recall, the two terms belonging to a given unitary stimulus,
eventually they almost always paired them correctly; this was definitely
not the case with non-unitary terms.
These observations raise an unexpected question: Do unitary per-
cepts produce invariable coherence? Or, do the terms of a unit persist
and disappear jointly? If this were the case, we would have a more
thoroughgoing difference between stimulus conditions than those pre-
viously described (although one that is still logically compatible with
a single-process interpretation). The question is also of interest aside
from its theoretical implications. It is of consequence to know whether
invariable coherence is a fact, and if so, what the responsible stimulus
conditions are. (It should be noted that invariable coherence is only
Perceptual Conditions of Association : 199
superficially similar to the older idea of "indissoluble" associations,
which was based on the operation of frequency.)
The question cannot be answered at this time. (For a discussion of
the evidence necessary for a decision, see reference 1, p. 37.) But it
suggests a more concrete way of characterizing the difference between
unitary and non-unitary (or between non-associative and associative)
coherence. Briefly, the recall of an association between heterogeneous
units often requires two steps: a step of recognition (of one of the
terms), followed by associative recall proper, or the transition to the
other term. The recall of a unit may collapse these steps into one; that
is, recognition of one of the terms may produce an unfailing transition
to the other term.
The traditional study of associations has limited observation to the
coherence established between units, neglecting relations internal to
units. Employing the technique of the association experiment, we
varied systematically the perceptual relations between the parts of a
stimulus distribution. The relations studied were those of constitution
or composition, of figure and ground, of inclusion, and of contiguity
between heterogeneous units. The relations were either phenomenally
unitary or non-unitary.
The capacity to recall jointly the parts or aspects of a stimulus dis-
tribution was markedly and significantly superior when these belonged
to the same phenomenal unit Errors were regularly more frequent
when the terms belonged to distinct units. These findings persisted as
the conditions of learning and recall varied widely. The evidence is
convincing that the effect occurs at the time of primary experience.
We conclude that there is a direct relation between perceptual unity
and coherence.
The question was considered whether unitary and non-unitary co-
herence differs in process, or whether we may have to distinguish be-
tween associative coherence and a coherence that is non-associative.
This investigation, which demonstrates a direct effect of organiza-
tional conditions on coherence, questions the traditional divorce be-
tween the regions of association and perception. This separation has
rested on a doctrine of psychological elements that preceded the study
of perception, and that has become untenable in the light of Gestalt
investigations. In this sense the present inquiry provides convincing evi-
dence for the theory of associations formulated by Kohler. It also
demonstrates that to take the relations between terms into account in
the study of associations is both necessary and fruitful. The procedure
200 : Solomon E. Asch
here adopted permits an independent and objective characterization of
stimulus properties, to which data at the level of recall can be sys-
tematically related. Indeed, it is perhaps not too much to hope that
perceptual data may be a source of solutions to some questions in the
theory of association.
The present investigation raises a host of questions that require
further study, ( 1 ) The range of relations here considered is far from
inclusive. In particular, they are distinct from the grouping principles
of Wertheimer, nor do they include the part-whole relation. 2 (2) The
psychological constitution of objects may be a function of their unit
character. Traditionally the formation of an object has been considered
as the product of associations between sense data that were distinct
units. We need to ask what the consequences are of the fact that the
data in question refer to the same object. (3) The issues here con-
sidered cannot be restricted to perceptual organization; their relevance
may extend to organizational processes generally* To give one illustra-
tion, the relation of meaning may be an instance of conceptual unit
formation; the contribution of meaning, which has been frequently
confirmed, may be a function of the unit relation. 3
REFERENCES
1. Asch, S. E,, Ceraso, J., and Heimer, W. Perceptual conditions of associa-
tion. Psychol. Monogr., 1960, 74, No. 3.
2. Kaswan, L Association of nonsense-figures as a function of fittingness
and intention to learn, Amer, J. Psychol, 1957, 70, 447-450.
3- Kohler, W. Gestalt psychology. New York: Liveright, 1929.
4. Kohler, W. On the nature of association. Proc. Amer. phil. Soc., 1941,
84, 489-502.
5, Prentice, W, C. H., and Asch, S. E. Paired association with related and
unrelated pairs of nonsense-figures. Amer. J. Psychol., 1958, 71, 247-
254.
2 A pronounced facilitating effect of relations of fittingaess on coherence in
recall was demonstrated by Prentice and Asch (1958); see also Kaswan (1957).
A recent investigation (unpublished) of Asch and Ceraso has demonstrated a
similar effect of perceptual unity when it is based on the part-whole relation.
3 For a recent investigation of the role of meaning in association, see W. Ep-
stein, I. Rock, and C. B. Zuckerman, Meaning and familiarity in associative
learning, Psychological Monographs, 1960, Vol. 74, No. 491.
PART IV
SOCIAL PSYCHOLOGY
AND MOTIVATION
Wolfgang Kohler
DARTMOUTH COLLEGE
PSYCHOLOGICAL REMARKS
ON SOME QUESTIONS
OF ANTHROPOLOGY
In his investigation of primitive mentality Levy-Bruhl is led to the con-
clusion that in primitive man mental operations are essentially different
from those with which we are familiar. "Les representations collec-
tives," all the notions with which, during childhood and adolescence,
an individual is imbued by the surrounding group, are "mystiques" in
the case of primitive tribes. Where we apply the laws of logic "la men-
talite prelogique" applies "la loi de participation." The author seems
to conclude that in many cases the anthropologist's attempt really to
understand and to explain primitive behavior will necessarily end in
failure. We even have difficulties in defining the term "mystique."
Again, who can formulate clearly what is implied in the "loi de par-
ticipation"? Still, when we find that one more primitive belief or custom
is mystical and that it follows the law of participation, this is, according
to the French sociologist, perhaps all that we can do about the case
in question.
Often, I am afraid, Levy-BruhTs skepticism is justified in practice,
Reprinted with permission from the American Journal of Psychology, Vol.
50 (Golden Jubilee Volume), 1937.
204 : Wolfgang Kohler
though perhaps not for the reasons which he indicates. The origin of
some tribal beliefs and institutions may be so complicated historically
that there simply is no unitary sense in the phenomena themselves. To
that extent, of course, we cannot fully "understand" their nature,
whether or not primitive mentality differs essentially from our own.
Besides, several primitive cultures are known only from reports in
which the white observer's preconceived notions seem to be hope-
lessly intermingled with the ethnological material as such. Not a few
of these cultures disintegrate quickly, if they have not actually disap-
peared. Since the opportunity for more objective investigation is gone,
in such cases the available data themselves will remain ambiguous.
Much of their obscurity may be due to serious misunderstandings rather
than to an alleged impermeable nature of primitive mentality.
Though some anthropologists do not like to admit it, psychological
principles play an important role in the interpretation of anthropologi-
cal facts. Theoretical difficulties may, therefore, arise quite as easily
from inadequate psychological notions as from the strange ways of
primitive mentality. As Levy-Bruhl has pointed out, the psychology
which Tylor and his school applied to anthropological problems can
hardly be regarded as an adequate tool in this field. In the meantime,
however, psychology has again changed a great deal. Perhaps some of
the facts in anthropology which, to the French author, seem to be
outside the white man's understanding will prove more approachable
if we apply to them recently developed psychological ideas. In the
following pages a modest and hesitant attempt will be made in this
direction. It must be hesitant because, for the two reasons given in the
last paragraph, we should not try to find more psychological sense in
our material than it can be expected really to contain. Our attempt
must be modest since no outsider can hope to have a sufficient survey
of anthropological data for a more comprehensive interpretation of
primitive cultures. It ought to be modest in another sense, too, because
some of the ideas which I shall bring forward have, at least implicitly,
been used before by experts. I should be pleased, however, if psycholo-
gists would give some attention to the following discussion. lit our de-
sire to be thoroughly exact, we are in the present phase of psychology
apt to concentrate our efforts on a few special problems. In conse-
quence of this we begin to lose sight of most questions for which the
social sciences expect answers from psychology. If we do not answer
because our outlook is too narrow, discontent, of which there are some
signs already, will doubtless grow steadily. It may be that the following
Psychological Remarks on Some Questions of Anthropology : 205
discussion will satisfy neither psychologists nor anthropologists. In this
case it might still have some value, if it induces other psychologists to
do better work in the same field.
Many examples may be given in which the behavior of primitive
man certainly looks incomprehensible at first. May I select two at
random?
Moffat reports from South Africa that once a woman came to him
in order to get the medicine which he had prepared for her sick hus-
band. When he made it clear to her that half of the potion should be
taken at once and the rest several hours later, she asked whether it
would not be possible to drink all of it at once. Upon his assent the
woman suddenly swallowed the medicine herself. To the missionary's
protest she answered by asking whether it would not cure her husband
if she drank the medicine. 1 Many Africans, it is true, like to take our
drugs because of their "power." It would, however, hardly be fair to
assume that this factor suffices as an explanation. Anthropology knows
of too many cases in which what we might call medical treatment or
measures of hygiene apply to the family group almost as much as to
an individual member upon whom our physicians would concentrate
their efforts. Neither is it satisfactory to say that among primitives the
individual is less a separate entity than in our society; or that, for them,
the individual's boundaries are not the same as for us. Both statements
may be true. But in such matters our own point of view seems to us so
obvious and necessary that we fail to understand on what basis other
views could ever develop.
I have the following report from a missionary who is working in East
Africa. While sleeping in his hut a man dreams that he is in another
village and that there he commits a crime which, according to the con-
victions of the tribe, deserves death punishment. Frightened, he tells his
people about the dream. Unfortunately something of the kind really
happens in that other village and approximately at the same time. The
man is indicted, he confesses and is killed, although he was miles away
from that other village when the crime was committed. Such things
will occur less often the more European officials and European schools
begin to influence the natives, but my informant, a perfectly calm and
sober man, told me that he has direct knowledge of more such cases.
Primitive mentality does move in strange ways; it does not seem to mind
1 R. Moffat, Missionary labours and scenes in Southern Africa. London:
L Snow, 1842, pp. 59 If.
206 : Wolfgang Kohler
contradiction. According to our thinking the man could not be in an-
other village while he was sleeping in his hut. This is exactly, however,
what the natives believe and what the poor victim of such logic ap-
parently believes himself.
I have gradually become convinced that in many such cases our
difficulties are due to a special coordinate system of our own no less
than to peculiarities of primitive mentality. In a way scientifically
trained people of our time may be particularly unfit for the understand-
ing of less intellectual cultures. What are the objective facts of nature?
What is the best access to objective knowledge in this sense? What in-
fluences on the other hand are apt to hamper our progress in this field?
Since the seventeenth century such questions have gradually intro-
duced a definite set of values which is now so dominant that far beyond
the circle of scientists proper the outlook of civilized people is thor-
oughly governed by these particular ideals. A sober attitude towards
a real world is instilled in children of our civilization by the words and
actions of their parents. Long ago the most basic convictions of sci-
entific culture lost the character of theoretically formulated sentences.
Gradually they have become aspects of the world as we perceive it; the
world looks today what our forefathers learned to say about it; we act
and we speak accordingly. In this form the consequences of a few cen-
turies of science are present in the remotest corners of the civilized
world.
The ideals which I mentioned, the special interest which they repre-
sent, and the particular aspect which they have given to the world, are
doubtless very apt to further quick progress of natural science and
practical behavior in contact with reality. I wonder whether to some
degree they do not disturb the advance of anthropology. Physicists
have gradually separated their real world from the realm of mere per-
cepts. In doing this they have also learned to distinguish between phe-
nomena which tell the truth about the real world and others which are
illusory. On this basis percepts and their traits are good or valueless
according to this distinction. Similarly, properties of percepts which
are easily understood in their relation to physical stimulation will be
readily accepted in sense physiology. Whatever does not fit into this
scheme will be neglected and despised because of its "subjective" na-
ture. Under the influence of scientific culture the very appearance of the
world around us is being transformed. It will not only be "intellectu-
alized." Such characteristics of the world will become less conspicuous,
they will gradually recede into the background, for which the physicist
Psychological Remarks on Some Questions of Anthropology : 207
and the sense physiologist have definitely no use. Those aspects on the
other hand are made outstanding which we regard as representatives
of reality.
For natural science, I repeat, this development may mean strict prog-
ress. I shall not deny that the modern aspect of the world around us
agrees better with our picture of physical reality than older views did.
But I insist upon the fact that before the era of scientific culture the
perceptual world could not have looked as it appears to us now. Psy-
chologically speaking our view of the "given" world is the product of
complicated historical processes. Judged from a merely psychological
standpoint and without any reference to physical reality, this aspect of
the world is therefore no more genuine than any previous appearance
of man's perceived environment. To some extent even more "natural"
traits of perception will be found where science has not exerted its
modifying influence. They would be most prominent in the primitive's
world.
Their study, however, is difficult. The psychologist no less than other
people has learned to see the world through the eyeglasses of natural
science. What he beholds in this manner seems to him so obvious, he
is so little aware of his eyeglasses, that any other view will appear to
him as a distortion of the genuine aspect of the world. If primitives have
such another view, their mystical thinking will be made responsible
for it. They have changed the genuine appearance of the world which
is, of course, originally the same everywhere and identical with our own
view.
It seems to me that the opposite judgment would be at least as cor-
rect. Primitive perception, to be sure, contains a great deal of intel-
lectual elaboration. It is improbable, however, that such influences
should ever have been more systematic and intense than have been the
effects of scientific learning in our own case. It would therefore be of
great advantage if, in studying the appearance which the world has for
primitives, we could overcome the tendency to use our own view as a
norm. Anthropologists have generally been at pains to follow this
principle. More might be done, however, if even in the application of
psychological thought to primitive perception the white man's custom-
ary outlook were recognized as a disturbing factor. Recently we learned
that a definite type of scientific analysis had nearly made us unable to
apprehend most common sides of perception and of mental life in gen-
eral. We shall try to apply this lesson to anthropological questions.
Almost 50 years ago von Ehrenfels pointed out that percepts have
208 : Wolfgang Kohler
properties which cannot be derived from so-called sensations. A sensa^
tion has a good physiological standing. It is the phenomenal correlate
of a well-defined stimulus. The properties of percepts which von Ehren-
fels had in mind seem, however, to occur only where groups of stimuli
cooperate in causing one phenomenal datum. The stimuli in question
may be spatially or temporally distributed. Correspondingly, von Eh-
renfels' qualities are the attributes of percepts which extend through
areas in space or through stretches of time. They are not popular in
sense physiology, which likes to study the reactions of single receptor
cells to single stimuli. Properties which are characteristic of larger
entities and only of these will evidently not fit into this scheme. A
well-known example is the "major" or the "minor" character of a
melody or of a chord. No single tone has minor character, whereas it
is an essential trait of many tunes and chords. Although quite familiar
to musicians, among psychologists such traits of auditory percepts had
never before found the attention which they deserve. After what has
been said above we need not discuss the reasons for such aloofness.
It would not be difficult to show that all percepts have such qualities.
When things are called "tall" or "bulky," persons "slender" or "stout,"
movements "clumsy" or "graceful," reference is made to definite von
Elirenfels qualities. When we describe events as "sudden" or "smooth,"
"jerky" or "continuous," we refer to the same class. Esthetically they
are doubtless of paramount importance, but they are hardly less so, I
think, in biological contexts. The color of the hair, the pitch of the
voice may have much to do with the strong impression which specific
persons of one sex make upon specific persons of the other. Still, prop-
erties of shape, of gesture, and of general movement are generally
found to be at least as dangerous and these are von Ehrenfels qual-
ities.
In a certain sense such qualities of percepts are "subjective." As we
perceive them they are functions of nervous processes; but this is true
of all properties of all percepts. They seem more "subjective" than
"green" or "blue" because no simple physical stimuli are responsible
for their occurrence. It may be doubted whether, despite this serious
disadvantage, they are not often particularly important as witnesses of
objective reality. 2 Beyond any doubt they appear quite as much as
properties of the world around us as, for instance, colors and smells.
Phenomenally, they are objective.
2 W. Kohler, Gestalt psychology. New York: Liveright, 1929, pp. 174 ff.
Psychological Remarks on Some Questions of Anthropology : 209
If I am present when two cars crash at a crossing, and if then I speak
about the "vehement" or "violent" impact of one upon the other, one
might say that in this description I attribute to what I see traits which
are borrowed from my own muscular or other subjective experiences.
My answer would be that this is not the point which we are discussing
at present. Whether the explanation by "empathy" is right or wrong,
we have to distinguish between the functional genesis of phenomenal
characters and their phenomenal appearance and localization as such.
Supposing even that von Ehrenfels qualities like "vehement" have a
subjective origin, from a descriptive point of view they are therefore
no less objectively localized, are no less aspects of events themselves.
Clumsy dancing looks clumsy there, outside, in a man or woman whom
I see moving before me. It is again Fred Astaire, not I, who performs
unheard-of von Ehrenfels qualities in dancing; it is on the screen that
I perceive them, not in my muscles. If genetically I should be respon-
sible for all this, I am at least in simple experience unaware of my
authorship. After all, it is a theory, not an observed fact, that such
traits of percepts are always due to subjective sources. For unsophisti-
cated perception vast numbers of von Ehrenfels qualities appear pre-
cisely where colors and other attributes appear with which customary
thinking is better acquainted: in and on things, persons and events. It
follows that the less sophisticated people are, the more will they accept
such properties of percepts at their face value.
Primitives are, of course, naive realists. They cannot possibly dis-
tinguish between "mere percepts" and independent physical realities.
A thing-percept is for them the very prototype of reality, although
they believe in further realities besides. In fact, percepts are for them
as for children so real that they would hesitate to accept the name,
if they could clearly understand its meaning. Why should they call a
tree a "percept"? There simply is a tree before them to which, it is true,
more or less attention may be given. That this tree which we call a
percept remains unaltered in its place even when nobody is present who
might have the percept, this is, for a naive view of the world, the most
natural thing. When before us, the tree certainly shows no striking
symptoms of depending upon any subjects for its existence. One has to
be far ahead in physics, physiology, or philosophy before anything like
a problem can be discovered in this connection.
It is again well known that without such sophistication no doubt will
ever arise about the independent reality of all the sensory qualities
which the tree or any other object may possess. We have seen, however,
210 : Wolfgang Kohler
that the von Ehrenf els qualities of percepts are localized in and on these
percepts. Primitives have no physics, no physiology, and no epistemol-
ogy which could throw any shadow of suspicion upon these qualities.
Consequently their naive realism will apply to such characteristics of
the world as it applies to any others. For them, things and events will
have these properties independently and objectively. Thus their en-
vironment will have many aspects to which they attribute full objective
value, while our scientific civilization denies such value and accordingly
tends to impoverish our perception.
Many of these qualities are distinctly dynamic. What is "sudden"
may, with increasing intensity, become "startling," "frightening" and
"threatening." Such terms imply the subject's emotional reaction. But
they also imply that in certain events or things there is a perceptual
basis for the reaction. Here, as so often, much phenomenological truth
is contained in language.
Not always is the "threatening" a high degree of the "sudden." While
climbing once in the Alps I beheld, on stepping cautiously around a
corner of the rocks, a big dark cloud which moved slowly and silently
towards me along the slope. Nothing could look more sinister and more
threatening. Genetically this might have been a case of empathy; but
for my awareness the menace was certainly in the cloud. I could perhaps
persuade myself that a cloud as such is an indifferent percept. If, how-
ever, I had been a primitive, no reason whatsoever could have given me
such sober consolation. The threatening character of the cloud itself
would have remained just as "objective" as its ugly dark color. Simi-
larly the ocean itself is "wild" in a gale; and a mountain which appears
high above the other tops in strange illumination is itself "majestic" or
"forbidding." Primitives have no science of physics in our sense of the
word, but those opinions about the world which they develop instead
will necessarily differ widely from our own views. If the expression is
admissible, it must be a curious "physics" in which the menacing char-
acter of a cloud, the wildness of a gale, the majesty of a mountain per-
cept have at least the same rank and objectivity as, with us, measured
distances, velocities, and weights. It is in the nature of such "physics"
that it can never be a calm and critical survey of facts as our science
likes to be. The primitive's reality is apt to be too exciting for sober ap-
preciation, since not only its effect on the subject, but also its own ap-
pearance is so often emotional. Awe and respect are more likely to
characterize the primitive's attitude in such matters than is a spirit of
Psychological Remarks on Some Questions of Anthropology : 211
critical investigation. Whatever else may follow, from this starting point
ways may even lead to religion.
In passing, a remark should be made on an old philosophical con-
troversy. David Hume denies that any of our "impressions" contain
the quality of power or force. His opinion can only be held so long as,
in a quiet environment, we discuss phenomenological questions in
abstracto. At his writing desk the philosopher is apt to see the world as
static and tranquil. It is, however, the sailor in a small ocean-going ves-
sel, the airplane pilot, the tramp left to his own resources in any weather,
who can tell us best whether there are any experiences of power, force,
or dynamic traits in general. For us it is, of course, another question
whether such dynamic traits of experience have any physical counter-
parts; but on the level of naive realism this question has no sense. Here
the objective world exhibits forces directly and not at all infrequently.
Not all dynamic qualities of things are startling. There is "heaviness"
in a middle-sized stone, "hardness" in the same object, "elasticity"
not as a property defined by the physicists but as a perceived quality
in a young tree which my arm bends. Everyday physics of the layman
deals with such properties of things as he is directly aware of them. This
is true in Central Africa, but also in civilized countries, whenever peo-
ple do not actually use the more sophisticated concepts of physical sci-
ence which is of course only done by a few. Regularly and with many
objects this procedure is quite safe, because in many respects such
simple von Ehrenfels qualities correspond f airly well with certain con-
cepts of physics. So long as this is the case nobody objects and nobody
would speak of mystical ideas.
Now and again, however, the situation acquires a slightly different
character; namely, where the dynamic side of things and events grows
beyond our feeling of control. Instead of a young sapling a primitive
may have bent an older and stronger tree. If he had any physics, he
would look upon the situation with some knowledge of objective dan-
ger, otherwise with a sober and calm mind but he has no physics in
our meaning of the word. Instead, there is now in the tree a perceived
power beyond the limits of what the man is able to handle safely. He
will respect such power. It may be actually frightening when suddenly
the tall stem breaks its bonds and jumps fiercely up into space. There
is an awe-inspiring quality in the thing, just as there is in toe sudden
rush of high wind which often precedes a thunderstorm, or in the light-
ning which accompanies it.
212 : Wolfgang Kohler
Even when enhanced beyond the common and concentrated to fear-
ful intensity, such characteristics will for naive realism still remain ob-
jective properties of things and events themselves. There is no reason
why primitive man should not behave correspondingly. From the more
trivial power qualities a nearly continuous series will extend for him
through more striking cases to those where he is overwhelmed by fear.
It is only natural that in his environment nothing should appear to him
more important and impressive than unusual power in this sense, par-
ticularly since he knows so little of its rules and whims.
Here, it seems to me, we have the perceptual foundation of what
some Melanesians call "mana," the Crow Indians of North America
"maxpe," and some tribes in West Africa "njomm." 3 Mana at least is
a notion in which the thinking of widely differing peoples seems to
agree. When Codrington first mentioned it, it appeared as a rather
enigmatic term. If the phenomenological side of psychology had then
been further developed the notion would have seemed less puzzling.
In earlier phases of anthropology, before mana became better known,
the notion was probably often misunderstood to mean a deity, a spirit,
or a soul in things. Though it seems possible that on a certain level the
concept of a soul develops as a special case of the general class mana,
this term and its synonyms cannot originally and generally have had
the special meaning of our word "soul." Tribes in many parts of the
world find mana in certain things and events; but among these tribes
there are some in which not even the human person himself has a soul
comparable in meaning with the soul of Christian doctrine or Descartes'
philosophy. 4 Few factors in our civilization make the anthropologist's
work so difficult as our popularized mind-body dualism.
I should also hesitate to interpret mana as though it meant "super-
natural." Where a thing, a person or an event has mana it remains for
primitive man on the same general level of existence as other parts of
his environment, although it is more conspicuous by its peculiar ap-
pearance or inherent power. Using our word "supernatural" in this
connection we are apt to lose sight of those obvious phenomenal facts
upon which, if I am right, the notion of mana is ultimately founded. To
some extent all things may have their mana. Innocent enough as simple
3 R. H. Lowie, Primitive religion. New York: Boni and Liveright, 1924, pp.
4 D. Westermann, Die Kpelle: Bin Negerstamm in Liberia. Gottingen; Vanden-
hoeck, 1921, pp. 174 Iff.
Psychological Remarks on Some Questions of Anthropology : 213
heaviness of a stone, it may still be the same in principle where its
intensity or particular behavior arouses respect and fear.
I mentioned above that in ordinary life our own treatment of many
things refers to their power qualities as we perceive them naively. This
attitude, however, is rather strictly limited by the scientific atmosphere
in which we live. We have learned that most mana is merely a "subjec-
tive" phenomenon. No such restraint disturbs the naive realism of prim-
itive man. He remains consistent in handling things according to their
inherent dynamic qualities. Consequently some of his activities will
seem to us quite natural and sober, others altogether strange because
they are contrary to the lessons of science. Here we begin to speak about
mystical thinking and about magic. Westennann, however, denies that
for the Kpelle themselves there is any sharp dividing line between
"normal" activities and everyday magic. This may not apply to the
Indians of North America; but among Africanists the same opinion
seems to win more and more adherents. Take the special case of med-
ical practices. "The word musamo . . . connotes . . . not only var-
ious medicinal remedies proper, but also, and much more, many things
whose power we should call magical. The difficulty is to separate the
two. From the native point of view there is no difference." 5 When the
African takes a hearty meal and thereby "fills himself with power," we
are not surprised. If he is convinced, however, that by eating the meat
of a courageous animal he will himself acquire more courage, we speak
about superstition. As a matter of fact, we have no evidence that for
the native there is any essential difference between these cases.
With this example, it is true, we have touched upon a new aspect of
power qualities. Even where for us their functional subjectivity seems
altogether obvious, people without any science will still remain naive
realists. For an object to have mana it is by no means necessary that its
peculiar characteristic appear under all circumstances. The new thing,
an event which does not usually happen, is for this extraneous reason
apt to be mana. Certainly, the unexpected looks different; it has its
peculiar appearance even for us; surprise, as other emotions, has its
phenomenally objective counterpart. But we subtract easily from "the
real facts" what, for our critical attitude, reveals its subjective genesis
so easily. Not so the native. Objects may have for him the very highest
degrees of mana which are for us as trivial as possible. He does not yet
5 E. W. Smith and A. M. Dale, The Ila-speaking peoples of Northern Rhodesia.
London: MacmiUan, 1, 1920, p. 222.
214 : Wolfgang Kohler
realize clearly that the "striking" in the new thing is due to its newness
only. Strangeness is a property of the thing itself; awe is the conse-
quence, and action will be taken corespondingly.
On the other hand, if an object has once exhibited power, such ex-
perience has an aftereffect. It will look its power or have the flavor of
it, even under conditions where, without such previous experience,
nobody would find its mana aspect very impressive. If in our own psy-
chology we had no examples of "sign Gestalt," as Tolman calls it, we
could hardly understand much primitive behavior. It is this principle
which imbues thousands of objects with power which for the neutral
observer may seem quite incapable of having any particular mana.
When the courageous animal has been killed, its meat certainly does
not look courageous to those who do not know from what it has been
taken. But as the living animal will be regarded as courageous even
at times when it is peacefully at rest, so its meat will still contain the
valuable power. We add skeptically: for those whose apprehension of
the object is distorted under the influence of previous experience. Yet
the natives are not so sophisticated. So long as naive realism is quite
undisturbed the effect of previous experience prevails, while little at-
tention is given to its psychological history. After all, it can happen here.
The place where yesterday a cruel murder was committed would have
today a strange look for most of us, even if no visible trace of the crime
were left; many would feel uneasy in such a locality, and few would
choose it as their living place.
For the native, mana qualities not only exist in his environment; they
also have definite forms of behavior. At least one of these is easily
understood, if only we begin with examples where our own judgment
agrees with that of primitive man. What is a smell? Very few people,
those who have studied this special part of sense physiology, can give a
clear answer. On the level of naive realism, on which to some degree we
all remain in everyday life, a smell is a quality that tends to spread from
one object and one place to others, as though it had the properties of
an intangible light liquid. We do not put kerosene near butter or bread,
lest the kerosene smell penetrate into the food. With onions the same
precaution has to be taken, although here it is another quality which
spreads to surrounding objects.
Smell is by no means the only entity that shows this behavior. From
a fire the quality of heat migrates in all directions. The rain, the fog,
and the thunderstorm spread from one place to the other. Most impor-
tant of all, many a sickness creeps through the community from hut to
Psychological Remarks on Some Questions of Anthropology : 215
hut; and an uncanny change which has just come over one man is soon
discernible in others. Qualities of all kinds seem thus to travel through
space, even though we cannot trace their ways in every detail. Should
we object to this description of facts? One hundred years ago heat was
an unweighable stuff even in orthodox physics; and its behavior was
akin to that of smells and of infectious diseases. In a more general for-
mulation: Qualities are transmitted from one object to another; and
for primitive thinking such transmission seems to be the basic form of
causation.
The same idea lingers still in our thought. It may also be entitled to
more attention than it receives in modern epistemology. Even now our
theorists are under the spell of David Hume's powerful work. Trans-
mission of qualities has no place in our positivistic concept of causality.
For our present purpose this concept may be sufficiently characterized
as follows: An object has the property x, while in the same object and
in its immediate environment we have the conditions a, b, c. . . .
When a, b, c . . . change into a 1 , b 1 , c 1 . . . , instead of x the object
assumes the property y. It is, since Hume, customary to say that no
internal relation connects the nature of the change x - y with the na-
ture of the change a, b, c . . . ->a*, b 1 , c 1 . . . . Facts of coexistence
and of sequence are mere facts, which means that a priori any datum
might coexist with or follow any other. In principle, we do not get more
out of nature than incomprehensible rules.
One sees at once that between this idea of causation and causation
by transmission there is a striking difference. According to the positiv-
istic notion, no matter how long we consider causes, we can never derive
from this inspection any prediction as to the nature of the effect. Con-
trariwise, if causation is transmission of an identical quality, state, or
entity from one locality, it follows necessarily that the effect, the ap-
pearance of that entity in the second place, must be similar to the cause
which is the same entity as it appeared in its first place. Causa aequat
effectum.
It seems appropriate at this point to say one more word about a
famous period in the history of European philosophy. Why should
Descartes' school be so uneasy about the idea that mind and body in-
teract? The reason seems to be simple. In his time causation was still
widely conceived of as transmission of some quality from one thing to
another. This primitive but rather plausible idea had much power on
the best minds as late as 1650. Two incomparable substances, they
say, cannot influence each other. A modern positivist might ask: Where
216 : Wolfgang Kohler
is the difficulty? But for the older view the problem is doubtless serious.
If mind and body are really incomparable substances, nothing can be
transmitted from one to the other. Otherwise the second would, by the
fact of transmission, become similar to the first, which is impossible,
since they are supposed to be incomparable forever. Thus neither can
exert an influence upon the other. "Influence" is just the right word.
No other term could express more characteristically what causation
normally meant before the concept was so utterly changed by Hume.
Primitive people could point to many examples in which observation
agrees well with the concept of causation by transmission. Effects are
so often similar to causes that modern epistemology would do well to
give the fact at least some attention. Take any of the cases which I have
mentioned above. What is first in one place soon appears as something
similar in the neighborhood. Some qualities do travel like thin fluids.
But primitive man never formulates a definite notion of causality. It is
probably not by a great number of confirming experiences that he is
induced implicitly to favor just this particular notion. Rather its intrin-
sic plausibility seems to make it dominant so dominant in fact that
where the primitive is definitely wrong the white man's appeal to con-
tradictory experience is generally of no avail.
Very often the primitive is wrong. Unacquainted with any of our
critical procedures he extends his naive realism to all possible qualities
of things. Quite as uncritically he applies to all these qualities his un-
formulated principle of causation by transmission. We accept some of
the qualities and neglect many in our practical behavior; for we are
trained by science. We are not very clear about causality; but our tacit
acknowledgment of causation by transmission is definitely checked,
again by the scientific attitude of our culture. Superadded criticism
distinguishes our view of the world from that of primitive man, not
mental traits of his which we could not understand at all. Once more:
It seems a misconception to say that in some domains primitive man acts
soberly, as we do, while in others he relies on peculiar mystical prac-
tices which constitute the sphere of magic. So long as no scientific
criticism is possible the practices of magic are probably as natural as
any activities of which we approve. Both spring from the same sources:
naive realism and causation by transmission.
The fundamental rule of magic follows from these premises. If you
wish to produce a definite effect in an object or in a person, you have
to bring near him or, better, into him such things as are conspicuous by
the quality in question. When we are cold we approach the fire or take
Psychological Remarks on Some Questions of Anthropology : 217
a hot drink. This is not regarded as magic. Apply the principle gen-
erally and civilized people will say that you perform magical practices.
The primitive's child is slow in acquiring speech. The child's tongue
lacks the necessary agility. Where do the parents find agility? In the
lizard's tongue. Consequently the child will be fed lizard's tongues. The
white critic might say that agility is not like calories which can really be
transmitted from one object to another. Why not? The strange power of
fire can thus be transmitted. We light a match, hold it near the ciga-
rette, and at once the burning quality is transferred to the cigarette.
For people who have no science there is no essential difference between
one case and the other. Among the Thonga in southeast Africa a mother
who has not enough milk for her baby will eat euphorbia plants. These
contain the necessary milkiness. 6 Without any chemistry people will
rely on perceptual characteristics as they are directly given. Among
the Spaniards of Tenerif e, men who are losing their hair will eat the
hairy stems of certain plants. The "hairiness" of these will make the
hair grow. Primitive mentality need certainly not be essentially dif-
ferent from our own in order to develop such ideas. It has only to remain
naive and, in its basic primitive conceptions, thoroughly consistent.
Where power qualities are not absorbed in the form of food, but sup-
posed to spread through space, magic seems to us more strangely magic.
Still, the underlying principle remains the same and agrees with our
experience in the case of smell and heat. Instead of eating the strongest
parts of a strong animal a person may wear them somewhere on his
body. From this amulet the power quality will permeate the owner.
Since there are many capacities which a rich man will wish to possess
he may be forced to carry along a bundle of different power substances,
one for each purpose as he really does in West Africa. Contact is, of
course, better than mere neighborhood. A few years ago they had the
first national sweepstake in a country of Western Europe. A sailor won
the big prize. When he came to the capital in order to get his money,
the girls in the office crowded around the man and touched his body
with their fingers. Apparently they felt that some of his luck-power
might be conducted to themselves.
For reasons which I have mentioned above, considerable power may
be contained in an object which, to European eyes, does not look con-
spicuous. This is generally true of those substances which are the main
parts of so-called fetiches in West Africa. It is widely recognized at
6 H. Junod, The life of a South African tribe. (2nd ed.) London: Macmillan,
1927, pp.47 ff.
218 : Wolfgang Kohler
present that as a rule these objects are not idols or images of gods, but
mana-batteries like our storage cells. Their preparation by special ex-
perts involves the same principle of causation by transmission. In a
special hut, protected against disturbing influences, the prospective
new fetich is put near an old powerful battery and thus gradually
charged. 7 It is in line with such ideas that when wrongly treated even
the best fetich may lose much of its power. If that happens it is recharged
or sold at a lower rate.
Even "imitative" magic, at first one of the more surprising activities
of primitive man, seems to follow from the same principles: naive
realism in an extreme form and causation by transmission. At the time
when rain should fall certain Mexican Indians carry a baby around the
fields. The child is known to contain the power of nearly inexhaustible
wetness. Apparently this quality will spread and soon it will begin to
rain. I hope I do not exaggerate if I say that most monographs on prim-
itive peoples contain reports about analogous practices. Often imitative
magic produces the characteristics of an event rather than of any ob-
ject. We know of several tribes where lightning or thunder is imitated as
representative of thunderstorms and heavy rains.
I am not overlooking the fact that definite formulae and chants are
generally quite as essential for magical purposes as powerful objects
and performances. What can mere language have to do with mana?
The obvious answer is that for naive perception language may contain
at least as much mana as any objects; and certainly it spreads. Once
more our lack of understanding is self -produced. We know about sound
waves, we have learned that in language indifferent auditory phenomena
are merely associated with meanings. Such knowledge makes us unfit
to realize what language is for the unsophisticated mind: a stream of
often forceful content that issues from the mouth of man, full of hatred
in one case, of love in another, of despair in a third. How could prim-
itive people come upon the theory that language consists of sounds to
which the hearer adds a subjective interpretation? Whatever may be
true functionally, from a phenomenological point of view even we must
acknowledge that this is not a correct description of language, that for
naiVe apprehension meanings are in the words and sentences of com-
mon speech. Small wonder that the native's naive realism remains con-
sistent in this case.
May I take an example from one of Malinowski's books. When
7 E. Pechiiel-Loesche, Volkskunde von Loango. Stuttgart: Strecker, 1907,
p. 366.
Psychological Remarks on Some Questions of Anthropology : 219
among the Trobriand people a woman expects a baby much has to be
done for the child's health and beauty. One practice is this: Other
women prepare a special skirt or coat for the mother. During their
work definite formulae, sweet words about beauty, are spoken literally
into the coat. Carefully folded, so that the wind cannot reach the inside
and thus remove the charm, the cloth is carried to the mother who,
with the coat, covers herself with the mana of those words.
In anthropological literature similar ideas are even applied to ques-
tions of human personality. Phenomenologically this seems a sound
procedure. Some persons have a strikingly powerful appearance; and
many people are conspicuous for their particular varieties of perceptible
personality-mana. When we think of unusual people, perhaps for a
moment only, of Mussolini, of Lloyd George, of Greta Garbo, there is
before us, often with a visual image, a characteristic and distinctive
quality of being or living in each case. We cannot define it, yet it is ob-
vious enough; it has the earmarks of a von Ehrenfels quality, compara-
ble with the Mozart-character of some music and the BraJhms-quality
of another.
Since even common people talk and act in their individual manner,
in all of them there is some mana which gives them or, perhaps, is their
particular kind of aliveness. At certain times, however, this power does
not show so clearly; apparently it is absent. This would not be more
surprising than that heat or a smell are here now and elsewhere later.
Take the case of the man who is sleeping in his hut. Compared with
what he is in daytime, he seems to be completely lacking in his dis-
tinctive personal essence. In the morning when he awakens it will be
evident again. Where, then, has it been in the meantime? The man had
a dream in which "he" was in another village. Thus we see where his
personal power roamed while his body lay inert in the hut. When tem-
porarily outside the body, that power will of course behave as human
persons do. The dream tells us about its concrete activities. If these
were of a criminal nature, punishment will seem to be fully justified.
Levy-Bruhl assumes that for native belief a man may be in two places
at the same time, and that we can do no more than accept such an ex-
traordinary contention. I hesitate to adopt this skeptical attitude so
long as, in the manner just indicated, a more plausible interpretation
is available.
Is this one more attempt to offer a standard solution for all anthro-
pological enigmas? It is not meant to be. Criticizing another thesis,
Lowie has once said, "that its very simplicity should militate against its
220 : Wolfgang Kohler
acceptance." 8 1 admit that the principles which I have tried to apply to
primitive behavior are simple. But, then, only the very simplest forms
of such behavior have been given an interpretation; and the discussion
ends at a point where both problems and possible explanations begin to
assume a higher degree of complexity. Hardly a word has been said
about primitive religion, none about primitive art, again none about
social life with all its institutions and ramifications. I am convinced that
in these fields, too, psychology can be of more help to anthropology
now than, say, thirty or even twenty years ago. But for such help other
principles besides those of the present essay will have to be introduced.
Even the problems which have been treated present certain aspects
which can hardly be understood on such a narrow psychological basis.
Causation by transmission, for example, is doubtless an essential factor
in primitive thinking. But we know many primitive beliefs which, al-
though somewhat similar to those discussed above, can nevertheless
not be reduced to this principle. Levy-Bruhl is right, I believe, in his
assumption that typical cases of "participation" do not involve what
we should call causation in any of its forms.
A more serious objection might be raised precisely on the ground
that primitive society as such and its overwhelming influence on the
individual have not been mentioned at all. No special form of magic,
for instance, to which I have ref ered should, according to this criticism,
be interpreted in terms of "general" psychology. Only in the group, as
products of its creative faculty, can such practices ever arise; to the
group the individual owes his magical beliefs; and as the nature of
group life varies from one part of the world to another, the importance
of magic and its special forms vary correspondingly.
This objection, it seems to me, is partly justified insofar as no attempt
has really been made to explain any differences which, in such matters,
are found between the various peoples of the globe. These differences
are remarkable and of the greatest interest for anthropology. The reason
why they have not been discussed is quite simple. One cannot do it
without at once becoming involved in all the major disputes of present
anthropology. On the other hand such a discussion was not strictly
necessary. To some extent magic exists in practically every society.
If my interpretation is right, at least some of its major premises are not
peculiar to a few specific tribes, but are the common property of all
mankind below a certain high level of sophistication. From this com-
8 R. H. Lowie, op. cit., p. 259.
Psychological Remarks on Some Questions of Anthropology : 221
mon stock, which general developmental psychology is entitled to study
and to explain, different societies, with their different environments
and histories, have in fact derived different varieties of actual practices.
I do not believe that we can fully understand the origin of such varieties
before we know on what ground magic in general grows. That this
ground is the group as such, and that the individual is no more than an
empty container for the products of group mentality, seems to me an
unacceptable thesis. If the individual were intrinsically incapable of
such attitudes as fear, respect, aversion, love, and joy, of such functions
as perception, learning, and memory, no influences of the group, how-
ever strong, could ever produce in him these mental events. It is my
contention that the same applies to the psychological sources of magic.
What springs from these sources is codetermined by the particular tra-
dition of the group. Much magic also applies to group relationships as
such, but the group contains no ultimate source of magic which is not
also found in the individual.
Solomon E. Asch
SWARTHMORE COLLEGE
EFFECTS OF
GROUP PRESSURE UPON
THE MODIFICATION AND
DISTORTION OF JUDGMENTS
We shall here describe in summary form the conception and first find-
ings of a program of investigation into the conditions of independence
and submission to group pressure. This program is based on a series
of earlier studies conducted by the writer while a Fellow of the John
Simon Guggenheim Memorial Foundation. The earlier experiments
and the theoretical issues which prompted them are discussed in a
forthcoming work by the writer l on social psychology.
Our immediate object was to study the social and personal condi-
tions that induce individuals to resist or to yield to group pressures when
the latter are perceived to be contrary to fact. The issues which this
problem raises are of obvious consequence for society; it can be of de-
cisive importance whether or not a group will, under certain condi-
tions, submit to existing pressures. Equally direct are the consequences
Reprinted with permission from Groups, Leadership and Men, edited by
Harold Guetzkow. Pittsburgh: Carnegie Press, 1951.
1 S. E. Asch, Social psychology. New York: Prentice-Hall, 1952. Ed,
Group Pressure and the Modification of Judgments : 223
for individuals and our understanding of them, since it is a decisive fact
about a person whether he possesses the freedom to act independently,
or whether he characteristically submits to group pressures.
The problem under investigation requires the direct observation of
certain basic processes in the interaction between individuals, and be-
tween individuals and groups. To clarify these seems necessary if we
are to make fundamental advances in the understanding of the forma-
tion and reorganization of attitudes, of the functioning of public opin-
ion, and of the operation of propaganda. Today we do not possess an
adequate theory of these central psychosocial processes. Empirical in-
vestigation has been predominantly controlled by general propositions
concerning group influence which have as a rule been assumed but not
tested. With few exceptions investigation has relied upon descriptive
formulations concerning the operation of suggestion and prestige, the
inadequacy of which is becoming increasingly obvious, and upon sche-
matic applications of stimulus-response theory.
The bibliography lists articles representative of the current theoreti-
cal and empirical situation. Basic to the current approach has been the
axiom that group pressures characteristically induce psychological
changes arbitrarily, in far-reaching disregard of the material properties
of the given conditions. This mode of thinking has almost exclusively
stressed the slavish submission of individuals to group forces, has neg-
lected to inquire into their possibilities for independence and for pro-
ductive relations with the human environment, and has virtually denied
the capacity of men under certain conditions to rise above group pas-
sion and prejudice. It was our aim to contribute to a clarification of
these questions, important both for theory and for their human impli-
cations, by means of direct observation of the effects of groups upon the
decisions and evaluations of individuals.
THE EXPERIMENT AND FIRST RESULTS
To this end we developed an experimental technique which has served
as the basis for the present series of studies. We employed the procedure
of placing an individual in a relation of radical conflict with all the
other members of a group, of measuring its effect upon him in quanti-
tative terms, and of describing its psychological consequences. A group
of eight individuals was instructed to judge a series of simple, clearly
structured perceptual relations to match the length of a given line
224 : Solomon E. Asch
with one of three unequal lines. Each member of the group announced
his judgments publicly. In the midst of this monotonous "test" one in-
dividual found himself suddenly contradicted by the entire group, and
this contradiction was repeated again and again in the course of the
experiment. The group in question had, with the exception of one mem-
ber, previously met with the experimenter and received instructions to
respond at certain points with wrong and unanimous judgments.
The errors of the majority were large (ranging between % in. and 1%
in. ) and of an order not encountered under control conditions. The out-
standing person the critical subject whom we had placed in the
position of a minority of one in the midst of a unanimous majority
was the object of investigation. He faced, possibly for the first time in
his life, a situation in which a group unanimously contradicted the evi-
dence of his senses.
This procedure was the starting point of the investigation and the
point of departure for the study of further problems. Its main features
were the following: ( 1 ) The critical subject was submitted to two con-
tradictory and irreconcilable forces the evidence of his own experi-
ence of an utterly clear perceptual fact and the unanimous evidence of
a group of equals. (2) Both forces were part of the immediate situa-
tion; the majority was concretely present, surrounding the subject phys-
ically. (3) The critical subject, who was requested together with all
others to state his judgments publicly, was obliged to declare himself
and to take a definite stand vis-a-vis the group. (4) The situation pos-
sessed a self-contained character. The critical subject could not avoid
or evade the dilemma by reference to conditions external to the experi-
mental situation. (It may be mentioned at this point that the forces
generated by the given conditions acted so quickly upon the critical
subjects that instances of suspicion were rare.)
The technique employed permitted a simple quantitative measure
of the "majority effect" in terms of the frequency of errors in the direc-
tion of the distorted estimates of the majority. At the same time we were
concerned from the start to obtain evidence of the ways in which the
subjects perceived the group, to establish whether they became doubt-
ful, whether they were tempted to join the majority. Most important,
it was our object to establish the grounds of the subject's independence
or yielding whether, for example, the yielding subject was aware of
the effect of the majority upon him, whether he abandoned his judgment
deliberately or compulsively. To this end we constructed a compre-
hensive set of questions which served as the basis of an individual in-
Group Pressure and the Modification of Judgments : 225
terview immediately following the experimental period. Toward the
conclusion of the interview each subject was informed fully of the
purpose of the experiment, of his role and of that of the majority. The
reactions to the disclosure of the purpose of the experiment became
in fact an integral part of the procedure. We may state here that the
information derived from the interview became an indispensable source
of evidence and insight into the psychological structure of the experi-
mental situation, and in particular, of the nature of the individual dif-
ferences. Also, it is not justified or advisable to allow the subject to
leave without giving him a full explanation of the experimental condi-
tions. The experimenter has a responsibility to the subject to clarify
his doubts and to state the reasons for placing him in the experimental
situation. When this is done most subjects react with interest and many
TABLE 1
LENGTHS OF STANDARD AND COMPARISON LINES
Trials
Length
of Stand-
ard Line
(in inches)
Comparison Lines
(hi inches)
Correct
Response
Group
Response
Majority
Error
(hi inches)
1
2
3
1
10
8%
10
8
2
2
2
2
2
1
1%
1
1
3
3
3%
4^4
3
3
1*
+%
4
5
5
4
6 ] / 2
1
2*
1.0
5
4
3
5
4
3
3
6
3
3%
4&
3
3
2*
+iy 4
7
8
6%
8
6%
2
3*
-1%
8
5
5
4
6%
1
3*
+iy 2
9
8
6%
8
6%
2
1*
-1%
10
10
8%
10
8
2
2
11
2
2
1
1%
1
1
12
3
3%
4%
3
3
1*
+%
13
5
5
4
6%
1
2*
-1.0
14
4
3
5
4
3
3
15
3
3%
4&
3
3
2*
+1%
16
8
6%
8
6%
2
3*
-1%
17
5
5
4
6%
1
3*
+1H
18
8
6%
8
6%
2
1*
-l%
* Starred figures designate the erroneous estimates by the majority.
226 : Solomon E. Asch
express gratification at having lived through a striking situation which
has some bearing on wider human issues.
Both the members of the majority and the critical subjects were male
college students. We shall report the results for a total of fifty critical
subjects in this experiment. In Table 1 we summarize the successive
comparison trials and the majority estimates.
The quantitative results are clear and unambiguous.
1. There was a marked movement toward the majority. One-third
of all the estimates in the critical group were errors identical with or in
the direction of the distorted estimates of the majority. The significance
of this finding becomes clear in the light of the virtual absence of errors
in control groups, the members of which recorded their estimates in
writing. The relevant data of the critical and control groups are sum-
marized in Table 2.
2. At the same time the effect of the majority was far from complete.
TABLE 2
DISTRIBUTION OF ERRORS IN EXPERIMENTAL AND
CONTROL GROUPS
Number of
Critical Errors
Critical Group *
(JV = 50)
Control Group
(N = 37)
F
F
13
35
1
4
1
2
5
1
3
6
4
3
5
4
6
1
7
2
8
5
9
3
10
3
11
1
12
Total
50
37
Mean
3.84
0.08
* All errors in the critical group were in the direction of the majority
estimates.
Group Pressure and the Modification of Judgments : 227
The preponderance of estimates in the critical group (68 per cent) was
correct despite the pressure of the majority.
3. We found evidence of extreme individual differences. There were
in the critical group subjects who remained independent without ex-
ception, and there were those who went nearly all the time with the
majority. (The maximum possible number of errors was 12, while the
actual range of errors was 0-11.) One-fourth of the critical subjects
was completely independent; at the other extreme, one-third of the
group displaced the estimates toward the majority in one-half or more
of the trials.
The differences between the critical subjects in their reactions to the
given conditions were equally striking. There were subjects who re-
mained completely confident throughout. At the other extreme were
those who became disoriented, doubt-ridden, and experienced a power-
ful impulse not to appear different from the majority.
For purposes of illustration we include a brief description of one
independent and one yielding subject.
Independent. After a few trials he appeared puzzled, hesitant. He
announced all disagreeing answers in the form of "Three, sir; two, sir";
not so with the unanimous answers. At trial 4 he answered immediately
after the first member of the group, shook his head, blinked, and whis-
pered to his neighbor: "Can't help it, that's one." His later answers
came in a whispered voice, accompanied by a deprecating smile. At
one point he grinned embarrassedly, and whispered explosively to his
neighbor: "I always disagree darn it!" During the questioning, this
subject's constant refrain was: "I called them as I saw them, sir." He
insisted that his estimates were right without, however, committing
himself as to whether the others were wrong, remarking that "that's the
way I see them and that's the way they see them." If he had to make a
practical decision under similar circumstances, he declared, "I would
follow my own view, though part of my reason would tell me that I
might be wrong." Immediately following the experiment the majority
engaged this subject in a brief discussion. When they pressed him to
say whether the entire group was wrong and he alone right, he turned
upon them defiantly, exclaiming: "You're probably right, but you may
be wrong!" To the disclosure of the experiment this subject reacted
with the statement that he felt "exultant and relieved," adding, "I do
not deny that at times I had the feeling: 'to heck with it, I'll go along
with the rest.' "
Yielding This subiect went with the maioritv in 11 out of 12 trials.
225 : Solomon E. Asch
He appeared nervous and somewhat confused, but he did not attempt to
evade discussion; on the contrary, he was helpful and tried to answer to
the best of his ability. He opened the discussion with the statement:
"If Td been the first I probably would have responded differently"; this
was his way of stating that he had adopted the majority estimates. The
primary factor in his case was loss of confidence. He perceived the
majority as a decided group, acting without hesitation: "If they had
been doubtful I probably would have changed, but they answered with
such confidence/' Certain of his errors, he explained, were due to the
doubtful nature of the comparisons; in such instances he went with the
majority. When the object of the experiment was explained, the subject
volunteered: "I suspected about the middle but tried to push it out
of my mind." It is of interest that his suspicion was not able to restore
his confidence and diminish the power of the majority. Equally striking
is his report that he assumed the experiment to involve an "illusion"
to which the others, but not he, were subject. This assumption too did
not help to free him; on the contrary, he acted as if his divergence from
the majority was a sign of defect. The principal impression this subject
produced was of one so caught up by immediate difficulties that he lost
clear reasons for his actions, and could make no reasonable decisions.
A FIRST ANALYSIS OF INDIVIDUAL DIFFERENCES
On the basis of the interview data described earlier, we undertook to
differentiate and describe the major forms of reaction to the experi-
mental situation, which we shall now briefly summarize.
Among the independent subjects we distinguished the following
main categories:
L Independence based on confidence in one's perception and ex-
perience. The most striking characteristic of these subjects is the vigor
with which they withstand the group opposition. Though they are sen-
sitive to the group and experience the conflict, they show a resilience in
coping with it, which is expressed in their continuing reliance on their
perception and the effectiveness with which they shake off the oppres-
sive group opposition.
2. Quite different are those subjects who are independent and with-
drawn. These do not react in a spontaneously emotional way, but rather
on the basis of explicit principles concerning the necessity of being an
individual.
3. A third group of independent subjects manifest considerable ten-
Group Pressure and the Modification of Judgments : 229
sion and doubt, but adhere to their judgments on the basis of a felt
necessity to deal adequately with the task.
The following were the main categories of reaction among the yield-
ing subjects, or those who went with the majority during one-half or
more of the trials.
/. Distortion of perception under the stress of group pressure. In
this category belong a very few subjects who yield completely, but are
not aware that their estimates have been displaced or distorted by the
majority. These subjects report that they came to perceive the majority
estimates as correct.
2. Distortion of judgment. Most submitting subjects belong to this
category. The factor of greatest importance in this group is a decision
the subjects reach that their perceptions are inaccurate, and that those
of the majority are correct. These subjects suffer from primary doubt
and lack of confidence; on this basis they feel a strong tendency to join
the majority.
3. Distortion of action. The subjects in this group do not suffer a
modification of perception nor do they conclude that they are wrong.
They yield because of an overmastering need not to appear different
from or inferior to others, because of an inability to tolerate the appear-
ance of defectiveness in the eyes of the group. These subjects suppress
their observations and voice the majority position with awareness of
what they are doing.
The results are sufficient to establish that independence and yielding
are not psychologically homogeneous, that submission to group pres-
sure (and freedom from pressure) can be the result of different psy-
chological conditions. It should also be noted that the categories de-
scribed above, being based exclusively on the subjects' reactions to the
experimental conditions, are descriptive, not presuming to explain why
a given individual responded in one way rather than another. The
further exploration of the basis for the individual differences is a sepa-
rate task upon which we are now at work.
EXPERIMENTAL VARIATIONS
The results described are clearly a joint function of two broadly differ-
ent sets of conditions. They are determined first by the specific external
conditions, by the particular character of the relation between social
evidence and one's own experience. Second, the presence of pro-
nounced individual differences points to the important role of personal
250 : Solomon E. Asch
factors, of factors connected with the individual's character structure.
We reasoned that there are group conditions which would produce
independence in all subjects, and that there probably are group condi-
tions which would induce intensified yielding in many, though not in
all. Accordingly we followed the procedure of experimental variation,
systematically altering the quality of social evidence by means of sys-
tematic variation of group conditions. Secondly, we deemed it reason-
able to assume that behavior under the experimental social pressure is
significantly related to certain basic, relatively permanent characteris-
tics of the individual. The investigation has moved in both of these di-
rections. Because the study of the character qualities which may be
functionally connected with independence and yielding is still in prog-
ress, we shall limit the present account to a sketch of the representative
experimental variations,
The Effect of Non-unanimous Majorities
Evidence obtained from the basic experiment suggested that the con-
dition of being exposed alone to the opposition of a "compact major-
ity" may have played a decisive role in determining the course and
strength of the effects observed. Accordingly we undertook to investi-
gate in a series of successive variations the effects of non-unanimous
majorities. The technical problem of altering the uniformity of a major-
ity is, in terms of our procedure, relatively simple. In most instances
we merely directed one or more members of the instructed group to
deviate from the majority in prescribed ways. It is obvious that we can-
not hope to compare the performance of the same individual in two
situations on the assumption that they remain independent of one an-
other. At best we can investigate the effect of an earlier upon a later
experimental condition. The comparison of different experimental sit-
uations therefore requires the use of different but comparable groups
of critical subjects. This is the procedure we have followed. In the
variations to be described we have maintained the conditions of the
basic experiment (e.g., the sex of the subjects, the size of the majority,
the content of the task, and so on) save for the specific factor that was
varied. The following were some of the variations we studied:
1 . The presence of a "true partner." ( a) In the midst of the majority
were two naive, critical subjects. The subjects were separated spatially,
being seated in the fourth and eighth positions, respectively. Each
therefore heard his judgment confirmed by one other person (provided
Group Pressure and the Modification of Judgments : 231
the other person remained independent) , one prior to, the other subse-
quently to announcing his own judgment. In addition, each experienced
a break in the unanimity of the majority. There were six pairs of critical
subjects, (b) In a further variation the "partner" to the critical subject
was a member of the group who had been instructed to respond cor-
rectly throughout. This procedure permits the exact control of the
partner's responses. The partner was always seated in the fourth po-
sition; he therefore announced his estimates in each case before the
critical subject.
The results clearly demonstrate that a disturbance of the unanimity
of the majority markedly increased the independence of the critical sub-
jects. The frequency of pro-majority errors dropped to 10.4 per cent
of the total number of estimates in variation (a), and to 5.5 per cent in
variation (b). These results are to be compared with the frequency of
yielding to the unanimous majorities in the basic experiment, which
was 32 per cent of the total number of estimates. It is clear that the
presence in the field of one other individual who responded correctly
was sufficient to deplete the power of the majority, and in some cases
to destroy it. This finding is all the more striking in the light of other
variations which demonstrate the effect of even small minorities pro-
vided they are unanimous. Indeed, we have been able to show that a
unanimous majority of three is, under the given conditions, far more
effective than a majority of eight containing one dissenter. That critical
subjects will under these conditions free themselves of a majority of
seven and join forces with one other person in the minority is, we be-
lieve, a result significant for theory. It points to a fundamental psycho-
logical difference between the condition of being alone and having a
minimum of human support. It further demonstrates that the effects
obtained are not the result of a summation of influences proceeding
from each member of the group; it is necessary to conceive the results
as being relationally determined.
2. Withdrawal of a "true partner." What will be the effect of pro-
viding the critical subject with a partner who responds correctly and
then withdrawing him? The critical subject started with a partner who
responded correctly. The partner was a member of the majority who
had been instructed to respond correctly and to "desert" to the majority
in the middle of the experiment. This procedure permits the observa-
tion of the same subject in the course of transition from one condition
to another. The withdrawal of the partner produced a powerful and
unexpected result. We had assumed that the critical subject, having
232 : Solomon E. Asch
gone through the experience of opposing the majority with a minimum
of support, would maintain bis independence when alone. Contrary to
this expectation, we found that the experience of having had and then
lost a partner restored the majority effect to its full force, the proportion
of errors rising to 28.5 per cent of all judgments, in contrast to the pre-
ceding level of 5.5 per cent. Further experimentation is needed to estab-
lish whether the critical subjects were responding to the sheer fact of
being alone, or to the fact that the partner abandoned them.
3. Late arrival of a "true partner" The critical subject started as a
minority of one in the midst of a unanimous majority. Toward the con-
clusion of the experiment one member of the majority "broke" away
and began announcing correct estimates. This procedure, which re-
verses the order of conditions of the preceding experiment, permits the
observation of the transition from being alone to being a member of a
pair against a majority. It is obvious that those critical subjects who
were independent when alone would continue to be so when joined by
another partner. The variation is therefore of significance primarily for
those subjects who yielded during the first phase of the experiment. The
appearance of the late partner exerts a freeing effect, reducing the level
to 8.7 per cent. Those who had previously yielded also became mark-
edly more independent, but not completely so, continuing to yield more
than previously independent subjects. The reports of the subjects do
not cast much light on the factors responsible for the result. It is our
impression that having once committed himself to yielding, the indi-
vidual finds it difficult and painful to change his direction. To do so is
tantamount to a public admission that he has not acted rightly. He
therefore follows the precarious course he has already chosen in order
to maintain an outward semblance of consistency and conviction.
4. The presence of a "compromise partner' 9 The majority was con-
sistently extremist, always matching the standard with the most un-
equal line. One instructed subject (who, as in the other variations, pre-
ceded the critical subject) also responded incorrectly, but his estimates
were always intermediate between the truth and the majority position.
The critical subject therefore faced an extremist majority whose una-
nimity was broken by one more moderately erring person. Under these
conditions the frequency of errors was reduced but not significantly.
However, the lack of unanimity determined in a strikingly consistent
way the direction of the errors. The preponderance of the errors, 75.7
per cent of the total, was moderate, whereas in a parallel experiment
in which the majority was unanimously extremist (i.e., with the "com-
Group Pressure and the Modification of Judgments : 233
promise" partner excluded), the incidence of moderate errors was re-
duced to 42 per cent of the total. As might be expected, in a unani-
mously moderate majority, the errors of the critical subjects were
without exception moderate.
The Role of Majority Size
To gain further understanding of the majority effect, we varied the
size of the majority in several different variations. The majorities, which
were in each case unanimous, consisted of 16, 8, 4, 3, and 2 persons,
respectively. In addition, we studied the limiting case in which the
critical subject was opposed by one instructed subject. Table 3 con-
tains the means and the range of errors under each condition.
TABLE 3
ERRORS OF CRITICAL SUBJECTS WITH UNANIMOUS MAJORITIES
OF DIFFERENT SIZE
Size of majority
Control
1
2
3
4
8
16
N
37
10
15
10
10
50
12
Mean number of
errors
0.08
0.33
1.53
4.0
4.20
3.84
3.75
Range of errors
0-2
0-1
0-5
1-12
0-11
0-11
0-10
With the opposition reduced to one, the majority effect all but dis-
appeared. When the opposition proceeded from a group of two, it
produced a measurable though small distortion, the errors being 12.8
per cent of the total number of estimates. The effect appeared in full
force with a majority of three. Larger majorities of four, eight, and
sixteen did not produce effects greater than a majority of three.
The effect of a majority is often silent, revealing little of its operation
to the subject, and often hiding it from the experimenter. To examine
the range of effects it is capable of inducing, decisive variations of
conditions are necessary. An indication of one effect is furnished by
the following variation, in which the conditions of the basic experiment
were simply reversed. Here the majority, consisting of a group of six-
teen, was naive; in the midst of it we placed a single individual who
responded wrongly according to instructions. Under these conditions
234 : Solomon E. Asch
the members of the naive majority reacted to the lone dissenter with
amusement and disdain. Contagious laughter spread through the group
at the droll minority of one. Of significance is the fact that the members
lack awareness that they draw their strength from the majority, and
that their reactions would change radically if they faced the dissenter
individually. In fact, the attitude of derision in the majority turns to
seriousness and increased respect as soon as the minority is increased
to three. These observations demonstrate the role of social support as a
source of power and stability, in contrast to the preceding investigations
which stressed the effects of withdrawal of social support, or to be more
exact, the effects of social opposition. Both aspects must be explicitly
considered in a unified formulation of the effects of group conditions
on the formation and change of judgments.
The Role of the Stimulus Situation
It is obviously not possible to divorce the quality and course of the
group forces which act upon the individual from the specific stimulus
conditions. Of necessity the structure of the situation molds the group
forces and determines their direction as well as their strength. Indeed,
this was the reason that we took pains in the investigations described
above to center the issue between the individual and the group around
an elementary and fundamental matter of fact. And there can be no
doubt that the resulting reactions were directly a function of the contra-
diction between the objectively grasped relations and the majority
position.
These general considerations are sufficient to establish the need of
varying the stimulus conditions and of observing their effect on the
resulting group forces. We are at present conducting a series of investi-
gations in which certain aspects of the stimulus situation are systemati-
cally altered.
One of the dimensions we are examining is the magnitude of dis-
crepancies above the threshold. Our technique permits an easy varia-
tion of this factor, since we can increase or decrease at will the devia-
tion of the majority from the given objective conditions. Hitherto we
have studied the effect of a relatively moderate range of discrepancies.
Within the limits of our procedure we find that different magnitudes of
discrepancy produce approximately the same amount of yielding. How-
ever, the quality of yielding alters: as the majority becomes more
extreme, there occurs a significant increase in the frequency of "com-
Group Pressure and the Modification of Judgments : 235
promise" errors. Further experiments are planned in which the dis-
crepancies in question will be extremely large and small.
We have also varied systematically die structural clarity of the task,
including in separate variations judgments based on mental standards.
In agreement with other investigators, we find that the majority effect
grows stronger as the situation diminishes in clarity. Concurrently,
however, the disturbance of the subjects and the conflict quality of the
situation decrease markedly. We consider it of significance that the ma-
jority achieves its most pronounced effect when it acts most painlessly.
SUMMARY
We have investigated the effects upon individuals of majority opinions
when the latter were seen to be in a direction contrary to fact. By means
of a simple technique we produced a radical divergence between a ma-
jority and a minority, and observed the ways in which individuals
coped with the resulting difficulty. Despite the stress of the given con-
ditions, a substantial proportion of individuals retained their independ-
ence throughout. At the same time a substantial minority yielded, mod-
ifying their judgments in accordance with the majority. Independence
and yielding are a joint function of the following major factors: (1)
The character of the stimulus situation. Variations in structural clarity
have a decisive effect: with diminishing clarity of the stimulus condi-
tions the majority effect increases. (2) The character of the group
forces. Individuals are highly sensitive to the structural qualities of
group opposition. In particular, we demonstrated the great importance
of the factor of unanimity. Also, the majority effect is a function of the
size of group opposition. (3) The character of the individual. There
were wide, and indeed, striking differences among individuals within
the same experimental situation. The hypothesis was proposed that
these are functionally dependent on relatively enduring character dif-
ferences, in particular those pertaining to the person's social relations.
REFERENCES
1. Asch, S. E. Studies in the principles of judgments and attitudes: II.
Determination of judgments by group and by ego-standards. J. soc.
PsychoL, 1940, 12, 433-465.
2. Asch, S. E. The doctrine of suggestion, prestige and imitation in social
psychology. PsychoL Rev., 1948, 55, 250-276.
3. Asch, S. E., Block, H., and Hertzman, M. Studies in the principles
236 : Solomon E. Asch
of judgments and attitudes: I. Two basic principles of judgment. /.
PsychoL, 1938, 5, 219-251.
4. Coffin, E. E. Some conditions of suggestion and suggestibility: A study
of certain attitudinal and situational factors influencing the process of
suggestion. PsychoL Monogr., 1941, 53, No. 4.
5. Lewis, H. B. Studies in the principles of judgments and attitudes: IV.
The operation of prestige suggestion. 7. soc. PsychoL, 1941, 14, 229-
256.
6. Lorge, I. Prestige, suggestion, and attitudes. /. soc. PsychoL, 1936, 7,
386-402.
7. Miller, N. E., and Dollard, J. Social learning and imitation. New Haven:
Yale University Press, 194L
8. Moore, H- T. The comparative influence of majority and expert opinion.
Arner. /. PsychoL, 1921, 32, 16-20.
9. Sherif, M. A study of some social factors in perception. Arch. PsychoL,
1935, No. 187.
10. Thorndike, E. L. The psychology of wants, interests, and attitudes.
New York: Appleton-Century, 1935.
Solomon E. Asch
SWARTHMORE COLLEGE
FORMING IMPRESSIONS
OF PERSONALITY
We look at a person and immediately a certain impression of his charac-
ter forms itself in us. A glance, a few spoken words are sufficient to tell
us a story about a highly complex matter. We know that such im-
pressions form with remarkable rapidity and with great ease. 1 Subse-
quent observation may enrich or upset our first view, but we can no
more prevent its rapid growth than we can avoid perceiving a given
visual object or hearing a melody. We also know that this process,
though often imperfect, is also at times extraordinarily sensitive.
This remarkable capacity we possess to understand something of the
character of another person, to form a conception of him as a human
being, as a center of life and striving, with particular characteristics
forming a distinct individuality, is a precondition of social life. In
what manner are these inipressions established? Are there lawful prin-
ciples regulating their formation?
One particular problem commands our attention. Each person con-
fronts us with a large number of diverse characteristics. This man is
Reprinted with permission from the Journal of Abnormal and Social Psy-
chology, Vol. 41, No. 3, July, 1946.
1 The present investigation was begun in 1943 when the writer was a Fellow
of the John Simon Guggenheim Memorial Foundation.
238 : Solomon E. Asch
courageous, intelligent, with a ready sense of humor, quick in his move-
ments, but he is also serious, energetic, patient under stress, not to men-
tion his politeness and punctuality. These characteristics and many
others enter into the formation of our view. Yet our impression is from
the start unified; it is the impression of one person. We ask: How do
the several characteristics function together to produce an impression
of one person? What principles regulate this process?
We have mentioned earlier that the impression of a person grows
quickly and easily. Yet our minds falter when we face the far simpler
task of mastering a series of disconnected numbers or words. We have
apparently no need to commit to memory by repeated drill the various
characteristics we observe in a person, nor do some of his traits exert an
observable retroactive inhibition upon our grasp of the others. Indeed,
they seem to support each other. And it is quite hard to forget our view
of a person once it has formed. Similarly, we do not easily cpnf use the
half of one person with the half of another. It should be of interest to
the psychologist that the far more complex task of grasping the nature
of a person is so much less difficult.
There are a number of theoretical possibilities for describing the
process of forming an impression, of which the major ones are the
following:
1. A trait is realized in its particular quality. The next trait is simi-
larly realized, etc. Each trait produces its particular impression. The
total impression of the person is the sum of the several independent
impressions. If a person possesses traits a, b, c, d, e, then the impression
of him may be expressed as:
I. Impression = a + b + c + d + e
Few if any psychologists would at the present time apply this formu-
lation strictly. It would, however, be an error to deny its importance for
the present problem. That it controls in considerable degree many of
the procedures for arriving at a scientific, objective view of a person
(e.g., by means of questionnaires, rating scales) is evident. But more
pertinent to our present discussion is the modified form in which Propo-
sition I is applied to the actual forming of an impression. Some psy-
chologists assume, in addition to the factors of Proposition I, the opera-
tion of a "general impression." The latter is conceived as an affective
force possessing a plus or minus direction which shifts the evaluation
of the several traits in its direction. We may represent this process as
follows:
Forming Impressions of Personality : 239
la*- Impression =
a
To the sum of the traits there is now added another factor, the general
impression.
2. The second view asserts that we form an impression of the entire
person. We see a person as consisting not of these and those inde-
pendent traits (or of the sum of mutually modified traits), but we try
to get at the root of the personality. This would involve that the traits
are perceived in relation to each other, in their proper place within
the given personality. We may express the final impression as
II. Impression =
It may appear that psychologists generally hold to some form of the
latter formulation. The frequent reference to the unity of the person,
or to his "integration," implying that these qualities are also present in
the impression, point in this direction. The generality of these expres-
sions is, however, not suitable to exact treatment. Terms such as unity
of the person, while pointing to a problem, do not solve it. If we wish
to become clear about the unity in persons, or in the impression of per-
sons, we must ask in what sense there is such unity, and in what manner
we come to observe it. Secondly, these terms are often applied inter-
changeably to Propositions TL and la. It is therefore important to state
at this point a distinction between them.
For Proposition II, the general impression is not a factor added to
the particular traits, but rather the perception of a particular form of
relation between the traits, a conception which is wholly missing in la.
Further, Proposition la conceives the process in terms of an imposed
240 : Solomon E. Asch
affective shift in the evaluation of separate traits, whereas Proposition
II deals in the first instance with processes between the traits each of
which has a cognitive content.
Perhaps the central difference between the two propositions becomes
clearest when the accuracy of the impression becomes an issue. It is
implicit in Proposition II that the process it describes is for the subject
a necessary one if he is to focus on a person with maximum clarity. On
the other hand, Proposition la permits a radically different interpreta-
tion. It has been asserted that the general impression "colors" the par-
ticular characteristics, the effect being to blur the clarity with which
the latter are perceived. In consequence the conclusion is drawn that
the general impression is a source of error which should be supplanted
by the attitude of judging each trait in isolation, as described in Propo-
sition L This is the doctrine of the "halo effect" (9) .
With the latter remarks, which we introduced only for purposes of
illustration, we have passed beyond the scope of the present report. It
must be made clear that we shall here deal with certain processes in-
volved in the forming of an impression, a problem logically distinct
from the actual relation of traits within a person. To be sure, the man-
ner in which an impression is formed contains, as we shall see, definite
assumptions concerning the structure of personal traits. The validity
of such assumptions must, however, be established in independent
investigation.
The issues we shall consider have been largely neglected in investiga-
tion. Perhaps the main reason has been a one-sided stress on the sub-
jectivity of personal judgments. The preoccupation with emotional
factors and distortions of judgment has had two main consequences for
the course investigation has taken. First, it has induced a certain lack
of perspective which has diverted interest from the study of those
processes which do not involve subjective distortions as the most de-
cisive factor. Secondly, there has been a tendency to neglect the fact
that emotions too have a cognitive side, that something must be per-
ceived and discriminated in order that it may be loved or hated. On
the other hand, the approach of the more careful studies in this region
has centered mainly on questions of validity in the final product of
judgment. Neither of the main approaches has dealt explicitly with the
process of forming an impression. Yet no argument should be needed
to support the statement that our view of a person necessarily involves
a certain orientation to, and ordering of, objectively given, observable
characteristics. It is this aspect of the problem that we propose to study.
Forming Impressions of Personality : 241
The plan followed in the experiments to be reported was to read to
the subject a number of discrete characteristics, said to belong to a per-
son, with the instruction to describe the impression he formed. The sub-
jects were all college students, most of whom were women. 2 They were
mostly beginners in psychology. Though they expressed genuine in-
terest in the tasks, the subjects were not aware of the nature of the prob-
lem until it was explained to them. We illustrate our procedure with one
concrete instance. The following list of terms was read: energetic as-
sured talkative cold ironical inquisitive persuasive. The read-
ing of the list was preceded by the following instructions:
1 shall read to you a number of characteristics that belong to a particular
person. Please listen to them carefully and try to form an impression of the
kind of person described. You will later be asked to give a brief characteriza-
tion of the person in just a few sentences. I will read the list slowly and will
repeat it once.
The list was read with an interval of approximately five seconds be-
tween the terms. When the first reading was completed, the experi-
menter said, "I will now read the list again," and proceeded to do so.
We reproduce below a few typical sketches written by subjects after
they heard read the list of terms:
He seems to be the kind of person who would make a great impression
upon others at a first meeting. However as time went by, his acquaintances
would easily come to see through the mask. Underneath would be revealed
his arrogance and selfishness.
He is the type of person you meet all too often: sure of himself, talks too
much, always trying to bring you around to his way of thinking, and with not
much feeling for the other fellow.
He impresses people as being more capable than he really is. He is popular
and never ill at ease. Easily becomes the center of attraction at any gathering.
He is likely to be a jack-of-all-trades. Although his interests are varied, he is
not necessarily well-versed in any of them. He possesses a sense of humor.
His presence stimulates enthusiasm and very often he does arrive at a posi-
tion of importance.
2 The writer wishes to express his gratitude to the following colleagues for
their help in the performance of these experiments in their classes: Drs. B. F.
Riess, L. Welch, V. J. McGill, and A. Goldenson of Hunter College; Drs. M.
Blum and A. Mintz of the College of the City of New York; Dr. Lois Adams, Mr,
Michael Newman, and Mr. Herbert Newman of Brooklyn College.
242 ; Solomon E. Asch
Possibly he does not have any deep feeling. He would tend to be an oppor-
tunist. Likely to succeed in things he intends to do. He has perhaps married
a wife who would help him in his purpose. He tends to be skeptical.
The following preliminary points are to be noted:
1 . When a task of this kind is given, a normal adult is capable of re-
sponding to the instruction by forming a unified impression. Though
he hears a sequence of discrete terms, his resulting impression is not
discrete. In some manner he shapes the separate qualities into a single,
consistent view. All subjects in the following experiments, of whom
there were over 1,000, fulfilled the task in the manner described. No
one proceeded by reproducing the given list of terms, as one would in
a rote memory experiment; nor did any of the subjects reply merely
with synonyms of the given terms.
2. The characteristics seem to reach out beyond the merely given
terms of the description. Starting from the bare terms, the final account
is completed and rounded. Reference is made to characters and situa-
tions which are apparently not directly mentioned in the list, but which
are inferred from it.
3. The accounts of the subjects diverge from each other in important
respects. This will not be surprising in view of the variable content of
the terms employed, which permits a considerable freedom in interpre-
tation and weighting.
In the experiments to be reported the subjects were given a group of
traits on the basis of which they formed an impression. In view of the
fact that we possess no principles in this region to help in their sys-
tematic construction, it was necessary to invent groupings of traits. In
this we were guided by an informal sense of what traits were consistent
with each other.
The procedure here employed is clearly different from the everyday
situation in which we follow the concrete actions of an actual person.
We have chosen to work with weak, incipient impressions, based on
abbreviated descriptions of personal qualities. Nevertheless, this pro-
cedure has some merit for purposes of investigation, especially in ob-
serving the change of impressions, and is, we hope to show, relevant to
more natural judgment.
More detailed features of the procedure will be described subse-
quently in connection with the actual experiments. We shall now in-
quire into some of the factors that determine the content and alteration
of such impressions.
Forming Impressions of Personality : 243
I. CENTRAL AND PERIPHERAL CHARACTERISTICS
VARIATION OF A CENTRAL QUALITY
Observation suggests that not all qualities have the same weight in
establishing the view of a person. Some are felt to be basic, others
secondary. In the following experiments we sought for a demonstration
of this process in the course of the formation of an impression.
Experiment 1
Two groups, A and B, heard read a list of character qualities, identi-
cal save for one term. The list follows:
A. intelligent skillful industrious warm determined practical
cautious
B . intelligent skillful industrious cold determined practical
cautious
Group A heard the person described as "warm"; Group B, as "cold."
Technique. The instructions were as described above. Following the
reading, each subject wrote a brief sketch.
The sketches furnish concrete evidence of the impressions formed.
Their exact analysis involves, however, serious technical difficulties.
It seemed, therefore, desirable to add a somewhat simpler procedure for
the determination of the content of the impression and for the purpose
of group comparisons. To this end we constructed a check list con-
sisting of pairs of traits, mostly opposites. From each pair of terms in
this list, which the reader will find reproduced in Table 1, the subject
was instructed to select the one that was most in accordance with the
view he had formed. Terms were included which were quite different
from those appearing in the basic list, but which could be related to
them. Of necessity we were guided in the selection of terms for the
check list (as well as for the experimental lists) by an informal sense
of what was fitting or relevant. Some of the terms were taken from
written sketches of subjects in preliminary experiments. In the exami-
nation of results we shall rely upon the written sketches for evidence
of the actual character of the impressions, and we shall supplement
these with the quantitative results from the check list.
There were 90 subjects in Group A (comprising four separate class-
room groups), 76 subjects in Group B (comprising four separate class-
room groups).
244 : Solomon E. Asch
TABLE 1
CHECK LIST I
1. generous ungenerous 10. ruthless humane
2. shrewd wise 11. good-looking unattractive
3. unhappy happy 12. persistent unstable
4. irritable good-natured 13. frivolous serious
5. humorous humorless 14. restrained talkative
6. sociable unsociable 15. self-centered altruistic
7. popular unpopular 16. imaginative hard-headed
8. unreliable reliable 17. strong weak
9. important insignificant 18, dishonest honest
Results. Are the impressions of Groups A and B identical, with the
exception that one has the added quality of "warm," the other of
"cold"? This is one possible outcome. Another possibility is that the
differentiating quality imparts a general plus or minus direction to the
resulting impression. We shall see that neither of these formulations
accurately describes the results.
We note first that the characteristic "warm-cold" produces striking
and consistent differences of impression. In general, the A-impressions
are far more positive than the B-impressions. We cite a few representa-
tive examples:
Series A ("warm")
A person who believes certain things to be right, wants others to see his
point, would be sincere in an argument, and would like to see his point won.
A scientist performing experiments and persevering after many setbacks.
He is driven by the desire to accomplish something that would be of benefit.
Series B ("cold")
A very ambitious and talented person who would not let anyone or any-
thing stand in the way of achieving his goal. Wants his own way, he is deter-
mined not to give in, no matter what happens.
A rather snobbish person who feels that his success and intelligence set
him apart from the run-of-the-mill individual. Calculating and unsympa-
thetic.
This trend is fully confirmed in the check-list choices. In Table 2 we
report the frequency (in terms of percentages) with which each term
in the check list was selected. For the sake of brevity of presentation
Forming Impressions of Personality : 245
we state the results for the positive term in each pair; the reader may
determine the percentage of choices for the other term in each pair by
subtracting the given figure from 100. To illustrate, under Condition
A of the present experiment, 91 per cent of the subjects chose the
designation "generous"; the remaining 9 per cent selected the designa-
tion "ungenerous." Occasionally, a subject would not state a choice
for a particular pair. Therefore, the number of cases on which the
figures are based is not always identical; however, the fluctuations were
minor, with the exception of the category "good-looking unattrac-
tive," which a larger proportion of subjects failed to answer.
TABLE 2
CHOICE OF FITTING QUALITIES (PERCENTAGES)
Experiment 1
Experiment 2
Experiment 3
"Warm"
N = 9Q
"Cold"
AT = 76
Total
tf=56
"Warm"
N = 23
"Cold"
N=33
"Polite"
N = 20
"Blunt"
N=26
1. generous
91
8
55
87
33
56
58
2. wise
65
25
49
73
33
30
50
3. happy
90
34
71
91
58
75
65
4. good-natured
94
17
69
91
55
87
56
5. humorous
77
13
36
76
12
71
48
6. sociable
91
38
71
91
55
83
68
7. popular
84
28
57
83
39
94
56
8. reliable
94
99
96
96
97
95
100
9. important
88
99
88
87
88
94
96
10. humane
86
31
64
91
45
59
77
11. good-looking
77
69
58
71
53
93
79
12. persistent
100
97
98
96
100
100
100
13. serious
100
99
96
91
100
100
100
14. restrained
77
89
82
67
94
82
77
15. altruistic
69
18
44
68
27
29
46
16. imaginative
51
19
24
45
9
33
31
17. strong
98
95
95
94
96
100
100
18. honest
98
94
95
100
92
87
100
We find:
1. There are extreme reversals between Groups A and B in the
choice of fitting characteristics. Certain qualities are preponderantly
assigned to the "warm" person, while the opposing qualities are equally
prominent in the "cold" person. This holds for the qualities of (1)
generosity, (2) shrewdness, (3) happiness, (4) irritability, (5) hu-
mor, (6) sociability, (7) popularity, (10) ruthlessness, (15) self-
centeredness, (16) imaginativeness.
246 : Solomon E. Asch
2. There is another group of qualities which is not affected by the
transition from "warm" to "cold," or only slightly affected. These are:
(8) reliability, (9) importance, (11) physical attractiveness, (12)
persistence, (13) seriousness, (14) restraint, (17) strength, (18)
honesty.
These results show that a change in one character quality has pro-
duced a widespread change in the entire impression. Further, the writ-
ten sketches show that the terms "warm-cold" did not simply add a new
quality, but to some extent transformed the other characteristics. With
this point we shall deal more explicitly in the experiments to follow.
That such transformations take place is also a matter of everyday
experience. If a man is intelligent, this has an effect on the way in which
we perceive his playfulness, happiness, friendliness. At the same time,
this extensive change does not function indiscriminately. The "warm"
person is not seen more favorably in all respects. There is a range of
qualities, among them a number that are basic, which are not touched
by the distinction between "warm" and "cold." Both remain equally
honest, strong, serious, reliable, etc.
The latter result is of interest with reference to one possible inter-
pretation of the findings. It might be supposed that the category "warm-
cold" aroused a "mental set" or established a halo tending toward a
consistently plus or minus evaluation. We observe here that this trend
did not work in an indiscriminate manner, but was decisively limited at
certain points. If we assume that the process of mutual influence took
place in terms of the actual character of the qualities in question, it is
not surprising that some will, by virtue of their content, remain un-
changed. 3
The following will show that the subjects generally felt the qualities
"warm-cold" to be of primary importance. We asked the subjects in
certain of the groups to rank the terms of Lists A and B in order of
their importance for determining their impression. Table 3, containing
the distribution of rankings of "warm-cold," shows that these qualities
ranked comparatively high. At the same time a considerable number
of subjects relegated "cold" to the lowest position. That the rankings
are not higher is due to the fact that the lists contained other central
traits.
These data, as well as the ranking of the other traits not here repro-
duced, point to the following conclusions:
3 This by no means excludes the possibility that the nuances of strength, hon-
esty, etc., do change in relation to "warm-cold."
Forming Impressions of Personality : 247
1. The given characteristics do not all have the same weight for the
subject. He assigns to some a higher importance than to others.
2. The weight of a given characteristic varies within limits from
subject to subject.
TABLE 3
RANKINGS OF "WARM" AND "COLD": EXPERIMENT 1
"Warm"
"Cold"
Rank
N
Percentage
N
Percentage
1
6
14
12
27
2
15
35
8
21
3
4
10
1
2
4
4
10
2
5
5
4
10
3
7
6
3
7
2
5
7
6
14
13
33
42
100
41
100
Certain limitations of the check-list procedure need to be consid-
ered: (1) The subject's reactions are forced into an appearance of
discreteness which they do not actually possess, as the written sketches
show; (2) the check list requires the subject to choose between extreme
characteristics, which he might prefer to avoid; (3) the quantitative
data describe group trends; they do not represent adequately the form
of the individual impression. Generally the individual responses exhibit
much stronger trends in a consistently positive or negative direction.
For these reasons we employ the check-list results primarily for the
purpose of comparing group trends under different conditions. For this
purpose the procedure is quite adequate.
OMISSION OF A CENTRAL QUALITY
That the category "warm-cold" is significant for the total impression
may be demonstrated also by omitting it from the series. This we do in
the following experiment.
Experiment 2
The procedure was identical with that of experiment 1, except that
the terms "warm" and "cold" were omitted from the list read to the
subject (intelligent skillful industrious determined practical
248 : Solomon E. Asch
cautious) . Also, the check list was identical with that of experiment 1,
save that "warm-cold" was added as the last pair. There were three
groups, consisting of a total of 56 subjects.
Under these conditions the selection of fitting characteristics shows a
significant change. The distribution of choices for the total group (see
Table 2, column labeled "Total") now falls between the "warm" and
"cold" variations of experiment 1. It appears that a more neutral im-
pression has formed.
The total group results are, however, largely a statistical artifact.
An examination of the check-list choices of the subjects quickly re-
vealed strong and consistent individual differences. They tended to be
consistently positive or negative in their evaluations. It will be recalled
that the terms "warm-cold" were added to the check list. This per-
mitted us to subdivide the total group according to whether they judged
the described person on the check list as "warm" or "cold." Of the
entire group, 23 subjects (or 41 per cent) fell into the "warm" cate-
gory. Our next step was to study the distribution of choices in the two
subgroups. The results are clear: the two subgroups diverge consist-
ently in the direction of the "warm" and the "cold" groups, respectively,
of experiment 1. (See Table 2.) This is especially the case with the
two "warm" series, which are virtually identical.
It is of interest that the omission of a term from the experimental list
did not function entirely as an omission. Instead, the subjects inferred
the corresponding quality in either the positive or negative direction.
While not entirely conclusive, the results suggest that a full impression
of a person cannot remain indifferent to a category as fundamental as
the one in question, and that a trend is set up to include it in the im-
pression on the basis of the given data. In later experiments too we have
found a strong trend to reach out toward evaluations which were not
contained in the original description.
VARIATION OF A PERIPHERAL QUALITY
Would a change of any character quality produce an effect as strong
as that observed above? "Warm" and "cold" seem to be of special im-
portance for our conception of a person. This was, in fact, the reason
for selecting them for study. If there are central qualities, upon which
the content of other qualities depends, and dependent qualities which
are secondarily determined, it should be possible to distinguish them
objectively. On this assumption the addition or omission of peripheral
Forming Impressions of Personality : 249
qualities should have smaller effects than those observed in experiment
1. We turn to this question in the following experiment.
Experiment 3
The following lists were read, each to a different group:
A. intelligent skillful industrious polite determined practical
cautious
B. intelligent skillful industrious blunt determined practical
cautious
The A group contained 20, the B group 26 subjects.
TABLE 4
RANKINGS OF "POLITE" AND "BLUNT": EXPERIMENT 3
Rank
A: "Polite"
B: "Blunt"
N
Percentage
N
Percentage
1
2
4
15
3
3
12
4
2
10
5
19
5
3
16
6
23
6
4
21
1
4
7
10
53
7
27
19
100
26
100
The changes introduced into the selection of fitting characteristics
in the transition from "polite" to "blunt" were far weaker than those
found in experiment 1 (see Table 2) . There is further evidence that the
subjects themselves regarded these characteristics as relatively periph-
eral, especially the characteristic "polite." If we may take the rankings
as an index, then we may conclude that a change in a peripheral trait
produces a weaker effect on the total impression than does a change in
a central trait. (Though the changes produced are weaker than those
of experiment 1, they are nevertheless substantial. Possibly this is a
consequence of the thinness of the impression, which responds easily to
slight changes.)
250 ; Solomon E. Asch
TRANSFORMATION FROM A CENTRAL TO A PERIPHERAL QUALITY
The preceding experiments have demonstrated a process of discrimi-
nation between central and peripheral qualities. We ask: Are certain
qualities constantly central? Or is their functional value, too, dependent
on the other characteristics?
Experiment 4
We selected for observation the quality "warm," which was demon-
strated to exert a powerful effect on the total impression (experiments 1
and 2). The eSect of the term was studied in the following two series:
A, obedient weak shallow warm unambitious vain
B. vain shiewd unscrupulous warm shallow envious
Immediately "warm" drops as a significant characteristic in relation
to the others, as the distribution of rankings appearing in Table 5
shows* (Compare Table 3 of experiment 1.)
TABLE 5
RANKINGS OF "WARM" AND "COLD": EXPERIMENT 4
"Warm"
"Cold"
Series A
Series B
Series C
Rank
N
Percentage
N
Percentage
N
Percentage
1
L
4
1
5
2
3
2
9
1
5
3
15
4
5
27
4
19
2
10
5
7
30
4
19
1
5
6
7
30
12
57
2
10
7
11
55
23
100
21
100
20
100
More enlightening are the subjects' comments. In Series A the qual-
ity "warm" is now seen as wholly dependent, dominated by others far
more decisive.
I think the warmth within this person is a warmth emanating from a fol-
lower to a leader.
Forming Impressions of Personality : 251
The term "warm" strikes one as being a dog-like affection rather than a
bright friendliness. It is passive and without strength.
His submissiveness may lead people to think he is kind and warm.
A more extreme transformation is observed in Series B. In most in-
stances the warmth of this person is felt to lack sincerity, as appears in
the following protocols:
I assumed the person to appear warm rather than really to be warm.
He was warm only when it worked in with his scheme to get others over
to his side. His warmth is not sincere.
A similar change was also observed in the content of "cold" in a
further variation. The subject heard List B of experiment 1 followed
by Series C below, the task being to state whether the term "cold" had
the same meaning in both lists.
C. intelligent skillful sincere cold conscientious helpful modest
All subjects reported a difference. The quality "cold" became pe-
ripheral for all in Series C. The following are representative comments:
The coldness of 1 (experiment 1) borders on ruthlessness; 2 analyses
coldly to differentiate between right and wrong.
1 is cold inwardly and outwardly, while 2 is cold only superficially.
1 : cold means lack of sympathy and understanding; 2 : cold means some-
what formal in manner.
Coldness was the foremost characteristic of 1. In 2 it seemed not very
important, a quality that would disappear after you came to know him.
That "cold" was transformed in the present series into a peripheral
quality is also confirmed by the rankings reported in Table 5.
We conclude that a quality, central in one person, may undergo a
change of content in another person, and become subsidiary. When
central, the quality has a different content and weight than when it is
subsidiary.
Here we observe directly a process of grouping in the course of which
the content of a trait changes in relation to its surroundings. Secondly,
we observe that the functional value of a trait, too whether, for
252 : Solomon E. Asch
example, it becomes central or not is a consequence of its relation
to the set of surrounding traits. At the same time we are able to see more
clearly the distinction between central and peripheral traits. It is inade-
quate to say that a central trait is more important, contributes more
quantitatively to, or is more highly correlated with, the final impression
than a peripheral trait. The latter formulations are true, but they fail
to consider the qualitative process of mutual determination between
traits, namely, that a central trait determines the content and the func-
tional place of peripheral traits within the entire impression. In Series
A, for example, the quality "warm" does not control the meaning of
"weak," but is controlled by it.
The evidence may seem to support the conclusion that the same
quality which is central in one impression becomes peripheral in an-
other. Such an interpretation would, however, contain an ambiguity.
While we may speak of relativity in the functional value of a trait within
a person, in a deeper sense we have here the opposite of relativity. For
the sense of "warm" (or "cold") of experiment 1 has not suffered a
change of evaluation under the present conditions. Quite the contrary;
the terms in question change precisely because the subject does not see
the possibility of finding in this person the same warmth he values so
highly when he does meet it (correspondingly for coldness) .
Experiment 5
The preceding experiments have shown that the characteristics form-
ing the basis of an impression do not contribute each a fixed, independ-
ent meaning, but that their content is itself partly a function of the
environment of the other characteristics, of their mutual relations. We
propose now to investigate more directly the manner in which the con-
tent of a given characteristic may undergo change.
Lists A and B were read to two separate groups (including 38 and
41 subjects, respectively). The first three terms of the two lists are
opposites; the final two terms are identical.
A. kind wise honest calm strong
B. cruel shrewd unscrupulous calm strong
The instructions were to write down synonyms for the given terms.
The instructions read: "Suppose you had to describe this person in the
same manner, but without using the terms you heard, what other terms
would you use?" We are concerned with the synonyms given to the two
final terms.
Forming Impressions of Personality : 253
In Table 6 we list those synonyms of "calm" which occurred with
different frequencies in the two groups. It will be seen that terms appear
in one group which are not at all to be found in the other; further, some
terms appear with considerably different frequencies under the two
conditions. These do not, however, include the total group of syno-
nyms; many scattered terms occurred equally in both groups.
TABLE 6
SYNONYMS OF "CALM": EXPERIMENT 5
"Kind" Series
"Cruel" Series
serene
18
3
cold, frigid, icy, cool, calculating, shrewd,
nervy, scheming, conscienceless
20
soothing, peaceful, gentle, tolerant, good-
natured, mild-mannered
11
poised, reserved, restful, unexcitable, un-
shakable
18
7
deliberate, silent, unperturbed, masterful,
impassive, collected, confident, relaxed,
emotionless, steady, impassive, com-
posed
11
26
We may conclude that the quality "calm" did not, at least in some
cases, function as an independent, fixed trait, but that its content was
determined by its relation to the other terms. As a consequence, the
quality "calm" was not the same under the two experimental condi-
tions. In Series A it possessed an aspect of gentleness, while a grimmer
side became prominent in Series B. 4
Essentially the same may be said of the final term, "strong." Again,
some synonyms appear exclusively in one or the other groups, and in
the expected directions. Among these are:
Series A: fearless helpful just forceful courageous reliable
Series B: ruthless overbearing overpowering hard inflexible un-
bending dominant
4 In an earlier investigation the writer (2) has dealt with basically the same
question though in a very different context. It was there shown that certain phe-
nomena of judgment, which appeared to be due to changes of evaluation, were
produced by a shift in the frame of reference.
254 : Solomon E. Asch
The data of Table 6 provide evidence of a tendency in the described
direction, but its strength is probably underestimated. We have already
mentioned that certain synonyms appeared frequently in both series.
But it is not to be concluded that they therefore carried the same mean-
ing. Doubtless the same terms were at times applied in the two groups
with different meanings, precisely because the subjects were under the
control of the factor being investigated. To mention one example: the
term "quiet" often occurred as a synonym of "calm" in both groups,
but the subjects may have intended a different meaning in the two cases.
For this reason Table 6 may not reveal the full extent of the change
introduced by the factor of embedding.
The preceding experiments permit the following conclusions:
1. There is a process of discrimination between central and pe-
ripheral traits. All traits do not have the same rank and value in the
final impression. The change of a central trait may completely alter
the impression, while the change of a peripheral trait has a far weaker
effect (experiments 1, 2, and 3).
2. Both the cognitive content of a trait and its functional value are
determined in relation to its surroundings (experiment 4).
3. Some traits determine both the content and the function of other
traits. The former we call central, the latter peripheral (experiment 4) .
H. THE FACTOR OF DIRECTION
If impressions of the kind here investigated are a summation of the
effects of the separate characteristics, then an identical set of charac-
teristics should produce a constant result. Is it possible to alter the im-
pression without changing the particular characteristic? We investigate
this question below.
Experiment 6
The following series are read, each to a different group:
A. intelligent industrious impulsive critical stubborn envious
B. envious stubborn critical impulsive industrious intelligent
There were 34 subjects in Group A, 24 in Group B.
The two series are identical with regard to their members, differing
only in the order of succession of the latter. More particularly, Series A
opens with qualities of high merit (intelligent industrious), proceeds
to qualities that permit of a better or poorer evaluation (impulsive
Forming Impressions of Personality : 255
critical stubborn), and closes with a dubious quality (envious). This
order is reversed in Series B.
A considerable difference develops between the two groups taken
as a whole. The impression produced by A is predominantly that of an
able person who possesses certain shortcomings which do not, however,
overshadow his merits. On the other hand, B impresses the majority as
a "problem," whose abilities are hampered by his serious difficulties.
Further, some of the qualities (e.g., impulsiveness, criticalness) are
interpreted in a positive way under Condition A, while they take on,
under Condition B, a negative color. This trend is not observed in all
subjects, but it is found in the majority. A few illustrative extracts
follow:
Series A
A person who knows what he wants and goes after it. He is impatient at
people who are less gifted, and ambitious with those who stand in his way.
Is a forceful person, has his own convictions and is usually right about
things. Is self-centered and desires his own way.
The person is intelligent and fortunately he puts his intelligence to work.
That he is stubborn and impulsive may be due to the fact that he knows what
he is saying and what he means and will not therefore give in easily to some-
one else's idea which he disagrees with.
Series B
This person's good qualities such as industry and intelligence are bound to
be restricted by jealousy and stubbornness. The person is emotional. He is
unsuccessful because he is weak and allows his bad points to cover up his
good ones.
This individual is probably maladjusted because he is envious and im-
pulsive.
In order to observe more directly the transition in question, the
writer proceeded as follows. A new group (N = 24) heard Series B,
wrote the free sketch, and immediately thereafter wrote the sketch in
response to Series A. They were also asked to comment on the relation
between the two impressions. Under these conditions, with the transi-
tion occurring in the same subjects, 14 out of 24 claimed that their
impression suffered a change, while the remaining 10 subjects reported
no change. Some of the latter asserted that they had waited until the
256 : Solomon E. Asch
entire series was read before deciding upon their impression. The fol-
lowing are a few comments of the changing group:
You read the list in a different order and thereby caused a different type of
person to come to mind. This one is smarter, more likeable, a go-getter,
lively, headstrong, and with a will of his own; he goes after what he wants.
The first individual seems to show his envy and criticism more than the
second one.
This man does not seem so bad as the first one. Somehow, he seems more
intelligent, with his critical attitude helping that characteristic of intelligence,
and he seems to be industrious, perhaps because he is envious and wants to
get ahead.
The check-list data appearing in Table 7 furnish quantitative sup-
port for the conclusions drawn from the written sketches.
TABLE 7
CHOICE OF FITTING QUALITIES (PERCENTAGES)
Experiment 6
Experiment 7
Intelligent -
Envious
(N = 34)
Envious -
Intelligent
(JV = 24)
Intelligent -
Evasive
(AT = 46)
Evasive -
Intelligent
(JV = 53)
1. generous
24
10
42
23
2. wise
18
17
35
19
3. happy
32
5
51
49
4. good-natured
18
54
37
5. humorous
52
21
53
29
6. sociable
56
27
50
48
7. popular
35
14
44
39
8. reliable
84
91
96
94
9. important
85
90
77
89
10. humane
36
21
49
46
11. good-looking
74
35
59
53
12. persistent
82
87
94
100
13. serious
97
100
44
100
14. restrained
64
9
91
91
15. altruistic
6
5
32
25
16. imaginative
26
14
37
16
17. strong
94
73
74
96
18. honest
80
79
66
81
Forming Impressions of Personality : 257
Under the given conditions the terms, the elements of the description,
are identical, but the resulting impressions frequently are not the same.
Further, the relations of the terms to one another have not been dis-
turbed, as they may have been in experiments 1 and 2, with the addi-
tion and omission of parts. How can we understand the resulting
difference?
The accounts of the subjects suggest that the first terms set up in
most subjects a direction which then exerts a continuous effect on the
latter terms. When the subject hears the first term, a broad, uncrystal-
lized but directed impression is born. The next characteristic comes
not as a separate item, but is related to the established direction.
Quickly the view formed acquires a certain stability, so that later char-
acteristics are fitted if conditions permit 5 to the given direction.
Here we observe a factor of primacy guiding the development of an
impression. This factor is not, however, to be understood in the sense
of Ebbinghaus, but rather in a structural sense. It is not the sheer tem-
poral position of the item which is important as much as the functional
relation of its content to the content of the items following it. 6
TABLE 8
RANKING OF "ENVIOUS": EXPERIMENT 6
Intelligent -> Envious Envious - Intelligent
Rank
N
Percentage
N
Percentage
1
2
3
4
5
6
5
4
5
3
4
13
15
11
15
9
11
39
7
4
5
2
2
4
29
17
21
8
8
17
34
100
24
100
6 For an instance in which the given conditions may destroy the established
direction, see experiment 8 below.
6 In accordance with this interpretation the effect of primacy should be abol-
ished or reversed if it does not stand in a fitting relation to the succeeding
qualities, or if a certain quality stands out as central despite its position. The latter
was clearly the case for the quality "warm-cold" in experiment 1 (see Table 1 )
which, though occupying a middle position, ranked comparatively high.
The distinction between the two senses of primacy could be studied experi-
mentally by comparing the recall of an identical series of character qualities in
255 : Solomon E. Asch
Some further evidence with regard to this point is provided by the
data with regard to ranking. We reproduce in Table 8 the rankings of
the characteristic "envious" under the two conditions.
Experiment 7
It seemed desirable to repeat the preceding experiment with a new
series. As before, we reversed the succession of terms. Unlike the pre-
ceding series, there is no gradual change in the merit of the given char-
acteristics, but rather the abrupt introduction at the end (or at the
beginning) of a highly dubious trait* The series were:
A. intelligent skillful industrious determined practical cautious
evasive
B. evasive cautious practical determined industrious skillful
intelligent
While the results are, for reasons to be described, less clear than in
the experiment preceding, there is still a definite tendency for A to
produce a more favorable impression with greater frequency. We
report below the more extreme protocols in each series.
Series A
He seems to be a man of very excellent character, though it is not unusual
for one person to have all of those good qualities.
A scientist in an applied field, who does not like to discuss his work before
it is completed. Retiring and careful but brilliant. Works alone, does not
like to be annoyed with questions. A very dynamic man.
A normal, intelligent person, who sounds as if he would be a good citizen,
and of value to all who know him.
He seems to have at least two traits which are not consistent with the rest
of his personality. Being cautious and evasive contradicts his positive quali-
ties. Altogether, he is a most unattractive person the two above-mentioned
traits overbalancing the others.
Series B
This is a man who has had to work for everything he wanted therefore
he is evasive, cautious and practical. He is naturally intelligent, but his strug-
gles have made him hard.
two groups, one of which reads them as a discrete list of terms, the other as a
set of characteristics describing a person.
Forming Impressions of Personality : 259
He is out for himself, is very capable but tends to use his skill for his own
benefit.
He is so determined to succeed that he relies on any means, making use
of his cunning and evasive powers.
Questioning disclosed that, under the given conditions, the quality
"evasive" produced unusual difficulty. Most subjects in both groups
felt a contradiction between it and the series as a whole. In response to
the question, "Were there any characteristics that did not fit with the
others?" 1 1 out of 27 in Group A mentioned "evasive" while it was
mentioned by 1 1 out of a total of 30 in Group B.
It is of interest to observe how this crucial term was dealt with by
individual subjects. Some in Group A felt unable to reconcile it with the
view they had formed; consequently they relegated it to a subsidiary
position and, in the most extreme cases, completely excluded it. Others
reported the opposite effect: the final term completely undid their im-
pression and forced a new view. The following comments are illus-
trative:
Series A
I put this characteristic in the background and said it may be a dependent
characteristic of the person, which does not dominate his personality, and
does not influence his actions to a large extent.
I excluded it because the other characteristics which fitted together so well
were so much more predominant. In my first impression it was left out
completely.
It changed my entire idea of the person changing his attitude toward
others, the type of position he'd be likely to hold, the amount of happiness
he'd have and it gave a certain amount of change of character (even for
traits not mentioned), and a tendency to think of the person as somewhat
sneaky or sly.
Similar reactions occur in Group B, but with changed frequencies.
The importance of the order of impressions of a person in daily ex-
perience is a matter of general observation and is perhaps related to
the process under investigation. It may be the basis for the importance
attached to first impressions. It is a matter of general experience that
we may have a "wrong slant" on a person, because certain characteris-
tics first observed are given a central position when they are actually
subsidiary, or vice versa.
260 : Solomon E. Asch
Experiment 8
We studied the factor of direction in yet another way. Series A of
experiment 6 was divided in two parts and presented to a new group
as a description of two persons. The new series were:
A, intelligent industrious impulsive
B. critical stubborn envious
Procedure. ( 1 ) Series A was read to this group (Group 1 ) , followed
by the written sketch and the check list. (2) The subjects were in-
structed that they would hear a new group of terms describing a second
person. Series B was read and the usual information was obtained. (3)
Upon completion of the second task the subjects were informed that
the two lists described a single person. They were instructed to form
an impression corresponding to the entire list of terms. Certain ques-
tions were subsequently asked concerning the last step, which will be
described below. A control group (Group 2) responded only to the
entire list of six terms (as in Series A of experiment 6), and answered
some of the final questions.
We are concerned mainly to see how Group 1 dealt with the final
task, the establishing of an impression based on the two smaller series.
That Lists A and B were widely different will be clear in the check-list
results of Table 9.
Most subjects of Group 1 expressed astonishment at the final in-
formation (of Step 3) and showed some reluctance to proceed. In
response to the question, "Did you experience difficulty in forming an
impression on the basis of the six terms," the majority of Group 1 (32
out of 52) replied in the affirmative. The reasons given were highly
uniform: the two sets of traits seemed entirely contradictory.
I had seen the two sets of characteristics as opposing each other. It was
hard to envision all these contradictory traits in one person.
The person seemed to be a mass of contradictions.
He seemed a dual personality. There are two directions in this person.
On the other hand, only a minority in Group 2 (9 out of 24) report any
difficulty. Further, the reasons given by the latter are entirely different
from those of Group L These subjects speak in very general terms, as:
These characteristics are possessed by everyone in some degree or other.
The terms do not give an inclusive picture.
Forming Impressions of Personality : 261
Only two subjects in Group 2 mentioned contradiction between traits
as a source of difficulty.
TABLE 9
CHOICE OF FITTING QUALITIES: EXPERIMENT 8 (PERCENTAGES)
Intelligent Industrious Critical Stubborn
Impulsive (N = 52) Envious (N = 52)
1. generous
87
6
2. wise
48
3
3. happy
84
4. good-natured
74
3
5. humorous
87
12
6. sociable
89
24
7. popular
94
9
8. reliable
85
47
9. important
90
24
10. humane
87
19
11. good-looking
81
36
12. persistent
85
67
13. serious
87
83
14. restrained
16
37
15. altruistic
66
16. imaginative
65
15
17. strong
94
50
18. honest
100
58
The formation of the complete impression proceeds differently in
the two groups. Series A and B are at first referred, in Group 1, to
entirely different persons. Each is completed in its direction, and the
fact that they come successively seems to enhance the contrast between
them. It is therefore difficult for them to enter the new impression.
Some subjects are unable to reconcile the two directions completely; in
consequence their divergence becomes the paramount fact, as the fol-
lowing protocols illustrate:
The directions reacted on each other and were modified, so that the pull
in each direction is now less strong. This gives a Jekyll and Hyde appearance
to this person.
I applied A to the business half of the man as he appeared and acted
during working hours. B I referred to the man's social life.
262 : Solomon E. Asch
The independent development of A and B is on the other hand pre-
vented in Group 2, where they function from the start as parts of one
description. 7
This conclusion is in general confirmed by the following observa-
tion. To the question: "Did you proceed by combining the two earlier
impressions or by forming a new impression?" the following responses
are obtained: 33 of 52 subjects answer that they formed a new im-
pression, different from either A or B; 12 subjects speak of combining
the two impressions, while 7 subjects assert that they resorted to both
procedures. The following are typical responses in the first subgroup:
I couldn't combine the personalities of A and B. I formed an entirely new
impression.
I can conceive of the two sets of characteristics in one person, but I cannot
conceive of my impressions of them as belonging to one person.
As I have set down the impressions, one is exactly the opposite of the
other. But I can fit the six characteristics to one person.
That the terms of Series A and B often suffered considerable change
when they were viewed as part of one series becomes evident in the
replies to another question. The subjects were asked, "Did the terms of
the series A and B retain for you their first meaning or did they
change?" Most subjects describe a change in one or more of the traits,
of which the following are representative:
In A impulsive grew out of imaginativeness; now it has more the quality
of hastiness.
Industriousness becomes more self-centered.
Critical is now not a derisive but rather a constructive activity.
Stubborn had an entirely personal meaning; now it refers to being set in
one's ideas.
The tenor of most replies is well represented by the following com-
ment:
7 The procedure of "successive impressions" here employed might be extended
to the study of the effect of early upon later impressions. For example, the ini-
pression resulting from the sequence (A) + (B) might be compared with the
reverse sequence (B) + (A), and each of these with the sequence (A -f B) or
/T> I A *\
(B+A).
Forming Impressions of Personality : 263
When the two came together, a modification occurred as well as a limiting
boundary to the qualities to which each was referred.
m. STRONGLY SIMPLIFIED IMPRESSIONS
To a marked degree the impressions here examined possess a strongly
unified character. At the same time they lack the nuances and dis-
criminations that a full-fledged understanding of another person pro-
vides. Therefore they can be easily dominated by a single direction. We
propose now to observe in a more direct and extreme manner the
formation of a global impression.
Experiment 9
We select from the series of experiment 1 three terms, intelligent
skillful -warm, all referring to strong positive characteristics. These
form the basis of judgment. The results appear in the first column of
figures in Table 10.
TABLE 10
CHOICE OF FITTING QUALITIES: EXPERIMENT 9 (PERCENTAGES)
Intelligent
Skillful Warm
(N = 34)
Warm
(AT = 22)
Cold
(N = 33)
1. generous
100
100
12
2. wise
97
95
11
3. happy
100
100
10
4. good-natured
100
100
8
5. humorous
100
100
12
6. sociable
100
100
9
7. popular
100
100
6
8. reliable
100
100
87
9. important
84
68
54
10. humane
97
100
17
11. good-looking
72
95
57
12. persistent
100
78
97
13. serious
100
68
97
14. restrained
66
41
97
15. altruistic
97
91
3
16. imaginative
82
95
9
17. strong
97
74
87
18. honest
100
100
81
264 : Solomon E. Asch
There develops a one-directed impression, far stronger than any
observed in the preceding experiments. The written sketches, too, are
unanimously enthusiastic. The impression also develops effortlessly.
Negative characteristics hardly intrude. That this fails to happen
raises a problem. Many negative qualities could quite understandably
be living together with those given. But the subjects do not as a rule
complete them in this direction. This, indeed, they seem to avoid.
Experiment 9a
The next step was to observe an impression based on a single trait.
There are two groups; one group is instructed to select from the check
list those characteristics which belong to a "warm" person, the second
group those belonging to a "cold" person. The results appear in the
last two columns of Table 10.
In order to show more clearly the range of qualities affected by the
given terms, we constructed a second check Ust (Check List II) to
which the subjects were to respond in the manner already described.
The results are reported in Table 1 1 .
A remarkably wide range of qualities is embraced in the dimension
"warm-cold." It has reference to temperamental characteristics (e.g.,
TABLE 11
CHECK LIST II: CHOICE OF FITTING QUALITIES:
EXPERIMENT 9a (PERCENTAGES)
Warm
(N = 22)
Cold
(N = 33)
C
Warm
Cold
(N = 33)
1. emotional
100
12
unemotional
88
2. practical
40
73
theoretical
60
27
3. optimistic
95
17
pessimistic
5
83
4. informal
95
formal
5
100
5. cheerful
100
18
sad
82
6. short
91
8
tafl
9
92
7. modest
86
9
proud
14
91
8. imaginative
95
28
unimaginative
5
72
9. thin
15
93
stout
85
7
10. intelligent
81
96
unintelligent
19
4
11. brave
91
74
cowardly
9
26
12. pale
15
97
ruddy
85
3
Forming Impressions of Personality : 265
optimism, humor, happiness), to basic relations to the group (e.g.,
generosity, sociability, popularity), to strength of character (e.g., per-
sistence, honesty). It even includes a reference to physical characteris-
tics, evident in the virtually unanimous characterizations of the warm
person as short, stout, and ruddy, and in the opposed characterizations
of the cold person.
The differences between "warm" and "cold" are now even more con-
siderable than those observed in experiment 1. No qualities remain
untouched. But even under these extreme conditions the characteriza-
tions do not become indiscriminately positive or negative. "Warm"
stands for very positive qualities, but it also carries the sense of a certain
easy-goingness, of a lack of restraint and persistence, qualities which
are eminently present in "cold." A simplified impression is not to be
simply identified with a failure to make distinctions or qualifications.
Rather, what we find is that in a global view the distinctions are drawn
bluntly.
The consistent tendency for the distribution of choices to be less ex-
treme in experiment 1 requires the revision of an earlier formulation.
We have said that central qualities determine the content and func-
tional value of peripheral qualities. It can now be seen that the central
characteristics, while imposing their direction upon the total impres-
sion, were themselves affected by the surrounding characteristics.
Upon the conclusion of the experiments, the subjects were asked to
state the reason for their choice of one predominant direction in their
characterizations. All agreed that they felt such a tendency. Some can-
not explain it, saying, in the words of one subject: "I do not know the
reason; only that this is the way it 'hit' me at the moment"; or: "I did
not consciously mean to choose the positive traits." Most subjects, how-
ever, are explicit in stating that the given traits seemed to require com-
pletion in one direction. The following statements are representative:
These qualities initiate other qualities. A man who is warm would be
friendly, consequently happy. If he is intelligent, he would be honest.
The given characteristics, though very general, were good characteristics.
Therefore other good characteristics seemed to belong. When, for example,
I think of a person as warm, I mean that he couldn't be ugly.
This was the tenor of most statements. A few show factors at work of
a somewhat different kind, of interest to the student of personality, as:
266 : Solomon E. Asch
I naturally picked the best trait because I hoped the person would be that
way.
I went in the positive direction because 7 would like to be all those things.
It is of interest for the theory of our problem that there are terms
which simultaneously contain implications for wide regions of the per-
son. Many terms denoting personal characteristics show the same prop-
erty. They do not observe a strict division of labor, each pointing neatly
to one specific characteristic; rather, each sweeps over a wide area and
affects it in a definite manner. 8
Some would say that this is a semantic problem. To do so would be,
however, to beg the question by disposing of the psychological process
that gives rise to the semantic problem. What requires explanation is
how a term, and a highly "subjective" one at that, refers so consistently
to so wide a region of personal qualities. It seems similarly unfruitful
to call these judgments stereotypes. The meaning of stereotype is itself
badly in need of psychological clarification. Indeed, in the light of our
observations, a stereotype appears (in a first approximation) to be a
central quality belonging to an extremely simplified impression.
We propose that there is, under the given conditions, a tendency to
grasp the characteristics in their most outspoken, most unqualified
sense, and on that basis to complete the impression. The subject aims
at a clear view;- he therefore takes the given terms in their most com-
plete sense. (What is said here with regard to the present experiment
seems to apply also to the preceding experiments. In each case the sub-
ject's impression is a blunt, definite characterization. It lacks depth but
not definiteness. Even when the view is of a mediocre character, it is
outspokenly so.) The comments of the subjects are in agreement with
the present interpretation.
8 On the basis of the last findings an objection might be advanced against our
earlier account of the distinction between central and peripheral traits. If, as has
just been shown, "warm" refers to such a wide range of qualities, then the force
of the demonstration (see experiment 1) that it exerts a great effect on the final
impression seems to be endangered. Is it to be wondered at that this quality,
which is single only in a linguistic sense, but psychologically plural, should be so
effective? And should not the distinction be drawn rather between qualities which
contain many other qualities and qualities such as "politeness" that are much
more specific in range?
The objection presupposes that a quantitatively larger number of qualities will
exert a greater effect than a smaller number. But this assumption is precisely what
needs to be explained. Why does not the more inclusive term provide a greater
number of occasions for being affected by other terms? What the assertion fails
to face is that there is a particular direction of forces.
Forming Impressions of Personality : 267
IV. SIMILARITY AND DIFFERENCE OF IMPRESSIONS
The preceding discussion has definite consequences for the perception
of identity and difference between the characteristics of different per-
sons. Of these the most significant for theory is the proposition that a
given trait in two different persons may not be the same trait, and con-
trariwise, that two different traits may be functionally identical in two
different persons. We turn now to an investigation of some conditions
which determine similarity and difference between personal qualities.
Experiment 10
I. The group has before it Sets 1, 2, 3, and 4 with instructions to
state (1) which of the other three sets most resembles Set 1, and (2)
which most resembles Set 2.
Setl Set 2 Set 3 Set 4
quick quick slow slow
skillful clumsy skillful clumsy
helpful helpful helpful helpful
One quality "helpful" remains constant in all sets. The other two
qualities appear in their positive form in Set 1, and are changed to their
opposites singly and together in the three other sets.
A remarkable uniformity appears in the findings, reported in Table
12.
TABLE 12
RESEMBLANCE OF SETS: EXPERIMENT 10
Set 1 Resembles Set 2 Resembles
Set N
Percentage
N
Percentage
1
2 10
3 68
4
13
87
7
5
66
9
6
85
78
100
78
100
Set 1 is equated with Set 3 in 87 per cent of the cases, while its simi-
larity to Set 2 is reported in only 13 per cent of the cases. Similarly, Set
2 is asserted to resemble Set 4 in 85 per cent of the cases, while the
resemblance to Set 1 drops to 9 per cent.
265 : Solomon E. Asch
The choice of similar sets cannot in this case be determined merely
on the basis of the number of "identical elements," for on this criterion
Sets 2 and 3 are equally similar to 1, while Sets 1 and 4 are equally
similar to 2. What factors may be said to determine the decisions with
regard to similarity and difference?
We come somewhat closer to an answer in the replies to the following
question: "Which characteristics in the other sets resemble most closely
(a) 'quick' of Set 1? (b) 'quick' of Set 2? (c) 'helpful' of Set 1? (d)
'helpful' of Set 2?" The results appear in Table 13.
TABLE 13
RESEMBLANCE OF QUALITIES: EXPERIMENT 10
"Quick" of Set 1 Resembles
"Quick" of Set 2 Resembles
N
Percentage
N
Percentage
"quick" of set 2
11
22
"quick" of set 1
10
24
"helpful" of set 2
1
2
"slow" of set 4
21
51
"slow" of set 3
16
32
"clumsy" of set 4
7
17
"skillful" of set 3
21
42
"slow" of set 3
2
5
"helpful" of set 3
1
2
"helpful" of set 3
1
3
50
100
41
100
"Helpful" of Set 1 Resembles
"Helpful" of Set 2 Resembles
N
Percentage
N
Percentage
"helpful" of set 2
7
15
"helpful" of set 1
5
11
"helpful" of set 4
2
4
"quick" of set 1
2
4
"helpful" of set 3
33
68
"slow" of set 4
2
4
"skillful" of set 3
6
13
"helpful" of set 4
30
65
"clumsy" of set 4
4
9
"slow" of set 3
3
7
48
100
46
100
We see that qualities which, abstractly taken, are identical, are in-
frequently equated, while qualities which are abstractly opposed are
equated with greater frequency. For example, the quality "quick" of
Sets 1 and 2 is matched in only 22 and 24 per cent of the cases, re-
spectively, while "quick" of Set 1 is, in 32 per cent of the cases, matched
Forming Impressions of Personality : 269
with "slow" of Set 3, and "quick" of Set 2 with "slow" of Set 4 in 51
per cent of the cases. 9
At this point the reports of the subjects become very helpful. They
were requested at the conclusion to state in writing whether the quality
"quick" in Sets 1 and 2 was identical or different, together with their
reasons, and similarly to compare the quality "slow" in Sets 3 and 4.
The written accounts permit of certain conclusions, which are stated
below.
The content of the quality changes with a change in its environment.
The protocols below, which are typical, will show that the "quicks" of
Sets 1 and 2 are phenomenally different, and similarly for the "slows"
of Sets 3 and 4.
The quickness of 1 is one of assurance, of smoothness of movement; that
of 2 is a forced quickness, in an effort to be helpful.
1 is fast in a smooth, easy-flowing way; the other (2) is quick in a bustling
way the kind that rushes up immediately at your request and tips over the
lamps.
3 takes his time in a deliberate way; 4 would like to work quickly, but
cannot there is something painful in his slowness.
3 is slow in a methodical, sure way, aiming toward perfection; in 4 it im-
plies a certain heaviness, torpor.
The dynamic sources of the quality are relationally determined.
In the protocols we observe a process of mutual determination between
traits. They are grasped as not simply contiguous to one another but
in dynamic relation, in which one is determined by, or springs from,
the other.
1 is quick because he is skillful; 2 is clumsy because he is so fast.
Great skill gave rise to the speed of 1, whereas 2 is clumsy because he does
everything so quickly.
9 In a forthcoming publication the writer will deal with theoretically similar
issues in the context of a problem in social psychology. This will be the report
of an investigation of changes in the content of identical social assertions when
they function as part of different frames of reference. [S. E. Asch, Social psy-
chology. New York: Prentice-Hall, 1952. Chap. 15. Ed.]
270 : Solomon E. Asch
The quality slow is, in person 3, something deliberately cultivated, in order
to attain a higher order of skill.
In 3 slowness indicates care, pride in work weE done. Slowness in 4 indi-
cates sluggishness, poor motor coordination, some physical retardation.
Speed and skill are not connected as are speed and clumsiness. With-
out exception, "quick" is perceived to spring from skill (skillful ->
quick) ; but the vector in Set 2 is reversed, "clumsy" becoming a con-
sequence of speed (clumsy < quick). While Sets 1 and 3 are identical
with regard to the vectors, Set 2 is not equivalent to 4, the slowness and
clumsiness of 4 being sensed as part of a single process, such as slug-
gishness and general retardation (slow ^ clumsy).
Dynamic consequences are grasped in the interaction of qualities.
"Quick" and "skillful" (as well as "slow" and "skillful") are felt as
cooperating, whereas "quick" and "clumsy" cancel one another.
2 drops everything fast. He is fast but accomplishes nothing. The clumsy
man might be better off if he were slow.
The second person is futile; he is quick to come to your aid and also quick
to get in your way and under your hair.
1 can afford to be quick; 2 would be far better off if he took things more
slowly. 10
In the light of these comments, which are representative, we are able
to formulate the prevailing direction of the relations within the sets.
In Sets 1 and 3 the prevailing structure may be represented as:
helpful helpful
quick skillful slow skillful
"Quick-slow" derive their concrete character from the quality "skill-
ful"; these in turn stand in a relation of harmony to "helpful," in the
sense that they form a proper basis for it and make it possible.
10 Parallel experiments in which the last term of the sets was changed to "not
helpful" gave results essentially identical with the above.
Forming Impressions of Personality : 271
In Sets 2 and 4 the characteristic structures are as follows:
helpful helpful
4
quick clumsy slow
But now these stand in a relation of inherent contradiction to the
quality "helpful," the fulfillment of which they negate.
Our results contain a proportion of cases (see Tables 12 and 13)
that are contrary to the described general trend. These do equate the
characteristic of 1 and 2 and of 3 and 4. They require explanation. It
is especially important to decide whether the disagreements are capri-
cious or whether they have an understandable basis. As a rule we find
in these cases that the given quality is viewed in a narrower, more lim-
ited way. For example, these subjects view "quick" of Sets 1 and 2 in
terms of sheer tempo, deliberately excluding for the moment considera-
tions of fitness. The following protocols are illustrative:
These persons' reactions to stimuli are both quick, even though the results
of their actions are in opposite directions.
They are both quick, but they differ in the success of their actions.
The two terms are basically the same, for both would execute their tasks
with their individual maximum speed.
II. The reader will readily think of other sets of characteristics in-
volving similar processes. In view of the fact that such analyses have
not been previously reported, we select for brief description a few
additional examples.
The task was to state whether the term "aggressive" was alike or
different in Sets 1 and 2, and 3 and 4, respectively. This example will
be of particular interest to psychologists, in view of current discussions
of aggressiveness.
Setl Set 2 Set 3 Set 4
active
helpful
aggressive
lazy
unhelpful
aggressive
weak
sensitive
aggressive
strong
self-centered
aggressive
4
stand
272 : Solomon E. Asch
Nineteen out of 20 subjects judge the term to be different in Sets 1
and 2; 17 out of 20 judge it to be different in Sets 3 and 4. Some repre-
sentative reports follow:
The aggressiveness of 1 is friendly, open, and forceful; 2 will be aggressive
when something offends him.
The aggressiveness of 1 is an expression of confidence in his abilities, of
his strength of will and mind; in 2 it is a defensive measure to cover sensitivity.
3 will be aggressive to try to hide his weakness. The aggressiveness of 4 is
a natural result of his strength and self-centeredness.
is aggressive because he has needs to be satisfied and wishes nothing to
d in his way; 3 has the aggressiveness of self-pity and indecision.
In nearly all cases the sources of aggression and its objects are sensed
to be different. In consequence, the form it takes and its very psycho-
logical content become different in the series compared.
Substantially the same results are observed in another group in the
comparison of "unaggressive" in Sets 1 and 2 below.
Set 1 Set 2
active weak
helpful sensitive
unaggressive unaggressive
Twenty-eight out of 30 subjects call "unaggressive" different in the
two series. Some of their reasons follow:
Unaggressive in 1 might mean that he does not push or force his way into
things. In the second case it may mean meekness or fear of people.
1 does not care to be aggressive; 2 lacks the stamina for it.
2 does not fight back at the world nor try to rise above his weaknesses.
The word "aggressive" must have the same connotations in both cases;
otherwise why not use different terms to express different things?
HI. The second and third terms in Sets 1 and 2 below were com-
pared, respectively.
Forming Impressions of Personality : 273
Setl
Set 2
intelligent
critical
impulsive
critical
stubborn
stubborn
All subjects in a group of 3 1 judged the term "critical" to be different
in the two sets; while 19 (or 61 per cent) judged "stubborn" as differ-
ent. A few of the remarks follow:
Critical:
1 is critical because he is intelligent; 2 because he is impulsive.
The intelligent individual is critical in a constructive manner; the im-
pulsive one probably hurls criticism unthinkingly.
The intelligent person may be critical in a completely impersonal way; 2
may be critical of people, their actions, their dress, etc.
Stubborn:
The stubbornness of an intelligent person is more likely to be based on
reason and it can be affected by reasoning.
The intelligent person might be stubborn about important things, things
that mean something to him, that he knows something about; whereas an im-
pulsive person might be stubborn just to be contrary.
An intelligent person may be stubborn because he has a reason for it and
thinks it's the best thing to do, while an impulsive person may be stubborn
because at the moment he feels like it.
Some representative statements defending the identity of "stubborn" in
the two series follow:
Stubbornness to me is the same in any language. Of course, an intelligent
person may have a better reason for being stubborn than an impulsive one,
but that does not necessarily change the degree of stubbornness.
Both refuse to admit to anything that does not coincide with their opinion.
In my opinion there is only one kind of stubbornness an unswerving
desire either to do or not to do a certain thing.
IV. In the following series the second and third terms were to be
compared:
274 : Solomon E. Asch
Set 1 Set 2
warm cold
witty witty
persuasive persuasive
Twenty-seven of 30 subjects judged "persuasive" as different; all
judged "witty" to be different. A few of the comments follow:
Witty:
1 laughs with the audience; 2 is either laughing at or trying to make others
laugh at some one. 2 is satirical, not humorous.
1 has a jolly and happy-go-lucky wit. 2 will use wit as one uses a bow and
arrow with precision. He will have a target which will not be missed.
The wit of the warm person touches the heart. The cold person's wit is
touched with irony.
Persuasive:
1 is persuasive in trying to help others; 2 in trying to help himself.
2 may persuade through fear.
2 would be detached in his arguments; 1 would appeal more to the inner
emotional being of others.
V. The term "gay" was compared in the following series:
Setl Set 2
intelligent stupid
industrious lazy
Twenty-seven of 30 subjects call "gay" different. Some representa-
tive reasons follow:
They may both be equally gay, but the former is different. The stupid
person can be gay over serious, sad matters, while the intelligent person is
gay with reason.
The first person's gaiety comes from fullness of life; 2 is gay because he
knows no better.
1 knows when to be gay and when not to be.
The gaiety of 1 is active and energetic; the gaiety of 2 is passive.
Forming Impressions of Personality : 275
The intelligent person is gay in an intelligent way.
They are the same gaiety has no relation to intelligence and industrious-
ness.
The foregoing observations describe a process of relational deter-
mination of character qualities. A given quality derives its full concrete
content from its place within the system formed by the relations of the
qualities. Some qualities are seen as a dynamic outgrowth of determin-
ing qualities. Qualities are seen to stand in a relation of harmony or
contradiction to others within the system. These processes set require-
ments for the comparison of impressions. Identical qualities in different
structures may cease to be identical: the vectors out of which they grow
may alter, with the consequence that their very content undergoes radi-
cal change. In the extreme case, the same quality in two persons will
have different, even opposed, meanings, while two opposed qualities
will have the same function within their respective structures.
DISCUSSION
The investigations here reported have their starting point in one prob-
lem and converge on one basic conclusion. In different ways the obser-
vations have demonstrated that forming an impression is an organized
process; that characteristics are perceived in their dynamic relations;
that central qualities are discovered, leading to the distinction between
them and peripheral qualities; that relations of harmony and contra-
diction are observed. To know a person is to have a grasp of a particular
structure.
Before proceeding it may be helpful to note two preliminary points.
First: For the sake of convenience of expression we speak in this dis-
cussion of forming an impression of a person, though our observations
are restricted entirely to impressions based on descriptive materials.
We do not intend to imply that observations of actual persons would
not involve other processes which we have failed to find under the
present conditions; we are certain that they would (see pp. 282 ff.).
But we see no reason to doubt that the basic features we were able to
observe are also present in the judgment of actual persons. Secondly:
We have not dealt in this investigation with the role of individual
differences, of which the most obvious would be the effect of the sub-
ject's own personal qualities on the nature of his impression. Though
the issue of individual differences is unquestionably important, it
276 ; Solomon E. Asch
seemed desirable to turn first to those processes which hold generally,
despite individual differences. A proper study of individual differences
can best be pursued when a minimum theoretical clarification has been
reached.
Let us briefly reformulate the main points in the procedure of our
subjects:
1. There is an attempt to form an impression of the entire person.
The subject can see the person only as a unit; 11 he cannot form an im-
pression of one-half or of one-quarter of the person. This is the case
even when the factual basis is meager; the impression then strives to
become complete, reaching out toward other compatible qualities. The
subject seeks to reach the core of the person through the trait or traits.
2. As soon as two or more traits are understood to belong to one
person, they cease to exist as isolated traits, and come into immediate
dynamic interaction. 12 The subject perceives not this and that quality,
but the two entering into a particular relation. There takes place a
process of organization in the course of which the traits order them-
selves into a structure. It may be said that the traits lead an intensely
social life, striving to join each other in a closely organized system. The
representation in us of the character of another person possesses in a
striking sense certain of the qualities of a system.
3. In the course of this process some characteristics are discovered
to be central. The whole system of relations determines which will be-
come central. These set the direction for the further view of the person
and for the concretization of the dependent traits. As a rule the several
traits do not have equal weight. And it is not until we have found the
center that we experience the assurance of having come near to an
understanding of the person.
4. The single trait possesses the property of a part in a whole. A
change in a single trait may alter not that aspect alone, but many others
at times all. As soon as we isolate a trait we not only lose the dis-
tinctive organization of the person; the trait itself becomes abstract.
11 To be sure, we do often react to people in a more narrow manner, as when
we have dealings with the ticket collector or bank teller. It cannot, however, be
said that in such instances we are primarily oriented to the other as a person. The
moment our special attitude would give way to a genuine interest in the other
the point stated above would fully apply.
12 We cannot say on the basis of our observations whether exceptions to this
statement occur, e.g., whether some traits may be seen as accidental, having no
relation to the rest of the person. It seems more likely that even insignificant traits
are seen as part of the person.
Forming Impressions of Personality : 277
The trait develops its full content and weight only when it finds its place
within the whole impression.
5 . Each trait is a trait of the entire person. It refers to a characteristic
form of action or attitude which belongs to the person as a whole. In
this sense we may speak of traits as possessing the properties of Ehren-
fels qualities. Traits are not to be considered as referring to different
regions of the personality, on the analogy of geographical regions which
border on another.
6. Each trait functions as a representative of the person. We do not
experience anonymous traits, the particular organization of which con-
stitutes the identity of the person. Rather the entire person speaks
through each of his qualities, though not with the same clearness.
7. In the process of mutual interaction the concrete character of
each trait is developed in accordance with the dynamic requirements
set for it by its environment. There is involved an understanding of
necessary consequences following from certain given characteristics
for others. The envy of a proud man is, for example, seen to have a
different basis from the envy of a modest man.
8. On this basis consistencies and contradictions are discovered.
Certain qualities are seen to cooperate; others to negate each other. But
we are not content simply to note inconsistencies or to let them sit
where they are. The contradiction is puzzling, and prompts us to look
more deeply. Disturbing factors arouse a trend to maintain the unity
of the impression, to search for the most sensible way in which the
characteristics could exist together, 13 or to decide that we have not
found the key to the person. We feel that proper understanding would
eliminate, not the presence of inner tensions and inconsistencies, but
of sheer contradiction. (It may be relevant to point out that the very
sense of one trait being in contradiction to others would not arise if
we were not oriented to the entire person. Without the assumption of a
unitary person there would be just different traits.)
9. It follows that the content and functional value of a trait change
with the given context. This statement expresses for our problem a
principle formulated in Gestalt theory with regard to the identity of
parts in different structures (8, 10). A trait central in one person may
be seen as secondary in another. Or a quality which is now referred to
13 Indeed, the perception of such contradiction, or of the failure of a trait to
fit to the others, may be of fundamental importance for gaining a proper view.
It may point to a critical region in the person, in which things are not as they
should be.
278 : Solomon E. Asch
the person may in another case be referred to outer conditions. (In the
extreme case a quality may be neglected, because it does not touch
what is important in the person.)
We conclude that the formation and change of impressions consist
of specific processes of organization. Further, it seems probable that
these processes are not specific to impressions of persons alone. It is a
task for future investigation to determine whether processes of this
order are at work in other important regions of psychology, such as in
forming the view of a group, or of the relations between one person
and another.
It may be of interest to relate the assumptions underlying the naive
procedure of our subjects to certain customary formulations. (1) It
should now be clear that the subjects express certain definite assump-
tions concerning the structure of a personality. The gaining of an im-
pression is for them not a process of fixing each trait in isolation and
noting its meaning. If they proceeded in this way the traits would re-
main abstract, lacking just tie content and function which make them
living traits. In effect our subjects are in glaring disagreement with the
elementaristic thesis which assumes independent traits (or traits con-
nected only in a statistical sense) of constant content. (2) At the same
time the procedure of our subjects departs from another customary
formulation. It is equally far from the observed facts to describe the
process as the forming of a homogeneous, undifferentiated "general im-
pression." The unity perceived by the observer contains groupings the
parts of which are in more intimate connection with each other than
they are with parts of other groupings. 14 Discrimination of different
aspects of the person and distinctions of a functional order are essential
parts of the process. We may even distinguish different degrees of unity
in persons. Increasing clearness in understanding another depends on
the increased articulation of these distinctions. But in the process these
continue to have the properties of parts in a single structure.
If we may for the purpose of discussion assume that the naive pro-
cedure is based on a sound conception of the structure of personality,
it would by no means follow that it is therefore free from misconcep-
tions and distortions. But in that case the nature of errors in judgment
i* jf we ma y assume that the situation in the observed person corresponds to
this view, an important conclusion follows for method, namely, that we can
study characteristics of persons without an exhaustive knowledge of the entire
person.
Forming Impressions of Personality : 279
would have to be understood in a particular way. It would be necessary
to derive the errors from characteristics of the organizational processes
in judgment. The present investigation is not without some hints for
this problem. It points to the danger of forcing the subject to judge
artificially isolated traits a procedure almost universally followed in
rating studies and to the necessity of providing optimal conditions for
judging the place and weight of a characteristic within the person (un-
less of course the judgment of isolated traits is required by the particu-
lar problem). Under such conditions we might discover an improve-
ment in the quality of judgment and in agreement between judges. At
the same time this investigation contains some suggestions for the study
of errors in factors such as oversimplification leading to "too good" an
impression, viewing a trait outside its context or in an inappropriate
context.
Returning to the main theoretical conceptions described earlier (see
pp. 238-240) it is necessary to mention a variant of Proposition I,
which we have failed so far to consider and in relation to which we
will be able to state more precisely a central feature of Proposition II.
It would be a possible hypothesis that in the course of forming an im-
pression each trait interacts with one or more of the others, and that
the total impression is the summation of these effects. The impression
would accordingly be derived from the separate interaction of the com-
ponents, which might be represented as follows:
a + c
16. Impression^ a ^ d or a^^fcT^c^ ~*VT+ e
a = e
b + c
etc.
It is important to note that this formulation is in a fundamental re-
gard different from Proposition II. The latter proposition asserts that
each trait is seen to stand in a particular relation to the others as part
of a complete view. The entire view possesses the formal properties of
a structure, the form of which cannot be derived from the summation
of the individual relations. 15 In the same manner that the content of
15 For a basic treatment of the concept of structure the reader is referred to
M.Wertheimer(lO).
280 : Solomon E. Asch
each of a pair of traits can be determined fully only by reference to
their mutual relation, so the content of each relation can be determined
fully only with reference to the structure of relations of which it is a
part. This we may illustrate with the example of a geometrical figure
such as a pyramid, each part of which (e.g., the vertex) implicitly refers
to the entire figure. We would propose that this is the basis for the
discovery of central and peripheral traits and for assertions such as
that a given person is "integrated," "restricted," etc.
On the other hand, the notion of structure is denied in all proposi-
tions of the form I, including Ib. In the latter, an assumption is made
concerning the interaction of qualities, which has the effect of altering
the character of the elements. Once we have taken account of this
change, we have in the final formulation again a sum of (now changed)
elements:
Ib. Impression = a + b + c + d + e
In still another regard there is a difference between Propositions II
and Ib. This has to do with the nature of the interaction between the
traits. In terms of Proposition II the character of interaction is deter-
mined by the particular qualities that enter into the relation (e.g.,
"warm-witty" or "cold-witty"). It is doubtful, however, whether a
theory which refuses to admit relational processes in the formation of
a whole impression would admit the same relational processes in the
interaction of one trait with another.
In view of the fact that Proposition Ib has not, as far as we know,
been explicitly formulated with reference to the present problem, it
becomes necessary to do so here, and especially to state the process
of interaction in such a manner as to be consistent with it. This we
might do best by applying certain current conceptions. We could speak
of traits as "conditioned verbal reactions," each of which possesses a
particular "strength" and range of generalization. Interaction between
traits would accordingly be assimilated to the schema of differential
conditioning to single stimuli and to stimuli in combination, perhaps
after the manner of the recent treatment of "stimulus configurations"
by Hufl (4, 5). 16
16 Proceeding in the same manner, it would be possible to restate some of our
observations in terms such as the following: (1) the distinction between central
and peripheral traits would be referred to a difference between conditioned re-
actions of greater and lesser strengths; (2) the change from a central to a periph-
eral trait could be explained by the displacement of a response by other, stronger
responses; (3) the factor of direction might be dealt with in terms of changes in
the temporal appearance of stimuli; (4) strongly unified impressions could be
Forming Impressions of Personality : 281
How consistent would this interpretation be with the observations
we have reported? It seems to us that there are grave difficulties in the
way of such an interpretation. Insofar as the terms of conditioning are
at all intelligible with reference to our problem, the process of inter-
action can be understood only as a quantitative increase or diminution
in a response. This is not, however, the essential characteristic of inter-
action as we have observed it, which consists in a change of content
and function. The gaiety of an intelligent man is not more or less than
the gaiety of a stupid man; it is different in quality. Further, the condi-
tioning account seems to contain no principle that would make clear
the particular direction interaction takes.
Here we may mention a more general point. We have referred earlier
to the comparative ease with which complex situations in another per-
son are perceived. If traits were perceived separately, we would expect
to encounter the same difficulties in forming a view of a person that
we meet in learning a list of unrelated words. That we are able to en-
compass the entire person in one sweep seems to be due to the structured
character of the impression. 17 In terms of an interaction theory of com-
ponent elements, the difficulty in surveying a person should be even
greater than in the formulation of Proposition I, since the former must
deal with the elements of the latter plus a large number of added factors.
In order to retain a necessary distinction between the process of
forming an impression and the actual organization of traits in a person,
we have spoken as if nothing were known of the latter. While we can-
not deal with the latter problem, one investigation is of particular rel-
evance to the present discussion. We refer to the famous investigation
of Hartshorne and May (3), who studied in a variety of situations the
tendencies in groups of children to act honestly in such widely varied
matters as copying, returning of money, correcting one's schoolwork,
etc. The relations between the actions of children in the different situa-
tions were studied by means of statistical correlations. These were gen-
erally low. On the basis of these results the important conclusion was
drawn that qualities such as honesty are not consistent characteristics
an expression of highly generalized reactions; etc. Such formulations would,
however, fail to deal adequately with the central feature of our findings, namely,
changes in the quality of traits and the organized form of the impression.
17 It should not, however, be concluded that our views of persons are crystal
clear. In fact, they lack the precision with which we grasp a mathematical theo-
rem. We rarely feel that we have exhausted our understanding of another person.
This has partly to do with the fact that the person is in constant change.
252 : Solomon E. Asch
of the child but specific habits acquired in particular situations, that
"neither deceit, nor its opposite, honesty, are unified character traits,
but rather specific functions of life situations," Having accepted this
conclusion, equally fundamental consequences were drawn for char-
acter education of children.
Abstracting from the many things that might be said about this
work, we point out only that its conclusion is not proven because of
the failure to consider the structural character of personality traits. As
G. W. Allport ( 1, pp. 250 ff.) has pointed out, we may not assume that
a particular act, say the clandestine change by a pupil of an answer
on a school test, has the same psychological meaning in all cases. 18
Once this point is realized, its consequences for the thesis of Hartshorne
and May become quite threatening. Let us consider a few of the pos-
sibilities in the situation, which would be classified as follows by Hart-
shorne and May:
Honest:
1. The child wants to alter his answer on a test but fears he will be caught.
2. He does not change because he is indifferent to the grade.
Dishonest:
1. The child changes his answer because he is devoted to his teacher and
anxious not to lose her regard.
2. He cannot restrain the impulse to change the wrong answer into the an-
swer he now knows to be correct.
Psychologically, none of these acts are correctly classified. Further,
two of these are classified in precisely the wrong way. The child who
wishes to cheat but is afraid does not belong in the honest category,
while the child who cannot bear to leave the wrong answer uncorrected
does not necessarily deserve to be called dishonest. We do not intend
to say that the psychological significance of the reactions was as a rule
misinterpreted; for the sake of illustration we have chosen admittedly
extreme examples. But the failure to consider the psychological content
introduces a serious doubt concerning the conclusions reached by Hart-
shorne and May.
A far richer field for the observation of the processes here considered
would be the impressions formed of actual people. Concrete experience
with persons possesses a substantial quality and produces a host of ef-
18 See also discussion by D. W. MacKinnon (7, pp. 26 ff.) .
Forming Impressions of Personality : 283
fects which have no room for growth in the ephemeral impressions of
this investigation. The fact that we are ourselves changed by living
people, that we observe them in movement and growth, introduces
factors and forces of a new order. In comparison with these, momentary
impressions based on descriptions, or even the full view of the person
at a given moment, are only partial aspects of a broader process.
In such investigation some of the problems we have considered would
reappear and might gain a larger application. Other problems, which
were of necessity excluded from the present investigation, could be
clarified in such an approach. We mention one which is of particular
importance. It was a constant feature of our procedure to provide the
subject with the traits of a person; but in actual observation the dis-
covery of the traits in a person is a vital part of the process of establish-
ing an impression. Since observation gives us only concrete acts and
qualities, the application of a trait to a person becomes itself a problem.
Is characterization by a trait, for example, a statistical generalization
from a number of instances? Or is it the consequence of discovering a
quality within the setting of the entire impression, which may therefore
be reached in a single instance? In the latter case, repeated observation
would provide not simply additional instances for a statistical conclu-
sion, but rather a check on the genuineness of the earlier observation, as
well as a clarification of its limiting conditions. Proceeding in this man-
ner, it should be possible to decide whether the discovery of a trait itself
involves processes of a structural nature. Only direct investigation
based on the observation of persons can furnish answers to these ques-
tions*
In still another regard did our investigation limit the range of ob-
servation. In the views formed of living persons past experience plays a
great role. The impression itself has a history and continuity as it ex-
tends over considerable periods of time, while factors of motivation
become important in determining its stability and resistance to change.
Even within the limits of the present study factors of past experience
were highly important. When the subject formed a view on the basis
of the given description, he as a rule referred to a contemporary, at no
time to characters that may have lived in the past; he located the per-
son in this country, never in other countries. Further, experiments we
have not here reported showed unmistakably that an identical series
of traits produced distinct impressions depending on whether we iden-
tified the person as a man or woman, as a child or adult. Distinctions
of this order clearly depend on a definite kind of knowledge obtained in
284 : Solomon E. Asch
the past. Indeed, the very possibility of grasping the meaning of a trait
presupposes that it had been observed and understood.
That experience enters in these instances as a necessary factor seems
clear, but the statement would be misleading if we did not add that the
possibility of such experience itself presupposes a capacity to observe
and realize the qualities and dynamic relations here described. The as-
sertion that the properties of the impression depend on past experience
can only mean that these were once directly perceived. In this connec-
tion we may refer to certain observations of Kohler (6, pp. 234 flE.)
concerning our understanding of feelings in others which we have not
observed in ourselves, or in the absence of relevant previous experi-
ences. In his comprehensive discussion of the question, G. W. Allport
(1, pp. 533 flE.) has equally stressed the importance of direct percep-
tion of a given structure in others, of our capacity for perceiving in
others dynamic tendencies.
Nor do we consider it adequate to assert that in the present investiga-
tion our subjects were merely reproducing past observations of qualities
and of the ways in which they modify each other. When the subject se-
lected a certain trait as central (or when he deposed a once central trait
to a minor role within a new context) it is by no means clear that he was
guided by specific, acquired rules prescribing which traits will be cen-
tral in each of a great number of constellations. It seems more in ac-
cordance with the evidence to suppose that the system of the traits itself
points to a necessary center. And as we have mentioned earlier, the
interaction between two traits already presupposes that we have dis-
covered whether in the past or in the present the forces that work
between them. Given the quality "quick" we cannot unequivocally infer
the quality "skillful"; 19 but given "quick-skillful" we try to see how one
grows out of the other. We then discover a certain constancy in the
relation between them, which is not that of a constant habitual con-
nection.
While an appeal to past experience cannot supplant the direct grasp-
ing of qualities and processes, the role of past experience is undoubt-
edly great where impressions of actual people extending over a long pe-
riod are concerned. Here the important question for theory is whether
the factors of past experience involve dynamic processes of the same
order that we find at work in the momentary impression, or whether
19 That it is at times difficult to infer qualities on the basis of central traits is
due to such factors as the lability of the person, the degree to which the actions of
a person are directed by a single center, as well as situational forces.
Forming Impressions of Personality : 285
these are predominantly of the nature of associative bonds. It seems to
us a useful hypothesis that when we relate a person's past to his present
we are again relying essentially on the comprehension of dynamic
processes.
REFERENCES
1. Allport, G. W. Personality; a psychological interpretation. New York:
Holt, 1937.
2. Asch, S. E. Studies in the principles of judgments and attitudes: II.
Determination of judgments by group and by ego standards. /. soc.
PsychoL, 1940, 12, 433-465.
3. Hartshorne, H., and May, M. A. Studies in deceit. New York: Macmillan,
1928; Hartshorne, H., May, M. A., and Mailer, J. B. Studies in service
and self-control New York: Macmillan, 1929; Hartshorne, H., May,
M. A., and Shuttleworth, F. K. Studies in the organization of character.
New York: Macmillan, 1930. The three constitute Studies in the nature
of character by the Character Education Inquiry, Teachers College,
Columbia University, in cooperation with the Institute of Social and
Religious Research.
4. Hull, C. L. Principles of behavior. New York: Appleton-Century, 1943.
5. Hull, C. L. The discrimination of stimulus configurations and the hypoth-
esis of afferent neural interaction. PsychoL Rev., 1945, 52, 133-142.
6. Kohler, W. Gestalt psychology. New York: Liveright, 1929.
7. MacKinnon, D. W. The structure of personality. In J. McV. Hunt (Ed.) ,
Personality and the behavior disorders, Vol. I. New York: Ronald
Press, 1944.
8. Ternus, J. Experimentelle Untersuchungen liber phanornenale Identitat.
Psych. Forsch., 1926, 7, 81-136.
9. Thorndike, E. L. A constant error in psychological rating. /. appl.
PsychoL, 1920, 4, 25-29.
10. Wertheimer, M. Productive thinking. New York: Harper, 1945.
Mary Henle
GRADUATE FACULTY, NEW SCHOOL FOR SOCIAL RESEARCH
ON FIELD FORCES
INTRODUCTION
Although a number of authors have called attention to field forces, the
concept has found no place in motivational theory. It is the purpose of
the present paper to describe a method of demonstrating field forces
and to indicate their importance for a psychology of motivation.
The term "field forces" is here used to refer to motivational forces
arising from parts of the psychological field other than the individual.
Since the concept has not found general acceptance, I shall start with
a brief discussion of it and with examples of the operation of field forces.
Kohler (10) has pointed to the necessity for distinguishing between
the locus and the reference of experiences. While all experiences pre-
suppose physiological processes in the organism, they need not all refer
to the self; their reference may be to the perceived environment. As
Asch remarks in a discussion of this distinction (2, p. 292) : "All ex-
periences are, to be sure, mine, but only a small fraction of them refer
to me; I am right to locate my toothache in my body and to locate my
country outside it. 9 '
Wertheimer (12) and Asch have applied this same distinction to
problems of motivation. Motives, like other psychological processes,
presuppose physiological events in the individual's nervous system;
Reprinted with permission from the Journal of Psychology, Vol. 43, Second
half, April, 1957. Copyright by The Journal Press.
On Field Forces : 287
their locus is thus the organism. But they may refer to parts of the psy-
chological field other than the ego; forces may arise from the experi-
enced requirements of a situation, of persons and events outside the
self. For the sake of convenience, the latter forces will be designated
field forces, while the term ego forces will be applied to those motiva-
tional forces arising in the person. 1
Field forces seem to be important in the initiation of behavior when-
ever we feel that there are situations which call for action, as contrasted
with those in which we need to act. For example, the response to the
request from a stranger, "Got a match, bud?" is one in which the person
acts to meet the requirements of a perceived situation rather than those
of an existing ego need. Or we right an object which is about to fall.
Again, we say that a person or a scene compels our attention, that a
man needs help, that an activity attracts us. We think of a person as
caught in the web of circumstances.
In all these cases the forces that initiate action seem to arise mainly
from the perceived circumstances. It must be understood, however, that
where field forces are operating (as also, of course, in the case of ego
forces) both personal and environmental conditions are involved in
the determination of behavior. Where forces arise from the demands of
a situation, these demands must be perceived by the individual, and
his resources must be brought into play in their service. A situation may
cry out for action, but if this cry falls on deaf ears, or on the sensitive
ears of a person incapable of acting, there will be no action (cf . 7 ) . Illus-
trations of the ways in which demands of the field must be related to
those of the self to be effective will be given below.
Thus the operation of field forces, as of ego forces, presupposes not
only a role of the experienced situation in the determination of behav-
ior, but also a role of the person (forces from the ego as well as the non-
ego in the person, as also his abilities, attitudes, and resources). But
the important point is that the role of each is different in the two cases.
This difference is of considerable interest for theory. Where ego forces
1 Actually both terms are misnomers. The ego is, of course, part of the psycho-
logical field. Thus the name "field forces" is an abbreviation for the expression
"forces arising in the psychological field exclusive of those arising in the person."
Again, it cannot be assumed that all motivational forces arising in the person are
ego forces in any precise sense; much of our motivation stems from parts of the
person which are not ego. Since, however, even these forces require the coopera-
tion of the ego for their expression, and since their reference is to the person, the
term "ego forces" will be extended to include them. Perhaps more accurate terms
would be "field-referent forces" and "ego-referent forces."
288 : Mary Henle
are primary, the role of the environment is to provide incentive objects
and means and barriers to the satisfaction of the person's needs. In in-
stances in which field forces initiate action, if we are to believe our
phenomenal experience, it is not mainly the ego that uses its surround-
ings; it seems, rather, to lend itself to them.
Although phenomenal experience argues eloquently for a role for
field forces in the initiation of behavior, there have been a number of
reasons for its neglect:
1. First, it must be admitted that this experience is ambiguous. For
example, social and objective interest, interest in the people, things, and
events in the world around us, seems to offer some of the most convinc-
ing evidence that forces may arise from parts of the psychological field
outside ourselves. But such facts have been handled by the assumption
of a gregarious need, a curiosity drive, a need for understanding, and
the like. Such an assumption presupposes, of course, that the forces in
question refer to the person. Again, a heroic act of self-sacrifice to res-
cue another may be saved for the hedonistic principle by the supposi-
tion of the prospect of reward in heaven; or else masochistic or other
unconscious tendencies may be held responsible. 2
While other difficulties with this kind of solution will be mentioned
below, it may be stated here that it does no more than to push the prob-
lem to a different level. As has been pointed out elsewhere (8), even
if all motivational forces are viewed as referring to the individual, it is
still important to distinguish between those ego forces in the service of
a situation and those in the service of the individual (cf . also 2, p. 3 1 8 ) .
2. A second reason for the neglect of field forces is that special rela-
tions quickly become established between the individual and the in-
centive object or activity. Take a case in which we may assume that an
object or activity attracts the individual: the object quickly becomes
transformed through the person's occupation with it ( 8 ) . For example,
a scientist undertakes a particular line of work because he sees that
certain issues remain unsettled; let us suppose that he has no other stake,
conscious or unconscious, in the problem. Soon it becomes his prob-
lem: criticism of the work may become criticism of him, etc. So it is
2 Of course, instances of behavior which appear to go against the interests of
the ego are not, in themselves, evidence that field forces have initiated the be-
havior. In the present example, there may indeed be neurotic reasons for the
sacrifice, or some other need of the ego may be stronger than self-preservation.
It cannot be established in advance, without intimate knowledge of the circum-
stances, that any particular act is predominantly in the service of the ego or pre-
dominantly in the service of the situation.
On Field Forces : 289
when we form sentiments. A system comes to be established including
the needs both of the individual and of the object. The relations be-
tween field forces and ego forces will be discussed more specifically
below. The point of present interest is that the fact that such relations
are quickly established obscures the question of the source of the action.
3. Perhaps the most compelling reason for the neglect of field forces
is that it has not been possible in research to separate them from ego
forces. It is scarcely possible to set up in the laboratory situations whose
demands are so compelling that the person is caught up in them regard-
less of his own needs. In the typical experiment on motivation a need or
intention is aroused in the subject to perform some activity. The activity
or the experimental situation may have particular demands of its own,
demands which would produce certain effects. But these cannot be sep-
arated from the consequences of the need which had to be aroused to
get the individual to participate in the experimental activity in the first
place. Not only does the task need to be done (a field force) ; the indi-
vidual needs to do it (an ego force). With the interrupted task tech-
nique, for example, the subject both perceives the interrupted activity
perceives that the demands of the task have not been met and is
himself interrupted. Heretofore it has not been possible experimentally
to separate these two kinds of demands. Of the two kinds of forces op-
erating in such a situation, ego forces alone have received adequate at-
tention, in accordance with current theory. The requirements of the
task have been overlooked. (The importance of the perceived demands
of a task for its resumption has been suggested elsewhere; cf. 8.)
THE OBSERVER TECHNIQUE
Before a strong case can be made out for the importance of field forces
for motivational theory, it seems necessary to be able to demonstrate
them. To this end it is necessary to separate the two kinds of motivation
field forces and ego forces. For this purpose the "observer technique"
has been devised.
In the "observer experiment" one individual we will call him the
observer (O) merely watches a subject performing a series of tasks,
but is not himself permitted to participate. The performing S is known
little or not at all to O, so that the latter has no personal interest in S's
successes and failures. If, now, some of the tasks are interrupted, then
in the case of O there will be operating essentially no forces deriving
from his own needs to complete the tasks, only forces deriving from the
290 : Mary Henle
perceived incompleteness of the activities (or the perceived lack of
satisfaction of S's need) that is to say, only field forces. If these latter
are important in the determination of behavior, then O should show
such effects of interruption as the superiority of incomplete tasks in
recall ( 13 ) or increased attractiveness of the interrupted activities (4) .
On the other hand, if it is the unsatisfied (ego) need which is responsi-
ble for these effects, incomplete tasks should be treated no differently
from completed ones by O.
Experiments performed under the conditions described have shown
that Os do recall incomplete tasks (i.e., tasks which S has had to leave
unfinished) better than completed ones (3). Furthermore, when of-
fered a choice among tasks similar to those which they have just seen
S performing, they choose tasks like those on which S has been inter-
rupted significantly more often than chance; the attractiveness of the
interrupted tasks has been enhanced for the Os. Some force must exist
which favors the interrupted tasks in the case of the Os. The forces re-
sponsible must arise from O's cognitive grasp of the experimental situa-
tion. Since his own needs are not primarily involved since he is not
interrupted nor allowed to complete tasks, and since he himself runs
no risk of failure ego-referent forces cannot be mainly responsible
for the obtained differences between interrupted and completed tasks.
The results speak, rather, for the operation of field forces.
It might, of course, be argued that seeing S fail in an activity may
play into an O's own deep-seated feelings of insecurity, and that his
choice of the interrupted task represents an effort to cope with the
aroused anxiety. While the experimental situation minimizes such a
possibility, 3 it cannot be eliminated in all cases. But even in such cases
it must be remembered that the forces responsible for the effect include
field-referent forces arising from the perceived failure, which cooperate
with tendencies in O himself. Again, it might be held that O identifies
with the (unknown) S, and that his ego needs are thus vicariously in-
volved. This seems to me to be begging the question. We hardly seem
to have gained in understanding by subsuming the observer effect under
the little-understood processes of identification and empathy. On the
contrary, the observer technique seems to offer the possibility of some-
day investigating these very processes.
3 Since the experiments are concerned with the effects of interruption, not of
failure, cases of obvious failure are eliminated from the tabulated results.
On Field Forces : 291
RELATIONS OF FIELD FORCES AND EGO FORCES
Since the experimental isolation of field forces has been stressed, it
must now be repeated that, except under the most artificial experimental
conditions, field-referent forces operate always in conjunction with
ego forces. Although the specific relations between the two kinds of
forces require investigation, we may begin to describe some of the ways
in which they operate in relation to each other.
1 . Certain essential demands of the person must be met before there
can be any real turning outward of interest. In other words, a certain
level of satisfaction of ego needs is necessary before the individual can
respond to field forces. An impressive illustration is found in Sir Ernest
Shackleton's account of his effort to reach the South Pole (quoted by
Leeper, 11, p. 26):
During the last weeks of the journey outwards, and the long march back,
we really thought of little but food. The glory of the great mountains that
towered high on either side, the majesty of the enormous glacier up which
we traveled so painfully, did not appeal to our emotions to any great
extent. . . .
Practically the whole imaginative life of semistarved men may thus
be preoccupied with thoughts and fantasies about food and eating; and
there is no reason to think that the situation is different in the case of the
emotionally starved.
In this same connection an observation of Asch's is pertinent (2,
p. 304) : "In a developed individual the presence of a coherent ego is
a necessary condition of action; the carrying out of complex actions
requires a knowledge of and a reference to the self."
2. Under usual conditions there seems to be a genuine cooperation
between field forces and ego forces, the one reinforcing the other. Take,
for example, the question of ethnic prejudices. A good deal of recent
investigation has looked for personality correlates of such prejudices,
displaced hostility for example; in other words it has looked to ego
forces for an explanation. On the other hand, we may not forget that
a stereotype or image of the minority group exists in the culture. If the
group really possessed the characteristics attributed to it, one could not
help disliking its members. And similarly, if a person unthinkingly ac-
cepts this cultural image, he cannot help disliking the group. In other
words, forces making for the prejudice may arise from the cultural
image of the minority group as well as from tendencies in the prejudiced
292 : Mary Henle
individual's personality; we must consider field forces as well as ego
forces.
For the sake of argument let us grant both positions. Suppose, now,
that we have established in a given case that displaced aggression is
responsible for the prejudice. It will still be true that the individual finds
some particular target for his hostility rather than some other one; he
hates the minority group instead of kicking his dog. Thus forces arising
out of the properties of the cultural image cooperate with his own tend-
ency. Or we may find that a person unthinkingly accepts the cultural
image of the minority group, which image is then the principal source
of the forces responsible for the prejudice. But it is still worth investi-
gating whether he may not find something in the image to hate in
particular so that forces in himself cooperate with those arising from
the (here wrongly) perceived object.
Indeed, so close is the cooperation between field forces and ego
forces that it is perhaps more accurate to speak of two aspects of all
motivation rather than of two kinds of forces.
3. Again, conflicts may arise between the demands of the ego and
those of the field. 4 Conflicts between pleasure and duty may be (but
need not be) of this kind. 5
4. A special case in which field forces and ego forces are in opposi-
tion is that in which the one blinds the person to the requirements of
the other, so that no conflict is experienced. For example, an individual
may be so involved in a personal problem that he cannot attend prop-
erly to his work, or he misses an appointment, or he may be immune to
other demands of the field. Or else one may be so caught in an activity
that legitimate ego needs may be neglected. For example, we read until
too late, not because we want to continue reading, but because we can-
not put the book down. In a very important sense, it is the book that
holds us, not we who hold the book. 6
4 Of course conflicts also exist between various personal forces as well as
between simultaneously active field forces.
15 It is important not to confuse the present problem with a moral one, although
the considerations discussed here have application to questions of ethics (9, 12) .
To distinguish between the needs of the person and the requirements of the situa-
tion does not imply that the former are selfish. Not everything in the interests of
the person is egotistical. The concept of field forces carries no implication that
these should take precedence over ego forces. The opposite is equally often the
case. And there are instances in which the needs of the person are the most im-
portant objective requirements of the field (cf. 2, p. 312 f.).
In this example, it will be noted, forces from the field blind the reader to
certain personal needs, but require the cooperation of others. Thus the absorb-
ing book must be absorbing to him.
On Field Forces : 293
FIELD FORCES IN A THEORY OF MOTIVATION
If we incorporate the concept of field force into a theory of motivation,
several problems seem to be simplified, and others appear in a differ-
ent perspective. The enormous diversity of adult human motivation has
presented difficulties to need or instinct theories as well as to S-R ac-
counts (cf. 1). Likewise Allporfs solution of functional autonomy of
motives (1) has not escaped criticism, and different interpretations of
the evidence Allport adduces in support of this doctrine have been sug-
gested (2, pp. 343 ff.). The concept of field force offers another solu-
tion to this problem. It proposes to deal with the tremendous variety of
human motivation in terms of properties of force in the perceived ob-
jects and events outside the self. It seems difficult to account for our
specifically appropriate response to the request "Got a match?" in
terms of a gregarious propensity or a need for affiliation or nurturance
alone. But if this request itself is perceived as eliciting forces which
cooperate with tendencies in the individual, the difficulties seem to
disappear.
This example suggests another, closely related question whose solu-
tion is facilitated by the assumption of field forces: the question of how
specific or how inclusive a segment of activity to employ in our concept
of needs. Are we to invoke a need for offering matches to account for
the corresponding action? It was seen above that a gregarious need does
not account for the specificity of the behavior. But if we regard needs as
specific enough to account for each act, we have lost the advantages of
a need theory in bringing order to the diversity of behavior. Again the
concept of field forces offers a third possibility which seems to avoid
the difficulties of the other two. Specificity is introduced if perceived
objects and events of the environment possess the attributes of force;
and we can retain a concept of need inclusive enough to organize the
data of behavior.
In the study of field forces we are investigating forces arising from the
cognitive appreciation of the situation confronting the individual. This
is one aspect of a much larger problem, that of the cognitive side of
motives and emotions in general. Motives and emotions are not purely
irrational; they have their cognitive aspect as well. This problem has
been raised by Wertheimer and carried farther by Asch; elsewhere it
has been fairly generally neglected. Nevertheless I believe this point
is implicit in every theory of motivation and affect. It is implicit in
294 : Mary Henle
every theory that takes for granted a certain appropriateness of the
emotion to the situation in which it arises; it is silently contained in
every view that speaks of pathology where this appropriateness is lack-
ing. 7
This general problem of the cognitive aspect of motives and emotions
has been neglected for precisely the same reason that our particular
problem of field forces has been neglected: it has been impossible to
separate the affective aspect of the experience from the cognitive. At the
same time that the individual perceives the situation confronting him,
something is happening to him. It may be that we are now in possession
of a conceptual tool and a method which will permit investigation of
this larger problem.
Finally, the concept of field forces suggests a new view of certain
problems in the phenomenology of motivation an area of problems
that seems to have got lost in the interest in unconscious motivation.
I mentioned earlier that common experience is ambiguous as evidence
for the initiation of action from the field. It must now be added that
there is a further ambiguity. At times we incorrectly refer the source
of action or experience to the experienced situation or to the self. The
concept of projection in part describes cases of the first kind; for ex-
ample, self-rejection may be translated into rejection by another.
Conversely, an experience may erroneously be referred to the self; to
use a parallel example, rejection by an important person may be trans-
lated in part into rejection by the self. The facts described as introjec-
tion and identification are relevant in this connection. In dealings with
others we may erroneously take or assign credit or blame. How are we
to understand these instances? If they are simply errors in locating the
source of motivational forces if the reference of our phenomenal ex-
perience to ourselves or to the perceived surroundings is not to be relied
on it might be argued that much of the case for field forces collapses.
Actually, these are probably limiting cases, and are probably not
simply errors in the reference of an experience. We need to know more
of the psychological processes involved before the bearing of these
instances on the present problem will become clear. Meanwhile, it
seems safe to say that both field forces and ego forces are operating in
7 Cf. Anna Freud (5, p. 42) : "We expect children normally to react to these
particular occurrences with these specific affects. But, contrary to expectation,
observation may show us a very different picture. For example, a child may
exhibit indifference when we should have looked for disappointment, exuberant
high spirits instead of mortification, excessive tenderness instead of jealousy. In
all these cases something has happened to disturb the normal process. . . "
On Field Forces : 295
all these cases and that, in ways that still have to be understood, their
interaction results in errors of emphasis.
In the case of projection, except in the most extreme instances, we
are probably never dealing simply with an incorrect view of the sources
of forces. There may be, rather, an oversensitization, because of a
problem in the individual, to something that actually exists in the
other. We do not project at random. 8 (The incorrect emphasis on the
field forces in these instances remains, of course, to be explained. If
an ego need can blind us to the requirements of the field, there is no
reason why one cannot blind us to other forces in the person.)
The case of erroneous reference to the self is at present more difficult
to understand. Two directions of thinking suggest themselves. First,
while a force originating outside the self may be referred to the self,
its outside reference may also be present. If tie rejection by the other
is translated into self -rejection, this need not obscure the original rejec-
tion. It seems plausible, then, that intervening cognitive processes may
account in part for the new reference of the experience. The rejection
by the other may lead to a search for and emphasis upon the deficiencies
of the self. Or it may play into existing feelings of self -rejection.
Perhaps a more significant, but more obscure, point in the same con-
nection concerns the nature of relations. The other person represents,
fulfills, brings out, reflects, or in some way "fits" part of the individual.
The other is not simply an outer figure, but an inner figure as well (one
who corresponds more or less closely to the outer one or to aspects of
the outer one) if he is to stand in these relations to the self. Thus it seems
likely, in our example (though the processes involved are by no means
clear), that rejection by the other is rejection by the self. The same
issue is seen more clearly in cases involving projection: we may refuse
to have anything to do with a person who appeals to, supports or in
some way represents a part of ourselves that we will have nothing to
do with. Here rejection of the (suitable) other is rejection of part of
the self; again ego forces and field forces are at work.
The same kinds of considerations seem to apply even more obviously
to the example of the wrong assignment of credit or blame. In joint
undertakings responsibility probably never lies only in the self or only
in the (objective) other.
8 Cf . Freud (6, p. 236) : "We begin to see that we describe the behavior of
both jealous and persecuted paranoiacs very inadequately by saying that they
project outwards onto others what they do not wish to recognize in themselves.
"Certainly they do this; but they do not project it into the sky, so to speak,
where there is nothing of the sort already. . . ." (Cf. also 5, p. 130.)
296 : Mary Henle
We have been dealing in these examples with one aspect of the prob-
lem of how we experience our own motivation viz., the reference of
forces to the person or the field. With this as a starting point, other
aspects of this problem may become clear. For example, A and B have
a falling out. A considers B responsible, while B thinks that an out-
sider, C, made the trouble between them. Each may then structure the
relevant facts of the case about his own view of the reference of the dif-
ficulty, so that the two may come out with entirely different versions.
SUMMARY
It has been suggested that it is time to revise current ego-centered the-
ories of motivation to give a place to field forces, those motivational
forces arising from the experienced requirements of situations and per-
sons outside the self. A method of demonstrating field forces, the ob-
server technique, has been described, and some of the relations between
field forces and ego forces have been indicated. The place of the con-
cept of field force in a theory of motivation was discussed. The assump-
tion offers a new solution to the problem of the diversity and specificity
of human motivation. In addition, it provides one approach to the
neglected problem of the cognitive side of motivation and emotions.
Finally, certain problems of the phenomenology of motivation were
discussed in the light of the concept of field force.
REFERENCES
1. Allport, G. W. Personality. New York: Holt, 1937.
2. Asch, S. E. Social psychology. New York: Prentice-Hall, 1952.
3. Baltimore, G., et al. Some cognitive aspects of a motivational field.
Unpublished research, New School for Social Research, 1953.
4. Cartwright, D. The effect of interruption, completion, and failure upon
the attractiveness of activities. /. exp. PsychoL, 1942, 31, 1-16.
5. Freud, A. The ego and the mechanisms of defense. London: Hogarth
Press and the Institute of Psycho-Analysis, 1937.
6. Freud, S. Certain neurotic mechanisms in jealousy, paranoia and homo-
sexuality. In Collected papers. London: Hogarth Press and the Insti-
tute of Psycho-Analysis, 1924, Vol. H, pp. 232-243.
7. Henle, M. On activity in the goal region. PsychoL Rev., 1956, 63, 299-
302.
8. Henle, M., and Aull, G. Factors decisive for resumption of interrupted
activities: the question reopened. PsychoL Rev., 1953, 60, 81-88.
9. Kohler, W. The place of value in a -world of facts. New York: Liveright,
1938.
On Field Forces : 297
10. Kohler, W. Gestalt psychology. (Rev. ed.) New York: Liveright, 1947.
11. Leeper, R. Psychology of personality. Eugene: University of Oregon
Press, 1947.
12. Wertheimer, M. Some problems in the theory of ethics. Soc. Res., 1935,
2, 353-367. [See pp. 29-41 of the present book.]
13. Zeigarnik, B. Das Behalten erledigter und unerledigter Handlungen.
Psychol Forsch. t 1927, 9, 1-85.
PART V
PSYCHOLOGY OF
EXPRESSION AND ART
Rudolf Arnheim
SARAH LAWRENCE COLLEGE
THE GESTALT THEORY
OF EXPRESSION
What is the exact location and range of the territory covered by the term
"expression"? Thus far, no generally accepted definition exists. In order
to make clear what is meant by expression in the present paper, it is
therefore necessary to indicate (1) the kind of perceptual stimulus
which involves the phenomenon in question, and (2) the kind of
mental process to which its existence is due. This delimitation of our
subject will show that the range of perceptual objects which carry ex-
pression according to Gestalt theory is unusually large and that expres-
sion is defined as the product of perceptual properties which various
other schools of thought consider nonexistent or unimportant.
(1) In present-day usage, the term "expression" refers primarily
to behavioral manifestations of the human personality. The appearance
and activities of the human body may be said to be expressive. The
shape and proportions of the face or the hands, the tensions and the
rhythm of muscular action, gait, gestures, and other movements serve
as objects of observation. In addition, expression is now commonly
understood to reach beyond the observed person's body. The "projec-
tive techniques" exploit characteristic effects upon, and reactions to,
the environment. The way a person dresses, keeps his room, handles the
Reprinted with permission from the Psychological Review, Vol. 56, No. 3,
May, 1949.
502 : Rudolf Arnheim
language, the pen, the brush; the colors, flowers, occupations he pre-
fers; the meaning he attributes to pictures, tunes, or inkblots; the story
he imposes on puppets; his interpretation of a dramatic part these
and innumerable other manifestations can be called "expressive" in
that they permit conclusions about the personality or the temporary
state of mind of the individual. Gestalt psychologists extend the range
of expressive phenomena beyond this limit. For reasons which will be
discussed, they consider it indispensable to speak also of the expression
conveyed by inanimate objects, such as mountains, clouds, sirens, ma-
chines.
(2) Once the carrier of expression is determined, the kind of mental
process must be indicated which is charged with producing the phe-
nomenon. It is the contention of Gestalt psychology that the various
experiences commonly classified under "perception of expression" are
caused by a number of psychological processes, which ought to be dis-
tinguished from each other for the purpose of theoretical analysis.
Some of these experiences are partly or wholly based upon empirically
acquired knowledge. The mere inspection of many half-smoked ciga-
rettes in an ashtray would suggest no connection with nervous tension
to a visitor from a planet inhabited by nonsmokers. The letters EVVIVA
GUERRA and EWIVA DON pio scribbled all over the walls of an Italian
village will reveal the mentality of the natives only to someone who
happens to know that these words pay homage to a champion cyclist
and the village priest. For the purpose of the present paper, the use of
past experience for the interpretation of perceptual observations will
be excluded from the field of expression and referred to the psychology
of learning. We shall be concerned only with instances in which, ac-
cording to Gestalt psychology, sensory data contain a core of expres-
sion that is perceptually self-evident. The way a person keeps his lips
tightly closed or raises his voice or strokes a child's head or walks
hesitatingly is said to contain factors whose meaning can be understood
directly through mere inspection. Instances of such direct expression
are not limited to the appearance and behavior of the subject's own
body. They are also found in such "projective" material as the stirring
red of a woman's favorite dress or the "emotional" character of the
music she prefers. In addition, inanimate objects are said to convey
direct expression. The aggressive stroke of lightning or the soothing
rhythm of rain impress the observer by perceptual qualities which ac-
cording to Gestalt psychology must be distinguished theoretically from
the effect of what he knows about the nature of these happenings. It is
The Gestalt Theory of Expression : 303
assumed, however, that practically every concrete experience combines
factors of both kinds.
Procedures and findings. What is expression, and what enables the
observer to experience it? By means of which perceptual factors and
in what way do stimulus configurations evoke such experiences in the
onlooker? During the last twenty-five years or so, numerous experi-
mental investigations have been devoted to the phenomena of expres-
sion, but hardly any of them have tried to answer our questions. Limited
as they were to the connection between how a person behaves and what
happens in him psychologically, they centered upon the certainly im-
portant problem: To what extent are observers, untrained or trained,
gifted or average, capable of getting valid information about a person's
temporary state of mind or his more permanent psychical constitution
from an inspection of his face, voice, gait, handwriting, etc.?
This is true for the various matching experiments, which are con-
veniently summarized by Woodworth (24, pp. 242-256) and by All-
port and Vernon (1, pp. 3-20). Similarly, in the field of the projective
techniques psychologists have looked for correlations between per-
sonality traits and reactions to environmental stimuli. Almost invari-
ably, these stimuli contain factors of the kind which concern the present
paper. However, thus far, little explicit discussion has been devoted to
the question why and how the given percepts provoke the observed
reactions. There is evidence that the whole structure of a face rather
than the sum of its parts determines expression (2). But which struc-
tural features make for what expression and why? In the Rorschach
test, the typical reactions to color are probably based on expression.
But why are emotional attitudes related to color rather than shape?
Ernest G. Schachtel has done pioneer work in this field, pointing out,
for instance, that responses to colors and to affect-experiences are both
characterized by passive receptivity (19). On the whole, however,
questions of this kind have been answered thus far by summary and
scantily supported theoretical assertions.
A few remarks are in order on the investigations which have tested
the accomplishments of observers. A glance at the results reveals a
curious contrast. One group of experimenters reports essentially nega-
tive findings. Another, consisting mainly of Gestalt psychologists, as-
serts that observers judge portraits, handwritings, and similar material
with a measure of success that clearly surpasses chance. Pessimistic
generalizations have been drawn from the studies of the first type. The
subject of expression is sometimes treated with the buoyant unkindness
304 : Rudolf Arnheim
that distinguished the early behavioristic statements on introspection.
This attitude has not encouraged research.
The main reason for the conflicting results can be found in differ-
ences of approach. The investigators of the first type asked: How vaHdly
can the bodily expression of the average person or of a random member
of a particular group of people be interpreted? In other words, they
focused on the important practical question of the extent to which ex-
pression can be relied upon in everyday life. On the other hand, the
Gestalt psychologists preferred the common scientific procedure of
purifying as carefully as possible the phenomenon under investigation.
They searched for the most favorable condition of observation. A major
part of their efforts was spent in selecting and preparing sets of speci-
mens which promised to demonstrate expression clearly and strongly
(2, p. 8).
Some of the factors which may account for the often disappointing
results obtained in experiments with random material are the following,
(a) Everyday observation suggests that the structural patterns of char-
acter, temperament, mood, are not equally clear-cut in all people.
While some individuals are pronouncedly depressed or lighthearted,
strong or weak, harmonious or disharmonious, warm or cold, others
strike us as indefinite, lukewarm, fluid. Whatever the exact nature of
such indefiniteness, one would expect the corresponding faces, gestures,
handwritings to be equally vague in form and therefore in expression.
When one examines material of this kind, one notices in some cases that
the decisive structural features are not sharply defined. In other cases,
factors which are clear-cut in themselves add up to something that
shows neither harmony nor conflict but a lack of unity or relatedness,
which renders the whole meaningless, inexpressive. Many telling ex-
amples can be found among the composite faces made up by the sum-
mation of unrelated parts for experimental purposes. If observers can
cope with such material at all, they do so presumably by guessing what
these artifacts are meant to mean rather than by having the experience
of live expression, (b) The presence of a portrait photographer's cam-
era tends to paralyze a person's expression, and he becomes self-con-
scious, inhibited, and often strikes an unnatural pose, (c) Candid shots
are momentary phases isolated from a temporal process and a spatial
context. Sometimes they are highly expressive and representative of
the whole from which they are taken. Frequently they are not. Further-
more the angle from which a shot is made, the effect of lighting on
shape, the rendering of brightness and color values, as well as modifica-
The Gestalt Theory of Expression : 305
tions through retouching, are factors which make it impossible to ac-
cept a random photograph as a valid likeness, (d) If for purposes of
matching experiments a number of samples is combined at random,
accidental similarities of expression may occur, which will make dis-
tinction difficult, even though every specimen may be clear-cut in itself.
Further reasons for the lack of consistent results are discussed by
Wolff (23, p. 7). x
The conclusion seems to be that the recognition of expression has
been proven to be reliable and valid under optimal conditions. For the
average face, voice, gesture, handwriting, etc., the results are likely
to be less positive. However, in order to establish this fact trustworthily,
the additional obstacles created by unsuitable experimental conditions
will have to be reduced.
Associationist theories. What enables observers to judge expression?
The traditional theory, handed down to our generation without much
questioning, is based on associationism. In his essay on vision Berkeley
(4, 65) discusses the way in which one sees shame or anger in the
looks of a man.
Those passions are themselves invisible: they are nevertheless let in by the
eye along with colours and alterations of countenance, which are the im-
mediate object of vision, and which signify them for no other reason than
barely because they have been observed to accompany them: without which
experience, we should no more have taken blushing for a sign of shame, than
of gladness.
Darwin, in his book on the expression of emotions, devoted a few
pages to the same problem (7, pp. 356-359). He considered the rec-
ognition of expression to be either instinctive or learned. "Children, no
doubt, would soon learn the movements of expression in their elders
in the same manner as animals learn those of man," namely, "through
their associating harsh or kind treatment with our actions."
Moreover, when a child cries or laughs, he knows in a general manner what
he is doing and what he feels; so that a very small exertion of reason would
1 Since there is no reason to expect that every photograph will reproduce essen-
tial features of expression, it would be interesting to know by which criterion
the photographs for the Szondi test (18) have been selected. If an integral
feature of the test consists in establishing the reactions of people to the personali-
ties of homosexuals, sadistic murderers, etc., two questions arise. (1) Is there a
complete correlation between these pathological manifestations and certain clear-
cut personality structures? (2) Are the latter suitably expressed in the photo-
graphs? These problems are avoided if the test is meant simply to investigate
people's responses to a given set of portraits, whatever their origin.
306 : Rudolf Arnheim
tell him what crying or laughing meant in others. But the question is, do our
children acquire their knowledge of expression solely by experience through
the power of association and reason? As most of the movements of expres-
sion must have been gradually acquired, afterwards becoming instinctive,
there seems to be some degree of a priori probability that their recognition
would likewise have become instinctive.
In Darwin's view, the relationship between expressive bodily behavior
and the corresponding psychical attitude was merely causal. Expressive
gestures were either remnants of originally serviceable habits or due
to "direct action of the nervous system." He saw no inner kinship be-
tween a particular pattern of muscular behavior and the correlated
state of mind.
A variation of the associationist theory contends that judgments of
expression are based on stereotypes. In this view, interpretation does
not rely on what belongs together according to our spontaneous insight
or repeated observation but on conventions, which we have adopted
ready-made from our social group. We have been told that aquiline
noses indicate courage and that protruding lips betray sensuality. The
promoters of the theory generally imply that such judgments are wrong,
as though information not based on first-hand experience could never
be trusted. Actually, the danger does not lie in the social origin of the
information. What counts is that people have a tendency to acquire
simply structured concepts on the basis of insufficient evidence, which
may have been gathered first-hand or second-hand, and to preserve
these concepts unchanged in the face of contrary facts. While this may
make for many one-sided or entirely wrong evaluations of individuals
and groups of people, the existence of stereotypes does not explain the
origin of physiognomic judgments. If these judgments stem from tradi-
tion, what is the tradition's source? Are they right or wrong? Even
though often misapplied, traditional interpretations of physique and
behavior may still be based on sound observation. In fact, perhaps they
are so hardy because they are so true.
Empathy. The theory of empathy holds an intermediate position
between the traditional and a more modern approach. This theory is
often formulated as a mere extension of the association theory, designed
to take care of the expression of inanimate objects. When I look at the
columns of a temple, I know from past experience the kind of mechan-
ical pressure and counterpressure that occurs in the column. Equally
from past experience I know how I should feel myself if I were in the
The Gestalt Theory of Expression : 307
place of the column and if those physical forces acted upon and within
my own body. I project my feelings into the column and by such anima-
tion endow it with expression. Lipps, who developed the theory, stated
that empathy is based on association (16, p. 434). It is true, he also
says, that the kind of association in question connects "two things be-
longing together, or being combined by necessity, the one being im-
mediately given in and with the other." But he seems to have conceived
of this inner necessity as a merely causal connection, because immedi-
ately after the statement just quoted he denies explicitly that the rela-
tionship between the bodily expression of anger and the angry person's
psychical experience could be described as an "association of similarity,
identity, correspondence" (p. 435). Like Darwin, Lipps saw no in-
trinsic kinship between perceptual appearance and the physical and
psychological forces "behind" it. However, he did see a structural simi-
larity between physical and psychological forces in other respects.
After discussing the mechanical forces whose existence in an inanimate
object is inferred by the observer through past experience, Lipps writes
the following remarkable passage:
And to (the knowledge of these mechanical forces) is furthermore attached
the representation of possible internal ways of behavior of my own, which
do not lead to the same result but are of the same character. In other words,
there is attached the representation of possible kinds of my own activity,
which in an analogous fashion, involves forces, impulses, or tendencies, freely
at work or inhibited, a yielding to external effect, overcoming of resistance,
the arising and resolving of tensions among impulses, etc. Those forces and
effects of forces appear in the light of my own ways of behavior, my own
kinds of activity, impulses, and tendencies and their ways of realization (16,
p. 439).
Thus Lipps anticipated the Gestalt principle of isomorphism for
the relationship between the physical forces in the observed object and
the psychical dynamics in the observer; and in a subsequent section of
the same paper he applies the "association of similarity of character"
even to the relationship between the perceived rhythm of musical tones
and the rhythm of other psychical processes that occur in the listener.
Which means that in the case of at least one structural characteristic,
namely rhythm, Lipps realized a possible inner similarity of perceptual
patterns and the expressive meaning they convey to the observer.
The Gestalt approach. The Gestalt theory of expression admits that
correspondences between physical and psychical behavior can be dis-
covered on the basis of mere statistical correlation but maintains that
308 : Rudolf Arnheim
repeated association is neither the only nor the common means of ar-
riving at an understanding of expression. Gestalt psychologists hold
that expressive behavior reveals its meaning directly in perception. The
approach is based on the principle of isomorphism, according to which
processes which take place in different media may be nevertheless sim-
ilar in their structural organization. Applied to body and mind, this
means that if the forces which determine bodily behavior are struc-
turally similar to those which characterize the corresponding mental
states, it may become understandable why psychical meaning can be
read off directly from a person's appearance and conduct.
It is not the aim of this paper to prove the validity of the Gestalt
hypothesis. 2 We shall limit ourselves to pointing out some of its impli-
cations. Only brief presentations of the theory are available so far.
However, Kohler's (12, pp. 216-247) and Koffka's (10, pp. 654-
661) remarks about the subject are explicit enough to indicate that
isomorphism on only two levels, namely the psychical processes which
occur in the observed person and the corresponding behavioral activity,
would be insufficient to explain direct understanding of expression
through perception. In the following an attempt will be made to list a
number of psychological and physical levels, in the observed person
and in the observer, at which isomorphic structures must exist in order
to make the Gestalt explanation possible.
Let us suppose that a person A performs a "gentle" gesture, which
is experienced as such by an observer B. On the basis of psychophysical
parallelism in its Gestalt version it would be assumed that the tender-
ness of A's feeling (Table 1, level I) corresponds to a hypothetical
process in A's nervous system (level II) and that the two processes,
the psychical and the physiological, are isomorphic, that is to say,
similar in structure.
The neural process will direct the muscular forces which produce
the gesture of A's arm and hand (level HI) . Again it must be assumed
that the particular dynamic pattern of mechanical action and inhibi-
2 For that purpose, observations of infants are relevant. Even in his day, Dar-
win was puzzled by the fact that young children seemed directly to understand a
smile or grief "at much too early an age to have learnt anything by experience"
(7, p. 358) . According to Biihler (6, p. 377) , "the baby of three or four months
reacts positively to the angry as well as to the kind voice and look; the five-to-
seven-months-old baby reflects the assumed expression and also begins to cry at
the scolding voice and threatening gesture" on the basis of "direct sensory in-
fluence." Further evidence will have to come from detailed demonstrations of
structural similarities. (Cf. pp. 319-321.)
The Gestalt Theory of Expression : 309
tion in A's muscles corresponds structurally to the configuration of
physiological and psychical forces at the levels II and I. The muscular
action will be accompanied with a kinesthetic experience (level IV),
which again must be isomorphic with the other levels. The kinesthetic
experience need not always take place and is not strictly indispensable.
However, the structural kinship of the experienced gentleness of Ms
gesture and the equally experienced gentleness of his mood will make A
feel that his gesture is a fitting manifestation of his state of mind.
TABLE 1
ISOMORPHIC LEVELS
A. Observed person
I. State of mind Psychological
II. Neural correlate of I Electrochemical
III. Muscular forces Mechanical
IV. Kinesthetic correlate of III Psychological
V. Shape and movement of body Geometrical
B. Observer
VI. Retinal projection of V Geometrical
VII. Cortical projection of VI Electrochemical
VIII. Perceptual correlate of VII Psychological
Finally, the muscular forces of level III will cause A's arm and hand
to move in a, say, parabolic curve (level V) ; and again the geometric
formation of this curve would have to be isomorphic with the structure
of the processes at the previous levels. An elementary geometrical ex-
ample may illustrate the meaning of this statement. Geometrically, a
circle is the result of just one structural condition. It is the locus of all
points that are equally distant from one center. A parabola satisfies
two such conditions. It is the locus of all points that have equal distance
from one point and one straight line. The parabola may be called a
compromise between two structural demands. Either structural con-
dition yields to the other. 3 Is there any possible connection between
these geometrical characteristics of the parabola and the particular
configuration of physical forces to which we attribute gentleness? One
may point to the kind of physical process that produces parabolic pat-
3 One can express this also in terms of projective geometry by saying that the
parabola as a conic section is intermediate between the horizontal section, namely
the circle, and the vertical section, the straight-edged triangle.
310 : Rudolf Arnheim
terns. In ballistics, for instance, the parabolic curve of a trajectory is
the result of a "compromise" between the direction of the original im-
pulse and the gravitational attraction. The two forces "yield" to each
other. 4
At this point the description must shift from the observed person A
to the observer B. B's eyes receive an image (level VI) of the gesture
performed by A's arm and hand. Why should this image produce in B
the impression that he is observing a gentle gesture? It may be true that
the geometrical pattern of the gesture as well as the configuration of
muscular forces which has created this pattern can both be character-
ized structurally as containing compromise, flexibility, yielding. But
this fact in itself is not sufficient to explain the direct experience which
B is said to receive by his perceptual observation. It becomes clear at
this point that the Gestalt theory of expression is faced not only with
the problem of showing how psychical processes can be inferred from
bodily behavior, but that the primary task consists in making plausible
the fact that the perception of shape, movement, etc., may convey to the
observer the direct experience of an expression which is structurally
similar to the organization of the observed stimulus pattern.
A's gesture is projected on the retinae of B's eyes 5 and, by way of
the retinal images, on the visual cortex of B's cerebrum (level VII).
Correspondingly, B perceives A's gesture (level VIII). Is there a pos-
4 One of the principles on which the analysis of handwritings is based indicates
that the script pattern reflects dynamic features of the writer's motor behavior,
which in turn is produced by a characteristic configuration of muscular forces.
The same isomorphism of muscular behavior and resulting visible trace has found
applications in the technique of drawing. Langfeld (15, p. 129) quotes Bowie
(5, pp. 35 and 77-79) concerning the principle of "living movement" (Set Do)
in Japanese painting: "A distinguishing feature in Japanese painting is the strength
of the brush stroke, technically called fude no chikara or fude no ikioi. When
representing an object suggesting strength, such, for instance, as rocky cliff, the
beak or talons of a bird, the tiger's claws, or the limbs and branches of a tree,
the moment the brush is applied the sentiment of strength must be invoked and
felt throughout the artist's system and imparted through his arm and hand to the
brush, and so transmitted into the object painted."
5 At this stage a number of factors may interfere with the adequate projection
of decisive characteristics of body A on the receptor organ of B. In our specific
example it will depend, for instance, on the angle of projection, whether or not
the perspective retinal image will preserve the essential structural features of the
parabolic movement or transform it into a stimulus trace of unclear or clearly
different structure. (In photographs and motion pictures such factors influence
the kind of expression obtained from the reproduction of physical objects.) Simi-
lar factors will influence the veracity of other perceptual qualities which carry
expression.
The Gestalt Theory of Expression : 311
sible similarity of the geometrical structure of the stimulus configura-
tion and the structure of the expression which it conveys to the ob-
server? We may go back to our mathematical analysis of the circle and
the parabola. Simple experiments confirm what artists know from ex-
perience, namely that a circular curve looks "harder," less flexible, than
a parabolic one. In comparison with the circle the parabola looks more
gentle. One could try to explain this finding by assuming that the ob-
server knows, through past experience, the geometrical characteristics
of such patterns or the nature of the physical forces which frequently
produce them. This would take us back to the associationist theory.
Along Gestalt lines another explanation suggests itself.
The projection of the perceptual stimulus on the visual cortex can
be assumed to create a configuration of electrochemical forces in the
cerebral field. The well-known Gestalt experiments in perception sug-
gest that retinal stimulations are subjected to organizational processes
when they reach the cortical level. As a result of these processes the
elements of visual patterns are perceived as being grouped according
to Wertheimer's rules. Furthermore, any visual pattern appears as an
organized whole, in which some predominant elements determine the
over-all shape and the directions of the main axes, while others have
subordinate functions. For the same reasons, modifications of objective
shape and size are perceived under certain conditions.
It will be observed that all these experimental findings focus upon
the effects of the strains and stresses which organize the cortical field.
Is there any reason to assume that only the effects of these dynamic
processes, namely the groupings, the hierarchies of structural functions,
and the modifications of shape and size, are reflected in perceptual ex-
perience? Why should not the strains and stresses of the cortical forces
themselves also have their psychological counterpart? It seems plau-
sible that they represent the physiological equivalent of what is ex-
perienced as expression.
Such a theory would make expression an integral part of the elemen-
tary processes of perception. Expression, then, could be defined as the
psychological counterpart of the dynamic processes which result in
the organization of perceptual stimuli. While concrete verification is
obviously far away, the basic assumption has gained in concreteness
since Kohler and Wallach (14) have explained phenomena of per-
ceptual size, shape, and location through tie action of electrochemical
forces. The future will show whether the theory can be extended tg
covering the phenomena of expression,
312 : Rudoli Arnheim
It is possible now to return to the question of how the perception of
shape, movement, etc., may convey to an observer the direct experience
of an expression which is structurally similar to the organization of the
observed stimulus pattern. We referred previously to the constellations
of physical forces which will induce an object to pursue a parabolic
path. The physicist may be able to tell whether the example from bal-
listics is invertible. Will a parabolic pattern, such as the one projected
on the cortical field, under certain conditions set off a configuration of
forces which contains the structural factors of "compromise" or "yield-
ing"? If so, isomorphism of the cortical forces and those described as
levels I-V could be established.
This brings the description of isomorphic levels to an end. If the
presentation is correct, the Gestalt-theoretical thesis would imply that
an observer will adequately gauge another person's state of mind by
inspection of that person's bodily appearance if the psychical situation
of the observed person and the perceptual experience of the observer
are structurally similar by means of a number of intermediate isomor-
phic levels.
Expression as a perceptual quality. The definition which was given
above suggests that expression is an integral part of the elementary
perceptual process. This should not come as a surprise. Perception is
a mere instrument for the registration of color, shape, sound, etc., only
as long as it is considered in isolation from the organism of which it is
apart. In its proper biological context, perception appears as the means
by which the organism obtains information about the friendly, hostile,
or otherwise relevant environmental forces to which it must react.
These forces reveal themselves most directly by what is described here
as expression.
There is psychological evidence to bear out this contention. In fact,
the observations on primitives and children cited by Werner (21, pp.
67-82) and Kohler (13) indicate that "physiognomic qualities," as
Werner calls them, are even more directly perceived than the "geomet-
ric-technical" qualities of size, shape, or movement. Expression seems
to be the primary content of perception. To register a fire as merely a
set of hues and shapes in motion rather than to experience primarily
the exciting violence of the flames presupposes a very specific, rare, and
artificial attitude. Even though the practical importance of, and hence
the alertness to, expression has decreased in our culture, it cannot be
maintained that a basic change has taken place in this respect. Darwin
(7, pp. 359-360) noted that people sometimes observe and describe
The Gestalt Theory of Expression : 313
facial expression without being able to indicate the features of form,
size, direction, etc., which carry it. In experimental work, one notices
that even with the object directly in front of their eyes, subjects find it
a hard and uncomfortable task to take note of the formal pattern. They
constantly fall back upon the expressive characteristics, which they
describe freely and naturally. Everyday experience shows that people
may clearly recall the expression of persons or objects without being
able to indicate color or shape. Asch observes: "Long before one has
realized that the color of the scene has changed, one may feel that the
character of the scene has undergone change" (3, p. 85 ) . Finally, there
is the fact that the artist's, writer's, musician's approach to their sub-
ject is principally guided by expression. 6
Generalized theory. Thus far, the phenomenon of expression has been
discussed essentially in its best-known aspect, namely, as a physical
manifestation of psychical processes. However, some of the foregoing
considerations implied that expression is a more universal phenom-
enon. Expression does not only exist when there is a mind "behind" it,
a puppeteer that pulls the strings. Expression is not limited to living
organisms, which possess consciousness. A flame, a tumbling leaf, the
wailing of a siren, a willow tree, a steep rock, a Louis XV chair, the
cracks in a wall, the warmth of a glazed teapot, a hedgehog's thorny
back, the colors of a sunset, a flowing fountain, lightning and thunder,
the jerky movements of a bent piece of wire they all convey expres-
sion through the various senses. The importance of this fact has been
concealed by the popular hypothesis that in such cases human expres-
sion is merely transferred to objects. If, however, expression is an in-
herent characteristic of perceptual factors, it becomes unlikely that
nonhuman expression should be nothing but an anthropomorphism, a
"pathetic fallacy." Rather will human expression have to be considered
a special case of a more general phenomenon. The comparison of an
object's expression with a human state of mind is a secondary process
(cf. p. 315) . A weeping willow does not look sad because it looks like
a sad person. It is more adequate to state that since the shape, direction,
and flexibility of willow branches convey the expression of passive
hanging, a comparison with the structurally similar psychophysical
pattern of sadness in humans may impose itself secondarily.
Expression is sometimes described as "perceiving with imagination."
In doing so Gottshalk (9) explains that "something is perceived as if it
6 This has led to the erroneous notion that all perception of expression is
aesthetic.
314 : Rudolf Arnheim
were actually present in the object of perception, although literally it is
only suggested and not actually there. Music is not literally sad or gay
or gentle; only sentient creatures or creatures with feeling, such as
human beings, could be that." If our language possessed more words
which could refer to kinds of expression as such, instead of naming
them after emotional states in which they find an important applica-
tion, it would become apparent that the phenomenon in question is
"actually present in the object of perception" and not merely associated
with it by imagination.
Even with regard to human behavior, the connection of expression
with a corresponding state of mind is not as compelling and indis-
pensable as is sometimes taken for granted. Kohler (12, pp. 260-264)
has pointed out that people normally deal with and react to the expres-
sive physical behavior in itself rather than being conscious of the psy-
chical experiences reflected by such behavior. We perceive the slow,
listless, "droopy" movements of one person as against the brisk,
straight, vigorous movements of another, but do not necessarily go
beyond the meaning of such appearance by thinking explicitly of the
psychical weariness or alertness behind it. Weariness and alertness are
already contained in the physical behavior itself; they are not distin-
guished in any essential way from the weariness of slowly floating tar or
tiie energetic ringing of the telephone bell.
This broader conception has practical consequences. It suggests,
for instance, that the phenomenon of expression does not belong pri-
marily under the heading of the emotions or personality, where it is
commonly treated. It is true that the great contributions which the
study of expression has in store for these fields of psychology are thus
far almost untapped. However, the experience of the last decades shows
that little progress is made unless the nature of expression itself is clari-
fied first. 7
Secondary effects. Strictly speaking, the phenomenon of expression
is limited to the levels V-VIII of Table 1. That is, the term "expres-
sion," as used in this paper, refers to an experience which takes place
when a sensory stimulus affects the visual cortex of an observer's brain.
The processes which may have given rise to the stimulus as well as those
7 Once this is done, it will be possible and necessary to approach the further
problem of the influences which the total personality exerts upon the observation
of expression. To Vincent van Gogh, cypress trees conveyed an expression which
they do not have for many other people. Cf. Kofflca (10, p. 600).
The Gestalt Theory of Expression ; 315
which the cortical stimulation provokes in other brain centers are
supplementary.
Once perceptual stimulation has taken place, a number of secondary
happenings may follow. (1) The observer B may deduce from the ex-
pression of B's bodily behavior that particular psychical processes are
going on in A's mind; that is, through the perception of level V the
observer gains knowledge about level I. The observation of a gentle
gesture leads to the conclusion: B is in a gentle mood. This conclusion
may be based on an isomorphic similarity between the observed be-
havior and a state of mind known or imaginable to the observer. In
other cases, the conclusion may rely on past experience. Yawning, for
instance, conveys the direct expression of sudden expansion; but the
connection between yawning and fatigue or boredom is discovered by
learning. The same seems to be true for the spasmodic outbursts of
sound which we call laughter and which in themselves are so far from
suggesting mirth that they remain permanently incomprehensible to
the chimpanzee, who otherwise "at once correctly interprets the slight-
est change of human expression, whether menacing or friendly" (11,
p. 307) . It is important to realize that an expression may be correctly
perceived and described, yet the inferences derived from it may be
wrong. If, in an experiment, 80 per cent of the observers agree on an
"erroneous" attribution, it is not sufficient to dismiss the result as an
instance of failure. The high amount of agreement represents a psycho-
logical fact in its own right. The reliability of the observers' responses
to a perceptual stimulus is a problem quite different from the validity of
such responses, i.e., the question whether the observers' diagnosis is
"true."
(2) The observed expression may bring about the corresponding
state of mind in B. In perceiving A's gentle behavior, the observer him-
self may experience a feeling of tenderness. (Lipps speaks of "sympa-
thetic empathy" as distinguished from "simple empathy" [16, p. 417].)
(3) The observed expression may provoke the corresponding kines-
thetic experience, e.g., a feeling of relaxed softness. The effects de-
scribed under (2) and (3) may be instances of a kind of "resonance"
based on isomorphism. Just as a sound calls forth a vibration of similar
frequency in a string, various levels of psychological experience, such
as the visual, the kinesthetic, the emotional seem to elicit in each other
sensations of similar structure. (4) The perceived expression may re-
mind B of other observations in which a similar expression played a
role. Thus past experience is considered here not as the basis for the
316 : Rudolf Arnheim
apperception of expression; instead, the direct observation of expres-
sion becomes the basis for comparison with similar observations in the
past.
The role of past experience. While there is no evidence to support
the hypothesis that the central phenomenon of expression is based on
learning, it is worth noting that in most cases the interpretation of the
perceived expression is influenced by what is known about the person
or object in question and about the context in which it appears. Mere
inspection will produce little more than over-all impressions of the
forces at work, strong and clear-cut as such an experience may be. In-
creasing knowledge will lead to more differentiated interpretations,
which will take the particular context into account. (As an example,
one may think of the expression conveyed by the behavior of an animal
whose habits one does not know and the changes that occur with closer
acquaintance.) Knowledge does not interfere with expression itself, it
merely modifies its interpretation, except for cases in which knowledge
changes the appearance of the carrier of expression, that is, the per-
ceptual pattern itself. For instance, a line figure may change its per-
ceptual structure and therefore its expression if it is suddenly seen as a
human figure. A lifted eyebrow is seen as tense because it is perceived
as a deviation from a known normal position. The expression of Mon-
golian eyes or Negro lips is influenced, for a white observer, by the fact
that he conceives them as deviations from the normal face of his own
race.
In Gestalt terms, past experience, knowledge, learning, memory are
considered as factors of the temporal context in which a given phe-
nomenon appears. Like the spatial context, on which Gestaltists have
concentrated their attention during the early development of the theory,
the temporal context influences the way a phenomenon is perceived. An
object looks big or small depending on whether it is seen, spatially, in
the company of smaller or larger objects. The same is true for the
temporal context. The buildings of a middle-sized town look tall to a
farmer, small to a New Yorker, and correspondingly their expression
differs for the two observers. Mozart's music may appear serene and
cheerful to a modern listener, who perceives it in the temporal context
of twentieth-century music, whereas it conveyed the expression of vio-
lent passion and desperate suffering to his contemporaries against the
background of the music they knew. Such examples do not demonstrate
that there is no intrinsic connection between perceptual patterns and
the expression they convey but simply that experiences must not be
The Gestalt Theory of Expression : 317
evaluated in isolation from their spatial and temporal whole-context.
Knowledge often merges with directly perceived expression into a
more complex experience. When we observe the gentle curve of a
coachman's whip while being aware at the same time of the aggressive
use of the object, the resulting experience clearly contains an element
of contradiction. Such contradictions are exploited by artists; compare,
in motion pictures, the uncanny effect of the murderer who moves softly
and speaks with a velvety voice.
Finally, the perceptual experience of expression can be influenced
by the kind of training which in artistic and musical instruction is
known as making students "see" and "hear." By opening people's eyes
and ears to what is directly perceivable, they can be made to scan the
given sensory pattern more adequately and thus to receive a fuller ex-
perience of its expression. A neglected or misled capacity for respond-
ing perceptually can be revived or corrected.
The role of kinesthesia. Frequently people feel that another person,
whom they are observing, behaves physically the way they themselves
have behaved before. They get this impression even though at that time
they probably did not watch themselves in the mirror. It may be that
they compare their own state of mind as they remember it from the
former occasion with the expression conveyed by the bodily behavior
of the other person and/or with the state of mind reflected in that be-
havior. Probably the kinesthetic perception of one's own muscular be-
havior plays an important part in such situations. If muscular behavior
and kinesthetic experience are isomorphic, it becomes explainable why
at times one is so keenly aware of one's own facial expression, posture,
gestures. One may feel, for instance: Right now, I look just like my
father! The most convincing example is furnished by actors and danc-
ers, whose bodily performance is created essentially through kinesthetic
control. And yet their gestures are understandable to the audience
visually. This suggests that there is a valid correspondence between
bodily behavior and the related kinesthetic perception. The problem
of what enables an infant to imitate an observer who smiles or shows
the tip of his tongue belongs in the same category. Of particular in-
terest is the fact that the blind express their feelings even though im-
perfectly in spite of their inability to observe expression in others
visually. The blind also understand certain gestures on the basis of
their own kinesthetic experiences.
The blind man, like the person who sees, is aware of the gestures he makes
when under the influence of various emotions. He shrugs his shoulders and
318 : Rudolf Arnheim
raises his arms to express his disdain and amazement. The same gestures
recognized by him in a statue will evoke within him the same sentiments
(20, p. 320).
Isomorphism would seem to account also for the fact that it often
suffices to assume a particular posture (levels III and IV) in order to
enter into a corresponding state of mind (level I). Bending the head
and folding the hands is more than an accidentally chosen posture of
praying, which derives its meaning merely from tradition. The kines-
thetic sensation which accompanies this posture is structurally akin
to the psychical attitude called devotion. "Bowing" to a superior pow-
er's will is a mental condition so directly related to the corresponding
bodily gesture that its common linguistic description uses the physical
to describe the psychological. Rituals not only express what people feel
but also help them to feel the way the situation requires. By straighten-
ing our backbones we produce a muscular sensation which is akin to
the attitude of pride, and thus introduce into our state of mind a notice-
able element of bold self-sufficiency. 8
Even the "practical" motor activities are accompanied more or less
strongly by structurally corresponding states of mind. For instance,
hitting or breaking things normally seems to evoke the emotional over-
tone of attack. To assert merely that this is so because people are ag-
gressive would be an evasion of the problem. But if the dynamic charac-
ter of the kinesthetic sensation which accompanies hitting and break-
ing resembles the emotional dynamics of attack, then the one may be
expected to evoke the other by "resonance" (cf. p. 315). (This kin-
ship makes it possible for aggressiveness, wherever it exists, to express
itself through such motor acts.) Probably this parallelism holds true
for all motor activity. Muscular behavior such as grasping, yielding,
lifting, straightening, smoothing, loosening, bending, running, stopping
seems to produce mental resonance effects constantly. (In consequence,
language uses all of them metaphorically to describe states of mind.)
The psychosomatic phenomena of pathological "organ-speech" ("I
cannot stomach this!") may be considered the most dramatic examples
of a universal interdependence. The range and the importance of the
phenomenon are not acknowledged as long as one studies expression
only in motor activities that are not, or not any more, serviceable. It
8 James's theory of emotion is based on a sound psychological observation.
It fails where it identifies the kinesthetic sensation with the total emotional ex-
perience instead of describing it as a component which reinforces and sometimes
provokes emotion because of the structural similarity of the two.
The Gestalt Theory of Expression : 319
seems safe to assert that all motor acts are expressive, even though in
different degrees, and that they all carry the experience of correspond-
ing higher mental processes, if ever so faintly. Therefore, it is inade-
quate to describe expressive movements as mere atavisms, the way
Darwin did. They are physical acts which take place because of their
inner correspondence with the state of mind of the person who per-
forms them. To use one of Darwin's examples: a person who coughs in
embarrassment is not simply the victim of a meaningless association
between a state of mind and a physical reaction, which was or can be
serviceable under similar circumstances. Rather does he produce a
reaction which he experiences to be meaningfully related to his state
of mind. The bodily accompaniment completes the mental reaction.
Together they form an act of total psychophysical behavior. The hu-
man organism always functions as a whole, physically and psychically.
This view permits an application to the theory of art. It highlights
the intimate connection of artistic and "practical" behavior. The
dancer, for instance, does not have to endow movements with a sym-
bolic meaning for artistic purposes, but uses in an artistically organized
way the unity of psychical and physical reaction which is characteristic
for human functioning in general.
In a broader sense, it is the direct expressiveness of all perceptual
qualities which allows the artist to convey the effects of the most uni-
versal and abstract psychophysical forces through the presentation of
individual, concrete objects and happenings. While painting a pine
tree, he can rely on the expression of towering and spreading which
this tree conveys whenever it is seen by a human eye, and thus can span
in his work the whole range of existence, from its most general princi-
ples to the tangible manifestations of these principles in individual
objects.
An illustration. It has been pointed out in the beginning that experi-
menters have been concerned mostly with the question whether and to
what extent observers can judge a person's state of mind from his physi-
cal appearance. In consequence, the psychological literature contains
few analyses of perceptual patterns with regard to the expression they
convey. As an example of the kind of material which is badly needed in
this field, Efron's study on the gestures of two ethnical groups (8) may
be cited. He describes the behavior of Eastern Jews and Southern Ital-
ians in New York City by analyzing the range, speed, plane, coordina-
tion, and shape of their movements. A comparison of these findings
with the mentalities of the two groups would probably produce excel-
320 : Rudolf Arnheim
lent illustrations of what is meant by the structural similarity of psychi-
cal and physical behavior. Among the experimental investigations,
Lundholm's early study (17) may be mentioned. He asked eight lay-
men in art to draw lines, each of which was to express the affective tone
of an adjective given verbally. It was found, for instance, that only
straight lines, broken by angles, were used to represent such adjectives
as exciting, furious, hard, powerful, while only curves were used for
sad, quiet, lazy, merry. Upward direction of lines expressed strength,
energy, force; downward direction, weakness, lack of energy, relaxa-
tion, depression, etc. Recently Willmann (22) had thirty-two musi-
cians compose short themes, meant to illustrate four abstract designs.
Some agreement among the composers was found concerning the
tempo, meter, melodic line, and amount of consonance, chosen to
render the characteristics of the drawings. Subsequently the designs
and compositions were used for matching experiments.
Because of the scarcity of pertinent material, it may be permissible
to mention here an experiment which is too limited in the number of
cases and too subjective in its method of recording and evaluating the
data to afford a proof of the thesis we are discussing. It is presented
merely as an example of the kind of research which promises fruitful
results. 9 Five members of the student dance group of Sarah Lawrence
College were asked individually to give improvisations of the following
three subjects: sadness, strength, night. Rough descriptions of the dance
patterns which resulted were jotted down by the experimenter and
later classified according to a number of categories. Table 2 presents
the findings in an abbreviated form. The numerical agreement is high
but obviously carries little weight. As a point of method, it may only
be mentioned that instances of disagreement cannot be taken simply
to indicate that there was no reliable correspondence between task and
performance. Sometimes, the task allows more than one valid interpre-
tation. For instance, "strength" expresses itself equally well in fast and
in slow movement. "Night" is less directly related to one particular dy-
namic pattern than "sadness" or "strength."
Most tempting is the comparison between the movement patterns
and the corresponding psychical processes. Such comparison cannot
be carried through with exactness at this time mainly because psy-
chology has not yet provided a method of describing the dynamics of
9 The data were collected and tabulated by Miss Jane Binney, a student at
Sarah Lawrence College.
The Gestalt Theory of Expression : 321
TABLE 2
ANALYSIS OF DANCE MOVEMENTS IMPROVISED BY FIVE SUBJECTS
Sadness
Strength
Night
Speed: 5: slow
2: slow
1: very fast
1: medium
1: decrescendo
5: slow
Range:
5:
small, enclosed
5:
large, sweeping
3:
small
2:
large
Shape:
3:
round
5:
very straight
5:
round
2:
angular
Tension:
4:
little tension
5:
much tension
4:
little tension
1:
inconsistent
1:
decrescendo
Direction:
5:
indefinite, chang-
5:
precise, sharp,
3:
indefinite, chang-
ing, wavering
mostly forward
ing
2:
mostly downward
Center:
5:
passive, pulled
5:
active, cen-
3:
passive
downward
tered in body
2:
from active to
passive
states of mind in a way which would be more exact scientifically than
the descriptions offered by novelists or everyday language. Neverthe-
less, it can be seen from our example that the dynamic patterns of ex-
pressive behavior permit relatively concrete and exact descriptions in
terms of speed, range, shape, etc. Even the crudely simplified charac-
terizations given in the table seem to suggest that the motor traits
through which the dancers interpreted sadness reflect the slow, lan-
guishing pace of the psychological processes, the indefiniteness of aim,
the withdrawal from the environment, the passivity all of which dis-
tinguish sadness psychologically. The fact that expressive behavior is
so much more readily accessible to concrete scientific description than
the corresponding psychical processes deserves attention. It suggests
that in the future the study of behavior may well become the method of
choice, when psychologists undertake the task of reducing emotions
and other psychical processes to configurations of basic forces. Already
the analysis of handwriting has led to a number of categories (pressure,
size, direction, proportion, etc.) which invite a search for the corre-
sponding psychological concepts.
Our example will also show why it is fruitless to dismiss the phe-
322 : Rudolf Arnheim
nomena of expression as "mere stereotypes." If it can be demonstrated
that the dynamics of psychical and physical processes are structurally
interrelated and that this interrelation is perceptually evident, the ques-
tion of whether and to what extent the performance and its interpreta-
tions are based on social conventions loses importance.
REFERENCES
1. Allport, G. W., and Vernon, P. E. Studies in expressive movement. New
York: Macmfflan, 1933.
2. Arnheim, R. Experimentell-psychologische Untersuchungen zum Aus-
drucksproblem. Psychol. Forsch., 1928, 11, 2-132.
3. Asch, S. E. Max Wertheimer's contribution to modern psychology. See.
Res., 1946, 13, 81-102.
4. Berkeley, G. An essay toward a new theory of vision. New York: Dutton,
1934.
5. Bowie, H. P. On the laws of Japanese painting. San Francisco: Elder,
191L
6. Biihler, C. The social behavior of children. In C. Murchison (Ed.),
A handbook of child psychology. Worcester, Mass.: Clark University
Press, 1933. Pp. 374-416.
7. Darwin, C. The expression of the emotions in man and animals. New
York: Appleton, 1896.
8. Efron, D. Gesture and environment. New York: King's Crown, 1941,
9. Gottshalk, D. W. Art and the social order. Chicago: University of Chicago
Press, 1947.
10. Koffka, K. Principles of Gestalt psychology. New York: Harcourt,
Brace, 1935.
11. KQhler, W. The mentality of apes. New York: Harcourt, Brace, 1925.
12. Kohler, W. Gestalt psychology. New York: Liveright, 1929.
13. Kohler, W. Psychological remarks on some questions of anthropology.
Amer. J. Psychol, 1937, 50, 271-288. [See pp. 203-221 of the present
book.]
14. Kohler, W., and Wallach, H. Figural after-effects: An investigation of
visual processes. Proc. Amer. phil. Soc., 1944, 88, 269-357.
15. Langfeld, H. S. The aesthetic attitude. New York: Harcourt, Brace,
1920.
16. Lipps, T. Aesthetische Einfiihlung. Z. Psychol., 1900, 22, 415-450.
17. Lundholm, H. The affective tone of lines. Psychol. Rev., 1921, 28,
43-60.
18. Rapaport, D. The Szondi test. Bull. Menninger Clin., 1941, 5, 33-39.
19. Schachtel, E. G. On color and affect. Psychiat., 1943, 6, 393-409.
20. Villey, P. The world of the blind. London: Duckworth, 1930.
2 1 . Werner, H. Comparative psychology of mental development. New York :
Harper, 1940.
The Gestalt Theory oj Expression : 323
22. Willmann, R. R. An experimental investigation of the creative process
in music. Psychol. Monogr., 1944, 57, No. 261.
23. Wolff, W. The expression of personality. New York: Harper, 1943.
24. Woodworth, R. S. Experimental psychology. New York: Holt, 1938.
Solomon E. Asch
SWARTHMORE COLLEGE
THE METAPHOR:
A PSYCHOLOGICAL
INQUIRY
AN OBSERVATION
When we turn to the study of cognition of persons, 1 it is natural to ask
about the differences between things and persons as psychological
stimuli. We do this in order to state more sharply the problems peculiar
to interpersonal cognition. Proceeding in this way, one soon reaches a
conclusion of importance. The study of person cognition is, in good
part, the study of the ways in which we observe and take into account
Reprinted from Person Perception and Interpersonal Behavior, edited by
Renato Tagiuri and Luigi Petrullo, with the permission of the publishers, Stanford
University Press. Copyright 1958 by the Board of Trustees of Leland Stanford
Junior University.
[Additional data are to be found in the following source: Solomon E. Asch. On
the use of metaphor in the description of persons. In H. Werner (Ed.), On ex-
pressive language. Worcester, Mass.: Clark University Press, 1955. Ed.]
1 The writer wishes to thank Dr. Helen Peak and Dr. A. Irving Hallowell for
their thoughtful criticisms of this paper at the Symposium on Person Percep-
tion,
This study was done with the help of a grant from the Ford Foundation.
The Metaphor: A Psychological Inquiry : 325
perceptions, intentions, thoughts, and passions of others. These phe-
nomena we find in persons, not as a rule in things.
This is a sound and fruitful starting point. But as soon as we agree
to this, another question obtrudes itself. Persons and things are also
similar in many respects; there is much they share in common. At this
point, we might be tempted to conclude that those features present in
things and in persons may well be reserved for a general psychology of
cognition; we would thus free ourselves to concentrate on the problems
specific to this area. This would be a sensible conclusion if it did not
obscure a range of questions that may be fundamental. It is possible
that the similarities between persons and things are of definite concern
to the cognition of persons.
This discussion takes its start from the following observation of our
language: when we describe the workings of emotions, ideas, or trends
of character, we almost invariably employ terms that also denote prop-
erties and processes observable in the world of nature. Terms such as
warm, hard, straight refer to properties of things and of persons. We
say that a man thinks straight; that he faces a hard decision; that his
feelings have cooled. We call persons deep and shallow, bright and dull,
colorful and colorless, rigid and elastic. Indeed, for the description of
persons we draw upon the entire range of sensory modalities.
This dual function is not restricted to property terms, or adjectives*
Action names or verbs also possess this dual function. We hunger and
thirst for knowledge; we carry thoughts in our minds; our hopes are
kindled and shattered. We find the same dual function in noun terms;
so, we speak of a flash of wit, of winds of doctrine, and climates of
opinion. "Depth" psychology is today nearly a technical term.
Finally, the language of social experience and action reveals the
same characteristic. We are joined to people with ties and bonds; classes
are high and low; groups exert pressure, maintain distance from other
groups, and possess atmosphere.
There is apparently no aspect of nature that does not serve to express
psychological realities. Light and darkness express the conditions of
knowledge and ignorance, while the action of rivers and storms and
the change of the seasons are the images in which we describe the vicis-
situdes of life and the data of inner experience. Conversely, there are,
it seems, hardly any psychological terms sui generis, denoting psycho-
logical operations exclusively. (Whether there are any terms with an
exclusively psychological reference is a question of interest that we
have not investigated.) Our language has not established a distinct
326 : Solomon E. Asch
vocabulary pertaining exclusively to psychological phenomena. To be
sure, we possess many terms that have become specialized in a psycho-
logical direction (e.g., hope, jealousy, wrath). However, their etymol-
ogy reveals that they once had a clear physical reference.
This way of speaking has, for most of us, every mark of being literal.
Hard describes a person as directly as a rock. Indeed, as a rule, we are
not aware of the double function of terms; they develop their concrete
meaning in each context in apparent independence of the meaning they
have in other contexts. Further, when we do become aware of the
duality, it strikes us, members of the same language community, as
singularly appropriate. It makes considerable sense to speak of a spot-
less character, or of black treason. This is the language we employ
spontaneously to describe the life of feelings and ideas.
These observations bear a curious relation to a long-standing con-
cern of psychologists. We have disciplined ourselves to an attitude of
distrust toward what we call anthropomorphism. Yet here we find, ap-
parently, the reverse tendency developed to the fullest extent. Every
man employs the language of naive physics when he is talking about
psychological matters, including his private experiences.
What is the nexus that unites under the same term, such as straight,
a property of surfaces, paths, movements, and of certain operations of
thinking or trends of character? We propose to explore the grounds of
this phenomenon and to examine its relevance to human thinking,
A QUESTION
What was said so far leads to the following empirical question: does
the characteristic under discussion belong to human languages gener-
ally, or is it restricted to those from which the present observations
started? The question is a twofold one: (a) Do historically independent
languages employ the same morphemes to designate physical and psy-
chological properties? (b) If so, do languages belonging to different
families also agree in the detailed couplings they make?
The questions prompted the investigation now to be described. We
drew up a list of adjectives which, in English, have the double function
mentioned above, in order to establish whether there are morphemes in
other historically unrelated languages that refer to the same sets of
physical and psychological properties. Included among the terms were
the following: warm, cold, hot; right, left; dull, bright, pale, shining;
straight, twisted, crooked; sweet, bitter; colorful, colorless, white, black
The Metaphor: A Psychological Inquiry : 327
(and some other color terms) ; rough, smooth, slippery; dry, wet; clear,
cloudy; deep, shallow; high, low; broad, rounded, sharp; hard, soft.
Our purpose required the examination of languages belonging to
different families and, as far as possible, separated in time and space.
Among the languages selected were: (a) Old Testament Hebrew, be-
longing to the Semitic branch of the Semito-Hamitic family, and dating
back to approximately 1,000 B.C.; (b) Homeric Greek, a member of
the Indo-European family dating back to approximately 800 B.C.; (c)
Chinese, a member of the Sino-Tibetan family (the examples for
Chinese were either colloquial or literary; it was often not possible to
distinguish between them) ; (d) Thai, a member of the Sino-Tibetan
family; (e) Malayalam, a member of the Dravidian family, spoken in
southwestern India; (f) Hausa, a member of the Sudanese family,
spoken in Western Africa by approximately thirteen million people;
(g) Burmese, also a member of the Sino-Tibetan family.
The informants were either scholars or native speakers. The in-
formant was given the English term and asked to state the morpheme
which, in the given language, described the same or most similar physi-
cal property. (Where possible, the morpheme was transcribed phone-
mically; if a phonemic analysis was not available, phonetic transcrip-
tions were used. ) He was then invited to give a few instances of phrases
or sentences in which the morpheme referred to physical properties;
these were transcribed, and the literal translation into English of each
morpheme was obtained, followed by the English equivalent of the en-
tire expression. This first step had the purpose of establishing the dis-
tribution of the term in physical contexts, and of permitting us to de-
cide whether it corresponded to the English equivalent. Thereupon, the
informant was asked whether the same morpheme referred also to psy-
chological properties. If it did, he illustrated the usage with phrases or
sentences; these were first translated literally and then idiomatically. 2
Our first finding is that all the languages here examined contain
terms that simultaneously describe both physical and psychological
qualities.
To what extent is there agreement among languages in the pairing
of physical and psychological properties? For illustration we have se-
lected the results obtained for the terms sweet, bitter, sour; these appear
below.
Each of these languages possesses some morphemes that refer to the
2 1 wish to acknowledge the indispensable aid of Dr. Joseph R. Applegate in
obtaining data from informants and in the work of transliteration.
325 : Solomon E. Asch
Sweet
Hebrew:
Greek:
Chinese:
Thai:
Hausa:
Burmese:
Bitter
Hebrew:
Greek:
Chinese:
Thai:
Hausa:
Burmese:
Sour
Hebrew:
Chinese:
Burmese:
sweet to the soul (said of pleasant words) (Prov. 16:24)
sweet laughter, voice (etymologically linked with the verb of
"please")
a sweet smile (colloquial); sweet, honeyed words = specious
words
to be sweet is to faint; to be bitter is medicine = beware of peo-
ple with whom you have relations
I don't feel sweetness = I don't feel well
face sweet = pleasant-faced; voice sweet = pleasant voice;
speech sweet = pleasant speech
I will complain in the bitterness of my soul (Job 7:11)
bitter pain, bitter tears
bitter fate = hard lot in life (literary and colloquial)
as in English
bitterness of character = an unpleasant disposition; he felt (or
perceived) the bitterness of this talk = he was very upset by
this talk
to speak bitterly = to speak in an unfriendly manner
for my heart was soured (Psalms 73:21)
sour man = a misanthrope; sour heart = sick at heart, grieved
I am very sour toward that person = I detest that person
same paired properties as those found in the other languages. In this
respect, there is impressive agreement. The usages recorded above will
not offer difficulty to an English-speaking person.
The following question may be raised concerning the significance of
the data here presented. Perhaps the terms sweet, bitter, sour merely
designate generally positive and negative qualities on the psycho-
logical side. If so, they would point only grossly to their intended
meanings, and would fail to represent more differentiated characteris-
tics. Closer study does not sustain this conclusion. Sweet does not stand
for just any positive psychological quality; it is not employed, for exam-
ple, to describe courage or honesty. It seems appropriate to conclude
that it describes, in the main, those psychological characteristics that
we may call soothing. Similarly, bitter and sour are not synonymous
with any negative quality. Our records do not contain reference to bitter
or sour fear. 3
3 Related evidence in the same direction will be found in S. E. Asch (3).
The Metaphor: A Psychological Inquiry : 329
There are also divergences in the data, (a) We often found that a
morpheme of a given language denotes only a physical quality. The
reasons for this restriction of meaning we have not had the opportunity
to study; we can, therefore, offer only some partial conjectures. A lan-
guage lacking an extended psychological vocabulary will often not
name qualities of persons that are common currency in other languages.
If, for example, the category of intelligence is not of outstanding im-
portance, we will not find terms such as penetrating or bright to qualify
it. There may also be a failure to employ a given morpheme in the psy-
chological direction because another term already meets this need. In
such cases it would be of value to identify the other morpheme, and to
establish whether it too has a dual reference. Unfortunately we did not
realize the import of this question until late in the investigation. (6) A
given term may, in each language, develop a somewhat different range
of meanings. The data above offer only limited evidence for this state-
ment (e.g., the reference to sweet as specious, in the Chinese). The
evidence, as a whole, contains other such instances. For example, sharp
lips in the Chinese and Hausa stands for fluency or glibness. In these
cases, it is not difficult to discern that the several meanings are by no
means heterogeneous; they are specializations of a more general prop-
erty in which they all share. It is noteworthy that some terms, such as
straight and crooked, develop a remarkably uniform psychological
meaning, in contrast with others, such as hot and cold. 4 (c) In some
instances, by no means the most frequent, we cannot readily deduce or
fully understand after the fact the psychological sense of a term. In
Thai, for example, spoiled heart is to be sad, while heart spoiled is to
be discouraged. In these cases, we can only anticipate the most general
direction. A fuller understanding would perhaps require a familiarity
with the syntactical properties of the language.
We have not found instances that could be called contradictory. The
data permit us rather to speak of agreement and absence of agreement.
It is indeed hard at this point to specify strictly what a contradictory
finding would be. It might seem that directly opposed usages in two
languages would be an instance of contradiction; for example, if the
term designating physical straightness referred also to dishonesty. No
investigator would be content, however, to leave the matter at this
point. He would attempt to discover whether there was a relation be-
tween these meanings evident to the speakers of the language. If no
4 Ibid.
330 : Solomon E. Asch
such relation could be found, the issue would be left in doubt. If, how-
ever, the connection between the terms was clarified, the contradiction
would at the same time dissolve. Clarification, in this context, stands
for a connection that is understandable both to the speaker and to the
inquirer.
The present data are incomplete in one important respect. Since we
confined ourselves to a limited number of languages and to a limited
set of terms, we lack systematic information about the degree of agree-
ment among languages, or about the consistency among languages with
respect to particular terms. We cannot therefore account systematically
for the presence and absence of agreement between one language and
another, or for consistency in respect to any one term, or for the far
more extended incidence of double-function terms in some languages
than in others. Far more detailed investigations would be necessary for
this purpose. This limitation is a barrier to a complete theory of the
phenomena in question, but it does not, we would propose, throw
doubt on the positive findings we have reported. We conclude that
there is no inevitable agreement, but the agreement that occurs is law-
ful and substantial.
THE CONCEPTUAL BASIS OF DUAL TERMS
The substantial agreement we have found among languages establishes
that the dual function here examined is not a fortuitous product. From
the linguistic evidence alone, even if it were more complete, we could
not, of course conclude about the responsible operations. For this pur-
pose we need a psychological analysis. In what follows, we will attempt
to see how the agreements for which we found evidence might arise.
An explanation of the data requires an answer to two questions, one
general and the other special. We need, first, to clarify the fact that
there are terms denoting both physical and psychological properties.
In addition, we have to account for the particular joining of given
physical and psychological properties by an identical term.
That we describe psychological happenings in terms that have also
a physical reference should not raise a thorny issue. 5 We come to know
about persons by observation of their actions; this is the source of our
knowledge about the motives and thought of other persons. We have
5 It should be clear from the context that the terms "physical" and "psychologi-
cal" refer in this discussion throughout to phenomenal data. Both are part of the
phenomenal field. This discussion deals with the relations between experienced
physical and psychological data, not with the objective mind-body problem.
The Metaphor: A Psychological Inquiry : 331
no access to their experiences except through observation of what they
do, of their posture and of the expressions furnished by the changing
appearances of the face and the tone of voice. The observer and ob-
served are distinct systems between whom there is no psychological
continuity; our knowledge of others is mediated by the physical ener-
gies that leave them and reach us. 6 It is therefore by no means sur-
prising that the terms descriptive of experience are so often "physi-
calistic."
The second and main task is to account for the specific joining of a
given physical and psychological experience. We need to understand
how a term such as warm or hard comes to designate certain particular
properties of human action.
We see two main directions that an interpretation of dual terms may
take. One possibility is that we experience certain events in persons and
things as similar in some fundamental respects, and therefore describe
them with the same tenn. If so, the terms of dual reference name simi-
larities that we observe to be intrinsic. The significant alternative is one
in terms of association by contiguity. Dual terms may be the conse-
quence of stable associative connections established between dissimilar
physical and psychological conditions that regularly share some stimu-
lus properties.
Both alternatives, which are obviously not mutually exclusive, rely
on past experience but in ways that are fundamentally different. The
thesis of intrinsic similarity requires that a present datum, say the ob-
servation that a person is soft, make contact with the trace of earlier
experiences pertaining to softness in things. This step is necessary if
the identical term is to refer to both settings by similarity. However, the
contact between the present experience and the aftereffect of the earlier
experiences, or the grasp of their resemblance, is not itself learned, be-
ing the direct result of perceived similarity. In contrast, the associative
interpretation excludes the relation of similarity, relying instead on a
specific association at a given point of time between two heterogeneous
stimuli that need bear no resemblance to each other.
According to the contiguity interpretation, a psychological event is
designated by a physical term when the property corresponding to the
latter is one component of it. Thus we say that a person reddened to
indicate that he was embarrassed, or that he paled when we want to
convey that he was frightened.
Although language often designates an entire situation by naming
a part of it, it is doubtful whether this operation accounts for dual
For a fuller discussion of this point see S. E. Asch (2, Chap. 5) .
552 : Solomon E. Asch
terms completely or even in greater part. It is often not possible to find
the particular physical component in the psychological setting; the use
of the term colorful in relation to persons might serve as an example.
Further, the contiguity interpretation cannot account for the absence
of complete agreement among languages. These, as well as some other
reasons that might be mentioned, limit the value of this interpretation.
Any attempt at an interpretation in terms of stimulus similarity meets
at the outset the difficulty, mentioned earlier, that we often cannot find
in the psychological and physical settings the stimulus conditions that
they presumably share. At this point it becomes necessary to examine
more carefully the content of the terms in question.
What are we trying to say when we call a thing, say the surface of a
table, hard? We mean that it resists change when pushed or pressed,
that it supports other things placed upon it without changing its own
form. Hardness is resistance to change imposed by external forces; it
describes a mode of interaction. Correspondingly, what is soft takes on
the form of things acting upon it, as does the tablecloth that follows
the contours of a surface. What now is the sense of hard when it refers
to a person? It describes an interaction that is formally similar. We see
a man refusing the appeal of another. This interaction we experience
as a force proceeding from one person, having as its aim the production
of a change in the other, which, however, fails to move him, or which
produces resistance. The hardness of a table and of a person concerns
events radically different in content and complexity, but the schema of
interaction is experienced as dynamically similar, having to do with
the application of force and of resulting action in line with or contrary
to it. What holds in the preceding instance applies to the other terms
in the same category. Warm, aside from thermal qualities, stands for
bringing closer, or for drawing into a union, while cold excludes or
isolates. Colorful designates not only the possession of color but also
the presence of diversity capable of eliciting interest. 7
The conclusion we draw, and one we consider essential to a solution
of the present problem, is that the terms under discussion refer not
alone to unique sensory qualities, but to functional properties or modes
of interaction. They do not denote exclusively the "raw materials" of
experience; they are also the names of concepts. A sensationalistic psy-
chology, whose hold upon us has not completely relaxed, obscures this
point and hinders us from seeing that straightness t depth, sharpness en-
7 1 am indebted to Dr. Fritz Heider for an illuminating discussion of this point.
The Metaphor: A Psychological Inquiry : 333
compass far more than those selected aspects that lend themselves to
psychophysical investigation. We need to guard against the unwitting
assumption that these data are as narrow as their current technical
meanings; as a rule they include important dynamic and physiognomic
properties that are as yet less accessible to exact investigation.
The concepts in question have little in common with abstract logical
operations. They are not generalizations of what is common to an array
of different instances. Rather they are concrete cognitive operations in
terms of which we naively comprehend events and similarities between
them.
The conclusion we have reached is that when we describe psycho-
logical events in the same terms we employ for the description of the
forces of nature of fire, sea, wind we are referring to functional
properties they share. We see natural events as conductors of the same
fundamental forces that we find in the human sphere. Therefore we
speak spontaneously of seeing a point, of shedding light on or illuminat-
ing a problem, of penetrating to the heart of a matter. The dual terms
of this study derive from this source, being shorthand names for func-
tional relations and forces.
We have sketched a first approach to the experienced relation be-
tween physical and psychological events. It is helpful in raising a num-
ber of questions that can be studied empirically. It should, for example,
be possible to investigate, in a relatively straightforward way, the de-
velopment in children of the usages here described, and their role in
the mastery of language. This inquiry also suggests a point of departure
for thinking about the general problem of metaphor. It is, perhaps, not
too much to hope that the study of these questions could contribute to
our knowledge of cognitive functions and to a lessening of the gap that
has, too long, continued between psychology and the humanities.
REFERENCES
1. Aristotle. De Poetica.
2. Asch, S. E. Social psychology. New York: Prentice-Hall, 1952.
3. Asch, S. E. On the use of metaphor in the description of persons. In H.
Werner (Ed.), On expressive language. Worcester, Mass.: Clark Uni-
versity Press, 1955.
4. Hornbostel, E. M. The unity of the senses. Psyche, 1927, 1-7.
5. Klages, L. Vom Wesen des Bewusstseins. Leipzig: J. A. Earth, 1921.
6. Kohler, W. Gestalt psychology. New York: Liveright, 1929.
7. Vico, G. B. The new science. Translated from 3d Ed. (1744) by T. G.
Bergin and M. H. Fisch. Ithaca: Cornell University Press, 1948.
Rudolf Arnheim
SARAH LAWRENCE COLLEGE
EMOTION AND FEELING
IN PSYCHOLOGY
AND ART
In many of the more recent writings on art, words like "emotion" and
"feeling" have been made to work overtime. The burden of describing
the content and function of artistic activity has been carried by them
to a considerable degree. Clive Bell asserts that "the starting-point for
all systems of aesthetics must be the personal experience of a peculiar
emotion/' which he calls "the aesthetic emotion" (4, pp. 6, 7). Else-
where we read that art expresses emotion and requires emotion and
that beauty is an emotional element. There are also emotional aspects
of design as well as cosmic emotions; and poetic language is said to
express "feeling or emotion presented as the qualitative character of
imaginal content" (20, p. 145). When the meaning of such terms is
taken for granted, even the kind of sensitive description of which
aestheticians have given us many will not serve to clarify the nature of
artistic activity as compared to other activities of the human mind. The
implied definitions are either so broad that the statements lose concrete
sense, or so narrow that they make art look like an outlet for highstrung
ladies.
If the dissatisfied reader turns to the psychology of emotion he dis-
Reprinted with permission from Confinia Psychiatrica, Vol. 1, No. 2, 1958,
Emotion and Feeling in Psychology and Art : 335
covers that writers from Descartes to Claparede (21, p. 124) have
prefaced their expositions of the subject with the assertion that it makes
for the most deficient and confused chapter in the whole field. A few
decades ago textbook writers were at a loss as to what to say about
emotion (21, p. 18). This situation still prevails in academic psy-
chology, which has limited itself to some irritated discussion of whether
emotion is to signify everything or nothing and, for the rest, has con-
centrated on the interesting physiological aspects of the phenomenon.
At the same time clinical psychology, less inhibited by scruples about
theory, has set off a veritable inflation, comparable to that in aesthetics.
Professionals and laymen speak of "emotional adjustment" and "emo-
tional deprivation," "emotional conflict" and "emotional immaturity.,"
"uncanny emotions" and "emotional disorders." The term "covers" al-
most everything not strictly perceptual or intellectual wishes, atti-
tudes, judgments, reactions, intuitions, opinions, disturbances. The
present paper will suggest that an appropriate trimming of the term
"emotion" might discipline the thinking of the clinicians, relieve the
bad conscience of the theorists, and reopen to scientific research neg-
lected central areas of psychology. In the second part of the paper, an
examination of some of the psychological problems that are hidden in
aesthetics by a catch-all jargon will illustrate questions still to be an-
swered for psychology in general.
Emotion as a category per se. Most psychologists seem to agree with
the layman that there are "emotions," i.e., that among the genera of
mental state there is one called "emotion," which is made up of various
species. There is no agreement on the number and names of these
species. Some authors present lists of "primary emotions," from which
an indefinite number of secondary ones is to be derived. Others have
sample collections of "emotions." Woodworth (26, p. 410) groups
"feelings and emotions" under the following keynotes: pleasure, dis-
pleasure, mirth, excitement, calm, expectancy, doubt, surprise, desire,
aversion, anger. Leeper (17, p. 16) selects "fear, anger, feelings of
guilt, feelings of grief, affection, pride in the doing of good work, enjoy-
ment of beautiful music, and enjoyment of companionship." Here again
one notices that the only common denominator is a negative one:
"purely" motivational processes, such as instincts, and "purely" per-
ceptual and intellectual ones are excluded. Everything else is stored in
the large receptacle for used and discarded matter, labeled "emotion."
And there it rests.
336 : Rudolf Arnheim
The term "emotion," however, suggests a positive meaning; ety-
mologically it refers to agitation, physical, at first, and also mental,
later on. In fact, at an elementary level of psychological theorizing the
term is reserved for states of high-pitched agitation, such as rage and
panic. Hence the view that emotion is disruptive by definition. Once
theorists look beyond the more patent aspects of the phenomenon, they
discover that the spectacular extreme states are nothing but high de-
grees of the excitation inherent in all mental activity. This point has
been made so clearly by several writers that one wonders why it has
not yet become a common property of psychological thinking. Mc-
Dougall (18, p. 148) used the swatting of a fly as an example to show
that an initially almost neutral activity can work up to a clearly "emo-
tional" state with no other change but in the degree of excitement. Simi-
larly Meyer (19) and Duffy (6, 7, 8, 9) pointed out that emotion is
not an additional mode of experience but a component of all experi-
ence. Duffy proposed to drop the old term entirely and to speak instead
of the degree of excitation, that is, the extent of activation or arousal.
To her, this dimension of intensity is one of the two "cross-sectional
concepts" that apply to all behavior, the other being a motivational and
cognitive one, "goal direction (incl. responses to relationships)." Com-
pare here also Massennan's recent energetic statement (22, pp. 40 fL).
Apparently this more adequate view is hard to accept, even for psy-
chologists who endorse it in principle. Koffka, for instance (15, p.
401 ), cites McDougall with approval, insists that emotion must not be
treated as a thing, describes it as a dynamic characteristic applying to
"certain psychophysical processes," and continues a few lines later to
use the term "emotions" in the plural. Schlosberg (24), in presenting
what he calls the "activation theory of emotion," speaks at times as
though he recognized intensity as the dimension of human behavior
corresponding to what is commonly called "emotion" ("activation
would seem to be a very good name for what emotion does to us") but
proceeds to point out "one obvious failing" of his theory: "It deals only
with the intensitive dimension and takes no account of differentiation
among the various emotions."
Emotions as motives or cognitions. Authors who assume that "emo-
tions" represent a specific category of mental state run into instructive
difficulties. Leeper (17), after rejecting disruptiveness and disorgani-
zation as differentiating traits of emotion, presents a criterion of his
own. He describes emotions as a subclass of motive, distinguished from
the more elementary "bodily drives or physiological motives such as
Emotion and Feeling in Psychology and Art : 337
hunger, thirst, toothache, and craving for salt" by their complexity.
When animals become "more complex in their receptor equipment,
motor equipment, and capacity for learning," they develop fear based
on past experience, interest in the offspring not dependent on physio-
logical impulses, an urge to explore, etc. This is the kind of motive
Leeper calls "the emotions." His terminology accomplishes the feat of
denying the emotional quality to hunger and toothache and of identify-
ing emotions by means of properties that by no stretch of the term can
be described as specifically emotional, namely, the cognitive and moti-
vational capacities for learning, understanding, foresight, interest,
curiosity.
Attempts to describe emotions as a subclass of cognition do not seem
to fare better. According to Broad (5), an emotion is a cognition that
has an emotional quality. "To be fearing a snake . . . is to be cog-
nising something correctly or incorrectly as a snake, and for that
cognition to be toned with fearfulness." It will be noticed that in the
example the cognitive aspect of the reaction is limited to the identifica-
tion of the snake rather than including, as it surely must, the recognition
of its fearsomeness. Once this correction is made, nothing seems to
remain of the "emotional quality" but the sheer, unspecific excitement,
which accompanies the (cognitive) realization of danger and the (con-
ative) desire to escape. More in general, the reader of Broad's paper
finds that the characteristics attributed by the author to the "emotions"
(motived and unmotived, misplaced, appropriate and inappropriate,
etc.) do not apply to the emotional but to the cognitive aspects of the
reactions in question. Therefore it seems hardly justified to call these
reactions "emotions." On the other hand, calling them "cognitions"
leaves the description clearly incomplete.
A label that stops research. Academic psychology is driven to call
certain mental states "emotions" because it is accustomed to distribut-
ing all psychological phenomena into the three compartments of cog-
nition, motivation, and emotion instead of realizing that every mental
state has cognitive, motivational, and emotional components, and can-
not be defined properly by any one of the three. In the cases of cognition
and motivation the pigeonholing procedure has had grave conse-
quences for research. Only processes which without obvious distor-
tion of their nature can be described as pure percepts, pure acts of
thinking or learning, pure motives, etc., are being given more than
cursory attention by scientific psychologists. One can strip hunger of
its cognitive and emotional aspects and classify it as a motive, and a
338 : Rudolf Arnheim
glance at a blue triangle may not be seriously falsified by being called a
perception. But the more relevant psychological phenomena resist such
treatment. Hope or pride, for instance, do not fit into either box: they
inseparably combine certain views of people and situations with certain
needs. They are neither motives nor percepts, and therefore scientific
psychologists do not deal with them. They can speak about sex but have
little to say about love because love is no instinct. Hate, ambition,
honesty, grief, trust, happiness, courage, shame, admiration, modesty
things like these have always been considered the characteristic stir-
rings of the human mind; but their study has found no place in scien-
tific psychology. It is left to philosophers who still consider the analysis
of the mind as their business; to clinicians, who are forced by practical
necessity to deal with the essentials; and to common-sense talk inside
and outside of textbooks and lectures. Much of such discussion does
not meet scientific standards; but it is also true that scientific standards
will have to fit the essential tasks of psychology if we wish to develop a
discipline that, by the standards of the great poets and thinkers and by
those of the common man, deserves to be called a science of the human
mind.
Academic psychologists have not simply overlooked the existence of
the salient features of the mind. Since they could not fit them into the
two main boxes, "motivation" and "cognition," they put them in the
third and called them "emotions." The trouble with this third category
has been that, whereas there are enough mental states that can be
trimmed to look like pure motives or pure perceptions, the excitement
of emotion is dominant only in rare extremes and even then is nothing
but an unspecific by-product of what the person perceives, knows,
understands, and desires. Hence the embarrassed searching for some-
thing to talk about in the chapter on the emotions, and the attempts to
equip them with cognitive and conative features in order to give them
body, variety, function. Once such essentially perceptual experiences as
pleasure and pain, and such complex states as sadness and joy, had been
filed under "emotion" there was no stopping, so that by now we have
the above-cited lists of "emotions," which read like the inventory of a
storage vault.
We are not dealing here with a matter of mere terminology nor even
with one of conceptual tidiness alone. By pigeonholing the essential
objects of true psychological interest, academic psychologists have re-
moved them from the grasp of research. No better treatment than dis-
orderly enumeration can be given them as long as they remain labeled
Emotion and Feeling in Psychology and Art : 339
as "emotions" because the emotional component offers no understand-
ing of the processes that cause the excitement. The history of medicine
offers useful analogies. In the olden days people fell sick with "the
fever." What we recognize now as a symptom was then considered the
nature of the illness. Not before the causal processes of infection, etc.,
became the center of interest was it possible to understand the diseases
and to distinguish them adequately from each other. Nowadays people
have "emotional difficulties." But when, for instance, a college student
cannot use his intelligence and his interest, this does not happen be-
cause he is "emotionally upset," but he is emotionally upset because
such conative factors as thwarted wishes and such cognitive ones as his
particular perception of himself and the world around him interfere
with his work. And when a person "reacts emotionally" he does not re-
place reason with emotion but replaces one set of motives and percepts
with another, less appropriate one, for instance, by viewing a contro-
versy as a matter of personal attack. As long as the psychologist thinks
and acts as a clinician the crude labeling by symptom may matter little.
For the psychologist as an academic scientist, however, it has the effect
of putting the phenomenon on a sidetrack and thus removing it from
the work on motivation and cognition, by which alone it can profit and
to which, in turn, it gives its raison d'etre. Something is wrong if the
theorist and experimentalist wait for the clinician to figure out, under
the pressure of his practical obligations, what "anxiety" is. We must not
blame the clinician if our more relevant concepts give off a sickroom
smell; he has done more than his share.
Perhaps the academic psychologist will defend himself by saying
that scientific virtue requires him to limit himself to simpler problems
until he is ready to deal with the complex ones. But it is one thing to
proceed with caution and another not to envisage the goal. By hiding
the central processes under the cloak of emotion we help make our-
selves blind to the true object of our science. Recent attempts to put
the milled-out vitamins back in the bread are thus far conceived as
studies of the mutual influence of "motives" and "percepts." It remains
to be seen whether this promising development will lead to tie explora-
tion of mental states that cannot be reduced either to motives or to
percepts but are compounds of motivational, cognitive, and emotional
factors from the outset.
Models of the past. The historian will have to tell us who contributed
what to this development. Surely it would seem unjust to blame the
great masters of the seventeenth century for it. It is misleading, al-
340 : Rudolf Arnheim
though technically correct, for instance, to state that "Descartes speci-
fied six primary emotions" (27, p. 27) . In Les Passions de YAme, Des-
cartes divided the functions of the soul into active and passive ones
(actions and passions). Passions are affections of the soul, caused ei-
ther by the soul itself or by the body. Those caused by the body are
subjectively attributed either to outer objects (seeing the light of a
torch) or to the body (feeling hunger or pain) , or they seem to be pro-
duced by the soul itself. These latter, the sentiments of wonder, love,
hatred, desire, joy, and sadness, plus their derivations, are named pas-
sions in the more restricted sense. Of them Descartes says that they may
be called perceptions in that they are passive, or sentiments in that the
soul receives them in the same way as the objects of the outer senses,
or, "even better, one may call them emotions of the soul, not only be-
cause this name may be given to all changes that come about in the
soul , . . but particularly because among all the kinds of activity the
soul can have there are none that agitate and shake it so strongly as do
these passions" (article 28). "Emotions," then, means "agitations,"
and the term is used to describe one attribute of the mental reactions
Descartes continues to call "passions" throughout his treatise. In fact,
the first and foremost of them, namely wonder, is said to be devoid of
emotion because it does not regard good and evil but is merely the
surprised attention given to rare objects (articles 70, 71).
Spinoza also keeps the emotional aspect of human reactions in its
proper subordinate place. Where Descartes speaks of passions the
Ethics uses the term affectus, meaning on the one hand affections of the
body by which the power of action in the body is increased or dimin-
ished, aided or restrained, and on the other hand the conscious aware-
ness (ideae) of these affections (book III, def. 3). The basic affection
is conscious desire (cupiditas) which is experienced as pleasure (laeti-
tia) when it is furthered and as displeasure (tristitid) when it is ham-
pered in its striving for greater perfection (book HI, prop. 57). From
these primary affections derive the others, all of which are defined in
cognitive and motivational terms, for instance, love as pleasure that
is, unhampered stirring concomitant with the awareness of an ex-
ternal cause (book III, prop. 13). Primitive though such a definition
may be, it invites and permits further investigation, as distinguished
from our present conception of love as a kind of nondescript quiver.
Feeling. In the field of aesthetics, the generation of Bell (4) and Fry
(11) preferred to speak of "emotions" whereas writers such as Prall
and Langer rely heavily on the word "feeling." Prall (20, p. 147) de-
Emotion and Feeling in Psychology and Art : 341
fines aesthetic experience as "the full felt response to what is directly
given," and to Langer (16, p. 40) "art is the creation of forms sym-
bolic of human feeling." Psychologists speak of feelings as well as of
emotions but the distinction between the two terms is admittedly un-
clear. "Emotion" tends to describe agitation or to designate a state of
mind by the agitation it involves. "Feeling" centers about cognitive re-
actions that seem to defy further decomposition. Aveling, for instance
(21, p. 49), restricts the term "feeling" to pleasure-unpleasure because
they "seem to be kreducible to any other conscious experience whereas
all other 'feelings' can be reduced either to cognitive or to conative
processes." McDougall also appears unwilling to treat "pleasant and
unpleasant feelings" as compounds of percepts and strivings. He con-
cedes to them the status of one of the "three distinguishable but in-
separable aspects of all mental activity" knowing, striving, and feel-
ing (18, p. 146) whereas emotion is considered nothing but the
degree of excitement found in any such activity. To Claparede "feelings
are useful in our conduct while emotions serve no purpose" (21, p.
126) . He lists as feelings not only such percepts as pain but also such
"intellectual feelings" as surprise, curiosity, doubt, that is, the emotion-
ally toned reactions to intellectual or perceptual insight, as well as
James's "feelings of relations" (14, I, p. 245), those seemingly non-
intellectual and nonperceptual cognitions to which reference will be
made later. Here again psychology hardly can be said to offer much
clarification to the theorist of art.
A conceptual framework. The foregoing discussion will have shown
that the misuse of the terms "emotion" and "feeling" is only a local
consequence of a more general defect of psychological reasoning,
namely, the habit of defining any mental act by one of its components.
To remedy this situation, a first step has already been suggested: for
all but the most elementary purposes a mental process must be de-
scribed as a compound of phenomena that come under several basic
concepts. It seems necessary to go further, however, and to think of
these concepts as being subordinated rather than coordinated and as
mutually inclusive rattier than exclusive. What such a conceptual
framework might look like can be indicated here only in the crudest
tentative fashion.
The most inclusive concept is that of perception. It holds the entire
inventory of things and happenings of which a person is aware, either
consciously or unconsciously. We distinguish here, first of all, percepts
of the outer physical world (a horse, my hands, a thunderstorm) from
342 : Rudolf Arnheim
percepts of the inner physical world (a sweet taste, a tightening muscle,
a pain in the neck) . In its current usage the term "perception" is limited
to these two groups of experiences. But among the percepts of the inner
physical world there should also be included those that are not due to
the stimulation of sensory organs outside of the brain; we may call them
informally "intracerebral percepts" (thoughts, desires, memories, im-
agery) as distinguished from the "extracerebral percepts," which are
"sensed." A thought is no less directly perceived than a table. Per-
ception, then, is the top level of the psychological hierarchy.
At a second level, various categories of percepts can be distinguished
from each other. There is, to repeat, the basic distinction between extra-
cerebral and intracerebral percepts, plus the frequent interaction of
both (a listener tiiinks about what he is hearing; a painter organizes an
image of shapes and colors into a meaningful pattern) . Images may be
extracerebral or intracerebral, that is, sensed or imagined, and are dis-
tinguished from the abstract concepts of reasoning. Reasoning is distin-
guished from intuitive thinking ("feeling").
At a third level there are various attributes of percepts. We need to
mention only two, which are shared by all percepts, regardless of cate-
gory. First, all percepts possess directed tension. This attribute is
sensed, for instance, in the upward striving of a poplar tree, the rising
or falling of a musical scale, a cat's rubbing his fur against the leg of
his master, the impingements of colors, sounds, smells, or touch upon
the self, or the expansion of one's breathing chest; it is also perceived
intracerebrally when the self "sends out" its capacities of seeing, hear-
ing, touching for the sensory exploration of the outer and the inner
world, or when we yearn for a loved person or press a train of thought
towards its goal. This omnipresent quality of directed tension is com-
monly called "expression" when it is found in sensory percepts. In
intracerebral percepts it is known as "motivation."
All directed tension has an intensity which is the second universal
attribute of mental processes to be mentioned here. The impact of a
colored light, a noise, the ocean surf, or the sting of an injection needle
may be perceived as being strong or weak, and, similarly, goal-directed
strivings vary in intensity. When applied to intracerebral percepts the
level of intensity can be described as "emotion."
The conceptual framework that has been sketched here implies that
"emotions" and "feelings" are not independent entities but rather
terms referring to properties that are better defined by other psycho-
Emotion and Feeling in Psychology and Art : 343
logical concepts. This approach will now be applied to problems con-
cealed by the two words in the field of the arts.
Among the reasons why the words "emotion" and "feeling" are fre-
quently chosen to describe the artistic processes are the following three:
1. Art is said to be made and sought because it gives pleasure; and
pleasure is described as "an emotion."
2. The particular aspects of reality caught and reproduced by the
work of art are said to be accessible neither to sensory perception nor
to the intellect but instead to a third cognitive capacity, called "feeling."
3. The aspects of reality inherent in the work of art are not only
received as factual information but arouse states of mind that are called
"emotions" or "feelings."
1. Pleasure an unspecific criterion. Words such as "pleasure,"
"delight," "enjoyment" abound in the writings on art; Santayana, for
instance, defines beauty as "pleasure regarded as the quality of a thing"
(23, p. 49). Nevertheless, not much needs to be said about the hedo-
nistic theory.
Since little is gained by describing a mental phenomenon as "emo-
tion," that is, by its mere intensity, we ask instead: What kind of per-
cept is "pleasure"? We find it to be a sensory percept of the inner physi-
cal world, a state of well-being, experienced in the body. Typically the
sensation is connected with some directed tension, such as a drive of
the self toward the pleasure-giving object, or a tonic expansion of the
self, or a positively experienced let-up of pressure. In such a description
nothing can be found that leads to a distinction between pleasure de-
rived from art and pleasure derived from any other source, for instance,
food. The erroneous impression that there is a specific "aesthetic pleas-
ure" is due to the fact that a given component of a mental state receives,
from the total state, modifications that are easily attributed to the nature
of the component itself. If the pleasure derived from a piece of sculp-
ture "feels" different than that derived from food, the difference is due
to the context, since everything that can be cited to characterize it is a
part of the difference between looking at sculpture and tasting food.
Pleasure as such is no more specific than the purr of a cat
Pleasure always indicates that the given situation conforms to some
need of the organism a need for stimulation or distension. But this
attribute of pleasure is, again, not distinctive. Therefore, the hedonistic
definition, according to which art is what produces pleasure, conveys
344 : Rudolf Arnheim
nothing but the trivial fact that art satisfies some kind of need; it pro-
vides a pseudo-answer, which tends to dry up curiosity. Scores of
studies in so-called experimental aesthetics have missed their more
relevant data by being based on the question: "Do you like it?" rather
than: "What do you perceive?"
If we wish to go further, we must move beyond pleasure itself and
ask with what kinds of state pleasure is associated in the particular case
of art. Answers can be drawn from the whole range of percepts, those
of the outer as well as those of the inner world, sensory and nonsensory.
The harmony of a color scheme or a musical chord may give pleasure;
so may the well-balanced, integrated, and clearly directed movements
experienced by a dancer through the muscle tensions in his body. Mem-
ory joins sensory perception when a pleasurable correspondence is
discovered between the configuration of forces observed in a pattern
of shapes and colors and that inherent in a significant life situation.
These are not attributes but causes of pleasure. By examining the
conditions that produce pleasure, art can be distinguished from other
sources of satisfaction. It also becomes possible to make inferences
about the nature of the specific needs served by art. Such an approach
invites research rather than leading it into a dead end.
2. "Feeling" as perception and unconscious judgment. There is
common agreement on the fact that art presents sensory patterns,
images, and thoughts not for their own sake but as mere forms, capable
of transmitting something else. The final content of the work of art is
usually described as "emotions" or "feelings." Thus, Roger Fry calls
the work of art "an expression of emotions regarded as ends in them-
selves" (11, p. 29). In the older writers, assertions of this kind refer
mostly to internal states such as fear, joy, sadness, whereas the more
recent ones have broadened the offering. According to Prall, the feel-
ings conveyed by the artist "may range from the hard look of iron ma-
chines or polished brass railings to delicate shallows of light among
grass stems, from feelings of the lightest gaiety to feelings of fate and
doom" (20, p. 160).
Art, then, deals with the kind of mental state that as I showed
earlier psychology has failed to investigate. Psychologists have been
satisfied with attaching labels. Accordingly, when they experimented
with "expression," that is, with the manifestations of such mental states
in perceivable patterns (a human face, a piece of music) they did not
investigate these manifestations either. They concentrated on finding
out what qualities of sadness, or calmness, or passion observers were
Emotion and Feeling in Psychology and Art ; 345
willing to attribute to the object and to what extent this was done "cor-
rectly." But they rarely asked what particular audible or visible features
of the object determined the reactions of the observers. The question of
how observers accomplished their feats was not made a part of the
study but taken care of by some untested assumption. The traditional
belief, for example, that in such experiments observers rely on associa-
tions established in the past, makes it unnecessary to study the nature
of either the mental or the behavioral phenomena involved. Correla-
tions among unknowns will suffice. Aestheticians have hardly done
better, except that their answer to the question: How do people do it?
tends to be: By feeling! rather than: By association. They prefer to
attribute die accomplishments of the artist and his public to a cognitive
faculty sui generis, neither perceptual nor intellectual.
Suppose now we wished to find out why people see pride in the Napo-
leonic stance of an actor or a marble figure; and suppose we were un-
willing to gloss over the problem by simply relying on the assumption
that observers know from past experience how proud people carry
themselves. In examining the stimulus we would notice a head raised
above the plane of human interaction, eyes closed or looking upward, a
body stretched to its greatest height, a chest ostentatiously presented,
hands withholding acceptance and cooperation by hiding between the
buttons of the waistcoat or in the trouser pockets. These features would
not suggest any kinship with pride as long as pride were nothing but
an unknown "emotion-" But if we tried to do our psychological duty
by studying a proud person's state of mind, even the most primitive
examination would show that such a person perceived in the outer
world creatures so inferior to his own excellence that they could only
be dominated, not dealt with as equals. We would find him striving to
tower over them and to impress his superiority upon them. We might
also notice "emotion," that is, the degree of tension or excitement in-
volved, but it would tell us nothing specific. Rather would the places
and directions of the vectors observed in the outer physical world (the
image of the Napoleonic figure) be discovered to have an isomorphic
affinity with the vectors operating in the "intracerebral percepts" of
proud rising, withdrawing, displaying, etc. The obstructive split be-
tween outer dynamics and inner dynamics, caused by the separation of
"perception" from "motivation," would be overcome.
Once we examine concretely what people do when they apprehend
expression, it becomes apparent that the instrument they use is per-
ception, not some other mysterious cognitive faculty. Perception not
346 : Rudolf Arnheim
of the static aspects of shape, size, hue, or pitch, which can be measured
with some yardstick, but of the directed tensions conveyed by these
same stimuli. Which particular properties of visible and audible pat-
terns contain the dynamic features has been shown elsewhere (2, 3).
Decisive for the present discussion is that they are as immediately per-
ceptual as are dimensions, quantities, or locations. It is owing to a tra-
ditional preference for static conceptions that Western theorizing has
tended to exclude the dynamic aspects from perception and to assign
them either to a kind of internally generated projection or to the special,
negatively defined faculty of "feeling" ( 1 ) .
The fuller way of perceiving, which stresses directed tensions, is a
prerequisite for, but not a monopoly of, the aesthetic attitude. Once it
is understood that the capacity to apprehend artistic expression grows
out of unrestricted everyday observation, an artificial distinction will
break down, and research in the general psychology of perception will
cross-fertilize with that in the field of art.
Our assertion that it is uneconomical and misleading to assume the
existence of "feeling" has been derived so far from a discussion of the
relatively simplest processes, namely, the passive apprehension of per-
ceptual patterns, both sensory and nonsensory. But art involves more
than perceiving the dynamic quality of, say, rapid attack in the swoop
of a seagull or the argument of a trial lawyer. It requires, for instance,
an ability to judge the correctness of a compositional structure in the
visual arts or music as to balance, unity, rhythm. In such tasks the in-
tellect, which can apply conceptual rules such as those of proportion,
plays only a minor part. Rather is the judgment of right and wrong
performed by mere looking and listening, often without any awareness
of the criteria that determine it. It takes a stubborn effort successful
only in the good teacher or critic to make some such criteria explicit.
But again we should be most reluctant to attribute these "intuitive"
judgments to a cognitive faculty sui generis. What guides the artist or
connoisseur turns out to be the same tensions that we have recognized
as the very basis of perception. In a successful composition the forces
that make up the pattern are in balance; but, if the work is incomplete
or unsuccessful, pushes and pulls within the pattern indicate not only
that something is wrong but also where correction should apply and in
what directions it should proceed. Here again, artistic "feeling," to the
extent to which it is an intuitive response to the given structure, is
nothing but ordinary perception. It is a capacity not different in princi-
Emotion and Feeling in Psychology and Art : 347
pie from that displayed by a tightrope walker or even of a dog who
balances a stick in his mouth.
One step further, and we remember that art goes beyond the per-
ceivable pattern. Artistic expression has always a semantic function;
the painted and carved images stand for referents, and so do the shapes
of music and the events and thoughts described by the poet. Therefore,
art requires the judging of meaning, relevance, and truth, and again the
task is accomplished largely by the mysterious capacity of "feeling." Is
the artist's presentation relevant for, or true to, the kind of thing for
which it stands? Does the particular sunflower that models for the artist
or is seen on his canvas by the beholder look the way sunflowers look
to them? Does it conform to their conceptions of the nature of flowers in
general? Does its dynamic pattern of, say, graceful heaviness capture a
way of behavior worth capturing? These judgments are not simply per-
ceptual, although ultimately they refer to percepts. They are based on
the observer's entire life experience and involve his convictions, values,
biases, memories, preferences. They presuppose translations of intel-
lectual conclusions into perceptual images, and vice versa; and at the
moment at which they take place they necessarily require a lightning-
fast comparison of the given individual object with this complex pre-
cipitate of thought and vision.
Here again "feeling" is defined merely by exclusion. It describes a
cognitive capacity that is neither mere perceiving nor based on con-
scious logical operations of the intellect. Here, too, no guiding criteria
may be observable. How does the human mind perform such feats?
In searching for the answer, the student of art discovers one of the
most astonishing gaps in the program of modern psychology. If we
remember to how large an extent all human activity, from the driving of
a car and the adding up of digits to the dealing with other people and
the solution of creative problems in art and science, is done "intuitively"
or "mechanically," that is, without conscious awareness of the processes
that determine the action, it is hard to understand why psychologists
limit their interest in this basic aspect of our functioning to the special
case of the Freudian repression. The mere fact that so many of our
judgments of right and wrong are made before we know why we make
them should be sufficient to alert the profession. William James esti-
mated that "a good third of our psychic life consists in these rapid pre-
monitory perspective views of schemes of thought not yet articulate"
(14, 1, p. 253).
348 : Rudolf Arnheim
Judging by the accomplishments of these unconscious performances,
we must conclude that mental capacities of the highest order are at
work: reasoning, selecting, comparing, problem solving by restructur-
ing, etc. But are these processes the same as their counterparts, known
from conscious experience? And if so, do they function the same way?
What should be made of the ample anecdotal evidence according to
which scientific and artistic thinking of the highest caliber can solve,
under the protective cover of the unconscious, problems whose solu-
tions the conscious mind vainly struggled to find? (12, 13). Is perhaps
unconscious creation less subject to the sets or other constraints that
hamper conscious invention? Psychoanalysis has given illustrations of
what distinguishes conscious from unconscious reasoning but it has
also uncovered certain striking similarities. For instance, the dream
mechanisms of condensation, fusion, displacement of function, and
symbolism (10, chap. 6) closely resemble typical, and often conscious,
operations of the artist.
Unless concrete research suggests that the processes here involved
are fundamentally different from those operating in conscious experi-
ence there is no reason to imitate Moliere's doctoral candidate, who
explained the sleep-inducing effect of opium by a virtus dormitiva.
3. The aesthetic "experience" Up to this point, art has been dis-
cussed as though it were nothing but a transmitter of information about
visible objects or audible events, subject matter, and the significant pat-
terns of forces that are inherent in such material. But art not only
acknowledges the presence of, say, agitation in a piece of music or
bewildered wavering in the thoughts of Hamlet; it makes the artist and
his public "feel" these dynamic states as personal "experiences." Tol-
stoy defined art as "a human activity consisting in this that one man
consciously, by means of certain external signs, hands on to others
feelings he has lived through, and that others are infected by these
feelings and also experience them" (25, p. 123). What is the nature
of such experiences?
In the aesthetic discussions of this question the use of the word "feel-
ing" has made for a curious pseudo-problem. The notion was intro-
duced by the theory of empathy, according to which architecture or
music owed its expressiveness to past experiences of the observer, pro-
jected by him upon suitable objects. When the theory was abandoned
by all those who realized that properties of the percepts themselves are
responsible for the bulk of expression, the terminology remained and
with it the problem of how a lifeless object could manifest "feelings"
Emotion and Feeling in Psychology and Art : 349
if they were not cast upon it by some "pathetic fallacy." According to
Langer, who entitled her recent book on aesthetics Form and Feeling,
"The solution of the difficulty lies ... in the recognition that what
art expresses is not actual feeling, but ideas of feeling" (16, p. 59).
This formulation hides the problem by the word "expresses"; for either
Langer's "ideas" are received, the way ideas usually are, as intellectual
information, in which case the main difference between an artistic and
a purely informative statement is neglected by the formulation; or the
ideas are "felt," that is, are "feelings," in which case the dilemma re-
mains what it was.
Once the terminological camouflage is removed, it appears that three
different states or perhaps three degrees of one state must be dis-
tinguished in the phenomenon under discussion.
(a) At a first level, the difference between the mere apprehension of
information and the fuller artistic "experience" is identical with that
between the static and the dynamic properties of percepts. A piece of
music can be played or listened to in a way that conveys nothing but
an assembly of pitches and durations. The resulting sounds are "dead"
because by lacking dynamics they lack the main structural quality of
life; and they are no concern of anybody because only by exhibiting
patterns of forces that it shares with the human mind can music build
a connection between two media that otherwise are alien to each other:
the extracerebral world of pure sounds and the intracerebral one of
human striving. Similarly as long as Hamlet's monologues are only
understood but not perceived as a zigzag course of motivational vectors
they remain in the domain of the psychologist or historian. So far, then,
there is nothing in aesthetic "experience" that was not discussed in the
preceding section as a property of ordinary perception.
(b) Dynamic features will enliven not only the percepts of the outer
physical world, such as music, statues, or buildings, but also those of
the inner world, notably the kinesthetic sensations of the actor and the
dancer. As a carrier of expressive patterns the performer's body is a
perceptual object, just as removed from the self of the perceiver as a
painting he may be looking at; and he directs and controls his instru-
ment according to kinesthetic pushes and pulls just as the painter uses
visual dynamics to judge his composition on the canvas. In a similar
fashion the actor or dancer perceives and controls the dynamics of the
"intracerebral percepts" that are part of his performance, the reckless
striving of the tyrant, the panicky speculations of the fugitive, or the
passive yielding of the lover. Such a technique, which takes advantage
350 : Rudolf Arnheim
of the full perceptual experience, is clearly distinguished from the cold
display of appropriate gestures on the stage.
These examples show that the various functions of a person's body
or mind can serve him as vehicles for expressive dynamic patterns. The
technique of the performer helps to illustrate what happens when the
experience of looking at a painting or listening to music goes beyond
the perception of directed tensions in the visible and audible object:
while watching one of Picasso's "blue" figures the beholder may find
that the bending under pressure, so compellingly perceived in the paint-
ing, invades his own attitude of body and mind through a kind of
resonance. Yet just as the actor's perception of what is going on within
himself can leave him in the position of the detached observer and
director, so the personal participation in the dynamics found in the
painting or music does not necessarily entwine the beholder's or lis-
tener's self.
(c) The difference between "feeling" the part of Romeo and delud-
ing oneself into "being" Romeo has been amply discussed by the prac-
ticians and the theorists of the stage. But the use of a loose terminology
makes it hard to distinguish psychologically between cases in which the
self acts as a detached perceiver of a dynamic state and others in which
the self is the very center of such a state. The difference between feeling
fear within oneself and being afraid is illustrated perhaps by the well-
known effect of adrenalin injections. Yet the fact that a theatergoer may
tremble with Romeo and nevertheless enjoy the show at the same time
has been considered a troublesome contradiction.
What happens when the self does become the center of the dynamic
state it perceives? Do we have to assume that a still broader range of
the mind than that envisaged under (b) is seized by the perceived pat-
tern? Or is the process more in the nature of a contact established with
the affected regions? There is no answer as long as the lack of research
keeps the discussion at a crudely amateurish level.
The familiar notion of "aesthetic distance" rejects the involvement
of the self as discussed here under (c) and recommends an attitude
compatible with either (a) or (b). But when Tolstoy speaks of the
"feelings" by which the reader or listener is "infected" he means states
of the self. So, presumably, does Langfeld when he asserts that both
phylogenetically and ontogenetically "we can trace in the development
of aesthetic appreciation a gradual diminution of the emotional re-
sponse" (21, p. 350).
The range of responses to perceived dynamic patterns is a problem
Emotion and Feeling in Psychology and Art ; 351
not limited to matters of art. In general psychology it connects with the
study of a variety of phenomena, such as pity, sympathy, self-control,
self-estrangement, persona, spontaneity, or affectation. Here are tasks
that will be noticed and approached when "emotions" and "feelings"
will have ceased to obstruct the view.
SUMMARY
Many fundamental psychological processes have been filed away under
the heading "emotion" and thereby removed from the necessary and
fruitful contact with research on motivation and perception. "Emotion"
is not a kind of mental activity but merely the amount of excitation ac-
companying any mental activity at all. The psychology of art is used as
an example to show that the vague term "emotion" hides from sight a
number of basic problems, such as those of aesthetic pleasure, per-
ceived expression, intuitive judgment, and the resonance effect of dy-
namic states.
REFERENCES
1. Arnheim, R. The Gestalt theory of expression. Psychol. Rev., 1949, 56,
156-171. [See pp. 301-323 of the present book.]
2. Arnheim, R. Perceptual and aesthetic aspects of the movement response.
J.Pers., 1951, 19,265-281.
3. Arnheim, R. Art and visual perception. Berkeley and Los Angeles:
University of California Press, 1957.
4. Bell, C. Art. London: Chatto and Windus, 1931.
5. Broad, C. D. Emotion and sentiment. /. Aesthet. Art Crit., 1954, 13,
203-214.
6. Duffy, Elizabeth. Emotion: an example of the need for reorientation in
psychology. Psychol Rev., 1934, 41, 184-198.
7. Duffy, Elizabeth. The conceptual categories of psychology: a suggestion
for revision. Psychol Rev., 1941, 48, 177-203.
8. Duffy, Elizabeth. An explanation of "emotional" phenomena without
the use of the concept "emotion." /. gen. Psychol, 1941, 25, 283-293.
9. Duffy, Elizabeth. Leeper's "motivational theory of emotion." Psychol
Rev. t 1948, 55, 324-335.
10. Freud, S. The interpretation of dreams. New York: Macmillan, 1927.
11. Fry, R. Vision and design. London: Chatto and Windus, 1929.
12. Ghiselin, B. (Ed.) The creative process. Berkeley and Los Angeles:
University of California Press, 1952.
13. Hadamard, J. S. An essay on the psychology of invention in the mathe-
matical field. Princeton: Princeton University Press, 1945.
552 : Rudolf Arnheim
14. James, W. The principles of psychology. New York: Dover, 1950.
15. Koffka, K. Principles of Gestalt psychology. New York: Harcourt,
Brace, 1935.
16. Langer, Susanne K. Feeling and form. New York: Scribner, 1953.
17. Leeper, R. W. A motivational theory of emotion to replace "emotion as
disorganized response." PsychoL Rev., 1948, 55, 5-21.
18. McDougall, W. The energies of men. New York: Scribner, 1933.
19. Meyer, M. F. That whale among the fishes the theory of emotions.
PsychoL Rev., 1933, 40, 292-300.
20. Prall, D. W. Aesthetic analysis. New York: Crowell, 1936.
21. Reymert, M. L. (Ed.) Feelings and emotions: the Wittenberg Sympo-
sium. Worcester, Mass.: Clark University Press, 1928.
22. Reymert, M. L. (Ed.) Feelings and emotions: the Mooseheart Sympo-
sium. New York: McGraw-Hill, 1950.
23. Santayana, G. The sense of beauty. New York: Scribner, 1896.
24. Schlosberg, H. Three dimensions of emotion. PsychoL Rev., 1954, 61,
81-88.
25. Tolstoy, L. What is art? and Essays on art. New York: Oxford Univer-
sity Press, 1942.
26. Woodworth, R. S. Psychology. (4th ed.) New York: Holt, 1940.
27. Young, P. T. Emotion in man and animaL New York: Wiley, 1943.
si
128885
