You are on page 1of 6

History of Psychology Copyright 1998 by the Educational Publishing Foundation

1998, Vol. 1, No. 1,21-26 1093-4510/98/33.00

WOLFGANG KOHLER AND GESTALT THEORY:


An English Translation of Kohler's Introduction to Die
physischen Gestalten for Philosophers and Biologists
Rudolf Arnheim
Harvard University

This article presents an English translation (from the German) of one of gestalt
psychology's most significant documents, first published in 1920 in Wolfgang
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Kohler's Die physischen Gestalten in Ruhe und im stationaren Zustand (The


This document is copyrighted by the American Psychological Association or one of its allied publishers.

Physical Gestalten at Rest and in a Stationary State). The book it introduces both
embodies Kohler's extension of gestalt theory into new domains and did much to
ensure the broad impact of these ideas and approaches. This introduction itself well
illustrates Kohler's own thought processes both as his ideas emerged and as he
sought to convince his readers of their value. Despite the fact that they are more than
70 years old, Kohler's words have many implications for late 20th century
discussions of the relationships among psychology, physiology, and physics.

Translator's note. In 1920 Wolfgang Kohler, one of the pioneers of gestalt


theory, published his first book, which dealt with physical gestalten. He reasoned
that as long as the theory remained limited to psychology, its claim of describing a
general law of nature could not be verified. As a psychologist, trained also in
physics, he referred to fairly simple examples, such as the overall equilibrium of a
closed system or the stable flow of electricity in a circuit. Kohler's book was titled
Die physischen Gestalten in Ruhe und im stationaren Zustand. Literally translated,
it reads The Physical Gestalten at Rest and in a Stationary State. By the word Ruhe
Kohler intended an immobile state.1 The only English version of the book I know
of is contained in Willis D. Ellis's Sourcebook of Gestalt Psychology.2 It is not a
literal translation of Kohler's text but merely a rendition of his theoretical
discussion, including the two introductions. Of Kohler's actual text I have here
translated the first introduction, addressed to philosophers and biologists. (The
second is addressed to physicists.) It is a remarkable document in the history of
science, showing what is going on in the mind of the initiator of a new paradigm,
his hedging and hesitation, his scrutiny of all possible alternatives, but also his
certainty of the objective he seeks to make convincing. Beyond the historical
significance of the document it anticipates the discussion of the relations among
psychology, physiology, and physics, as relevant today as it was then. Kohler's text
is written in the highly intricate language tradition of German philosophy but also
with its lucid logic. It requires the patience of readers, as it did that of Kohler
himself. Given the purpose of this introduction, it is pitched toward the natural
sciences and does not call for an explicit discussion of the social sciences, but the
relevance for them of many points made here will be obvious.

Rudolf Arnheim, Professor Emeritus of the Psychology of Art, Harvard University.


Correspondence concerning this article should be addressed to Rudolf Arnheim, who is now at
1200 Earhart Road, Number 537, Ann Arbor, Michigan 48105-2768.

21
22 ARNHEIM

Introduction for Philosophers and Biologists


by Wolfgang Kohler
In 1890, Christian von Ehrenfels gave the name "gestalten" to psychological
states and events, whose characteristic properties and effects cannot be reduced to
the sum of their so-called parts.3 One came across gestalten in visual space, and the
name was then adopted for all other gestalten, not only in vision but also for
melodies or intellectual conceptions. Gestalten are fundamental patterns, whose
particular properties are by far superior to those of the materials of which they
consist. If one assumed that local color sensations or single sound sensations and
the meanings of single words are to be called "parts" of gestalten, then thoughts or
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

the particular impression of visual figures, the character of musical themes or the
This document is copyrighted by the American Psychological Association or one of its allied publishers.

meaning of a verbal sentence would certainly be more than the corresponding


spots of color, auditory sensations, or meanings of single words. The same spatial
gestalten can be presented in different colors and at different locations of the visual
field. The same musical theme can be given at different pitch levels. Therefore,
such "absolute elements" do not constitute the particular nature of the total
structures. If one looks at the meaning of a verbal sentence as a whole and
compares it then with the meaning of each word in isolation, one misses in the
latter version the meaning of the context.
Such structures, which as wholes have special properties and therefore must be
conceived for good reasons as units, are the ones that we call "gestalten." The
concept is of recent origin and therefore still quite unprecise, but even so its
scientific fruitfulness seems essentially greater than the principles that earlier were
considered fundamental in the life of the mind. All research approaches lead to the
same central concept, although from different sides and in different special fields.
In consequence, the characteristic features of gestalten will gradually become
more clearly prominent. Psychological research oriented in a forward direction can
turn the curious features that have been observed into objects of sharply defined
properties, subject to well-known laws and sensible subdivisions.
If psychology were the first empirical science mankind tried to apprehend and
human thinking were without fundamental ideas developed elsewhere, we would
have to consider this way of striving ahead, at least tentatively, as the only one, on
which the new conception could be brought to further concretenesswithout
bothering about further empirical support. This, however, is not the case. Instead,
other empirical sciences have preceded psychology for centuries. This is sufficient
to make us think constantly in terms of much older categories; and it opens a
second way of solidifying the gestalt concepts. This approach is geared toward the
past. If some of the features of the new psychology happen to exist already in the
older conceptions, we want to search for them, and if we find them we want to
establish the connection between the two areas. Only if such other areas do not
exist, are we left with the empirical forward-procedure. In that case, we have to put
up with the dissatisfying idea that whatever is empirically accessible does not
always constitute an inherently closed system, and that nothing in earlier thinking
corresponds to the fundamental psychological category now discovered.
In fact, a widely accepted view suggests precisely that this backward-oriented
approach is not feasible. Some of its proponents, probably owing to philosophical
premises, are inclined to confine the laws of mental life as much as possible to their
own kind. They point to gestalten and say: Look how the generative power of the
WOLFGANG KOHLER AND GESTALT THEORY 23

mind creates patterns of an entirely new kind, which have no parallel in any area of
nature; elsewhere things are so "mechanical"! Others are entirely convinced that
consistent reasoning has built up the system of older thought objects, and that
therefore newer conceptions should conform or be subordinated to the traditional
ones. Since they see no way the properties they assume do exist in gestalten can be
made to conform to proven categories, they cautiously keep away from this
innovation. They believe it might be a misinterpretation of a shaky and empirical
assumption. It may turn out to be a passing fashion.
In fact, even those who are already accustomed to working with the concept of
gestalten as something psychologically real, sometimes feel a slight impediment in
using this procedure, since it means dealing with objects that have little of an
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

empirical base and are not yet legitimized by theoretical support. Undoubtedly, it
This document is copyrighted by the American Psychological Association or one of its allied publishers.

does not come naturally to people to treat consciousness and mental behavior
scientifically. Using this approach they do not even feel safe with self-observation.
When one comes across something so foreign, one likes to turn to the sciences
concerned with the outer world and especially with inorganic nature. In those
sciences, the ground has long been more stable, and one wants to seek there a
control or at least an analogical confirmation. Not only does it seem possible, but
one feels the strongest need to explore gestalten, or at least something resembling
them, not just in ephemeral sensations but in the firm shapes of inorganic natural
events. If possible, one hopes to use for gestalt theory the clarity and definitude
that have been attained by mankind so much earlier in observing and thinking
about physical matters than it was in psychology.
A first glance at the natural sciences, however, is discouraging enough. Their
proudest distinction, exactitude, seems to reject right away anybody who searches
in their area for patterns that in some sense can be said to create a context and yet
are characterized as specific units. If furthermore one insists that these units
display properties and effects that go beyond the sum of the parts, that, in other
words, there are objects in natural science that are "wholes," "more than the sum
of their parts," then indeed one will be getting the feeling that one is asking for
something directly contrary to the foundations of the exact sciences.
For awhile, chemistry may raise something like a hope. After all, the
compound KCN [potassium cyanide] looks like a unit of the kind we are looking
for: It has precise properties, which certainly cannot be found in its elements,
potassium, carbon, and nitrogen. Some theorists have thought of chemistry as an
analogy to gestalt psychology, but they have found in it nothing better than a vague
image. Some of the most important properties of psychological gestalten cannot be
shown to exist in chemical compounds. Perhaps, however, with all the miraculous
discoveries in natural science, this additional one could happen, namely that the
apparently new pattern suggested in only a preliminary way by chemistrywhich
is barely a century oldwill reduce itself some day, as physics and chemistry
progress, to a fundamental physical principle. Therefore, if gestalten are now alien
to physics, we cannot be sure how long chemistry will still offer something
somewhat resembling it. In any case, the concept of a chemical compound itself is
still not clear enough, so that the analogy to gestalten could help us to obtain a
sufficiently stable specification of the psychological category.
It would be different, of course, in the realm of physics. Anybody who after
quickly surveying the basic conceptions of physics does not lose his courage,
24 ARNHEIM

could perhaps hope, by systematically scrutinizing its different areas and concepts,
to find something that, at least in some respects, had already been thoroughly
thought about and contained properties of gestalten. But the realm of physics is so
vast and looks so little like gestalten, especially as long as one is without any
helpful pointer, that one soon gets tired. The following consideration also turns one
away from such a systematic scrutiny.
Regardless of whether or not gestalten do in fact exist in physics, physicists so
far have not seen any reason to single out those particular instances from the
wealth of the whole field. Otherwise one would probably have become aware of
this special area, and it would have happened long agoto the benefit of
psychology. We would not have to search for it. As of now, neither the physicists'
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

customary approach, so successful in their own work, nor the terminology


developed in it, seem to provide anything like it. Without a particular lead, we
would handle the facts of inorganic nature with the methods and terminology of
physics, and since physics apparently does not point to what we are looking for,
gestalten belonging to the material of physics would escape us entirely. The
thinking and language of natural science are firmly controlled by other points of
view. They would be like railway tracks taking us straight away from our
destination, much though they fit the demands of the usual physical research.
There has hardly ever been a time like ours, in which the question has been
more insistently debated whether the essential traits of organic events can be
reduced completely to the laws of inorganic ones or whether special forces, found
only in organisms, play a part everywhere in chemical and physical functioning.
These latter forces would imprint the higher forms upon the organism as mere raw
material. If one asks the vitalists4 which phenomena make them adopt this second
view, it turns out that many are taken with the "compactness" and the behavior of
the organism, for the reason that it seems to contain its parts and organs not like the
elements of a sum. Instead, they seem to make the totality into a whole of new and
valuable properties. The most characteristic behavior of the organism does not
look like the resultant of an uninhibited chaos of many forces. It looks rather like
the orderly response of the unit to its actual vital conditions in the whole. Also the
striking appropriateness of the structure, the functioning, and the responses are
probably seen in the context of the fundamental character of the organic. This
raises for biology the general problem that resembles somewhat the general
problem in psychology. The affinity becomes evident also when the vitalists single
out the nature and effect of higher psychical faculties to explain the strange
behavior of biological forces. The fundamental form and character of those higher
mental patterns turns out more and more to be the typical gestalt. Already by now,
gestaltlike behavior is considered explicitly as the essential quality of the
organism.
This means that a need to make the gestalt type of patterns and behavior define
conceptions of thought exists everywhere in biology. The specifically vitalist
answer to this demand, however, would exclude the existence of truly physical
gestalten.
And yet, if one does search for them, one seems to find that the typical traits of
organic behavior suggest the direction in which one has to look in physics to find
gestalten. In that area the physical patterns would show something of the orderly
units of the organic world. This direction has been taken repeatedly for biological
WOLFGANG KOHLER AND GESTALT THEORY 25

purposes; but experience has shown that it does not get the research very far. It
yields all sorts of vague analogies, such as that between crystals and organisms.
General biology and psychology come closest to each other in the functioning
of the nervous system, especially with regard to the physical foundation of
consciousness. At this point we cannot get around the imperative need to think of
organic behavior as gestalten, because it is expected to correspond directly to the
higher levels of psychological behavior. The consequence and outstanding
significance of this demand was clearly recognized first by Max Wertheimer,5 who
attributed to gestalten a degree of reality not allowed them before to any extent.
More recently Kurt Koffka6 insisted with Wertheimer that the central physical
processes be not viewed as sums of single stimulations but as wholes.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

If, for the reasons mentioned above, the discussion of the problem is directed
specifically to physical gestalten, one might present the gist of the physiological
and the physical question most fruitfully as follows: If physical gestalten exist,
there is a well-founded hope of understanding central physiological processes as
special cases of those physical gestalten. Conversely, every concrete case in which
gestalten are experienced and which therefore, according to Wertheimer's and
Koffka's postulate, are accompanied by physical gestalten, offer specific, definite
suggestions as to which particular kinds of process need to be considered under the
given circumstances. Such a case tells us, therefore, where in physics to look for
gestalten.
This is what we needed. Instead of the general but only vaguely defined task
we have now a much more specific and also more concrete undertaking. Nothing is
lost, if for the time being we limit ourselves to this more specific task.
For one thing, the first step of the theory becomes something like a discovery
of physical gestalten, at least for the corresponding physiological correlates. It
safeguards our category by connecting it with older and more mature thinking; and
it is connected with a step forward in physiology.
Secondly, the most general properties of gestalten differ until now so strangely
from the usual objects of our thinking that one can already foresee: anytime we
become aware of similar examples in physics, we must have hit the principle of the
total solution. Either there exist in physics no examples of gestalten at all, or when
those particular instances have convinced us that they do, we can derive from them
insight into the more general problem. For, since physics as an abundantly rich
system is evident to the eyes of all of us, the question can only be, under what
conditions does it contain gestalten. This will permit us to discover the point of
view from which to recognize gestalten in physics.
If this is correct, one can expect, thirdly, that the special route from examples
of gestalt processes found in the nervous system to physics will eventually lead to
other and broader roads. These will get us back to biology and to a more
comprehensive physical treatment of neural and generally organic gestalt proc-
esses.
Somebody who has laboriously tried to at least start to make this procedure
come into existence knows best that such a program cannot become scientific
reality in one swoop. I hope, however, to have been successful in this book to the
extent that the first difficulties can be considered overcome. Of course, this
increases the number of research questions, which now have been rendered more
concrete. Exactly those phenomena with which we have long been familiar and
26 ARNHEIM

which therefore look all too obvious are the ones that soon reacquire the more
vivid and attractive colors of unknown things.
I was not able to entirely foresee all the consequences. Even after the first steps
have been taken care of, there are others whose implications I more or less
recognize; however, they will be better dealt with when they no longer look just
unfamiliar but instead compelling.
Notes
1. Wolfgang Kohler, Die physischen Gestalten in Ruhe und im stationdren Zustand
(The Physical Gestalten at Rest and in a Stationary State) (Erlangen, Germany: Verlag der
Philosophischen Akademie, 1920). For information on the creation and history of Die
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

physischen Gestalten, and in particular the decision to publish two separate introductions
This document is copyrighted by the American Psychological Association or one of its allied publishers.

to the book, see Siegfried Jaeger, ed., Briefe Wolfgang Kohlers an Hans Geitel 1907-1920
(Germany: Passau, 1989); and Mitchell G. Ash, Gestalt Psychology in German Culture,
1890-1967: Holism and the Quest for Objectivity (Cambridge, England: Cambridge
University Press, 1995) p. 168ff.
2. Willis D. Ellis, A Sourcebook of Gestalt Psychology (New York: Harcourt Brace,
1939).
3. Christian von Ehrenfels, "Uber Gestaltqualtitaten," Vierteljahrsschriftfur wissen-
schaftliche Philosophic 14 (1890): 249-292.
4. Vitalism was a biological theory, introduced in the 18th century and still quite in
favor at Kohler's time. It claimed that life derives from a special vital force not explainable
by chemistry or physics.
5. Max Wertheimer, "Ueber Gestalttheorie" ("Gestalt Theory") (Erlangen, Ger-
many: Verlag der Philosophischen Akademie). The English translation by N. Nairn-Allison
was published 1925. In Social Research, 11 (1, February 1944).
6. Kurt Koffka, "Zur Grundlegung der Wahrnehmungspsychologie: Eine Ausein-
andersetzung mit V. Benussi," Zeitschrift fiir Psychologic 73 (1915): 11-90, portions of
which are summarized in Ellis, A Sourcebook of Gestalt Psychology (see note 2). See also
Kurt Koffka, Principles of Gestalt Psychology. (New York: Harcourt Brace, 1935). (All
notes were added by the translator.)
Received April 29, 1997
Revision received September 15, 1997
Accepted September 15, 1997