Sherrill on Gestalt Therapy/Psychology

This article originally appeared in the Fall, 1986, issue of The Gestalt Journal.  It was drawn from Sherrill's doctoral thesis and appeared first in the Memorial Festschrift for James Solomon Simkin published by the Gestalt Therapy Institute of  Los Angeles in 1984


Robert E. Sherrill

Jim Simkin sometimes told a story from his early years in New York City with Fritz Perls, of going with Fritz to hear a lecture by Kurt Goldstein. Fritz had been Goldstein's assistant in 1926 at a neurological institute, and credited Goldstein with his first exposure to Gestalt principles as applied to organismic functioning. After the lecture, Fritz went up to greet his former mentor. Jim remembered that Goldstein's return greeting was lukewarm.

Goldstein's unenthusiastic response to Perls has been repeated by the academic Gestalt psychologists who have studied the Gestalt therapy literature. Gestalt therapists frequently claim descent from Gestalt psychology by quoting some of the classical Gestalt experiments in figure/ground information and figure/ground reversal, and relating those to Perls' description of needs altering an organism's perceptions. They note that the writings of Kohler, Koffka, Wertheimer and their students include applications of Gestalt principles to personality functioning. Yet Gestalt psychologists have consistently denied any close kinship between their system of theory and research, and the writings of Gestalt therapists. Kohler's window wrote, "I know that his attitude toward Gestalt psychotherapy was negative" (L. Kohler, personal communication, January 10, 1974). Kohler's colleagues Rudolph Arnheim and Mary Henle have been vehemently negative. Arnheim (1974) wrote: "I can see Max Wertheimer fly into one of his magnificent rages, had he lived to see Perls' Ego, Hunger, and Aggression (1947) dedicated to him as though he were the father of it all" (p. 570). Henle (1978) concluded her critique of Perls' later writings by saying, "Fritz Perls has done -- `his thing'; whatever it is, it is not Gestalt psychology" (p. 32).

Gestalt therapists see close kinship between the two Gestalt systems; Gestalt psychologists deny any meaningful similarity. How has this disjunction of perception occurred? I believe it is due in part to Gestalt psychologists having recognized that Perls made frequent errors in describing Gestalt psychology. Another part has been Perls, et al.'s confusion between Gestalt psychology and the systems of Lewin and Goldstein. Finally, there is an important theoretical difference between Gestalt therapy and Gestalt psychology, that of the degree to which organismic variables influence perception. I will describe each of these parts briefly, then review how Gestalt therapy can ground itself more adequately as a descendant of the Gestalt movement.

Perls (1969a) stated explicitly that his study of Gestalt psychology was confined to some papers of Kohler, Wertheimer, and Lewin (unpaginated). His later works contained obvious errors in describing Gestalt psychology. For example, in one passage, Perls (1969a) scolded Gestalt psychologists for concepts which they never endorsed:

(our answer) comes from a direction which never claimed the status of a philosophy...Gestalt psychology....The formulation as expressed by the Gestaltists cannot possibly be correct. They say that the whole is more than the parts. In other words, something is added to the world simply by configuration. This would ruin our picture of the energy balance of the universe. Shall we then let the Gestaltists attribute to Gestalt formulation more power than our pious ancestors gave to God? (unpaginated)

Gestalt psychologists would recognize quickly the errors in this. Kohler (1930/1971) discussed energy balance in detail, and insisted that all Gestalten obeyed the physical laws of conservation-of-energy (p. 237). And the notion that a Gestalt is something which is added to a summation-of-parts was one portion of earlier theories of perception which the Gestaltists explicitly denied in their reformulation of what Gestalt meant. (What Kohler and his colleagues did say was that the whole is different from the sum of its parts).

The quotation above also reveals Perls' lack of awareness that the Gestalt psychologists considered their ideas to be not a specialized school of psychology, but rather an approach to science which contributes to diverse fields of study, including philosophical problems of ethics. It was for this reason that they referred to themselves as Gestalt theorists. They were passionately concerned with applying the rigor of scientific truth-seeking to human problems. Kohler titled one of his books The Place of Value in a World of Facts (1934/1966). He theorized that values in the form of "requiredness" exert a force upon the nervous system (pp. 270-271). Wertheimer (1935/1961) demonstrated how scientifically derived "requiredness" contributes to ethical understanding (p.29)

Gestalt theorists have reacted both to Perls' errors in describing Gestalt theory as non-philosophical and to his tone of moral relativism (Henle, 1978). Perls' description of right and wrong as being "...always a matter of boundary, of which side of the fence I am on" (1969b, p. 9) is alien to Kohler and Wertheimer's insistence upon the objective aspects of morality. Consequently, Gestalt theorists have attempted to disassociate their work from Gestalt therapy.

Perls mentioned the work of Lewin and Goldstein often in his later books, and appropriately so. Both men took ideas from Gestalt theory, expanded their meanings, and applied them to problems of motivation and personality. Because Lewin and Goldstein broadened the definitions of Gestalt concepts, neither was a part of the core Gestalt theory group. Lewin was a junior colleague of Kohler and Wertheimer at the University of Berlin in the 1920's. He had enormous respect for the Gestalt theorists, but became dissatisfied with their emphasis upon perception, and began designing experiments which would permit direct inference to dynamics of personality. Dr. Edwin Newman remembers,

Kohler had a very high standard for himself and his students in terms of precision of concepts and evidence. During the Berlin years there was almost a running battle between the students of Lewin and Kohler, and the latter disapproved strongly of some of the Lewin theses. (personal communication, December 8, 1974)

By the time Lewin came to the United States in 1933, he had named his own theoretical system "topological psychology." He and Kohler continued to express appreciation for each other's work, (Lewin, 1936, preface), although both apparently agreed that Lewin was not an orthodox Gestalt theorist. In 1966, Kohler remarked, "...we do not yet know how Lewin's important work is related to Gestalt psychology..." (1966/1969, p. 120).

Goldstein had extensive contact with Kohler, Koffka and Wertheimer in Germany; praised their work; and used Gestalt terminology in his writings. But he considered himself a holist rather than a Gestalt psychologist, according to Mary Henle (personal communication, October 30, 1974). By the 1940's both Kohler and Wertheimer had adopted a very cool tone toward him. Like Lewin, Goldstein expanded the meaning of Gestalt terms. He named his own adaptation of Gestalt principles "organismic biology."

An example of how Lewin and Goldstein expanded Gestalt definitions was their use of figure and ground. For the Gestalt theorists, Gestalt was a generic term for all naturally occurring organized entities or functional wholes. A Gestalt could be a visual percept, a temporal pattern such as a melody or a dance, an organized memory trace, or a concept such as truth or democracy. A Gestalt is segregated from surrounding perceptual or cognitive activity by virtue of its internal organization and coherence. Figure/ground formation is a special Gestalt process; in addition to its internal organization as a Gestalt, a figure possesses a definite shape, a prominent contour, and the appearance of greater "thing-character" than the ground.

Kohler and Wertheimer used figure/ground terminology only for visual entities. Koffka was more liberal, possibly because he attempted to write a comprehensive Gestalt description of mental functioning. He considered it likely that the figure/ground distinction exists for all the senses, although the quality of the ground could be described well only for audition: "...we can hear speech on the background of the patter of the rain..." (1935, pp. 220-201). He added that the grounds for potentially all the senses combine to form a general ground which is super-sensory.

I had found only one instance in which a Gestalt theorist used figure/ground terminology to describe an internal event such as cognition. Duncker (1945) explained the process of restructuration in problem-solving: "Parts and elements of the situation which, psychologically speaking, were either hardly in existence or remained in the background unthematic suddenly emerge, become the main point, the theme, the `figure'..." (p. 29). Dunker may have put "figure" in quotation marks to acknowledge its use in a non-perceptual context, or to indicate he was using the term metaphorically. Gestalt theorists generally were careful to use figure/ground terms only for Gestalten in which the special-process qualities of shape, contour, and thing-character could be identified.

Lewin (1935) mentioned that he used the term figure broadly, and implied that ground is a more active and more general process than the Gestalt theorists would have accepted: "...the whole dynamic of sensory psychological process depends upon the ground and beyond it upon the structure of the whole surrounding field" (p. 40). This was an important extension of Gestalt concepts, since the Gestalt theorists paid relatively little attention to the dynamics of the ground (Gottschaldt, cited in Koffka, 1935, p. 156). They did not develop Koffka's tentative suggestion of a generalized, super-sensory ground for all our perception.

Goldstein certainly did. He explicitly extended figure/ground to cover all perception, cognition, and emotion: "Any excitation in the nervous system has the character of a figure/ground process. Any performance invariably shows this figure/ground character...Figure and background can be discriminated as readily in speaking, thinking, feeling, etc" (1940/1963, pp. 12-13). Such a definition of figure/ground is far broader than that which Kohler and Wertheimer would have accepted. And it is obviously closest to Perls, et al.'s (1951) use of figure/ground: "...the forming of a figure of interest against a ground or context of the organism/environment field" (p.231). Perls, et al., grouped Lewin and Goldstein with the Gestalt theorists, on the basis of their common concern with naturally occurring wholes, and the use of figure/ground terminology. In one article, Perls (1948) named the three great Gestalt psychologists as "Kohler, Wertheimer, and Kurt Goldstein" (p.569). Perls, et al. (1951), included Lewin and Goldstein in their description of Gestalt psychology (Introduction). Such groupings overlooked how the Gestalt theorists have seen important differences between their rigor of definition and the broader use of terms in the Gestalt-derived systems of Lewin and Goldstein. Consequently, Gestalt theorists have considered Perls, et al.'s use of figure/ground to be unfamiliar, and fuzzy.

An important theoretical difference between Gestalt theory and Gestalt therapy is the degree of influence of organismic variables such as needs upon perception. Gestalt theorists stressed that figures emerge because of the objective properties of their internal organization. They opposed earlier theories which described the contents of perception as a rather arbitrary result of the viewer's direct consciousness. Their emphasis upon the objective autonomy of Gestalten led them to ascribe relatively little importance to organismic energy, attention, or needs in perception, except as special-case factors. Koffka (1935) wrote that if the objective stimulus properties were weak, an attitude of active searching could cause an undifferentiated part of the field to coalesce into a figure (p. 149). He described how extreme fatigue or boredom could cause a visual percept to lose good figure/ground articulation, and revert to a condition of uniformity (p. 173). He allowed some influence of interest upon figure formation: "Figures become the objects of our interests, and where our interests lie, figures are likely to emerge" (p. 197). And one Gestalt experiment showed that needs influence figure-formation, when the stimulus is so ambiguous that it can be organized in two very different ways. When presented with a reversing figure which could be seen either as a young woman or as an old hag, young men tended to see the organization of the young woman first. (This reversing figure is discussed in Perls, et. al., (1951), p. 27.)

With the exception of such special cases, Gestalt theorists maintained that organismic needs rarely influence the actual formation of percepts. Rather, they asserted that needs cause certain already-existing perceptual organizations to become endowed with qualities of attraction or repulsion. They described these qualities as "demand character" (a term which was picked up by Lewin and later by Perls). Koffka (1935) remarked that demand character might at times participate in the formation of a figure, but since there was no exact knowledge of the conditions under which this occurs, he neglected this possibility.

Gestalt theorists acknowledged readily that needs influence the organism's problem-solving. Kohler and Wertheimer argue that stresses inherent in the structure of a problem cause corresponding stresses in the cognitions of the problem-solver. Insight is a process of re-organizing one's cognitions to relieve this internal stress, and thereby to solve the problem. Kohler and Wertheimer added that intense organismic need-states often interfere with an accurate sensing of one's cognitive stresses resulting from a problem, and consequently interfere with the discovery of an elegant correct solution. They acknowledged the role of intense needs in psychopathology, but spoke of "transformations of the cognitive field." They were careful to describe both organismic needs and the organism's cognitions as Gestalten, rather than as figures-on-grounds, since both are internal events.

In contrast, Perls, et al. (1951) used figure and Gestalt almost interchangeably, and extended figure to include internal events: "The figure (Gestalt) in awareness is a clear, vivid perception, image, or insight...." (p. 231). Perls granted organismic needs a major role in the formation of figures, such as in his example (1969b) of walking through the desert and being thirsty: "...suddenly in this undifferentiated general world something emerges as a Gestalt, as a foreground, namely, let's say, a well with water...." (p. 14). In this way, as in his use of figure/ground terminology, Perls was far closer to Goldstein's concepts than to those of the Gestalt theorists.

I believe that we can ground Gestalt therapy more adequately in the Gestalt tradition by remembering one of Jim Simkin's favorite admonitions, that of starting with what is. What is, is that Gestalt therapy has expanded the meanings of important Gestalt terms, and blurred distinctions among concepts which Gestalt theorists were careful to maintain. We Gestalt therapists have used figure and Gestalt almost interchangeably. We have broadened figure/ground to include internal events such as images and ideas. And we have ignored the distinction between figure formation and selective attention, and how organismic states might have very different degrees of influence upon each. Given these changes we have made, it is unlikely that traditional Gestalt theorists will accept Gestalt therapy as a direct descendant of their system; and I suggest that we do not try to assert such a close relationship. We can acknowledge frankly that we have made a value choice different than that of the Gestalt theorists, and similar to that made by Lewin and Goldstein. As Koffka's widow wrote:

The founders of the school were afraid of any "premature" application of their ideas to other disciplines "premature" meaning before the scientific groundwork had been firmly established. They may easily have been over-conscientious, for it limited the scope of their searchings and immediate influence...(personal communication, January 29, 1974)

We have chosen to broaden concepts, and risk imprecision, in the hope of fruitfulness of understanding the broader-scale functioning of persons.

At the same time that we acknowledge our differences with traditional Gestalt theorists, we can still assert them to be perhaps second-generation spiritual ancestors. We can do this by relating Gestalt therapy concepts not just to the classic perceptual experiments on changing figure/ground organizations, but more specifically to what the Gestalt theorists did say about personality functioning. Koffka devoted two chapters of his Principles of Gestalt Psychology (1935) to the ego, adjusted behavior, attitudes and the will. Van Hornbostel's delightful article, "The Unity of the Senses" (1927/1938) described the interrelationships of perception, body and spirit as an expression of an identity of process in the neural substratum. Schulte (1924/1938) analyzed the development of a delusional relationship in paranoia. Although Wertheimer avoided the general study of motivation out of concern for subjectivistic misinterpretations, (Kohler, 1944, p. 145) he described (Luchins, 1970) how anxiety or bodily needs created a psychological rigidity which interfered with a recognition of simple solutions to problems (Vol. 2, p. 168 and Vol. 3, p. 60). Finally, Kohler (1958/1971) analyzed the intrapsychic dynamics of obsessions, and described how they could transform the cognitive field both in negative and in socially-productive ways.

We can also relate Gestalt therapy more specifically to our immediate ancestors, the Gestalt-derived systems of Lewin and Goldstein. Lewin and Perls complement each other in an interesting way: Lewin wrote at length on the internal dynamics of the subsystems of the psyche, and on the structure of the psychological environment. He mentioned the organism/environment contact boundary, but only briefly. Perls, on the other hand, analyzed the dynamics and pathologies of the contact boundary in great detail, but spoke of the self as if it were relatively undifferentiated. he described five layers of personality, but mentioned little of how the layers interact with one another. Some fruitful points of contact in Lewin's work for Gestalt therapy include:

1.Lewin's assertion that an increase in a specific tension causes a high selectivity in perception (Marrow, 1969, p. 32).

2.His hypothesis (1935) that we must often encapsulate tense psychological subsystems. This permits one tension system to be discharged at a time; the resulting orderliness is necessary for biological survival. However, highly encapsulated subsystems could be pathological (p. 55).

3.Lewin (1935) related the dynamics of internal structures to the ease with which the organism can concentrate voluntarily. (p. 234)

4.He detailed (1935) how the Self is a specialized sub-structure at the "core" of all the Gestalten which compose a human personality (p. 61).

Perls, et al., borrowed several of Goldstein's concepts, including those of organismic self-regulation and the dynamic of unfulfilled needs in the form of "pathological Gestalten" pushing into conscious awareness. Goldstein's writings contain many other concepts to which Gestalt therapy could be related. Perls apparently was not aware that his former mentor had written an article about organismically-based psychotherapy! Some of these concepts are:

1.Goldstein (1939) showed that any perceptual figure must be evaluated against a background of the whole organism (p. 253).

2.He described (1963) how anxiety could de-center the organism and cause various phenomena of neurosis, such as compulsiveness (pp. 115, 155, 156).

3.He explained (1974) that a psychotherapist must be a participant, not just an observer, and must allow a state of "communion" to develop with the client (p. 729).

4.He compared (1974) this state of communion with Martin Buber's "I-Thou" experience (p. 736).

Points of contact not only with historical theories, but also with contemporary experimental psychology, will emerge as Gestalt therapy becomes more widely known. Persons familiar with laboratory experimentation will experience directly in Gestalt therapy how their anxiety disrupts their perceptions and cognitions, and will be challenged to specify how their experiences in therapy relate to the general study of perception. An example is Ralph Hefferline, who was an experimental psychologist. His interest in Gestalt therapy led him to utilize self-paced "Informal Experiments in Self-Awareness" devised by Elliott Shapiro as experiential learning for his students. These self-experiments, and the students' discoveries from them, were incorporated into the book Gestalt Therapy (1951).

Hefferline eventually came to believe that it was misleading to use the name "Gestalt" for his therapy: he felt (1962) that the two Gestalt movements shared little other than terms such as figure/ground (p. 124). However, he remained impressed with how an attitude of directed awareness can cause intense proprioceptive sensations to emerge suddenly from a part of the body of which the person had not been conscious for years. He suggested (1955) that the boundary between overt and covert behavior is one of particular vulnerability for psychopathology (p. 376). His interest in how proprioceptions could be kept from reaching conscious awareness led him into laboratory studies of electromyography. He explored (1963) how muscle twitches below the level of awareness can be conditioned without the subject being aware of them (pp. 834-835). These studies have influenced mainstream experimental science, such as biofeedback.

As Gestalt therapy concepts are translated into a variety of experimental procedures, the concepts will be defined more exactly. This eventually will permit a new, more careful consideration of how Gestalt therapy might be a "good continuation" of Gestalt theory, in its development of how perception is affected by the state of the whole organism. Even severe critics of Perls such as Henle and Arnheim state that organismic needs are important in perception and cognition; they have objected to Gestalt therapy in part because clinical insights have not (yet) been confirmed by careful definition and study. Henle (1955) addressed the subject of needs by describing twelve possible effects of motivational processes upon cognition, and reviewing the experimental literature which supported each effect (pp. 423-432). Arnheim wrote that for historical reasons the Gestalt Theorists concentrated upon the more "objective" aspects of perception. He then added,

But [the Founders] would have been the first to agree that in order to make any sense of perception you have to embed it in the need structure of the organism and indeed treat it as an outgrowth of those basic needs. Otherwise, what sort of a Gestaltist are you? Personal communication, December 8, 1974)

The confirmation and refinement of Gestalt therapy concepts in the experimental laboratory is a task for the future. For now, we can acknowledge our spiritual ancestry in Gestalt Theory, as well as our differences with it. And we can say "thank you" to people such as Jim Simkin, who have adapted ideas from the Gestalt tradition in ways which have enriched psychotherapy.


Arnheim, R (1974). `Gestalt' misapplied [Letter to the editor]. Contemporary Psychology, 19, 570.

Duncker, K. (1945). On problem-solving. Psychological Monographs, 58, 29.

Goldstein, K. (1939). The organism. New York: American Book.

Goldstein, K. (1963). Human nature in the light of psychopathology. New York: Schocken. (Originally published [1940] Cambridge, MA: Harvard).

Goldstein, K. (1974). The organismic approach. In S. Arieti (Ed.), American handbook of psychiatry, revised edition. (Vol. 1) New York: Basic.

Hefferline, R. (1955). Communication theory. (2 articles) Quarterly Journal of Speech, 41, 223-233 and 365-376.

Hefferline, R. (1962). Learning theory and clinical psychology. In A. Bachrach (Ed.), Experimental foundations of clinical psychology. New York: Appleton-Century.

Hefferline, R., & Perera, T.B. (1963). Proprioceptive discrimination of a covert operant without its observation by the subject. Science, 139, 834-835.

Henle, M. (1955). Some effects of motivational processes on cognition. Psychological Review, 62, 423-432.

Henle, M. (1978). Gestalt psychology and Gestalt therapy. Journal of the History of the Behavioral Sciences, 14, 23-32.

Koffka, K. (1935). Principles of Gestalt psychology. New York: Harcourt, Brace & World.

Kohler, W. (1944). Max Wertheimer, 188-1943. Psychological Review, 51, 143-146.

Kohler, W. (1966). The place of value in a world of facts. New York: New American Library. (Originally presented as the William James lectures at Harvard University, 1934, and published in 1938 [New York: Liveright].)

Kohler, W. (1969). The task of Gestalt psychology. Princeton, NJ: Princeton. (Originally presented as the Langfeld lectures at Princeton University, 1966.)

Kohler, W. (1971). The obsessions of normal people. In M. Henle (Ed.), The selected papers of Wolfgang Kohler (pp. 398-412). New York: Liveright. (Originally presented as an address at Brandeis University, 1958.)

Kohler, W. (1971). The new psychology and physics. In M. Henle (Ed.), The selected papers of Wolfgang Kohler (pp. 237-251). New York: Liveright. (Reprinted from The Yale Review, 1930, 19, 560-576.)

Lewin, K. (1935). A dynamic theory of personality. New York: McGraw-Hill.

Lewin, K. (1936). Principles of topological psychology. New York: McGraw-Hill

Luchins, A. (1970). Wertheimer's seminars revisited: Problem solving and thinking. Albany, NY: SUNY/Albany.

Perls, F., Hefferline, R.F., & Goodman, P. (1951). Gestalt therapy. New York: Julian.

Perls, F. (1969a). In and out the garbage pail. Lafayette, CA: Real People.

Perls, F. (1969b). Gestalt therapy verbatim. Lafayette, CA: Real People.

Schulte, H. (1938). An approach to a Gestalt theory of paranoic phenomena. In W. Ellis (Ed.), A source book of Gestalt psychology (pp. 362-369). Reprinted from Psychol. Forschung, 1924, 5, 1-23.)

Von Hornbostel, E. (1938). The unity of the senses. In W. Ellis (Ed.), A source book of Gestalt psychology (pp. 210-216). London: Routledge & Kegan Paul. (Reprinted from Psyche, 1927, 7, 83-89.)

Wertheimer, M. (1961). Some problems in the theory of ethics. In M. Henle (Ed.), Documents of Gestalt psychology (pp. 29-41). (Reprinted from Social Research, 1935, 2.)

Return to The Gestalt Therapy Home Page