You are here

Hyman - Why Psychoanalysis Is Not a Health Care Profession

Why Psychoanalysis Is Not a Health Care Profession

by Marvin Hyman, Ph.D.


    The relationship between psychoanalysis and medicine has been a continuing issue in the history of analysis. From the very outset, psychoanalysis was considered as a treatment for a neurological disorder, hysteria, notwithstanding that the treatment was "Psychical" in nature rather than physical. Simultaneously, even though many of the earliest psychoanalysts were not medically trained, it was not considered amiss that they under-took the treatment of conditions that were supposedly in the medical domain. Of course, no one took the position that the successful modification of such conditions by Psychical means implied that they were phenomena substantially different from those which were appropriately defined as medical. Legal and political circumstances forced a public discussion of this relationship between psychoanalysis and medicine when allegations of quackery were made against Theodore Reik for his practicing medicine without medical training or license by psychoanalyzing. In response, Freud (1926) wrote "The Question of Lay Analysis." In that essay and its postscript, he argued that "psychoanalysis is a part of Psychology; not of medical psychology in the old sense, not of the psychology of morbid processes, but simply of psychology" (p. 262). Nonetheless, he also argued that in the absence of purely psychoanalytic training at that time, medical training was the best available at the moment. At best, as we consider his arguments, we can see that Freud was taking the position that, because of its special nature, "lay' persons should be allowed to practice this medical specialty. Inasmuch as psychoanalysis was considered to be a health care profession, the question of the relationship between psychoanalysis and health care, as it is being considered today, never arose in Freud's essay.

    Another discussion of the relationship between medicine and psychoanalysis appeared in Szasz's (1964) book The Myth of Mental Illness. Szasz took the position that "mental illness is a myth" and that in actual practice we "deal with personal, social, and ethical problems in living" (p. 308). Further, he argued that the so-called symptom is not a product of a disease process, but, rather, is a form of protolanguage, a system of representational signs or picture language, rebuslike in its form and message. From this view, psychoanalysis is not the treatment of disease, but the semiotical analysis of verbal behavior (associations) and bodily language (e.g., hysterical symptoms).

    In 1965, Eissler noted the extraordinary success psychoanalysis had achieved in the United States- it had become the dominant viewpoint in psychiatry and in psychiatric education. He lamented that this success had been achieved, however, at the cost of altering psychoanalysis conceptually and technically in the service of the aims of biological science, that is, reductionistic explanation and the search for cure. As an example of that kind of alteration, Eissler cited Alexander's proposal that psychoanalysis make the "corrective emotional experience" its therapeutic aim. Eissler went on to argue that the phenomena which psychoanalysis investigates cannot, for the most part, be understood through application of the physical sciences-biology, chemistry, and physiology. He contended, therefore, that psychoanalysis should better consider itself an "anthropic" science, dedicated to the study of "man," and that it should designate itself as an "academy for the science of man."

    Now, more than 30 years later, it is ironic that we are lamenting the disappearance of psychoanalysis from psychiatric practice and education and its replacement by the biological sciences. And, with equal irony, we observe psychoanalysis struggling to be considered a medical endeavor or, at least, a health care profession. It was not so long ago that Rubenstein (1965) was advocating that psychoanalytic conceptualization become more protoneurophysiological so as to interface better with biological knowledge, which, perhaps reductionistically, was conceptualized as underlying the phenomena studied by psychoanalysis. Now psychoanalysis and its theory and practice have been rejected by the medical and health care establishments. Thus, for those groups, it is irrelevant whether or not psychoanalysis chooses to become more biological.

    While it is understandable that, because of political, economic, and organizational pressures, psychoanalysis, until very recently, had not considered fully its relationship to health care, I think that it is important that we begin to do so, systematically and on a continuing basis. Only through an ongoing review of the effects on psychoanalysis of its putative medical affiliation can we gauge the extent to which psychoanalytic thinking has been affected by largely unquestioned and unexamined medical models and practices.

    I do not think that many of us in psychoanalysis go to work each day in order to stamp out mental disease. We do not work in the medical model, yet we are often perceived as doing just that. What is the difference? (In what follows, I am going to use the terms health care and medicine as synonymous and the terms physician and health care provider or professional in a like manner, since I believe the models that govern their functioning are identical and defining.)


    People present themselves for medical care because thev have a "problem.' The problem is usually, a symptom, a manifestation that gives them grounds for believing something is amiss. Pain, fatigue, functional incapacitv are examples. Frequently, the symptoms are accompanied by a sign or signs, observed by the physician, such as jaundice or pallor or a particular odor. The symptom reported and the signs observed all are defined as departures from a norm of physical function and, therefore, are deserving of further investigation. Symptoms, in this model, are conceptualized to be the result of some pathological process ongoing in the patient, (e.g., infection), on the basis of which conceptualization further examination is conducted to determine the nature of the pathology. 

    As pathology is conceived to be causal to symptomatology, so too is etiology causal to pathology. Thus, for example, infection is caused by a particular etiological agent, say, streptococcus B, which is deemed to be the fundamental basis for the manifestations that have appeared in the diagnostic study of the individual and that define that individual as a patient.

    This model of a hierarchical relationship of symptomatology pathology, and etiology has important characteristics. First, it constitutes a theoretical ideal of the diagnostic process. If one can know the three elements of the model, one can understand fully the multiple departures from the norm of the physical processes involved in any disease entity. Further, the model provides an ideal that parallels the ideal model of treatment for disease. One can treat symptoms palliatively, for example, using analgesics for pain. Or one can treat pathology in a general way," for example, using broad-spectrum antibiotics for an infection. Finally, the knowledge of the etiology enables the selection of a specific treatment choice as well as the development of preventative methods for the protection of individuals and, ultimately, the elimination of the etiological agent from nature.

    In the treatment model that interfaces the diagnostic model I have just reviewed, the patient is understood to be without responsibility for any of the three aspects of the disease. To be sure, a patient may be responsible for creating conditions of living that make a disease process possible, but once it has begun, the patient's responsibility for it and the power to alter it diminish or disappear. At that point, the health care professional assumes responsibility for prescribing the treatment and implementing its application. The patient, in this model, is limited to complying or not complying with what has been prescribed.

    These medical or health care models of diagnosis and treatment work rather well in the case of physical disease processes. A detailed description will enable comparison of these models with alternative ones. Indeed, in those comparisons, the devil is in the details.


    Consider first the diagnostic model and its utility, in approaching "problems" that are presented to the psychoanalytic practitioner. From one psychoanalytic point of view, these problems are not problems- they, are solutions to the intrapsychic, unconscious dilemmas the individual is attempting to solve. But, even if this viewpoint is not adopted, problems or symptoms are not defined by some departure from a societal, statistical, or ideal standard of behavioral functioning; they are defined for the individual by the experience of distress that accompanies them. Problems manifestly presented by analysands are what they state them to be, notwithstanding that the analyst may view them as something more, particularly as indications of some unconscious process, content, or both. Thus, anything can be a symptom" so long as it is experienced as a discontinuity between the conscious intentions of the individual and the outcome of efforts to implement those intentions. It is part of the analytic process that we demonstrate that those "undesired" effects the individual is experiencing are the product of seemingly "unknown" intentions.

    Pathology, as I have discussed it as part of the medical model, is a concept that has plagued psychoanalysis (and psychiatry) for a long time. Defined as any deviation from a healthy, normal, or efficient condition, it challenges the psychoanalytic view of the human condition in a variety of ways. Freud (1901) called all those manifestations that were unconsciously motivated "psychopathology," as in the "Psychopathology of Everyday Life." Simply put, he took the position that behaviors that are consciously motivated are "normal" and those that are unconsciously motivated are pathological." In so doing, he left us a legacy that troubles us to this very day. Today, I think that there is general agreement that what has been termed psychopathology is better conceived of as psychodynamics. Whatever psychoanalytic orientation we hold, we probably all agree that the dynamic interplay of psychic forces, however conceptualized (e.g., as instincts, defenses, relations to self and object), is an important area of psychoanalytic study in practice, theory, and research. This dynamic interplay of forces is ubiquitous and part of everyday life, including many of its more creative aspects and, therefore, not usefully considered pathological.

    The confounding of psychodynamics and psychopathology has led to sometimes odd situations. Take homosexuality, for example. Before 1978, it was considered pathological. One psychoanalyst argued that in his work with homosexuals he had discovered psychodynamic processes that his analysands had in common. He then argued that since psychodynamics equals psychopathology, homosexuality is a disease or, at least, a psychiatric disorder, that is, a manifestation of pathology. Parenthetically, even if his discovery of common dynamics in homosexuals were valid, it would not justify the equating of psychodynamics with psychopathology. It was for this reason, as well as many others, that the American Psychiatric Association disqualified homosexuality as a disorder in 1978.

    The skepticism about what is pathological that occasioned that decision in 1978 has, unfortunately, not persisted. At the present time, our society is only too willing to pathologize every behavior with which it is uncomfortable. In a court case, a defendant argued that she was sexually involved with 13-year-old boys because of her "dependent personality,' the implication being that she was therefore not guilty of statutory rape by reason of insanity. Appropriately and surprisingly, her argument was rejected in court. I say appropriately because, even if she was psychodynamically a dependent personality, the absence of responsibility that has come to be seen as the product of pathology was defined by the jury. Yet cases such as this, which argue that, because a person is motivated by psychodynamic factors that person is suffering from a pathological condition and is, therefore, without responsibility for the motivated actions, illustrate the weird logic that results in the unjustified disrepute that psychoanalysis has been experiencing.

    That which is psychodynamic is not psychopathological. Psychodynamic processes are certainly not "deviations," which are unhealthy, abnormal, or inefficient. People are not sick because they have an Oedipus complex. On the contrary, psychodynamic functions lead to manifestations that are ingenious, creative attempts to adapt to unconscious disturbances in emotional life. To no small degree, psychodynamic processes have been shown to contribute to art, humor, literature, and other such endeavors which our society has, for the most part, valued (e.g., Faber, 1970; Freud, 1905; and Sterba, 1954).

    As pathology underlies symptomatology in the medical model, so psychodynamics underlies the discontinuities experienced by those who come for analysis. While psychodynamics must be inferred from the material provided by the analysand rather than observed directly, it does not follow that psychoanalysis is more an art and less a science. Probably it is both, as art and science come to be seen more and more as interwoven approaches to knowledge. And this is perhaps even more the case when we come to a consideration of etiology as viewed psychoanalytically.

    How do people come to have the psychodynamics they are inferred to have and that are discovered in the course of a psychoanalysis? Are there etiological agents that underly the pathology and symptomatology supposedly presented by those we analyze? It used to be that we thought so, and in some cases still do. Just as Stephen Sondheim had the gang members in West Side Story proclaiming that they "were depraved because they were deprived," so too did we entertain for a time the concept of the "schizophrenogenic mother," whose characteristic child-rearing practices caused the malignant pathology and bizarre symptomatology of her off-spring. Perhaps we still want to think in that etiological fashion as is implied by such current concepts as "the unempathic selfobject" exerting a significant influence during crucial early child development.

    Current concepts of "etiology" emphasize "experiences," defined as current reports of "memories," the actuality of which is irrelevant to the analysis since the reports are viewed as thoughts that have come to the analysand's mind in the present moment in order to communicate some ongoing current dynamic. "Experiences" are thus contrasted with "events," defined as actual occurrences that, in the mind of the analyst and the analysand, are remembered exactly as they took place in the "original" situation.

    "Experiences" are elements in the analysand's memory and mind, and we seek to learn about such mental events through the experiences of analyst and analysand in the analysis. I do not think that analysis is any longer the search for the "truth" of the "psychic traumas" that analysands are presumed to have undergone in childhood, if for no other reason than that truth is unknowable. Besides, what the analysand thinks and feels and experiences in the analysis constitutes the essential truth of the clinical moment and has embedded in it all the antecedents that come to the analysis with the analysand.

    It may well be that in some future "brave new world" we will learn the genetic, chemical, physical, and neurophysiological bases of the infinite varieties of behaviors and mental experiences of which human beings are capable. It may even be that we will learn all the effects of developmental events on the future life of the individual. None of that will, however, provide us with any causal understanding of the specific, concrete, infinitely complex, myriad happenings of the analytic moment as experienced by analyst and analysand. Nor do we need any such causal under-standing. The conclusions as to causality at which the participants in the analysis arrive together suffice: for them, for the moment, for the analysis, and for however long they wish.

    In its essence, psychoanalysis is simultaneously both a "diagnostic and a "therapeutic" process. The so-called therapy consists in the participants' making together and continuously a so-called diagnosis of each of the phenomena that make their appearance in the analytic situation. The therapy consists of making diagnoses. It is for that reason that the legendary analyst who, when asked the diagnosis of his analysand, replied, "I don't know. We haven't finished the analysis yet."

    Psychoanalysis as a therapy is radically different from the health care model of treatment, and not only because it proceeds from a different diagnostic framework. As I have noted, the patient seeking health care consciously defines himself or herself as not responsible for the condition presented, without sufficient knowledge and training to understand it, and incapable of doing anything about it. In most instances, the health care professional shares that point of view. That professional makes the diagnosis and may or may not share that "truth" with the patient. The professional may also be sympathetic about the diagnosis, because the patient is perceived to be a "victim" of the disease process that has been visited upon her. Thus, patient and health care professional collude to define the patient as incompetent to understand what has been discovered and to decide what to do about it. It follows, therefore, that the treatment process proceeds by means of the professional's doing to and for the patient. Prescriptions are made, procedures are performed, and instructions are given, with each of which the patient is expected to comply.

    I believe that most psychoanalytic practitioners do not share this point of view about the status of the people with whom they are doing analysis. I am aware that some analysts believe that certain of their patients suffer from the results of permanently damaging etiological agencies, such as developmental arrests, infantile traumas, genetically endowed incapacities of ego function, or anomalies of the central nervous system. These analysts conclude, compassionately, that they have to provide reparative services to those people, however limited in effect those services must be.

    Other analysts, using almost the same model, believe that they are the knowers of "truth" about their patients, the analytic situation, and the world in general. This superior knowledge of reality is proven, for them, by their conviction of their natural endowment, by their possession of diplomas and certificates, and by their conclusion that transference consists of "distortions of the reality of the analyst" as a result of infantile trauma. Such analysts see correction of those "distortions" as one of their major analytic functions. They hold the view that the so-called corrections are not the responsibility of the analysand and that the analysand cannot use the analytic process to know and effect the needed "corrections," if he or she chooses.

    Most of us, however, view the analysand as collaborator, someone with whom the analyst works as a "colleague," each doing her or his task of the analysis. In this view, the analysand is equally responsible for the analysis, something to which we sometimes give insufficient weight as we consider the analytic situation. When we stop to think of it, analysands' responsibilities are many and substantial. They include being responsible for making the contact, for deciding to enter and continue in analysis, and for meeting the minimum requirements of "coming to the session," talking" in it, and paying for it. Other significant responsibilities of analysands are choosing the manifest verbal material and their behaviors, which the participants analyze together; utilizing the interventions of the analyst to effect change, or choosing not to do so; and preserving those important effects of the analytic effort beyond the conclusion of the analysis. The analysand, then, is not perceived as or expected to be the passive, incompetent recipient of those services which are provided by a health care system and its personnel.

    In analysis, the status of the analyst is reciprocal to that of the analysand. In contrast to the position held by most physicians and other health care professionals, the analyst is not responsible for the analysand or the analysand's life, however much the latter would like to transfer that responsibility. Trying to assume that responsibility gratuitously risks derogating the analysand by conveying the analyst's belief that the analysand cannot do for himself or herself. People who need another to be responsible for them and for their lives don't need an analyst-they need a guardian. What the analyst is responsible for is analyzing, however that may be individually defined theoretically, technically, or stylistically.

    Throughout any analysis, the analyst does more than just analyze. She does not treat the greetings that begin and end the session as thoughts that came to the analysand's mind. She offers appropriate congratulations and condolences, as well as greetings of the season, information of which only she is possessed, and needed assistance when none can be obtained anywhere else. She treats such departures as social conventions that may also be psychodynamically significant but are at the moment not analyzed. Richard Sterba (personal communication) used to compare the analytic method to distilled water in that both could be perceived ideally as being totally pure: the water being without impurities and the analysis being free of anything other than the analysand's associating and the analyst's interpreting, period. Sterba went on, however, to point out that distilled water does not carry an electrical current; it needs impurities in it if it is to do so. The analysis, analogously, needs and has appropriate departures from "just analyzing" if it is to carry an "emotional' current. Thus, even in the most conservative conception of the analytic method, there is a place for human interaction as an integral part of the method. This aspect of analytic work, however, does not support the conclusion of some critics that analysis consists essentially of chatting with the worried well about their sexual hangups. On the other hand, it also does not support the view that analysis is advice-giving by a knowledgeable authority and thus consistent with the medical treatment model.

    Psychoanalysis is currently in a state of productive theoretical, conceptual, and technical ferment. Many argue that the effects of psychoanalysis are the result of the analyst's first formulating the psychodynamics that are inferable from the associations in the analytic hour and then presenting them in verbal form to the analysand. Others take the position that the analytic relationship, in its myriad affective and experiential forms, produces the analytic results. Most analysts, in general, believe that it is both. Whatever it is that the analyst provides, interpretation and relationship the analysand is the only intermediary between that which is provided by the analyst and the results that ensue. And whatever the results may be and however they are described, they occur only with the analysand's active, albeit unconscious, use of that which he encounters in the analysis. This essential aspect of the analysand's active involvement in and responsibility for portions of the processes in the psychoanalysis might appear to the health care professional as improbable as a surgical patient's assisting the surgeon in performing the operation. It is, I believe, one of the aspects of psychoanalysis that is so enigmatic to those who evaluate it using the standards of the currently prevailing health care models.

    In much of what I have presented thus far there is an implication that the outcome of the analytic work is essentially the responsibility of the analysand. Further, the only responsibility the analyst can assume in regard to outcome is to analyze it. Most analysands state that they experience significant and early relief from the subjective distress they brought to the analysis. With that objective attained, and as they become better acquainted with the process, their conscious objectives seem to expand in the direction of achieving change in themselves and their lives. Also, they adopt the objective of continuing the satisfaction that is a part of the ongoing analytic work. The outcome of analysis is, therefore, an entirely subjective matter and quite different from that which is expected in a medical model, viz., conformity to a generally accepted state of normality or, at least, a cessation of complaints and symptoms. Indeed, I would argue that one would have to deviate significantly from a psychoanalytic ethos and frame of reference in order to conduct outcome studies of psychoanalysis in a manner that would be acceptable to a health care system and congruent with its model. I urge, therefore, that we be fully content with any belief that we have that our work in a particular analysis has or has not been productive, which belief may or may not be held by the analysand, and that we dismiss any accusations that such conclusions are subjective and, therefore, impossible of validation. If deciding to begin analysis is the prerogative only of the participants, individually and collectively, so too should be any evaluation of outcome. And, finally, it seems to me that any need to consider outcome should better be analyzed, rather than regarded as a truth to be verified.

    By way of summary and in support of my arguments that psychoanalysis is not a health care profession, I would like to contrast the health care model and the psychoanalytic one:

Health care is concerned with symptoms; psychoanalysis addresses the whole of an individual's    mental life.

Health care pertains to pathology; psychoanalysis, to psychodynamics.

Health care seeks physical, objective, and historical fundamental causes; psychoanalysis studies subjective current experience.

Health care seeks the cure of illness; psychoanalysis only examines the subjective experiences of mental life, including the idea of being ill.

Health care professionals assume responsibility for their patients; psychoanalysts assume responsibility for the analysis.

Health care seeks to reduce the human condition to the biological and chemical; psychoanalysis is content with the psychical.

Health care sees mental illness as an objective fact; psychoanalysis recognizes it as a metaphor.

In health care, the diagnosis determines the treatment; in psychoanalysis the "treatment" is "diagnosing" and vice versa.

Health care emphasizes outcome; psychoanalysis emphasizes process.

In health care, the professional is the responsible authority; in psychoanalysis, the participants are collaborators and share responsibility for the analytic process.

    I have understood my assignment in writing this chapter to espouse a particular point of view rather than develop a careful presentation of all sides of the issue. I am cognizant of other positions, but I have left those to the other contributors. In somewhat the same vein, I am mindful of the significance that attaches to being a health care profession in today's world . The organizational, political, legal, and economic benefits that accompany the designation of health care profession are many. I would note, also, that those benefits are costly since they subject the profession to the strangling effects of ever-increasing and enforced rules, regulations, laws, prohibitions, standards, evaluations, intrusions, and the like that compromise and diminish its status and integrity. But discussion of the benefits and costs of being a health care profession should be ongoing for all of us.

(This article "Why Psychoanalysis is Not Health Care", from the book:  Psychoanalytic Therapy as Health Care:  Effectiveness and Economics in the 21st Century, edited by Hariette Kaley, Morris Eagle, and David Wolitzky (1999) is being electronically reprinted by permission of The Analytic Press.   Copyright © 1999 by The Analytic Press, Inc., Hillsdale, NJ.)


Eissler, K R- (1964), Medical Orthodoxy and the Future of Psychoanalysis. New York: International Universities Press.

Faber, M. D. (Ed.). (1970) The Design Within: Psychoanalytic Approaches to Shakespeare. New York: Science House.

Freud, S. (1901), The psychopathology of everyday life. Standard Edition, 6:1-290. London: Hogarth Press, 1960.

- (1905), Jokes and their relation to the unconscious. Standard Edition, 8:1-258. London: Hogarth Press, 1960.

- (1926), The question of lay analysis. Standard Edition, 20:179-258. London: Hogarth Press, 1959.

Rubenstein, B. (1965), Psychoanalytic theory and the mind-body problem. In Psychoanalysis and Current Biological Thought, ed. N. S. Greenfield & W. C. Lewis. Madison: University of Wisconsin Press, pp. 35-56.

Sterba, E. & Sterba, R. (1954), Beethoven and his Nephew. New York: Pantheon.

Szasz, T. S. (1964), The Myth of Mental Illness. New York: Hoeber-Harper.

Abbreviated Curriculum Vita

Psychoanalytic Education
   Classes, Seminars, Tutorials and Supervision with several senior analysts, notably Siegfried Bernfeld and Richard Sterba.  Personal analysis.   Education self-directed.

Academic Appointments
    Associate Professor, Department of Psychiatry and Behavioral Neuroscience, Wayne State University School of Medicine, Detroit, Michigan (Retired 31, December 1998).

Organizational Positions
    Former President, International Federation for Psychoanalytic Education.
    Former President, Division of Psychoanalysis, American Psychological Association
    Former President, Psychologist-Psychoanalyst Practitioners
    Former President, Michigan Society for Psychoanalytic Psychology
    Former President, Michigan Psychological Association

Current Professional Activities
    Private Practice of Psychoanalysis, West Bloomfield, Michigan (Retired 31 December 1998)
    Faculty, Center for Psychoanalytic Studies, Michigan Society for Psychoanalytic Psychology

    Co-Author (with B.F. Auld):  Resolution of Inner Conflict:   An Introduction to Psychoanalytic Therapy