Telematic Review

Area Problems of Psychotherapy (editor: Paolo Migone, M.D.)


The effect of the structure of the ego
on psychoanalytic technique (1953)
Kurt R. Eissler

Introduction by Paolo Migone

This is K.R. Eissler's 1953 classic paper on "parameter" of technique, of fundamental importance because for the first time the criteria for differentiating between psychoanalysis and psychotherapy were coherently spelled out according to ego psychology. Influenced by the theoretical advances on the concept of ego, Eissler systematized with a coherent theory the modifications that more and more analysts at that time were introducing in classical analytic techinque. These modifications were caused, in part, by the more severe diagnoses of patients undergoing psychoanalytic treatment, due, in turn, to "the widening scope of indications for psychoanalysis" (L. Stone, 1954) of the '40s and '50s. Eissler theorized, heuristically, a "basic model technique": an ideal model, not easily applicable in real practice, in which the analyst is supposed to work with a patient with an intact ego, and in which the main intervention is verbal interpretation without modifications of the classical ground rules (the couch, four weekly sessions, etc.). In such cases the patient's ego is so strong that is capable to tolerate and work through the meanings of interpretations. Since it soon became apparent that many patients (especially phobics, schizophrenics, delinquents, etc.) could not tolerate this technique, it became indispensable to modify some aspects of it. Eissler coined the term "parameter of technique" to describe a modification of technique that was made necessary because of the patient's ego deficits. Such modifications may include various kinds of interventions different from interpretation - e.g., reassurance, advise, return to the vis-à-vis position, behavioral prescription such as the suggestion to be exposed to a phobic object, the decision to terminate analysis on a given date in order to mobilize possible resistances (these two last techniques were employed by Freud [1914] himself with the "Wolfman"), and so on. But according to Eissler (1953) - and this is his most important contribution - a technique can still be called "psychoanalysis" when a parameter is introduced according the four following criteria: "(1) a parameter must be introduced only when it is proved that the basic model technique does not suffice; (2) the parameter must never transgress the unavoidable minimum; (3) a parameter is to be used only when it finally leads to its self-elimination; that is to say, the final phase of the treatment must always proceed with a parameter of zero" (p. 111), and (4) "the effect of the parameter on the transference relationship must never be such that it cannot be abolished by interpretation" (p. 113). In this way he suggested a coherent differentiation between psychoanalysis and psychotherapy: in order to modify ego structure, the parameter that was introduced needs to be subsequently eliminated, otherwise we run an important risk, rightly emphasized by Eissler (1953): "Every introduction of a parameter incurs the danger that a resistance has been temporarily eliminated without having been properly analyzed" (p. 127). Hence, from a psychoanalytic viewpoint, according to Eissler by psychotherapy proper we mean a therapy based on parameters that are not analyzed nor eliminated (constituting some forms of "acting in", we might say), steadily introduced in the therapeutic environment (hence, to be precise, they are not "parameters" in the way used by Eissler, but "variations"; see Eissler, 1958). This technique could be beneficial to the patient. For example, it could reassure him, but, by not allowing him to internalize specific functions (i.e., keeping him dependent on some characteristics of the environment in order to function), nor to acquire insight on why he feels reassured by the parameter, he cannot modify his ego structure which is the goal of psychoanalytic treatment (for some clinical examples of use of parameters, see Migone, 1993, 1995a pp. 58-62 [see 1992], 1995b, 1999).

As we have said, Eissler's theorization is coherent. In the following years, when gradually the interpersonal perspective in psychoanalysis begun to be more and more diffused, many authors begun to criticize an important theoretical premise of Eissler's theory, namely the fact that the "basic model technique" must necessarily be characterized by the "extrinsic" criteria established by the classical tradition (the couch, four weekly sessions, etc.). One of these authors, surely the most authoritative, was the late Merton Gill (see his 1984 paper, and the discussion on Internet edited by Favero & Migone [1999]), who believed that any kind of setting evokes a type of transference, so that we cannot say that a given setting (e.g., the classical one) is able to evoke a "pure" or "genuine" transference, since in any case it is the transference itself that determines the meanings of the setting for a given patient, being these menaings specific to him or her. In other words, Gill's theoretical revision denies to classic psychoanalytic technique the role of "gold standard", both in the sense of the pure "gold" of psychoanalysis (compared to the "copper" of psychotherapy) and in the sense of a measure to which all other "therapeutic currencies" can be referred to. The positions taken by the late Gill (1984), that mark a clear revision of the oft-quoted positions Gill himself had taken 30 years earlier (Gill, 1954), can be seen as within the "interpersonal" or "relational" tradition of psychoanalysis, that in recent years received a growing attention within the psychoanalytic mainstream. From this point of view, I think we can say that Gill's revision does not represent a real advance in the history of psychoanalytic ideas, but simply a return (albeit theorized in a very lucid and coherent way) to the intuitions made by Sullivan, Fromm-Reichmann and the other North American "interpersonalists" or "revisionists" formulated already in the '20s and '30s, who represented the first main institutional split of American psychoanalysis. It is not possible here to further discuss the theoretical and sociological implications of these interesting developments (for an interpretation, see Migone, 1995a cap. 4, 1992). Gill's critique, however, does not concern, as we have said, the inner coherence of Eissler's conception of parameter, which remains valid and can be applied to any kind of setting, be it "orthodox" or "unorthodox".

Kurt R. Eissler, one of the most important analysts of his time, died at 90 years old in New York on Feb. 17, 1999, on Ash Wednesday. He was born in Vienna on July 2, 1908. As J. Stone tells us in the obituary in The Independent of March 2, 1999, "he received a Ph.D. in Psychology in 1934 and an M.D. in 1937 from the University of Vienna. His Ph.D. thesis under Professor Karl Bühler was on the constancy of visual configurations in the variations of objects and their representation. He underwent psychoanalytic training at the Vienna Psychoanalytic Institute in its most fertile period among colleagues such as August Aichorn, Paul Federn, and Richard Sterba, and became a member of the Vienna Psychoanalytic Society in 1938". When Austria was annexed to Germany he fled to the United States, leaving behind unfortunately his brother Erik who eventually was killed in a concentration camp. Once in the US he went to Chicago, where he qualified for the American Board of Psychiatry and Neurology and soon became known for his strong critique to Franz Alexander, then the leading figure of the Chicago school of psychoanalysis (his well know paper "The Chicago Institute of Psychoanalysis and the sixth period of the development of psychoanalytic technique", in which he criticized Alexander's concept of "corrective emotional experience", was published in 1950). Later Eissler moved to New York where he lived for the rest of his life.

In 1943 he left his psychoanalytic practice to volunteer in the US army, where he served as Captain of the Medical Corps. In 1952, with other colleagues, he founded the Sigmund Freud Archives, of which he was the director for many years until, in the early '80s, he was painfully involved with the Masson case (Malcolm, 1983, 1984; see Migone, 2002). In the Sigmund Freud Archives he collected thousands of interviews, documents, and letters by and to Freud, which he deposited at the Library of Congress in Washington, D.C. In such endeavor he was much helped by Anna Freud, of whom he was close friend since the Vienna times. Eissler helped also Ernest Jones (1953-57) in his writing of the 3 volumes of Freud's biography, and also James Strachey for the 24 volumes of the Standard Edition, the official English translation of Freud's psychological works. In 1952 Eissler founded the Anna Freud Foundation to collect tax-free American contributions for the Hampstead Child Therapy Clinic which she had just set up.

Eissler wrote about a hundred papers and a dozen of books, dealing with the most varied topics: problems of psychoanalytic technique (suffice to think to his aforementioned 1950 paper against Alexander, or to his 1953 paper on parameter that here we publish), treatment of psychotics, delinquency, art, psychoanalytic interpretation of history, etc. Gifted with a typically European culture, he wrote also important psycho-biographies of Leonardo (Leonardo da Vinci: Psychoanalytic Notes on the Enigma, of 1962), Goethe (Goethe: a Psychoanalytic Study, 1775-1786, of 1963, a two-volume set of 1538 pages), Hamlet (Discourse on Hamlet and HAMLET: A Psychoanalytic Inquiry, of 1971), and so on (see Eissler's full bibliography on the web site on K.R. Eissler by J. Stone, editor of Freudpsa E*Archive and author of the above mentioned obituary in The Independend of March 2, 1999; see the obituary by Clifford Yorke in The Guardian of March 23, 1999; see also the official web site devoted to Kurt R. Eissler,, edited by dr. Emanuel E. Garcia, admistrator of Eissler's literary estate [1525 Locust Street, Floor 19, Philadelphia, PA 19102, USA, tel. 215-9852688, 215-4712000; home: 2120 Race Street, Philadelphia, PA 19103, USA, tel. 215-5684745]).

Eissler was a protagonist of the best classical tradition, which saw him committed in several battles in defense of the founder of psyhoanalysis, but in the same time, as J. Stone wrote in the above mentioned obituary, "His 'irreverence' - his own word - for rigid orthodoxies defined his consistent stance as a classical psychoanalyst impatient with institutional inflexibilities, which he believed endangered the future of psychoanalysis". His personality made him a fascinating person: apparently shy at times reserved in public, he was able to open up with few people showing extraordinary warmth and a strong sense of humor (regarding this 1953 paper on parameter, for example, I still have a letter of him in which he says that he was ashamed for having become so famous simply for having invented a new word of the psychoanalytic jargon). Not infrequently he was quite ironical and unpredictable.

In Vienna he had been friend and student of August Aichhorn, who was very close to Freud (in the occasion of Aichhorn's 70th birthday, on July 27, 1948, Eissler dedicated to him a book he edited, Searchlights on Delinquency, of 1949). In 1965, always coherently with Freud's thought, he wrote an important book in favor of lay analysis, Medical Orthodoxy and the Future of Psychoanalysis, in which he opposed the medical position of North American psychoanalytic establishment.

Eissler was living in one of those beautiful buildings in Central Park West, at no. 300, with large windows hovering all over Central Park. In the last part of his life he remaoned alone, having lost in 1989 his wife Ruth Selke, whom he had married in 1936, another well known analyst and founder of the prestigious journal The Psychoanalytic Study of the Child. One of the last times I went to visit him, since he had reminded me that the previous time was two years earlier, amusingly I said that I then will come back in another two year period, and he replied that he would have considered himself lucky if he would be still alive. Incidentally, another important interest of Eissler's life was the psychotherapeutic approach to the dying patient and thanatology (the study of the various religious, psychological, and philosophical approaches toward death, a topic he dealt with in a book of 1955, The Psychiatrist and the Dying Patient, still considered an important reference for the researchers of the field). It was just few days before his death that he wrote to me giving the permission, that I had asked him, to publish on Internet some of his papers. The publication of this paper, than, has also a particular meaning.

This classic paper by K.R. Eissler, "The effect of the structure of the ego on psychoanalytic technique", first appeared in the Journal of the American Psychoanalytic Association, 1953, 1, 1: 104-143, and is reprinted here with permission of International Universities Press, Inc. We are grateful to Emanuel E. Garcia, M.D., admistrator of Eissler's literary estate, who paied the required permission fee for this publication in PSYCHOMEDIA. It was transalted into Italian ("Effetto della struttura dell'Io sulla tecnica psicoanalitica ") in the journal Psicoterapia e Scienze Umane, 1981, XV, 2: 50-79 (published by Franco Angeli of Milan), and later also in Celestino Genovese (editor), Setting e processo psicoanalitico, Milano: Raffaello Cortina, 1988, pp. 3-35. In this Internet edition we have maintained the paragraphs of the original edition, and the footnotes (with original numbers) are introduced within the text (this article, as well as all articles published since 1920 in the journals The International Journal of Psychoanalysis, The Journal of the American Psychoanalytic Association, The Psychoanalytic Quarterly, The Psychoanalytic Study of the Child, The International Review of Psychoanalysis, and Contemporary Psychoanalysis, is in a CD-ROM of Psychoanalytic Electronic Publishing [PEP]).


Alexander F., French T.M. et al. (1946). Psychoanalytic Therapy: Principles and Applications. New York: Ronald Press (trad. it. dei capitoli 2, 4 e 17: La esperienza emozionale correttiva. Psicoterapia e Scienze Umane, 1993, XXVII, 2: 85-101.

Eissler K.R., editor (1949). Searchlights on Delinquency: New Psychoanalytic Studies. Dedicated to Prof. August Aichhorn on the occasion of his seventieth birthday, July 27, 1948 (managing editor: K.R. Eissler; chairman of the editorial board: Paul Federn). New York: International Universities Press.

Eissler K.R. (1950). The "Chicago Institute of Psychoanalysis" and the sixth period of the development of psychoanalytic technique. Journal of General Psychology, 42: 103-157 (trad. it: Il Chicago Institute of Psychoanalysis e il sesto periodo dello sviluppo della tecnica psicoanalitica. Psicoterapia e Scienze Umane, 1984, XVIII, 3: 5-33 [I parte], e 4: 5-35 [II parte]. Edizione su Internet:

Eissler K.R. (1955). The Psychiatrist and the Dying Patient. New York: International Universities Press.

Eissler K.R. (1958). Remarks on some variations in psychoanalytical technique. Int. J. Psycho-Anal., 39: 222-229 (trad. it.: Osservazioni su alcune variazioni nella tecnica psicoanalitica. In Genovese C., a cura di, Setting e processo psicoanalitico. Milano: Cortina, 1988, pp. 186-199).

Eissler K.R. (1962). Leonardo da Vinci: Psychoanalytic Notes on the Enigma. New York: International Universities Press.

Eissler K.R. (1963). Goethe: A Psychoanalytic Study 1775-1786 (Two Volumes). Detroit: Wayne State University Press.

Eissler K.R. (1965). Medical Orthodoxy and the Future of Psychoanalysis. New York: International Universities Press (trad. it.: Ortodossia medica e futuro della psicoanalisi. Roma: Armando, 1979).

Eissler K.R. (1971). Discourse on Hamlet and HAMLET: A Psychoanalytic Inquiry. New York: International Universities Press.

Favero A.M. & Migone P., a cura di (1999). Dibattito sull'articolo di Merton M. Gill "Psicoanalisi e psicoterapia: una revisione" avvenuto nella Lista "Psicoterapia" di PSYCHOMEDIA (PM-PT) da febbraio a giugno 1999 (Interventi di Adriano Alloisio, Andrea Angelozzi, Tullio Carere, Gaetano Dell’Anna, Licia Filingeri, Wilfredo Galliano, Tullio Garau, Gaetano Giordano, Renzo Giraldi, Marco Longo, Salvatore Manai, Fabrizio Marcolongo, Paolo Migone, Luca Panseri, Piero Porcelli, Ermete Ronchi, Gian Paolo Scano, Daniele Toffoletto). PSYCHMEDIA,

Freud S. (1914 [1918]). Dalla storia di una nevrosi infantile (Caso clinico dell'Uomo dei lupi). Opere, 7, 483-593. Torino: Boringhieri, 1975.

Gill M.M. (1954). Psychoanalysis and exploratory psychotherapy. J. Am. Psychoanal. Ass., 2: 771-797.

Gill M.M. (1984). Psychoanalysis and psychotherapy: a revision. Int. Rev. Psychoanal., 11: 161-179 (trad. it.: Psicoanalisi e psicoterapia: una revisione. In Del Corno F. & Lang M., a cura di, Psicolgia Clinica. Vol. 4: Trattamenti in setting individuale. Milano: Franco Angeli, 1989, pp. 128-157). Edizione su Internet:

Jones E. (1953-57). The Life and Work of Sigmund Freud (3 volumes). New York: Basic Books, 1953 (vol. I), 1955 (vol. II), 1957 (vol. III) (trad. it.: Vita e opere di Freud. Milano: Il Saggiatore, 1962).

Migone P. (1992). Esiste ancora una differenza tra psicoanalisi e psicoterapia psicoanalitica? Il Ruolo Terapeutico, 59: 4-14.

Migone P. (1993). Riflessioni cliniche sul lavoro del "Psychotherapy Research Group" di San Francisco guidato da Weiss e Sampson. Il Ruolo Terapeutico, 62: 55-58. Edizione su Intenet:

Migone P. (1995a). Terapia psicoanalitica. Milano: Franco Angeli.

Migone P. (1995b). L'elaborazione della fine della terapia come intervento terapeutico. Il Ruolo Terapeutico, 68: 41-44.

Migone P. (1999). L'uso del lettino in psicoanalisi: un esempio clinico. Gli argonauti (edizione telematica),

Migone P. (2002). Storia dello scandalo Masson. Il Ruolo Terapeutico, 89 (I parte), 90 (II parte), 91 (III parte).

Stone L. (1954). The widening scope of indications for psychoanalysis. J. Am. Psychoanal. Ass., 2: 567-594.

The effect of the structure of the ego on psychoanalytic technique

Kurt R. Eissler

(Journal of the American Psychoanalytic Association, 1953, 1, 1: 104-143)


The subject matter of this paper is closely related to a problem which has occupied the minds of analysts for decades. It is within the scope of the question which Freud raised at the Berlin Psychoanalytic Convention (1922) when he asked: "What is the relationship between psychoanalytic technique and psychoanalytic theory?" Freud's question encompassed a vast area, only one portion of which pertains to the subject matter of this paper.

Freud's question of 1922 will interest us today particularly in conjunction with the structure of the ego. During the last two decades a certain sentiment has spread which might be formulated as follows: If our knowledge of the structure of the ego were complete, then a variety of techniques—ideally adapted to the requirements of the individual disturbance—could be perfected; thus we could assure definite mastery of the ego over those areas in which it had suffered defeat, that is to say, assure complete recovery. Like all sentiments, this one does not adequately reflect objective reality, but it is probably correct to say that greatly deepened, almost complete, insight into the structure of the ego would multiply the clinical effectiveness of psychoanalytic techniques. The pessimist's claim, however, must be conceded: Full knowledge of the structure of the ego would, no doubt, make the task of changing that structure appear in its true and gigantic proportion, inducing us as a consequence to withdraw modestly from such heroic attempts.

Before delving into the subject matter, I want to exclude two variables which have great bearing on psychoanalytic techniques. Such delimitation will facilitate a more precise formulation of basic issues and avoid the bewilderment which might arise in view of the great variety of problems involved. Psychoanalytic techniques depend chiefly on three variables: the patient's disorder and personality, the patient's present life circumstances, and the psychoanalyst's personality. In the following discussion the last two variables will be excluded. It is assumed that the living conditions of the patient and the personality of the analyst are both ideal; that is, entirely favorable to the analytic process. Thus in our assumption no disturbance of the psychoanalytic process originates either from the patient's actual life circumstances or the analyst's personality.

Failure to distinguish these variables has considerably lowered the standards of discussions of psychoanalytic technique [Footnote 1: Following Freud's (1910b) concept of wild analysis one may really talk of wild discussions on psychoanalytic technique in this context]. Clinical reality, of course, is so highly varied and provides so many unforeseen situations that it is impossible to set up a standard technique which would meet all exigencies of practice. This is also true of other specialties. All accepted rules of asepsis are thrown aside in some emergency situations; nevertheless, when operating under optimal conditions, the surgeon follows these rules faithfully, and they are still taught in approved medical schools although the teacher knows well the many situations in which there will be no occasion for their application.

While the patient's particular life circumstances may necessitate a certain technical measure, it is a grave mistake to conclude that this measure has general validity because it has proved its usefulness under special conditions. To overlook the specificity of variables to which a technical measure is correlated means to discard sound scientific standards.

For the purpose of demonstrating the errors we may fall into when we do not distinguish between the variables of the technique, I wish to cite only one example. In discussing the principle of flexibility, Alexander, French et al. (1946) quote Freud's technical advice that at certain points of the treatment phobic patients should be urged to expose themselves to the fear-arousing situation. Alexander uses this technical device as an additional argument in favor of his technique of giving his patients ample advice and encouragement. However, Freud's technical measure, as will be seen presently, does not lend itself at all to generalization if it is seen in its true proportion and context; that is to say, if it is correlated with that variable which forced its introduction.

Another general remark comes to mind. I mentioned earlier as one variable of psychoanalytic technique the psychoanalyst's personality. Freud reported some of the subjective factors which influenced the evolving of his technique. For example, in explaining his request that the patient take the supine position during analysis, Freud (1913) mentions his dislike of being stared at for several hours. And he goes on to add other reasons which make the supine position preferable.

When she discussed her deviation from classical psychoanalysis, analyzing in a face-to-face situation, Frieda Fromm-Reichmann (1950, p. 11) supported her argument by quoting Freud's idiosyncrasy. Her argument is out of place. An analyst may be an exhibitionist and may therefore prefer a face-to-face technique. Whatever technique a therapist may devise can be used in the service of his pleasure principle. The value of a technical measure must rest on objective factors. If it coincides with the therapist's pleasure, all the better, but this coincidence is not a decisive factor in judging and evaluating the given technique.

Fromm-Reichmann calls the reader's attention to another factor which deserves consideration when we speak of Freud's aversion to being gazed at for eight hours. She claims that the therapist of the time was "liable to share the embarrassment of his patient while listening to difficult communications" (1950, p. 11) and this made him prefer the patient's supine position [Footnote 2: Fromm-Reichmann's reasoning in favor of face-to-face interviews cannot be thoroughly discussed here; therefore I will limit myself to one statement. If the author means to identify Freud with the above-quoted statement, as the context suggests, she was mistaken. Freud (1924b, p. 42) reported that he had used the supine position before he had discovered the sexual etiology of the neuroses, and further that he had gained the full conviction regarding the correctness of the theory from his interviews with neurasthenics whose sexual life he investigated in a face-to-face position]. Fromm-Reichmann's reasoning, whether it is correct or not, brings into the picture a cluster of factors which I have deliberately deleted from my earlier enumeration of variables, namely, the historical situation. There have been attempts to correlate all kinds of historical factors with classical psychoanalysis: Victorianism and anti-Victorianism, feudalism, Puritanism, etc. No individual can divorce himself from the historical period in which he is living any more than he can put himself beyond time or space. Valuable as the sociology of science is, it does not decide which scientific finding is correct and which is not. The historical viewpoint can be applied to any of the so-called modern psychoanalytic innovations. Let us consider, for example, the technique of the staggering of interviews which is now so often advised.

As is well known, Freud attributed to the constancy and continuity of the technique—that is to say, to the technique of daily interviews—great importance, whereas some contemporary psychoanalysts believe the frequency of interviews should be adapted to the therapeutic needs of the patient; that is to say, he should be seen less often when it is desirable to increase his emotional participation and more often to assuage anxiety. The resulting technique accustoms the patient to see his analyst some times rarely and other times frequently. A historical evaluation of this technique will show that the living pattern of many analysts, certainly of those who are nationally prominent, is quite different from that of Freud. They are prominent figures on the national scene, being called to Washington as government advisers, serving on numerous committees, at all times of the year, lecturing at places hundreds or thousands of miles apart, participating in conventions—in short, they are kept busy in many extracurricular activities, so to speak. Can such analysts indulge in the luxury of daily interviews for ten months of the year without sinking into national oblivion? I have mentioned only the crudest historical reasons for the technique of staggered interviews; there are other, more subtle, ones. A historical factor may well be a valid aspect of research. But we must remember that although historical factors may be easily correlated with the techniques of a given period, the correctness or incorrectness of any technique is not decided by such correlation. Everything that is created by man must be deeply imbued by the historical climate at the time of its creation. In considering the creations of scientists we observe that at certain times and under certain circumstances the historical climate has led to a correct interpretation of reality; at other times it has led to an incorrect interpretation. Since it is idle to raise the historical argument in weighing the pros and cons of a scientific proposition, I have omitted the historical factor as one of the variables of psychoanalytic technique to be discussed here.

To return to the discussion of the effect of ego structure on technique: I will begin with a clinical example in which the psychoanalytic technique can be applied with the fewest complications. The basic model of the psychoanalytic technique can be discussed with relative ease in a case of hysteria. In such a case, we assume—and in this abstract context it is unimportant whether this assumption is clinically correct or not—that the hysteric patient has reached the phallic level and that his ego has all the potentialities for developing into an organization which can maintain an adequate relationship to reality. The task of therapy, at this point, is to give the patient that support which is necessary for the attainment of the genital level and to make possible the realization of those potentialities of the ego which have been held in abeyance chiefly because of traumatic experiences. Such a patient is informed of the basic rule and of his obligation to follow it. He adheres to it to the best of his ability, which is quite sufficient for the task of achieving recovery. The tool with which the analyst can accomplish this task is interpretation, and the goal of interpretation is to provide the patient with insight. Insight will remove the obstacles which have so far delayed the ego in attaining its full development. The problem here is only when and what to interpret; for in the ideal case the analyst's activity is limited to interpretation; no other tool becomes necessary.

In order to avoid misunderstandings, I want to stress that I do not discuss here the problem of what is therapeutically effective in the analysis of a neurosis. The therapeutically effective factors are, of course, of a far greater diversity than interpretation; among many others there is, for example, transference. It would, however, be a mistake to consider transference a tool of therapy, particularly in a case of hysteria. Transference in this instance is a source of energy which if properly used leads to recovery through the application of interpretation.

Another point should be clarified. There are other therapeutically effective factors which may look like tools, such as the denial of wish fulfillment, to which the patient must submit throughout the treatment or, more generally, the psychoanalytic therapeutic attitude. I believe that these factors are secondary; that is to say, they are the necessary consequences when interpretation is the only tool of the analyst. Similarly, working through is a specific technique for using interpretation.

I have left out one tool which is indispensable for the basic model technique. It is doubtful if any person was ever analyzed without being asked questions by the analyst in the course of the psychoanalytic treatment. Indeed, I think that the question as a type of communication is a basic and therefore indispensable tool of analysis, and one essentially different from interpretation. Unfortunately this tool has been taken for granted. The principal investigations at the present time pertain to the proper use of questions in interviews and not in the psychoanalytic process itself (F. Deutsch, 1939, 1949). The psychology of "question" in terms of structural psychology has not yet been written. But though it offers a most challenging task, I shall not consider it further in this paper but shall proceed to another neurosis, investigating the minimum tools necessary in the case of a phobia.

The technique required in a classical case of phobia is likely to be surprising at one point. The treatment begins and proceeds for a long time like that of a hysteria; that is to say, the analyst uses interpretation as the exclusive tool of therapy. However, in the treatment of some cases a point is reached when it becomes evident that interpretation does not suffice as a therapeutic tool and that pathogenic material is warded off despite the analysis of all those resistances which become clinically visible. In other words, despite maximum interpretation, the pathogenic area cannot be tapped. Even if all resistances are interpreted and every reconstruction obtainable from the material is conveyed to the patient, and even if the patient ideally adheres to the basic rule, the area constituting the core of the psychopathology will not become accessible to the analyst. At that moment a new technical tool becomes necessary. As is well known, this new tool is advice or command.

The analyst must impose on the patient a command: to expose himself to the dreaded situation despite his fear of it and regardless of any anxiety which might develop during that exposure. In extreme cases it may become necessary to threaten the discontinuance of treatment unless the patient takes on the burden of voluntarily suffering anxiety. Advising the patient to perform a certain action or even forcing him indirectly to do it is beyond the scope of interpretation and introduces a therapeutic tool which is of an entirely different type. In order to facilitate communication I introduce here the term parameter of a technique. I define the parameter of a technique as the deviation, both quantitative and qualitative, from the basic model technique, that is to say, from a technique which requires interpretation as the exclusive tool. In the basic model technique the parameter is, of course, zero throughout the whole treatment. We therefore would say that the parameter of the technique necessary for the treatment of a phobia is zero in the initial phases as well as in the concluding phases; but to the extent that interpretation is replaced by advice or command in the middle phase, there is a parameter which may, as in the instance cited here, be considerable, though temporary.

The justification of introducing a parameter into the treatment of phobia is based exclusively on clinical observation. Early experience demonstrated that the basic model technique had led to a stalemate. It became clear to Freud that if phobias were to be treated at all by psychoanalysis, he had to deviate from the basic technical position; namely, not to impose advice or command on a patient after treatment has started. The parameter which he introduced was the minimum, without which no progress could be made. The great advantage of this parameter was that it needed to be used for only a short time, that once it had proved its usefulness it could be dispensed with, and the treatment could proceed with the basic model technique. The parameter introduced into the psychoanalysis of phobia may serve as a model from which the ideal conditions which a parameter ought to fulfill can be deduced. We formulate tentatively the following general criteria of a parameter if it is to fulfill the conditions which are fundamental to psychoanalysis: (1) A parameter must be introduced only when it is proved that the basic model technique does not suffice; (2) the parameter must never transgress the unavoidable minimum; (3) a parameter is to be used only when it finally leads to its self-elimination; that is to say, the final phase of the treatment must always proceed with a parameter of zero. These three conditions are ideally fulfilled by the parameter which has become part and parcel of the analytic treatment of phobic patients.

If we now turn to the next group of neuroses, the compulsive-obsessive ones, we encounter a still different situation. Here we can take the history of the "Wolfman" as paradigmatic (Freud, 1914) [Footnote 3: This patient has been diagnosed in various ways. Before his analysis some authorities claimed manic-depressive insanity (Freud, 1914, p. 488). Freud's diagnosis was "a condition following upon an obsessional neurosis which has come to an end spontaneously, but has left a defect behind it after recovery" (Freud, 1914, p. 474), but another passage suggests that Freud may have considered the patient an obsessional neurotic: "It [the patient's intestinal trouble] represented the small trait of hysteria which is regularly found at the root of an obsessional neurosis" (Freud, 1914, p. 552). Freud (1937, p. 318) later referred to the paranoic character of some of the patient's symptoms. Yet the problems of technique which Freud discussed in the original paper pertained to those which are generally encountered in analyses of compulsive neuroses. I do not agree with Binswanger (1945) who considers the early history of the "Wolfman" as typical of childhood schizophrenia]. As far as can be seen, for the greater part of the treatment Freud used the basic model technique. Toward the end—"not until trustworthy signs had led me to judge that the right moment had come" (Freud, 1914, p. 478)—Freud introduced two parameters. One of them is well known: He appointed a fixed time for the conclusion of the treatment. The second, rarely mentioned, impresses me as even more consequential: "I promised the patient a complete recovery of his intestinal activity" (Freud, 1914, p. 552). The patient must have experienced this as a definite surrender of the analytic reserve and as the analyst's admission and promise of omnipotence; hence the resurgence of the disease when the analyst became sick and proved himself to be not omnipotent (Brunswick, 1928, p. 442).

These two parameters are of a different order than that encountered in the treatment of phobias. They fulfill the first demand we put upon a parameter; they were introduced when it was proved that the basic model technique would not bring about the patient's recovery. It is questionable whether they fulfill the second demand, of presenting the indispensable minimum deviation. Undoubtedly they do not fulfill the third requirement; they are not self-eliminating for two reasons: (1) Since the patient is to be dismissed at an appointed time, there is no time left for a concluding phase in which the basic model technique will be the exclusive one. (2) The other parameter, the promise of omnipotence, extends vastly over the termination of the treatment and seems to have been, in the case of the "Wolfman," a precondition necessary for the patient's maintenance of mental health during the years following his analysis. This deviation is of interest for other reasons: It is possible that the introduction per se of some parameters has a lasting effect on the patient's transference, an effect which cannot be undone by interpretation. Deviations from the basic model technique are occasionally lightheartedly suggested by some analysts under the assumption that the effect of any therapeutic measure can be "analyzed" later. As a general statement this is definitely wrong. Unfortunately, the boundaries have not yet been ascertained beyond which therapeutic measures create an irreparable damage to the transference relationship. One must expect individual variations from one patient to another in this respect [Footnote 4: Of the many examples which could be cited I arbitrarily select one. There are patients in whom the slightest deviation of the rule of conducting the treatment in the situation of frustration can have an extremely detrimental effect and the fulfillment of a wish, trivial as the request for a cigarette, may endanger the further course of the treatment by establishing a fixed fantasy inaccessible to further analysis. Other patients, and I believe they are the majority, are less rigid. Whatever may, in them, be evoked in terms of transference formation by trivial wish fulfillments can easily be analyzed and does not become a stumbling block to further treatment].

Freud made a definitive statement with regard to this problem when he discussed the treatment of negative therapeutic reactions. After describing the parameter of the technique which he would have to introduce if he were to effect clinical recovery in patients showing negative therapeutic reactions, he went on to state clearly and emphatically that this particular parameter is irreconcilable with psychoanalytic technique because it would convert transference into a relationship per se inaccessible to psychoanalytic interpretations [Footnote 5: See Freud (1922, footnote on p. 72). The content of this footnote is of formidable importance. Perusal of most books on psychoanalytic technique of recent date will show that Freud's spirit of intellectual honesty has largely been lost. Technical innovations are introduced in large number and are supported by the simple-minded justification that the innovator has noticed subsequent disappearance of symptoms. The question of "at what cost to and limitation of the ego" is no longer asked; instead pride in the alleged superiority of the contemporary analysts' knowledge makes many authors believe that Freud's safeguards against the effect of the therapist's personality—in situations where a structural change, induced by the analytical process, ought to take place—have become superfluous]. Thus a fourth proposition must be introduced in order to delineate the conditions which a parameter must fulfill if the technique is to remain within the scope of psychoanalysis: The effect of the parameter on the transference relationship must never be such that it cannot be abolished by interpretation.

Returning to Freud's technique in the treatment of the "Wolfman," I want to re-emphasize what is generally known: that neither of the technical innovations in this case has become an integral part of analysis. We do not have a technique for this type of compulsive neurosis which is comparable in either adequacy or precision to the parameter of technique used in the treatment of phobias.

If we now approach the other two groups of disorders, the schizophrenias and delinquencies, the situation becomes infinitely more complicated. The technique of free association cannot be applied in either group. In the schizophrenias the patient would be incapable of co-operating; moreover, the technique might precipitate regressions. In the delinquencies the basic rule is inapplicable because of the patient's intentional and adamant refusal to follow it. In these two groups, not only is the basic rule inapplicable, but simultaneously the main tool of interpretation is thrown out of gear, and insight cannot be conveyed to these patients by verbal interpretation—at least not in the initial phase of treatment. Therefore the parameters of the necessary techniques cannot be used to alter the basic model technique at certain spots nor can they be introduced as new devices in certain phases, as in the neuroses just mentioned: In the schizophrenias and the delinquencies the whole technique must be changed in all essential aspects.

Nevertheless the four criteria just formulated, which a parameter must fulfill if a technique is to be accepted as psychoanalytic, are valid also for these two groups [Footnote 6: The fourth condition, that the parameter must not give the transference a lasting direction, will be difficult to fulfill during the acute phases of the disease. If it has happened that a parameter has influenced the transference in a way which cannot be undone by interpretation, a change of analyst may become necessary].

It is impossible to demonstrate here the consequences which must follow when the basic model technique is adapted to the necessities of such grave disorders as the schizophrenias always are and the delinquencies in almost all instances prove to be, but I do want to stress that, despite the claim to the contrary by a few analysts, I am convinced that it has not yet been proved that schizophrenic patients ever reach a state in which they can be treated in accordance with the basic model technique. This is, to a certain extent, coincidental with a doubt that schizophrenic patients can be "cured" by psychoanalysis in that sense in which we commonly say neuroses can be cured. This statement is not to be construed as a denial of the effectiveness of psychoanalysis in the treatment of schizophrenic patients.

To return to the neuroses: We have taken the minimum requirements of a case of hysteria as our basic model and have compared them with the minimum requirements of other disorders. For historical reasons hysteria can be taken as the base line of the psychoanalytic therapy, since Freud demonstrated the basic technique and the basic concepts of psychoanalysis in conjunction with his clinical experiences with hysterias. However, there is also an intrinsic reason why psychoanalysis was evolved in the course of the treatment of hysterias. Tentatively I would say that the discovery of psychoanalysis would have been greatly impeded, delayed, or even made impossible if in the second half of the nineteenth century the prevailing neurosis had not been hysteria. Notwithstanding some inaccuracy, it can be said that the earliest psychoanalytic model of hysteria pertains to an ego which has suffered that minimum of injury without which no neurosis would develop at all. It is of interest to peruse the earliest publication by Breuer & Freud (1892-95) from this point of view.

In the paper of 1892, two different functions were assigned to the ego in the course of the development of hysteria: (a) Most hysterical symptoms were recognized as being consequences of traumata. Any experience which could elicit intensive painful affects could become a trauma, depending upon the sensitivity of the ego. This sensitivity was the only factor through which the ego contributed to the development of the disease, but nothing further was said about it. The psychic trauma had penetrated the patient like a foreign body and accomplished—well protected in its hideout—the whole variety of hysterical symptomatology. This early theory comes close to picturing the disease process as an event in which a part of reality has intruded into the psychic organism and pushed aside for a while the normal personality. Because of this interpretation of clinical observations, the structure of the ego did not need consideration. (b) That part of reality which remains isolated in the patient obtains its privileged function from the lack of affective abreaction, which would have been necessary in order to assimilate it. Two groups of factors were held responsible for the want of abreaction. First, the patient did not abreact because the nature of the trauma made it impossible or because the patient did not want to take notice of the trauma; that is to say, he did not want to abreact. The second group of factors concerned the state of the ego at the time when the trauma occurred. The ego was either paralyzed by an inordinately strong affect or by a hypnoid state and therefore was incapable of contributing that amount of work which would have been necessary for rendering innocuous the poisonous effect of reality—if this kind of imagery is permitted [Footnote 7: One receives the impression that these early theories might have been influenced by contemporary concepts of internal medicine regarding the origin of infectious disorders].

All these explanations had one point in common: They conjured away the bearing of the ego upon the disease process; the ego does not want to, or cannot, function and thus an area is established in which the ego-alien puts forth roots and flourishes [Footnote 8: To be sure, one of the authors must have seen beyond these conceptions, since at one point he speaks of "a hysterical individuality."]. This disregard of the ego shows up also in the therapy, which was based on a maximum paralysis of the ego, induced by means of hypnosis.

We are not, in this paper, concerned with investigating the extent to which these conceptions correctly reflected clinical reality, but only with noting that in hysteria it was apparently possible to study the disease process quite apart from the rest of the personality. Consequently a technique was evolved which permitted concentration on the clinically most conspicuous part of the disease process and which succeeded in eliminating it, at least temporarily [Footnote 9: Cf. for the foregoing A. Freud (1936, pp. 11 f.)].

A late statement of Freud's may now be brought into genetic connection with early theories. In Analysis Terminable and Interminable, Freud (1937, p. 321) reported that complete recovery could be achieved with relative ease in patients whose pathology was caused mainly by traumata. Despite their symptomatology, the ego had not been noticeably modified in such patients. In view of this statement the following conclusion can be drawn: The basic model technique, without emendations, can be applied to those patients whose neurotic symptomatology is borne by an ego not modified to any noteworthy degree. In other words, if the ego has preserved its integrity, it will make maximum use of the support it receives from the analyst in the form of interpretation. The exclusive technical problem in such instances is simply to find that interpretation which will provide the ego, in the respective phases of the treatment, with maximum support [Footnote 10: I will pursue this problem further in connection with Freud's concept of the hypothetically normal ego. The question of whether such a technique, based principally on interpretations, leads to intellectualization and lack of emotional participation on the patient's side will not be discussed here. Freud's papers on metapsychology and on the technique of psychoanalysis refute this argument. See also Alexander's (1925) incisive criticism of Ferenczi's and Rank's (1923) book The Development of Psychoanalysis].

For the type of phobia which I presented initially, this diagram must be slightly changed. Despite maximum assistance by means of interpretations, the ego cannot recover from the damage caused by the past. I believe that this fact has not made us marvel enough. It is still a riddle why a human being should refuse to make maximal use of the insight which is conveyed to him. It must be remembered that the insight offered him comprehends not only the history of his sickness but also all those resistances checking his recovery which manifest themselves during this treatment phase. Nonetheless, though recovery does not follow the offer of maximum insight, the process of recovery can be initiated after the patient has been forced to expose himself to the very danger he is so much afraid of. The patient behaves like someone who has in his grasp all the riches of the world but who refuses to take them and must be forced by threat to do so.

Of course we know some of the reasons which make it necessary in such cases to deviate from the model technique and to demand of the patient that he expose himself to the feared situation. The prospect of anxiety is such a deterrent that it cannot be overcome unless the patient is threatened with the even greater pain of losing a beloved object. But this does not explain why such an ego can give up a resistance and turn toward pathogenic, repressed material only when it is re-exposed to the pain of a dreaded anxiety. It is necessary to conclude that this ego had lost its capacity of adjustment to a larger extent than the ego of a hysteric patient. The ego organization in phobia must be significantly different from that of hysteria.

At this point it is wise to remember that the state of affairs alluded to above is not encountered in all patients suffering from phobias. Some of them recover without having recovery imposed upon them. Therefore it may be appropriate to say that it is not so much the particular combination of symptoms and defenses—that is to say, the structure of the symptom—which necessitates the specific technique but the ego organization in which the particular symptom is embedded. We must also remember that the pattern of the basic model technique does not always suffice in the treatment of hysterias; sometimes a technique resembling that for phobias becomes necessary. A hysterical patient who consistently consults internists for the treatment of conversion symptoms or who uses physical means of therapy may have to be told that he must either abstain from such escapes or face a discontinuance of his psychoanalysis. In such instances we must assume that the ego has become modified to a larger extent than was to be expected from the classical description of the dynamics of hysteria. On the other hand, as Freud (1909) has shown in the history of the "Ratman," it is possible to have a compulsive-obsessive neurosis which will subside on the mere application of the basic model technique, and a comparison of the "Ratman's" history with that of the "Wolfman" will show that fairly similar symptoms may be combined with two entirely different ego organizations—one barely, the other severely, modified.

It may be worth while to demonstrate how little or how much a mechanism or a symptom as such may count, depending on the all-inclusive ego organization in which it occurs. In his essay on Leonardo da Vinci Freud (1910a) investigated the circumstances which may have been responsible for the relative lack of artistic productivity which became increasingly noticeable in Leonardo's life. Science and scientific research gradually gained ground over artistic achievements. Freud thought that Leonardo's hesitation was caused by a lack of power to isolate the work of art by pulling it out of a broader context. Leonardo's want of capacity to isolate was correlated with a craving to express all of the associations linked with the artistic intention.

In contrast to this example of the lack of capacity to isolate, I want to cite a passage from a letter (21 November 1782) which Goethe wrote to a friend during a time when he was overburdened with administrative work as a Privy Councilor at the Court of Weimar:

"I have totally separated (externally of course) my political and social life from my moral and poetic one and in this way I feel at my very best. … I leave separated the Privy Councilor and my other self without which the Privy Councilor can exist very well. Only in my innermost plans and purposes and endeavors do I remain mysteriously self-loyal and thus tie my social, political, moral, and poetic life again together into a hidden knot."

Here isolation—and quantitatively a rather extensive isolation, cutting across Goethe's existence—functions as a truly life-saving device. I cannot go into the details of this period in the life of Goethe; suffice it to say that it was an extremely critical one, and that without some very felicitous circumstances he might have suffered injuries which could have endangered his future as an artist. Isolation was one of the mechanisms which enabled him to survive this period in a way which was of greatest benefit to him. I want to point out that the isolation of which Goethe speaks here is dangerous, and one which can be found in cases suffering from severe psychopathology; nevertheless, the mechanism which Goethe described cannot be classified clinically as part of a disease. Fortunately, Goethe made another remark which enlightens us as to the reason why isolation did not lead to psychopathology. He mentioned briefly his loyalty to himself and the hidden knot by dint of which the isolated activities were tied again into a unit; that is to say, the powerful isolation was counterpoised by an unusually strong capacity for synthesis. This mysterious knot of which Goethe speaks is really the subject matter of my exposition.

My third example is a patient whose whole personality organization was interwoven with the effects of isolation, the mechanism which dominated her life. For her time had separated into isolated moments and her childhood recollections were remembered as disconnected flashes of an ego-alien past. Likewise her various contemporary activities were isolated from each other, and probably body space, too, had fallen apart into disconnected space units as evidenced by her difficulty in distinguishing right from left. Time and space had become an aggregate. Isolation had achieved its maximum effect. As can be easily foreseen, a most severe psychopathology must flourish on such fertile ground. Interestingly enough, the patient had no feeling of suffering about this part of her psychopathology.

These three clinical examples show us three totally different effects of the mechanism of isolation [Footnote 11: Of course, such variety of effects could be shown for any defense mechanism; it is not true of isolation only]: first, a deficit in the capacity for isolation, leading to a deficit in artistic creativity; second, a profusion preserving the continuity of manifold functions; and third, an excessive growth decomposing the ego into innumerable fragments.

In view of the relative independence of ego structure and mechanism the following conclusion can be drawn: The behavior of the ego in the situation of the basic model technique is specific. Here is the crucial point where it can be determined whether or not the ego has suffered a modification. The symptoms or behavior deviations do not necessarily betray the true structure of the ego organization. This was brought conspicuously to my attention during the analysis of a patient who had spent one and a half years in voluntary commitment. At times this patient was believed to be schizophrenic because of the bewildering variety of bizarre behavior patterns. She made an astounding improvement under a technique which, with rather rare exceptions, followed the rules of classical analysis. Very much to my surprise, the many bizarre features melted away under the impact of a purely interpretative technique and a relatively unharmed ego strongly interested in and strongly attached to the world appeared from behind the maze of symptoms.

The rule that symptoms can only remotely be correlated with ego organization is true also for the allegedly symptom-free ego. I once had an opportunity to analyze a person who gave the impression, in the two initial interviews, of being relatively symptom-free and well adjusted, in accordance with his own claim of wanting treatment only for professional reasons. After several months the treatment was discontinued at the patient's request. It was my impression that the patient was not analyzable. His acting out, under the guise of conforming with the necessities of reality, firmly rooted mechanisms which compensated for an excessive castration fear and his tenuous object relationships, not unlike those often found in schizophrenics, both made me decide for a time that I would never again try the analysis of a "normal" person.

The problem which deserves keenest attention concerns the concept of ego modifications [Footnote 12: Dr. Hartmann suggests in a personal communication "ego deformation" instead of "ego modification." The latter term follows Joan Riviere's translation of Freud's (1937) Analysis Terminable and Interminable]. It must first be differentiated from an ego change. The ego, as are all parts of the personality, is constantly changing. Through new perceptions and by the acquisition of new knowledge and the formation of new memories it enriches itself; by the interplay of defense mechanisms it tries to discard part of these new acquisition: The ever-changing constellations of reality and the unceasing rhythm of biological processes make it face an infinite variation of tasks. It is correct to say that while two cross-sections of the ego are never identical, nevertheless it is always the same. It shares this property with most organisms which can maintain their identity and constancy by means of constant and rapidly occurring changes. All of these changes—primarily changes of content—do not add up to modifications of the ego. However, some of these changes may in certain frames of reference be looked upon as modifications. In the state of sleep, for example, a profound—perhaps the greatest possible—reorganization of the ego occurs. If the state of sleep is considered as a preparation for the return to a temporarily relinquished state characterized by a series of constant indices, then it will be called an ego change. If, however, we are investigating thought processes—which appear greatly modified during sleep as compared with the thought processes of the state of wakefulness—or the laws of dream work or the shifts of cathexis which take place during sleep, then the sleeping ego will be classified as a highly modified one. Disregarding the rather special question of whether the biologically enforced fluctuations of ego states ought to be called changes or modifications, it can be said that if an ego is not normal it has been modified. But what is a normal ego? Freud answered this question in the course of devising a conceptual scale which permits the ranging of all possible ego modifications from a zero point (the hypothetically normal ego) to an absolute maximum (the psychotic ego). According to Freud's definition, a normal ego is one which "would guarantee unswerving loyalty to the analytic compact" (Freud, 1937, p. 342), and since such an ego is a theoretical construction, he called it a hypothetically normal ego (fiktives Normal-Ich).

In my estimation, such an ego would be one which would react properly to the basic model technique as outlined above. It would be an ego—and this I think is a crucial point—not characterized by specific defenses, attitudes, functions or by any other structural property, but characterized exclusively by a certain mode of behavior in the comprehensive situation of the analytic treatment situation [Footnote 13: This definition presupposes optimal conditions in regard to the analyst's conduct of treatment as well as to the external conditions under which the treatment takes place].

According to Freud's definition, the hypothetically normal ego is an ego which uncompromisingly assists in the psychoanalytic therapy. It surrenders, so to speak, to the voice of reason and unflinchingly makes maximum use of the help proffered during the treatment. This description of the hypothetically normal ego (though never encountered in clinical reality) introduces a new concept into psychoanalysis. The whole troublesome question of normal behavior is thus taken out of the contexts in which it has been discussed up to now. It is no longer a question of whether or not a person has adjusted to reality, whether or not he has integrated current value systems or has achieved mastery over his biological needs. The whole question of symptomatology has been brushed aside with one stroke, and all current static definitions of normality superseded by a new dynamic definition. Freud's separation of the concepts of "normality" and "health" and their redefinition were great steps forward and ought to facilitate communication. Here Freud laid the foundation of a metapsychology of the psychoanalytic technique in structural terms [Footnote 14: For a discussion of the concept of mental health from the point of view of ego psychology see Hartmann (1939)]. Thus a normal ego is one which, notwithstanding its symptoms, reacts to rational therapy with a dissolution of its symptoms.

It is necessary to follow up the implications of this concept. It implies that the normal ego also may suffer disease. The ego of the child, because of its weakness, cannot help building up defenses and in most instances cannot escape the formation of symptoms. Indeed, Freud's definition implies that under certain circumstances neurosis is a "normal" phenomenon. Once the seed of a psychopathological disorder has been planted in the child's personality, the later, adult ego has no other choice under certain stresses but to fall back on the earlier adjustive process (Hartmann, 1939). The discovery that, in some of its most important aspects, the ego also is unconscious adds to the plausibility of this description. However, an ego which has thus been coerced into falling back on inappropriate solutions may have preserved its "normality" if it is still endowed with the capacity for capitalizing on proper help. If we may assume that psychoanalytic treatment is the most comprehensive psychological therapy because its goal is to provide the ego with all the knowledge and all the support it needs to regain its full competency, then psychoanalysis becomes the only procedure by which "normality" can be gauged.

In more general terms, the concept of the hypothetically normal ego presupposes that a childhood neurosis has developed as a result of the ego's infantile incapability of mastering the tasks put upon it by external as well as by internal reality. Yet, despite those neurotic solutions which were imposed upon the childhood ego, the development and the maturation of the ego organization were not essentially delayed or injured. Owing to the inheritance of childhood, the adult ego has not acquired its full freedom, but when it is brought into a situation in which it can obtain the assistance needed, it fights against this inheritance, and the ego's potentialities, which have been unharmed by past traumata, achieve full realization; in other words, one of the significant features of an ego essentially unharmed by traumata or constitutional factors or archaic fixations of libido is its responsiveness to rational, verbal communications which do not contain more than interpretations [Footnote 15: See A. Freud (1945) for a brilliant clinical application of this theoretical problem].

I think this conception of the normal ego is in substantial agreement with a profound thought which Goethe expressed (probably in conjunction with an experience of impotence): "The disease only avers the healthy" (Die Krankheit erst bewähret den Gesunden). Illness thus becomes the unavoidable accident of life; that is to say, it is a manifestation of life itself, and the ego's reaction to the sickness is the exclusive frame of reference of health.

At the other end of the scale is the ego of the psychotic, with whom the analytic compact is impossible (Freud, 1937, p. 337). There is scarcely anything to say about this end of the scale aside from a historical remark: When Freud was sketching the maximal ego modification, he probably had in mind the acute hallucinatory confusion which so often had served him as the prototype of psychosis. [Footnote 16: See Freud (1894, 1917). See also Freud (1923) where he calls Meynert's amentia, the acute hallucinatory confusion, the most extreme and striking form of psychosis]. Indeed, during the acute phase of a psychosis, psychoanalysis in its usual form is of no therapeutic avail. At that stage the ego is at least temporarily "modified" in such a way as to make direct psychoanalytic interference impossible. [Footnote 17: This remark does not militate different kind]. In the case of the acute hallucinatory confusion, the ego derives all its wish fulfillments out of itself even in its state of wakefulness. Tension is removed by the hallucination of instinctual gratification. The ego falsifies reality in accordance with its own wishes and can thus dispense with reality. The analyst has lost every avenue of approach because the ego has become inaccessible. It is exclusively engaged in its subservience to the id. Since the normal ego is fictitious, it is evident that in clinical reality elements of the other extreme are always intermixed [Footnote 18: I think that the clinical varieties of ego modification cannot be completely dissolved into varying mixtures of these two extremes but require a third component whose extreme is the criminal. The analyst cannot establish with him any kind of alliance for which the hypothetically normal ego is ideally adapted].

As Freud said,

"…every normal person is only approximately normal; his ego resembles that of the psychotic in one point or another, in a greater or lesser degree, and its distance from one end of the scale and proximity to the other may provisionally serve as a measure of what we have indefinitely spoken of as 'modification of the ego'" (Freud 1937, p. 337).

This view of normality as cursorily outlined by Freud seems to coincide remarkably well with views advanced by modern biologists and physiologists. In describing the variety of meanings which the concept of normality has in biology, Ivy mentions the non-arbitrary statistical view which maintains

"that a sharp distinction between 'normal' and 'abnormal' does not exist for a group or even an individual. … It recognizes that degrees of normality and abnormality exist. … It permits an absolute diagnosis of abnormality only when death occurs." (Ivy, 1944).

Freud set up a series from the acute hallucinatory psychosis to the hypothetically normal ego and thereby established a point at one end of the scale of "absolute diagnosis of abnormality" from which various degrees of normality and abnormality lead to the hypothetically normal ego at the other end of the scale.

In an attempt to clarify the meanings which the concept of normality should have in physiology, Ivy makes a significant remark about physiological processes in disease. Defensive processes, he says, such as fever or leucocytosis are normal although their effects may be abnormal. The defensive processes "are the usual physiological responses to insult. The process concerned is a statistically and physiologically normal response. The response, however, may produce abnormal effects in certain instances and hence is physiologically abnormal." Here, the problem of the "normal disease" is solved by differentiating process from response, the latter leading sometimes to abnormal effects. For example, the abnormal effect of a rise in body temperature is seen in its eventual disturbance of "other functions and margins of safety," when the fever reaches a certain height.

Likewise, according to Freud's conceptual frame of reference, psychogenic symptoms must be viewed as the logical and unavoidable consequences of the impact of external and internal reality on the childhood ego, still weak because still immature; thus symptoms may be the signs of the ego's basic health. The "physiologically normal response" would, then, become abnormal as soon as it led to an ego modification. Freud's concepts:—(1) the hypothetically normal ego as defined by the response in the situation of the basic model technique; (2) a scale leading by degrees to a state of absolute unresponsiveness to the analytic compact; and (3) the intervening variety of ego modifications to which a variety of techniques must be correlated—provide, in my estimation, a system which is ideally flexible and superbly adaptable to actual clinical work. Their heuristic value impresses me as enormous and—finally, but of greatest importance—these concepts ought to bring some rationale and order into psychoanalytical discussions of technique and thus end the contemporary arguments, most of which are based exclusively on utilitarian viewpoints. Points of expediency will always deflect the strict course which practice should take according to theory, but psychoanalysis will lose its standing as a science if problems of technique are discussed exclusively from the viewpoint of expediency.

I am aware that I follow a thought of Freud's, possibly too rigorously, by insisting that the baseline of psychoanalytic technique is one which uses a single technical tool, to wit, interpretation. In support of my contention, repeated clinical experience shows that there is a group of patients whose treatment does need scarcely more than interpretation to usher in the process of recovery and to lead the ego to the therapeutic goal. Clinical experience shows also that this group has an important structural factor in common—a relatively unmodified ego. Furthermore, it can be demonstrated that the introduction of an additional tool, one which will play a prominent part in the analytic technique, is necessitated by a structural defect in the ego. Therefore, we are warranted in classifying personality structures in accordance with the techniques required to deal adequately with their defects. This aspect justifies our assigning a special place to a purely interpretative technique [Footnote 19: This is not the place to discuss the epistemology of interpretation. For a comprehensive treatise on interpretation, see a paper by Bernfeld (1932) which regrettably is not available in English. See also Waelder (1939)].

It is well known that the proper use of interpretation is difficult and complicated. But so central is this tool that any proposed variation or addition should be scrutinized with the greatest care. The introduction of parameters, even of such simple ones as are necessary in some cases of phobia, contains dangers which must not be overlooked. Each parameter increases the possibility that the therapeutic process may be falsified, inasmuch as it may offer the patient's ego the possibility of substituting obedience for a structural change.

The term obedience, not entirely an accurate one, is used here to designate all those improvements which a patient may show under the pressure of the therapy but which are not based on a dissolution of the corresponding conflicts. A patient often prefers to produce adjusted behavior instead of a structural change [Footnote 20: This is one of the many reasons why in psychotherapy one so frequently has an opportunity of enjoying bristling clinical successes and why the proper psychoanalytic technique always works against a much heavier resistance than any other technique].

Every introduction of a parameter incurs the danger that a resistance has been temporarily eliminated without having been properly analyzed. Therefore, after an obstacle has been removed by the use of a parameter, the meaning which this parameter has had for the patient and the reasons which necessitated the choice of the parameter must retrospectively be discussed; that is to say, interpretation must become again the exclusive tool to straighten out the ruffle which was caused by the use of a parameter.

At this point I must strongly emphasize that in my use of the term interpretation I always presuppose the proper use of this technique. It would, of course, be foolish to suggest that just any kind of interpretation, or the mere act of interpreting, will do. Again, this paper is not the place for a discussion of what a proper interpretative technique is; it is mandatory, however, that a warning be raised against the quick introduction of parameters under the justification that interpretations have been of no avail. There is a great temptation to cover up, by the introduction of parameters, one's own inability to use properly the interpretative technique.

In view of the paramount importance which one must attribute to ego modification as an obstacle to psychoanalytic therapy and therefore to recovery, the question of the cause of ego modification must be raised. It is again Freud who gives us the answer by delineating the twofold effect which defense mechanisms may have on the ego. They may protect the ego or they may destroy the ego. "The purpose of the defense mechanisms is to avert dangers. It cannot be disputed that they are successful; it is doubtful whether the ego can altogether do without them during its development, but it is also certain that they themselves may become dangerous. Not infrequently it turns out that the ego has paid too high a price for the services which these mechanisms render" (Freud, 1937, p. 340); (see also A. Freud, 1936, p. 54). And Freud suggests that we consider these deleterious effects of the defenses upon the ego as ego modification.

I want to illustrate such an effect briefly with a clinical example. A three-year-old girl was awakened one morning by her mother, who carried a newborn baby, the girl's sister, in her arms, saying: "Look here, Mary, this is Marguerit. Isn't she sweet?" Showing all the signs of enjoyment, the little girl agreed that Marguerit was sweet. Twenty-eight years later, in her analysis, the girl described the incident, reporting that her mother had descended with incredible swiftness upon her and complaining that her mother had no reason to assume that Mary knew Marguerit was the name of a girl. She claimed she was given no choice but to show the same emotion as her mother and that to meet the situation "requested" by her mother, she had to rally an incredible amount of energy. At the time of the analysis the patient reported that after a day of responsible work, which she performed to the satisfaction of her superiors, she would return home in a state of complete exhaustion. However, she was not exhausted by the exigencies of her work, but because she must rally an incredible amount of energy to show the emotion requested by her environment. To say good morning to her co-workers, to "pitch in with the elevator man" when he made a trivial remark about the weather, absorbed her energy. She found it necessary to brace herself constantly when in the company of others in order to respond appropriately to the respective social realities. She really would have liked to sit in a rocking chair alone in her room and hold her head in her arms.

Here we see that the defense accomplished its goal ideally in the little child in terms of facilitating social behavior, and this patient's history may serve as an example that the relatively symptom-free child is often the most endangered one. The whole jealousy, the whole terrific anger about her mother's unfaithfulness and rejection were excluded from consciousness and were replaced by the socially required admiration of and love for the baby [Footnote 21: I delete from this description the severe anxiety this patient suffered as a child and delineate only the effect which the defense had on the child's external behavior]. The little sister soon became the patient's favorite, and she cheerfully spent all her free time with the new companion, developing with surprising quickness strong maternal attitudes. None of her recollections indicated a disturbance of behavior or outward signs of ambivalence toward the baby. The ideal result of the defense, however, must be viewed simultaneously with the catastrophic effect it had on the ego organization. It seems that the defense devoured the ego like a cancerous growth devours the organisms harboring it.

A high degree of modification has occurred in the schizophrenic ego. The single defense mechanism and the individual patterns of defense mechanisms do not assist the ego but are destructive; they burden the ego to such an extent that it is constantly on the verge of breaking off its relationship with reality. This appears to be a complete contradiction to what is usually described as the fundamental process in schizophrenia, that is to say, the ego's being made subservient to the id. No doubt, in most phases of the schizophrenic psychosis, wish fulfillments of the id play a great role, but the function of defense was not deleted from Freud's (1924) metapsychological diagram of psychosis.

"… the falsifications of memory, the delusional systems and hallucinations, in so many cases and forms of psychosis are of a most distressing character and are bound up with a development of dread, which is surely a sign that the whole process of remodeling reality is conducted in the face of most strenuous opposition. We may, for our own purposes, reconstruct the process on the model of a neurosis, which is more familiar to us. Here we see that a reaction of anxiety takes place every time the repressed instinct makes a move forward, and that the result of the conflict is after all but a compromise and as a satisfaction is incomplete. Probably in a psychosis the rejected part of reality reasserts its claim, just as in neurosis the repressed instinct does; hence the consequences in both are the same." (Freud, 1924, p. 280).

Here one of the defensive purposes is clearly described. The ego of the psychotic must defend itself constantly against the perception, recognition, and acknowledgment of objective reality. As the ego can postpone awakening by responding to the arousal stimulus by an arousal dream, so can the psychotic prevent the intrusion of objective reality by the maintenance of his own self-created reality. The arousal dream requires a minimum of cathexis, and the unmodified ego suffers the pain of having overruled reality when, upon awakening, it pays the price for having unduly indulged in the desire to sleep. The psychotic ego must garner a very large amount of energy in order to feed constantly its own reality, and the unceasing struggle against the pain which would be aroused by the perception of objective reality in turn results in pain.

In a thoughtful study, Katan (1950) clearly describes the defensive function of one of Schreber's hallucinations of little men descending upon his head and then perishing after a short period of existence. Schreber reached a stage in his psychosis where he could masturbate without erection and emission:

"…The hallucination occurs instead of the excitement. …In the hallucination, sexual excitement does not occur at all, and instead of the idea of Schreber perishing, we find the idea of the other men losing their lives" (Freud, 1910c, p. 34).

By hallucination, the ego anticipated danger and warded it off. The remarkable feature, however, is that a fantasy or daydream or a passing thought of the same (though usually less bizarre) content may occur in the neurotic for the same purpose and with the same effect of sparing himself anxiety or excitement. Indeed, one may even venture to say that an ego relatively free of symptoms may maintain its functional organization by a passing thought of such a kind. Yet in Schreber the defensive process, leading to a hallucination, imposed itself upon the total ego, absorbed all its functions, and took full possession of the visual apparatus. It is permissible to say that at that moment the ego could not do anything else but hallucinate or, in other words, that the defense process had spread itself out at the cost of the rest of the ego.

In the case of Schreber's hallucination, the content against which the defense was directed (passive homosexual wishes) was not different from those which one finds quite often in neurotics, if Katan's reconstruction is correct. Yet sometimes the contents which the schizophrenic tries to ward off are quite surprising. The schizophrenic patient I have mentioned before assured me for years that there was only hatred in her, that she would like to see the people killed with whom she had to deal and that she was incapable of feeling any interest in or longing for any human being. However, when she started to tell me the daydreams which filled her mind during the hour it took her to fall asleep—until then she had chiefly reported the feelings and fantasies which she had in the company of others—I was amazed to hear of a fantasy in which she took care of a crippled and mentally disturbed girl whom she knew. With great skill and tact she got me acquainted in her fantasy with the child and arranged for the child's treatment and cure. Aside from the narcissistic-erotic elements which undoubtedly are to be found in that daydream, there was expressed a core of real warmth and affection. There was no doubt that this patient had kept repressed her social inclinations and that her elaborate fantasies of killing also served the purpose of denying her sociability [Footnote 22: It is questionable whether my description of the patient's repressing her social tendencies is correct. One of her problems was to be different from her mother. Since her mother was social, the patient had to add to any expression of friendliness the feeling that this was pretense only. The feeling of hatred was the last anchor left to her for the purpose of making sure that she was not identical with her mother]. This paradoxical constellation is not essentially different from that which Freud described when he wrote:

"I recollect a case of chronic paranoia in which after each attack of jealousy a dream conveyed to the analyst a correct picture of the cause, free from any trace of delusion. An interesting contrast was thus brought to light: For, while we are accustomed to discover from the dreams of neurotic patients jealousies which are alien to their waking lives, in the present psychotic case the delusion which dominated the patient's daytime existence was corrected by a dream" (Freud, 1938, p. 78).

A few factors may be mentioned which are involved in the ego's being victimized by its defensive apparatus. I believe that all defense mechanisms are initially fed by energy which has not been neutralized, just as the child's early thought processes are closer to the primary process than to the secondary. In the course of development the defense mechanisms are subjected to a process which is comparable to the change from the primary to the secondary thought processes. The energy which they consume is delibidinized and freed of primary aggression. The schizophrenic ego does not achieve this [Footnote 23: I am continuing here Hartmann's (1950, p. 88) suggestions regarding the energetics of defense mechanisms]. Its defenses are driven by passion and destruction. The use of destructive energy seems to explain why the schizophrenic ego is primarily a masochistic or self-destructive one; the use of libidinal energies would explain why some schizophrenics in certain phases of their disorder can substitute defensive processes for the whole sexual gratification [Footnote 24: Cf. Katan (1950)].

In the unmodified ego the whole apparatus of defense mechanisms functions vis-à-vis internal stimulation in the same way that the stimulus barrier functions to prevent overstimulation by external stimuli. In the schizophrenic the defensive apparatus does not possess the firmness necessary for this function. Therefore the ego is forced to respond in its totality without being able to channel adequately the internal or external demands, which threaten to engulf the whole ego. The world—the external as well as the internal one—always descends with incredible swiftness upon the schizophrenic.

The defense mechanisms, however, become particularly visible after the acute schizophrenic symptomatology has vanished and there emerges the organization of the ego which lies behind the picturesque schizophrenic symptomatology per se. Then one can observe the excessive demand which the schizophrenic puts upon the synthetic function (Nunberg, 1931). One can also see his inability to bear up under the impact of internal contradictions and his desperate fight for an ego purified of contradictory feelings; that is to say, for a purified pleasure ego—the only ego state known when we were completely one with ourselves, a state accessible to and desired by the adult chiefly in sleep. The fact that the defense mechanisms of the schizophrenic are still set in motion by, and use up, instinctual energy greatly reduces the seeming contradiction which I initially mentioned, the contradiction between the two metapsychological formulations: (1) That in schizophrenia the ego loses territory to the id, (2) that the ego is devoured by its defenses. The vast majority of schizophrenic patients who are clinically observed are in a stage in which the defenses are still working, but independently of and unchecked by a comprehensive, over-all ego organization; yet, since these defense mechanisms work in close co-operation with the id owing to energic conditions, it is also correct to speak of the id's encroachment on the ego.

The hypothesis that in schizophrenia the defensive apparatus is kept in motion by energy which has not been desexualized or neutralized must not be confused with another psychoanalytic proposition; namely, that many defense mechanisms may lead to instinctual gratifications, despite their defensive function. The degree of gratification, of course, varies. We are accustomed to find this coincidence of gratification and defense in the neurotic symptom, but it is also true of some defense mechanisms [Footnote 25: See Waelder (1936), Nunberg (1932, p. 94), and others. The quantitative relationship between gratification and defense is probably quite different in symptom and defense mechanism]. This does not, however, mean that a defense mechanism—which, apart from its effect as a tool of warding off instincts, leads to a partial discharge of id energy—is itself cathected by id energy. I think that at this point one must distinguish with particular exactness between the accomplishment of a defense mechanism within the personality and the cathexis of the defense mechanism per se. Projection leads always to the transfer of a content from within the personality to without, but a comparison of the sporadic, neurotic projections in a hysteria and the stable, rigid projections in a paranoid psychosis shows that the energic factors are quite different. The hypothesis that such differences are also based on the difference of the energies used by the mechanisms per se, facilitates, I think, the understanding of the ego modification encountered in schizophrenia.

Be this as it may, it is important to keep in mind Freud's statement that the defense mechanisms themselves present only one of the difficulties to be overcome in analysis. If the effect of these defense mechanisms has resulted in a modification of the ego, the analysis will face even greater difficulties, necessitating deviations from the basic model technique.

Unfortunately we do not yet have an adequate conceptual frame of reference to describe these ego modifications although we are constantly struggling with them in most patients who now come for analysis. Freud compared ego modifications with "dislocation or crippling," [Footnote 26: See Freud, 1932, p. 321). The German words are Verrenkung and Einschränkung. See also A. Freud, 1936, p. 54): "Thus repression becomes the basis of compromise-formation and neurosis. The consequences of the other defensive methods are not less serious but, even when they assume an acute form, they remain more within the limits of the normal. They manifest themselves in innumerable transformations, distortions and deformities of the ego, which are in part the accompaniment of and in part substitutes for neurosis."] but the metapsychology of such ego modifications has scarcely been established [Footnote 27: See A. Freud (1936, pp. 100-113, particularly p. 111) for an attempt in that direction]. Observing a patient, we watch the defense mechanisms and their interplay. We see single functions like judgment and perception and note their bearing upon each other. We observe some of the results such as the identifications and projections, but we are not able to perceive the ego organization which underlies them, the mysterious knot of which Goethe spoke and which makes a human being more than the aggregate of his defense mechanisms and functions. Indeed, it is most tantalizing to know of a problem, to observe its manifestation in clinical reality, but to be unable to evolve an adequate conceptual framework necessary for its solution.

Since the ego modification presents itself most conspicuously in the schizophrenias, one is forced to return to this group of disorders when discussing the chances and limitations of psychoanalysis in grappling with this sector of psychopathology [Footnote 28: See Freud (1915-17 pp. 357-373, and 1932 pp. 82-112) for Freud's remarks about psychoses in general and schizophrenia in particular as sources of insight into the structure of the ego].

Despite our great ignorance one statement can be made with certainty. The parameters necessary in the psychoanalysis of schizophrenia will be most extensive and numerous. The most remarkable difference, of course, concerns the essentially different technique of handling the transference [Footnote 29: See Waelder (1925) for a comparison of the techniques in the treatment of neurosis and psychosis]. In most neuroses the transference develops spontaneously, and the technical problem consists of converting transference into a helper of the analytic process by means of interpretation, while in some phases of the treatment of the schizophrenic transference must be produced by action, gesture, or words, and for long stretches the chief technical problem consists of manipulating the therapeutic situation in such a way as to effect, quantitatively and qualitatively, the proper accretion of transference.

In discussing parameters enforced by the ego modification which is prevalent in schizophrenia, a therapeutic task must be mentioned which holds no place, or only a very subordinate one, in the treatment of neurotic ego modifications. The schizophrenic must acquire a capacity which the neurotic possesses fully unless he is temporarily deprived of it under the onslaught of an acute, emotional upsurge. I refer to the capacity for putting a mental distance between oneself and the phenomena of the mind, whether these are correlated to external or internal stimulation. It is the privilege of man to possess the antithetical capabilities of feeling at one with his experiences and also of elevating himself above them. What may be now an experience which fills out completely the borders of his consciousness, may become at any moment a content of observation, judgment, or evaluation. The schizophrenic, however, has lost this capacity in respect to certain contents, although the function per se is not destroyed. But to one sector of his life at least he is bound so firmly that he is incapable of elevating himself beyond its sphere. This lack is one of the most significant indices of the profound modification which the ego of the schizophrenic has suffered [Footnote 30: For a description and discussion of this problem see the consequential thoughts about a typology of psychopathology as presented by Waelder (1934). See also Freud (1932, p. 84) and Sterba (1934)].

A schizophrenic once impressively said: "I could rather believe that you or the world around me do not exist than assume that the voices I hear are not real." The schizophrenic has lost the ability to differentiate between the possible and the real in certain sectors of reality [Footnote 31: I take this formulation from Waelder (1934, p. 477)]. This incapacity to lift himself out of the context of phenomena at one point at least must make the technique of treating schizophrenics essentially different from that of neurotics if one extends the treatment to the treatment of the ego modification. It is strange to notice that this technical problem which is most typical of the treatment of schizophrenics is barely mentioned in the contemporary literature on the psychotherapy of schizophrenia [Footnote 32: Fromm-Reichmann (1950) seems to claim that there is essentially no difference between the technique of treatment of schizophrenics and neurotics, a point of view which in my opinion is tenable only if the field of therapeutic action is limited to the patient's interpersonal relationships with disregard of the patient's ego modification]. The analyst meets in this instance a task of formidable extent which cannot be sufficiently discussed here. It can only be said that sometimes when one succeeds in demonstrating to the schizophrenic that the symptom is a derivative of bodily sensations, one may reach a point where the schizophrenic can extend his faculty of objectivation also to this sector of his psychopathology.

In the following brief description, I have arbitrarily chosen two parameters which fairly regularly play a role in the treatment of schizophrenics: (1) goal construction, and (2) reduction of symptomatology.

1. The goal toward which psychoanalytic treatment strives is implicitly, though vaguely, represented in the neurotic's mind. The schizophrenic is deprived of such an integrated and elaborated goal. He must be provided with a diagram of the unmodified ego. Since the patient often does not know how such an ego does function, it is up to the analyst to provide the frame of reference, which often may be an entirely new one to the patient [Footnote 33: Personal communication of Dr. Edith Jacobson].

Certainly the argument will be raised that such measures do not fall within the scope of psychoanalysis but belong to education or instruction or correction. Yet I wonder whether this parameter necessarily leads us outside psychoanalysis. Education is essentially a technique which tries to force the ego to assimilate the ego alien or, in other words, to convert the ego alien into the ego syntonic [Footnote 34: I am aware of the insufficiency of such a broad and vague statement, but to clarify it would necessitate the enumeration of its many exceptions]. The parameter which I am briefly mentioning here concerns the reconstruction of a viable ego. It concerns a goal which the patient once upon a time had aspired to though probably he had never reached it. Education always tries to implant values; this parameter is essentially divorced from any value system, although admittedly if it is not used wisely, it may become tainted by the tacit application of value systems.

Education always restricts the ego in some way despite the accretion of content which it provides. This parameter, however, must never lead to a restrictive process within the ego. In other words, the reconstructive processes initiated by this parameter must lay the foundation for the later education of the ego. Therefore I would rather say that this parameter is essentially outside of education.

2. One of the most difficult tasks is to find and to demonstrate to the patient which function or which functions of the ego have been impaired and in what way. The dysfunctions which can be clinically seen on the surface are, of course, not the primary ones. The production of a delusion may occur owing to the injury of quite different functions of the ego. If the modification of the ego is to be undone, the specific function which is disturbed must be drawn into the treatment and brought to the patient's attention [Footnote 35: See also Waelder (1925)].

This parameter partly overlaps one which may occur in the analysis of a neurotic. In the neurotic, however, the parameter will usually not transgress interpretation whereas in the treatment of the schizophrenic it concerns a tool which is essentially beyond the scope of interpretation. The disturbed function must be isolated from interplay with others, and the patient must learn to study the way in which this particular function becomes altered under the impact of specific conditions. If a disturbed function is discussed while it is still riveted to others, interpretation will be far less successful than when the disturbed function has been presented to the patient in isolation. In one instance, distortions of reality which appeared like true delusions based on projection turned out to be supplements and confirmations of certain delusions which the patient had formed about herself. The complicated symptomatology could be reduced finally to a certain annoying bodily feeling on which the delusion of self was based. The distortions of reality were only a secondary formation produced by the patient's need (and her fear) of finding supported by external evidence that which she had assumed beforehand to be true regarding a process belonging to internal reality.

I want to try to narrow down as much as possible the problem on which, in my opinion, the chief question of psychoanalytic theory as well as practice hinges today. It goes back to a point which Freud advanced for the first time in 1920 and which was taken up and continued by Alexander (1927, p. 5) seven years later before being discussed once more by Freud (1937) in Analysis Terminable and Interminable when he spoke of the resistance against the uncovering of resistances. These secondary resistances become noticeable in the course of psychoanalytic treatment when the analyst tries to make the patient's consciousness focus on those resistances which ward off id impulses [Footnote 36: For the stratification of defenses, see Gero (1951)]. Then it becomes surprisingly evident that the modified ego is highly disinclined not only to become aware of id contents but also to become aware of some of those processes and contents which occur within its own boundaries [Footnote 37: See Freud (1925, pp. 25-27). Helene Deutsch (1939, p. 11) describes the narcissistic gratification which some patients derive from their defenses. This of course reduces the motivation to give up these defenses]. Yet these secondary resistances are also active outside of the analytic situation, just as the primary resistances (directed against the id) are constantly active although they become palpable mainly in the psychoanalytic situation. What is the function of these secondary resistances? The primary ones protect the ego against the spread of the id, and one of the functions of the secondary resistances is to prevent the spread of the primary defenses [Footnote 38: See Rapaport (1951, p. 692) for the effect of delay in the formation of psychic structure]. They also would like to arrogate to themselves the maximum territory as would the eternally insatiable id drives. Under normal conditions—that is to say, in an unmodified ego—they utilize neutralized energy and are fully occupied with their work against the id. In the modified ego, however, they turn also against the ego. In extreme instances the secondary defenses are swept away, and there is no barrier against that cancerous growth of which I have spoken figuratively before.

A clinical example may illustrate how the secondary resistances can make themselves noticeable in the treatment. A patient of superior intelligence, with unusually strong pregenital fixations but well-preserved psychosexual genital activity, filled long stretches of his analysis with repetitive complaints about trivial matters regarding his wife. He did not show any understanding of the obvious fact that the discrepancy between the intensity of his complaints and the triviality of their content required a discussion and explanation. One day he reported, somewhat abruptly, that he enjoyed his wife's doing the very things he had always complained of and that he knew how secretly to manipulate situations in such a way as to make his wife act the way he had considered so obnoxious and which gave him occasion to be cold and unfriendly to her.

When the sadistic, aggressive nature of this impulse was explained, he acknowledged it and even volunteered that he had known this for a long time. He showed some understanding of the uncanny sadistic technique with which he maneuvered his wife into the situation of a helpless victim without giving her an opportunity of defending herself.

The sadistic impulse had been warded off by means of denial and substitution of the opposite, since the patient tried to prove to himself and to the analyst that he was not cruel, but that he deserved pity owing to his wife's deficiencies. I tried to show the patient that his incessant complaining had also served the purpose of assuaging his feelings of guilt. The more successful he was in gratifying his sadism in the camouflaged way he used so expertly, the more he had to present himself the next day as injured and unjustly treated by fate in being married to an allegedly unsatisfactory partner. This interpretation was not accepted by the patient. He could not understand it; he could not follow me; and he insisted upon the validity of his complaints, although he had just agreed that he himself secretly induced his wife to behave in the manner about which he habitually complained to me the following day.

Here we encounter the paradoxical situation in which a patient accepts the interpretation of an id impulse and admits its existence but shows an excessive resistance against the interpretation of the corresponding defense mechanism. There are several reasons for the latter type of resistance. The patient's complaining was done with much emotion. The defense had become partly cathected with instinctual energy. Furthermore, in some instances it is questionable which is easier for the ego to give up, the gratification of an id impulse or the defense. I believe that this patient had come to a point when he would more readily forego the sadistic gratification and acquire mastery over this force than he would sacrifice the feeling of being unjustly treated by fate. Indeed, there is some wisdom in the paradox. As long as he holds to the defense which consists of his playing the role of the victim, there is hope that possibly in the future he can permit himself sadistic gratifications. Only after he has discarded this defense would his conscience no longer tolerate the camouflaged enjoyment of sadistic pleasures. In this instance the defense provided masochistic gratifications which rooted the mechanism with particular firmness in the ego. I have the impression that it is usually the masochism of the ego which makes the interpretation of defense mechanisms excessively difficult. The ego seems to feel particularly safe when the masochistic gratification is achieved in a process which genuinely wards off another drive.

In accordance with well-known features of the pathology of repression one can safely assume that the pathology of the secondary resistances will take one of two forms: It will be too strongly or too weakly cathected. Tentatively I would like to suggest that possibly the neurotic ego modification belongs to the former and the psychotic to the latter group. However, it is not probable that clinical reality would follow such neatly drawn lines. Be this as it may, in the patient just mentioned, the excessive growth of a defense mechanism can be observed although the ego modification was not of the schizophrenic type. A less modified ego would have received the interpretation with some relief and the resistance against the interpretation of the id impulse would have been much sharper.

The secondary defenses which do their main work subterraneously and which can be gauged predominantly by the study of their effect on the primary defenses probably form a part of a special organization within the ego and—depending on these secondary defenses—ego modification can, or cannot, be altered by psychoanalysis. There can be no doubt that the neurotic ego modification, such as that found in phobia, can be altered by psychoanalysis. In certain compulsive-obsessive neuroses of long standing, the possibility is questionable.

Although techniques have been devised to undo, at least temporarily, acute schizophrenic symptoms, it is highly debatable whether the modification which the schizophrenic ego so impressively shows can be altered by psychoanalysis. The psychotherapeutic techniques which are applied most commonly in the treatment of schizophrenics do not add substantially to our knowledge and understanding of schizophrenia, since most of them disregard the clinical fact that the problem of the therapy of schizophrenia is essentially the problem of undoing an ego modification. Many a psychotherapist takes on the schizophrenic to demonstrate his psychotherapeutic courage. He will not hesitate to apply any psychotherapeutic tool so long as it gives hope of forcing the schizophrenic out of his acute condition. In so far as such endeavors are heavily sprinkled with pseudo-analytic interpretations, one must call these techniques "wild" psychoanalysis. I think that the concept of a parameter and adherence to the four rules I mentioned may prevent us from falling into wild analysis, which is particularly tempting in the case of schizophrenia. In general, I think, one can say that the most promising source of knowledge of the structure of the ego will be found in an exact description and a justification—both in terms of metapsychology—of any deviation from the basic model technique whenever such deviation becomes necessary.


Alexander F. (1925). Review of Ferenczi S. and Rank O., "The Development of Psychoanalysis". Internat. Ztschr. f. Psychoanal, 11: 113-122.

Alexander F. (1927). The Psychoanalysis of the Total Personality. New York and Washington: Nerv. and Ment. Dis. Publ. Co.

Alexander F., French T.M. et al. (1946). Psychoanalytic Therapy: Principles and Applications. New York: Ronald Press (chapters 2, 4 e 17: The corrective emotional experience; Italian transl.: La esperienza emozionale correttiva. Psicoterapia e Scienze Umane, 1993, XXVII, 2: 85-101).

Bernfeld S. (1932). Der Begriff der "Deutung" in der Psychoanalyse. Ztschr. f. angewandte Psychol., 42: 448-497.

Binswanger L. (1945). Zur Frage der Häufigkeit der Schizophrenie im Kindesalter. Ztschr. f. Kinderpsychiat., 12: 33-50.

Breuer J. & Freud S. (1892-95 [1893]). The psychic mechanism of hysterical phenomena. Studies in Hysteria. New York and Washington: Nerv. and Ment. Dis. Publ. Co., 1936 (Italian transl.: Comunicazione preliminare. Sul meccanismo psichico dei fenomeni isterici. Opere di Sigmund Freud, 1: 175-188. Torino: Boringhieri, 1967).

Brunswick R.M. (1928). Supplement to Freud's "History of an Infantile Neurosis". Internat. J. Psycho-Anal., 9: 439-476.

Deutsch F. (1939). The associative anamnesis. Psychoanal. Quart., 8: 354-381.

Deutsch F. (1949). Applied Psychoanalysis. New York: Grune & Stratton.

Deutsch H. (1939). Ueber bestimmte Widerstandsformen. Internat. Ztscher. f. Psychoanal. u. Imago, 24: 10-20.

Eissler K.R. (1950). The "Chicago Institute of Psychoanalysis" and the sixth period of the development of psychoanalytic technique. Journal of General Psychology, 42: 103-157 (Italian transl.: Il Chicago Institute of Psychoanalysis e il sesto periodo dello sviluppo della tecnica psicoanalitica. Psicoterapia e Scienze Umane, 1984, XVIII, 3: 5-33 [I parte], e 4: 5-35 [II parte]).

Eissler K.R. (1955). The Psychiatrist and the Dying Patient. New York: International Universities Press.

Ferenczi S. & Rank O. (1923). Entwicklungsziele der Psychoanalyse. Wien: Internationaler Psychoanalytischer Verlag, 1924 (trad. inglese di Caroline Newton: The Development of Psychoanalysis. New York: Nervous and Mental Disease Publishing Co., 1925; New York: Dover Publications, 1956; trad. it. dei capitoli 1, 3 e 5 dalla traduz. francese: Prospettive di sviluppo della psicoanalisi. In: Fondamenti di psicoanalisi, Vol. III. Rimini: Guaraldi, 1974, pp. 175-193; riedizione in: Opere, 1919/1926, vol. III. Milano: Raffaello Cortina, 1992, pp. 201-216. Nuova trad. it. completa dall'originale tedesco, con una Postfazione di Michael Turnheim: Psicoterapia e Scienze Umane, 2012, XLVI, 4: 487-538).

Freud A. (1936). Das Ich und die Abwehrmechanismen. Wien: Internationaler Psychoanalytischer Verlag (English transl.: The Ego and the Mechanisms of Defense. New York: International Universities Press, 1946; Italian transl.: L'io e i meccanismi di difesa. Firenze: Martinelli, 1967. Also in: Scritti, Vol 1: 1922-1943. Torino: Boringhieri, 1978).

Freud A. (1945). Indications for child analysis. The Psychoanalytic Study of the Child, 1: 127-149 New York: Internat. Univ. Press.

Freud S. (1894). The defence neuro-psychoses. Coll. Papers, 1: 59-75 London: Hogarth Press1924 (Italian transl.: Le neuropsicosi da difesa. Opere di Sigmund Freud, 2: 121-134. Torino: Boringhieri, 1968).

Freud S. (1909). Notes upon a case of obsessional neurosis. Coll. Papers, 3: 293-383 (Italian transl.: Osservazioni su un caso di nevrosi ossessiva [Caso clinico dell'Uomo dei topi]. Opere di Sigmund Freud, 6: 1-124. Torino: Boringhieri, 1974).

Freud S. (1910a). Leonardo da Vinci. New York: Dodd, Mead, 1932 (Italian transl.: Un ricordo d'infanzia di Leonardo da Vinci. Opere di Sigmund Freud, 6: 213-276. Torino: Boringhieri, 1974).

Freud S. (1910b). Observations on "wild" psycho-analysis. Coll. Papers, 2: 297-304. London: Hogarth Press, 1924 (Italian transl.: Psicoanalisi "selvaggia". Opere di Sigmund Freud, 6: 325-331. Torino: Boringhieri, 1974).

Freud S. (1910c [1911]). Psychoanalytic notes on an autobiographical account of a case of paranoia (dementia paranoides). S.E., 12: 3-84 (Italian transl.: Osservazioni psicoanalitiche su un caso di paranoia [dementia paranoides] descritto autobiograficamente [Caso clinico del presidente Schreber]. Opere di Sigmund Freud, 6: 335-406. Torino: Boringhieri, 1974).

Freud S. (1913). Further recommendations in the technique of psycho-analysis: On beginning the treatment. Coll. Papers, 2: 342-365 (Italian transl.: Nuovi consigli sulla tecnica della psicoanalisi: 1. Inizio del trattamento. Opere di Sigmund Freud, 7: 333-352. Torino: Boringhieri, 1975).

Freud S. (1915-17 [1916-17]). A General Introduction to Psychoanalysis. New York: Garden City Publ. Co., 1935 (Italian transl.: Introduzione alla psicoanalisi. Opere di Sigmund Freud, 8: 191-611. Torino: Boringhieri, 1975)

Freud S. (1917). Metapsychological supplement to the theory of dreams. Coll. Papers, 4: 137-151 London: Hogarth Press,1925 (Italian transl.: Metapsicologia. Supplemento metapsicologico alla teoria del sogno. Opere di Sigmund Freud, 8: 89-101. Torino: Boringhieri, 1976).

Freud S. (1914 [1918]). From the history of an infantile neurosis. Coll. Papers, 3: 473-605 (Italian transl.: Dalla storia di una nevrosi infantile [Caso clinico dell'Uomo dei lupi]. Opere di Sigmund Freud, 7, 483-593. Torino: Boringhieri, 1975).

Freud S. (1920). Beyond the Pleasure Principle. London: Hogarth Press,1922 (Italian transl.: Al di là del principio di piacere. Opere di Sigmund Freud, 9: 189-249. Torino: Boringhieri, 1977).

Freud S. (1922 [1923]). The Ego and the Id. London: Hogarth Press, 1949 (Italian transl.: L'Io e l'Es. Opere di Sigmund Freud, 9: 471-520. Torino: Boringhieri, 1977).

Freud S. (1923 [1924]). Neurosis and psychosis. Coll. Papers, 2: 250-254. London: Hogarth Press, 1924 (Italian transl.: Nevrosi e psicosi. Opere di Sigmund Freud, 9: 611-615. Torino: Boringhieri, 1977).

Freud S. (1924a). The loss of reality in neurosis and psychosis. Coll. Papers, 2: 277-282 (Italian transl.: La perdita di realtà nelle nevrosi e nelle psicosi. Opere di Sigmund Freud, 10: 39-43. Torino: Boringhieri, 1978).

Freud S. (1924b [1925]). An Autobiographical Study. London: Hogarth Press, 1950 (Italian transl.: Autobiografia. Opere di Sigmund Freud, 10: 71-141. Torino: Boringhieri, 1978).

Freud S. (1925 [1926]). The Problem of Anxiety. New York: W. W. Norton & Co.,1936 (Italian transl.: Inibizione, sintomo e angoscia. Opere di Sigmund Freud, 10: 233-317. Torino: Boringhieri, 1978).

Freud S. (1932 [1933]). New Introductory Lectures on Psychoanalysis. New York: W. W. Norton & Co.,1933 (Italian transl.: Introduzione alla psicoanalisi [nuova serie di lezioni]. Opere di Sigmund Freud, 11: 117-284. Torino: Boringhieri, 1979).

Freud S. (1937). Analysis terminable and interminable. Coll. Papers, 5: 316-357. London: Hogarth Press, 1950 (Italian transl.: Analisi terminabile e interminabile. Opere di Sigmund Freud, 11: 497-535. Torino: Boringhieri, 1979).

Freud S. (1938 [1940]). An Outline of Psychoanalysis. New York: W. W. Norton and Co., 1950 (Italian transl.: Compendio di psicoanalisi. Opere di Sigmund Freud, 11: 569-634. Torino: Boringhieri, 1979).

Fromm-Reichmann F. (1950). Principles of Intensive Psychotherapy. Chicago: Univ. of Chicago Press (Italian transl.: Princìpi di psicoterapia. Milano: Feltrinelli, 1981).

Gero G. (1951). The concept of defense. Psychoanal. Quart., 20: 565-578

Goethe J.W. (1792). Das Tagebuch Sopienausgabe. Vol. 5/2, p. 345.

Hartmann H. (1939). Psycho-analysis and the concept of health. Internat. J. Psycho-Anal., 20: 308-321 (Italian transl.: Psicoanalisi e concetto di salute. In: Saggi sulla Psicologia dell'Io. Torino: Boringhieri, 1976, cap. 1, pp. 18-33).

Hartmann H. (1959). Comments on the psychoanalytic theory of the ego. The Psychoanalytic Study of the Child, 5: 74-96. Also in: Essays on Ego Psychology. New York: International Universities Press, 1964 (Italian transl.: Considerazioni sulla teoria psicoanalitica dell'Io. In: Saggi sulla Psicologia dell'Io. Torino: Boringhieri, 1976, cap. 7, pp. 129-157).

Ivy A. C. (1944). What is normal or normality. Quart. Bull. Northwest. Univ. Med. School, 18: 22-32.

Katan M. (1950). Schreber's hallucinations about the "little men". Internat. J. Psycho-Anal., 31: 32-35.

Nunberg H. (1932). Allgemeine Neurosenlehre. Bern: Huber.

Nunberg H. (1931). The synthetic function of the ego. Internat. J. Psycho-Anal., 12: 123-140 (Italian transl.: La funzione sintetica dell'Io. Psicoterapia e Scienze Umane, 1990, XXIV, 4: 91-108).

Rapaport D. (1951). Organization and Pathology of Thought. New York: Columbia Univ. Press.

Sterba R. (1934). Das Schicksal des Ichs im therapeutischen Varfharen. Int. Z. Psychoanal., 20: 66-73 (English transl.: The fate of the Ego in analytic therapy. Int. J. Psycho-Anal., 1934, 15: 117-126; Italian transl.: Il destino dell'Io nella terapia analitica. Psicoterapia e Scienze Umane, 1994, XXVIII, 2: 109-118).

Waelder R. (1925). The psychoses: their mechanisms and accessibility to influence. Internat. J. Psycho-Anal., 6 259-281.

Waelder R. (1936). On the principle of multiple function and overdetermination. Psychoanal. Quart., 5: 43-62 (Italian transl.: Il principio della funzione multipla. Osservazioni sulla sovradeterminazione. Psicoterapia e Scienze Umane, 1990, XXIV, 1: 107-123).

Waelder R. (1934). Das Freiheitsproblem in der Psychoanalyse. Imago, 20 467-484.

Waelder R. (1939). Kriterien der Deutung. Internat. Ztschr. f. Psychoanal. u. Imago, 24 136-145.