The Ego and the Mechanisms of Defense
Anna Freud (1936)

Chapter 1- The Ego as the Seat of Observation

DEFINITION OF PSYCHOANALYSIS
There have been periods in the development of psychoanalytic science when the theoretical study of the individual ego was distinctly unpopular. Somehow or other, many analysts had conceived the idea that, in analysis, the value of the scientific and therapeutic work was in direct proportion to the depth of the psychic strata upon which attention is focused. Whenever interest was shifted from the deeper to the more superficial psychic strata--whenever, that is to say, research was deflected from the id to the ego--it was felt that here was a beginning of apostasy from psychoanalysis as a whole. The view held was that the term psychoanalysis should be reserved for the new discoveries dating to the unconscious psychic life, i.e., the study of repressed instinctual impulses, affects, and fantasies. With problems such as that of the adjustment of children or adults to the outside world, with concepts of value such as those of health and disease, virtue or vice, psychoanalysis was not properly concerned. It should confine its investigations exclusively to infantile fantasies carried on into adult life, imaginary gratifications, and the punishments apprehended in retribution for these.
     Such a definition of psychoanalysis was not infrequently met with in analytic writings and was perhaps warranted by the current usage, which has always treated psychoanalysis and depth psychology as synonymous terms. Moreover, there was some justification for it in the past, for it may be said that from the earliest years of our science its theory, built up as it was on an empirical basis, was pre-eminently a psychology of the unconscious or, as we should say today, of the id. But the definition immediately loses all claim to accuracy when we apply it to psychoanalytic therapy. From the beginning analysis, as a therapeutic method, was concerned with the ego and its aberrations: the investigation of the id and of its mode of operation was always only a means to an end. And the end was invariably the same: the correction of these abnormalities and the restoration of the ego to its integrity.
     When the writings of Freud, beginning with Group Psychology and the Analysis of the Ego (1921) and Beyond the Pleasure Principle (1920), took a fresh direction, the odium of analytic unorthodoxy no longer attached to the study of the ego and interest was definitely focused on the ego institutions. Since then the term "depth psychology" certainly does not cover the whole field of psychoanalytic research. At the present time we should probably define the task of analysis as follows: to acquire the fullest possible knowledge of all the three institutions of which we believe the psychic personality to be constituted and to learn what are their relations to one another and to the outside world. That is to say: in relation to the ego, to explore its contents, its boundaries, and its functions, and to trace the history of its dependence on the outside world, the id, and the superego; and, in relation to the id, to give an account of the instincts, i.e., of the id contents, and to follow them through the transformations which they undergo.

THE ID, THE EGO, AND THE SUPEREGO IN SELF-PERCEPTION
We all know that the three psychic institutions vary greatly in their accessibility to observation. Our knowledge of the id--which was formerly called the system Ucs.--can be acquired only through the derivatives which make their way into the systems Pcs. and Cs. If within the id a state of calm and satisfaction prevails, so that there is no occasion for any instinctual impulse to invade the ego in search of gratification and there to produce feelings of tension and unpleasure, we can learn nothing of the id contents. It follows, at least theoretically, that the id is not under all conditions open to observation.
     The situation is, of course, different in the case of the superego. Its contents are for the most part conscious and so can be directly arrived at by endopsychic perception. Nevertheless, our picture of the superego always tends to become hazy when harmonious relations exist between it and the ego. We then say that the two coincide, i.e., at such moments the superego is not perceptible as a separate institution either to the subject himself or to an outside observer. Its outlines become clear only when it confronts the ego with hostility or at least with criticism. The superego, like the id, becomes perceptible in the state which it produces within the ego: for instance, when its criticism evokes a sense of guilt.

THE EGO AS OBSERVER
Now this means that the proper field for our observation is always the ego. It is, so to speak, the medium through which we try to get a picture of the other two institutions.
     When the relations between the two neighboring powers--ego and id--are peaceful, the former fulfills to admiration its role of observing the latter. Different instinctual impulses are perpetually forcing their way from the id into the ego, where they gain access to the motor apparatus, by means of which they obtain gratification. In favorable cases the ego does not object to the intruder but puts its own energies at the other's disposal and confines itself to perceiving; it notes the onset of the instinctual impulse, the heightening of tension and the feelings of unpleasure by which this is accompanied and, finally, the relief from tension when gratification is experienced. Observation of the whole process gives us a clear and undistorted picture of the instinctual impulse concerned, the quantity of libido with which it is cathected, and the aim which it pursues. The ego, if it assents to the impulse, does not enter into the picture at all.
     Unfortunately the passing of instinctual impulses from one institution to the other may be the signal for all manner of conflicts, with the inevitable result that observation of the id is interrupted. On their way to gratification the id impulses must pass through the territory of the ego and here they are in an alien atmosphere. In the id the so-called "primary process" prevails; there is no synthesis of ideas, affects are liable to displacement, opposites are not mutually exclusive and may even coincide, and condensation occurs as a matter of course. The sovereign principle which governs the psychic processes is that of obtaining pleasure. In the ego, on the contrary, the association of ideas is subject to strict conditions, to which we apply the comprehensive term "secondary process"; further, the instinctual impulses can no longer seek direct gratification--they are required to respect the demands of reality and, more than that, to conform to ethical and moral laws by which the superego seeks to control the behavior of the ego. Hence these impulses run the risk of incurring the displeasure of institutions essentially alien to them. They are exposed to criticism and rejection and have to submit to every kind of modification. Peaceful relations between the neighboring powers are at an end. The instinctual impulses continue to pursue their aims with their own peculiar tenacity and energy, and they make hostile incursions into the ego, in the hope of overthrowing it by a surprise attack. The ego on its side becomes suspicious; it proceeds to counterattack and to invade the territory of the id. Its purpose is to put the instincts permanently out of action by means of appropriate defensive measures, designed to secure its own boundaries.
     The picture of these processes transmitted to us by means of the ego's faculty of observation is more confused but at the same time much more valuable. It shows us two psychic institutions in action at one and the same moment. No longer do we see an undistorted id impulse but an id impulse modified by some defensive measure on the part of the ego. The task of the analytic observer is to split up the picture, representing as it does a compromise between the separate institutions, into its component parts: the id, the ego, and, it may be, the superego.

INROADS BY THE ID AND BY THE EGO CONSIDERED AS MATERIAL FOR OBSERVATION
In all this we are struck by the fact that the inroads from the one side and from the other are by no means equally valuable from the point of view of observation. All the defensive measures of the ego against the id are carried out silently and invisibly. The most that we can ever do is to reconstruct them in retrospect: we can never really witness them in operation. This statement applies, for instance, to successful repression. The ego knows nothing of it; we are aware of it only subsequently, when it becomes apparent that something is missing. I mean by this that, when we try to form an objective judgment about a particular individual, we realize that certain id impulses are absent which we should expect to make their appearance in the ego in pursuit of gratification. If they never emerge at all, we can only assume that access to the ego is permanently denied to them, i.e., that they have succumbed to repression. But this tells us nothing of the process of repression itself.
     The same is true of successful reaction formation, which is one of the most important measures adopted by the ego as a permanent protection against the id. Such formations appear almost unheralded in the ego in the course of a child's development. We cannot always say that the ego's attention had previously been focused on the particular contrary instinctual impulse which the reaction formation replaces. As a rule, the ego knows nothing of the rejection of the impulse or of the whole conflict which has resulted in the implanting of the new characteristic. Analytic observers might easily take it for a spontaneous development of the ego, were it not that definite indications of obsessional exaggeration suggest that it is of the nature of a reaction and that it conceals a long-standing conflict. Here again, observation of the particular mode of defense does not reveal anything of the process by which it has been evolved.
     We note that all the important information which we have acquired has been arrived at by the study of inroads from the opposite side, namely, from the id to the ego. The obscurity of a successful repression is only equaled by the transparency of the repressive process when the movement is reversed, i.e., when the repressed material returns, as may be observed in neurosis. Here we can trace every stage in the conflict between the instinctual impulse and the ego's defense. Similarly, reaction formation can best be studied when such formations are in the process of disintegration. In such a case the id's inroad takes the form of a reinforcement of the libidinal cathexis of the primitive instinctual impulse which the reaction formation concealed. This enables the impulse to force its way into consciousness, and, for a time, instinctual impulse and reaction formation are visible within the ego side by side. Owing to another function of the ego--its tendency to synthesis--this condition of affairs, which is particularly favorable for analytic observation, lasts only for a few moments at a time. Then a fresh conflict arises between id derivative and ego activity, a conflict to decide which of the two is to keep the upper hand or what compromise they will adopt. If through reinforcement of its energic cathexis the defense set up by the ego is successful, the invading force from the id is routed and peace reigns once more in the psyche--a situation most unfruitful for our observations.

Chapter 2- The Application of Analytic Technique to the Study of the Psychic Institutions

In my first chapter I have described the conditions under which psychoanalytic observation of the psychic processes has had to be conducted. In what follows I propose to give account of the way in which our analytic technique, as has developed, has accommodated itself to these conditions.

HYPNOTIC TECHNIQUE IN THE PREANALYTIC PERIOD
In the hypnotic technique of the preanalytic period the role of the ego was still entirely negative. The purpose of the hypnotist was to arrive at the contents of the unconscious and he regarded the ego merely as a disturbing factor in his work. It was already known that by means of hypnosis it was possible to eliminate, or at any rate to overpower, the patient's ego. The new feature in the technique described in Studies on Hysteria (1893-1895) was this: that the physician took advantage of the elimination of the ego to gain access to the patient's unconscious--now known as the id--the way to which had hitherto been blocked by the ego. Thus the goal aimed at was the revelation of the unconscious; the ego was a disturbing factor and hypnosis was a means of getting rid of it temporarily. When a piece of unconscious material came to light in hypnosis, the physician introduced it to the ego, and the effect of thus forcibly bringing it into consciousness was to clear up the symptom. But the ego took no part in the therapeutic process. It tolerated the intruder only so long as it was itself under the influence of the physician who had induced hypnosis. Then it revolted and began a new struggle to defend itself against that element of the id which had been forced upon it, and so the laboriously achieved therapeutic success was vitiated. Thus it came about that the greatest triumph of hypnotic technique--the complete elimination of the ego during the period of investigation--proved prejudicial to permanent results and disillusionment as to the value of the technique set in.

FREE ASSOCIATION
Even in free association--the method which has since replaced hypnosis as an aid to research--the role of the ego is at first still a negative one. It is true that the patient's ego is no longer forcibly eliminated. Instead, it is required to eliminate itself, to refrain from criticizing the associations, and to disregard the claims of logical connection, which are at other times held to be legitimate. The ego is, in fact, requested to be silent and the id is invited to speak and promised that its derivatives shall not encounter the usual difficulties if they emerge into consciousness. Of course, it is never promised that, when they make their appearance in the ego, they will attain their instinctual aim, whatever that may be. The warrant is valid only for their translation into word representations: it does not entitle them to take control of the motor apparatus, which is their real purpose in emerging. Indeed, this apparatus is put out of action in advance by the strict rules of analytic technique. Thus we have to play a double game with the patient's instinctual impulses, on the one hand encouraging them to express themselves and, on the other, steadily refusing them gratification--a procedure which incidentally gives rise to one of the numerous difficulties in the handling of analytic technique.
     Even today many beginners in analysis have an idea that it is essential to succeed in inducing their patients really and invariably to give all their associations without modification or inhibition, i.e., to obey implicitly the fundamental rule of analysis. But, even if this ideal were realized, it would not represent an advance, for after all it would simply mean the conjuring up again of the now obsolete situation of hypnosis, with its one-sided concentration on the part of the physician upon the id. Fortunately for analysis such docility in the patient is in practice impossible. The fundamental rule can never be followed beyond a certain point. The ego keeps silence for a time and the id derivatives make use of this pause to force their way into consciousness. The analyst hastens to catch their utterances when the ego bestirs itself again, repudiates the attitude of passive tolerance which it has been compelled to assume, and by means of one or other of its customary defense mechanisms intervenes in the flow of associations. The patient transgresses the fundamental rule of analysis, or, as we say, he puts up "resistances." This means that the inroad of the id into the ego has given place to a counterattack by the ego upon the id. The observer's attention is now diverted from the associations to the resistance, i.e., from the content of the id to the activity of the ego. The analyst has an opportunity of witnessing, then and there, the putting into operation by the latter of one of those defensive measures against the id which I have already described and which are so obscure, and it now behoves him to make it the object of his investigation. He then notes that with this change of object the situation in the analysis has suddenly changed. In analyzing the id he is assisted by the spontaneous tendency of the id derivatives to rise to the surface: his exertions and the strivings of the material which he is trying to analyze are similarly directed. In the analysis of the ego's defensive operations there is, of course, no such community of aim. The unconscious elements in the ego have no inclination to become conscious and derive no advantage from so doing. Hence any piece of ego analysis is much less satisfactory than the analysis of the id. It has to proceed by circuitous paths, it cannot follow out the ego activity directly, the only possibility is to reconstruct it from its influence on the patient's associations. From the nature of the effect produced--whether it be omission, reversal, displacement of meaning, etc.--we hope to discover what kind of defense the ego has employed in its intervention. So it is the analyst's business first of all to recognize the defense mechanism. When he has done this, he has accomplished a piece of ego analysis. His next task is to undo what has been done by the defense, i.e., to find out and restore to its place that which has been omitted through repression, to rectify displacements, and to bring that which has been isolated back into its true context. When he has re-established the severed connections, he turns his attention once more from the analysis of the ego to that of the id.
     We see then that what concerns us is not simply the enforcement of the fundamental rule of analysis for its own sake but the conflict to which this gives rise. It is only when observation is focused now on the id and now on the ego and the direction of interest is twofold, extending to both sides of the human being whom we have before us, that we can speak of psychoanalysis, as distinct from the one-sided method of hypnosis.
     The various other means employed in analytic technique can now be classified without difficulty, according to whether the attention of the observer is directed to one side or the other.

INTERPRETATION OF DREAMS
The situation when we are interpreting our patient's dreams and when we are listening to his free associations is the same. The dreamer's psychic state differs little from that of the patient during the analytic hour. When he obeys the fundamental rule of analysis he voluntarily suspends some functions of the ego; in the dreamer this suspension takes place automatically under the influence of sleep. The patient is made to lie at rest on the analyst's couch, in order that he may have no opportunity to gratify his instinctual wishes in action; similarly, in sleep, the motor system is brought to a standstill. And the effect of the censorship, the translation of latent dream thoughts into manifest dream content, with the distortions, condensations, displacements, reversals, and omissions which this involves, corresponds to the distortions which take place in the associations under the pressure of some resistance. Dream interpretation, then, assists us in our investigation of the id, insofar as it is successful in bringing to light latent dream thoughts (id content), and in our investigation of the ego institutions and their defensive operations, insofar as it enables us to reconstruct the measures adopted by the censor from their effect upon the dream thoughts.

INTERPRETATION OF SYMBOLS
One by-product of dream interpretation, namely, the understanding of dream symbols, contributes largely to the success of our study of the id. Symbols are constant and universally valid relations between particular id contents and specific word or thing representations. The knowledge of these relations enables us to draw reliable inferences from conscious manifestations as to the unconscious material behind them, without having first laboriously to reverse some measure which the ego has adopted in defense. The technique of translating symbols is a short cut to understanding, or, more correctly, a way of plunging from the highest strata of consciousness to the lowest strata of the unconscious without pausing at the intermediate strata of former ego activities which may in time past have forced a particular id content to assume a specific ego form. The knowledge of the language of symbols has the same sort of value for the understanding of the id as mathematical formulae have for the solution of typical problems. Such formulae may be used with advantage. It does not matter if one is ignorant of the way in which they were originally arrived at. But, though they help to solve the problems, they do not contribute to our understanding of mathematics. In the same way, by translating symbols we may reveal the contents of the id without really gaining any deeper psychological understanding of the individual with whom we are dealing.

PARAPRAXES
From time to time we obtain further glimpses of the unconscious in another way, in those irruptions of the id which are known as parapraxes. As we know, these irruptions are not confined to the analytic situation. They may occur at any time when, in some special circumstances, the vigilance of the ego is relaxed or diverted and an unconscious impulse again owing to some special circumstances) is suddenly reinforced. Such parapraxes, especially in the form of slips of the tongue and forgetting, may of course occur in analysis, when they illuminate as with a flash of lightning some part of the unconscious which we have perhaps long been endeavoring to interpret analytically. In the early days of analytic technique such windfalls were welcomed as afford a well-nigh irrefutable proof of the existence of the unconscious to patients who tended to be impervious to analytic insight. Then, too, analysts were glad to be able to demonstrate by means of easily understood examples various mechanisms, such as those of displacement, condensation, and omission. But, generally speaking, the importance of these chance occurrences for analytic technique dwindles in comparison with that of those irruptions of the id which are deliberately brought in to assist our analytic work.

TRANSFERENCE
The same theoretical distinction between observation of the id on the one hand and observation of the ego on the other may be drawn in the case of that which is perhaps the most powerful instrument in the analyst's hand: the interpretation of the transference. By transference we mean all those impulses experienced by the patient in his relation with the analyst which are not newly created by the objective analytic situation but have their source in early--indeed, the very earliest--object relations and are now merely revived under the influence of the repetition compulsion. Because these impulses are repetitions and not new creations they are of incomparable value as a means of information about the patient's past affective experiences. We shall see that we can distinguish different types of transference phenomena according to the degree of their complexity.

Transference of Libidinal Impulses
The first type of transference is extremely simple. The patient finds himself disturbed in his relation to the analyst by passionate emotions, e.g., love, hate, jealousy, and anxiety, which do not seem to be justified by the facts of the actual situation. The patient himself resists these emotions and feels ashamed, humiliated, and so forth, when they manifest themselves against his will. Often it is only by insisting on the fundamental rule of analysis that we succeed in forcing a passage for them to conscious expression. Further investigation reveals the true character of these affects--they are irruptions of the id. They have their source in old affective constellations, such as the oedipus and the castration complex, and they become comprehensible and indeed are justified if we disengage them from the analytic situation and insert them into some infantile affective situation. When thus put back into their proper place, they help us to fill up an amnestic gap in the patient's past and provide us with fresh information about his infantile instinctual and affective life. Generally he is quite willing to cooperate with us in our interpretation, for he himself feels that the transferred affective impulse is an intrusive foreign body. By putting it back into its place in the past we release him from an impulse in the present which is alien to his ego, thus enabling him to carry on the work of analysis. It should be noted that the interpretation of this first type of transference assists in the observation of the id only.

Transference of Defense
The case alters when we come to the second type of transference. The repetition compulsion, which dominates the patient in the analytic situation, extends not only to former id impulses but equally to former defensive measures against the instincts. Thus he not only transfers undistorted infantile id impulses, which become subject to a censorship on the part of the adult ego secondarily and not until they force their way to conscious expression; he also transfers id impulses in all those forms of distortion which took shape while he was still in infancy. It may happen in extreme cases that the instinctual impulse itself never enters into the transference at all but only the specific defense adopted by the ego against some positive or negative attitude of the libido, as, for instance, the reaction of flight from a positive love fixation in latent female homosexuality or the submissive, feminine-masochistic attitude, to which Wilhelm Reich (1933) has called attention in male patients whose relations to their fathers were once characterized by aggression. In my opinion, we do our patients a great injustice if we describe these transferred defense reactions as "camouflage" or say that the patients are "pulling the analyst's leg" or purposely deceiving him in some other way. And indeed we shall find it hard to induce them by an iron insistence on the fundamental rule, that is to say, by putting pressure upon them to be candid, to expose the id impulse which lies hidden under the defense as manifested in the transference. The patient is in fact candid when he gives expression to the impulse or affect in the only way still open to him, namely, in the distorted defensive measure. I think that in such a case the analyst ought not to omit all the intermediate stages in the transformation which the instinct has undergone and endeavor at all costs to arrive directly at the primitive instinctual impulse against which the ego has set up its defense and to introduce it into the patient's consciousness. The more correct method is to change the focus of attention in the analysis, shifting it in the first place from the instinct to the specific mechanism of defense, i.e., from the id to the ego. If we succeed in retracing the path followed by the instinct in its various transformations, the gain in the analysis is twofold. The transference phenomenon which we have interpreted falls into two parts, both of which have their origin in the past: a libidinal or aggressive element, which belongs to the id, and a defense mechanism, which we must attribute to the ego--in the most instructive cases, to the ego of the same infantile period in which the id impulse first arose. Not only do we fill in a gap in the patient's memory of his instinctual life, as we may also do when interpreting the first, simple type of transference, but we acquire information which completes and fills in the gaps in the history of his ego development or, to put it another way, the history of the transformations through which his instincts have passed.
     The interpretation of the second type of transference is more fruitful than that of the first type, but it is responsible for most of the technical difficulties which arise between analyst and patient. The latter does not feel the second kind of transference reaction to be a foreign body, and this is not surprising when we reflect how great a part the ego plays--even though it be the ego of earlier years--in its production. It is not easy to convince him of the repetitive nature of these phenomena. The form in which they emerge in his consciousness is ego syntonic. The distortions demanded by the censorship were accomplished long ago and the adult ego sees no reason for being on its guard against their making their appearance in his free associations. By means of rationalization he easily shuts his eyes to the discrepancies between cause and effect which are so noticeable to the observer and make it evident that the transference has no objective justification. When the transference reactions take this form, we cannot count on the patient's willing cooperation, as we can when they are of the type first described. Whenever the interpretation touches on the unknown elements of the ego, its activities in the past, that ego is wholly opposed to the work of analysis. Here evidently we have the situation which we commonly describe by the not very felicitous term "character analysis."
     From the theoretical standpoint, the phenomena revealed by interpretation of the transference fall into two groups: that of id contents and that of ego activities, which in each case have been brought into consciousness. The results of interpretation during the patient's free association may be similarly classified: the uninterrupted flow of associations throws light on the contents of the id; the occurrence of a resistance, on the defense mechanisms employed by the ego. The only difference is that interpretations of the transference relate exclusively to the past and may light up in a moment whole periods of the patient's past life, while the id contents revealed in free association are not connected with any particular period and the ego's defensive operations, manifested during the analytic hour in the form of resistance to free association, may belong to his present life also.

Acting in the Transference
Yet another important contribution to our knowledge of the patient is made by a third form of transference. In dream interpretation, free association, the interpretation of resistance, and in the forms of transference hitherto described, the patient as we see him is always inside the analytic situation, i.e., in an unnatural endopsychic state. The relative strength of the two institutions has been altered: the balance is weighted in favor of the id, in the one case through the influence of sleep and, in the other, through the observance of the fundamental rule of analysis. The strength of the ego factors when we encounter them--whether in the form of the dream censorship or in that of resistance to free associations--has always been impaired and their influence diminished, and often it is extremely difficult for us to picture them in their natural magnitude and vigor. We are all familiar with the accusation not infrequently made against analysts--that they may have a good knowledge of a patient's unconscious but are bad judges of his ego. There is probably a certain amount of justification in this criticism, for the analyst lacks opportunities of observing the patient's whole ego in action.
     Now an intensification of the transference may occur, during which for the time being the patient ceases to observe the strict rules of analytic treatment and begins to act out in the behavior of his daily life both the instinctual impulses and the defensive reactions which are embodied in his transferred affects. This is what is known as acting in the transference--a process in which, strictly speaking, the bounds of analysis have already been overstepped. It is instructive from the analyst's standpoint, in that the patient's psychic structure is thus automatically revealed in its natural proportions. Whenever we succeed in interpreting this "acting," we can divide the transference activities into their component parts and so discover the actual quantity of energy supplied at that particular moment by the different institutions. In contrast to the observations that we made during the patient's free associations, this situation shows us the absolute and the relative amount naturally contributed by each institution.
     Although in this respect the interpretation of "acting" in the transference affords us some valuable insight, the therapeutic gain is generally small. The bringing of the unconscious into consciousness and the exercise of therapeutic influence upon the relations between id, ego, and superego clearly depend upon the analytic situation, which is artificially produced and still resembles hypnosis in that the activity of the ego institutions is curtailed. As long as the ego continues to function freely or if it makes common cause with the id and simply carries out its behests, there is but little opportunity for endopsychic displacements and the bringing to bear of influence from without. Hence this third form of transference, which we call acting, is even more difficult for the analyst to deal with than the transference of the various modes of defense. It is natural that he should try to restrict it as far as possible by means of the analytic interpretations which he gives and the nonanalytic prohibitions which he imposes.

THE RELATION BETWEEN THE ANALYSIS
OF THE ID AND THAT OF THE EGO
I have described in considerable detail how the phenomena of transference come under three headings: transference of libidinal tendencies, transference of defensive attitudes, and acting in the transference. My purpose has been to show that the technical difficulties of analysis are relatively less when it is a question of bringing the id derivatives into consciousness and that they are at their greatest when we have to grapple with the unconscious elements in the ego. This might be better expressed as follows: the difficulty is not inherent in our analytic technique as such; that is no less well adapted to bring into consciousness the unconscious part of the ego than to bring into consciousness the unconscious part of the id or superego. Only, we analysts are less familiar with the difficulties of ego analysis than with those of id analysis. Analytic theory has ceased to hold that the concept of the ego is identical with that of the system of perceptual consciousness; that is to say, we have realized that large portions of the ego institutions are themselves unconscious and require the help of analysis in order to become conscious. The result is that analysis of the ego has assumed a much greater importance in our eyes. Anything which comes into the analysis from the side of the ego is just as good material as an id derivative. We have no right to regard it as simply an interruption to the analysis of the id. But of course anything which comes from the ego is also a resistance in every sense of the word: a force directed against the emerging of the unconscious and so against the work of the analyst. It is our ambition to learn to handle the analysis of a patient's ego, even though it has to be carried through against that ego's will, with at least no less sure a touch than the analysis of his id.

ONE-SIDEDNESS IN ANALYTIC TECHNIQUE AND
THE DIFFICULTIES TO WHICH IT LEADS
We know from what has already been said that, if we devote our attention to our patients' free associations and latent dream thoughts to the translation of symbols and the contents of the transference, whether fantasied or acted, we may make progress in our investigation of the id, but the analysis is one-sided. On the other hand, the study of resistances, of the work of the dream censorship and the various transferred modes of defense against instinctual impulses and fantasies will assist our investigation of the unknown activities of the ego and the superego, but this
method is equally one-sided. If it is true that only a combination of the two lines of research, without a bias in either direction, can produce a complete picture of the analysand's inner situation, then it must also be the case that, if we give the preference to any one of the means of analytic investigation at the cost of all the others, the result will inevitably be a distorted or at least an incomplete picture of the psychic personality--a travesty of the reality.
     For instance, a technique which confined itself too exclusively to translating symbols would be in danger of bringing to light material which consisted, also too exclusively, of id contents. Anyone employing such a technique would naturally be inclined to neglect or at all events to attach less importance to those unconscious elements in the ego institutions which can be brought into consciousness only by some of the other means at our disposal in analysis. One might seek to justify such a technique by saying that there was really no need for it to take the circuitous route by way of the ego, seeing that it could reach the repressed instinctual life directly. Nevertheless, its results would still be incomplete. Only the analysis of the ego's unconscious defensive operations can enable us to reconstruct the transformations which the instincts have undergone. Without a knowledge of these we may, indeed, discover much about the contents of the repressed instinctual wishes and fantasies, but we shall learn little or nothing about the vicissitudes through which they have passed and the various ways in which they enter into the structure of the personality.
     A technique which inclined too far in the other direction, so that the foreground was occupied exclusively by the analysis of the patient's resistances, would also be defective in its results, but on the opposite side. This method would give us a picture of the whole structure of the analysand's ego, but depth and completeness in the analysis of his id would have to be sacrificed.
     The results of a technique which concentrated too much on the transference would be similar. There is no doubt that patients, when in the state of intensified transference which such a method would foster, produce abundant material from the deepest strata of the id. But, in so doing, they overstep the bounds of the analytic situation. The ego no longer remains outside, its energies diminished, its strength reduced, its attitude that of objective observation, with no active part in what is going on. It is caught up, overwhelmed, swept into action. Even though, under the domination of the repetition compulsion, it behaves wholly as an infantile ego, this does not alter the fact that it is acting instead of analyzing. But this means that such a technique, embarked upon with high hopes of attaining a more profound knowledge of our patients, may end in all those disappointments from the therapeutic standpoint which on theoretical grounds we should naturally expect to result from acting in the transference.
     Again, the technique of child analysis which I myself have advocated (1926-1927) is a good example of the dangers of one-sidedness. If we must give up free association, make but a sparing use of the interpretation of symbols, and begin to interpret the transference only at an advanced stage in the treatment, three important avenues to the discovery of id contents and ego activities are closed to us. The question then arises which I propose to answer in the next chapter: how can we make good these deficiencies and, in spite of all, pass beyond the superficial strata of psychic life?