Excerpts from Must Read Books & Articles on Mental Health Topics
Articles- Part XXII

Sigmund Freud (1926)
Parts I-VII

IN the description of pathological phenomena, linguistic usage enables us to distinguish symptoms from inhibitions, without, however, attaching much importance to the distinction. Indeed, we might hardly think it worth while to differentiate exactly between the two, were it not for the fact that we meet with illnesses in which we observe the presence of inhibitions but not of symptoms and are curious to know the reason for this.
     The two concepts are not upon the same plane. Inhibition has a special relation to function. It does not necessarily have a pathological implication. One can quite well call a normal restriction of a function an inhibition of it. A symptom, on the other hand, actually denotes the presence of some pathological process. Thus, an inhibition may be a symptom as well. Linguistic usage, then, employs the word inhibition when there is a simple lowering of function, and symptom when a function has undergone some unusual change or when a new phenomenon has arisen out of it. Very often it seems to be quite an arbitrary matter whether we emphasize the positive side of a pathological process and call its outcome a symptom, or its negative side and call its outcome an inhibition. But all this is really of little interest; and the problem as we have stated it does not carry us very far.
     Since the concept of inhibition is so intimately associated with that of function, it might be helpful to examine the various functions of the ego with a view to discovering the forms which any disturbance of those functions assumes in each of the different neurotic affections. Let us pick out for a comparative study of this kind the sexual function and those of eating, of locomotion and of professional work.

(a) The sexual function is liable to a great number of disturbances, most of which exhibit the characteristics of simple inhibitions. These are classed together as psychical impotence. The normal performance of the sexual function can only come about as the result of a very complicated process, and disturbances may appear at any point in it. In men the chief stages at which inhibition occurs are shown by: a turning away of the libido at the very beginning of the process (psychical unpleasure); an absence of the physical preparation for it (lack of erection); an abridgement of the sexual act (ejaculatiopraecox), an occurrence which might equally well be regarded as a symptom; an arrest of the act before it has reached its natural conclusion (absence of ejaculation); or a non-appearance of the psychical outcome (lack of the feeling of pleasure in orgasm). Other disturbances arise from the sexual function becoming dependent on special conditions of a perverse or fetishist nature.
     That there is a relationship between inhibition and anxiety is pretty evident. Some inhibitions obviously represent a relinquishment of a function because its exercise would produce anxiety. Many women are openly afraid of the sexual function. We class this anxiety under hysteria, just as we do the defensive symptom of disgust which, arising originally as a deferred reaction to the experiencing of a passive sexual act, appears later whenever the idea of such an act is presented. Furthermore, many obsessional acts turn out to be measures of precaution and security against sexual experiences and are thus of a phobic character.
     This is not very illuminating. We can only note that disturbances of the sexual function are brought about by a great variety of means. (1) The libido may simply be turned away (this seems most readily to produce what we regard as an inhibition pure and simple); (2) the function may be less well carried out; (3) it may be hampered by having conditions attached to it, or modified by being diverted to other aims; (4) it may be prevented by security measures; (5) if it cannot be prevented from starting, it may be immediately interrupted by the appearance of anxiety; and (6), if it is nevertheless carried out, there may be a subsequent reaction of protest against it and an attempt to undo what has been done.
     (b) The function of nutrition is most frequently disturbed by a disinclination to eat, brought about by a withdrawal of libido. An increase in the desire to eat is also a not uncommon thing. The compulsion to eat is attributed to a fear of starving; but this is a subject which has been but little studied. The symptom of vomiting is known to us as a hysterical defence against eating. Refusal to eat owing to anxiety is a concomitant of psychotic states (delusions of being poisoned).
     (c) In some neurotic conditions locomotion is inhibited by a disinclination to walk or a weakness in walking. In hysteria there will be a paralysis of the motor apparatus, or this one special function of the apparatus will be abolished (abasia). Especially characteristic are the increased difficulties that appear in locomotion owing to the introduction of certain stipulations whose non-observance results in anxiety (phobia).
     (d] In inhibition in work--a thing which we so often have to deal with as an isolated symptom in our therapeutic work--the subject feels a decrease in his pleasure in it or becomes less able to do it well; or he has certain reactions to it, like fatigue, giddiness or sickness, if he is obliged to go on with it. If he is a hysteric he will have to give up his work owing to the appearance of organic and functional paralyses which make it impossible for him to carry it on. If he is an obsessional neurotic he will be perpetually being distracted from his work or losing time over it through the introduction of delays and repetitions.

      Our survey might be extended to other functions as well; but there would be nothing more to be learnt by doing so. For we should not penetrate below the surface of the phenomena presented to us. Let us then proceed to describe inhibition in such a way as to leave very little doubt about what is meant by it, and say that inhibition is the expression of a restriction of an ego-function. A restriction of this kind can itself have very different causes. Some of the mechanisms involved in this renunciation of function are well known to us, as is a certain general purpose which governs it.
     This purpose is more easily recognizable in the specific inhibitions. Analysis shows that when activities like playing the piano, writing or even walking are subjected to neurotic inhibitions it is because the physical organs brought into play--the fingers or the legs--have become too strongly erotized. It has been discovered as a general fact that the ego-function of an organ is impaired if its erotogenicity--its sexual significance--is increased. It behaves, if I may be allowed a rather absurd analogy, like a maid-servant who refuses to go on cooking because her master has started a love-affair with her. As soon as writing, which entails making a liquid flow out of a tube on to a piece of white paper, assumes the significance of copulation, or as soon as walking becomes a symbolic substitute for treading upon the body of mother earth, both writing and walking are stopped because they represent the performance of a forbidden sexual act. The ego renounces these functions, which are within its sphere, in order not to have to undertake fresh measures of repression--in order to avoid a conflict with the id.
     There are clearly also inhibitions which serve the purpose of self-punishment. This is often the case in inhibitions of professional activities. The ego is not allowed to carry on those activities, because they would bring success and gain, and these are things which the severe super-ego has forbidden. So the ego gives them up too, in order to avoid coming into conflict with the super-ego.
     The more generalized inhibitions of the ego obey a different mechanism of a simple kind. When the ego is involved in a particularly difficult psychical task, as occurs in mourning, or when there is some tremendous suppression of affect or when a continual flood of sexual phantasies has to be kept down, it loses so much of the energy at its disposal that it has to cut down the expenditure of it at many points at once. It is in the position of a speculator whose money has become tied up in his various enterprises. I came across an instructive example of this kind of intense, though short-lived, general inhibition. The patient, an obsessional neurotic, used to be overcome by a paralysing fatigue which lasted for one or more days whenever something occurred which should obviously have thrown him into a rage. We have here a point from which it should be possible to reach an understanding of the condition of general inhibition which characterizes states of depression, including the gravest form of them, melancholia.
     As regards inhibitions, then, we may say in conclusion that they are restrictions of the functions of the ego which have been either imposed as a measure of precaution or brought about as a result of an impoverishment of energy; and we can see without difficulty in what respect an inhibition differs from a symptom: for a symptom cannot any longer be described as a process that takes place within, or acts upon, the ego.


THE main characteristics of the formation of symptoms have long since been studied and, I hope, established beyond dispute. A symptom is a sign of, and a substitute for, an instinctual satisfaction which has remained in abeyance; it is a consequence of the process of repression. Repression proceeds from the ego when the latter--it may be at the behest of the superego--refuses to associate itself with an instinctual cathexis which has been aroused in the id. The ego is able by means of repression to keep the idea which is the vehicle of the reprehensible impulse from becoming conscious. Analysis shows that the idea often persists as an unconscious formation.
     So far everything seems clear; but we soon come upon difficulties which have not as yet been solved. Up till now our account of what occurs in repression has laid great stress on this point of exclusion from consciousness. But it has left other points open to uncertainty. One question that arose was, what happened to the instinctual impulse which had been activated in the id and which sought satisfaction? The answer was an indirect one. It was that owing to the process of repression the pleasure that would have been expected from satisfaction had been transformed into unpleasure. But we were then faced with the problem of how the satisfaction of an instinct could produce unpleasure. The whole matter can be clarified, I think, if we commit ourselves to the definite statement that as a result of repression the intended course of the excitatory process in the id does not occur at all; the ego succeeds in inhibiting or deflecting it. If this is so the problem of 'transformation of affect' under repression disappears. At the same time this view implies a concession to the ego that it can exert a very extensive influence over processes in the id, and we shall have to find out in what way it is able to develop such surprising powers.
     It seems to me that the ego obtains this influence in virtue of its intimate connections with the perceptual system--connections which, as we know, constitute its essence and provide the basis of its differentiation from the id. The function of this system, which we have called Pcpt.-Cs., is bound up with the phenomenon of consciousness. It receives excitations not only from outside but from within, and endeavours, by means of the sensations of pleasure and unpleasure which reach it from these quarters, to direct the course of mental events in accordance with the pleasure principle. We are very apt to think of the ego as powerless against the id; but when it is opposed to an instinctual process in the id it has only to give a 'signal of unpleasure' in order to attain its object with the aid of that almost omnipotent institution, the pleasure principle. To take this situation by itself for a moment, we can illustrate it by an example from another field. Let us imagine a country in which a certain small faction objects to a proposed measure the passage of which would have the support of the masses. This minority obtains command of the press and by its help manipulates the supreme arbiter, 'public opinion', and so succeeds in preventing the measure from being passed.
     But this explanation opens up fresh problems. Where does the energy come from which is employed for giving the signal of unpleasure? Here we may be assisted by the idea that a defence against an unwelcome internal process will be modelled upon the defence adopted against an external stimulus, that the ego wards off internal and external dangers alike along identical lines. In the case of external danger the organism has recourse to attempts at flight. The first thing it does is to withdraw cathexis from the perception of the dangerous object; later on it discovers that it is a better plan to perform muscular movements of such a sort as will render perception of the dangerous object impossible even in the absence of any refusal to perceive it--that it is a better plan, that is, to remove itself from the sphere of danger. Repression is an equivalent of this attempt at flight. The ego withdraws its (preconscious) cathexis from the instinctual representative that is to be repressed and uses that cathexis for the purpose of releasing unpleasure (anxiety). The problem of how anxiety arises in connection with repression may be no simple one; but we may legitimately hold firmly to the idea that the ego is the actual seat of anxiety and give up our earlier view that the cathectic energy of the repressed impulse is automatically turned into anxiety. If I expressed myself earlier in the latter sense, I was giving a phenomenological description and not a metapsychological account of what was occurring.
     This brings us to a further question: how is it possible, from an economic point of view, for a mere process of withdrawal and discharge, like the withdrawing of a preconscious ego-cathexis, to produce unpleasure or anxiety, seeing that, according to our assumptions, unpleasure and anxiety can only arise as a result of an increase in cathexis? The reply is that this causal sequence should not be explained from an economic point of view. Anxiety is not newly created in repression; it is reproduced as an affective state in accordance with an already existing mnemic image. If we go further and enquire into the origin of that anxiety--and of affects in general--we shall be leaving the realm of pure psychology and entering the borderland of physiology. Affective states have become incorporated in the mind as precipitates of primaeval traumatic experiences, and when a similar situation occurs they are revived like mnemic symbols. I do not think I have been wrong in likening them to the more recent and individually acquired hysterical attack and in regarding them as its normal prototypes. In man and the higher animals it would seem that the act of birth, as the individual's first experience of anxiety, has given the affect of anxiety certain characteristic forms of expression. But, while acknowledging this connection, we must not lay undue stress on it nor overlook the fact that biological necessity demands that a situation of danger should have an affective symbol, so that a symbol of this kind would have to be created in any case. Moreover, I do not think that we are justified in assuming that whenever there is an outbreak of anxiety something like a reproduction of the situation of birth goes on in the mind. It is not even certain whether hysterical attacks, though they were originally traumatic reproductions of this sort, retain that character permanently.
     As I have shown elsewhere, most of the repressions with which we have to deal in our therapeutic work are cases of after-pressure. They presuppose the operation of earlier, primal repressions which exert an attraction on the more recent situation. Far too little is known as yet about the background and preliminary stages of repression. There is a danger of overestimating the part played in repression by the super-ego. We cannot at present say whether it is perhaps the emergence of the super-ego which provides the line of demarcation between primal repression and after-pressure. At any rate, the earliest outbreaks of anxiety, which are of a very intense kind, occur before the super-ego has become differentiated. It is highly probable that the immediate precipitating causes of primal repressions are quantitative factors such as an excessive degree of excitation and the breaking through of the protective shield against stimuli.
     This mention of the protective shield sounds a note which recalls to us the fact that repression occurs in two different situations--namely, when an undesirable instinctual impulse is aroused by some external perception, and when it arises internally without any such provocation. We shall return to this difference later. But the protective shield exists only in regard to external stimuli, not in regard to internal instinctual demands.
     So long as we direct our attention to the ego's attempt at flight we shall get no nearer to the subject of symptom-formation. A symptom arises from an instinctual impulse which has been detrimentally affected by repression. If the ego, by making use of the signal of unpleasure, attains its object of completely suppressing the instinctual impulse, we learn nothing of how this has happened. We can only find out about it from those cases in which repression must be described as having to a greater or less extent failed. In this event the position, generally speaking, is that the instinctual impulse has found a substitute in spite of repression, but a substitute which is very much reduced, displaced and inhibited and which is no longer recognizable as a satisfaction. And when the substitutive impulse is carried out there is no sensation of pleasure; its carrying out has, instead, the quality of a compulsion.
     In thus degrading a process of satisfaction to a symptom, repression displays its power in a further respect. The substitutive process is prevented, if possible, from finding discharge through motility; and even if this cannot be done, the process is forced to expend itself in making alterations in the subject's own body and is not permitted to impinge upon the external world. It must not be transformed into action. For, as we know, in repression the ego is operating under the influence of external reality and therefore it debars the substitutive process from having any effect upon that reality.
    Just as the ego controls the path to action in regard to the external world, so it controls access to consciousness. In repression it exercises its power in both directions, acting in the one manner upon the instinctual impulse itself and in the other upon the [psychical] representative of that impulse. At this point it is relevant to ask how I can reconcile this acknowledgement of the might of the ego with the description of its position which I gave in The Ego and the Id. In that book I drew a picture of its dependent relationship to the id and to the superego and revealed how powerless and apprehensive it was in regard to both and with what an effort it maintained its show of superiority over them. This view has been widely echoed in psycho-analytic literature. Many writers have laid much stress on the weakness of the ego in relation to the id and of our rational elements in the face of the daemonic forces within us; and they display a strong tendency to make what I have said into a corner-stone of a psycho-analytic Weltanschauung. Yet surely the psycho-analyst, with his knowledge of the way in which repression works, should, of all people, be restrained from adopting such an extreme and one-sided view.
     I must confess that I am not at all partial to the fabrication of Weltanschauungen. Such activities may be left to philosophers, who avowedly find it impossible to make their journey through life without a Baedeker of that kind to give them information on every subject. Let us humbly accept the contempt with which they look down on us from the vantage-ground of their superior needs. But since we cannot forgo our narcissistic pride either, we will draw comfort from the reflection that such 'Handbooks to Life' soon grow out of date and that it is precisely our short-sighted, narrow and finicky work which obliges them to appear in new editions, and that even the most up-to-date of them are nothing but attempts to find a substitute for the ancient, useful and all-sufficient Church Catechism. We know well enough how little light science has so far been able to throw on the problems that surround us. But however much ado the philosophers may make, they cannot alter the situation. Only patient, persevering research, in which everything is subordinated to the one requirement of certainty, can gradually bring about a change. The benighted traveller may sing aloud in the dark to deny his own fears; but, for all that, he will not see an inch further beyond his nose.


To return to the problem of the ego. The apparent contradiction is due to our having taken abstractions too rigidly and attended exclusively now to the one side and now to the other of what is in fact a complicated state of affairs. We were justified, I think, in dividing the ego from the id, for there are certain considerations which necessitate that step. On the other hand the ego is identical with the id, and is merely a specially differentiated part of it. If we think of this part by itself in contradistinction to the whole, or if a real split has occurred between the two, the weakness of the ego becomes apparent. But if the ego remains bound up with the id and indistinguishable from it, then it displays its strength. The same is true of the relation between the ego and the super-ego. In many situations the two are merged; and as a rule we can only distinguish one from the other when there is a tension or conflict between them. In repression the decisive fact is that the ego is an organization and the id is not. The ego is, indeed, the organized portion of the id. We should be quite wrong if we pictured the ego and the id as two opposing camps and if we supposed that, when the ego tries to suppress a part of the id by means of repression, the remainder of the id comes to the rescue of the endangered part and measures its strength with the ego. This may often be what happens, but it is certainly not the initial situation in repression. As a rule the instinctual impulse which is to be repressed remains isolated. Although the act of repression demonstrates the strength of the ego, in one particular it reveals the ego's powerlessness and how impervious to influence are the separate instinctual impulses of the id. For the mental process which has been turned into a symptom owing to repression now maintains its existence outside the organization of the ego and independently of it. Indeed, it is not that process alone but all its derivatives which enjoy, as it were, this same privilege of extra-territoriality; and whenever they come into associative contact with a part of the ego-organization, it is not at all certain that they will not draw that part over to themselves and thus enlarge themselves at the expense of the ego. An analogy with which we have long been familiar compared a symptom to a foreign body which was keeping up a constant succession of stimuli and reactions in the tissue in which it was embedded. It does sometimes happen that the defensive struggle against an unwelcome instinctual impulse is brought to an end with the formation of a symptom. As far as can be seen, this is most often possible in hysterical conversion. But usually the outcome is different. The initial act of repression is followed by a tedious or interminable sequel in which the struggle against the instinctual impulse is prolonged into a struggle against the symptom.
     In this secondary defensive struggle the ego presents two faces with contradictory expressions. The one line of behaviour it adopts springs from the fact that its very nature obliges it to make what must be regarded as an attempt at restoration or reconciliation. The ego is an organization. It is based on the maintenance of free intercourse and of the possibility of reciprocal influence between all its parts. Its desexualized energy still shows traces of its origin in its impulsion to bind together and unify, and this necessity to synthesize grows stronger in proportion as the strength of the ego increases. It is therefore only natural that the ego should try to prevent symptoms from remaining isolated and alien by using every possible method to bind them to itself in one way or another, and to incorporate them into its organization by means of those bonds. As we know, a tendency of this kind is already operative in the very act of forming a symptom. A classical instance of this are those hysterical symptoms which have been shown to be a compromise between the need for satisfaction and the need for punishment. Such symptoms participate in the ego from the very beginning, since they fulfill a requirement of the super-ego, while on the other hand they represent positions occupied by the repressed and points at which an irruption has been made by it into the ego-organization. They are a kind of frontier-station with a mixed garrison. (Whether all primary hysterical symptoms are constructed on these lines would be worth enquiring into very carefully.) The ego now proceeds to behave as though it recognized that the symptom had come to stay and that the only thing to do was to accept the situation in good part and draw as much advantage from it as possible. It makes an adaptation to the symptom--to this piece of the internal world which is alien to it--just as it normally does to the real external world. It can always find plenty of opportunities for doing so. The presence of a symptom may entail a certain impairment of capacity, and this can be exploited to appease some demand on the part of the super-ego or to refuse some claim from the external world. In this way the symptom gradually comes to be the representative of important interests; it is found to be useful in asserting the position of the self and becomes more and more closely merged with the ego and more and more indispensable to it. It is only very rarely that the physical process of 'healing' round a foreign body follows such a course as this. There is a danger, too, of exaggerating the importance of a secondary adaptation of this kind to a symptom, and of saying that the ego has created the symptom merely in order to enjoy its advantages. It would be equally true to say that a man who had lost his leg in the war had got it shot away so that he might thenceforward live on his pension without having to do any more work.
     In obsessional neurosis and paranoia the forms which the symptoms assume become very valuable to the ego because they obtain for it, not certain advantages, but a narcissistic satisfaction which it would otherwise be without. The systems which the obsessional neurotic constructs flatter his self-love by making him feel that he is better than other people because he is specially cleanly or specially conscientious. The delusional constructions of the paranoic offer to his acute perceptive and imaginative powers a field of activity which he could not easily find elsewhere.
     All of this results in what is familiar to us as the '(secondary) gain from illness' which follows a neurosis. This gain comes to the assistance of the ego in its endeavour to incorporate the symptom and increases the symptom's fixation. When the analyst tries subsequently to help the ego in its struggle against the symptom, he finds that these conciliatory bonds between ego and symptom operate on the side of the resistances and that they are not easy to loosen.
     The two lines of behaviour which the ego adopts towards the symptom are in fact directly opposed to each other. For the other line is less friendly in character, since it continues in the direction of repression. Nevertheless the ego, it appears, cannot be accused of inconsistency. Being of a peaceable disposition it would like to incorporate the symptom and make it part of itself. It is from the symptom itself that the trouble comes. For the symptom, being the true substitute for and derivative of the repressed impulse, carries on the role of the latter; it continually renews its demands for satisfaction and thus obliges the ego in its turn to give the signal of unpleasure and put itself in a posture of defence.
     The secondary defensive struggle against the symptom takes many shapes. It is fought out on different fields and makes use of a variety of methods. We shall not be able to say much about it until we have made an enquiry into the various different instances of symptom-formation. In doing this we shall have an opportunity of going into the problem of anxiety--a problem which has long been looming in the background. The wisest plan will be to start from the symptoms produced by the hysterical neurosis; for we are not as yet in a position to consider the conditions in which the symptoms of obsessional neurosis, paranoia and other neuroses are formed.


LET us start with an infantile hysterical phobia of animals-- for instance, the case of 'Little Hans' [1909b], whose phobia of horses was undoubtedly typical in all its main features. The first thing that becomes apparent is that in a concrete case of neurotic illness the state of affairs is much more complex than one would suppose so long as one was dealing with abstractions. It takes a little time to find one's bearings and to decide which the repressed impulse is, what substitutive symptom it has found and where the motive for repression lies.
     'Little Hans' refused to go out into the street because he was afraid of horses. This was the raw material of the case. Which part of it constituted the symptom? Was it his having the fear? Was it his choice of an object for his fear? Was it his giving up of his freedom of movement? Or was it more than one of these combined? What was the satisfaction which he renounced? And why did he have to renounce it?
     At a first glance one is tempted to reply that the case is not so very obscure. 'Little Hans's' unaccountable fear of horses was the symptom and his inability to go out into the streets was an inhibition, a restriction which his ego had imposed on itself so as not to arouse the anxiety-symptom. The second point is clearly correct; and in the discussion which follows I shall not concern myself any further with this inhibition. But as regards the alleged symptom, a superficial acquaintance with the case does not even disclose its true formulation. For further investigation shows that what he was suffering from was not a vague fear of horses but a quite definite apprehension that a horse was going to bite him. This idea, indeed, was endeavouring to withdraw from consciousness and get itself replaced by an undefined phobia in which only the anxiety and its object still appeared. Was it perhaps this idea that was the nucleus of his symptom?
     We shall not make any headway until we have reviewed the little boy's psychical situation as a whole as it came to light in the course of the analytic treatment. He was at the time in the jealous and hostile Oedipus attitude towards his father, whom nevertheless--except in so far as his mother was the cause of estrangement--he dearly loved. Here, then, we have a conflict due to ambivalence: a well-grounded love and a no less justifiable hatred directed towards one and the same person. 'Little Hans's' phobia must have been an attempt to solve this conflict. Conflicts of this kind due to ambivalence are very frequent and they can have another typical outcome, in which one of the two conflicting feelings (usually that of affection) becomes enormously intensified and the other vanishes. The exaggerated degree and compulsive character of the affection alone betray the fact that it is not the only one present but is continually on the alert to keep the opposite feeling under suppression, and enable us to postulate the operation of a process which we call repression by means of reaction-formation (in the ego). Cases like 'Little Hans's' show no traces of a reaction-formation of this kind. There are clearly different ways of egress from a conflict due to ambivalence.
     Meanwhile we have been able to establish another point with certainty. The instinctual impulse which underwent repression in 'Little Hans' was a hostile one against his father. Proof of this was obtained in his analysis while the idea of the biting horse was being followed up. He had seen a horse fall down and he had also seen a playmate, with whom he was playing at horses, fall down and hurt himself. Analysis justified the inference that he had a wishful impulse that his father should fall down and hurt himself as his playmate and the horse had done. Moreover, his attitude towards someone's departure on a certain occasion makes it probable that his wish that his father should be out of the way also found less hesitating expression. But a wish of this sort is tantamount to an intention of putting one's father out of the way oneself--is tantamount, that is, to the murderous impulse of the Oedipus complex.
     So far there seem to be no connecting links between 'Little Hans's' repressed instinctual impulse and the substitute for it which we suspect is to be seen in his phobia of horses. Let us simplify his psychical situation by setting on one side the infantile factor and the ambivalence. Let us imagine that he is a young servant who is in love with the mistress of the house and has received some tokens of her favour. He hates his master, who is more powerful than he is, and he would like to have him out of the way. It would then be eminently natural for him to dread his master's vengeance and to develop a fear of him--just as 'Little Hans' developed a phobia of horses. We cannot, therefore, describe the fear belonging to this phobia as a symptom. If 'Little Hans', being in love with his mother, had shown fear of his father, we should have no right to say that he had a neurosis or a phobia. His emotional reaction would have been entirely comprehensible. What made it a neurosis was one thing alone: the replacement of his father by a horse. It is this displacement, then, which has a claim to be called a symptom, and which, incidentally, constitutes the alternative mechanism which enables a conflict due to ambivalence to be resolved without the aid of a reaction-formation. Such a displacement is made possible or facilitated at 'Little Hans's' early age because the inborn traces of totemic thought can still be easily revived. Children do not as yet recognize or, at any rate, lay such exaggerated stress upon the gulf that separates human beings from the animal world. In their eyes the grown man, the object of their fear and admiration, still belongs to the same category as the big animal who has so many enviable attributes but against whom they have been warned because he may become dangerous. As we see, the conflict due to ambivalence is not dealt with in relation to one and the same person: it is circumvented, as it were, by one of the pair of conflicting impulses being directed to another person as a substitutive object.
     So far everything is clear. But the analysis of 'Hans's' phobia has been a complete disappointment in one respect. The distortion which constituted the symptom-formation was not applied to the [psychical] representative (the ideational content) of the instinctual impulse that was to be repressed; it was applied to a quite different representative and one which only corresponded to a reaction to the disagreeable instinct. It would be more in accordance with our expectations if 'Little Hans' had developed, instead of a fear of horses, an inclination to ill-treat them and to beat them or if he had expressed in plain terms a wish to see them fall down or be hurt or even die in convulsions ('make a row with their feet'). Something of the sort did in fact emerge in his analysis, but it was not by any means in the forefront of his neurosis. And, curiously enough, if he really had produced a hostility of this sort not against his father but against horses as his main symptom, we should not have said that he was suffering from a neurosis. There must be something wrong either with our view of repression or with our definition of a symptom. One thing, of course, strikes us at once: if 'Little Hans' had really behaved like that to horses, it would mean that repression had in no way altered the character of his objectionable and aggressive instinctual impulse itself but only the object towards which it was directed.
     Undoubtedly there are cases in which this is all that repression does. But more than this happened in the development of 'Little Hans's' phobia--how much more can be guessed from a part of another analysis.
     As we know, 'Little Hans' alleged that what he was afraid of was that a horse would bite him. Now some time later I was able to learn something about the origin of another animal phobia. In this instance the dreaded animal was a wolf; it, too, had the significance of a father-substitute. As a boy the patient in question--a Russian whom I did not analyse till he was in his twenties--had had a dream (whose meaning was revealed in analysis) and, immediately after it, had developed a fear of being devoured by a wolf, like the seven little goats in the fairy tale. In the case of 'Little Hans' the ascertained fact that his father used to play at horses with him doubtless determined his choice of a horse as his anxiety-animal. In the same way it appeared at least highly probable that the father of my Russian patient used, when playing with him, to pretend to be a wolf and jokingly threaten to gobble him up. Since then I have come across a third instance. The patient was a young American who came to me for analysis. He did not, it is true, develop an animal phobia, but it is precisely because of this omission that his case helps to throw light upon the other two. As a child he had been sexually excited by a fantastic children's story which had been read aloud to him about an Arab chief who pursued a 'ginger-bread man' so as to eat him up. He identified himself with this edible person, and the Arab chief was easily recognizable as a father-substitute. This phantasy formed the earliest substratum of his auto-erotic phantasies.
     The idea of being devoured by the father is typical age-old childhood material. It has familiar parallels in mythology (e.g. the myth of Kronos) and in the animal kingdom. Yet in spite of this confirmation the idea is so strange to us that we can hardly credit its existence in a child. Nor do we know whether it really means what it seems to say, and we cannot understand how it can have become the subject of a phobia. Analytic observation supplies the requisite information. It shows that the idea of being devoured by the father gives expression, in a form that has undergone regressive degradation, to a passive, tender impulse to be loved by him in a genital-erotic sense. Further investigation of the case history leaves no doubt of the correctness of this explanation. The genital impulse, it is true, betrays no sign of its tender purpose when it is expressed in the language belonging to the superseded transitional phase between the oral and sadistic organizations of the libido. Is it, moreover, a question merely of the replacement of the [psychical] representative by a regressive form of expression or is it a question of a genuine regressive degradation of the genitally-directed impulse in the id? It is not at all easy to make certain. The case history of the Russian 'Wolf Man' gives very definite support to the second, more serious, view; for, from the time of the decisive dream onward, the boy became naughty, tormenting and sadistic, and soon afterwards developed a regular obsessional neurosis. At any rate, we can see that repression is not the only means which the ego can employ for the purpose of defence against an unwelcome instinctual impulse. If it succeeds in making an instinct regress, it will actually have done it more injury than it could have by repressing it. Sometimes, indeed, after forcing an instinct to regress in this way, it goes on to repress it.
    The case of the 'Wolf Man' and the somewhat less complicated one of 'Little Hans' raise a number of further considerations. But we have already made two unexpected discoveries. There can be no doubt that the instinctual impulse which was repressed in both phobias was a hostile one against the father. One might say that that impulse had been repressed by the process of being transformed into its opposite. Instead of aggressiveness on the part of the subject towards his father, there appeared aggressiveness (in the shape of revenge) on the part of his father towards the subject. Since this aggressiveness is in any case rooted in the sadistic phase of the libido, only a certain amount of degradation is needed to reduce it to the oral stage. This stage, while only hinted at in 'Little Hans's' fear of being bitten, was blatantly exhibited in the 'Wolf Man's' terror of being devoured. But, besides this, the analysis has demonstrated, beyond a shadow of doubt, the presence of another instinctual impulse of an opposite nature which had succumbed to repression. This was a tender, passive impulse directed towards the father, which had already reached the genital (phallic) level of libidinal organization. As regards the final outcome of the process of repression, this impulse seems, indeed, to have been the more important of the two; it underwent a more far-reaching regression and had a decisive influence upon the content of the phobia. In following up a single instinctual repression we have thus had to recognize a convergence of two such processes. The two instinctual impulses that have been overtaken by repression--sadistic aggressiveness towards the father and a tender passive attitude to him--form a pair of opposites. Furthermore, a full appreciation of 'Little Hans's' case shows that the formation of his phobia had had the effect of abolishing his affectionate object-cathexis of his mother as well, though the actual content of his phobia betrayed no sign of this. The process of repression had attacked almost all the components of his Oedipus complex--both his hostile and his tender impulses towards his father and his tender impulses towards his mother. In my Russian patient this state of affairs was much less obvious.
     These are unwelcome complications, considering that we only set out to study simple cases of symptom-formation due to repression, and with that intention selected the earliest and, to all appearances, most transparent neuroses of childhood. Instead of a single repression we have found a collection of them and have become involved with regression into the bargain. Perhaps we have added to the confusion by treating the two cases of animal phobia at our disposal--'Little Hans' and the 'Wolf Man'--as though they were cast in the same mould. As a matter of fact, certain differences between them stand out. It is only with regard to 'Little Hans' that we can say with certainty that what his phobia disposed of were the two main impulses of the Oedipus complex--his aggressiveness towards his father and his over-fondness for his mother. A tender feeling for his father was undoubtedly there too and played a part in repressing the opposite feeling; but we can prove neither that it was strong enough to draw repression upon itself nor that it disappeared afterwards. 'Hans' seems, in fact, to have been a normal boy with what is called a 'positive' Oedipus complex. It is possible that the factors which we do not find were actually at work in him, but we cannot demonstrate their existence. Even the most exhaustive analysis has gaps in its data and is insufficiently documented. In the case of the Russian the deficiency lies elsewhere. His attitude to female objects had been disturbed by an early seduction and his passive, feminine side was strongly developed. The analysis of his wolf-dream revealed very little intentional aggressiveness towards his father, but it brought forward unmistakable proof that what repression overtook was his passive tender attitude to his father. In his case, too, the other factors may have been operative as well; but they were not in evidence. How is it that, in spite of these differences in the two cases, almost amounting to an antithesis, the final outcome--a phobia--was approximately the same? The answer must be sought in another quarter. I think it will be found in the second fact which emerges from our brief comparative examination. It seems to me that in both cases we can detect what the motive force of the repression was and can substantiate our view of its nature from the line of development which the two children subsequently pursued. This motive force was the same in both of them. It was the fear of impending castration. 'Little Hans' gave up his aggressiveness towards his father from fear of being castrated. His fear that a horse would bite him can, without any forcing, be given the full sense of a fear that a horse would bite off his genitals, would castrate him. But it was from fear of being castrated, too, that the little Russian relinquished his wish to be loved by his father, for he thought that a relation of that sort presupposed a sacrifice of his genitals--of the organ which distinguished him from a female. As we see, both forms of the Oedipus complex, the normal, active form and the inverted one, came to grief through the castration complex. The Russian boy's anxiety-idea of being devoured by a wolf contained, it is true, no suggestion of castration, for the oral regression it had undergone had removed it too far from the phallic stage. But the analysis of his dream rendered further proof superfluous. It was a triumph of repression that the form in which his phobia was expressed should no longer have contained any allusion to castration.
     Here, then, is our unexpected finding: in both patients the motive force of the repression was fear of castration. The ideas contained in their anxiety--being bitten by a horse and being devoured by a wolf--were substitutes by distortion for the idea of being castrated by their father. This was the idea which had undergone repression. In the Russian boy the idea was an expression of a wish which was not able to subsist in the face of his masculine revolt; in 'Little Hans' it was the expression of a reaction in him which had turned his aggressiveness into its opposite. But the affect of anxiety, which was the essence of the phobia, came, not from the process of repression, not from the libidinal cathexes of the repressed impulses, but from the repressing agency itself. The anxiety belonging to the animal phobias was an untransformed fear of castration. It was therefore a realistic fear, a fear of a danger which was actually impending or was judged to be a real one. It was anxiety which produced repression and not, as I formerly believed, repression which produced anxiety.
     It is no use denying the fact, though it is not pleasant to recall it, that I have on many occasions asserted that in repression the instinctual representative is distorted, displaced, and so on, while the libido belonging to the instinctual impulse is transformed into anxiety. But now an examination of phobias, which should be best able to provide confirmatory evidence, fails to bear out my assertion; it seems, rather, to contradict it directly. The anxiety felt in animal phobias is the ego's fear of castration; while the anxiety felt in agoraphobia (a subject that has been less thoroughly studied) seems to be its fear of sexual temptation--a fear which, after all, must be connected in its origins with the fear of castration. As far as can be seen at present, the majority of phobias go back to an anxiety of this kind felt by the ego in regard to the demands of the libido. It is always the ego's attitude of anxiety which is the primary thing and which sets repression going. Anxiety never arises from repressed libido. If I had contented myself earlier with saying that after the occurrence of repression a certain amount of anxiety appeared in place of the manifestation of libido that was to be expected, I should have nothing to retract today. The description would be correct; and there does undoubtedly exist a correspondence of the kind asserted between the strength of the impulse that has to be repressed and the intensity of the resultant anxiety. But I must admit that I thought I was giving more than a mere description. I believed I had put my finger on a metapsychological process of direct transformation of libido into anxiety. I can now no longer maintain this view. And, indeed, I found it impossible at the time to explain how a transformation of that kind was carried out.
     It may be asked how I arrived at this idea of transformation in the first instance. It was while I was studying the 'actual neuroses', at a time when analysis was still a very long way from distinguishing between processes in the ego and processes in the id. I found that outbreaks of anxiety and a general state of preparedness for anxiety were produced by certain sexual practices such as coitus interruptus, undischarged sexual excitation or enforced abstinence--that is, whenever sexual excitation was inhibited, arrested or deflected in its progress towards satisfaction. Since sexual excitation was an expression of libidinal instinctual impulses it did not seem too rash to assume that the libido was turned into anxiety through the agency of these disturbances. The observations which I made at the time still hold good. Moreover, it cannot be denied that the libido belonging to the id-processes is subjected to disturbance at the instigation of repression. It might still be true, therefore, that in repression anxiety is produced from the libidinal cathexis of the instinctual impulses. But how can we reconcile this conclusion with our other conclusion that the anxiety felt in phobias is an ego anxiety and arises in the ego, and that it does not proceed out of repression but, on the contrary, sets repression in motion? There seems to be a contradiction here which it is not at all a simple matter to solve. It will not be easy to reduce the two sources of anxiety to a single one. We might attempt to do so by supposing that, when coitus is disturbed or sexual excitation interrupted or abstinence enforced, the ego scents certain dangers to which it reacts with anxiety. But this takes us nowhere. On the other hand, our analysis of the phobias seems to admit of no correction. Non liquet ['It is not clear.' An old legal verdict used when the evidence was inconclusive; compare the Scottish 'not proven'.]


WE set out to study the formation of symptoms and the secondary struggle waged by the ego against symptoms. But in picking on the phobias for this purpose we have clearly made an unlucky choice. The anxiety which predominates in the picture of these disorders is now seen as a complication which obscures the situation. There are plenty of neuroses which exhibit no anxiety whatever. True conversion hysteria is one of these. Even in its most severe symptoms no admixture of anxiety is found. This fact alone ought to warn us against making too close a connection between anxiety and symptom-formation. The phobias are so closely akin to conversion hysteria in every other respect that I have felt justified in classing them alongside of it under the name of 'anxiety hysteria'. But no one has as yet been able to say what it is that determines whether any given case shall take the form of a conversion hysteria or a phobia--has been able, that is to say, to establish what determines the generating of anxiety in hysteria.
     The commonest symptoms of conversion hysteria--motor paralyses, contractures, involuntary actions or discharges, pains and hallucinations--are cathectic processes which are either permanently maintained or intermittent. But this puts fresh difficulties in the way. Not much is actually known about these symptoms. Analysis can show what the disturbed excitatory process is which the symptoms replace. It usually turns out that they themselves have a share in that process. It is as though the whole energy of the process had been concentrated in this one part of it. For instance, it will be found that the pains from which a patient suffers were present in the situation in which the repression occurred; or that his hallucination was, at that time, a perception; or that his motor paralysis is a defence against an action which should have been performed in that situation but was inhibited; or that his contracture is usually a displacement of an intended innervation of the muscles in some other part of his body; or that his convulsions are the expression of an outburst of affect which has been withdrawn from the normal control of the ego. The sensation of unspleaure which accompanies the appearance of the symptoms varies in a striking degree. In chronic symptoms which have been displaced on to motility, like paralyses and contractures, it is almost always entirely absent; the ego behaves towards the symptoms as though it had nothing to do with them. In intermittent symptoms and in those concerned with the sensory sphere, sensations of unpleasure are as a rule distinctly felt; and in symptoms of pain these may reach an extreme degree. The picture presented is so manifold that it is difficult to discover the factor which permits of all these variations and yet allows a uniform explanation of them. There is, moreover, little to be seen in conversion hysteria of the ego's struggle against the symptom after it has been formed. It is only when sensitivity to pain in some part of the body constitutes the symptom that that symptom is in a position to play a dual role. The symptom of pain will appear no less regularly whenever the part of the body concerned is touched from outside than when the pathogenic situation which it represents is associatively activated from within; and the ego will take precautions to prevent the symptom from being aroused through external perceptions. Why the formation of symptoms in conversion hysteria should be such a peculiarly obscure thing I cannot tell; but the fact affords us a good reason for quitting such an unproductive field of enquiry without delay.
    Let us turn to the obsessional neuroses in the hope of learning more about the formation of symptoms. The symptoms belonging to this neurosis fall, in general, into two groups, each having an opposite trend. They are either prohibitions, precautions and expiations--that is, negative in character--or they are, on the contrary, substitutive satisfactions which often appear in symbolic disguise. The negative, defensive group of symptoms is the older of the two; but as illness is prolonged, the satisfactions, which scoff at all defensive measures, gain the upper hand. The symptom-formation scores a triumph if it succeeds in combining the prohibition with satisfaction so that what was originally a defensive command or prohibition acquires the significance of a satisfaction as well; and in order to achieve this end it will often make use of the most ingenious associative paths. Such an achievement demonstrates the tendency of the ego to synthesize, which we have already observed. In extreme cases the patient manages to make most of his symptoms acquire, in addition to their original meaning, a directly contrary one. This is a tribute to the power of ambivalence, which, for some unknown reason, plays such a large part in obsessional neuroses. In the crudest instance the symptom is diphasic: an action which carries out a certain injunction is immediately succeeded by another action which stops or undoes the first one even if it does not go quite so far as to carry out its opposite.
     Two impressions at once emerge from this brief survey of obsessional symptoms. The first is that a ceaseless struggle is being waged against the repressed, in which the repressing forces steadily lose ground; the second is that the ego and the super-ego have a specially large share in the formation of the symptoms.
     Obsessional neurosis is unquestionably the most interesting and repaying subject of analytic research. But as a problem it has not yet been mastered. It must be confessed that, if we endeavour to penetrate more deeply into its nature, we still have to rely upon doubtful assumptions and unconfirmed suppositions. Obsessional neurosis originates, no doubt, in the same situation as hysteria, namely, the necessity offending off the libidinal demands of the Oedipus complex. Indeed, every obsessional neurosis seems to have a substratum of hysterical symptoms that have been formed at a very early stage. But it is subsequently shaped along quite different lines owing to a constitutional factor. The genital organization of the libido turns out to be feeble and insufficiently resistant, so that when the ego begins its defensive efforts the first thing it succeeds in doing is to throw back the genital organization (of the phallic phase), in whole or in part, to the earlier sadistic-anal level. This fact of regression is decisive for all that follows.
     Another possibility has to be considered. Perhaps regression is the result not of a constitutional factor but of a time-factor. It may be that regression is rendered possible not because the genital organization of the libido is too feeble but because the opposition of the ego begins too early, while the sadistic phase is at its height. I am not prepared to express a definite opinion on this point, but I may say that analytic observation does not speak in favour of such an assumption. It shows rather that, by the time an obsessional neurosis is entered upon, the phallic stage has already been reached. Moreover, the onset of this neurosis belongs to a later time of life than that of hysteria--to the second period of childhood, after the latency period has set in. In a woman patient whose case I was able to study and who was overtaken by this disorder at a very late date, it became clear that the determining cause of her regression and of the emergence of her obsessional neurosis was a real occurrence through which her genital life, which had up till then been intact, lost all its value.
     As regards the metapsychological explanation of regression, I am inclined to find it in a 'defusion of instinct', in a detachment of the erotic components which, with the onset of the genital stage, had joined the destructive cathexes belonging to the sadistic phase.
     In enforcing regression, the ego scores its first success in its defensive struggle against the demands of the libido. (In this connection it is of advantage to distinguish the more general notion of' 'defence' from 'repression'. Repression is only one of the mechanisms which defence makes use of.) It is perhaps in obsessional cases more than in normal or hysterical ones that we can most clearly recognize that the motive force of defence is the castration complex and that what is being fended off are the trends of the Oedipus complex. We are at present dealing with the beginning of the latency period, a period which is characterized by the dissolution of the Oedipus complex, the creation or consolidation of the super-ego and the erection of ethical and aesthetic barriers in the ego. In obsessional neuroses these processes are carried further than is normal. In addition to the destruction of the Oedipus complex a regressive degradation of the libido takes place, the super-ego becomes exceptionally severe and unkind, and the ego, in obedience to the super-ego, produces strong reaction-formations in the shape of conscientiousness, pity and cleanliness. Implacable, though not always on that account successful, severity is shown in condemning the temptation to continue early infantile masturbation, which now attaches itself to regressive (sadistic-anal) ideas but which nevertheless represents the unsubjugated part of the phallic organization. There is an inherent contradiction about this state of affairs, in which, precisely in the interests of masculinity (that is to say, from fear of castration), every activity belonging to masculinity is stopped. But here, too, obsessional neurosis is only overdoing the normal method of getting rid of the Oedipus complex. We once more find here an illustration of the truth that every exaggeration contains the seed of its own undoing. For, under the guise of obsessional acts, the masturbation that has been suppressed approaches ever more closely to satisfaction.
     The reaction-formations in the ego of the obsessional neurotic, which we recognize as exaggerations of normal character-formation, should be regarded, I think, as yet another mechanism of defence and placed alongside of regression and repression. They seem to be absent or very much weaker in hysteria. Looking back, we can now get an idea of what is peculiar to the defensive process in hysteria. It seems that in it the process is limited to repression alone. The ego turns away from the disagreeable instinctual impulse, leaves it to pursue its course in the unconscious, and takes no further part in its fortunes. This view cannot be absolutely correct, for we are acquainted with the case in which a hysterical symptom is at the same time a fulfillment of a penalty imposed by the superego; but it may describe a general characteristic of the behaviour of the ego in hysteria.
     We can either simply accept it as a fact that in obsessional neurosis a super-ego of this severe kind emerges, or we can take the regression of the libido as the fundamental characteristic of the affection and attempt to relate the severity of the superego to it. And indeed the super-ego, originating as it does from the id, cannot dissociate itself from the regression and defusion of instinct which have taken place there. We cannot be surprised if it becomes harsher, unkinder and more tormenting than where development has been normal.
     The chief task during the latency period seems to be the fending-off of the temptation to masturbate. This struggle produces a series of symptoms which appear in a typical fashion in the most different individuals and which in general have the character of a ceremonial. It is a great pity that no one has as yet collected them and systematically analysed them. Being the earliest products of the neurosis they should best be able to shed light on the mechanisms employed in its symptom-formation. They already exhibit the features which will emerge so disastrously if a serious illness follows. They tend to become attached to activities (which would later be carried out almost automatically) such as going to sleep, washing, dressing and walking about; and they tend also to repetition and waste of time. Why this should be so is at present not at all clear; but the sublimation of anal-erotic components plays an unmistakable part in it.
     The advent of puberty opens a decisive chapter in the history of an obsessional neurosis. The genital organization which has been broken off in childhood starts again with great vigour. But, as we know, the sexual development in childhood determines what direction this new start at puberty will take. Not only will the early aggressive impulses be re-awakened; but a greater or lesser proportion of the new libidinal impulses--in bad cases the whole of them--will have to follow the course prescribed for them by regression and will emerge as aggressive and destructive tendencies. In consequence of the erotic trends being disguised in this way and owing to the powerful reaction-formations in the ego, the struggle against sexuality will henceforward be carried on under the banner of ethical principles. The ego will recoil with astonishment from promptings to cruelty and violence which enter consciousness from the id, and it has no notion that in them it is combating erotic wishes, including some to which it would not otherwise have taken exception. The over-strict super-ego insists all the more strongly on the suppression of sexuality, since this has assumed such repellent forms. Thus in obsessional neurosis the conflict is aggravated in two directions: the defensive forces become more intolerant and the forces that are to be fended off become more intolerable. Both effects are due to a single factor, namely, regression of the libido.
     A good deal of what has been said may be objected to on the ground that the unpleasant obsessive ideas are themselves quite conscious. But there is no doubt that before becoming conscious they have been through the process of repression. In most of them the actual wording of the aggressive instinctual impulse is altogether unknown to the ego, and it requires a good deal of analytic work to make it conscious. What does penetrate into consciousness is usually only a distorted substitute which is either of a vague, dream-like and indeterminate nature or so travestied as to be unrecognizable. Even where repression has not encroached upon the content of the aggressive impulse it has certainly got rid of its accompanying affective character. As a result, the aggressiveness appears to the ego not to be an impulsion (but, as the patients themselves say, merely a 'thought' which awakens no feeling. But the remarkable thing is that this is not the case. What happens is that the affect left out when the obsessional idea is perceived appears in a different place. The super-ego behaves as though repression had not occurred and as though it knew the real wording and full affective character of the aggressive impulse, and it treats the ego accordingly. The ego which, on the one hand, knows that it is innocent is obliged, on the other hand, to be aware of a sense of guilt and to carry a responsibility which it cannot account for. This state of affairs is, however, not so puzzling as it would seem at first sight. The behaviour of the super-ego is perfectly intelligible, and the contradiction in the ego merely shows that it has shut out the id by means of repression while remaining fully accessible to the influence of the super-ego. If it is asked why the ego does not also attempt to withdraw from the tormenting criticism of the super-ego, the answer is that it does manage to do so in a great number of instances. There are obsessional neuroses in which no sense of guilt whatever is present. In them, as far as can be seen, the ego has avoided becoming aware of it by instituting a fresh set of symptoms, penances or restrictions of a self-punishing kind. These symptoms, however, represent at the same time a satisfaction of masochistic impulses which, in their turn, have been reinforced by regression.
     Obsessional neurosis presents such a vast multiplicity of phenomena that no efforts have yet succeeded in making a coherent synthesis of all its variations. All we can do is to pick out certain typical correlations; but there is always the risk that we may have overlooked other uniformities of a no less important kind.
     I have already described the general tendency of symptom-formation in obsessional neurosis. It is to give ever greater room to substitutive satisfaction at the expense of frustration. Symptoms which once stood for a restriction of the ego come later on to represent satisfactions as well, thanks to the ego's inclination to synthesis, and it is quite clear that this second meaning gradually becomes the more important of the two. The result of this process, which approximates more and more to a complete failure of the original purpose of defence, is an extremely restricted ego which is reduced to seeking satisfaction in the symptoms. The displacement of the distribution of forces--in favour of satisfaction may have the dreaded final outcome of paralysing the will of the ego, which in every decision it has to make is almost as strongly impelled from the one side as from the other. The over-acute conflict between id and superego which has dominated the illness from the very beginning may assume such extensive proportions that the ego, unable to carry out its office of mediator, can undertake nothing which is not drawn into the sphere of that conflict.


IN the course of these struggles we come across two activities of the ego which form symptoms and which deserve special attention because they are obviously surrogates of repression and therefore well calculated to illustrate its purpose and technique. The fact that such auxiliary and substitutive techniques emerge may argue that true repression has met with difficulties in its functioning. If one considers how much more the ego is the scene of action of symptom-formation in obsessional neurosis than it is in hysteria and with what tenacity the ego clings to its relations to reality and to consciousness, employing all its intellectual faculties to that end--and indeed how the very process of thinking becomes hypercathected and erotized--then one may perhaps come to a better understanding of these variations of repression.
     The two techniques I refer to are undoing what has been done and isolating. The first of these has a wide range of application and goes back very far. It is, as it were, negative magic, and endeavours, by means of motor symbolism, to 'blow away' not merely the consequences of some event (or experience or impression) but the event itself. I choose the term 'blow away' advisedly, so as to remind the reader of the part played by this technique not only in neuroses but in magical acts, popular customs and religious ceremonies as well. In obsessional neurosis the technique of undoing what has been done is first met with in the 'diphasic' symptoms [p. 113], in which one action is cancelled out by a second, so that it is as though neither action had taken place, whereas, in reality, both have. This aim of undoing is the second underlying motive of obsessional ceremonials, the first being to take precautions in order to prevent the occurrence or recurrence of some particular event. The difference between the two is easily seen: the precautionary measures are rational, while trying to get rid of something by 'making it not to have happened' is irrational and in the nature of magic. It is of course to be suspected that the latter is the earlier motive of the two and proceeds from the animistic attitude towards the environment. This endeavour to undo shades off into normal behaviour in the case in which a person decides to regard an event as not having happened. But whereas he will take no direct steps against the event, and will simply pay no further attention to it or its consequences, the neurotic person will try to make the past itself non-existent. He will try to repress it by motor means. The same purpose may perhaps account for the obsession for repeating which is so frequently met with in this neurosis and the carrying out of which serves a number of contradictory intentions at once. When anything has not happened in the desired way it is undone by being repeated in a different way; and thereupon all the motives that exist for lingering over such repetitions come into play as well. As the neurosis proceeds, we often find that the endeavour to undo a traumatic experience is a motive of first-rate importance in the formation of symptoms. We thus unexpectedly discover a new, motor technique of defence, or (as we may say in this case with less inaccuracy) of repression.
      The second of these techniques which we are setting out to describe for the first time, that of isolation, is peculiar to obsessional neurosis. It, too, takes place in the motor sphere. When something unpleasant has happened to the subject or when he himself has done something which has a significance for his neurosis, he interpolates an interval during which nothing further must happen--during which he must perceive nothing and do nothing. This behaviour, which seems strange at first sight, is soon seen to have a relation to repression. We know that in hysteria it is possible to cause a traumatic experience to be overtaken by amnesia. In obsessional neurosis this can often not be achieved: the experience is not forgotten, but, instead, it is deprived of its affect, and its associative connections are suppressed or interrupted so that it remains as though isolated and is not reproduced in the ordinary processes of thought. The effect of this isolation is the same as the effect of repression with amnesia. This technique, then, is reproduced in the isolations of obsessional neurosis; and it is at the same time given motor reinforcement for magical purposes. The elements that are held apart in this way are precisely those which belong together associatively. The motor isolation is meant to ensure an interruption of the connection in thought. The normal phenomenon of concentration provides a pretext for this kind of neurotic procedure: what seems to us important in the way of an impression or a piece of work must not be interfered with by the simultaneous claims of any other mental processes or activities. But even a normal person uses concentration to keep away not only what is irrelevant or unimportant, but, above all, what is unsuitable because it is contradictory. He is most disturbed by those elements which once belonged together but which have been torn apart in the course of his development--as, for instance, by manifestations of the ambivalence of his father-complex in his relation to God, or by impulses attached to his excretory organs in his emotions of love. Thus, in the normal course of things, the ego has a great deal of isolating work to do in its function of directing the current of thought. And, as we know, we are obliged, in carrying out our analytic technique, to train it to relinquish that function for the time being, eminently justified as it usually is.
     We have all found by experience that it is especially difficult for an obsessional neurotic to carry out the fundamental rule of psycho-analysis. His ego is more watchful and makes sharper isolations, probably because of the high degree of tension due to conflict that exists between his super-ego and his id. While he is engaged in thinking, his ego has to keep off too much--the intrusion of unconscious phantasies and the manifestation of ambivalent trends. It must not relax, but is constantly prepared for a struggle. It fortifies this compulsion to concentrate and to isolate by the help of the magical acts of isolation which, in the form of symptoms, grow to be so noticeable and to have so much practical importance for the patient, but which are, of course, useless in themselves and are in the nature of ceremonials.
     But in thus endeavouring to prevent associations and connections of thought, the ego is obeying one of the oldest and most fundamental commands of obsessional neurosis, the taboo on touching. If we ask ourselves why the avoidance of touching, contact or contagion should play such a large part in this neurosis and should become the subject-matter of complicated systems, the answer is that touching and physical contact are the immediate aim of the aggressive as well as the loving object-cathexes. Eros desires contact because it strives to make the ego and the loved object one, to abolish all spatial barriers between them. But destructiveness, too, which (before the invention of long-range weapons) could only take effect at close quarters, must presuppose physical contact, a coming to grips. To 'touch' a woman has become a euphemism for using her as a sexual object. Not to 'touch' one's genitals is the phrase employed for forbidding auto-erotic satisfaction. Since obsessional neurosis begins by persecuting erotic touching and then, after regression has taken place, goes on to persecute touching in the guise of aggressiveness, it follows that nothing is so strongly proscribed in that illness as touching nor so well suited to become the central point of a system of prohibitions. But isolating is removing the possibility of contact; it is a method of withdrawing a thing from being touched in any way. And when a neurotic isolates an impression or an activity by interpolating an interval, he is letting it be understood symbolically that he will not allow his thoughts about that impression or activity to come into associative contact with other thoughts.
     This is as far as our investigations into the formation of symptoms take us. It is hardly worth while summing them up, for the results they have yielded are scanty and incomplete and tell us scarcely anything that we do not already know. It would be fruitless to turn our attention to symptom-formation in other disorders besides phobias, conversion hysteria and obsessional neurosis, for too little is known about them. But in reviewing those three neuroses together we are brought up against a very serious problem the consideration of which can no longer be put off. All three have as their outcome the destruction of the Oedipus complex; and in all three the motive force of the ego's opposition is, we believe, the fear of castration. Yet it is only in the phobias that this fear comes to the surface and is acknowledged. What has become of it in the other two neuroses? How has the ego spared itself this fear? The problem becomes accentuated when we recall the possibility, already referred to, that anxiety arises directly, by a kind of fermentation, from a libidinal cathexis whose processes have been disturbed. Furthermore, is it absolutely certain that fear of castration is the only motive force of repression (or defence)? If we think of neuroses in women we are bound to doubt it. For though we can with certainty establish in them the presence of a castration complex, we can hardly speak with propriety of castration anxiety where castration has already taken place.


LET us go back again to infantile phobias of animals; for, when all is said and done, we understand them better than any other cases. In animal phobias, then, the ego has to oppose a libidinal object-cathexis coming from the id--a cathexis that belongs either to the positive or the negative Oedipus complex--because it believes that to give way to it would entail the danger of castration. This question has already been discussed, but there still remains a doubtful point to clear up. In 'Little Hans's' case--that is, in the case of a positive Oedipus complex--was it his fondness for his mother or was it his aggressiveness towards his father which called out the defence by the ego? In practice it seems to make no difference, especially as each set of feelings implies the other; but the question has a theoretical interest, since it is only the feeling of affection for the mother which can count as a purely erotic one. The aggressive impulse flows mainly from the destructive instinct; and we have always believed that in a neurosis it is against the demands of the libido and not against those of any other instinct that the ego is defending itself. In point of fact we know that after 'Hans's' phobia had been formed, his tender attachment to his mother seemed to disappear, having been completely disposed of by repression, while the formation of the symptom (the substitutive formation) took place in relation to his aggressive impulses. In the 'Wolf Man' the situation was simpler. The impulse that was repressed--his feminine attitude towards his father--was a genuinely erotic one; and it was in relation to that impulse that the formation of his symptoms took place.
     It is almost humiliating that, after working so long, we should still be having difficulty in understanding the most fundamental facts. But we have made up our minds to simplify nothing and to hide nothing. If we cannot see things clearly we will at least see clearly what the obscurities are. What is hampering us here is evidently some hitch in the development of our theory of the instincts. We began by tracing the organization of the libido through its successive stages--from the oral through the sadistic-anal to the genital--and in doing so placed all the components of the sexual instinct on the same footing. Later it appeared that sadism was the representative of another instinct, which was opposed to Eros. This new view, that the instincts fall into two groups, seems to explode the earlier construction of the successive stages of libidinal organization. But we do not have to break fresh ground in order to find a way out of the difficulty. The solution has been at hand for a long time and lies in the fact that what we are concerned with are scarcely ever pure instinctual impulses but mixtures in various proportions of the two groups of instincts. If this is so, there is no need to revise our view of the organizations of the libido. A sadistic cathexis of an object may also legitimately claim to be treated as a libidinal one; and an aggressive impulse against the father can just as well be subjected to repression as a tender impulse towards the mother. Nevertheless we shall bear in mind for future consideration the possibility that repression is a process which has a special relation to the genital organization of the libido and that the ego resorts to other methods of defence when it has to secure itself against the libido on other levels of organization. To continue: a case like 'Little Hans's' does not enable us to come to any clear conclusion. It is true that in him an aggressive impulse was disposed of by repression, but this happened after the genital organization had been reached.
      This time we will not lose sight of the part played by anxiety. We have said that as soon as the ego recognizes the danger of castration it gives the signal of anxiety and inhibits through the pleasure-unpleasure agency (in a way which we cannot as yet understand) the impending cathectic process in the id. At the same time the phobia is formed. And now the castration anxiety is directed to a different object and expressed in a distorted form, so that the patient is afraid, not of being castrated by his father, but of being bitten by a horse or devoured by a wolf. This substitutive formation has two obvious advantages. In the first place it avoids a conflict due to ambivalence (for the father was a loved object, too), and in the second place it enables the ego to cease generating anxiety. For the anxiety belonging to a phobia is conditional; it only emerges when the object of it is perceived--and rightly so, since it is only then that the danger-situation is present. There is no need to be afraid of being castrated by a father who is not there. On the other hand one cannot get rid of a father; he can appear whenever he chooses. But if he is replaced by an animal, all one has to do is to avoid the sight of it--that is, its presence--in order to be free from danger and anxiety. 'Little Hans', therefore, imposed a restriction upon his ego. He produced the inhibition of not leaving the house, so as not to come across any horses. The young Russian had an even easier time of it, for it was hardly a privation for him not to look at a particular picture-book any more. If his naughty sister had not kept on showing him the book with the picture of the wolf standing upright in it, he would have been able to feel safe from his fear.
     On a previous occasion I have stated that phobias have the character of a projection in that they replace an internal, instinctual danger by an external, perceptual one. The advantage of this is that the subject can protect himself against an external danger by fleeing from it and avoiding the perception of it, whereas it is useless to flee from dangers that arise from within. This statement of mine was not incorrect, but it did not go below the surface of things. For an instinctual demand is, after all, not dangerous in itself; it only becomes so inasmuch as it entails a real external danger, the danger of castration. Thus what happens in a phobia in the last resort is merely that one external danger is replaced by another. The view that in a phobia the ego is able to escape anxiety by means of avoidance or of inhibitory symptoms fits in very well with the theory that that anxiety is only an affective signal and that no alteration has taken place in the economic situation.
     The anxiety felt in animal phobias is, therefore, an affective reaction on the part of the ego to danger; and the danger which is being signalled in this way is the danger of castration. This anxiety differs in no respect from the realistic anxiety which the ego normally feels in situations of danger, except that its content remains unconscious and only becomes conscious in the form of a distortion.
     The same will prove true, I think, of the phobias of adults, although the material which their neuroses work over is much more abundant and there are some additional factors in the formation of the symptoms. Fundamentally the position is identical. The agoraphobic patient imposes a restriction on his ego so as to escape a certain instinctual danger--namely, the danger of giving way to his erotic desires. For if he did so the danger of being castrated, or some similar danger, would once more be conjured up as it was in his childhood. I may cite as an instance the case of a young man who became agoraphobic because he was afraid of yielding to the solicitations of prostitutes and of contracting a syphilitic infection from them as a punishment.
     I am well aware that a number of cases exhibit a more complicated structure and that many other repressed instinctual impulses can enter into a phobia. But they are only tributary streams which have for the most part joined the main current of the neurosis at a later stage. The symptomatology of agoraphobia is complicated by the fact that the ego does not confine itself to making a renunciation. In order to rob the situation of danger it does more: it usually effects a temporal regression to infancy (in extreme cases, to a time when the subject was in his mother's womb and protected against the dangers which threaten him in the present). Such a regression now becomes a condition whose fulfillment exempts the ego from making its renunciation. For instance, an agoraphobic patient may be able to walk in the street provided he is accompanied, like a small child, by someone he knows and trusts; or, for the same reason, he may be able to go out alone provided he remains within a certain distance of his own house and does not go to places which are not familiar to him or where people do not know him. What these stipulations are will depend in each case on the infantile factors which dominate him through his neurosis. The phobia of being alone is unambiguous in its meaning, irrespective of any infantile regression: it is, ultimately, an endeavour to avoid the temptation to indulge in solitary masturbation. Infantile regression can, of course, only take place when the subject is no longer a child.
     A phobia generally sets in after a first anxiety attack has been experienced in specific circumstances, such as in the street or in a train or in solitude. Thereafter the anxiety is held in ban by the phobia, but it re-emerges whenever the protective condition cannot be fulfilled. The mechanism of phobia does good service as a means of defence and tends to be very stable. A continuation of the defensive struggle, in the shape of a struggle against the symptom, occurs frequently but not invariably.
     What we have learnt about anxiety in phobias is applicable to obsessional neuroses as well. In this respect it is not difficult for us to put obsessional neuroses on all fours with phobias. In the former, the mainspring of all later symptom-formation is clearly the ego's fear of its super-ego. The danger-situation from which the ego must get away is the hostility of the superego. There is no trace of projection here; the danger is completely internalized. But if we ask ourselves what it is that the ego fears from the super-ego, we cannot but think that the punishment threatened by the latter must be an extension of the punishment of castration. Just as the father has become depersonalized in the shape of the super-ego, so has the fear of castration at his hands become transformed into an undefined social or moral anxiety. But this anxiety is concealed. The ego escapes it by obediently carrying out the commands, precautions and penances that have been enjoined on it. If it is impeded in doing so, it is at once overtaken by an extremely distressing feeling of discomfort which may be regarded as an equivalent of anxiety and which the patients themselves liken to anxiety.
     The conclusion we have come to, then, is this. Anxiety is a reaction to a situation of danger. It is obviated by the ego's doing something to avoid that situation or to withdraw from it. It might be said that symptoms are created so as to avoid the generating of anxiety. But this does not go deep enough. It would be truer to say that symptoms are created so as to avoid a danger-situation whose presence has been signalled by the generation of anxiety. In the cases that we have discussed, the danger concerned was the danger of castration or of something traceable back to castration.
     If anxiety is a reaction of the ego to danger, we shall be tempted to regard the traumatic neuroses, which so often follow upon a narrow escape from death, as a direct result of a fear of death (or fear for life) and to dismiss from our minds the question of castration and the dependent relationships of the ego. Most of those who observed the traumatic neuroses that occurred during the last war took this line, and triumphantly announced that proof was now forthcoming that a threat to the instinct of self-preservation could by itself produce a neurosis without any admixture of sexual factors and without requiring any of the complicated hypotheses of psycho-analysis. It is in fact greatly to be regretted that not a single analysis of a traumatic neurosis of any value is extant. And it is to be regretted, not because such an analysis would contradict the aetiological importance of sexuality--for any such contradiction has long since been disposed of by the introduction of the concept of narcissism, which brings the libidinal cathexis of the ego into line with the cathexes of objects and emphasizes the libidinal character of the instinct of self-preservation--but because, in the absence of any analyses of this kind, we have lost a most precious opportunity of drawing decisive conclusions about the relations between anxiety and the formation of symptoms. In view of all that we know about the structure of the comparatively simple neuroses of everyday life, it would seem highly improbable that a neurosis could come into being merely because of the objective presence of danger, without any participation of the deeper levels of the mental apparatus. But the unconscious seems to contain nothing that could give any content to our concept of the annihilation of life. Castration can be pictured on the basis of the daily experience of the faeces being separated from the body or on the basis of losing the mother's breast at weaning. But nothing resembling death can ever have been experienced; or if it has, as in fainting, it has left no observable traces behind. I am therefore inclined to adhere to the view that the fear of death should be regarded as analogous to the fear of castration and that the situation to which the ego is reacting is one of being abandoned by the protecting super-ego--the powers of destiny--so that it has no longer any safeguard against all the dangers that surround it. In addition, it must be remembered that in the experiences which lead to a traumatic neurosis the protective shield against external stimuli is broken through and excessive amounts of excitation impinge upon the mental apparatus; so that we have here a second possibility--that anxiety is not only being signalled as an affect but is also being freshly created out of the economic conditions of the situation.
     The statement I have just made, to the effect that the ego has been prepared to expect castration by having undergone constantly repeated object-losses, places the question of anxiety in a new light. We have hitherto regarded it as an affective signal of danger; but now, since the danger is so often one of castration, it appears to us as a reaction to a loss, a separation. Even though a number of considerations immediately arise which make against this view, we cannot but be struck by one very remarkable correlation. The first experience of anxiety which an individual goes through (in the case of human beings, at all events) is birth, and, objectively speaking, birth is a separation from the mother. It could be compared to a castration of the mother (by equating the child with a penis). Now it would be very satisfactory if anxiety, as a symbol of a separation, were to be repeated on every subsequent occasion on which a separation took place. But unfortunately we are prevented from making use of this correlation by the fact that birth is not experienced subjectively as a separation from the mother, since the foetus, being a completely narcissistic creature, is totally unaware of her existence as an object. Another adverse argument is that we know what the affective reactions to a separation are: they are pain and mourning, not anxiety. Incidentally, it may be remembered that in discussing the question of mourning we also failed to discover why it should be such a painful thing.

Click here to go to Part VIII