Beyond the Pleasure Principle and Other Writings

by Sigmund Freud

  As a rule, a phobia only emerges after the individual has first been stricken by fear in a particular set of circumstances, for example when he is in the street, or in a train, or on his own. The fear is thereupon shut out, but reappears whenever the protective stratagem cannot be maintained. The phobia mechanism does splendid service as a means of defence, and tends to exhibit marked stability. In many cases – though not in all – the defensive battle is carried a stage further, being henceforth directed against the symptom.

  What we have learnt about fear in the context of phobias also remains relevant to obsessional neurosis. It is not difficult to reduce the overall situation characteristic of obsessional neurosis to that characteristic of phobias. Here, the motor driving all subsequent symptom-formation is plainly the ego's fear vis-á-vis the super-ego. The hostility of the super-ego constitutes the danger situation that the ego must fight shy of. There is not the least semblance of projection here: the danger is wholly internalized. But if we ask ourselves what the ego is so fearful of suffering at the hands of the super-ego, then we are compelled to conclude that the punishment threatened by the latter is simply a refined version of the punishment of castration. Just as the super-ego is the father in depersonalized form, so too the specific fear of being castrated by him has changed into an indefinite social or consciential fear.36 But this fear remains latent; the ego keeps it at a safe distance by obediently carrying out whatever commands, prescriptions and penances are imposed upon it. If it is prevented from so doing, then the most extreme discomfiture immediately ensues, which we may reasonably regard as the equivalent of fear, and which patients themselves equate with fear. We thus arrive at the following conclusion: fear consists in a reaction to a particular danger situation; the ego saves itself from this fear by taking action to withdraw from the situation or avoid it altogether. Now one might be tempted to say that symptoms are produced in order to avoid any fear being generated in the first place – but this does not really get us very far. It is more accurate to say that symptoms are produced in order to avoid the danger situation signalled by the fear that has already been generated. In the cases considered so far, however, this danger lay in castration or in something that can be traced back to castration.

  If fear is the ego's reaction to danger, then it may seem logical to construe traumatic neurosis - which so often ensues where the individual has survived a life-threatening danger - as a direct result of fear of death or fear for life and limb,37 thereby disregarding castration and the dependent status of the ego.38 This indeed is what most of those with clinical experience of the traumatic neuroses of the last war39 have done, and the triumphant cry has gone up that we now have proof positive that a threat to the self-preservation drive can produce a neurosis without any involvement on the part of sexuality, and without paying the slightest heed to the complex processes posited by psychoanalysis. It really is extremely regrettable that there exists not a single reliable analysis of a traumatic neurosis. Our regrets relate not to the denial of the aetiological importance of sexuality, for any such denial has long since been refuted by the introduction of the concept of narcissism, which places libidinal cathexis of the ego in the same category as object-cathexes and stresses the libidinal nature of the self-preservation drive; rather, we regret this analytical deficit because we have thereby lost a priceless opportunity to gain crucial information about the relationship between fear and symptom-formation. In the light of everything we know about the more straightforward neuroses of everyday life, it seems highly unlikely that any neurosis could be brought about solely by the presence of objective danger, without any involvement on the part of the deeper unconscious layers of the psychic apparatus. However, there is nothing within the unconscious capable of giving substance to our notion of the extinction of life. Whereas castration is rendered imaginable to us by our daily experience of being separated from the contents of our bowels, and by the loss of the maternal breast that we experience when weaned, experiences akin to death have never happened to us, or else – like fainting-fits – have left no identifiable trace. I therefore hold to the supposition that the fear of death has to be understood as an analogue of the fear of castration, and that the situation to which the ego reacts is that of being abandoned by its guardian the super-ego – that is, by the forces that rule our destiny – and hence deprived for ever of the shield safeguarding it from dangers all and sundry.40 It also needs to be borne in mind that in experiences leading to traumatic neurosis the barrier that normally provides protection against external stimuli is breached, and excessive quanta of excitation descend upon the psychic apparatus; this means that here we encounter a second possibility, namely that fear is not only signalled as an affect, but is also created anew out of the economic conditions of the situation.

  In asserting just now that the ego becomes habituated to the notion of castration by regularly experiencing object-loss, we have arrived at a new concept of fear. Whereas hitherto we have regarded fear as a signal of affect indicating danger, it now appears to us – since it so often involves the danger of castration – to constitute reaction to a loss, a separation. While numerous considerations might instantly seem to gainsay this conclusion, we none the less cannot help being struck by a most remarkable similarity: birth constitutes the first experience of fear, at any rate for human beings, and in objective terms signifies separation from the mother; it might therefore be likened to castration of the mother (based on the equation ‘child = penis’). Now it would be highly gratifying if fear were subsequently to be repeated as a symbol of separation every time an actual separation occurred; unfortunately, however, we cannot build an argument on this similarity given the fact that birth is of course not subjectively experienced as separation from the mother, the thoroughly narcissistic foetus being altogether unaware of the mother as object. A further objection can also be raised namely that the affective reactions to separation are well known to us, and that we experience them as pain and sorrow, not as fear. At the same time, however, we are mindful that in our discussion of sorrow we were at a loss to understand why it should be so painful.41

  VIII

  It is time for us to pause for thought. What we are looking for is plainly some decisive insight that will reveal to us the whole nature of fear, a clear perspective on the problem that will neatly separate truth from error. But that is difficult to achieve; fear is not easy to pin down. So far we have managed to come up with nothing but contradictory possibilities, none of which could be preferred over any other except on the basis of prejudice. I suggest that we now take a different approach: let us impartially rehearse all our arguments on the subject of fear, and in so doing abandon any expectation of arriving at an all-embracing new synthesis.

  Fear, then, is first and foremost something that is felt. We call it a ‘state of affect’, even though we don't actually know what an affect is. This feeling is blatantly unpleasurable in nature, but that is not a sufficient description of it, for not every form of unpleasure may be termed fear. There are other feelings of an unpleasurable kind, such as tension, pain and sorrow – and fear must accordingly have other characteristics besides this quality of unpleasure. Question: will we ever succeed in understanding the differences between these various unpleasurable states of affect?

  None the less there is one thing we can deduce from the feeling of fear: its quality of unpleasure appears to have a distinct character of its own (although probable, this is difficult to prove, as there would seem to be nothing very obvious to go on). But quite apart from this special quality of unpleasure that is so difficult to isolate, we also perceive more specific physical sensations when fear is present, which we connect with specific organs. Since the physiology of fear is not our concern here, it will be sufficient for us to mention just one or two representative examples of these sensations, the most frequent and most obvious being those involving the respiratory organs and the heart. They afford us proof that motor innervations, in other words release processes, play
a part in the overall phenomenon of fear. Our analysis of the state of fear thus reveals the following features: 1) a specific quality of unpleasure; 2) actions involving release; 3) perception of these actions.

  Straightaway, points 2 and 3 show us a distinct difference between fear and other states, e.g. those of sorrow and pain. In these latter cases, any motor manifestations that happen to occur are not directly integral to the phenomenon itself; where they exist, they clearly stand out as being not constituent elements of the whole process, but consequences of it or reactions to it. Fear is thus a particular state of unpleasure, with release actions that follow specific pathways. In accordance with our general philosophy in these matters we are inclined to think that the root of fear lies in an increase in excitation that on the one hand generates unpleasure of a particular kind, and on the other hand relieves it by means of the above-mentioned release processes. This purely physiological summary is scarcely going to satisfy us, however: we are tempted to suppose that a historical factor is at work here, linking the sensations and innervations of fear firmly together. Our supposition, in other words, is that the state of fear constitutes the reproduction of a prior experience containing the necessary conditions for such an increase in stimulus and for release via specific pathways, and that this is how the unpleasure of fear acquires its specific character. In the case of human beings, birth suggests itself to us as being just such a paradigmatic experience, and we are accordingly disposed to regard the state of fear as a reproduction of the trauma of birth.42

  In saying this we have adduced nothing that might serve to grant fear a special place amongst the states of affect. We take the view that the other affects are also reproductions of ancient, perhaps pre-individual events of life-and-death importance, and may be regarded as universal, typical, inborn hysterical attacks, in contradistinction to the attacks characteristic of hysterical neurosis, which develop at a much later stage and on an individual basis, and whose genesis and significance as memory-symbols have become clear to us thanks to psychoanalysis. It would be highly desirable, of course, if we could substantiate this hypothesis in respect of a whole series of other affects – but at present we are still very far from achieving this.

  Our contention that fear may be traced back to the experience of birth obliges us to deal with a number of self-evident objections. Fear is a reaction that is probably common to all organisms, at any rate all the higher ones, but birth is experienced by mammals alone, and it is questionable whether birth is traumatic for all of the latter. Fear thus exists without the paradigm of birth. But this objection ignores the boundaries dividing psychology from biology. Precisely because fear – as the reaction to danger – fulfils a biologically indispensable function, it may well be differently constituted in different organisms. Furthermore, we do not know whether it has the same repertoire of sensations and innervations in organisms far removed from human beings as it does in humans themselves. This accordingly presents no obstacle to the supposition that in human beings the paradigm for fear is the birth process.

  If the structure and origins of fear are indeed thus, then the question immediately arises: what is its function, and in what circumstances is it reproduced? The answer seems both obvious and compelling: fear arises in the first place as a reaction to a danger situation, and is then regularly reproduced whenever a situation of the same kind recurs.

  This calls for further comment, however. The innervations involved in the original state of fear were probably also senseful and purposive, just as the muscle actions in an initial hysterical attack are. So if we want to understand the hysterical attack, then of course we simply need to look for a situation in which the relevant movements formed part of some apt and necessary set of actions. We accordingly find that during birth the innervation process, by being directed at the respiratory organs, probably served to prepare the lungs for action, while the acceleration of the heart-beat was intended to counteract any potential poisoning of the bloodstream. Needless to say, this purposive element does not come into the picture whenever the original fear-state is subsequently reproduced as an affect, just as it is also conspicuously absent whenever the initial hysterical attack is repeated. Therefore if an individual encounters a new danger situation, it can easily be counter-purposive for him to respond by reproducing the original fear-state – namely the reaction to a previous danger – instead of reacting in a manner appropriate to the current one. The purposive element re-emerges, however, if the danger situation is perceived as impending rather than present, and is duly signalled by an attack of fear. More appropriate responses can then instantly take the place of this fear. It thus becomes clear at once that fear can emerge in two quite different ways: one that is counter-purposive, where there is a new situation of actual danger, and one that is purposive, namely one aimed at signalling an impending danger and preventing it from becoming a reality.

  But what constitutes a ‘danger’? In the birth process there is an objective danger to life; we know what that means in terms of physical reality – but in psychological terms it means nothing at all. The danger posed by birth does not at the time impinge in any way on the psyche. We surely cannot attribute to the foetus any knowledge whatsoever that the process might end in the extinction of its own life. All that the foetus is capable of registering is a massive upset in the economy of its narcissistic libido. Large quantities of excitation come surging into it, producing unpleasurable sensations of a new kind; various organs peremptorily achieve increased cathexis, which amounts to a sort of prelude to the process of object-cathexis that is soon to commence. But which element in all of this will be put to use as a marker indicative of a ‘danger situation’?

  Unfortunately we know far too little about the psychic make-up of the newborn to be able to answer this question directly. I cannot even vouch for the validity of the description I have just offered. It is easy enough to assert that the newborn child will repeat the affect of fear in all subsequent situations that remind it of its birth; the crucial issue, however, is the question as to what serves to remind it, and what it is reminded of.

  There is really only one course open to us, and that is to study the circumstances in which babies and somewhat older children show a readiness to generate fear. In his book Das Trauma der Geburt [The Trauma of Birth] (1924), Rank tried very hard to prove a link between the earliest phobias in children and their experience of the birth process but in my view he did not succeed. One can offer two objections to his argument. The first is that it rests on the supposition that in the course of their birth children receive specific sense impressions, particularly visual ones, any recurrence of which can call forth the memory of the birth trauma and therewith the associated fear reaction. This assumption is totally unproven and highly improbable: it is simply not credible that children retain any sensations from the birth process other than tactile ones and ones of a very generalized nature. Thus if a child subsequently exhibits fear of small animals that disappear down holes or emerge from them, this is due in Rank's view to its perception of an analogy, but one of which it necessarily remains unaware. The second objection is that in his analysis of these subsequent fear situations Rank picks and chooses according to the needs of his own argument in deciding which is the operative factor in any given case – the child's memory of its blissful intra-uterine existence, or its memory of the traumatic disruption of that existence. To do this is to open the floodgates to sheer arbitrariness of interpretation. Some instances of childhood fear are flatly unamenable to the application of Rank's principle. If a child is put on its own in a dark place, then on Rank's view we might expect it to welcome this restoration of its situation in the womb, whereas it is precisely in such circumstances that it reacts with fear and when we hear this being attributed to the child's memory of the disruption of uterine bliss by the birth process, we can no longer fail to recognize the factitiousness of this would-be explanation.

  I have to conclude that the phobias of very early childhood ca
nnot be directly attributed to the experience of birth, and indeed have so far defied all attempts to account for them. A certain degree of apprehension43 is unmistakably evident in babies. Rather than - say - being at its strongest just after birth and then slowly abating, it only manifests itself later on, when the psyche begins to develop, and carries on throughout a certain period of the individual's childhood. If early phobias of this kind persist beyond that period, then we tend to suspect the presence of a neurotic disorder, even though we have no idea what its relationship might be to the later and clear-cut neuroses of childhood.

  The manifestation of fear in young children is readily comprehensible to us only in very few circumstances, and these are the ones we need to focus on – as when the child finds itself alone, or in darkness, or faced with a stranger instead of the person intimately familiar to it (that is, its mother). These three circumstances boil down to a single determining factor: distress at the absence of the loved (and longed for) person. Once we appreciate this, however, we are well on the way to achieving a real understanding of the phenomenon of fear, and to resolving the apparent contradictions connected to it.