Perhaps the most frightening and disturbing aspect of anger, even to therapists, is destructiveness. It seems to fall outside the idea of a healthy anger, that is, an anger which, when expressed, frees up the individual and allows access to feelings and energy which have been inhibited, and to establish more honest relationships with others. In working therapeutically with anger, great emphasis is placed upon the safety of the therapeutic environment: there may be big soft cushions to bash, and ground rules will be in place about violence, so there can be plenty of noise and heat – but no smashing things to bits, except in fantasy. In a sense, all these very necessary arrangements which make the work with anger possible, also deny the reality of destructive feelings which actually do want to hurt, mess, break, or distress somebody else. Of course, this reality is not denied in words, nor in the models for psychotherapeutic working, but it is – necessarily – denied in fact. It is made very clear that these destructive aspects of our anger are fantasies – our sadistic fantasies. It seems to go without saying that we are not really like that! Or if we are, it was when we were very, very young and could yet not separate the reality of a self from mother.
In the light of this seeming denial within the practice of psychotherapy. I was especially interested to read the following description of a piece of work in Winnicott:
“For my first child analysis I chose a delinquent. This boy attended regularly for a year and the treatment stopped because of the disturbance that the boy caused in the clinic. I could say that the analysis was going well, and its cessation caused distress both to the boy and to myself in spite of the fact that on several occasions I got badly bitten on the buttocks …”
I put this phrase in italics because it so struck me when I read it. There was something most extraordinary in Winnicott’s being bitten – not just once, which might slip past his guard, but ‘on several occasions’. It contrasted very sharp ly with a passage in Melanie Klein about the importance of preventing actual damage to the therapist from taking place:
”I have usually been able to convey to the child that I would not tolerate physical attacks on myself. This attitude not only protects the psychoanalyst but is of importance for the analysis as well. For such assaults, if not kept within bounds, are apt to stir up excessive guilt and persecutory anxiety in the child and therefore add to the difficulties of the treatment. I have sometimes been asked by what method I prevented physical attacks, and I think the answer is that I was very careful not to inhibit the child’s aggressive phantasies; in fact, he was given the opportunity to act them out in other ways, including verbal attacks on myself. The more I was able to interpret in time the motives of the child’s aggressiveness, the more the situation could be kept under control. But with some psychotic children it has occasionally been difficult to protect myself against their aggressiveness.” 
Plainly Klein’s point about not stirring up excessive guilt and not exciting persecutory anxiety in the client is vital, and there is I believe no disagreement about this between herself and Winnicott, in the ordinary course of things. Winnicott describes over and over again in his work how important it is for first the mother and then the analyst to survive these aggressive attacks which are linked to the earliest of feeding experiences for the child.
Klein’s argument centres around the developmental need for the child (or client) to overcome the need to attack the loved object (eg the breast); she proposes the model which I feel is now generally accepted, of the angry, destructive (aggressive) feeding/feeling leading to fear of retaliation and feelings of guilt, the so-called ‘depressive position’, which then resolves into a more creative relationship with the object/person:
“I have found that the child’s attitude towards a toy he has damaged is very revealing. He often puts aside such a toy, representing for instance a sibling or a parent, and ignores it for a time. This indicates dislike of the damaged object, due to the persecutory fear that the attacked person (represented by the toy) has become retaliatory and dangerous. The sense of persecution may be so strong that it covers up feelings of guilt and depression which are also aroused by the damage done. Or guilt and depression may be so strong that they lead to a reinforcing of persecutory feelings. However, one day the child may search in his drawer for the damaged toy. This suggests that by then we have been able to analyse some important defences, thus diminishing persecutory feelings and making it possible for the sense of guilt and the urge to make reparation to be experienced.”
Winnicott is well aware of this developmental structure:
“It was Mrs Klein who took up the destructiveness that there is in human nature and started to make sense of it in psychoanalytic terms… Ideas of destroying an object turn up, a sense of guilt appears, and constructive work results. But what is found is very much more complex…” 
It is interesting that he is much less concerned with the client’s ‘fear of retaliation’ than Klein, and I might almost suggest that he sees the whole relationship with the client in a very different way. Klein is plainly excellent on boundaries, because she sees that they can enable the client/child to experience in fantasy or play the dangerous and ambivalent feelings which he is developmentally driven to overcome. Her rules reinforce that drive, almost in the way a moral code might be considered to reinforce particular social values. In a truly maternal way, it seems to me, she is concerned to bring the child up to be a well-adjusted adult. Going through a phantasy phase of destructiveness is part of it. In short, I think she probably disapproves of it, but can tolerate it on therapeutic grounds. Perhaps that is quite similar to the way many therapists feel about destructiveness.
But Winnicott seems to take a radically different view, one which means that he doesn’t regard it as critical when his bottom is bitten by the boy – the actual bites did not make him feel the analysis was going badly. He was himself ‘distressed’ when the analysis had to be terminated because of the huge amount of disruption which the delinquent child was able to create at the clinic, not all of which was directed at Winnicott himself. Winnicott connects ‘the antisocial tendency’ and its destructiveness which he is trying to work with in this child, to deprivation:
“A child becomes a deprived child when deprived of certain essential features of home life… The antisocial tendency is characterized by an element in it which compels the environment to be important.” 
Clearly this is a very different view of the child’s destructiveness from that taken by Klein; whereas she sees a person struggling to develop into relationship and away from violent infantile phantasy, Winnicott views the destructive delinquent as involved already in a ferocious struggle to wring what he needs from the (failing) environment. Winnicott can see ‘hope’ in the child’s fierce struggle that the environment will not always fail – and he therefore places actual value on the ‘nuisance’ of the symptoms, the thieving, the messing, the violent attacks. Essentially, he views it as the child’s right to get what he needs from the environment which has failed (deprived) him and to be a thoroughly antisocial nuisance in the process. Later on he adds:
”At the basis of the antisocial tendency is a good early experience that has been lost. Surely it is an essential feature that the infant has reached to a capacity to perceive that the cause of the disaster lies in an environmental failure. Correct knowledge that the cause of the depression or disintegration is an external one, and not an internal one, is responsible for the personality distortion and for the urge to seek for a cure by new environmental provision.”
Whereas Klein’s analysis entails accepting the predominance of phantasy in the child’s developing relationship with the world, Winnicott’s view, which extends rather than contradicts Klein’s, also includes the possibility that the child may be reacting to environmental deprivation, and not just struggling with the contradictory forces of instinctive human nature inside himself. I suppose that this dichotomy could be reminiscent of the famous dichotomy in Freud’s view of his clients who claimed to have been seduced; that at first he believed them but then - realizing what a can of worms he might be opening, say his detractors – revising his opinion and say ing that these stories were fantasies which reflected the unconscious desires of the clients. It might be more thoughtful if, instead of considering this shift of view to be dichotomous, we could see it as a process of ‘over-determination’, in which both interpretations could be true and simultaneous, and not necessarily resoluble into one another. Winnicott’s choice of a delinquent boy to work with was disastrous in one sense, as he says himself – without secure environmental provision (which was not provided by the clinic), the analysis could not be carried on. But in another sense, unless he could work with such a child and even perhaps unless he was bitten by him, the point of the destructiveness, its ’nuisance value’, could not be seen.
In a slightly later essay than ‘The Antisocial Tendency’, Winnicott returns to the question of destructiveness itself. Under a title which deliberately uses three words which are standard vocabulary in Klein – ‘Aggression, Guilt and Reparation’ – he begins by drawing attention to the structure which she hypothesized and which I quoted above. His work in this essay is to draw out the ‘much more complex’ findings about the relationship between constructive and destructive forces in his clients. Originally this was a talk given to The Progressive League, and not to an audience of fellow analysts. Perhaps this different context enabled Winnicott to reach into his clinical experience in a slightly different way – less technical and more concerned with identifying aspects of aggression which are both very ordinary and everyday and also pathological. The cases he cites in this essay are not ‘delinquent’ but normal, voluntary clients. In each case he identifies a particular process which looks regressive: the client reaches to a ‘platform of generosity’ (for instance, one client hopes that the work Winnicott does with him ‘will be of value to the world’, while another acknowledges the good work of the therapist) and then switches into real, vital and honest destructiveness. He notes that this destructiveness is therefore specifically directed against a good object, in fact against the very person who was being appreciated a moment before. Winnicott goes on:
”Naturally the fact that the patient was becoming conscious of the destructiveness made possible the constructive activity… But it is the other way round that I want you to see it just now. The constructive and creative experiences were making it possible for the child to get to the experience of her destructiveness.”
For Winnicott that is an experience of authenticity, of the real self. This gives it paramount value, not remotely connected to approval. And to me, the symbol of that would perhaps be his ability to allow himself, in very exceptional circumstances, to be bitten. The real difficulty of working with this primary destructiveness, which is part of ‘primitive loving’, is highlighted. Winnicott points out how hard we find it to stand this destructiveness in ourselves, let alone in others, and so “the result is either depression or a search for relief by the discovery of destructiveness elsewhere, that is to say by the mechanism of projection.” It may be glib of me, but I couldn’t help thinking of the recent killings in Columbine High School in Denver, Colorado, where the pampered and loved children of rich middle-class parents found no ‘relief for their nasty, messy, horrible and real destructiveness which nobody (including themselves) could stand.
1. D.W. Winnicott, The Antisocial Tendency’ (1956) in Deprivation and Delinquency, ed Clare Winnicott et al, Tavistock/Routledge, 1984
2. Melanie Klein, The Psycho-analytic Play Technique’ 1955, in The Selected Melanie Klein, ed Juliet Mitchell, Penguin 1986.
3. D.W. Winnicott, ‘Aggression, Guilt and Reparation’ (1960) op cit.
4. D.W. Winnicott, ‘The Antisocial Tendency’ op cit.