The Writings of Professor Robert M. Young
Mental Space
| Contents | Preface | Acknowledgements | Chapter:1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | BibliographyChapter Four
ANALYTIC SPACE: COUNTERTRANSFERENCE
The analytic space is the mental space within which psychoanalysis and psychotherapy occur. It is a space shared between patient and therapist. In this chapter I shall characterise the two main processes which occur in the analytic space - transference and countertransference. While doing so I shall carry further my philosophical reflections on the relations between psychoanalysis and the modern world view. I wish to begin with an important distinction - that between didactic and evocative knowledge. Didactic knowledge is imparted, while evocative knowledge is elicited or brought forth. Everything with which I am concerned in this chapter comes from what may at first glance appear to be a relatively trivial technical problem in psychoanalysis: the unconscious feelings stirred up in the therapist by the patient. I wish to argue, however, that it leads to the very heart of the analytic process and, beyond that, to our conception of human nature and how we may fruitfully think about how we come to know - the theory of knowledge or epistemology. This is a considerable weight to place on the concept of countertransference, but I shall try to argue that it can bear it. My story has a nicely linear plot, taking us from the simple to the complex and on to the interactive and the dialectical. I want to start with the traditional stance of the therapist in the analytic session - that of neutrality, holding up a mirror to the patient. But in offering the image of a mirror Freud did not mean that one should not be human. He was not urging the therapist to be inanimate glass and silver nitrate; he was saying that one should not tell about oneself. There was a tendency among the early psychoanalysts to be self-revealing. The mirror was an image in the service of the rule of abstinence: speak to the patient only about himself or herself and about characters who inhabit the patient's inner world. Before turning to countertransference, we must consider transference. Freud said, 'What are transferences? They are the new editions or facsimiles of the impulses and phantasies which are aroused and made conscious during the progress of the analysis; but they have this peculiarity, which is characteristic for their species, that they replace some earlier person by the person of the physician. To put it another way: a whole series of psychological experiences are revived, not as belonging to the past, but as applying to the person of the physician at the present moment' (Freud, 1905, p. 116). In a way, then, transference is a mistake, and, as Freud was to learn, the analytic process is based on this mistake - that the patient experiences the analyst as someone else. The patient, according to Racker, '"displaces" or "transfers" infantile and internal conflicts to current situations and objects which are out of place and inappropriate' (Racker, 1968, p. 15). A transference interpretation is basically pointing this out to the patient. Fenichel said, 'The process that requires demonstrating to patients the same thing again and again at different times or in various connections, is called, following Freud, "working through... Again and again the patient must in analysis re-experience 'there too' and 'there again'"'(Fenichel, 1941, pp. 78-80, quoted in Searles, 1978-9, p. 176). Fenichel's view of the task in working through is this: 'An analyst giving a transference interpretation says, schematically: "It is not I toward whom your feelings are directed; you really mean your father." But there are many patients who know about transference and defend themselves against emerging emotional excitement by referring to its transference nature. In such instances the "reversed transference interpretation" is necessary; "You are aroused at this moment not about your father but about me"' (Fenichel, 1945, p. 522, quoted in Searles, 1978-9, p. 176). All of this seems relatively straightforward, even mechanical. The concept of transference became increasingly enriched, however, so that it was eventually perceived that what is transferred is the total situation, a relationship or objects in a context, and not merely an individual. Moreover, the objects transferred are not external ones but internal objects (Klein, 1952; Joseph, 1983). As a consequence, ideas of the transference became broadened into a wider context and deepened into the object relations of the inner world. Turning to countertransference, you may think of it as an arcane topic; it is certainly an unwieldy word, one which conjures up the most abstract of latter-day metapsychological conceptualisation's. In fact, it arose very early and was very immediate: it is why Freud's first collaborator, Joseph Breuer, gave up. He ran away from Anna O because she aroused him. If transference is projection, countertransference is projective identification - something elicited by the patient in the therapist: evocative knowledge. Anna O elicited in Breuer a sexual excitement which he found unacceptable and was unbearable to himself and his wife, so he abandoned the work (Gay, 1988, pp. 63-9). For Freud the transference went from being an annoying interference to an instrument of great value to the main battlefield of the analysis. An analogous story can be told about the countertransference, but it is a story with profound implications. Now, to define countertransference. Freud rarely discussed the topic; he saw countertransference as the patient's influence on the analyst's unconscious. He said that no analyst could go farther than he or she had progressed in their own analysis, so the analyst's analysis was all-important. He first mentions the concept in 1910: 'We have become aware of the "countertransference", which arises in [the analyst] as a result of the patient's influence on his unconscious feelings, and we are almost inclined to insist that he shall recognise this countertransference in himself and overcome it. Now that a considerable number of people are practising psychoanalysis and exchanging their observations with one another, we have noticed that no psychoanalyst goes further than his own complexes and internal resistances permit; and we consequently require that he shall begin his activity with a self-analysis and continually carry it deeper while he is making his own observations on his patients. Anyone who fails to produce results in a self-analysis of this kind may at once give up any idea of being able to treat patients by analysis' (Freud, 1910, pp. 144-5). Later, of course, Freud also insisted on a training analysis conducted by a suitable analyst. It is often thought that Freud held a limited view of countertransference, and he certainly had little to say on the topic. Even so, I would argue that the following quotation, properly contemplated and making due allowance for the technological imagery of his day, contains all we need to know: 'To put it into a formula: [the analyst] must turn his own unconscious like a receptive organ toward the transmitting unconscious of the patient. He must adjust himself to the patient as a telephone receiver is adjusted to the transmitting microphone. Just as the receiver converts back into sound waves the electric oscillations in the telephone line which were set up by sound waves, so the doctor's unconscious is able, from the derivatives of the unconscious which are communicated to him, to reconstruct that unconscious, which has determined the patient's free associations' (Freud, 1912, pp. 115-6). This quotation takes us much deeper; it is much more resonant and in touch with primitive processes than the previous one, even though it was penned only two years later. With respect to the subsequent history of ideas about countertransference, Laplanche and Pontalis trace three successive positions on the subject, which I shall characterise in my own terms: (1) Get rid of it by means of one's own analysis, and concentrate on the patient's transference. (2) Exploit it in a controlled way, using the therapist's unconscious as an instrument for fathoming the patient's unconscious. (3) Go with it, treating the resonances from unconscious to unconscious as the only authentically psychoanalytic form of communication (Laplanche and Pontalis, 1983, pp. 92-3). Before embarking on that history, I want to say something about projective identification, since we will need this concept as we go along. I shall discuss this concept at length in chapters six and seven, but for the present I want to point out that countertransference is an aspect of projective identification. In the countertransference relationship, the patient puts something into the therapist which the therapist experiences as his or her own. That's not a bad definition of one of the forms of projective identification, in which the patient splits off an unacceptable or undesirable (or otherwise uncontainable) part of the self and puts it into another person. That person must have, if only to a very small degree, the potential to identify with and express that feeling. It rises up from the general repertoire of that person's potential feelings and gets exaggerated and expressed. The projector can then feel: 'It's not me; it's him', while the process of identification in the recipient may yield a bewildering feeling, reaction or act (Hinshelwood, 1991, pp. 179-208). In an attentive therapist, interrogating the countertransference leads to a fruitful interpretation. We can now see this congruence in action in a passage where Freud is quite explicit about a concept usually seen as originating with Klein. He is talking about the projections of jealous and persecuted paranoiacs, of whom it is said that they project onto others that which they do not wish to recognise in themselves. He continues, 'Certainly they do this; but they do not project it into the blue, so to speak, where there is nothing of the sort already. They let themselves be guided by their knowledge of the unconscious, and displace to the unconscious minds of others the attention which they have withdrawn from their own. Our jealous husband perceived his wife's unfaithfulness instead of his own; by becoming conscious of hers and magnifying it enormously he succeeded in keeping his own unconscious' (Freud, 1922, p. 226). Fenichel writes in the same vein: 'It is interesting to note that the hatred is never projected at random but is felt usually in connection with something that has a basis in reality. Patients with persecutory ideas are extremely sensitive to criticism and use the awareness of actual insignificant criticisms as the reality basis for their delusions. This basis has, of course, to be extremely exaggerated and distorted in order to be made available for this purpose... The paranoid individual is particularly sensitised to perceive the unconscious of others, where such perceptions can be utilised to rationalise his tendency toward projection' (Fenichel, 1945, p. 428, quoted in Searles, 1978-9, p. 177). Anyone who has ever worked in a mental hospital will recognise this instantly. Paranoids are geniuses at getting under the skin and ferreting out latent tendencies in others, especially staff. But, of course, this is only a relatively apparent exaggeration of the norm, as any of us can attest from our experiences of how much can get tangled up in a moment during telephone calls with prospective patients, and in relations with lovers or one's own children. Projection, introjection, exaggeration, reprojection - these are norms of social interaction. It is all a matter of degree. Nevertheless, as in the rest of life, everything can depend on matters of degree. Projective identification is a normal mechanism, but when employed excessively or virulently, it lies at the heart of paranoid processes, racism, narcissism, and innumerable other pathological conditions (see below, ch. 7). When employed excessively, it is also central to pathological conformism and ruthless ambition and acts as a defence against schizophrenic breakdown (Meltzer, 1992). Relinquishing its excessive use is essential to becoming a decent person. There is a rich history of ideas of countertransference, some phases of which I will not spell out except to list familiar names, since the relevant papers are competently reviewed in a collection edited by Edmund Slakter (1987): Stern (1924); Deutsch (1926); Glover (1927); Sharpe (1930); Hann-Kende (1933); W. Reich (1933); Strachey (1934); Low (1935). There are other overviews, for example, by Kohon (1986) and Orr (1988), a collection of Essential Papers on Countertransference (Wolstein, 1988) and a growing number of monographs and papers, which, as I write, has brought forth the inevitable Beyond Countertransference (Natterson, 1991; cf. Alexandris and Vaslamatzis, 1993). I want to begin my own story of recent work with a paper by Winnicott, startlingly entitled 'Hate in the Counter-transference' (1947). I am re-entering the history of ideas at the point where the transition is occurring between countertransference as 'that which is to be got rid of', to 'that which is to be made something of' or exploited. Winnicott said that to feel hate, when it has been projected into you and evoked by the patient, is part of the therapist's proper responsiveness. When the patient seeks the therapist's hate, the therapist must be able to make contact with it, to bear it without retaliating and to contain it; otherwise the analysis fails. Two years after Winnicott made this point, Paula Heimann took up the topic and began with the traditional view: 'I have been struck by the widespread belief amongst candidates that the countertransference is nothing but a source of trouble' (Heimann, 1949-50, p. 73). She takes a contrary position: 'My thesis is that the analyst's emotional response to his patient within the analytic situation represents one of the most important tools for his work. The analyst's countertransference is an instrument of research into the patient's unconscious' (p. 74). 'Our basic assumption is that the analyst's unconscious understands that of his patient' (p. 75). She claims that 'the analyst's countertransference is not only part and parcel of the analytic relationship, but it is the patient's creation, it is a part of the patient's personality' (p. 77). Winnicott was writing about psychotics and borderline personalities; Heimann extended the point to include work with all patients. However, when she returns to the topic a decade later, she reflects at length on how to handle countertransference material and takes up a cautious position, eschewing self-revelation. But in her conclusion, she reverts to the early view of Freud. In so doing, it seems to me, she fails to see the larger potential of the concept. She says, in her last paragraph, 'In conclusion, Freud's injunction that the countertransference must be recognised and overcome is as valid today as it was fifty years ago. When it occurs, it must be turned to some useful purpose. Continued self-analysis and self-training will help to decrease incidents of countertransference' (p. 160). For her, countertransference was still a hot potato - something to learn from when it occurs but preferably to be got rid of. Putting the point at its starkest, we will see below that to 'reduce the incidents of countertransference' is no longer a goal among many therapists. Another writer in this period, Roger Money-Kyrle, takes up a position that acknowledges that countertransference can be both useful and a serious impediment. As his title implies, countertransference was coming to be seen as a normal part of the analytic process: 'Normal Counter-transference and Some of its Deviations' (1956). He draws our attention to the fine texture of the process, whereby something is put into the analyst, re-projected in an interpretation and then taken in by the patient. If the receiving parts of the analyst are intact, what gets reprojected is likely to be helpful, but if the projection falls afoul of unresolved issues in the analyst's unconscious, a mess can be created. Everyone would be happy if 'normal countertransference' was all there was. 'Unfortunately, it is normal only in the sense of being an ideal. It depends for its continuity on the analyst's continuous understanding. But he is not omniscient. In particular, his understanding fails whenever the patient corresponds too closely with some aspect of himself which he has not yet learnt to understand' (Money-Kyrle, 1956, p. 361). In such cases, what ensues is a reprojection of something which is not helpful, not congruent with the patient's needs. The analyst may recover then or in the next session, but the patient is not simply waiting for him to get it right. He or she may well have moved on and be relating to the analyst as a damaged object as a result of the distorted interpretation. The result will take some sorting out. What is attractive about Money-Kyrle's reflections is the light they shed on the process - the minute details - of these interrelations. A sense that countertransference was, in the 1950s and 1960s, still basically seen as something to be wary of has been driven home for me by a revealing incident in the publishing history of the work of Margaret Little. In 1950, she wrote a paper on 'Countertransference and the Patient's Response to It'. It was duly reprinted in a collection of her writings, Transference Neurosis and Transference Psychosis in l981. But when the book came out in a paperback edition five years later, she added an intriguing footnote: 'I read this paper in 1950, almost exactly three years after my analyst's death. I could not then give the true account, but disguised it as best I could. (That version has rightly been questioned.) I have given this account in chapter 21 [a dialogue between her and Robert Langs], and also in my paper 'Winnicott Working in Areas where Psychotic Anxieties Prevail: A Personal Record' in Free Associations (1985) 3:9-42' (Little, 1986, p. 33n). The last-mentioned article had been submitted to the International Journal of Psycho-Analysis and rejected with the suggestion that it should be put in a vault until a suitable date well in the future. One of the referees told me that he had urged its sequestration because it revealed that a practising training analyst continued to work while she was very ill, that her analyst, Donald Winnicott, permitted this and had broken various boundaries with respect to the analytic frame, and that knowledge of this would be too distressing to her former patients, including some analysts still in practice whose training analyst she had been. When, after the most careful consideration, the article was published in Free Associations, Margaret Little received a gracious note from the Editor of the International Journal, saying how pleased he was that the article had been sufficiently modified so as to make it suitable for publication. Yet not a word had been altered from her original submission. My justification for telling this rather gossipy tale about the context of publishing these pieces is that it nicely illustrates that when we are dealing with the complex, subtle and primitive processes which occur in transference and countertransference, we are moving about on a very perilous terrain, one which puts at risk the boundaries of the analytic frame, the maintenance of which are essential to the successful conduct of psychoanalytic work. One view of the analytic frame is that it contains the psychotic parts of the relationship, thus allowing the work to proceed (Blejer, 1966). Taking full account of countertransference - and not confining attention to transference, keeping the therapist's unconscious processes out of the question - means that the analytic frame is constantly at risk of being violated. Many feel that the relations between Margaret Little and D. W. Winnicott went beyond the appropriate boundaries. I believe the debate about the relationship between countertransference and the analytic frame has an important bearing on how we do analytic work and how we conceptualise the maintenance of analytic integrity. Margaret Little has stressed in these and subsequent writings (1957, 1987, l989, 1990) that the relations between analyst and patient are much more complex and interactive, both consciously and unconsciously, than is acknowledged by orthodox models of the therapeutic process. A strong reaction against her views came from an orthodox Freudian, Annie Reich (1951), who argued that countertransference was not a therapeutic tool and was not useful for understanding or communicating with the patient. Slakter comments that Annie Reich was defending a conception of psychoanalysis as the analysis of the ego's defences and rejecting 'the seductive, regressive pull of id psychology' (Slakter, 1987, p. 23). At this point I want to interrupt my narrative of the history of countertransference and to broaden the terms of reference to revert to the philosophical issues raised in chapter three about the ways we picture the mind. I suggest that in the history of psychoanalysis there has been a subtle tension between 'picture thinking', on the one hand, and personal, evocative, story-telling accounts, on the other. We've seen that these two ways of representing things are intermingled in two of the classic texts, ''Project for a Scientific Psychology' (1895) and The Interpretation of Dreams (1900), and one can almost feel Freud's relief when he abandons the imagery of the telescope and the diagram and replaces them with stories. After chapter seven of The Interpretation of Dreams spatial representations and pictorial diagrams are rare, e.g., the rendering of the structural point of view in The Ego and the Id (Freud, p. 24) and the oval diagram in Lecture XXXI of The New Introductory Lectures (Freud, 1933, p. 78), and when they do appear, they are rather more metaphorical than truly spatial. As we have seen, in the traditional pictorial approach the knowing subject is at one end of an investigative instrument - typically a telescope or microscope. What is essential about this way of representing the mind and the process of knowing is the spatial gap. The subject is at one end, while the object is at the other end or 'out there'. The subject is the knower; the object is to be known. The object is open to scrutiny, and the subject is not. I now want to tell a story in which this representation of the problem of knowing within and between people is increasingly complicated, starting with the picture-thinking view and moving on to an interactive and then to a dialectical one. I am a subject. You are an object. I am here; my essence is in here. You are there - out there, across a physical and epistemological space. I can infer that you are also a subject - by eye contact, by tone of voice, by analogy to my own experience and by other cues - but you easily revert to being experienced as an object, and I may easily lapse into treating you as one - as someone who does not touch me: alien. You may sense this and be alienated from me. But the situation can be seen as much, much more complicated. I may experience myself as alienated from myself - as a thing, as a bug, as ugly, as dead. This is of the essence of the concept of alienation, where a person or a worker can experience himself or herself as alienated from the product, the means or instruments of production, from fellow workers and from one's own humanity (or 'species being'). Alienation is the subjective moment of the objective condition of exploitation at work, just as an analogous estrangement can occur in bad human relationships or in psychotic moments or states. Rather than experience yourself as dead, you may put that feeling into me by projection, and if I take up the projection (unconsciously) and display it, a successful projective identification is in being between us. There are other forms of projective identification which do not depend on what I feel or display, so that you can be in projective identification with me in other ways, depending on your inner state and mine. As we enrich the model, mental and interpersonal space lose the quality of a picture with simple locations. There are no simple spaces in this enriched account, one which can be called interactive, phenomenological or dialectical, depending on how mutually constitutive the relationships are conceived as being. In an interaction things get batted back and forth. In a phenomenological description you are no longer an 'it' but a 'thou', a person with whom I can identify and empathise, as I am for you. (It could be argued that the concept of 'thou' only makes sense in the light of at least a tacit understanding of projective identification.) In a dialectical account, there are many layers and reverberations. I am here and there at once, as you are. You are in me. I can expel a part of myself. You can take it in and re-expel it, and I can experience it as you, while, in another part of my mind, knowing where that part came from. (It should be obvious that I am not listing the permutations systematically. I only want to make the point that interpersonal relationships are much richer and multi-layered than the subject-object 'picture' account allows.) Moving on to related expressions: I enter you. You withdraw. Or you may contain my distress. I push through your defences. We become one and then separate but feel love, hatred or ambivalence or frequent oscillations among these at many levels: a relationship. If we move on further to part-object relationships and on still further to tenderness, envy, spite and reparation, it soon becomes obvious that the simple subject-object model is a reductio ad absurdum of human relationships. Returning to my main theme, I suggest that the history of ideas of countertransference is a progressive closing of the spatial gap between therapist and patient. It is a turning away from pictorial models toward story-telling ones, in which knowledge is not seeing but evocation. I offer two longish examples to drive my point home. The first is from Tom Main's essay on 'Some Psychodynamics of Large Groups': 'Although projective processes are primitive attempts to relieve internal pains by externalising them, assigning or requiring another to contain aspects of the self, the price can be high: for the self is left not only less aware of its whole but, in the case of projective identification, is deplenished [sic] by the projective loss of important aspects of itself. Massive projective identification of - for instance - feared aggressive parts of the self leaves the remaining self felt only to be weak and unaggressive. Thereafter, the weakened individual will remain in terror about being overwhelmed by frightening aggressive strength, but this will now be felt only as belonging to the other. Depending on the range of this projective fantasy the results will vary from terrified flight, appeasement, wariness and specific anxieties about the other, even psychotic delusions about his intentions. 'The above instance concerns only the projector's side of the projective relationship: but projective processes often have a further significance. What about the person on the receiving end of the projection? In simple projection (a mental mechanism) the receiver may notice that he is not being treated as himself but as an aggressive other. In projective identification (an unconscious fantasy) this other may find himself forced by the projector actually to feel his own projected aggressive qualities and impulses which are otherwise alien to him. He will feel strange and uncomfortable and may resent what is happening, but in the face of the projector's weakness and cowardice it may be doubly difficult to resist the feelings of superiority and aggressive power steadily forced into him. Such disturbances affect all pair relationships more or less. A wife, for instance, may force her husband to own feared and unwanted dominating aspects of herself and will then fear and respect him. He in turn may come to feel aggressive and dominating toward her, not only because of his own resources but because of hers, which are forced into him. But more: for reasons of his own he may despise and disown certain timid aspects of his personality and by projective identification force these into his wife and despise her accordingly. She may thus be left not only with timid unaggressive parts of herself but having in addition to contain his. Certain pairs come to live in such locked systems, dominated by mutual projective fantasies with each not truly married to a person, but rather to unwanted, split off and projected parts of themselves. Both the husband, dominant and cruel, and the wife, stupidly timid and respectful, may be miserably unhappy with themselves and with each other, yet such marriages although turbulent are stable, because each partner needs the other for pathologically narcissistic purposes. Forcible projective processes, and especially projective identification, are thus more than an individual matter: they are object-related, and the other will always be affected more or less' (Main, 1975, pp. 100-01). This is an excellent exposition of some of the complexities of unconscious processes at work in everyday life. I trust that the analogy to transference and countertransference is obvious. In case it is not, I want to follow this example with a lovely account of the power of countertransference and the use that can be made of it in a clinical setting. I shall excerpt the relevant passage from a dense and illuminating account of a clinical case of Margaret Rustin's, which, for reasons of confidentiality, I do not wish to quote in detail. What is of interest in the present context comes out very clearly in the therapist's reflections, which is the point at which I shall begin quoting: 'There is much to explore in these associations, but I now want to add an important fact about this session which I was not able to make use of at the time. I myself was having two experiences in addition to the conversation I have reported. I was struggling with a frustrating conviction that I could not properly get hold of the transference situation in the session... Much more uncomfortable than this intellectual frustration was a state of irritable anxiety which was building up, particularly focused on an urgent desire to suck or bite my fingers. Trying to understand these feelings and impulses is the process required of the therapist to work through the countertransference. I am here using the term countertransference to refer not to the neurotic response of analyst to patient, but to the broader current conception of countertransference which pays close attention to the feelings stirred up in the analyst by the patient's material. 'In the following session, the meaning of this projection began to emerge.' There follows more material in which alcoholism and stealing drink loom large. Rustin continues, 'I found myself plagued by similar surges of anxious discomfort to those of last week's session...' Then more clinical material, including the alcoholic's sitting there with thumb in mouth saying how delicious the stolen drink was, which enraged and overwhelmed the patient. The account continues, 'At this moment, I felt the relief of illumination. The image of X with... thumb in... mouth linked with my impulse to suck and chew my fingers during these last two sessions, an impulse which I felt was being irresistibly projected into me. Now I knew where this was coming from. So I gathered together the threads I could now follow...' (Rustin, 1989, p. 315). The author's account makes admirably clear the central importance of the countertransference to the interpretation and its evident usefulness to the patient. There is an equally graphic - and, in this case, excruciating - account of projective identification and countertransference, in which the therapist found herself unaccustomedly and unaccountably buying and cooking squid, only to find to her chagrin and amusement, as she contemplated the cut up pieces frying in oil, that she was retaliating against a particularly murderous 'prick' of a suicidal patient (Eigner, 1986). I have chosen the foregoing examples to convey the power of the projective processes involved in countertransference. The person who has looked into these most extensively is Harold Searles. The collection of his papers on Countertransference and Related Topics (1979) contains what I regard as two profound essays on the subject, while the dialogue between Searles and Robert Langs explores in a very illuminating way the details of the interrelations between analyst and patient (Langs and Searles, 1980). In 1949 - just when Winnicott and Little were challenging the orthodoxy in Britain - Harold Searles, an American psychoanalyst of remarkably independent spirit and originality, sought to publish a paper which significantly broadened the clinical importance of the countertransference, but it was rejected by both of the psychoanalytic journals to which he submitted it. It was only published in the wake of his achievements as an analyst of schizophrenics. In it he anticipates much of his subsequent work on the real basis, in the analyst's personality, for transference phenomena, phenomena which appropriately evoke the countertransference. He summarises his article as follows: '...transference phenomena constitute projections, and... all projective manifestations - including transference reactions - have some real basis in the analyst's behavior and represent, therefore, distortions in degree only. The latter of these two suggestions implies a degree of emotional participation by the analyst which is not adequately described by the classical view of him as manifesting sympathetic interest, and nothing else, toward the patient. It has been the writer's experience that the analyst actually does feel, and manifests in various ways, a great variety of emotions during the analytic hour' (Searles, 1978-9, p. 165). In his papers on 'The Patient as Therapist to His Analyst' (1975) and 'Transitional Phenomena and Therapeutic Symbiosis' (1976), as well as in his dialogue with Langs, Searles drives home again and again the centrality, the normality, the basic and essential utility of countertransference. Langs grants its ubiquity, but - if I read him aright - still wants to master and minimise it (e.g., Langs and Searles, 1980, pp. 96-7). Searles glories in its omnipresence and rich potential. One of the affinities of his ideas is the interpersonal psychiatry of Harry Stack Sullivan, but the main source is his extensive psychoanalytic work with schizophrenics (he has worked with one woman for more than thirty years). He also makes alliances with the (independently developed) ideas of Winnicott and Little. I commend his cornucopia of examples to the reader. After reflecting upon them I would be surprised if anyone could retain the traditional view of the analyst as mirror or as a subject looking at the patient across physical or metaphorical space at an object to be known by peering, as it were, through some sort of technological instrument. (I have written at some length about Searles' insights - Young, 1992). I now have to draw breath and speak about the analytic frame. You will recall that this topic was raised in the context of discussing Margaret Little's work with Winnicott and Annie Reich's orthodox objections. Lurking around my whole account has been a whiff of scandal - a suspicion that if we get too involved with the countertransference, there's no telling where it will all end. Annie Reich feared that it would end in the id, not in the ego, where she seemed to assume that good analytic work is done. It is noteworthy, then, that the writers who have set out to broaden and deepen the concept of countertransference have been people who were exploring primitive, psychotic processes: Winnicott on borderline and schizophrenic patients, Little on her own psychotic illness and Searles, who has worked a great deal with schizophrenics and borderlines. The analytic space is bounded by the analytic frame; it is the emotional environment in which it is seemly and safe enough to conduct the therapeutic alliance. It is a container, and containment is its essence. It is made up of a set of conventions quite mundane ones, but they are under constant threat. The session starts and ends on time; confidentiality is total; you never take notes (though many do); no interruptions are permitted; no personal information about the therapist should be made available (or discernible in the paraphenalia in the room); accounts should be presented on the same day of the month; there should be due warning for breaks; other missed sessions should be minimised and announced well in advance; patients should not be touched. (My analyst shook hands at the end of each term. I emulated him until two patients missed menstrual periods after I shook hands with them.) Others would extend this list in various ways. Practically all would say that social relations between therapist and patient should be taboo, and most would say that those with ex-patients should be minimised. Sexual relations are strictly taboo. Others would make a distinction between current patients, recent ex-patients and ex-patients some years later (I would not). These aspects of the frame are important, but the essence is an attitude of abstinence and containment. If that is right, and if the essence is internalised, it is silly to make a long list of prohibitions. As the cellist Tortellier was fond of saying, one must be pure but not puriste. The frame must provide a bounded space in which it is bearable to do the work - for the patient to be safe enough to explore what is unsafe, that is, defences built up and maintained over a lifetime. Praising, blaming, encouraging - all such dimensions of normal social relations are eschewed in a strict interpretation of the analytic frame. What is on offer is interpretation, the understanding that the patient can take away and treat as food for thought. In their dialogue, one point on which Langs and Searles agree is that 'the therapist's appropriate love is expressed by maintaining the boundaries' (Langs and Searles, 1980, p. 130). Langs' view is that as soon as you modify the frame, the likelihood of a misalliance or pathological symbiosis is greatly increased (pp. 44, 127). As he says, 'frame' is 'a nonhuman term for a very human set of tenets and functions. It serves to hold and to contain, to establish boundaries and conditions of relatedness and communication' (p. 179). There are those who advocate occasional suspension of what can be seen as rigid or strict maintenance of the boundaries of the frame - what Christopher Bollas has called 'Expressive Uses of the Countertransference' (Bollas, 1987, ch. 12). Related views have been expressed by Symington (1986), Little (see above) and - perhaps most notoriously - by Nina Coltart, in a lovely essay entitled '"Slouching Toward Bethlehem"... or Thinking the Unthinkable in Psychoanalysis' (1986), in which she tells a gripping tale, the denouement of which is shouting at her patient to what appears to be good effect. Symington and Bollas tell similar stories. Indeed, Bollas tells us that on one occasion he quite deliberately and temperately said to a patient, 'You know, you are a monster' (Bollas, 1989, p. 38), and it turns out that she did know and in due course professed to be relieved that he could say so and that the relationship could survive his acknowledging the fact. Searles also owns up to revealing, at selected moments, aspects of his subjective feelings toward patients, though much more often with schizophrenics than with others (Langs and Searles, 1980, pp. 123-4). As I mentioned above, Paula Heimann counselled against such self-revelations and criticised Margaret Little for advocating them (Heimann, 1959-60, p.156). Rayner (1991) reports that the approaches of Coltart and Bollas are widespread among members of the Independent or Middle Group in the British Psycho-Analytical Society, while it is usual among Kleinians to eschew such self-revelations. Among the authors I have mentioned, however, it is common ground that such practices are open to abuse, and great care must be taken to avoid 'acting out in the countertransference' (Heimann, 1959-60, p.157). While there are important differences in the degree to which various practitioners may be willing to express their countertransference, it is my impression that there is a growing consensus that being closely attuned to it is a, if not the, basis for knowing what is going on and for making interpretations. I want to leave this issue open, while making clear that my own bias is against expressive uses of the countertransference (which is not to say that I have never done it and never will). The tendency to 'get rid of it' is certainly waning among the writers whom I am examining, while more and more is being made of it. My best experiences in supervision have resulted from the supervisor asking me what I was feeling at a particular moment - usually a moment when I felt I did not understand the material. I would go so far as to say that this has never failed to provide at least some enlightenment. Interrogating the countertransference must not be seen as seeking a fact which is available on the surface of the mind. Countertransference is as unconscious as transference is. Understanding it is an interpretive task. I want now to move to the third of Laplanche and Pontalis' renderings of the countertransference: the injunction, not merely to exploit it but to 'go with it'. The experience of countertransference is, in the first instance, apprehensible but not comprehensible. What is occurring between patient and therapist is not merely interactive; it is interpenetrative or dialectical. Much, often most, of what goes on in an analytic session is non-verbal and atmospheric, and one could not say how it is imparted. The atmosphere may be soporific, tense, comforting, assaultive, arousing. I had a patient who spoke so horribly and in such a sustained way in one session that she filled the room with her (symbolic) vomit and had to flee, since, if she opened her mouth again, she would have to take in some of her own spew. I was able to make an interpretation in these terms, because I was feeling nauseous. I had another patient who spent many sessions at the beginning of our work standing on the threshold of the room. He had panic attacks. It took me the longest time to figure out that he was imparting to me the cliff-hanging feeling that was characteristic of his attacks. One reason I could not figure it out, by the way, was because he was a training patient, and I was in a panic that I might lose him or that colleagues might see him hovering there. When I belatedly made the interpretation, fruitful work began. Another patient would come to a session, never looked at me, would speak one or two sentences and often remain silent for the rest of the period. It eventually dawned on me that she unconsciously wanted me to feel starved the way her mother had made her feel. I had been feeling that way, but it took some time to convert that sense into a thought. When I did make that interpretation, she slowly began to give more, though she remained likely to revert to sullenness and withholding. Yet underneath this mean exterior was a longing and warmth and gratitude that no camera could detect but which I came to know and to find sustaining in innumerable bleak sessions. A patient can rob one of the ability to think. Indeed, there was one in a group I conducted who was able to project her sexuality so powerfully that, on occasion, no one in the group, including me, could think of anything but her breasts and legs. As Bion said, 'Refuge is sure to be sought in mindlessnsess, sexulaization, acting out and degrees of stupor' (Bion, 1970, p.126). A paper by Irma Brenman Pick takes the normality of countertransference to its logical extreme, without a trace of seeing it as something to be got rid of. She carefully considers it as the basis of understanding throughout the session: 'Constant projecting by the patient into the analyst is the essence of analysis; every interpretation aims at a move from the paraniod/schizoid to the depressive position' (Brenman-Pick, 1985, p. 158). She makes great play of the tone, the mood and the resonances of the process: 'I think that the extent to which we succeed or fail in this task will be reflected not only in the words we choose, but in our voice and other demeanour in the act of giving an interpretation...' (p. 161). Most importantly, she emphasises the power of the projections and what they evoke countertransferentially: 'I have been trying to show that the issue is not a simple one; the patient does not just project into an analyst, but instead patients are quite skilled at projecting into particular aspects of the analyst. Thus, I have tried to show, for example, that the patient projects into the analyst's wish to be a mother, the wish to be all-knowing or to deny unpleasant knowledge, into the analyst's instinctual sadism, or into his defences against it. And above all, he projects into the analyst's guilt, or into the analyst's internal objects. 'Thus, patients touch off in the analyst deep issues and anxieties related to the need to be loved and the fear of catastrophic consequences in the face of defects, i.e., primitive persecutory or superego anxiety' (p. 161). As I see it, the approach adopted by Brenman Pick takes it as read and as normal that these powerful feelings are moving from patient to analyst and back again, through the processes of projection, evocation, reflection, interpretation and assimilation. Moving on from the more limited formulations of an earlier period in the writings of Winnicott, Heimann and even Money-Kyrle, these feelings are all normal, as it were, in the processes of analysis. More than that, as she puts it, they are the essence. Kleinians have not always taken this view of countertransference. Klein had begged Heimann not to deliver her first paper on countertransference and told Tom Hayley in the late 1950s that she thought countertransference interferes with analysis and should be the subject of lightning self-analysis (Grosskurth, 1985, p. 378). According to Spillius, 'Klein thought that such extension would open the door to claims by analysts that their own deficiencies were caused by their patients' (Spillius, 1992, p. 61). Having said this, it is important not to be too literal. about the use of the term 'countertransference'. Klein's subtle interpretations of her patients' inner worlds - especially their preverbal feelings and ideas - only make sense in the light of her ability to be resonant with their most primitive feelings, and Bion's injunction to 'abandon memory and desire' is made in the name of countertransference, whatever term we attach to the process. Indeed, it can be said that his writings are about little else. Implicit in the way I have been writing about the phenomena of countertransference is a model for knowledge - that the way we really learn is from the Other's response to what we convey. We learn by evoking and provoking. We do not learn by imparting but by re-experiencing what we have projected and has then been passed through another human being (though that person may be held in imagination). We learn by putting something out and finding out what comes back. Our relationship with the world is a phenomenological 'I-thou', not a scientistic 'I-it'. It is evocative knowledge. It may be thought that this model for knowledge is appropriate to relations between people (and perhaps pets) but that it in no way applies to knowledge of the external world. Some such distinction would seem to be common sense. However, it does not take into account recent thinking in the history, philosophy and social studies of science which argues that we project onto nature particular versions of reality and frame it according to the prevailing value systems and preoccupations - the 'world view' or weltangschauung of a period or subculture or discipline. What is true for a particular version of the world is also true for the individuals who inhabit it. Jerome Bruner (1951) has shown this with respect to children's perceptions of ordinary objects: what they see - even the size of coins - is dependent on their social location. M. L. J. Abercrombie (l960) has shown it for the anatomical and scientific perceptions of medical students: the most mundane observations only make sense in the light of unconscious forces. Donna Haraway (1989, 1991) has shown it for various fields of scientific research, particularly the social construction of primatology, providing a pedigree for our humanity (see Young, 1992a). Other versions of this position are now commonplace among students of scientific thought, e.g., Figlio, 1978, 1979, 1985, 1990; Hesse, 1980; Young, 1977, 1981, 1985, 1990, 1994, 1994a-d). In the clinical realm, Searles' first book was a major study of how schizophrenics projectively perceive and treat the external environment (1960). This provides an interesting link between views of the inner world and ideas of the outer one: both worlds are highly interpretative. Karl Figlio (1990) has generalised this view to nature as projectively experienced by nuclear disarmers and members of the peace movement. These 'friends of the earth' relate to the planet as a significant Other - a thou. Moving beyond our culture, we have seen that the history of social anthropology can be seen as a case study of my thesis, as the work of Mary Douglas exemplifies (above, ch. 2). Similarly, philosophers now argue that truth is made, not found (Rorty, 1980, 1982, 1989). Those who reflect on the philosophical implications of the belief systems of different epochs, tribes and disciplines point out that each of these social groups has its own cosmology, which articulates more or less well with that of other tribes (Horton, 1967, Bloor, 1976; Douglas, 1975). Ordinary, didactic imparting of knowledge and learning from teachers and from the media do not thereby cease to occur; they become special, limited cases of a richer model for the process of knowing. The integration of psychoanalytic theory with developments such as those outlined here is, in my opinion, an important desideratum. What I have provided here is the barest sketch, in the hope that it will make attractive the project of bringing together a social, and cultural account of ways of knowing (epistemology) with the philosophical bearings of recent developments in psychoanalysis. Aspects of the work of Winnicott, Klein, Bion, and Meltzer seem to me to lie at the centre of this project. I have in mind, in particular, the concept of transitional space (chapter 6) and the notion that all experience is mediated through primitive processes and known through the mother's body (chapters 3, 7; cf. Young, 1986a, 1989c). Returning to the psychoanalytic sphere, the weight I have put on the concept of countertransference need not be borne by that concept alone; it can be shared by ways of thinking across a broad range of disciplines. In the analytic relationship, it turns out that the real justification for the free-floating attention that is characteristic of psychoanalysis is that it makes our minds available for the patient's projections and facilitates their search for the resonances in us for what they feel. Freud said, 'He should simply listen, and not bother about whether he is keeping anything in mind' (Freud, 1912a, p. 112). Bion put it poetically in his injunction that the analyst should 'impose upon himself the positive discipline of eschewing memory and desire. I do not mean that "forgetting" is enough: what is required is a positive act of refraining from memory and desire' (Bion, 1970, p. 31). If this sounds a bit mystical, so be it. Racker shares an appropriately Oriental parable: One day an old Chinese sage lost his pearls. 'He therefore sent his eyes to search for his pearls, but his eyes did not find them. Next he sent his ears to search for the pearls, but his ears did not find them either. Then he sent his hands to search for the pearls, but neither did his hands find them. And so he sent all of his senses to search for his pearls but none found them. Finally he sent his not-search to look for his pearls. And his not-search found them '(Racker, 1968, p. 17). Once one is in this state, one is open to the patient's unconscious and to the injunction that 'Constant projecting by the patient into the analyst is the essence of analysis' (Brenman Pick, 1985, p. 158). And at the other end of the analysis lies the ability of the patient to take back the projections. This is an important criterion of improvement. Bearing projections is the whole basis of containment: the therapist can bear to take in and contain the projections, to hold them and give them back, in due course, in the form of accessible interpretations.
I am suggesting that countertransference - as an aspect of projective identification - is not only the basis for analytic work but central to the basic process in all human communication and knowing. We only know what is happening because we are moved from within by what we have taken in and responded to from our own deep feelings. The space between people is filled - when it is and to the extent it is - by what we evoke in one another.