Several times she had left home, taking the
daughters
with her but leaving the sons behind.
A-Secure-Base-Bowlby-Johnf
Our task, as I see it, is to apply our method as skilfully as we can, on the one hand believing that the area of reliable knowledge will
103/362
? ? ? thereby be expanded and, on the other, accepting that there are likely always to remain still greater areas lying beyond the scope of any existing mode of scientific enquiry.
To many of you, I am afraid, engaged in thera- peutic practice but also hoping to contribute to the advance of psychoanalytic science, the con- trasts I am drawing between the roles of practi- tioner and scientist will hardly be welcome. Yet I believe that it is only by recognizing these differ- ences and acting accordingly that the strengths of each role can be used to fullest advantage--or that any one person can occupy both of them with any hope of success. As practitioners we deal in complexity; as scientists we strive to simplify. As practitioners we use theory as a guide; as sci- entists we challenge that same theory. As practi- tioners we accept restricted modes of enquiry; as scientists we enlist every method we can.
Earlier I remarked on the need for every devel- oping science to devise new methods for obtain- ing data. The reason for this is that, however pro- ductive any one method may be, it is bound to have its limitations, whilst there is always a pro- spect that some other method may compensate for them. Thus the new method may be in no way
104/362
? ? ? superior to the old; indeed it may have great lim- itations. Its usefulness lies simply in the fact that its strengths and its limitations are different. Per- haps I can illustrate the point by reference to my own work.
When I qualified in psychoanalysis in 1937, members of the British Society were occupied in exploring the fantasy worlds of adults and chil- dren, and it was regarded as almost outside the proper interest of an analyst to give systematic attention to a person's real experiences. That was a time when Freud's famous about-turn of 1897 regarding the aetiology of hysteria1 had led to the view that anyone who places emphasis on what a child's real experiences may have been, and per- haps still are being, was regarded as pitifully nai? ve. Almost by definition it was assumed that anyone interested in the external world could not be interested in the internal world, indeed was al- most certainly running away from it.
To me as a biologist this contrast of internal with external, of organism with environment, never appealed. Furthermore, as a psychiatrist engaged in work with children and families and deeply influenced by the insights of two analytic- ally oriented social workers, I was daily
105/362
? ? ? confronted with the impact on children of the emotional problems from which their parents suffered. Here are two examples I still recall vividly. In one a father was deeply concerned about his 8-year-old son's masturbation and in reply to my enquiries explained how, whenever he caught him with his hand on his genitals, he put him under a cold tap. This led me to ask fath- er whether he himself had ever had any worry about masturbation, and he launched into a long and pathetic tale of how he had battled with the problem all his life. In another case a mother's punitive treatment of her 3-year-old's jealousy of the new baby was as quickly traced to the prob- lem she had always had with her own jealousy of a younger brother.
Observations of these kinds led me to conclude that it is just as necessary for analysts to study the way a child is really treated by his parents as it is to study the internal representations he has of them, indeed that the principal focus of our studies should be the interaction of the one with the other, of the internal with the external. Be- lieving that that would be possible only if we had far more systematic knowledge about the effects on a child of the experiences he has during his
106/362
? ? ? early years within his family, I concentrated my attention on this area. The reasons that I selected as my special field of study the removal of a young child from his home to a residential nurs- ery or hospital rather than the broader field of parent-child interaction were several. First it was an event that I believed could have serious ill ef- fects on a child's personality development. Se- condly there could be no debate whether it had occurred or not, in this regard contrasting strongly with the difficulty of obtaining valid in- formation about how a parent treats a child. Thirdly it appeared to be a field in which prevent- ive measures might be possible. And perhaps I should add, fourthly, that I was stimulated by the sheer incredulity with which my views were met by some, though by no means all, of my col- leagues when I first advanced them just before the war.
The results of our ensuing studies, undertaken by two researchers both of whom subsequently qualified as analysts, James Robertson and Christoph Heinicke, are now well-known; and I believe them to have had a significant effect on psychoanalytic thinking. The points I wish to make now, however, concern research strategy.
107/362
? ? ? Despite the pioneer work of such distinguished analysts as Anna Freud, Rene? Spitz, Ernst Kris, Margaret Mahler, and others, for long there has been a tendency in analytic circles to regard the direct observation of young children and the re- cording of what they say as no more than an aux- iliary method of research, the results of which are of interest when they confirm conclusions already reached by the traditional method of treating pa- tients but which are unable to contribute any- thing original. The notion that the direct observa- tion of children--in and out of the family set- ting--is not only a valuable method for advancing psychoanalytic science but is indispensable to it has been slow to be accepted.
The principal contributions of these direct studies, I believe, are to cast light on how chil- dren develop emotionally and socially, on what the ranges of variation are in respect to a very large number of relevant parameters, and what types of family experience tend to influence chil- dren to develop in one way rather than another. Let me give some examples of findings by col- leagues working in our sister sciences of ethology and developmental psychology that I believe to be highly germane to our clinical understanding.
108/362
? ? ? The first example is from the work of Mary Sal- ter Ainsworth (1977), formerly at Johns Hopkins University and now at the University of Virginia. Trained initially as a clinical psychologist, Mary Ainsworth worked with us at the Tavistock dur- ing the early 50s and then spent a couple of years studying mothers and infants in Uganda. Her definitive study has concerned the development of mother-infant interaction during the first year of life in white middle-class homes in Baltimore, Maryland. She has had a personal analysis and is keenly alive to the types of problem analysts re- gard as important.
During her study of mothers and infants in Uganda Ainsworth was struck how infants, once mobile, commonly use mother as a base from which to explore. When conditions are favourable an infant moves away from mother on explorat- ory excursions and returns to her again from time to time. By eight months of age almost every in- fant observed who had had a stable mother-figure to whom to become attached showed this beha- viour; but, should mother be absent, such organ- ized excursions became much less evident or ceased. As a result of these and similar findings, both for human and for monkey infants, the
109/362
? ? ? notion has been developed that an ordinary de- voted mother provides a child with a secure base from which he can explore and to which he can return when upset or frightened. Similar observa- tions, of course, have been made by Margaret Mahler (Mahler, Pine, and Bergman, 1975), though she interprets them in terms of a theoret- ical framework different to the one which Ainsworth and I use. This concept of the secure personal base, from which a child, an adolescent, or an adult goes out to explore and to which he returns from time to time, is one I have come to regard as crucial for an understanding of how an emotionally stable person develops and functions all through his life.
In her project in Baltimore, Ainsworth was not only able to study this kind of behaviour more closely but described many individual variations of it to be seen in a sample of 23 infants at 12 months of age. Observations were made of the in- fants' exploratory and attachment behaviour, and the balance between them, both when the infants were at home with mother and also when they were placed in a slightly strange test situation. In addition, having obtained data on the type of mothering each infant had been receiving
110/362
? ? ? throughout his first year of life (by means of pro- longed observation sessions at three weekly inter- vals in the child's home), Ainsworth was in a pos- ition to propose hypotheses linking certain types of emotional and behavioural development at 12 months with certain types of preceding mother- ing experience.
The findings of the study (see the review by Ainsworth, 1977) show that the way a particular infant of 12 months behaves with and without his mother at home and the way he behaves with and without her in a slightly strange test situation have much in common. Drawing on observations of behaviour in both types of situation it is then possible to classify the infants into three main groups, according to two criteria: (a) how much or how little they explore when with mother or without her, and (b) how they treat moth- er--when she is present, when she departs, and, especially, when she returns.
There were eight children whose overall beha- viour at their first birthday Ainsworth was dis- posed to regard as promising well for the future. Such infants explored actively, especially in mother's presence, and used mother as a base by keeping note of her whereabouts, exchanging
111/362
? ? ? glances, and from time to time returning to her to share in enjoyable mutual contact. When mother had been absent for a short time she was greeted warmly on her return. I will call these group X.
There were no less than eleven children whose overall behaviour gave cause for concern and whom I will call group Z. Three of them were passive, both at home and in the test situation; they explored little and, instead, sucked a thumb or rocked. Constantly anxious about mother's whereabouts, they cried much in her absence but were contrary and difficult on her return. The other eight in this group alternated between ap- pearing very independent and ignoring mother altogether, and then suddenly becoming anxious and trying to find her. Yet, when they did find her, they seemed not to enjoy contact with her, and often they struggled to get away again. In fact, they presented a classical picture of ambivalence.
The remaining four of the 23 children studied were judged to occupy a position intermediate between those given a good prognosis on their first birthday and those given a guarded one. I will call them group Y.
112/362
? ? ? Since every three weeks throughout these in- fants' short lives the researchers had spent a three-hour session in the child's home observing and recording mother and infant behaviour, they had much first-hand data from which to rate a mother's behaviour towards her child. In making these ratings Ainsworth used four distinct nine- point rating scales; but, since ratings on these scales intercorrelate highly, for present purposes one scale is sufficient--a scale that measures the degree of sensitivity or insensitivity that a mother shows to her baby's signals and communications. Whereas a sensitive mother seems constantly to be 'tuned in' to receive her baby's signals, is likely to interpret them correctly, and to respond to them both promptly and appropriately, an in- sensitive mother will often not notice her baby's signals, will misinterpret them when she does no- tice them, and will then respond tardily, inappro- priately, or not at all. When the ratings on this scale for the mothers of infants in each of the three groups are examined, it is found that the mothers of the eight infants in group X are rated uniformly high (range 5. 5 to 9. 0), those of the el- even infants in group Z are rated uniformly low (range 1. 0 to 3. 5), and those of the four in group
113/362
? ? ? Y are in the middle (range 4. 5 to 5. 5). Differences are statistically significant.
Plainly a great deal of further work is necessary before it is possible to draw conclusions with any high degree of confidence. Nevertheless, the overall patterns of personality development and of mother-child interaction visible at 12 months are sufficiently similar to what is seen of person- ality development and of parent-child interaction in later years for it to be plausible to believe that the one is the forerunner of the other. At the least, Ainsworth's findings show that an infant, whose mother is sensitive, accessible, and re- sponsive to him, who accepts his behaviour and is co-operative in dealing with him, is far from be- ing the demanding and unhappy child that some theories might suggest. Instead, mothering of this sort is evidently compatible with a child who is developing a limited measure of self-reliance by the time of his first birthday combined with a high degree of trust in his mother and enjoyment of her company.
Conversely mothers who are insensitive to their children's signals, perhaps because they are preoccupied and worried about other things, who ignore their children, or interfere with their
114/362
? ? ? activities in an arbitrary way, or simply reject them, are likely to have children who are un- happy or anxious or difficult. Anyone who has worked in a clinic seeing disturbed children or adolescents will hardly be surprised by that.
Although Ainsworth's finding of a correlation between a mother's responsiveness to her infant and the infant's way of behaving towards her at 12 months is highly significant statistically and has been confirmed by subsequent studies, it is always possible to argue that the partner who plays the greater role in determining whether in- teraction develops happily or not is the infant and not the mother. Some infants are born difficult, so the argument runs, and the mothers' adverse reactions to them are only to be expected.
I do not think the evidence supports this view. For example, the observations made during the first three months of these infants' lives showed no correlation between the amount of crying a baby did and the way his mother was treating him; whereas by the end of the first year mothers who had attended promptly to their crying babies had babies who cried much less than did the ba- bies of mothers who had left them to cry.
115/362
? ? ? There are other findings, some of which are re- ferred to in Lecture 6, that also support the view that in all but a small minority of cases it is the mother who is mainly responsible for how inter- action develops.
Drawing on her own home observations, Ainsworth has given a graphic account of what can happen. For example, she describes how she has sat in homes hearing a baby crying and crying and counted the minutes until the mother has re- sponded. In some cases a mother sits it out as long as she can bear to, believing that it would be bad for the baby and make him cry more were she to attend to him--a belief that Ainsworth's findings firmly disprove. In other cases a mother may be too engaged in something else to go. In yet others it appears as though a mother has alto- gether failed to register that her baby is crying--a situation an observer finds extremely painful to sit through. Usually these are women suffering from anxiety and depression and who are really incapable of attending to anything else.
Now it will be evident to everyone that detailed and accurate observations of these kinds, which demonstrate how enormously different the ex- periences of different children can be, are
116/362
? ? ? obtainable only by the methods used by these re- searchers. Had the observers not been present to see and hear what was going on but had relied in- stead on what the mothers told them, the pictures they would have got would in many cases have been entirely false; and all hope of finding signi- ficant correlations between the way a child devel- ops and the way he is treated by his mother and father would have vanished. Yet, as we have seen, when reliable methods of observation are used, even with quite small samples, highly significant correlations are found.
In emphasizing the determining role a mother plays in setting the pattern of interaction with her baby, to which I believe the evidence clearly points, I lay no blame. Looking after a baby, or a toddler, or an older child for that matter, is not only a skilled job but also a very hard and exact- ing one. Even for a woman who has had a happy childhood and who is now enjoying the help and support of her husband, and perhaps also of her own mother, and who has not been filled with mistaken advice about the dangers of spoiling her baby, it is a taxing one. That a woman with none of these advantages gets into an emotional hassle is hardly surprising and certainly not an occasion
117/362
? ? ? for blame. Yet there now seems little doubt that when infants and young children are the subjects of insensitive mothering, mixed perhaps with oc- casions of outright rejection, and later to separa- tions and threats of separation the effects are de- plorable. Such experiences greatly increase a child's fear of losing his mother, increase his de- mands for her presence and also his anger at her absences, and may also lead him to despair of ever having a secure and loving relationship with anyone.
Although ideas of this sort are much more fa- miliar and also more acceptable in analytic circles today than they were a generation ago, thanks to the influence of Balint, Fairbairn, Winnicott, and many others, I am inclined to think that their im- plications, both for theory and for practice, are still a long way from being digested.
Let me illustrate the point by considering the aetiological and therapeutic problems presented by the type of patient who in the United Kingdom is likely to be described as a schizoid personality (Fairbairn, 1940) or as having a false self (Win- nicott, 1960) and in North America as being a borderline personality or suffering from
118/362
? ? ? pathological narcissism (e. g. Kohut, 1971; Kern- berg, 1975).
The picture such a person presents is one of as- sertive independence and emotional self-suffi- ciency. On no account is he going to be beholden to anyone and, in so far as he enters into relation- ships at all, he makes sure he retains control. For much of the time he may appear to manage won- derfully well, but there may be times when he be- comes depressed or develops psychosomatic symptoms, often for no reason he knows of. Only should symptoms or a bout of depression become severe is there any possibility of his seeking treat- ment, and then more likely than not he will prefer drugs to analysts.
When such a person does come for analysis he is careful to keep the analyst at arm's length and to control what happens. What he tells us is lucid, but he avoids any reference to feeling, except per- haps to say how bored he gets. Holidays or other interruptions he welcomes as saving his time. Perhaps he finds the analysis an 'interesting exer- cise'; although he is not convinced it is much use. And in any case he could probably do a better job by analysing himself!
119/362
? ? ? There is, of course, a large literature discussing the psychopathology of these conditions and the therapeutic problems they present; but on whatever issues there may be agreement there is none on aetiology. To take two contrasting view- points: whereas Winnicott (1960, 1974) attributes the condition squarely to early environmental failure in the form of 'not good enough mother- ing', Kernberg (1975) in his systematic treatise gives no more than a couple of easily missed paragraphs to the possible role that mothering plays in influencing development, and only a few passing references to the inadequate mothering certain of his patients may have received. That early experience may play the key role in determ- ining these conditions is not seriously examined by him.
Plainly it is of the greatest importance that in due course we should reach some consensus about this matter; and in debating the issues I be- lieve we should be foolish not to take account of data from as many sources as we can tap. For some conditions epidemiological surveys are now proving informative but I doubt whether they have anything yet to tell us about this one. At present therefore we have to make do with data
120/362
? ? ? from our two familiar sources: (a) the analytic treatment of patients, (b) the direct observation of young children with their mothers.
As regards data obtained during treatment, I suspect it would be fruitful for some open- minded person to survey the psychoanalytic liter- ature and draw together all the case reports which record information about the childhood experiences of these patients. My guess is that, in so far as any information is given, it would strongly support Winnicott's view that these pa- tients have had disturbed childhoods in which in- adequate mothering in one form or another--and it can take many--bulks large. Since I have made no such survey, I can do no more than illustrate the kind of findings that I would confidently ex- pect. The following details come from case re- ports published by three analysts each much in- fluenced by Winnicott's views.
One report is by Donald Winnicott's widow Clare Winnicott (1980). The patient, a profes- sional woman of 41, presented a classical picture of the emotionally self-sufficient personality who recently had developed a variety of psychosomat- ic symptoms. Only after a good deal of analysis did she divulge the events of her childhood. Since
121/362
? ? ? her mother was in full-time work, she was looked after by a German girl who left suddenly when the patient was 21/2. Then, after six months of un- certainty, she was taken by her mother to have tea with a friend and later found her mother had disappeared and she was alone in a strange bed. Next day she was taken to the boarding school where her mother's friend worked as matron, and she stayed there till she was 9, usually spending the holidays there also. She appears to have settled in well (ominous words! ) and coped very successfully; but from that time forward her emo- tional life had dried up.
A second report, by Jonathan Pedder (1976), is of a young teacher in her mid-twenties whose personality and symptomatology bear a strong resemblance to Clare Winnicott's patient. Al- though at the initial interview she had given an idealized picture of her childhood, it soon emerged that at the age of 18 months she had been sent to stay with an aunt during her moth- er's next pregnancy. After six months there she had come to feel that her aunt was more of a mother to her than was her real mother and she had found returning home a painful experience. Thereafter, until she was 10, she had been
122/362
? ? ? terrified of another separation; but then she had 'switched off' her anxiety 'like a tap', as she put it, and with the anxiety had disappeared most of her emotional life as well.
The third report, by Elizabeth Lind (1973), con- cerns a young graduate of 23 who, though severely depressed and planning suicide, main- tained that his state of mind was less an illness than 'a philosophy of life'. He was the eldest of a large family; and by the time he was 3 two sib- lings had already been born. His parents, he said, quarrelled both frequently and violently. When the family was young, father had been working long hours away from home training for a profes- sion. Mother was always unpredictable. Often she was so distraught by her quarrelling children that she would lock herself in her room for days on end.
Several times she had left home, taking the daughters with her but leaving the sons behind.
He had been told that he had been an unhappy baby, a poor feeder and sleeper, who had often been left alone to cry for long periods. His crying, it was said, had been just an attempt to gain con- trol of his parents and to be spoilt. On one occa- sion he had had appendicitis and he remembered lying awake all night moaning; but his parents
123/362
? ? ? had done nothing and by next morning he was seriously ill. Later, during therapy, he recalled how disturbed he used to be at hearing his young- er brothers and sisters being left to cry and how he hated his parents for it and felt like killing them.
He had always felt like a lost child and had been puzzled to understand why he had been re- jected. His first day at school, he said, had been the worst in his life. It had seemed a final rejec- tion by his mother; all day he had felt desperate and had never stopped crying. After that he had gradually come to hide all desires for love and support: he had refused ever to ask for help or to have anything done for him.
Now, during therapy, he was frightened he might break down and cry and want to be mothered. This would lead his therapist, he felt sure, to regard him as a nuisance and his beha- viour simply as attention-seeking; and, were he to say anything personal to her, he fully expected her to be offended and perhaps lock herself in her room.
In all three cases the patient's recent break- down had followed the collapse of a significant but fragile relationship about which each
124/362
? ? ? member of the pair had had reservations and to the ending of which each of the patients had, him or herself, obviously contributed.
In treating these patients all three analysts ad- opted Winnicott's technique of permitting the free expression of what are traditionally termed 'dependency feelings', with the result that each patient in due course developed an intense and anxious attachment to his or her analyst (to use the terminology I prefer, (Bowlby, 1969, 1973)). This enabled each patient to recover the emotion- al life he or she had lost during childhood and with it to recover a sense of 'real self'. Therapeut- ically the results were good.
Admittedly, the findings from these three cases prove nothing. Nevertheless they are suggestive and, so far as they go, support Winnicott's theory of aetiology. Even so it is always open to critics to cast doubt on the validity of what a patient recalls about his childhood and to question whether the sequence of events recounted had the effect on his feeling life that he so explicitly claims. (It is worth nothing that the events that each of these three patients held to be a turning point had oc- curred after their second birthdays. )
125/362
? ? ? Now it seems clear that the controversies about aetiology are never going to be settled as long as we rely solely on the retrospective and perhaps biased evidence derived from the analyses of pa- tients, whether they be adults or children. What is needed is evidence of a different kind to provide some sort of cross-check. This is where I believe the direct observations of young children and their mothers are potentially so useful. Is there any evidence from that source that suggests that a child's feeling life can become numbed by the types of experience described? The answer, of course, is that there is a great deal.
Here naturally I point first to the observations made by James Robertson (1953) and confirmed later by Christoph Heinicke and Ilse Westheimer (1965) on how children between the ages of 12 and 36 months behave when removed from home to the care of strange people in a strange place, such as a residential nursery or hospital, with no one person to act as a mother-substitute. In such conditions a child comes in time to act as if neither mothering nor contact with humans has much significance for him. As his caretakers come and go he ceases to attach himself to any- one and after his return home stays remote from
126/362
? ? ? his parents for days, and perhaps for much longer if he is treated unsympathetically.
There is reason to believe, moreover, that a young child can develop this kind of defensive numbing in response to a mother who rejects him and without any major separation. Examples of this sequence are to be found in observations re- corded by Mahler (1971). More definitive findings are reported by Mary Main (1977), a colleague of Mary Ainsworth's, who has made a special study of a group of children in the age-range 12 to 20 months, each of whom not only failed to greet his mother after she had left him with a stranger for a few minutes but deliberately avoided her. View- ing some of Main's videotaped records I was as- tonished to see to what lengths some of these children went. One approached her mother briefly but with head averted and then retreated from her. Another, instead of approaching his mother, placed himself facing into the corner of the room, as though complying with a punish- ment, and then knelt down with his face to the floor. In every case videotaped records of these mothers playing with their toddlers during a later session showed them to differ from the mothers of non-avoidant toddlers: they appeared 'angry,
127/362
? ? ? inexpressive and disliking of physical contact with the infant'. Some scolded in angry tones, some mocked, others spoke sarcastically to or about their child. An obvious possibility is that by keeping away from his mother in this way a child is avoiding being treated in a hostile way again.
Thus, so far as the cross-checks provided by direct observations of young children and their mothers go, they tend to support a Winnicott- type theory. Put briefly, and in my own words, the child, and later the adult, becomes afraid to allow himself to become attached to anyone for fear of a further rejection with all the agony, the anxiety, and the anger to which that would lead. As a result there is a massive block against his ex- pressing or even feeling his natural desire for a close trusting relationship, for care, comfort, and love--which I regard as the subjective manifesta- tions of a major system of instinctive behaviour.
An explanation of this kind, although much less complex than some proposed in the literat- ure, accounts well for how these people behave both in the world at large and with ourselves as analysts. Inevitably they bring their fear of enter- ing into a trusting relationship with them to ana- lysis, which we experience as a massive
128/362
? ? ? resistance. Then, when at length their feelings are recovered, they more than half expect that we shall treat them as they recall being treated by their parents. In consequence they live in dread of being rejected and become intensely angry should they suspect us of deserting them. Not in- frequently, moreover, the way they treat us--with abuse and rejection--is found to be a version of the treatment to which they recall having them- selves been subjected as children.
You will see that in the explanation of how these patients behave during analysis I have ad- vanced a number of interlocking hypotheses. In a research programme each requires scrutiny and testing in the light of further data. Among the many methods that I would expect to prove of value is the study, in a therapeutic setting, of par- ents and children interacting with one another. In addition, there remains an important place for further observations to be made during the ana- lysis of individual patients; though I believe that, if clinical research is to yield its full potential, it has to be pursued in a far more systematic and directed way than hitherto.
To give an example: it would be of value were a detailed record to be kept of the responses of one
129/362
? ? ? or more of these patients before and after each successive weekend, each vacation, and each un- expected interruption of the sessions, with an equally detailed record of how the analyst dealt with them. This would enable us to know the rep- ertoire of responses a given patient presents on these occasions, and also the changes in response he presents over time. It would also be especially valuable if we were to have a detailed account of the conditions in which a major therapeutic change occurs. If, perhaps in a collaborative pro- gramme, records could be kept on a number of such patients, it might be possible to discover whether a frank and detailed discussion of the painful experiences a patient recalls having had in his relationships with his parents and the ef- fects these appear to have had and still to be hav- ing on the ways he treats other people, including of course ourselves, promotes therapeutic change, as I predict, or hinders it, as is believed by some analysts.
Naturally, in embarking on this or any other research programme an analyst must bear in mind his professional responsibilities; for with patients who present a false self these can be very onerous. Winnicott describes the 'period of
130/362
? ? ? extreme dependence' through which such pa- tients go during therapy and gives warning that 'analysts who are not prepared to go and meet the heavy needs of patients who become dependent in this way must be careful so to choose their cases that they do not include false self types'.
This brings me back to the art of therapy. To provide, by being ourselves, the conditions in which a patient of this kind can discover and re- cover what Winnicott calls his real self, and I call his attachment desires and feelings, is not easy. On the one hand, we have really to be trustworthy and we have also genuinely to respect all those yearnings for affection and intimacy that each of us has but which in these patients have become lost. On the other, we must not offer more than we can give and we must not move faster than the patient can bear. To achieve this balance requires all the intuition, imagination, and empathy of which we are capable. But it also requires a firm grasp of what the patient's problems are and what we are trying to do. This is why it is so very important that the problems of aetiology and psy- chopathology should be clarified as far as the ap- plication of scientific method makes possible and, further, that analysts should be thoroughly
131/362
? ? ? informed about the whole range of family experi- ences, from birth on through adolescence, that, increasing evidence shows, affect how a child's emotional life develops. Only when we become armed with that and much further knowledge shall we be in a position to meet the exacting de- mand that Freud makes in one of the last papers he wrote, in which he draws attention to 'the ker- nel of truth' in a patient's symptoms and to the therapeutic value of constructions in analysis (1937). In it he writes: 'What we are in search of is a picture of the patient's forgotten years that shall be alike trustworthy and in all essential re- spects complete. '
1 Efron (1977) has discussed the circumstances in which Freud's abrupt change of mind took place.
4
PSYCHOANALYSIS AS A NATURAL SCIENCE
In the autumn of 1980 I was appointed Freud Memorial Visiting Professor of Psychoanalysis at University College, London. In my inaugural lecture I returned to the theme I had spoken on in Canada two years previously. Having always believed that the body of knowledge labelled psychoanalysis should become a part of natural science, I was distressed by the pressure of the opposition. To accept that psychoanalysis should abandon its aim of becoming a natural science and instead should regard itself as a hermeneut- ic discipline has seemed to me to be not only a result of obsolete ideas about science but also a counsel of despair; because, in a hermeneutic discipline, there are no criteria by the applica- tion of which it is ever possible to resolve disagreement.
A problem encountered by every analyst who has proposed new theoretical ideas is the
? 133/362
? ? ? criticism that the new theory is 'not psychoana- lysis'. Such criticism turns, of course, on how we define psychoanalysis. Most unfortunately, de- fining it in terms of Freud's theories is all too common. This is in contrast to the definitions ad- opted by academic disciplines which are always in terms of the phenomena to be studied and the problems to be solved. In such disciplines pro- gress is frequently signalled by changes of the- ory, sometimes of a revolutionary kind. So long as analysts continue to define psychoanalysis in terms of any particular theory, they must not complain that their discipline is cold-shouldered by academics. Furthermore by so defining it they are condemning it to frozen inertia.
The following version of this lecture differs in a number of ways from the original, in particu- lar by abbreviating the discussion of issues already dealt with in the preceding lectures.
From 1895, when Freud made his first attempt to sketch a theoretical framework for psychoanalys- is, until 1938, the year before he died, Freud was determined that his new discipline should con- form to the requirements of a natural science. Thus, in the opening sentence of his Project he writes: 'The intention is to furnish a psychology
134/362
? ? ? that shall be a natural science. . . . ' (Freud, 1950, 295); whilst in the Outline we find a passage in which he asserts that, once the concept of psych- ical processes being unconscious is granted, 'psy- chology is enabled to take its place as a natural science like any other' (Freud, 1940, 158).
Admittedly during the intervening years Freud's ideas about the scope of his science had changed considerably from his early ambition 'to represent psychical processes as quantitatively determinate states of specifiable material particles' (1950, written in 1895) to his later definition of psychoanalysis as 'the science of un- conscious mental processes' (1925). But from first to last there can be no doubt what sort of discipline Freud intended psychoanalysis to be.
Nevertheless, despite Freud's unwavering in- tention, the scientific status of psychoanalysis re- mains equivocal. On the one hand philosophers of science have dubbed it a pseudoscience on the grounds that, however large a measure of truth they may contain, psychoanalytic theories are cast in so elastic a form that they are unfalsifi- able. On the other many psychoanalysts, disillu- sioned by the inadequacies of Freud's metapsy- chology and preoccupied with the personal
135/362
? ? ? perspective which is unquestionably required in clinical work, have abandoned Freud's aims and claims and have declared that psychoanalysis is miscast as a science and should be conceived in- stead as one of the humanities, e. g. Home (1966), Ricoeur (1970), and others in Europe. Both Schafer (1976) and George Klein (1976), espous- ing this view, have advanced proposals alternat- ive to Freud's: but each of their reformulations, different though they be, seems a version of Hamlet without the Prince. Gone are all concepts of causality and theories of biologically rooted impulse and, in Schafer's version, gone also are concepts of repression and unconscious mental activity.
Melanie Klein has made very different propos- als, ones which certainly do not suffer from these defects; but it would be difficult to claim that the form they take or the research they have en- gendered meet scientific requirements.
Yet by no means do all analysts despair of de- veloping their discipline as a natural science. Alive to the deficiencies of Freud's metapsycho- logy, especially his concepts of psychic energy and drive, a few are attempting to replace it with a new conceptual framework consistent with
136/362
? ? ? current scientific thinking. Central to these new proposals are ideas drawn from systems theory and the study of human information processing. Those active in this enterprise include Rubinstein (1967), Peterfreund (1971, 1982), Rosenblatt and Thickstun (1977), Gedo (1979), and myself (1969, 1980). Meanwhile there are also a number of analysts who have been seeking to extend the dis- cipline's database by studying children's social and emotional development using direct observa- tions. Some of these studies have been atheoretic- al, e. g. Offer (1969). The authors of others have attempted to put new empirical wine into the old theoretical bottles, e. g. Spitz (1957), Mahler (Mahler, Pine, and Bergman, 1975); whilst others again, e. g. Sander (1964, 1980), Stern (1977), and myself (1958, 1969, 1973), have searched for new theoretical models. My own search has led not only to control theory and information pro- cessing but also to the biologically rooted discip- lines of ethology and comparative psychology. Thus there is no lack of new initiatives and it will take time to see which of them, or perhaps which combination of them, proves most productive of scientific advance. Here my aim is to describe one
137/362
? ? ? such initiative, my own, and why I think it promising.
In the preceding lecture I described the cir- cumstances which led me to choose as a field of research the responses of a young child to being removed from his or her mother and placed for a time in a strange place with strange people, and how these observations led on to the formulation of attachment theory. Among essential features of this are that the human infant comes into the world genetically biased to develop a set of beha- vioural patterns that, given an appropriate envir- onment, will result in his keeping more or less close proximity to whomever cares for him, and that this tendency to maintain proximity serves the function of protecting the mobile infant and growing child from a number of dangers, amongst which in man's environment of evolu- tionary adaptedness the danger of predation is likely to have been paramount.
A concept that emerged early from ethologic- ally oriented studies of mother-child relation- ships (Ainsworth, 1967) and which has proved of great clinical value is that of a mother, or mother- substitute, providing a child with a secure base from which he can explore. By the last months of
138/362
? ? ? the first year of life an infant brought up in an or- dinary affectionate home is very clear whom he prefers to care for him, a preference especially evident should he be tired, frightened, or sick. Whoever that may be, and it is usually his moth- er, is then able by her very presence, or ready ac- cessibility, to create the conditions which enable him to explore his world in a confident way. At the time of his second birthday, for example, a healthy child whose mother is resting on a garden seat will make a series of excursions away from her, each time returning to her before making the next excursion. On some occasions, when he re- turns, he simply smiles and makes his number; on others he leans against her knee; on yet others he wants to climb on her lap. But never does he stay for long unless he is frightened or tired or thinks she is about to leave. Anderson (1972), who made a study of this sort in a London park, observed that during the second and third years it is very rare for a child to go further than two hundred feet before returning. Should he lose sight of his mother, exploration is forgotten. His top priority then is to regain her, in an older child by searching and in a younger one by howling.
139/362
? ? ? It is evident that there is no way of explaining this type of behaviour in terms of a build-up of psychic energy which is then discharged. An al- ternative model (already described in earlier lec- tures) is to think of the proximity-keeping of a child as being mediated by a set of behavioural systems organized cybernetically. Activation is intensified in conditions of pain, fatigue, and anything frightening; and reduced by proximity to or contact with the mother-figure. We can then postulate that the behaviour that takes him away from his mother into the wide world, which is conveniently termed exploratory behaviour, is in- compatible with attachment behaviour and has a lower priority. It is thus only when attachment behaviour is relatively inactive that exploration occurs.
As an individual grows older his life continues to be organized in the same kind of way though his excursions become steadily longer both in time and space. On entering school they will last for hours and later for days. During adolescence they may last for weeks or months, and new at- tachment figures are likely to be sought. Throughout adult life the availability of a re- sponsive attachment figure remains the source of
140/362
? ? ? a person's feeling secure. All of us, from the cradle to the grave, are happiest when life is or- ganized as a series of excursions, long or short, from the secure base provided by our attachment figure(s).
In terms of the theoretical model proposed, the pronounced changes in the organization of at- tachment behaviour that occur during individual development are regarded as being due, in part, to the threshold for its activation being raised (perhaps through changes in endocrine levels) and, in part, to the control systems becoming in- creasingly sophisticated, in particular by their coming to incorporate representational models of the environment and important people in it and also of the self as a living active person.
The development during ontogeny of a set of systems of the kind described in humans, as well as in individuals of many other species, is attrib- uted to the action of natural selection, namely to individuals well endowed with the potential to develop such systems having survived and bred more successfully than those less well endowed, in other words to Darwinian evolution. Since a disposition to show attachment behaviour in cer- tain circumstances is regarded as an intrinsic
141/362
? ? ? part of human nature, reference to it as 'depend- ency' is not only misleading but seriously inap- propriate because of the word's pejorative overtones.
Once attachment behaviour and other forms of biologically determined behaviour are conceived in forms of control theory, the problem of the purposiveness of behaviour is solved without abandoning the concept of causation. Further- more the distinction between causation and func- tion, sadly neglected in traditional psychoanalytic theory, becomes explicit. Activation, in which emotional arousal and expression play a critical part, and subsequently termination and change of emotional state are caused when a system con- structed in a particular way receives information of particular sorts. Of the various consequences to which activation leads, the one postulated as its biological function is the one that, evidence suggests, has led to the system having evolved during phylogeny. In the case of attachment be- haviour the function postulated is that of dimin- ishing the risk of the individual coming to harm.
At this level of analysis the question of whether an individual is aware of what he is doing, let alone why he is doing it, has no relevance, in fact
142/362
? ? ? no more relevance than has the question of whether an individual is aware that he is breath- ing and, if so, realizes why he should be doing so. Biological systems serving vital functions, wheth- er at a behavioural or a physiological level, must be capable of operating automatically. Neverthe- less in the case of a human child awareness of what he is doing, and more particularly aware- ness of the conditions that will terminate his be- haviour, soon emerges, certainly by the end of his first year, and is a factor of great importance. For, once it is clear that a child is aware of the condi- tions that will terminate his behaviour, we begin speaking of intention, of his desire to achieve a certain goal, of his being satisfied and happy when he does so and frustrated, anxious, or angry when he fails, of satisfaction bringing pleasure and frustration the reverse.
At this point I wish to emphasize the sharp dis- tinction between conditions necessary to termin- ate a certain form of behaviour, commonly re- ferred to as its goal, and the biological function the behaviour serves. In the case of attachment behaviour in childhood, whereas we commonly expect both mother and child to be aware of the conditions necessary to terminate it, for example,
143/362
?
103/362
? ? ? thereby be expanded and, on the other, accepting that there are likely always to remain still greater areas lying beyond the scope of any existing mode of scientific enquiry.
To many of you, I am afraid, engaged in thera- peutic practice but also hoping to contribute to the advance of psychoanalytic science, the con- trasts I am drawing between the roles of practi- tioner and scientist will hardly be welcome. Yet I believe that it is only by recognizing these differ- ences and acting accordingly that the strengths of each role can be used to fullest advantage--or that any one person can occupy both of them with any hope of success. As practitioners we deal in complexity; as scientists we strive to simplify. As practitioners we use theory as a guide; as sci- entists we challenge that same theory. As practi- tioners we accept restricted modes of enquiry; as scientists we enlist every method we can.
Earlier I remarked on the need for every devel- oping science to devise new methods for obtain- ing data. The reason for this is that, however pro- ductive any one method may be, it is bound to have its limitations, whilst there is always a pro- spect that some other method may compensate for them. Thus the new method may be in no way
104/362
? ? ? superior to the old; indeed it may have great lim- itations. Its usefulness lies simply in the fact that its strengths and its limitations are different. Per- haps I can illustrate the point by reference to my own work.
When I qualified in psychoanalysis in 1937, members of the British Society were occupied in exploring the fantasy worlds of adults and chil- dren, and it was regarded as almost outside the proper interest of an analyst to give systematic attention to a person's real experiences. That was a time when Freud's famous about-turn of 1897 regarding the aetiology of hysteria1 had led to the view that anyone who places emphasis on what a child's real experiences may have been, and per- haps still are being, was regarded as pitifully nai? ve. Almost by definition it was assumed that anyone interested in the external world could not be interested in the internal world, indeed was al- most certainly running away from it.
To me as a biologist this contrast of internal with external, of organism with environment, never appealed. Furthermore, as a psychiatrist engaged in work with children and families and deeply influenced by the insights of two analytic- ally oriented social workers, I was daily
105/362
? ? ? confronted with the impact on children of the emotional problems from which their parents suffered. Here are two examples I still recall vividly. In one a father was deeply concerned about his 8-year-old son's masturbation and in reply to my enquiries explained how, whenever he caught him with his hand on his genitals, he put him under a cold tap. This led me to ask fath- er whether he himself had ever had any worry about masturbation, and he launched into a long and pathetic tale of how he had battled with the problem all his life. In another case a mother's punitive treatment of her 3-year-old's jealousy of the new baby was as quickly traced to the prob- lem she had always had with her own jealousy of a younger brother.
Observations of these kinds led me to conclude that it is just as necessary for analysts to study the way a child is really treated by his parents as it is to study the internal representations he has of them, indeed that the principal focus of our studies should be the interaction of the one with the other, of the internal with the external. Be- lieving that that would be possible only if we had far more systematic knowledge about the effects on a child of the experiences he has during his
106/362
? ? ? early years within his family, I concentrated my attention on this area. The reasons that I selected as my special field of study the removal of a young child from his home to a residential nurs- ery or hospital rather than the broader field of parent-child interaction were several. First it was an event that I believed could have serious ill ef- fects on a child's personality development. Se- condly there could be no debate whether it had occurred or not, in this regard contrasting strongly with the difficulty of obtaining valid in- formation about how a parent treats a child. Thirdly it appeared to be a field in which prevent- ive measures might be possible. And perhaps I should add, fourthly, that I was stimulated by the sheer incredulity with which my views were met by some, though by no means all, of my col- leagues when I first advanced them just before the war.
The results of our ensuing studies, undertaken by two researchers both of whom subsequently qualified as analysts, James Robertson and Christoph Heinicke, are now well-known; and I believe them to have had a significant effect on psychoanalytic thinking. The points I wish to make now, however, concern research strategy.
107/362
? ? ? Despite the pioneer work of such distinguished analysts as Anna Freud, Rene? Spitz, Ernst Kris, Margaret Mahler, and others, for long there has been a tendency in analytic circles to regard the direct observation of young children and the re- cording of what they say as no more than an aux- iliary method of research, the results of which are of interest when they confirm conclusions already reached by the traditional method of treating pa- tients but which are unable to contribute any- thing original. The notion that the direct observa- tion of children--in and out of the family set- ting--is not only a valuable method for advancing psychoanalytic science but is indispensable to it has been slow to be accepted.
The principal contributions of these direct studies, I believe, are to cast light on how chil- dren develop emotionally and socially, on what the ranges of variation are in respect to a very large number of relevant parameters, and what types of family experience tend to influence chil- dren to develop in one way rather than another. Let me give some examples of findings by col- leagues working in our sister sciences of ethology and developmental psychology that I believe to be highly germane to our clinical understanding.
108/362
? ? ? The first example is from the work of Mary Sal- ter Ainsworth (1977), formerly at Johns Hopkins University and now at the University of Virginia. Trained initially as a clinical psychologist, Mary Ainsworth worked with us at the Tavistock dur- ing the early 50s and then spent a couple of years studying mothers and infants in Uganda. Her definitive study has concerned the development of mother-infant interaction during the first year of life in white middle-class homes in Baltimore, Maryland. She has had a personal analysis and is keenly alive to the types of problem analysts re- gard as important.
During her study of mothers and infants in Uganda Ainsworth was struck how infants, once mobile, commonly use mother as a base from which to explore. When conditions are favourable an infant moves away from mother on explorat- ory excursions and returns to her again from time to time. By eight months of age almost every in- fant observed who had had a stable mother-figure to whom to become attached showed this beha- viour; but, should mother be absent, such organ- ized excursions became much less evident or ceased. As a result of these and similar findings, both for human and for monkey infants, the
109/362
? ? ? notion has been developed that an ordinary de- voted mother provides a child with a secure base from which he can explore and to which he can return when upset or frightened. Similar observa- tions, of course, have been made by Margaret Mahler (Mahler, Pine, and Bergman, 1975), though she interprets them in terms of a theoret- ical framework different to the one which Ainsworth and I use. This concept of the secure personal base, from which a child, an adolescent, or an adult goes out to explore and to which he returns from time to time, is one I have come to regard as crucial for an understanding of how an emotionally stable person develops and functions all through his life.
In her project in Baltimore, Ainsworth was not only able to study this kind of behaviour more closely but described many individual variations of it to be seen in a sample of 23 infants at 12 months of age. Observations were made of the in- fants' exploratory and attachment behaviour, and the balance between them, both when the infants were at home with mother and also when they were placed in a slightly strange test situation. In addition, having obtained data on the type of mothering each infant had been receiving
110/362
? ? ? throughout his first year of life (by means of pro- longed observation sessions at three weekly inter- vals in the child's home), Ainsworth was in a pos- ition to propose hypotheses linking certain types of emotional and behavioural development at 12 months with certain types of preceding mother- ing experience.
The findings of the study (see the review by Ainsworth, 1977) show that the way a particular infant of 12 months behaves with and without his mother at home and the way he behaves with and without her in a slightly strange test situation have much in common. Drawing on observations of behaviour in both types of situation it is then possible to classify the infants into three main groups, according to two criteria: (a) how much or how little they explore when with mother or without her, and (b) how they treat moth- er--when she is present, when she departs, and, especially, when she returns.
There were eight children whose overall beha- viour at their first birthday Ainsworth was dis- posed to regard as promising well for the future. Such infants explored actively, especially in mother's presence, and used mother as a base by keeping note of her whereabouts, exchanging
111/362
? ? ? glances, and from time to time returning to her to share in enjoyable mutual contact. When mother had been absent for a short time she was greeted warmly on her return. I will call these group X.
There were no less than eleven children whose overall behaviour gave cause for concern and whom I will call group Z. Three of them were passive, both at home and in the test situation; they explored little and, instead, sucked a thumb or rocked. Constantly anxious about mother's whereabouts, they cried much in her absence but were contrary and difficult on her return. The other eight in this group alternated between ap- pearing very independent and ignoring mother altogether, and then suddenly becoming anxious and trying to find her. Yet, when they did find her, they seemed not to enjoy contact with her, and often they struggled to get away again. In fact, they presented a classical picture of ambivalence.
The remaining four of the 23 children studied were judged to occupy a position intermediate between those given a good prognosis on their first birthday and those given a guarded one. I will call them group Y.
112/362
? ? ? Since every three weeks throughout these in- fants' short lives the researchers had spent a three-hour session in the child's home observing and recording mother and infant behaviour, they had much first-hand data from which to rate a mother's behaviour towards her child. In making these ratings Ainsworth used four distinct nine- point rating scales; but, since ratings on these scales intercorrelate highly, for present purposes one scale is sufficient--a scale that measures the degree of sensitivity or insensitivity that a mother shows to her baby's signals and communications. Whereas a sensitive mother seems constantly to be 'tuned in' to receive her baby's signals, is likely to interpret them correctly, and to respond to them both promptly and appropriately, an in- sensitive mother will often not notice her baby's signals, will misinterpret them when she does no- tice them, and will then respond tardily, inappro- priately, or not at all. When the ratings on this scale for the mothers of infants in each of the three groups are examined, it is found that the mothers of the eight infants in group X are rated uniformly high (range 5. 5 to 9. 0), those of the el- even infants in group Z are rated uniformly low (range 1. 0 to 3. 5), and those of the four in group
113/362
? ? ? Y are in the middle (range 4. 5 to 5. 5). Differences are statistically significant.
Plainly a great deal of further work is necessary before it is possible to draw conclusions with any high degree of confidence. Nevertheless, the overall patterns of personality development and of mother-child interaction visible at 12 months are sufficiently similar to what is seen of person- ality development and of parent-child interaction in later years for it to be plausible to believe that the one is the forerunner of the other. At the least, Ainsworth's findings show that an infant, whose mother is sensitive, accessible, and re- sponsive to him, who accepts his behaviour and is co-operative in dealing with him, is far from be- ing the demanding and unhappy child that some theories might suggest. Instead, mothering of this sort is evidently compatible with a child who is developing a limited measure of self-reliance by the time of his first birthday combined with a high degree of trust in his mother and enjoyment of her company.
Conversely mothers who are insensitive to their children's signals, perhaps because they are preoccupied and worried about other things, who ignore their children, or interfere with their
114/362
? ? ? activities in an arbitrary way, or simply reject them, are likely to have children who are un- happy or anxious or difficult. Anyone who has worked in a clinic seeing disturbed children or adolescents will hardly be surprised by that.
Although Ainsworth's finding of a correlation between a mother's responsiveness to her infant and the infant's way of behaving towards her at 12 months is highly significant statistically and has been confirmed by subsequent studies, it is always possible to argue that the partner who plays the greater role in determining whether in- teraction develops happily or not is the infant and not the mother. Some infants are born difficult, so the argument runs, and the mothers' adverse reactions to them are only to be expected.
I do not think the evidence supports this view. For example, the observations made during the first three months of these infants' lives showed no correlation between the amount of crying a baby did and the way his mother was treating him; whereas by the end of the first year mothers who had attended promptly to their crying babies had babies who cried much less than did the ba- bies of mothers who had left them to cry.
115/362
? ? ? There are other findings, some of which are re- ferred to in Lecture 6, that also support the view that in all but a small minority of cases it is the mother who is mainly responsible for how inter- action develops.
Drawing on her own home observations, Ainsworth has given a graphic account of what can happen. For example, she describes how she has sat in homes hearing a baby crying and crying and counted the minutes until the mother has re- sponded. In some cases a mother sits it out as long as she can bear to, believing that it would be bad for the baby and make him cry more were she to attend to him--a belief that Ainsworth's findings firmly disprove. In other cases a mother may be too engaged in something else to go. In yet others it appears as though a mother has alto- gether failed to register that her baby is crying--a situation an observer finds extremely painful to sit through. Usually these are women suffering from anxiety and depression and who are really incapable of attending to anything else.
Now it will be evident to everyone that detailed and accurate observations of these kinds, which demonstrate how enormously different the ex- periences of different children can be, are
116/362
? ? ? obtainable only by the methods used by these re- searchers. Had the observers not been present to see and hear what was going on but had relied in- stead on what the mothers told them, the pictures they would have got would in many cases have been entirely false; and all hope of finding signi- ficant correlations between the way a child devel- ops and the way he is treated by his mother and father would have vanished. Yet, as we have seen, when reliable methods of observation are used, even with quite small samples, highly significant correlations are found.
In emphasizing the determining role a mother plays in setting the pattern of interaction with her baby, to which I believe the evidence clearly points, I lay no blame. Looking after a baby, or a toddler, or an older child for that matter, is not only a skilled job but also a very hard and exact- ing one. Even for a woman who has had a happy childhood and who is now enjoying the help and support of her husband, and perhaps also of her own mother, and who has not been filled with mistaken advice about the dangers of spoiling her baby, it is a taxing one. That a woman with none of these advantages gets into an emotional hassle is hardly surprising and certainly not an occasion
117/362
? ? ? for blame. Yet there now seems little doubt that when infants and young children are the subjects of insensitive mothering, mixed perhaps with oc- casions of outright rejection, and later to separa- tions and threats of separation the effects are de- plorable. Such experiences greatly increase a child's fear of losing his mother, increase his de- mands for her presence and also his anger at her absences, and may also lead him to despair of ever having a secure and loving relationship with anyone.
Although ideas of this sort are much more fa- miliar and also more acceptable in analytic circles today than they were a generation ago, thanks to the influence of Balint, Fairbairn, Winnicott, and many others, I am inclined to think that their im- plications, both for theory and for practice, are still a long way from being digested.
Let me illustrate the point by considering the aetiological and therapeutic problems presented by the type of patient who in the United Kingdom is likely to be described as a schizoid personality (Fairbairn, 1940) or as having a false self (Win- nicott, 1960) and in North America as being a borderline personality or suffering from
118/362
? ? ? pathological narcissism (e. g. Kohut, 1971; Kern- berg, 1975).
The picture such a person presents is one of as- sertive independence and emotional self-suffi- ciency. On no account is he going to be beholden to anyone and, in so far as he enters into relation- ships at all, he makes sure he retains control. For much of the time he may appear to manage won- derfully well, but there may be times when he be- comes depressed or develops psychosomatic symptoms, often for no reason he knows of. Only should symptoms or a bout of depression become severe is there any possibility of his seeking treat- ment, and then more likely than not he will prefer drugs to analysts.
When such a person does come for analysis he is careful to keep the analyst at arm's length and to control what happens. What he tells us is lucid, but he avoids any reference to feeling, except per- haps to say how bored he gets. Holidays or other interruptions he welcomes as saving his time. Perhaps he finds the analysis an 'interesting exer- cise'; although he is not convinced it is much use. And in any case he could probably do a better job by analysing himself!
119/362
? ? ? There is, of course, a large literature discussing the psychopathology of these conditions and the therapeutic problems they present; but on whatever issues there may be agreement there is none on aetiology. To take two contrasting view- points: whereas Winnicott (1960, 1974) attributes the condition squarely to early environmental failure in the form of 'not good enough mother- ing', Kernberg (1975) in his systematic treatise gives no more than a couple of easily missed paragraphs to the possible role that mothering plays in influencing development, and only a few passing references to the inadequate mothering certain of his patients may have received. That early experience may play the key role in determ- ining these conditions is not seriously examined by him.
Plainly it is of the greatest importance that in due course we should reach some consensus about this matter; and in debating the issues I be- lieve we should be foolish not to take account of data from as many sources as we can tap. For some conditions epidemiological surveys are now proving informative but I doubt whether they have anything yet to tell us about this one. At present therefore we have to make do with data
120/362
? ? ? from our two familiar sources: (a) the analytic treatment of patients, (b) the direct observation of young children with their mothers.
As regards data obtained during treatment, I suspect it would be fruitful for some open- minded person to survey the psychoanalytic liter- ature and draw together all the case reports which record information about the childhood experiences of these patients. My guess is that, in so far as any information is given, it would strongly support Winnicott's view that these pa- tients have had disturbed childhoods in which in- adequate mothering in one form or another--and it can take many--bulks large. Since I have made no such survey, I can do no more than illustrate the kind of findings that I would confidently ex- pect. The following details come from case re- ports published by three analysts each much in- fluenced by Winnicott's views.
One report is by Donald Winnicott's widow Clare Winnicott (1980). The patient, a profes- sional woman of 41, presented a classical picture of the emotionally self-sufficient personality who recently had developed a variety of psychosomat- ic symptoms. Only after a good deal of analysis did she divulge the events of her childhood. Since
121/362
? ? ? her mother was in full-time work, she was looked after by a German girl who left suddenly when the patient was 21/2. Then, after six months of un- certainty, she was taken by her mother to have tea with a friend and later found her mother had disappeared and she was alone in a strange bed. Next day she was taken to the boarding school where her mother's friend worked as matron, and she stayed there till she was 9, usually spending the holidays there also. She appears to have settled in well (ominous words! ) and coped very successfully; but from that time forward her emo- tional life had dried up.
A second report, by Jonathan Pedder (1976), is of a young teacher in her mid-twenties whose personality and symptomatology bear a strong resemblance to Clare Winnicott's patient. Al- though at the initial interview she had given an idealized picture of her childhood, it soon emerged that at the age of 18 months she had been sent to stay with an aunt during her moth- er's next pregnancy. After six months there she had come to feel that her aunt was more of a mother to her than was her real mother and she had found returning home a painful experience. Thereafter, until she was 10, she had been
122/362
? ? ? terrified of another separation; but then she had 'switched off' her anxiety 'like a tap', as she put it, and with the anxiety had disappeared most of her emotional life as well.
The third report, by Elizabeth Lind (1973), con- cerns a young graduate of 23 who, though severely depressed and planning suicide, main- tained that his state of mind was less an illness than 'a philosophy of life'. He was the eldest of a large family; and by the time he was 3 two sib- lings had already been born. His parents, he said, quarrelled both frequently and violently. When the family was young, father had been working long hours away from home training for a profes- sion. Mother was always unpredictable. Often she was so distraught by her quarrelling children that she would lock herself in her room for days on end.
Several times she had left home, taking the daughters with her but leaving the sons behind.
He had been told that he had been an unhappy baby, a poor feeder and sleeper, who had often been left alone to cry for long periods. His crying, it was said, had been just an attempt to gain con- trol of his parents and to be spoilt. On one occa- sion he had had appendicitis and he remembered lying awake all night moaning; but his parents
123/362
? ? ? had done nothing and by next morning he was seriously ill. Later, during therapy, he recalled how disturbed he used to be at hearing his young- er brothers and sisters being left to cry and how he hated his parents for it and felt like killing them.
He had always felt like a lost child and had been puzzled to understand why he had been re- jected. His first day at school, he said, had been the worst in his life. It had seemed a final rejec- tion by his mother; all day he had felt desperate and had never stopped crying. After that he had gradually come to hide all desires for love and support: he had refused ever to ask for help or to have anything done for him.
Now, during therapy, he was frightened he might break down and cry and want to be mothered. This would lead his therapist, he felt sure, to regard him as a nuisance and his beha- viour simply as attention-seeking; and, were he to say anything personal to her, he fully expected her to be offended and perhaps lock herself in her room.
In all three cases the patient's recent break- down had followed the collapse of a significant but fragile relationship about which each
124/362
? ? ? member of the pair had had reservations and to the ending of which each of the patients had, him or herself, obviously contributed.
In treating these patients all three analysts ad- opted Winnicott's technique of permitting the free expression of what are traditionally termed 'dependency feelings', with the result that each patient in due course developed an intense and anxious attachment to his or her analyst (to use the terminology I prefer, (Bowlby, 1969, 1973)). This enabled each patient to recover the emotion- al life he or she had lost during childhood and with it to recover a sense of 'real self'. Therapeut- ically the results were good.
Admittedly, the findings from these three cases prove nothing. Nevertheless they are suggestive and, so far as they go, support Winnicott's theory of aetiology. Even so it is always open to critics to cast doubt on the validity of what a patient recalls about his childhood and to question whether the sequence of events recounted had the effect on his feeling life that he so explicitly claims. (It is worth nothing that the events that each of these three patients held to be a turning point had oc- curred after their second birthdays. )
125/362
? ? ? Now it seems clear that the controversies about aetiology are never going to be settled as long as we rely solely on the retrospective and perhaps biased evidence derived from the analyses of pa- tients, whether they be adults or children. What is needed is evidence of a different kind to provide some sort of cross-check. This is where I believe the direct observations of young children and their mothers are potentially so useful. Is there any evidence from that source that suggests that a child's feeling life can become numbed by the types of experience described? The answer, of course, is that there is a great deal.
Here naturally I point first to the observations made by James Robertson (1953) and confirmed later by Christoph Heinicke and Ilse Westheimer (1965) on how children between the ages of 12 and 36 months behave when removed from home to the care of strange people in a strange place, such as a residential nursery or hospital, with no one person to act as a mother-substitute. In such conditions a child comes in time to act as if neither mothering nor contact with humans has much significance for him. As his caretakers come and go he ceases to attach himself to any- one and after his return home stays remote from
126/362
? ? ? his parents for days, and perhaps for much longer if he is treated unsympathetically.
There is reason to believe, moreover, that a young child can develop this kind of defensive numbing in response to a mother who rejects him and without any major separation. Examples of this sequence are to be found in observations re- corded by Mahler (1971). More definitive findings are reported by Mary Main (1977), a colleague of Mary Ainsworth's, who has made a special study of a group of children in the age-range 12 to 20 months, each of whom not only failed to greet his mother after she had left him with a stranger for a few minutes but deliberately avoided her. View- ing some of Main's videotaped records I was as- tonished to see to what lengths some of these children went. One approached her mother briefly but with head averted and then retreated from her. Another, instead of approaching his mother, placed himself facing into the corner of the room, as though complying with a punish- ment, and then knelt down with his face to the floor. In every case videotaped records of these mothers playing with their toddlers during a later session showed them to differ from the mothers of non-avoidant toddlers: they appeared 'angry,
127/362
? ? ? inexpressive and disliking of physical contact with the infant'. Some scolded in angry tones, some mocked, others spoke sarcastically to or about their child. An obvious possibility is that by keeping away from his mother in this way a child is avoiding being treated in a hostile way again.
Thus, so far as the cross-checks provided by direct observations of young children and their mothers go, they tend to support a Winnicott- type theory. Put briefly, and in my own words, the child, and later the adult, becomes afraid to allow himself to become attached to anyone for fear of a further rejection with all the agony, the anxiety, and the anger to which that would lead. As a result there is a massive block against his ex- pressing or even feeling his natural desire for a close trusting relationship, for care, comfort, and love--which I regard as the subjective manifesta- tions of a major system of instinctive behaviour.
An explanation of this kind, although much less complex than some proposed in the literat- ure, accounts well for how these people behave both in the world at large and with ourselves as analysts. Inevitably they bring their fear of enter- ing into a trusting relationship with them to ana- lysis, which we experience as a massive
128/362
? ? ? resistance. Then, when at length their feelings are recovered, they more than half expect that we shall treat them as they recall being treated by their parents. In consequence they live in dread of being rejected and become intensely angry should they suspect us of deserting them. Not in- frequently, moreover, the way they treat us--with abuse and rejection--is found to be a version of the treatment to which they recall having them- selves been subjected as children.
You will see that in the explanation of how these patients behave during analysis I have ad- vanced a number of interlocking hypotheses. In a research programme each requires scrutiny and testing in the light of further data. Among the many methods that I would expect to prove of value is the study, in a therapeutic setting, of par- ents and children interacting with one another. In addition, there remains an important place for further observations to be made during the ana- lysis of individual patients; though I believe that, if clinical research is to yield its full potential, it has to be pursued in a far more systematic and directed way than hitherto.
To give an example: it would be of value were a detailed record to be kept of the responses of one
129/362
? ? ? or more of these patients before and after each successive weekend, each vacation, and each un- expected interruption of the sessions, with an equally detailed record of how the analyst dealt with them. This would enable us to know the rep- ertoire of responses a given patient presents on these occasions, and also the changes in response he presents over time. It would also be especially valuable if we were to have a detailed account of the conditions in which a major therapeutic change occurs. If, perhaps in a collaborative pro- gramme, records could be kept on a number of such patients, it might be possible to discover whether a frank and detailed discussion of the painful experiences a patient recalls having had in his relationships with his parents and the ef- fects these appear to have had and still to be hav- ing on the ways he treats other people, including of course ourselves, promotes therapeutic change, as I predict, or hinders it, as is believed by some analysts.
Naturally, in embarking on this or any other research programme an analyst must bear in mind his professional responsibilities; for with patients who present a false self these can be very onerous. Winnicott describes the 'period of
130/362
? ? ? extreme dependence' through which such pa- tients go during therapy and gives warning that 'analysts who are not prepared to go and meet the heavy needs of patients who become dependent in this way must be careful so to choose their cases that they do not include false self types'.
This brings me back to the art of therapy. To provide, by being ourselves, the conditions in which a patient of this kind can discover and re- cover what Winnicott calls his real self, and I call his attachment desires and feelings, is not easy. On the one hand, we have really to be trustworthy and we have also genuinely to respect all those yearnings for affection and intimacy that each of us has but which in these patients have become lost. On the other, we must not offer more than we can give and we must not move faster than the patient can bear. To achieve this balance requires all the intuition, imagination, and empathy of which we are capable. But it also requires a firm grasp of what the patient's problems are and what we are trying to do. This is why it is so very important that the problems of aetiology and psy- chopathology should be clarified as far as the ap- plication of scientific method makes possible and, further, that analysts should be thoroughly
131/362
? ? ? informed about the whole range of family experi- ences, from birth on through adolescence, that, increasing evidence shows, affect how a child's emotional life develops. Only when we become armed with that and much further knowledge shall we be in a position to meet the exacting de- mand that Freud makes in one of the last papers he wrote, in which he draws attention to 'the ker- nel of truth' in a patient's symptoms and to the therapeutic value of constructions in analysis (1937). In it he writes: 'What we are in search of is a picture of the patient's forgotten years that shall be alike trustworthy and in all essential re- spects complete. '
1 Efron (1977) has discussed the circumstances in which Freud's abrupt change of mind took place.
4
PSYCHOANALYSIS AS A NATURAL SCIENCE
In the autumn of 1980 I was appointed Freud Memorial Visiting Professor of Psychoanalysis at University College, London. In my inaugural lecture I returned to the theme I had spoken on in Canada two years previously. Having always believed that the body of knowledge labelled psychoanalysis should become a part of natural science, I was distressed by the pressure of the opposition. To accept that psychoanalysis should abandon its aim of becoming a natural science and instead should regard itself as a hermeneut- ic discipline has seemed to me to be not only a result of obsolete ideas about science but also a counsel of despair; because, in a hermeneutic discipline, there are no criteria by the applica- tion of which it is ever possible to resolve disagreement.
A problem encountered by every analyst who has proposed new theoretical ideas is the
? 133/362
? ? ? criticism that the new theory is 'not psychoana- lysis'. Such criticism turns, of course, on how we define psychoanalysis. Most unfortunately, de- fining it in terms of Freud's theories is all too common. This is in contrast to the definitions ad- opted by academic disciplines which are always in terms of the phenomena to be studied and the problems to be solved. In such disciplines pro- gress is frequently signalled by changes of the- ory, sometimes of a revolutionary kind. So long as analysts continue to define psychoanalysis in terms of any particular theory, they must not complain that their discipline is cold-shouldered by academics. Furthermore by so defining it they are condemning it to frozen inertia.
The following version of this lecture differs in a number of ways from the original, in particu- lar by abbreviating the discussion of issues already dealt with in the preceding lectures.
From 1895, when Freud made his first attempt to sketch a theoretical framework for psychoanalys- is, until 1938, the year before he died, Freud was determined that his new discipline should con- form to the requirements of a natural science. Thus, in the opening sentence of his Project he writes: 'The intention is to furnish a psychology
134/362
? ? ? that shall be a natural science. . . . ' (Freud, 1950, 295); whilst in the Outline we find a passage in which he asserts that, once the concept of psych- ical processes being unconscious is granted, 'psy- chology is enabled to take its place as a natural science like any other' (Freud, 1940, 158).
Admittedly during the intervening years Freud's ideas about the scope of his science had changed considerably from his early ambition 'to represent psychical processes as quantitatively determinate states of specifiable material particles' (1950, written in 1895) to his later definition of psychoanalysis as 'the science of un- conscious mental processes' (1925). But from first to last there can be no doubt what sort of discipline Freud intended psychoanalysis to be.
Nevertheless, despite Freud's unwavering in- tention, the scientific status of psychoanalysis re- mains equivocal. On the one hand philosophers of science have dubbed it a pseudoscience on the grounds that, however large a measure of truth they may contain, psychoanalytic theories are cast in so elastic a form that they are unfalsifi- able. On the other many psychoanalysts, disillu- sioned by the inadequacies of Freud's metapsy- chology and preoccupied with the personal
135/362
? ? ? perspective which is unquestionably required in clinical work, have abandoned Freud's aims and claims and have declared that psychoanalysis is miscast as a science and should be conceived in- stead as one of the humanities, e. g. Home (1966), Ricoeur (1970), and others in Europe. Both Schafer (1976) and George Klein (1976), espous- ing this view, have advanced proposals alternat- ive to Freud's: but each of their reformulations, different though they be, seems a version of Hamlet without the Prince. Gone are all concepts of causality and theories of biologically rooted impulse and, in Schafer's version, gone also are concepts of repression and unconscious mental activity.
Melanie Klein has made very different propos- als, ones which certainly do not suffer from these defects; but it would be difficult to claim that the form they take or the research they have en- gendered meet scientific requirements.
Yet by no means do all analysts despair of de- veloping their discipline as a natural science. Alive to the deficiencies of Freud's metapsycho- logy, especially his concepts of psychic energy and drive, a few are attempting to replace it with a new conceptual framework consistent with
136/362
? ? ? current scientific thinking. Central to these new proposals are ideas drawn from systems theory and the study of human information processing. Those active in this enterprise include Rubinstein (1967), Peterfreund (1971, 1982), Rosenblatt and Thickstun (1977), Gedo (1979), and myself (1969, 1980). Meanwhile there are also a number of analysts who have been seeking to extend the dis- cipline's database by studying children's social and emotional development using direct observa- tions. Some of these studies have been atheoretic- al, e. g. Offer (1969). The authors of others have attempted to put new empirical wine into the old theoretical bottles, e. g. Spitz (1957), Mahler (Mahler, Pine, and Bergman, 1975); whilst others again, e. g. Sander (1964, 1980), Stern (1977), and myself (1958, 1969, 1973), have searched for new theoretical models. My own search has led not only to control theory and information pro- cessing but also to the biologically rooted discip- lines of ethology and comparative psychology. Thus there is no lack of new initiatives and it will take time to see which of them, or perhaps which combination of them, proves most productive of scientific advance. Here my aim is to describe one
137/362
? ? ? such initiative, my own, and why I think it promising.
In the preceding lecture I described the cir- cumstances which led me to choose as a field of research the responses of a young child to being removed from his or her mother and placed for a time in a strange place with strange people, and how these observations led on to the formulation of attachment theory. Among essential features of this are that the human infant comes into the world genetically biased to develop a set of beha- vioural patterns that, given an appropriate envir- onment, will result in his keeping more or less close proximity to whomever cares for him, and that this tendency to maintain proximity serves the function of protecting the mobile infant and growing child from a number of dangers, amongst which in man's environment of evolu- tionary adaptedness the danger of predation is likely to have been paramount.
A concept that emerged early from ethologic- ally oriented studies of mother-child relation- ships (Ainsworth, 1967) and which has proved of great clinical value is that of a mother, or mother- substitute, providing a child with a secure base from which he can explore. By the last months of
138/362
? ? ? the first year of life an infant brought up in an or- dinary affectionate home is very clear whom he prefers to care for him, a preference especially evident should he be tired, frightened, or sick. Whoever that may be, and it is usually his moth- er, is then able by her very presence, or ready ac- cessibility, to create the conditions which enable him to explore his world in a confident way. At the time of his second birthday, for example, a healthy child whose mother is resting on a garden seat will make a series of excursions away from her, each time returning to her before making the next excursion. On some occasions, when he re- turns, he simply smiles and makes his number; on others he leans against her knee; on yet others he wants to climb on her lap. But never does he stay for long unless he is frightened or tired or thinks she is about to leave. Anderson (1972), who made a study of this sort in a London park, observed that during the second and third years it is very rare for a child to go further than two hundred feet before returning. Should he lose sight of his mother, exploration is forgotten. His top priority then is to regain her, in an older child by searching and in a younger one by howling.
139/362
? ? ? It is evident that there is no way of explaining this type of behaviour in terms of a build-up of psychic energy which is then discharged. An al- ternative model (already described in earlier lec- tures) is to think of the proximity-keeping of a child as being mediated by a set of behavioural systems organized cybernetically. Activation is intensified in conditions of pain, fatigue, and anything frightening; and reduced by proximity to or contact with the mother-figure. We can then postulate that the behaviour that takes him away from his mother into the wide world, which is conveniently termed exploratory behaviour, is in- compatible with attachment behaviour and has a lower priority. It is thus only when attachment behaviour is relatively inactive that exploration occurs.
As an individual grows older his life continues to be organized in the same kind of way though his excursions become steadily longer both in time and space. On entering school they will last for hours and later for days. During adolescence they may last for weeks or months, and new at- tachment figures are likely to be sought. Throughout adult life the availability of a re- sponsive attachment figure remains the source of
140/362
? ? ? a person's feeling secure. All of us, from the cradle to the grave, are happiest when life is or- ganized as a series of excursions, long or short, from the secure base provided by our attachment figure(s).
In terms of the theoretical model proposed, the pronounced changes in the organization of at- tachment behaviour that occur during individual development are regarded as being due, in part, to the threshold for its activation being raised (perhaps through changes in endocrine levels) and, in part, to the control systems becoming in- creasingly sophisticated, in particular by their coming to incorporate representational models of the environment and important people in it and also of the self as a living active person.
The development during ontogeny of a set of systems of the kind described in humans, as well as in individuals of many other species, is attrib- uted to the action of natural selection, namely to individuals well endowed with the potential to develop such systems having survived and bred more successfully than those less well endowed, in other words to Darwinian evolution. Since a disposition to show attachment behaviour in cer- tain circumstances is regarded as an intrinsic
141/362
? ? ? part of human nature, reference to it as 'depend- ency' is not only misleading but seriously inap- propriate because of the word's pejorative overtones.
Once attachment behaviour and other forms of biologically determined behaviour are conceived in forms of control theory, the problem of the purposiveness of behaviour is solved without abandoning the concept of causation. Further- more the distinction between causation and func- tion, sadly neglected in traditional psychoanalytic theory, becomes explicit. Activation, in which emotional arousal and expression play a critical part, and subsequently termination and change of emotional state are caused when a system con- structed in a particular way receives information of particular sorts. Of the various consequences to which activation leads, the one postulated as its biological function is the one that, evidence suggests, has led to the system having evolved during phylogeny. In the case of attachment be- haviour the function postulated is that of dimin- ishing the risk of the individual coming to harm.
At this level of analysis the question of whether an individual is aware of what he is doing, let alone why he is doing it, has no relevance, in fact
142/362
? ? ? no more relevance than has the question of whether an individual is aware that he is breath- ing and, if so, realizes why he should be doing so. Biological systems serving vital functions, wheth- er at a behavioural or a physiological level, must be capable of operating automatically. Neverthe- less in the case of a human child awareness of what he is doing, and more particularly aware- ness of the conditions that will terminate his be- haviour, soon emerges, certainly by the end of his first year, and is a factor of great importance. For, once it is clear that a child is aware of the condi- tions that will terminate his behaviour, we begin speaking of intention, of his desire to achieve a certain goal, of his being satisfied and happy when he does so and frustrated, anxious, or angry when he fails, of satisfaction bringing pleasure and frustration the reverse.
At this point I wish to emphasize the sharp dis- tinction between conditions necessary to termin- ate a certain form of behaviour, commonly re- ferred to as its goal, and the biological function the behaviour serves. In the case of attachment behaviour in childhood, whereas we commonly expect both mother and child to be aware of the conditions necessary to terminate it, for example,
143/362
?
