common and have, until recently, been neglected in the
psychotherapeutic
literature.
A-Secure-Base-Bowlby-Johnf
For someone who had a happy childhood no obstacles are likely to prevent free access to both the emotional and the cognitive aspects of such information.
For someone who suffered much unhappiness or whose parents forbade him or her to notice or to remember adverse events, ac- cess is painful and difficult, and without help may indeed be impossible.
Nevertheless, however she may accomplish it, when a woman manages either to retain or to regain access to such un- happy memories and reprocess them in such a way that she can come to terms with them, she is found to be no less able to respond to her child's attachment behaviour so that he develops a
273/362
? ? ? secure attachment to her than a woman whose childhood was a happy one. This is a finding to give great encouragement to the many therapists who for long have sought to help mothers in just this kind of way. Further reference to techniques for helping disturbed mothers is made at the end of Lecture 8.
PATHWAYS TO PERSONALITY DEVELOPMENT
There is one further way in which attachment theory differs from traditional types of psycho- analytic theory, namely its rejection of the model of development in which an individual is held to pass through a series of stages in any one of which he may become fixated or to which he may regress, and its replacement by a model in which an individual is seen as progressing along one or another of an array of potential developmental pathways. Some of these pathways are compat- ible with healthy development; others deviate in one or another direction in ways incompatible with health.
All variants of the traditional model invoking phases of development are based on the assump- tion that, at some phase of normal development, a child shows psychological features that, in an
274/362
? ? ? older individual, would be regarded as signs of pathology. Thus a chronically anxious and cling- ing adult might be regarded as being fixated in or having regressed to a postulated phase of orality or of symbiosis; whilst a deeply withdrawn indi- vidual might be regarded as having regressed to a postulated phase of autism or of narcissism. Sys- tematic and sensitive studies of human infants, such as those reported by Stern (1985), have now rendered this model untenable. Observations show that infants are socially responsive from birth onwards. Healthily developing toddlers do not show anxious clinging except when they are frightened or distressed; at other times they ex- plore with confidence.
The model of developmental pathways regards an infant at birth as having an array of pathways potentially open to him, the one along which he will in fact proceed being determined at every moment by the interaction of the individual as he now is with the environment in which he happens then to be. Each infant is held to have his own in- dividual array of potential pathways for personal- ity development which, except for infants born with certain types of neurological damage, in- clude many that are compatible with mental
275/362
? ? ? health and also many that are incompatible. Which particular pathway he proceeds along is determined by the environment he meets with, especially the way his parents (or parent substi- tutes) treat him, and how he responds to them. Children who have parents who are sensitive and responsive are enabled to develop along a healthy pathway. Those who have insensitive, unrespons- ive, neglectful, or rejecting parents are likely to develop along a deviant pathway which is in some degree incompatible with mental health and which renders them vulnerable to breakdown, should they meet with seriously adverse events. Even so, since the course of subsequent develop- ment is not fixed, changes in the way a child is treated can shift his pathway in either a more fa- vourable direction or a less favourable one. Al- though the capacity for developmental change di- minishes with age, change continues throughout the life cycle so that changes for better or for worse are always possible. It is this continuing potential for change that means that at no time of life is a person invulnerable to every possible ad- versity and also that at no time of life is a person impermeable to favourable influence. It is this
276/362
? ? ? persisting potential for change that gives oppor- tunity for effective therapy.
1 In earlier publications I have sometimes used the term 'representational model' as a synonym for 'work- ing model' because representation has been the more familiar concept in clinical literature. In a dynamic psychology, however, working model is the more ap- propriate term, and it is also the term that is now com- ing into use among cognitive psychologists (e. g. Johnson-Laird, 1983). Within the attachment frame- work the concept of working model of an attachment figure is in many respects equivalent to, and replaces, the traditional psychoanalytic concept of internal object.
2 In further examination of the data it has been found that all these correlations also hold true for fathers (Main, personal communication).
8
ATTACHMENT, COMMUNICATION, AND THE THERAPEUTIC PROCESS
In the second part of my 1976 Maudsley Lecture, 'The making and breaking of affectional bonds' (1977), I described some of my ideas on the therapeutic implications of attachment theory. Much that has been learned since then has strengthened my confidence in the approach. The present account therefore should be re- garded as an amplification of the earlier one. In it I give more detailed attention to the ways a patient's earlier experiences affect the transfer- ence relationship and discuss further the therap- ist's aim as being that of enabling his patient to reconstruct his working models of himself and his attachment figure(s) so that he becomes less under the spell of forgotten miseries and better able to recognize companions in the present for what they are.
? 278/362
? ? ? a thing which has not been understood inevitably reappears; like an unlaid ghost, it cannot rest until the mystery has been resolved and the spell broken.
Sigmund Freud 1909
Those who cannot remember the past are condemned to repeat it. George Santayana 1905
FIVE THERAPEUTIC TASKS
The theory of personality development and psy- chopathology outlined above can be used as a framework to guide each one of the three princip- al forms of analytic psychotherapy in use today--individual therapy, family therapy, and group therapy. Here I deal only with the first.
A therapist applying attachment theory sees his role as being one of providing the conditions in which his patient can explore his representational models of himself and his attachment figures with a view to reappraising and restructuring them in the light of the new understanding he ac- quires and the new experiences he has in the therapeutic relationship. In helping his patient towards this end the therapist's role can be de- scribed under five main heads.
The first is to provide the patient with a secure base from which he can explore the various un- happy and painful aspects of his life, past and
279/362
? ? ? present, many of which he finds it difficult or per- haps impossible to think about and reconsider without a trusted companion to provide support, encouragement, sympathy, and, on occasion, guidance.
A second is to assist the patient in his explora- tions by encouraging him to consider the ways in which he engages in relationships with significant figures in his current life, what his expectations are for his own feelings and behaviour and for those of other people, what unconscious biases he may be bringing when he selects a person with whom he hopes to make an intimate relationship and when he creates situations that go badly for him.
A particular relationship that the therapist en- courages the patient to examine, and that consti- tutes the third task, is the relationship between the two of them. Into this the patient will import all those perceptions, constructions, and expecta- tions of how an attachment figure is likely to feel and behave towards him that his working models of parents and self dictate.
A fourth task is to encourage the patient to consider how his current perceptions and expect- ations and the feelings and actions to which they
280/362
? ? ? give rise may be the product either of the events and situations he encountered during his child- hood and adolescence, especially those with his parents, or else as the products of what he may repeatedly have been told by them. This is often a painful and difficult process and not infrequently requires that the therapist sanction his patient to consider as possibilities ideas and feelings about his parents that he has hitherto regarded as un- imaginable and unthinkable. In doing so a pa- tient may find himself moved by strong emotions and urges to action, some directed towards his parents and some towards the therapist, and many of which he finds frightening and/or alien and unacceptable.
The therapist's fifth task is to enable his patient to recognize that his images (models) of himself and of others, derived either from past painful experiences or from misleading messages eman- ating from a parent, but all to often in the literat- ure mislabelled as 'fantasies', may or may not be appropriate to his present and future; or, indeed, may never have been justified. Once he has grasped the nature of his governing images (mod- els) and has traced their origins, he may begin to understand what has led him to see the world and
281/362
? ? ? himself as he does and so to feel, to think, and to act in the ways he does. He is then in a position to reflect on the accuracy and adequacy of those im- ages (models), and on the ideas and actions to which they lead, in the light of his current experi- ences of emotionally significant people, including the therapist as well as his parents, and of himself in relationship to each. Once the process has star- ted he begins to see the old images (models) for what they are, the not unreasonable products of his past experiences or of what he has repeatedly been told, and thus to feel free to imagine altern- atives better fitted to his current life. By these means the therapist hopes to enable his patient to cease being a slave to old and unconscious stereo- types and to feel, to think, and to act in new ways.
Readers will be aware that the principles set out have a great deal in common with the prin- ciples described by other analytically trained psy- chotherapists who regard conflicts arising within interpersonal relationships as the key to an un- derstanding of their patient's problems, who fo- cus on the transference and who also give some weight, albeit of varying degree, to a patient's earlier experience with his parents. Among the many well-known names that could be
282/362
? ? ? mentioned in this context are those of Fairbairn, Winnicott, and Guntrip in Britain, and Sullivan, Fromm-Reichmann, Gill, and Kohut in the Un- ited States. Among recently published works that contain many of the ideas prescribed here are those by Peterfreund (1983), Casement (1985), Pine (1985), and Strupp and Binder (1984), and also those of Malan (1973) and Horowitz et al. (1984) in the field of brief psychotherapy. In par- ticular, I wish to draw attention to the ideas of Horowitz and his colleagues who, in their de- scription of the treatment of patients suffering from an acute stress syndrome, employ a concep- tual framework closely similar to that presented here. Although their technique is aimed to help patients recover from the effects of a recent severely stressful event, I believe the principles informing their work are equally applicable to helping patients recover from the effects of a chronic disturbance resulting from stressful events of many years ago, including those that oc- curred during their earliest years.
Although in this exposition it is convenient to list the therapist's five tasks in a logical way, so inter- related are they that in practice a productive ses- sion is likely to involve first one task, then
283/362
? ? ? another. Nevertheless, unless a therapist can en- able his patient to feel some measure of security, therapy cannot even begin. Thus we start with the role of the therapist in providing his patient with a secure base. This is a role very similar to that described by Winnicott as 'holding' and by Bion as 'containing'.
In providing his patient with a secure base from which to explore and express his thoughts and feelings the therapist's role is analogous to that of a mother who provides her child with a se- cure base from which to explore the world. The therapist strives to be reliable, attentive, and sympathetically responsive to his patient's ex- plorations and, so far as he can, to see and feel the world through his patient's eyes, namely to be empathic. At the same time he is aware that, be- cause of his patient's adverse experiences in the past, the patient may not believe that the therap- ist is to be trusted to behave kindly or to under- stand his predicament. Alternatively the unex- pectedly attentive and sympathetic responses the patient receives may lead him to suppose that the therapist will provide him with all the care and affection which he has always yearned for but never had. In the one case therefore the therapist
284/362
? ? ? is seen in an unduly critical and hostile light, in the other as ready to provide more than is at all realistic. Since, it is held, both types of misunder- standing and misconstruction, and the emotions and behaviour to which they give rise, are central features of the patient's troubles, a therapist needs to have the widest possible knowledge of the many forms these misconstructions can take and also of the many types of earlier experience from which they are likely to have sprung. Without such knowledge a therapist is poorly placed to see and feel the world as his patient is doing.
Even so, a patient's way of construing his rela- tionship with his therapist is not determined solely by the patient's history: it is determined no less by the way the therapist treats him. Thus the therapist must strive always to be aware of the nature of his own contribution to the relationship which, amongst other influences, is likely to re- flect in one way or another what he experienced himself during his own childhood. This aspect of therapy, the counter-transference, is a big issue of its own and the subject of a large literature. Since it is not possible to deal with it further here, I want to emphasize not only the importance of
285/362
? ? ? the counter-transference but also that the focus of therapy must always be on the interactions of patient and therapist in the here and now, and that the only reason for encouraging the patient at times to explore his past is for the light it throws on his current ways of feeling and dealing with life.
With that proviso firmly in mind, let us con- sider some of the commoner forms that a pa- tient's misconstructions can take and how they are likely to have originated. This is the aspect of therapy in which the work of a therapist who ad- opts attachment theory is likely to differ most from one who adopts certain of the traditional theories of personality development and psycho- pathology. Thus, for example, a therapist who views his patient's misperceptions and misunder- standings as the not unreasonable products of what the patient has actually experienced in the past, or has repeatedly been told, differs sharply from one who sees these same misperceptions and misunderstandings as the irrational offspring of autonomous and unconscious fantasy.
In what follows I am drawing on several dis- tinct sources of information: studies by epidemi- ologists; the studies by developmental
286/362
? ? ? psychologists already referred to; observations made during the course of family therapy; and not least what I have learned from patients whom I have treated myself and from those whose ther- apy I have supervised.
INFLUENCE OF EARLIER EXPERIENCES ON THE TRANSFERENCE RELATIONSHIP
It not infrequently happens that a patient is acutely apprehensive lest his therapist reject, cri- ticize, or humiliate him. Since we know that all too many children are treated in this way by one or other, or both, of their parents, we can be reas- onably confident that that has been our patient's experience. Should it seem likely that the patient is aware of how he is feeling and how he expects the therapist to treat him, the therapist will indic- ate that he also is aware of the problem. How soon the therapist can link these expectations to the patient's experiences of his parents, in the present perhaps as well as the past, turns on how willing the patient is to consider that possibility, or whether, by contrast, he insists that his par- ents' treatment of him is above criticism. Where the latter situation obtains, there is the prior problem of trying to understand why the patient
287/362
? ? ? should insist on retaining this favourable picture when such evidence as is available points to its being mistaken.
It happens in some families that one or other parent insists that he or she is an admirable par- ent who has always done everything possible for the child and that, in so far as friction is present, the fault lies exclusively with the child. This atti- tude of the parent all too often cloaks behaviour that, by ordinary standards, has been far from perfect. Yet, since the parent insists that he or she has given the child constant affection and that the child must have been born bad and ungrateful, the child has little option but to accept the pic- ture, despite being aware somewhere in his mind that the picture is hardly fair.
An added complication arises when a patient has, as a child, been subjected to the strongest of instructions from a parent on no account to tell anyone of certain happenings within the family. These are usually quarrels in which the parent is aware that his or her behaviour is open to criti- cism; for example, quarrels between the parents, or between a parent and a child, during which dreadful things have been said or done. The more insistent a therapist is that his patient tell
288/362
? ? ? everything, the more distressing the dilemma is for his patient. Injunctions to silence are not un- common in families and have been much neg- lected as sources of what has traditionally been called resistance. It is often useful for a therapist to enquire of a patient whether he may have been subjected to such pressures and, if so, to help him resolve the dilemma.
So far we have been considering cases in which a patient is in some degree aware of his expecta- tions of being rejected, criticized, or humiliated. Not infrequently, however, a patient seems wholly unaware of any such feelings despite his attitude to the therapist exuding distrust and eva- sion. Evidence shows that these states of mind occur especially in those who, having developed an anxiously avoidant pattern of attachment dur- ing early years, have striven ever since to be emotionally self-contained and insulated against intimate contacts with other people. These pa- tients, who are often described as being narciss- istic or as having a false self, avoid therapy as long as they can and, should they undertake it, keep the therapist at arm's length. If allowed to, some will talk incessantly about anything and everything except emotionally charged
289/362
? ? ? relationships, past or present. Others will explain that they have nothing to talk about. One young woman, whose every move indicated deep dis- trust of me, spent the time boasting of her delin- quent exploits, many of them fictitious I suspec- ted, and pouring contempt on what she insisted was my dull and narrow life. To treat such deeply distrustful people was compared many years ago by Adrian Stephen (1934) with trying to make friends with a shy or frightened pony: both situ- ations require a prolonged, quiet, and friendly patience. Only when the therapist is aware of the constant rebuffs the patient is likely to have been subjected to as a child whenever he sought com- fort or help, and of his terror of being subjected to something similar from the therapist, can the latter see the situation between them as his pa- tient is seeing it.
Another and quite different cause of wariness of any close contact with a therapist for the pa- tient is dread lest the therapist trap him into a re- lationship aimed to serve the therapist's interests rather than his own. A common origin of such fear is a childhood in which a parent, almost al- ways mother, has sought to make the child her own attachment figure and caregiver, that is, has
290/362
? ? ? inverted the relationship. Very often this is done unconsciously and using techniques that, to an uninformed eye, may appear to be overindul- gence but that are really bribes to retain the child in a caregiving role.
Not infrequently a patient shifts during the therapy from treating his therapist as though he was one or other of his parents to behaving to- wards him in the way one of his parents had treated him. For example, a patient who has been subjected to hostile threats as a child may use hostile threats to his therapist. Experiences of scornful contempt from a parent may be re-en- acted as scornful contempt of the therapist. Sexu- al advances from a parent may reappear as sexual advances to the therapist. Such behaviour may be understood in the following way. During his childhood a person learns two principal forms of behaviour and builds in his mind two principal types of model. One form of behaviour is, of course, that of a child, namely himself, interact- ing with a parent, his mother or his father. The corresponding working models he builds are those of himself as a child in interaction with each parent. The other form of behaviour is that of a parent, namely his mother or his father,
291/362
? ? ? interacting with a child, himself. The correspond- ing models he builds are those of each parent in interaction with himself. Therefore, whenever a therapist is puzzled by, or resentful of, the way he is being treated by a patient, he is always wise to enquire when and from whom the patient may have learned that way of treating other people. More often than not it is from one of his parents. 1
With some patients the therapeutic relation- ship is one in which anxiety, distrust, and criti- cism, and sometimes also anger and contempt, are overt and predominate, and the therapist seen in dark colours. Such sentiments as gratit- ude for the therapist's efforts or respect for his competence are conspicuous by their absence. The task then is to help the patient grasp that much of his present resentment stems from past mistreatment at the hands of others and that, however understandable his anger may be as a result, to continue fighting old battles is unpro- ductive. To accept that an unhappy past cannot be changed is usually a bitter pill.
With other patients the situation is reversed: the transference relationship becomes one in which overt gratitude, admiration, and affection are readily expressed, and the therapist seen in a
292/362
? ? ? glow of rosy perfection. Dissatisfaction and criti- cism are notably absent, and anger at the therap- ist's shortcomings, especially absences, unima- ginable. Such idealization of the therapist springs, I believe, partly from unrealistic hopes and expectations of what the therapist is able and willing to provide, and partly from a childhood in which criticism of a parent is forbidden and com- pliance enforced, either by some guilt-inducing technique or else by sanctions such as threats not to love, or even to abandon, the child. With this type of childhood experience, the patient's un- conscious assumption is that the therapist will expect the same degree of obedience as his par- ents had expected and will enforce it by tech- niques or by threats similar to the ones they used.
Unfortunately there has been a tendency in some quarters to confuse the theory advanced here, which regards the way certain parents treat their children as being a major cause of mental ill- health, with an attitude of mind that simply blames parents. No one who works in the fields of child psychiatry and family therapy is likely to make this mistake. On the contrary, as already remarked, it has long been recognized that the misguided behaviour of parents is more often
293/362
? ? ? than not the product of their own difficult and unhappy childhood. As a result much skilled time has always been devoted to helping parents es- cape from the adverse influences of their own childhoods.
Furthermore, during the treatment of an indi- vidual (of any age) who has suffered at the hands of his parents, the therapist, whilst accepting the patient's account, avoids moral judgement. On the contrary, whenever opportunity offers, he will encourage his patient to consider how and why the parent under discussion may have behaved as he or she has done. In raising these questions, it is always useful for the therapist to enquire of the patient what he knows of the childhood experi- ences that the parent in question may have had. Not infrequently this leads the patient to gain some understanding of how things had developed and, from understanding, often to move on to a measure of forgiveness and reconciliation. In family sessions it can be especially valuable if a parent can be encouraged to give an account of his or her childhood. This enables all those present--the parent him- or herself, the spouse, the children, and the therapist--to gain some in- sight into how and why family life has developed
294/362
? ? ? as it has and how each can best help improve it. As already mentioned (p. 150), this strong tend- ency for attachment problems to be transmitted across generations, through the influence on par- enting behaviour of relationship problems stem- ming from the parent's own childhood, is at last receiving the research attention it deserves.
SOME PATHOGENIC SITUATIONS AND EVENTS OF CHILDHOOD
A therapist, I believe, cannot be too well in- formed about the disguised and distorted rela- tionships that can occur in some families, and the terrible things that can happen in others, for it is only if he is so informed that he can have a reas- onably clear idea of what probably lies behind his patient's defences, or of the origins of his anxiety, anger, and guilt. Once he is adequately informed, he is well placed not only to appreciate the truth of what his patient may be describing as having happened to him but also to broach, more or less tentatively, some of the kinds of situation to which the patient may well have been exposed but to which he may be either unable or unwilling to refer. In listing the following situations I am doing no more than indicating some that are both
295/362
? ? ?
common and have, until recently, been neglected in the psychotherapeutic literature. 2
Threats not to love a child used as a means of control
It is easy for a mother to say to a child that she will not love him if he behaves in such and such a way. What this means is that the mother is threatening not to provide affection or comfort at times when her child is upset, frightened, or dis- tressed, and not to provide help or encourage- ment at other times. If such threats are used sys- tematically by either parent, or both, the child in- evitably grows up intensely anxious to please and guilt-prone.
Threats to abandon a child
Threats to abandon are a degree more frightening to a child than threats no longer to love him. This is especially so if the parent enacts a threat, per- haps by disappearing herself for a few hours or by packing her child's suitcase and walking him up the street, allegedly to the home for bad boys. Since threats to abandon often take a highly idio- syncratic form, a patient may deny that he was ever subjected to them. In such cases the truth
296/362
? ? ? may emerge with its accompanying emotion only when the special way in which the threat was phrased is recollected. One example is a mother who had concocted the story that a yellow van would draw up and take her son away. Another is a father whose story was that his daughter would be sent to a school on a remote rock surrounded by sharks (Marrone, 1984). Thus, in the first case, all the mother had to say was 'Well the yellow van will come', and, in the second, for the father to say 'Then it will be the rock school', for the child instantly to desist from whatever he or she was doing. In a third case, the code word was 'mar- garine', the mother having coupled her threat to send her son away to a children's home with an insistence that he would have to eat margarine there. For these patients a general phrase like 'threat to abandon' had failed to ring a bell. Only when the code word was unearthed was the ori- ginal terror experienced afresh and the source of the separation anxiety located.
Threats to commit suicide
Sometimes a distraught parent threatens to com- mit suicide if some distressing situation contin- ues. This may occur during quarrels between
297/362
? ? ? parents, which the child overhears, or may be dir- ected at the child himself. In either case such threats strike terror. One lesson to be learned from these cases is that, whenever a patient refers to his parents as having quarrelled, the therapist should always enquire 'What did they say to each other? ' Not infrequently a patient blocks at this. In a fit of temper quarrelling parents may say ap- palling things to each other. This is bad enough. What makes it far worse is when, after cooling down, they disclaim having said any such thing.
Disclaimers and disconfirmations
Examples of disclaimers by a parent of what he or she has said or done, and persistent efforts to dis- confirm what a child has seen or heard, are given in some detail in Lecture 6 and the adverse ef- fects on personality development of such pres- sures emphasized. During therapy these effects emerge as great uncertainty in a patient as to whether some family episode did or did not occur and guilt about adumbrating it. Here, as so often, a key role for the therapist is to sanction the pa- tient's exploration of all the various possibilities, both those favourable to his parents and those unfavourable to them, and to encourage him to
298/362
? ? ? weigh the evidence available, whilst he (the ther- apist) remains resolutely open-minded as to where the truth may lie.
Thus far in this exposition I have not considered the critical issue of how far we can and should rely on the validity of our patients' reports. Memories are certainly fallible, and there are various occasions when an experienced therapist will rightly question the truth of what his patient is saying. What then are the criteria by which we should judge?
First, broad generalizations about the kind of parent the mother or father was and about the kind of parenting received are never to be given credence unless and until they are supported by detailed examples of how each treated the patient as a child in particular situations. For example, the glowing account of a wonderful mother may well go unsupported when detail becomes avail- able. Valid accounts of affectionate parenting not only give plenty of favourable details but are likely also to be interlaced with occasional criti- cism, so that the parent can be seen in the round. Similarly, disparaging accounts of parents in uni- formly adverse terms need close examination. In- valid accounts of either sort tend to be sweeping
299/362
? ? ? and extreme, to be either white or black. Detail is either lacking or, should it be given, is at variance with the portrait presented. By contrast, whenev- er plenty of consistent detail is given and the pic- ture that emerges conforms to what we know, from other sources, does happen in other famil- ies, and also to the known antecedents of the types of problem besetting the patient, it is ab- surd to doubt its overall validity, even if some points remain in question.
The origin of these extremes is not infrequently external pressure. For example, one parent has insisted that the child take sides with him or her against the other parent who is represented as being all bad. Or else a parent who has many shortcomings insists that he or she is above criticism.
Another occasion when a therapist is right to doubt the patient's story is when there is reason to suspect that the patient is a pathological liar. Such cases are comparatively rare and, if only for that reason, may go undetected for a time. Sooner or later, however, mounting inconsistencies and improbabilities, as well as the way the patient tells his story, engender first doubt and, later, certainty that the patient is not to be believed.
300/362
? ? ? Apart from these exceptions I believe patients' accounts are sufficiently trustworthy that a ther- apist should accept them as being reasonable ap- proximations to the truth;3 and furthermore that it is anti-therapeutic not to do so. Constantly to query the validity of the patient's story, even if only by implication, and insistence on the distort- ing role of imagination or fantasy, is the reverse of empathic. It conveys to the patient that the therapist
does not understand him and may indeed con- vince him that the therapist is behaving exactly as his parent had predicted. Thus some parents, having insisted that their child not tell of something the parent is ashamed of, may then add that, in any case, were he to do so no one would believe him.
Among the large range of adverse events and situations not so far mentioned in this lecture that a therapist should have in his mind as likely to have occurred in the life of one patient or an- other are the following:
a child may never have been wanted by one or both parents;
a child may be of the wrong sex in a family in which parents had set their hearts on a boy or a girl;
301/362
? ? ? a child may have been made the family scapegoat, sometimes as a result of a family tragedy that, with greater or less plausibility, has always been at- tributed to him;
a parent may have used guilt-inducing techniques to control a child, for example, frequent claims that the child's behaviour makes mother ill;
a parent may have sought to make one of her children her attachment fig- ure by discouraging him from exploring the world away from her and from believing that he will ever be able to make his way on his own;
a child's unusual role in a family may be the result of his mother having had an extra-marital affair during her marriage so that the child's putative father is not his real father;
another cause of a child's unusual role is when one or other parent identi- fies one child with a relative, often one of the child's grandparents, with whom he or she has had a difficult relationship, and who then re-enacts that relationship with the child;
a child may have been the target of more or less serious physical abuse from a parent or step-parent;
a child may have been involved in sexual abuse from a parent, step-par- ent, or older sibling for short or long periods of time.
For those unaware of the commoner effects on personality development of exposure to situ- ations of these kinds, a number of references are given in an Appendix to references (pp. 194-5).
Inevitably the influential events of an individu- al's first two or three years will either never have been registered in his memory or else cannot now be recalled. Here, of course, the best a therapist can do is to infer, on the basis of the transference situation and of such information as the patient has gleaned about his early years, combined with
302/362
? ? ? such wider knowledge of personality develop- ment as the therapist has acquired, what the nature of those events may have been. In other words, he resorts to reconstruction; but in doing so he can in future draw on a much wider and more reliable knowledge of family influences on personality development than has traditionally been available to analytically trained psychotherapists.
THE THERAPIST'S STANCE
In this account of therapeutic principles, therap- ists will recognize much that has long been famil- iar, though often under a different name. The therapeutic alliance appears as a secure base, an internal object as a working, or representational, model of an attachment figure, reconstruction as exploring memories of the past, resistance (sometimes) as deep reluctance to disobey the past orders of parents not to tell or not to remem- ber. Among points of difference is the emphasis placed on the therapist's role as a companion for his patient in the latter's exploration of himself and his experiences, and less on the therapist in- terpreting things to the patient. Whilst some tra- ditional therapists might be described as
303/362
? ? ? adopting the stance 'I know; I'll tell you', the stance I advocate is one of 'You know, you tell me'. Thus the patient is encouraged to believe that, with support and occasional guidance, he can discover for himself the true nature of the models that underlie his thoughts, feelings, and actions and that, by examining the nature of his earlier experiences with his parents, or parent substitutes, he will understand what has led him to build the models now active within him and thus be free to restructure them. Fortunately the human psyche, like human bones, is strongly in- clined towards self-healing. The psychotherap- ist's job, like that of the orthopaedic surgeon's, is to provide the conditions in which self-healing can best take place.
Amongst those who have recently given de- tailed accounts of the special value of adopting a modest and tentative approach are Peterfreund (1983) and Casement (1985).
In the foregoing description the therapist's role has been likened to that of a mother who provides her child with a secure base from which to explore. This means, first and foremost, that he accepts and respects his patient, warts and all, as a fellow human being in trouble and that his
304/362
? ? ? over-riding concern is to promote his patient's welfare by all means at his disposal. To this end the therapist strives to be reliable, attentive, em- pathic, and sympathetically responsive, and also to encourage his patient to explore the world of his thoughts, feelings, and actions not only in the present but also in the past. Whilst always en- couraging his patient to take the initiative, the therapist is in no sense passive. On the one hand he tries to be attentive and sensitively responsive. On the other, he recognizes that there are times when he himself should take the initiative. For example, when a patient wastes time talking about everything and anything except his thoughts and feelings about people, it will be ne- cessary to draw his attention to his avoidance of the area, and perhaps also to his deep distrust of the therapist's efforts to be helpful or of his capa- city to keep confidences. With another patient, who perhaps is very willing to explore memories of childhood, there will be many occasions when a therapist can usefully ask for more detail or raise questions about situations of childhood that the patient has so far not referred to directly, but which seem plausible possibilities in the light of what he has been describing, and in the light also
305/362
? ? ? of the particular problems from which the patient is suffering. In doing so, moreover, the therapist must never forget that his patient may still be strongly influenced by his parents' injunctions not to know about events he is not supposed to know about and not to experience feelings he is not supposed to experience.
An interruption to therapy probably always generates some reaction in a patient; sometimes it is conscious, at others unconscious, but non- etheless evident. When conscious, it may take the form of overt complaint or angry protest; when unconscious it may manifest itself by the patient disparaging therapy or missing a session or two before the break. How a therapist evaluates these reactions and responds to them will reflect his theoretical position. Someone who adopts attach- ment theory will respect his patient's distress or anger about the separation and will regard them as the natural responses of someone who has be- come attached to another--a respect that will be implicit in anything he says or does. At the same time he will give attention to the form his pa- tient's reaction takes. If openly expressed, he will be sympathetic and may be able to ease the pa- tient's distress by giving him information about
306/362
? ? ? how he could communicate during the break. In addition, the therapist will consider how the pa- tient is construing the interruption and, should there be evidence of misconstruction, will at- tempt to discover how it may have originated. If, for example, the patient is apprehensive that the therapist will not return, the possibility of the pa- tient having been exposed to threats by a parent to abandon him might be explored. In cases where the interruption is due to the therapist be- ing unwell, he will be alert to the possibility that the patient may be apprehensive lest something he (the patient) has done or said is responsible. Were that to be so, the therapist would explore whether one of the patient's parents had sought to control him by claiming that the way he be- haved was making mother or father ill.
Similarly, should a patient react to an interrup- tion by disparaging therapy or missing a session, a therapist who adopts attachment theory would ask himself why his patient is afraid to express his feelings openly and what his childhood exper- iences may have been to account for his distrust.
It is not unlikely that the description just given of a therapist's mode of responding to his pa- tient's reactions to an interruption will contrast
307/362
? ? ? with that of a therapist who adopts and applies one or another of the traditional psychoanalytic theories. For example, one such therapist might regard his patient's reactions as being rather childish, even infantile, and as indicating that the patient was fixated in an oral or a symbiotic phase. What the therapist then might say, and es- pecially the way he might say it, could well be ex- perienced by the patient as lacking in respect for his (the patient's) current feelings of attachment, distress, or anger. Here again there would be danger that the therapist might appear to be re- sponding in a cold unsympathetic way and all too like one or other of the patient's parents. Were that so the exchange would be anti-therapeutic.
How far a therapist can wisely go in meeting a patient's desire to keep in communication during breaks, e. g. by telephone, and for comforting when distressed during a session, turns on many personal factors in their relationship. On the one hand, there is danger of the therapist's appearing to lack sympathy for the patient's distress or even to seem rejecting. On the other is the risk of his appearing to offer more than he is prepared to give. There are occasions when it would be inhu- man not to allow a distressed patient to make
308/362
? ? ? some form of physical contact: the roles are then explicitly comforter and comforted. Yet there is always danger that physical contact can elicit sexual feelings, especially when sexes are differ- ent. Depending on the situation each therapist must make his own decisions and draw his own lines. The more alive to such issues a therapist is the better will he be able to avoid the pitfalls.
EMOTIONAL COMMUNICATIONS AND THE RESTRUCTURING OF WORKING MODELS
When a therapist utilizes the kind of technique advocated here, it can sometimes happen that therapy gets into a rut in which the patient per- sists endlessly in describing what a terrible time he had as a child and how badly his parents treated him, without any progress being made. One cause of such perseveration, I suspect, is that the patient is convinced that his therapist does not accept the truth of what he is saying: hence his endless repetition of it. This may be due to the patient having always been scoffed at by those to whom he has told the story in the past or, and perhaps more commonly, to the therapist himself having indicated scepticism or disbelief. This can be done in a myriad of ways, by tone of voice, by
309/362
? ? ? querying the details, by failing to attach any par- ticular weight to what the patient describes.
Evidently when the problem lies in the therap- ist's incredulity the way out is for him to make it plain that he knows all too well that such things do happen to children and has no reason to doubt the patient's account. Even so the impasse may continue: the story is told and retold in a flat cyn- ical way with no show of feeling whatever.
This situation has been discussed by Selma Fraiberg who, with colleagues, set out to help vul- nerable mothers at risk of either neglecting or ab- using their infants (Fraiberg, Adelson, and Sha- piro, 1975). They describe making visits to the homes of two such mothers and listening to the distressing tales these women had to tell. Each told a story of gross cruelty during child- hood--being subjected to violent beatings, being locked out in the cold, often deserted by mother, being shunted from one place to another, and of having no one to go to for help or comfort. Neither gave a hint of how they might have felt nor what they may have felt like doing. One, a girl of 16 who avoided touching or holding her baby (who screamed hopelessly), insisted: 'But what's the use of talking? I always kept things to myself.
310/362
? ? ? I want to forget. I don't want to think. ' This was the point at which the therapist intervened--by herself expressing all the feelings that any and every child would be expected to have in the situ- ations described: how frightened, how angry, how hopeless one would feel, and also how one would long to go to someone who would understand and provide comfort and protection. In doing so the therapist not only showed an understanding of how the patient must have felt, but communic- ated in her manner that the expression of such feeling and desire would be met with a sympath- etic and comforting response. Only then was it possible for the young mother to express all the grief, the tears, 'and the unspeakable anguish for herself as a cast-off child' that she had always felt but had never dared express.
In this account of Fraiberg's methods of help- ing a patient express the emotions she dares not show I have deliberately emphasized the link between emotion and action. Failure to express emotion is due very largely to unconscious fear lest the action of which the emotion is a part will lead to a dreaded outcome. In many families an- ger with an adult leads to punishment which can sometimes be severe. Moreover a tearful appeal
311/362
? ? ? for comfort and help can lead to rejection and hu- miliation. It is perhaps too often forgotten by clinicians that many children, when they become distressed and weepy and are looking for com- fort, are shooed off as intolerable little cry-babies. Instead of the comforting provided by an under- standing and affectionate parent, these children meet with an unsympathetic and critical rebuff. No wonder therefore if, should this pattern pre- vail during childhood, the child learns never to show distress or seek comfort and, should he un- dertake therapy, assumes that his therapist will be as intolerant of anger and tears as his parents always were.
Every therapist who adopts a psychoanalytic per- spective has long recognized that, to be effective, therapy requires that a patient not only talks about his memories, his ideas and dreams, his hopes and desires, but also expresses his feelings. The discussion of Fraiberg's technique for help- ing a cynical and frozen young woman to discover the depth of her feelings and to express them freely to her therapist is therefore a fitting note on which to end. In writing this lecture I have throughout been aware that, by using terms such as 'information', 'communication', and 'working
312/362
? ? ? models', it would be easy for an unwary reader to suppose that these terms belong within a psycho- logy concerned only with cognition and one bereft of feeling and action. Although for many years it was all too common for cognitive psycho- logists to omit reference to emotion, it is now re- cognized that to do so is artificial and unfruitful (Hinde, Perret-Clermont, and Stevenson-Hinde, 1985). There are, in fact, no more important com- munications between one human being and an- other than those expressed emotionally, and no information more vital for constructing and re- constructing working models of self and other than information about how each feels towards the other. During the earliest years of our lives, indeed, emotional expression and its reception are the only means of communication we have, so that the foundations of our working models of self and attachment figure are perforce laid using information from that source alone. Small won- der therefore, if, in reviewing his attachment re- lationships during the course of psychotherapy and restructuring his working models, it is the emotional communications between a patient and his therapist that play the crucial part.
313/362
? ? ? 1 Within traditional theory this shift of role by a pa- tient is likely to be termed a case of identification with the aggressor.
2 Since in previous publications I have given much at- tention to the ill effects on personality development of bereavements and prolonged separations, these themes are omitted from what follows.
3 For research purposes, however, criteria for accept- ing retrospective information as valid must be much stricter.
REFERENCES
Adams-Tucker, C. (1982) 'Proximate effects of sexual abuse in childhood: a report on 28 children', American Journal of Psychiatry, 139: 1252-6.
Ainsworth, M. D. (1962) 'The effects of maternal deprivation: a review of findings and controversy in the context of research strategy' in: Deprivation of maternal care: a reassessment of its effects, Public Health Papers no. 14, Geneva: World Health Organisation.
Ainsworth, M. D. (1963) 'The development of infant- mother interaction among the Ganda' in B. M. Foss (ed. ) Determinants of infant behaviour, vol. 2, London: Methuen; New York: Wiley.
Ainsworth, M. D. S. (1967) Infancy in Uganda: infant care and the growth of attachment, Baltimore: Johns Hopkins University Press.
Ainsworth, M. D. S.
273/362
? ? ? secure attachment to her than a woman whose childhood was a happy one. This is a finding to give great encouragement to the many therapists who for long have sought to help mothers in just this kind of way. Further reference to techniques for helping disturbed mothers is made at the end of Lecture 8.
PATHWAYS TO PERSONALITY DEVELOPMENT
There is one further way in which attachment theory differs from traditional types of psycho- analytic theory, namely its rejection of the model of development in which an individual is held to pass through a series of stages in any one of which he may become fixated or to which he may regress, and its replacement by a model in which an individual is seen as progressing along one or another of an array of potential developmental pathways. Some of these pathways are compat- ible with healthy development; others deviate in one or another direction in ways incompatible with health.
All variants of the traditional model invoking phases of development are based on the assump- tion that, at some phase of normal development, a child shows psychological features that, in an
274/362
? ? ? older individual, would be regarded as signs of pathology. Thus a chronically anxious and cling- ing adult might be regarded as being fixated in or having regressed to a postulated phase of orality or of symbiosis; whilst a deeply withdrawn indi- vidual might be regarded as having regressed to a postulated phase of autism or of narcissism. Sys- tematic and sensitive studies of human infants, such as those reported by Stern (1985), have now rendered this model untenable. Observations show that infants are socially responsive from birth onwards. Healthily developing toddlers do not show anxious clinging except when they are frightened or distressed; at other times they ex- plore with confidence.
The model of developmental pathways regards an infant at birth as having an array of pathways potentially open to him, the one along which he will in fact proceed being determined at every moment by the interaction of the individual as he now is with the environment in which he happens then to be. Each infant is held to have his own in- dividual array of potential pathways for personal- ity development which, except for infants born with certain types of neurological damage, in- clude many that are compatible with mental
275/362
? ? ? health and also many that are incompatible. Which particular pathway he proceeds along is determined by the environment he meets with, especially the way his parents (or parent substi- tutes) treat him, and how he responds to them. Children who have parents who are sensitive and responsive are enabled to develop along a healthy pathway. Those who have insensitive, unrespons- ive, neglectful, or rejecting parents are likely to develop along a deviant pathway which is in some degree incompatible with mental health and which renders them vulnerable to breakdown, should they meet with seriously adverse events. Even so, since the course of subsequent develop- ment is not fixed, changes in the way a child is treated can shift his pathway in either a more fa- vourable direction or a less favourable one. Al- though the capacity for developmental change di- minishes with age, change continues throughout the life cycle so that changes for better or for worse are always possible. It is this continuing potential for change that means that at no time of life is a person invulnerable to every possible ad- versity and also that at no time of life is a person impermeable to favourable influence. It is this
276/362
? ? ? persisting potential for change that gives oppor- tunity for effective therapy.
1 In earlier publications I have sometimes used the term 'representational model' as a synonym for 'work- ing model' because representation has been the more familiar concept in clinical literature. In a dynamic psychology, however, working model is the more ap- propriate term, and it is also the term that is now com- ing into use among cognitive psychologists (e. g. Johnson-Laird, 1983). Within the attachment frame- work the concept of working model of an attachment figure is in many respects equivalent to, and replaces, the traditional psychoanalytic concept of internal object.
2 In further examination of the data it has been found that all these correlations also hold true for fathers (Main, personal communication).
8
ATTACHMENT, COMMUNICATION, AND THE THERAPEUTIC PROCESS
In the second part of my 1976 Maudsley Lecture, 'The making and breaking of affectional bonds' (1977), I described some of my ideas on the therapeutic implications of attachment theory. Much that has been learned since then has strengthened my confidence in the approach. The present account therefore should be re- garded as an amplification of the earlier one. In it I give more detailed attention to the ways a patient's earlier experiences affect the transfer- ence relationship and discuss further the therap- ist's aim as being that of enabling his patient to reconstruct his working models of himself and his attachment figure(s) so that he becomes less under the spell of forgotten miseries and better able to recognize companions in the present for what they are.
? 278/362
? ? ? a thing which has not been understood inevitably reappears; like an unlaid ghost, it cannot rest until the mystery has been resolved and the spell broken.
Sigmund Freud 1909
Those who cannot remember the past are condemned to repeat it. George Santayana 1905
FIVE THERAPEUTIC TASKS
The theory of personality development and psy- chopathology outlined above can be used as a framework to guide each one of the three princip- al forms of analytic psychotherapy in use today--individual therapy, family therapy, and group therapy. Here I deal only with the first.
A therapist applying attachment theory sees his role as being one of providing the conditions in which his patient can explore his representational models of himself and his attachment figures with a view to reappraising and restructuring them in the light of the new understanding he ac- quires and the new experiences he has in the therapeutic relationship. In helping his patient towards this end the therapist's role can be de- scribed under five main heads.
The first is to provide the patient with a secure base from which he can explore the various un- happy and painful aspects of his life, past and
279/362
? ? ? present, many of which he finds it difficult or per- haps impossible to think about and reconsider without a trusted companion to provide support, encouragement, sympathy, and, on occasion, guidance.
A second is to assist the patient in his explora- tions by encouraging him to consider the ways in which he engages in relationships with significant figures in his current life, what his expectations are for his own feelings and behaviour and for those of other people, what unconscious biases he may be bringing when he selects a person with whom he hopes to make an intimate relationship and when he creates situations that go badly for him.
A particular relationship that the therapist en- courages the patient to examine, and that consti- tutes the third task, is the relationship between the two of them. Into this the patient will import all those perceptions, constructions, and expecta- tions of how an attachment figure is likely to feel and behave towards him that his working models of parents and self dictate.
A fourth task is to encourage the patient to consider how his current perceptions and expect- ations and the feelings and actions to which they
280/362
? ? ? give rise may be the product either of the events and situations he encountered during his child- hood and adolescence, especially those with his parents, or else as the products of what he may repeatedly have been told by them. This is often a painful and difficult process and not infrequently requires that the therapist sanction his patient to consider as possibilities ideas and feelings about his parents that he has hitherto regarded as un- imaginable and unthinkable. In doing so a pa- tient may find himself moved by strong emotions and urges to action, some directed towards his parents and some towards the therapist, and many of which he finds frightening and/or alien and unacceptable.
The therapist's fifth task is to enable his patient to recognize that his images (models) of himself and of others, derived either from past painful experiences or from misleading messages eman- ating from a parent, but all to often in the literat- ure mislabelled as 'fantasies', may or may not be appropriate to his present and future; or, indeed, may never have been justified. Once he has grasped the nature of his governing images (mod- els) and has traced their origins, he may begin to understand what has led him to see the world and
281/362
? ? ? himself as he does and so to feel, to think, and to act in the ways he does. He is then in a position to reflect on the accuracy and adequacy of those im- ages (models), and on the ideas and actions to which they lead, in the light of his current experi- ences of emotionally significant people, including the therapist as well as his parents, and of himself in relationship to each. Once the process has star- ted he begins to see the old images (models) for what they are, the not unreasonable products of his past experiences or of what he has repeatedly been told, and thus to feel free to imagine altern- atives better fitted to his current life. By these means the therapist hopes to enable his patient to cease being a slave to old and unconscious stereo- types and to feel, to think, and to act in new ways.
Readers will be aware that the principles set out have a great deal in common with the prin- ciples described by other analytically trained psy- chotherapists who regard conflicts arising within interpersonal relationships as the key to an un- derstanding of their patient's problems, who fo- cus on the transference and who also give some weight, albeit of varying degree, to a patient's earlier experience with his parents. Among the many well-known names that could be
282/362
? ? ? mentioned in this context are those of Fairbairn, Winnicott, and Guntrip in Britain, and Sullivan, Fromm-Reichmann, Gill, and Kohut in the Un- ited States. Among recently published works that contain many of the ideas prescribed here are those by Peterfreund (1983), Casement (1985), Pine (1985), and Strupp and Binder (1984), and also those of Malan (1973) and Horowitz et al. (1984) in the field of brief psychotherapy. In par- ticular, I wish to draw attention to the ideas of Horowitz and his colleagues who, in their de- scription of the treatment of patients suffering from an acute stress syndrome, employ a concep- tual framework closely similar to that presented here. Although their technique is aimed to help patients recover from the effects of a recent severely stressful event, I believe the principles informing their work are equally applicable to helping patients recover from the effects of a chronic disturbance resulting from stressful events of many years ago, including those that oc- curred during their earliest years.
Although in this exposition it is convenient to list the therapist's five tasks in a logical way, so inter- related are they that in practice a productive ses- sion is likely to involve first one task, then
283/362
? ? ? another. Nevertheless, unless a therapist can en- able his patient to feel some measure of security, therapy cannot even begin. Thus we start with the role of the therapist in providing his patient with a secure base. This is a role very similar to that described by Winnicott as 'holding' and by Bion as 'containing'.
In providing his patient with a secure base from which to explore and express his thoughts and feelings the therapist's role is analogous to that of a mother who provides her child with a se- cure base from which to explore the world. The therapist strives to be reliable, attentive, and sympathetically responsive to his patient's ex- plorations and, so far as he can, to see and feel the world through his patient's eyes, namely to be empathic. At the same time he is aware that, be- cause of his patient's adverse experiences in the past, the patient may not believe that the therap- ist is to be trusted to behave kindly or to under- stand his predicament. Alternatively the unex- pectedly attentive and sympathetic responses the patient receives may lead him to suppose that the therapist will provide him with all the care and affection which he has always yearned for but never had. In the one case therefore the therapist
284/362
? ? ? is seen in an unduly critical and hostile light, in the other as ready to provide more than is at all realistic. Since, it is held, both types of misunder- standing and misconstruction, and the emotions and behaviour to which they give rise, are central features of the patient's troubles, a therapist needs to have the widest possible knowledge of the many forms these misconstructions can take and also of the many types of earlier experience from which they are likely to have sprung. Without such knowledge a therapist is poorly placed to see and feel the world as his patient is doing.
Even so, a patient's way of construing his rela- tionship with his therapist is not determined solely by the patient's history: it is determined no less by the way the therapist treats him. Thus the therapist must strive always to be aware of the nature of his own contribution to the relationship which, amongst other influences, is likely to re- flect in one way or another what he experienced himself during his own childhood. This aspect of therapy, the counter-transference, is a big issue of its own and the subject of a large literature. Since it is not possible to deal with it further here, I want to emphasize not only the importance of
285/362
? ? ? the counter-transference but also that the focus of therapy must always be on the interactions of patient and therapist in the here and now, and that the only reason for encouraging the patient at times to explore his past is for the light it throws on his current ways of feeling and dealing with life.
With that proviso firmly in mind, let us con- sider some of the commoner forms that a pa- tient's misconstructions can take and how they are likely to have originated. This is the aspect of therapy in which the work of a therapist who ad- opts attachment theory is likely to differ most from one who adopts certain of the traditional theories of personality development and psycho- pathology. Thus, for example, a therapist who views his patient's misperceptions and misunder- standings as the not unreasonable products of what the patient has actually experienced in the past, or has repeatedly been told, differs sharply from one who sees these same misperceptions and misunderstandings as the irrational offspring of autonomous and unconscious fantasy.
In what follows I am drawing on several dis- tinct sources of information: studies by epidemi- ologists; the studies by developmental
286/362
? ? ? psychologists already referred to; observations made during the course of family therapy; and not least what I have learned from patients whom I have treated myself and from those whose ther- apy I have supervised.
INFLUENCE OF EARLIER EXPERIENCES ON THE TRANSFERENCE RELATIONSHIP
It not infrequently happens that a patient is acutely apprehensive lest his therapist reject, cri- ticize, or humiliate him. Since we know that all too many children are treated in this way by one or other, or both, of their parents, we can be reas- onably confident that that has been our patient's experience. Should it seem likely that the patient is aware of how he is feeling and how he expects the therapist to treat him, the therapist will indic- ate that he also is aware of the problem. How soon the therapist can link these expectations to the patient's experiences of his parents, in the present perhaps as well as the past, turns on how willing the patient is to consider that possibility, or whether, by contrast, he insists that his par- ents' treatment of him is above criticism. Where the latter situation obtains, there is the prior problem of trying to understand why the patient
287/362
? ? ? should insist on retaining this favourable picture when such evidence as is available points to its being mistaken.
It happens in some families that one or other parent insists that he or she is an admirable par- ent who has always done everything possible for the child and that, in so far as friction is present, the fault lies exclusively with the child. This atti- tude of the parent all too often cloaks behaviour that, by ordinary standards, has been far from perfect. Yet, since the parent insists that he or she has given the child constant affection and that the child must have been born bad and ungrateful, the child has little option but to accept the pic- ture, despite being aware somewhere in his mind that the picture is hardly fair.
An added complication arises when a patient has, as a child, been subjected to the strongest of instructions from a parent on no account to tell anyone of certain happenings within the family. These are usually quarrels in which the parent is aware that his or her behaviour is open to criti- cism; for example, quarrels between the parents, or between a parent and a child, during which dreadful things have been said or done. The more insistent a therapist is that his patient tell
288/362
? ? ? everything, the more distressing the dilemma is for his patient. Injunctions to silence are not un- common in families and have been much neg- lected as sources of what has traditionally been called resistance. It is often useful for a therapist to enquire of a patient whether he may have been subjected to such pressures and, if so, to help him resolve the dilemma.
So far we have been considering cases in which a patient is in some degree aware of his expecta- tions of being rejected, criticized, or humiliated. Not infrequently, however, a patient seems wholly unaware of any such feelings despite his attitude to the therapist exuding distrust and eva- sion. Evidence shows that these states of mind occur especially in those who, having developed an anxiously avoidant pattern of attachment dur- ing early years, have striven ever since to be emotionally self-contained and insulated against intimate contacts with other people. These pa- tients, who are often described as being narciss- istic or as having a false self, avoid therapy as long as they can and, should they undertake it, keep the therapist at arm's length. If allowed to, some will talk incessantly about anything and everything except emotionally charged
289/362
? ? ? relationships, past or present. Others will explain that they have nothing to talk about. One young woman, whose every move indicated deep dis- trust of me, spent the time boasting of her delin- quent exploits, many of them fictitious I suspec- ted, and pouring contempt on what she insisted was my dull and narrow life. To treat such deeply distrustful people was compared many years ago by Adrian Stephen (1934) with trying to make friends with a shy or frightened pony: both situ- ations require a prolonged, quiet, and friendly patience. Only when the therapist is aware of the constant rebuffs the patient is likely to have been subjected to as a child whenever he sought com- fort or help, and of his terror of being subjected to something similar from the therapist, can the latter see the situation between them as his pa- tient is seeing it.
Another and quite different cause of wariness of any close contact with a therapist for the pa- tient is dread lest the therapist trap him into a re- lationship aimed to serve the therapist's interests rather than his own. A common origin of such fear is a childhood in which a parent, almost al- ways mother, has sought to make the child her own attachment figure and caregiver, that is, has
290/362
? ? ? inverted the relationship. Very often this is done unconsciously and using techniques that, to an uninformed eye, may appear to be overindul- gence but that are really bribes to retain the child in a caregiving role.
Not infrequently a patient shifts during the therapy from treating his therapist as though he was one or other of his parents to behaving to- wards him in the way one of his parents had treated him. For example, a patient who has been subjected to hostile threats as a child may use hostile threats to his therapist. Experiences of scornful contempt from a parent may be re-en- acted as scornful contempt of the therapist. Sexu- al advances from a parent may reappear as sexual advances to the therapist. Such behaviour may be understood in the following way. During his childhood a person learns two principal forms of behaviour and builds in his mind two principal types of model. One form of behaviour is, of course, that of a child, namely himself, interact- ing with a parent, his mother or his father. The corresponding working models he builds are those of himself as a child in interaction with each parent. The other form of behaviour is that of a parent, namely his mother or his father,
291/362
? ? ? interacting with a child, himself. The correspond- ing models he builds are those of each parent in interaction with himself. Therefore, whenever a therapist is puzzled by, or resentful of, the way he is being treated by a patient, he is always wise to enquire when and from whom the patient may have learned that way of treating other people. More often than not it is from one of his parents. 1
With some patients the therapeutic relation- ship is one in which anxiety, distrust, and criti- cism, and sometimes also anger and contempt, are overt and predominate, and the therapist seen in dark colours. Such sentiments as gratit- ude for the therapist's efforts or respect for his competence are conspicuous by their absence. The task then is to help the patient grasp that much of his present resentment stems from past mistreatment at the hands of others and that, however understandable his anger may be as a result, to continue fighting old battles is unpro- ductive. To accept that an unhappy past cannot be changed is usually a bitter pill.
With other patients the situation is reversed: the transference relationship becomes one in which overt gratitude, admiration, and affection are readily expressed, and the therapist seen in a
292/362
? ? ? glow of rosy perfection. Dissatisfaction and criti- cism are notably absent, and anger at the therap- ist's shortcomings, especially absences, unima- ginable. Such idealization of the therapist springs, I believe, partly from unrealistic hopes and expectations of what the therapist is able and willing to provide, and partly from a childhood in which criticism of a parent is forbidden and com- pliance enforced, either by some guilt-inducing technique or else by sanctions such as threats not to love, or even to abandon, the child. With this type of childhood experience, the patient's un- conscious assumption is that the therapist will expect the same degree of obedience as his par- ents had expected and will enforce it by tech- niques or by threats similar to the ones they used.
Unfortunately there has been a tendency in some quarters to confuse the theory advanced here, which regards the way certain parents treat their children as being a major cause of mental ill- health, with an attitude of mind that simply blames parents. No one who works in the fields of child psychiatry and family therapy is likely to make this mistake. On the contrary, as already remarked, it has long been recognized that the misguided behaviour of parents is more often
293/362
? ? ? than not the product of their own difficult and unhappy childhood. As a result much skilled time has always been devoted to helping parents es- cape from the adverse influences of their own childhoods.
Furthermore, during the treatment of an indi- vidual (of any age) who has suffered at the hands of his parents, the therapist, whilst accepting the patient's account, avoids moral judgement. On the contrary, whenever opportunity offers, he will encourage his patient to consider how and why the parent under discussion may have behaved as he or she has done. In raising these questions, it is always useful for the therapist to enquire of the patient what he knows of the childhood experi- ences that the parent in question may have had. Not infrequently this leads the patient to gain some understanding of how things had developed and, from understanding, often to move on to a measure of forgiveness and reconciliation. In family sessions it can be especially valuable if a parent can be encouraged to give an account of his or her childhood. This enables all those present--the parent him- or herself, the spouse, the children, and the therapist--to gain some in- sight into how and why family life has developed
294/362
? ? ? as it has and how each can best help improve it. As already mentioned (p. 150), this strong tend- ency for attachment problems to be transmitted across generations, through the influence on par- enting behaviour of relationship problems stem- ming from the parent's own childhood, is at last receiving the research attention it deserves.
SOME PATHOGENIC SITUATIONS AND EVENTS OF CHILDHOOD
A therapist, I believe, cannot be too well in- formed about the disguised and distorted rela- tionships that can occur in some families, and the terrible things that can happen in others, for it is only if he is so informed that he can have a reas- onably clear idea of what probably lies behind his patient's defences, or of the origins of his anxiety, anger, and guilt. Once he is adequately informed, he is well placed not only to appreciate the truth of what his patient may be describing as having happened to him but also to broach, more or less tentatively, some of the kinds of situation to which the patient may well have been exposed but to which he may be either unable or unwilling to refer. In listing the following situations I am doing no more than indicating some that are both
295/362
? ? ?
common and have, until recently, been neglected in the psychotherapeutic literature. 2
Threats not to love a child used as a means of control
It is easy for a mother to say to a child that she will not love him if he behaves in such and such a way. What this means is that the mother is threatening not to provide affection or comfort at times when her child is upset, frightened, or dis- tressed, and not to provide help or encourage- ment at other times. If such threats are used sys- tematically by either parent, or both, the child in- evitably grows up intensely anxious to please and guilt-prone.
Threats to abandon a child
Threats to abandon are a degree more frightening to a child than threats no longer to love him. This is especially so if the parent enacts a threat, per- haps by disappearing herself for a few hours or by packing her child's suitcase and walking him up the street, allegedly to the home for bad boys. Since threats to abandon often take a highly idio- syncratic form, a patient may deny that he was ever subjected to them. In such cases the truth
296/362
? ? ? may emerge with its accompanying emotion only when the special way in which the threat was phrased is recollected. One example is a mother who had concocted the story that a yellow van would draw up and take her son away. Another is a father whose story was that his daughter would be sent to a school on a remote rock surrounded by sharks (Marrone, 1984). Thus, in the first case, all the mother had to say was 'Well the yellow van will come', and, in the second, for the father to say 'Then it will be the rock school', for the child instantly to desist from whatever he or she was doing. In a third case, the code word was 'mar- garine', the mother having coupled her threat to send her son away to a children's home with an insistence that he would have to eat margarine there. For these patients a general phrase like 'threat to abandon' had failed to ring a bell. Only when the code word was unearthed was the ori- ginal terror experienced afresh and the source of the separation anxiety located.
Threats to commit suicide
Sometimes a distraught parent threatens to com- mit suicide if some distressing situation contin- ues. This may occur during quarrels between
297/362
? ? ? parents, which the child overhears, or may be dir- ected at the child himself. In either case such threats strike terror. One lesson to be learned from these cases is that, whenever a patient refers to his parents as having quarrelled, the therapist should always enquire 'What did they say to each other? ' Not infrequently a patient blocks at this. In a fit of temper quarrelling parents may say ap- palling things to each other. This is bad enough. What makes it far worse is when, after cooling down, they disclaim having said any such thing.
Disclaimers and disconfirmations
Examples of disclaimers by a parent of what he or she has said or done, and persistent efforts to dis- confirm what a child has seen or heard, are given in some detail in Lecture 6 and the adverse ef- fects on personality development of such pres- sures emphasized. During therapy these effects emerge as great uncertainty in a patient as to whether some family episode did or did not occur and guilt about adumbrating it. Here, as so often, a key role for the therapist is to sanction the pa- tient's exploration of all the various possibilities, both those favourable to his parents and those unfavourable to them, and to encourage him to
298/362
? ? ? weigh the evidence available, whilst he (the ther- apist) remains resolutely open-minded as to where the truth may lie.
Thus far in this exposition I have not considered the critical issue of how far we can and should rely on the validity of our patients' reports. Memories are certainly fallible, and there are various occasions when an experienced therapist will rightly question the truth of what his patient is saying. What then are the criteria by which we should judge?
First, broad generalizations about the kind of parent the mother or father was and about the kind of parenting received are never to be given credence unless and until they are supported by detailed examples of how each treated the patient as a child in particular situations. For example, the glowing account of a wonderful mother may well go unsupported when detail becomes avail- able. Valid accounts of affectionate parenting not only give plenty of favourable details but are likely also to be interlaced with occasional criti- cism, so that the parent can be seen in the round. Similarly, disparaging accounts of parents in uni- formly adverse terms need close examination. In- valid accounts of either sort tend to be sweeping
299/362
? ? ? and extreme, to be either white or black. Detail is either lacking or, should it be given, is at variance with the portrait presented. By contrast, whenev- er plenty of consistent detail is given and the pic- ture that emerges conforms to what we know, from other sources, does happen in other famil- ies, and also to the known antecedents of the types of problem besetting the patient, it is ab- surd to doubt its overall validity, even if some points remain in question.
The origin of these extremes is not infrequently external pressure. For example, one parent has insisted that the child take sides with him or her against the other parent who is represented as being all bad. Or else a parent who has many shortcomings insists that he or she is above criticism.
Another occasion when a therapist is right to doubt the patient's story is when there is reason to suspect that the patient is a pathological liar. Such cases are comparatively rare and, if only for that reason, may go undetected for a time. Sooner or later, however, mounting inconsistencies and improbabilities, as well as the way the patient tells his story, engender first doubt and, later, certainty that the patient is not to be believed.
300/362
? ? ? Apart from these exceptions I believe patients' accounts are sufficiently trustworthy that a ther- apist should accept them as being reasonable ap- proximations to the truth;3 and furthermore that it is anti-therapeutic not to do so. Constantly to query the validity of the patient's story, even if only by implication, and insistence on the distort- ing role of imagination or fantasy, is the reverse of empathic. It conveys to the patient that the therapist
does not understand him and may indeed con- vince him that the therapist is behaving exactly as his parent had predicted. Thus some parents, having insisted that their child not tell of something the parent is ashamed of, may then add that, in any case, were he to do so no one would believe him.
Among the large range of adverse events and situations not so far mentioned in this lecture that a therapist should have in his mind as likely to have occurred in the life of one patient or an- other are the following:
a child may never have been wanted by one or both parents;
a child may be of the wrong sex in a family in which parents had set their hearts on a boy or a girl;
301/362
? ? ? a child may have been made the family scapegoat, sometimes as a result of a family tragedy that, with greater or less plausibility, has always been at- tributed to him;
a parent may have used guilt-inducing techniques to control a child, for example, frequent claims that the child's behaviour makes mother ill;
a parent may have sought to make one of her children her attachment fig- ure by discouraging him from exploring the world away from her and from believing that he will ever be able to make his way on his own;
a child's unusual role in a family may be the result of his mother having had an extra-marital affair during her marriage so that the child's putative father is not his real father;
another cause of a child's unusual role is when one or other parent identi- fies one child with a relative, often one of the child's grandparents, with whom he or she has had a difficult relationship, and who then re-enacts that relationship with the child;
a child may have been the target of more or less serious physical abuse from a parent or step-parent;
a child may have been involved in sexual abuse from a parent, step-par- ent, or older sibling for short or long periods of time.
For those unaware of the commoner effects on personality development of exposure to situ- ations of these kinds, a number of references are given in an Appendix to references (pp. 194-5).
Inevitably the influential events of an individu- al's first two or three years will either never have been registered in his memory or else cannot now be recalled. Here, of course, the best a therapist can do is to infer, on the basis of the transference situation and of such information as the patient has gleaned about his early years, combined with
302/362
? ? ? such wider knowledge of personality develop- ment as the therapist has acquired, what the nature of those events may have been. In other words, he resorts to reconstruction; but in doing so he can in future draw on a much wider and more reliable knowledge of family influences on personality development than has traditionally been available to analytically trained psychotherapists.
THE THERAPIST'S STANCE
In this account of therapeutic principles, therap- ists will recognize much that has long been famil- iar, though often under a different name. The therapeutic alliance appears as a secure base, an internal object as a working, or representational, model of an attachment figure, reconstruction as exploring memories of the past, resistance (sometimes) as deep reluctance to disobey the past orders of parents not to tell or not to remem- ber. Among points of difference is the emphasis placed on the therapist's role as a companion for his patient in the latter's exploration of himself and his experiences, and less on the therapist in- terpreting things to the patient. Whilst some tra- ditional therapists might be described as
303/362
? ? ? adopting the stance 'I know; I'll tell you', the stance I advocate is one of 'You know, you tell me'. Thus the patient is encouraged to believe that, with support and occasional guidance, he can discover for himself the true nature of the models that underlie his thoughts, feelings, and actions and that, by examining the nature of his earlier experiences with his parents, or parent substitutes, he will understand what has led him to build the models now active within him and thus be free to restructure them. Fortunately the human psyche, like human bones, is strongly in- clined towards self-healing. The psychotherap- ist's job, like that of the orthopaedic surgeon's, is to provide the conditions in which self-healing can best take place.
Amongst those who have recently given de- tailed accounts of the special value of adopting a modest and tentative approach are Peterfreund (1983) and Casement (1985).
In the foregoing description the therapist's role has been likened to that of a mother who provides her child with a secure base from which to explore. This means, first and foremost, that he accepts and respects his patient, warts and all, as a fellow human being in trouble and that his
304/362
? ? ? over-riding concern is to promote his patient's welfare by all means at his disposal. To this end the therapist strives to be reliable, attentive, em- pathic, and sympathetically responsive, and also to encourage his patient to explore the world of his thoughts, feelings, and actions not only in the present but also in the past. Whilst always en- couraging his patient to take the initiative, the therapist is in no sense passive. On the one hand he tries to be attentive and sensitively responsive. On the other, he recognizes that there are times when he himself should take the initiative. For example, when a patient wastes time talking about everything and anything except his thoughts and feelings about people, it will be ne- cessary to draw his attention to his avoidance of the area, and perhaps also to his deep distrust of the therapist's efforts to be helpful or of his capa- city to keep confidences. With another patient, who perhaps is very willing to explore memories of childhood, there will be many occasions when a therapist can usefully ask for more detail or raise questions about situations of childhood that the patient has so far not referred to directly, but which seem plausible possibilities in the light of what he has been describing, and in the light also
305/362
? ? ? of the particular problems from which the patient is suffering. In doing so, moreover, the therapist must never forget that his patient may still be strongly influenced by his parents' injunctions not to know about events he is not supposed to know about and not to experience feelings he is not supposed to experience.
An interruption to therapy probably always generates some reaction in a patient; sometimes it is conscious, at others unconscious, but non- etheless evident. When conscious, it may take the form of overt complaint or angry protest; when unconscious it may manifest itself by the patient disparaging therapy or missing a session or two before the break. How a therapist evaluates these reactions and responds to them will reflect his theoretical position. Someone who adopts attach- ment theory will respect his patient's distress or anger about the separation and will regard them as the natural responses of someone who has be- come attached to another--a respect that will be implicit in anything he says or does. At the same time he will give attention to the form his pa- tient's reaction takes. If openly expressed, he will be sympathetic and may be able to ease the pa- tient's distress by giving him information about
306/362
? ? ? how he could communicate during the break. In addition, the therapist will consider how the pa- tient is construing the interruption and, should there be evidence of misconstruction, will at- tempt to discover how it may have originated. If, for example, the patient is apprehensive that the therapist will not return, the possibility of the pa- tient having been exposed to threats by a parent to abandon him might be explored. In cases where the interruption is due to the therapist be- ing unwell, he will be alert to the possibility that the patient may be apprehensive lest something he (the patient) has done or said is responsible. Were that to be so, the therapist would explore whether one of the patient's parents had sought to control him by claiming that the way he be- haved was making mother or father ill.
Similarly, should a patient react to an interrup- tion by disparaging therapy or missing a session, a therapist who adopts attachment theory would ask himself why his patient is afraid to express his feelings openly and what his childhood exper- iences may have been to account for his distrust.
It is not unlikely that the description just given of a therapist's mode of responding to his pa- tient's reactions to an interruption will contrast
307/362
? ? ? with that of a therapist who adopts and applies one or another of the traditional psychoanalytic theories. For example, one such therapist might regard his patient's reactions as being rather childish, even infantile, and as indicating that the patient was fixated in an oral or a symbiotic phase. What the therapist then might say, and es- pecially the way he might say it, could well be ex- perienced by the patient as lacking in respect for his (the patient's) current feelings of attachment, distress, or anger. Here again there would be danger that the therapist might appear to be re- sponding in a cold unsympathetic way and all too like one or other of the patient's parents. Were that so the exchange would be anti-therapeutic.
How far a therapist can wisely go in meeting a patient's desire to keep in communication during breaks, e. g. by telephone, and for comforting when distressed during a session, turns on many personal factors in their relationship. On the one hand, there is danger of the therapist's appearing to lack sympathy for the patient's distress or even to seem rejecting. On the other is the risk of his appearing to offer more than he is prepared to give. There are occasions when it would be inhu- man not to allow a distressed patient to make
308/362
? ? ? some form of physical contact: the roles are then explicitly comforter and comforted. Yet there is always danger that physical contact can elicit sexual feelings, especially when sexes are differ- ent. Depending on the situation each therapist must make his own decisions and draw his own lines. The more alive to such issues a therapist is the better will he be able to avoid the pitfalls.
EMOTIONAL COMMUNICATIONS AND THE RESTRUCTURING OF WORKING MODELS
When a therapist utilizes the kind of technique advocated here, it can sometimes happen that therapy gets into a rut in which the patient per- sists endlessly in describing what a terrible time he had as a child and how badly his parents treated him, without any progress being made. One cause of such perseveration, I suspect, is that the patient is convinced that his therapist does not accept the truth of what he is saying: hence his endless repetition of it. This may be due to the patient having always been scoffed at by those to whom he has told the story in the past or, and perhaps more commonly, to the therapist himself having indicated scepticism or disbelief. This can be done in a myriad of ways, by tone of voice, by
309/362
? ? ? querying the details, by failing to attach any par- ticular weight to what the patient describes.
Evidently when the problem lies in the therap- ist's incredulity the way out is for him to make it plain that he knows all too well that such things do happen to children and has no reason to doubt the patient's account. Even so the impasse may continue: the story is told and retold in a flat cyn- ical way with no show of feeling whatever.
This situation has been discussed by Selma Fraiberg who, with colleagues, set out to help vul- nerable mothers at risk of either neglecting or ab- using their infants (Fraiberg, Adelson, and Sha- piro, 1975). They describe making visits to the homes of two such mothers and listening to the distressing tales these women had to tell. Each told a story of gross cruelty during child- hood--being subjected to violent beatings, being locked out in the cold, often deserted by mother, being shunted from one place to another, and of having no one to go to for help or comfort. Neither gave a hint of how they might have felt nor what they may have felt like doing. One, a girl of 16 who avoided touching or holding her baby (who screamed hopelessly), insisted: 'But what's the use of talking? I always kept things to myself.
310/362
? ? ? I want to forget. I don't want to think. ' This was the point at which the therapist intervened--by herself expressing all the feelings that any and every child would be expected to have in the situ- ations described: how frightened, how angry, how hopeless one would feel, and also how one would long to go to someone who would understand and provide comfort and protection. In doing so the therapist not only showed an understanding of how the patient must have felt, but communic- ated in her manner that the expression of such feeling and desire would be met with a sympath- etic and comforting response. Only then was it possible for the young mother to express all the grief, the tears, 'and the unspeakable anguish for herself as a cast-off child' that she had always felt but had never dared express.
In this account of Fraiberg's methods of help- ing a patient express the emotions she dares not show I have deliberately emphasized the link between emotion and action. Failure to express emotion is due very largely to unconscious fear lest the action of which the emotion is a part will lead to a dreaded outcome. In many families an- ger with an adult leads to punishment which can sometimes be severe. Moreover a tearful appeal
311/362
? ? ? for comfort and help can lead to rejection and hu- miliation. It is perhaps too often forgotten by clinicians that many children, when they become distressed and weepy and are looking for com- fort, are shooed off as intolerable little cry-babies. Instead of the comforting provided by an under- standing and affectionate parent, these children meet with an unsympathetic and critical rebuff. No wonder therefore if, should this pattern pre- vail during childhood, the child learns never to show distress or seek comfort and, should he un- dertake therapy, assumes that his therapist will be as intolerant of anger and tears as his parents always were.
Every therapist who adopts a psychoanalytic per- spective has long recognized that, to be effective, therapy requires that a patient not only talks about his memories, his ideas and dreams, his hopes and desires, but also expresses his feelings. The discussion of Fraiberg's technique for help- ing a cynical and frozen young woman to discover the depth of her feelings and to express them freely to her therapist is therefore a fitting note on which to end. In writing this lecture I have throughout been aware that, by using terms such as 'information', 'communication', and 'working
312/362
? ? ? models', it would be easy for an unwary reader to suppose that these terms belong within a psycho- logy concerned only with cognition and one bereft of feeling and action. Although for many years it was all too common for cognitive psycho- logists to omit reference to emotion, it is now re- cognized that to do so is artificial and unfruitful (Hinde, Perret-Clermont, and Stevenson-Hinde, 1985). There are, in fact, no more important com- munications between one human being and an- other than those expressed emotionally, and no information more vital for constructing and re- constructing working models of self and other than information about how each feels towards the other. During the earliest years of our lives, indeed, emotional expression and its reception are the only means of communication we have, so that the foundations of our working models of self and attachment figure are perforce laid using information from that source alone. Small won- der therefore, if, in reviewing his attachment re- lationships during the course of psychotherapy and restructuring his working models, it is the emotional communications between a patient and his therapist that play the crucial part.
313/362
? ? ? 1 Within traditional theory this shift of role by a pa- tient is likely to be termed a case of identification with the aggressor.
2 Since in previous publications I have given much at- tention to the ill effects on personality development of bereavements and prolonged separations, these themes are omitted from what follows.
3 For research purposes, however, criteria for accept- ing retrospective information as valid must be much stricter.
REFERENCES
Adams-Tucker, C. (1982) 'Proximate effects of sexual abuse in childhood: a report on 28 children', American Journal of Psychiatry, 139: 1252-6.
Ainsworth, M. D. (1962) 'The effects of maternal deprivation: a review of findings and controversy in the context of research strategy' in: Deprivation of maternal care: a reassessment of its effects, Public Health Papers no. 14, Geneva: World Health Organisation.
Ainsworth, M. D. (1963) 'The development of infant- mother interaction among the Ganda' in B. M. Foss (ed. ) Determinants of infant behaviour, vol. 2, London: Methuen; New York: Wiley.
Ainsworth, M. D. S. (1967) Infancy in Uganda: infant care and the growth of attachment, Baltimore: Johns Hopkins University Press.
Ainsworth, M. D. S.
