No More Learning

John Bowlby and Attachment
Theory
John Bowlby (1907-1990) has been described as 'one of the three or four most important psychiatrists of the twentieth century'. In this book Jeremy Holmes provides a focused and coherent account of Bowlby's life and work, based on his writings and those of the 'post-Bowlbians', as well as interviews with members of his family and with psychoanalysts who knew him.
Bowlby's Attachment Theory is one of the major theoretical developments in psychoanalysis this half-century. Combining the rigorous scientific empiricism of ethology with the subjective insights of psychoanalysis, it has had an enormous impact in the fields of child development, social work, psychology, psychotherapy and psychiatry. Jeremy Holmes examines the origins of Bowlby's ideas, and presents the main features of Attachment Theory and their relevance to contemporary psychoanalytic psychotherapy. He looks at the processes of attachment and loss, and reviews recent experimental evidence linking secure attachment in infancy with the development of 'autobiographical competence'. He also provides fascinating insights into the history of the psychoanalytic movement, and considers the ways in which Attachment Theory can help in understanding society and its problems.
John Bowlby and Attachment Theory will be essential reading for all students of psychotherapy, counselling, social work, psychology and psychiatry, and for professionals working in those fields.
Jeremy Holmes is Consultant Psychiatrist/Psychotherapist at the North Devon District Hospital.
The Makers of Modern Psychotherapy
Series editor: Laurence Spurling
This series of introductory, critical texts looks at the work and thought of key contributors to the development of psychodynamic psychotherapy. Each book shows how the theories examined affect clinical practice, and includes biographical material as well as a comprehensive bibliography of the contributor's work.
The field of psychodynamic psychotherapy is today more fertile but also more diverse than ever before. Competing schools have been set up, rival theories and clinical ideas circulate. These different and sometimes competing strains are held together by a canon of fundamental concepts, guiding assumptions and principles of practice.
This canon has a history, and the way we now understand and use the ideas that frame our thinking and practice is palpably marked by how they came down to us, by the temperament and experiences of their authors, the particular puzzles they wanted to solve and the contexts in which they worked. These are the makers of modern psychotherapy. Yet despite their influence, the work and life of some of these eminent figures is not well known. Others are more familiar, but their particular contribution is open to reassessment. In studying these figures and their work, this series will articulate those ideas and ways of thinking that practitioners and thinkers within the psychodynamic tradition continue to find persuasive.
Laurence Spurling
John Bowlby and
Attachment Theory
Jeremy Holmes
London and New York
First published 1993
by Routledge
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada by Routledge
29 West 35th Street, New York, NY 10001
This edition published in the Taylor & Francis e-Library, 2001.
"To purchase your own copy of this or any of Taylor & Francis or Routledge's
collection of thousands of eBooks please go to www. eBookstore. tandf. co. uk. "
(C) 1993 Jeremy Holmes
All rights reserved. No part of this book may be reprinted Or reproduced or utilised in any form or by any electronic, Mechanical, or other means, now known or hereafter Invented, including photocopying and recording, or in any Information storage or retrieval system, without permission In writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
A catalog record for this book is available from the Library of Congress
ISBN 0-203-13680-2 Master e-book ISBN
ISBN 0-203-17585-9 (Adobe eReader Format) ISBN 0-415-07729-X (hbk)
ISBN 0-415-07730-3 (pbk)
To Jacob, Matthew, Lydia and Joshua; also to Ben, Polly, Matilda and Flora; and in memory of Tabitha.
By the same author
The Values of Psychotherapy (with R. Lindley), (1991 [1989]),Oxford University Press.
A Textbook of Psychotherapy in Psychiatric Practice (Editor),(1991), Churchill Livingstone.
Between Art and Science: Essays in Psychotherapy and Psychiatry
(1992), Routledge.
The Good Mood Guide (with R. Holmes), (1993), Orion.
Contents
List of illustrations ix Acknowledgements x 1 Introduction 1 Part I Origins
2 Biographical 13 3 Maternal deprivation 37 Part II Attachment Theory
4 Attachment, anxiety, internal working models 61 5 Loss, anger and grief 86
6 Attachment Theory and personality development: the research evidence 103
Part III Implications
7 Bowlby and the inner world: Attachment Theory and psychoanalysis 127
8 Attachment Theory and the practice of psychotherapy 149 9 Attachment Theory and psychiatric disorder 177 10 Attachment Theory and society 200 11 Epilogue 210
Glossary of terms relevant to Attachment Theory 217
viii John Bowlby and Attachment Theory
Chronology of John Bowlby 225 Bibliography 228 Index 244
Illustrations
FIGURES
3. 1 Developmental pathways from maternal deprivation 52 4. 1 The attachment behavioural system 76 4. 2 Patterns of insecure attachment 80 6. 1 The evolution of attachment patterns 118 6. 2 Anna and the holding environment 121 7. 1 The anxiously attached infant 142 TABLES
6. 1 The continuity of secure and insecure attachment 115
7. 1 Classical, Object-Relations and Attachment Theories compared 133
8. 1 Clinical aspects of insecure-avoidant and dismissive attachment 163
Acknowledgements
The origins of this book can be traced to a phone call from Laurie Spurling asking me whom I would like to write about. John Bowlby sprang instantly to mind - so I must start by acknowledging my gratitude to him for that invitation, and for his subsequent editorial help and suggestions.
The next 'without whom . . . ' is to the Wellcome Trust, who granted me a six-month Research Fellowship which enabled me to concentrate exclusively on the book, free from my clinical and administrative responsibilities. Sabbaticals are becoming rare enough in universities, and in the National Health Service are almost unheard of. Mine was a blissful period which not only enabled me to devote myself to the book, but also provided a perspective from which I realised how stressful and exhausting most NHS consultant posts are. So the Wellcome has saved me from burn-out (or postponed it for a while) as well as enabling the book to get written. Never was a psychotherapist so professionally indebted to a pharmaceutical firm!
Next I must express my gratitude to my colleagues in the Department of Psychiatry at North Devon District Hospital, and especially to Drs Roberts, Sewell and Van Buren, who gave their blessing to my absence, even though it meant more work for them, also to Dr Simon Nicholson, who cheerfully and efficiently took my place while I was away.
I am grateful to the Institute of Psycho-Analysis for granting me access to their archives.
Much gratitude is due to the many friends and colleagues who have taken time and effort to discuss Attachment Theory and to read all or part of the manuscript and have made many helpful suggestions which have improved its quality: Anthony Bateman, Mary Boston, Peter Fonagy, Dorothy Heard, Matthew Holmes, Sebastian Kraemer, Brian Lake, Richard Lindley, Pat Millner,
Acknowledgements xi
Oliver Reynolds, Glenn Roberts, Charles Rycroft, Anthony Storr, and Robert and Lorraine Tollemache. Needless to say, the defects of the book are entirely my responsibility.
I am very grateful for the help and encouragement I have had from members of the Bowlby family. By agreeing to be interviewed, and with their written comments, Ursula Bowlby, Mary Gatling, Sir Henry and Lady Phelps Brown, Juliet Hopkins and Marjorie Durbin gave me a fascinating window of reminiscence into their family and social life in the early part of the century.
My secretary, Patricia Bartlett, helped as always in countless ways to lighten the burden of my work and made my absence from the hospital possible by her combination of good humour, vigilance and efficiency.
Alison Housley and her staff in the library at North Devon District Hospital have been tireless in the promptness and enthusiasm with which they responded to my endless requests for references. Their contribution to making our hospital a true 'periphery of excellence' is incalculable.
Finally, I am grateful to Ros and Joshua Holmes, who, by providing me with a loving and secure base, enabled me to explore the world of Attachment Theory, and by expecting - without demanding - me to provide one for them, offered the necessary distraction without which one's work becomes stale and unbalanced.
Chapter 1 Introduction
When people start writing they think they've got to write something definitive . . . I think that is fatal. The mood to write in is 'This is quite an interesting story I've got to tell. I hope someone will be interested. Anyway it's the best I can do for the present. ' If one adopts that line one gets over it and does it.
(Bowlby in Hunter 1991)
This book has four main aims. The first, and simplest, is to present John Bowlby's story of attachment - and we shall hear much about stories in the course of the book - in a condensed and coherent way. Bowlby was a lucid and prolific exponent of his own views, but the very comprehensiveness of his work, described by one critic as having a 'Victorian monumentality about it' (Rycroft 1985), can be daunting. Despite the clarity of his thought and the charm and epigrammatic flair of his literary style, the 1,500 pages of the Attachment and Loss trilogy (Bowlby, 1969b, 1973a, 1980), covering as they do every aspect of the subject in immense detail, are hard going for the fainthearted reader. His later works, The Making and Breaking of Affectional Bonds (Bowlby 1979) and A Secure Base (Bowlby 1988), are more accessible, but as collections of essays they do not necessarily pull all his theories together into a whole. So there is a need for a survey of Bowlby's work, and also - here is a second objective - given that it is well over half a century since he published his first papers, a need to take an historical perspective on the evolution of his ideas.
The past thirty years have seen a second, and more recently a third, generation of researchers stimulated by Bowlby's seminal ideas first published in the 1960s. The
2 John Bowlby and Attachment Theory
'post-Bowlbians' - Ainsworth, Main, Bretherton, Marris and Sroufe in the United States; the Grossmanns in Germany; Parkes, Hinde, Byng-Hall and Heard in Britain - have developed Attachment Theory into a major framework of developmental psychology in ways that are highly relevant to psychotherapy. Many of their findings have been collected into two important research symposia (Parkes and Stevenson-Hinde 1982; Parkes et al. 1991), but there is no single volume explicitly devoted to the exposition of contemporary Attachment Theory and little concerted effort has been made to consider its implications for psychotherapeutic theory and practice. The need for such a work is a third objective and justification for the present work.
A fourth and more compelling reason than these worthy but perhaps mundane considerations informs much of the purpose of this book. This is the attempt to come to grips with a curious enigma which surrounds Bowlby and his work. Apart from Freud and Jung, Bowlby is one of the few psychoanalysts who have become household names and whose ideas have entered the vernacular. The ill effects (or otherwise) of maternal deprivation; the importance of bonding between parents and children; the need for a secure base and to feel attached; the realisation that grief has a course to run and can be divided into stages - these are concepts with which people far removed from the worlds of psychology and psychotherapy are familiar. All may be traced, in whole or in part, to the work of John Bowlby.
Yet Bowlby's familiarity and acceptance by the general public and his influence in a number of specialist fields such as pediatrics, developmental psychology, social work and psychiatry have never been matched within the domain of psychotherapy. In his chosen profession of psychoanalysis his influence is honoured more often in the breach than the observance. Between his papers delivered in the late 1950s and early 1960s to psychoanalytic societies in Britain and the States, and polite obsequies of the early 1990s, there has for the most part been a resounding silence from the psychoanalytical movement in response to the challenges and opportunities which his work represents. A major aim of this book is to try to understand this discrepancy between public recognition and professional avoidance, and the
Introduction 3
attempt to remedy it by showing how Attachment Theory can inform the practice of adult psychotherapy.
The details of the relationship between Bowlby and psychoanalysis will emerge gradually in the course of the book but, as an overture, a brief summarising overview will now follow. The answers to the riddle of Bowlby's rift with the psychoanalytic movement can be found at three distinct but interrelated levels: Bowlby's own personality, background and outlook; the atmosphere within the psychoanalytic society just before and in the aftermath of Freud's death; and the social and intellectual climate in the years surrounding and including the 1939-45 world war.
John Bowlby, described in an obituary as 'one of the three or four most important psychiatrists of the twentieth century' (Storr 1991), came from a conventional upper-middle class background. Cambridge educated, very 'English' in his reserve and empiricism, a 'nineteenth century Darwinian liberal' (Rycroft, 1985), he entered a psychoanalytical society in the 1930s that was riven between two warring factions led by Melanie Klein on the one hand and Anna Freud on the other (Grosskurth 1986). Melanie, a Berliner, divorced and separated from her children, the great innovator, faced the unmarried Viennese Anna, devoted to the orphans in her Hampstead nursery, defender of the true Freudian faith. Klein was powerful and domineering, but with a helpless side that meant that she depended on utter loyalty from her supporters. Anna Freud was shy and diffident, but with a steely determination not to be done down, and the confidence of her father's blessing.
The battle between the two women was ostensibly about theory. For Freud the Oedipus complex was the 'kernel of the neuroses' and he had had little to say about the early mother- infant relationship. Klein put the mother on the psychoanalytic map, arguing for the importance of phantasy in the early weeks and months of life; the primacy of the death instinct as an explanation for infantile aggression; and the need in therapy to lay bare and put into words these primitive impulses of infancy. Anna Freud - Antigone to her father's Oedipus - questioned Klein's speculations about the mind of the infant, continued to see the Oedipus complex arising at the age of two to three years as the starting point for the neuroses, and saw the role of therapy as strengthening the ego in its efforts to reconcile id and superego. Bowlby struggled
4 John Bowlby and Attachment Theory
to find his bearings in the charged atmosphere of the psychoanalytic society created by the rivalrous antagonism of these two daughters of the psychoanalytical movement, each vying for supremacy. With characteristic independence he steered a course between them, trying to work things out for himself. He took his stand on two main battlefields: the scientific status of psychoanalysis, and the role of the environment in the causation of neurosis.
Although both sides invoked the name of science in support of their ideas, this was, in Bowlby's view, little more than a genuflection to Freud's insistence that psychoanalysis should take its place as a new science of the mind. Bowlby saw both women and most of their followers as hopelessly unscientific. Neither Klein nor Anna Freud had any kind of scientific background. Both argued from intuition and authority rather than subjecting their claims to empirical testing. Neither had made any attempt to keep abreast of contemporary developments in science, or to revise Freudian metapsychology in the light of the emerging ideas about information processing and feedback which were to have such an impact on academic psychology and ethology. In rejecting Bowlby, his psychoanalytic critics on the other hand felt that by restricting himself to a narrow definition of science - to what could be observed and measured - Bowlby was missing the whole point of psychoanalysis. Any so-called 'science' of the mind which did not take account of the inner world of phantasy was worthless and certainly had no place within psychoanalytic discourse.
A similar polarisation took place around the role of the environment in the causation of neurosis. Bowlby was struck by the extent to which his patients had suffered from privation and loss, and horrified by the apparent disregard of real trauma as compared with an emphasis on the importance of autonomous phantasy in the Kleinian approach. Matters came to a head when Bowlby presented to the psychoanalytic society his famous film made with James Robertson (Robertson 1952; Bowlby and Robertson 1952b) documenting the distress shown by a small girl when separated from her parents on going into hospital. While Anna Freud endorsed Bowlby's views, the Kleinians in the audience were unimpressed, and felt that the girl's distress was due more to her unconscious destructive phantasies towards her pregnant mother's unborn baby than to the separation itself.
Bowlby was in an unusual position within the psychoanalytic society in that he was someone with non-Kleinian views who had been analysed
Introduction 5
and supervised by members of the Kleinian group (Joan Riviere and Miss Klein herself). Finding himself stuck in his analysis he decided to change to a non-Kleinian analyst, but extreme pressure was placed on him not to do so, to which, uncharacteristically, he submitted (Grosskurth 1986). He was cited by the Kleinians as evidence that they were not out to brainwash or convert all psychoanalytical candidates to their persuasion. As someone with evident ability and reputation he would have been quite a catch for whichever group he chose to join.
But both sides had reckoned without Bowlby's originality and ambition and preparedness to go out on a limb on his own. His discovery of ethology in the early 1950s provided the opportunity he was looking for to put psychoanalysis on a sound scientific footing. His World Health Organisation monograph (Bowlby 1951) and later observations of children separated from their parents enabled him to establish once and for all the importance of environmental trauma as a cause of neurosis and character disturbance. Attachment Theory was born, but rather than illuminating and strengthening Object Relations Theory as Bowlby had hoped, it was perceived by many analysts as a threat or even a betrayal. Bowlby had hoped to reconcile the warring factions within the society with his new theory, but instead they were for the most part united in either outright opposition or polite indifference to his ideas. Bowlby gradually drifted away from the society, and Attachment Theory came to stand as a discipline in its own right, owing much to psychoanalysis, but with links also to systems theory and cognitive psychology, and making a contribution as much to family and cognitive therapies as to psychoanalysis.
In retrospect the splits within the British Psycho-Analytical Society seem comparatively trivial. As Pedder (1987) puts it:
an innocent might . . . ask what all the fuss was about. Because really it could be argued that there was not a lot of disagreement. They argued about phantasy: how wide the concept should be. . . . They argued about . . . how early the Oedipus complex starts, whether at two or three or sooner. . . . They argued about the emphasis that should be placed on aggression and the death instinct, and whether neurosis is precipitated by the frustration of libido, as the Viennese thought, or [as the Kleinians saw it] by the awakening of aggression. . . . All these could be seen as matters of degree which you might think reasonable people could well discuss.
(Pedder 1987)
6 John Bowlby and Attachment Theory
But as every psychotherapist should know, things are rarely that simple. The psychoanalytical movement was still struggling with the death of its founder, searching for a direction in which to go. The polarisation between those who idealised the dead leader (the Anna Freudians) and those who dealt with their depression by a kind of manic triumphalism, a celebration of the new (the Kleinians), can be understood in terms of the very concepts that those two groups espoused. A female principle was needed to balance the phallocentrism of the earlier Freudian movement. 'The King is dead, long live the Queen' might have been their slogan. But which queen should it be? The battle for psychoanalysis was going on against a backdrop of world war, of death, dislocation and genocide. The Kleinian emphasis on autonomous phantasy, on the death instinct, on the power of psychoanalysis to heal, irrespective of environmental factors, can be seen as a desperate attempt to bring some sense of order and the possibility of control - at times omnipotently - into a world in which one could not but feel powerless and helpless. Anna Freud's emphasis on the need to strengthen the ego was an effort to hold on to reason and sanity in the face of the irrational destructiveness unleashed by war.
Bowlby was perhaps the perfect scapegoat, with his cool Englishness, his social and intellectual powers, his espousal of a narrow version of science that could not encompass the cultural breadth of the Jewish-European intelligentsia, his comparative insulation from the full horrors of war, and his Whiggish belief in the possibilities of progress based on social and scientific reason. His attempt to open out psychoanalysis to ethology and contemporary science was premature. He threatened the closed world of psychoanalysis and, offered a cold shoulder, like others before him (Jung, Adler, Ferenczi, Reich), he gave up the fight after a while and moved away to follow his own interests.
The loss was both his and that of psychoanalysis. There is something in the kernel of psychoanalysis which Bowlby seems not to have fully assimilated. In comparison with Freud's and Klein's passionate world of infantile sexuality, Attachment Theory appears almost bland, banal even. An appreciation of the power of phantasy, and the complexity of its relationship with external reality, is somehow lacking in his work. It is not loss alone that causes disturbance, but the phantasies stirred up by loss - the lack of this appreciation makes Bowlby appear at times simplistic
Introduction 7
in his formulations. But in eschewing the scientific rigour which Bowlby saw it so badly needed, psychoanalysis was held back in its development as a discipline and a therapy, a setback from which it is only just beginning to recover (see Peterfreund 1983; Stern 1985). Perhaps there was something in the climate of the 1950s which made such a split inevitable. The divide between the 'two cultures' epitomised by the belief in the possibility of progress based on science advocated by C. P. Snow, and Leavis's moral condemnation of an illiterate and degenerate society was just too great to bridge (Holmes 1992). Psychoanalysis became increasingly identified with 'culture' - with the imagination, linguistics and the moral and aesthetic dimension (Rycroft 1985; Rustin 1991), while Attachment Theory gathered momentum as a part of scientific psychology, taking root in the United States in a way that seemed less possible in a Britain that was so split in its educational and bureaucratic structures between art and science.
But times have changed. The old certainties no longer hold. Psychoanalysis has lost its dogmatism and is much more open to empirical evidence and to cross-disciplinary influence. The Berlin wall which separated psychoanalysis from the superficiality but also the stimulus of other disciplines has come down. The debate about the scientific status of psychoanalysis, and the role of the environment in neurosis, continues, but it is no longer a matter of life and death. Each side can claim partial victories. Klein was right in her emphasis on the early weeks and months of life - there is abundant evidence of psychic life from the moment of birth (Stern 1985). She was probably wrong in her insistence on the universality of the paranoid-schizoid position - it seems likely that splitting and projection predominate only in anxiously attached infants. She was right to emphasise loss and separation as central themes in character formation from the earliest years, but wrong in the concreteness of her thinking - she believed that bottle feeding could never substitute for the breast and that the events surrounding weaning were critical determinants of character. It seems likely that it is the style and general handling of the infant that matters, not the specific events, unless these are overwhelmingly traumatic. In therapy she was right to emphasise the central importance of the relationship between therapist and patient, but wrong in her belief that only 'deep', 'Kleinian' interpretations would be effective: the strength of the therapist- patient attachment is a crucial determining factor in the outcome
8 John Bowlby and Attachment Theory
of therapy, but the nature of the interpretations, as long as they are reasonably sensible, coherent and brief, is not (Holmes 1991). As increasing evidence of early trauma appears in the histories of patients with major character disorder (Grant 1991), Bowlby's emphasis on the importance of the environment as a determinant of pathology appears to be vindicated, but he also tended to be too concrete and specific in his hypotheses - it is not the loss of a parent in itself that is traumatic but the family discord or disruption surrounding it that causes the damage (Rutter 1981).
Klein showed how an individual's inner world shapes their perception of the object, and how, through projective identification, the object is coerced into feeling and behaving according to the projections it receives. In contrast with this near solipsistic account, Bowlby is concerned primarily with the impact of the object on the self. The self, which in his theories tends to be almost passive, is moulded by the inadequacies and absences of the object. We shall explore how the interactive view of self and object postulated by Winnicott (1965) and Bion (1978) and observed by developmental psychologists like Stern (1985) and Brazelton and Cramer (1991) offers the possibility of a long overdue climate of reconciliation and new understanding.
Bowlby was always careful to distinguish between the scientific and therapeutic aspects of psychoanalysis. As a scientist he was struggling for simplicity and clarity and for general principles, while therapy inevitably concerns itself with complexity and concreteness of the individual case. Much of the disagreement between Bowlby and psychoanalysis appears to rest on a confusion of these two aspects. Bowlby's main concern was to find a firm scientific underpinning to the Object Relations approach, and Attachment Theory, with its marrying of ethology to the developmental ideas of psychoanalysis, can be seen in that light. Although couched in the language of science, psychoanalytic therapy has come increasingly to be seen as a hermeneutic discipline, more concerned with meanings than mechanism, in which patient and therapist collaboratively develop a coherent narrative about the patient's experience. Such objectification and coherence are in themselves therapeutic, irrespective of the validity or otherwise of the meanings that are found. An extreme illustration of this comes from the finding that schizophrenic patients with complex and coherent delusional systems are better able to function socially than those who lack such meanings,
Introduction 9
however idiosyncratic (Roberts 1992). Bowlby's work has ensured that clinical hypotheses based on Object Relations Theory with a scientific underpinning of Attachment Theory are unlikely to be far removed from the truth, or to be tainted by totally unjustified speculation.
As we shall see in Chapter 6, recent developments in Attachment Theory suggest an exciting bridge between the narrative approach of contemporary psychoanalysis and the science of developmental psychology. There is a strong link between the kinds of attachment patterns found in infancy and the narratives that people tell about themselves several years later. Put briefly, securely attached children tell coherent stories about their lives, however difficult they have been, while insecurely attached children have much greater difficulty in narrative competence, either dismissing their past or remaining bogged down in it, and in neither case being able to talk objectively about it. The therapeutic implications of this are self-evident. Good therapy, like good parenting, provides the security and space within which a healing narrative can begin to emerge.
1 The anxiously attached infant 142 TABLES
6. 1 The continuity of secure and insecure attachment 115
7. 1 Classical, Object-Relations and Attachment Theories compared 133
8. 1 Clinical aspects of insecure-avoidant and dismissive attachment 163
Acknowledgements
The origins of this book can be traced to a phone call from Laurie Spurling asking me whom I would like to write about. John Bowlby sprang instantly to mind - so I must start by acknowledging my gratitude to him for that invitation, and for his subsequent editorial help and suggestions.
The next 'without whom . . . ' is to the Wellcome Trust, who granted me a six-month Research Fellowship which enabled me to concentrate exclusively on the book, free from my clinical and administrative responsibilities. Sabbaticals are becoming rare enough in universities, and in the National Health Service are almost unheard of. Mine was a blissful period which not only enabled me to devote myself to the book, but also provided a perspective from which I realised how stressful and exhausting most NHS consultant posts are. So the Wellcome has saved me from burn-out (or postponed it for a while) as well as enabling the book to get written. Never was a psychotherapist so professionally indebted to a pharmaceutical firm!
Next I must express my gratitude to my colleagues in the Department of Psychiatry at North Devon District Hospital, and especially to Drs Roberts, Sewell and Van Buren, who gave their blessing to my absence, even though it meant more work for them, also to Dr Simon Nicholson, who cheerfully and efficiently took my place while I was away.
I am grateful to the Institute of Psycho-Analysis for granting me access to their archives.
Much gratitude is due to the many friends and colleagues who have taken time and effort to discuss Attachment Theory and to read all or part of the manuscript and have made many helpful suggestions which have improved its quality: Anthony Bateman, Mary Boston, Peter Fonagy, Dorothy Heard, Matthew Holmes, Sebastian Kraemer, Brian Lake, Richard Lindley, Pat Millner,
Acknowledgements xi
Oliver Reynolds, Glenn Roberts, Charles Rycroft, Anthony Storr, and Robert and Lorraine Tollemache. Needless to say, the defects of the book are entirely my responsibility.
I am very grateful for the help and encouragement I have had from members of the Bowlby family. By agreeing to be interviewed, and with their written comments, Ursula Bowlby, Mary Gatling, Sir Henry and Lady Phelps Brown, Juliet Hopkins and Marjorie Durbin gave me a fascinating window of reminiscence into their family and social life in the early part of the century.
My secretary, Patricia Bartlett, helped as always in countless ways to lighten the burden of my work and made my absence from the hospital possible by her combination of good humour, vigilance and efficiency.
Alison Housley and her staff in the library at North Devon District Hospital have been tireless in the promptness and enthusiasm with which they responded to my endless requests for references. Their contribution to making our hospital a true 'periphery of excellence' is incalculable.
Finally, I am grateful to Ros and Joshua Holmes, who, by providing me with a loving and secure base, enabled me to explore the world of Attachment Theory, and by expecting - without demanding - me to provide one for them, offered the necessary distraction without which one's work becomes stale and unbalanced.
Chapter 1 Introduction
When people start writing they think they've got to write something definitive . . . I think that is fatal. The mood to write in is 'This is quite an interesting story I've got to tell. I hope someone will be interested. Anyway it's the best I can do for the present. ' If one adopts that line one gets over it and does it.
(Bowlby in Hunter 1991)
This book has four main aims. The first, and simplest, is to present John Bowlby's story of attachment - and we shall hear much about stories in the course of the book - in a condensed and coherent way. Bowlby was a lucid and prolific exponent of his own views, but the very comprehensiveness of his work, described by one critic as having a 'Victorian monumentality about it' (Rycroft 1985), can be daunting. Despite the clarity of his thought and the charm and epigrammatic flair of his literary style, the 1,500 pages of the Attachment and Loss trilogy (Bowlby, 1969b, 1973a, 1980), covering as they do every aspect of the subject in immense detail, are hard going for the fainthearted reader. His later works, The Making and Breaking of Affectional Bonds (Bowlby 1979) and A Secure Base (Bowlby 1988), are more accessible, but as collections of essays they do not necessarily pull all his theories together into a whole. So there is a need for a survey of Bowlby's work, and also - here is a second objective - given that it is well over half a century since he published his first papers, a need to take an historical perspective on the evolution of his ideas.
The past thirty years have seen a second, and more recently a third, generation of researchers stimulated by Bowlby's seminal ideas first published in the 1960s. The
2 John Bowlby and Attachment Theory
'post-Bowlbians' - Ainsworth, Main, Bretherton, Marris and Sroufe in the United States; the Grossmanns in Germany; Parkes, Hinde, Byng-Hall and Heard in Britain - have developed Attachment Theory into a major framework of developmental psychology in ways that are highly relevant to psychotherapy. Many of their findings have been collected into two important research symposia (Parkes and Stevenson-Hinde 1982; Parkes et al. 1991), but there is no single volume explicitly devoted to the exposition of contemporary Attachment Theory and little concerted effort has been made to consider its implications for psychotherapeutic theory and practice. The need for such a work is a third objective and justification for the present work.
A fourth and more compelling reason than these worthy but perhaps mundane considerations informs much of the purpose of this book. This is the attempt to come to grips with a curious enigma which surrounds Bowlby and his work. Apart from Freud and Jung, Bowlby is one of the few psychoanalysts who have become household names and whose ideas have entered the vernacular. The ill effects (or otherwise) of maternal deprivation; the importance of bonding between parents and children; the need for a secure base and to feel attached; the realisation that grief has a course to run and can be divided into stages - these are concepts with which people far removed from the worlds of psychology and psychotherapy are familiar. All may be traced, in whole or in part, to the work of John Bowlby.
Yet Bowlby's familiarity and acceptance by the general public and his influence in a number of specialist fields such as pediatrics, developmental psychology, social work and psychiatry have never been matched within the domain of psychotherapy. In his chosen profession of psychoanalysis his influence is honoured more often in the breach than the observance. Between his papers delivered in the late 1950s and early 1960s to psychoanalytic societies in Britain and the States, and polite obsequies of the early 1990s, there has for the most part been a resounding silence from the psychoanalytical movement in response to the challenges and opportunities which his work represents. A major aim of this book is to try to understand this discrepancy between public recognition and professional avoidance, and the
Introduction 3
attempt to remedy it by showing how Attachment Theory can inform the practice of adult psychotherapy.
The details of the relationship between Bowlby and psychoanalysis will emerge gradually in the course of the book but, as an overture, a brief summarising overview will now follow. The answers to the riddle of Bowlby's rift with the psychoanalytic movement can be found at three distinct but interrelated levels: Bowlby's own personality, background and outlook; the atmosphere within the psychoanalytic society just before and in the aftermath of Freud's death; and the social and intellectual climate in the years surrounding and including the 1939-45 world war.
John Bowlby, described in an obituary as 'one of the three or four most important psychiatrists of the twentieth century' (Storr 1991), came from a conventional upper-middle class background. Cambridge educated, very 'English' in his reserve and empiricism, a 'nineteenth century Darwinian liberal' (Rycroft, 1985), he entered a psychoanalytical society in the 1930s that was riven between two warring factions led by Melanie Klein on the one hand and Anna Freud on the other (Grosskurth 1986). Melanie, a Berliner, divorced and separated from her children, the great innovator, faced the unmarried Viennese Anna, devoted to the orphans in her Hampstead nursery, defender of the true Freudian faith. Klein was powerful and domineering, but with a helpless side that meant that she depended on utter loyalty from her supporters. Anna Freud was shy and diffident, but with a steely determination not to be done down, and the confidence of her father's blessing.
The battle between the two women was ostensibly about theory. For Freud the Oedipus complex was the 'kernel of the neuroses' and he had had little to say about the early mother- infant relationship. Klein put the mother on the psychoanalytic map, arguing for the importance of phantasy in the early weeks and months of life; the primacy of the death instinct as an explanation for infantile aggression; and the need in therapy to lay bare and put into words these primitive impulses of infancy. Anna Freud - Antigone to her father's Oedipus - questioned Klein's speculations about the mind of the infant, continued to see the Oedipus complex arising at the age of two to three years as the starting point for the neuroses, and saw the role of therapy as strengthening the ego in its efforts to reconcile id and superego. Bowlby struggled
4 John Bowlby and Attachment Theory
to find his bearings in the charged atmosphere of the psychoanalytic society created by the rivalrous antagonism of these two daughters of the psychoanalytical movement, each vying for supremacy. With characteristic independence he steered a course between them, trying to work things out for himself. He took his stand on two main battlefields: the scientific status of psychoanalysis, and the role of the environment in the causation of neurosis.
Although both sides invoked the name of science in support of their ideas, this was, in Bowlby's view, little more than a genuflection to Freud's insistence that psychoanalysis should take its place as a new science of the mind. Bowlby saw both women and most of their followers as hopelessly unscientific. Neither Klein nor Anna Freud had any kind of scientific background. Both argued from intuition and authority rather than subjecting their claims to empirical testing. Neither had made any attempt to keep abreast of contemporary developments in science, or to revise Freudian metapsychology in the light of the emerging ideas about information processing and feedback which were to have such an impact on academic psychology and ethology. In rejecting Bowlby, his psychoanalytic critics on the other hand felt that by restricting himself to a narrow definition of science - to what could be observed and measured - Bowlby was missing the whole point of psychoanalysis. Any so-called 'science' of the mind which did not take account of the inner world of phantasy was worthless and certainly had no place within psychoanalytic discourse.
A similar polarisation took place around the role of the environment in the causation of neurosis. Bowlby was struck by the extent to which his patients had suffered from privation and loss, and horrified by the apparent disregard of real trauma as compared with an emphasis on the importance of autonomous phantasy in the Kleinian approach. Matters came to a head when Bowlby presented to the psychoanalytic society his famous film made with James Robertson (Robertson 1952; Bowlby and Robertson 1952b) documenting the distress shown by a small girl when separated from her parents on going into hospital. While Anna Freud endorsed Bowlby's views, the Kleinians in the audience were unimpressed, and felt that the girl's distress was due more to her unconscious destructive phantasies towards her pregnant mother's unborn baby than to the separation itself.
Bowlby was in an unusual position within the psychoanalytic society in that he was someone with non-Kleinian views who had been analysed
Introduction 5
and supervised by members of the Kleinian group (Joan Riviere and Miss Klein herself). Finding himself stuck in his analysis he decided to change to a non-Kleinian analyst, but extreme pressure was placed on him not to do so, to which, uncharacteristically, he submitted (Grosskurth 1986). He was cited by the Kleinians as evidence that they were not out to brainwash or convert all psychoanalytical candidates to their persuasion. As someone with evident ability and reputation he would have been quite a catch for whichever group he chose to join.
But both sides had reckoned without Bowlby's originality and ambition and preparedness to go out on a limb on his own. His discovery of ethology in the early 1950s provided the opportunity he was looking for to put psychoanalysis on a sound scientific footing. His World Health Organisation monograph (Bowlby 1951) and later observations of children separated from their parents enabled him to establish once and for all the importance of environmental trauma as a cause of neurosis and character disturbance. Attachment Theory was born, but rather than illuminating and strengthening Object Relations Theory as Bowlby had hoped, it was perceived by many analysts as a threat or even a betrayal. Bowlby had hoped to reconcile the warring factions within the society with his new theory, but instead they were for the most part united in either outright opposition or polite indifference to his ideas. Bowlby gradually drifted away from the society, and Attachment Theory came to stand as a discipline in its own right, owing much to psychoanalysis, but with links also to systems theory and cognitive psychology, and making a contribution as much to family and cognitive therapies as to psychoanalysis.
In retrospect the splits within the British Psycho-Analytical Society seem comparatively trivial. As Pedder (1987) puts it:
an innocent might . . . ask what all the fuss was about. Because really it could be argued that there was not a lot of disagreement. They argued about phantasy: how wide the concept should be. . . . They argued about . . . how early the Oedipus complex starts, whether at two or three or sooner. . . . They argued about the emphasis that should be placed on aggression and the death instinct, and whether neurosis is precipitated by the frustration of libido, as the Viennese thought, or [as the Kleinians saw it] by the awakening of aggression. . . . All these could be seen as matters of degree which you might think reasonable people could well discuss.
(Pedder 1987)
6 John Bowlby and Attachment Theory
But as every psychotherapist should know, things are rarely that simple. The psychoanalytical movement was still struggling with the death of its founder, searching for a direction in which to go. The polarisation between those who idealised the dead leader (the Anna Freudians) and those who dealt with their depression by a kind of manic triumphalism, a celebration of the new (the Kleinians), can be understood in terms of the very concepts that those two groups espoused. A female principle was needed to balance the phallocentrism of the earlier Freudian movement. 'The King is dead, long live the Queen' might have been their slogan. But which queen should it be? The battle for psychoanalysis was going on against a backdrop of world war, of death, dislocation and genocide. The Kleinian emphasis on autonomous phantasy, on the death instinct, on the power of psychoanalysis to heal, irrespective of environmental factors, can be seen as a desperate attempt to bring some sense of order and the possibility of control - at times omnipotently - into a world in which one could not but feel powerless and helpless. Anna Freud's emphasis on the need to strengthen the ego was an effort to hold on to reason and sanity in the face of the irrational destructiveness unleashed by war.
Bowlby was perhaps the perfect scapegoat, with his cool Englishness, his social and intellectual powers, his espousal of a narrow version of science that could not encompass the cultural breadth of the Jewish-European intelligentsia, his comparative insulation from the full horrors of war, and his Whiggish belief in the possibilities of progress based on social and scientific reason. His attempt to open out psychoanalysis to ethology and contemporary science was premature. He threatened the closed world of psychoanalysis and, offered a cold shoulder, like others before him (Jung, Adler, Ferenczi, Reich), he gave up the fight after a while and moved away to follow his own interests.
The loss was both his and that of psychoanalysis. There is something in the kernel of psychoanalysis which Bowlby seems not to have fully assimilated. In comparison with Freud's and Klein's passionate world of infantile sexuality, Attachment Theory appears almost bland, banal even. An appreciation of the power of phantasy, and the complexity of its relationship with external reality, is somehow lacking in his work. It is not loss alone that causes disturbance, but the phantasies stirred up by loss - the lack of this appreciation makes Bowlby appear at times simplistic
Introduction 7
in his formulations. But in eschewing the scientific rigour which Bowlby saw it so badly needed, psychoanalysis was held back in its development as a discipline and a therapy, a setback from which it is only just beginning to recover (see Peterfreund 1983; Stern 1985). Perhaps there was something in the climate of the 1950s which made such a split inevitable. The divide between the 'two cultures' epitomised by the belief in the possibility of progress based on science advocated by C. P. Snow, and Leavis's moral condemnation of an illiterate and degenerate society was just too great to bridge (Holmes 1992). Psychoanalysis became increasingly identified with 'culture' - with the imagination, linguistics and the moral and aesthetic dimension (Rycroft 1985; Rustin 1991), while Attachment Theory gathered momentum as a part of scientific psychology, taking root in the United States in a way that seemed less possible in a Britain that was so split in its educational and bureaucratic structures between art and science.
But times have changed. The old certainties no longer hold. Psychoanalysis has lost its dogmatism and is much more open to empirical evidence and to cross-disciplinary influence. The Berlin wall which separated psychoanalysis from the superficiality but also the stimulus of other disciplines has come down. The debate about the scientific status of psychoanalysis, and the role of the environment in neurosis, continues, but it is no longer a matter of life and death. Each side can claim partial victories. Klein was right in her emphasis on the early weeks and months of life - there is abundant evidence of psychic life from the moment of birth (Stern 1985). She was probably wrong in her insistence on the universality of the paranoid-schizoid position - it seems likely that splitting and projection predominate only in anxiously attached infants. She was right to emphasise loss and separation as central themes in character formation from the earliest years, but wrong in the concreteness of her thinking - she believed that bottle feeding could never substitute for the breast and that the events surrounding weaning were critical determinants of character. It seems likely that it is the style and general handling of the infant that matters, not the specific events, unless these are overwhelmingly traumatic. In therapy she was right to emphasise the central importance of the relationship between therapist and patient, but wrong in her belief that only 'deep', 'Kleinian' interpretations would be effective: the strength of the therapist- patient attachment is a crucial determining factor in the outcome
8 John Bowlby and Attachment Theory
of therapy, but the nature of the interpretations, as long as they are reasonably sensible, coherent and brief, is not (Holmes 1991). As increasing evidence of early trauma appears in the histories of patients with major character disorder (Grant 1991), Bowlby's emphasis on the importance of the environment as a determinant of pathology appears to be vindicated, but he also tended to be too concrete and specific in his hypotheses - it is not the loss of a parent in itself that is traumatic but the family discord or disruption surrounding it that causes the damage (Rutter 1981).
Klein showed how an individual's inner world shapes their perception of the object, and how, through projective identification, the object is coerced into feeling and behaving according to the projections it receives. In contrast with this near solipsistic account, Bowlby is concerned primarily with the impact of the object on the self. The self, which in his theories tends to be almost passive, is moulded by the inadequacies and absences of the object. We shall explore how the interactive view of self and object postulated by Winnicott (1965) and Bion (1978) and observed by developmental psychologists like Stern (1985) and Brazelton and Cramer (1991) offers the possibility of a long overdue climate of reconciliation and new understanding.
Bowlby was always careful to distinguish between the scientific and therapeutic aspects of psychoanalysis. As a scientist he was struggling for simplicity and clarity and for general principles, while therapy inevitably concerns itself with complexity and concreteness of the individual case. Much of the disagreement between Bowlby and psychoanalysis appears to rest on a confusion of these two aspects. Bowlby's main concern was to find a firm scientific underpinning to the Object Relations approach, and Attachment Theory, with its marrying of ethology to the developmental ideas of psychoanalysis, can be seen in that light. Although couched in the language of science, psychoanalytic therapy has come increasingly to be seen as a hermeneutic discipline, more concerned with meanings than mechanism, in which patient and therapist collaboratively develop a coherent narrative about the patient's experience. Such objectification and coherence are in themselves therapeutic, irrespective of the validity or otherwise of the meanings that are found. An extreme illustration of this comes from the finding that schizophrenic patients with complex and coherent delusional systems are better able to function socially than those who lack such meanings,
Introduction 9
however idiosyncratic (Roberts 1992). Bowlby's work has ensured that clinical hypotheses based on Object Relations Theory with a scientific underpinning of Attachment Theory are unlikely to be far removed from the truth, or to be tainted by totally unjustified speculation.
As we shall see in Chapter 6, recent developments in Attachment Theory suggest an exciting bridge between the narrative approach of contemporary psychoanalysis and the science of developmental psychology. There is a strong link between the kinds of attachment patterns found in infancy and the narratives that people tell about themselves several years later. Put briefly, securely attached children tell coherent stories about their lives, however difficult they have been, while insecurely attached children have much greater difficulty in narrative competence, either dismissing their past or remaining bogged down in it, and in neither case being able to talk objectively about it. The therapeutic implications of this are self-evident. Good therapy, like good parenting, provides the security and space within which a healing narrative can begin to emerge.
Psychoanalysis, perhaps more than it would care to admit, is influenced by the prevailing cultural climate. The Oedipus complex with its emphasis on castration anxiety reflected the patriarchy of its day. With the weakening of paternal power within the family came the rise of the female principle within psychoanalysis. The Society which Bowlby joined in the 1930s was dominated by strong women: Melanie Klein, Anna Freud, Joan Riviere, Sylvia Payne, Susan Isaacs, Paula Heimann and many others. Ernest Jones's power was waning, and Glover's grip on the Society was gradually being loosened. The main theorists of the post-war period - Klein, Bion, Winnicott and Bowlby - were all concerned with the role of the mother. A new phase of deconstruction has begun which emphasises the reciprocities of reader and writer, social, cultural and racial pluralism. We are entering an era of therapeutic co-constructionism - far removed from the ex- cathedra interpretations of classical therapy - where therapist and patient collaboratively build up a picture of their world and history. We have moved, perhaps, from the father principle, through the maternal, to the era of the sibling, in which, however different in their roles, there is a fundamental symmetry between patient and therapist.
This brings us to a concluding note about the nature of biography. A biographer is, in a sense, both patient and therapist
10 John Bowlby and Attachment Theory
in relation to the person he chooses to write about. As Gathorne- Hardy (1992) points out, there is an inevitable positive transference to one's subject; how else could one justify the long hours spent (far exceeding any psychoanalysis, however interminable) reading, studying, thinking about them? Biographers identify with their subjects, just as patients identify with their therapists, and see them in a way that is inevitably influenced and may be biased by their own themes and preoccupations. At the same time, biographers as 'therapists' have an opportunity to see their subjects as they really are: but with that privilege must also take into account their own counter-transferential tendencies towards voyeurism, prurience, envy, denigration and idealisation. Biographers should approach their subjects in the same spirit in which therapists see their patients: compassionately without becoming over-involved, with objectivity but without excessive detachment, with a sense of the uniqueness and specialness of the individual but without indulgence. The aim of this book is to form a working alliance with Attachment Theory and its originator - to see them in their strengths and limitations, their possibilities and blind spots. Although this book is not primarily a biography, some biographical preliminaries are therefore inevitably needed and it is to them that we must now turn.
Part Origins
Chapter 2 Biographical
Parents, especially mothers, are much-maligned people. (Bowlby 1988a)
A family photograph, taken just before the First World War, shows Lady Bowlby surrounded by her six children. Her husband, Sir Anthony, the King's surgeon, is not there - he is, as usual, at work. She is flanked by her two favourite sons, John and Tony, aged about four and five, looking boldly and brightly into the camera. On her lap sits the baby Evelyn. The two older girls, aged eight and eleven, stand dutifully and demurely to one side. Finally there is two-year-old Jim, the weak member of the family, dubbed a 'late developer', lacking the physical and intellectual vigour of his brothers and sisters. A hand appears around his waist, partly propping him up. But whose hand can it be? Is it his mother's? No, hers are firmly around the baby - a rare moment of physical closeness, as it turned out. Can it be one of his older sisters? No, their hands are politely by their sides. It is in fact the hand of an invisible nurse, crouching behind the tableau vivant, the tiny and perfectionist 'Nanna Friend' who, with the nursemaids and governess, provided the child care in this fairly typical example of the English haute bourgeoisie on the threshold of the modern era.
Bowlby was notoriously reticent about his background and early family life. In the Adult Attachment Interview (see Chapter 6) he might have been rated as 'dismissive', giving the kind of response in which a person describes their childhood as 'perfectly all right' and refuses to be drawn further - a pattern that is strongly correlated with an insecure-avoidant pattern of attachment. But, as his book on Darwin testifies, Bowlby found the task of psychobiography worthwhile, making a strong case
14 Origins
for considering Darwin's recurrent anxiety attacks as a manifestation of his inability to grieve loss, the pattern for which was set by his mother's death when he was eight. Whereas the main purpose of this book is an exposition of Attachment Theory, the aim of this chapter is to consider Bowlby's life and personality as a background to his ideas and to explore the relationship between them. The chapter is divided into three parts: the first is a chronological account of his life and career, touching on much that will be developed in subsequent chapters; the second consists of an assessment of his character, based on reminiscences of his family, friends and colleagues; and the third considers some of the major personal themes and preoccupations which inform Bowlby's work.
BOWLBY'S LIFE Childhood and youth
Edward John Mostyn Bowlby was born on 26 February 1907. His father, whom he resembled in many ways, was Major-General Sir Anthony Bowlby (1855-1929), a successful London surgeon who had operated on one of Queen Victoria's sons, and was rewarded with a knighthood for his appointments as Royal Surgeon to King Edward VII and King George V, and a baronetcy on becoming President of the Royal College of Surgeons in 1920. John's grandfather, 'Thomas Bowlby of The Times', was a foreign correspondent for The Times who was murdered in Peking in 1861 during the Opium Wars when Sir Anthony was a small child. Anthony felt responsible for his mother, who did not remarry, and he only began to look for a wife after her death when he was forty. He was introduced by a mutual friend at a house party to the well-connected May Mostyn, then thirty, and pursued her (mainly on bicycles - they shared a love of the countryside and outdoor life) until they were married less than a year later.