the manifold social and
psychological
conditions that influence its development for better or worse.
A-Secure-Base-Bowlby-Johnf
There are probably many reasons for this.
One is that initially the data drawn on ap- peared to be unduly behavioural.
Another is that clinicians are very busy people who are naturally reluctant to spend time trying to master a new and strange conceptual framework until they have strong reasons for believing that to do so will improve their clinical understanding and therapeutic skills.
For those who have decided the time has come to sample what this new per- spective has to offer I hope the lectures gathered here may provide a convenient introduction.
ACKNOWLEDGEMENTS
During the past ten years I have had the great be- nefit of frequent communication with staff and students at the Tavistock Clinic and also with a number of colleagues engaged in pioneering studies of how patterns of attachment develop during infancy and childhood. To all of them I owe a deep debt of gratitude, often for useful sug- gestions, sometimes for necessary corrections, and always for stimulation and encouragement. To my secretary, Dorothy Southern, I also owe a deep debt of gratitude for many years of devoted service during which she has made my interests her own.
For editorial assistance in preparing these lec- tures for publication and for constructing the in- dex my thanks are due to Molly Townsend.
The first six lectures in this book have ap- peared in other publications and I am grateful to the publishers concerned for permission to repro- duce them here. Lecture 1 was chapter 18 in Par- enthood: A Psychodynamic Perspective edited by Rebecca S. Cohen, Bertram J. Cohler, and Sidney H. Weissman, the Guilford Press, New York
12/362
? ? ? (1984); Lecture 2 was 'Attachment and loss: ret- rospect and prospect', American Journal of Orthopsychiatry 52: 664-78 (1982); Lecture 3 was 'Psychoanalysis as art and science', Interna- tional Review of Psychoanalysis 6: 3-14 (1979); Lecture 4 was 'Psychoanalysis as a natural sci- ence', International Review of Psychoanalysis 8: 243-56 (1981); Lecture 5 was 'Violence in the family as a disorder of the attachment and care- giving systems', The American Journal of Psy- choanalysis 44: 9-27 (1984); Lecture 6 was chapter 6 in Cognition and Psychotherapy edited by Michael J. Mahoney and Arthur Freeman, Plenum Publishing Corporation, New York and London (1985), expanded from 'On knowing what you are not supposed to know and feeling what you are not supposed to feel', Canadian Journal of Psychiatry 24: 403-8 (1979).
JOHN BOWLBY
PREFACE TO THE ROUTLEDGE CLASSICS EDITION
A Secure Base first appeared in 1988 when John Bowlby was in his 81st year. Although there was, amazingly, one more book to come--the Darwin biography--it is his final contribution to Attach- ment Theory, the discipline which he, with Mary Ainsworth's help, founded nearly half a century earlier. Thus A Secure Base has a valedictory feel to it--a summation of a life's work, but also a tribute and a handing over to the next generation of attachment researchers and clinicians.
In it are to be found all the familiar Bowlbian themes--theoretical, etiological, methodological, clinical, and political. He restates the conceptual foundation stones of his ideas: the primacy of the attachment behavioural response and its role in protection from predation; sensitive care-giving as a foundation for psychological health; the con- tinuing importance of attachment throughout the life cycle. He argues powerfully for the role of real-life adversity--emotional deprivation, un- mourned bereavement, rejection, obfuscation, neglect, physical and sexual abuse--as the origin of subsequent psychopathology, as opposed to
14/362
? ? ? putative endo-psychic entities such as 'death instinct'.
Methodologically he emphasizes the import- ance of systematic scientific observation of chil- dren and parents, as opposed to speculative re- constructions from the couch. Clinically he sees the therapist as providing a secure base for her patients, a springboard from which they can be- gin to develop the free flowing discourse of emo- tion that is characteristic of those who are se- curely attached.
Finally, there is the figured base of Bowlby's social philosophy lying at the heart of his work: 'man and woman power devoted to the produc- tion of happy, healthy, and self-reliant children . . . does not count at all. We have created a topsy- turvy world' (p. 2).
The past 25 years has seen an explosion of in- terest in attachment theory, culminating in the landmark volume Handbook of Attachment (Cassidy & Shaver, 1999), an exhilarating sum- mation of how Bowlby and Ainsworth's acorn, nurtured by them into a sturdy sapling, has seeded a whole forest of developments, applica- tions, and ideas. In this brief introduction I shall pick out three of these growth points, or 'recent
15/362
? ? ? advances' in attachment theory, hinted at by Bowlby, but greatly expanded since A Secure Base was first published: the role of fathers in creating secure attachments; mentalization and 'theory of mind' as a developmental achievement; and psychotherapy as an interpersonal enterprise.
Attachment to Fathers. As implied in the quo- tation above, Bowlby was always insistent that mothers and fathers mattered when it came to providing a secure base. One of the crucial planks in the argument that security of attachment is an interpersonal, interactive phenomenon, and not simply a matter of the child's inborn tempera- ment, is that one and the same child can be clas- sified in the Strange Situation as secure with one parent and insecure with another. Nevertheless, attachment theory, both in its research and clin- ical guises, has tended to be a somewhat mater- nocentric enterprise and it has not been easy to pin down precisely what the father's contribution to security of attachment comprises.
Recent work (Grossman, Grossman, & Zim- merman, 1999; Grossman, Grossman, & Zindler, 2005) has begun to throw light on this topic. When Bowlby was writing the essays and lectures
16/362
? ? ? that make up this book in the 1980s, longitudinal studies of attachment phenomena were them- selves in their infancy. We now have 20-year pro- spective studies that look at measures of attach- ment security, parental sensitivity, exploration, relational competence, and their mental repres- entations throughout childhood. These can now be correlated with attachment disposition in young adulthood as manifested in attitudes to- wards romantic relationships and the Adult At- tachment Interview.
These studies show that paternal contributions are indeed vital to secure, stable, exploratory, balanced, verbally fluent attachment dispositions in adulthood. However, this contribution to psy- chological health is not predominantly mediated via security of attachment as measured in the Strange Situation. The role of fathers, rather, makes itself felt via the exploratory dimension of the attachment/exploration dichotomy and is eli- cited in the 'SCIP' (Sensitive and Challenging In- teractive Play) measure (Grossman, Grossman, & Zindler, 2005), which observes and rates parents interacting with their children in a 10-minute play session.
17/362
? ? ? In general, combined mother-father scores along multiple dimensions of attachment in childhood are far more predictive of security or insecurity of attachment representation in adult- hood than those of either parent alone. However, the preoccupied dimension--adults who give confused, affect-laden, unstructured responses to probes--is strongly correlated with father rejec- tion and insensitivity in mid-childhood, with the maternal contribution being relatively weak. Thus it seems that good-enough fathers help their children to develop clarity of thought and to be able to face up to negative emotions without feeling overwhelmed.
Like mothers, fathers need to be sensitive, but this takes the form of praise, encouragement, and the capacity to sustain positive affect in their off- spring. As they help their children to cope with curiosity-wariness conflicts, this protective, chal- lenging, 'you can do it' father diverges markedly from the castrating picture of classical psycho- analytic theory, which might apply more accur- ately to insensitive fathers who fail intuitively to grasp the basic message of attachment: that achievement is always predicated on security.
18/362
? ? ? Mentalization and theory of mind. Bowlby lived long enough to appreciate the enormous significance of Mary Main's contribution to At- tachment Theory and in particular the possibilit- ies represented by her development of the Adult Attachment Interview (AAI) (Hesse, 1999). He quotes her finding, confirmed by others, that mothers who showed insecurity on the AAI were more likely to have insecure children as meas- ured in the Strange Situation. He saw insecure at- tachment in terms of psychological defences--ne- cessary for emotional survival (and in the envir- onment of evolutionary adaptiveness in which our species evolved, physical survival)--but also as a constricting factor that excludes the insecure individual from the possibility of processing ad- verse experience.
A crucial finding in this early work was that 're- flexive function' (Fonagy et al. , 2002) as meas- ured in the AAI, appears to be a protective factor that despite adverse childhood experience such as parental separation, bereavement or even neg- lect and abuse, enables individuals to remain se- cure themselves and to provide security for their offspring. Put simply, the ability to 'talk about it' mitigates the long-term negative consequences of
19/362
? ? ? childhood trauma. 'Reflexive function' can be seen as tapping into internal speech, or mental representation of experience that underlies this capacity for external interactive story-telling.
Fonagy and his co-workers (Fonagy, Gergely, Jurist, & Target, 2002; Bateman & Fonagy, 2004) have extended these early findings with their new concept of 'mentalization'. Drawing on the philo- sophical tradition of 'theory of mind' they suggest that there are crucial developmental processes which enable infants to begin to appreciate the fact that they themselves, and those around them, have 'minds'--i. e. the ability to represent the world and to have projects, beliefs, and de- sires. Mentalization enables us to differentiate between 'reality' and our perspective or appreci- ation of reality, and also to grasp the fact that dif- ferent people view the world in differing ways. Thus mentalization, which could be seen as an extension of Bowlby's notion of internal working models, is anti-narcissistic and, so Fonagy et al. argue, a vital component of the ability to interact socially, including the ability to survive the vicis- situdes of mis-attunement, alliance ruptures, and the minor failures that are a normal part of
20/362
? ? ? 'good-enough parenting', or even 'not-good- enough' parenting.
This set of ideas leads to a rather different take on the evolutionary significance of attachment, which Bowlby, drawing on an ethological per- spective, always insisted was that of protection from predation. Fonagy et al. (2002) suggest that the emotional and physical proximity which at- tachment ensures also equips infants with the ca- pacity to understand themselves and those around them. Secure attachment enables us to 'read' people--including ourselves. Recent work in developmental psychopathology has begun to look at the developmental processes which lead to successful mentalization--'mirroring' between mother and child, 'marking' clear boundaries between 'pretence' and reality--and the ways in which these can break down. These studies sug- gest that insecure attachment, especially disor- ganised attachment (Holmes, 2004), is a likely predisposing factor for the subsequent develop- ment of personality difficulties in adult life (Bate- man & Fonagy, 2004), especially Borderline Per- sonality Disorder (described in this volume by Bowlby in traditional psychoanalytic terms as
21/362
? ? ? cases of 'false self', schizoid personality or patho- logical narcissism).
Interpersonal aspects of psychotherapy. Bowlby regularly emphasized the parallels between secure parenting and good psychother- apy (see Chapter 8). Just as he rehabilitated the role of real trauma, as opposed to phantasy, as the pathogenic factor in psychological difficulties, so he called for 'greater emphasis [to be] placed on the contribution of the therapist's role as a companion for his patient in the latter's explora- tion of himself and his experiences, and less on the therapist interpreting things to the patient' (p. 151).
Bowlby himself clearly was able to provide a secure base for his patients, co-workers, and stu- dents, inspiring a huge amount of affection and admiration amongst them. His writings, and life- story, are imbued with his capacity to balance the 'maternal' qualities of sensitivity and responsive- ness with 'paternal' challenge and support (note that I am here utterly repudiating my earlier in- accurate and impertinent description of him as 'dismissing', Holmes, 1993). He may perhaps not have appreciated, however, just how hard won these qualities are for some therapists. Recent
22/362
? ? ? work has begun to tease out the separate contri- butions of therapist and patient to the therapeut- ic dyad and to suggest that the quality of the ther- apy, and probably ultimately its outcome, is a function not of each alone, but of the interaction or 'fit' that exists between them. Dozier et al. (1999) looked at the attachment classification of therapists and their clients, and found that insec- ure therapists tend to reinforce the insecure at- tachment patterns of their clients, in avoidant cli- ents resulting in yet more hypoactivation of at- tachment behaviours, and in their preoccupied counterparts greater clinging and affective dys- regulation. Secure therapists by contrast tend to redress the balance between avoidance and pre- occupation and to push their clients towards more secure patterns of relating.
With the introduction of the Patient-Therapist Adult Attachment Interview (PT-AAI) Diamond and her co-workers (Diamond et al. , 2003) have taken this line of investigation one step further. Here the Adult Attachment Interview procedure is applied to the therapeutic relationship it- self--both therapist and patient are asked to provide adjectives to describe one another, to- gether with supporting anecdotes, and are asked
23/362
? ? ? speculate about why they think their patient/ therapist behaves in a particular way. The inter- view is then transcribed and rated in a similar way to the AAI, generating attachment categories and also measures of Reflexive Function in rela- tion to the therapeutic process. A number of in- teresting findings are beginning to emerge.
First, as might be hoped and expected, Reflex- ive Function improves in the course of therapy. Second, good outcomes seem to be associated with therapists who are neither too far behind nor too far ahead of their clients in the PT-AAI scores. A degree of distance between therapist and client is needed, but it must not be too far--for effectiveness therapy should not be too cosy, nor too astringent. Third, the therapist's ca- pacity for Reflexive Function varies from patient to patient. Each therapist-patient pair appears to generate its own particular attachment atmo- sphere and capacity for mentalization, or the lack of it.
All this implies a much more complex and dy- namic relational culture between caregiver and care-receiver (whether parent and child, or ther- apist and patient) than perhaps Bowlby's sugges- tions in the last chapter in this volume suggest.
24/362
? ? ? Tracking the vicissitudes of this relationship is a huge challenge for developmentalists, psycho- therapy researchers, and clinicians wishing to practice and teach their craft. Bowlby was an ec- lecticist par excellence. He had a remarkable ca- pacity to bring together differing discip- lines--psychoanalysis, cognitive science, child de- velopment, ethology, cybernetics--and to weld them into a coherent story.
This eclecticism makes attachment theory hugely attractive to some, but also off-putting to clinicians in search of truths deriving from but one 'God' alone. For our discipline to advance further, another effort of synthesis will be needed, bringing together ideas from neurobio- logy, neuroimaging, linguistics, ecology, and the mathematics of complex systems such as chaos theory. Forging such creative links is a task for the future--one which Bowlby would surely have happily endorsed and be sad to have missed.
JEREMY HOLMES
REFERENCES
Bateman, A. & Fonagy, P. (2004) Psychotherapy for Borderline Personality Disorder. Oxford: Oxford University Press.
25/362
? ? ? Cassidy, J. & Shaver, P. (1999) Handbook of Attach- ment. London: Guilford.
Diamond, D. , Stovall-McClough, C. , Clarkin, J. , & Levy, K. (2003) Patient-therapist attachment in the treatment of Borderline Personality Disorder. Bul- letin of the Menninger Clinic 76, 227-259.
Dozier, M. , Chase Stowall, K. , & Albus, K. (1999) At- tachment and psychopathology in adulthood. In Handbook of Attachment (Eds J. Cassidy & P. Shaver). London: Guilford.
Fonagy, P. , Gergely, G. , Jurist, E. , & Target, M. (2002)
Affect regulation, mentalization, and the develop-
ment of the self. New York: Other Press. Grossman, K. , Grossman, K. , & Zimmerman, P. (1999) A Wider view of attachment and exploration: sta- bility and change during the years of immaturity. In Handbook of Attachment (Eds J. Cassidy & P.
Shaver). London: Guilford.
Grossman, K. , Grossman, K. , & Kindler, H. (2005)
Early care and the roots of attachment and part- nership representation in the Bielefeld and Re- gensburg longitudinal studies. In Attachment from infancy to adulthood: the major longitudin- al studies. New York: Guilford.
Hesse, E. (1999) The Adult Attachment Interview: His- torical and Current Developments. In Handbook of Attachment (Eds J. Cassidy & P. Shaver). Lon- don: Guilford.
26/362
? ? ? Holmes, J. (1993) John Bowlby and Attachment The- ory. London: Routlege.
Holmes, J. (2004) Disorganised attachment and Borderline Personality Disorder: a clinical per- spective. Attachment and Human Development 6, 181-190.
1
CARING FOR CHILDREN
During the early months of 1980 I was giving lectures in the United States. Amongst invita- tions reaching me was one from the psychiatric staff of the Michael Reese Hospital in Chicago to address a conference on parenting.
AN INDISPENSABLE SOCIAL ROLE
At some time of their lives, I believe, most human beings desire to have children and desire also that their children should grow up to be healthy, happy, and self-reliant. For those who succeed the rewards are great; but for those who have children but fail to rear them to be healthy, happy, and self-reliant the penalties in anxiety, frustration, friction, and perhaps shame or guilt, may be severe. Engaging in parenthood therefore is playing for high stakes. Furthermore, because successful parenting is a principal key to the mental health of the next generation, we need to know all we can both about its nature and about
? 28/362
? ? ?
the manifold social and psychological conditions that influence its development for better or worse. The theme is a huge one and all I can do in this contribution is to sketch the approach that I myself adopt in thinking about these issues. That approach is an ethological one.
Before I go into detail, however, I want to make a few more general remarks. To be a successful parent means a lot of very hard work. Looking after a baby or toddler is a twenty-four-hour-a- day job seven days a week, and often a very wor- rying one at that. And even if the load lightens a little as children get older, if they are to flourish they still require a lot of time and attention. For many people today these are unpalatable truths. Giving time and attention to children means sac- rificing other interests and other activities. Yet I believe the evidence for what I am saying is un- impeachable. Study after study, including those pioneered in Chicago by Grinker (1962) and con- tinued by Offer (1969), attest that healthy, happy, and self-reliant adolescents and young adults are the products of stable homes in which both par- ents give a great deal of time and attention to the children.
29/362
? ? ? I want also to emphasize that, despite voices to the contrary, looking after babies and young chil- dren is no job for a single person. If the job is to be well done and the child's principal caregiver is not to be too exhausted, the caregiver herself (or himself) needs a great deal of assistance. From whom that help comes will vary: very often it is the other parent; in many societies, including more often than is realized our own, it comes from a grandmother. Others to be drawn in to help are adolescent girls and young women. In most societies throughout the world these facts have been, and still are, taken for granted and the society organized accordingly. Paradoxically it has taken the world's richest societies to ignore these basic facts. Man and woman power devoted to the production of material goods counts a plus in all our economic indices. Man and woman power devoted to the production of happy, healthy, and selfreliant children in their own homes does not count at all. We have created a topsy-turvy world.
But I do not want to enter into complex politic- al and economic arguments. My reason for rais- ing these points is to remind you that the society we live in is not only, in evolutionary terms, a
30/362
? ? ? product of yesterday but in many ways a very pe- culiar one. There is in consequence a great danger that we shall adopt mistaken norms. For, just as a society in which there is a chronic insuf- ficiency of food may take a deplorably inadequate level of nutrition as its norm, so may a society in which parents of young children are left on their own with a chronic insufficiency of help take this state of affairs as its norm.
AN ETHOLOGICAL APPROACH
I said earlier that my approach to an understand- ing of parenting as a human activity is an etholo- gical one. Let me explain.
In re-examining the nature of the child's tie to his1 mother, traditionally referred to as depend- ency, it has been found useful to regard it as the resultant of a distinctive and in part prepro- grammed set of behaviour patterns which in the ordinary expectable environment develop during the early months of life and have the effect of keeping the child in more or less close proximity to his mother-figure (Bowlby, 1969). By the end of the first year the behaviour is becoming organ- ized cybernetically, which means, among other things, that the behaviour becomes active
31/362
? ? ? whenever certain conditions obtain and ceases when certain other conditions obtain. For ex- ample, a child's attachment behaviour is activ- ated especially by pain, fatigue, and anything frightening, and also by the mother being or ap- pearing to be inaccessible. The conditions that terminate the behaviour vary according to the in- tensity of its arousal. At low intensity they may be simply sight or sound of the mother, especially effective being a signal from her acknowledging his presence. At higher intensity termination may require his touching or clinging to her. At highest intensity, when he is distressed and anxious, nothing but a prolonged cuddle will do. The bio- logical function of this behaviour is postulated to be protection, especially protection from predators.
In the example just given the individuals con- cerned are a child and his mother. It is evident, however, that attachment behaviour is in no way confined to children. Although usually less read- ily aroused, we see it also in adolescents and adults of both sexes whenever they are anxious or under stress. No one should be surprised there- fore when a woman expecting a baby or a mother caring for young children has a strong desire to
32/362
? ? ? be cared for and supported herself. The activation of attachment behaviour in these circumstances is probably universal and must be considered the norm. 2
A feature of attachment behaviour of the greatest importance clinically, and present irre- spective of the age of the individual concerned, is the intensity of the emotion that accompanies it, the kind of emotion aroused depending on how the relationship between the individual attached and the attachment figure is faring. If it goes well, there is joy and a sense of security. If it is threatened, there is jealousy, anxiety, and anger. If broken, there is grief and depression. Finally there is strong evidence that how attachment be- haviour comes to be organized within an indi- vidual turns in high degree on the kinds of exper- ience he has in his family of origin, or, if he is un- lucky, out of it.
This type of theory I believe to have many ad- vantages over the theories hitherto current in our field. For not only does it bring theory into close relationship with observed data but it provides a theoretical framework for the field compatible with the framework adopted throughout modern biology and neurophysiology.
33/362
? ? ? Parenting, I believe, can usefully be ap- proached from the same ethologically inspired viewpoint. This entails observing and describing the set of behaviour patterns characteristic of parenting, the conditions that activate and ter- minate each, how the patterns change as a child grows older, the varying ways that parenting be- haviour becomes organized in different individu- als, and the myriad of experiences that influence how it develops in any one person.
Implicit in this approach is the assumption that parenting behaviour, like attachment beha- viour, is in some degree preprogrammed and therefore ready to develop along certain lines when conditions elicit it. This means that, in the ordinary course of events, the parent of a baby experiences a strong urge to behave in certain typical sorts of way, for example, to cradle the in- fant, to soothe him when he cries, to keep him warm, protected, and fed. Such a viewpoint, of course, does not imply that the appropriate beha- viour patterns manifest themselves complete in every detail from the first. Clearly that is not so, neither in man nor in any other mammalian spe- cies. All the detail is learned, some of it during in- teraction with babies and children, much of it
34/362
? ? ? through observation of how other parents be- have, starting during the parent-to-be's own childhood and the way his parents treated him and his siblings.
This modern view of behavioural development contrasts sharply with both of the older paradigms, one of which, invoking instinct, over- emphasized the preprogrammed component and the other of which, reacting against instinct, over- emphasized the learned component. Parenting behaviour in humans is certainly not the product of some unvarying parenting instinct, but nor is it reasonable to regard it as the product simply of learning. Parenting behaviour, as I see it, has strong biological roots, which accounts for the very strong emotions associated with it; but the detailed form that the behaviour takes in each of us turns on our experiences--experiences during childhood especially, experiences during adoles- cence, experiences before and during marriage, and experiences with each individual child.
Thus I regard it as useful to look upon parent- ing behaviour as one example of a limited class of biologically rooted types of behaviour of which attachment behaviour is another example, sexual behaviour another, and exploratory behaviour
35/362
? ? ? and eating behaviour yet others. Each of these types of behaviour contributes in its own specific way to the survival either of the individual or his offspring. It is indeed because each one serves so vital a function that each of these types of beha- viour is in some degree preprogrammed. To leave their development solely to the caprices of indi- vidual learning would be the height of biological folly.
You will notice that in sketching this frame- work I am making a point of keeping each of these types of behaviour conceptually distinct from the others. This contrasts, of course, with traditional libido theory which has treated them as the varying expressions of a single drive. The reasons for keeping them distinct are several. One is that each of the types of behaviour men- tioned serves its own distinctive biological func- tion--protection, reproduction, nutrition, know- ledge of the environment. Another is that many of the detailed patterns of behaviour within each general type are distinctive also: clinging to a par- ent is different from soothing and comforting a child; sucking or chewing food is different from engaging in sexual intercourse. Furthermore, factors which influence the development of one of
36/362
? ? ? these types of behaviour are not necessarily the same as those that influence the development of another. By keeping them distinct we are able to study not only the ways in which they differ but also the ways in which they overlap and interact with each other--as it has long been evident they do.
INITIATION OF MOTHER-INFANT INTERACTION
During the past decade or so there has been a dramatic advance in our understanding of the early phases of mother-infant interaction, thanks to the imaginative research of workers on both sides of the Atlantic. The studies of Klaus and Kennell are now well known. Of special interest here are their observations of how mothers be- have towards their newborns when given free- dom to do what they like after delivery. Klaus, Trause, and Kennell (1975) describe how a moth- er, immediately after her infant is born, picks him up and begins to stroke his face with her finger tips. At this the baby quietens. Soon she moves on to touching his head and body with the palm of her hand and, within five or six minutes, she is likely to put him to her breast. The baby responds with prolonged licking of the nipple.
37/362
? ? ? 'Immediately after the delivery', they noted, 'the mothers appeared to be in a state of ecstasy', and, interestingly enough, the observers became elated too. From the moment of birth attention becomes riveted on the baby. Something about him tends to draw not only the mother and father but all those present to the new arrival. Given the chance, a mother is likely during the next few days to spend many hours just looking at her new possession, cuddling him, and getting to know him. Usually there comes a moment when she feels the baby is her very own. For some it comes early; perhaps when she first holds him or when he first looks into her eyes. For a large minority of primaparae who are delivered in hospital, however, it may be delayed for up to a week, of- ten until they are home again (Robson and Ku- mar, 1980).
Phenomena of the greatest importance to which recent research has drawn attention are the potential of the healthy neonate to enter into an elemental form of social interaction and the potential of the ordinary sensitive mother to par- ticipate successfully in it. 3
When a mother and her infant of two or three weeks are facing one another, phases of lively
38/362
? ? ? social interaction occur, alternating with phases of disengagement. Each phase of interaction be- gins with initiation and mutual greeting, builds up to an animated interchange comprising facial expressions and vocalizations, during which the infant orients towards his mother with excited movements of arms and legs; then his activities gradually subside and end with the baby looking away for a spell before the next phase of interac- tion begins. Throughout these cycles the baby is likely to be as spontaneously active as his mother. Where their roles differ is in the timing of their responses. Whereas an infant's initiation and withdrawal from interaction tend to follow his own autonomous rhythm, a sensitive mother reg- ulates her behaviour so that it meshes with his. In addition she modifies the form her behaviour takes to suit him: her voice is gentle but higher pitched than usual, her movements slowed, and each next action adjusted in form and timing ac- cording to how her baby is performing. Thus she lets him call the tune and, by a skilful interweav- ing of her own responses with his, creates a dialogue.
The speed and efficiency with which these dia- logues develop and the mutual enjoyment they
39/362
? ? ? give point clearly to each participant being pread- apted to engage in them. On the one hand is the mother's intuitive readiness to allow her inter- ventions to be paced by her infant. On the other is the readiness with which the infant's rhythms shift gradually to take account of the timing of his mother's interventions. In a happily developing partnership each is adapting to the other.
Very similar alternating sequences have been recorded in other quite different exchanges between mother and child. For example, Kaye (1977), observing the behaviour of mother and in- fant during feeding, has found that mothers tend to interact with their infants in precise synchrony with the infant's pattern of sucking and pausing. During bursts of sucking a mother is generally quiet and inactive; during pauses she strokes and talks to her baby. Another example of a mother taking her cue from her infant, in this case an in- fant within the age range 5 to 12 months, is re- ported by Collis and Schaffer (1975). A mother and her infant are introduced to a scene in which there are a number of large brightly coloured toys which quickly seize their visual attention. Obser- vation of their behaviour then shows two things. First, both partners as a rule are looking at the
40/362
? ? ? same object at the same time. Secondly, examina- tion of the timing shows almost invariably that it is the baby who leads and the mother who fol- lows. The baby's spontaneous interest in the toys is evidently closely monitored by his mother who almost automatically then looks in the same dir- ection. A focus of mutual interest having been es- tablished, mother is likely to elaborate on it, com- menting on the toy, naming it, manipulating it. 'A sharing experience is then brought about, instig- ated by the infant's spontaneous attention to the environment but established by the mother al- lowing herself to be paced by the baby'.
Yet another example, also reported by Schaffer (Schaffer, Collis, and Parsons, 1977), concerns vocal interchange between mother and child at a preverbal level. In a comparison of two groups of children, aged 12 and 24 months, it was found that the ability of the pair to take turns and to avoid overlapping was not only strikingly efficient but as characteristic of the younger as of the older infants. Thus, long before the appearance of words, the pattern of turn-taking so characteristic of human conversation is already present. Here again the evidence suggests that, in ensuring the
41/362
? ? ? smooth transitions from one 'speaker' to the oth- er, mother is playing the major part.
My reason for giving these examples at some length is that I believe they illustrate some basic principles both about parenting and about the nature of the creature who is parented. What emerges from these studies is that the ordinary sensitive mother is quickly attuned to her infant's natural rhythms and, by attending to the details of his behaviour, discovers what suits him and behaves accordingly. By so doing she not only makes him contented but also enlists his co-oper- ation. For, although initially his capacity to adapt is limited, it is not absent altogether and, if al- lowed to grow in its own time, is soon yielding re- wards. Ainsworth and her colleagues have noted that infants whose mothers have responded sens- itively to their signals during the first year of life not only cry less during the second half of that year than do the babies of less responsive moth- ers but are more willing to fall in with their par- ent's wishes (Ainsworth et al. , 1978). Human in- fants, we can safely conclude, like infants of other species, are preprogrammed to develop in a so- cially cooperative way; whether they do so or not turns in high degree on how they are treated.
42/362
? ? ? This is a view of human nature, you will notice, radically different from the one that has long been current in western societies and that has permeated so much of the clinical theory and practice we have inherited. It points, of course, to a radically different conception of the role of parent.
ROLES OF MOTHERS AND FATHERS: SIMILARITIES AND DIFFERENCES
In the examples given so far, the parent con- cerned has been the mother. This is almost inev- itable because for research purposes it is relat- ively easy to recruit samples of infants who are being cared for mainly by their mother, whereas infants being cared for mainly by their father are comparatively scarce. Let me therefore describe briefly one of several recent studies which, to- gether, go some way to correct the balance.
Several hundred infants have now been studied by means of the strange situation procedure de- vised by Ainsworth (Ainsworth et al. , 1978) which gives an opportunity to observe how the infant responds, first in his parent's presence, next when he is left alone, and later when his parent returns. As a result of these observations infants
43/362
? ? ? can be classified as presenting a pattern either of secure attachment to mother or of one of two main forms of insecure attachment to her. Since these patterns have been shown to have consider- able stability during the earliest years of life and to predict how a nursery-school child in the age range 41/2 to 6 years will approach a new person and tackle a new task (Arend, Gove, and Sroufe, 1979), the value of the procedure as a method of assessing an infant's social and emotional devel- opment needs no emphasis.
Hitherto almost all the studies using this pro- cedure have observed infants with their mothers. Main and Weston (1981), however, extended the work by observing some 60 infants, first with one parent and, six months later, with the other. One finding was that, when looked at as a group, the patterns of attachment that were shown to fath- ers resembled closely the patterns that were shown to mothers, with roughly the same per- centage distribution of patterns. But a second finding was even more interesting. When the pat- terns shown by each child individually were ex- amined, no correlation was found between the pattern shown with one parent and the pattern shown with the other. Thus one child may have a
44/362
? ? ? secure relationship with the mother but not with the father, a second may have it with the father but not with the mother, a third may have it with both parents, and a fourth may have it with neither. In their approach to new people and new tasks the children represented a graded series. Children with a secure relationship to both par- ents were most confident and most competent; children who had a secure relationship to neither were least so; and those with a secure relation- ship to one parent but not to the other came in between.
Since there is evidence that the pattern of at- tachment a child undamaged at birth develops with his mother is the product of how his mother has treated him (Ainsworth et al. , 1978), it is more than likely that, in a similar way, the pat- tern he develops with his father is the product of how his father has treated him.
This study, together with others, suggests that, by providing an attachment figure for his child, a father may be filling a role closely resembling that filled by a mother; though in most, perhaps all, cultures fathers fill that role much less fre- quently than do mothers, at least when the chil- dren are still young. In most families with young
45/362
? ? ? children the father's role is a different one. He is more likely to engage in physically active and novel play than the mother and, especially for boys, to become his child's preferred play com- panion. 4
PROVISION OF A SECURE BASE
This brings me to a central feature of my concept of parenting--the provision by both parents of a secure base from which a child or an adolescent can make sorties into the outside world and to which he can return knowing for sure that he will be welcomed when he gets there, nourished phys- ically and emotionally, comforted if distressed, reassured if frightened. In essence this role is one of being available, ready to respond when called upon to encourage and perhaps assist, but to in- tervene actively only when clearly necessary. In these respects it is a role similar to that of the of- ficer commanding a military base from which an expeditionary force sets out and to which it can retreat, should it meet with a setback. Much of the time the role of the base is a waiting one but it is none the less vital for that. For it is only when the officer commanding the expeditionary force is
46/362
? ? ? confident his base is secure that he dare press forward and take risks.
In the case of children and adolescents we see them, as they get older, venturing steadily further from base and for increasing spans of time. The more confident they are that their base is secure and, moreover, ready if called upon to respond, the more they take it for granted. Yet should one or other parent become ill or die, the immense significance of the base to the emotional equilib- rium of the child or adolescent or young adult is at once apparent. In the lectures to follow evid- ence is presented from studies of adolescents and young adults, as well as of school children of dif- ferent ages from nursery school up, that those who are most stable emotionally and making the most of their opportunities are those who have parents who, whilst always encouraging their children's autonomy, are none the less available and responsive when called upon. Unfortunately, of course, the reverse is also true.
ACKNOWLEDGEMENTS
During the past ten years I have had the great be- nefit of frequent communication with staff and students at the Tavistock Clinic and also with a number of colleagues engaged in pioneering studies of how patterns of attachment develop during infancy and childhood. To all of them I owe a deep debt of gratitude, often for useful sug- gestions, sometimes for necessary corrections, and always for stimulation and encouragement. To my secretary, Dorothy Southern, I also owe a deep debt of gratitude for many years of devoted service during which she has made my interests her own.
For editorial assistance in preparing these lec- tures for publication and for constructing the in- dex my thanks are due to Molly Townsend.
The first six lectures in this book have ap- peared in other publications and I am grateful to the publishers concerned for permission to repro- duce them here. Lecture 1 was chapter 18 in Par- enthood: A Psychodynamic Perspective edited by Rebecca S. Cohen, Bertram J. Cohler, and Sidney H. Weissman, the Guilford Press, New York
12/362
? ? ? (1984); Lecture 2 was 'Attachment and loss: ret- rospect and prospect', American Journal of Orthopsychiatry 52: 664-78 (1982); Lecture 3 was 'Psychoanalysis as art and science', Interna- tional Review of Psychoanalysis 6: 3-14 (1979); Lecture 4 was 'Psychoanalysis as a natural sci- ence', International Review of Psychoanalysis 8: 243-56 (1981); Lecture 5 was 'Violence in the family as a disorder of the attachment and care- giving systems', The American Journal of Psy- choanalysis 44: 9-27 (1984); Lecture 6 was chapter 6 in Cognition and Psychotherapy edited by Michael J. Mahoney and Arthur Freeman, Plenum Publishing Corporation, New York and London (1985), expanded from 'On knowing what you are not supposed to know and feeling what you are not supposed to feel', Canadian Journal of Psychiatry 24: 403-8 (1979).
JOHN BOWLBY
PREFACE TO THE ROUTLEDGE CLASSICS EDITION
A Secure Base first appeared in 1988 when John Bowlby was in his 81st year. Although there was, amazingly, one more book to come--the Darwin biography--it is his final contribution to Attach- ment Theory, the discipline which he, with Mary Ainsworth's help, founded nearly half a century earlier. Thus A Secure Base has a valedictory feel to it--a summation of a life's work, but also a tribute and a handing over to the next generation of attachment researchers and clinicians.
In it are to be found all the familiar Bowlbian themes--theoretical, etiological, methodological, clinical, and political. He restates the conceptual foundation stones of his ideas: the primacy of the attachment behavioural response and its role in protection from predation; sensitive care-giving as a foundation for psychological health; the con- tinuing importance of attachment throughout the life cycle. He argues powerfully for the role of real-life adversity--emotional deprivation, un- mourned bereavement, rejection, obfuscation, neglect, physical and sexual abuse--as the origin of subsequent psychopathology, as opposed to
14/362
? ? ? putative endo-psychic entities such as 'death instinct'.
Methodologically he emphasizes the import- ance of systematic scientific observation of chil- dren and parents, as opposed to speculative re- constructions from the couch. Clinically he sees the therapist as providing a secure base for her patients, a springboard from which they can be- gin to develop the free flowing discourse of emo- tion that is characteristic of those who are se- curely attached.
Finally, there is the figured base of Bowlby's social philosophy lying at the heart of his work: 'man and woman power devoted to the produc- tion of happy, healthy, and self-reliant children . . . does not count at all. We have created a topsy- turvy world' (p. 2).
The past 25 years has seen an explosion of in- terest in attachment theory, culminating in the landmark volume Handbook of Attachment (Cassidy & Shaver, 1999), an exhilarating sum- mation of how Bowlby and Ainsworth's acorn, nurtured by them into a sturdy sapling, has seeded a whole forest of developments, applica- tions, and ideas. In this brief introduction I shall pick out three of these growth points, or 'recent
15/362
? ? ? advances' in attachment theory, hinted at by Bowlby, but greatly expanded since A Secure Base was first published: the role of fathers in creating secure attachments; mentalization and 'theory of mind' as a developmental achievement; and psychotherapy as an interpersonal enterprise.
Attachment to Fathers. As implied in the quo- tation above, Bowlby was always insistent that mothers and fathers mattered when it came to providing a secure base. One of the crucial planks in the argument that security of attachment is an interpersonal, interactive phenomenon, and not simply a matter of the child's inborn tempera- ment, is that one and the same child can be clas- sified in the Strange Situation as secure with one parent and insecure with another. Nevertheless, attachment theory, both in its research and clin- ical guises, has tended to be a somewhat mater- nocentric enterprise and it has not been easy to pin down precisely what the father's contribution to security of attachment comprises.
Recent work (Grossman, Grossman, & Zim- merman, 1999; Grossman, Grossman, & Zindler, 2005) has begun to throw light on this topic. When Bowlby was writing the essays and lectures
16/362
? ? ? that make up this book in the 1980s, longitudinal studies of attachment phenomena were them- selves in their infancy. We now have 20-year pro- spective studies that look at measures of attach- ment security, parental sensitivity, exploration, relational competence, and their mental repres- entations throughout childhood. These can now be correlated with attachment disposition in young adulthood as manifested in attitudes to- wards romantic relationships and the Adult At- tachment Interview.
These studies show that paternal contributions are indeed vital to secure, stable, exploratory, balanced, verbally fluent attachment dispositions in adulthood. However, this contribution to psy- chological health is not predominantly mediated via security of attachment as measured in the Strange Situation. The role of fathers, rather, makes itself felt via the exploratory dimension of the attachment/exploration dichotomy and is eli- cited in the 'SCIP' (Sensitive and Challenging In- teractive Play) measure (Grossman, Grossman, & Zindler, 2005), which observes and rates parents interacting with their children in a 10-minute play session.
17/362
? ? ? In general, combined mother-father scores along multiple dimensions of attachment in childhood are far more predictive of security or insecurity of attachment representation in adult- hood than those of either parent alone. However, the preoccupied dimension--adults who give confused, affect-laden, unstructured responses to probes--is strongly correlated with father rejec- tion and insensitivity in mid-childhood, with the maternal contribution being relatively weak. Thus it seems that good-enough fathers help their children to develop clarity of thought and to be able to face up to negative emotions without feeling overwhelmed.
Like mothers, fathers need to be sensitive, but this takes the form of praise, encouragement, and the capacity to sustain positive affect in their off- spring. As they help their children to cope with curiosity-wariness conflicts, this protective, chal- lenging, 'you can do it' father diverges markedly from the castrating picture of classical psycho- analytic theory, which might apply more accur- ately to insensitive fathers who fail intuitively to grasp the basic message of attachment: that achievement is always predicated on security.
18/362
? ? ? Mentalization and theory of mind. Bowlby lived long enough to appreciate the enormous significance of Mary Main's contribution to At- tachment Theory and in particular the possibilit- ies represented by her development of the Adult Attachment Interview (AAI) (Hesse, 1999). He quotes her finding, confirmed by others, that mothers who showed insecurity on the AAI were more likely to have insecure children as meas- ured in the Strange Situation. He saw insecure at- tachment in terms of psychological defences--ne- cessary for emotional survival (and in the envir- onment of evolutionary adaptiveness in which our species evolved, physical survival)--but also as a constricting factor that excludes the insecure individual from the possibility of processing ad- verse experience.
A crucial finding in this early work was that 're- flexive function' (Fonagy et al. , 2002) as meas- ured in the AAI, appears to be a protective factor that despite adverse childhood experience such as parental separation, bereavement or even neg- lect and abuse, enables individuals to remain se- cure themselves and to provide security for their offspring. Put simply, the ability to 'talk about it' mitigates the long-term negative consequences of
19/362
? ? ? childhood trauma. 'Reflexive function' can be seen as tapping into internal speech, or mental representation of experience that underlies this capacity for external interactive story-telling.
Fonagy and his co-workers (Fonagy, Gergely, Jurist, & Target, 2002; Bateman & Fonagy, 2004) have extended these early findings with their new concept of 'mentalization'. Drawing on the philo- sophical tradition of 'theory of mind' they suggest that there are crucial developmental processes which enable infants to begin to appreciate the fact that they themselves, and those around them, have 'minds'--i. e. the ability to represent the world and to have projects, beliefs, and de- sires. Mentalization enables us to differentiate between 'reality' and our perspective or appreci- ation of reality, and also to grasp the fact that dif- ferent people view the world in differing ways. Thus mentalization, which could be seen as an extension of Bowlby's notion of internal working models, is anti-narcissistic and, so Fonagy et al. argue, a vital component of the ability to interact socially, including the ability to survive the vicis- situdes of mis-attunement, alliance ruptures, and the minor failures that are a normal part of
20/362
? ? ? 'good-enough parenting', or even 'not-good- enough' parenting.
This set of ideas leads to a rather different take on the evolutionary significance of attachment, which Bowlby, drawing on an ethological per- spective, always insisted was that of protection from predation. Fonagy et al. (2002) suggest that the emotional and physical proximity which at- tachment ensures also equips infants with the ca- pacity to understand themselves and those around them. Secure attachment enables us to 'read' people--including ourselves. Recent work in developmental psychopathology has begun to look at the developmental processes which lead to successful mentalization--'mirroring' between mother and child, 'marking' clear boundaries between 'pretence' and reality--and the ways in which these can break down. These studies sug- gest that insecure attachment, especially disor- ganised attachment (Holmes, 2004), is a likely predisposing factor for the subsequent develop- ment of personality difficulties in adult life (Bate- man & Fonagy, 2004), especially Borderline Per- sonality Disorder (described in this volume by Bowlby in traditional psychoanalytic terms as
21/362
? ? ? cases of 'false self', schizoid personality or patho- logical narcissism).
Interpersonal aspects of psychotherapy. Bowlby regularly emphasized the parallels between secure parenting and good psychother- apy (see Chapter 8). Just as he rehabilitated the role of real trauma, as opposed to phantasy, as the pathogenic factor in psychological difficulties, so he called for 'greater emphasis [to be] placed on the contribution of the therapist's role as a companion for his patient in the latter's explora- tion of himself and his experiences, and less on the therapist interpreting things to the patient' (p. 151).
Bowlby himself clearly was able to provide a secure base for his patients, co-workers, and stu- dents, inspiring a huge amount of affection and admiration amongst them. His writings, and life- story, are imbued with his capacity to balance the 'maternal' qualities of sensitivity and responsive- ness with 'paternal' challenge and support (note that I am here utterly repudiating my earlier in- accurate and impertinent description of him as 'dismissing', Holmes, 1993). He may perhaps not have appreciated, however, just how hard won these qualities are for some therapists. Recent
22/362
? ? ? work has begun to tease out the separate contri- butions of therapist and patient to the therapeut- ic dyad and to suggest that the quality of the ther- apy, and probably ultimately its outcome, is a function not of each alone, but of the interaction or 'fit' that exists between them. Dozier et al. (1999) looked at the attachment classification of therapists and their clients, and found that insec- ure therapists tend to reinforce the insecure at- tachment patterns of their clients, in avoidant cli- ents resulting in yet more hypoactivation of at- tachment behaviours, and in their preoccupied counterparts greater clinging and affective dys- regulation. Secure therapists by contrast tend to redress the balance between avoidance and pre- occupation and to push their clients towards more secure patterns of relating.
With the introduction of the Patient-Therapist Adult Attachment Interview (PT-AAI) Diamond and her co-workers (Diamond et al. , 2003) have taken this line of investigation one step further. Here the Adult Attachment Interview procedure is applied to the therapeutic relationship it- self--both therapist and patient are asked to provide adjectives to describe one another, to- gether with supporting anecdotes, and are asked
23/362
? ? ? speculate about why they think their patient/ therapist behaves in a particular way. The inter- view is then transcribed and rated in a similar way to the AAI, generating attachment categories and also measures of Reflexive Function in rela- tion to the therapeutic process. A number of in- teresting findings are beginning to emerge.
First, as might be hoped and expected, Reflex- ive Function improves in the course of therapy. Second, good outcomes seem to be associated with therapists who are neither too far behind nor too far ahead of their clients in the PT-AAI scores. A degree of distance between therapist and client is needed, but it must not be too far--for effectiveness therapy should not be too cosy, nor too astringent. Third, the therapist's ca- pacity for Reflexive Function varies from patient to patient. Each therapist-patient pair appears to generate its own particular attachment atmo- sphere and capacity for mentalization, or the lack of it.
All this implies a much more complex and dy- namic relational culture between caregiver and care-receiver (whether parent and child, or ther- apist and patient) than perhaps Bowlby's sugges- tions in the last chapter in this volume suggest.
24/362
? ? ? Tracking the vicissitudes of this relationship is a huge challenge for developmentalists, psycho- therapy researchers, and clinicians wishing to practice and teach their craft. Bowlby was an ec- lecticist par excellence. He had a remarkable ca- pacity to bring together differing discip- lines--psychoanalysis, cognitive science, child de- velopment, ethology, cybernetics--and to weld them into a coherent story.
This eclecticism makes attachment theory hugely attractive to some, but also off-putting to clinicians in search of truths deriving from but one 'God' alone. For our discipline to advance further, another effort of synthesis will be needed, bringing together ideas from neurobio- logy, neuroimaging, linguistics, ecology, and the mathematics of complex systems such as chaos theory. Forging such creative links is a task for the future--one which Bowlby would surely have happily endorsed and be sad to have missed.
JEREMY HOLMES
REFERENCES
Bateman, A. & Fonagy, P. (2004) Psychotherapy for Borderline Personality Disorder. Oxford: Oxford University Press.
25/362
? ? ? Cassidy, J. & Shaver, P. (1999) Handbook of Attach- ment. London: Guilford.
Diamond, D. , Stovall-McClough, C. , Clarkin, J. , & Levy, K. (2003) Patient-therapist attachment in the treatment of Borderline Personality Disorder. Bul- letin of the Menninger Clinic 76, 227-259.
Dozier, M. , Chase Stowall, K. , & Albus, K. (1999) At- tachment and psychopathology in adulthood. In Handbook of Attachment (Eds J. Cassidy & P. Shaver). London: Guilford.
Fonagy, P. , Gergely, G. , Jurist, E. , & Target, M. (2002)
Affect regulation, mentalization, and the develop-
ment of the self. New York: Other Press. Grossman, K. , Grossman, K. , & Zimmerman, P. (1999) A Wider view of attachment and exploration: sta- bility and change during the years of immaturity. In Handbook of Attachment (Eds J. Cassidy & P.
Shaver). London: Guilford.
Grossman, K. , Grossman, K. , & Kindler, H. (2005)
Early care and the roots of attachment and part- nership representation in the Bielefeld and Re- gensburg longitudinal studies. In Attachment from infancy to adulthood: the major longitudin- al studies. New York: Guilford.
Hesse, E. (1999) The Adult Attachment Interview: His- torical and Current Developments. In Handbook of Attachment (Eds J. Cassidy & P. Shaver). Lon- don: Guilford.
26/362
? ? ? Holmes, J. (1993) John Bowlby and Attachment The- ory. London: Routlege.
Holmes, J. (2004) Disorganised attachment and Borderline Personality Disorder: a clinical per- spective. Attachment and Human Development 6, 181-190.
1
CARING FOR CHILDREN
During the early months of 1980 I was giving lectures in the United States. Amongst invita- tions reaching me was one from the psychiatric staff of the Michael Reese Hospital in Chicago to address a conference on parenting.
AN INDISPENSABLE SOCIAL ROLE
At some time of their lives, I believe, most human beings desire to have children and desire also that their children should grow up to be healthy, happy, and self-reliant. For those who succeed the rewards are great; but for those who have children but fail to rear them to be healthy, happy, and self-reliant the penalties in anxiety, frustration, friction, and perhaps shame or guilt, may be severe. Engaging in parenthood therefore is playing for high stakes. Furthermore, because successful parenting is a principal key to the mental health of the next generation, we need to know all we can both about its nature and about
? 28/362
? ? ?
the manifold social and psychological conditions that influence its development for better or worse. The theme is a huge one and all I can do in this contribution is to sketch the approach that I myself adopt in thinking about these issues. That approach is an ethological one.
Before I go into detail, however, I want to make a few more general remarks. To be a successful parent means a lot of very hard work. Looking after a baby or toddler is a twenty-four-hour-a- day job seven days a week, and often a very wor- rying one at that. And even if the load lightens a little as children get older, if they are to flourish they still require a lot of time and attention. For many people today these are unpalatable truths. Giving time and attention to children means sac- rificing other interests and other activities. Yet I believe the evidence for what I am saying is un- impeachable. Study after study, including those pioneered in Chicago by Grinker (1962) and con- tinued by Offer (1969), attest that healthy, happy, and self-reliant adolescents and young adults are the products of stable homes in which both par- ents give a great deal of time and attention to the children.
29/362
? ? ? I want also to emphasize that, despite voices to the contrary, looking after babies and young chil- dren is no job for a single person. If the job is to be well done and the child's principal caregiver is not to be too exhausted, the caregiver herself (or himself) needs a great deal of assistance. From whom that help comes will vary: very often it is the other parent; in many societies, including more often than is realized our own, it comes from a grandmother. Others to be drawn in to help are adolescent girls and young women. In most societies throughout the world these facts have been, and still are, taken for granted and the society organized accordingly. Paradoxically it has taken the world's richest societies to ignore these basic facts. Man and woman power devoted to the production of material goods counts a plus in all our economic indices. Man and woman power devoted to the production of happy, healthy, and selfreliant children in their own homes does not count at all. We have created a topsy-turvy world.
But I do not want to enter into complex politic- al and economic arguments. My reason for rais- ing these points is to remind you that the society we live in is not only, in evolutionary terms, a
30/362
? ? ? product of yesterday but in many ways a very pe- culiar one. There is in consequence a great danger that we shall adopt mistaken norms. For, just as a society in which there is a chronic insuf- ficiency of food may take a deplorably inadequate level of nutrition as its norm, so may a society in which parents of young children are left on their own with a chronic insufficiency of help take this state of affairs as its norm.
AN ETHOLOGICAL APPROACH
I said earlier that my approach to an understand- ing of parenting as a human activity is an etholo- gical one. Let me explain.
In re-examining the nature of the child's tie to his1 mother, traditionally referred to as depend- ency, it has been found useful to regard it as the resultant of a distinctive and in part prepro- grammed set of behaviour patterns which in the ordinary expectable environment develop during the early months of life and have the effect of keeping the child in more or less close proximity to his mother-figure (Bowlby, 1969). By the end of the first year the behaviour is becoming organ- ized cybernetically, which means, among other things, that the behaviour becomes active
31/362
? ? ? whenever certain conditions obtain and ceases when certain other conditions obtain. For ex- ample, a child's attachment behaviour is activ- ated especially by pain, fatigue, and anything frightening, and also by the mother being or ap- pearing to be inaccessible. The conditions that terminate the behaviour vary according to the in- tensity of its arousal. At low intensity they may be simply sight or sound of the mother, especially effective being a signal from her acknowledging his presence. At higher intensity termination may require his touching or clinging to her. At highest intensity, when he is distressed and anxious, nothing but a prolonged cuddle will do. The bio- logical function of this behaviour is postulated to be protection, especially protection from predators.
In the example just given the individuals con- cerned are a child and his mother. It is evident, however, that attachment behaviour is in no way confined to children. Although usually less read- ily aroused, we see it also in adolescents and adults of both sexes whenever they are anxious or under stress. No one should be surprised there- fore when a woman expecting a baby or a mother caring for young children has a strong desire to
32/362
? ? ? be cared for and supported herself. The activation of attachment behaviour in these circumstances is probably universal and must be considered the norm. 2
A feature of attachment behaviour of the greatest importance clinically, and present irre- spective of the age of the individual concerned, is the intensity of the emotion that accompanies it, the kind of emotion aroused depending on how the relationship between the individual attached and the attachment figure is faring. If it goes well, there is joy and a sense of security. If it is threatened, there is jealousy, anxiety, and anger. If broken, there is grief and depression. Finally there is strong evidence that how attachment be- haviour comes to be organized within an indi- vidual turns in high degree on the kinds of exper- ience he has in his family of origin, or, if he is un- lucky, out of it.
This type of theory I believe to have many ad- vantages over the theories hitherto current in our field. For not only does it bring theory into close relationship with observed data but it provides a theoretical framework for the field compatible with the framework adopted throughout modern biology and neurophysiology.
33/362
? ? ? Parenting, I believe, can usefully be ap- proached from the same ethologically inspired viewpoint. This entails observing and describing the set of behaviour patterns characteristic of parenting, the conditions that activate and ter- minate each, how the patterns change as a child grows older, the varying ways that parenting be- haviour becomes organized in different individu- als, and the myriad of experiences that influence how it develops in any one person.
Implicit in this approach is the assumption that parenting behaviour, like attachment beha- viour, is in some degree preprogrammed and therefore ready to develop along certain lines when conditions elicit it. This means that, in the ordinary course of events, the parent of a baby experiences a strong urge to behave in certain typical sorts of way, for example, to cradle the in- fant, to soothe him when he cries, to keep him warm, protected, and fed. Such a viewpoint, of course, does not imply that the appropriate beha- viour patterns manifest themselves complete in every detail from the first. Clearly that is not so, neither in man nor in any other mammalian spe- cies. All the detail is learned, some of it during in- teraction with babies and children, much of it
34/362
? ? ? through observation of how other parents be- have, starting during the parent-to-be's own childhood and the way his parents treated him and his siblings.
This modern view of behavioural development contrasts sharply with both of the older paradigms, one of which, invoking instinct, over- emphasized the preprogrammed component and the other of which, reacting against instinct, over- emphasized the learned component. Parenting behaviour in humans is certainly not the product of some unvarying parenting instinct, but nor is it reasonable to regard it as the product simply of learning. Parenting behaviour, as I see it, has strong biological roots, which accounts for the very strong emotions associated with it; but the detailed form that the behaviour takes in each of us turns on our experiences--experiences during childhood especially, experiences during adoles- cence, experiences before and during marriage, and experiences with each individual child.
Thus I regard it as useful to look upon parent- ing behaviour as one example of a limited class of biologically rooted types of behaviour of which attachment behaviour is another example, sexual behaviour another, and exploratory behaviour
35/362
? ? ? and eating behaviour yet others. Each of these types of behaviour contributes in its own specific way to the survival either of the individual or his offspring. It is indeed because each one serves so vital a function that each of these types of beha- viour is in some degree preprogrammed. To leave their development solely to the caprices of indi- vidual learning would be the height of biological folly.
You will notice that in sketching this frame- work I am making a point of keeping each of these types of behaviour conceptually distinct from the others. This contrasts, of course, with traditional libido theory which has treated them as the varying expressions of a single drive. The reasons for keeping them distinct are several. One is that each of the types of behaviour men- tioned serves its own distinctive biological func- tion--protection, reproduction, nutrition, know- ledge of the environment. Another is that many of the detailed patterns of behaviour within each general type are distinctive also: clinging to a par- ent is different from soothing and comforting a child; sucking or chewing food is different from engaging in sexual intercourse. Furthermore, factors which influence the development of one of
36/362
? ? ? these types of behaviour are not necessarily the same as those that influence the development of another. By keeping them distinct we are able to study not only the ways in which they differ but also the ways in which they overlap and interact with each other--as it has long been evident they do.
INITIATION OF MOTHER-INFANT INTERACTION
During the past decade or so there has been a dramatic advance in our understanding of the early phases of mother-infant interaction, thanks to the imaginative research of workers on both sides of the Atlantic. The studies of Klaus and Kennell are now well known. Of special interest here are their observations of how mothers be- have towards their newborns when given free- dom to do what they like after delivery. Klaus, Trause, and Kennell (1975) describe how a moth- er, immediately after her infant is born, picks him up and begins to stroke his face with her finger tips. At this the baby quietens. Soon she moves on to touching his head and body with the palm of her hand and, within five or six minutes, she is likely to put him to her breast. The baby responds with prolonged licking of the nipple.
37/362
? ? ? 'Immediately after the delivery', they noted, 'the mothers appeared to be in a state of ecstasy', and, interestingly enough, the observers became elated too. From the moment of birth attention becomes riveted on the baby. Something about him tends to draw not only the mother and father but all those present to the new arrival. Given the chance, a mother is likely during the next few days to spend many hours just looking at her new possession, cuddling him, and getting to know him. Usually there comes a moment when she feels the baby is her very own. For some it comes early; perhaps when she first holds him or when he first looks into her eyes. For a large minority of primaparae who are delivered in hospital, however, it may be delayed for up to a week, of- ten until they are home again (Robson and Ku- mar, 1980).
Phenomena of the greatest importance to which recent research has drawn attention are the potential of the healthy neonate to enter into an elemental form of social interaction and the potential of the ordinary sensitive mother to par- ticipate successfully in it. 3
When a mother and her infant of two or three weeks are facing one another, phases of lively
38/362
? ? ? social interaction occur, alternating with phases of disengagement. Each phase of interaction be- gins with initiation and mutual greeting, builds up to an animated interchange comprising facial expressions and vocalizations, during which the infant orients towards his mother with excited movements of arms and legs; then his activities gradually subside and end with the baby looking away for a spell before the next phase of interac- tion begins. Throughout these cycles the baby is likely to be as spontaneously active as his mother. Where their roles differ is in the timing of their responses. Whereas an infant's initiation and withdrawal from interaction tend to follow his own autonomous rhythm, a sensitive mother reg- ulates her behaviour so that it meshes with his. In addition she modifies the form her behaviour takes to suit him: her voice is gentle but higher pitched than usual, her movements slowed, and each next action adjusted in form and timing ac- cording to how her baby is performing. Thus she lets him call the tune and, by a skilful interweav- ing of her own responses with his, creates a dialogue.
The speed and efficiency with which these dia- logues develop and the mutual enjoyment they
39/362
? ? ? give point clearly to each participant being pread- apted to engage in them. On the one hand is the mother's intuitive readiness to allow her inter- ventions to be paced by her infant. On the other is the readiness with which the infant's rhythms shift gradually to take account of the timing of his mother's interventions. In a happily developing partnership each is adapting to the other.
Very similar alternating sequences have been recorded in other quite different exchanges between mother and child. For example, Kaye (1977), observing the behaviour of mother and in- fant during feeding, has found that mothers tend to interact with their infants in precise synchrony with the infant's pattern of sucking and pausing. During bursts of sucking a mother is generally quiet and inactive; during pauses she strokes and talks to her baby. Another example of a mother taking her cue from her infant, in this case an in- fant within the age range 5 to 12 months, is re- ported by Collis and Schaffer (1975). A mother and her infant are introduced to a scene in which there are a number of large brightly coloured toys which quickly seize their visual attention. Obser- vation of their behaviour then shows two things. First, both partners as a rule are looking at the
40/362
? ? ? same object at the same time. Secondly, examina- tion of the timing shows almost invariably that it is the baby who leads and the mother who fol- lows. The baby's spontaneous interest in the toys is evidently closely monitored by his mother who almost automatically then looks in the same dir- ection. A focus of mutual interest having been es- tablished, mother is likely to elaborate on it, com- menting on the toy, naming it, manipulating it. 'A sharing experience is then brought about, instig- ated by the infant's spontaneous attention to the environment but established by the mother al- lowing herself to be paced by the baby'.
Yet another example, also reported by Schaffer (Schaffer, Collis, and Parsons, 1977), concerns vocal interchange between mother and child at a preverbal level. In a comparison of two groups of children, aged 12 and 24 months, it was found that the ability of the pair to take turns and to avoid overlapping was not only strikingly efficient but as characteristic of the younger as of the older infants. Thus, long before the appearance of words, the pattern of turn-taking so characteristic of human conversation is already present. Here again the evidence suggests that, in ensuring the
41/362
? ? ? smooth transitions from one 'speaker' to the oth- er, mother is playing the major part.
My reason for giving these examples at some length is that I believe they illustrate some basic principles both about parenting and about the nature of the creature who is parented. What emerges from these studies is that the ordinary sensitive mother is quickly attuned to her infant's natural rhythms and, by attending to the details of his behaviour, discovers what suits him and behaves accordingly. By so doing she not only makes him contented but also enlists his co-oper- ation. For, although initially his capacity to adapt is limited, it is not absent altogether and, if al- lowed to grow in its own time, is soon yielding re- wards. Ainsworth and her colleagues have noted that infants whose mothers have responded sens- itively to their signals during the first year of life not only cry less during the second half of that year than do the babies of less responsive moth- ers but are more willing to fall in with their par- ent's wishes (Ainsworth et al. , 1978). Human in- fants, we can safely conclude, like infants of other species, are preprogrammed to develop in a so- cially cooperative way; whether they do so or not turns in high degree on how they are treated.
42/362
? ? ? This is a view of human nature, you will notice, radically different from the one that has long been current in western societies and that has permeated so much of the clinical theory and practice we have inherited. It points, of course, to a radically different conception of the role of parent.
ROLES OF MOTHERS AND FATHERS: SIMILARITIES AND DIFFERENCES
In the examples given so far, the parent con- cerned has been the mother. This is almost inev- itable because for research purposes it is relat- ively easy to recruit samples of infants who are being cared for mainly by their mother, whereas infants being cared for mainly by their father are comparatively scarce. Let me therefore describe briefly one of several recent studies which, to- gether, go some way to correct the balance.
Several hundred infants have now been studied by means of the strange situation procedure de- vised by Ainsworth (Ainsworth et al. , 1978) which gives an opportunity to observe how the infant responds, first in his parent's presence, next when he is left alone, and later when his parent returns. As a result of these observations infants
43/362
? ? ? can be classified as presenting a pattern either of secure attachment to mother or of one of two main forms of insecure attachment to her. Since these patterns have been shown to have consider- able stability during the earliest years of life and to predict how a nursery-school child in the age range 41/2 to 6 years will approach a new person and tackle a new task (Arend, Gove, and Sroufe, 1979), the value of the procedure as a method of assessing an infant's social and emotional devel- opment needs no emphasis.
Hitherto almost all the studies using this pro- cedure have observed infants with their mothers. Main and Weston (1981), however, extended the work by observing some 60 infants, first with one parent and, six months later, with the other. One finding was that, when looked at as a group, the patterns of attachment that were shown to fath- ers resembled closely the patterns that were shown to mothers, with roughly the same per- centage distribution of patterns. But a second finding was even more interesting. When the pat- terns shown by each child individually were ex- amined, no correlation was found between the pattern shown with one parent and the pattern shown with the other. Thus one child may have a
44/362
? ? ? secure relationship with the mother but not with the father, a second may have it with the father but not with the mother, a third may have it with both parents, and a fourth may have it with neither. In their approach to new people and new tasks the children represented a graded series. Children with a secure relationship to both par- ents were most confident and most competent; children who had a secure relationship to neither were least so; and those with a secure relation- ship to one parent but not to the other came in between.
Since there is evidence that the pattern of at- tachment a child undamaged at birth develops with his mother is the product of how his mother has treated him (Ainsworth et al. , 1978), it is more than likely that, in a similar way, the pat- tern he develops with his father is the product of how his father has treated him.
This study, together with others, suggests that, by providing an attachment figure for his child, a father may be filling a role closely resembling that filled by a mother; though in most, perhaps all, cultures fathers fill that role much less fre- quently than do mothers, at least when the chil- dren are still young. In most families with young
45/362
? ? ? children the father's role is a different one. He is more likely to engage in physically active and novel play than the mother and, especially for boys, to become his child's preferred play com- panion. 4
PROVISION OF A SECURE BASE
This brings me to a central feature of my concept of parenting--the provision by both parents of a secure base from which a child or an adolescent can make sorties into the outside world and to which he can return knowing for sure that he will be welcomed when he gets there, nourished phys- ically and emotionally, comforted if distressed, reassured if frightened. In essence this role is one of being available, ready to respond when called upon to encourage and perhaps assist, but to in- tervene actively only when clearly necessary. In these respects it is a role similar to that of the of- ficer commanding a military base from which an expeditionary force sets out and to which it can retreat, should it meet with a setback. Much of the time the role of the base is a waiting one but it is none the less vital for that. For it is only when the officer commanding the expeditionary force is
46/362
? ? ? confident his base is secure that he dare press forward and take risks.
In the case of children and adolescents we see them, as they get older, venturing steadily further from base and for increasing spans of time. The more confident they are that their base is secure and, moreover, ready if called upon to respond, the more they take it for granted. Yet should one or other parent become ill or die, the immense significance of the base to the emotional equilib- rium of the child or adolescent or young adult is at once apparent. In the lectures to follow evid- ence is presented from studies of adolescents and young adults, as well as of school children of dif- ferent ages from nursery school up, that those who are most stable emotionally and making the most of their opportunities are those who have parents who, whilst always encouraging their children's autonomy, are none the less available and responsive when called upon. Unfortunately, of course, the reverse is also true.
