Part I
Attachment and loss
1 Review I: Attachment and love
Happy he
With such a mother! Faith in womankind
Beats with his blood, and trust in all things high
Comes easy to him, and thoâ he trip and fall
He shall not blind his soul with clay.
Alfred, Lord Tennyson
The Princess (1847) pt 7, song l.308
Science and love
In recent years a large amount of research has been carried out into the patterns of attachment made between parents and children in childhood and these, as we shall see, have been found to influence the patterns of attachment, not only to parents but to others in later childhood and adult life. They also profoundly influence the ways in which people view themselves and the world at large.
In this chapter theories about the nature of human attachments will be considered in the light of recent scientific research. In Chapter 2 we go on to examine the relevance of this work for understanding the reaction to the loss of a loved person. The rest of the book draws on the experience of people with problematic reactions to bereavement to fill in the intervening links in the chain of causation between patterns of love and patterns of grief. What emerges is a new understanding of the anatomy of love.
Ever since Breuer and Freud developed the âtalking cureâ (1893) the influence of events occurring in a personâs early childhood has been recognised as likely to contribute to psychiatric problems in later life. In fact the whole field of psychoanalysis takes this assumption as its basis. During the first half of the twentieth century numerous theories were formulated to explain how parents could damage their children. Freud, Jung and Klein led the way but many others followed in their train and the field was rived with disagreement.
Freudâs theory of repression, which claimed that painful memories and ideas were âforgottenâ and transferred to âthe unconsciousâ, ruled out ordinary introspection as a valid method of research. Each school of psychoanalysis relied on the interpretations that its proponents made of the memories, dreams and free associations made by psychiatric patients. But these claims proved controversial and nobody came up with a satisfactory way of finding out who was right and who wrong.
To this day, within the main stream of psychiatry, psychoanalysis is highly suspect, but this does not prevent most psychiatrists from acknowledging that childhood influences are important, if only because of the significance attached to them by our patients. The âeclecticâ view held by most psychiatrists since the mid-twentieth century is that psychiatric disorders can only be explained by taking into account many factors which each contribute to decide why this person, at this moment in their lives, is suffering this particular combination of symptoms and problems. Genetic influences, childhood influences and the later problems and traumas to which we are exposed throughout our lives, all need to be taken into account. This psycho-biological approach, of which Adolf Meyer was the most influential exponent (Muncie 1948), laid emphasis on the importance of taking a detailed life history from each patient at the end of which the concatenation of problems would be summarised as a âpsychiatric formulationâ and a treatment plan proposed.
Though less speculative than the psychoanalytic method, this approach also suffered from the lack of any satisfactory way of deciding which of the many circumstances and events that patients recalled were contributing significantly to their current difficulties and what should be done about them. Again the door was open to numerous theories and prejudices.
Not that there was any shortage of research. Great strides had been made in genetics, neuro-anatomy, neuro-physiology, psychology, psychiatry, ethology, sociology and neuro-pharmacology and each of these disciplines has something important to contribute. But, as is usually the case with scientists, workers in each discipline tend to act in isolation from the rest and to develop their own language and frame of reference rather than to make links with other disciplines. Few people attempt to cross the boundaries in order to produce an integrated theory.
More recently the information explosion has made it more difficult than ever to keep up to date with all the literature. Renaissance Man is dead and we are all afraid of being labelled dilettantes. Yet the pay-off when people take the time and trouble to cross the boundaries between fields of study is very great and modern techniques of multivariate analysis do enable us to study more than one variable at a time.
One who succeeded in bridging the gaps between disciplines was John Bowlby, the originator and pioneer of attachment theory. Having obtained a first-class honours degree at Trinity College, Cambridge in natural sciences and psychology he went on to study medicine and psychoanalysis. His scientific training made him critical of many of the theories of his fellow analysts and caused him to look beyond the confines of that discipline in his attempts to understand the problems that he met.
After the Second World War, when many children had been evacuated from danger zones and separated from one or both of their parents, Bowlby was invited to carry out a review, for the World Health Organisation, of empirical studies of the effects of maternal deprivation. This was published in 1953 as Child Care and the Growth of Love and it established, beyond reasonable doubt, the damage that could be done to small children by the absence of or rejection by a mother or mother-substitute in early childhood. It also established Bowlbyâs reputation as someone who could draw together and integrate the findings of research from many sources.
During 1951 Bowlby was looking for a theoretical explanation for these empirical findings. The answer came to him âin a flashâ after he read a draft of Konrad Lorenzâs King Solomonâs Ring (published in 1952). Lorenz was the founder of ethology, the study of animal behaviour, and his seminal book enabled Bowlby to explain, in evolutionary terms, the mechanisms by which mothers become attached to their children, and the consequences which arise when they are separated. These ideas were worked out in more detail during a fruitful year (1958) at the Center for Advanced Studies in Stanford, California. They formed the basis of his major work, the three volumes on Attachment and Loss, which took him another 22 years to complete (Vol. I Attachment 1969; Vol. II Separation 1973a; Vol. III Loss 1980). Between them these provide a body of well-argued scientific evidence in support of a new understanding of parentâchild relationships and much else beside.
In Attachment (1969), and in an earlier paper which appeared in 1958, Bowlby addressed the problem of the nature of the childâs tie to its mother. He had, by this time, recognised that the primary attachment was not always to the childâs biological mother and he used the term âmother-figureâ for this person. He saw this tie as rooted in instinct and much of the book consists of a detailed scrutiny of the complex interaction between instinct and learning which underlies all human behaviour and emotion. He described the âinternal working modelsâ of the world, which each child builds up and which are then used as a means of orientation and planning.
Bowlby reviewed the attachments of infants and mothers of non-human animal species, including the fascinating and important concept of âimprintingâ. This term was coined by Heinroth from observation of greylag geese who, when hatching from the egg, become attached to the first large moving object that they see. In the wild this is likely to be the mother, but in a laboratory situation it may be a man in a white coat. Konrad Lorenz was fond of walking into a lecture theatre followed by a line of goslings. He would then hand his coat to his assistant who would lead the goslings from the room. This attachment, once formed, was difficult to change and gave rise to the concept of âfixed action patternsâ, which could only be learned during âcritical learning periodsâ. Many other examples of imprinting have now been discovered in a variety of species and are most likely to arise shortly after birth (McFarland 1981: 303â305).
Bowlby then turned his attention to human infants and described the sequence of behaviours by which the attachment to the mother-figure is developed and expressed during the first two years of life. These âattachment behavioursâ include sucking, crying, smiling, clinging and following. Each of these is modified, from the time of its inception, by the behaviour of the mother-figure so that, by the end of the second year, large differences are already evident between the patterns of attachment exhibited by different infants. These differences, in turn, influence the internal models of the world as seen by each child.
An explanation for many of the important differences was given in Bowlbyâs second volume, Separation: Anxiety and Anger (1973a). In this he showed how temporary separations from mother-figures can evoke a distinctive type of anxiety, âseparation anxietyâ, and anger, both of which can give rise to a second level of problems such that lasting difficulties in relationships and personality development may persist even after the return of the mother-figure. He referred to the intense but anxious attachments made by children whose mothers have stayed away too long and showed how clinging may itself evoke the very behaviour that it is intended to prevent, rejection.
During this very productive period Bowlby attracted to him, at the Tavistock Institute of Human Relations, a team of researchers whose work enabled him to flesh out the bones of the theory that he was developing. James Robertson made films of children in institutional care with and without their mothers (Robertson and Bowlby 1952) and later, with his wife Joyce, was able to demonstrate that many of the damaging effects of separation from mother could be prevented by the provision of sensitive foster care (Robertson and Robertson 1967â73). Tony Ambrose carried out systematic studies of the smiling responses of young babies and showed how easily they could be augmented or extinguished by interaction with smiling or non-responsive adults (1961).
Attachment patterns in early childhood
Another of Bowlbyâs trainees was the American psychologist, Mary Ainsworth, who, after a brief spell at the Tavistock, applied Bowlbyâs theories to studying motherâchild interaction in native Ghanaians. She it was who made the important distinction between strength of attachment and security of attachment. She asked âIs the child who clings to his mother â who is afraid of the world and the people in it, and who will not move off to explore other things or other people â more strongly attached or merely more insecure?â (1963).
Ainsworth deduced that a way of studying love would be to observe the effects of separation. Returning to the USA she achieved considerable distinction by developing a systematic method of observing and classifying the patterns of attachment between infants and mothers, her Strange Situation Test (SST). It was this test which, more than anything else, placed the study of parentâinfant love on a firm scientific footing and showed how the particular ways in which mothers love their babies can have a profound effect on the ways those babies come to view themselves and their world.
In the classic SST mothers and infants, in the second year of life, are observed through a one-way mirror before, during, and after a brief period of separation in a strange room. Ainsworth described one pattern of secure attachment and two patterns of insecure attachment (Ainsworth et al. 1978). Her colleague Mary Main subsequently added a third insecure pattern following further research (Main and Goldwyn 1984; Main and Hesse 1990; Main and Solomon 1990) and Ainsworth accepted the validity of this modification, which will be included in the version that follows. Their research also showed that each pattern of attachment is associated with a particular pattern of parenting. Their categories of attachment, as observed in the SST, together with the pattern of parenting that has been found by George and Solomon (1989, 1996) to be associated with each of them, can be summarised:
1 Secure
2 Insecure
⢠Anxious/Ambivalent
⢠Avoidant
⢠Disorganised/Disoriented
Secure (Ainsworthâs Category B)
Parents whose sensitivity and responsiveness to their infantâs needs for security and a safe base from which to explore their world are adequate, or âgood enoughâ, have children who tolerate brief separations without great distress and respond rapidly and warmly to their mothersâ comforting behaviour when she returns. Subsequent research shows that, although some of these mothers may have experienced problems with their own parents, they are aware how their past has influenced the present and can describe and accept their parents in a realistic and credible way. In other words they have overcome any attachment problems of their own. It comes as no surprise to find that their marriages are also less likely to be conflicted than those of the parents of insecurely attached children (Simpson and Rholes 1994).
Insecure
Anxious/ambivalent (Ainsworthâs Category C)
Mothers who are over-anxious, insensitive to their infants and discourage exploration, have children who, in the SST, show great distress during the period of separation and who both cling and cry angrily when she returns. Their distress continues after reunion for much longer than that of the securely attached infants.
Avoidant (Ainsworthâs Category A)
The children of mothers who do not show feelings, cannot tolerate closeness and/or punish the childâs attachment behaviour, learn to inhibit their tendencies to cling and to cry. When, in the SST, mother leaves the room they appear indifferent and uncaring. When she returns they often continue playing, ignore or turn away from her.
In the early stages of her research Ainsworth saw these children as âdetachedâ, but further investigation showed that, however uncaring they may appear to be, they are in fact physiologically aroused, as reflected in a rapid heart rate, during the period of separation and for long afterwards; their indifference is more apparent than real (Sroufe and Waters 1977).
A more recent study by Belsky et al. (1984) showed that many mothers of avoidant infants are responsive to their child at low levels of stress but become less responsive if the stress level rises. This reversal of the usual pattern would seem to defeat the object of the care which is, presumably, to provide protection and security when it is most needed and to encourage autonomy when it is not.
Disorganised/disoriented (Main and Ainsworthâs Category D)
This group of children exhibit disorganised and contradictory activity. They may cry during separation but avoid the mother when she returns, or they may approach the mother, then âfreezeâ or fall on the floor; some show stereotyped behaviour, rocking to and fro or repeated...