CHAPTER 1
PERSONAL INTRODUCTION TO ARCHETYPE: A NATURAL HISTORY OF THE SELF (1982)
On the slopes of Mount Parnis just outside Athens there is an unusual institution for unwanted children called the Metera Babiesâ Centre. The name âMeteraâ (âmotherâ in modern Greek) was chosen carefully, for it expresses the conviction of those who work there that a residential centre for the care of infants can only succeed if it performs the functions of a mother. From the time that the Metera opened in the mid-1950s, its declared policy was to provide every child, for as long as the Metera was its home, with a substitute mother with whom he might share that warm, intimate, continuous relationship which was acknowledged as being indispensable to normal human development.
When I went to Greece in 1966 to study the formation of attachment bonds in infancy, there were as many nurses in residence at the Metera as there were children: approximately 100 infants were being looked after by a staff of 36 qualified and 60 student nurses, all products of the Centreâs own School of Nursing. But despite these fairly large numbers, it was a friendly, intimate place. The soulless anonymity of traditional institutions was avoided by splitting up the community of nurses and children into small, relatively autonomous groups, each centred on one of eight separate pavilions. Each pavilion contained twelve children. Their cots were arranged in four compartments, which were divided from one another by partitions about three feet high. To each compartment was allocated one of the four graduate nurses who lived in the pavilion with the children. A graduate was known as a mother-nurse, and she was expected to devote herself exclusively to the three children in her compartment or âboxâ, as the nurses preferred to call it, using the English word.
At that time, the Metera had an English matron, who was a firm believer in the box system. Not only did she appoint a mother-nurse to each box, but she also allocated two student nurses to assist the mother-nurse specifically in the care of her three children. A brisk, upright lady with aquiline features, quick bird-like movements and sharp, twinkling eyes, the matron resembled a vigilant, though not unkindly, eagle: on her daily rounds, she would swoop down on each pavilion, checking that it was working to her satisfaction, and administering sharp pecks to any nurse she found attending to children other than those from her own box. As a result, I noted, both medical and senior nursing staff appeared satisfied that each child was receiving intensive care from a small number of women â much as a normal family-reared Greek child might be looked after by its mother, grandmother and eldest sister. It seemed an admirable arrangement.
However, within days of beginning my research it became clear to me that the only occasion on which the box system seemed to work satisfactorily was when the matron made her rounds. As soon as she took off from a pavilion, her neat theoretical arrangement invariably dissolved into a general free-for-all. Nurses and children became interchangeable to an extraordinary degree, so that during the course of a few hours â provided that the matron remained in the fastness of her eyrie â each nurse came into contact with practically every child in the pavilion. A form of maternal Marxism reigned in which caretaking was shared â from each according to her ability, to each according to his need. If a child had to be fed, comforted or have its nose wiped, as often as not it was the nearest nurse with her hands free who coped with the situation, and not the nurse officially designated as âmotherâ.
I decided to keep a careful check on what was happening, and found that in one month every child at the Metera, irrespective of the pavilion it was in, had been fed, on average, by fifteen different nurses, bathed by seven nurses, changed by fifteen nurses, put to bed in the evening by ten nurses, and lifted again in the morning by ten different nurses. It was evident that these infants were receiving multiple mothering on a scale which was possibly unprecedented in the existence of our species. Never in the whole history of mothering had so few received so much from so many.
This discovery excited me. For I realized that I had fallen into a situation which was perfectly set up to test two rival theories which were then the subject of heated and unresolved controversy. It was a stroke of good fortune such as seldom occurs in the life-time of a researcher, and when I went to the Metera I had no inkling that it was about to happen to me. Let me explain.
The theories in dispute related to the manner in which children may be supposed to become attached to those who look after them. Right up to the end of the 1950s, it had been accepted that infant attachment behaviour, like practically all other forms of human behaviour, was learned through a form of âoperant conditioningâ associated with natural rewards and punishments, the caretakerâs presence and nurturant behaviour being experienced as rewarding, and her absence or lack of maternal attention being experienced as punishing. As with most theories espoused by academic psychologists at that time, the primary reward held to be responsible for eliciting infant attachment behaviour was food, and, as a consequence, it came to be known as the âcupboard loveâ theory. Practically all psychologists, psychiatrists and psychoanalysts accepted the cupboard love theory as accounting for the facts, and it went unquestioned for decades.
Then, in 1958, the British psychiatrist John Bowlby published a now famous paper entitled âThe nature of the childâs tie to his motherâ, in which he attacked the cupboard love theory and suggested instead that infants become attached to their mothers, and mothers to their infants, not so much through learning as by instinct. Mothers and infants had no need to learn to love one another: they were innately programmed to do so from birth. The formation of mother-infant attachment bonds is a direct expression of the genetic heritage of our species.
It would be inaccurate to describe the academic reaction to Bowlbyâs paper as one of critical interest; fury would be closer to the mark. His theory outraged too many cherished assumptions for it to be received with equanimity. In the first place, the term âinstinctâ had become unacceptable to academic psychologists, who insisted that innate factors played little or no part in the behaviour of human beings. And secondly, in advancing his theory, Bowlby drew parallels between human attachment behaviour and that observed among mammals and birds. Such comparisons, many argued, were unwarranted, since human behaviour is too plastic, and too susceptible to environmental factors, to bear any resemblance to the behaviour of lesser breeds.
It was the second factor that most upset the academics, who objected to the readiness with which Bowlby borrowed concepts from the relatively new science of ethology (the study of behaviour patterns in organisms living in their natural environments) and applied them to human psychology. But Bowlby was adamant that such comparisons between different species were biologically justifiable; and in attacking the cupboard love theory he was able to cite many examples from the ethological literature of the existence of strong infant-mother bonds which had been formed through mechanisms bearing no relation to feeding gratification and which developed in the absence of any conventional rewards such as those postulated by the learning theorists.
By the time I took up my research appointment at the Metera, however, several workers had published findings which were in line with Bowlbyâs theory. For example, Dr Mary Ainsworth (1964) of Virginia University and Dr Rudolf Schaffer and Dr P. E. Emerson (1964) of Glasgow independently described the formation of strong attachments by human infants to familiar persons who played no part in feeding them. But, on the whole, the denizens of University Departments of Psychology throughout the world remained resistant to Bowlbyâs ideas, preferring to believe with Freudâs eminent daughter, Anna, that a human child learns to display attachment to its mother because she is its primary source of oral satisfaction: âWhen its powers of perception permit the child to form a conception of the person through whose agency it is fedâ, she wrote (1946), âits love is transferred to the provider of the foodâ. This view had been fully endorsed by the American psychiatrists, Dollard and Miller, who wrote (1950):
This view was still extremely influential.
These, then, were the rival theories prevailing at the time. Although Bowlbyâs âethologicalâ theory was gaining ground, the cupboard love theory still had the greater number of adherents. It was against the background of this controversy that I recognized my luck in arriving at the Metera Babiesâ Centre at that particular time. I saw at once that the richly polymatric (= many-mothered) environment prevailing at the Metera would provide me with a unique opportunity to test the relative validity of the two theories.
My reasoning ran like this: if the cupboard love theory were valid it must follow that multiply-mothered children would form multiple attachments. Metera children would become attached to all the nurses who regularly cared for them. Moreover, the nurses to whom a child attached itself would necessarily be arranged in a hierarchy of preference, the nurses at the top of the hierarchy being those who fed it the most.
If, on the other hand, Bowlbyâs theory were valid, the outcome would be very different. In the circumstances in which our species evolved (what ethologists call âthe environment of evolutionary adaptednessâ) the women responsible for an infantâs care would be few in number (usually the mother and a few close relatives) and the innate mechanism controlling the development of attachment would tend to focus on only one or two figures. The tendency for an innately determined behavioural system to take as its goal a particular individual or a small group of individuals Bowlby believed to be a biological characteristic of our species, and he gave it a name: he called it monotropy. If Bowlby was right, therefore, a Metera child would not become attached to the great majority of his caretakers as the cupboard love theory would predict, but would come to demonstrate a clear preference for one nurse above all the rest.
It was wonderfully straightforward. All I had to do was select a group of infants and make regular observations of their social progress. Quickly I chose twenty-four unattached children, aged three months and above, and began recording their interactions with their nurses. Within six months I had collected enough data to establish beyond doubt that far from becoming attached to all their nurses, three-quarters of the children became specifically attached to one nurse, who was preferred way above all the rest. Even by the strictest statistical criteria (allowing for the small size of the sample) Bowlbyâs monotropic principle was confirmed.
Most children established their preference by eight or nine months of age (i.e. at about the same age that family-reared children show unequivocal signs of specific attachment to, and conscious recognition of, their mothers). Six of the children did not become specifically attached, it is true, but this was probably because most of them left the Metera for adoption before they reached the age at which specific attachment becomes obvious.
Worse still for the cupboard love theory was my finding that no less than a third of the children became attached to nurses who had done little or nothing in the way of routine caretaking of the child before the attachment bond had been formed. Thereafter, the nurse invariably did a lot more for the child â usually because she came to reciprocate the attachment, but also because the child would often refuse to be tended by any other nurse when âhisâ or âherâ nurse was in the pavilion. The crucial factors leading up to the pairing off of a particular nurse with a particular child were not so much linked with routine feeding as with play, physical contact and social interaction; the whole process was more akin to failing in love through mutual delight and attraction than to âoperant conditioningâ. I was also fascinated to discover that in few cases did a child become primarily attached to the nurse whom the matron had officially designated as its âmother-nurseâ. Attachments, it seems, cannot be made to order. One cannot legislate in matters of the heart.
The intellectual consequences of this research were for me far more important than any contribution that my work may have made to the sum of human knowledge. It convinced me that human psychology, like animal psychology, is dependent on genetic as well as environmental factors. We are programmed from birth to form attachments; and I believe that the programme operates on the a priori assumption that when our birth takes place it will be into the bosom of a family, where the primary caretaker and principal object of attachment will be mother.
Far from wishing to lend my voice to the chorus raised in protest against the application of ethological concepts to the study of human behaviour, I began to see the ethological approach as the means by which psychology might liberate itself from the behaviourists and neobehaviourists in whose thraldom it had languished for over half a century, and it made me want to cheer.
In my twenties, I had spent four precious years of my life in the mainly dispiriting study of academic (âexperimentalâ) psychology. Like many idealistic young people, I had gone to the university believing that âthe proper study of mankind is manâ and that the modern science of psychology must hold some of the keys; and like most of them, I suspect, I was disappointed. Psychology, we discovered, possessed few keys, and the doors which they opened led into rather dismal chambers. Despite more than 70 years of effort, psychology had still fallen short of its ambition to become a respectable science, largely because it lacked any sound foundation on which to base itself, but also because it attempted to emulate the wrong models, conceiving of itself more as an offshoot of physics than as an integral part of biology. Consequently, decades of experiment had yielded a wealth of disparate findings, but, as yet, there existed no coherent thread to tie them all together.
Possessing no solid basis in biology, psychology could only lend itself the semblance of coherence by adopting the quasi-theological subterfuge of establishing dogmas that it became heresy for the faithful to deny. In the psychology laboratories and lecture rooms of university departments throughout the Western world a form of neobehaviourist fundamentalism prevailed which it was academic suicide to question. Thus, it was an article of faith that organisms should be studied as âblank slatesâ (i.e., unstructured, unprogrammed, owing little to heredity and practically everything to experience) and that no determining principles existed in life apart from a handful of âdrivesâ and the famous âlaws of learningâ.
Fortunately, I was lucky in my Professor of Psychology at Oxford, Carolus Oldfield, who, though a behaviourist, had sufficient vision to look forward to a time when psychology could be established as a science compatible with Darwinism. Carolus was well disposed to my work at the Metera and it was he who introduced me to John Bowlby, who became one of my supervisors for my D.M. dissertation.
As I completed my research project at the Metera I felt greatly encouraged by the thought that the ethologists could help one to throw off the ideological straitjacket in which the behaviourists had bound us.
And, as I was soon to discover, I was not alone in this. As the 1960s drew to a close, ethology achieved a tremendous popular success, which persisted throughout the 1970s. Books by Konrad Lorenz, the father of modern ethology, went to the top of the best-seller lists, while the works of such writers as Desmond Morris and Robert Ardrey brought an appetizing blend of ethological fact and speculation to a vast and appreciative audience. There was, apparently, no shortage of people who were intrigued by the thought that human nature and animal nature were linked through a common ancestry and that many features of contemporary human behaviour might be understood in terms of their evolutionary origins.
It was clear that the application of ethology to human psychology need not be restricted to the development of intimacy between mothers and children. There were equally good reasons for examining the possibility that we are innately territorial, inclined to mate for life, potentially co-operative with allies and hostile to foes, prone to congregate in hierarchically organized communities, and so on, much in the same way as many other mammalian and primate species. To many this seemed a highly original attempt to understand the extraordinary ways in which human beings conduct their affairs. It was both stimulating and amusing to stand back and view ourselves as one animal among many with a repertoire of behaviours which are at once characteristic of our species and at the same time traceable to earlier evolutionary forms.
Easily the most important contribution of ethology was the brilliant demonstration that behaviour can be studied comparatively â in precisely the same way as anatomy. Just as two bones, the radius and ulna, have been shown by anatomists to be homologous in the wing of the bird, the foreleg of the mammal, and the forearm of man, so ethologists began to trace the evolution of patterns of behaviour by describing homologous behaviour patterns in different animal species of a...