The disruption of harmony

Most people are aware of many controversies surrounding autism today, as well as those that abounded in the 1960s asserting the fault of mothers in causing the condition. Other major controversies have centred on the MMR vaccine and the use of mercury in vaccines. More recently, debates have exploded over whether autism can truly be defined as an illness or medical condition, or whether it is in fact merely a variation in the normal human condition. Thus, when the history of the concept of autism is fully explored and examined, one should not be surprised at the level of controversy and the impassioned arguments that have been, and still are, made about it. Autism, and its association with a wider model of social development in children, has always lent itself to controversy and intense debate about what those who purport to advance any theory of it are actually arguing. It has been open to such debate because it defines such an important and vital part of the theory of psychological development. It has always been a conduit for wider social anxieties because of the presumptions made about its potential to describe the origins of social development. To take on so much theoretical baggage could never have been easy.

The first major controversies over the theory of autism and social development in children emerged in the midst of war. By the mid-1930s, political posturing over which European nation was the strongest had intensified as Britain and other countries jostled to display their imperial power and national prowess. Children were often used as examples of the thriving British race, as in the 1937 Festival of Youth in which the king and queen, along with 60,000 others, watched a 10,000-strong display of British youth at Wembley Arena.¹ With the outbreak of the Second World War in 1939, displays of prowess turned to outright military conflict. In Britain, the Chamberlain government shifted its focus to external pressures. This had a major impact on how children’s welfare services were run. It also had an impact on who was available to run those services, and it heightened theoretical tensions over the best way to conceptualise and treat psychological problems in children. It led to a number of major scientific controversies about the way to collect and employ scientific data when making claims about children’s early development.

As discussed in Chapter 1, the first autism was adopted into psychological theory in Britain primarily via major mental health institutions, child guidance clinics and progressive and permissive schools. It was not integrated via the mainstream education system, nor via the Board of Control authorities who preferred to work with more political ideas of social development and the presumed inability of certain groups to achieve this. One important consequence of this was that the children to whom the autism concept was first applied were always those who came within a normal range of intelligence and who had not been singled out as a supposed threat to society and placed in institutional care. As part of a growing concern with children’s rights, psychological clinics collaborated with social workers in order to help gather knowledge of family life, as part of a growing network of social investigators. Although ‘deficient’ children were disregarded, other children’s home lives were coming under intense scrutiny.

Shortly before the Second World War broke out, the Child Guidance Council and the National Committee for Mental Hygiene were consulted for a report on voluntary mental health services. The Feversham Report planned to extend child guidance services and child mental health services as part of a wider programme of preventing child ‘maladjustment’, a broad term used to define a child whose instincts were not adjusted to their environment. However, these plans were all put on hold as the pressures of war increased.² Whilst the war led to the closure of many child guidance clinics as many of the staff were drafted into the armed forces,³ the conflict also stimulated the development of some forms of child observation and psychotherapeutic techniques by enabling an unfortunate testing ground for theories on the psychological impact of family breakdown, as many children were separated from their families during the mass evacuations from London during the Blitz.⁴ In just the first phase of evacuation in 1939, 750,000 schoolchildren and 542,000 mothers with small children were removed to the countryside.⁵ In London, according to official figures, half the entire school population was evacuated.⁶ Susan Isaacs, Lucy Fildes and John Bowlby, a young medically trained psychologist who had studied with most major figures in English psychology and psychoanalysis at the time, including Melanie Klein and Cyril Burt, were amongst a growing group of psychologists and psychoanalysts who studied the responses of children to these difficult circumstances and the impact on their mental health, culminating in the Cambridge Evacuation Survey of 1941.⁷

During the war, huge controversies and vitriolic arguments raged about how psychologists gathered evidence to make claims about early child development. Whereas the interwar period had witnessed a rather blasé attitude towards the integration of intelligence tests and other scientific evidence as the basis for changes in government policy towards children with ‘mental defect’, the Second World War brought new suspicions about science, and in particular mistrust of claims made by eugenicists. In the interwar period, ingrained prejudice manifested itself in the belief that certain national and racial groups were superior to others, a belief that often found its way into language about the scourge of ‘mental defectives’, and which by the Second World War had become increasingly disconcerting. Furthermore, the idea that psychological health necessarily led to a healthy society seemed at odds with the mass violence that was spreading across Europe. The war ushered in a deeper awareness of the way that political and military powers could adopt theories regarding hereditary transmission and exploit them for their own ends. The seamless rhetoric that had united psychological and social theory was then placed on increasingly shaky ground.

Instincts, science and proof

Because the first theory of autism was so bound up with a theory of human instincts and drives, any threats to a science of human instinct were necessarily also a threat to the idea of autism as an early form of thought characterised by hallucinatory wishes. Although most child psychological sciences were based on instinct theory, some psychologists had not been impressed with it. Conwy Lloyd Morgan (1852–1936), a British experimental psychologist, had rejected the concept of instinctive ‘drives’ and the ‘hormic’ theory of action, which had been advanced by psychologists such as McDougall. Morgan thought that instinctive acts should not be conceived as the result of impulsive forces or powers because these forces were metaphysical entities. He claimed that the concept of ‘behaviour’ should be used to describe the actions of both humans and animals because it was more scientific and devoid of any metaphysical claims.⁸

Although Morgan had little support amongst child development specialists in Britain, in the USA, ‘behaviourism’, as the support of behavioural concepts came to be known, was advanced enthusiastically by John Broadus Watson at Johns Hopkins University and used to develop a theory of child learning.⁹ Behaviourism gained many followers in the USA, such as Robert Yerkes, a psychologist and primatologist who was president of the American Psychological Association. These psychologists used experimental studies of learning expounded by E. L. Thorndike from the Teachers College at Columbia University and also adopted the work of the Russian physiologist, I. Pavlov, which generated a terminology for describing the processes whereby children learned based on stimulus, response and reinforcement, which came to be known as ‘conditioning’. The early behaviourists prided themselves on basing their theories only on overt, visible, measurable phenomena.¹⁰ They shunned the study of subjectivity and selfhood in psychology.

Isaacs and other British child psychologists criticised the enthusiasm for behaviourism that was emanating from the USA, as they had criticised social development scales. As Isaacs and others pointed out, because of the approach that they took, behaviourists cleverly dodged the central problems plaguing those involved in developmental psychology, namely, the way that an infant developed a relationship to reality and a relationship to society. The description and conceptualisation of autism, primary narcissism and autoerotism, and their relation to childhood schizophrenia and psychosis, was at the heart of these problems, which behaviourists happily bypassed in their focus on behaviour management.

The 1930s had seen increasing support for instinct theory in Britain, and an increasing body of literature that described the ‘conflicts’ that emerged when animal instincts were confronted with social norms. In the twenty-third edition of An Introduction to Social Psychology, published in 1936, McDougall continued to defend his theory of instincts, stating that after the sale of 62,000 copies of his book in the English language, with more in translation, ‘I am ever more convinced that these principles are valid.’¹¹ Furthermore, in 1938, Freud and his family moved from Vienna to London in order to escape persecution. Freud’s psychological take on instincts, society and individualism was becoming increasingly well known in Britain, in particular his work on Civilisation and Its Discontents (1930), which used the theory of unconscious instinctive drives to explain the psychological conflict that civilised societies inflicted on individuals driven by animal desire.¹² McDougall’s Psychoanalysis and Social Psychology (1936) developed a similar argument, ensuring that instinct theory was recognised as paramount to understanding individual behaviour and its conflicts with law and social norms.

Following the outbreak of war, English psychologists became increasingly defensive of their model of human instincts and its potential for making political statements about the need for psychological support to assist humans to adapt to the civilised world. At the same time, behaviourists were forcing instinct theorists to generate more substantive evidence for the claims they made about human nature as all psychologists felt intimidated by the prospect that their theories on society and social development may fall on deaf ears after the war. In the early 1940s, a number of arguments and controversies arose amongst psychologists, psychoanalysts, educationalists and other professional groups working with children in Britain that all centred on the innate properties – ‘impulses’, ‘instincts’ and ‘drives’ – that were thought to structure the thoughts of children. The theory of autism, autoerotism, primary narcissism and related concepts was at the heart of these debates that interrogated the evidential base for a theory of instinctive drives in children and their impact on early thought.

In November 1941, Cyril Burt made the first contribution to a major symposium on Instincts that involved many major British psychologists such as T. H. Pear, Professor of Psychology at the University of Manchester, James Drever, Professor of Psychology at the University of Edinburgh, and the founder of the British Psychological Society, C. S. Myers, which was being published in the British Journal of Educational Psychology.¹³ Elsewhere, at the British Psychoanalytic Society, concerns surrounding the scientific description of instincts in psychology were also raised. In February 1942, just three months after the debate on instincts erupted in the British Journal of Educational Psychology, a major crisis about evidence and instincts also exploded within the British Psychoanalytic Society between Melanie Klein and Anna Freud. The story of the ‘controversial discussions’ has often been told against a background of personal infighting following the death of Freud, sparked by disagreements between the established group of psychoanalysts in Britain and the Viennese émigrés whom Ernest Jones, the president of the Society, had invited to Britain following the Nazi occupation of Vienna.¹⁴ However, the passionate debates that took place between members of the British Psychoanalytical Society were strikingly similar to the debates that were also taking place between educational psychologists and teachers sparked by the introduction of behaviourist and social work methods from the USA. All of these discussions centred on the problem of how to produce convincing scientific evidence for the existence of instinctive drives in infants and children. This necessarily encompassed a defence of all the psychological concepts that spun off from instinct theory, including the theory of autism, autoerotism and primary narcissism as the most primary, and developmentally significant, states of mind.

Burt was the most vociferous supporter of instinct theory and argued strongly that statistical psychology, based on Galtonian principles, should be employed in conjunction with instinct theory in order to gain evidence about child development. He claimed the human organism purposefully strove towards particular aims and this gave human behaviour its essential characteristics. Instincts provided the force that lay behind this determination, giving these aims ‘energy’. Burt’s description of instinctive drives drew from earlier models of the transfer of energy that he regarded as the best working model for the science of instincts. He also adopted Freudian terminology and concepts, claiming that instinctive ‘unconscious motives’ were biologically the most important drives in man. He argued that the body contained ‘reserves of extra energy’ that were ‘automatically released at times of special crisis’ to support instinctive motivations and aims.¹⁵ He thought that behaviourists disregarded these fundamental neurological facts. Although the science of psychology needed development, he claimed that behaviourist models did not solve any problematic issues but merely shelved them. Instead of abandoning the foundations of psychology, Burt argued that psychologists and psychiatrists needed to conduct further factor-analytic studies that could then be used to generate a list of the fundamental human instincts that structured character traits in individuals.

Burt set out a very clear method for the collection and analysis of data to ensure that the existence of unconscious drives could be verified. He argued that infants displayed a wide range of behaviours in the first two or three years of life that could be easily predicted and recorded: ‘Even when he [the baby] seems to be simply wriggling, his hand is more likely to move to his mouth than to any other part; his thumb more frequently approaches his first finger than any of the other three; his vocal organs give vent to certain melodic patterns and to certain articulate syllables more often than others.’¹⁶ Older children will also produce certain types of behaviour that can be predicted such as ‘running away, hiding, keeping perfectly still, crying out in a particular way, producing a particular expression on the face’, etc. If data regarding these reactions, as well as their stimuli, were collected from large populations and then categorised into specific behavioural and emotional reactions, then, Burt claimed, one would find complete support for instinct theory and unconscious motivation. This would also lead to the identification of children who appeared to lack certain fundamental human instincts. Second, if one then tried to find correlations between the reactions of particular individuals and their parents, one could show that patterns of instincts were inherited across generations. His method was a ‘double factorial analysis’ that studied correlations in individuals and then correlations between individuals of the ‘same family or species’. For Burt this factor-analytic approach would directly challenge all of the behaviourists’ claims and reassert instinct theory in psychology. Although environmental and learned behaviours should be taken into consideration, the aim of psychologists would be to search for individual types with a particular biological make-up that made them susceptible to particular types of behaviour. Psychologists should then be able to predict the likely outcome for every child.¹⁷ It was with this model in mind that he attempted to prove the existence of unconscious instinctive forces in the child’s body that defined the child’s identity. In Burt’s view, an individual’s unconscious could be quantified and analysed. He was convinced that statistical methods would make this possible.

Burt claimed that he wanted to replace all behaviourists’ claims to science with his own branch of psychology: ‘In the present state of our knowledge I hold that there is room and need for a separate branch of psychology which I would call “psycho-dynamics,” having much the same relation to “neuro-dynamics” as the classical dynamics of masses has to the atomic physics of quantum theory. Like the former, it would be essentially statistical.’¹⁸ This new psychological science was needed because the measurement of intelligence was not enough to determine the mental functions and the potential social role of infants and children. For Burt, an instinct was a ‘complex inherited tendency’ that impelled an individual ‘to perceive and pay attention to certain objects or situations’ and to ‘become pleasurably or unpleasurably excited about those objects whenever they are perceived’. These tendencies, which were ‘cognitive, affective and conative’, influenced the development of individual subjectivity and character.¹⁹ Any lack or deficiency in particular instinctive drives would affect the development of the child’s individuality. Burt’s aspiration to use statistical sciences in conjunction with instinct theory in order to generate the basic facts of child development was supported by E. L. Thorndike.²⁰ However, James Drever thought that Burt was being ‘unduly optimistic’ about the potential uses of factor analysis in supporting instinct theory.²¹ Phillip Vernon, a pioneer in intelligence testing and psychological research adviser to the War Office during the Second World War, pointed out the danger of generating scientific proofs about children’s characters that could then be used to legitimise apathy regarding social reform. Any administrative system would thus apply false labels to children that would not be helpful for their development. Factorial methods of analysis, he claimed, could only be used to measure objective phenomena and dabbling in them to explain the unconscious and its role in social development was not only dangerous but also bad science. This was somewhat ironic given the marked social apathy that existed around ‘mentally defective’ children; however, Vernon’s point was that instinctive drives, the basis of unconscious thought, could not be measured in the same way as intelligence.

Although Gesell’s social development scales and Isaacs’ theory of social development were creating new models of psychological development in infancy and childhood and the formation of subjectivity and subjective awareness, no one had attempted to validate the claims made within them using statistical psychology in the same way that they had used statistical psychology to validate the theory of intelligence. Although Burt’s work on education was accepted within government departments at the time, his view that the doctrine of instincts could be used in the construction of a comprehensive ‘insurance’ system for all children reflected these interests. By quantifying the exact proportion of children who had particular tendencies, one could draw up plans to administer and manage those children in the education and health care systems, as well as in the growing system of educational psychology that would sit uneasily between these two. In Burt’s view, the collection of databanks of information for generating proofs about unconscious forces that were otherwise invisible and unknowable was essential to progress. Theories derived from them could then be used in order to structure future psychological and educational services for these children in Britain.

One thing was clear, in the heart of national and political crisis, psychologists were in desperate need of new forms of evidence that could justify their existence and support their work in a post-war administration. Behaviouristic theories were not enough to explain the development of children’s self-identity and their contact with reality. There was a critical need for a new science of child development and its pathological manifestations that could form the basis for political reform.

Infants, the unconscious, autism, autoerotism and primary narcissism

At the British Psychoanalytic Society, other controversial discussions ensued over how to describe the development of subjectivity in infants and children. Although the precise concept of ‘autism’ was rarely mentioned, the descriptive concepts of ‘autoerotism’ and ‘primary narcissism’, a term that had been developed by Sigmund Freud as a response to Bleuler’s concept of autism, were discussed frequently. These discussions of whether, and at what age, one could attribute desires and thoughts to infants were never concluded. Yet they continued to exist as important conceptual problems in child psychology because they concerned the origins of relational thought.

Susan Isaacs supported a method for the scientific defence of instinct theory that was based not on statistics but on detailed observations of the play of individual infants. Isaacs, and the Kleinian group of analysts whom she supported, thought that their scientific observations provided evidence of subjective responses to instincts in infants. These presupposed the existence of complex mental mechanisms that could control, redirect and repress unconscious instinctual urges from the very first moments of life. For example, a phenomenon that they frequently drew attention to was that of feeding difficulties in very young infants. How could these be explained, they asked, were it not for the infant’s subjective thoughts that led him to restrain his instincts to consume? Paula Heimann, a loyal Kleinian, claimed that the infant prohibited his desires and impulses in these very early stages of life because he felt himself to be overly greedy and ‘cannibalistic’ and had the capacity to recognise this fact and correct it unconsciously.²²

Early in 1943, Isaacs described in detail her thesis that children experienced ‘phantasies’ in relation to their early experiences that led them to repress or divert internal forces and drives, which could manifest problems in later life. Isaacs claimed that from the moment an infant experienced an instinctual urge, he also had the capacity to think about that urge and to imagine the direction it may take. If an instinctual drive was frustrated, then the infant would also attach a meaning to this experience of frustration through his phantasies. As Isaacs put it: ‘Phantasy is the mental corollary, the psychic representative of instinct … Every impulse, every feeling, every mode of defence is expressed and experienced in such a specific phantasy, which gives it mental life and shows its specific direction and purpose.’²³ Isaacs then gave examples of the types of phantasies that infants may have in the first few months of life. These were not only libidinal instincts or drives but also destructive instincts and impulses. For example, if the child was feeling ‘desires towards his mother’, he would experience these, in Isaac’s words, as: ‘I want to suck the nipple, to stroke her face, to eat her up, to keep her inside me, to bite the breast, to tear her to bits, to drown and burn her, to throw her out of me.’²⁴ If these phantasies and desires then stirred up anxiety in the child because he thought that by doing this he may in some way destroy or provoke the mother with his aggression, he could then use other phantasies as defences against these anxieties or as a way of controlling instinctual urges. As Isaacs put it: ‘When he feels anxiety, stirred up by an aggressive wish, he feels, “I shall be bitten or cut up by my mother”. When he feels loss and grief, he experiences (as Freud taught us) “my mother has gone forever”. When he wants to restore his mother, he feels “I want to make her better, to feed her, to put the bits together again” and so on.’²⁵ Although Isaacs was not claiming that the child experienced these phantasies in words, it is clear that she was claiming that the child, through these psychic experiences, actually felt and imagined that he was engaging in these activities. The infant thus felt this in an omnipotent way, as if every wish that he or she had actually came true.

In 1943, Isaacs published a paper on ‘An Acute Psychotic Anxiety Occurring in a Boy of Four Years’. Her understanding of ‘psychosis’ did not relate to the boy’s symptoms, which she described as ‘periodic and severe attacks of rage when he is frustrated’.²⁶ Rather, it referred to his internal state of mind, which she claimed to have uncovered through observation of the boy’s play. Isaacs thought that his play during analytic sessions showed ‘psychotic anxieties’:

This description of a ‘psychotic’ state was premised on the idea that similar thought processes occurred in the early stages of life, before the infant developed relationships to other people. As in the work of Klein, the concept of psychosis was fused with all early thought processes, so that even pre-social or pre-relational thinking was characterised by a complex phantasy life.²⁸

Taking a term from Ferenczi, and highlighting the fact that the first desires that an infant has are oral, Isaacs and the Kleinians claimed that the infant used his early experiences and phantasies as the basis to build up a ‘rudimentary ego’ or sense of self. The infant would ‘introject’ good experiences that came to form a part of the self, and ‘project’ painful experiences that were not associated with this ego. Along with this rudimentary ego, the child would also develop a rudimentary ‘super-ego’, which could invoke feelings of anxiety and guilt and censure the instinctive desires of the child. The Kleinians claimed that these psychic mechanisms developed after the child first experienced a feeling of loss and unconsciously attributed this loss to his own actions. These were termed ‘depressive’ feelings. The group were never entirely clear about when exactly these feelings first took place, though in a paper given in March 1944, Klein stated that ‘the assumption seems justified, that the seeds of depressive feelings, in so far as birth gives rise to feelings of loss, are there from the beginning of life’.²⁹ As Freud’s daughter, Anna, and others would often point out, the Kleinians were essentially positing an early stage of development that structured the mind, and restricted instinctual drives, in a similar way to the Oedipus complex, but which occurred in the very early stages of life.³⁰ In other words, it was a model of how laws and social forces impacted on thinking, occurring from the first moments of birth; infants were always social beings.

The critics of the Klein group argued that they were attributing advanced psychical processes to the infant without giving thorough evidence and explanation for these claims. Kate Friedlander, a colleague of Anna Freud who had also studied juvenile delinquency in children, claimed that Kleinian theory was incompatible with brain anatomy because ‘the myelinisation of the white matter of the brain and the formation of the cortex are by no means finished even at the end of the first year of life’. She claimed that ‘the formation of conceptions and abstract thinking as well as the faculties which we ascribe to the function of the superego’ was dependent upon a fully grown cortex and it was ludicrous to suggest that infants could engage in such advanced thought. As Friedlander put it, ‘we would not say that the child does not start to walk in the first weeks of life because it is prevented from doing so by some mental process, but on account of the fact that the motor tracts are not fully developed yet’. If Klein’s mental functions were translated into motor functions, ‘it would mean that the child starts to walk within the first weeks of life, before it has been sitting up and crawling’.³¹

One of the most vociferous opponents to Kleinian ideas about children’s mental development was Edward Glover. Glover was highly respected in medical circles and served on the Psychological Committee of the British Medical Association in the planning of a national psychiatry service during the early 1940s.³² Glover accused Isaacs and the Kleinian group of misunderstanding the first stages of life that formed the basis for the infant’s comprehension of reality and for all scientific thinking. He argued that they were ‘addicted to a sort of psychic anthropomorphism’ that led them to confuse concepts of the psychic apparatus with actual psychic mechanisms, and then to claim that both were ‘phantasy’.³³ In Glover’s view, the infant built up the means for reality testing before he started to phantasise about the possible directions that his instincts might take. If this was not fully taken into account then there was a danger that psychoanalytic theories would revert to, and support, pre-scientific models of thinking. Glover argued that in the early stages of life, the child builds up ‘memory traces’ in the mind that are associated with either pleasurable or painful experiences. The function of these images is that of adaptation, and the image associations develop a reality value. When the child is frustrated, it seeks out images associated with pleasurable experiences and avoids those associated with pain. It is when these images become associated with actual objects that the child is able to prove the existence of reality, and to correlate the subject, the aim, and the object of any instinct that arises in him or her. These early memory traces cannot be associated with phantasies as it is only when the child has an awareness of the subject, aim, and object that he is able to phantasise about these things. To claim that the child phantasised from the beginning of life denied the stages of basic reality testing in infancy.³⁴

Anna Freud argued that after an infant was born there was a period of roughly six months when the child was inherently ‘narcissistic and autoerotic’. During this phase, the aim of an instinct was fundamental but the object of that instinct was ‘only dimly taken into account’. In this state, ‘satisfaction counts for everything and objects count for nothing’. The child had no awareness of the effects of his actions on others, had no sense of guilt or anxiety over his actions and no sense of loss.³⁵ In 1936, Anna Freud had produced a classic text, The Ego and the Mechanisms of Defence, translated into English the following year, which described the developmental stages of the ego’s formation in a clear and lucid way and which served as the model for understanding the ego’s defence mechanisms of repression, projection and sublimation. She always maintained that infants could not relate to others in any way in these early stages and therefore these early stages of thinking could not be related to later emotional disturbances. She thus supported a pure psychic state before the onset of phantasy concerning other objects and other people.

Unperturbed by such criticisms, the Kleinians claimed that they could observe the unconscious thought processes of small infants by observing their behaviour. Isaacs, for example, gave several illustrations of the way that children’s behaviour was supposedly indicative of early phantasies. These came in the form of general observations about an infant’s behaviour such as their ability to recognise and react to their mother’s voice at around one month old and their interest in watching bodies in the room and reacting with happiness or distress to particular people. ‘In these observations as to the infant’s active concern with his mother’s body – both loving and aggressive – we have full confirmation of Melanie Klein’s theories’, she claimed. In addition, in the observation of their later play activities, such as ‘putting one thing inside another, e.g. poking the finger into a hole of a bead, digging nails into soap, putting finger or hand into a can … pounding, tearing, throwing, opening boxes’, after around six months old, Isaacs claimed that one could find ‘evidence’ of children acting out instinctive phantasies of getting inside and attacking objects, which they developed in early infancy.

The analysis of children would be conducted in relation to these behaviours, which Kleinians associated directly with ‘free association’ in adults. This technique was based on interpreting the child’s ‘transference’ relation, i.e. ‘the feelings, phantasies, and sexual desires’ towards the therapist, which she assumed were re-enactments of the child’s relation to objects in very early infancy, in particular the mother. Klein claimed that ‘the transference situation permeates the whole actual life of the child during analysis’; thus any material that the child brought to analysis was immediately interpreted in relation to this transference relationship, i.e. the child’s relationship to the analyst as a person.³⁶ Glover and others claimed that both the observation and analysis of children using the Kleinian technique was self-perpetuating and prejudiced. Using this method, anything could be offered up as ‘clinical evidence’ that Kleinian theories were correct.³⁷ Marjorie Brierley argued that the Kleinians were guilty of taking analogies literally, which put limitations on ‘understanding the subjective meaning of our unconscious preconceptions about mental processes and to equate these with the processes themselves’.³⁸ Many accused the Kleinians of developing religious doctrines and then pretending they were science. Anna Freud went further, arguing that if heretical psychoanalytic techniques and theories were adopted without question, then one could end up with a psychotherapeutic institute such as that set up in Berlin under the express wishes of the Nazi regime.³⁹

Although the infighting sometimes clouded the clarity of the debate, in all of these discussions it was clear that very important questions were being raised about how scientific evidence could be generated in order to explain psychological development and the formation of relationships in, and also to justify psychological work with, infants and children. The Kleinians did develop quite sophisticated theories to explain forces of energy as they travelled through the body and were then repressed and redirected. In 1943, Susan Isaacs and Paula Heimann wrote an influential paper on the nature of ‘regression’, in which they outlined their understanding of the way that instinctual drives were frustrated or blocked and then regressed to earlier stages of development. Sigmund Freud’s theories on infant sexuality had taught that during child development, the life instinct attached itself to different parts of the body. At first it was attached to the oral zone during suckling, then the anal zone during potty training, and then the phallic, or genital, zone, during the Oedipus complex. In regressive states the force of this instinct, the libido, could revert to one of the earlier ‘fixation points’ on the body, or to a similar acute phase of attachment towards a particular ‘object’.⁴⁰ This occurred to some degree in all mental pathologies. Freud would often discuss the libido in terms of dynamic flows, as he also discussed the movement of mental energy. Isaacs and Heimann used these ideas as the basis for their new metapsychology, but instead of focusing purely on the libido, they also began to theorise the ‘death instinct’ as if it were also an active force that manifested itself as aggression. They claimed that when there was an excess of aggressive impulses, the libido would get ‘dammed up’ and then both instincts would ‘defuse’, causing a backward flow of the instincts towards earlier fixation points. The libido could only overcome such regressive forces by its attempts to ‘drain the sources of destructive impulses’ and to master, or control, them. In particular, the libido would have to master the early destructive oral impulses from the first stage of life that were felt to be particularly strong.⁴¹ This detailed explanation of regression via bodily processes offered a concrete explanation of what was otherwise a very problematic concept, yet one that was often vital in the diagnosis of schizophrenia and psychosis in childhood.

Through these descriptions of energy ‘draining’ from one source to another, Isaacs and Heimann’s theories appeared to support the existence of some kind of primordial substance such as McDougall’s neurin. In their view the drainage and transmission of energy took place between the instinctual forces of life and death. In addition, although they described regression of genital impulses back towards the anal and the oral stages, it was clear that they considered the destructive oral impulses as primary. As Glover pointed out, this implied that there was a primary independent core to the unconscious through which all instincts had to pass.⁴² It was clear that the Kleinian theory of instincts envisaged the human being as something like a hydraulic system with a central control device. Instincts were channelled through the body after being adapted by a central unconscious gauge that could determine ‘good’ and ‘bad’ forces and deal with them appropriately. The Kleinians’ tendencies to simplify the vicissitudes of both instincts and mental energy lay behind Glover and Freidlander’s claims that Klein and her adherents were not true adherents of ‘dynamic’ psychology because their description of instinctive drives was too ‘primitive’.⁴³ Klein’s theory was controversial because it attempted to explain the origins of all subjective thinking processes. In any case, it was the Kleinians who developed the most comprehensive theory of the role of thinking processes in the very earliest stages of socialisation and this is why their work received so much attention and interest. It was a comprehensive theory of how infants developed relationships with others in the earliest stages of life via a theory of instinctive motivation.

Nevertheless, as demonstrated in Chapter 1, although Klein, Isaacs and other child psychologists went into the depths of psychological theory to develop a complete theory of human relationships, they were still blinkered by the fact that they did not include the mentally defective populations. Their ‘style of reasoning’ would thus always be limited when it came to postulating comprehensive theories of the development of all children as a collective mass. Even Burt’s arguments that factor analysis was the only way to prove that subjective development occurred in a similar fashion across populations did not intend to encompass defective populations. It was thus restricted in its approach. Defences for instinct theory and the concepts of developing subjectivity that it supported, including autism, autoerotism and primary narcissism, were thus all put forward without any consideration for a major section of the population and their psychological experiences. It was particularly important to include this section of the population when these psychological theories were going on to influence the management of child populations at policy and government level. The Second World War was fuelling anxieties about what would become of the child population. When the debate on instincts and the development of infant subjectivity erupted in Britain, the administration of social welfare was under reconstruction. In 1942, the Beveridge Report on Social Insurance and Allied Services was published, which alerted everyone to potential drastic changes to social insurance. In particular, the Beveridge Report laid out plans for the establishment of the National Health Service (NHS). The report set in motion discussions between medical doctors and government departments over the way in which a national health service should function once it was established. It was because of this that in July 1943, John Bowlby sought to suspend the controversial discussions to insist that the Psychoanalytic Society set aside their disagreements and immediately establish a ‘Medical Committee’ and a ‘Child Welfare Committee’ in order to consult with government departments over the future of state services.⁴⁴ Theories of child development that had been forged in the throes of war, and without any regard for children with low levels of measured intelligence, then began to form the basis for post-war government policy towards children. These practicalities of management meant that the passionate arguments died down, but the fundamental problems inherent in the description of developing subjectivity in infants and children had not gone away.

Experimentation and childhood schizophrenia in the USA

Whilst debates raged in Britain over the development of subjective awareness in infancy, in the USA, several clinicians working in the field of child psychiatry thought through similar issues but employed different methods. Lauretta Bender from the Bellevue Hospital in New York was an early supporter of Kleinian theory. She was married to Paul Schilder, who had been a member of the Viennese Psychoanalytic Society and an associate of Freud.⁴⁵ Schilder was also important in developing the neurologist Henry Head’s work on the ‘body schema’, which he called a ‘surface schema’ in which vision and cutaneous experience enable one to develop a model of the self.⁴⁶ Bender had always been interested in psychoanalytic theory, as evidenced clearly in the title of one of her early papers: ‘The Anal Component in Persecutory Delusions’. In 1938, Bender developed a method for testing the ‘perceptual problems of schizophrenic children’, in which she gave children gestalt figures to draw.⁴⁷ From the late 1930s, she began to test the responses of ‘schizophrenic’ children, arguing that their responses showed an ‘accelerated impulse to motion, action, whirling, dancing, and aggression’ and an inability to understand the physical ‘boundaries and peripheries’ of the body. These children would sometimes draw extra heads and limbs. Bender argued that these images revealed ‘a motor compliance and cohesiveness between the boundaries of two objects’ that reflected the child’s ‘fluid ego boundary’.⁴⁸ Bender also developed a ‘whirling’ test in the late 1930s in which a doctor would ask a child to stand in front of him with his arms stretched out and eyes closed and would then turn the head of the child on the neck. According to Bender, normal children responded by turning their bodies to move in line with the head, whereas ‘the schizophrenic child responds with a graceful fluid whirling which he quickly accepts as a new pattern of activity’. Such ‘rotating and whirling play in motor activity’ was thought to reflect the child schizophrenic’s ‘relationship to the reality of the outer world’.⁴⁹

Bender drew substantially from the work of Melanie Klein because it described the early stages of infantile thought as a bodily experience in immense depth, and it linked this directly to schizophrenic or ‘psychotic’ thinking. Later she also drew from the work of Gesell and Amatruda, who were interested in documenting normal developmental stages drawing on biological concepts. In their classic text The Embryology of Behavior (1945), they maintained that children’s behaviour evolved in a foetal infant through stages from homeostasis of vasovegetative control; respiratory patterns; sleep and wakefulness; tissue tonus; and motor activity patterned on a primitive tonic neck reflex behaviour. Bender defined childhood schizophrenia as a ‘pathology in behaviour at every level and in every area of integration or patterning within the functioning of the central nervous system, be it vegetative, motor, perceptual, intellectual, emotional or social’.⁵⁰ She claimed that childhood schizophrenia struck ‘at the substratum of integrative functioning or biologically patterned behaviour’, and that it could present as a ‘dramatic emergence’ of ‘artistic, philosophic or linguistic preoccupations’ rather than a simple loss of mental functions.⁵¹ She drew from Klein to argue that child schizophrenics became fixated on the stage of infancy characterised by ‘internalised objects’.⁵² She also claimed that they were particularly driven by infantile aggression. This caused ‘condensation or the superimposing of many levels of thinking and psychological problems’.⁵³ Whereas in normal children symbolism became abstract and appeared only in dreams, fantasies and fairy tales, the symbolic thought of schizophrenic children remained concrete and structured their entire thought disorder.⁵⁴ Bender later used Kleinian theory to argue that ‘psychotic’ children retained ‘primitive homeostatic control’ and ‘primitive patterns of sleep and wakefulness with waning states of consciousness’.⁵⁵

Other child psychiatrists in the USA interested in the problem of childhood schizophrenia in the early 1940s included Frances Cottington, also from Bellevue, Louise Despert from Cornell University Medical College and Charles Bradley from the Emma Pendleton Bradley Home, Rhode Island.⁵⁶ In the USA, several medically trained psychiatrists readily adopted psychoanalytic theory in their understanding of childhood schizophrenia and saw no problem in integrating it into a model of severe psychopathology. Margaret Mahler, a doctor based at Columbia University and the New York Psychoanalytic Institute, wrote widely on her therapeutic work to encourage psychotic children to engage with ‘outer reality’, as she described it.⁵⁷ She developed a concept of ‘symbiotic infantile psychosis’ that she said occurred between the ages of two and five when, she claimed, children developed symptoms in order to avert the ‘catastrophic anxiety’ of separation.⁵⁸ She argued that Klein’s concepts of ‘introjection’ and ‘projection’ were advanced psychical mechanisms and followed Anna Freud in her descriptions of ego development in children and its failure in child psychosis.⁵⁹ Many US researchers regarded childhood psychosis as a problem of the formation of the concept of selfhood.⁶⁰ J. Cotter Hirschberg, director of the Department of Child Psychiatry at the Menninger Foundation in Topeka, Kansas, argued that schizophrenic children demonstrated severe forms of ‘emotional disturbance’ that could be treated using directed education based on ‘ego development’ as understood by Anna Freud.⁶¹ Unlike Creak, many US child psychiatrists saw little problem with employing Kleinian concepts of psychosis, schizophrenia and child psychoanalysis and fusing these with the idea that childhood schizophrenia was a form of extreme psychopathology with biochemical correlates. This also encouraged a much more experimental treatment culture in relation to children than was seen in Britain.

As Edward Shorter and David Healy have argued, the Second World War stalled much European research and experimentation using insulin coma, and the focus then shifted to the USA.⁶² In Britain, a drive towards modernisation had brought insulin treatments to the Maudsley Hospital, where William Sargant and Russell Fraser pioneered them in the late 1930s with adults.⁶³ Prior to this, Maudsley doctors, such as Emslie Hutton, had experimented with thyroid shock treatment, but this was a much less dramatic treatment and was not intended to induce coma.⁶⁴ As Jack Pressman has argued, North American psychiatrists in the 1930s and 1940s sought desperately to dissociate themselves from the tarnished image of state-managed asylums that served primarily as dumping grounds for the insane, in which treatment was minimal and neglect was common.⁶⁵ These psychiatrists jumped at the prospect of presenting themselves as enlightened medical scientists engaging in cutting-edge research rather than merely asylum footmen. It was in this context that it is possible to understand the rapid growth of psychosurgery and other dramatic physical interventions in the treatment of mental illness in the 1940s, such as electroconvulsive therapy and insulin shock treatment. The use of amphetamine for children had been trialled by Charles Bradley at the Emma Pendleton Bradley Home, Rhode Island in the late 1930s, and Bender also employed it directly for childhood schizophrenia. In fact, the late 1930s and 1940s saw much experimentation with different kinds of shock therapy for mental illness in the USA, and children were subjected to very invasive treatments during this period.⁶⁶ Walter Freeman, head of neurology at George Washington University, who was well known for conducting lobotomy in adult patients, also conducted several experimental lobotomies on children diagnosed with ‘schizophrenia’ in the 1940s, some as young as four years old.⁶⁷ Frances Cottington at Bellevue, New York, had also started ‘treating’ childhood schizophrenia with Metrazol shock therapy by 1941 and these experiments were continued by Lauretta Bender.⁶⁸ Bender also engaged in a mass programme at Bellevue to administer electric shock therapy to one hundred ‘schizophrenic’ children aged four to twelve.⁶⁹ It was only later that children suffering from the after-effects of Bender’s experiments convinced other doctors to strongly warn against its use. There were reports that the children later described the experiences as horrific, one recounting that he was ‘scared to death’ and ‘felt like a bunch of rocks were going around in my head’.⁷⁰ Nevertheless, these invasive treatments for children unlucky enough to have been diagnosed with schizophrenia were sadly quite common in the 1940s.

It is in the context of increasing controversy over the description of subjective development and the formation of relationships in infants and children, as well as a strong North American focus on invasive treatments, that the publication of Leo Kanner’s 1943 article on ‘Autistic Disturbances of Affective Contact’ in 1943 must be viewed. Kanner claimed that he had identified a ‘unique “syndrome” not heretofore reported’, which was ‘inborn’ and characterised by ‘extreme autism, obsessiveness, stereotypy, and echolalia’ in children. He claimed that these symptoms brought ‘the total picture into relationship with some of the basic schizophrenic phenomena’. Kanner initially distinguished his own group of eleven cases from those of ‘dementia infantilis’ by claiming that his group of children had not experienced any period of normal development but had ‘all shown their extreme aloneness from the very beginning of life’.⁷¹ However, he later revised this theory, claiming that ‘autistic’ children were not always ‘autistic’ from birth.⁷²

What was unique about Kanner’s 1943 article was precisely the fact that it did not employ any particular psychoanalytic, psychological or psychiatric framework, nor make any claims about treatment efficacy. Instead, he simply described a group of cases in which he had observed similar symptomatology. Kanner had conducted very close observations of these children and his writing style reflected this. A typical description went as follows:

When Kanner employed the concept of autism in his article, he made no claims about the symbolic phantasy life of the children he was describing. He noted that this group of children tended to use language in a very literal fashion and that they failed to relate to other people physically. However, he also pointed out that if they were not ‘dumped in a school for the feeble-minded’, they were able to progress and develop in a way that ‘refute[d]‌ the earlier impression of extreme limitation in the child’s ideational content’.⁷⁴

Kanner’s description of ‘autistic disturbances of affective contact’ followed descriptions of autism as both an early stage of thinking and as a unique clinical ‘syndrome’ that was a subgroup of schizophrenia, so there was very little that was new in either of those ideas. What was new was his ability to capture these ideas in a circumscribed group of child cases. Kanner’s legitimacy as a writer stemmed from his pivotal place in the newly developing field of child psychiatry, as writer of the first English-language textbook on the subject. Kanner’s 1943 work has received much attention within popular understanding of the history of autism research.⁷⁵ However, his work was not widely accepted at the time and many other child psychological professionals in both Britain and the USA continued to employ the concepts of autism in conjunction with autoerotism, primary narcissism and symbolic thinking to understand infantile psychopathology and problems with developing relationships. Bender, for example, always maintained that Kanner’s ‘autism’ was simply a description of amentia or ‘mental defect’.⁷⁶ Of course, C. J. C. Earl had pointed out long before that ‘mental defectives’ were all ‘autistic’. Kanner’s definitions of autism would also be frequently challenged in Britain using these older theories of autism as part of social development and its antithesis.⁷⁷

It is true that Kanner highlighted the ‘affective’ aspects of the condition, arguing that the primary problem was one concerned with emotions and affects rather than just intellectual defect. However, as discussed earlier, the relationship between intellectual development and social and emotional development was only just beginning to be mapped out in individual and statistical studies of child development. Kanner’s work presented the topic of atypical or abnormal emotional development as something that could precede intellectual development, but in fact this had been an interest of researchers in defective institutions for a long time, and interest into this had been stimulated precisely by Bleuler’s concepts of schizophrenia and autism and Freud’s concept of primary narcissism. They had been inspired by an interest in the early stages of development and their impact on later thought through the work of Piaget. What Kanner did was present very clear and concise case histories that ostensibly carried very little theoretical baggage.

English theories of normal social development and its relation to autism, autoerotism and primary narcissism in children existed alongside the idea that Kanner’s autism was an inborn condition. This co-existence kept alive the idea that autism was a normal stage of early thinking that preceded socialisation and that some children simply could not move beyond this stage. The idea that childhood psychosis, and autism within it, was always a pathological manifestation and a shift away from normal development was promoted in some psychiatric circles, for example, by practitioners such as Creak, but the focus on the autistic aspects of childhood schizophrenia tended to shift attention away from the study of disease and pathology. The inherent contradiction between Bleuler and Piaget’s claim that autism was a normal stage of all child development, versus the claim that autism was ‘inborn’ in a select group of children, would continue to lead to many clashes and confrontations between different disciplinary groups and theoretical advocates in the post-war period. At the same time, the fact that ‘mental defectives’ were still being shunned as subjects for in-depth psychological research did not help matters.

Bowlbyism and its discontents

As the war drew to a close in Britain, fear, anxiety and uncertainty turned to optimism as new institutions for social reform were established. Beveridge’s plans were enacted in a range of new policies, including the Family Allowances Act 1945, the National Insurance Act 1946 and the Children’s Act 1948, which led to the introduction of family allowances, comprehensive welfare assistance and the establishment of local authority children’s departments in every locality, which had the right to receive children ‘in need of care and protection’. New administrative centres were set up to cater for the basic problems of malnutrition, childcare and family management.⁷⁸ The National Health Service Act 1945 placed responsibility for all children’s medical treatment on regional hospital boards rather than local authorities. These changes helped to develop medical and psychological specialities by shifting responsibility away from specialist institutions such as the Maudsley, which had previously had to deal with the needs of children suffering from malnutrition, neglect or other indications of poverty, as well as serving as an evaluation centre for schools. Both the Maudsley Hospital and the Tavistock Clinic then began to achieve a higher level of autonomy in the research they conducted and the cases they received. In 1946, John Bowlby was appointed to head the Child Department of the Tavistock, the same year that the institution came under the auspices of the NHS and expanded its training and services.⁷⁹

When considering the impact of child psychologists on British childcare policy, it is hard to underestimate the significance of John Bowlby. His work was so influential that historians have since referred to the phenomenon of ‘Bowlbyism’ as a wide-ranging social tendency to support the place of mothers in the home environment. One well-known aspect of ‘Bowlbyism’ was a tendency to blame mothers for any complications with their child’s development. It is important to point out that this applied not only to the development of ‘autism’ in children but also to every kind of psychological ‘problem’ that a child may have developed or displayed. Nevertheless, as many historians have argued, Bowlbyism was not simply a ploy to blame mothers for everything. It was part of a much wider movement in the development of child welfare, and a far more complex story about the development of children’s rights, than is often acknowledged by those who characterise it as merely a dark period in the history of autism.

Denise Riley pointed out long ago that although Bowlby’s theories were clearly influential, they were in many ways simply a reassertion of wider policies of post-war pro-natalism in which women were encouraged to resume their roles in the home environment following the huge increase in female employment during the war.⁸⁰ Bowlby had collaborated with important political figures, such as the Labour politician Evan Durbin. In their joint publication, Personal Aggressiveness and War (1939), a watered-down version of Kleinian theory was employed to emphasise the importance of mothers to the ‘emotional education’ of their children.⁸¹ As Harry Hendrick has argued, reformers of the 1940s extended the need for emotional love and stability to their model of the welfare state, which they viewed as an organic whole that could stave off rampant individualism.⁸² As the Second World War came to a close, it became almost a cliché in many official documents to claim that motherhood and housework was ‘essential work’ and that the government should recognise this by ‘enabling’ women to return to the home. Fabian socialists were the strongest supporters of a model society in which women stayed at home to look after their children whilst they were young. Their ideals were supported in official policy when state nurseries that incentivised female factory work during the war were closed.⁸³

As Mathew Thomson has recently argued, however, Bowlby’s interest in demonstrating the ill effects of maternal deprivation was part of a wider critique of state institutional care for children, and also a critique of the rather tactless way in which the state often intervened in removing children from their homes.⁸⁴ By the end of the war, the separation of children from their families in the interests of the nation had reached considerable proportions. When the Curtis Committee reported the number of UK children resident in institutional or foster care away from their families in 1946, they gave a figure of 124,900.⁸⁵ The removal of children from the home had become almost a staple of British life, and was even extended in the case of mass child migrations to new care homes in Canada, Australia and elsewhere.⁸⁶ This is a very important point because there were then few people making arguments against child institutional care or removal from the home. Bowlby’s work played a major role in the post-war construction of welfare services for children. For example, the Children Act 1948 affirmed the importance of stable family environments for children and established a centralised structure to childcare. Local authority children’s officers oversaw children’s psychological development and sought to intervene where they saw necessary, replacing the more haphazard model of child removal that had preceded this.⁸⁷ However, ‘defective’ children were not integrated in this new-found model of child freedom that prevailed after the war. The idea of ‘mental defect’ still had so much stigma attached to it that any children classified as such were not given the same rights as other children under the 1948 Act. This cultural anomaly existed up until the end of the 1950s, and overshadowed most attempts to construct comprehensive theories of social, emotional and psychological development in children.

Bowlby had a unique way of gathering evidence for his theories of maternal deprivation and its impact on the young child’s developing mind. He did not rely on scientific models of ‘instinct theory’, as Isaacs, Klein and Anna Freud had done, but instead drew very squarely from statistical and social scientific studies documenting children’s behaviour following deprivations, in particular from studies conducted in the USA.⁸⁸ The use of this kind of evidence was anathema to a significant majority of the psychoanalytic community, who accused him of misunderstanding the unconscious. He also provoked controversy when he drew from animal studies of behaviour, in particular the work of Konrad Lorenz and Harry Harlow, to demonstrate the effects of maternal deprivation. These animal studies gave Bowlby’s work scientific legitimacy from the biological sciences and helped him create a model of mother love as ‘natural’, likening the need for maternal love to the need for vitamins.⁸⁹ They were his response to the growing criticism of instinct theory as a model for understanding child development in and of itself.

Bowlby wanted clear evidence for his claims that would be accepted in the wider scientific and social scientific community. After joining the Tavistock Clinic, he appointed Esther Bick to organise the training of child psychotherapists and she instituted methods of infant and child direct observation that were later replicated internationally.⁹⁰ After the war, professionals based at the Tavistock became dominant in the development of the ‘human relations’ and ‘object relations’ school of psychoanalysis in Britain through training programmes that they ran, along with associated courses at the London School of Economics (LSE).⁹¹ The term ‘object relations’ began to be used increasingly in Britain from the 1950s in psychoanalytic writing as a shorthand to describe early thought processes prior to the development of the ego. Klein’s work was important in generating a language with which to describe such early mental states, and later psychoanalytic writers, such as W. D. Fairbairn, highlighted that early infantile instinctive drives were oriented towards ‘objects’, rather than merely towards pleasure-seeking.⁹² The Tavistock Clinic became an important institutional outlet for ‘object relations’ theory. The Tavistock was involved in the national post-war training of psychiatric social workers and other professionals specialising in social work. During the war, Susan Isaacs’ involvement with the Mental Health course for social workers at the LSE had led to Kleinian approaches being integrated into the training. This trend would continue and in 1950 the Advanced Course in Social Case Work was established at the LSE, led by Bowlby’s department at the Tavistock Clinic.

In the 1940s and 1950s, Bowlby achieved something like celebrity status through his engagement with the media at a time when media outlets were rare. He came to define a particular approach to child development that quelled the controversies over the finer points of instinct theory and the early stages of ego development by presenting a plain and lucid explanation for the cause of most childhood psychological problems. In his celebrated 1951 report for the World Health Organization, Bowlby largely referenced studies on the direct observation of mother/child separations that focused on children in institutions with little one-to-one contact.⁹³ Gesell, for example, found that these children demonstrated marked developmental lags such as ‘diminished interest and reactivity’, ‘excessive preoccupation with strange persons’ and ‘channelization and stereotypies of sensori-motor behaviour’.⁹⁴ These findings supported those reported by other researchers, such as Rene Spitz from the University of Colorado School of Medicine, who had studied hospitalised children,⁹⁵ as well as Anna Freud and Dorothy Burlingham’s studies of children displaced during wartime and the Cambridge evacuation surveys. Donald Winnicott, a paediatrician from Paddington Green Children’s Hospital who developed a strong interest in psychoanalysis following an analysis with James Strachey, had also drawn attention to the plight of institutionalised infants who were denied the attention and love of a mother.⁹⁶ Bowlby claimed that every researcher in the field, ‘with monotonous regularity, each put their finger on the child’s inability to make relationships as being the central feature from which all other disturbances sprang’.⁹⁷ Bowlby did not claim that institutional deprivation caused autism per se, although he did discuss ‘autistic’ states as a developmental stage as Piaget had done, drawing directly from a 1929 study by Marguerite Loosli-Usteri, who was then based at the Institut Jean-Jacques Rousseau with Piaget.⁹⁸ What Bowlby did claim was that institutionalisation caused problems for children in forming relationships. In 1951, particularly in Britain, this was not yet synonymous with the term ‘autism’, although it would later become so.

Winnicott and the psychoanalytic theory of autism

Bowlby was not the only psychoanalyst to become a towering figure in the popularisation of child psychoanalysis. Winnicott also took on this role with aplomb, appearing on numerous radio shows for the BBC, which were broadcast across the UK from the mid-1940s. For example, in 1949, the BBC producer Isa Benzie commissioned a series of radio programmes by Winnicott on ‘The Ordinary Devoted Mother’, which propelled his theories and parenting advice into the public arena.⁹⁹ As Michal Shapira has argued, Winnicott’s radio work was pivotal in the popularisation of psychoanalysis and in the promotion of the centrality of the mother–child bond to the development of emotional stability in infants and children.¹⁰⁰

In the early 1950s, Winnicott began to develop the specific language of ‘autistic states’ in infants as states in which an infant or child was ‘in defence against the terrible anxieties of the paranoid state’. The ‘autistic state’, he argued, had been given various names, but it was essentially a psychological state of ‘defensive pathological introversion’ in which ‘the infant lives permanently in his or her inner world which is not, however, firmly organised’.¹⁰¹ By the early 1950s, object relations theorists had thus developed quite a sophisticated language with which to describe early emotional attachments to internal objects that they claimed resembled adult psychopathology, and they used the term ‘autism’ to explain this. As Winnicott argued, ‘emotional development in its primitive or earliest stages concerns exactly the same phenomena that appear in the study of adult schizophrenia’, and can also be seen in ‘organized defences against confusion and un-integration’.¹⁰² Winnicott, in particular, used the term autism, as opposed to ‘primary narcissism’ or ‘autoerotism’, in his description of these states, and he was very clear on how to prevent them developing. As he put it: ‘I realise that much that I shall say is controversial. Nevertheless, it is necessary to explore the possibility that mental health in terms of lessened ability to develop schizoid states and to schizophrenia is laid down in the very earliest stages, when the infant is being introduced gradually to external reality.’¹⁰³ Winnicott argued that if one could intervene in infancy, one could therefore prevent schizophrenia in adulthood. This was a new departure for psychoanalysts who had previously been more interested in developing the language to describe internal mental states in great detail. Winnicott began to use the term ‘good enough mother’ in the early 1950s as a response to Klein’s discussions of the ‘good’ and ‘bad’ mother, which she used in the abstract to discuss an infant’s ‘internal objects’, rather than the actual mother herself. Winnicott used the term ‘good enough mother’ to discuss real women, bringing the descriptive terminology of psychoanalysis into the domain of parenting advice.¹⁰⁴

Similarly, Edward Glover argued that ‘the most difficult and responsible task with which the child psychiatrist is confronted is the recognition of psychotic reactions in childhood’.¹⁰⁵ Similarly to Winnicott, Glover asserted that concepts from adult psychiatry were useless because what the psychoanalyst was searching for was more like a ‘psychotic predisposition’ or a ‘pre-psychotic formation’ in infancy that could later develop into a multitude of symptom formations including ‘psycho-somatic disorders, psychopathic reactions, sexual perversions’ and ‘psychoneurotic types of mechanisms’. He was interested in reality testing more than anything else because ‘the main criterion of all psychoses, whether childhood or of adult life, lies in the degree of mal-development of or injury to the sense of reality-proving’, so he wanted to know if, when and how that mal-development took place. He argued that even in the first eighteen months of life, a child’s ‘incapacity to endure stress’ suggested a ‘constitutional predisposition’ to psychosis. For him, this was what to look out for: ‘Persistent disturbances of sleep, excessive anxiety reactions of a primary type, constant disturbances of body function, especially of oral and gastro-intestinal processes, marked skin sensitiveness, violent muscular crises of the tantrum type or persistent inertia are also significant.’¹⁰⁶ Furthermore, if the child had major ‘outbursts of violence’ following delay in gratification or ‘rapid diminution of an existing capacity for emotional enjoyment’, this suggested inadequacy in the ‘development of object relations’. For Glover, problems with object relations were most problematic if they were expressed as ‘incapacity to stand the strain of object-contact during periods of gratification’, ‘incapacity to sustain a state of object recognition in the intervals between gratification’ and ‘incapacity or delay in making a distinction between the “self” and the “non-self”’ combined to form ‘what is generally called an “autistic” reaction’.¹⁰⁷

With regard to the difficult question of hallucination, phantasy and its relation to schizophrenia in childhood, Glover argued that ‘the existence of phantasy products is not in itself pathognomonic’ and that ‘the content and bearing of the phantasies on object relations is the essential point to determine’.¹⁰⁸ Of course, all children phantasised, but one had to be particularly wary of ‘those persistent negative transferences that generally accompany the psychoses’.¹⁰⁹ Drawing from both ego psychology and Kleinian theory, he argued that the analysis and treatment of all children was possible depending ‘on the degree of synthesis of the originally fragmented ego’.¹¹⁰

Some other specialist child psychiatrists in Britain readily adopted Kleinian perspectives. For example, Louise Eickhoff, Senior Lecturer in Child Psychiatry at the University of Leeds, drew from Bender and Klein to argue that schizophrenic children demonstrated ‘emotional fixation at the infant to toddler level’, in which they fixated on the ‘oral to anal level of eroticism’, the ‘practice of bodily movements’ or the ‘flights in fantasy’ of these early stages, often employing bodily comfort mechanisms like nursery children. Eickhoff was referred a child via the local education authorities who was ‘backward, maladjusted and showed abnormal sexual advances’. Eickhoff claimed the girl was schizophrenic because she showed ‘bizarre behaviour’ and ‘difficulty with abstract thought’. Furthermore, ‘She frequently demonstrated this inability to appreciate the realities of her environment, pointing [sic] Melanie Klein’s statement of schizophrenic children in general that the ego has ceased to form a relationship to reality.’¹¹¹ Eickhoff argued that schizophrenic children showed ‘a delay in the formation of the body image’ and ‘delay in general image formation’ leading to ‘retardation or arrest in development of abstract thinking’. This meant that the ‘formation of the body image is at an intuitive rather than a perceptual level’ and does not respond to sensations registered by nerve endings in the muscles and the skin. Instead, she claimed, it ‘travels to the higher centres via the posterior columns and the thalamus’, leading to inadequate perceptions of the external world. She argued that it was possible that ‘an overactive mother might affect adversely the neurological “finish” of her child’ through her impact on foetal development.¹¹² The cure, she claimed, was as follows: ‘On this hypothesis I can conceive a scheme of treatment which would be on the lines of remedial sensation development, involving mild massage, almost fondling, of the whole body daily, gymnastics, dancing, play in water, graduated from the passive to the active.’¹¹³ She claimed to have treated three children using this method in the early 1950s, in collaboration with a physiotherapy department at Birmingham Hospital, a complete child guidance clinic team, the head teacher and staff of a school and the ‘devoted mothers’ of the children in question. Eickhoff claimed that ‘it is possible to undo the schizophrenic process in children’.¹¹⁴

Many other psychologists in Britain employed Kleinian theory to try to explain psychosis as a form of severe maladjustment. In 1948, Elizabeth Norman, a child psychologist at Guy’s Hospital, London, published a series of pictures drawn by a ‘psychotic’ child who had attended Emanuel Miller’s clinic in East London, arguing that these pictures reflected the child’s ‘lack of reality determination in the manner of play’, his ‘lack of unity and coherence in play objects’ and his ‘composite or mutilated objects, symbols, inappropriate to the reality theme of the play’.¹¹⁵ She became especially interested in the play of ‘psychotic’ children and was later referred more cases by Dr König from the Camphill Rudolf Steiner schools and Dr Howells from the Department of Child Psychiatry at Ipswich. As with many English thinkers, Norman saw ‘early infantile autism’ as an early form of schizophrenia. Drawing from the work of Klein and Bender, she thought that the infant’s suckling stage, and the inner changes relating to it, enabled the infant to ‘form a unity that precedes to integration of his own individuality as a unitary of experience, action and body-boundary’.¹¹⁶ She argued that ‘failure in development of such integration’ and an ‘impoverishment of the ego’ led to the development of ‘a relationship to objects’ that was dominated by ‘projective identification’. Building on Winnicott’s work on primitive emotional development, she claimed that ‘psychotic’ children were thus not able to integrate internal affective experiences with bodily experiences.¹¹⁷ Norman pointed out the general consensus was that ‘schizophrenic children are mute because they are self sufficient in a world of fantasy’, although it was also true that some tried to communicate. Shifting from the intra-psychic world to the problem in relationships, as was typical at the time, she argued that ‘Crucial to the withdrawal and the disturbance in social behaviour was the failure of the children to respond emotionally to the emotional behaviour of others, the failure in empathy or sympathetically induced emotion.’¹¹⁸ Norman took the last phrase from McDougall. All in all, Norman’s arguments were largely negative expressions of the stages that normal infants were supposed to go through in early development as outlined by Klein and Winnicott. As she understood it, if children failed to go through these normal psychological developmental stages then they would become psychotic. This is why the ‘schizophrenic’ child was an object of such intense scrutiny and interrogation. ‘Schizophrenic’ and ‘psychotic’ children represented the failure of all human development. They had a lot to answer for.

Glover, Winnicott and Bowlby were all considered part of the ‘independent’ group when it came to the disagreements between Anna Freud and Melanie Klein, but they developed their own unique arguments that explained all children’s psychological problems as problems of emotional attachment. They all had wider areas in which they wanted to cast their nets. Glover was interested in juvenile justice and Bowlby and Winnicott sought to turn psychoanalysis into a theory of general child development in order to guide parents and to direct public policy. Although they did not take sides on psychoanalytic controversies, they all remained committed to a theory of childhood psychosis, schizophrenia and autism that was structured around a theory of ego development that psychoanalysts could engage with therapeutically and potentially alter. As Winnicott described it, he was putting forward the contention that ‘disturbances which can be recognized and labelled as psychotic have their origin in distortions in emotional development arising before the child has clearly become a whole person capable of total relationships with whole persons’.¹¹⁹ This was tied in with a general treatise on the best family situations that infants should experience in order to develop well. Psychosis and autism were thereby drawn firmly into a general theory of child development in a way that was unique to the post-war era. For Winnicott, the development of the child’s relationship to reality was the development of the child’s relationship to other people.¹²⁰ This fitted well with the school of human relations in which he worked and also with the modelling of psychoanalysis as a total human science akin to anthropology. Perhaps most importantly, it supported his belief that English paediatricians should be involved in promoting general mental health. He thought they were a significant group involved in national health and were responsible for preventing psychosis within the nation. As he put it, ‘prophylaxis against psychosis is therefore the responsibility of the paediatricians, did they not know it’.¹²¹

At the same time, the post-war period in Britain also saw the wider conceptualisation of schizophrenia as a problem caused by culture or society. The flurry of interest in childhood schizophrenia in the early 1950s was part of a much wider interest in schizophrenia and development in the fields of psychiatry, psychoanalysis, psychology, anthropology and other human sciences of the period. Many child psychologists in Britain, including Margaret Lowenfeld, Susan Isaacs and Lydia Jackson, quoted the North American anthropologist Margaret Mead in their work on child development. In fact, in the early 1950s, Mead communicated directly with Lowenfeld over her concept of a ‘primary system’ that referred to the infant’s primary drive towards patterning, grouping, synthesising and reproducing experiences. Mead argued that the concept of a ‘primary system’ would lead to confusion with Sigmund Freud’s ‘primary processes’, which described early imagistic thinking that could tolerate contradiction and was not oriented to reality, and so Lowenfeld renamed it a ‘protosystem’.¹²² Mead, as with many anthropologists of the period, was well versed in psychoanalysis and other ‘psy’ subjects. Child psychologists, psychiatrists and anthropologists claimed they could uncover the ‘primitive’ stages of human thought as a form of psychosis.¹²³

The post-war period also saw the development of therapeutic psychiatry that played a large role in framing understandings of mental illness generally and schizophrenia in particular. During the war, Joshua Bierer began group treatments at Runwell Hospital, near London, and later Maxwell Jones developed ‘therapeutic communities’ at Mill Hill, leading to an expansion of the project in the post-war era. These unsurprisingly encouraged social and cultural explanations for the development of schizophrenia. From the late 1940s, a growing number of psychiatrists, social scientists and anthropologists in both Britain and North America explained mental illness as the result of social sickness.¹²⁴ Many of these expanded on Sigmund Freud’s Civilisation and Its Discontents. After collaborating with Margaret Mead, his first wife, on the schizophrenic nature of Balinese culture, the Cambridge-educated all-round human scientist, Gregory Bateson, published on ‘The Social Matrix of Psychiatry’, proclaiming that the psychiatric study of individuals was irrelevant because individuals were part of a wider social matrix in which changes in communication technology were transforming human behaviour and selfhood.¹²⁵

These ideas were taken up directly by child psychologists interested in childhood psychosis. For example, Lydia Jackson, an Oxford-trained psychologist who had published a best-selling book on play therapy, argued that whether classed with ‘childhood schizophrenia, infantile autism or pre-psychotic state’, all of these children did not speak and were ‘characterized by severe regression, and regression means retreat before a danger or an obstacle’.¹²⁶ She claimed that the psychopathology they presented represented a primitive response to danger observed in other animals, namely, fight, flight or feign death. As she put it, ‘silence and the absence of relationships with others are, after all, a kind of death, non existence as a social being’.¹²⁷ She also thought this was supported by anthropological work by Mead that demonstrated similar responses in Balinese children, following early frustration and teasing within the culture as a whole. She therefore encouraged ‘permissive’ techniques in enabling the child to play and explore, in a mode similar to that adopted by Isaacs and Lowenfeld. Techniques such as Lowenfeld’s sand tray were employed. Later, the therapist works to form a relationship to the child so that child views them ‘as a person’, and finally by helping the child to broaden his or her circle of relationships by encouraging group activity and learning.¹²⁸ Jackson employed the term ‘emotional and behaviour disorders’ to encompass everything that could possibly go wrong in a child’s psychological development and included conditions labelled as autistic, psychotic and schizophrenic.¹²⁹

This wider cultural shift to understand ‘schizophrenia’ as a social sickness and to relate it to particular social contexts helps to explain its rather liberal use not only in relation to infants and children but also in wider contexts. It also helps to explain how the doctrine of maternal deprivation rose so rapidly as an explanation for all forms of mental illness. In fact, it was largely historical chance that linked a lack of maternal care to all forms of psychological instability, of which childhood schizophrenia was touted as the pinnacle.

Whereas Bleuler, Piaget and Klein used the terms schizophrenia, psychosis and autism to describe internal psychic processes only, when ‘psychosis’ and ‘autism’ entered the dialogue of everyday childcare and mothering, they were used as tools to criticise parenting and the institutional practices of childcare; that is, they were used politically. Although employed for the purposes of guiding a newborn, needy, flailing welfare state, the concepts therefore became increasingly controversial, not only amongst infighting psychoanalysts, or mental defect specialists, but also amongst the wider public. At the same time, the use of such concepts, and the presumed judgements associated with their use, became increasingly common and pervasive.

In the late 1940s, child psychiatrists such as Mildred Creak, working largely out of the public eye, continued to focus on ‘schizophrenia’ and ‘psychosis’ as a psychiatric and medical diagnostic category, whilst child psychoanalysts seeking to devise comprehensive theories of social development, and to popularise such ideas, instead focused on ‘schizophrenia’ and ‘psychosis’ as a normal stage of child development. It was entirely problematic that both were using the same terms of psychosis, schizophrenia and autism to describe both normal and atypical development. That psychologists such as Klein, Bowlby, Winnicott and others appeared to be unaware of this can only be explained by the fact that their work occupied a unique cultural zone in which the social development of a large section of the child population was not regarded as significant to the development of policies for those children thought to be in the ‘normal’ range.

Other post-war institutes did begin to develop new ways to measure and monitor subjective development using statistical methods, in a manner similar to that which Burt had proposed during the controversial discussions. In 1949, Alan Moncrieff, director of the Institute of Child Health, and Dorothy Gardner, head of the Department of Child Development at the Institute of Education, established a Child Study Research Project with the aim of conducting longitudinal studies of child development that tracked a group of individuals from infancy to adulthood, and that attempted to ‘integrate knowledge on the physical, intellectual and emotional aspects of personality’. The Child Study Research Project also intended to compare results on the development of ‘normal’ children with the development of ‘abnormal’ children, both physically and mentally, and the researchers were in contact with Creak and others based at Great Ormond Street Hospital, where Creak had moved after the war.¹³⁰ Furthermore, they aimed to draw international comparisons with similar institutions in France, Switzerland, Sweden and the USA. With regard to psychological development, their aim was ‘at various age levels to assess, so far as is possible, the child’s developmental level, including cognitive abilities and, more particularly, certain personality characteristics’.¹³¹ Moncrieff, Gardner and the wider research team decided that, in comparison to Gesell and Amatruda, who had focused on cognitive development, they would focus instead on ‘the orectic side of the personality’; in other words, ‘the emotional characteristics, social behaviour and personal relationships’ that differentiated children ‘in another dimension’. Developmental scales were used only as yardsticks with which to measure this other ‘less well charted dimension’. The development of speech, manipulation and other characteristics were dependent on ‘strength of drive and attitude’ and ‘opportunity for learning’, and ‘personality development’ was dependent on ‘the relationship the child makes with his parents and siblings in a particular social context’.¹³²

This use of statistical models to track child development built on Cyril Burt’s dream to use statistics to explain all social development in children. However, this project was obviously hampered by the fact that the models for analysing, describing and detailing child development had still not effectively integrated mental models that could describe the experiences of those classed with ‘mental defect’. Even though the absence of this group was creating major rifts and disagreements amongst psychologists, problematic theories were re-hashed, allegiances were formed and no one seemed to realise that if childhood psychosis, schizophrenia, autism and mental defect were all the obverse of normal social development, then ‘normal social development’ must be a pretty problematic concept in itself.

Nevertheless, in the 1940s and early 1950s, largely due to the popularisation of psychoanalysis, it was widely accepted that during the early stages of social development, infants experienced mental states that resembled schizophrenia and psychosis in adulthood. These mental states were characterised by phantasies and hallucinations concerning people, or parts of other people such as ‘the breast’, and were often terrifying. They were associated with a complete lack of contact with any form of reality and also existed before the onset of any kind of relational or social thinking. These states were described widely as ‘autistic’, ‘schizophrenic’ and ‘psychotic’.

At the Tavistock Clinic, the Children’s Department was renamed the Department of Children and Parents in 1953, and psychotherapists aimed to integrate whole families into the treatment process. Klein’s theories of infantile phantasy and Winnicott’s work on ‘transitional objects’ helped to establish British ‘object relations’ theory as a dominant discourse within both psychological and political circles and helped to establish the ‘Tavi’ model of child psychotherapy, in which clinicians employed play therapy and transference models to elicit and expel pathological memories.¹³³ All of this helped to support the work of Bowlby, Winnicott and Glover on the importance of early mothering to stable mental health. Nikolas Rose has argued that this model of ‘human relations’ psychology was part of a wider political project to promote individual freedom, whilst simultaneously ensuring that this model of individuality mapped neatly onto the goals of British post-war objectives to establish social stability within a welfare state. Public officials would therefore use psychology to ensure that industrial workers remained contented, that women adopted their roles as Winnicott’s ‘good enough mothers’ and that social workers remained convinced that adapting and interfering in the structure and organisation of family life was a necessary task. Whilst this is largely true, it is imperative to acknowledge that this Tavistock-inspired model of psychology constructed a unique model of the development of subjectivity and individuality that was structured around the idea that early infancy was characterised by complex thinking patterns and affective bonds that pre-existed the establishment of human relationships and social relationships. This model of early infantile thought was necessary to justify the social interventions that sought to relieve and help to stabilise it.

By the early 1950s, the Tavistock model of child psychology presented a model of relationship formation that was structured on theories of pre-social unconscious object relations and pre-ego development that did not include mental deficiency, yet which privileged models of adult mental illness, in particular schizophrenia, which had at its core ideas about persecutory delusional states. However, this model of child development and human relations would come to be increasingly challenged as the social model that excluded ‘mental deficiency’, and other atypical forms of infantile thought began to be acknowledged as legitimate parts of a comprehensive social model. In fact, the controversies that existed around the description of infantile thought in the 1940s and 1950s were always constrained by the fact that ‘defective’ thought was off the table for discussion. The fact that these discussions were so exclusive only helped to fuel the disputes over who could present a complete model of children’s psychological development that was comprehensive. Unfortunately, few people were able to acknowledge the glaring omissions that had made that project impossible in the 1940s.