Introduction

Given their concern with the history of individuals, the practices of child psychotherapy and related disciplines have paid remarkably little attention to their own history. Within psychoanalysis, of course, there is a strong tradition of returning to Freud in introducing theoretical and technical innovations. But the focus on internal forms of historical inquiry, or on what John Forrester (1986) has described as 'psychoanalytic' histories, has been at the expense of asking what may be prior questions: How was a place produced or carved out for recognizing and treating emotional problems in childhood? How did the social context influence the development of theory and technique? How, too, has the study of emotional dynamics associated with the therapeutic professions become so sharply divided from the so-called normal development studied by academic developmental psychologists?

These questions have become particularly pertinent over the last decade. Within academic developmental psychology, child psychotherapy and psychiatry, psychoanalysis, and allied disciplines, there is a growing interest in intellectual roots and the conditions of their production. Several factors have contributed to this interest. One may be related to a general preoccupation with history and heritage, reflected in films, literature, television and other aspects of popular culture at the present time; this appears to be associated with the marked social, economic and political upheavals of the last ten years (Raphael Samuel, 1987). Another factor may be put down to the age of the traditions of psychoanalysis, psychology and psychiatry. Within psychoanalysis, for example, sufficient time has elapsed to allow for the publication of new biographies, such as Phyllis Gross -kurth's (1986) work on Melanie Klein, new accounts of influences on Sigmund Freud, such as Frank Sulloway's (1979) Freud, Biologist of the Mind, and re-examinations of old controversies, such as Jeffrey Masson's (1985) equally controversial account of Freud's abandonment of the seduction theory (Freud, 1896). Together these works broaden the basis for understanding the evolution of psychoanalysis to include not only the personalities of individual pioneers but also the social and political context of the times, and the effects of rivalries between different disciplines and professional groups.

Where children are the central focus, the evolution of the very idea of what a child is has itself demanded historical investigation. From the extreme form of the thesis that childhood is a social invention, particularly associated with Philippe Ariès's (1962) Centuries of Childhood, there has been a widespread acceptance of the idea that childhood itself is a concept with a history. That is, what children are taken to be varies between cultures and across historical epochs (Hoyles, 1979; Pollack, 1983). Within complex Western societies, specialized knowledges on child development have played a particularly important role in producing generally shared assumptions about children's emotional, physical and intellectual needs and notions of normal development. Recent analyses of particular social practices suggest that these effects are achieved through the intimate relation between the production of specialized knowledges and the ways in which institutions associated with such professions as social work, psychiatry, and education function in social regulation and the management of deviance (Donzelot, 1979; Foucault, 1977; Ingleby, 1986; Rose, 1985; Urwin, 1985a; Walkerdine, 1984).

A fourth factor promoting interest in history is more obviously related to the current political and economic climate. Restrictions in available resources and cuts in public expenditure have direct effects on professionals involved in such child-centred professions as child health, education and social work. These professionals include child psychotherapists working in the public sector who may be called upon to justify their existence. Paradoxically, economic restrictions are occurring just as developmental psychology may be beginning to recognize a lack of adequate tools for conceptualizing the contribution of emotional processes to cognitive development and symbolic functioning, raising questions about whether research psychologists can engage productively with, for example, the theory and practice of psychoanalysis (Boston, 1975, 1987; Stern, 1985; Sylvester Bradley, 1988; Urwin, 1986). The relative reduction in or pressure on existing resources is also occurring in spite of the apparently overwhelming evidence indicating persistent if not increasing emotional problems in children and young people. This is illustrated, for example, in the media coverage given to child abuse, drug dependency, depression and violence inflicted on the self or others.

As Dilys Daws (1987) has pointed out, the profession of child psychotherapy has always been a relatively small one and, perhaps because of this, it has probably always regarded itself as under threat. But whether or not the profession could or should make claims about understanding or ameliorating such manifestations of emotional turmoil cannot be answered by referring to what goes on in the consulting room between child and therapist, or even to what constitutes relief or cure in any absolute or timeless way. If it is recognized that what counts as a problem, or the forms in which emotional problems manifest themselves, vary historically and socially, then it becomes imperative to ask not only what the relation is between socially defined manifestations of unhappiness or maladaptation and the inner world of the child, but also, what role is played by the psychotherapeutic tradition in the social construction of norms of normality and pathology, either independently or through its relation to other agencies, such as medicine, education, social work and the law.

In examining how and why child psychotherapy became possible, the period between the two world wars stands out as particularly significant. At that time a number of social and administrative changes associated with the post-war reconstruction brought the twentieth-century child into the centre of the stage. These changes included developments in medicine and welfare provision aimed at infants, young children and their mothers. Concern over high infant-mortality rate and the health of children had, in fact, become marked at the turn of the century when the poor state of health amongst a large proportion of the population had been demonstrated through the large number of potential recruits who had to be turned down for service in the Boer War because of ill health (Lewis, 1980). This concern was increased after the First World War when, in addition to the considerable loss of young men, the national population was declining (Rose, 1985). The response took the form of active research into infant feeding and nutrition, the widespread dissemination of 'mothercraft' and the instigation of a large number of epidemiological studies. These aimed to demonstrate or uncover links between diseases such as tuberculosis and correlates in the social environment and, ultimately, the family. Within medicine this marked the beginnings of a shift of focus on to community care, providing a point of entry for the philosophies of early detection and prevention (Armstrong, 1983; Lewis, 1980).

Over the same time period, parallel changes were taking place in education. Though compulsory education was introduced in the last century, the educational reforms and expansions promised in the 1918 Education Act occurred in the face of overcrowded conditions in schools and limited economic resources (Simon, 1974). School inspectors, welfare workers and teachers were directly confronted with the problem of how to deal with children who would not easily fit in, and who had earlier been described as the 'potentially imbecile', the 'potentially criminal' and the 'potentially insane' (Warner, 1890; Keir, 1952). It was, of course, in the context of the problem of identifying and possibly treating the child who did not fit that the educational psychology promulgated particularly by Cyril Burt over the first decades of the century found a foothold, giving rise to the concepts of educational subnormality, delinquency and maladjustment (Burt, 1917a, 1917b, 1925; Keir, 1952; Rose, 1985).

These developments occurred at a time when for many individuals in the population a major task was to recover from the disillusion and horrors of the First World War. It was also a time when the implications of the Russian Revolution of 1917 were felt internationally. Preoccupations with Empire and with problems of social control were clearly behind many government interventions. So, too, was the deliberate aim of inculcating positive attitudes towards British democracy (Simon, 1974; Walkerdine, 1984). Yet it was also a time of potentially radical intervention, and of educational reforms directed not just at who should get taught, but what and how. This was the 'New Era' in education, the era of experiments, the foundation of Summerhill, Dartington Hall, Bedales, and the Caldecott community, and the introduction of Montessori's methods. By 1933 the government-commissioned Hadow report had endorsed the 1918 Education Act's insistence on the need for widespread expansion in nursery education (Consultative Committee of the Board of Education, 1933). Though there were substantial differences in the positions of the new generation of 'progressives', they shared a commitment to moving away from old-style, rigid and autocratic teaching methods. For many, these were associated with Prussianism or the Germanic tradition, and were seen as playing a causative role in the instigation and course of the First World War. Within this movement, providing environments which would foster emotional as well as intellectual growth became a priority. This was linked, as for Froebel and Pestalozzi, with explicitly libertarian or democratic political ideals, fuelled by a concern to promote world peace. Among the new progressives, many drew explicitly on psychoanalytic assumptions (Selleck, 1972; Walkerdine, 1984).

Though apparently radical, the idea of freeing children from oppression and repression was not entirely lost on a population increasingly exposed to discussions on 'nerves' and coping with stress. Over the first two decades of the twentieth century psychological ideas to some extent became part of everyday parlance. Of particular importance in the British context was the relatively successful treatment of victims of shell shock sustained during the war by, broadly speaking, psychodynamic methods (see Crichton-Miller, 1920). Some success had also been claimed in the elucidation of psychological concomitants to malingering, or the reluctance to return to industrial work as well as to the front. In 1920 Hugh Crichton-Miller, a doctor and psychiatrist by training, established what later became the Tavistock Clinic, London, with the idea of expanding the possibilities of psychodynamic work in peacetime into the wider community. As Dicks (1970) has emphasized, one of the first patients treated in this clinic was, in fact, a child. A children's department was rapidly established, its development fuelled in part by the relative success that could be claimed in treating children exhibiting various forms of antisocial behaviour (Dicks, 1970). By 1927, the first experimental Child Guidance Clinic had been established in this country, and the National Child Guidance Council had been founded (Burke and Miller, 1929; Keir, 1952).

The work of Crichton-Miller and his colleagues and of the Child Guidance Council was theoretically eclectic. For example, many of Crichton-Miller's associates drew their ideas from the British dynamic psychology movement rather than from psychoanalysis per se. This movement reflected a shared commitment to the idea of 'deep' strata of the mind and the power of unconscious mental forces which demanded expression, whether in adaptive behaviour or in symptomatology. The membership varied in how far they accepted Freud's tenets, such as the importance of infantile sexuality. Nevertheless, the foundation of the new clinics marked some degree of receptivity to psychoanalytic ideas. By 1926 Melanie Klein had established residence in London. Donald Winnicott began using psychoanalytic ideas within his paediatric practice at around the same time, and Anna Freud arrived in this country as a practising child psychoanalyst in 1938, shortly before the outbreak of the Second World War.

The names of these pioneers in child psychotherapy are well known throughout the Western world. A much less well-known figure is the paediatrician and child psychiatrist, Margaret Lowenfeld. Yet, in the history of child psychotherapy in the United Kingdom, her role was particularly interesting if not of crucial significance. In the inter-war years she was at the spearhead of many of the shifts of emphasis mentioned above, but always maintained a position that was tangential. This eventually allowed her to develop a form of therapy and an approach to children's minds which remains unique. Here her route was not through psychoanalysis, though she was strongly influenced by the 'New Psychology' associated with the shell-shock work mentioned previously. Rather, Lowenfeld's first commitment was to medicine and to the health of children. After training in medicine through the First World War and further wartime experience in Eastern Europe, by the mid 1920s she had moved into epidemiological research and medical practice as a paediatrician. Through this route, she was one of the first to make explicit links between emotional and physical processes as they emerge in the psychosomatic disorders of childhood (Lowenfeld, 1930). In general terms, opening up relations between mind and body was a major influence in widening the acceptance of the importance of psychological processes by the medical profession, contributing to changing the face of modern medicine and psychiatry (Armstrong, 1983; Dicks, 1970). It was while she was engaged in research on infant feeding and the problems of child care that Lowenfeld began to make observations which led her to recognize the significance of children's play. By 1928 she had founded a Children's Clinic for the Treatment and Study of Nervous and Difficult Children, a venture which was both therapeutically and research-oriented.

Following Freud (1920), at this time the child psychoanalysts were also insisting on the importance of children's play. But for Lowenfeld, major influences included the progressive educationists, the early work of Piaget, and of other proponents of the child study movement, or the movement committed to the scientific study of child development (Riley, 1983; Urwin, 1986; Walkerdine, 1984). This movement had flourished throughout Europe and the United States since the turn of the century. It actually began in the last century under the influence of Charles Darwin, whose biographical sketch of his own infant served as a kind of rubric for future researchers (Darwin, 1877; see Urwin, 1986). But by the 1920s the push to discover the regularities of child behaviour, growth and development had been fuelled by the assumption that such knowledge could provide an appropriate basis for management and intervention. The introduction of mothercraft and the rationale behind psychological testing are obvious examples of this.

Initially Lowenfeld shared a commitment to the value of recording behaviour and development according to methods which had proved so successful in the biological sciences. However, her sensitivity to what she was observing rapidly led her to the emotional life of the child. From here she developed a form of psychotherapy which differed in emphasis from psychoanalytic therapies in that play was viewed not only as a medium through which children can work through or master emotional conflicts and confusions, but also as a cognitive process, giving access to the ways in which children think. In therapy itself the emphasis was less on using the emotional relationship between child and therapist than on enhancing the child's ability to organize and make sense of his or her own experience. Similarly, in interpreting symbolic material, Lowenfeld prioritized understanding the meaning from the point of view of the child currently producing it, and stressed the need to suppress pre-existing theoretical prejudices.