Child care and early years education: the ideas of Susan Isaacs, Melanie Klein and Anna Freud

A historical survey of the development, and contestation, of the key person approach

This is an extract from a piece of research which I undertook a couple of years back. It is an attempt to describe how the key person approach developed through the ideas of Susan Isaacs, Melanie Klein and Anna Freud, and the significant contribution made by later practitioners and writers.

From the 1920s onwards, psychoanalytic thinking began to influence theory and practice in English nursery childcare, an attention to children’s emotional development which would eventually inform the development of the key person system.  Susan Isaacs’s Malting House School (open from 1924-1929) for children aged from 2 upwards provided the context for the development of some of the key ideas in the Kleinian school of psychoanalysis (Mitchell, 1986). Isaacs provided the children with an environment in which there were more opportunities for free play and fewer prohibitions, to enable “an all-round lessening of the degree of inhibition of children’s impulses" (Isaacs, quoted in Drummond: 2000). In this context, aspects of the child’s anxiety could be expressed symbolically, and this symbolisation was seen to support the child’s ego-development and the "greater dramatic vividness of … social and imaginative and intellectual life as a whole" (Isaacs, quoted in Drummond: 2000).

This development did not just depend on the conditions of free play, however; in the Isaacs/Klein model, the infant needs a consistently available adult figure. In a 1945 paper written for the Home Office, Isaacs (1945: 212) despairs of the impersonal “rigid routine and emotionally barren life in an institution” and proposes that instead each child “should feel himself to be the member of a small family group”  (Isaacs, 1945: 225) in care settings.

Similarly, Anna Freud arranged for each child in her wartime Hampstead Nursery to have a constant “maternal figure” (1974: XIX), contending that  “repeated experience proves the importance of the introduction of this substitute mother relationship into the life of the residential nursery. A child who forms this kind of relationship to a grown up not only becomes amenable to educational influence in a very welcome manner, but shows more vivid and varied facial expressions, develops individual qualities, and unfolds his whole personality in a surprising way” (Freud, 1974: 590).

Anna Freud (1974: XIX) proposes a model for the child’s development in which there is what she terms an “intimate interchange of affection” between the child and the “maternal figure”. Both Isaacs and Anna Freud emphasize this inner life of the child, the social, imaginative and creative life, which is developed only in the conditions of interchange: in other words it depends both on the child’s inner development, and the conditions and relationships provided by the adult. However, for Anna Freud the process is one of successive disengagement from relationships as the infant becomes more independent, a process which she does not significantly elaborate or theorise (Elliot, 2002: 27). Isaacs and Klein place significantly more emphasis on the inner psychic structure and functioning of the infant. They see the infant as splitting the world into good and bad, and feeling persecuted or even destroyed by the bad. Hence Isaacs (1945: 218) argues that if a child is in an institution where there is inadequate care, “this does not mean to him the mere absence of the good he requires, a merely neutral place; it means the actual presence of positive evil.”

Whilst for Anna Freud, the assigned carer enables a process of ego development, for Klein and Isaacs, the assigned carer creates a relationship through which a drama of love and hate, care and persecution, may be played out with the desired development of the infant taking up what Klein calls the “depressive position”, in which “the ego becomes able to take in the whole person, to see that good and bad can exist in the same person” (Mitchell, 1986: 20). Space does not allow for a full discussion of these ideas; for this paper there is one particularly salient point. For Anna Freud, the care offered by the adult to the child is absolutely real: real kindness, feeding, cleaning, and loving interaction, creating a climate in which the child can develop independence and an independent personality. For Klein and Isaacs, there is more of a dynamic relationship between the real world of physical needs and care, and the child’s inner world, which Klein calls “phantasy”. Juliet Mitchell (1986: 23) summarises that “phantasy emanates from within and imagines what is without, it offers an unconscious commentary on instinctual life and links feelings to objects and creates a new amalgam: the world of imagination”.

In the 1970s, a seminal piece of action research study was carried out in a London day nursery by Alastair Bain and Lynne Barnet, published in 1980 as The design of a day care system in a nursery setting for children under five. This is a very detailed study of relationships in a nursery setting, from a mainly Kleinian theoretical perspective. Bain and Barnet (1980:122) argue that allocating a carer for each child is not in itself adequate because of what they term “discontinuity of care provided by a single caretaker” (underlining in the original). They argue that the child who identifies with the nursery nurse’s discontinuous care can experience no more “than a momentary sense of self in relationship to the world” (Bain and Barnett, 1980: 123). The discontinuous care from the nursery nurse relates dynamically to the child’s own inner fragmentation: the phantasy’s hypothesis of existence being fragmented and persecuting is reinforced by experience. The children therefore experience discontinuous care as something more like persecution than simple neglect. In response to this serious problem, Bain and Barnett (1980: 72-73) propose a “care assignment system” which they describe in the following terms: “each child was predominantly cared for during the day by his nurse whom he could turn to for love, attention and help, at meal times, in play, when he needed comfort and affection, being changed, being helped on the lavatory, and washing”.

A more accessible and systematic elaboration of this idea was subsequently provided in Elinor Goldschmied and Sonia Jackson’s book People Under Three (1994); this outlines practical steps for implementing and managing an assigned care system, which they term the “key person” approach (1994: 35-51). However, Goldschmied and Jackson (1994) emphasize physical proximity and care, in contrast to Bain and Barnett (1980) and Hopkins (1988), all based at the Tavistock Centre, who follow in the Kleinian tradition of emphasizing psychic care and the inner development of the child’s personality, intelligence, playfulness and imagination. For Hopkins (1988: 99), it is essential that the nursery nurse offers “personal and playful interaction”, whereas Jackson and Goldschmied (1994: 8) consider play as a category which is linked to adult interaction, but discrete. They comment that “play is only one element in child development; much more crucial is adult concern and attention”.

A contested approach to the care of young children

The whole project of care assignment or key person work has been consistently, and robustly, challenged over the last three decades by a significant group of researchers, most based at the Institute of Education. Jack and Barbara Tizard’s report on residential nurseries notes that "no attempt was made in any of the nurseries we visited to reduce the number of adults handling each child by assigning the care of particular children within the group to particular members of staff” (Tizard, J and Tizard, B, 1971: 158-159).  They conclude by agreeing with the matrons they spoke to, that “close relationships should not be allowed to develop, because these were potentially damaging to the children and created difficulties for the staff.” (Tizard, J and Tizard, B, 1971: 159-160).

Helen Penn (1997: 134) shares the Tizards’ conclusion that the key person system could not work in practice; and she extends the critique by claiming that the key person system is really about “the surveillance and monitoring of individual children” (Penn, 1997: 52). Moss (2006), building on Penn’s (1997) research, locates the key person approach in a wider “maternalist regime” which, he argues, “remains dominant in many countries, still productive of ‘attachment pedagogy’ and the worker understood as substitute mother and sustaining highly gendered workforces"  (Moss, 2006: 37).

These conclusions are, I would argue, contestable. The term “attachment pedagogy” was coined by Singer (1998), as part of a wider consideration of how seemingly “normal” ideas are actually the products of specific social and political conditions. Singer (1998: 68) argues that “attachment theory offers a culturally specific model for regulating emotions and behaviour and internalizing moral concepts…such a model can justly be called a pedagogy”. Rejecting what she argues is the privileging of the close, dyadic adult:child relationship, Singer (1998: 78) proposes that there are other ways to understand how children respond to group care, noting that “many children adapt themselves in a few days and are happy to explore the new environment; but for some children this process of adaptation can last several weeks and, in exceptional cases, more than five months. Nevertheless, they all manage to feel comfortable after a while.” With a similar focus on this process of adaptation, Penn (1997: 68) describes how the staff in an Italian nursery treat “a boy who is very weepy. On and off all day, tears silently roll down his cheeks, and he stands apathetically doing nothing." She notes that the nursery staff understand and sympathise with his distress, and decide that the best idea is to "be realistic about what they can offer him. They feel that they have to help him come to terms with his misery."

Both Penn and Singer reject “attachment pedagogy” and “key working”; their views and observational data are incorporated into the arguments of Dahlberg, Moss and Pence (1999: 37) who further argue that close observation and attention to children acts as “a technology of normalization determining how children should be”; Dahlberg, Moss and Pence (2007: 38-42) extend this further with a wider critique of the modernist project to categorise, and therefore in a sense construct, the child in the discourse of developmental psychology. They propose instead an attempt to put ethics before truth and see each interaction with the child as an encounter with the Other.

Yet it is arguable that in this rejection of the normalising discourses of psychology, there is an implicit acceptance of a normalising practice in nursery care. Singer (1998) proposes that the child will adapt – eventually. This is somewhat at variance with Lévinas’s concept of an ethical encounter in which the demands and the suffering of “the Other” take primacy, in which, as Derrida (1978: 110) writes, “the face does not signify”. It is notable that for Penn the crying face does signify – weepiness and apathy. But it is not allowed to demand. It is best ignored.

If there is more discontinuity than is readily apparent in the narrative of argument that Dahlberg et al (2007) advance, then there are also sometimes surprising similarities in the research on key person care from very different standpoints. Penn (1997) identifies the significant gap between the supposed theory and the actual operation of key person work in the English nurseries, commenting that, despite good ratios, it "did not work in practice" (Penn, 1997: 88). Penn (1997: 103) observes that “physical affection was rare” and that the nursery nurse she is watching "is not deliberately unkind but … does not appear in any way to be engaged". In this respect, her observations and conclusions are strikingly similar to those of Bain and Barnett (1980) and Elfer and Dearnley (2007), and reminiscent of the powerful opening sentence of Juliet Hopkins’s study (1988: 99): “the gap between ideals and practice in child care is often inexplicably wide”.

Where Dahlberg et al (2007: 52) propose that the child should be understood as a “co-constructor of knowledge, identity and culture” rather than a “’poor’ child, weak and passive, incapable and under-developed, dependent and isolated”, there is an unexplored possible link to the work of Melanie Klein and Susan Isaacs, who, as discussed above, view the child in dynamic relationships with the external world, and who emphasize play and creativity. Isaacs’s pedagogical approach brought together a very rich environment together with an open-ness on the part of staff to respond to children’s interests, leading Mary-Jane Drummond (2000) to claim that the children could be “more active, more curious, more creative, more exploratory, and more inventive than they could have been in any ordinary school.”

As discussed above, there has been considerable empirical research into the experience of children in nursery childcare, as well as considerable theoretical elaboration; and whilst the innovative adoption of post-modern perspectives on early childhood education and care by Dahlberg et al (2007) has been portmanteau-ed with a rejection of the key person approach, there are other, possible readings in which the key person system could be re-imagined rather than rejected. However, the construction of the role and identity of the nursery nurse in the key person system and nursery childcare remains comparatively unexplored, though there are some exceptions. Stathan and Mooney (2006: 89) argue against the notion of the adult providing intimacy, cosiness and “false closeness” in the nursery. Moss (2006: 34) claims that “the early childhood worker as substitute mother produces an image that is both gendered and assumes that little or no education is necessary to undertake the work, which is understood as requiring qualities and competencies that are either innate to women (‘maternal instinct’) or else are acquired through women’s practice of domestic labour (‘housework skills’).” From an alternative perspective, promoting the development of intimate relationships between nursery nurses and their key children, Manning-Morton’s (2006: 50) training programme for nursery nurses is intended to remedy the “historical deficit view of … early years practitioners” and promote instead a new “ professional identity of a critically reflexive, theoretical boundary crosser”.


Read more posts about Susan Isaacs:
A chance encounter: Susan Isaacs and Jean Piaget
Susan Isaacs: a life freeing the minds of children