Becoming a problem: Behaviour and reputation in the early years classroom

Introduction

How does it happen that some children acquire a reputation as a ‘problem’ in school?

In this article we discuss some of the findings of a qualitative study involving children in the Reception year—i.e., the year of entry into compulsory education in England at age four to five.¹ The research, located in four primary/infant schools, focused on problematic behaviour as this emerged within, and was shaped by, the culture of the classroom. We started from the premise that securing a successful reputation as a ‘good’ pupil, or acquiring a negative one as a ‘problem’, is never the sole responsibility of the individual child. Our aim here is to introduce and discuss some of the key issues around behaviour and reputation which surfaced in the research, using selected extracts from the project data to exemplify our arguments. A key question for the research was: what makes it difficult for some children to be, and to be recognised as, good students?

Background to the research

Behaviour in school is an enduring and contested issue, framed by shifting theoretical and professional discourses, and beset by ‘chronic definition difficulties’ (Visser, 2003, p. 10).² Problem behaviour has been associated with poor impulse control, motivation and concentration; the inability to cooperate and anticipate consequences; low empathy and self-esteem; ‘language delay’; and deficiencies in ‘emotional literacy’. Its origins have been attributed to maturational factors (Herbert, 1991), family practices (Roffey, 1992), or a combination of both (Rutter et al., 1970; Tizard et al., 1988). Current policy in England frames behaviour as one of a trilogy of ‘skills’, linked in the phrase ‘social, emotional and behavioural skills’, to be addressed within the SEAL component (Social and Emotional Aspects of Learning) of the National Strategies (DCSF, 2009, p. 2).

Whatever their differences, such perspectives often locate responsibility for behaviour either with the individual child, or with her social milieu. To the extent that education is implicated, its role is usually envisaged as that of compensating for a deficit that lies elsewhere (‘in’ the child or home). Such perspectives may yield important insights; but they overlook the potentially constitutive role of schooling itself in producing behaviour and reputation.³ Our research, by contrast, focused on problematic behaviour as it emerged within, and was shaped by, the culture of the classroom and by wider educational and social discourses. We started from the premise that children must not only act appropriately but must be recognised as having done so. Reputation is a public matter: a child becomes a problem in the eyes of others (teachers, school staff, classmates and other parents). Children's actions and demeanour are calibrated against powerful definitions, overt and tacit, of what counts as ‘normal’ development, orderly conduct and the ‘proper’ child. Yet, children are not entirely passive subjects: their interactions contribute to the fabrication of identity and reputation in the milieu of the classroom.

Theoretical framework

The theoretical framework underpinning the research was informed by discourse analysis and poststructuralist theory (Foucault, 1977; Butler, 1997, 2004; Britzman, 2000; MacLure, 2003), and particularly by work in early childhood (Davies, 1994; Walkerdine, 1999; Laws & Davies, 2000; Brown & Jones, 2002; Dahlberg & Moss, 2005; Burman, 2007). Discourse-based approaches conceptualise subjectivity as produced in the discursive practices that make up the social world. Britzman (2000, p. 36) describes discourses thus:

We were interested, then, in the ways that ‘communities of consent and dissent’ emerged in the reception classroom, and how they operated to sustain a consensus as to which children were becoming a ‘problem’.

From such a perspective, acquiring and maintaining an identity is never the sole responsibility of the child. Definitions of the ‘good’ child, ‘proper’ conduct, or ‘normal’ development are shaped by social and institutional discourses that profoundly affect ‘who’ a child will become. As Walkerdine (1999, p. 2) argues, these discourses are strongly informed by developmental psychology, and ‘privilege a particular model of normality, to the extent that it is certain children, who are “Othered”, who become the object of pathologisation discourses’.

These pathologising discourses are, Butler argues, central to the formation of the subject. Individuals acquire and maintain their status as proper, recognisable subjects under constant apprehension of the threat of becoming the ‘bad’ subject. ‘To become a “subject” is thus to have been presumed guilty, then tried and declared innocent’, Butler writes. Moreover the achievement of subjectivity requires constantly repeated acts of vigilance: it is a matter of being ‘continuously in the process of acquitting oneself of the accusation of guilt’ (1997, pp. 118–119). The spectre of the ‘bad’ subject thus haunts, and is centrally implicated in, the production of the well-formed subject. As Laws and Davies note in a study of the formation of the ‘behaviourally disturbed’ child, schools are replete with instructions and expectations as to what pupils need to do in order to perform the identity of the ‘good school student’ (2000, p. 209), and with identifications of the characteristics of the bad student, whom pupils are expected to repudiate. The research described below was alert therefore to instances of ‘pathologising’ as a resource for establishing normative pupil identities and thereby for maintaining classroom order.

According to Laws and Davies, pathologising begins early. For the young child starting school, the achievement of recognition as a good student is a matter of high stakes and short timescales:

Our research confirms the broad contours of this description. We noted the pervasive deployment of a ‘developmental/categorising psychological discourse’ to ‘explain’ the apparent failure of some children to achieve recognition as a good student, and observed extensive remedial and support mechanisms designed to equip children with the behavioural, social and emotional ‘skills’ that they were considered to lack. However as we discuss below, while the failure to become a good student was indeed ultimately attributed to the child, behaviour was construed within ‘discursive frames’ that could also make reference to a child's parents or community as a cause of her perceived failure to conform.

In summary, the research was informed by post-structural critiques of humanist notions of the ‘natural’ child. After Walkerdine (1999, p. 13), we started from the assumption that such a child ‘is not discovered but produced in regimes of truth created in those very practices which proclaim the child in all [her] naturalness’.

Categorisations of the child as ‘natural’, ‘normal’, or otherwise must nevertheless be actualised in the specifics of the interactions through which participants interpret, categorise and judge one another, and establish their own identities. Subjectivity is neither wholly pre-determined nor fully self-actualising. In Butler's terms, it is ‘a practice of improvisation within a scene of constraint’ (2004, p. 1). In order to capture more of the fine grain of the interactions through which subjectivity is produced and contested, the analytic framework also drew on conversation analysis (CA) (see Sacks et al., 1974; Drew & Heritage, 1992; Hutchby & Wooffitt, 1998). CA is attuned to the details of what participants actually say and do in the course of their interactions, and how they build their social world through their ongoing talk. For conversation analysis, as for ethnomethodology, from which it developed, social structure and social categories—‘behaviour’, ‘the family’, ‘the teacher’, etc—are not a matter of external social constraints, roles or functions imposed on hapless individuals, but are interactionally produced through the cultural and interpretive practices that people mobilise to make sense of the world and render it mutually available. Social order is not an abstract set of constraints therefore, but the outcome of practical action in situ: i.e., an ongoing accomplishment of members acting in concert. As Macbeth notes, CA is interested in how ‘ordinary worlds [are] set in place not from afar, as though distant hands were animating anonymous action, but locally, contextually, and interactionally’ (Macbeth, 2003, p. 242).⁴

Educational identities such as ‘child’, ‘adult’, ‘teacher’ and so on are thus ‘cultural events that members make happen’ (Watson, 1992, p. 262). Through the design and the sequential co-ordination of their turns at talk, participants display their orientations to one another as teacher and pupil, novice and expert, peers, etc. In whole-class lessons for instance, the conventions for turn taking and for managing interactions involving large numbers of people ensure that the teacher usually initiates topics, allocates the next speaker and takes the majority of turns at talk. This is evident in the extensively-documented three-part sequence called ‘initiation-response-evaluation’ (IRE), in which the teacher asks a question—frequently one to which she knows the answer—a student responds in the next turn, and the teacher then indicates the adequacy of the response in the third turn (see Mehan, 1979; Richards, 2006)). For example (from our project data):

By participating in IRE sequences, the participants display their orientation to differentials of power, status and knowledge. They ‘make happen’ the asymmetrical roles of teacher and pupil, and assemble the instructional order of the classroom. Failure to observe the conventions of the IRE sequence may be signalled as an ‘issue’ in the course of the interaction, as in the following example, where the teacher indicates that the second turn slot is not available for students to ask questions (from Richards, 2006):

Children must learn the interactional conventions for taking part competently in classroom talk, and thus for acting ‘properly’ as a student. Failure to do so—for instance by assuming the right to question the teacher's reasons—may draw down censure:

Macbeth describes the IRE as part of ‘a venerable pedagogy for the teaching of novices’ (2003, p. 244). Classroom talk, especially in early years settings is, he argues, dedicated to ‘the work of making “knowledge” public, witnessable, and observable from any chair in the room’ (p. 258). It may be that the public character of ‘pedagogy for novices’ is unavoidable, and in many ways productive, as Macbeth asserts—an efficient means of ‘“installing”’ knowledge in the room’ (p. 258). Yet it is precisely the public nature of classroom interaction that links it directly to reputation: when children fail to be recognised as competent interactants, they fail in full view of the class.

The research also drew on a particular strand of research within CA that is concerned with ‘membership categorisation analysis’. Stemming from the ethnomethodological work of Sacks (1972), this focuses specifically on participants’ methods for assigning one another, and themselves, to relevant identity categories (mother, parent, teacher, child, etc; see reviews in Hester & Eglin, 1997; Schegloff, 2007). People do not simply ‘belong’ to such identity categories; rather, identity is built and ratified in and through the talk itself. It is not the case that one simply ‘is’ a teacher or a student. It is necessary to ‘bring off’ one's identity as a practical accomplishment, repeatedly, every time afresh.

Membership categories are linked (by participants) to ‘category bound activities’—that is, ‘activities or actions or forms of conduct taken by the common-sense or vernacular culture to be specially characteristic of a category's members’ (Schegloff, 2007, p. 470). So for example, in a study of parent-teacher consultations, MacLure (2003, Chapter 3) shows how the identity categories of student, teacher and parent were negotiated and contested, according to tacit or explicit descriptions of the activities that members of these categories should characteristically exhibit. In the course of such identity work, blame, justification and self-esteem were at stake, as the participants sought to defend their status as good student/responsible parent/competent teacher.

Membership categorisation is thus a ‘moral order’ (Jayyusi, 1984), fraught with consequence for the participants. As Schegloff points out, failure to measure up to an identity category does not generally lead to an expansion of the scope of the category; rather, it is likely to lead to pathologising, self-censure or perceptions of inauthenticity:

Membership categorisation provided a powerful resource for investigating the ways in which the children in the project classrooms were inducted into the behaviours that characterise the good student, and the sanctions and exclusions that could arise for children who failed to ‘pass’ as the proper child.

Conversation analysis thus provides a powerful means of tracing the jeopardy and the complexity involved in the myriad daily acts of ‘doing/being a “good school student”’ (Laws & Davies, 2000, p. 207). Waksler (1991) found that ‘making it’ as a good pupil in a US kindergarten classroom was a complex accomplishment, requiring watchfulness, the ability to derive unspoken rules from the flux of classroom activity, intersubjective awareness of self as seen through the eyes of others, and the ability to modify one's own behaviour through continuously-adjusted comparisons with that of other children in the class.

In the remainder of this article we describe how, for the children in our research, being good involved mobilising complex and subtle interpretive and interactional resources, and adjusting these continuously to changing circumstances. Children needed to be able to identify the ‘category bound activities’ that characterise the good student—such as ‘being kind’, ‘being helpful’, ‘joining in’, keeping quiet, ‘being sensible’, ‘sitting properly’, ‘good listening’, ‘sharing’ and making the teacher ‘happy’—and display them through their participation in classroom interaction. They also needed to display the emotional registers that are judged appropriate for their age and status. They had to be able to negotiate conflicting expectations—for instance between competition and sharing; taking part and keeping quiet; self-reliance and compliance to authority—and know which to mobilise in a given situation, through competent participation in the turn-taking systems of classroom talk. Moreover children were required not only to accomplish these complex tasks, but crucially, to secure public recognition for having done so—an outcome that was never solely within the compass of the individual child.

Outline of the research

The research was based in one reception class (4- to 5-year-olds) in each of four primary/infant schools in Greater Manchester: a ‘faith’ school with students of mainly white-British heritage and high entitlement to free meals; an inner-city school with a multi-ethnic intake including asylum seekers and refugees; a school in a ‘leafy suburb’ of moderately affluent homes; and a city school in an area of social deprivation served by a Sure Start children's centre. The research adopted an ethnographic orientation (Hammersley & Atkinson, 2007) that sought depth of understanding of the cultures and contexts within which behaviour assumed significance for the participants. The project team worked closely with teachers and other staff, visiting each school once a week from the start of the first term in the reception class, continuing through the reception year and into the first term of the following school year. Field notes and video recordings were made of all aspects of school life, from the daily minutiae of classroom and playground activities to assemblies, concerts and parties. Interviews were held with the class teachers, and two workshops were convened at which the participating teachers met with the project team to review video and field note data and discuss emergent themes. Field notes were collated for analysis, and interviews, video recordings and meetings were transcribed. Toward the end of the project a video for use by practitioners and teacher educators was made, incorporating video and field note excerpts exemplifying key themes such as those addressed in this article, with discussion of these themes by the participating teachers and project team.

The field notes and transcribed video data were analysed using principles derived from conversation analysis and poststructuralist work on discourse, as described above. This involved careful attention to the sequential organisation of the talk, and to the categorisation of behaviour and identity by the participants. The discussion below is supported by excerpts from these data. The names of children, teachers and schools are pseudonyms.

Offending behaviour

Two broad types of behaviour were of particular concern to teachers and other school staff. Firstly, there were actions that might be considered ‘traumatic’ in their immediate physical impact, such as biting, kicking, punching or slapping. Secondly, persistent failure to comply with adult requests was a recurring issue, especially where this offended against the collective ‘rules’ that are a major focus in the reception year. It is this kind of ‘offending’ behaviour that we focus on in this article. It included ‘calling out’ or not sitting ‘properly’ in whole-class sessions, and apparent failure to listen or concentrate. Being noisy or restless in queues, assemblies and other whole-school events might also attract attention. Rules were reiterated in many forms: through explicit statements, identification of infractions, circle games, and elaborate systems of reward (such as stickers, certificates, tokens and house points). This did not mean that the rules were necessarily easy for children to decode and apply, as we discuss below. Moreover, since discipline was pre-eminently a public matter, enacted in plain view of the class, there were potentially significant implications for the formation of status and reputation.

The discursive ‘framing’ of problem behaviour

Behaviour that might have been categorised as resistant or aggressive did not inevitably result in a poor reputation. To congeal into reputation, behaviour must be inserted into a discursive frame that grants meaning and duration to a child's conduct.⁵ Within a discursive frame, individual actions come to be read as ‘signs’ of a more enduring condition or disposition. Family and community provided one such framing device, in the form of narratives of the neglectful, indulgent, anxious, uncooperative or interfering parent. For instance, one teaching assistant remarked: ‘Home lives are very haphazard for some of these children. The behaviours aren't their fault. I blame the parents… They need to live by the rules. Basically they're feral aren't they?’ (to the class teacher). To take another example: Brent's teacher called his mother into school on several occasions to talk about issues such as getting wet on the way to school. ‘Actually in a way it is his mum who is getting taught the lesson’, she remarked. Some parents were considered to have ‘spoiled’ their children by failing to teach self-discipline or consideration for others. The terminology of ‘spoiling’, or the ‘feral’ child, implies that a child has been rendered unfit or unready in some way for the demands of schooling. The school's mission becomes one of civilising children whose parents have not fulfilled their obligations adequately.

Another framing device was medicalisation—the attribution of offending behaviour to underlying physical or psychological causes such as autism or attention deficit disorder. The search for underlying causes was also evident, on a more bureaucratic scale, in the form of psychometric and behavioural instruments for baseline assessment on entry to school, to identify and support children whose behaviour or performance was judged to fall outside normal ranges. Within the frame of medicalisation, non-compliant behaviour—such as repeated failure to ‘sit properly’, pay attention or obey requests—was judged to be beyond the child's control, and staff often relaxed the rules. As we discuss further below, while this can be seen as a flexible response to the needs of individual children, it also marks them publicly as ‘different’.

The discursive framing of reputation could also invoke attributes and dispositions of the child herself. Some children were judged to be ‘manipulative’, others to be self-centred. ‘It's all about me’, one adult said, summing up Chloe's disposition; another described her as ‘a little madam’. Joe was reported to have ‘the concentration of a gnat’. Hugo's teacher felt there was ‘something about him’:

Once reputations have begun to circulate in staffrooms and dinner halls, it may be very difficult for a child's behaviour not to be interpreted as a ‘sign’ of a more pervasive problem. Reputations may be mobilised by any of the adults who work in a school setting, teachers, assistants, lunch staff, playground supervisors, etc. Indeed ancillary staff, who often live locally, may have personal knowledge of children and families, which then becomes available for the framing of reputation. Some children in the study had entered reception year with reputations already partially in place. Carter (at Dronsfield) had been assigned a personal key worker in nursery school because of his behaviour. Hugo's teacher in the extract above refers to opinions held by the nursery staff.

Discursive frames are underpinned by the assumption, derived from developmental psychology, of a normal developmental course—one of the strongest regulatory discourses of early childhood, as noted above (Walkerdine, 1988). Children who fail to act, or to be recognised as acting, in accordance with what is normal and expected for children of their age are at risk of being judged a problem. As Burman notes, developmental discourses ‘pathologise’ differences from assumed norms, and render parents subject to blame and scrutiny (2007, p. 50). The disciplinary effects of the discourse of normal development are also felt by practitioners, who risk censure (including self-censure) if ‘their’ pupils are not seen to be acting within the range that defines normal behaviour and competence.

There is an inevitable interpretive circularity in the discourse of normal development: specific child behaviours come to be read as signs of deviation from the normal path; yet the integrity of the normal path is consolidated by the identification of deviations. Discrete acts by children are read metonymically, as ‘standing for’ a bigger problem; while the status of the bigger problem is consolidated through the iterations of the acts that ‘exemplify’ it. This has practical implications for children whose actions become subject to such framing, as it may lead teachers and other arbitrators to orient to ‘offending’ behaviours and to pay less attention to those that stand ‘outside’ the frame. Children who have acquired a strong reputation may therefore find it harder to be good—or to be recognised as good.

Teachers who took part in the research were aware of the potentially harmful effects of reputation. One had tried not to ‘learn too much’ from feeder nurseries, or to convey negative information at transfer meetings with Year 1 teachers. Alan's teacher, concerned when a colleague referred to him as ‘your naughty one’, remarked to the researcher: ‘it's there now, you can't take it back’. Jake's teacher tried to interrupt the spread of his reputation by finding pretexts to talk with his mother once other parents had left for the day. Such awareness may help to interrupt the formation of negative reputations. However precisely because the formation of reputation is tied to powerful discourses about development, and by the need for collective order in the classroom, many of the processes that are involved in forming reputation are not easily available to self-scrutiny or reflection.

It is important to emphasise that the frames that make sense of problem behaviour are discursive and not causal: one cannot directly predict that the occurrence of certain family characteristics, parenting practices or behaviours will lead to a problematic reputation. Some children with comparable experiences, background or behaviour did not earn poor reputations. It is possible their behaviour was being filtered through a different set of discursive frames which constructed them as predominantly ‘good’, and that this led teachers and others to orient to positive behaviours and fail to ‘see’ the more negative ones. This is by no means to suggest that reputations are merely a matter of opinion, entirely disconnected from children's actions and their material effects. However it is the discursive interpretation/explanation of behaviour that grants meaning and significance to difference.

The pathologising of difference

Very significant amounts of time were devoted to teaching and modelling the rules and conventions for maintaining classroom order, across all the project schools. Explicit formulations and rationales were commonplace:

While rule breaking was sometimes overlooked or dealt with unobtrusively, ‘disciplinary talk’ was frequently done in front of the class and served an ‘exemplary’ purpose. There were occasional instances where a child was explicitly ‘made an example of’:

Daniel is publicly excluded from belonging with the ‘good’ children. He is again excluded, implicitly, in the following example from his Year 1 class:

‘Exemplary’ discipline helps to form a crowd of children into a class. However it also works to marginalise some children. Indeed it could be said to work by marginalising a minority, so that ‘normal’ children can recognise themselves in their difference. Lucy and Caroline have learned to distinguish themselves from Chloe:

A different form of ‘exemplary’ practice was observed with Jamie, Ishmael and Matt. All three boys (from different schools) were thought to display autistic behaviours, and all were granted exemption from some of the usual classroom routines. Ishmael was allowed to take a soft toy into assembly, and to work or play on his own during whole-class work. Matt's teachers allowed him to behave in ways that would not be tolerated with other children: for example sprawling on the floor during assembly. Jamie was seldom included in whole-class activities and was left to sit amongst the children, silent and unengaged.

While such exemptions can be understood as flexible responses to children who seem to have difficulty coping with classroom life, they may exert a cost, in terms of such children's status and identity within the class. Jamie and Ishmael fulfilled an ‘exemplary’ function for the preponderance of the class, who were able, and indeed encouraged to distinguish themselves from the child in question, and to position themselves as more mature:⁶

Tessa is asked to distinguish herself from Ishmael and to ‘grow up’—i.e., to place herself at a more advanced point on the developmental course. In the following example, we can see the children working alongside Jamie doing similar developmental ‘placing’ of themselves:

The girls’ positioning of Jamie as less competent provides an occasion for them to rehearse their own positions in the hierarchy of age and ability.

Jamie also provided occasions for other children to rehearse and claim the moral virtues that are strongly promoted in early years education, such as being kind and helpful.

Jamie's classmates were thus encouraged by teachers and assistants to ‘practice’ the virtues of kindness and helpfulness. While undoubtedly well-intentioned, this positions Jamie in a subordinate position, as a resource upon which other children can exercise their developing social and moral competence. Jamie's subordinate position was also marked in other ways, as the above examples demonstrate. He was frequently referred to by others in the third person—i.e., talked ‘about’ rather than addressed directly. And he was often subject to physical interventions—touching, repositioning, leading by the wrist, etc. The administration and distribution of touch is, we suggest, a marker of developmental or behavioural status within the early years classroom: children who are perceived to be less mature than their peers are more likely be touched and manipulated. As a previous ESRC investigation noted (Piper et al., 2006), children must ‘earn’ exemption from unsolicited touch by demonstrating that they have reached an appropriate developmental stage.

Difference thus had a complex status in the production of classroom order and the fashioning of reputation. Tolerance of difference was low: in all four classrooms, norms and conventions for behaviour, feelings and moral conduct were continually emphasised. Teachers and other adults devoted significant effort to helping children who appeared to have trouble being ‘good’; but the aim of such effort was generally to help such children become less different. Yet difference also seems paradoxically necessary to the maintenance of classroom order. A few children are granted a status as exceptions and examples, against which the normal order may recognise itself.

The difficulties of being good

Being good is not a straightforward matter: children need to do interpretive work to understand what they are expected to do, or refrain from doing, in order for their behaviour to be assigned to the category of ‘good’, or to the various related categories such as ‘sitting beautifully’/‘properly’/‘nicely’, ‘good listening’, being ‘sensible’, not being ‘silly’, putting hands up and waiting to be chosen to speak, and not speaking while the teacher is speaking. As noted, children are also expected to display a range of social, attitudinal and moral qualities—being kind, being helpful, working hard, being polite, sharing, etc. Again, it is not self-evident what has to be done in order to earn these attributions. Moreover, they may be granted or withheld by adults on grounds that are not entirely within a child's control.

Some rules concerning behaviour are relatively easy to trace ‘back’ to the requisite action (though not necessarily easy to comply with), such as a request to put one's hand up; but others may need more sophisticated ‘categorisation work’ (Baker, 2000). Sarah needs to examine her own posture and relate this to her knowledge of the rules for correct sitting in order to know how she has offended and what she needs to do to ‘sit properly’:

Being ‘sensible’—a common term for appropriate behaviour—may be significantly open to interpretation, as may be the range of behaviours that will be judged not to be sensible. Ellie, below, must inspect her own past behaviour and future intentions, and identify the nature of Mrs F's dissatisfaction with her, in order to know what she will have done in order to ‘behave more sensibly’:

Part of the problem is that evaluations are by definition retrospective: children must read ‘back’ from the adult's assessment to the behaviour which has earned it. At Chesterfield, ‘certificate assemblies’ were held weekly. These celebrated a wide range of behaviours and competences. For instance:

For the recipients, interpretive work would be required to identify what they had done in the past week that counted as ‘fantastic joining in on the carpet’, ‘settling in well’, or ‘being kind and helpful’.

The space between evaluations and the behaviours to which they retrospectively refer may be large enough for the evaluation to be withdrawn. Brent's teacher, as noted above, was angry with him (and his mother) for coming to school ‘soaked’. As the class sit on the carpet before assembly, Ms M picks up a (blank) certificate:

Brent's offending behaviour (coming to school ‘soaked’) is retrospectively identified as a breach of the ‘good listening’ for which he was prospectively to be commended, although he was not aware of the impending commendation until the point at which it was withdrawn. Evaluations of behaviour may exist in a strange ‘future pluperfect’ time in which the import of children's own actions will have been deferred, or even altered, by unforeseen events and unpredicted interpretations by others. In this case, the behaviour in question, ‘listening’, becomes a site of difficult meanings. There is a complex moral economy at work, within which the kudos attaching to one act of listening can be erased by another one, or measured against an entirely different action (getting soaked); and where a child can be held responsible for his mother's actions. It is also unclear whether Brent's teacher is displeased with the nature of his listening, or angry at him or his mother. Further, it may not be clear to Brent whether he is being admonished for not doing what he was told, or for not exercising self-discipline. As Millei (2005) notes, teacher-led control operates against self-discipline, since the former is exerted externally and the latter is supposed to operate internally.

The orthopaedics of affect

Early years education, as we have noted, is not only about ‘academic’ subjects such as literacy and numeracy, but also about moral conduct (how one should act towards others), and about emotions—i.e., what it is possible, and appropriate to feel. Some sessions were explicitly devoted to modelling feelings and attitudes—for instance circle games such as ‘passing the smile’ to one's neighbour. In sessions such as these children are shown what kinds of feelings and opinions are appropriate, and encouraged to rehearse them. In the early years classroom feelings can be taught, appraised, and even modified by authoritative adults. For instance in a circle game where each child had to say what her neighbour was ‘good at’, the teacher not only accepted but evaluated the children's contributions, and was especially positive about those that coincided with key values, such as sharing (‘that's a really important one, well done’).

We have called this the ‘orthopaedics of affect’—that is, the moulding of children's feelings, sensations and dispositions into a repertoire of discrete emotions. It can be seen in the next field note also:

Through the teacher's feedback/evaluations, the children are helped to discriminate appropriate and inappropriate feelings. Several ‘lessons’ are combined here: the kinds of feelings that would be appropriate in a particular circumstance; the importance of pleasing adults through good behaviour, the correctness of the answers given, and the teacher's right to evaluate these. Emotions, discipline and knowledge are again intertwined. The teacher evaluates the children's answers as she might in a literacy lesson (‘I was thinking of another word’). Emotion ‘work’ is not always clearly distinguished therefore from other classroom priorities. Indeed it may not always be interpreted as something that relates to a child's ‘own’ feelings. Hamid, for instance, did not seem to connect the morning routine of attaching one's name card onto one of four ‘mood’ faces to his own emotional state (cf. MacLure et al., 2010b):

Passing as the ‘proper child’

The discourse of the reception classroom offers children an idealised (though not always consistent) version of the good or ‘proper’ child, and encourages them to approximate to the behaviour, emotions and cognitive abilities of this imagined child. Being good is thus, we suggest, a matter of ‘passing’. In order to pass—i.e., to perform a reasonably convincing imitation—children must be able to recognise something of themselves and their experiences in this idealised child. It may be difficult for some children to achieve that recognition, if the repertoire of acceptable emotions or moral virtues does not match those that they experience—such as anger, boredom or envy (Adams et al., 2004). Children may also have experiences and wisdom that are not addressed by this idealised version of the proper child. Daniel was considered too adult-like in his interactional manner, as a result of his home circumstances. Chelsea's contribution, below, suggests a more nuanced understanding of the significance of letters than the Postman Pat stories can accommodate:

In order to be (seen to be) good, children need therefore to ‘pass’ as the sort of proper child that is fabricated in the texture of classroom interaction and educational discourse.

There may be many reasons why individual children are unable or unwilling to perform the mimicry that this requires. Some may be less astute than others at reading the interactional conventions that regulate definitions of good behaviour; or less able to handle the frustrations and disappointments that are inevitably involved. Others, especially if they have life skills, wisdom and experiences that do not fit the versions of children and families that prevail in early childhood education, may find themselves judged for having progressed too quickly along the path that calibrates the development of the proper child. Such children may find it hard to recognise themselves in the simulacrum of the proper child, and may become cynical about the moral economy of happiness, sharing, kindness and mutual regard that operates in the classroom.

Implications

Some of the practices that are central to education in the early years of school, and indeed highly valued, nevertheless also contribute to the marginalisation of some children. Indeed these practices may actually depend upon the constitution of such a minority, against which the preponderance may recognise itself as ‘normal’. Some children are materialised as ‘abject’, in Butler's (1993) terms, forming the ‘constitutive outside’ that renders the status quo intelligible.⁷ The public nature of discipline is an important factor here. There are undoubtedly good reasons for classroom rules—collective responsibility, courtesy, democratic participation, safety, a congenial learning environment. But this means that very significant amounts of classroom interaction may be devoted to spelling out behavioural rules and identifying violations, with resulting marginalisation of some children. The increase in whole-class teaching (Galton et al., 1999; Myhill, 2006; Wyse et al., 2008), prioritised in primary education policy over the last 15 years, will have contributed to the public visibility of behaviour.

The ‘discursive framing’ of children's reputation by reference to their homes and families raises issues relating to home-school relations—a cornerstone of UK primary education. The staff involved in the research drew sensitively on their knowledge of children's home backgrounds in making decisions about how best to meet their needs and understand their current capabilities. But there is a downside too: the humane ‘interest’ that school staff take in children's parents, community and prior experiences may also, less productively, furnish resources for ‘deficit’ views (cf. Pollard & Filer, 1999; Hughes, 2004).

The ‘holistic’ ethos of early years education, which integrates intellectual, social, moral and emotional development, requires children to decode complex, tacit and contextually variable messages about what counts as good behaviour (cf. Pryor & Torrance, 2000). Moreover the association of behaviour with emotions, and the coupling of academic performance with winning teachers’ approval, means that early years classrooms can be places of uneven emotional temperature, where praise, censure, heightened interest and unveiled indifference are enacted on shifting grounds that are not understood by all.

The ‘orthopaedics of affect’ that we identified in the research chimes with contemporary work that has identified an ‘affective turn’ (Conroy et al., 2008) or an ‘emotional turn’ (Burman, 2006) in early years pedagogy and in wider public discourses. Burman argues that ‘the emotional turn imports a fundamental ambiguity between care and control which… thus may be less cosy and friendly than is usually assumed’ (p. 316). Our research begins to show how that ambiguity between care and control is played out in the fine grain of interactions between adults and children in the classroom.

The discourse of normal development continues to ‘discipline’ notions of appropriate behaviour, obliging early years practitioners to calibrate children in terms of their approximation to the normal child. Indeed the disciplinary force of the discourse of development has been intensified by the introduction of audit procedures and base-line assessment in schools, and the specification of age-related statements and goals within the Early Years Foundation Stage⁸ and the Every Child Matters framework.⁹ It is, we suggest, becoming harder for teachers to adopt a reflective, ‘wait and see’ approach with children whose behaviour is starting to raise questions. Early interventions, in the form of IEPs (individual education plans), key workers, behaviour targets, and special groupings were common across all the schools. These provide additional support and resources for children and staff, but also intensify communal perceptions of ‘difference’ from assumed norms.

It is difficult to make practical suggestions for interrupting the production of negative reputations, since the problematic does not only operate at a personal or individual level, but is affected by deep-seated discourses of child development and early years practice. We have seen that problem behaviour is a product of educational aims that are in other ways productive: the imperative to form children into a collectivity; the inculcation of respect and tolerance; the desire to be responsive to children's differing circumstances and dispositions. The practices that produce reputation would be difficult to change through self-reflection, and it is important to emphasise that the teachers who took part in the project were caring and reflective about children's well-being and educational prospects. We saw many instances of generosity of spirit, affection and dedication in teachers’ and other adults’ interactions with children. The processes that produce winners and losers in the reputational stakes happen behind the backs and the best intentions of the individuals involved.

We might however offer some tentative suggestions. Firstly, professionals might consider trying not to intervene too early with explanations and ‘solutions’ for children who are beginning to emerge as a problem. It might also be worthwhile to reflect on the developmental maps, tacit and explicit, that underpin early years professional practice, and how these may generate ‘deficit’ views of some children, parents and families. This will of course be a difficult undertaking, given that teachers are held publicly accountable through inspection reports and league tables for children who are not seen to be operating within ‘normal’ tolerances of behaviour.

The public nature of discipline also merits consideration. Are there ways of achieving the goal of forming children into a collectivity, and maintaining and monitoring learning and attainment, that would rely less on the public administration of praise and reprimands? This would involve either reducing the overall amount of whole-class teaching (an argument already being made on the grounds of learning: Myhill, 2006; Wyse et al., 2008), or a radical reconfiguration of whole-class pedagogy. Whole-class teaching under contemporary conditions has particular rigours for very young pupils, as it submits them to extreme body discipline, requiring them to sit erect and immobile—i.e., ‘properly’—on the floor for substantial amounts of time.

We also wonder, perhaps controversially, whether it is time to contemplate trying to reduce the emotional quotient of classroom life. Boler (1999) notes that it is difficult to argue against arguments for tolerance, fairness, kindness, sharing, etc, but adds that it is also important to know what is done in the name of such values, and how they might be promoted by other means.

Perhaps the most intransigent problem, for educators and researchers alike, is the strength of received notions of ‘the child’. It is difficult to see children outside the frames provided by normal development, professional practice and liberal humanist views of the proper person. Can we free ourselves up so that we might see and think children otherwise? In follow-up work, we have been pursuing methodologies for awakening the indolence of our own ethnographic regard (Jones et al., 2010a; MacLure et al., 2010b). This has included making a film that disperses excerpts from the classroom ‘data’ amongst many other images, sounds and text fragments (MacLure et al., 2010a). The aim is to ‘animate’ adult responses to children and behaviour, and block easy exit to judgement and received wisdom, our own included.

Some such shake-up is needed, we suggest, in order to grasp how some key principles of early years education continuously produce unintended consequences in terms of behaviour and reputation. It is worth emphasising, in conclusion, that the discourse processes and interactional strategies that were identified in this research were in many respects similar across all sites, even though the four project schools were located in very different catchment areas. It is especially significant that one of the schools, Chesterfield, operates an explicitly innovative organisation involving ‘free flow’ and pupil choice of activities. Yet during plenary sessions (whole-class lessons, collective worship, etc) and teacher-led group activities, the interactional demands and the disciplinary strategies used by staff were indistinguishable from those in the other schools. This suggests that interactions with, and perceptions of reception-age children are indeed regulated by deep-seated assumptions and discourses that may over-ride differences of organisation, ethos, professional commitment and personal conviction.