Engaging with Wellbeing through Restorative Justice Practices: insights from staff at a pupil referral unit

(1)

Engaging with Wellbeing

through Restorative Justice

Practices:

insights from staff at a pupil referral unit

Anthony Ten Jet Foei – 11692359

Master Thesis Political Science

Track: Public Policy and Governance

Research Project: Protracted Violence, Mixed-Emotions and The

Search for Wellbeing

Supervisor: Dr Nel Vandekerckhove

Second Reader: Dr Polly Pallister-Wilkins

(2)

4.2.3. Reflexivity ... 45 4.2.4. Learner Agency ... 47 4.2.5. Judgement ... 49 4.2.6. Understanding... 51 4.2.7. Flexibility ... 52 4.3. Wellbeing ... 53 4.3.1. Building Relationships ... 54 4.3.2. Environment ... 57 4.3.3. Intersectionality ... 58 4.3.4. Communication ... 59 5. Conclusion ... 63 6. List of References ... 68

(4)

(5)

1. Introduction

The school acts an institution that serves a crucial function of socialising the youngest members of society. The environment is tailored to education both in terms of knowledge and behaviour. In this thesis I will discuss how students are disciplined, how ‘inappropriate behaviour’ is dealt with and how this affects the relationship between pupil and teacher. In doing so I will examine two distinct modes of behaviour management. Punitive justice refers to discipline where consequences for misbehaviour are intended to deter wrongdoing (Hopkins, 2002, p.144). Conversely, restorative justice emphasises repairing harm to relationships, involving all parties concerned with an incident who will together find an acceptable solution (Ibid.).

Punitive justice and strict forms of discipline are considered the traditional means of instruction in schools (Foucault, 1977, p.170). It generally relies on a teacher’s authority to give instructions (Tirri and Puolimatka, 2000). However, when confrontation occurs in the classroom behaviour management can result in exclusion of students. This can have consequences such as resentment or retaliation (Graham, 2018, p.1253).

In contrast a restorative approach focuses on repairing harms to relationships, engaging the school and the surrounding community, and providing agency to students who participate in the restorative process (Hopkins, 2002). The approach represents a potential paradigm shift with respect to discipline in education (Idem. 2002, p.144) in response to similar practice in the criminal justice system (McCluskey, Lloyd, Kane, et al., 2008, p.405).

Restorative justice is often presented as an alternative to the punitive approach (e.g. Suvall, 2009, pp.547–548), creating a binary opposition. However, moving away from focusing on ‘justice’ and towards practice, can provide an understanding of restorativity that does more than police unacceptable behaviour (McCluskey, Lloyd, Stead, et al., 2008, pp.212–213).

I will be examining the attitudes of staff in a pupil referral unit (PRU), a school that educates students who for one reason or another have been deemed as unfit for mainstream education. A pupil referral unit therefore educates students who are some of the most vulnerable in society and acts as extreme case of education provision in England. The purpose of this research is to answer the question ‘what meanings are conveyed by the use and discussion of Shine College’s discipline policy?’ Discussing the policy and two methods of

(6)

behaviour management reveals several themes relating to systemic influences and the location of the school in the educational landscape.

Drawing from a diverse body of literature I will examine disciplinary action in schools by combining concepts such as authority, exclusion, inclusion, and belonging. This will allow for a more holistic analysis of discipline in schools than each concept does individually. Coupled with a conceptualisation of restorative justice centring on social relationships, reveals a process that engages with the multiple dimensions of wellbeing (White, 2010). Restorative justice can then be used as a theoretical lens, and method of practice, for school staff to engage with wellbeing.

The results of this research indicate that there is considerable systemic pressure on students who are disciplined with punitive measures, but also pressure on institutions on how to enact discipline. Restorative practices can in part address these systemic pressures through flexibility and communication.

Moreover, policy implications arise concerning education for students in PRUs and their reintegration in the mainstream. While enhanced communication and alignment between education providers can remove artificial distinctions concerning the nature of education provision. Within institutions, a flexible approach to behaviour management, and specific teacher training, may facilitate restorative processes and engagement with student wellbeing. While placing the systemic influence of punitive authority in focus can address issues of inequality and systemic racism in English education.

This thesis is divided into three chapters, first a discussion of theory concerning wellbeing, discipline, and restorative practice in schools. Then a discussion of methodology. Data analysis expands on the views of staff at the PRU, and the themes emerging from interviews. The conclusion brings the discussion to a close highlighting the policy implications. I will therefore turn to the state of the debate concerning discipline and wellbeing in schools.

(7)

2. Theoretical Framework

In this chapter I will begin by providing an overview of the education system in the UK and the context of pupil referral units providing education outside of the mainstream. I will then discuss the concept of wellbeing from various perspectives and its relation to the educational setting. After, I will briefly look at the philosophical underpinnings of education and the nature of the school as a social institution. I will then consider two forms of behaviour management in schools, punitive and restorative justice, which are often presented as opposite ends of a spectrum (Suvall, 2009, pp.547–548).

However, in practice there will often be a mix of styles depending on the specific characteristics of the school, circumstances surrounding the behaviour, and personalities of the individuals involved. Placing the practices in focus, with a focus on how they engage with wellbeing allows a conceptualisation of restorativity that does more than delineate between acceptable and unacceptable behaviour (McCluskey, Lloyd, Stead, et al., 2008, pp.212–213). I should also note that when discussing discipline, I may be referring to restorative justice and I will use the terms discipline and behaviour management interchangeably.

In discussing the consequences of punitive justice, I examine the concept of authority before considering how exclusion is used as a punishment. Then, looking at restorative justice, I will discuss more democratic forms of authority and teaching, before considering inclusion and belonging. The various bodies of literature I will discuss all concern the educational setting and use abstract concepts such as belonging. The purpose of my discussion is to demonstrate the various perspectives on the ideas but also to link them together. Drawing from diverse sources allows me to bring together concepts that are academically separated but in practice will concurrently influence students and staff alike.

I will also discuss the importance of social relations for different styles of behaviour management in school and how these in turn can be considered a measure of wellbeing. Before I treat these more abstract aspects of the theoretical framework, I will provide a very brief overview of the context in which my research exists, the education system in the UK.

(8)

2.1. Education in the UK

2.1.1. A Brief Contextualisation

The first point I should note is that the UK government has devolved the jurisdiction over education to the constituent nations, excluding England. Therefore, although this section is titled Education in the UK, I will only be discussing educational provision in England and the specific characteristics may not apply to the whole of the UK. With that said, the statutory basis for education provision in the England lies largely in the Education Reform Act (ERA) 1988, having been built on the foundations of the earlier Education Act 1944 (Harland and Heppenstall, 2011, pp.482–483). The changes introduced by the ERA 1988 began the shift towards performance and inspections fuelling competition between schools. Subsequent legislation has created an environment where “schools operate in a quasi-competitive market, with competition between schools for pupils who in turn have a choice of the school they wish to attend” (Harland and Heppenstall, 2011, p.483).

The term quasi-competitive market refers to the role of government finance and regulation in the system of competition amongst educational providers (Lucas and Mace 1999, cited in Lucas and Crowther, 2016, p.588). Lucas and Crowther (2016, p.588) argue that a free market does not exist in the context of further education in the England because “there was no free market in education as the relationship was not fundamentally determined by price and the funding was controlled by the state”. The term further education (FE) is used to mean education after secondary school that is not an undergraduate degree, thus including A-Levels and technical qualifications (UK government, 2014).

Criticisms were made against 1944 Education Act. For example a lack of clarity regarding specific frameworks for the structure of new schools allowed for a multitude of school types, each with “different characteristics such as admission policies” (Harland and Heppenstall, 2011, p.483).

But what are the consequences of this quasi-market framework? According to Hicks (2015, p.328), the arrangement can produce both inequality of education as well as increased satisfaction with the school system. From an institutional perspective, the dominating operating logic of the marketised system has shifted the focus of FE colleges from education provision to financial considerations (Lucas and Crowther, 2016, pp.588–589). The marketised

(9)

logic may also block the use of other logics or discourses such as teaching and learning, professionalism, or what should be taught, meaning that stakeholders in the system lack a shared understanding of these issues (Idem. 2016, p.584). However, Ball (2009, p.83) argues that the privatised system allows for localised logics such as ‘partnership’ to emerge within the dominant operating logic. And that connecting schools with private enterprise is “part of a process of enabling organisations and their actors to think about themselves and what they do, differently” (Idem. 2009, pp.96–97).

In terms of assessment practices, the marketisation of the English education system affects the social student-teacher relationship in primary schools by commoditising pupils’ results (Pratt, 2016). While Torrance (2007, p.282) argues that assessment practices have

move[d] from assessment of learning, through . . . assessment for learning, to assessment as learning, where assessment procedures and practices come completely to dominate the learning experience, and ‘criteria compliance’ comes to replace ‘learning’.

Hence the quasi-competitive market for education in England has pervasive effects at the systemic, institutional, and interpersonal levels.

2.1.2 The Educational Landscape

According to the UK government website, there are roughly 30,000 schools in England, of multiple categories, providing education from ages 5-18 (UK government, n.d.). State schools provide free education for pupils between the ages of 5 and 16, receiving funding from local or national government (UK government, n.d.). State school can be of different types, with varying institutional structures, management characteristics, and organisational discretion in education delivery (Ibid. n.d.).

For example community schools are maintained and run by the local government authority (Ibid. n.d.). Meanwhile, foundation schools are also funded by local government but the board of governors holds the land on which the school operates and has control over admissions and staff employment (Ibid. n.d.). Academies and Free schools are operated by not-for-profit trusts and the local government has no involvement in the running of the school (Ibid. n.d.). These are all examples of state school structures. Of course there are also schools

(10)

that are not government funded and charge for attendance (UK government, n.d.), Eton College being a famous example.

This discussion serves to highlight that schools vary widely in their institutional character, how they are funded, and how they are managed. Furthermore, the level of state involvement in education seems to have been reduced since the enactment of the Education Act 2011. Local government are now required to consider proposals for new schools to be created as academies, and to inform the Secretary of State for Education of such proposals (Education Act, 2011 s37).

One type of educational provider central to my discussion is for students who have been turned away from ‘mainstream’ schools. So called ‘pupil referral units’ (PRU), Shine is an example of this type of school. However, a further compilation is that Shine College exists as a not-for-profit business, and does not technically fall under any of the categories of school discussed above. The consideration of the marketised education system then becomes even more salient with businesses engaged in educational provision for the most vulnerable students.

I will now briefly analyse literature on education in PRUs, as the context and dynamic within schools providing education for students outside the mainstream is worthy of specific attention.

2.1.3 Pupil Referral Units

A school considered a pupal referral unit is one that

[T]each[es] children who aren’t able to attend school and may not otherwise receive suitable education. This could be because they have a short- or long-term illness, have been excluded or are a new starter waiting for a mainstream school place. (UK government, n.d.)

The extent to which education for such pupils should be provided separately to the main student body is contested (Norwich, 2008, p.142). Considering that students of Black, Asian, mixed race, and other students of ethnic minority backgrounds (BAME) are overrepresented in PRUs compared to the percentage of the national population (Department for Education, 2016; Office for National Statistics, n.d.), I support this contention.

(11)

Within PRUs, Mainwaring and Hallam (2010, p.162) argue that students are less likely “to be able to generative a positive possible self, [and] had fewer strategic plans for attaining their positive possible self”. The authors argue that students with previous negative experiences may bring a focus on possible negative futures, without positive self-images in the same context as a counterbalance (Idem. 2010, p.166). While Solomon and Rogers (2001, p.344) argue that the tendency of PRU staff of aiming to raise self-esteem generally is ineffective due to self-esteem being derived from context specific situations. Hart (2013, p.206) presents an argument for what can benefit students in PRUs, suggesting that attachment relationships, support from adults, and personalised learning were crucial for creating an environment where pupil’s needs could be met and engagement with learning encouraged. Hart (Idem. 2013, p.208) also offers resilience as a key tool to achieving positive outcomes.

Other researchers also find that positive relationships with teachers and personalised learning are desired by students at PRUs and are beneficial to their outcomes (Michael and Frederickson, 2013, p.481; Lawrence, 2011, p.224). Michael and Frederickson (2013, p.418) identify extracurricular activities and support from peers and family as motivating factors for students at PRUs. While Lawrence (2011, p.224) argues that communication and a close working relationship between PRUs and mainstream schools, holistic decision making, and a multi-agency approach greatly benefit students reintegrating into the mainstream. In discussing the implications of their research Michael and Frederickson (2013, p.420) argue that changing the learning environment by taking care to build relationships and introducing flexible, personalised learning can prove beneficial. While practitioners could also consider involving students in planning, implementing, and monitoring their own learning (Ibid.). The latter point I will return to below when discussing democratic education. Discussing the sense of self of teachers at a PRU, Farouk (2014, p.28) suggests that “self-understanding has a determining influence on [teachers’] approach to education and how they teach their students”.

Thus, the provision of education in PRUs is a complex task, existing in England’s already complex environment of marketised education. The literature shows that relationships, a positive atmosphere, and self-image are crucial to students at PRUs. These aspects can all be considered part of a framework of wellbeing. To which I will now turn.

(12)

2.2. Wellbeing in Education

2.2.1. What is Wellbeing?

White (2010, p.163) distinguishes three dimensions of wellbeing; a material, a social, and a subjective. Briefly summarising each dimension, the material relates to aspects such as income levels and employment. The social dimension includes interpersonal relationships, and relationships with the state. Finally, the subjective dimension includes perceptions concerning the previous two dimensions. While White considers education part of the “humandimension” – her terminology is somewhat inconsistent but it equates with the subjective dimension – I consider education to be part of the material dimension since levels of achievement can be used as a resource, for example to gain access to higher paying jobs. However, the distinction may be irrelevant as White (Idem. 2010, p.165) stresses that the three dimensions are interrelated with wellbeing being the process of interaction between them.

White (Idem. 2010, pp.164, 168) also emphasises that relations are central to wellbeing and argues for using a notion of collective wellbeing to address concerns that the concept could be co-opted by individualism or political implications of its application (Idem. 2010, pp.166–167). Meanwhile Smith and Reid (2018, pp.817–819) offer a similar view, conceptualising wellbeing as a continually evolving phenomenon that exists in a particular place and time and dependent on a person’s relation to their environment. Both articles treat relationality as central to the concept of wellbeing, which is considered a process involving the space and time a person occupies. This focus on relationality and interaction between various aspects of the environment will be central to my analysis, rather than just focusing on one person’s feelings.

2.2.2 How is Wellbeing Treated in an Educational Context?

Of note, authors writing about wellbeing in an educational context argue that an emphasis solely on subjective wellbeing is lacking, as it may be used to justify liberalism or to build ‘one size fits all’ policies (Warnick, 2009). Other authors turn to culture as an authority for wellbeing (White, 2007). Price and McCallum (2016, pp.5–6) present an encompassing definition that includes individual and collective elements relating to values, opportunities and

(13)

contexts. The authors also note links between wellbeing and achievement in education, stating that the two are closely related arguing that a holistic approach to wellbeing will enable educational attainment (Idem. 2016, pp.7–8). While other authors have argued that teachers meeting students’ emotional needs is a strong predictor of learning and motivation (Bainbridge Frymier and Houser, 2009, p.216).

This discussion of wellbeing demonstrates that the concept has many different interpretations, all attempting to capture the complexity of human experience. Relevant to this paper is the relationality of wellbeing, both between people and their environment, how this affects their education, and how their educational setting provides wellbeing. I shall mainly base my discussion of wellbeing in White’s (2010), Smith and Reid’s (2018) and Price and McCallum’s (2016) definitions. I will also draw on theories of exclusion and belonging as to consider the social element of wellbeing. Furthermore, formal education generally depends on a teacher student relationship, with the teacher being tasked to manage the behaviour of the students. I will therefore examine two opposing approaches to behaviour management, both of which generally argue that they provide students with elements of wellbeing.

But first, it is also important to discuss the purposes of education and the underlying philosophies, which I will now turn to.

2.3. Philosophies of Education

I will focus my discussion concerning the nature of education on Dewey’s arguments in Democracy and Education (2004) and Foucault’s Discipline and Punish (1977), while also drawing from Bourdieu’s Outline of a Theory of Practice (1977). By placing these sources in conversation with each other, I intend to link the notions of education as an action and as an institution. Both are generative of particular values, beliefs, and modes of being.

According to Dewey (2004, pp.3–4) education is necessary for social life to reproduce. Without educating the young in the experiences and knowledge of the mature, civilisation could not last. Moreover Dewey argues that society exists and is reproduced ‘in communication’ (Idem. 2004, p.4 emphasis in original). It is this communication that passes on shared goals, values, and understanding, creating a community of like-minded individuals (Idem. 2004, pp.4–5).

(14)

Thus communication, which Dewey (Idem. 2004, p.7) states is educative by nature, also creates an environment which serves to instil the particular beliefs, dispositions, and behaviours in the receiver of communication (Idem. 2004, p.12). Bourdieu (1977, pp.78–79) makes a similar argument, proposing that the structure in which one exists defines the social conditions that produce certain practices and cognitive structures – what Bourdieu calls habitus – which then naturalise certain social conditions. This environment causes individuals to react in particular ways, which instils certain instincts in an individual. However, to be properly socialised the aims of individuals must be brought in line with the aims of a wider group (Dewey, 2004, pp.14–15). It is this process that is truly educative.

Dewey (Idem 2004, pp.20–21) states the environment has a mediating effect on education, with schools having been specifically designed with educative effect in mind. Foucault (1977, pp.135–138) also argues that schools are environments which are designed to transmit specific learning, where power exercised over bodies trains obedience. Both argue that the school environment serves to eliminate morals or behaviours that are undesirable (Dewey, 2004, p.22; Foucault, 1977, pp.200–201). If school environments are specifically designed to educate certain behaviours and morals into people, the modes of behaviour management in schools, and the values upon which they are based, are likely to be reproduced in the pupils who attend those institutions.

Different modes of behaviour management will focus on different behaviours, values, and be based on different relationships, meaning that attention is paid to different things. In short, the environment is different, which will have an effect on the way those raised in that environment behave. And within that environment “no detail is unimportant, but not so much for the meaning that it conceals within it as for the hold it provides for the power that wishes to seize it” (Foucault, 1977, p.140)

Foucault (Idem. 1977, p.170) also argues against the idea that strict discipline is the correct way of training, stating that disciplinary power is exercised in various ways over “docile bodies” and in doing so “makes’ individuals”. Thus, I will now examine this method of behaviour management, including the authority used in discipline as well as some consequences of the system I am calling punitive justice.

(15)

2.4. Traditional Forms of Discipline in Schools

2.4.1. Punitive Justice

Punitive justice can be considered the traditional mode of behaviour management in education settings (Ibid.). Foucault (Ibid.) writes that strict forms of discipline are intended to train and in doing so separate and emphasise multiple individualities. Thereby making individuals. However, Poplin et al. (2011) argue that firm and direct instruction generate respect and leads to academic achievement. Meanwhile Graham (2018) questions the effectiveness of strict discipline. While Teasly (2014, p.131) specifically links the approach with increased likelihood of encounters with the criminal justice system. And Morris (2005, p.45) links systems of strict discipline with reproductions of race and class inequality, arguing that “the emphasis on regulating students into embodying dominant modes of dress and comportment only seems to bolster perceptions of poor and minority youth as flawed in some way”.

Reminiscent of Dewey’s (2004, p.6) argument that education occurs in communication and is therefore by nature a collaborative exercise, Graham (2018, p.1246) uses the concept of ‘cumulative continuity’ which relates to the transactional or cumulative effects of an environment on the emotional state and development of a child. Noting the transactional nature of behaviour management, Graham (Idem. 2018, p.1247) argues that no-excuses discipline places the responsibility of poor behaviour on the student, although teachers may equally contribute to pupil’s misdeeds; “ineffective teaching contributes to classroom disruption, even when students are largely compliant and trying to participate in learning”. Macbeth (1991, p.306) also discusses the interactional nature of authority, stating that when students react to punishment, the interaction with the teacher may itself become the reason for admonishment rather than an assertion of authority. I will therefore turn to a discussion of how authority in education is constructed and used.

2.4.2. Authority

Tirri and Puolimatka (2000) distinguish between the authority to give orders (deontic authority) and knowledge of a particular subject (epistemic authority). They argue that the

(16)

ends of education rests on the existence of both of these types of authority (Idem. 2000, p.159). Elliot (2009) echoes this position in his findings that professional expertise contributes in large parts to a teacher’s authority, indicating that epistemic authority derives not only from subject knowledge but also from experience. Meanwhile Macbeth (1991) argues that a major component of teacher authority is located in practical action, in interactions with students. In a practice as transient as addressing a student, teachers may establish their authority with both the individual and other members of the cohort (Idem. 1991, p.288). Furthermore the practice of addressing a student may indicate not only the behaviour to be corrected by the student, but to project authority into the future by indicating consequences for non-compliance (Idem. 1991, pp.294–295).

Macbeth does not seem to share the concern about the misuse of authority that Tirri and Puolimatka have. But the authors similarly call for teacher training, indicating a shared desire for reflexivity on the part of educators concerning their powers over their students.

Thus, the no-excuses type discipline relies on authority that is used to direct and control the activities of pupils but does not recognise the impact of teacher’s own conduct. Nor does this style of behaviour management take responsibility for the power of the teacher. Graham (2018, p.1253) argues that students having sole responsibility for their behaviour, which may be influenced by the quality of teaching – a factor beyond their control – is a flaw underpinning the justification for the no-excuses approach. Thus, the no-excuses discipline oversimplifies the issue of student misbehaviour and may be counterproductive, perhaps causing resentment or retaliatory behaviours in the future which may lead to exclusion (Idem. 2018, pp.1253–1254).

Here we can return to Foucault’s (1977, p.170) claim that strict discipline “makes individuals”. The system of zero-tolerance focuses on the individual, thereby casting them in a role of accountability for their actions regardless of other influences. So, again following Foucault’s thread, it is crucial to consider unequal dynamics of power when considering student discipline, as well as potential consequences of such power. For example, exclusion is often used as a punishment and according to Tirri and Puolimatka (2000, p.157), punishments given by teachers are related to the most problematic conflicts in schools.

(17)

2.4.3 Exclusion

When discussing the power of teachers to affect the educational future of students Macrae and colleagues (2003) conceptualise social exclusion. They distinguish between exclusion focusing only on the excluded – weak exclusion – and exclusion that considers the mechanisms of power that can operate to exclude – strong exclusion. The authors note that a number of powerful gatekeeps, such as headteachers, have significant influence in the decision to exclude students but decisions are also affected by governmental policy which may penalise schools for high rates of poor behaviour (Idem. 2003, p.95). This latter issue relates to the above discussion of the quasi-competitive market for education, wherein schools are incentivised to perform well based on certain assessment criteria. The need to adhere to such criteria, supposedly to ensure quality, can significantly affect the educational future of individual students.

Moreover, while strict discipline may work well with some students, for those that it does not work, the system characteristically finds blame. The exclusion resulting from such punishments can result in harms including decreasing connectedness (White, 2010) and, particularly for adolescents, pain on a neurological level (Sapolsky, 2017, p.154). Psychological experiments have also demonstrated that social exclusion decreases self-regulation (Baumeister et al., 2005) and prosocial behaviour (Twenge et al., 2007). This discussion is also particularly relevant for the consequences for students who are continuously excluded from education, having an increased likelihood of engaging with the criminal justice system (Meiners, 2011; Roffey, 2013; Teasley, 2014).

From a wellbeing perspective then, the students who do not respond to strict discipline may have their material wellbeing affected by having decreased access to a certain level of education. Their social wellbeing is also affected as access to peer groups is cut off, as well as their acting in a less prosocial manner. These effects are the results of punishments given by teachers who have relatively unchecked power to do so, so called overt authority (Tirri and Puolimatka, 2000, p.159).

The above discussion highlights the notion that relations of power shaping the course of an individual’s future go far beyond the behaviour of one pupil, teacher, or school. Focusing only on behaviour that should be punished risks failing to consider the influence and attitudes

(18)

of institutional actors (Macrae et al., 2003, p.99). In other words, individualised responsibility makes those with the least power accountable for the failures of systemic influence.

Having discussed strict discipline and some of its consequences, I will now turn to a system of behaviour management that attempts to ensure strong relationships between individuals and communities.

2.5. An Alternative Approach

2.5.1. Restorative Justice

According to Hopkins (2002, p.144), restorative justice – which she conceptualises as a set of processes, a set of skills, and a philosophy – challenges conventional western ideas concerning justice, misbehaviour, and punishment. Hopkins proposes involving the school community, beyond just teachers and students, when taking a restorative approach. This may enable schools to demonstrate their restorative values, and create a community consciousness around the process, skills, and philosophy (Idem. 2002, p.145). Hopkins (Idem. 2002, p.148) calls this the “whole school approach”.

Other authors base their definition of restorative justice in a value set, also arguing that community-based relationship building is key (Teasley, 2014, p.132). Wearmouth et al. (2007, p.48) also call for involvement from the community in taking an restorative approach, including parents and carers. Meanwhile, Haft (2000, p.812) specifically mentions victim-offender mediation as central to restorative justice but likewise places emphasis on community involvement in the justice process. However this mediation process must be careful to avoid misplacing shame and to acknowledge the complexity of life and human emotions, as well as the influence group dynamics such as systemic racism or sexism (McCluskey, Lloyd, Stead, et al., 2008, pp.205–206). McClusky and colleagues (Idem. 2008, pp.212–213) also advocate for involving the whole school community, but stress a wider definition of restorativity that does not rely on the concept of justice as a way to sanction unacceptable behaviour. I will therefore be using the terms restorative justice, restorative practice, and restorative approach(es) interchangeably throughout.

The consensus on restorative justice is that it is a process that involves people, both those immediately affected by an incident and those who have an interest in the outcome.

(19)

Concerning the implementation of restorative justice in education, Suvall (2009, p.565) notes a tension when introducing the approach with more traditional punitive measures, both within a school and in wider society where norms of punitive criminal justice prevail. Current institutional norms are identified as a barrier to implementing restorative practices elsewhere, as is viewing restorative approaches as just one of many valid techniques (McCluskey, Lloyd, Kane, et al., 2008, pp.413–415).

Despite its difficulties, the restorative justice approach bases its ethos in relationships. This focus on interpersonal connections, as well as connections with a wider community, demonstrate an attention to White’s (2010) version of collective wellbeing that seems absent in the punitive justice approach. Indeed, McClusky et al. (2008, p.213) argue that restorative approaches:

admit the centrality of power relations and the complexity of social structures, offering the opportunity for all those involved to explore more deeply the relationship between the internal and external tensions of schools and to focus on how and where possible solutions might lie.

Thus, interactions between teachers and students progress differently using a restorative approach, requiring me to return to a discussion of authority in the classroom.

2.5.2. Shared Authority and Democratic Education

Harjunen’s (2011) article on pedagogical authority treats the concept as relational, with student consent allowing power sharing in the classroom that is conducive to teaching and learning. Harjunen (Idem. 2011, p.405) distinguishes between three types of interaction in the classroom, didactic, deontic, and pedagogical. The didactic and deontic interactions, relating roughly to communicating morals and duty respectively, would correspond to Dewey’s conception of education (Dewey, 2004, p.10). Harjunen (2011, pp.418–422) argues that students have various demands, which differ depending on the nature of the interaction. The author concludes that teachers’ authority is not complete without the consent of students, and reaching a power sharing arrangement can prove empowering for pupils (Idem. 2011, p.421). Like Macbeth (1991), Harjunen demands (2011, p.419) argues that students’ demands are made within interactions, but also notes that what teachers say and do can become a source of student.

(20)

Harjunen (Idem. 2011, p.420) argues that students’ demands and consent to authority correspond to more democratic ways of teaching, which according to Brubaker (2016, p.162), involve a clear pedagogical vision, where a social order is established that empowers learners to act on their own behalf and that of others. Considering the similar aims of both restorative justice and democratic teaching, both concerned with the sharing of power between students and teachers I will now briefly discuss democratic education.

Democratic education requires teachers to reflect on their own experiences as well as consider the type of teacher they wish to be (Ibid.). Furthermore, democratic teaching also requires a balance between cultural and institutional demands, while being comfortable with teaching in novel ways (Ibid.).

Brubaker’s conclusion seems to be built on what he has learned from his experience as a democratic teacher, meaning that the relations of power between teacher and student are more closely considered than in traditional pedagogical styles. Darling-Hammond (1996) also advocates for democratic modes of education, arguing for a coordination between teaching, research and policy, and the inclusion of the diverse voices and perspectives of teachers, students, parents and communities. Darling-Hammond (1996, p.14) argues for centring school operations on the needs of students, which can be achieved by school structures giving teachers more time with each student, displaying students’ work so as to demonstrate the values of the school, ensuring teacher collaboration focused on pupil’s learning, and structures for shared decision making (Idem. 1996, pp.13–14).

The democratic approach also seems consistent with restorative practices, although the two are not necessarily linked. However, a school with democratic teaching may have an ethos of student participation that enables easier adoption a whole school restorative approach. However, in order to do this, some measures must be taken to actively include people.

2.5.3. Inclusion and Belonging

Patton et al. (2006) discuss the effects of actively taking measures to promote inclusion in schools, their results showing that interventions to promote inclusion are associated with lower rates of health-risk behaviour (substance abuse, anti-social behaviour, and early initiation of sexual intercourse) compared to no intervention. However, the research showed

(21)

no difference in emotional problems (Idem. 2006, p.1585). The authors note limitations to their study, including potentially having a weak operationalisation of school social inclusion, but argue that their findings can be used to bring attention to the effect of the school social environment on health and learning (Idem. 2006, pp.1586–1587). Meanwhile Crouch et al. discuss school inclusion for children with disabilities. The authors argue that for new students with disabilities, positive relationships with teachers is associated with increased feelings of belonging while negative relationships were associated with decreased feelings of belonging (Crouch et al., 2014, pp.26–27). Crucially the authors found that students and teachers reported the same frequency of positive interactions, but students reported higher frequency of negative interactions than teachers (Idem. 2014, p.27).

Also discussing students with special educational needs (SEN) Norwich discusses some current thinking concerning where SEN students should be educated. Presenting a spectrum from ‘most separate’ to ‘most included’, Norwich (2008, pp.136–137) argues that government policy positions may indicate a need for an inclusive system, but do not necessarily mean that ordinary schools are inclusive for SEN students. Norwich then argues that to provide inclusive education for all students and their diverse needs, education practitioners must conceptualise inclusion involving multiple dimensions (Idem. 2008, pp.141–142). Norwich’s arguments are relevant to the extent that my case involves students with ‘behavioural difficulties’ whose education provision can be passed off to ‘special facilities’, whose remit is specifically for those considered ‘problem children’. However Crouch et al.’s (2014, p.27) study underscores the importance of belonging for all pupils.

Roffey (2013) also writes on the importance of a feeling of belonging for students in schools, and notes that current policy approaches emphasise schools that use strong discipline – which relies on exclusion as a tool for behaviour management – as ‘good’ and desirable. This way of thinking, which is closely aligned with the weak exclusion or strong exclusion discussed above (Macrae et al., 2003), can again locate the responsibility to adapt with students rather than considering institutional or policy approaches as an issue (Roffey, 2013, p.47). Roffey employs a useful conceptualisation of belonging based on social capital theory, distinguishing between exclusive and inclusive belonging. The former relies on a sense of belonging that non-group members may be excluded from, while the latter is characterised by more flexible and inclusive group membership (Roffey, 2013, pp.40–41).

(22)

Meanwhile, Ma (2003) argues that student self-esteem is a strong predictor of student belonging for students, also identifying school climate (including disciplinary climate) as a crucial factor for belonging. Conversely Cemalcilar’s (2010) research shows satisfaction with interpersonal relations as a stronger indicator of a sense of belonging than satisfaction with school environment. Conducting a meta-analysis of studies on belonging, Allen et. al (2018) identify 10 factors at the student level influencing belonging at school including parent, peer, and teacher support, emotional stability, and extra-curricular activities. The authors identify teacher support and positive personal characteristics as the most influential factors for a feeling of school belonging (Allen et al., 2018). Allen and Bowels (2012, p.113) note that there is a lack of consistent terminology in research on belonging, moreover there is a lack of attention paid to belonging in practice which may be due to a lack of a clear framework. Thus, belonging seems to have been identified as a crucial factor for academia, enough to devote considerable resources on research, yet a fragmented understanding of the concept may be a hinderance to schools paying the same attention to belonging. Regardless of vague conceptualisation, belonging appears to align with social and subjective wellbeing, meaning it is useful for my purposes of analysing school discipline.

2.6 Discipline and Wellbeing

My discussion on the two forms of discipline has served to highlight the complexities of providing education, in managing student behaviour, in having authority in the classroom, as well as for considering the wellbeing of the school community. Clearly educating does not solely depend on the transfer of information. Relational factors influence learners and teachers alike, which can impact on the quality of a classroom experience. As I argued above the attainment of education is part of the material dimension of wellbeing, considering that one’s salary is generally tied to one’s educational credentials. However, academic achievement also depends on one’s social connections with individual teachers and the wider school community (Price and McCallum, 2016, p.8).

However, I have demonstrated that concepts concerning education which are generally treated as separate, are often concurrently present when dealing with discipline and also relate to wellbeing. Authority is a relational notion that depends on the wellbeing of the relationship between students and teachers (on both individual and collective levels). Similarly, exclusion

(23)

can be a painful experience that causes decreased social connections, as well as affecting learning. The individual relationships involved in this process of exclusion are also influenced by wider policy positions and other relations of power. Therefore, concepts used when considering education relate back to wellbeing in multiple ways.

It is important also to bear in mind that influences derive from beyond just the individual teacher or school. The provision of education in the UK operates in a particular system, which induces particular pressures and inclinations. Using Bourdieu’s language, those who operate within a particular education system will have a particular habitus. Therefore, there will also be issues of collective wellbeing to consider.

Looking at behaviour management’s interactions with wellbeing allows us to think more closely about why a particular style is used and the consequences thereof. Just as Allen and Bowles (2012) state the concept of belonging may have limited influence in everyday practice due to a lack of a clear framework, the same may be true for wellbeing. But, due their impact on all three dimensions of wellbeing, methods of discipline can offer both an analytical tool and a practical guide with which to engage wellbeing. Punitive approaches influence perceptions, relationships, and potentially levels of attainment for those excluded (Graham, 2018, pp.1253–1254). It can also individualise responsibility to students, thus not engaging with collective wellbeing (Idem. 2018, p.1253) Restorative factors can promote collective bonds between student and school (Suvall, 2009, p.569) and could bridge gaps between teacher’s attribution of behaviour (Miller, 1995) and those of students (Miller et al., 2000). Methods of behaviour management are also part of the environment in which the process of wellbeing take place (Graham, 2018, pp.1245–1247). Considering a school’s discipline style can then provide a method to engage with the concept of wellbeing.

(24)

3. Methodology

This thesis was initially conceived as an entirely different project. Originally, I was to conduct an ethnographic investigation at a pupil referral unit in Oxford aiming to understand how students experienced living in the city and what the nature of their relationship with the police was like. In the interest of confidentiality, I have anonymised the name of the pupil referral unit and will refer to my case as Shine College, or Shine throughout. As a pupil referral unit in Oxford that deals mostly with students of an ethnic minority background, Shine offers unique insights into social and educational inequalities. The cohort of students live in one of the most famous cities in the world, and their daily lives are vastly different to that of the academics of the acclaimed university. These circumstances would provide the backdrop to conducting research on the experience of police prejudice in England, using intersectional and black feminist theory. Being situated in the middle of the school-to-prison pipeline (Meiners, 2011), research at Shine would provide unique perspectives on the experience of systemic violence in England. This research would have taken place in April 2020 and I had made arrangements with the school to sit in on lessons and conduct focus groups with students. These plans were in place until mid-March 2020 when the Coronavirus crisis escalated to the point that conducting field research in a school no longer seemed feasible.

The second conception of my research project was to conduct a policy analysis comparing schools whose approaches to student discipline were vastly opposing. I analysed the policies of various schools that used either the punitive or the restorative approaches. In the meantime I attempted to contact staff members at these schools so that I could interview them on their understanding of their school’s policies, using Yanow’s (2000) methodology for interpretative policy analysis.

In this comparative analysis I also intended to include Shine, as I consider its policy to be a mix of the punitive and restorative approaches. After many unsuccessful attempts to contact staff from other schools I realised this project would not enable to me to collect enough data to write my final thesis. So, I decided to focus on Shine as a case study.

Again, Shine is rather uniquely situated as a pupil referral unit in a poorer area of Oxford, whose methods incorporate restorative philosophy and practice, yet with a policy document largely reminiscent of a punitive system. This focus on a single institution could also be considered a case study, which Flyvbjerg (2011) argues is a sound methodology, producing

(25)

robust data. Flyvbjerg (Idem. 2011, p.302) argues that context-dependent knowledge is highly valuable and is the only type of knowledge available considering the nature of human experience. Moreover, case studies can fulfil generalisability criteria and are in fact critical for generating understanding a wider class of cases (Idem. 2011, pp.304–305). This justifies my choice of a single organisation within which to perform a policy analysis.

Conducting an interpretative policy analysis that takes into account “’local knowledge’ – the very mundane, expert understanding of and practical reasoning about local conditions derived from lived experience.” (Yanow, 2000, p.6). The benefits of this methodology are that it allows the researcher to analyse beliefs and feelings, which ultimately influence what is considered meaningful to the relevant stakeholders interacting with the policies (Ibid. 2000). This allows for open textured understanding of policy, with data constructed from participants’ experiences allowing for an appreciation of why a certain disciplinary approach may be preferred over another. Furthermore, Lipsky (2010) argues that policy is actually made within the discretionary actions of individuals delivering public services. The particularities of public service delivery mean that a more realistic understanding of how policy is enacted must be gained from the bottom up. Thus, when considering school policy positions, taking an interpretative analysis of the actions and perspectives of teachers, youth workers, and welfare officers is crucial. In my case, PRU staff opinions on behaviour management informs understanding of educational practice with vulnerable or disadvantaged pupils.

During the data-collection process I conducted semi-structured interviews with eight staff members of Shine Community College. An interview guide is attached in appendix.1. Before each interview began, I confirmed the participants’ consent to be recorded and to have their anonymised data used in my thesis. I also confirmed that they were aware of the research aims, how their data would be used, and where it would be stored. Interviews lasted between 30 minutes and an hour, owing to the semi-structured nature and follow up questions I asked. The responses to my questions form the majority of data collected. However, I have also analysed the discipline policy of Shine. Data therefore comes in the form of a policy document, opinions about that document, anecdotal data of behaviour management, and perceptions of two styles of behaviour management.

Interpretative policy analysis is based on assumptions that knowledge is acquired through subjective interpretation based on a person values, feelings, experiences (Yanow, 2000, p.8). Thus, actors in a policy situation will create meaning based on their own particular

(26)

qualities. The same is true for researchers (Ibid. 2000). This methodology analyses what policies mean, as well as how actions carrying out these policies communicate meaning (Idem. 2000, p.10). Yanow notes that interpretative policy analysis focuses on two types of ‘puzzle’. First, a puzzle of what a researcher expects to find and what they actually find. Second, a puzzle of inconsistency between policy documents and policy actions (Idem. 2000, pp.10–12). The former puzzle can help a researcher identify the various meanings that a policy holds for differing groups of stakeholders based on the differing views and feelings (Idem. 2000, pp.10– 11). The latter accounts for policy meanings that are created in action by relevant groups who implement or enact policies, thus demonstrating that a policy can have one meaning as written and another in practice (Idem. 2000, pp.11–12).

Yanow (Idem. 2000, p.14) states that the central question for interpretative policy analysts “How is the policy issue being framed by the various parties to the debate?”. My study began with this question informing my interview guide on how staff members understood their behaviour policy. However, I soon realised that focusing solely on the policy does not adequately provide understanding of the policy issue in my research. That of the tension between punitive and restorative justice. I therefore expanded my interview questions to include feelings concerning these two systems, as well as perceptions of the policy.

Yanow (Idem. 2000, p.17) states that policy analysis means focusing on the symbols that represent the more abstract elements of a policy, with symbols being something that represents something else and allows a uniting of individuals or groups who share meanings. Policy artifacts (language, objects, actions) can be considered concrete manifestations of an abstract value, belief, or meaning and thus have meanings embedded in them (Idem. 2000, p.18) while use of artifacts also creates meaning itself (Idem. 2000, p.19). And as actors will bring different interpretations to their reading or enacting of a policy, meaning is often found in these varying perspectives (Idem. 2000, pp.20–21). Yanow (Idem. 2000, p.23) calls these different groups interpretative communities, each of which may hold a different meaning of policy artifacts. I therefore focused on the meanings that staff held concerning the behaviour policy and different disciplinary styles, as well as the meanings conveyed by their actions.

The policy document I analysed was relatively short and vague, so the meaning of the text itself is relatively sparse. Because of this I focused more on the meaning of staff actions, as well as drawing on thematic analysis to code data from the opinions expressed to me. I identified recurring themes and compared the codes with the whole of the collected data, in

(27)

order to understand subjective meaning and approximate social reality (Vaismoradi et al., 2016, p.101). Thematic analysis operates on a more abstract level to content analysis (Idem. 2016, p.102), making it appropriate to research intangible concepts such as punitive justice and wellbeing.

Vaismoradi et al. (Idem. 2016, p.103) recommend four phases of research involving analysing, coding, reflecting, recoding, and relating themes to established knowledge before finalising a storyline. This allows for a high degree of familiarity with the data, as well as allowing for organising of conceptually similar themes (Idem. 2016, pp.103, 105). Then, distancing and reflecting on the data may allow new themes to come to light, or allow the researcher to verify their existing codes (Idem. 2016, p.106). This method of thematic analysis attempts to thoroughly engage with a data set to ensure systematic analysis, providing rigour, quality, and understanding (Idem. 2016, p.108).

As noted above, I have also drawn on interpretative policy analysis for my methodology as the themes identified amount to organisational communication. I therefore incorporated some of the above elements without my framework. I have organised the data into sub-themes which in turn relate to overarching themes. I will identify the overarching themes within each section.

In discussing the way staff manage behaviour, and their relationships with students many told me stories of interactions they have had with students. While the staff members never revealed names or any identifying information, I imagine that if the student in question found out their story had been told to a researcher without their knowledge or consent, they would feel hurt and lose trust in that staff member. My participants mention trust between staff and students and an issue to be specifically aware of thus although it is not direct harm to the student or the staff member, a relationship between the two could be conceivably damaged. Particularly considering the vulnerability of many students at Shine. In this type of research such discussions between researcher and participants are inevitable, however had I had access to the field I could have built rapport with the students and given then the opportunity to share their perspectives.

I have taken care to ensure the privacy of my participants, giving them aliases in my discussion. Using their initials could have still identified them as their information is publicly available on the staff page of Shine’s website. This is also why I have anonymised the name of the college. Below is a table of the participants in my research.

(28)

Alias

Job Title

Danish CEO and Principal

Inaya Head of 6th_form

Salim Teacher

Joel Teacher

Makena Welfare Officer

Alexandra Welfare Support

Michael Youth Worker

Hanifa Administrative Worker

(29)

4. Data Analysis

In this chapter I will analyse data from the interviews I conducted with Shine College staff members. The chapter will progress along three themes. I will first analyse punitive justice, then restorative justice, before turning to wellbeing in education. I have organised the analysis in these sections in keeping with the construction of my theoretical framework. The themes I will be discussing emerged from the interview process but are of course influenced by prior theoretical reading and the construction of my questions.

4.1. Punitive Justice

I think that’s what’s failed all these kids in the first place (Inaya, head of 6th_form)

In this section I will discuss the views of Shine staff towards the system of punitive justice. The staff at Shine College were all generally in agreement concerning the use of punitive justice. For example, the above quote indicates a feeling that systemic issues have influenced the educational trajectory of students who attend Shine. The location of the college in the UK’s educational landscape certainly informs the views of staff, both in terms of taking circumstances into account and of a tension between punitive justice and its limits. Furthermore, when discussing punitive justice, many staff members would talk in relation to more restorative measures meaning my analysis of both systems will contain some overlap. Due to the hybrid nature of Shine’s discipline policy this may be expected. However, it is also indicative of how staff feel about punitive measures, that they were usually discussed with qualified support or outright rejection. I will also compare testimony from staff to the substantive policy document to highlight tension between policy and practice.

To begin, I will discuss the behaviour policy, how it is applied and how it is perceived by staff. I will then analyse the limits of punitive justice, before turning to the sub-theme of circumstance when applying a punishment. After, I will look at how authority is used by teachers when disciplining students, before turning to an organising theme of systemic influence.

(30)

4.1.1. The Behaviour Policy and Staff Perceptions

Shine’s (2017) discipline policy has a three-stage structure. Fig.1, taken from Shine’s discipline policy document, shows the process visually. The first level is based on an informal discussion between students and staff, with level two escalating to a one-day exclusion and writing a letter of apology. The final level, reserved for gross misconduct or regular contravention of the behaviour policy sees the student meeting with a disciplinary committee. This framework seems to allow for some measures of restorative practice, with informal discussion between teacher and students being the first response to an incident. However, with the introduction of formal discipline, the policy also employs punitive justice methods, with a relatively vague threshold of behaviour continually failing to reach ‘the required standard’. So, the policy as identified by a teacher, Joel is “to some extent halfway between a punitive and a restorative approach.” Alternatively it can be considered a pyramid approach, starting with more restorative methods and escalating severity of responses if unsuccessful (Suvall, 2009, pp.565–566)

(31)

However according to staff, in practice behaviour management does not closely reflect the policy document.

[Y]ou spoke about the policy. At Shine . . . this is something that . . . we’re always wrestling with in terms of trying to fine tune it to our students. Because… taking a behaviour policy, a standard behaviour policy let’s say from a secondary school or something like that, we realised just doesn’t work at Shine. And it doesn’t work with me as the person that I am as well actually. So I think currently . . . we have this behaviour policy . . . like a traffic light system. . . We have that on paper but the reality of it is- I think, is slightly different. (Salim, teacher)

This view on the influence of the behaviour policy was expressed by multiple staff members. As noted above, it is not unusual for practitioner’s actions to differ from an organisation’s official policy. Furthermore Lipsky (2010, p.13) would say that it is in this difference, in the exercise of discretion and autonomy, that policy is made. Considering the lack of influence that Shine’s policy document has on the daily lives of staff, it seems that policy is indeed made in individual situations. As one staff member stated that they had not been given an induction regarding the disciplinary process, it would seem that individual choices influence institutional policy to a greater extent than the formal document. It should also be noted that Shine’s discipline policy is in a process of review.

I think one thing to probably say is it’s kind of an evolving understanding. And I think where we’re trying to get to is a greater shift around relationships so more around how do we develop better relationships than how we manage negative behaviour. That’s a shift. And the reason why we’re trying to shift towards that place is because the question we’re trying to answer is, if we improve the relationship between the students or the staff and the student in question, would that then shift the behaviour? Rather than saying actually we need to focus on the negative behaviour specifically and throw out consequences. So you don’t turn up, therefore you’ll be excluded or something more straightforward. So there… the shift is around what are the relationships? And the reason we’ve done that is when we opened the sixth-form initially, we found that the learners we engaged were the learners that weren’t being engaged well at other institutions. The slightly more challenging learners or learners with different needs. And then when we were using some of our adult teachers, we found that we had lots of challenges because there was almost like a miscommunication between the young person and the teacher.

(32)

And maybe a lack of understanding of when a young person’s behaving in a negative manner, what does that really mean? Is it because they’re upset or is something deeper and that’s going on? (Danish, Principal and CEO)

Thus, the lack of influence of Shine’s current policy document, with its focus on discipline, can also be understood as shifting organisational priorities. Considering the location of Shine in England’s educational landscape, the focus on relationships could be the result of multiple influences. As Danish noted there is a hope that by focusing on relationships, behavioural issues will improve. This factor could itself be motivated by systemic circumstances as Shine is a pupil referral unit, meaning that the students who attend are often those who have already been excluded from mainstream schools. Thus, students having already faced punitive measures in the past, staff feel that it is inappropriate for them to be treated in the same way. Moreover, were Shine to employ strictly punitive discipline they would likely have the same results as mainstream schools, resulting in exclusion. Suffice it to say that Shine’s policy can be considered a response to a culture of punitive discipline in English schools.

Although Shine’s goal of a relationship-based policy is a progressing development, in practice there will inevitably be situations that staff must handle. I will therefore now turn to staff perceptions of the punitive approach and its limits.

4.1.2. Punitive Approach and Limits

Despite the presence of punitive measures in the behaviour policy, there was a divide between respondents. Some felt that punitive justice should be never be used, or only as a last resort, while others felt some need for a punitive approach. For Salim punitive justice was appropriate after learners had taken individual responsibility, depending on the particularities of the situation.

[O]nce they’ve taken some individual responsibility then it becomes- then they’re invested in that relationship. Then they’re invested in their own learning and their own progression . . . if I simplify it using carrot and stick sort of thing, then if you start applying the stick they’ll see the benefit. They’ll see that . . . once they’re invested and they come in an hour late, and then you hold them back. They’ll almost see it as in “yeah I need to do this because I’m the one that’s losing out”. The problem with the punitive approach is if you do- if you do it straight away . . .

(33)

and there’s no . . . responsibility from the individual learner. Then it’s just… Then it’s like a warden jailor situation. And no one wants to be in jail. (Salim, teacher)

Salim’s use of the warden/prisoner analogy is reminiscent of Foucault (1977), who argued that constant observation and discipline in public institutions such as schools are akin to jailors surveilling inmates. Another teacher, Joel, echoed Salim’s comments on individual responsibility by stating:

I try and get to a point where the students take responsibility for their own behaviour. In that my classes and how I run my lessons tends to be very much is “by the end of this lesson, this is what we have to do”. How we get there is largely learner driven. If however you’re constantly disrupting or are disrupting yourself, you’re still gonna have to complete that work. So, the longer you persist with any disruptive behaviour, the more likely it is you’re gonna have to take work home.

In both cases the teachers feel the need for some system of punishments, as well as students taking responsibility for themselves. However, they do not consider a strict punitive approach appropriate and Joel feels that punitive measures such as detention are not actually educative. Joel bases his desire for student-driven working as preparation for the workplace. A rationale reminiscent of Foucault’s (1977, p.138) discussion of producing trained docile bodies as a means of achieving economic utility and political obedience. Other respondents identified schooling as preparation for a career, thus Foucault’s argument may still hold value despite his own (Foucault, 2009, pp.48–49) qualification of the idea.

Defaulting to a punishment, particularly for students in a pupil referral unit, is not deemed acceptable by the respondents. And while their comments concerning pupils’ responsibility for learning stray close to the individualised responsibility of strict discipline discussed above (Graham, 2018; Macrae et al., 2003), more positivist research has indicated that use of punitive measures in classroom discipline can cause an inhibition of personal responsibility in students (Lewis, 2001). It is important to note, the respondents’ comments were also qualified with both teachers being against a purely punitive approach. This perspective, of a need for some punishment but without it being the first-and-only measure, is reflected in the discipline policy.

Engaging with Wellbeing through Restorative Justice Practices: insights from staff at a pupil referral unit