Student experiences of democratic education and the implications for social justice

Using data from a case-study school as a springboard, this article explores how enactments of democratic education might both problematise and illuminate new possibilities for the way we conceptualise social justice in education. Nancy Fraser’s tripartite framework of social justice is used to analyse in-depth interviews with students aged 14–16 from a democratic school in the United Kingdom. The article makes two key arguments: first, it highlights the interdependence of ‘recognition’ and ‘representation’ and, consequently, calls on mainstream policy and practice to make a substantive commitment to participatory democracy as part of the ‘inclusive education’ agenda. Second, it points to the tensions between ‘redistributive’ justice and other social justice aims which may be particularly stark in democratic education (and other progressive education) spaces. The article suggests that a strengthened relationship between democratic schools and research communities would offer a crucial contribution to collective critical reflection on social justice in education.


Introduction
Typically, in times [of crisis], one does look to grassroots movements -to people, often young people, who are willing to think out of the box, who don't have a big investment in existing power structures [. . .]. That is where I look for transformative visions and energies. [. . .] I think we're getting very good at critique and we're not so good at imagining alternatives. (Nancy Fraser, 2016) Social justice in education can be defined in myriad ways: from the recognition of marginalised people, voices and knowledges, to the equitable distribution of educational resources, opportunities and outcomes; from building relationships rooted in equality and partnership, to restructuring systems of access, assessment or governance. Regardless of what definition of 'social justice' one uses, academic literature is full of examples of mainstream education's failure to deliver it (e.g. Mills, 2012a, 2012b;Reay, 2012). When it comes to 'imagining alternatives', theoretical critique is immeasurably enriched by research on real-life practice (Cribb and Gewirtz, 2005;Francis and Mills, 2012a) -particularly that which takes place outside of 'existing [educational] power structures' (Fraser, 2016). Democratic education -and the schools and educational communities inspired by it -can be seen as one such example of alternative practice. This article analyses data from interviews with seven students from a 'democratic school' in the United Kingdom ('Harbourside' 1 ) in relation to Nancy Fraser's tripartite conceptualisation of social justice as comprising recognition, representation and redistribution. The article's main aim is to demonstrate how analysis of democratic schools like Harbourside might offer important insights into how social justice is best conceptualised and enacted in education -crucially, in ways that go beyond mainstream approaches.
The structure and scope of the article are as follows: in the section below, a definition of democratic education (and, connectedly, democratic schools) is presented and related to existing associated literature. The following section explains the methodology of the study on which this article is based. In the subsequent 'Findings' section, participant data are presented in relation to each component of Fraser's framework. I will argue that the data offer valuable insights into how Harbourside realises recognition and representation in ways that deviate from mainstream education practice. I will also argue that elements of Harbourside's approach -not least connected to its enactment of recognition -raise concerns about the school's capacity to fulfil redistributive justice aims. The 'Discussion' section builds on the findings to suggest that Harbourside's practice demonstrates the deep interrelationships between Fraser's components of justice in two core ways, which are not always accounted for in mainstream contexts. First, I will argue that pursuing recognition in the form of 'inclusive' learning cannot be separated from a substantive commitment to student representation and shared decision-making power. Second, I will argue that there are specific tensions arising between redistribution and other social justice aims in school contexts (tensions which may be particularly stark in progressive schools). I will argue that redistributive strategies based on blind adherence to 'conventional educational success' risk undermining other core elements of justice and that a critical balance must be struck, in line with Fraser's own calls for 'transformative' strategies of redistribution.

Democratic education and democratic schools
Democratic education can be broadly understood as educational approaches, systems or practices that centre democracy or democratic principles. However, there is considerable diversity in how 'democratic education' is interpreted in existing literature and in practice. For instance, Sant's contemporary theoretical review covers 377 articles utilising the concept in three distinct ways. Most focus on what Sant (2019) calls 'education within democracy' (p. 682) -competing stances on the desirable relationship between governments, (adult) citizens and education provision at a structural or policy-making level, for example, the debate over 'school choice' mechanisms. A second iteration is what Sant (2019) calls 'education for democracy' -a commitment to curriculum and pedagogical approaches that promote 'democratic character' (p. 681) -for example, the so-called 'citizenship education'. A final iteration is what Sant (2019) calls 'education through democracy' (p. 682) -the modelling of democratic processes and values in educational communities, primarily by 'involving the members of the community in the process of decision making ' (p. 682). It is this latter iteration of democratic education that best maps onto the practices of 'democratic schools'. Democratic school networks the European Democratic Education Community (EUDEC, n.d.) and the International Democratic Education Network (IDEN, n.d.) both suggest that democratic schools are characterised by this core commitment to participatory democracy, as well as other associated principles. These can be broadly summarised as follows: Schools that enact this interpretation of democratic education as participatory democratic governance (Sant, 2019) and the associated principles identified by EUDEC and IDEN above (henceforth 'democratic schools') are relatively rare; both EUDEC and IDEN maintain global 'directories' which put the number in the hundreds. A small handful of these schools are in the United Kingdom, perhaps the most famous example of which is AS Neill's (1960) Summerhill, founded in 1921 (see Vaughan et al., 2006). In the United Kingdom, democratic schooling has also been associated with the wave of progressive free schools founded in the 1960s and 1970s, such as Risinghill School (Berg, 1968) and The White Lion Street Free School (Wright, 1994) (both now closed), which were particularly striking for their non-fee-paying status. There are some extremely valuable existing studies on democratic schools (e.g. Apple and Beane, 1999;Baroutsis et al., 2015;Hope, 2012); nonetheless, democratic schools -and democratic education more broadly -remain surprisingly under-researched. It is particularly difficult to find research comprehensively examining the above iteration of democratic education in relation to theories of social justice. Many studies of democratic schools focus on specific learning outcomes such as 'intrinsic motivation' (Berg and Corpus, 2013) or 'cross-age interaction' (Gray and Feldman, 2004). Those which discuss social justice often do so within the context of a much broader analysis of 'alternative' schools (e.g. Mills and McGregor, 2014) and thus pay less attention to what democratic schools specifically offer to the debate. Much of the remaining well-known literature on democratic schools is written by practitioners in the form of in-depth descriptive case studies; while excellent guides, these are rarely mapped onto theoretical critique (e.g. Gribble, 1998;Neill, 1960;Shotton, 1993). One notable exception is the extensive research that has been carried out on a group of state-funded schools known as the 'Citizen Schools' in Porto Alegre, Brazil, which can be considered at least partially democratic and which are explicitly oriented around social justice goals (e.g. Gandin and Apple, 2002, 2004. However, analysis of these schools tends to focus on the localised challenges or efficacy of their practice, or on implications for concepts connected to social justice -such as teacher education or democracy -rather than being explicitly linked to social justice theorising more broadly. 2 There is a much more extensive body of literature exploring mainstream schools that have sought to bolster forms of 'student voice' (e.g. Mitra, 2003Mitra, , 2004SooHoo, 1993), including some which explicitly engages with social justice issues -for instance, by exploring the significance of forms of democratic participation for marginalised groups (e.g. McCowan, 2010;McMahon et al., 2012). However, such studies tend to focus on schools that are in most respects conventional but which have 'added in' initiatives to increase (certain) students' participation in (certain) aspects of institutional decisionmaking. Many commentators have noted the limitations of mainstream 'student voice' initiatives in engendering substantive democratic empowerment for students (e.g. Cook-Sather et al., 2014;Fielding, 2004b;McCowan, 2010). Often, this is attributed to issues around superficiality and tokenism (e.g. Maitles and Duchar, 2006 ;Whitty and Wisby, 2007) and/or dissonance with the broader institutional culture of mainstream schooling (Mills and McGregor, 2014); for instance, mainstream schooling tends to be based on forms of hierarchy and traditional authority (McCowan, 2010), as well as marketised priorities such as competition and attainment (Fielding, 2001), which are seen to clash with a commitment to substantive power-sharing and/or lead to the reduction in 'student voice' to a form of 'consumer feedback' (Fielding, 2001). Thus, the majority of case studies focusing on 'student voice' are not examining 'democratic education' in the way it is outlined above.
In summary, there is some research on democratic schools, and a growing body of research on mainstream schools committed to aspects of 'student voice'. However, there is relatively little research on spaces adopting a more radical definition of democratic education as 'education through democracy'. Given the apparent ongoing failures of mainstream education in supporting social justice, alternative paradigms such as this seem worthy of critical analysis, especially those that may help build both conceptual and practical clarity around what we are aiming for when we speak of realising social justice in education. Democratic schools may have a particularly important contribution to make in this regard -not least because they orient themselves around principles with deep relevance for social justice (such as participation and inclusion, recognition and equality). This article, drawing on data from a case-study school, offers one example of such a contribution.

Nancy Fraser's theory of social justice
The article is organised around a critical dialogue with Nancy Fraser's framework of social justice. Throughout, Fraser's framework is used to illuminate key social justice tensions in relation to both the theory and practice of democratic education. Fraser's (1997) framework began with two major components: 'redistribution' and 'recognition'. These relate to a common distinction between two dimensions of justice: justice as concerned with the way goods are distributed, and justice as concerned with how we treat one another (Gewirtz, 1998). Thus, the redistribution element of Fraser's framework is 'concern[ed] with ensuring that goods (material and otherwise) are distributed in a just manner' (Pratt-Adams et al., 2010: 58), while 'recognition' is about striving for 'a difference-friendly world' based on respect for diverse identities (Fraser, 2003: 7). Non-recognition can take many forms, including 'cultural domination' -being forced to acquiesce to dominant norms that are 'alien and/or hostile to one's own' -or 'being routinely maligned or disparaged', either in the public sphere or 'in everyday life interactions' (Fraser, 2003: 13). Non-recognition could be applied on an individual level -that is, the failure to recognise the needs or identity of a specific person. However, calls for 'recognition' are generally framed as a critique of liberal individualism (Fraser, 2003), by recognising the pervasive influence of -and thus the importance of deconstructinggroup-level injustices (Fraser, 2003;Wang, 2016;Young, 1990). Fraser's (2008) later work introduced a third component of justice -'representation' -which relates to the extent to which people are involved in decision-making processes governing their community or society. Fraser's (2008) reason for this later addition is that although redistribution and recognition affect a person's capacity to engage meaningfully in governance, neither alone can ensure substantive political representation.
Fraser's framework is neither uncontested nor unrivalled. However, it is commonly applied in contemporary educational research (e.g. see Baroutsis et al., 2015;Keddie, 2012;McCluskey et al., 2015;Wang, 2016) and is widely acclaimed for its comprehensiveness (Keddie, 2012). A primary reason for its use in this article is that the three principles of 'democratic education' identified earlier have an uncanny overlap with Fraser's framework: participatory structures of governance and a 'democratic ethos' of respect and inclusivity relate strongly to Fraser's concepts of representation and recognition, respectively. Conversely, and as explored later, progressive education (including democratic education) has been routinely criticised for a perceived failure to consider the requirements of economic redistribution. In the 'Findings' section, empirical data from a case-study school are critically analysed in relation to each component of Fraser's framework, and implications for theory and practice are explored.

The research site
Harbourside is a democratic secondary school composed of around 70 students aged 11-17 and 20 staff. The school's philosophy can be condensed into three principles that closely mirror the definition of democratic education outlined above: (1) participatory democratic decision-making structures; (2) an inclusive ethos oriented around equality and mutual respect; and (3) a commitment to giving each student a high degree of control over their own learning. Harbourside is a private day school; the majority of its places are funded by fees, and accordingly, most of its students are from relatively wealthy backgrounds. There are important questions to be asked about whether fee-paying schools can ever be described as 'democratic'. Nonetheless, it is worth noting that democratic schools in the United Kingdom are generally ineligible for full public funding because of their opposition to enacting the mainstream schooling practices (such as compulsory testing) on which such funding is still made conditional. Moreover, Harbourside is not academically selective and also receives a number of (state-funded) local authority referrals of students whose needs are not being met by local schools. Indeed, it is well-known for providing an educational alternative for young people who have been let down by mainstream approaches. Despite their problematic positioning, then, democratic schools like Harbourside may have important perspectives to offer to the debate on social justice in education, perhaps precisely because -to return to Fraser's (2016) words earlier -they offer a much-needed alternative to dominant educational 'power structures'.

The participants
The study on which this article draws focused on the experiences and insights of seven Harbourside students aged 14-16: Ash, Dylann, Fern, Gilgamesh, 3 Katy, Liam and Nancy. Participants ranged in their time at the school from a few months to a few years and had a variety of educational backgrounds (all but two had spent time in mainstream school(s) before coming to Harbourside). As a group, participants identified a range of complex educational needs and life circumstances. Many participants identified needs/ conditions that fall under the conventional umbrella of 'SEND' (Special Educational Needs and Disabilities), including autism spectrum conditions (Liam, Fern and Gilgamesh), dyslexia (Fern) and dyscalculia (Fern). Participants also identified a number of mental health conditions, such as obsessive compulsive disorder (Katy), anxiety (Gilgamesh and Nancy) and depression (Gilgamesh). The remaining participants -Dylann and Ash -were experiencing challenging life circumstances at the time of their interviews, which, although not related to specific needs or conditions, were nonetheless having a significant impact on their lives. Dylann was dealing with the sudden death of a close friend in a drug-related incident. Ash was experiencing substantial difficulties at home because of her mother's opposition to her exploration of alternative gender identities.
To some extent, the fact that the study's participants have a complex range of educational and emotional needs is incidental; the study's analysis and conclusions are hopefully relevant for a much wider group of people. However, in discussions of social justice in education, it is important to consider the experiences of groups who are particularly vulnerable to marginalisation (Keddie, 2012). The participant sample mirrors the predominantly White and middle-class demographic of the school, and consequently this study's data on two core forms of injustice (race-and class-based) are limited. However, young people with SEND or mental health conditions, or who identify as LGBTQ+, are examples of groups who face varying forms of injustice in education -particularly around non-/mis-recognition. Thus, I believe the specific experiences and insights of the study's participants add an important group-based justice component to the analysis.

Data collection and analysis
Data collection was conducted over the course of a full week spent at Harbourside. Semistructured interviews (lasting 60-90 minutes) were carried out with seven student participants aged 14-16, and observation notes were recorded based on virtually uninterrupted participation in lessons, democratic meetings, mealtimes and ad hoc conversations with students and staff. Prior to data collection, I spent a day at Harbourside explaining the research and inviting students to participate. Prospective participants were provided with an information sheet, and staff sent a follow-up email to parents/guardians/carers. To ensure a diverse sample, I personally approached three students who I believed -due to both positionality and experience -had particularly important contributions to make. Thus, sampling was partly purposive. Six students volunteered to take part during my recruitment visit and one during data collection. All participants signed a consent form, and those under 16 also provided signed parent/guardian/carer consent forms. Interviews were conducted in an on-site private space of each participant's choice (generally an empty classroom). The information sheet was verbally summarised at the start of each interview.
The study's original research question was: how have a diverse 4 range of students experienced being at a democratic school, and how do they view the implications for their empowerment? However, during analysis it became apparent that Fraser's social justice framework offered a useful lens through which to critically explore the school's philosophy and practice. Thus, the analytical focus of this article shifted to the question: how have a diverse range of students experienced being at a democratic school, and what might the implications be for social justice? Interview questions did not focus on 'justice in education' explicitly; however they were oriented around various strongly associated themes, such as relationships, wellbeing, structures of democratic decisionmaking and academic attainment. The interview was also designed to be flexible and open so as to enable participants to identify factors of most significance to them. For instance, participants were asked how they would define Harbourside's approach and what implications this had had for them personally. Participants were also asked whether they identified as having any specific 'needs' -educationally or emotionally -and, if so, the extent to which they felt Harbourside was meeting those needs.
My starting point for analysis was grounded theory. Data were thus analysed partly in real time: as interesting themes emerged, I amended interview questions and reoriented the focus of field-notes to draw out discourses and events which appeared to best illuminate Harbourside's 'basic social processes' (Charmaz, 2014: 34). Interview recordings were transcribed strictly verbatim. However, small changes were made to extracts included in the write-up (e.g. deletions of stammers and word-repetition) to protect the fluidity of participants' narratives. Major alterations are indicated by square-bracketed ellipses. Data were initially coded through handwritten annotations before being sorted into dominant themes through physically cutting and arranging sections of paper transcript. It was at this point that I noticed a potentially interesting convergence between the emerging themes and Fraser's theory of social justice. The fact that Fraser's theory was applied post hoc likely means that certain opportunities were lost during data collection. However, an advantage of this approach is that data collection was less influenced by a desire for a 'fit' with a preempted theoretical model. This allowed 'complications' to come to the fore which might have otherwise been missed. Following the decision to use Fraser's framework, the initial coding process was digitally replicated -this time emphasising the most relevant themes for the new theoretical focus.
It should be acknowledged that data collection took place in the context of an existing relationship with both Harbourside and democratic education more broadly. Two years prior to this study, I had undertaken a number of voluntary teaching placements at democratic schools, including Harbourside. The placement at Harbourside involved 6 weeks of working closely with a number of students, some of whom stayed in contact. Thus, four out of seven participants for this study already knew me relatively well. This existing connection with Harbourside, coupled with the fact that I attended a different democratic school for 2 years as a student and have maintained contact with democratic education networks since, meant that I was positioned as a partial 'insider'. This generated practical and ethical advantages as well as challenges. For instance, while democratic schools are often wary of outsiders (due to a historically mixed relationship with the media and the public) my connection to Harbourside and democratic education engendered trust from the outset. The school community was immeasurably accommodating throughout the data collection process; I was left to undertake the research as I saw fit and felt treated as part of the community. Another consequence of this was that the 'formality' of the interviews melted away relatively quickly, facilitating greater depth of exchange. My positionality also generated a weighty responsibility: I wanted to avoid contributing to harmful or inaccurate perceptions of Harbourside while nonetheless pursuing honest analysis. Yet my wish to contribute productively to democratic education through research is motivated in part by a desire for more critical debate by advocatesincluding reflection on the challenges generated by current approaches. Thus, while this article seeks in part to shed light on a school whose practice I believe has positive connotations for how we conceptualise and enact social justice in education, it also aims to grapple with some of the practical and theoretical tensions that its approach brings to the fore.

Findings
Recognition Fraser (2003) defines recognition as the striving for a 'difference-friendly world where assimilation to majority or dominant cultural norms is no[t] [. . .] the price of equal respect' (p. 7). This has been interpreted in relation to education as the creating of 'inclusive [. . .] learning environments' (Keddie, 2012: 267) -those which value and affirm the experiences and needs of marginalised groups, or people who are 'different' from dominant norms. Descriptions of the supportive and inclusive ethos of Harbourside were strongly foregrounded in participant narratives and can be seen to mirror this idea of recognition as based on mutually respectful relationships and learning environments that celebrate 'difference'. Ash, for instance, described how the school provided 'an open space' in which to explore her gender identity, explaining that 'at [Harbourside] you get this kind of opportunity just to be able to explore without [. . .] criticism'. This was in sharp contrast to home, where Ash has struggled with her mum's 'very traditional' attitudes towards gender. Though assigned 'male' at birth, Ash had been exploring alternative gender identities for around a year and uses female pronouns. She described how the school supported her in the wake of a disagreement with her mother when she began wearing make-up: Harbourside's inclusive ethos was also seen as manifesting in the form of acceptance and forgiveness of people displaying 'challenging' behaviours. For instance, when a community member breaks rules or behaves in ways which damage others' well-being, participants felt that the main priority of the community was to try to support that person: I think the whole philosophy is based around imperfection, in a way. Like it's all to do with learning through your mistakes. [. . .] I think that's why so many people who really struggle [in mainstream school] come here -because you're allowed to struggle here. (Nancy) Some participants suggested that Harbourside's acceptance of 'imperfection' is in part facilitated by its lack of preoccupation with academic outcomes (in sharp contrast to their perceptions of mainstream schools): The chance to fail, I think that's actually quite a large thing in Harbourside. [. . .] [Mainstream schools] don't support failure -they don't allow failure I guess, because it's all about getting the highest grades. (Gilgamesh) This decentring of traditional academic expectations was also evident in participants' descriptions of the school's emphasis on mental health. Harbourside's relaxed attitude to both school and lesson attendance was highlighted by most participants as particularly key in this regard. For instance, 3 months before our interview, one of Dylann's best friends, Jackson, died in a drug-related incident. Dylann explained how Harbourside actively sought to minimise any pressure in relation to attendance and academic progress, despite the fact that she was in the midst of GCSE (General Certificate of Secondary Education) preparation: 'the amount of support I've had [at Harbourside] is amazing. It's like being at home'. Dylann contrasted this with her brother's experience of being at the local comprehensive, which Jackson also attended, where Jackson's friends were told 4 days after his death that absence from lessons would no longer be tolerated. In a similar vein, Katy explained how she never felt she had to 'justify' her mental health needs, meaning she can be open and honest with the school, helping them -in turn -to know how best to support her: [A] lot of students who have really bad mental health are forced to go to school every day, and they have no say in that. [ A final aspect of Harbourside related to recognition is the nature of student-staff relationships, which participants described as positive and largely egalitarian. Just as Ash highlighted the support of staff as crucial in her exploration of her gender identity, Gilgamesh suggested that the overwhelming anxiety and depression he used to experience eased off 'partly because of the support of all of the amazing teachers', who 'just treat you as a human rather than a pupil'. In describing their relationships with staff, many participants invoked a 'family-like' or 'friendship' dynamic. Often, this was explicitly contrasted with negative experiences of -or perceptions of -student-staff relations in mainstream schools: There is no, like, hiding from the fact that they're adults and we're kids, for sure, but I think, yeah it's like family, it's like an elders respect thing. [. . .] And the respect is reciprocal. Which doesn't happen very much at other schools. (Nancy) Harbourside students feeling recognised -as Gilgamesh put it above, feeling treated as 'human' rather than part of some alternate (inferior) category of 'pupil' -can be reasonably construed as a form of group-based justice in and of itself. However, this intersects with other forms of group-based recognition implicit in participants' experiences, such as the school's affirmation of Ash's gender identity, participants' sense that their mental health needs were being prioritised, or descriptions of an ethos of support rather than disciplinary sanction for 'challenging' behaviours. Conversely, participants who spent time in mainstream schools prior to Harbourside cited their experiences as predominantly negative, describing persistent bullying, drug-use, panic attacks or repeatedly being placed in 'isolation'. The extent to which participants' experiences of 'non-recognition' in previous school contexts can be understood as explicitly caused by misrecognition or denigration of their specific educational, mental health or other needs is difficult to say. The important point, however, is that for the participants, Harbourside was somewhere they finally felt able to be themselves, rather than being forced to conform to conventional expectations around what learning -and, crucially, 'learners'should look like. As Nancy put it: [Y]ou're seen, like, as the individual that you are, so then people can [. . .] adapt to your learning style. And I think that's why it works for so many people, because it's not, like, this one set philosophy.
It is one of the paradoxes of group-based recognition that it sometimes demands transcending group-based assumptions and seeing people as equally valued individuals. Thus, for Nancy, recognition at Harbourside is based on an absence of homogenising assumptions -whether about students in general or particular groups of students -and a willingness to adapt to a diverse range of people and needs.

Representation
Despite the relatively recent push towards bolstering 'student voice' through measures ranging from fora for student feedback to 'citizenship education', most mainstream schools are 'policed by powerful neo-liberal discourses' oriented around authoritarianism, marketisation and standardisation, which 'do not sit comfortably with a democratic ethos' (Mills and McGregor, 2014: 58). Harbourside appears to tell a very different story: structural representation of student interests constitutes a central part of school life. Like many democratic schools, Harbourside's primary decision-making body is the weekly meeting at which the whole community decides on a wide range of issues. All students and staff have one vote, anyone can add to the agenda in advance, and discussion points range from the relatively mundane (e.g. a proposed switch from paper to reusable towels) to the serious or contentious (e.g. how Harbourside can contribute to climate activism, how to respond to a student behaving violently or how to respond to a feeling that older students are so distracted by exams that they are unable to contribute to the community). Only a few areas -such as safeguarding and finances -are determined by staff alone. In addition to the school meeting, there is an elected 'Council' -the primary conflict-resolution forum -composed of six to eight students and usually one staff member. The Council is non-punitive and discussion-based, drawing on principles of restorative justice.
In line with Dewey's (1916) claim that, to become a 'habit', democracy must be practised firsthand rather than taught in the abstract, Harbourside provides a space to experience democratic participation in tangible ways. This includes grappling with the challenges that democracy entails. Harbourside is not a community underpinned by an unqualified zeal for democratic process. Rather, it actively recognises both democracy's benefits and its tensions and tries to work with and through them. One 'democratic tension' highlighted by participants related to the way in which -even in allegedly egalitarian spaces -hierarchies still emerge. Gilgamesh pointed out that some individuals will always be looked up to, even if just by virtue of their confidence or personality, and suggested that this is exacerbated by norms around 'authority' in wider society: I think the biggest problem is definitely people with really quite deep-seated notions of who has the power, who has the influence. [ Participants described how these latent hierarchies sometimes threaten the democratic ethos of the school. For instance, Nancy explained that staff have greater control over certain areas of school life, such as curriculum development and safeguarding. She did not see this as problematic in itself; however, she felt that greater staff authority meant that 'lines can get blurred occasionally. And then that's when it starts spiraling into [being] non-democratic'. Both Nancy and other participants noted, however, that when this 'spiral' occurs, it generates a responsibility for students to mount a challenge to the prevailing distribution of power. What is crucial about Harbourside is that there are structurally embedded -and institutionally powerful -fora for recourse through which community members can challenge emerging inequities or concerns. This helps ensure that 'representation' is not something that is merely paid lip-service (e.g. in a school's Code of Conduct), or 'kettled off' in a space marred by weighty discrepancies in interpersonal or 'legislative' power (such as a principal's office or mainstream school council). Rather, students at Harbourside realise that they both need to and can try to redress the balance, and take action accordingly. For instance, Fern explained how: As Fielding (2004b) notes, in most mainstream schools 'there are no spaces, physical or metaphorical, where staff and students meet one another as equals, as genuine partners' (p. 309). Thus, attempts to value student representation 'however committed they may be, will not of themselves achieve their aspirations unless [. . .] [they] provide the organisational structures [. . .] to make their desired intentions a living reality' (Fielding, 2004a: 202). At Harbourside, the school meeting and the Council -and the space they provide for both discursive reflections on authority and the explicit power to make decisions and pass motions -constitute tangible structures of representation that empower students to find ways of tackling problems alongside staff.
At this juncture, it is worth expanding on the point above about 'latent hierarchies' by emphasising that participatory spaces existing is not sufficient to ensure everyone actually is represented. There is a rich body of critical work on 'student voice' pointing out that 'who is heard' in school settings is inevitably shaped by powerful currents of marginalisation, particularly along lines of race and class (Orner, 2014). That this was not explicitly raised in the dataset could be a result of methodology (e.g. question focus), but is also likely a reflection of the predominantly White, middle-class demographic of the school. Conceptualisations of social justice in education must evidently include such critiques if they are to meaningfully support the enactment of representation and avoid simply reproducing existing inequities in 'speaking and listening' (McLeod, 2011;Silva, 2001). Nonetheless, I would argue that if open participatory structures do not even exist in the first instance -if student participation is reduced to a handful of voices or enacted tokenistically -then we are even less likely to achieve representation in a substantive sense. Indeed, more traditional, limited forms of student participation are not only problematic in their own right, but potentially particularly damaging for already-marginalised students, not least because of evidence that who gets selected/elected to the elite position of student representation/council involvement is itself affected by privilege and positionality (Young and Jerome, 2020).

Redistribution
The final component of Fraser's justice framework is 'redistribution'. Redistribution can relate to many kinds of 'goods and resources' -not just economic capital; thus, those applying Fraser's framework to the field of education often link redistribution to the various 'educational resources' that schools provide (Wang, 2016), and that influence 'students' subsequent access to the labour market' (Keddie, 2012: 266). These could range from 'conventional educational capital', which might be understood as grades or qualifications, to broader iterations focused on skills and outcomes linked to economic empowerment but which are not well captured by current assessment practice. I assume that Harbourside actively pursues redistribution in relation to this broader iteration; the school's emphasis on recognition and representation arguably engenders outcomessuch as self-esteem, empathy and group problem-solving skills -that employers are increasingly calling for (see, for example, Bughin et al., 2018), but which it is widely suggested mainstream schooling is failing to inculcate (All-Party Parliamentary Group for Education (APPGfE), 2017; Damianidou and Phtiaka, 2016;Hutchings and Kazmi, 2015). However, due to limitations of space, this section only focuses on data that help assess Harbourside's redistributive impact in relation to 'conventional educational capital'. In a paper about challenging mainstream paradigms of education, such a narrow focus may seem unexpected. However, it is important for three reasons. First, conventional educational capital, however narrow, remains an important passport for 'access to material wealth' and socioeconomically privileged spaces (Francis and Mills, 2012a: 257) -particularly for socioeconomically marginalised students, who may not have other forms of capital (such as inherited wealth or family networks) to fall back on (Bourdieu and Passeron, 1990). Second and connectedly, progressive education (including democratic education) has been widely criticised for its lack of focus on conventional educational capital on the grounds that this damages opportunities for educational success, especially for working-class students. Although it is worth noting that Harbourside's demographic is predominantly middle-class -and therefore the impact of its 'goodsdistribution' is arguably not as high-stakes -this article seeks to consider the social justice implications of Harbourside's practice beyond its own community and, therefore, these critiques need to be met head on. Finally, and most importantly, a number of the study's participants raised concerns about what they perceived as Harbourside's lack of focus on conventional educational capital. It is important -particularly in a study promoting student representation -to take those concerns seriously.
Participants were clear that academic engagement is by no means debased or discouraged at Harbourside; many described feeling intellectually challenged and academically supported. However, as explored in the section on recognition, participants highlighted how conventional notions of academic success are often 'decentred'. Some felt this could be problematic for those who either want or need to 'achieve' in this way, but who can only do so with more conventional academic structure and/or teacher-led approaches -neither of which are foregrounded at Harbourside: Concerns about the implications for social justice of educational paradigms that decentre conventional academic achievement are not new. Since the influential 'Black Papers' (Cox and Dyson, 1971), there has been growing support for the idea that child-centred or 'progressive' approaches that allow students to 'opt out' of conventional academic engagement constitute a neglect of duty that perpetuates disadvantage for those who are already socioeconomically marginalised. We can relate this in social justice terms to what Fraser (1997) calls 'deprivation' -a version of 'maldistribution' summarised by Wang (2016) as when 'particular social groups [. . .] are denied adequate social goods' to participate fully in society (p. 330). The influence of such critiques is well-demonstrated by the growing prevalence of so-called 'No Excuses' schools, which cater almost exclusively to socioeconomically marginalised students (Cheng et al., 2017;Golann, 2015;Golann and Torres, 2018). 'No Excuses' (NE) schools (also sometimes called Knowledge is Power (KIP) schools) maintain that one of the most effective ways to break cycles of inequality is to equip disadvantaged students with educational capital that enables them to 'get ahead' -or, as Nancy puts it above, 'get to their next level'. Such schools see it as imperative that their students achieve academically and, ideally, go to elite universities (Casey Carter, 2000;Ellison, 2012) and often rely on rigid top-down disciplinary structures to achieve these aims (Golann and Torres 2018;Kerstetter, 2016). 5 They are thus very different in structure and ethos to a school like Harbourside.
While most Harbourside students are relatively socioeconomically privileged, and thus can arguably more easily 'take the risk' of rejecting conventional educational capital, not all Harbourside students are in this position and, in any case, the critique raises concerns for the broader social justice implications of the school's approach. These are explored in greater depth below.

Discussion
The findings reported above highlight the way Harbourside can be seen to actively foreground practices of recognition and representation as well as its problematic relationship with redistribution. These are helpful insights in their own right. However, the analysis also has important implications for how we conceptualise social justice in education. These are explored below.

Recognition and representation -A false divide?
As discussed earlier, Fraser introduced 'representation' relatively late on in her body of work. This was justified on the basis that 'misrepresentation can occur even in the absence of the [other] injustices' -that is, misrecognition and maldistribution -even though 'it is usually intertwined with them' (Fraser, 2008: 279). That is, a given social space could, in theory, both provide a person with resources and affirm their identity while nonetheless depriving them of proper political representation. Evidently, this distinction can be analytically useful. But, as with all theoretical devices, real life muddies the waters. The experiences of the study's participants strain against such neat bracketing, suggesting that recognition and representation are more deeply intertwined than Fraser's framework might suggest.
Participants clearly foregrounded the significance of the school's democratic processes for building strong interpersonal relationships based on equality and mutual respect. 6 For instance, it was suggested that the democratic meeting provides a space in which students see the school's commitment to student empowerment actually manifesting, which in turn gave them the confidence to speak up and engage in wider community life in the full knowledge that their needs and opinions will be taken seriously: I think [the meeting] really has given me this kind of spirit. It's kind of [. . .] given me courage and it's given me . . . like heart, I guess. And it's made me feel like my actions do matter. (Gilgamesh) Participants also pointed to how the democratic meeting helps to foster a general sense of empathy in the wider community: In short, participants seemed to suggest that, at Harbourside, inclusivity and democratic empowerment work together simultaneously and symbiotically; one cannot be realised without the other. Interestingly, in Fraser's earlier work -before she identified representation as an independent component of justice -representation was essentially treated as a 'version' of recognition. Fraser (2008) distinguished between two 'models' of recognition: an 'identity model' (p. 131) and a 'status model ' (p. 135). The 'identity model' aligns with the (common) interpretation of Fraser's notion of recognition which has been used in this article so far -that is, affirmation and acceptance of group/individual identities and associated needs. The 'status model', by contrast, emphasises the extent to which one is 'recognised' as an actor 'via the authoritative representational, communicative and interpretative practices of one's culture' (Fraser, 2008: 14). This could be understood in myriad ways, but readings of Fraser seem to view the 'status model' of recognition as emphasising power over attributes (e.g. see Keddie, 2012) -the role or authority one is assigned in a given context rather than simply the extent to which facets of one's identity are accepted or affirmed. Such status might manifest through informal or relational practices -such as language, day to day interaction and portrayals in media and cultural discourse -that render you either influential or marginal. But status might also be determined by formal power structures -for example, whether or not you are represented, directly or indirectly, on boards, in governments, committees, juries, institutional meetings and so on.
The latter model feels near-identical to what later became Fraser's third component of justice: representation. Given that, again, it is perfectly possible to imagine a situation in which someone's identity is affirmed in a relational sense, but in which they are given no 'status' in a (formal or informal) representational sense, it is understandable why Fraser eventually chose to cordon off representation as an independent component of justice. This does achieve some conceptual clarity. But it also obscures the way in which representation is deeply connected to recognition -as explored above. Fraser's earlier work suggests that feeling that you are treated by others in ways that reject your identity, or marginalise your needs, is a violation of recognitional justice in its own right. But even where this is absent -that is, even where you are treated with apparent respect and esteem -if you are excluded from the systems and practices in a given environment that determine, in a structural, institutional or 'official' sense who is represented, what gets communicated and what counts as true, good and desirable, then you are still facing a form of 'misrecognition'. Contra Fraser's (2008) later claims, this analysis implies that it does not make sense to argue that 'misrepresentation can occur even in the absence of the [other] injustices ' (p. 279). Rather, where misrepresentation occurs, misrecognition occurs also, because the absence of representation actually damages the extent to which recognition can be realised. Thus, although I agree with Fraser that there are analytical advantages in stopping short of collapsing recognition and representation into one another, the fact that mis-/non-representation can be meaningfully conceptualised (even by Fraser herself!) as a form of mis-/non-recognition highlights the extent to which these two components can be seen as not simply 'intertwined' (Fraser, 2008: 279), but as mutually reliant. Fraser's earlier work's ambiguity over the distinction between recognition and representation may actually have got closer to the truth of their lived relationship.
One might, at this juncture, field an accusation of semantic quibbling. Fraser maintains that all three components of her framework must be present for social justice to be realised. So perhaps exploring the nuances of the connections between the components adds little to the debate; if all three must be present anyway, why does it matter which labels we use? The problem is that, in reality, people and institutions make bold claims to aspects of justice without considering what is missing. It is common in education policy and discourse, for instance, to find powerful emotive language strongly connected to the idea of recognition -language around 'inclusive education' in particular -which does not accompany what must be seen (if we accept the above analysis) as a concomitant commitment to substantive, structural participatory democracy in schools. Thus, the popular, hopeful language of inclusion often obscures the deeply partial ways in which it is implemented in practice. By reconceptualising how we understand recognition, we can ensure it is not normatively acceptable to make claims about recognition in the sense of 'inclusive' school environments without also considering how such environments are contingent on institutional representation -on considering who actually gets to make the decisions and what formal accountability structures facilitate or prevent the governed from being 'in the room where it happens'.
Far from mere theoretical gymnastics, this analysis has rather radical connotations for education policy. It means that making 'nice' schools with an inclusive ethos, but where students are structurally excluded from decision-making fulfils neither recognition nor representation. It means that a school allegedly committed to recognitional justice, but which does not even attempt to treat students as equal partners in core decision-making processes is pursuing a very thin version of recognition indeed.

Redistribution and critiques of progressivism
In the 'Findings' section, I highlighted concerns -connecting participants' narratives to contemporary critiques of progressive education -about the decentring of conventional academic success in progressive schools and the broader implications of this for redistributive justice in education. However, there are powerful counterarguments to the definition of redistribution that such a critique relies upon. These are briefly explored below.
Undermining recognition and representation? Writing off Harbourside's approach in favour of a tighter focus on conventional educational capital raises further problems when it comes to promoting social justice in education. Fraser differentiates between two main 'strategies' for pursuing redistribution: (a) 'affirmative' and (b) 'transformative'. The former focuses on 'correcting inequitable outcomes [. . .] without disturbing the underlying framework that generates them' (Fraser, 1997: 23). In the case of education, this means promoting conventional 'academic success' without questioning that definition of 'success'; it is a 'surface'-level strategy (Fraser, 1995: 85) which says to young people: your best chance of empowerment is to win within the game as it stands. Such a strategy may seem admirably pragmatic, until we consider whether it is compatible with a broadly articulated vision of justice like Fraser's. Indeed, 'affirmative' redistributive strategies in education may actually risk undermining both recognition and representation. Fraser's (2008) work regularly grapples with tensions between 'distribution' and her other components of justice and many commentators have directly responded. For instance, Axel Honneth (2002) contests what he sees as 'Fraser's assumption' that processes of redistribution and recognition can be usefully divided into the spheres of 'material goods ' and 'identity', respectively (p. 54). Rather, he sees processes of redistribution -far from merely 'material' -as deeply implicated in practices of non-/recognition, arguing that ''conflicts over distribution [. . .] are always symbolic struggles over the legitimacy of [. . .] [particular] activities [and] attributes' (Honneth, 2002). That is, distributive processes, in normalising the proliferation of particular kinds of capital, help to determine what counts as 'good' or 'desirable' in education -what is, in short, afforded recognition. It is fairly widely accepted that current educational assessment/qualification-acquisition practice is underpinned by universalist assumptions of value that valorise a narrow range of skills and disciplines (Hutchings and Kazmi, 2015;Robinson and Aronica, 2015) and prioritise that which can be quantified by 'simple [. . .] categories of judgement' (Ball, 2003: 217). Accordingly, there is a wealth of evidence that current assessment approaches are damaging and ill-matched to fulfilling the needs and interests of a vast number of children and young people (Francis and Mills, 2012a;Hutchings and Kazmi, 2015;McNeil, 2000). Conventional notions of academic achievement too often stigmatise difference, rather than celebrating it. The upshot is that an 'affirmative' redistributive strategy -in the current educational paradigm at least -is mutually exclusive with a true 'politics of recognition'.
There are also, I believe, clear tensions between redistribution and representation, although these are less widely theorised. As Francis and Mills (2012a) argue, '[s] chools as they are currently run have a closer resemblance to authoritarian regimes than they do democracies, especially for those young people who do not conform easily to school expectations' (p. 264). This is not incidental, but part and parcel of an approach oriented around the pursuit of a narrow set of 'socioeconomically valuable' outcomes, leaving little room for challenge. Thus, when we stick to the non-critical redistribution of 'conventional educational capital', the possibilities for structural empowerment are undermined -whether that manifests as a narrowed scope for what a community can debate and decide on (to protect against decisions impinging on 'academic work') or, connectedly, less time and flexibility available in the school day to enable those deliberative processes to occur when needed and play out organically. Moreover, as was argued earlier, limiting representation has an adverse impact on the recognition potential of a school's practice. Thus, the curtailment of student agency precipitated by an affirmative redistributive strategy risks further exacerbating the erosion of school-based recognition.
The bottom line is that it is difficult to generate an ethos oriented around recognition and representation while also seeking to ensure that all students walk away from school armed with the same plethora of (currently) socioeconomically worthwhile educational capital. Because it is precisely that definition of 'worthwhile' which is contributing to so many young people feeling that they are worthless in school. And it is precisely the time and energy that 'academic achievement' demands which can too easily push other educational priorities -such as self-care or democratic process -to the sidelines. The result is too often that core elements of recognition and representation are sacrificed in the name of one conception of 'distributive justice'. These concerns are not merely hypothetical. Those schools (such as NE/KIP schools) that effectively promote an 'affirmative strategy' of redistribution have been widely criticised for relying on an authoritarian ethos that undermines both student agency and wellbeing (Ben-Porath, 2013;Golann, 2015). Claims that this approach is in 'disadvantaged' young people's long-term socioeconomic interest -that it does not matter that they miss out on recognition and representation now so long as they have economically powerful capital laterare arguably grounded in a form of utilitarianism that feels incompatible with the human rights and needs discourse embedded in most theories of social justice (e.g. Clément, 2018;Fraser, 1989;Hibbert, 2017).
'Transformative' approaches: An alternative way forward? Precisely because of the shortcomings of 'affirmative' redistributive strategies, Fraser (2008) calls for a reconceptualisation of redistribution based on 'transformation' (p. 28). Transformative strategies of redistribution seek to critically challenge the 'deep structures' (Fraser, 1995: 85) of value that define and distribute capital; in the case of education, this would mean questioning -and actively modelling alternatives to -what currently 'counts' as desirable educational capital. This does not mean ignoring all critiques -a concern that democratic schools' resistance to convention may inadvertently reproduce the economic marginalisation of those who cannot afford to reject it is a concern shared by many of their advocates (e.g. Apple and Beane, 1999). According to Fraser (2008), though, truly transformative approaches are characterised by balance. One way to envisage such balance in education might be as a 'conditional' commitment to the distribution of conventional capital -one that seeks to support students in learning to 'play the game' but that does so while (a) using critical pedagogical tools that challenge the parameters of what counts as valuable 'knowledge' (e.g. Freire, 1970;hooks 1994) and 'educational success' and (b) within limits, always drawing a line at the point where the strategic pursuit of conventional capital begins to undermine either recognition or student representation. It would thus seek to protect student wellbeing and empowerment in the here and now, as well as their future options in a world still wedded to traditional conceptions of educational capital. This strategy is not dissimilar from the strategies of Porto Alegre's democratic Citizen Schools described by Gandin and Apple (2004) as attempting to educate socioeconomically disadvantaged students in a way that provides them with 'better chances in the paid labor market' while also centring democratic empowerment and the critical deconstruction of 'elite knowledge ' (p. 195).
Such a 'conditional' commitment would look very different from NE/KIP schools; maintaining a commitment to recognition means continuing to take all students' priorities, needs and interests seriously -even if this detracts from their capacity to chase a string of top results. Substantively embedding representation means taking the collective decisions of the community seriously, and protecting time for debate and decision-making -again, even where this detracts from 'academic focus'. Authentic engagement with deconstructing conventional knowledge means avoiding an 'exam factory' model and taking time to explore other ways of 'knowing'. Thus, such an approach remains vulnerable to the critiques explored earlier; a refusal to pursue conventional academic success 'at all costs' means it still risks compromising certain forms of future socioeconomic empowerment. But what is protected is a commitment to recognising and representing young people -to active, critical questioning of educational truth-claims, and to pursuing a more humane and diverse form of educational community.
While such a balance sounds nice in theory, it is no small feat in practice. It is unsurprising that, for some students, Harbourside doesn't always get this balance right. Attempting transformative strategies of redistribution raises a number of complex and difficult-to-answer questions: how do we support young people who are in some way disadvantaged by the present system to learn how to navigate it, without inadvertently endorsing the damaging practices of the status quo? How do we take their needs and interests seriously without undermining their chances of 'conventional success'? Exploring these critical questions 7 is a crucial avenue for further research on democratic schools. As Fraser (2008) writes, 'the [. . .] dilemma is real' (p. 39) when it comes to balancing competing social justice demands. This is compounded by what Keddie (2012) describes as 'the lack of shared understanding about issues of justice within schools and amongst teachers' (p. 264). Clearer, collective understanding of how best to conceptualise justice may lead to more effective practice in pursuing it. This article has used Harbourside's practice as a springboard for making two main critical points about how we might better conceptualise social justice in education:

Conclusion
• • Recognition requires representation. Harbourside's inclusive ethos is to a large extent facilitated by its commitment to structural student empowerment. While it can make sense to distinguish between recognition and representation for the sake of analytical clarity, in real life, feeling that your voice matters and that you have substantive power to influence the community you are part of is integral to feeling recognised. The apparent practical interrelationship between recognition and representation demonstrated by this article suggests the need for greater nuance in the way we conceptualise social justice and, subsequently, policy discourse: widespread calls for such recognition-oriented practice as 'inclusive education' are incomplete unless they include serious, non-tokenistic commitments to student participation in governance. • • Redistribution sits in particularly uneasy relation to recognition and representation. Harbourside foregrounds recognition and representation in powerful ways, in part facilitated by what can be understood as a rejection of 'affirmative' redistributive strategies. The school is thus vulnerable to the criticism that its decentring of 'conventional educational capital' disadvantages young people (especially those who are socioeconomically marginalised). Having rejected 'affirmative' strategies of redistribution on the basis that they reinforce damaging status quo assumptions around educational 'value' and are subsequently incompatible with Fraser's other social justice components, this article has advocated a 'transformative' strategy. Such a strategy would seek to balance a 'conditional' commitment to the distribution of conventional educational capital with both the requirements of recognition and representation and a deconstructive approach to existing education truth-claims.
More critical dialogue on how we might conceptualise social justice in education in a way that achieves the challenging task of foregrounding recognition and representation, without sacrificing important aspects of redistribution, is crucial. This article has attempted to demonstrate the valuable contribution of democratic schools to such dialogue. It is important that we strengthen (currently scant) links between democratic schools and research communities; practitioners and students tend to be extremely busy with the day-to-day demands of educational spaces, while researchers have time and resources to identify tensions and collate existing solutions according to the priorities of those on the ground. This article has also suggested that insights from democratic schools are of critical relevance to the wider world of education policy and practice; research communities can help disseminate findings beyond philosophical silos, and help to foster a richer collective debate on how the complex -and sometimes competing -demands of social justice might be realised for all.