Why inclusive education falters: a Bernsteinian analysis

ABSTRACT

The inclusive education policy agenda has not made a significant impact on the global problem of educational exclusion. Explanations for this lack of impact include inadequate teacher education, lack of resources, negative attitudes, and a policy-practice gap. This paper takes a different turn and, using the concept of classification, argues that the challenge to achieve more inclusive education is more fundamental than has been previously articulated. Key tenets of the inclusive education agenda demand a weakening of the insulation between categories that are sustained and advanced by current marketised and standards-driven education systems. Inclusive schooling weakens spatial insulation, collaboration weakens professional insulation, transformability weakens ability insulation, intersectionality weakens identity insulation, and inclusive pedagogy weakens pedagogical insulation. When inclusive education is mapped onto strongly classified education systems, limited instantiations of inclusive education are inevitable, and difference and exclusion are re-inscribed. Change is possible if 1. Those advancing the inclusive education agenda acknowledge the identities and defences that classification constructs. 2. The workings of power that sustain insulation between categories in education are identified. 3. Counter-hegemonic action that weakens insulation and blurs boundaries is encouraged.

KEYWORDS:

• Inclusive education
• classification
• collaboration
• transformability
• intersectionality
• inclusive pedagogy

The problem: educational exclusion

Two hundred and fifty eight million children and youth aged between six and seventeen years, most of them girls, are not in school (UNESCO 2019). Children and young people with disabilities are disproportionately represented in the out-of-school count (UNESCO 2018a). Being in school, though, is no guarantee of learning. While school enrolment internationally is increasing, levels of learning are not (World Bank 2018). Many children and young people are not achieving grade appropriate curriculum outcomes, or, as framed by Morrow (2007), they have formal access to schooling but limited access to the knowledge goods of education. Poor teaching, inadequate infrastructure and resources and various other exogenous factors result in the ‘silent exclusion’ (Lewin 2009, 24) of many of the world’s children. Some children find themselves restrained at school, sent to ‘inclusion rooms’ for disciplinary infractions and sent home early. Others endure school days of bullying by peers and teachers, sexual assault and trauma. Still others are schooled in segregated institutions, isolated from their peers and delivered a curriculum of low expectations and inequitable outcomes. Instead of serving as the great equaliser, education has been shown to perpetuate inequality, reproducing rather than eliminating the widening gap between rich and poor (Hoadley 2008). The problem of educational exclusion is not new, nor has it been ignored by academics and policy makers. One of the educational reform movements of the twentieth century, mobilised to address exclusion in and from schools, is inclusive education.

The solution (?): inclusive education

The idea(l) of inclusive education has become well-known internationally. Its origins can be traced to the efforts of parents of disabled children in the Global North to challenge the discriminatory and inferior educational provision of separate special education services (Dyson and Forlin 1999; Thomas and Vaughn 2004). In the Global South, there is evidence that many pre-colonial communities were disability-inclusive in their education practices (Duke et al. 2016; Mahlo 2017). While initially conceptualised in terms of realising the right to unconditional access and full participation for disabled children and young people in mainstream education, inclusive education now concerns itself with identifying and addressing exclusionary practices and pressures experienced by those with a variety of identities that are marginalised (UNESCO 2018b). Inclusive education has been taken up by a ‘second generation’ of countries, particularly in the Global South, and has found its way into the policy and legislative structures of education systems throughout the world (Kozleski, Artiles, and Waitoller 2011, 4). Successive United Nations initiatives, including the Salamanca Statement (UNESCO 1994) have paved the way for inclusive education to be confirmed as a right in the Convention of the Rights of Persons with Disabilities (CRPD) (United Nations 2006). General Comment 4 on Article 24 of the CRPD (UNESCO 2016, Section 2) says, ‘Inclusive education is central to achieving high quality education for all learners, including those with disabilities, and for the development of inclusive, peaceful and fair societies’. Given the moral and political impetus for inclusive education, it must be questioned why educational exclusion prevails, and why an educational initiative so grounded in a rights discourse (Dyson 1999) has not compelled educational change.

Inclusive education has failed to gain a long-term foothold in many countries (Marchesi 2019; Pather 2019), including in England. The promise of an inclusive education system in England was ushered in by the Warnock Report of 1978 and the Education Act of 1981 (Hodkinson 2010). But, as Hodkinson (2010) notes, the twin-track approach that sustained special schools was not abandoned, with the government continuing to promote and develop ‘the orthodoxy of segregation within the loci of special schools’ (62). Special education has flourished in England in the twenty-first century, and both the number and proportion of pupils attending special schools is increasing (Shaw 2017; Warnes, Done, and Knowler 2022). This situation has led the United Nations Committee on the Rights of Persons with Disabilities to register concern at ‘The persistence of a dual education system that segregates children with disabilities in special schools, including based on parental choice’ and ‘The increasing number of children with disabilities in segregated education environments’ (United Nations 2017, Section 52 (a) and (b)) in the English context.

Various reasons are given for superficial and distorted instantiations of inclusive education, and its non-implementation, even in countries where inclusion is mandated by legislation and policy. General Comment 4 (UNESCO 2016, section 4) lists reasons such as a failure to adopt a human rights model of disability; discrimination; a lack of political will; technical knowledge and capacity, including inadequate teacher education; inadequate funding; and lack of legal redress. Other reasons are that inclusive education has been inadequately defined, leading to confusion about what it means in practice (Haug 2017); the reduction of inclusion to an accountability metric (Hodkinson 2010); inspection regimes and the standards agenda in England that encourage head teachers to encourage transfer to special schools (Shaw 2017); large classes that result in teachers regarding disabled students as ‘an additional workload’ (Srivastava, de Boer, and Pijl 2015, 189). Concern has also been raised that for developing countries, inclusive education represents yet another imposition from the Global North and does not consider local contexts (Duke et al. 2016; Sharma, Loreman, and Macanawai 2016; Walton 2018). The so-called policy practice gap is also blamed, where the intentions of policies are thwarted at various levels of the education system, and thus do not achieve their desired outcome. This paper follows the ‘theory adaptation’ tradition in conceptual research to offer a novel perspective on an existing conceptualisation of a problem (Jaakkola 2020; MacInnis 2011). Using Basil Bernstein’s (2000) concept of classification, I suggest that these reasons for the non-implementation of inclusive education, while valid, are symptoms of a more fundamental challenge.

Theoretical approach

Inclusion, exclusion and marginalisation in education are key concerns for Bernstein. He introduces his book Pedagogy, symbolic control and identity (2000) by announcing conditions for an effective democracy. Meeting these conditions requires that three interrelated rights are realised in schools. These rights are ‘enhancement’, ‘inclusion’ and ‘participation’. Bernstein’s main concern is with social class, and he asks, ‘What limitation of pedagogic democratic rights, for whom and where, is a given society prepared to tolerate and, at any one time, accept?’ (Bernstein 2000, xxv). Power and control are a starting point in Bernstein’s quest to understand principles of communication and how these bring about a differential pedagogical consciousness. He defined two concepts – classification for the translation of power relations, and framing for the translation of control relations (5).

The concept of classification will be put to work (Moore 2013) in this article to explain why inclusive education falters. Bernstein’s theoretical contribution is rich, expansive and generative. But, like Pandraud (2011, 176), I need to offer a ‘selective reading’ and restrict my focus to one concept, while acknowledging that other concepts, like framing, would also be valuable and lead to further insights. Bernstein (2000) took the term classification from Durkheim and used it to refer to the relations between different categories of groups and the extent to which the boundaries between these categories insulate them. He also used the term to apply to relations within categories. Three aspects of classification are significant for the discussion that follows. First is that there can be strong or weak classification, depending on how insulated the categories are. Strong classification occurs where impermeable boundaries between categories are created, and weak classification is found with porous boundaries and open lines between categories. Where there is strong classification, ‘each category has its unique identity, its unique voice, its own specialised rules of internal relations’ (Bernstein 2000, 7). Second, power relations create, legitimise and reproduce boundaries between different categories of groups. The insulation between categories is maintained by the workings of power. However, classification tends to obscure power relations, such that the principles of classification come to be seen as the ‘natural order’ and ‘the identities that it constructs are taken as real, as authentic’ (Bernstein 2000, 7). Third, the potential to change the classification by changing the extent of insulation between categories will be determined by the ‘power relations on which the classification is based and which it reproduces’ (Bernstein 2000, 7). Because the principles of classification construct identities, individuals will seek to maintain the integrity of categories and to forestall the possibility of weakening insulation.

Conceptual language is valuable for ‘ … its ability to generate strong principles of description for empirical studies’ (Bernstein 2000, 97). Theory thus enables description and analysis of what is observed in the real world. But the specific contribution of theory from Bernstein’s perspective is that it offers conditions of possibility for what Muller (2004, 3) calls ‘political choice and action’. This means that it not only describes what is, but what could be. Herein lies the force of a Bernsteinian analysis of why inclusive education falters – it offers the possibility of a variant educational world. The analysis points to the possibilities of change, by acknowledging the identities that classification constructs and identifying and exposing the workings of power that sustain and reproduce insulation between categories.

Strong classification in education

Schools and educational institutions are characterised by strong boundaries in several domains. Bernstein (2000) gives various examples, including the strong classification between official school knowledge and everyday knowledge, and between departments within schools. Cazden (2001, 61), in her tribute to Bernstein, identifies classification in a reading programme as, ‘Strong within-school as child is taken to a separate room with a separate teacher and a special curriculum, all different from the regular classroom’. From this, it is possible to think about the following five domains that maintain strong classification: spatial arrangements, professional roles and work, student ability, student identity, and pedagogy. These domains echo the triad of place, people and pedagogy, which has been used to examine aspects of education (e.g. McCowan, Janmaat, and Rao (2016) and Wade (2012)). I will offer reasons for the strong classification within the five domains, and given the theme of this Special Issue, I will be paying particular attention to the ways in which strong classification is reinforced by pervasive high-stakes testing and accountability regimes.

The spatial arrangement of schooling

Traditionally, regular and special schools and schooling have been classified with strong spatial insulation. This has meant that special education takes place in separate locations, either in separate classrooms, or in different premises. The rationale for this spatial insulation ranges from the concentration of specialist support providers, the demands of building and facility modification, student instructional needs or behaviours and parental choice (Farrell 2001; Kauffman et al. 2019; Shaw 2017; Veck 2014). The demands of high-stakes testing have also been used to justify strong spatial insulation. Referrals to ‘alternate education paths’ are, according to Hamre, Morin, and Ydesen (2018, x), ‘intimately connected to ubiquitous testing practices’. Kauffman et al. (2020, 259) argue that testing places pressure on ‘general education teachers’, for whom disabled students represent a ‘significant addition’ to their responsibilities. Furthermore, the competitiveness of high-stakes testing perversely incentivises schools to game the system by referring students who might compromise test results to special schools (Brzyska 2018; Thomson 2020; Tomlinson 2017).

Professional roles and work

Professional expertise tends to be strongly classified in educational provision. This can be seen in subject or grade specialisation in schools, and Bernstein (2000) uses school subjects (departments) to show how classification works in an institution with ‘staff tied to their category and its organisational base’ (10). When it comes to teaching and learning support, strong classification works to insulate the work and expertise of general educators, special educators, and allied health professionals (psychologists, speech pathologists, occupational therapists etc.). Each speciality is accorded its separate ‘field of practice’, that is, the ‘specialised practical contexts’ (Young and Muller 2014, 14) in which professional knowledge and judgment is exercised. The practical context is, as discussed in the previous section, often spatially insulated. Disabled students and those with other learning difficulties are commonly pulled out of lessons to receive support or remediation from these experts (Theoharis and Causton 2014; van Garderen, Stormont, and Goel 2012). Cook and Schirmer (2003) confirm the specialised expertise of the special educator operating in specific settings. These authors refer to a range of unique practices which have been ‘developed by special educators for students with disabilities’ (202), noting that ‘these techniques are, de facto, unique to special education settings, unique to students with disabilities, and unique to the competencies of special education teachers’ (203). Through various testing and diagnostic regimes, the role of special educators is reified and entrenched, and the professional expert has become the undisputed authority on pathology and intervention (Slee and Allan 2001). Powerful ‘vested groups in special education’ (Tomlinson 2002, 130) ensure that these roles remain strongly classified within the system.

Student ability

Education systems tend to organise themselves based on the bell-curve of scholastic ability (Florian 2014a; 2014b). This presumes that there is a natural distribution of the student population into distinct ability groups and that there are strong differences in the educational potential of children (Tomlinson 2017). The assumed hierarchy of ability levels is deemed the natural order of things (Francis et al. 2017; Spina 2019). This is a three-level hierarchy: a below-average or low ability group, an average ability group (the ‘norm’), and an above-average or high ability (gifted) group. These ability groupings are believed to reflect innate intelligence, environmental influence, and or scholastic performance with learning difficulties and differences in achievement being attributed to the fundamentally fixed nature of ability (Hamilton and O’Hara 2011; Hart et al. 2004; Tomlinson 2017). Research further reifies the tripartite classification of ability levels as it is used as a stable construct to investigate intervention effects (see, for example, Steenbergen-Hu, Makel, and Olszewski-Kubilius (2016) and Booij, Leuven, and Oosterbeek (2017)). In other words, ability levels are assumed to be strongly classified. This strong classification is reflected in various practices that seek to cater for these different groups – enrichment for the ‘gifted’, remediation for those who are deemed to be academically weak, and regular teaching for the average (Roberts-Holmes and Kitto 2019). These practices include pull-out programmes for specialised instruction (mentioned above), streaming (setting/tracking), homogenous ability grouping for co-operative learning, and tasks differentiated with variable demands, all of which have been associated with inequitable outcomes.

Testing regimes contribute to the construction, normalisation and legitimation of strongly classified ability groupings and entrench these in educational discourses, policy and practice (Hardy and Woodcock 2018). Florian (2018, 250) explains the danger of the imbrication of testing regimes and the construction of ability groupings as

National testing systems, intended to raise the academic standards of all students, may actually impeded the progress of some when they are used to make decisions about schooling practices such as tracking, streaming, banding, or setting that exacerbate inequalities … 

Standardised tests are perceived to be fair, objective, and credible (Ydesen and Au 2018) and they thereby legitimate instructional and other practices that differentiate among students (Giersch 2018; Spina 2019). In Australian schools, Spina (2019, 344) shows how ‘the rise of the quantification of education and evidence-informed practices’ may contribute to ‘a rise in the orthodoxy of ability grouping’. The practice of ability grouping is highly resistant to change as it is discursively connected to ‘standards’ in education (Francis et al. 2017), and standards are defined by testing. In a self-legitimising cycle, standardised assessments are used to categorise students by ability, and then ability groupings are used for targeted instruction directed towards improving the publicly reported results on high-stakes testing (Spina, Harris, and Jaremus 2021).

Student identity

Strongly classified student identities often form the basis for educational provision. These identities are often forged through diagnoses or appellation of the ‘Special Educational Needs’ (SEN) label, but may also be associated with language proficiency (English as a Second Language provision), gender (monastic schools) or religion (faith-based provision). The division of special and regular/ordinary/mainstream education is made on the strength of the classification between students with special education(al) needs and those without. England’s SEND Code of Practice (Department for Education (DfE) 2015, 15) identifies students as having ‘SEN if they have a learning difficulty or disability which calls for special educational provision to be made’ for them. This identification is made with reference to ‘the majority of others of the same age’ and the required provision ‘is additional to or different from that made generally for other children or young people of the same age’ (16). The scholarly literature rehearses and entrenches the binary distinction between those ‘with SEN’ and those without (see as recent examples, Schwab (2019) and LeRoy et al. (2019)). Not only are SEN and non-SEN students strongly classified, there is also strong classification between diagnostic categories within the SEN category.

The proliferation of diagnostic categories in special education was identified nearly thirty years ago (Forness and Kavale 1994; Pincus et al. 1992) and continues to be reified in policy and in textbooks (Brantlinger 2006; Walton 2016). England recognises four ‘broad areas of need’ for SEN pupils: Communication and interaction, cognition and learning, social, emotional and mental health difficulties, and sensory and/or physical needs. Within each of these areas, different conditions, disabilities and difficulties are itemised (DfE 2015). The annual school census counts pupils in 13 separate categories, with Speech, Language and Communications Needs, Moderate Learning Difficulty, Social, Emotional and Mental Health, Specific Learning Difficulty, and Autistic Spectrum Disorder being the most common type of need ‘among SEN pupils’ (DfE 2019). Research confirms that support provision for children tends to be driven by diagnostic categories, rather than their actual individual learning needs (Dockrell et al. 2019) with assessment data used to sort students, rather than to support their learning needs (Florian 2018).

Pedagogy

The strong classification of ability and disability is associated with a strongly classified pedagogy, in other words, ‘special’ pedagogy for ‘special’ children. The underlying assumption is that disabled children and those deemed to have ‘special needs’ have distinctive educational needs which need to be served with pedagogical approaches that can be distinguished from ‘regular’ pedagogy. Confirming this, Cook and Schirmer (2003, 204) say ‘there will always be children and youth whose learning needs and characteristics deviate from the norm in meaningful ways and who, therefore, will require some form of special education to receive appropriate instruction’. What makes a special pedagogy distinctive, say these authors, is a range of unique practices which have been ‘developed by special educators for students with disabilities’ (202). While these authors do concede that the boundaries between special and regular education may be permeable, in that ‘Techniques that are effective for students with disabilities are generally effective for all students’ (202), they do emphasise that these techniques are not used in regular education, making them unique. Strongly classified pedagogy is evident beyond the regular/special education bifurcation. Teachers adjust their teaching for students deemed to be of lower ability and who are placed in lower streams, sets or tracks, often on the basis of the results of standardised testing. These students receive ‘different pedagogies’ (Spina, Harris, and Jaremus 2021, 126) with less cognitively demanding instruction and less rigorous content (Giersch 2018).

These five domains are imbricated, and the strong classification of each domain mutually reinforces the others. This makes it very difficult for inclusive education, which is premised on weakening classification in these domains, to gain traction.

How inclusive education weakens classification

Inclusive schooling, collaboration, transformability, intersectionality, and inclusive pedagogy are key concepts in inclusive education, and these all represent a weakening of classification.

Inclusive schooling: weakened spatial insulation

Inclusive education challenges the strong classification of regular and special schools and classrooms. It maintains that most, if not all students should be educated together in a classroom made up of diverse students with a range of learning needs. General Comment 4 (UNESCO 2016, section 12c) confirms that inclusive education ‘ … commits to ending segregation within educational settings by ensuring inclusive classroom teaching in accessible learning environments with appropriate supports’. In other words, inclusive education entails the weakening of spatial insulation between special and regular schools and classrooms, and making the boundaries between these more porous. Special education advocates also recognise that inclusive education disrupts the strong classification of regular and special education, with Fuchs, Fuchs, and Stecker (2010) writing about the ‘blurring’ of special education into regular education in the No Child Left Behind reforms in the United States. They say, ‘At the building level, blurring means that special educators should abandon resource rooms and self-contained classes and take up residence in regular classrooms’ (Fuchs, Fuchs, and Stecker 2010, 306).

Collaboration: weakened professional insulation

Inclusive education challenges professional insulation and advances a weakening of the classification of professional roles in student learning and support. This does not mean expertise is not valued. Tancredi et al. (2020) describe a range of knowledge and skills that different professionals bring to an inclusive school, and argue for an appreciation of the contribution of all professional roles. But this contribution is expected to be made in collaborative teams which enhance student experience and contribute to their inclusion, participation and success. Instead of pulling students out of classes, inclusive education calls for push-in models, where other adults work together in classrooms. These models (which are not uncontested in the literature) include co-teaching, collaborative planning, coaching and instruction with a small group of students or individuals (Kilanowski-Press, Foote, and Rinaldo 2010; Naraian, Ferguson, and Thomas 2012; Tancredi et al. 2020). In characterising inclusive pedagogy (which is discussed further in a later section), Florian (2014a, 19) points to the need for teachers to work ‘with and through other adults in ways that respect the dignity of learners as full members of the community of the classroom’ and in partnership with ‘other adults who work alongside them in the classroom’ (Emphases mine).

Transformability: weakened ability insulation

Inclusive education challenges the classification of children and young people according to notions of fixed ability. It rejects ‘deterministic beliefs about ability as being fixed’ (Florian and Black-Hawkins 2011, 819). The transformability of learning capacity weakens ability classification with the ‘conviction that there is the potential for change in current patterns of achievement and response’ (Hart et al. 2004, 168). Inclusive education is based on the belief that that ‘all learners can make progress’ with learning potential being open-ended (Florian 2014b, 290). On this basis, inclusive education challenges practices based on homogenous ability grouping (Florian 2014b), drawing on the body of research which shows inequitable outcomes of these practices (Mijs 2016).

Intersectionality: weakened identity insulation

Inclusive education demands a recognition of intersectionality (Thomas and Macnab 2022), which challenges strongly classified categories of identity. Intersectionality foregrounds not only the multiplicity of identities, but also draws attention to the axes of oppression that differentially advantage and disadvantage students (Collins and Bilge 2020). It disrupts single factor analyses of educational failure, recognising that ableism and disablism are necessary, but often insufficient explanations for the exclusion that occurs in and through education. Students, as Artiles (2004, 552) notes, are ‘suspended in a web of historical, ideological, and cultural forces’ which must be acknowledged in educational provision. An inclusive approach to education recognises these forces and disrupts single-identity, strongly classified constructions of children and young people that determine provision and support (Bešić 2020).

Inclusive pedagogy: weakened pedagogical insulation

The strong classification of regular and special pedagogy has been challenged by Lewis and Norwich (2004) in their edited book Special teaching for special children? They draw a distinction between knowledge, curriculum and pedagogic strategies, and they tease these out in terms of whether they could be common to all students or distinctive for some. Norwich and Lewis (2004, 9) assert that pedagogic strategies are not distinctive for children ‘with learning difficulties’, but recognise that ‘there may be specific knowledge about pupils in the exceptional range that is required to apply common strategies’. In the final chapter of their book, Lewis and Norwich (2004) conclude by explicating a continuum, with differentiation or specialisation being a process of intensification of pedagogic strategies, rather than unique strategies.

An inclusive pedagogy rejects a ‘most’ and ‘some’ approach by,

Extending what is ordinarily available for all learners (creating a rich learning community) rather than using teaching and learning strategies that are suitable for most alongside something ‘additional’ or ‘different’ for some who experience difficulties (Spratt and Florian 2015, 92).

Inclusive pedagogy thus weakens the pedagogical insulation of special and regular teaching by seeking approaches that benefit everyone in the classroom. For Florian (2015), this does not mean that individual differences are not important in teaching and learning, but that these differences are routinely accounted for in lessons that are planned for all.

Why inclusive education falters: the persistence and defence of boundaries

The strong classification evident in the spatial arrangements of schooling, professional roles and work, student ability, student identity, and pedagogy is highly resistant to weakening. Power (often invisible) is at work to sustain these boundaries, and ensure that they appear as the ‘natural order’ of things. The status quo tends to be conserved by dominant groups (Case 2014) who have vested social, cultural and economic interests in maintaining strong insulation between categories. Evidence of this can be found in beliefs and practices at various levels of the education system. Parents in different contexts have been found to be invested in the preservation of the strong spatial classification of schooling by maintaining the segregation of other children (Brantlinger 2003; Kern 2020). Some regular and special education teachers reject efforts to weaken the insulation between their respective professional roles and identities through moves towards inclusive education (De Boer, Pijl, and Minnaert 2011; Maag, Kauffman, and Simpson 2019). Medical and therapeutic personnel with the power to name and treat the increasing proliferation of disorders have career and financial interests in maintaining the strong insulation between diagnostic categories (Slee 2011), as do academics whose research and teaching depend on strongly classified specialist knowledge in the field. Powerful Multi-Academy Trusts in England say ‘they have no room for students with special needs’ (Thomson 2020, 160) and head teachers in Cyprus are reluctant to challenge the ‘discriminatory and ineffective special education practices adopted by special teachers’ (Liasidou and Antoniou 2013, 498). In these and other ways, those with power and personal or institutional vested interests defend the strong classification of education and make it resistant to the weakening of boundaries that inclusive education represents.

Inclusive education, premised on weakened boundaries, falters when attempts are made to map it onto a strongly classified system that is buttressed by testing regimes. Because of different classification logics, inclusive education is, in many ways, incompatible with the current system. This incompatibility gives rise to limited or distorted instantiations of inclusive education, as policy-makers and practitioners try to patch inclusion into the system. This is exemplified in Cyprus as Liasidou and Antoniou (2013, 496) note that ‘the notion of inclusion is reduced to the propagation of assimilationist practices’ resulting in the marginalisation of some students in mainstream settings. In England, the government has made teachers responsible for inclusion within an ‘unchanged educational culture that is far from inclusive’ (Warnes, Done, and Knowler 2022, 41). These examples are symptoms of the deeper problem, which is that inclusive education demands a fundamental change to the ‘architecture’ of schooling (Slee 2011, 171).

Conclusion: the possibility of change

The possibility of change comes first with an acknowledgment of the identities that classification constructs and the defences that individuals build to sustain categories. Weakening insulation between boundaries may threaten individual identities, which may have been forged through years of training and practice. Discounting the threat to identity risks alienating people who have much to contribute to the dismantling of exclusionary pressures and practices. Second, change demands the identification and exposure of the workings of power that sustain and reproduce insulation between categories. The crucial and critical question of ‘who benefits’ from the hegemony of strong classification in the five domains discussed in this paper must be asked and answered. The benefits of a strongly classified education system accrue to some across a lifetime, and are implicated in the perpetuation of inequality. Finally, while acknowledging the power and persistence of the boundaries described in this paper, there are possibilities for ‘counter-hegemonic action’ (Apple, Au, and Gandin 2009, 4) to blur boundaries and weaken insulation at various levels of the system. Instances of inclusive schooling, with collaborative practice, recognition of transformability and intersectionality, and inclusive pedagogy give ‘reasons to be cheerful’ (Slee 2011, 13) in the quest for quality inclusive education for all.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Elizabeth Walton

Elizabeth Walton is a Professor of Education at the University of Nottingham, UK and a visiting Professor at the School of Education, University of the Witwatersrand in Johannesburg. She is the co-convenor of the UNESCO Chair in Teacher Education for Diversity and Development, and her research interests include teacher education for inclusive education, the field of inclusive education (its knowledge and knowers) and identifying and addressing educational exclusion, with a particular focus on the Global South.