SPECIAL SCHOOLS: What future for special schools and inclusion? Conceptual and professional perspectives


  • Brahm Norwich

Brahm Norwich
School of Education and Lifelong Learning
University of Exeter
Heavitree Rd
Exeter EX1 2LU
Email: b.norwich@ex.ac.uk


In this article, Brahm Norwich, Professor of Education at the University of Exeter, examines the roles that special schools can play within inclusive educational systems. He notes that the percentage of young people in special schools in England has remained broadly stable over a number of years, despite inclusive policy initiatives. Brahm Norwich suggests that policy makers and practitioners have found it hard to understand how a broad and shifting notion like inclusion should be operationalised, especially when valued positions, such as meeting individual needs and providing a sense of belonging and participation, can appear to generate such tensions and contradictions. Brahm Norwich summarises findings on teachers' attitudes towards this crucial ‘dilemma of difference’ from three countries and argues that it is time to develop more sophisticated ways of thinking about provision. Rather than insisting on locating ‘mainstream’ and ‘special’ at opposite ends of a one-dimensional placement continuum, Brahm Norwich puts forward a multi-dimensional model in which a number of attributes can be considered when analysing provision. The ‘flexible interacting continua’ provided in this model concern identification, participation, placement, curriculum and teaching and governance and Brahm Norwich shows how schools, whether mainstream or special, need to strive towards commonality in terms of all five dimensions rather than simply in terms of placement. Policy makers as well as staff in both mainstream and special schools will be interested in exploring the implications of these ideas.


With the historical movement for comprehensive schools and the vision of a school for all, irrespective of gender, social class or ability, it followed that more provision was to be made for children with special educational needs/disabilities in ordinary schools. This meant that special schools came to be seen as less favourable settings for specialist services. It was in the wake of the Warnock Report (DES, 1978) that the Government adopted an official policy of integration in the 1981 Education Act. Children with special educational needs were to be educated in ordinary schools, subject to several conditions: (1) that the child received the provision needed; (2) that others' education was not disrupted by integration arrangements; (3) that parents were supportive; and (4) that the arrangements were consistent with the ‘efficient use of resources’. This legislative framework made it possible to have a mixed model of provision, where special schools still existed, but there was also scope for increasing provision in ordinary schools. That framework still operates, despite several legislative changes and the reduced number of conditions enabling separate special provision. The current conditions for ordinary school placement are: (1) that it is supported by parents; and (2) that it does not disrupt the education of other children.

This movement towards more provision for children with learning difficulties and disabilities in ordinary schools was underpinned by what has been known as a continuum of provision. This continuum has taken various forms, but has mainly concerned the range of placements, from the most separate to the most included, in ordinary settings. Figure 1 shows one way of representing the continuum and where special schools fit into this scheme.

Figure 1.

Continuum of special education provision

From 1983 (when the 1981 legislation came into operation) until 2001, there was a decrease in the percentage of children in special schools, from 1.87% to 1.30% (Norwich, 2002). The decrease was greater in the 1980s than the 1990s, and there are indications that since the year 2000 the percentage has remained at around the 1.2% to 1.3% level.

Current debates

The arguments against special schools that emerged with the growth of the comprehensive movement and commitment tointegration centred partly on the experiences of stigma and devaluation associated with going to special schools, and partly on the poor quality of provision available in these schools. The argument for integration was that without clear evidence of the learning advantages of special schools, and given the commitment to equality of opportunities, the presumption was towards ordinary schools (Galloway & Goodwin, 1987). Research reviews over the last few decades have been interpreted as indicating no clear support for the positive academic or social effects of either inclusion or separate schooling (Lindsay, 2007).

Since the 1990s, the movement towards increasing provision for children with special educational needs in ordinary schools has been promoted in terms of inclusion rather than integration. This can be seen in the context of wider policies promoting social inclusion, which involves the participation of ‘vulnerable’ members of society in a range of social activities and settings, not just those with special educational needs/disabilities in ordinary schools. Inclusion has also been promoted as a rights issue and as about changing the system (in this case, ordinary schools) to make them more accommodating of those who are ‘different’ (Ainscow, 1999). Integration then came to be seen as about placing children in ordinary schools without this kind of organisational accommodation. So, for some people, inclusion also took on the connotation of being about social acceptance and instilling a sense of belonging to common institutions (Thomas & Loxley, 2001).

It is clear that inclusion represents a broad and significant social and political value commitment, which, when applied to inclusive education, has come to be defined as ‘the participation in the cultures, curricula and communities of local schools’ (Booth, Ainscow, Black-Hawkins, Vaughn & Shaw, 2000; emphasis added).

However, the problem with a multi-faceted and general notion like that of inclusive education (Lunt & Norwich, 1999) is that it has been hard to know how to apply it in the everyday world of policy and educational practice. This problem was recognised in the recent House of Commons Education Select Committee Report on Special Educational Needs (House of Commons, 2006). In grappling with the meaning and policy implications of a commitment to inclusion, the Select Committee urged the Government to ‘work harder to define exactly what it means by inclusion’ (House of Commons, 2006). It is interesting to note that the Government response produced a concept of inclusion that still did not clarify the extent to which children with special educational needs are to be placed in local ordinary schools. The DfES proposed the following:

‘The Government shares the Committee's view that inclusion is about the quality of a child's experience and providing access to the high quality education which enables them to progress with their learning and participate fully in the activities of their school and community.’

(DfES, 2006, section 28)

Although this complex definition refers to various aspects of education (quality of experience, access to quality education, learning progress and participation in school activities and community), it does not deal with the critical question of placement in local schools. The problem is that this DfES definition is consistent with an emphasis on an inclusive system, which could involve special schools, rather than on inclusive ordinary schools.

Even if inclusion is about academic and social participation in local ordinary schools, it is still unclear whether ‘participation in local schools’ requires children with difficulties in learning to be in ordinary classrooms for most of the time. It is interesting that the Centre for the Study of Inclusive Education (CSIE), a pressure group known for its strong commitment to inclusive education, accepts in its Charter that some children with special educational needs can spend part of their time outside the ordinary classroom:

‘Time spent out of the ordinary classroom for appropriate individual or group work on a part-time basis is not segregation. Neither is removal for therapy or because of disruption, provided it is time-limited, for a specified purpose . . . Any time-out from the ordinary classroom should not affect a student's right to full membership of the mainstream.’

(Thomas & Vaughn, 2004, p. 137)

This concession to part-time withdrawal raises a further question. Are part-time placements in off-site settings for appropriate and time-limited learning compatible with inclusive education, if the children are still members of ordinary schools and classes? The CSIE concession is a step towards justifying separation, and shows an inter-connection between included and separate provision.

Apart from these questions about placement and social belonging, there are still other problems with the concept of inclusion. Some reject the idea that inclusion is about location or placement at all. Mary Warnock (2005), the chair of the 1978 Government committee that endorsed integration, has recently rejected educational inclusion as being ‘all children under the same roof’. She prefers a learning concept of inclusion, which is about ‘including all children in the common educational enterprise of learning, wherever they learn best’ (Warnock, 2005). Her version of inclusion reflects views that are shared by some teachers who work with children with special educational needs. They see some degree of withdrawal to a separate setting as being inclusive in the sense of making it possible for certain children to engage in learning the same curriculum as other children (Norwich, 2007a). It is clear that there are different and sometimes conflicting notions of inclusion.

These confusions are affected by political divides concerning the future of special schools. While on the one hand, the Government has been ‘duplicitous’ (House of Commons, 2006) in its policies, the Conservative Party has taken a stand in favour of more special schools in its recent Policy Commission on Special Needs in Education (Conservative Party, 2007). As the Commission Chairman Bob Balchin said:

‘. . . closing special schools and sending large numbers of children with special education needs to mainstream schools has caused great distress. This is why we need to recreate many more special school places.’

Administrators' and teachers' views about placement and the future role of special schools

One way to improve our understanding of the tensions that arise over separate settings for children with special educational needs/disabilities is to consider some of the values that drive the organisation of school systems. Two values in particular are relevant: educational provision to meet individual needs, and instilling a sense of belonging and acceptance in ordinary schools for all children. However, the extent to which these can be combined and co-exist depends on the diversity of children's educational needs and the flexibility of schools to offer quality and differentiated provision. However, there may arise situations where there is a tension between these values, leading to a dilemma about placement or the location of provision (Norwich, 2007a). This placement dilemma relates to children with more severe disabilities/special educational needs in the following ways:

  • If children with more severe disabilities (needing special education) are taught in general classrooms, then they are less likely to have access to scarce and specialist services and facilities.
  • If children with more severe disabilities (needing special education) are not taught in general classrooms, then they are more likely to feel excluded and not be accepted by other children.

This dilemma was examined as part of an international study of 132 policy-makers and teachers in the USA, the Netherlands and England to explore positions about various aspects of difference or differentiation (the three aspects were identification, curriculum and placement). These participants were from two distinct areas in each country (urban and rural) and included national and local administrators; class and senior teachers in ordinary and special schools; and advisory and support professionals. Through semi-structured interviewing, they were asked to what extent they recognised this dilemma (using a rating scale) and the reasons for their position. If they did recognise the presented dilemma, they were asked to what extent they could resolve it (using a rating scale) and for the specific ways of doing so. Semi-structured interviews were used because this gave the opportunity to explore the meanings that participants attributed to terms like ‘severe disability’ and the presented general consequences. Interviews were recorded, transcribed and the reasons (for recognition/non-recognition of the dilemma) and ways of resolving/not resolving the dilemma were analysed qualitatively for emerging themes. These emergent recognition themes were then analysed to determine to what extent they corresponded with categories that reflected possible responses to such a dilemma (for example, themes about ‘tensions’; ‘denial or moderation of negative consequences’ in the dilemma; ‘other consequences’ of placement options; and current ways of ‘resolving tensions’ about placement). For more details of the study, findings and discussion, see Norwich (2007a).

Reasons for recognising/not recognising placement dilemma

The most frequent response was the recognition of a significant placement dilemma across each country. A large majority of the participants (76% to 86%) across the three countries reported recognising this dilemma to some extent.

Figure 2 summarises an analysis of participants' responses to the dilemma that were common across the three countries. There were also responses that were specific to each country group, but they were in a minority and are not represented here. At the centre of the figure are the key placement options for children with more severe disabilities. The two solid lines point to the presented negative consequences of two options. The arrows point to emergent reasons that were shared by participants across two or all three countries. The most frequently occurring theme for recognising the dilemma was ‘tensions’, with eight specific sub-themes that were shared across the country groups. (The number of sub-themes corresponds approximately with the frequency of using the theme.) The other two more frequently used themes for recognising the dilemma were ‘other negative aspects of inclusion’ and ‘resolved tensions’. The former theme identifies other negative consequences not presented in the dilemma, while the latter theme acknowledges the tension, but identifies a working resolution. The next most frequently occurring theme identified ‘other consequences’ of separation/inclusion, not presented in the dilemma. Of these emergent themes, ‘positive aspects of separation’ was used more than ‘positive aspects of inclusion’. The minority of participants who did not see a dilemma, or saw only a marginal dilemma, tended to either deny or see moderated negative consequences (see right-hand side of figure). This denial or moderation of the ‘less access to specialist services/facilities’ negative consequence was more frequent than the denial/moderation of the ‘feel excluded and not accepted by others’ negative consequence. The sub-themes used by those questioning the dilemma were, on the one hand, about building specialist capacity in ordinary classes and this being easier in the early and primary years; and on the other hand, that few children were aware of or cared about being apart. Some participants in all three countries also responded by explaining that there were already ways of resolving the placement dilemma: ‘resolved tensions’ which involved different ways of combining included and separate settings.

Figure 2.

Placement dilemma – shared recognition themes across the three countries

Resolving the placement dilemma and the future of special schools

For those recognising the dilemma, the most frequent response was to identify a significant resolution. A very large majority saw some degree of resolution (90% to 100% across the three countries).

Figure 3 shows that the most frequently used resolutions that were common across the three country groups were ‘balance included/separate provision’ and ‘enhance flexible services and staffing’. Less frequently used were those themes that identified persistent problems: ‘limits to inclusion’ and ‘continuing issues’. The least frequent shared theme was ‘systems/local change needed’, although this was used much more frequently in country-specific ways.

Figure 3.

Placement dilemma – common resolution themes across the three countries

In their resolutions, participants also revealed their views on the future of special schools in their own national settings. Some participants referred to a ‘reluctant future for special schools’ when talking about the rationale for special schools. There was some reference to reducing the number of special schools, particularly by Dutch and English participants. Dutch participants tended to refer to special schools as part of a continuum of provision, while English participants referred more to co-located special schools (special schools on the same campus and with links to ordinary schools). There was also support for special schools dependent on various factors: for example, special schools had to provide quality provision that maximised learning progress and more inclusion depended on appropriate provision in ordinary school, including teachers who were willing to accommodate exceptional needs. Some of the comments in each country also revealed that tensions were evident. For some participants from the USA and the Netherlands, this was captured by the theme that there were ‘two sides to the debate’.

Figure 4 shows the five shared reasons across the three countries for a future role for special schools. The most frequent reason related to what was required for children with certain ‘children's characteristics’. There was some agreement on these characteristics across the country groups, though terminology differed. In the USA they were referred to as ‘significant care needs’; in the Netherlands as ‘very severe and complex difficulties’ or ‘very low intellectual levels’; and in England as ‘profound disabilities’. There was also reference to children with severe emotional and behavioural problems and to the related theme about ‘challenging behaviour that is a threat to others’.

Figure 4.

Common reasons for special schools across three countries

The second most frequent reason for having special schools was about the ‘provision available’ in such schools. This involved beliefs that special schools were ‘justified if they[children]can move back to regular settings’, on one hand, and that special schools provided a ‘safer/happy environment’, on the other. The third reason for having special schools was in terms of ‘professional and school specialisation’. One version of this was about the special school outreach and support services for ordinary schools. The other was about special schools having ‘specialist professionals’. ‘Stakeholders' interests’ was the fourth reason justifying special schools, though not a frequently used one. This shared reason was about parental interests: ‘some parents want them’. The fifth reason was about economic factors; that special schools were a more economic use of resources (though this was used by few participants).


The summary of the findings presented here, considered alongside the other findings of the larger study of dilemmas of difference (see Norwich, 2007a), shows that a range of professionals and administrators in three countries see tensions about differentiation. The specific reasons and arguments for recognising and ways of resolving the placement dilemma are interesting in themselves. They show how those directly involved in the field think and argue in common ways about these questions across three countries, despite some country-specific aspects. These positions also show a level of balanced and sophisticated thinking that goes beyond some of the recent policy debates that have arisen over the future role of special schools (Norwich, 2007b). The study can also be interpreted as evidence for the relevance of taking account of dilemmas that arise in the field generally and about special schools in particular.

It is revealing that some participants responded to the placement dilemma by suggesting that there were tensions that had been resolved to some extent. For some, this was a way of justifying only a marginal or no recognition of the dilemma, though it was also used by participants who thought that some placement tensions remained. For those who held this view, what was involved in the ‘resolved tension’ was some balancing; for example, maximising the child's time in ordinary class or providing services in additionally resourced mainstream schools. These views reveal a balancing of separate and included provision. This balancing was also the most frequently used way of resolving the dilemma. As Figure 3 shows, there were many specific resolutions, although the most frequent method, which was common across the three countries, was to have a ‘mixed model’– mixing some ordinary class with some separation/withdrawal, an approach which could be implemented in various ways, some of which were specific to particular countries. Across the three countries, it was apparent that this balancing required more flexibility about services and staffing. This shared approach to resolving the placement dilemma was also evident in general views about the future of special schools, although most of those who expressed such a view endorsed a reduction in the number of special schools. Some participants admitted that they were reluctant to hold the view that there was a future for special schools, while others continued to see two sides to the debates about separate settings. These responses can be seen as reflecting expressions of the continuing tensions experienced in relation to this placement dilemma. The general reasons for having special schools were broadly related to the interaction of specific characteristics of children and provision factors moderated by stakeholders' interests and economic factors.

Dilemmas arise because we live in a society where there are plural values that sometimes conflict and need to be balanced with each other (Billig, Condor, Edwards, Gane, Middleton & Radley, 1988). Seeing decisions in this field in terms of dilemmas of difference (or differentiation) is useful because it helps us to understand the nature of policy and practice decisions and the need to co-ordinate the different values involved. The implication is that decisions can involve different balances between included and separate provision. The more included options involve some aspects of separate provision, while the more separate options involve some aspects of included provision. This avoids the oversimplified splits or dichotomies that are often found in debates and discussions over provision, and the futile pursuit of ideological purity.

Concluding implications

Many of the administrators and teachers across the three countries recognised a reduced but nonetheless persisting role for special schools that are inter-connected with ordinary schools. When this finding is viewed in the context of the wider findings from the international study that found similar professional recognition of other aspects of differentiation, it can be concluded that there is a need for a more sophisticated multi-dimensional model of provision than the one represented in Figure 1 above. Designing educational provision for all children, including those with difficulties and disabilities, involves balancing common and different aspects, and this can sometimes lead to difficult choices, as this study has indicated. This requires further development of the notion of a continuum of provision. Rather than the traditional uni-dimensional approach in terms of placement, continua might best be specified in terms of various specific dimensions –‘flexible interacting continua of provision’ (Norwich & Gray, 2007) – where the following questions need to be considered:

  • identification – how can children with special educational needs or disabilities be identified?
  • placement – where do children/young people learn?
  • curriculum – what do children/young people learn?
  • level of governance about educational provision – which agency decides about provision?

Options within these dimensions are illustrated in Figure 5.

Figure 5.

Options within flexible interacting continua of provision

This conception of ‘flexible interacting continua of provision’ differs from the traditional continuum of special provision in its analysis of several key dimensions and the fact that different options in these dimensions are inter-connected. But it also differs in setting limits to these options, with some options not being included. For example, in the placement continuum, separate special schools that are not linked to ordinary schools are not included, as these do not represent a balance between common and separate provision. Curriculum aims that are disconnected from general common curriculum aims are not an option in the curriculum dimension, as they also fail to represent sufficient balance between common and differentiated curriculum content.

A commitment to inclusion means that progress in developing educational provision in terms of these five dimensions (shown in Figure 5) consists of a move towards greater commonality. This view reflects a specific political/ideological position, and of course there is scope for others to have a different emphasis and take a different stance. Nevertheless, I would argue that the question of inclusion should be considered in terms of all five dimensions – and not just one or two dimensions; for example, the curriculum and social participation, while leaving out placement, as some do (Warnock, 2005). I also realise that in some circumstances there may be limits to the extent of commonality; for example, separate settings for teenagers involved in violent crime. But even here, such separate settings would be connected in various ways with general provision. The overall implication of this study is that the future role for special schools is limited and is inter-connected with ordinary schooling and educational processes.