The modern state has traditionally viewed disabled people as passive victims of personal tragedies, fit mainly for medical intervention or institutionalization (Oliver 1990). That view has been challenged, particularly in North America and Western Europe, by activists, scholars and organizations. The consequences of the transformation that has been, and still is, taking place are not yet fully known, however. At the beginning of the twenty-first century, what it means to secure equal rights for disabled people is far from clear, and the discourse in which such rights can be justified, established, clarified and elaborated is still taking shape. One of the key challenges that faces the disability rights advocates, lawyers and policymakers who produce such discourse is that of defining what ‘equality’ means for people with very different abilities and potential. People with very different impairments are discriminated against in very different ways by social, physical and economic arrangements, and ensuring equality is not simply a matter of removing barriers or eliminating prejudice.

Over the last few decades, comprehensive bills that define disability as a matter of universal human (or civil) rights and thus as a suitable target for anti-discrimination measures have nevertheless been introduced in a number of countries and territories. These include, but are not limited to, the United States (1990), Australia (1992), the United Kingdom (1995), and Ontario in Canada (2002). Quite recently, the Norwegian Department of Children and Equality introduced a legislative proposal entitled the Discrimination and Accessibility Act, (hereafter, the DAA), which entered into law on 1 January 2009. It is this law that will serve as the focus of analysis for this article, and as a basis for asking what the relationship between grand universal rights and everyday particular accommodations is like.

One way of rephrasing this question is to ask how the provision of equal opportunity for people with impairments in Norwegian society is tied to discourses that originate outside the disability field – in architecture, in public works, in the health professions, etc. From this it follows that we need to talk about what is still, according to Tobin Siebers (2008), the ‘emerging field of disability studies’. This field, emerging though it may be, is in constant danger of turning in on itself – of ‘producing the illness it seeks to cure’, in the words of one reviewer of Siebers’ book (Davidson 2008). One such illness, to go with the reviewer's metaphor, is the overindulgence in reductive theoretical models and the dichotomies between such models.

Disability is a diverse and ill-defined phenomenon, which already sprawls across and is interpreted in a number of fields. Multiple models such as the British social model (see Oliver [1996b] for an introduction and Shakespeare [2006] for criticism), the American minority model (Hahn 1996) and the Scandinavian relational or gap model (Tøssebro 2004), have gained significant influence in academic and activist circles, and to some extent in social policy, though none of them have become hegemonic. The government of disability is of course still a matter of medicine as well (Tremain 2005), but it would be inaccurate to say that there is a medical model that is on par with the others. I am especially concerned to show that any dichotomy between a medical model and other models of disability is less than helpful when engaging in disability studies, because it will inevitably gloss over the fact that the medical model has few defenders on a conceptual level, even though it may be alive and well in, for example, professional practices.

There is another borderline-activist point to this article, which also has to do with disability studies: in order for social justice for people with impairments to come about, it is crucial that the comprehension of both general principles of anti-discrimination and specific needs for accommodation are embedded within as wide a range of discourses as possible. The field of disability studies should emerge not as a self-contained arena, but as one that influences and remains in constant dialogue with other academic and political fields.

Case in point: the law that is discussed in this article forms part of a larger, ongoing debate on how disability should be defined for political purposes. The outcome of that debate will influence almost every aspect of the lives of many people, and in order for the best outcome to be achieved, a multiplicity of perspectives and contributions will be needed. My own contribution comes with the caveat that I am neither a legal scholar nor a policy specialist. I am aiming here to present issues that are and will continue to be important to disability and disability studies, issues that have to do with how disability is framed and understood.

To that end, I believe the framework of critical discourse analysis (CDA) can be useful. In Part 2 of this paper, I present (very briefly) some tools from that framework. In Part 3, I discuss the discourse of the DAA and its provisions. Parts 4–7 address the relationship between medical knowledge and sociopolitical goals, as well as the role of civil society and non-governmental organizations (NGOs) in carrying out the policies of the DAA.

A note on method: CDA

A key aspect of CDA is the view of language as both a symptom and cause of social change (Fairclough 2005; Fairclough, Cortese, and Ardizzone 2007; Wodak and Chilton 2005; Wodak and Meyer 2001). Studies in CDA have addressed themselves variously to ideologies of news media (van Dijk 1988), Thatcherist and neo-liberal discourse in the UK (Fairclough 1992 ,1989/2001) and anti-immigration discourse in Austria (Wodak 1999; Wodak and Pelinka 2002).

Disability is somewhat under-represented in the CDA tradition, but it would not be an exaggeration to say that key figures in disability studies have employed methods that fit well with CDA. A classical example is the British sociologist Mike Oliver's rephrasing of survey questions:

 … used in the 1990 OCPS Disability Surveys with the social model approach he preferred. For example, instead of asking ‘What is wrong with you?’ (medical model), Oliver offers ‘What is wrong with society?’ (social model). (Shakespeare 2006, 28)

Oliver directs the reader's attention to social constraints instead of intrinsic properties (see Oliver [1990] for the full list). In doing so, he critically examines the relationship between language and the social world. The language of the survey is exposed as obscuring the many disabling factors of society and the environment. This method – showing the inadequacy of certain forms of powerful language in adequately representing the world – is at the heart of CDA.

Material: the DAA

I will begin this section with an overview of the DAA, which is a 6-page law supported by a 270-page advisory statement (hereafter, the AS). Thematically, it ranges over a wide variety of topics. This range follows from its central concern with equality:

The purpose of the law is to promote equality and equal worth, to ensure equal opportunity and equal rights, and to prevent discrimination due to impairment. (Barne- og likestillingsdepartementet 2008, 12)1

The range of the DAA follows from how the concept of equality gets translated into anti-discrimination measures that are specific to the disability field. As Tom Shakespeare has pointed out, preventing discrimination against people with impairment presents challenges that are unique, and very different from those that relate to anti-discrimination measures in the fields of gender and ethnicity:

Women and men may be physiologically and physically different, but it is no longer possible to argue that women are made less capable by their biology. … Similarly, only racists would see the biological differences between ethnic communities as the explanation for their social differences. Nor is it clear why being lesbian or gay would put any individual at a disadvantage, in the absence of prejudice and discrimination. But even in the absence of social barriers or oppression, it would still be problematic to have an impairment, because many impairments are limiting or difficult, not neutral. (Shakespeare 2006, 41)

Discrimination on the basis of gender (or ethnicity) may frequently be described as a form of negative intervention – an imposition of social barriers or oppression, in Shakespeare's terms. Discrimination on the basis of impairment, in contrast, is frequently a lack of positive intervention. Hence, there is a need not only to claim that discrimination has taken place, but to explain precisely what the act of ‘discrimination due to impairment’ consists of.

Section 2 of the DAA states that the Act will apply in ‘every area of society, excepting family life and other relationships of a personal nature’, prohibiting discrimination on the basis of impairment in all such areas. Specifically, it contains provisions that relate to:

  • •  Employment, education and social services (Section 3 and Section 12).
  • •  Harassment (Section 6).
  • •  Universal design standards for publicly accessible buildings and concerns (including public transport) as well as information technology (Section 9, Section 10, and Section 11).
  • •  Housing and rental regulations.

Trying to fix, in Mike Oliver's self-consciously general phrase, ‘what is wrong with society’, leads to the question of how the particular problems that need fixing are going to be identified. In some of the areas, the responsibility clearly lies with public authorities, corporations, and other institutional agents to conform to the DAA by changing their structures or procedures. Universal design is the most relevant case. In other areas, however, it is disabled people themselves (or other outside agents) who will have to report violations of the law, as with harassment or the lack of provisions in education. This means that the anti-discrimination measures in the DAA are sites of complex negotiations between parties with different interests. The space in which the sites are distributed can be described with dimensions of directness and indirectness, as well as individual protection or structural change (see Table 1).

Table 1. Policy measures in the Discrimination and Disability Act.

Measures of individual protection         Anti-harassment measures
    Workplace adaptation (in general)      
    Adjustments to social services etc.      
      Adjustments to educational programmes    
    Workplace adapatation w/ mandatory reporting (50+ employees)   Equal opportunity hiring practices  
Measures of structural change Universal design        
  Indirect discrimination       Direct discrimination

Table 1 is meant to illustrate two things: firstly, that discrimination due to impairment is represented as a complex phenomenon in the DAA. The general goal leads to the multiplicity of particular changes in the social fabric. I have placed universal design in the lower left corner because the rules and regulations intended to implement it are aimed at institutional practices that discriminate indirectly; it is the architects and builders that are required to comply with the standards. Anti-harassment measures in the upper right corner, on the other hand, will require direct action from disabled people themselves in order to be effective, and may or may not address the underlying causes of harassment.

Secondly, most of the DAA measures cluster in the centre of the table. They are neither blanket requirements directed at institutions, nor are they primarily directed at overt forms of discrimination. They are justified as means of compensation for inherent or pre-existing disadvantage, and their extent is qualified or limited to the scope of reasonable individual accommodations. They largely avoid matters of structural change, but propose adjustments to established institutional practices that disadvantage people with various impairments. As such they will enter into a web of pre-existing power relations. Their justifications are frequently of the form ‘if people with impairments are disadvantaged in situation X, individual accommodations must be made’. Changing what is wrong with society, in other words, implies finding out what is ‘wrong’ with the people in it. This, of course, means that people will have to admit that something is wrong in order for society to change. What are the requirements for this to happen?

What is wrong with you? Disability, discourse and identity

Who counts themselves as being disabled? Who will (no pun intended) stand up to be counted? It has been repeatedly established that being ‘a disabled person’ is something that varies with historical epochs, cultural environments and situational factors (Snyder and Mitchell 2006; Stiker 1982/1999). It is also well known that the proportion of any population that counts as disabled varies greatly according to definitional criteria and measurement techniques, as well as context. In recent decades, many attempts have been made at minimizing or marginalizing the medical aspects of disability definitions, in order to link the goal of equality to political and legal discourse.

Some sort of division between impairments and disabilities nevertheless seems essential to any definition; legally and politically because it separates medically defined diagnoses from social or economic disadvantage, practically because people with vastly different impairments frequently share experiences of oppression and exclusion. The recently issued United Nations Convention on the Rights of Persons with Disabilities states that:

Persons with disabilities include those who have long-term physical, mental, intellectual or sensory impairments which in interaction with various barriers may hinder their full and effective participation in society on an equal basis with others. (United Nations 2007)

Such a definition is deeply problematic on (at least) two main counts:

  • 1.   Duration and severity of the impairment, which is sometimes presupposed as a stable factor. But what counts as ‘long-term’? And when, precisely, is a body impaired? Labour market research from Statistics Norway suggests that approximately 15% of the Norwegian population consider themselves to be impaired (Olsen and Van 2007), but the researchers note that fully 35% of respondents provide different answers in different studies.
  • 2.   Causality. Is ‘full and effective participation’ hindered by the impairment or by the barriers? This has been a point of contention for most, if not all, modern writers on disability, because it carries strong implications for the direction of political efforts. At one extreme, being unable to walk is the reason why citizen A cannot get a university degree – not the fact that all the university buildings have steps. At the other, ‘disablement has nothing to do with the body’ (Oliver 1996a). The UN definition places itself somewhere in the middle, where impairments and barriers share causality.

Finding out who is disabled is not a matter of counting broken limbs or counting the marginalized part of the population. From the complex interactions between people with impairments and various social institutions, different discourses of disability arise. In the pre-modern age, religion played a crucial role (Stiker 1982/1999), as in the social interpretation and construction of mental illness (Foucault 1973), but for the last few centuries, medical science has provided the singularly most important source of disability discourses. Given that there is a statistically average or normal human body, and given that some people display significant negative deviations from the norms of bodily function, society may classify these deviations as impairments and react in a number of different ways. But disability has relatively rarely been a positive identity championed from within, compared with how frequently it has been a negative identity imposed from without.

The question is not clear-cut. Public or private charitable institutions may use impairments as the basis for morally legitimate needs, making disability a superior alternative to delinquency – for some people. Institutions of state welfare may define impairment as the threshold beyond which an individual is unable to work, and may be relegated from the wage economy to the needs-based economy (Gleeson 1999a ,b ,2001; Stone 1984), and this may be desirable – for some people.

A medical bureaucracy may certify this inability, and subsequently provide people with impairments with identities that make them legitimate2 objects for intervention and rehabilitation, namely techniques of normalization. Identity-defining discourses of charity, welfare and rehabilitation, then, all originate in part from the medical model. And these discourses remain alive, even in a document as progressive as the DAA.

Marginalizing medical knowledge

Another way of talking about the aforementioned discourses is to say that they are grounded in medical topoi, or places of argument, which serve as the legitimizations for interventions geared toward achieving normality. The topos is useful in the context of CDA because it is a mainstay of both rhetorical and argumentation theory. It has been understood in a number of ways, but for the purposes of the following I will use the definition of the Austrian discourse analyst Ruth Wodak:

Within argumentation theory, ‘topoi’ or ‘loci’ can be described as parts of argumentation which belong to the obligatory, either explicit or inferable premises. They are the content-related warrant or ‘conclusion rules’ which connect the argument or arguments with the conclusion, the claim. As such, they justify the transition from the argument or arguments to the conclusion. (Wodak and Meyer 2001, 74; emphasis added)

Topoi are particularly salient in grounding key concepts because they provide the link to the more specific arguments in which those concepts will be employed. As such, they have direct bearing on identity construction. Medical topoi may be conclusion rules such as ‘because disability describes a medical property of individuals, the consequences of disability should be addressed on an individual basis’ and ‘because disability is caused by bodily impairment, attention should be directed towards remedying or ameliorating the effects of these impairments’. Such rules of thought connect a host of particular facts about the socioeconomic marginalization of disabled people with impairments to conclusions about the necessity of charity, welfare and rehabilitation in order to improve their situation.

On the other hand, there are social topoi such as ‘because disability is a property of social situations, attention should be paid to how these situations arise from the organization of society’ and ‘because disabling situations are caused by barriers and institutions, the barriers should be removed and the institutions reformed’. These topoi are grounds for connecting the same observations about the socioeconomic marginalization of people with impairments to conclusions about the necessity of sociopolitical change.

The facts themselves can be integrated with either discourse – the situation of the people in question is identical – but the drive towards political action will run in very different directions. Similarly, recurring medical or social topoi will sustain an identity framework in which disabled people are, respectively, marked primarily by their impairment or primarily by their social exclusion.

Which topoi are employed in the DAA? We've already seen that the law is presented, in its most programmatic sections, as a weapon for securing equality and preventing discrimination – for fixing what is wrong with society:

The purpose of the law is to promote equality and equal worth, to ensure equal opportunity and equal rights, and to prevent discrimination due to impairment. (Barne- og likestillingsdepartementet 2008, 12)

In fact, talking about what is wrong with people is explicitly marked as undesirable. From the AS:

The Department does not support the ranking of different impairments. The safeguards against discrimination will not be reserved for a particular group of people with impairments. The decisive factor is whether discrimination due to impairment has taken place. (Barne- og likestillingsdepartementet 2008, 91)

Elsewhere, the AS argues that disabilities arise when there is a gap between the abilities of the individual and the demands for functional ability in a specific situation (Barne- og likestillingsdepartementet 2008, 90). This is essentially a variant of the Scandinavian relational model, which incorporates both medical impairments and social arrangements. On the whole, however, there has been a shift of attention away from the properties of individuals when definitions of disability are being attempted.

This reflects recent developments, in which the relational model is being increasingly influenced by more strictly social model elements. By way of comparison: in the late 1970s the relational model still contained strong elements of ‘personal tragedy’. The following quote is from a white paper titled On Disabled People in Society:

A person is disabled when he, because of chronic illness, injury or impairment, or deviations of a social nature, is significantly restrained in his practical way of living, relative to the society that surrounds him. This may apply to education and work, as well as physical or cultural activity. (Stortingsmelding1977–1978)

By 1983, however, following the United Nations’ International Year of Disabled Persons in 1981, disabling environmental factors were being taken more strongly into account. The relational model is compatible with both medical and social topoi (because it acknowledges both individual/bodily and social causes of disability) and over the course of the next two decades, it became more explicit about the role of society's demands in producing the disability gap (Tøssebro 2004).

Currently, it seems that social factors are being prioritized more heavily in more prominent and visible sections of public discourse. The DAA intertwines definitions from the gap model with markers of Anglo-American socio-legal discourse, and employs formulations such as these:

It is unacceptable that certain groups have their opportunities and rights in Norwegian society curtailed because they are exposed to discrimination, e.g. in the form of direct and indirect discrimination, harassment, etc., or in the form of discrimination due to lack of accessibility. (Barne- og likestillingsdepartementet 2008, 11)

The pan-disabled experience of discrimination and exclusion is privileged over impairment-specific experiences (for criticism of this privileging, see Thomas (1999) and Wendell (1996)).

Building new arenas for negotiation

What does the absence of medical topoi in the programmatic sections of the DAA mean in policy terms? The act makes clear that the absence of a strict definition of what constitutes an impairment and of a list of recognized impairments is a deliberate choice – a choice motivated by data from other countries. The American and British anti-discrimination laws in particular were interpreted in the courts as having established a relatively narrow class of people to be protected from discrimination, and wound up being significantly less effective in socioeconomic terms than their framers intended (Colker 2005; Krieger 2003). They are cited as examples to be avoided.

The Department, in its attempt to shift focus towards the act of discrimination rather than the subjects of discrimination, cites gender as a category that is comparable to impairment. The implied topoi are ones of social justice or equal treatment before the law, as in ‘if different treatment takes places on the basis of social or biological difference, it counts as discrimination’. The purpose of the Norwegian gender provisions is not to protect the category of women or men, but to prevent discrimination because of gender.

In theory, therefore, the DAA represents a decisive victory for social topoi. It suggests that disability is a neutral identity marker, and that making use of the law does not entail declaring membership of a particularly stigmatized group. But this is clearly an illusion. Let me return to the introductory quote from Tom Shakespeare:

But even in the absence of social barriers or oppression, it would still be problematic to have an impairment, because many impairments are limiting or difficult, not neutral. (Shakespeare 2006, 41)

If real equality is to be achieved – if the full spectrum of provisions in the DAA, both direct and indirect, both individual and systemic, are to be employed – then there is no way to avoid a discussion of specific impairments. There is still a need for medical knowledge and medical topoi.

The social model of disability has been criticized elsewhere for lacking clear policy implications (Samaha 2007). That lack is conceptually related to the absence of impairment-specific discussions. Policy recommendations are frequently a matter of making provisions for people with specific impairments. Almost all positive interventions, and even several negative interventions, differ vastly according to different conditions.

In terms of topoi, it is very difficult to articulate justifications such as ‘because disability correlates with disadvantage X, policy intervention Y is needed’. The problem is that the number of general or blanket disability policy interventions is relatively small. There is always a need for negotiation: the DAA text uses modifiers like ‘reasonable’ in order to qualify measures in every area except harassment. Section 4 states that ‘different treatment which is necessary in order to achieve a valid purpose and which is not a disproportionate intervention for the person or persons being treated differently, will not be considered discrimination under the law.’

The organs made responsible for the adjudication of conflicts covered by the DAA will have considerable influence in their interpretations of grounds for discrimination as well as reasonable accommodation. Chapter 14 of the proposal discusses the institutions that are charged with enforcing the law. These are the Equality and Anti-Discrimination Ombud (a 37-person division within the Ministry of Children and Equality, offering fact-finding, mediation and non-binding rulings), The Norwegian Equality Tribunal (an eight person body capable of ordering fines), and ultimately the courts. The lack of precedents in the area is openly acknowledged: the Anti-Discrimination Ombud itself (on its webpage, under the heading ‘What is a disability?’) notes that ‘it is difficult to define what constitutes a disability until the courts have decided what the term is meant to cover’. This means that the consequences of the DAA will depend on a process of negotiation in which people must come before the courts as disabled people. That is, the law will both shape and be shaped by the public identities of disabled people as they make use of its provisions.

Aside from the courts and public agencies, who will be in a position to negotiate the definitions of disability under the provisions of the DAA? Chapter 12 of the AS discusses provisions under which particular organizations may be provided with the power of attorney on behalf of individuals who want to bring a civil suit. The organization should, according to Section 15 of the DAA, be one which ‘in full or in part works to counter discrimination based on disability.’ This constitutes a circular definition, but also allows for a structure in which any organization that can convincingly argue that it represents disabled people will be in a position to argue cases. However, in the Norwegian system, Section 15 will de facto place a significant amount of both agency and responsibility with disability NGOs such as the Federation of Disabled People's Organization (FFO) and its constituent members.

These organizations will be in a position to demonstrate actual disadvantage. This will require the judicious display and framing of impairment-specific knowledge, and will make the organizations in question key sites for the production of disability-defining discourse. There is, however, already a model of sorts for the connections between impairment-specific and socio-political knowledge, because the NGOs that will participate in negotiating the provisions of the DAA are already obliged to embody such connections. This applies to multiple organizations in the Norwegian civil sector. Because of its size and prominence, and because it is an umbrella organization which organizes 70 impairment-specific organizations, FFO will serve as the best exemplar.

FFO states, in its general policy documents, that it has an anti-discrimination brief:

FFO bases its definition of disability on the understanding of disability as a conflict between the capabilities of the individual to function and the demands for function made by the surroundings. … The conflicts, or matters of discrimination, are experienced differently by different individuals, but are united by the fact that they affect aspects of life taken for granted by most people. … The relations of conflict faced by disabled people therefore centre on the basic interests of human being – human rights. (From FFO's definition of disability )

The federation has traditionally assumed a coordinating role for its member organizations. It has assumed coordination and top-level communication responsibilities, and deals directly with government representatives in negotiating the level of welfare benefits. Many of the individual people that belong to its member organizations, however, do not necessarily consider themselves disabled in a social or even technical-legal sense. They belong to organizations that represent particular medical diagnoses.

The DAA, through its language and its provisions, is establishing a corresponding arena in which socio-legal and medical discourse will be more closely intertwined than before. It is arguably part of an ongoing process in which a new disabled public identity is developed, one that will allow individuals to construe their impairments as socio-politically relevant and strategically functional, but not dominant. In a sense, this is a logical development from the increased public visibility of disabled people across the world over the course of the last decades (or, for that matter, century). It also means that the disability NGOs are partly responsible for bridging the gap between medical and sociopolitical identities.

The public disabled identity

What is the future of the public disabled identity? In Norway, as in many other countries, it is already a hybrid identity that implies both demanding one's rights and claiming welfare state provisions. It contains elements of stigma and pride, depending on the context. By declaring oneself to be disabled, a person will effectively, at the very least, do one or more of the following:

  • 1.   Identify as a member of a class that is deemed vulnerable (under welfare state regulations).
  • 2.   Identify as a member of a class that suffers from discrimination (under the DAA).
  • 3.   Identify as a member of a class that suffers stigma and prejudice (under the DAA, where ‘discrimination due to presumption of impairment’ is deemed grounds for action).

Awareness of the effects of (3) is of course fully established with disabled people, along with the knowledge that the public disabled identity is a deeply problematic one. Generally speaking, disability correlates with disadvantage, because it is by any definition linked to a negative deviation from a norm or standard of functions of the body. In this way, it is fundamentally different from other identity categories associated with stigma and prejudice, for example, gender and ethnicity.

The link between disability and negative deviation is problematic, i.e. identity-threatening for many individuals. Some data suggest that there is a threshold of declaring one's disability, of admitting that something is wrong, which has to do with being sufficiently inconvenienced by one's impairment (Shakespeare 2006).3 People who are able to ‘pass’, that is, who see themselves as capable of avoiding the imposition of a disabled identity, may choose to do so as long as the benefits of avoiding stigma outweigh the benefits that would be gained through a disabled identity.

The threshold of declaration will obviously vary in height in different societies. The factors that influence that decision lie outside the scope of this article, but let me suggest a few candidates:

  • 1.   Stigma;
  • 2.   direct or indirect discrimination;
  • 3.   impairment effects such as pain and fatigue;
  • 4.   degree of adaptation in the physical environment in general, including access to public transportation;
  • 5.   welfare benefits;
  • 6.   anti-discrimination measures; and
  • 7.   the right to individual accommodation in education and employment situations.

While the confluence of significant impairment effects, strong welfare provisions and a low level of public stigma might encourage ‘coming out’ as disabled, the confluence of a high level of stigma, weak anti-discrimination measures and manageable impairment effects would encourage trying to ‘pass’–pass, alternatively invoking the category of temporary illness rather than permanent impairment (cf. Goffman 1963). Passing remains a legitimate goal for many disabled people, and it is probably no accident that the most strident activists for disability rights are frequently people who have minimal, if any, chances of passing whatsoever.

The previous view holds particularly true when disability is conceived as a global or dominant identity. However, there is another way to think about the threshold at which people will decide to pursue a disabled identity, a way that: (a) views that identity primarily as a situational, strategic asset; and (b) supports that view by embedding disability knowledge deeply in various social fields. The inherent negative connotations of disability and its synonyms – the connotations of negative deviation – may, perhaps, be countered through the strengthening of sites in which specific impairments may be made temporarily relevant.

An example: some airlines, concert venues, cinemas etc. allow its customers to specify that they have impairments. Some provide only for a general declaration of disability. The first option is often unsatisfactory because there is insufficient information about the facilities in question, while the second is all but useless, since it inevitably requires additional information from the patron or passenger. However, when the company in question chooses to embed impairment knowledge in its booking systems, allowing people to specify whether they wanted a seat or a wheelchair parking spot, whether the venue has telecoil capacity, and so on, a number of problems go away.

The previous example can be extended mutatis mutandi to most of the areas in which laws such as the DAA are intended to affect changes. The implementation of the goals of the DAA will hinge on the willingness of people with impairments to pursue their means of legal recourse. This in turn depends on their assuming the law's notion of a disabled identity. This identity must be equipped with provisions that constitute a real form of protection for those who need it the most, but it cannot be constructed in such a way as to de-motivate or exclude people who might gain comparatively minor but nevertheless significant benefits.

The advisory statement of the DAA consistently frames disability as a consequence of direct or indirect discrimination. Its express purpose is to strengthen the legal safeguards against such discrimination. At the same time, the people who are to benefit from these safeguards are identified as those who suffer discrimination because of having (or being presumed to have) medically diagnosed impairments or chronic illness. In other words, the cause of the problem of discrimination is framed in terms of medical discourse, while the solution is framed in terms of socio-legal discourse.

Conclusion: a paradox of disability discourse

The paradox of disability discourse consists of the following: in order to benefit from sociopolitical anti-discrimination measures, a person must first self-identify as being medically impaired. The fight against discrimination becomes to some extent a reification of the basis for that discrimination. If there's nothing wrong with you, how can you identify society's particular shortcomings?

Sociopolitical disability activism can therefore achieve its goals only by: (a) appropriating the medical knowledge that it has historically strived to suppress, embedding it in particular policy fields; and (b) actively redefining discrimination so that it effectively matches policy needs. The work of (a) is, in Norway as elsewhere, partly being done by the ever-increasing number of impairment organizations, for whom education and the spread of information are high priorities. The achievement of (b), however, raises considerable long-term problems. How can the concept of discrimination be made thoroughly relevant to disability? In the DAA, the purpose of the law is stated as being equality and the prevention of discrimination:

The Department advises that the purpose of the law shall be to ensure equality and equal worth, ensure equal opportunities for and right to participation in society for all, independently of functional ability, and to prevent discrimination on the basis of impairment. (Barne- og likestillingsdepartementet 2008, 76)

Prevention of discrimination, then, does not necessarily secure true equality. That is a matter of greater social change, which must take place in every area of society. Long-term social change depends on the wide dissemination of both the sociopolitical and medical aspects of disability. This means destabilizing the category further in both situational terms (allowing for the fact that people are disabled relative to some, but not all activities, situations etc.) and essential terms (allowing for the fact that some impairments may legitimately be considered curable diseases).

This presents a challenge with respect to both classically medical and strictly social definitions, because a weakening of the exclusion-based, social definition of disability identity may lead to a resurgence of older, essentialist medical categories. This will in turn strengthen the notion that disability issues may mostly be resolved on an individual level, by means of medical intervention.

Disability should, in discourse terms, be linked as closely as possible to other forms of knowledge. The more actively sociopolitical disability provisions are included in, for example, architecture, urban planning, social science, and, indeed, medicine, and particularly in the educational programmes associated with those fields, the less problematic the public disabled identity becomes.

Disability studies, then, is a field ultimately working towards its own dispersion, and it must, crucially, be an interdisciplinary one. Much in the same way that a disabled identity should be something that can be voluntarily assumed in situations in which it is strategically useful, disability discourse should attach itself to as many other forms of discourse as possible, in order to extend awareness of the true range of human variation.