Disability as a category: the medical model, the social model and their critics

Ask a design team why a booking form excludes a wheelchair user. The modern answer is environmental: the form was built for the wrong body, and the body is not the problem. That answer is the social model, and it has won.

This first of ten units asks what the category means, and the nine that follow apply it. The model was built to move attention from the body to the barrier. It left the field without a settled vocabulary for the parts of disability that are not barriers. A tremor makes a target hard to hit. Chronic pain ends a session early. Neither is an artefact of a badly built world, and the model that made them awkward still governs practice. The field now runs on a middle position it rarely states. Disability is produced where a body meets a built world, and the pure social model survives mainly as a slogan from a fight already won.

The founding distinction: impairment is one thing, disability another

The social model begins with a definition. In 1975 and 1976 the Union of the Physically Impaired Against Segregation set out the distinction on which the tradition rests:

In our view, it is society which disables physically impaired people. Disability is something imposed on top of our impairments, by the way we are unnecessarily isolated and excluded from full participation in society. Disabled people are therefore an oppressed group in society. (UPIAS, 1976)

Two moves are packed into that passage. The first separates impairment, a bodily state, from disability, a social relation, so the two need not travel together. The second names the resulting position as oppression rather than misfortune, which turns access from a private trouble into a public claim. Oliver (1990) made the distinction systematic: disability is produced by the social and economic organisation of society rather than by the body. The reframing converts a design failure from a problem that lives in a person into a decision that lives in a system.

The model turned barrier removal into a design task

If disability is in the environment, the environment can be changed. Barrier removal becomes a product requirement: a step becomes a ramp, a small target a larger one, a missing label a described image, a page that fails on a slow connection a lighter one. The model also supplies the argument for involving disabled people rather than treating them as subjects. Mankoff, Hayes and Kasnitz (2010) show both the pull and the gap. They review the key literature of disability studies for an assistive technology audience, and they argue the two fields share goals but not problem definitions. Assistive technology therefore keeps reproducing a medical view of the people it builds for.

The model was asked to explain more than barriers

The criticism is not that the distinction is false. Shakespeare and Watson (2002) set out three objections to the British social model. It cannot account for the experience of impairment. The impairment and disability dualism is unstable. Its account of identity is too thin. Anastasiou and Kauffman (2013) argue from philosophy that reducing disability to a single social dimension downplays biological and mental conditions. A political account taken as a complete one can end up denying the needs it was meant to serve. Hviding (2013) presses a related case against critical disability studies. He asks what its ideas can carry when put to ethical and political use.

The response from inside the tradition is not a retreat to the medical model. Siebers (2008) proposes complex embodiment. On that account the effects of a disabling environment and the effects that derive from the body are both real and reciprocal. Kafer (2013) rejects the medical model because it imagines only cure or failure. She rejects the impairment and disability distinction as unhelpful, because what counts as an impairment shifts with context. She proposes a political and relational model in which built environments and social patterns produce disability. Both keep the model's political commitment and drop its silence about the body. Siebers's instance is the person whose chronic pain or fatigue, which ageing makes commoner, is not caused by a barrier and is not removed by redesigning one.

The field runs on a middle position, and writes it into its instruments

Read together, these writers converge on a position the field uses without naming. Thomas (2004) points out it is not new: the social relational understanding developed by Finkelstein and Hunt in the 1970s was overtaken by its own offspring, the social model, and she recommends reviving it. Her point is methodological as much as political, because the definition a researcher starts from determines which questions can then be asked. Shakespeare (2006) developed the critique into a systematic position from a critical realist starting point, arguing that the social model had reached a dead end and that disability is best treated as the product of an interaction between a body and a society, both treated as real. It is the position most of the field's own writers now hold, and the one against which design claims should be tested. Reindal (2008) shows what the position does where a field cannot avoid categories. Special needs education has to name a pupil's functional difficulty to build an individual education plan, yet the social model's critique makes that naming look like the labelling it condemned. She defends a social relational model that keeps oppression in the frame and still allows the need to be identified. The test of a model then becomes what it lets a practitioner do. The reason to care which model is in play is that models travel. Bickenbach and colleagues (1999) review the frameworks of disablement and the instruments built from them. They argue for the universalist account the WHO classification later embodied as the biopsychosocial model. A model written into an instrument survives the argument that produced it.

Contested question: is the social model a failed theory or a tool misread as one?

The literature disagrees about what to do with the model. One position holds that it over-reached and should be replaced by a relational account. Shakespeare and Watson (2002) argue impairment must be brought back into the frame. Anastasiou and Kauffman (2013) hold that the separation of impairment from disability is conceptually unstable, and that it can harm people with significant impairments by denying their needs. Hviding (2013) adds that the tradition's later normative claims outrun what a political account can support. Siebers (2008) and Kafer (2013) supply the constructive alternatives. On this reading the model's silence about the body is a substantive failure.

The opposing position holds that the model was never a theory of impairment, so the charge mistakes its function. Oliver (2013) restates what he meant it to be, a tool for improving the lives of disabled people, and argues the criticisms have had damaging consequences at a time of cuts to benefits and services. Thomas (2004) supports a version of this reading. On her account the relational understanding existed before the model, and the model did not replace it so much as eclipse it.

The two sides rest on evidence of different kinds. The replacement case rests on conceptual argument, namely three named objections to a definition and two accounts of embodiment built from the experience of chronic pain. The tool case rests on the author's own statement of intent and on a reading of the tradition's history.

The strongest objection is that the unit's middle position carries no test. A relational account can absorb almost any finding, because the body and the barrier can each be named after the fact. A claim that cannot fail is not yet a theory a designer can put to work.

What would settle it is not another exchange of definitions but evidence about consequences. The testable question is whether decisions made on a barrier-removal frame serve people with significant impairments as well as decisions made on a relational frame. Bickenbach and colleagues (1999) offer a partial lever, because instruments built on one framework or the other produce different counts and support different allocations. The literature has not run that comparison.

What to do while the dispute is open

While the dispute is open, name the model the design rests on and the people it leaves out. Where a need is not a barrier, record the need beside the barrier removed, and test the result against an ageing hand or a reader in a second language.

Where the argument is heading

Shew (2020) argues ableism shapes technological imagination. She coins technoableism to name the assumption that disability is a defect to be engineered away. That assumption now shapes artificial intelligence, where the body and mind to be optimised are imagined before any user is met. Spiel, Gerling and Bennett (2020) point at the profession rather than the product. They argue disabled scholars remain marginal in the research community that studies them. A claim about disability as a category is finally a claim about who is allowed to make it. The next unit asks what a single design can deliver for a category this contested.

The sources this unit rests on

• Union of the Physically Impaired Against Segregation (1976). Fundamental Principles of Disability. London: UPIAS. https://disability-studies.leeds.ac.uk/wp-content/uploads/sites/40/library/UPIAS-fundamental-principles.pdf Records the 1975 discussion between UPIAS and the Disability Alliance and the Union's founding statement. It supplies the founding distinction between impairment, which is bodily, and disability, imposed by society.
• Oliver, M. (1990). The Politics of Disablement. London: Macmillan. doi:10.1007/978-1-349-20895-1 Argues disability is produced by the social and economic organisation of society rather than by the body, and that disabled people are an oppressed group. It founded the British social model and moved attention from the individual to the barrier.
• Bickenbach, J. E., Chatterji, S., Badley, E. M. and Ustun, T. B. (1999). Models of disablement, universalism and the international classification of impairments, disabilities and handicaps. Social Science & Medicine. doi:10.1016/s0277-9536(98)00441-9 Reviews the models of disablement and their instruments, and argues universalism serves disabled people better than a minority-group approach. It shows the revised international classification embodied a biopsychosocial synthesis, so model choice survives inside the instrument.
• Shakespeare, T. and Watson, N. (2002). The social model of disability: An outdated ideology? Research in Social Science and Disability. doi:10.1016/s1479-3547(01)80018-x Sets out three criticisms of the British social model, over impairment, the impairment and disability dualism, and identity. It proposes an embodied account and argues the time has come to move beyond the model.
• Thomas, C. (2004). How is disability understood? An examination of sociological approaches. Disability & Society. doi:10.1080/0968759042000252506 Compares sociological accounts and finds a social relational understanding across disability studies and medical sociology. It holds the relational understanding of Finkelstein and Hunt was lost as the social model rose, and calls for its revival.
• Reindal, S. M. (2008). A social relational model of disability: a theoretical framework for special needs education. European Journal of Special Needs Education. doi:10.1080/08856250801947812 Argues for a social relational model of disability as the platform for special needs education, keeping oppression central while still allowing a pupil's needs to be identified without the harms of labelling. It shows the relational position at work in a field that cannot avoid categories.
• Shakespeare, T. (2006). Disability Rights and Wrongs. London: Routledge. doi:10.4324/9780203640098 Argues from a critical realist position that the social model has reached a dead end and promotes a pluralist account. It criticises the polarities of medical against social, impairment against disability, and disabled against non-disabled.
• Siebers, T. (2008). Disability Theory. Ann Arbor: University of Michigan Press. doi:10.3998/mpub.309723 Argues the social model needs a theory of complex embodiment to account for chronic pain, secondary health conditions and ageing. It treats the relation between social representation and the body as reciprocal.
• Mankoff, J., Hayes, G. R. and Kasnitz, D. (2010). Disability studies as a source of critical inquiry for the field of assistive technology. Proceedings of the 12th international ACM SIGACCESS conference on Computers and accessibility. doi:10.1145/1878803.1878807 Reviews the key literature of disability studies for an assistive technology audience and argues the two fields share goals but differ in what they treat as a problem. Two case studies show how that literature changed the problems the authors chose.
• Anastasiou, D. and Kauffman, J. M. (2013). The Social Model of Disability: Dichotomy between Impairment and Disability. Journal of Medicine and Philosophy. doi:10.1093/jmp/jht026 Critiques the social model's basic claims, arguing that reducing disability to a single social dimension downplays biological and mental conditions. It warns that denying those realities can harm disabled people.
• Hviding, K. (2013). Moral wrongs, disadvantages, and disability: a critique of critical disability studies. Disability & Society. doi:10.1080/09687599.2013.831751 Reviews the main approaches of critical disability studies and examines the normative issues it raises. It questions the ethical and political applicability of the approach rather than its descriptive claims.
• Kafer, A. (2013). Feminist, Queer, Crip. Bloomington: Indiana University Press. doi:10.2979/6841.0 Argues disability should be understood as political and offers a political and relational model in which the problem lies in built environments and social patterns. It rejects the impairment and disability distinction because what counts as an impairment shifts with context.
• Oliver, M. (2013). The social model of disability: thirty years on. Disability & Society. doi:10.1080/09687599.2013.818773 Reviews three decades of the social model and restates what the author meant it to be. It defends the model as a tool for political action rather than a complete account of experience, and warns the criticisms have had damaging consequences at a time of cuts.
• Shew, A. (2020). Ableism, technoableism, and future AI. IEEE Technology and Society Magazine. doi:10.1109/MTS.2020.2967492 Argues ableism shapes technological imagination, and coins technoableism to name the belief that disability should be engineered away. It shows those expectations shaping what is designed and how artificial intelligence is imagined.
• Spiel, K., Gerling, K. and Bennett, C. L. (2020). Nothing about us without us: investigating the role of critical disability studies in HCI. Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems. doi:10.1145/3334480.3375150 Takes stock of the role critical disability studies plays in HCI research, ten years after Mankoff called for disabled people to be involved. It aims to identify the systemic obstacles that keep disabled scholars at the margins.