The ‘disabilitization’ of medicine: The emergence of Quality of Life as a space to interrogate the concept of the medical model

Criticisms of medicine in the humanities and social sciences

The way in which western medicine ‘treats’ disease and disability has long been studied in fields such as medical anthropology, medical sociology, and disability studies, oftentimes from a critical perspective. For instance, following Arthur Kleinman’s (1988) classic distinction between ‘disease’ and ‘illness’, a particular strand in medical anthropology has shown us how people experience illness and suffering in daily life as distinct from disease in the clinic. Medical sociology has introduced the notion of ‘medicalization’ (Zola, 1972) to think through the ways in which non-medical domains of life have been brought under the jurisdiction of medicine. This has in turn paved the way for other concepts, such as ‘biomedicalization’ (Clarke et al., 2009), that have furthered our understandings of biomedical knowledge production. The strongest criticism of medicine, however, has come from within disability studies, as part of its ‘demedicalization’ of disability (Kasnitz and Shuttleworth, 2001). As disability scholar and activist Simi Linton writes, disability studies ‘arose in part, as a counterpoint to the medicalized perspectives of disability emanating from the applied fields’, such as health and occupational therapy, which resulted in what Linton calls ‘Not Disability Studies’ (1998: 132–3).

What are the epistemological differences between the ways in which medicine and disability studies define disability? ¹ First, medicine locates disability within the individual body, its diseases, its ‘lacks’, and its ‘abnormalities’, which it then subjects to treatment. Disability studies and politics, in contrast, insist that disability is a problem of society: disability emerges from the discriminatory attitudes, oppression, and barriers of a disabling society, not from impairments of the body. Second, medicine uses statistical methods and standards that situate the healthy/abled/sane body as normal, against which those falling outside that arbitrary median range are defined as ‘pathological’, ‘deviant’, ‘aberrant’, disabled. Disability studies, in contrast, theorizes disability as an aspect of human variation to be embraced, a resourcefulness to be learned from, and a socio-political identity to be celebrated. These differences constitute the binary framework of critical approaches to medicine in disability studies: the ‘medical’ (or ‘individual’) versus the ‘social’ model of disability (and, in conjunction with this, impairment versus disability). ² In this binary, medicine pathologizes – and thus individualizes – disability, while the social model focuses on the social causation of disability.

While feminist (Crow, 1996) and phenomenological (Hughes and Paterson, 1997) critiques of the impairment/disability binary, as well as critiques of the social model (Shakespeare and Watson, 2002) have proliferated since the binary was formulated, understandings of the medical model remain rather ossified. More recently, scholars have begun to identify problematic consequences of the social/medical framework. Tom Shakespeare, for example, argues that the medical model has become ‘a proxy for all that is wrong with traditional attitudes to disability’, from medicalization, to professional authority, to objectification, ableist ideas, and paternalistic research methods and practices (2006: 18). The result is that this ‘powerful symbol’ has become ‘nothing but a straw person’ (ibid.). In other words, the medical model risks reducing highly differentiated practices to a monolithic entity. The same is true of the ubiquitous use of related concepts, such as medicalization and the biomedical model of disease, in social science critiques of medicine. These too run the risk of reifying the conception of medicine.

In this article, I seek to explore the dangers of this ossification as a barrier to effective criticism, through approaches to medicine developed in the field of science and technology studies (STS). In particular, this (sub)field has long argued that medicine is not a monolithic entity, but comprises heterogeneous and often conflicting sets of practices that generate multiple objects, even though they are referred to medically as a single condition or disease (see Berg and Mol, 1998; Mol, 2002). This argument suggests that, as critics of medicine, we can become trapped in medicine’s own epistemologies. For instance, Mol argues that, following Foucault, we could instead look for ‘noncritical strategies for escaping dominant ways of thinking’, and suggests that ‘a good way to escape from a medicine founded on pathology [might be] to wonder whether, in practice, medicine is indeed founded on pathology. This implies that instead of criticizing pathology’s foundational role, we raise questions about it, we doubt it’ (2002: 47). Taking Mol’s suggestion as a point of departure, I ask: what if we do not take the medical model for granted but instead doubt it? Has medicine ever been a singular practice, with definitive and frictionless objects of knowledge, as the medical model assumes it to be? Or is medicine composed of heterogeneous practices that create multiple, unstable, and evolving objects, even if they go by the same name (such as ‘disability’ or a specific ‘disease’)? Does medicine, in practice, actually apply the medical model? Or is the model an invention of social scientists as they sought to know how medicine knows its objects of knowledge? ³ What strategies can we develop to escape dominant ways of thinking, other than the criticism offered by the medical model, which has perhaps become too dominant a criticism and may prevent us from exploring medicine’s evolving and multiple articulations of disease and disability?

There can be many ways to ‘doubt’ the medical model. In this article, I take as my entry point the emergence of QoL as a ‘matter of concern’ (Latour, 2004) in healthcare. QoL emerged within medicine and healthcare in the 1970s, and has since coalesced into a discourse in its own right. This history, I argue, gives us a more complex picture of medical knowledge than that offered by the model. First, however, I offer a brief account of where and how I encountered QoL, working at the intersection of disability studies, medical anthropology, social studies of medicine, and my own experience of chronic illness.

My encounter with QoL measures in the clinic

In an ethnography of invisible disabilities that I undertook in 2009–10, I worked with people living with disabilities related to rheumatoid arthritis (RA) – a disease that I also have – and filmed them performing daily household tasks. In the process, I observed that my participants came up with an incredibly creative set of survival techniques in their everyday routines. To think through these encounters, I engaged with James Gibson’s theory of affordances (1979) in ecological psychology, which considers the action possibilities that emerge from the relation between an organism and its environment. In developing an entirely new theory of affordances, informed by a critical disability and performance lens, I proposed the concept of ‘micro-activist affordances’. This term refers to the way in which the experience of disability can become a way of forging new organism–environment relations, of improvising creative affordances, which would not have been imaginable outside the experience of disability (Dokumacı, 2017).

My ethnographic research coincided with the time that I, as an RA ‘patient’, was asked to fill in various questionnaires at the clinic. These asked me to rate the severity of my difficulties, with such questions as ‘over the last week, were you able to dress yourself, including tying shoelaces and doing buttons?’ (HAQ), and during the past four weeks, ‘have you felt downhearted and blue?’ (SF-36). The questionnaires were designed to measure the outcomes of a medical intervention according to patients’ perceptions, including the ease with which they performed their activities of daily living (ADLs); their social and emotional functioning; their levels of pain and fatigue; their mood; and their overall health, well-being, and happiness. This contrasted with traditional ways of assessing health outcomes, which use clinical markers, laboratory and radiological results, mortality rates, and survival times.

As I engaged with QoL measurements as a ‘patient’, I was struck by a paradox. My instinct, as a scholar working at the intersections of disability studies and medical anthropology, was to find these instruments reductionist, seeking to represent on a five-point scale the heterogeneity of micro-activist affordances that I had studied. Still, they did not conform to traditional models of assessment either. These instruments did not seem to reduce disability to a disease or a pathology, but looked at how diseases manifested themselves outside the skin. They marked a rupture from traditional biomedical indicators of health and the broader medical model of disability. It was precisely this mismatch that led me to ask: how has medicine ended up developing measures for entirely subjective perceptions about the entirely ‘non-scientific’ phenomenon of everyday living? What does the medical model have to say about the emergence of a discourse that makes patients’ everyday experiences of living with diseases and treatments into a matter of formalized medical concern known as Quality of Life? What can the discursive formation of QoL tell us about the enunciative regularities and limits of concepts such as the ‘medical model’ that we are so familiar with? Might the emergence of QoL be a way to de-familiarize the familiar criticism?

QoL analysed from a social studies of medicine perspective

Even though QoL is a relatively new concept in medicine, various social studies of medicine scholars have taken it up (see Armstrong, 2009; Armstrong and Caldwell, 2004; Armstrong et al., 2007; Dokumacı, 2014; Wahlberg, 2018; Wahlberg and Rose, 2015). In their genealogy of the concept of health-related quality of life (HRQoL), Armstrong and his colleagues argue that ‘the conceptualisation and measurement of Quality of Life began to change the relationship between symptom and illness that had dominated the discourse of clinical practice since the 19th century’ (Armstrong et al., 2007: 581). After tracing a number of processes – from interwar developments in questionnaire technology to the postwar proliferation of symptoms checklists, pain questionnaires, and ADLs, and the eventual condensation of these domains into HRQoL – they note that ‘symptoms increasingly detached themselves from their pathological anchor and began new attachments to aspects of the patient’s psychosocial world’ (ibid.) – becoming what the authors call ‘distal symptoms’ (ibid.: 575).

Similarly, anthropologist Ayo Wahlberg examines a set of practices, including patient schools tailored to create ‘expert patients’, practical ‘living with…’ guides, and clinical trials measuring QoL, and the way in which these formulate life ‘not as an anatomical, cellular or molecular affair’, but as ‘something that is lived [and] experienced’ (2009: 166). He proposes the concept of ‘knowledge of living’, which comes from the study of ‘how it is to live with disease’ through the very methodologies used by medical anthropologists, as a methodologically distinct category from the biological ‘knowledge of life’, which involves the study of cells, molecules, organs, and DNA structures (2018: 729-30).

Both ‘distal symptoms’ and ‘knowledge of living’ are useful in thinking through transformations in medical perceptions since the emergence of QoL. Using the insights provided by these two concepts to push towards a ‘critique of a critique’, I propose that the historical emergence of QoL can be considered a process of what I term the disabilitization of medicine. ⁴ The ‘disability’ in disabilitization is, to be sure, not the same ‘disability’ as in disability studies (which is itself multiple). But, nor is it the ‘disability’ of which the medical model is presumed to be a model. Indeed, this is exactly the point I seek to foreground with the concept of disabilitization.

Disabilitization, in the way I propose the concept, is a way of historicizing QoL and thereby demonstrating that, as Mol and others have shown for other medical objects, QoL (and its conception of disability and disease) is or can be more heterogeneous than a single ‘model’ can address. Disabilitization, in this sense, refers to the way the medical model becomes disrupted and disarmed when confronted with histories, such as the emergence of QoL, that do not neatly fit within its binaries. These disruptions provide a basis for understanding, outside of them, the model’s (somewhat) overworked critique. In brief, to consider the emergence of QoL as a process of disabilitization is to consider medicine not as one thing, but many, and how attending to that plurality may keep us from over-consolidating or empowering medicine in the first place. In other words, rather than reproducing the medical model as an object of criticism, the notion of disabilitization is intended to create a space for ‘doubt’ in our use of critical concepts that have become (perhaps) too familiar.

Trajectory

Available histories suggest that early health status assessments paved the way for contemporary QoL measurements (see Bowling, 2001; McDowell, 2006; McHorney, 1997; Prutkin and Feinstein, 2002). These antecedents include symptoms checklists, pain questionnaires, ADLs, and visual analogue scales (see Armstrong et al., 2007). There are also various assessments of function that laid the groundwork for QoL (see Prutkin and Feinstein, 2002). These include: a four-grade categorization of disability to assess the medical needs of old-age assistance recipients in New York City (State of New York Department of Social Welfare, 1937); the PULHEMS system of functional classification developed by the Canadian Army during the Second World War to crossmatch available manpower with the range of military tasks necessary to achieve a ‘manpower economy’ (Canadian Army, 1943); a five-grade functional classification developed at the Home for Aged and Infirm Hebrews (New York City) to ‘give a complete picture of the condition’ of its residents (Zeman, 1947: 723); a four-grade classification of functional capacity to diagnose heart disease (Bainton, 1928); a four-grade functional classification for RA (Steinbrocker, Traeger, and Batterman, 1949); and a performance scale developed during a chemotherapy trial with nitrogen mustard in the aftermath of the Second World War (Karnofsky et al., 1948). In terms of the objects they generated, these instruments can all be considered precursors to QoL. I have chosen to focus on the last three, as these were developed strictly under medical auspices and were among the first precursors to QoL. ⁵ Whereas the first three assessments of function (emerging from welfare administration, military practices, and institutional geriatric care, respectively) might more readily relate to everyday living, the last three emerged from clinical research and practice. If one were to follow the axioms of the medical model, these would be the most likely to equate disease with mere pathology, lacking any consideration of patients’ everyday living with disease (hence reducing disability to a problem of pathology). And yet even at this point of origin, we find a disruption to a straightforward application of the medical model – a disruption that I seek to make evident here.

This article comprises four sections. First, I present a Foucauldian analysis of the three selected case studies. I then trace the history in which QoL was made into a ‘matter of concern’ (Latour, 2004) in both public life and healthcare. Third, I introduce the concept of disabilitization to describe the emergence of QoL as a new way of articulating disease and disability in medicine. Finally, I discuss how disabilitization can provide a space for thinking beyond the medical model.

Case study 1: Classification of functional capacity in patients with heart disease, 1928

In 1928, the Heart Committee of the New York Tuberculosis and Heart Association published Criteria for the Classification & Diagnosis of Heart Disease, which expanded the nomenclature for cardiac disease published in 1923. The foreword of this text identified ‘the establishment of definite criteria for diagnosis’ (Bainton, 1928: vi) as the next step in the development of the field, and highlighted ‘the fundamental difficulty…found in the definition of the diagnosis [of heart disease] itself’, as the same diagnostic term might mean different things to different physicians at different times and places (ibid.: ix). To navigate this ambiguity, the committee presented four types of criteria, each explained in a separate chapter. The chapters enumerated ‘etiological criteria’, which ranged from hypothyroidism to neoplasm; ‘anatomical criteria’, which included diseases of the aorta and pulmonary arteries, of the myocardium, of the endocardium and valves, and of the pericardium, as well as their corresponding signs and symptoms; and ‘physiological criteria’, the main categories of cardiac physiology. After 73 pages of dense medical terminology, the book went on to introduce its final criterion, ‘functional capacity’:

At the present time, there is no clinical test which will accurately measure the functional capacity of the heart. This section of the diagnosis refers, then, to the functional capacity of the cardiac patient, as modified by his cardiac disease. Only an approximate estimate of this functional capacity is possible, and the most useful guide is found in the patient’s ability to perform physical activity. (ibid.: 87; emphasis added)

While this ability to perform daily activities was a new criterion in the diagnosis of disease, it needed to be estimated using traditional clinical methods: by taking ‘a careful history of the patient’s symptoms on effort’, created by asking how ‘walking on the level or up a grade’, ascending stairs, or running affected the patient, and, where necessary, by directly observing the patient perform the exercise (Bainton, 1928: 88–9). When it came to rendering this knowledge enunciable, analysable, and measurable, however, something new was introduced into the diagnostic process: a four-grade classification used to rate the patient’s capacity to perform daily activities. ⁶ Just as earlier chapters mapped out anatomical criteria and ways of knowing their signs (such as tapping or listening), functional classification was presented as the diagnostic means of making functional capacity analysable and measurable. Through the production of a classification table, the patient’s ability to perform daily activities became as significant as clinical signs in the diagnosis of a disease.

In their concept of ‘distal symptoms’, Armstrong and his colleagues discuss how ‘the clinical gaze (Foucault, 1973), which for over a century had been firmly fixed on the pathological lesion, began to form new structures of perception, of organising and thinking about the nature of illness’ through the emergence of new tools for health assessment. In ‘pathological medicine’, pain in RA, for instance, would be a clinical symptom, a proximal indicator of an underlying pathology (ibid.: 575). With the emergence of ADL scales, however, the ‘inability to climb stairs’ or perform other daily activities became symptoms in and of themselves – symptoms as ‘more downstream effect[s] of the disease’, rather than ‘immediate manifestations of pathology’ such as pain (ibid.). Armstrong et al. locate the emergence of distal symptoms in the postwar period (particularly the 1970s onwards), when QoL was consolidated into a formal concept. I argue that this consolidation can be traced to a much earlier period, to the classification of functional capacity that I have just described (even though it used clinical methods to assess functional capacity rather than the patients’ own formalized estimates). It is precisely in the definition of this new diagnostic criterion, and its operationalization by way of a table, that we can identify ‘distal symptoms’. The functional capacity of the heart is not rendered enunciable through the workings of ‘the clinical gaze’, which looks at the ‘tangible space of the body’ to find hidden ‘secrets, invisible lesions, and the very mystery of origins’ (Foucault, 2003[1973]: 150). Instead, this capacity becomes knowable through its effects on the patient’s everyday life.

In defining the patient’s ability to perform activities of daily living as a diagnostic criterion in itself, and in instituting a classification system that rendered this criterion enunciable, the Criteria extended symptoms beyond the envelope of the skin to the realm of everyday life. When a functional criterion is added to the etiological, anatomical, and physiological criteria, the clinical gaze extends beyond the inner workings of the body to determine what this body can do in life. Moreover, functional capacity is not merely an add-on to ‘objective’ criteria (etiological, anatomical, or physiological) – it is a separate set of criteria that renders its own particular diagnostic information. As the Heart Committee wrote, functional classification ‘should not be influenced by the anatomical diagnosis or by the prognosis’, and ‘should depend solely on the functional capacity of the patient at the time of the examination’ (Bainton, 1928: 87). ⁷ Once functional capacity had been established as a new way of looking at diseases, and a diagnostic system had been put in place, it would be taken up by others (see Zeman, 1947: 721–2).

Case study 2: Classification of functional impairment in Rheumatoid Arthritis, 1949

The Criteria discussed above delineated patients’ ability to perform daily activities as both diagnostic and therapeutic criteria. This report did the opposite. In its recommendations for uniform therapeutic criteria for RA, the Classification considered ‘subjective symptoms’ as ‘unreliable’ (Steinbrocker, Traeger, and Batterman, 1949: 662), and noted that while functional capacity often correlated with disease activity, it ‘may vary considerably in spite of an unaltered rheumatoid process, or as a result of such different procedures as orthopaedic measures, physical therapy, psychotherapy, and many others which improve function without altering the activity of the disease’ (ibid.: 660). The report emphasized the importance of ‘distinguish[ing] between those therapeutic agents which show measurable objective effects and those which only influence subjective and/or functional features of the disease’. It concluded that ‘for that reason especially, the criteria must be based entirely on objective evidence’ (ibid.).

If functional capacity and subjective experiences were deliberately excluded from its treatment evaluation, why is the Classification of interest? The answer lies within the broader ‘system of classification and evaluation’ (Steinbrocker, Traeger, and Batterman, 1949: 662) in which this therapeutic criterion was to be incorporated. Like the committee on cardiac diseases, the committee on RA began by identifying ‘the manifest difficulty inherent in therapeutic evaluation in any disease of unknown causation with no specific treatment’ (ibid.: 659) alongside other confounding factors that impeded the evaluation of results, notably subjective factors, especially pain, psychogenic and psychological influences, and patients’ level of functioning. To provide a standardized process of treatment evaluation, the report proposed a four-grade system of therapeutic classification based on objective information only, namely clinical, laboratory, and roentgenic evidence. However, the authors then added that,

in the course of the Committee’s efforts to arrive at practical therapeutic criteria, it became increasingly clear that the effective use of such standards requires agreement on other preliminary considerations. These have been designated as supplementary aids to the therapeutic criteria. They consist of a definition of rheumatoid arthritis, a classification of the stages of rheumatoid arthritis, and a classification of functional impairment. (ibid.: 660)

OPEN IN VIEWER

Table 1.

Classification of functional capacity.

Class
I	Complete Ability to carry on all usual duties without handicaps
II	Adequate for normal activities Despite handicap of discomfort or limited motion at one or more joints
III	Limited Only to little or none of duties of usual occupation or self care
IV	Incapacitated, largely or wholly Bedridden or confined to wheelchair; little or no self care

Note: Steinbrocker, Traeger, and Batterman, 1949: 660. Reproduced with permission from Journal of the American Medical Association. Copyright © 1949 American Medical Association. All rights reserved.

Case study 3: The Karnofsky Performance Scale, 1948

The final case study belongs to an experimental chemotherapy trial undertaken at the Sloan-Kettering Institute for Cancer Research (SKI) in New York. ⁸ Historians of medicine note that until the beginning of the Second World War, chemotherapy was considered less scientific than surgery or radiation, and was often likened to quackery (see Bud, 1978: 440; Gaudillière, 2009: 498). In the wake of the war, this situation began to change, particularly in the United States. Chemotherapy afforded ways to connect experimental investigations with clinical applications – a strategy well suited to the postwar enthusiasm for ‘organized science’ (Bud, 1978: 429–35). The National Cancer Institute was established in 1937 and a rise in research funding for chemotherapy followed. Another impetus for the rise of chemotherapy was the translation of wartime research on poisonous gases and nutrition into a ‘model [for] civilian clinical research’ (Gaudillière, 2009: 498). This research model would provide ‘a firm basis for the development’ of early chemotherapy screening programmes (Zubrod et al., 1966: 350).

The KPS emerged from a screening programme initiated at the SKI. Its author, David Karnofsky, had studied the biological effects of mustard gases on goats as part of the Chemical Warfare Service (Burchenal, 1970: 549). Upon his discharge, Karnofsky went to work at the SKI, joining its director, Cornelius Packard Rhoads. ⁹

At the SKI, Karnofsky continued his investigations into the effects of HN-2 (nitrogen mustard) on humans, and in a 1946–48 study, he and his colleagues tested its potential as an anti-cancer drug on 35 patients with inoperable carcinoma of the lung (and, for comparison, 18 other patients with inoperable neoplasms). Recruited patients had not responded to or were considered unsuitable for roentgen-ray therapy, or were relapsing after a temporary response (Karnofsky et al., 1948: 634). In other words, this ‘highly experimental treatment’ was for palliative purposes, ‘a last attempt to intervene rather than let the disease take its course’ (Timmermann, 2012: 182).

To evaluate the effectiveness of this aggressive treatment as a potential antitumour agent, the researchers took four criteria into consideration. The first criterion, ‘subjective improvement’ (SI), was evaluated in terms of how the patients felt; whether appetite and strength were increased and whether they were relieved of symptoms (Karnofsky et al., 1948: 634). Instead of being measured objectively, these factors were assessed in general terms, marked as G (good), F (fair), or 0 (none) (Karnofsky and Burchenal, 1949: 194). The second criterion, ‘objective improvement’ (OI), involved ‘quantitatively measureable’ fields such as decrease in the size of lesions and nodes, and gain in weight (ibid.: 194–5), each graded as 0, 1+, or 2+. The third criterion, ‘duration of improvement’, was measured in weeks beginning from the administration of the agent to conclusive signs of relapse.

While subjective symptoms were thereby formalized as therapeutic criteria in their own right, the authors, like the committee on RA, considered these criteria to be ‘a notoriously poor method’, and prioritized objective measures over subjective ones as ‘the most substantial method of demonstrating activity’ (Karnofsky and Burchenal, 1949: 194). Thus, while SI and OI could occur simultaneously, OI alone was the yardstick for gauging treatment effectiveness. Of crucial importance, however, was a third possible scenario in the researchers’ study design, in which the patient improved in both subjective and objective terms, while the way he lived with the disease did not:

The fact that subjective and objective evidence of improvement can occur in a patient, while the patient remains bedridden, has suggested to us the need for another criterion of effect. This has been called the performance status, or PS. It is a numerical figure, in terms of percentage, describing the patient’s ability to carry on his normal activity and work, or his need for a certain amount of custodial care, or his dependence on constant medical care order to continue alive. These simple criteria serve a useful purpose, in our experience, in that they measure the usefulness of the patient or the burden that he represents to his family or society. (ibid.: 195–7; emphasis added)

To measure performance status (PS), the authors introduced a scale (Table 2) that expressed living with a disease and its treatment as varying degrees of ability to perform ‘normal’ daily activity, ranging from living as independently, effortlessly, and symptom-free as possible at one extreme, to no longer alive at the other. ¹¹ These four criteria (including PS) yielded the following results: 4 patients showed some improvement in PS but not in OI, and their SI was unchanged; 6 patients showed no improvement in PS, but some improvement in either or both of the two other fields; 3 showed some improvement in PS and in either OI or SI; 10 showed no improvement at all; and 14 improved in all three categories. ¹² Based on these outcomes, the authors concluded that HN-2 had an ‘immediate palliative effect of varying degree in 74 per cent’ of cases; this response was temporary, however, and there was no evidence that the medication had a significant impact on the course of the disease (Karnofsky et al., 1948: 653). In fact, given the lack of significant therapeutic effectiveness and the risks and side-effects involved, the authors cautioned that ‘HN-2 must not be used indiscriminately. Its use in a given case may be justified if there is some prospect that it might relieve discomfort or distressing symptoms or prolong useful life’ (ibid.: 655).

OPEN IN VIEWER

Table 2.

Karnofsky Performance Scale.

PERFORMANCE STATUS
Definition	%	Criteria
Able to carry on normal activity and to work. No special care is needed.	100	Normal; no complaints; no evidence of disease.
	90	Able to carry on normal activity; minor signs or symptoms of disease.
	80	Normal activity with effort; some signs or symptoms of disease.
Unable to work. Able to live at home, care for most personal needs. A varying amount of assistance is needed.	70	Cares for self. Unable to carry on normal activity or to do active work.
	60	Requires occasional assistance, but is able to care for most of his needs.
	50	Requires considerable assistance and frequent medical care.
Unable to care for self. Requires equivalent of institutional or hospital care. Disease may be progressing rapidly.	40	Disabled; requires special care and assistance.
	30	Severely disabled; hospitalization is indicated although death not imminent.
	20	Very sick; hospitalization necessary; active supportive treatment necessary.
	10	Moribund; fatal processes progressing rapidly.
	0	Dead.

Note: Karnofsky et al., 1948: 635. Copyright © 1948 American Cancer Society. Reproduced with permission from John Wiley and Sons.

Here again, the nature of the disease, much like that of RA, clearly complicated the process by which the treatment was going to be assessed for its outcomes, mainly because of the lack of definitive cure. Unlike the therapeutic criteria for RA, however, which was a generic framework for all treatment regimens, PS was developed to measure the effectiveness of a specific agent – one that was, in fact, toxic enough to owe its emergence to chemical warfare. This toxicity further complicated the process. The difficulties involved in assessing its medical outcomes become clear in the authors’ observations below:

If a drug is of curative value there should be relatively little difficulty in ascertaining this fact. Unfortunately, such drugs are not known, and most agents proposed for the treatment of cancer can only be expected to modify the course of the disease, or alleviate some of the symptoms. In evaluating drugs in this range of effectiveness, particularly in a disease as complex and variable as cancer, one is faced with a formidable task. It almost appears that the ease in determining the activity of a drug will vary directly with its true effectiveness. (Karnofsky and Burchenal, 1949: 191; emphasis added)

Furthermore, the KPS was not designed to measure a ‘distal symptom’, if this symptom is understood to be a linear consequence of an underlying disease progression or remission. What was to be made enunciable through the use of the scale did not have to be a direct correlation of (potential) tumour shrinkage or relapse. The authors emphasized that ‘while it is important to know that subjective and objective improvement have been produced, the picture is filled out if we also know whether the patient remained flat on his back or was able to return to work’ (Karnofsky and Burchenal, 1949: 197). The KPS was not developed merely to support what objective and subjective measures had already proven, but because it might contradict subjective and objective improvement. Just as functional classification of cardiac patients was designed to serve as a diagnostic criterion in itself (rather than a supplement to objective criteria), the KPS was designed to serve as an outcome measure that allowed the researchers to ‘fill out the picture’ and assess outcomes in new ways.

Returning to the medical model: The question of ‘function’

To discuss assessments of function as part of an attempt to question the medical model may appear counterintuitive or paradoxical. Disability scholarship has long taken issue with the way disability is traditionally defined by ‘functional limitation’ in practices ranging from rehabilitation programmes to official statistics and welfare services. As many have noted, this has to do with the history of defining disability in relation to the capacity to work, which was in turn rooted in the emergence of nation-states. ¹³ One of the main claims of early British disability studies is that the transformation to an industrialized capitalist mode of production resulted in ‘the creation of the disabled individual’, as, in the process, ‘what was essentially a labour market issue [was turned] into an individualized medical problem’ (Oliver and Barnes, 2012: 16). As Deborah Stone also shows, disability emerged as an administrative category in welfare states as a way to control labour supplies (1984: 26). Stone argues that two distributive systems, work-based and need-based, defined how wealth and services should be distributed in capitalist societies. But the question of how to determine who ‘truly’ belonged to which system disrupted this distributive logic. Disability was devised as a bureaucratic category to solve this irresolvable dilemma. ‘Validated’ by the clinical gaze, Stone claims, the category of disability would legitimize exemption from paid work and become the determinant of ‘true’ need for social aid. Assessments of function designed to crossmatch an impairment with the requirements of work or everyday living have since been used extensively by welfare bureaucracies, government programmes, and insurers to determine whether and to what extent a person is worthy of public assistance.

Given this history, functional criteria hardly seem worthy disruptors of the medical model. In fact, disability scholarship has long called to task such reductionist assessments of function and their incorporation into ‘fit-for-work’ evaluations, as well as the calculation of social security benefits, compensation, pensions, and disability living allowances. Identifying the ‘individual model’ of function as problematic, for instance, Mike Oliver writes that it ‘focuses on the functional limitations of individuals in attempting to use their own environment’. In contrast, the social model ‘sees disability as being created by the way housing is unsuited to the needs of particular individuals’ (1983: 25). From this perspective, the instruments discussed above do not appear to disrupt the medical model. Disablement as a social and political process is certainly not their concern. Furthermore, these instruments preserve the pathology-disease-disability causal link, and do not implicate inaccessible environments and discriminatory attitudes, thereby reducing disability to impairment, to a ‘problem’ of the individual body (as has long been argued with the concept of the medical model).

But what if the disease to which disability is reduced is not about biology, anatomy, or pathology either? This is the question I want to raise with the idea of disabilitization. In the classification of functional capacity in cardiac diseases, we are not dealing with heart rates. In the functional classification of impairment in RA, we are not talking about sedimentation numbers. In the KPS, we are not looking at tumour sizes. In short, we are not dealing with disease as characteristically defined in the medical model as strictly a ‘problem’ of the body, of its pathologies and abnormalities. I call attention to these three instruments for precisely this reason. No matter how narrow their scope, these tables and scales expanded the clinical gaze beyond the body to the everyday activities of that body. In so doing, they brought about a new way of looking at disease itself. This new way of seeing and knowing disease (alongside other historical contingencies that I will address later) would, over time, extend beyond how a body functioned to how ‘well’ it lived in the social world – a process of QoL development that I call disabilitization. My argument is that it is disabilitization, and the shift in medical perception that emerged from it – the shift from the inner depths of the biological body to the everyday lives of patients – that challenges the medical model. ¹⁴ Thus, I discuss these three instruments not because of what they do in themselves (that is, classify function) but because of the history they belong to – the development of QoL – and what that history of disabilitization does to the concept of the medical model.

I will now turn to this history and chart the emergence of QoL as a ‘matter of concern’ both in public life and in medicine. My argument is that either the history of QoL can trace its origins outside the medical model (in a way that challenges its validity), or, if it were to be incorporated into the model, the critiques embedded in it would become so dilated that it would not hold together with any coherence.

The making of Quality of Life into a ‘matter of concern’

Even though I discuss QoL in relation to medicine, the concept emerged, not from medicine, but from what came to be known as the ‘social indicators movement’, or rather, from the affluence ‘crises’ that have beset advanced industrialized societies since the 1960s (Armstrong and Caldwell, 2004; Noll, 2004; Rapley, 2003), as ‘developed’ nations began to face the societal and environmental costs of ‘progress’, the ‘great society’, and economic growth with rising crime, overpopulation, pollution, and housing problems. Amidst growing concerns over these crises, social scientists began to develop surveys, statistics, and indicators to measure subjective variables such as happiness, well-being, and life satisfaction. The idea of ‘Quality of Life’ became the springboard for these instruments, which were intended to monitor social progress, evaluate welfare, and, more broadly, create ‘an information base which supports the policy making process’ and the setting of priorities (Noll, 2004: 154). As doubts increased as to ‘whether “more” should continue to equal “better”’, the notion of Quality of Life emerged ‘as an alternative to the more and more questionable concept of material prosperity in the affluent society’ (ibid.: 3). Regardless of whether people thought society was headed towards progress or decline, ‘all agreed that quality of life was the goal and potential arbiter of the debate’ (Armstrong and Caldwell, 2004: 368). The concept was vague enough that differing viewpoints could find common ground and set ‘a common goal often across very different political programs’ (ibid.: 363). In its ambiguity, Quality of Life permeated modern societal domains, from advertising to public policy. This did not happen independently of other contemporaneous developments, however. With the rise of corporatism and consumer movements in the 20th century, concepts such as ‘service quality assessments’, ‘consumers’, ‘choice’, ‘value for money’, and ‘satisfaction’ entered into the rhetoric of governments, public policy, and service sectors, such that ‘the idea of quality of life has come to be intimately bound up with the broader discourses of managerialism and corporatism in contemporary Western societies’ (Rapley, 2003: 124–5).

Medicine was no exception to these broader developments. The postwar period saw an unprecedented rise in drug discoveries and medico-technological inventions, but not all of these advances delivered their purported benefits, or, when they did so, their biological benefits were not always accompanied by subjective improvements. On the contrary, saving or prolonging life (in the face of otherwise fatal diseases) could come at a high ‘human cost’ to patients (Armstrong and Caldwell, 2004: 364), as in the case of aggressive chemotherapy treatments such as the agent used in Karnofsky et al.’s study. At the same time, successful medical interventions and rising living standards contributed to the growth of ageing and chronically ill populations, including ‘people living with and beyond cancer’ (Macmillan Cancer Support, 2017), whose care called for a focus not only on survival, but on ‘living well’ (Petchey, 2016).

Meanwhile, in 1948 the World Health Organization’s constitution famously defined health as ‘a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity’. Studies in social medicine and epidemiology from the 1970s onwards began to call critical attention to medical outcomes and the lack of correlation between the amount or type of care provided and the level of improved health (see Bury, 1994; Timmermans and Berg, 2003: 15). These studies, Bury notes, coincided with a ‘restructuring of welfare’ and cost-containment efforts in late modern societies, particularly as healthcare became increasingly ‘extensive and expensive’ (1994: 123). The very idea that the outcomes of treatments could be assessed systematically, he adds, quickly became linked to ‘the issue of value for money’ (ibid.). As discourses of managerialism and corporatism began to enter into the governance of healthcare, the new ‘managers’ and ‘purchasers’ of services sought information and evidence on which to base decisions on resource allocation in overstrained health systems. With the infiltration of consumer culture, the category of patients evolved from recipients of care to active ‘consumers’, who could make informed ‘choices’ among treatment products (see Rapley, 2003). These consumers could hold healthcare providers and controllers accountable for their services and the outcomes they produced (Ware, 1984: 2316). With this ‘outcomes movement’ came the idea that changes in health resulting directly from antecedent medical care could be assessed so as to document the effectiveness of healthcare services, monitor quality of care, plan optimal resource use, and set future health policies.

It was within such ‘various fields of constitution and validity’ (Foucault, 2010[1972]: 4), and contingent histories ranging from economic concerns to policymaking and epidemiological changes, that QoL emerged as a new object of knowledge and a ‘matter of concern’ in medicine. QoL became a versatile tool to address a whole new set of tensions and questions, through which medical knowledge and healthcare were beginning to be articulated (see also Armstrong and Caldwell, 2004: 368). Where there was little or no chance of a cure, QoL provided medicine with new goals. When life-saving and life-prolonging treatments began to take their toll (with adverse effects like toxicity and nausea), QoL offered a new criterion for both drug development and approval, as well as for clinical decision-making. When alternative treatments had equivalent biological outcomes, QoL provided a new means of comparison for policymakers to find the ‘best buys’, or for the pharmaceutical industry to gain a competitive advantage. Where medicine’s professional authority was being called into question with the rise of an evidence-based paradigm and outcomes movement, QoL measures offered a means to oversee performance and assess quality of care. In an era of ageing and chronically ill populations, generic QoL measures enabled research in health services to make comparisons across different disease populations in terms of the ‘burden’ they posed. Preference-based QoL metrics like the Quality-Adjusted Life Year (QALY) allowed policymakers and economists to compare treatments targeting different diseases and their health ‘gains’. In sum, for all these diverse healthcare actors – from newly emerging consumer-patients who were expected to be able to make ‘informed’ choices between alternative treatments, to healthcare planners seeking ‘evidence’ for ‘best buys’; from regulatory authorities encouraging the use of QoL endpoints in clinical trials, to the pharmaceutical industry incorporating QoL data into product labelling – QoL proved itself to be a resourceful and adaptable tool, capable of addressing their vested interests and concerns.

The consolidation of QoL in medicine did not happen overnight. As Armstrong’s genealogy (2009: 114) suggests, it has taken decades of ‘stabilising’ efforts to make QoL into a ‘hard’ scientific fact, a measurable clinical endpoint, and the ‘thing’ that we all now have. First was the ‘advocacy phase’, in which QoL appeared mostly in discursive and educational publications that worked hard to ‘sell’ this rather vague notion to a doubtful medical profession as a ‘new goal for medicine’ (ibid.: 105). Once this new rhetoric had begun to gain currency, attention was turned to making this nebulous concept a measurable entity and an applicable clinical endpoint. From the 1980s, publications proliferated in which new QoL instruments were designed and tested, the methodologies of instrument development were addressed, and an increasing number of diseases and conditions were considered in terms of their effects on QoL. By the 1990s, QoL publications had come to be dominated by the area in which QoL is today mostly deployed: outcomes research (ibid.: 107–8). In this phase, now-formalized QoL instruments were used to evaluate the outcomes of medical interventions, most notably to compare treatment effectiveness in randomized clinical trials (ibid.: 109).

As new psychometric and other empirical methods are invented; as more ‘precise’ techniques of measuring subjective states are introduced; as the field of quantifiable and classifiable domains is expanded (to encompass, for example, pain, fatigue, moods, feelings, social interactions, well-being, and so on); as new diseases and conditions are rendered relevant in terms of their effects on daily living; and as more and more specialties (such as gerontology and psychiatry) and actors (such as clinical trial units, policymakers, regulators, healthcare funders, and governments) take interest in these instruments, QoL has become a discourse in its own right, and ‘an industry in itself’ (Bowling, 2001: 10). As Foucault reminds us, however, the autonomy of a discourse does ‘not give it the status of pure ideality and total historical independence’ (2010[1972]: 164). Indeed, as I summarized above, the production of QoL cannot be understood in isolation from other contingent events, processes, and practices. Some of these were ‘not themselves of a discursive order’ (ibid.: 164), such as public concerns about a looming social crisis, curbs on public expenditure, and the restructuring of welfare systems. Others were of a discursive order but not necessarily of medical origin, such as the functional assessment developed for recruits in wartime and measures of well-being and happiness introduced by the social indicators movement. Still others arose directly from medical discourse but predated the introduction of QoL, such as the three assessments of function analysed in this article.

From a contemporary perspective, with QoL established as a discourse (and an industry) in its own right, the instruments analysed in this paper hardly qualify as QoL instruments, as the discourse distances itself from crude assessments of physical function. Instead, it frames itself as a broad construct that also includes emotional, social, and cognitive functions, as well as domains such as well-being, life satisfaction, and happiness. Genealogically speaking, however, one could argue that it was because these crude instruments (alongside the aforementioned discursive and non-discursive developments) created certain possibilities and opened up new ways of mapping objects of knowledge that the discourse of QoL could have emerged. These assessments, no matter how narrow their scope, made it possible to move clinical perception beyond the body and situate it in patients’ everyday lives in formalized ways. Once this formalization had occurred, it could be rectified and remade within newly developed instruments, with which medical perception would look not only at how a body functions, but also at how ‘well’, ‘happy’, or ‘satisfied’ the patient feels, how ‘good’ her life is, and how much ‘quality’ it has. Certainly, the daily living measured and the disability made enunciable in the KPS is not the same as that rendered knowable through the use of, say, the EORTC QLQ-C30. ¹⁵ Nor is the formulation of the ability to perform everyday life as a therapeutic criterion in RA (1949) the same as the ability formalized within contemporary measures of QoL in RA. We are not, to quote Foucault, ‘dealing in each case with the same discursive event’ (2010[1972]: 143). From a genealogical lens, however, it was because these rudimentary instruments turned the body that performed functional tasks in daily life into a classifiable, knowable, and measurable category that this category could be remade into a body that has emotions and moods, gets tired, feels pain and symptoms, and is gradually transformed into ‘an individual’ that socializes and enjoys a happy, quality life.

In sum, it was, on the one hand, new ways of mapping diseases that were opened up by these crude classifications of function (as well as by social indicators movements and developments in psychiatric methods and questionnaire techniques); and it was, on the other, new matters of concern to medicine that arose from non-medical processes (such as economic, epidemiological, political, and social developments) that have made QoL into a new way of knowing and thinking in medicine. I argue that it is precisely this history of disabilitization, its conditions of possibility, and the shift that it incurs within medical objects of knowledge – from pathologies and lesions to physical, social, and emotional function, activities of daily living, happiness, well-being, and life satisfaction – that opens up a space in which to question the medical model and its basic tenets. ¹⁶ These questions are:

How can we explain the emergence of a construct characterized by a history of ‘distancing’ medical perception from intra-corporeal lesions to a patient’s everyday social worlds (Armstrong et al., 2007: 581) with the concept of a medical model that assumes medicine to reduce disability to disease, and disease to pathology? The making of QoL could in fact be read as a continuous manifestation of the medical model’s limitations rather than of its successful application.

These questions suggest that the concept of the medical model is but one way of understanding medicine’s complex relationship to disease and disability, and perhaps quite a limiting one. As I have tried to show, there are other histories to which the medical model cannot be as readily applied. It is to foreground this friction, and to tease out what it might offer to our current understanding of medicine within disability studies and related fields, that I consider the emergence of QoL a process of disabilitization of medicine.

The ‘disabilitization’ of medicine

As stated earlier, the ‘disability’ implied in disabilitization is not the same ‘disability’ that has been reclaimed and theorized within disability studies (namely, disability as an identity, a social justice issue, a cultural category, and a form of aesthetic and everyday creation). To claim that medicine has broadened its gaze to include patients’ ability to perform the everyday is not to argue that it has finally turned to social aspects of disablement and de-individualized disability in the meantime. But, as noted earlier, the ‘disability’ in disabilitization is also not the same ‘disability’ considered the subject of medicine by the medical model of disability. As the discourse of QoL has made it possible to expand medical perception beyond the inner depths of the body, and locate it within patients’ everyday worlds through standardized measurements, diseases would be configured, not as pathological lesions and anatomical abnormalities, but as ‘living with’ those states (Wahlberg, 2018). As a corollary, disability would be articulated not through its reduction to an idea of disease in the inner depths of the body, but through its expansion to an idea of disease that is experienced, embodied, and lived with in the everyday – an expansion that I term disabilitization. I use this provocative concept to refer to the historical process by which QoL emerged as a matter of concern in medicine and healthcare, and how this has brought about new articulations of disease and disability that the medical model remains too limited a tool to explain. As medicine became disability-sized throughout the development of QoL, diseases and disabilities were sought, not within the biological body – as is presumed to be characteristic of medicine in ideas of the medical model or the clinical gaze – but within the everyday where that body performs activities, feels emotions, engages with others, and, for better or worse, lives a life. I propose the concept of disabilitization as a way to capture precisely these new configurations of disease and disability that weaken the traditional understanding of medical perception within the medical model of disability.

Disabilitization, as I conceptualize the term, also refers to the relations between the discourse of QoL and the non-discursive events included in its conditions of possibility, as well as the other ways in which those relations might challenge the medical model. For example, according to the medical model, disability is a ‘deviance’, an out-of-the-ordinary body that sparks the curiosity of the medical gaze, which seeks to treat and correct this ‘aberration’. Disability becomes a ‘tragedy’, a ‘loss’ that needs to be avoided at all costs. While this argument can easily be applied to certain histories, it is harder to hold onto it in the face of a chronically ill and ageing population, in which QoL has emerged as a major concern for medicine. In this epidemiological landscape, being chronically ill or disabled does not figure as ‘extraordinary’, but as highly ordinary – states any of us might reach if we live long enough. Accordingly, the goals of medicine are directed not only at eradicating diseases or halting their progression, but also at providing patients with ‘a life that is as comfortable, functional and satisfying as possible’ (Sullivan, 1992, in Bowling, 2001: 11). The National Cancer Strategy of Ireland (2017–26) is a case in point: ‘Since many forms of cancer are chronic yet highly survivable, the definition of successful treatment can be seen to have shifted toward maximising the quality of life of individuals diagnosed with cancer for as long as they live. In short, it is not a question of “just surviving” the aim is to maximise quality of life’ (Department of Health, 2017: 109).

Second, the medical model is based on the presumption that medicine has a totalizing power and professional authority. The emergence of QoL makes this argument untenable, especially in view of the economic concerns, political events, and institutional practices that have been involved in it. In post-industrialized societies, the governance of healthcare is dispersed among various actors and practices. In these changing constellations of knowledge and power, we see the emergence of new categories such as patient experts, whose ‘knowledge and experience’ is considered ‘an untapped resource…that could greatly benefit the quality of patients’ care and ultimately their quality of life’ (British Department of Health, cited in Rapley, 2003: 139). We also see categories such as patients as survivors, who can be educated and guided towards self-managing their care and maximizing their QoL (Department of Health, 2017: 9). New actors are emerging in the restructuring of healthcare practices: the ‘consumers’, ‘buyers’, and ‘managers’ of healthcare. It is to these actors that ‘the use of performance and quality measures, including quality of life measures, appears to provide…an important “window” onto providers’ activities, and acts as a check on professional autonomy’ (Bury, 1994: 126). The development of QoL that I refer to as disabilitization thus points not to the exercise of a resolute knowledge or power, but to its diffusion, and the undermining of medicine’s professional autonomy.

What if we integrate a critique of QoL into the medical model?

To argue that QoL does not represent the workings of the medical model is not to say that it cannot be critically questioned. First, QoL instruments articulate disability in relation to an external ‘social’ world, rather than a ‘biological’ body. But, no matter how far this expansion goes, the link between disease and pathology is preserved, and ‘the issue of causality’ (Oliver, 1996: 31) – long criticized by disability scholars – is left intact. In other words, no matter how far QoL measures ‘distance’ their gaze from the body, the individual body remains their point of departure. This leads to the creation of a linear sequence: pathology –> disease –> disability. ¹⁷ Both disability and disease remain individualized, as the individual/medical model claims them to be.

Second, measuring living ‘well’ and a ‘quality’ life entails measuring essentially unquantifiable phenomena (that is, values, beliefs, and experiences). Scholars of disability and medical anthropology (see Hahn and Hegamin, 2001; Warren and Manderson, 2013) have already highlighted the importance of qualitative enquiries into experiences of living with chronic illness and disabilities, aspects of which might get lost when quantified.

Third, as I have argued with the notion of micro-activist affordances, chronically ill and disabled people may invent new ways of being in the everyday, which might not be imaginable in the absence of pain, illness, or disabling factors. Unlike the ADLs listed in QoL measures, these improvised affordance-creations cannot be abstracted from the locality of their occurrences and subjected to mathematization.

Fourth are the controversies surrounding QoL measures in policymaking and resource allocation decisions, particularly the use of metrics such as QALYs. These evaluations of ‘value for money’, and the utilitarianism undergirding them, have been debated (including in the field of QoL itself) especially for the bias they create against disabled, elderly, and poor populations. Hays, Hahn, and Marshall (2002: S5), for instance, point out that disability communities ‘fear that…the designation of quality of life (QOL) might be used as a threshold or triage principle in the allocation of resources’, with potential detrimental consequences to their lives. Disability rights activists Harlan Hahn (2002: 180) and David Pfeiffer (2000: 1082) have denounced metrics such as QALYs and disability-adjusted life years (DALYs) for equating disability with a ‘burden’ to be eliminated, and for opening the door to eugenics.

There are many other angles from which QoL discourse can be criticized: the fact that QoL rhetoric can function as a form of governmentality through which individuals self-rate, self-surveil, and become complicit in their own subjection; its problematic vocabulary, as evidenced by references to the ‘burden’ or ‘usefulness’ of patients; or the implicit assumption that a ‘quality’ life is one that is lived independently, productively, and self-sufficiently, contradicting the interdependence, vulnerability, and care ethics theorized within disability studies. ¹⁸

What do these criticisms of QoL mean for the medical model? First, to criticize QoL instruments for preserving the link between pathology and disability is not the same as criticizing them for reducing disability to mere function or for turning subjective experiences into numbers. Second, a conceptual criticism of QoL instruments is not comparable with a criticism of their fields of application (such as economic evaluations and resource allocation decisions). Each of these is a different criticism. Last, the limits of QoL methodology, its vocabulary, and its philosophical underpinnings are all different domains, requiring differently formulated critiques. The point I want to make is that just because there is always a way to contest or critically question heterogeneous sets of practices and knowledge formations, does not imply that the medical model is necessarily present. The possibility of criticizing QoL discourse from multiple angles does not make it the instantiation of an overarching model, nor of a totalizing history of this or that model of disease or disability. Room for criticism does not imply a space to be filled by the medical model. On the contrary, any critique of the disparate knowledge practices within QoL needs to be as specific, and as singular, as the object of its critique, and to take into account its particular conditions of possibility.

Disabilitization: A proposal to move beyond the medical model

Considering that objects of medical knowledge are incomplete, elusive, and undergoing ongoing transformation, concepts that lock this incompleteness into fixity and subsume all the tensions and differences within medicine under a totalizing rubric can hardly provide a productive framework. In this article, I have taken Mol’s suggestion ‘to doubt’ as my departure point for examining the limits of the traditional criticism of medicine, particularly as it is voiced in the medical model of disease and disability. In discussing how QoL discourse brings up discontinuities to the tenets of the medical model, I have sought to find occasions for not taking the model for granted, and to begin to ‘doubt’ it instead.

To be sure, to ‘doubt’ the medical model in no way denies the appalling and atrocious treatments to which disabled people have historically been subjected under the auspices of medicine, as in histories of institutionalization and eugenics. Nor does it mean that disability and disease have not been reduced to pathologies in certain histories and medical practices, or that disability has not been medicalized in other discursive formations. The question is: why should it? Why should ‘doubting’ the medical model equate to those things? Is there no way to describe medicine’s objects, concepts, and styles of enunciation without simultaneously adhering to the criticism promoted by the medical model of disability (or by ‘medicalization’, ‘normalization’, or ‘disciplining’, for that matter)? Or have certain concepts, while owing their origins to specific sets of practices, become institutionalized over time, and turned into discursive regulations in and of themselves, regulating the objects of which we, as disability studies or medical anthropology scholars, can speak, and allowing us only certain subject positions (and not others)? With the notion of disabilitization, I seek to offer one such space for doubting – a space that might allow us to think beyond concepts that have, perhaps, become too familiar.