Disability Through the Lens of Justice

NEW TOPICS IN APPLIED PHILOSOPHY

Series editor
Kasper Lippert-Rasmussen

This series presents works of original research on practical issues that are not yet
well covered by philosophy. The aim is not only to present work that meets
high philosophical standards while being informed by a good understanding of
relevant empirical matters, but also to open up new areas for philosophical
exploration. The series will demonstrate the value and interest of practical issues
for philosophy and vice versa.

    

Spying Through a Glass Darkly
The Ethics of Espionage and Counter-Intelligence
Cécile Fabre
Inheritance of Wealth
Justice, Equality, and the Right to Bequeath
Daniel Halliday
The Politics of Social Cohesion
Immigration, Community, and Justice
Nils Holtug
Not In Their Name
Are Citizens Culpable for their States’ Actions?
Holly Lawford-Smith
Sharing Territories
Overlapping Self-Determination and Resource Rights
Cara Nine
Exploitation as Domination
What Makes Capitalism Unjust
Nicholas Vrousalis
Disability Through the
Lens of Justice
JESSICA BEGON
 Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Jessica Begon 2023
The moral rights of the author have been asserted
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2022950547
ISBN 978–0–19–887561–1
DOI: 10.1093/oso/9780198875611.001.0001
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
 Contents

Acknowledgements ix
Introduction 1
1. What Disability is Not 10
1.1 The Problem of Defining Disability 10
1.2 Disability and Impairment 11
1.3 The Medical and Social Models of Disability 15
1.4 Beyond the Medical and Social Models 20
1.5 Species Norms and Impairment 24
1.6 Welfarist Accounts 28
1.7 Barnes’s Social Constructionist Approach 32
1.8 An Ameliorative Approach 36
1.9 Conclusion 39
2. Disability: A Justice-Based Account 41
2.1 Which (In)abilities Matter? 41
2.2 What Sort of Justice? 43
2.3 Feasibility, Levelling-Down, and Thinning-Out 45
2.4 Identifying Distributive Entitlements 48
2.5 Entitlements without Hierarchy 54
2.6 Beyond Minimal Functionings 57
2.7 What Counts? Who’s Disabled? 62
2.8 Conclusion 68
3. Disability and Distribution: A Capability Approach 69
3.1 Introduction 69
3.2 A Capability Theory of Justice 70
3.3 Capabilities or Functionings? 72
3.4 Capabilities: The Good, the Bad, and the Trivial 75
3.5 Against Resourcism: Means, Ends, and Conversion Factors 81
3.6 Resourcism: The Less Demanding Alternative? 84
3.7 Beyond Simple Resourcism 86
3.8 Conclusion 88
4. Capabilities for Control 90
4.1 Introduction 90
4.2 Having the Capability: From Formal Freedom to Freedom
as Control 91
4.2.1 Independence 94
4.2.2 Security 95
4.2.3 Robustness 96
4.2.4 Set-consistency 97
vi 

4.3 The Problem with Capabilities to Function 99

4.4 Capabilities to Control 106
4.5 Capabilities to Control: Too Little Freedom, or Too Much? 112
4.6 The Value of Capabilities 116
4.7 Why Capability Lists 123
4.8 Conclusion 126
5. Neutral Impairment, Disadvantageous Disability 128
5.1 Introduction 128
5.2 The Debate about Impairment 131
5.3 In Defence of Impairment 134
5.4 Impairment as Atypicality 137
5.5 The Over-Inclusiveness Objection 143
5.5.1 Functioning Better 146
5.5.2 Latent or Hidden Pathologies 149
5.5.3 Distinguishing ‘Genuine’ Disorders 151
5.5.4 Distinguishing Disability from Other Forms of Disadvantage 152
5.6 Mere Difference or Complex Difference? 154
5.7 Bad Difference or Complex Difference? 158
5.8 Conclusion 163
6. Disambiguating Adaptive Preferences: When, and
Why, Should Testimony be Trusted? 165
6.1 The Dilemma of Adaptive Preferences 165
6.2 Adaptation and the Political Project 168
6.3 Well-Being Adaptive Preferences 172
6.4 Why Identify Well-Being Adaptive Preferences? 178
6.5 Perfectionism, Justice, and the Need for a Substantive
Approach 181
6.6 Justice Adaptive Preferences 184
6.6.1 Non-Autonomous Character Planning 187
6.6.2 Justice Adaptive Choices 189
6.6.3 The Coherent Internalisation of Oppressive Norms 190
6.6.4 Justice and Well-Being Adaptive Preferences 191
6.7 Responding to Justice Adaptive Preferences 192
6.8 Conclusion: Distrust without Insult? 196
7. Don’t do it for my Sake: Providing Control, Avoiding
Paternalism, and Applying the Justice Account of Disability 199
7.1 Introduction 199
7.2 Defending Anti-Paternalism 202
7.3 State Assistance and the Anti-Paternalist Filter 211
7.4 Anti-Paternalism in a Theory of Justice 215
7.5 Intervention beyond Interference 220
7.5.1 Raising and Reducing Costs 221
7.5.2 Pre-emption: Information and Persuasion 222
7.5.3 Refusals to Assist 223
  vii

7.6 The Paternalism of Unwanted Offers 225

7.7 Specifying Capabilities without the Good 228
7.8 Soft Paternalism, Autonomy, and Background Conditions 232
7.9 Cures and Parental Paternalism 236
7.10 Conclusion 238
Conclusion 240

References 251
Index 267
 Acknowledgements

There are many people to whom I owe thanks for invaluable discussion,
feedback, and arguments, throughout the years it took to complete this
book. A no doubt incomplete list of these are: Elvio Baccarini, Alice Baderin,
Mike Begon, Paul Billingham, Ian Carroll, Brian Carey, Stephanie Collins,
Maria Dimova-Cookson, Adam Cureton, Cécile Fabre, Søren Flinch
Midtgaard, Chris Finlay, Tim Fowler, Ana Gavran Miloš, Alex Geddes, Anca
Gheaus, Rich Healey, Katherine Jenkins, Beth Kahn, Jan Kandiyali, Cécile
Laborde, Joseph Lacey, Gerald Lang, Kasper Lippert-Rasmussen, David Miller,
Chris Mills, Tom Parr, Jonathan Parry, Angie Pepper, Adam Slavny, Zofia
Stemplowska, Tony Taylor, Kristin Voigt, David Wasserman, Katy Wells, and
Nebojša Zelić. I would also like to thank audiences at the Association of
Social and Political Philosophy Conference at the University of Sheffield,
the CSSJ Seminar at the University of Oxford, the CELPA Seminar at the
University of Warwick, the Centre for Political Thought Seminar Series at
Durham University, the MANCEPT seminar at the University of Manchester,
the Nuffield Political Theory Workshop, the Paternalism, Health, and
Discrimination Workshop at Aarhus University, the Philosophy of Disability
Conference at the. University of Tennessee, the Political Philosophy Seminar
at the University of Newcastle, the Politics Department Seminar at the University
of Bristol, the Priority in Practice Conference at the University of Oxford, the
Royal Institute of Philosophy Seminar at the University of Nottingham,
the Society for Applied Philosophy Conference at University of Edinburgh and
Cardiff University, and the UCL Legal and Political Theory Seminar. Thanks,
too, to Nuffield College, Oxford, who provided the funding, environment, and
support for the period of research when this book began, though was not
completed. Finally, thanks to Carl Fox, for absolutely everything.
 Introduction

Disability remains a neglected topic in discussions of distributive justice.

Insofar as it is discussed at all, it is generally used as convenient shorthand
for disadvantage. Yet far from being an afterthought or appendage to theories
of justice, an analysis of disability can and should inform how we think about
and refine these theories.
My goal is to determine how disabled individuals should be justly treated in
the public policy of liberal democratic states. This, of course, requires an
account of what constitutes ‘just treatment’ in general. But such an account
should not be developed in a vacuum, or developed following reflection on
able-bodied, fully cooperating members of society, and simply imposed on
disabled persons as one ‘special case’. On the contrary, we should engage in a
process of reflective equilibrium, wherein our theory of justice informs how we
understand and respond to disability, and a consideration of disability pro-
vides insights into how we develop a theory of justice. Disability, then, should
play a central, rather than subsidiary, role in our approach to justice.
I also aim to challenge the preconception that paradigm disabilities entail
suboptimal functioning, and thus that disabled individuals are both unhealthy
and worse-off than the non-disabled. Indeed, disabled individuals most often
appear in theories of justice merely as paradigmatic, and apparently uncon-
troversial, cases of disadvantage. All that remains to be determined in such
accounts is what disabled individuals are owed to mitigate, remove, or com-
pensate for this disadvantage. Thus, theorists of justice have given little
consideration to the question of what just treatment of disabled individuals
entails if disability is not simply assumed to be always and necessarily
disadvantageous.
This is not to say that disability is unrelated to disadvantage. It is important
to consider disability in the context of distributive justice exactly because
disabled individuals are so often overrepresented amongst those who do not
have what they are entitled to. Indeed, on the definition of disability I defend,
to be disabled just is to be unable to exercise the control over our life that
everyone should be able to. More precisely, then, it is impairment—physical or
cognitive features causing atypical functioning—that is associated with, but

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0001
2      

not identical to, disadvantage. Human bodies and minds are heterogeneous,
and this can make the lives of those who function differently more difficult—
perhaps because their mode of functioning is not accommodated by society;
perhaps due to that mode of functioning’s intrinsic features; perhaps due to
interaction with some aspect of an individual’s physiology, personality, or
preferences; or perhaps due to a combination of these factors. Yet the possi-
bility that life can thus be made more difficult does not mean that a life with an
impairment is worse, or even always difficult.
There is nothing inconsistent in insisting that impairment as difference
need not imply deficiency and should not necessarily be eliminated, whilst also
acknowledging that those who function differently often do live worse lives
than their non-impaired counterparts. But it is essential that a distinction is
made between the mere fact of functional difference—lacking some sensory
abilities, being mobile without using one’s legs, processing social cues
differently—and the implications this has for individuals’ capabilities: specif-
ically, in the context of justice, the implications this has for the capabilities
they are entitled to. It is the latter that should be our focus. In other words, we
should be concerned with the kinds of lives individuals can lead, the oppor-
tunities they have, and their ability to choose between them, and not on
whether their functioning mode is typical: what matters is that people can
be mobile, form relationships, engage in leisure and cultural activities, and so
on, and not how they achieve these things.
The idea that impairment itself is not the problem, and that normalisation
should not be our goal, will seem obvious to some, yet highly implausible to
others. There is a similar divergence on the plausibility of the claims that
abstract theories of justice can contribute anything useful to the very practical
problems entailed by living with impairments. And, indeed, on whether the
experience of living with impairments can generate insights for abstract
theories of justice. Whether initially plausible or not, I aim to show that all
these claims are true.
I begin by considering conceptual questions, and defend a new account of
disability. Chapter 1 first outlines central features of accounts of disability on
which there is a broad philosophical consensus.¹ Crucially, this includes
rejecting both medical and social models, and adopting a hybrid of these
two extremes. Adopting this hybrid allows us to acknowledge that disability
involves a restriction in our abilities to perform tasks as a result of physical or

¹ E.g. Wolff 2009a; Kahane and Savulescu 2009; Shakespeare 2014; Terzi 2004; 2008; Daniels et al.
2009; WHO 2002: 8–9.
  3

cognitive features that cause deviation from a biological or statistical norm of

functioning for our species (impairment), as the medical model insists; but in
combination with external contextual factors, such as institutional design,
social norms, natural environment, available technology, individual resource
share, and access to support networks, as well as personal features, such as age,
education, social background, career and personal goals, as the social model
has demonstrated. In other words, we should take account of both physical
and cognitive difference itself, and what it means to have anomalous func-
tioning in a particular context.
Whilst I accept this broad approach, I contend that it leaves an important
question unanswered: which inabilities matter? Which of the many restrictions
in our abilities to perform tasks resulting from atypical functioning are
disabling? I consider and reject various competing answers to this question.
Thus, the opportunity losses that should be considered disabling are not those
entailed by ‘normal’ species functioning,² nor those that undermine well-
being,³ nor those considered relevant by disability activists.⁴ The latter, social
constructionist approach leaves disability radically underspecified, whilst on
both the species norm and welfarist approaches disability will be ubiquitous.
An account of disability is an essential prerequisite to formulating public
policies that safeguard the just treatment of disabled individuals, but an
account cannot form the basis of such policy if it sets the boundaries of the
category of disability so wide that everybody is included. As such, none of
these alternatives prove to be an adequate basis for public policy, political
campaigning, or activist movements. In other words, none provide an
adequate account of disability in the context of justice: one that will enable
us to identify and respond to the injustice, discrimination, and mistreatment
to which many disabled individuals are subject.
In Chapter 2 I present my own answer: that the relevant losses of oppor-
tunities are of those that individuals ought, as a matter of justice, to have access
to. More precisely, I suggest that we should understand disability as the
restriction in the ability to perform those tasks human beings are entitled to
be able to perform as a matter of justice (as the result of the interaction between
an individual’s impairment, their social, political, and material context, the
resources they have access to, and their other internal characteristics). An
implication of this view is that determining whether an impairment is

² Buchanan et al. 2000; Daniels 1985.

³ Kahane and Savulescu 2009; Savulescu and Kahane 2009; Terzi 2008; 2009; Wolff 2009a.
⁴ Barnes 2016a.
4      

disabling requires an account of what we are entitled to. However, even before
this is provided in subsequent chapters, the revisionary implications of this
account are clear. Individuals with many paradigmatic sensory, physical, and
cognitive impairments may not be considered disabled insofar as they still
possess the more general opportunities they are entitled to—in the same way
that short-sighted individuals may continue to have a visual impairment but
are not disabled if, wearing glasses, their access to their entitlements is not
restricted. Further, individuals with apparently minor or invisible impair-
ments may be disabled, and individuals’ status as disabled may change across
time and context, since what matters is not impairment per se but its impact
on our access to our entitlements. Whilst potentially counterintuitive, these
implications are a key advantage of my approach. By distinguishing disability
as necessarily disadvantageous (lacking opportunities all human beings are
entitled to), from a normatively neutral account of impairment (capturing the
broad heterogeneity of human bodies and capacities) we can shift the focus
from eliminating impairment to removing the disadvantages that can be
associated with them.
The next chapters then consider what individuals are entitled to, and argue
that this is, indeed, broad opportunities or capabilities. Chapter 3 argues that
reflection on impairment and disability can contribute to an argument in
favour of adopting a capability approach to distributive justice. Indeed, insofar
as disability has been considered in the context of distributive justice it is
usually by proponents of the capability approach.⁵ However, capabilities are
commonly understood as components of well-being, meaning that individuals
are considered to be entitled to those capabilities that will make them better
off. If incorporated into my justice-based account of disability this would
appear to entail that individuals are disabled whenever their impairments
contribute to a loss of opportunities that decreases well-being. I argue that
this is implausible both as an account of justice and as an account of disability,
since it will tend to justify an excessive degree of intervention in order
to eliminate any opportunity loss that reduces well-being. Not only will
this be overdemanding of the well-endowed, it will also push us towards
normalising—even perfecting—human functioning capability. I argue that we
should distinguish well-being from entitlements, where the latter may be nar-
rower than the former. I then outline some of the reasons for understanding our

⁵ E.g. Barclay 2010; 2018; Mitra 2017; Nussbaum 2006a; Terzi 2008; Wolff 2009a, 2009b; Wolff and
de-Shalit 2007; Wasserman 2006.
  5

entitlements in terms of capabilities, and delineate the core features of a

capability theory of justice.
Chapter 4 considers in greater detail what it means for an individual to have
a capability, and why this is valuable. Capabilities are standardly understood as
opportunities to perform (or not) particular ‘valuable’ functionings. This focus
on the ability to do or to be particular things can exclude many individuals
with impairments, who often cannot perform and do not value the function-
ings assumed to be central or essential to any decent human life. Individuals
with sensory impairments may be unable to use all five senses, for example.
Capabilities should, therefore, be understood instead as the ability to control
central domains of our life. Even when impairment limits access to specific
valuable functionings it need not prevent individuals from valuing, and being
enabled to have, control and choice in these parts of their life. Deafness and
blindness, for example, do not preclude deaf and blind individuals from
having valuable sensory experiences and being able to control the exercise of
these capacities.
Thus, rather than attempting to identify valuable functionings and ensure
everyone has identical access to them, we should focus on ensuring individuals
have an acceptable range of options and the ability to choose between them.
Rather than endorsing a hierarchy of functioning, inevitably grounded in the
functioning capacities of the able-bodied, we should accept that different
modes of functioning can be interchangeable and equally valuable: it is not
necessarily worse to achieve mobility in a wheelchair than by walking, to
communicate by Sign rather than speech, to read using Braille rather than
text, and so on. This approach implies that a broad array of functionings may
be valuable, that our entitlements can be fulfilled by multiple means, and that
individuals with impairments should be given a central role in determining the
different forms of functioning that may enable them to have the control they
are entitled to. This theory of justice can accommodate individuals with a
variety of needs, preferences, and abilities. It also implies that individuals are
not entitled to the ability to perform specific functionings (even when these
have some value), and are not disabled by an inability to do so as long as they
nonetheless have the capacity to exercise control.
It is worth highlighting that, whilst it is possible to adopt my account of
disability without also accepting my account of justice, I believe (unsurpris-
ingly!) that they work best in combination. As will become clear, narrower
conceptions of justice will lead to an implausibly narrow view of disability on
my view: if we are not entitled to much (just a basic bundle of resources,
perhaps), then individuals with severely constrained opportunities as a result
6      

of their impairment will not be deemed disabled. However, as reflection on

impairment will make clear such views should not simply be discarded due to
their lack of fit with my definition of disability, but also because they are
independently implausible as theories of justice.
Building on the contention that our goal should not be to ensure all
individuals can function normally by eliminating impairment as such,
Chapter 5 considers exactly how impairment should be understood. The
word itself naturally supports the assumption that it is something that war-
rants removal, which is further bolstered by its widespread definition as
deficient or suboptimal deviations from the species norm. This negative
account is controversial, however, since it is by no means straightforward to
determine which forms of functioning are worse. Further, this standard
negative conception of impairment risks undermining both the coherence of
disability pride (why be proud of a deficiency?), and the possibility of solidarity
(dividing the ‘genuinely’ deficient from those merely perceived as such). This
has led some to abandon the central distinction between disability and impair-
ment altogether.
I argue that, whilst the distinction should be retained, we should abandon
the negative account of impairment in favour of a neutral one. By this I mean
that impairment should be understood as anomalous rather than deficient
functioning. On this view, impairment is a broad church, and may include
conditions that are relatively inert or even beneficial in our current context, as
well as some that tend to make individuals’ lives much more difficult. As such,
I suggest that impairment is best described as a complex-difference, rather
than a mere-difference or a bad-difference. By this I mean that impairments
are neither necessarily harmful (i.e. negatively impacting individuals’ well-
being) nor necessarily disadvantageous (i.e. preventing individuals having
what they are entitled to), though they may be both.⁶ When understood as
difference rather than defect, and taking account of the complex effects of
difference, it is easier to see why our goal should be eliminating the capability
loss associated with impairment rather than eradicating anomalous function-
ing itself.
I contend that individuals with impairments should be given a voice in
determining both when the effects of these impairments are disadvantageous,
and the content of their entitlements. Such testimony supports the conclusion

⁶ Note that whilst disabilities are disadvantageous by definition on my view—they just are the subset
of impairments that prevent access to entitlements in a particular context—they also need not be
harmful. Some of the things we are entitled to—the ability for political participation, for example—will
have no impact on some people’s well-being.
  7

that many impairments do not, or need not, undermine access to entitlements.

However, this diverges from the widely held, and apparently common-sense
view of impairment as a paradigm case of disadvantage—indeed, as a tragedy.
Chapter 6 then considers how we should respond to this divergence of
views. We may be inclined to simply cede authority to those who have the
relevant lived experience, especially given they are frequently subject to unjus-
tified epistemic exclusion, silencing, and marginalisation. However, we should
be cautious of assuming that those with the most direct experience of oppres-
sion or deprivation are always those who are best placed to identify and object
to it. Part of the oppression impaired individuals face might be deprivation of
knowledge regarding how their circumstances might be improved. More
fundamentally, all individuals adapt to their circumstances, and may therefore
participate in and support oppressive norms, but this hardly constitutes
evidence that ableist—or patriarchal or racist—norms are justified.
We may worry, however, that the possibility of such unreliable adaptation
to circumstances—or ‘adaptive preferences’—simply provides a licence to
ignore marginalised voices. It is certainly possible to misapply the concept of
adaptive preferences, but this should not lead us to ignore the distortion that
adaptation can cause. If we rely solely on the oppressed to identify their
oppression, important sources of injustice will be ignored. Thus, this chapter
will argue, first, that a balance must be struck between taking individuals’
preferences and self-assessment as definitive at one extreme, and ignoring
them entirely at the other. The lesson of adaptive preferences is that we should
be cautious in our use of affected individuals’ testimony, not that it can be
justifiably ignored.
The second goal of this chapter is to resolve an ambiguity in current
accounts of adaptive preferences. By clarifying the various phenomena thus
categorised, I aim to show that correctly diagnosing adaptive preferences need
not constitute epistemic injustice, nor insultingly malign people’s rational
capacities. This is, in part, because unreliability often results from unjust social
circumstances, not an inherent lack of capacity. Being judged to have an
adaptive preference, then, need not always have the insulting implications
ordinarily supposed. Moreover, our account of adaptive preferences must be
sensitive to our particular goals and context: no single account of adaptive
preferences will be satisfactory for all the purposes to which it is put. In
particular, I distinguish between adaptive preferences that are an unreliable
guide to individuals’ interests (‘well-being adaptive preferences’), and those
that are a poor guide to just policy (‘justice adaptive preferences’). Individuals
may have adaptive preferences of the former but not the latter variety, and vice
8      

versa. Using my framework to determine when this is so allows us to avoid the

blanket exclusion of individuals’ voices on the basis of their potential unreli-
ability in some narrower domain, allowing for greater inclusion and better-
informed policy.
Finally, Chapter 7 begins to consider the implications of utilising my justice-
based account of disability. I have said that our focus should be accommodating
and ameliorating the disabling effects of impairments rather than removing
impairments themselves, and have assumed that it is the state who ought to
fulfil this obligation. However, there is a deep level of distrust of the state by
some sections of the disabled community, and for good reason: in the UK, for
example, funding has been cut to essential services, welfare sanctions have
been increased, the requirements that must be met to access what funding is
available have become narrow and exclusionary, as well as increasingly com-
plex and bureaucratic, and key changes in policy often are not communicated
in an accessible way.⁷ All of these problems and more have been exacerbated
by the response to the COVID-19 pandemic.⁸ Even when the state does make a
genuine attempt to advance the interests of disabled persons, this often
involves an assumption that it knows what these interests are, and the wishes
of intended beneficiaries are ignored.
Whilst I will not attempt to comment on the policies, motives, or prospects
of success of any particular state or government, I will argue that all state
action should be grounded in a commitment to anti-paternalism. Disabled
individuals should not be treated as passive beneficiaries, whose views and
voices are side-lined, but shown appropriate respect as autonomous agents.⁹
Often paternalism is objectionable because it fails by its own terms: interfering
with an individual to benefit them will frequently not actually promote their
interests. Yet I suggest that, even if successful, paternalism should be avoided.
I defend an authority account of anti-paternalism, according to which indi-
viduals’ will should be treated as authoritative in matters concerning their own
interests. Paternalism fails to respect this authority by making unauthorised

⁷ See e.g. Glasby et al. 2021; Institute for Government 2019; Morris 2011; Duffy 2013.
⁸ See e.g. Shakespeare et al. 2021; Institute for Government 2020; Kuper et al. 2020; Armitage and
Nellums 2020.
⁹ It is worth noting at the outset that, whilst I consider both bodily and cognitive impairments, I do
not consider those cognitive impairments that threaten individuals’ status as an autonomous agent.
Nor do I take a view on exactly who should be included within this category, beyond insisting that mere
cognitive impairment is not, in itself, sufficient to undermine this status. As ever, our focus must be on
the effects of the impairment and the degree to which it undermines someone’s ability to set and pursue
their own goals—that is, their ability to act autonomously—and not the mere possession of an
impairment. Clearly, no sharp line can be drawn, and many individuals can act autonomously at
some times and in some contexts and not others. I set these complications aside.
  9

appeals to individuals’ good as reasons for action, and thus potentially inter-
vening without adequate justification: for example, providing cochlear
implants to benefit deaf individuals, ignoring the heterogeneity of views
regarding whether such an option would be welcomed. I demonstrate that,
whilst some iterations of the capability approach may be guilty of paternalism
of this sort, my own version is not objectionably paternalist.
Yet paternalist redistribution is not merely objectionable because it misuses
individuals’ good, but because it mistakes the very aim of redistribution. As
I emphasise throughout the book, our goal should not be to ensure individuals
lead the best life possible, but that they are able to control the shape of their
life, deciding for themselves amongst an acceptable range of options. We
might not always agree with these decisions, but this is what respecting people
as autonomous agents demands. Our account of disabled individuals’ entitle-
ments should not entail the paternalistic imposition of policies on disabled
individuals as ‘passive beneficiaries’, nor assume that disabled individuals are
incapable of forming their own conceptions of the good, or of making autono-
mous decisions about how their own lives should go. The goal of just policy
should not be to enable identical functioning outcomes amongst all individ-
uals, disabled or not, but to enable individuals to exercise broadly specified
opportunities, however they choose.
This requires that we develop an account of justice that leaves space for all
individuals to set and pursue their own ends, and this cannot be achieved
without taking account of the lives, opinions, and needs of all individuals from
the outset. Our approach to disability must take seriously the diversity of
disabled lives, the justified pride many individuals feel in their disabled
identity, the abilities and capacities of disabled individuals (not just their
inabilities), and the importance of enabling disabled individuals to direct
their own lives. This has not been achieved by existing accounts of justice,
and will not for as long as disability is a mere afterthought.
 1
What Disability Is Not

1.1 The Problem of Defining Disability

My first task is to investigate how disability should—and should not—be

defined. It may seem that this is not a question in need of sustained philo-
sophical investigation. After all, most people have a clear sense of what they
mean by disability, and have little trouble identifying conditions they consider
disabling. Further, there is a broad consensus on some central features of an
account of disability (as §1.2–§1.4 will consider). In brief, these are, first, the
widely accepted distinction between impairment as physical or cognitive
features that cause deviation from a biological or statistical norm of function-
ing for our species, and disability as a restriction in our ability to perform
tasks.¹ There is considerable agreement, too, that disability is caused in part by
impairment, and in part by an individual’s social and political context and
infrastructure, and the resources they have access to.²
I do not intend to challenge the consensus on either of these points. Rather,
I ask: if disability is the restriction in our ability to perform tasks, then which
inabilities matter? In other words, which of the limitations that result from
individuals’ impairments, in combination with their context and resources,
should be considered disabling? We are all restricted by our bodies, and are
all incapable of performing some tasks, but most of these inabilities are not
considered disabilities. An account of disability is an essential prerequisite
to formulating public policies that safeguard the just treatment of disabled
individuals. But an account cannot form the basis of such policy if it
simply concludes that ‘everyone is disabled’. If, then, we are to avoid the
category of disability becoming overly broad—and thus politically and prac-
tically useless—we need some way of picking out which of these inabilities are
relevant to defining disability.

¹ E.g. Terzi 2004; Nussbaum 2006a; Shakespeare 2014; Smith 2001; Buchanan et al. 2000; Oliver
1996. Though some recent work questions this distinction (Barnes 2018; 2016a), as Chapter 5 will
discuss.
² E.g. Wolff 2009a; Kahane and Savulescu 2009; Shakespeare 2014; Shakespeare and Watson 2001;
Terzi 2004; Daniels et al. 2009; Vehmas 2004; Vehmas and Watson 2014; WHO 2002: 8–9.

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0002
     11

In the next chapter I defend a new account of disability, according to which

the relevant opportunity restrictions are those individuals are entitled to as a
matter of justice: to be disabled is to have these opportunities restricted. Here,
I lay the groundwork for this account by laying out central features of the
literature on disability, and then considering and rejecting alternative accounts
of disability. First, in §1.2, I sketch the widely drawn distinction between
disability and impairment.³ Then, in §1.3, I outline the current consensus
against pure versions of both the traditional medical model, and its rival, the
social model. The former suggests disability is straightforwardly caused by
impairment. The latter can seem to suggest that disability is solely the result of
unjust social structures. Clearly neither extreme is plausible, and, in §1.4, I lay
out a hybrid account, which takes the middle ground between these two poles.
Next, I argue that the opportunity losses that should be considered disabling
are not those entailed by ‘normal’ species functioning (§1.5),⁴ or those that
undermine well-being (§1.6),⁵ or those considered relevant by disability activ-
ists (§1.7).⁶ I demonstrate that the latter approach leaves disability radically
underspecified, whilst on the former two approaches, disability will be ubi-
quitous and hence meaningless. As such, all these alternatives prove to be an
inadequate basis for public policy, political campaigning, or activist move-
ments, and none can be used to identify and respond to the injustice, discrim-
ination, and mistreatment to which so many disabled individuals are subject.

1.2 Disability and Impairment

Most discussions of disability are premised on a distinction between impair-

ment and disability. Much can (and will) be said to clarify this terminology,
but roughly, impairment can be understood as atypical forms of physical or
cognitive functioning or ‘departure from human normality’.⁷ Impairments
might include blindness, deafness, and mobility impairments, as well as

³ These terms are investigated in much greater detail in Chapter 5.

⁴ Buchanan et al. 2000; Daniels 1985.
⁵ Kahane and Savulescu 2009; Savulescu and Kahane 2009; Harris 2000, 2001; Terzi 2008, 2009;
Wolff 2009a.
⁶ Barnes 2016a.
⁷ Terzi 2008: 43. Or, on the WHO’s (2002: 10) definition: an impairment is ‘a problem in body
function or structure’, where body functions are ‘physiological functions of body systems (including
psychological functions)’, and body structures are ‘anatomical parts of the body such as organs, limbs
and their components’. To avoid the normative connotations—that to be impaired is, itself, a
problem—it is preferable to talk in terms of deviation from the norm, as Chapter 5 will argue. The
WHO account of disability is discussed in this section. Impairment is considered further in §1.5.
12      

non-standard cognitive functionings experienced by individuals with autistic

spectrum conditions (ASCs), Down’s syndrome, or Alzheimer’s. Disabilities
are the ‘activity limitations, and participation restrictions’ that may result.⁸
Providing examples of disability is more complicated, given that determining
which of the restrictions in our abilities to perform tasks are disabling is
exactly what is at issue. However, some general points can be made.
First, it is usually assumed that these tasks should only be those that are
considered, in some sense, significant. Certainly, an account would be deeply
implausible if it implied that our trivial inabilities—being unable to raise an
eyebrow, or sing, or stand on one leg—count as disabilities. Second, whilst
impairment focuses on differences in our body and mind, disability tends to
focus on the social roles we are consequently unable to fulfil. For example,
blindness is an impairment of our sight, whilst an associated disability might
be our inability to drive a car or, even more specifically, to drive our child to
school. ASCs restrict our ability to read others’ emotional responses, whilst the
associated disability might be the resulting difficulty in, for example, working
in occupations that require understanding and effectively reacting to these
responses, such as being a social worker or car salesperson. In Jonathan
Wolff ’s terminology, disability is a restriction of ‘social functionings’.⁹
This distinction between disability and impairment is not employed uni-
versally. First, Elizabeth Barnes rejects this distinction entirely, referring only
to disability (see §1.7 and Chapter 5). Second, the WHO, whilst acknowledg-
ing the existence of impairment as a distinct category, nonetheless considers
this to be incorporated into its broader classification of disability. Thus, they
define disability as: ‘an umbrella term, covering impairments, activity limita-
tions, and participation restrictions’, where ‘an activity limitation is a difficulty
encountered by an individual in executing a task or action’, and ‘a participa-
tion restriction is a problem experienced by an individual in involvement in
life situations’.¹⁰ Leprosy, for example, is a disability that involves the impair-
ment of losing sensation in extremities, activity restrictions such as difficulties
in grasping objects, and participation limitations such as unemployment
resulting from the stigma surrounding leprosy.¹¹
In some ways, the terminological disagreement between the many theorists
who see impairment and disability as distinct categories and the WHO is just
that: a matter of semantics. Both acknowledge the same core components. On
one account, disability is seen as certain consequences (restrictions in activities
and participation) arising from a particular cause (impairment, plus external

⁸ WHO 2017. ⁹ Wolff 2009a: 125. ¹⁰ WHO 2017. ¹¹ WHO 2002: 17.
     13

circumstances); on the other, disability incorporates both part of the cause (the
impairment, not the circumstances) and the effect. Nonetheless, I believe there
are reasons to favour the former approach.
First, it is widely acknowledged (including by the WHO) that the restrictions
in individuals’ capabilities are not solely the result of their impairment, but also
depend on their social, political, and material environment.¹² As such, it seems
more natural to understand disability as the outcome of these various factors,
rather than folding one of the causes into the concept itself. Indeed, elsewhere
the WHO itself defines disabilities as ‘the outcomes of interactions between
health conditions (diseases, disorders and injuries) and contextual factors’.¹³
Second, whilst we might reasonably seek to eliminate the restrictions and
limitations that can result from impairments in some contexts, it is much
more controversial to suggest that impairments themselves should always be
eliminated. Distinguishing disability and impairment, then, allows us to cam-
paign against disability (as restriction) without suggesting that we should also
seek to normalise human bodies and minds (remove impairment).¹⁴ However,
if impairment is understood as part of disability, then reducing instances of
disability will necessarily require reducing instances of impairment: even if an
individual no longer faced restrictions in activity or participation, they would
still be disabled as long as they still have an impairment on the ‘umbrella view’.
This obscures the fact that in many cases the restrictions associated with
impairments can be removed by contextual changes alone, and that sometimes
the focus of policy ought to be contextual changes alone.
Third, then, distinguishing disability and impairment draws attention to the
contingency of the connection between bodily and cognitive difference and
having a disability, which also depends on context, resources, and personal
features (as Chapter 5 explores). Indeed, highlighting this contingency was a
major motivation behind the development of the social model by the disability
rights movement.¹⁵ Finally, as Chapter 5 also discusses, retaining a concept of
impairment allows for a sense of community and solidarity amongst those who
function in ways that are unusual, and often poorly accommodated by institu-
tions, norms, and infrastructure—regardless of whether this results in disability.
I will, therefore, continue to distinguish disability and impairment.
Nonetheless, objections may be raised against this terminology. First, whilst

¹² For example, the WHO notes that disability ‘is a complex phenomenon, reflecting the interaction
between features of a person’s body and features of the society in which he or she lives’ (WHO 2017).
¹³ WHO 2002: 10 (emphasis added). ¹⁴ As §1.7 will discuss.
¹⁵ The social model not only drew a conceptual distinction between impairment and disability but
also, in some iterations, disputed the causal connection between them (§1.3).
14      

common among disability scholars and activists, outside this community this
language might strike some as clumsy and counterintuitive. We are very used
to calling people with impairments—of vision, mobility, cognitive function—
disabled, without paying much attention to the restrictions in activity or
participation they may or may not face. It may seem awkward to insist that
an individual in a wheelchair—often the literal poster child of disability—is
merely impaired, and that this might not translate into a disability. Yet this
discomfort is often grounded in the unexamined assumption that differences
in physical and cognitive functionings must manifest themselves in restric-
tions of ability, and in the associated belief that the impaired body or mind is,
itself, the problem.
These intuitions and assumptions, however strongly held, ought to be
questioned, as the disability rights movement has long argued. We should
recognise that we all have impairments of various sorts—we all deviate from
the norm in some way or other—but this need not be disabling. Indeed, that
even those deviations most widely assumed to be disabling need not be so. Our
focus should be on significant restrictions in what individuals are able to do
and how they are able to participate in society, and not on how their body and
minds differ. We should not assume that the former straightforwardly follows
from the latter.
A second, more serious worry concerns the potential normative implica-
tions of labelling individuals ‘impaired’. This is not intended to be an evalu-
ative concept. It simply provides a way of distinguishing disability, which
(partially) results from impairment, from other forms of disadvantage, such as
that associated with some racial or gender identities, or sexual orientation.¹⁶
Further, (almost) everyone in some way deviates from the norm; hence, has an
impairment. Nonetheless, this term has connotations of deficiency, defect, and
imperfection. It may, therefore, seem more apt to refer to ‘difference’, ‘diversity’,
or ‘heterogeneity’. I will begin to talk in terms of ‘anomaly’ in Chapter 5 when
I defend a neutral account of impairment, but do not do so now in order to retain
some neutrality over the definition of impairment and to avoid the proliferation
of competing terminology. But to emphasise: it is my view that impairment
should be understood as synonymous with difference rather than deficiency.
This neutral account of impairment may seem to leave my approach
vulnerable to a third objection: that it will render the category of disability

¹⁶ This is in addition to the other noted benefits: highlighting both the contingency of the connec-
tion between impairment and disadvantage, and the shared experience of atypical functioning.
A further reason to accept this distinction is that I am not convinced by the arguments of its primary
opponent—Barnes (2018, 2016a: 23–8)—for its rejection. These I consider in Chapter 5.
     15

over-inclusive. If impairment implies mere atypicality (rather than deficiency)

then, it may be argued, it will include almost everyone. However, as will shortly
become clear, an expansive, neutral account of impairment does not necessitate
an all-encompassing concept of disability, and it is the latter prospect, not the
former, that should concern us. For those who remain unconvinced by the idea
of neutral impairment, this should not present a barrier to accepting my general
account of disability. It is possible to agree that individuals are disabled when
they lack the opportunities they are entitled to, but focus on cases in which this
results from functioning in a ‘deficient’ or ‘unhealthy’ way (i.e. negative impair-
ment). For now I merely defend the framework of my account.

1.3 The Medical and Social Models of Disability

The preceding discussion foreshadows much of the disagreement between pro-

ponents of the medical and social models of disability, since their core conflict
centres on disagreement over how the causal relationship between disability and
impairment should be understood. In brief, a strict medical model would suggest
that disability is straightforwardly caused by individual impairment, whilst a
strict social model would deny this causal relationship, and insist instead that
disability is solely the result of unjust social structures. There is now a broad
consensus that neither of these extreme positions is plausible (and, indeed, may
never have been endorsed in this form), and that we should instead adopt a
hybrid account. It is important to understand why this is so.
The problems with the medical model may, by now, seem obvious: it
ignores the ways in which social and institutional structures and the material
environment can restrict the opportunities of individuals with impairments
and create disability. As such, it fails to capture an important element
of disabled individuals’ experiences. Consider, for example, Alison Davis’s
account of living with spina bifida:

if I lived in a society where being in a wheelchair was no more remarkable

than wearing glasses and if the community was completely accepting and
accessible, my disability would be an inconvenience and not much more than
that. It is society which handicaps me, far more seriously and completely
than the fact that I have spina bifida.¹⁷

¹⁷ Davis in Newell 1999: 172.

16      

A purely medical model would simply fail to acknowledge the ways in which
Davis’s social context contributes to her disability. The various contextual
factors that may be disabling will be considered further below, but might
include architectural design inaccessible to wheelchair users, social norms
according to which wheelchair users are patronised, ignored, or even abused,
and a material environment, such as a hilly landscape, that makes wheelchair
use difficult. The medical model ignores the significance of such contextual
factors and identifies the ‘problem’ of disability as wholly internal to, and the
result of, individual impairment. Thus, the solution to this ‘problem’ must be
to cure this impairment: to eliminate impaired bodies and minds.
Not only is such an approach inaccurate, failing to capture much of what
causes disability, it is also considered by many to be offensive. Consider, for
example, Steven Smith’s claim that this model:

render[s] disabled people as passive and powerless targets of intervention

in the face of non-disabled expertise and in the process this reduces
the person and their experience to the ‘abnormal’ and ‘deficient’ medical
condition.¹⁸

Defenders of the medical model may object to some of these lines of criticism
and insist that the targets of Smith’s critiques are not the necessary outcomes
of the model, but rather the contingent consequence of the high-handed and
sometimes condescending way in which it has been applied. It is true that
the assumption that disability is caused by impairment does not necessarily
imply that those who have these impairments are passive or powerless, nor
that they are reduced to their condition. Nonetheless, this is likely to be a
consequence of seeing impairments as the problem, since the impairments
must then be the target of policies and interventions. Further, on a medical
model, it becomes hard to make sense of the testimony of individuals such as
Alison Davis, who see their impairment as no more than an inconvenience. If
disabled individuals’ testimony is deemed untrustworthy, it will likely be
ignored, and this may well render individuals passive and powerless (as
Chapters 6 and 7 will explore).
It is clear, then, that the medical model is deeply unsatisfactory, and much
of the progress towards recognising its limitations has been achieved by
proponents of the social model, who successfully drew attention to the
importance of contextual factors. In large part, this was the practical goal

¹⁸ Smith 2001: 24.

     17

they aimed to achieve.¹⁹ However, versions of the social model that see
disability as rooted in social factors alone, and claim that ‘disablement is
nothing to do with the body’, also do not provide a plausible account of
disability.²⁰ We can draw out three related lines of criticism. First, whilst social
structures can certainly be disabling, it is also clear that disability could not be
eliminated by social changes alone, and that it should not be reduced to its
social causes. Second, even when causes of disability are social they may be
unavoidable, and so not resolvable with merely social changes. Finally, not all
social causes are unjust, as some iterations of the social model claim.
First, then, some impairments—for example, chronic pain, severe paralysis,
or serious cognitive impairments—clearly restrict individuals’ opportunities
independently of the influence of social structures.²¹ This is not merely to
point out, as Barnes does, that if someone is in chronic pain, then ‘no amount
of social awareness would be able to fully alleviate the ways in which they
suffer from their disability’.²² This is undoubtedly true, but it is unfair to
proponents of the social model to reduce their concern with environment and
context to improving ‘social awareness’ and changing social norms. Norms,
biases, and stigma can certainly be disabling. This can be demonstrated, for
example, by the fact that unemployment is a consequence or feature of leprosy
(and, indeed, of a great many other disabilities, especially in developing
countries²³). However, the social model also draws attention to the influence
of broader social arrangements, such as how information is communicated,
the form of transportation that is common, architectural design, and legal
processes, which introduce a host of new ways in which impairments might be
disabling.
Yet even this may be too narrow, insofar as its focus is still restricted to
social, rather than broader ‘external’ or ‘contextual’, factors. Failing, for
example, to include features of the wider material environment, such as
physical terrain and climate: whether it is mountainous or flat, wet or dry,
warm or cold. Such features can significantly alter the impact of many
impairments. My goal here is not to determine what the boundaries of the
social model are, or should be. If nothing else, it would stretch our under-
standing of the term to describe these features as social. Yet maintaining that
disability is rooted solely in social causes would mean failing to account for the

¹⁹ E.g. see: Terzi 2008: 44–7; Shakespeare and Watson 2001: 9–12; Shakespeare 2014: 11–17.
²⁰ Oliver 1996: 35. Note that Mike Oliver’s view is complex, and may not be well represented by this
oft-used quote. However, my goal is to simply outline and critique one possible approach, rather than
to determine which authors it can be attributed to.
²¹ E.g. Barnes 2009b; Terzi 2004. ²² Barnes 2009b: 338. ²³ E.g. Mitra et al. 2011.
18      

ways in which the wider material environment can cause disability, and
expanding the account to include them seems consistent with the motivations
of many proponents of the model.
We have good reason, then, to take account of the impact of all external
factors, both social and natural. However, even this broader construal of
context is insufficient. Whilst many forms of chronic pain, for example, may
be exacerbated by such contextual factors—transportation that requires phys-
ical exertion, architecture that includes many stairs, open spaces without rest
areas, cold climates, or mountainous terrain—it is also clear that circum-
stances could not be designed to eliminate chronic pain in all its forms.
Thus, many impairments will continue to be disabling (to some degree) in
any likely social and material context. Whilst disability should not be reduced
to the effects of individual impairment, then, we also should not ignore the
directly disabling effects impairments may have. That is, ‘impairment effects’.
This relates to a second problem with the social model’s focus on social
causes of, and so social solutions to, disability: that there is no social environ-
ment that can be perfectly inclusive of all impairments. The previous point
drew attention to the fact that some disabilities will not be fully mitigated by
contextual factors, even broadly construed; this observes that even when
disabilities could be thus mitigated, it may be that there is no set of social
changes that would eliminate all disabilities. As Linda Barclay notes:

A society maximally accommodating of the blind will be different to one that

is maximally accommodating of the deaf . . . The shape of any social envir-
onment will always be more favourable to some people than it is to others,
even if only in some respects and in some contexts.²⁴

Or, as Tom Shakespeare points out:

People with mobility issues who do not use wheelchairs may find that
steps are safer and easier for them than ramps . . . Wheelchair users may
have problems with tactile paving which gives locational cues to visually
impaired people . . . Partially sighted people may request large text on white
background: people with dyslexia may prefer black print on yellow paper.
Some people will prefer rooms to be dim, others will prefer them to be
brightly lit.²⁵

²⁴ Barclay 2010: 161. ²⁵ Shakespeare 2014: 37.

     19

Deafness may only be disabling as the result of context, but if changing this
context were to make blindness more disabling, then deafness may effectively
be as intractably, and context-independently disabling as chronic pain.
This connects, in turn, to a final objection to a particular interpretation of
the social model, according to which disability is not merely externally caused,
but is the result of unjust social causes.²⁶ On this interpretation, the social
model does not merely claim that disability is caused by the social environ-
ment, but maintains that these ‘beliefs, attitudes, and social arrangements also
need to be mistaken and unjust’.²⁷ Further, that a more just social arrangement
is both realistic and possible.²⁸ However, this version of the social model is
vulnerable to the objection that at least some of the social causes of disability
are not unjust.²⁹
As the previous discussion has highlighted, some social causes may be
impossible to avoid. If this is so, then it may seem implausible to suggest
that settling on a social arrangement that is still disabling to a subset of
individuals is unjust. In some instances, this is because we lack the technology
or the resources to alter the social environment in ways that would accom-
modate some impairments. For example, speech-to-text programmes for
individuals with hearing-loss, and text-to-speech programmes for those with
visual impairments may significantly reduce, or eliminate, activity and par-
ticipation restrictions that would otherwise be associated with these impair-
ments. However, before the technology existed to provide useful versions of
these programmes, their absence need not imply the state has acted unjustly—
assuming sufficient effort has been made to research and develop them.
Equally, if certain states simply lack the resources to provide these accommo-
dations, their failure to do so may not imply they are guilty of injustice.³⁰

²⁶ Again, my goal is not to discuss the history of the social model, nor to take a view on which
interpretation should be taken as authoritative. For discussion of the provenance of the social model see
e.g. Shakespeare (2014: 11–46); Shakespeare and Watson (2001); Terzi (2004).
²⁷ Kahane and Savulescu 2009: 39. ²⁸ Kahane and Savulescu 2009: 40.
²⁹ Providing a complete account of which causes are unjust would require a complete theory of
justice. However, the general objection is likely to follow from any account.
³⁰ This may, of course, be an instance of global distributive injustice, insofar as unjust inequalities in
global resource distribution leave states in a position in which they cannot provide for their citizens.
However, I will not investigate this topic here. This point also foreshadows a possible objection to my
account: if these failures of accommodation are not unjust, and I suggest disability means lacking what
we are entitled to as a matter of justice, then it may seem ineliminable restrictions are not disabling on
my account. As §2.4 will argue, this is not so, since human beings may be entitled to opportunities or
resources that cannot be provided universally, and whose lack of provision is not the result of injustice.
Thus, insisting it is not unjust to uphold a social arrangement that unavoidably disadvantages some of
those who live under it need not commit us to also suggesting that such disadvantages are not,
themselves, unjust. In other words, a situation may be unjust though no agent has committed
injustice.
20      

In other cases, these accommodations cannot be made because the needs of

individuals with different impairments pull in different directions, and there is
no single social structure that is ideal for all. As noted, even if there are social
arrangements that would improve some individuals’ situation, this might
come at others’ expense. Thus, these cases can prove as intractable as those
in which we simply lack the ability to institute the relevant change. Yet the fact
that some social causes are unavoidable does not imply that the status quo
should not be challenged: that we should not work to change existing social
arrangements, and to shift the boundaries of the possible.
I by no means wish to claim that no social causes are unjust. Rather, I wish
to point out that the fact that a cause is social need not imply that it is unjust.
We can, therefore, allow that it may be unjust to select an arrangement that is
only designed for the able-bodied majority (or minority); that it may be unjust
to fail to make changes that reduce disability at minimal resource cost, and
without disabling others; that it may be unjust to fail to challenge discrimin-
atory social norms, biases, and stigma; and that it may be unjust to expend few
resources developing assistive technologies or design modifications, thus arti-
ficially maintaining the infeasibility of accommodating individuals with
impairments. However, if social arrangements are as good as they can be, by
whatever metric this is measured—perhaps reducing the instance of disability,
perhaps reducing the most severe disabilities, perhaps reducing disability
amongst the most vulnerable—they are not plausibly unjust. Yet they may
still be disabling.
The most plausible reading of the social model, then, is as an approach that
roots the cause of disability in the impact of contextual features on individuals
with impairments. This does not restrict our focus solely to social causes, nor
to those that are unjust and can be changed.³¹ Yet even the broader focus on all
contextual factors is insufficient, insofar as this ignores the direct impact that
our bodies and minds can have on what we are able to do, and how we are able
to live. Returning to the first critique: some impairments will be disabling in
any context.

1.4 Beyond the Medical and Social Models

Given the problems that plague both medical and social models, the right
approach is surely a hybrid: an account that acknowledges the influence of

³¹ This may mean that its title (the social model) begins to look increasingly inappropriate.
     21

contextual factors, without divorcing the concept of disability from the effects
of our bodies and minds. An approach, in other words, that takes on board the
above criticisms, and understands disability as a restriction in the ability to
perform tasks as the result of both internal and external factors. This is a view
that is now widely held.³² I turn in §1.5 to the various competing answers to
the question on which I believe most interesting debate turns: which inabilities
are disabling? But before doing so, I first delineate these internal and external
factors. Whilst the preceding discussion has given a sense of what these
might be, it is worth explicating the various ways in which impairments may
interact with individuals’ context to restrict their capacity for activity and
participation.
First, internal features of individuals distinct from their impairment may
contribute to the ways in which their impairment is disabling. This is not
clearly emphasised in either social or medical models, but does feature in
various hybrid accounts, especially those grounded in the capability
approach.³³ Evidently, the impact of an impairment on an individual will
depend not only on their context but also on how it interacts with their
other characteristics. The WHO names this category ‘personal factors’, and
draws attention to: gender, age, coping styles, social background, education,
profession, past and current experience, overall behaviour pattern, character
and other factors that influence how disability is experienced by the individual.³⁴
For example, cognitive impairments such as Down’s syndrome may be less
disabling in younger children when there is less of a discrepancy in function-
ing capacity and behaviour between them and their peers; mobility impair-
ments will be less disabling to an individual who wishes to pursue the
sedentary career of a philosopher than to one who wishes to pursue the active
life of a park ranger; and the ability to adapt to changes in one’s circumstances,
and modify one’s goals and conception of the good, has been shown to have a
significant impact on individuals’ ability to cope with a decrease in their
cognitive capacities following traumatic brain injury.³⁵
Second, a wide range of external factors can bear on whether impairments
result in individual limitations. Note that whilst some of these factors may be

³² See e.g. Wolff 2009a; Kahane and Savulescu 2009; Daniels et al. 2009; Terzi 2004; Shakespeare
2014; WHO 2002; Vehmas 2004; Riddle 2013. This is sometimes called the ‘biopsychosocial model’
(cf. Engel 1977, though Engel was developing a model of disease rather than disability), the ‘interactive
model’ (e.g. Wasserman 2006: 216), or the ‘interactional approach’ (Shakespeare 2014: 74–84).
³³ E.g. Wolff 2009a; Terzi 2008. ³⁴ WHO 2002: 10.
³⁵ On the latter example, see Daniels et al. 2009.
22      

unjust, they need not all be so. Drawing on and modifying the WHO’s
categories,³⁶ we can distinguish:

(a) Services, Systems, Policies, and Institutions. This includes:

(i) Formal legal and institutional policies. For example, anti-
discrimination legislation (or its absence), legal mandates that
reasonable accommodation should be made to ensure individuals
can access buildings and facilities, or schemes such as the UK’s
‘Disability Confident’ (formerly ‘Two Ticks’) to improve disabled
individuals’ access to the job market.³⁷
(ii) The provision of services. For example, appropriate educational
provision for those with special educational needs, and the pro-
vision of welfare payments and other benefits.
(b) Norms and Attitudes. This includes:
(i) Individual attitudes. For example, open hostility, mistreatment,
and abuse. This also includes the condescending and potentially
insulting behaviour of individuals who are ‘trying to help’, yet
assume that individuals with impairments are incapable when
they are not, and who provide this assistance without first asking
for permission.³⁸
(ii) Informal institutional practices. This incorporates both practices
that are clearly discriminatory and unjust, as well as those that
may contingently disadvantage individuals with certain impair-
ments. The former might include a prejudicial reluctance to
hire individuals with impairments; whilst the latter might include
the increasing availability of white-collar, rather than manual,
work as disadvantageous to those with learning difficulties, such
as dyslexia.
(c) Natural and Material Environment. This includes:
(i) Natural features of the landscape and climate. For example,
mountainous terrain making wheelchair use difficult, or a cold
and wet climate leading to icy pavements, treacherous for indi-
viduals with impairments involving their vision, mobility, or
balance.

³⁶ These are listed as: Products and Technology; Natural Environment and Human-Made Changes
to Environment; Support and Relationships; Attitudes; Services, Systems, and Policies (WHO 2002:
16). The content of these categories is not clearly elaborated.
³⁷ Although this scheme may not have had this effect in practice (Hoque et al. 2014).
³⁸ Cureton 2016. Also see Chapter 7.
     23

(ii) Human-made changes to this environment (or their absence).

For example, installing handrails on steep roads, including rest
areas in hilly terrain, or paving paths over rocky or uneven
ground (or failing to do so). This may also include environmental
changes that increase individual restrictions, such as using gravel
or sand on previously smooth ground.
(iii) The design of material infrastructure. For example, kerb cuts,
textured paving, providing alternatives to stairs such as lifts or
ramps, auditory as well as visual cues for pedestrian traffic
signals, and accessible entrances that do not require individuals
to travel long distances to reach them.³⁹ The architectural ideal is
surely universal design, but this may not always be feasible.
(d) Products, Technology, and Resources. This includes:
(i) The availability and creation of new products and technologies.
This requires funding for the research and innovation necessary
to develop new products, and the support necessary to ensure
such products are available, even if the market for them is
relatively small.
(ii) Individual access to these products and technologies. This may
be achieved either through direct state provision, or because
individuals possess sufficient personal resources to acquire what
they need.
(e) Support Networks, Relationships, and Community. This includes:
(i) Formal assistance. For example, access to home care workers,
nurses, and social workers. This may be provided by the state or,
when it is not, gaps may be filled by charities, church groups, and
other volunteer support workers.
(ii) Informal support networks of family and friends.

Both informal networks and formal assistance may have great value to
individuals, and are frequently relied upon. However, it should be noted that
such reliance may render individuals vulnerable when mechanisms are not in
place to ensure that this support is securely provided. For example, if the
provision of state assistance is seen as a matter of benevolence rather than
right, individuals may fear the withdrawal or reduction of this provision.

³⁹ Ashley Shew’s (2017) account of the impact of Virginia Tech’s failure to take accessibility into
account in a campus remodel demonstrates the importance of such features. As she says, ‘there's no
reason it should be this hard just to be on campus’.
24      

Third sector and informal networks may also be insecure, depending, for
example, on charities maintaining their funding, on individuals maintaining
relationships with friends and family who take on a caring role, or on these
individuals continuing to have the capacity to provide assistance. Even if these
dangers are never manifested, this vulnerability to the withdrawal of necessary
assistance may generate the same worries that accompany the threat of
arbitrary interference in republican thought. In brief, these worries are: the
anxiety over the possibility of withdrawal; strategic deference, anticipation,
and flattery in an attempt to avoid this outcome; and the undermining of the
equal social status when one individual is forced to ‘bow and scrape’ to
another.⁴⁰ This provides reasons to ensure that individuals’ access to networks
of support is made as secure as possible, where this will likely require institu-
tional protection: to ensure the formal provision is not withdrawn, and to act
as a failsafe against the breakdown of informal networks. (The importance of
such security is discussed further in §4.2.)

1.5 Species Norms and Impairment

I now turn to the various answers that have been given to the central question
of which restrictions in individuals’ activities and participation should be
considered disabling. One seemingly natural answer is that disability should
be understood as the loss of abilities that members of the species are normally
able to perform. These are ‘normal species functioning’, or ‘species norm’
approaches. This is the view of Allen Buchanan et al., who define disability as
being ‘unable to perform some significant range of tasks or functions that
individuals in someone’s reference group (e.g. adults) are ordinarily able to do,
at least under favourable conditions’.⁴¹ It also seems to be the view underlying
many folk intuitions regarding what it means to be disabled.
However, despite these common-sense roots, this account will not,
I contend, be able to capture all the cases we would want it to. Most obviously,
perhaps, such accounts will be over-inclusive since individuals depart from the
species norm in many trivial (or even beneficial) ways. Raising an eyebrow
might be a species-normal ability, but those who cannot do this are hardly
disabled by their inability. Buchanan et al. attempt to avoid this problem by

⁴⁰ Pettit 1997: 85–9. For further discussion of the harms of insecurity, even when these dangers are
not realised, see e.g. Wolff 2009c; Wolff and de-Shalit 2007: 65–73; Perry 2009.
⁴¹ Buchanan et al. 2000: 286. Also see Gregory 2020.
     25

distinguishing ‘impairment of normal species functioning’⁴² from disability as

the resulting inability to perform a significant range of tasks. To illustrate this
difference, they give an example of someone who cannot hear sound at a
particular range of frequencies, yet in whose environment nothing ‘she is likely
to be required to do or would benefit from doing’ ⁴³ requires hearing sound in
this range. This individual thus has an impairment, but this does not translate
into a disability.
However, this distinction between inabilities to perform species-normal
functionings, and inabilities to perform species-normal tasks is difficult to
draw—at least, without some additional account of which of these tasks are
‘significant’. Indeed, in describing their example, Buchanan et al. make refer-
ence to what someone might ‘be required’ to do or ‘would benefit’ from doing.
By doing so they can avoid the inclusion of trivial inabilities as disabilities.
However, this also suggests that their means of identifying the relevant
restrictions in our abilities is not purely based on a species norm. Thus, it is
not clear that this is any longer a species norm approach.⁴⁴
Buchanan et al. may respond that what they have in mind is species-normal
requirements or benefits. However, this would raise a dilemma for the
approach. On the one hand, these may be understood in a relatively thin,
and thus relatively uncontroversial, way—for example, focusing only on basic
human needs. Few would dispute that humans require, and would benefit
from, access to nutrition, clean water, and shelter. However, we would be left
with a very narrow account of disability if we were only considered disabled
when we are unable to fulfil this narrow set of needs. On the other hand, then,
we may wish to adopt a thicker account, perhaps including education, social
relationships and family life, political participation, leisure time, and the
chance for autonomous self-direction. Yet revision of this kind will move us
beyond a simple statistical analysis of what adult humans are normally able to
do, and into more complex and controversial normative terrain than is
generally acknowledged by advocates of the species norm approach.

⁴² Buchanan et al. 2000: 285. ⁴³ Buchanan et al. 2000: 287.

⁴⁴ Alex Gregory’s approach to this problem is to allow that views focusing on species atypical
inabilities (as his inability theory does), will indeed include trivial and insignificant inabilities as
disabilities, but to insist that they ‘are not appropriately described as such’ (Gregory 2020: 31, my
emphasis). However, this seems to shift rather than solve the problem. We are still left with the task of
determining which inabilities are appropriately described as disabilities, which raises the same dilemma
as for Buchanan et al. Either we devise a principle that determines which are ‘sufficiently life-changing’
(Gregory 2020: 31) in their effects (hence moving away from a species norm account), or leave this
open to idiosyncratic individual judgement without a normative guide. This would then open the door
to inclusion of trivial inabilities, as well as leaving the concept deeply indeterminate, which is a problem
at least in the context of theorising about justice and individual entitlements (as §1.7 will discuss).
26      

The point is not that this task is necessarily impossible, or has never been
attempted.⁴⁵ Rather, the objection is that the account now hangs on questions
concerning what is valuable or essential to human life (or, perhaps, a good or
flourishing human life), and not merely questions of what is normal. Thus,
opting for this latter horn of the dilemma will mean that the species norm
approach collapses into other, more substantive approaches: perhaps a wel-
farist one if the inabilities that matter are those that reduce human welfare, or
a justice-based approach if the inabilities that matter are those that humans are
entitled to. I discuss the former approach in the next section, and the latter in
the next chapter. Either way, this can no longer be plausibly interpreted as a
species norm approach.
A second problem for the approach is that its focus on ‘tasks’ may mean that
it is under-inclusive. Consider, for example, individuals who suffer from
chronic pain, fatigue, anxiety, or depression, but who are nonetheless high-
functioning and able to perform all species-normal tasks (or those they are
required to do, or that benefit them). These cases are analogous to that of the
individual who cannot hear particular sound frequencies: they may not func-
tion as normal, but this does not translate into a restriction in their abilities.
Yet it is implausible to suggest that individuals who struggle with pain, fatigue,
anxiety, or depression are not disabled until this manifests in an inability to
perform tasks.⁴⁶
Buchanan et al. could respond by adopting a more expansive understand-
ing of tasks, and thus of disability. They might point out that performing
tasks whilst suffering is not species-normal: we should not focus just on
what people can achieve but on whether their method of achievement is
normal. This response would, indeed, allow them to include individuals with
chronic pain, depression, and so on, but would come at the cost of also
including their hearing-impaired person, and other trivial cases. After all,
the species-normal way of performing the task of hearing includes, inter
alia, not being in pain and hearing sound frequencies within a certain range.

⁴⁵ Indeed, this is the focus of much of Nussbaum’s early, Aristotelian work (e.g. Nussbaum 1992,
1993).
⁴⁶ Gregory (2020: 28) insists that we can solve this problem with regards to fatigue, at least, since
this will be associated with ‘lower levels of ability at various temporally extended tasks’. Leaving aside
the potentially expansive implications of allowing that individuals are disabled whenever they have
lower levels of ability at particular tasks (not just absolute inability), we may also worry individuals
will not be included in cases where this does not actually impair someone’s abilities in sufficiently
significant ways, though it does increase the difficulty of performing tasks. For example, they are able
to work a forty-hour week, but then need to take significant time to rest and recuperate, undermining
their ability to pursue recreational activities. Further, Gregory’s response does not help with the other
problem cases.
     27

This fully inclusive approach is both implausible and contrary to their stated
view. The route out would involve an appeal to which of the atypical ways of
achieving a task are relevant or significant and this, as noted, takes us
beyond the appeal to mere species norms. (As §2.6 will consider, this
route is open to me since I am explicitly committed to a principle that
identifies which of the various ways of performing atypical functionings are
disabling.)
This gives us good reason to reject species norm accounts of disability,
but it does not follow from this that we must also reject accounts of
impairment grounded in species norms—indeed, I adopt just such an
account (§1.2; §5.3). This is, first, because impairment should be an expan-
sive category. Thus, we need not resist the inclusion of the individual with
the mild hearing impairment, and those with anxiety, depression, chronic
pain, and fatigue—and even those who cannot raise an eyebrow or, as we
will see, individuals who apparently function unusually well. Impairment,
then, will be a very broad church. This does not, however, imply that
disability must be similarly all-encompassing, since, unlike Buchanan et al.,
I explicitly offer a criterion to identify which of the associated restrictions
are disabling.
Second, if impairment is, as I suggest, understood to be normatively
neutral—synonymous with difference, not defect—it is not vulnerable to a
final objection often raised against species norm accounts of disability. This is
the worry that focusing on an idealised conception of normal species func-
tioning entails a corresponding disrespect of, and insult to, those who fail to
meet this ideal. For example, as Lorella Terzi notes: ‘the whole ideology of
normal functioning is seen as primarily constructed in order to control and
exclude disabled people from active and full participation in social and
institutional arrangements which have no interest in accommodating
them’.⁴⁷ This line of criticism is similar to those that draw attention to the
medical model’s inaccurate and insulting focus on apparent individual
deficiency, and insistence that this must be rectified. However, an account
of impairment that is indexed to species norms merely acknowledges that
many (or all) of us are not ‘normal’ in various ways and, whilst not prob-
lematic in itself, this can, in some circumstances, prevent us from having
the opportunities we are entitled to. The same can be said of gender and
racial identities and sexual orientation, without implying that any such
identity is deficient.

⁴⁷ Terzi 2009: 90.

28      

1.6 Welfarist Accounts

The second account of disability I will consider is the welfarist approach. This
is most clearly and explicitly defined and defended by Guy Kahane and Julian
Savulescu, according to whom disability is a ‘stable physical or psychological
property of subject S that tends to reduce S’s level of well-being in circum-
stances C, when contrasted with a realistic alternative, excluding the effect that
this condition has on well-being that is due to prejudice against S’.⁴⁸ Thus, the
relevant inabilities, on this view, are those that reduce welfare. Whilst other
welfarist views have been defended, I will focus on Kahane and Savulescu’s
version of the approach here.⁴⁹
Perhaps the most obvious problem with this account concerns the difficul-
ties in developing an account of welfare. Whilst Kahane and Savulescu aim to
remain neutral between alternative accounts of welfare, there are problems
with any that might be incorporated. This reflects the more general difficulties
with formulating a clear and satisfactory account of welfare, which I will not
rehearse here.⁵⁰ However, in brief: first, a subjective account of welfare, relying
on individuals’ self-assessment, will inevitably raise reliability concerns. The
various ways in which self-reported levels of happiness, contentment, or
preference-satisfaction can be misleading have been well documented.⁵¹ One
example is the problem of adaptive preferences, where individuals who have
been subject to mistreatment or deprivation adapt to their circumstances, and
no longer prefer what might seem like clearly superior alternatives, as when
people prefer to remain in an abusive relationship.⁵²

⁴⁸ Kahane and Savulescu 2009: 53.

⁴⁹ E.g. John Harris (2001: 283) insists that ‘to be disabled in any sense is not the same as being
differently abled’. Being deaf, for instance, positively ‘harms the individual relative to freedom from
deafness’ (Harris 2001: 283). (Also see Harris 2000.) Arguably, Terzi’s (2009, 2008) approach could also
be considered welfarist. Terzi adopts a capability approach to disability, and so understands impair-
ments as disabling insofar as they cause a loss of, or restriction in, capabilities. However, she
understands the capability approach as an account of subjective well-being, and so considers the loss
of capabilities to entail the loss of components of well-being. Thus, individuals are disabled by
capability loss because this renders them worse off. For example, she suggests that whether a wheelchair
user has a capability depends on her ‘most valuable ends’ and her ‘attitude’ (Terzi 2009: 99): if these are
such that she cannot fulfil her goals as the result of being in a wheelchair ‘her well-being appears to be
restricted . . . and hence the full set of capabilities available to this person is diminished’ (Terzi 2009:
100). If we interpret Terzi as implying individuals are disabled whenever their well-being is restricted
and capabilities diminished as a result of an impairment, then disability would be ubiquitous on her
account—just as on other welfarist views, as we will see.
⁵⁰ Some of the many contributions to this debate include: Scanlon 1975, 1998; Griffin 1986; Crisp
2006; Sumner 1996; Parfit 1984.
⁵¹ E.g. Kahneman et al. 1982; Kahneman 2011; Le Grand and New 2015; Thaler and Sunstein 2008.
⁵² E.g. Elster 1987; Khader 2011; Terlazzo 2016; Begon 2020, 2015; Chapter 6.
     29

Yet moving toward a more objective understanding of welfare brings its

own problems. These include the difficulty of finding a non-arbitrary way of
compiling a list of basic goods, or the worry that this list will entail the
paternalist imposition of a perfectionist conception of the good, or that it
will be elitist and exclusionary. Further, we may be sceptical about the impli-
cation that something can be considered good for someone independently of
their attitudes towards it, and indeed, despite their own insistence to the
contrary. This is particularly problematic in the context of social justice,
where there may be a divergence between state and individual assessments
of which conditions reduce welfare, and thus (on a welfarist account) of what
constitutes disability, and who is entitled to state assistance.⁵³
A further problem in the social justice context is the epistemic difficulty of
assessing welfare in a great many cases. For example, Kahane and Savulescu
discuss the ‘Ashley treatment’, named after a child with severe cognitive
impairments who was given treatment to restrict her further growth and
prevent puberty. They argue that, though this treatment moved Ashley further
from the species norm, it lessened her disability since it improved her welfare.
This stands in stark contrast to, for example, Eva Feder Kittay’s discussion of
the case, in part because Kittay employs a different understanding of Ashley’s
welfare. She contends that ‘[w]e take pleasure and pride in our bodies as they
grow and mature because . . . we do. Full stop. It needs no further justification.
It is constitutive of a thriving life’.⁵⁴ Yet evidently this would prove insufficient
to Kahane and Savulescu, who do not recognise this value. These epistemic
difficulties are especially intractable in cases such as Ashley’s, where it can be
hard to ascertain the view of the individual concerned, and far from clear
that those who have not experienced this condition can be in a position to
understand what would be required for her to thrive.
Thus, Kahane and Savulescu’s definition does not determine who is
disabled—who experiences a welfare deficit—but simply moves the debate to
a new context: finding a satisfactory account of welfare. This is an important
gap in their approach. However, the fact that they have not filled it does not
mean it cannot be satisfactorily filled. Further, a welfarist might respond that
disability should not be easy to define, and that my own approach will face
similar worries—there is, of course, no consensus on what justice demands.
However, these epistemic worries seem deeper for welfarist approaches—at
least for those that include an element of subjectivity in their account of

⁵³ Dworkin 2000; Williams 2002; Colburn 2014; Lazenby 2016; Begon 2021b; Chapter 7.
⁵⁴ Kittay 2011: 621.
30      

welfare—since they necessarily require a direct reliance on first person

testimony. Moreover, determining how the benefits and burdens of social
cooperation are to be distributed requires that we reach agreement on certain
essentials of a theory of justice, but not of welfare.⁵⁵ Lacking a working
conception of the demands of distributive justice would cause problems for
considerably more than my account of disability.
Finally, whilst we may not have reached agreement on a complete theory of
justice, it might be easier to agree on basic entitlements, and so basic injustices,
than it is to identify core instances of welfare deficit. In Rawlsian terminology,
the burdens of judgement mean that reasonable people will hold competing
conceptions of the good yet can still be part of the overlapping consensus on
the demands of justice.⁵⁶ For example, it is, perhaps, easier to agree that
individuals are entitled to equal political participation, opportunities for
mobility, and the ability to form relationships than it is to show that all
these capabilities will make all individuals better-off, given the existence of
the idiosyncratic preferences of the politically apathetic, sedentary, and mis-
anthropic. Thus, disagreements about what makes life go well may be harder
to settle even in apparently core cases, as examples like Ashley illustrate. The
lack of a widely accepted account of welfare thus provides reason for scepti-
cism about the possibility of operationalising the welfarist account. However,
this amounts to raising a worry that demands a response, rather than provid-
ing a decisive refutation of the welfarist view.
The deeper problem for this view is that, whatever account of welfare is
used, disability will be ubiquitous, and this will prove problematic from the
perspective of justice and public policy. As Kahane and Savulescu openly
acknowledge, disability will be a degree rather than a threshold concept.
Thus, ‘everyone has disabilities of one kind or another’.⁵⁷ I agree that there
can be a degree of arbitrariness in drawing the boundary between mild and
severe disabilities, and indeed between being disabled and able-bodied. I agree,
too, that we might have reasons to make this category more expansive than it
is at present (as §2.7 discusses). Nonetheless, we should be cautious of
expanding it so far that it becomes meaningless, as it will be if it is a category
in which everyone is included.
A related problem is not just that it includes too many, but that it includes
them for the wrong reasons. In some cases, individuals will count as disabled

⁵⁵ E.g. Rawls 1999: 5.

⁵⁶ Rawls’s argument is merely illustrative of this claim, so I will not explicate the details of this
position here.
⁵⁷ Kahane and Savulescu 2009: 30.
     31

on the basis of conditions that do not much seem like disabilities—for

example, because they have a ‘grumpy personality’.⁵⁸ More fundamentally,
even when it seems like the right cases are identified, their inclusion surely
should not depend on individuals’ welfare level—an individual with severe
paralysis is not disabled because (and only if) they are unhappy about this.⁵⁹
I agree that it is absolutely fundamental that we focus on the consequences of
impairment (for particular people, in particular contexts), but the relevant
consequences simply do not seem to be how happy or satisfied they feel. Nor
should we adopt an approach where coming to feel proud and happy about
one’s anomalous functioning means one ceases to be disabled. As the next
chapter argues, I believe we should focus, instead, on determining which
impairments prevent individuals from getting opportunities they are entitled
to. In other words, which individuals are actually disadvantaged in a way that
is the proper object of social concern.
Yet even leaving this point aside, the extensional inadequacy of welfarist
accounts remain. If every stable physical or psychological property that lowers
our welfare counts as disabling, then this definition does not seem to pick out
the right cases—or, not only the right cases. At least, not in the context of social
justice, where our concern is identifying disadvantage and determining how
the state should appropriately respond. It is certainly worth emphasising that
disability is a continuum, and that we are all restricted by the limitations of our
bodies and minds. Nonetheless, when fighting against ableism and injustice,
simply concluding that ‘we are all disabled’ is surely not a helpful approach.
Similarly, we may wish to reject binaries of sexual orientation and gender
identity, and deny that racial categorisation is possible, yet nonetheless find it
useful to refer to ‘women’, and ‘gay people’, and ‘black people’ as a means of
identifying, and fighting against, injustices against these groups. ‘Disability’,
I contend, has value and meaning as a way of identifying, and responding to,
disadvantage. The term has little function if entirely ubiquitous.⁶⁰

⁵⁸ Gregory 2020: 35. ⁵⁹ Also see Gregory 2020.

⁶⁰ Further, though we do not want to replicate a common-sense understanding of disability, it
should not be entirely divorced from our intuitions. Indeed, Kahane and Savulescu (2009: 19)
themselves object to the species norm approach on this basis. It might also be objected that the
welfarist account deviates from common sense in the other direction: by failing to count harmless
impairments as disabilities (Gregory 2020: 34–5). There are two ways to read this criticism. One is that
disability should not depend on welfare loss, and individuals with high welfare may still be disabled. As
noted in the text, I entirely agree. Alternatively, the concern may be that it is that it is ‘surprising at best’
that paradigm disabilities won’t be included, and that this is reason enough to reject an account with
this consequence (Gregory 2020: 35). A version of this criticism also applies to my own account: I will
implausibly fail to count impairments that do not diminish our entitlements as disabilities. However,
32      

1.7 Barnes’s Social Constructionist Approach

Barnes, like many others, rejects the social model’s exclusive focus on external
causes of disability, and emphasises that our understanding of disability
should be grounded in an assessment of individuals’ physical features—or,
‘what their body is (really) like’.⁶¹ However, she is keen to emphasise, too, that
particular qualities are considered significant not because of their objective
features, but because of the way we think about them. Barnes further argues
that those best placed to determine which particular features should be
considered disabling are disability activists. Thus, that ‘disability just is what-
ever the disability rights movement is promoting justice for’.⁶² Inclusion in the
category of disability therefore depends on ‘rule-based solidarity among
people with certain kinds of bodies’.⁶³ In other words, we are disabled on
Barnes’s view if our body meets the criteria used by the disability rights
movement to identify disability.
However, there may be problems with relying on rules devised by activists.
First, we should not unquestioningly accept the content of these rules. Whilst
the testimony of disabled individuals should undoubtedly play an important
role in developing an account of disability, it does not follow that the rules
adopted by the disability rights movement will be free of bias. Individuals who
already ‘count’ as disabled, and have paradigm conditions, may have vested
interests in excluding others from entering this group, and perhaps diluting
the strength of their claims with the accretion of many others. This need not be
a conscious process of exclusion. Individuals may simply tend to see their own
impairments as the paradigm, and be less inclined to include those they are
less familiar with. Barnes’s own focus on individuals with ‘certain kinds of
bodies’, rather than individuals with cognitive impairments or mental health
problems, might be an instance of this.
Disability, on any account, is a hugely heterogeneous category. Thus,
we should not assume that disabled individuals’ epistemically privileged
position regarding their own experience extends to understanding the diverse
experiences of all other disabled people.⁶⁴ Nor can we assume that a group
fighting against injustice will never unjustly exclude others from membership.

I believe this objection is misguided, as §2.7 argues at length. When a paradigm has arisen from
prejudice and ill-informed assumptions about disabled life, it is not in the least surprising that it will
require revision.
⁶¹ Barnes 2016a: 38. Barnes is focused only on physical, and not cognitive, disabilities.
⁶² Barnes 2016a: 43. ⁶³ Barnes 2016a: 46.
⁶⁴ For a discussion of the impact of restricted information on the reliability of individuals’ choices
and preferences see Begon (2020) and Chapter 6.
     33

Consider, for example, the exclusion of asexual individuals by some in the

LGBTQ+ community, or the exclusion of trans women by some feminists. The
point is not that such exclusion is necessarily unjust, but that the question of
injustice is not resolved by observing that it was existing members of the
community who determined these boundaries.⁶⁵
A related worry with Barnes’s approach is that, given the heterogeneity
that exists between and within such groups, disagreement about the content
and application of the rules of inclusion seems likely, and difficult to
resolve. Barnes acknowledges the diversity of disability rights movements,
and allows that it may be ‘simply vague or indeterminate which social group
is referred to . . . [as] “the disability rights movement”’.⁶⁶ Her response is to
insist that ‘this entire domain . . . is riddled with indeterminacy’,⁶⁷ and that it
would be ‘deeply implausible’ if it turned out that ‘any aspect of our complex,
multifaceted social reality had fully determinate boundaries’.⁶⁸
It is certainly true that any definition will likely include difficult or border-
line cases. Indeed, an account that provides a firm and unequivocal distinction
between those who are, and are not, disabled, seems to have missed something
important about this category. Yet even if the boundaries are somewhat fuzzy,
we still need some sense of where they lie. I am sceptical that the implicit rules
used by disability rights movements will be sufficient for this task,⁶⁹ and,
indeed, whether we should simply accept these rules even if they could be
clearly identified.
Further, I believe that there are deeper flaws in Barnes’s approach. Barnes is
not simply neutral about the content of an account of disability, but rather
outlines criteria that any successful account should meet.⁷⁰ I will object to two
of them. These are, first, that an account of disability should not prejudge
normative issues and, second, that it should be able to deliver the correct
verdict about paradigm cases. These criteria will not be met by my account,
nor by the welfarist account. Yet I contend that an account of disability need
not—and, indeed, should not attempt to—fulfil these criteria.⁷¹
Considering the latter point first, it is clear that the welfarist account will not
deliver the correct verdict about paradigm cases. Its failure to do so does not
come from the ubiquity of disability on this account: such ubiquity will

⁶⁵ Similar worries are raised by Campbell and Stramondo (2016), Wasserman (2018), Howard and
Aas (2018: 1127), and Gregory (2020: 38–9).
⁶⁶ Barnes 2016a: 49. ⁶⁷ Barnes 2016a: 49. ⁶⁸ Barnes 2016a: 50.
⁶⁹ Also see: Wasserman (2018: 254); Howard and Aas (2018: 1115); Francis (2018).
⁷⁰ Barnes 2016a: 10–13.
⁷¹ Thus, though for the reasons already given I think we should reject the welfarist approach, this is
not for the reasons Barnes would give.
34      

‘correctly’ identify paradigm disabilities, though it will also deliver the same
verdict for many more cases besides. The greater problem is that Kahane and
Savulescu allow that for particular people, in particular circumstances, condi-
tions such as deafness, dwarfism, and (voluntary) amputation may positively
improve well-being.⁷² These individuals will not be considered disabled and,
indeed, regaining one’s hearing, or becoming average height, or being forced to
keep one’s limbs may be disabling if, on balance, this lowers someone’s welfare.
Clearly, Barnes would not take this to be the ‘correct’ verdict about these
paradigm cases. My approach, too, will have revisionary implications (§2.7).
However, it is not clear why it would be a virtue of an account to be ‘correct’
about paradigm cases. Whilst we may worry about an approach that is
completely divorced from our understanding of disability—for example, an
account that sees homosexuality as a disability—we should also be aware that
our sense of what is ‘paradigm’ when it comes to disability is often informed by
intuitions that are based on prejudice and misunderstanding of what disabled
lives are like.⁷³ As noted, these worries may not be solved even if we allow
those who are already considered to be part of this ‘paradigm’ to be the arbiters
of who else will be included.
Next, Barnes argues that ‘[i]t should not be built into the very definition of
disability that disability is something that’s bad or suboptimal’.⁷⁴ Yet the
welfarist account is explicitly evaluative: ‘on our definition it’s tautological
that disability is bad for those suffering from it’.⁷⁵ Further, since to be disabled
on my view means to be deprived of opportunities all humans are entitled to,
then this is also an evaluative concept. However, it is only important that
disability be value-neutral if we follow Barnes in rejecting the distinction
between impairment and disability.⁷⁶ If we accept this distinction then we
can allow that disability is necessarily disadvantageous and still maintain a
normatively neutral account of impairment, capturing the heterogeneity of
human bodies and capacities, which may be disadvantageous in some con-
texts, but need not be so. Thus, a normatively loaded account of disability does
not undermine the possibility of being proud of our diverse bodies and minds.
However, by conflating these two terms Barnes makes it near impossible to
achieve the twin goals of recognising that diversity need not be negative, and
that it can be correlated with disadvantage, which must be identified to be
rectified.

⁷² Kahane and Savulescu 2009: 49–50.

⁷³ As §2.7 considers, and Barnes herself has pointed out (Barnes 2016a: 119–42; Barnes 2009a).
⁷⁴ Barnes 2016a: 11. ⁷⁵ Kahane and Savulescu 2009: 42.
⁷⁶ See Chapter 5 and §1.2 for a defence of this distinction.
     35

Barnes meets her own criteria, and so adopts a value-neutral model of

disability, according to which disability is a ‘mere difference’. Yet this does
not mean it has no effect on individuals’ lives. On the contrary, it ‘may be good
for you, it may be bad for you, it may be utterly indifferent for you’.⁷⁷ Indeed,
Barnes allows that being disabled may sometimes lower overall well-being,
that some aspects of some disabilities are bad differences,⁷⁸ and even that some
specific disabilities are bad simpliciter (something that makes ‘your life go
worse in virtue of it specifically’⁷⁹). By including diverse conditions within this
wider category, it becomes true that disability is not necessarily a bad differ-
ence, but this category still encompasses conditions (whether few or many)
that are bad differences. As such, Barnes’s conception of neutrality is very
weak.⁸⁰
Clearly, then, Barnes does not believe that ‘the positive effects [of disability]
have to balance the negative ones’, and accepts that ‘in the world we inhabit,
disability tends to significantly reduce one’s good options’—contra the claims
of some of her critics.⁸¹ Nonetheless, espousing a ‘value-neutral model’ or
‘mere-difference view’ of disability can be importantly misleading, especially if
this is used to ground general claims about the kinds of policies that should be
pursued with regards to disability. Certainly we should acknowledge that it is
‘simply untrue that, as a general rule, disabilities are bad for those who have
them’.⁸² Thus, claims grounded on the general assumption that someone’s life
will be worse simply because they have ‘a disability’ are unjustified.⁸³ However,
we should also avoid grounding claims about policy on the assumption that
all disabilities should be treated as neutral simply because this category
incorporates conditions with good, bad, and neutral effects. For example, we
may accept that disability is a mere difference in Barnes’s sense, but still
believe, contra Barnes, that it is sometimes wrong simpliciter to cause some
disabilities.⁸⁴ Amalgamating all disability into a general ‘value-neutral model’,
then, can obscure the disadvantage that can be associated with impairment.⁸⁵

⁷⁷ Barnes 2016a: 98. ⁷⁸ Barnes 2016a: 75. ⁷⁹ Barnes 2016a: 87, 102.
⁸⁰ See Wasserman 2018: 254–6; Hawkins 2018: 463–6; Dougherty 2014.
⁸¹ Kahane and Savulescu 2016: 777; 776. For Barnes’s response, see Barnes (2016b).
⁸² Campbell and Stramondo 2017: 176.
⁸³ More precisely, since unlike Barnes I distinguish impairment and disability: because they have an
impairment widely assumed to be disabling.
⁸⁴ Barnes 2014; 2016a: 167. For a response, see Begon 2018; Wasserman 2018: 255. Also §5.6
and §7.9.
⁸⁵ As such, Campbell and Stramondo’s (2017: 163–5) suggestion that disabilities are intrinsically
neutral, but not instrumentally, comparatively, or overridingly neutral, seems a more plausible and
nuanced view. Nonetheless, there may be reasons to worry that their account could also be employed in
a way that obscures disadvantages associated with impairments insofar as they tend to consider
36      

1.8 An Ameliorative Approach

How problematic this is may depend on the context in which our account of
disability is employed. My focus here is relatively narrow: aiming to develop an
account that can guide theory and policy that aims to identify and mitigate the
disadvantage associated with physical and cognitive difference. Such policies
should be nuanced and individualised, as I will argue. They must acknowledge
that some impairments, for some people, in some contexts, are not neutral—or
merely ‘a minority body’⁸⁶—and should not be treated as such. Thus, in the
context of justice, at least, it seems unwise to remain committed to a unified
category of disability so broadly construed—or, indeed, to an account on
which disability is ubiquitous.
My approach to defining disability and impairment, then, is ameliorative in
the sense that it seeks to answer the question of how disability should be
understood, as opposed to simply describing how the word is commonly used.
Further, like Haslanger my goal is to develop concepts that are ‘effective tools
in the fight against injustice’.⁸⁷ The appropriate tools will depend on our
purposes, so this leaves room for the possibility that there may be different
contexts in which a different account of disability will be more appropriate,
and I leave open the chance that social constructionist or welfarist accounts
could have a place here. Further, insofar as advocates of these approaches have
other goals they may not find some of my lines of criticism convincing
(though, of course, the difficulties of specifying the content of these
accounts—determining the rules of inclusion or a theory of welfare—remain).
Insofar as my approach has an explicitly political or normative goal, it may
appear that it could be vulnerable to criticisms levelled against the political
project underpinning the social model. It is first worth clarifying which
criticisms of the social model will not apply to my account—indeed,
I outline and endorse these very criticisms in §1.3, §1.4, and §5.2. These
start from the contention that social model theorists over-emphasised the
external causes of disability as a political strategy to counteract the medical
model’s single-minded focus on physical causes. Emphasising the importance
of, and need for, social accommodation undoubtedly proved liberatory
for many disabled people—‘They didn’t need to change: society needed to
change. They didn’t have to be sorry for themselves: they could be angry.’⁸⁸

disabilities ‘in isolation from their effects’ that are most obviously detrimental to well-being. It is not
clear that uncontroversially harmful elements of impairments such as pain, fatigue, or disturbed or
depressed emotional states can be meaningfully isolated from these conditions, as §5.6 will discuss.
⁸⁶ Barnes 2016a: 1. ⁸⁷ Haslanger 2000: 36. ⁸⁸ Shakespeare and Watson 2001: 11.
     37

However, first, the social model’s normative assumptions and political goals
were implicit, and hence unexamined and undefended.⁸⁹ Second, insofar as
the focus on social causes led to the occlusion of impairment effects it could be
counterproductive, undermining the very goal of improving disabled lives.
Failing to capture ‘the lived, embodied and visceral experiences of having an
impairment’⁹⁰ alienated many from the disability rights movement, and might
even imply cures or resource enhancements are never appropriate. Since my
own normative commitments are explicit and will have very different political
consequences, these criticisms of the social model will not have bite against my
account.
However, Christopher Riddle raises a more general objection to normative
approaches to defining disability that may seem to be applicable to my
account. Riddle objects that the social model’s political objectives lead to its
failure to provide ‘an accurate characterization of the reality of disability’.⁹¹ He
insists that ‘an accurate model of disability’ needs to ‘aptly characterize, what it
is like to live with a disability’.⁹² However, this does not seem the right way to
describe what has gone wrong in the social model’s failure to acknowledge the
embodied, messy reality of living with impairment. Beginning with the
assumption that there is a ‘reality’ of ‘what it is like to live with disability’
simply begs the question we are trying to answer. Disability is not a natural
kind but a socially constructed concept, so there is no context-independent
truth about its scope that our definition must reflect. As Shakespeare and
Watson say, ‘disability studies and disability politics share the assumption that
we know who the disabled subject is. Yet this cannot be taken for granted.’⁹³
Indeed, even if the scope of disability was determinate it would remain open
to debate which ‘experiences of disability’ are constitutive of the category.
Conceding that the ‘brute facts’ of impairment should be part of our account
of disability, for example, does not determine which facts about which experi-
ences should be counted amongst the ‘reality’ of disability. There are, of
course, biological facts about the various impairments archetypically con-
sidered disabling: arthritis is painful, individuals with Down’s syndrome
have an extra copy of chromosome 21, and so on. There is nothing about an
ameliorative approach that need deny these realities. The question that an
account of disability must answer is which of the restrictions, resulting from

⁸⁹ See Riddle 2013, 2020; Shakespeare and Watson 2001; Vehmas and Riddle 2020; Vehmas and
Watson 2014; Vehmas 2004; Watson 2002.
⁹⁰ Vehmas and Watson 2014: 641. ⁹¹ Riddle 2013: 384. ⁹² Riddle 2020: 1511.
⁹³ Shakespeare and Watson 2001: 23. Riddle’s concern here might echo Barnes’s contention that our
account should include paradigm cases.
38      

atypical functionings, are the disabling ones, and what unites these experiences
into a meaningful category. Chapter 5 will refute the claim that the unifying
feature of disability is defect, arguing that many disabling impairments com-
prise different, not deficient, modes of functioning. Yet even if we were to
grant that all disabled bodies and minds are united in falling below the
(notoriously impossible to identify) threshold of normal species functioning,
they are not unique in doing so (§1.5). Which of the many possible imperfec-
tions of human bodies and minds are considered significant and relevant
enough to count as disabling is not reflective of some deep and objective
truth, but determined in particular contexts depending on which functionings
are valued and what accommodations are made. To reiterate: there is no pre-
theoretical truth to appeal to, or a natural kind to identify. Rather, our answer
will depend on what we want from our account of disability.
In a broad sense, then, I believe the best definitions of disability—and other
socially constructed categories—must be ameliorative. However, it is also
worth noting that, whilst I share Haslanger’s broad goals, I do not advocate
a Haslangerian model of disability: that is, one derived from her accounts of
gender and race.⁹⁴ On one plausible iteration of a Haslangerian account, to be
disabled is to be in a bodily or psychological state assumed to be an impair-
ment in the prevailing ideology, where this marks one out for pity, stigma, and
exclusion, and where the fact one is in this state plays a role in one’s systematic
disadvantage.⁹⁵ Unlike the account I defend here, this approach is focused not
on the specific ‘valuable activities’ individuals are excluded from, but with the
cause of this exclusion: that it results from possession of a feature that, within
the dominant ideology, marks one for this exclusion. Thus, as will become
clear, this view will be both more and less inclusive than my own. On the one
hand, individuals may be disabled though they face very little actual
disadvantage—for example, Oscar Pistorius.⁹⁶ On the other hand, individuals
may not be deemed disabled on this view though they face a restriction in
opportunities they are entitled to due to atypical bodily or cognitive function-
ing, if this atypicality is not assumed to be an impairment in the prevailing
ideology—for example, the case of obesity discussed in §2.7, which though
clearly stigmatised is not generally pathologised as an impairment in Howard
and Aas’s sense.
I will defend how I draw the boundaries of inclusion in the next chapter. For
now, my point is simply that it makes little sense to talk of disability as a

⁹⁴ See Haslanger 2000, 2012; Barnes 2016a: 28–38; Howard and Aas 2018.
⁹⁵ Howard and Aas 2018: 1113, 1128–9. ⁹⁶ Howard and Aas 2018: 1129.
     39

category divorced from some social purpose. This does not mean our defin-
ition should be entirely untethered from standard usage—an account that
deems homosexuality a disability, or divorces disability from the experience,
and perhaps suffering, entailed by functioning atypically, for example, would
not be satisfactory. But this is because we could reasonably argue that this is
not how disability should be understood. When defining disability we should
engage in a process of reflective equilibrium, and if we are to conclude
that paradigm cases should be jettisoned when incompatible with our pre-
ferred definition then good reasons will be needed. However, such reasons
are available, and we should not leave uncontested the current scope of
disability—socially constructed, unexamined, indeterminate, and inconsistent
as our rules of inclusion are. The problem with the social model is not that it
seeks to answer the question of how disability should be understood, but that
the answer it gives is ill-supported and unconvincing. It lacks justification to
ignore the significance of impairment, and as a result fails by its own stand-
ards. We should not conclude from this that our approach should not be
normative. Indeed, it cannot be anything else.

1.9 Conclusion

This chapter began with areas of relative consensus in the debate on disability.
These were, first, the distinction between impairment as deviation from
species-normal physical or cognitive functioning, and disability as the restric-
tion in the ability to perform tasks; and second, that disability is caused in part
by individuals’ impairment, and in part, by other internal features of the
individual and external features of their circumstances, including policies
and institutions, norms and attitudes, the natural and material environment,
technology and resources, and relationships and community. I next outlined
and rejected three alternative accounts of which restrictions in the ability to
perform tasks should be considered disabilities. The species norm approach
will include trivial inabilities, or cease to be concerned merely with species
norms. The welfarist approach lacks a satisfactory account of welfare and will
render disability ubiquitous. The social constructionist approach places too much
weight on indeterminate and potentially unreliable rules of inclusion. It also
broadens the category of disability to include conditions with good, bad, and
neutral affects. Few conclusions about the just treatment of disabled individuals
can be drawn until we can adequately define and delineate these subcategories,
and identify which aspects of which disabilities are bad differences.
40      

All these approaches shift the focus of the debate rather than settling it, and
none settles the question of when disability is a disadvantage of the kind
the state should mitigate. Thus, none can form the basis of public policy or
political activism. I have emphasised the importance of having a value-neutral
account of physical and cognitive difference (which I call impairment).
However, it is important, too, to have a concept that can play a role in political
debates about when difference is disadvantageous, what constitutes injustice
towards disabled individuals, and the policies that should be pursued to
respond to this. Existing approaches do not provide us with the tools to do
so. My own account of disability, I argue, does.
 2
Disability: A Justice-Based Account

2.1 Which (In)abilities Matter?

Not all of the restrictions in individuals’ capacity for activity and participation,
resulting from impairment, internal conditions, and external circumstances
are disabilities: we are not disabled if we cannot sing, cannot stand on one leg,
or cannot raise an eyebrow. As I have argued, this is so even if these abilities
are species normal, even if restrictions in them decrease our welfare, and even
if they are the subject of the rules of inclusion employed by disability rights
movements. Instead, I argue that the relevant restrictions are in abilities that
individuals are entitled to be able to perform. More precisely, I suggest that we
should understand disability as the restriction in the ability to perform those
tasks human beings are entitled to be able to perform as a matter of justice,
resulting from the interaction between an individual’s impairment, their social,
political, and material context, the resources they have access to, and their other
internal characteristics.¹
To give an intuitive sense of what I have in mind, it will help to reflect on a
specific application of this approach. Consider, then, the case of deafness. The
impairment here is the inability to hear, whilst the disability is the reduction in
relevant opportunities that may arise when most people communicate via
speech, such as difficulties in political participation, access to appropriate
education, and forming meaningful relationships. Deaf individuals are not
disabled on the basis that they are unable to hear music, or the voices of their
loved ones. However, this need not be because opportunities for aesthetic
experience or forming relationships are not considered central entitlements.
On the contrary, they may well be central—we simply do not need to hear
music or voices to have them. Being entitled to opportunities in a domain need
not require having every opportunity (as Chapter 4 argues). As such, if deaf
individuals have access to other forms of aesthetic experience, their inability to
hear music is not a concern of justice. Indeed, if deaf individuals have all the
opportunities they are entitled to, they are not disabled.

¹ I have also outlined and defended this account in Begon (2021a).

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0003
42      

Individuals are disadvantaged when their access to their entitlements is

restricted, and disabled when this results from an impairment (in combination
with context, resources, and personal characteristics). My claim is that there is
no simple or necessary connection between impairment and disadvantage,
and hence disability. Thus, individuals with the same impairments may not be
identically disabled, and individuals’ status as disabled may change even if
their impairment does not alter. Rightly, then, we cannot know whether an
individual is disabled merely from information about their impairment. On
this view, to determine whether an impairment is disabling requires an
account of our just distributive entitlements, and the contextual information
necessary to determine whether an individual has access to them. However,
my goal in this chapter is not to identify the contents of these entitlements, but
to develop a framework into which various conceptions of distributive justice
can be incorporated.² Theorists who disagree over whether individuals should
be entitled to resources or capabilities, over which specific resources or
capabilities they are entitled to, and over how this content should be deter-
mined, can nonetheless agree that individuals are disabled when their access to
their entitlements is thus restricted. This is discussed in §2.2.
Then, in §2.3, I consider possible objections to the definition I have
sketched, according to which my approach will lead to an implausible account
of what constitutes disability, or of what individuals are entitled to, or both.
First, it may seem that if being disabled means lacking what we are entitled to,
then in cases where this cannot be rectified, this cannot be unjust. This implies
individuals with serious impairments would not be considered disabled on my
account. Second, if we adopt a universal conception of entitlements, such that
all individuals are entitled to the same opportunities (or resources), then it
may be objected that this does not properly reflect disabled individuals’ unique
experiences and ways of functioning. Finally, in part to avoid the former
objections, I argue we have reason to adopt a broadly specified and multiply
realisable conception of entitlements: for example, as opportunities for mobil-
ity, communication, or forming relationships. This leaves me vulnerable to the
further worry that individuals will not be considered disabled, or entitled to
assistance, even when unable to perform important functionings (walking,
hearing, easily reading social cues) if they have more general capabilities
in these domains. I respond to these objections in §2.4, §2.5, and §2.6,
respectively.

² I defend my preferred conception of distributive justice in the following two chapters.

 :  -  43

I finish, in §2.7, by considering the revisionary and counterintuitive

implications of my account. It will, for example, imply that individuals with
significant and visible impairments may not be disabled by them, that indi-
viduals with apparently minor or invisible impairments may be disabled, and
that individuals’ status as disabled may change across time and context.
However, I argue that far from giving us a reason to reject my approach,
this is one its key benefits. The intuitions with which this approach may clash
ought to be challenged: our understanding of disability should no longer focus
on bodily difference, or deviation from what we take to be the norm, and
should instead reflect a concern with the opportunities that all individuals
ought to have access to, and which are closed to so many.

2.2 What Sort of Justice?

I have called my account of disability ‘justice-based’, so it is worth saying a

word about the kinds of justice I have in mind. On the one hand, I intend the
account to be a framework into which various competing theories of distribu-
tive justice can be incorporated. Certainly, accepting the general claim that
individuals are disabled when their impairments restrict their access to forms
of activity and participation they are entitled to does not require accepting a
particular conception of entitlements. Yet, on the other hand, it is difficult
to defend this account in the abstract, and cumbersome to run through
every possible way to cash out individual entitlements, including those
I believe we have independent reasons to reject. Thus, I will tend to talk in
specific terms. As I will outline here, I believe that a consideration of
disability will give us reasons to adopt the kind of approach I employ, but
I largely defer my efforts to persuade readers that this is so until the next two
chapters. For now, those who disagree can substitute their preferred under-
standing of entitlements.
However, this does come with a caveat: there is one approach to distributive
justice that cannot be incorporated. My account distinguishes what individuals
are entitled to from what merely increases their welfare. As such, it cannot
accommodate those welfarist approaches that consider anything that lowers
individuals’ welfare—whether state oppression or stubbing our toe—to con-
stitute a relevant disadvantage. Subjective welfarist accounts, which do not
distinguish the different reasons for a decrease in hedonic state or the different
sorts of preferences individuals may care to have satisfied, would collapse into
the welfarist accounts discussed in §1.6 and so face the same worries about
44      

over-inclusiveness and the ubiquity of disability raised there.³ This does not
mean our theory of justice can have no place for welfare—for example,
deeming individuals to be disadvantaged and perhaps disabled as a result of
welfare deficits resulting from pain or fatigue (as §1.7 considers). Thus, I need
not rule out pluralist welfarist approaches, such as G. A. Cohen’s, which allow
that opportunities for welfare may be amongst the advantages individuals
should have access to.⁴ I believe there are, indeed, reasons to reject such
moderate versions of welfarism (see §3.4 and §4.6), but for now am only
setting aside the view—rather implausibly extreme—that subjective welfare is
the sole determinant of individuals’ entitlements.
My approach relies on the general intuition that ‘[s]ome deprivations of
capabilities [or resources] express greater disrespect than others, in ways any
reasonable person can recognize’.⁵ This distinction is employed widely in the
distributive justice literature, by theorists who disagree on much else, and has
strong intuitive pull.⁶ It is the suggestion that individuals should be entitled to
be able to vote in elections, to be mobile even if they are paraplegic, and to have
access to educational opportunities, but that they are not entitled to have
access to every ride in an amusement park, or a sports car, or an expensive
musical instrument—and that this is so even if they would prefer these latter
opportunities, and they would have a greater positive impact on their welfare.
In other words, this is the claim that only some domains of life are the concern
of justice, and so the state, and some are not, and that the scope of these
domains is not determined by individuals’ preferences.
Beyond this restriction, then, adopting a specific theory of distributive
justice is not a prerequisite to accepting my account of disability. However,
as I have said, I believe that reflection on disability gives us reasons to opt for
accounts with certain features. First, I will assume that individuals are entitled
to capabilities, rather than merely access to resources. The latter approach fails
to account for the different opportunity sets a bundle of resources can enable
given individuals’ heterogeneous capacities: a specific income level, for example,
will allow an able-bodied individual to do more than someone with complex,
expensive, and poorly accommodated health problems (see Chapter 3). Second,
our conception of entitlements should be relatively thick: not merely entailing

³ It is worth noting that it is both implausible that individuals are disabled by anything that lowers
their subjective welfare, and implausible that such individuals lack what they are entitled to as a matter
of justice, though I will not defend this stronger claim here (see Begon 2021b).
⁴ Cohen 2011. ⁵ Anderson 1999: 332.
⁶ E.g. Anderson 1999; Dworkin 2000: 61; Carter 2014: 84–5. I will not defend this distinction against
its welfarist opponents here (see Begon 2016a).
 :  -  45

the fulfilment of basic needs, but including access to opportunities such as

forming social relationships, engaging in leisure activities, and enjoying aesthetic
experiences.
Finally, I assume that individuals are entitled to broadly specified oppor-
tunities, and not particular or identical functionings: for example, opportun-
ities for aesthetic experiences, mobility, or communication, and not hearing
music, walking, or talking. In other words, individuals should be entitled to
capabilities understood as exercising control over certain domains of their life,
where this requires acceptable alternatives but not the capacity to perform any
valuable functioning.⁷ It follows from this commitment that individuals will
not necessarily be entitled to the option they would most prefer. Even if
we determine that individuals are entitled to leisure and aesthetic experiences,
for example, this does not entail that the state must enable us to pursue
our preferred hobby of being able to listen to live music (and not just if we
cannot hear).
Obviously, these criteria have not been selected at random. I believe there
are good reasons to opt for an account of distributive justice with these
features, and that reflection on disability can help demonstrate why this is
so, as much of the discussion in this and subsequent chapters will illustrate.
Approaches that fail to meet the criteria of thickness and generality will,
I argue, generate both an implausibly thin account of disability and, indeed,
an implausibly thin set of entitlements (as the concluding chapter considers).
However, I certainly cannot definitively refute all alternative understandings
of entitlements here, so will leave it to their adherents to demonstrate how they
can navigate the issues disability raises.

2.3 Feasibility, Levelling-Down, and Thinning-Out

I will now outline some possible objections to the account of disability I have
sketched, and, in responding to them in subsequent sections, will elucidate my
account. First, it may be objected that if I am suggesting that being disabled

⁷ Approaches that accept some version of the second criterion but not the third—i.e. an expansive,
but narrowly specified set of entitlements—are common (arguably including: Dworkin 2000;
Nussbaum 2000a; Cohen 2011; Rawls 1999; Anderson 1999). For example, a view that suggests
individuals should be entitled to the capabilities or resources necessary for a decent human life, but
provides a specific account of what these entails: those that enable individuals to walk, to exercise all
five senses, or achieve species-normal cognitive abilities. I have defended the third criterion elsewhere
(Begon 2017), and will do so again in Chapter 4. Also see Asch and Wasserman (2005) for an argument
that individuals need not have access to every opportunity.
46      

means lacking opportunities we are entitled to, then impairments that result in
the loss of significant opportunities will not be considered disabling if we
cannot prevent this loss, since we cannot be entitled to an opportunity it is
impossible to provide. Thus, if, due to lack of medical or technological
knowledge, or merely lack of resources, we cannot prevent individuals being
rendered immobile, unable to communicate, or to form relationships, say, as
the result of an impairment, then they cannot be entitled to these opportun-
ities, and cannot be considered disabled by their absence. If this were
so, serious impairments—perhaps the most serious impairments—would
not be considered disabling on my account. Clearly, this would be very
counterintuitive.
Further, this understanding of our entitlements presents a dilemma. On the
one hand, this may lead us to conclude that individuals with serious impair-
ments should simply have a different set of entitlements. That is, if some
individuals can only be provided with a restricted set of opportunities, then
their entitlements are correspondingly diminished. On the other hand, we may
wish to resist this tiered, non-universal approach to entitlements, and insist
instead that all humans are entitled to the same opportunities (though not, of
course, to the same outcomes). Yet this then seems to imply that the content of
everyone’s entitlements should reflect the unavoidable constraints on some
individuals’ functioning. In other words, that we can be entitled only to what
can be provided to all. If we grasp this latter horn, then the content of our
entitlements might start to look rather thin. This is implausible in itself and
will generate an implausible account of disability if my approach is employed.
For example, it implies that if we cannot currently enable everyone to control
their mobility then no one can be entitled to this opportunity, and individuals
whose control over their mobility is impaired are not disabled by their
inability.
However, this is an artificial dichotomy. We should reject an implausibly
strict egalitarian view, according to which everyone must be enabled to
perform an identical set of functionings. Instead, individuals should be entitled
to generalised opportunities—for example, mobility, not walking—i.e. the
third criterion outlined above. It then may seem that everyone can have the
same entitlements. Indeed, if we recognise that capabilities are multiply
realisable, and treat different functionings as interchangeable and equally
valuable, then many individuals with impairments can have the opportunities
they are entitled to even if there are some functionings they cannot perform.
For example, blind and deaf individuals can have opportunities for sensory
experience even if they cannot use all five senses, can have opportunities for
 :  -  47

aesthetic experiences even if the enjoyment of visual arts and music is limited
for each respectively, and can have opportunities to communicate and form
social relationships even if they cannot read visible social cues, in the former
case, or communicate verbally, in the latter. Thus, even if we were to limit our
entitlements to those that could be provided universally, this might lead to less
being taken off the table than it initially appeared. As such, it seems we can
largely avoid a tiered approach without levelling-down our entitlements.
Nonetheless, this response will not prove fully satisfactory. First, even if we
focus on a generalised, and multiply realisable understanding of the oppor-
tunities that individuals should have access to, there may be some opportun-
ities that cannot be made available to some individuals, with some
impairments, in any form. Returning to the example of mobility: though
individuals with many mobility impairments can be enabled to control their
mobility, this may not be possible for individuals with locked-in syndrome, at
least for now.
Moreover, the very focus on diversity in the expression of capabilities may
generate a further objection about the possible ‘thinning-out’ of the content of
individuals’ entitlements. It might seem that if it is ‘enough’ to use a wheel-
chair to have the capability to be mobile, then those who use wheelchairs
rather than their legs to be mobile will be considered to have all they are
entitled to from the state. Thus, if, for example, it were possible to restore
functioning to a paraplegic person’s legs, even at very little cost, they would
neither be entitled to demand this from the state nor considered disabled if the
treatment were not provided.⁸
This is an instance of the more general worry that sufficientarian
approaches, by only ensuring individuals reach some threshold of achieve-
ment, may fail to eliminate radical, and potentially unjust, inequalities above
this threshold. The same line of reasoning would apply to other impairments,
and potential treatments for them: if capabilities can be experienced in a
variety of ways, then individuals can have these capabilities even when there

⁸ Note that this objection is specific to conceptions of entitlements as general opportunities (for
mobility, communication, forming relationships, aesthetic experiences, and so on), rather than those
that aim to provide the resources, or substantive capabilities, necessary to perform more narrowly
specified functionings (walking, talking, using all five senses, easily understanding others’ expressions
of emotion). Given that I will defend a version of the former approach, it is worth considering possible
responses to this objection. Further, the latter approaches will face the additional worry that individuals
will be entitled to state assistance (and, if my framework is applied, considered disabled) on the basis
that they are mobile, or communicate, or enjoy aesthetic experiences ‘in the wrong way’. Although
these individuals have control and a reasonable range of options in this domain, their inability to
perform specific (often species-normal) functionings will be seen as disadvantageous nonetheless. This
seems to put the focus in the wrong place, to put it mildly.
48      

are specific, seemingly valuable functionings they cannot perform. Thus, deaf
individuals who have all the general opportunities they are entitled to would
not be entitled to cochlear implants, even if these were cheap and effective,
blind individuals will not be entitled to operations to restore their sight, and so
on. This implies individuals are neither considered disabled by their inability
to perform specific functionings (walking, hearing, seeing, talking) nor entitled
to be enabled to perform them, if they have more general capabilities (for
mobility, leisure, aesthetic experiences, communication) at some appropriate
level. If these objections were to hit their mark, this would certainly be
damning for my account. Thus, I will now provide a response to each in turn.

2.4 Identifying Distributive Entitlements

I turn, first, to the objection that individuals with serious impairments will not
be considered disabled if the opportunity loss associated with their condition
cannot be avoided; and, further, that we will therefore face the dilemma of
either decreasing entitlements for this subset of individuals or levelling-down
entitlements generally. This objection is grounded on the assumption, often
called ‘the feasibility constraint’, that the demands of distributive justice must
be realisable: that if an individual is entitled to something, it must be feasible to
provide it to them. However, we need not accept this assumption. If, as
I suggest, we reject it, then there will be some instances of injustice we are
currently unable to rectify.
Most significantly for our purposes, some individuals may not be able to
access opportunities that justice requires they have. If a just state of affairs is
not currently realisable then they cannot have a claim that someone should
provide these opportunities. They can, nonetheless, have a claim that we work
towards a situation in which their provision is feasible. In other words, even if
no one currently has the ability to provide these entitlements, it may be that
someone (or some collective) has the ‘indirect diachronic ability’ to provide
them: the ability to provide these opportunities later if they perform some
action now.⁹ If, for example, we believe all humans are entitled to control over
their mobility, then this cannot mean the state has an obligation to provide
this to someone with locked-in syndrome if this is impossible. However, they
do have an obligation to enable them to have such control if it were possible,
and to work towards realising this state of affairs. Further, the individual with

⁹ Gilabert and Lawford-Smith 2012: 811. Also see Lawford-Smith 2013: 249–50.
 :  -  49

locked-in syndrome is disabled because they cannot be provided with an

opportunity that they should have access to: that they should be enabled to
perform if this were possible.
On this view, then, a situation in which some individuals lack opportunities
that everyone is entitled to may be unjust even if no one is guilty of causing this
injustice and no one has a direct duty to rectify it. However, some find the
rejection of the feasibility constraint, and so the suggestion that entitlements
should be conditional in this sense, implausible. How can we be entitled to
something no one has a duty to provide? Can we really allow ‘the conceptual
possibility of an unjust society in which nobody is doing anything wrong’?¹⁰
The answer, I believe, is yes.
Whilst allowing for the possibility of injustice without perpetrators might
seem counterintuitive, accepting the feasibility constraint has similarly
unappealing implications. It implies that procedures with unfair outcomes
that cannot be improved cannot be described as unjust—for example, a penal
system that sometimes convicts the innocent and fails to identify the guilty.¹¹
And, more significantly, that ‘one must first establish what is feasible in order
to determine what is just’.¹² Contrary, for example, to Cohen’s argument that
our fundamental principles of justice should be fact insensitive, and should
take priority over practical, action-guiding principles.¹³ Similarly, Pablo
Gilabert distinguishes evaluative and prescriptive uses of justice claims. The
truth of evaluative claims does not depend on whether ‘the obligation they
mention cannot be fulfilled’.¹⁴ Whilst there are certainly contexts in which
prescriptive or action-guiding principles are needed, my claim is that the
broad content of our account of distributive justice—our list of central cap-
abilities, primary goods, human rights, or whatever else we think all individ-
uals are owed as a matter of justice—should be evaluative or fact insensitive. If
we were instead to apply the feasibility constraint in the context of developing
such a list of universal entitlements, then it seems that we can only include
items on this list if they can currently be provided to every individual. Thus, if
it is currently logistically impossible to secure nourishment for all, then even
something this basic cannot be amongst our entitlements. Or, at least, it will
only be unjust that such a basic need is not universally fulfilled if there is
someone who can meet it: on this view, ‘a society can be fully just even if some

¹⁰ Stemplowska and Swift 2012: 385. ¹¹ Gheaus 2013: 452.

¹² Gheaus 2013: 452. I leave aside here whether this also commits adherents to a violation of the
naturalistic fallacy, as Gheaus (2013: 458) has argued: ‘A reading of “ought implies can” to mean that, if
X cannot be achieved, then X is the wrong ideal, is to derive a normative claim from a factual one.’
¹³ Cohen 2008. ¹⁴ Gilabert 2011: 56.
50      

needs remain unmet, provided there is nothing we can do to change the

situation’.¹⁵
Some of the disagreement with the statement of my position may be
semantic rather than substantive: even those who agree that we can distinguish
fact-insensitive fundamental principles of justice from action-guiding prin-
ciples, may object that talk of entitlements brings to mind the latter rather than
the former. To be clear, then, by entitlements I mean what all individuals are
conditionally owed as a matter of justice—the list of capabilities (or primary
goods or resources or human rights) central to many theories of distributive
justice—whether or not they can press their claim against a specific duty-
bearer, and whether or not they can currently be provided with every item on
this list. Thus, we can speak about individuals being entitled to the capability
for bodily integrity, or freedom of movement, or the social bases of social
respect, even when these cannot be universally secured.
This language allows us to capture what seems so objectionable about such
situations. Individuals may not be wronged when they cannot be nourished, or
mobile, or secure, but it is wrong, and not merely regrettable, that they do not
have the opportunities (or resources, or rights) we think they ought to have—
that they are entitled to.¹⁶ In addition, this language helps to identify the
strength and nature of our duties in such cases: these are not mere wants or
desires, which the state can choose to provide or not. Whilst we cannot have a
duty to provide what is impossible, the state (or other relevant body) should, as
noted above, work towards making these opportunities available, and certainly
should not artificially limit the boundaries of the possible. We cannot know in
advance what it is possible to achieve. Consider, for example, the vast changes
in our views on the capability set feasible for individuals with Down’s syn-
drome. There has been a movement in just a few decades from a focus on
institutionalisation and medicalisation, to enabling many individuals to be
fully functioning and participating members of society.¹⁷ If currently unrecti-
fiable capability deprivations are not considered injustices, we may focus
instead on less urgent claims we know we can meet (‘actual’ injustices).
Thus, we may fail to ‘invest . . . resources in pushing further the limits of
what is feasible’.¹⁸
On my view, then, not all disability will be the result of unjust behaviour—
though, of course, much will be. Sometimes the state may unjustifiably

¹⁵ Gheaus 2013: 453.

¹⁶ Further, as Gheaus (2013: 254) observes, ‘isn’t it mysterious that the same features that make
improvable situations unjust should make non-improvable situations regrettable but not unjust?’
¹⁷ See Bérubé (1996) for a personal account. ¹⁸ Gheaus 2013: 456.
 :  -  51

fail to fund research to develop technologies that would increase impaired

individuals’ functioning capacity, or wrongly fail to ensure that individuals
have access to existing technologies, or design material infrastructure in
avoidably exclusionary ways. Indeed, sometimes disability may be the result
not merely of such unjust negligence, incompetence, or oversight, but arise
from more direct discrimination or mistreatment. Examples include the loss of
freedom of movement or the ability to engage in social interactions of autistic
individuals (or others) locked in solitary confinement,¹⁹ or the lack of oppor-
tunity for employment given prevalent attitudes about the competence of
individuals with impairments.²⁰ More benignly, sometimes the restrictions
to individuals’ entitlements may simply be ineliminable and unavoidable:
perhaps due to natural obstacles, perhaps due to competing claims on a limited
pool of resources, or to conflicts between different forms of accommodation
(certain policies may render some impairments non-disabling (indeed, may be
the only way to do so) whilst at the same time exacerbate others: see §1.3).
Centrally, though, the content of our universal distributive entitlements should
not be determined by what it is feasible to provide universally.
Understanding entitlements in this way means we do not need to grasp
either horn of the above dilemma: we can have a universal set of entitlements
that is not levelled-down to include only those opportunities that can be
achieved by all. However, this does not imply that the content of our entitle-
ments can or should take no account of feasibility whatsoever. We could, for
example, take the view that humans should be entitled to the opportunities
necessary for a decent life, where this will be informed by our reasonable
expectations of what such a life entails. On this view, controlling our mobility
might be central to such a life; the ability to fly would, of course, not be. Further,
the content of our reasonable expectations for a decent life can be predicted to
change over time: we can now, for example, reasonably anticipate levels of life
expectancy, health, nutrition, and education that would not have been con-
sidered essential to a decent life even in the relatively recent past.²¹
The point, here, is not to identify the specific content of our distributive
entitlements, nor even to outline the method by which this content would be

¹⁹ Gallagher 2019. ²⁰ I discuss the counterintuitive implications of some such cases in §2.7.
²¹ I do not consider here the ways in which expectations may differ by geographical location.
However, if individuals have lower expectations because they have adapted to opportunities that are
limited as the result of unjust mechanisms of global interaction and distribution then it is not clear that
this should be used to justify a more limited set of entitlements. Localised expectations might, though,
have an impact on which opportunities we expect within a domain such as mobility, aesthetic
experiences, or leisure activities. I will discuss adaptive preferences in Chapter 6, and consider how
the content of our capabilities should be cashed out in Chapter 7.
52      

determined. Rather, it is to note that this might depend on what it is possible

for humans to do, and on what we think humans ought to be able to do,
without implying that it must be possible that all these entitlements must
currently be achievable for every human. In other words, entitlements can be
indexed to what is feasible for most individuals, without requiring that they
can be provided to all (though we should strive to achieve this).
Two further objections may be raised against my approach here. First, if our
entitlements depend to some extent on our reasonable expectations about a
decent (or dignified, or flourishing) human life, and our reasonable expect-
ations are informed by a consideration of what humans, as a species, can
generally do, then this might seem to collapse into a species norm approach.
That is, we are entitled to perform species normal functionings, and disabled
when unable to do so as the result of an impairment. Yet this is not the case.
Even without specifying the precise content of our entitlements we can see that
these will be both narrower and broader than species-normal functionings.
They will be narrower, first, because on any plausible view, some relatively
trivial human capabilities will not be central entitlements in any form (the
ability to roll our tongue, to sing, or to stand on one leg). Individuals who thus
deviate from the species norm do not lack opportunities they are entitled to.
Our entitlements are narrower, too, if understood as general opportunities,
rather than the performance of specific functionings. An individual can have
what they are entitled to (the capability for mobility, say), even if there is a
species-normal functioning they cannot perform (walking or running using
their own legs).²²
Additionally, our entitlements will be broader than the category of species-
normal abilities, since we may be entitled to more opportunities than those
that are ‘natural’ or normal for our species.²³ As medical, social, and techno-
logical progress expands the scope of human capacities, our reasonable
expectations of a decent life, and the opportunities we should thus be entitled
to, will surely reflect this. For example, an account of species-normal mobility
will depend on what a human body can normally do, or is ‘designed’ to do:
walking at a speed of 3mph, running at a speed of 10mph, being able to travel
at least twenty miles in a day, being able to walk for at least a few hours without
rest. Yet our entitlement to mobility will surely incorporate much more than
this. For example, being able to make a hundred mile round-trip commute to

²² I respond to the apparent implausibility of this implication in §2.7.

²³ The species norm approach will be restricted to only ‘natural’ human abilities if we follow
Daniels’s (1985: 28) suggestion that it derives from the ‘design of the organism’, rather than ‘merely
a statistical notion’.
 :  -  53

work in a reasonable time, being able to travel around the country, or even
abroad, to visit friends and family, or for leisure.²⁴ The idea of being entitled to
easily travel a hundred miles in a day would have been deeply implausible even
in the recent past. Consider the discussion between Elizabeth Bennett and
Mr Darcy over whether spending half a day travelling fifty miles should be
deemed an ‘easy distance’.²⁵ Certainly it is not a journey either would expect to
be able to undertake twice daily, and nor would we consider them conse-
quently disadvantaged. Yet I would contend that if an individual in our society
were unable to engage in such travel, and accordingly could not take certain
jobs or see some of her friends or family, then she would lack something she is
entitled to—and if this is the (partial) result of an impairment, then she would
be disabled.²⁶
A second objection concerns the potential ambiguity regarding what it
means to say that it is feasible to provide an individual with an opportunity.
I have argued that we should reject the feasibility constraint, and hence that
individuals may be entitled to opportunities that cannot currently be provided,
but that these opportunities should, nonetheless, be provided if possible. But
what if this is only possible at radical expense of time or resources? There are
two broad forms of this worry. First, if we could all be enabled to fly, say—if the
government were to divert all research funding to this project—then should
this opportunity be an entitlement, in the absence of which we are disabled?
Clearly the answer is no. When this is a mere abstract possibility, such a
functioning is unlikely to feature in our reasonable conception of a decent life,
or what we can reasonably expect from the state. However, I would concede
that if a way of enabling humans to fly were discovered and made available to
the majority, then we might expand our conception of our entitlements to
include this.²⁷ In this imagined scenario this is no more a reductio ad absur-
dum of my approach than acknowledging that individuals’ capacity for mobil-
ity should take account of modern transportation and that, as such, someone
who could only get around at speeds and distances walking and running would
usually allow may not have what they are entitled to in this domain. To
emphasise: insisting that we can be entitled to opportunities it is infeasible

²⁴ Concerns about climate change may push against our entitlements to engage in foreign travel for
as long as this likely involves highly polluting air travel. Similarly, the harm we may do to others by
travelling widely during a global pandemic might also limit how much travelling we can reasonably
expect to do. I leave these concerns aside here.
²⁵ Austen 2004: 137.
²⁶ I discuss the contextual nature of my account of disability further in §2.7.
²⁷ Thanks to Paul Billingham for pushing me to clarify this point.
54      

to provide universally does not mean our entitlements must be entirely

unmoored from what is feasible for most.
Second, and perhaps more troubling, are cases in which an opportunity is
available to most individuals, but can only be provided to a further minority at
great expense. For example, if an individual with locked-in syndrome can
control their mobility only with an extremely expensive piece of equipment,
are they entitled to it? Our approach to such questions must involve weighing
up the capabilities provided to the minority against those lost by the majority.
Thus, if the cost of providing this equipment meant that others suffered
premature death or significant impairment as a result, then the individual
surely cannot be considered entitled to their mobility in the face of these costs.
In such a case, we can continue to say that it is impossible for the individual to
be mobile and they are unavoidably disabled, given that their entitlements can
only be provided at a prohibitive cost. There will of course be many difficult
cases, but these are the considerations we must try to balance.
This means that scarcity will not be wholly irrelevant to identifying our
entitlements. Again, I claim only that the possibility of universal provision is
not a prerequisite to something being an entitlement. Feasibility, including
considerations of scarcity, may play a role in determining where the bound-
aries of our entitlements lie: for example, as discussed, via informing our
reasonable expectations about the shape of a decent life. Returning to the
above example, if individual jet-pack technology were created but was pro-
hibitively expensive and/or relied on a resource too scarce for it to be made
available to all, then this is not likely to feature amongst our reasonable
expectations and thus our entitlements. In contrast, if this could be provided
near-universally, became the dominant mode of transport, and a prerequisite
for participation in various aspects of social and political life, then this may be
amongst our (conditional) entitlements.

2.5 Entitlements without Hierarchy

This understanding of entitlements allows us to avoid the above dilemma, and

adopt a universal conception of entitlements even if we cannot currently
provide individuals with all the opportunities they are entitled to. Thus, we
can avoid a hierarchical notion of entitlement, according to which those who
can be provided with less will be entitled to less, without a general levelling-
down of the scope of our entitlements. However, a further objection may be
raised against advocating a set of opportunities (or resources) all humans
 :  -  55

should be entitled to. This is, that it will entail forcing a conception of a good
life on disabled individuals ‘from the outside’, or insisting that the most
important opportunities in life are ones they cannot access or appreciate the
value of.
This is not the case, as subsequent chapters will elaborate. This worry does,
however, highlight some of the ways in which a consideration of disability
pushes us towards a particular conception of justice, as I will briefly outline
here. First, when entitlements are understood at an abstract or generalised
level then most individuals with most impairments can have access to them,
even when they cannot perform a particular functioning (being mobile whilst
being unable to walk, say). Thus, valuable opportunities are not generally
closed to individuals with impairments; they are simply experienced in differ-
ent ways. Further, individuals can recognise the value of an opportunity, even
if they are part of a minority who cannot exercise it. An individual with
locked-in syndrome, for example, need not dispute the value of mobility nor
deny that this should be a central human entitlement. This is so whether they
repudiate their condition and wish it could be cured, or are fully satisfied with
their mode of existence. Thus, individuals with impairments need not dispute
the value of certain capabilities, even if they are constrained from performing
them.²⁸
Indeed, far from imposing a conception of the good on disabled individuals,
I contend that understanding disability as a restriction in the ability to perform
a specific, but generalised, set of capabilities, in fact better captures the
concerns of disabled individuals than competing approaches. Many other
approaches will not be satisfied with providing an acceptable set of options,
and aim instead to provide individuals with an ever ‘broader range of oppor-
tunities for functionings’.²⁹ Welfarist approaches, for example, consider indi-
viduals to be disabled by the restriction in any functionings that may
contribute to an individual’s welfare. Egalitarian versions of the capability
approach would consider individuals to be disabled by any significant differ-
ence in their functioning capacities. Wolff, for example, argues that even a
wealthy paraplegic person is disabled because their mobility is not at ‘the level
enjoyed by others’.³⁰ What this means is slightly ambiguous. If ‘not being at
this level’ means that this individual is not able to be independently mobile,
then certainly they should be considered disabled (and would be on my

²⁸ The possibility of overlapping consensus on capabilities amongst individuals with different

capacities is discussed in Chapter 4.
²⁹ Terzi 2009: 103. ³⁰ Wolff 2009a: 125.
56      

account). A more plausible interpretation of the example, though, is to assume

that this individual has the level of mobility that a wealthy paraplegic person is
actually likely to have in our society. That is: they can be mobile, with the
use of a potentially expensive wheelchair or other form of mobility assistance,
have access to a modified vehicle or accessible public transport, are able to
take flights, and so on—but they are unable to walk or run. This is sufficient for
Wolff to deem them disabled since they do not have the same capabilities as
other, non-paraplegic, individuals.
Yet I would contend that insofar as they have control over their mobility,
this is sufficient: we should be entitled to acceptable alternatives, but not
identical options. This restriction to relevant opportunity loss seems to better
capture the claims and concerns of disabled individuals. Consider, for
example, Harriet McBryde Johnson’s account of the pleasures of wheelchair
use, and the misperceptions of those who do not use one:

I used to try to explain the fact that I enjoy my life, that it’s a great sensual
pleasure to zoom by power chair on these delicious muggy streets, that I have
no more reason to kill myself than most people. But it gets tedious . . . they
don’t want to know. They think they know everything there is to know just
by looking at me. That’s how stereotypes work. They don’t know that they’re
confused.³¹

Of course, such testimony does not in any sense prove that wheelchair users
should not be compensated. However, it does give us reason to question the
assumption that we must consider individuals disabled, and compensate them,
simply because they are mobile with the use of a wheelchair rather than their
legs. Given the prevalence of disability positive testimony this does not seem to
be an appropriate response to their demands, or reflective of many disabled
individuals’ experiences.
Yet, a critic might respond, what about those individuals who do not enjoy
using a wheelchair? Will my account cruelly fail to assist them? Broadly, the
answer is ‘yes’ (as §2.6 and Chapter 4 will discuss). However, we often fail to
compensate individuals for features of themselves they dislike. This case is
only especially ‘cruel’ if we assume it is especially disadvantageous. If it can be
shown to be disadvantageous the individual should be entitled to assistance,
but we should not assume this is so. I defer a full defence of my view of what
individuals are entitled to, and when they are relevantly disadvantaged, to the

³¹ McBryde Johnson 2003.

 :  -  57

following two chapters. For now, my point is simply that we should not
assume any non-standard functioning is disadvantageous, and that it is only
when it is disadvantageous (i.e. involves the deprivation of what we are
entitled to) that it is disabling. As Elizabeth Anderson points out:

It is useful to ask what the deaf demand on their own account, in the name of
justice. Do they bemoan the misery of not being able to hear, and demand
compensation for this lack? On the contrary . . . [t]hey insist that sign lan-
guage is just as valuable a form of communication as is speech and that the
other goods obtainable through hearing, such as appreciation of music, are
dispensable parts of any conception of good.³²

It is not the mere differences in their lives and ways of functionings that
individuals with impairments consider disabling, or demand rectification of.
Nor, in many cases, do individuals want the impairment itself to be eliminated.
Our focus—in activism for social change and in the development of policy—
should be on removing limitations on individuals’ access to the capabilities
(or resources) they are entitled to, not ensuring every individual exercises these
capabilities (or uses these resources) in an identical way. Accepting that not all
difference is disadvantageous, and only disadvantage is disabling, does not
require accepting that we should be entitled to multiply realisable opportun-
ities, only the general framework of my approach to disability.

2.6 Beyond Minimal Functionings

I now turn to an objection that is specific to my view that individuals are

entitled to general and multiply realisable opportunities: namely, that such an
account is far too minimal, and will mean that individuals will not be con-
sidered disabled even when they lack important opportunities. First, I consider
Anita Silvers’s objection to what she takes to be capability theorists’ approach
to educational provision for individuals with impairments. Silvers claims that
the approach would only provide ‘educational opportunity adequate for a
dignified life, which depending on cultural context may not even need to
rise to the level of average educational attainment’.³³ She discusses a child with
a hearing impairment, who struggles to catch all the information imparted in
class due to the difficulty of lip-reading in this context. Nonetheless, she is

³² Anderson 1999: 333. ³³ Silvers 2009: 181.

58      

denied a translator since her educational attainment is already above

average.³⁴ Silvers objects that it is unfair that children with impairments are
unable to fulfil their potential, simply because their educational environment
is not designed to support them. Further, it might seem that, on my view, these
children would (implausibly) not be considered disabled by this restriction in
their access to education because ‘they already have what they’re entitled to’.
However, contrary to Silvers’s view, a decent education surely requires more
than the opportunity for some minimal educative outcome, such as passing
grades.³⁵ More plausibly, this requires something like ‘the opportunity to
develop our potential to a similar degree to our peers’. This must be at least
part of the goal of education, and we would fail to show individuals equal
concern and respect if we provided this to some and not others.³⁶ As Silvers
puts it, we should aim to provide ‘equality of meaningful access to the content
of instruction appropriate to developing students’ talents whether or not they
have disabilities’.³⁷ Such an aim seems entirely compatible with any plausible
reading of the capability approach.
This is not to say that there will not be difficult cases, where the needs and
requirements of different children pull in different directions, and limitations
of resources mean that a learning environment cannot be provided that is ideal
for everyone. However, as has been discussed, the fact that something cannot,
currently, be universally provided does not mean that we cannot reasonably
include this amongst our entitlements. Individuals will not have what they are
entitled to, then, without equality of meaningful access to education.
Yet it may be objected that education is a special case. Whilst the very goal
of education involves enabling individuals to develop their talents and abilities,
having opportunities for mobility, communication, or leisure surely does not
require that we are able to satisfy our full potential in these domains. It seems
much more plausible to suggest (as I will do) that we are here entitled to
something like acceptable options and the capacity to meaningfully choose
between them. As such, individuals who achieve mobility with a wheelchair
rather than by walking, or communicate using Sign rather than speech, or look
at art because they are unable to listen to music, are not considered disabled by
this restriction in their ability to function; nor would they be entitled to an

³⁴ Silvers 2009: 177–82.

³⁵ The same is true if we focus on other criteria to identify entitlements, such as leading a dignified or
flourishing life. Indeed, if a criterion did not have this implication, this would surely count as a reason
against its adoption.
³⁶ See Terzi (2008) for a capability approach to education for children with impairments, grounded
in such equal concern and respect. (Also see Gutmann 1987.)
³⁷ Silvers 2009: 186.
 :  -  59

operation to restore the use of their legs, or cochlear implants to restore their
hearing, even if this would not be costly.
This bullet should be bitten, though only with some important caveats.
I would accept, then, that if we lived in circumstances in which individuals
with impairments had all the opportunities they are entitled to, then they
should neither be considered disabled, nor necessarily entitled to assistance
from the state. For example, if an individual with a mobility impairment could
exercise control over their mobility without being dependent on others for
help, or being excluded from access to the public, built environment, then they
may not be entitled to be enabled to walk.³⁸ Similarly, a deaf individual who
had access to leisure activities and aesthetic experiences (and every other
capability they are entitled to) would not be entitled to cochlear implants
that allowed them to hear music if these were available. Nor would these
individuals be considered disabled.
Such an implication will strike many as counterintuitive but, as §2.7 will
argue, I do not believe this should motivate rejecting this account. For now,
I will simply point to an analogous case that does not seem implausible:
individuals with visual impairments that can be accommodated with glasses
and contact lenses. Such individuals still have an impairment (short-
sightedness, perhaps). Further, many face restrictions in their activities: certain
occupations may be closed to them (being an airline pilot, for example), long
periods of reading or screen use may cause eye strain, they may be unable to
read if they do not have their glasses or contact lenses with them, and they
may be unable to participate in certain sports if contact lenses are insufficient
and glasses are prone to falling off or breaking. Nonetheless, we do not
generally consider these individuals to be disabled. Indeed, many would balk
at what would seem to be a wanton and unnecessary expansion of the
term. Other impairments may be less common, and we may be less used to

³⁸ Exactly how much of the environment would have to be inaccessible to conclude that an
individual lacks control over their mobility is ambiguous. At one extreme, imagine a situation in
which a few, privately owned houses exist that were built before concerns with accessibility, and have
narrow halls and staircases unsuitable for manoeuvring a wheelchair, and which could not easily be
altered. Individuals who use wheelchairs may find it difficult to visit friends in such houses, and when
purchasing a home, these would not be an acceptable alternative. This seems insufficient to conclude
that an individual lacks control over their mobility. The same restrictions might apply to an individual
with a cat-hair allergy (they cannot visit the homes of certain friends or, perhaps, purchase certain
homes where a cat has lived, or lives nearby). At the other end of the spectrum, imagine a historic town
centre, in which most central shops, offices, and homes are similarly inaccessible to those who use a
wheelchair. This level of restriction surely would prevent someone having control over their mobility,
since so many options they would reasonably wish to pursue are closed to them. I will not attempt to
settle how the various intermediary possibilities between these two extremes should be classified, which
would require a consideration of the particulars of each case.
60      

accommodating them, but we have no more reason to insist that these must be
disabling, even when individuals have all the opportunities they are entitled to.
Nonetheless, some caveats are needed. First, individuals with impairments
often do not actually have what they are entitled to. In many cases, the
restrictions in their capabilities are avoidable, and could be eliminated by
changing social infrastructure and norms. However, this is not always so. As
we have seen, in some instances this is because we lack the technology or the
resources to prevent some impairments from restricting individuals’ oppor-
tunities. In others, this is because individuals’ needs conflict, and there is no way
to structure the external environment that would eliminate the restrictions in
activity and participation for all individuals. As such, in many cases the appro-
priate way to mitigate the disadvantage associated with an impairment may not
be to change the environment, and this might be so even when the disadvantage
is a result of that environment. Thus, to conclude that if an impairment was not
disabling (did not result in relevant restrictions) an individual would not be
entitled to state assistance does not imply that now, or in the near future,
medical treatments that mitigate impairments would not be provided.
The second caveat is to point out that suggesting that such treatments may
not be a matter of entitlement does not imply that they are banned or
restricted. Any liberal approach to distributive justice will aim both to ensure
fairness in distribution, and to allow individuals the freedom to form and
pursue their own conception of the good. Thus, if an individual desires to
walk, or hear, or see, they will not be prevented from pursuing these goals.
Thus, these treatments would have the same status as laser eye surgery in our
society: individuals may choose to undergo it if they would prefer to see
without using glasses, but they are not entitled to it if their visual impairment
is not restricting their access to their entitlements.
There is one way in which this analogy breaks down, however: these
treatments, and the choices they bring, should not be open only to the
minority who can afford them. Indeed, in a situation in which individuals
had access to all of the central capabilities (or primary goods, or equality of
Dworkinian resources) there would be sufficient resource equality that indi-
viduals would have the ability to achieve these goals (especially on the above
assumption that these treatments are inexpensive). Indeed, my account does
not even rule out state provision or subsidy of these options on the grounds of
public choice, as long as the decision is procedurally just and does not divert
resources from the satisfaction of actual entitlements.³⁹

³⁹ Thanks to Carl Fox for helping me clarify this point.

 :  -  61

One final sense in which our entitlements might seem too minimal on my
approach relates to an objection earlier raised against the species norm
approach. This is, that by focusing on what individuals are able to do, I take
insufficient account of how much more difficult it may be to do certain things
when we have an impairment. Someone who uses a wheelchair, for example,
may have the capability for mobility, but this will often be restricted by both
the built and natural environment. Thus, they may find travelling much more
physically demanding than, for example, walking a similar distance. Or, an
individual in chronic pain may be able to perform all the functionings that a
person without this condition can, yet they must struggle against the pain and
fatigue that accompanies every functioning they perform.⁴⁰ I would certainly
want to resist the implication that the difficulty of performing various func-
tionings would not be considered a relevant dimension of disadvantage, and
that individuals who experience such difficulty would not be considered
disabled, nor entitled to assistance to mitigate it.
Whether this is indeed an implication of my approach depends on the
sensitivity of our conception of distributive entitlements not just to what we
can do, but also to the pain and effort that must be expended to do it. I would
contend that such concerns should feature in our understanding of individ-
uals’ entitlements.⁴¹ It is important to note that insisting that a theory of
distributive justice is sensitive to these considerations is not merely a post-hoc
modification to ensure that my definition of disability remains plausible. On
the contrary, the above examples give us reason to think that individuals
should be entitled not just to the mere capacity to perform some functionings,
but to be able to do so without associated pain, and without having to expend
massively more energy than others to achieve the same outcome. Thus, the
point is not that an account of our distributive entitlements that ignored the
pain and effort required to achieve them would generate an implausible
account of disability if my framework were employed, but that this would be
an independently implausible account of our entitlements. More generally,
I contend that a plausible understanding of our entitlements will generate a
plausible definition of disability on my account. This is illustrative of my
central theme: not only should our account of disability be informed by our
understanding of what justice requires, but an investigation of disability
should inform the theory of justice we adopt, as I will show.

⁴⁰ Thanks to Zofia Stemplowska for this example.

⁴¹ For approaches that do so, see Wolff and de-Shalit 2007; Wolff 2009a; 2009b; Cohen 2011;
Anderson 1999. It may be a problem with Nussbaum (2000a; 2006a; 2011a) and Sen’s (1999a; 2009)
approaches that they do not clearly incorporate such concerns.
62      

2.7 What Counts? Who’s Disabled?

I will finish by considering whether my account will prove to be too revision-

ary. Certainly, the impairments that would be considered disabling do not
neatly cohere with our intuitions in a number of cases. As emphasised from
the outset, my account will imply that individuals with significant and visible
impairments may not be disabled by them, that individuals with seemingly
minor or invisible impairments may be disabled, and that individuals’ status as
disabled may change across time and context. However, far from giving us
reason to reject this approach, this is one of its key benefits. As §1.7 discussed,
there is little point in devising a definition that captures all paradigm cases
when we have good reason to think the paradigm is informed by prejudice and
misinformation.
First, then, many paradigm impairments, such as deafness, blindness, para-
plegia, achondroplasia (a common form of dwarfism), ASCs, and Down’s
syndrome need not be disabling on my account. If individuals have the oppor-
tunities they are entitled to—perhaps as a result of adequate infrastructure,
access to resources, and lack of bias—they will not be considered disabled
merely on the basis of their impairment. Throughout this chapter I have
emphasised that disabilities should not be seen as simply any bodily or cognitive
difference, but rather as a particular subset of the opportunity losses that might
result from such difference. This does not imply that we cannot campaign for
the better treatment of individuals with impairments, or that they will not be
entitled to state assistance. However, we should not assume that having an
impairment, especially a visible impairment, must be disadvantageous. We
should, instead, focus on the specifics of each case, and determine whether an
impairment is actually associated with restrictions in individuals’ activity and
participation. As Chapter 6 will explore, the reality often will not accord with
our unexamined intuitions.⁴²
My approach will also lead to deviations from the paradigm in the other
direction: many non-paradigm cases, such as depression and anxiety, and even
obesity or shyness, might be disabling. Further, individuals who have body
integrity identity disorder (the strong desire to amputate a healthy limb) may
be disabled by being prevented from undergoing this amputation, insofar as
living with a limb they disassociate from disrupts their access to capabilities,

⁴² As discussed, part of the motive behind the introduction of the social model, and the rhetoric of
claiming that ‘disablement is nothing to do with the body’ (Oliver 1996: 35), was to resist the
medicalisation of disability and to highlight the contingency of the connection between impairment
and disadvantage.
 :  -  63

for example, by making social interaction difficult.⁴³ I will consider perhaps

the most controversial of these examples: obesity. I am aware that both
disabled and obese individuals will likely object to this implication, given the
unjustified stigma attached to both these labels.⁴⁴ It is worth emphasising,
then, that though I consider an obese body an impairment (after all, it is a
deviation from the species norm), I am not suggesting that it is worse, or
necessarily disadvantageous, to have such a body. Impairment is analogous to
‘different’ rather than ‘defective’.⁴⁵ However, this is a dimension of difference,
like many others, that may be associated with disadvantage: that is, the
restriction of important opportunities.⁴⁶
My focus here is not the possible (and sometimes controversial) correlation
between obesity and health conditions such as high blood pressure, diabetes,
or risk of heart attack or stroke. Even leaving these aside, merely having an
obese body can be disabling. Our built environment is made for human bodies
of a particular size and weight. As Roxane Gay says, ‘[t]here are very few
spaces where bodies like mine fit’.⁴⁷ One simple and important example is
chairs: aeroplane seats, restaurant booths, lecture theatres, cinema seats.
Again, Gay gives a powerful account of this problem:

Anytime I enter a room where I might be expected to sit, I am overcome by

anxiety. What kind of chairs will I find? Will they have arms? Will they be
sturdy? How long will I have to sit in them? If I do manage to wedge myself
between a chair’s narrow arms, will I be able to pull myself out? If the chair is
too low, will I be able to stand up on my own?⁴⁸

This lack of accommodation of individuals with larger bodies can be hugely

restrictive to those who possess them, limiting opportunities to be mobile, to

⁴³ For discussion of these cases, see Bayne and Levy 2005; Müller 2009; Blom et al. 2012.
⁴⁴ One possible objection to considering obese individuals disabled is that their bodies are a ‘choice’,
and their condition ‘self-inflicted’, unlike paradigm impairments. First, this often is not true: social
factors and upbringing, past abuse and mental health can all contribute to individuals’ eating habits.
Yet even if this were simply a ‘choice’, this does not obviously create a distinction from more standard
cases of impairment: individuals may choose to engage in risky sports, like sky diving or rugby, and so
be similarly responsible for their resulting impairment, or may choose not to have their impairment
eliminated (just as obese people may ‘refuse’ to diet). Thus, appeal to responsibility is not a promising
way to distinguish these cases, even without settling the question of whether responsibility provides a
reason to treat disadvantaged individuals differently.
⁴⁵ This idea was introduced in Chapter 1 and will be defended in Chapter 5.
⁴⁶ There are, of course, some differences correlated with disadvantage that we would not want to
describe as disabilities, such as our gender or racial identity, or sexual orientation. As §1.2 considered,
this need not be an implication of my view. (Also see §5.5.)
⁴⁷ Gay 2017: 184. Also, see West 2016. ⁴⁸ Gay 2017: 184–5.
64      

access education on an equal footing with their peers, or to engage in many

leisure activities.
In addition, an obese body may be restrictive independently of the design of
material infrastructure, rendering ‘impossible’ activities like standing for long
periods, walking a mile uphill, or hiking to an overlook with a great view.⁴⁹
Finally, the bias, stigma, and abuse obese individuals are often subject to, the
commentaries on their bodies by friends, family, medical professionals, and
strangers, and the assumptions made about their capacities may further limit
their capabilities. Options may be closed to them because others assume they
are incapable of exercising them (for example, not being offered certain jobs),
or they may lack internal capabilities (for example, not feeling qualified to
apply for these jobs in the first place).⁵⁰ In all these ways, then, obesity—like
other forms of physical and cognitive difference—can be disabling.
A further way in which my account may seem to be over-inclusive is by
incorporating those whose restrictions do not seem to follow ‘directly’ from
their impairment. As noted in §2.4, there are different reasons why individuals
with impairments may lack opportunities they are entitled to. In some cases
this opportunity loss is simply unavoidable. Hence, though the situation is not
ideally just, this is not because any individual or collective agent is guilty of
injustice. At other times, individuals may have their abilities restricted due to
some form of discrimination or mistreatment. This is surely not a dichotomy
but a continuum—from the unavoidable, to the unavoidable now, to the costly
to avoid, and so on. The seemingly problematic cases for my approach are those
on the avoidable end of the scale: restrictions that are avoidable because they are
artificial and externally imposed, and so much so that we might begin to doubt
whether any connection exists between the restriction and the impairment.
Consider the following succession of cases:

(1) It is possible to replace a stairway entry to a building with a ramp.

There is not space for both. This would improve accessibility for some
individuals, including wheelchair-users, but would exclude a greater
number of building-users whose mobility impairment makes stairs
easier to master. The decision is made to keep the stairs.⁵¹

⁴⁹ Gay 2017: 16.

⁵⁰ In Nussbaum’s terminology, internal capabilities are ‘developed states of the person herself that
are . . . sufficient conditions for the exercise of requisite functions’ Nussbaum 2000a: 84). Whether
restrictions in internal capabilities constitute a loss of entitlements depends on the conception of
entitlements we employ. As noted, I defer to Chapters 3 and 4 the task of defending a capability
approach to justice.
⁵¹ See, Shakespeare 2014: 37; §1.3.
 :  -  65

(2) Due to a bureaucratic error, a homeless wheelchair-user is provided

with an inaccessible hotel room as emergency accommodation.⁵² It is at
the top of a flight of stairs meaning they cannot leave without assist-
ance, and is designed such that they cannot comfortably move around,
get in and out of bed, or use the toilet alone. This individual thus lacks,
inter alia, freedom of movement, control over their environment, and
access to adequate shelter.
(3) An individual is similarly confined to an inaccessible hotel room, but in
this case as the result of the direct and conscious action of an employee
of the Department of Work and Pensions, despite accessible accom-
modation being available.
(4) A schizophrenic individual is confined to a cell as the result of a state
policy of imprisoning all individuals with cognitive impairments,
resulting in a similar opportunity loss to the individuals in (2) and (3).

Whilst the degree of injustice, and the extent to which the impaired
individual can press a claim against any agent, differs in the first three cases,
all are pretty straightforward examples of the ways in which individuals with
impairments might be disabled by some combination of institutions, norms,
environment, resources, and support networks (as §1.4 discussed).
The fourth case is rather more counterintuitive. The individual’s opportun-
ities are restricted as a result of their impairment—it is this, after all, that is
used to justify their imprisonment under this terribly unjust policy. However,
it might seem that their impairment does not ‘directly’ contribute to the
restriction in their mobility as it does in the former cases: whilst the hotel
rooms are inaccessible because of the wheelchair-users’ impairment, we might
say it is being imprisoned rather than being schizophrenic that causes the
restriction in this case. It may seem odd, then, to say that they are disabled as a
result of their imprisonment, whilst the able-bodied person in the neighbour-
ing cell—imprisoned due to a policy of race-based incarceration, say—is not
disabled by their identical ability loss. Nonetheless, I believe this is exactly
what we should say.
Norms or bias usually act in some way to exacerbate what might seem like
the ‘natural’ effects of an impairment, but we should not allow this to mislead
us. Impairments are functional atypicalities that are often disabling because an
individuals’ way of functioning does not fit well with their context. It is not
natural that an individual with Down’s syndrome is excluded from education,

⁵² E.g. see Youngs 2019.

66      

or that an individual with autism lacks opportunities to work because potential

employers find stimming or self-soothing behaviours offputting. Nonetheless,
it is common to see these as ways they are disabled by their impairment.
This is not to say that there is nothing distinctive about cases such as (4), in
which the ability loss associated with an impairment is entirely arbitrary. Such
cases are particularly objectionable, and often more unjust than the standard
cases: the losses are not unavoidable, nor the result of negligence, nor uncon-
scious bias, but a conscious decision to limit an individual’s opportunities
because of their impairment. However, this line is not a clear one: it is not clear
that (3) is markedly more just. Indeed, even negligence or failures to mitigate
the effects of our unconscious biases can also be deeply unjust. Thus, cases
such as (4) may be towards one end of a scale of the avoidability of the
restriction and, relatedly, the injustice done by imposing it. Nonetheless,
they are not qualitatively different from more familiar cases in which restric-
tions are imposed on individuals with impairments.
Moreover, I believe that including cases in which the connection between
the impairment and the opportunity loss seems wholly contingent better
achieves the aims that a definition of disability should achieve. Recall that
I take myself to be engaged in an ameliorative project, with my goal being to
develop an account of disability that is effective in our fight against injustice
(see §1.8). These contingent cases of direct discrimination against individuals
with impairments certainly constitute relevant opportunity losses from this
justice-oriented perspective. We will be better able to identify and resist such
injustices if our definition of disability (as disadvantage) includes all cases in
which individuals lack opportunities they are entitled to as a result of their
impairment. Even if it were possible, embarking on a project of solidifying the
vague and rather arbitrary line between ‘natural’ and ‘unnatural’ opportunity
losses seems likely to create unnecessary divisions within the disability rights
movement, and may potentially mean important injustices fail to be recognised.
The final counterintuitive implication is that what counts as a disability on
my view will be contextualised, and may change over time. This can happen in
at least three sorts of way. First, as society becomes better able to accommodate
impairments, and enable different forms of functioning, then these impair-
ments will cease to be disabling. This is not because these impairments are
‘cured’, but because their link with disadvantage is severed. For example, if the
oft-discussed examples of deafness and mobility impairments no longer lead
to capability loss, then individuals with these impairments may not be dis-
abled, though they still cannot hear or walk unaided. Whilst this may seem
counterintuitive to some, it accords with the experience of many disabled
 :  -  67

individuals, who do not regret, or demand compensation for, the different

ways they function (see §2.5). It is also in line with our approach to other
impairments that have been similarly accommodated: for example, individuals
with visual impairments who wear glasses or contact lenses.
Second, as our capacity to provide opportunities changes, individuals may
be entitled to more if the content of our entitlements depends on our reason-
able expectations or something similar. For example, we now expect leisure
time, aesthetic experiences, and a longer life than previously. Impairments that
restrict access to these new entitlements will then count as disabling, even
though individuals’ actual abilities have not changed. One example here might
be (some forms of) dyslexia. In a context in which many people are illiterate,
this is unlikely to be disabling. In a context in which a grasp of the written
word is a prerequisite for many other opportunities, it would be. Thus,
dyslexic individuals may become disabled over time, or by moving from one
context to another. Consider, for example, a dyslexic girl who emigrates from
rural Gambia, where illiteracy is rife and educational infrastructure under-
developed, to the UK, where she has better access to education, though her
dyslexia is not fully accommodated. Whilst she is now less disadvantaged (she
now has access to education), she may also become disabled insofar as it is
now her impairment that restricts her access to education (where previously it
may have been poverty or sexism). Again, this seems to be the right way to
understand such cases.
Lastly, individuals who have impairments, such as fatigue and pain condi-
tions, which are not experienced as constant over time, but involve periodic
and perhaps unpredictable flare-ups, may have a changeable disability status.
To pick a personal example, consider migraines. During the hours when I am
experiencing impaired vision, nausea, and pain, I am unable to drive, unable to
read, and sometimes unable to enter any brightly lit space. Given the infre-
quency and comparatively short duration of such flare-ups, combined with
personal features such as a flexible work schedule, these restrictions do not in
general amount to a limitation in the opportunities I am entitled to. Many
others are not so lucky. Some will be disabled only during a flare-up, some may
be disabled during periods when flare-ups are common, and for some the mere
possibility of a flare-up may render them consistently disabled. For example,
an individual with epilepsy may never be able to drive a car, or an individual
with ME may struggle to retain employment if their impairment necessitates
frequent absences.
Making a determination about any case will depend both on features of the
relevant impairment (for example, severity, duration, and frequency of pain,
68      

fatigue, or other symptoms) as well as contextual features (for example,

workplace accommodations, a supportive employer and colleagues, access to
resources, and a personal support structure). It is undoubetedly appropriate
to make such determinations on a case-by-case basis, rather than making a
blanket assumption that all individuals with some impairment (in all times
and places) are or are not disabled by it.

2.8 Conclusion

It is a consequence of my approach that individuals are not categorised as

disabled in the way many would expect, and that who is included in this
category may change. Yet it is surely right to set aside our preconceptions
about how bad some impairment must be, and concentrate instead on the
actual restrictions individuals face, and on which of these should be the focus
of our concern. What matters—what is disabling—is that individuals are
rendered (often contingently) unable to be mobile, to communicate, to form
social relationships, or to access education or meaningful work as a result of
their impairments. Not that they are unable to walk, unable to listen to music,
or unable to read emotional cues easily. Whatever account of distributive
justice is used, then, my approach to disability will be revisionary and contrary
to many common-sense intuitions about who is disabled. However, these
implications constitute a feature, not a bug. Our present intuitions concerning
what constitutes disability often reflect prejudice and bias, and should be
altered. Further, our account of disability should change over time, and reflect
what individuals are able to do, and what we think they ought to be able to do.
The problem of disability is the problem of losing significant opportunities—
and not the ‘problem’ of deviation from the species norm, or the loss of our
irrelevant opportunities—and our understanding of disability should be reori-
ented to reflect this.
 3
Disability and Distribution
A Capability Approach

3.1 Introduction

Disability appears frequently in examinations of distributive justice. Having

developed a theory of justice—often based on the consideration of standard or
idealised individuals, unburdened by many features of actual humans—
philosophers then consider how their theory will handle ‘hard cases’. And
what cases are harder than those of individuals with physical or cognitive
impairments? I share the goal of ascertaining how theories of justice inform
our understanding of, and response to, disability. However, I also believe that
this process must not be unidirectional: simply applying a theory of justice to
cases of disability and seeing what answers it generates. Rather, we should
engage in a process of reflective equilibrium, where our consideration of dis-
ability and impairment informs our approach to justice, as well as the reverse.
Certainly, my goal in a couple of chapters will not be to develop a wholly new
approach to distributive justice. Nor is it to rehearse every move in the extensive
debate on the metric of distributive justice (‘equality of what?’), or the appro-
priate distributive principle (‘equality, or what?’). Rather, my central aim is to
begin to demonstrate the ways in which reflecting on disability can inform our
approach to justice and, in particular, the nature and content of what individuals
should be entitled to as a matter of justice. I argue that considering disability will
lead us to conclude that our distributive entitlements should be conceptualised
as capabilities. In so doing, I defend a version of the capability approach: that is,
an approach that deems individuals to be entitled to the substantive ability and
real freedom to be able to do and be certain things—to be nourished, to be
healthy, to use their imagination and reason, to play, and so on.
The capability approach has been widely used by those considering disabil-
ity in the context of distributive justice, and for good reason.¹ Yet, in this

¹ E.g. Barclay 2010; 2018; Mitra 2017; Nussbaum 2006a; Terzi 2008; Wolff 2009a; 2009b; Wolff and
de-Shalit 2007; Wasserman 2006.

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0004
70      

context, the capability approach is often interpreted as an account of

well-being, rather than of what individuals are entitled to—where the latter
may be narrower than the former. This then implies that individuals should be
provided with any opportunities that contribute to their well-being, and sees
disadvantage in any restriction of these capabilities. This leaves the capability
approach open to the critique that it is over-demanding and will seek to rectify
all inequalities:² for example, compensating a tone-deaf individual for their
inability to appreciate music. This criticism can be avoided if the capability
approach is understood, instead, as an account of what individuals are entitled
to, where this need not be identical to what would increase their well-being.
Thus, I begin by outlining what it means to develop a capability theory of
justice in particular—as opposed to a more general capability approach, or a
capability theory for some other context (§3.2). I then outline two key features
that I argue a capability theory of justice—though not necessarily a capability
theory of well-being—should possess: a focus on capabilities rather than
functionings (§3.3) and a concern for particular, valuable capabilities (§3.4).
Next, I will discuss the benefits of the capability approach over resourcist
alternatives in the context of disability, and in particular its focus on what
individuals are actually able to do and be, given their particular features and
resources needs (§3.5), before responding to resourcists’ objection that the
capability approach is overly demanding of the well-endowed (§3.6). Finally,
I consider whether more complex resourcist accounts are able to accommo-
date diversity equally well as capability theories and, indeed, whether the two
sorts of approaches may simply collapse together. I conclude that they cannot,
and do not (§3.7). Having sketched some of the reasons for adopting a
capability theory, and outlined its core features, in the next chapter I will
consider in greater detail what it means for an individual to have a capability,
and why this is valuable.

3.2 A Capability Theory of Justice

The capability approach is concerned with the kinds of life individuals

are able lead: with their substantive opportunities (or ‘capabilities’) to perform
certain functionings (or ‘doings and beings’). A capability, then, is the real
freedom for individuals to achieve a certain outcome if they so choose.
Functionings involve either pursuing certain courses of action (‘doings’)—such

² See Pogge 2002; §3.6.

    71

as travelling, voting, caring for a child, engaging in leisure activities, forming

friendships, owning property, or accessing the labour market—or achieving
certain states (‘beings’)—for example, being healthy, sheltered, literate, well-
nourished, or part of a developed social network.
Functionings are ‘constitutive of human life’ in the sense that a range of
functionings make a human life both a life (not the existence of an inani-
mate object), and a distinctively human life (not that of a plant or animal).³
As we will see, this need not mean that the approach excludes or ignores the
lives or functionings of disabled individuals as not being fully human.
Whilst this may be a consequence of some capability theories, whether
this is so will depend on the functionings selected and the reason for
selection, and not the mere concept of human functionings. For example,
Nussbaum selects ten central capabilities on the basis that the ability to
perform certain functionings is essential to a decent human life.⁴ In earlier
iterations of her theory, at least,⁵ this required the performance of specific
functionings, such as using all five senses, achieving sexual satisfaction,
exercising complex reasoning, and employing developed social skills.⁶
This approach is open to the objection that it will be exclusionary to those
unable to perform these specific functionings who will be deemed incapable
of living a fully decent life until the impairment that prevents them from
functioning is removed or mitigated.⁷ Further, such specificity is implaus-
ible. A decent life surely does not require, for example, the use of every
sense. More plausibly, it might require access to, and control over, sensory
experience (as §4.4 will discuss). These problems, then, are not endemic to
any version of the capability approach, and the general claims—that human
life entails performing some human functionings, and that individuals are
entitled to control their functioning in certain important domains—need
not be exclusionary.
This is one example of why it is so important to distinguish the general
principles of the capability approach from their specification and application
in various capability theories, which cover diverse forms and fields of inves-
tigation. As Ingrid Robeyns has powerfully demonstrated, the capability
approach is ‘a flexible and multipurpose framework, rather than a precise
theory’,⁸ though this framework can be used to form the basis of many specific
capability theories. My goal here is not to provide an account or defence of the

³ Robeyns 2017: 39. ⁴ E.g. Nussbaum 1992; 2000a: 34–96; 2002a; 2011a: 17–45, 69–112.
⁵ Wasserman 2006: 220, 228–9. ⁶ See Begon 2017.
⁷ E.g. Barnes 2009a; Begon 2017; Terlazzo 2014, 2019. ⁸ Robeyns 2017: 24.
72      

capability approach as a framework, but to defend the use of a specific iteration

of the approach as the basis for a partial theory of justice. The capability theory
I will defend is not intended to be applicable in all other areas. For example, in
a capability theory of individual welfare the full spectrum of possible function-
ings would likely be relevant, whilst in a capability theory of distributive justice
(aiming to identify individuals’ entitlements) the focus may justifiably be
restricted to capabilities the state has an obligation to secure. Similarly, whilst
an accurate theory of well-being should take account both of what individuals
can do (their capabilities) and what they actually do (their achieved function-
ings), justice might only be concerned with individuals’ opportunities, and not
with whether they choose to exercise them in ways inimical to their well-being.
Both these points are central to my capability theory of justice, and will be
explored in §3.4 and §3.3, respectively.

3.3 Capabilities or Functionings?

Beginning, then, with the latter point: despite its name, there is much debate
regarding whether capability theories should promote individuals’ capabilities
to function, their actual achievement of functionings, or some combination of
the two. The distinctive focus of the capability approach is ‘on the state of the
person, distinguishing it both from the commodities that help generate that
state, and from the utilities generated by the state’.⁹ Crucially, then, the
capability approach draws attention to the fact that it is unsatisfactory to
focus either on the bundle of resources individuals have access to, without
taking account of what they are able to actually do or achieve with these
resources, or on their level of subjective welfare given how much this may
be influenced by potentially unjust background conditions, contextual
factors, and (potentially malformed) individual expectations. For example, a
pregnant woman and a man may have access to an identical amount of food.
However, given that the woman currently needs a greater caloric intake, she
may be less well-nourished. Further, if she does not expect to be adequately
nourished, and is used to doing without, she may consider her preferences
better satisfied or experience higher levels of contentment or happiness.
Nonetheless, despite her resource equality and higher welfare, it is surely the
woman who lacks something important in this case, and whom we might
consider entitled to access more food to ensure she has adequate nourishment.

⁹ Sen 1993: 43.

    73

However, determining that it is the state of the person that matters rather
than their resources or welfare does not settle whether we should focus on
capabilities (to be nourished, in this case) or functionings (actual nourish-
ment). The state of a person incorporates not only the activities and states of
being they have achieved but also those they are able to achieve. Thus, the
appropriate focus remains an open question, and as with so many questions
about the capability approach, the answer will depend on the context and
application of the theory.
Both of the capability approach’s primary proponents, Nussbaum and Sen,
seem to allow for the value of both opportunities and outcomes.¹⁰ On the one
hand, accepting the value of particular achieved outcomes (functionings)
prompts the worry that this may result in the paternalistic imposition of a
specific conception of the good.¹¹ Yet, on the other hand, a pure capabilities
focus has been criticised by those who insist that, if flourishing matters, we
should be concerned with outcomes—with the kinds of lives individuals
actually lead—and not merely with their opportunities.¹² After all, an individ-
ual’s quality of life depends on how their lives actually go, and not just on what
they are able to achieve through the expression of their own agency. However,
we might respond that ensuring individuals have a flourishing life should not
be our sole concern in the context of distributive justice. On the contrary, we
may instead believe that individuals should also be entitled to opportunities—
regardless of how many resources this requires, irrespective of the degree to
which this improves their welfare, and independent of ‘whether or not [they]
choose to make use of that opportunity’.¹³
The claim that justice should be concerned with individuals’ opportunities,
rather than their achievements, will be returned to multiple times in the
remainder of this chapter (and Chapters 4 and 7). However, it is worth
providing some motivation for this contention from the outset. The most
famous example used to make this point is that of a fasting and a starving
individual. Both have the same level of functioning (malnourishment), but
different capability sets: the former’s includes the opportunity to achieve
adequate nourishment, whilst the latter’s does not. The focus on opportunities
captures the intuition that only the starving person’s lack of functioning is
unjust, because they lack the capability to be nourished. There is no injustice

¹⁰ E.g. Nussbaum 2000a: 91; 2011b: 25; Sen 1993: 43–4.

¹¹ E.g. Nelson 2008; Carter 2014; Deneulin 2002; Claassen 2014a; Jaggar 2006.
¹² E.g. Richardson 2000; Arneson 2000a; Fleurbaey 2006; Cohen 1993, 1994; Vallentyne 2005, 2006.
¹³ Sen 2009: 235.
74      

when individuals voluntarily forgo an option, even if this makes their life go
less well than it might.
A more topical, but closely analogous example is the phenomenon of
‘poverty appropriation’ or ‘poverty tourism’, wherein wealthy individuals
choose to adopt the trappings of working-class or lower income lifestyles.
For example, high-income individuals who choose to downsize into a small
mobile home, as per the Tiny House Movement,¹⁴ or who frequent bars and
restaurants that attempt to emulate (and perhaps mock) traditionally low-
income cuisine:

Perhaps the most egregious example is San Francisco’s Butter Bar, a trendy
outpost that prides itself on being a true-blue, trailer park-themed bar,
serving up the best in ‘trashy’ cuisine and cocktails. With tater tots, micro-
waved food, and deep-fried Twinkies on the menu, the bar also serves
cocktails that contain cheap ingredients, such as Welch’s grape soda. The
bar has an actual trailer inside, and serves cans in paper bags, so that bar flies
can have a paid-for experience of being what the owners of this bar think of
when they think of trailer trash.¹⁵

Interestingly, the author of this article identifies the root of her discomfort
about this kind of poverty tourism very explicitly in the issue of choice.
Individuals who decide (perhaps temporarily) to live in the woods in a tiny
home, to live as a ‘freegan’, surviving off food salvaged from supermarket
dustbins, or to live off state support in an anarchist commune, are very
explicitly choosing to live this lifestyle (and, at any point, could make a
different choice), and this transforms the nature of the activity. An individual
in a ‘tiny home’ and a trailer may have very similar functionings, but their
options are very different: one is there by choice and could choose to leave; the
other may have no alternatives.
Of course, these examples bring additional complications. Most notably,
issues surrounding social status and cultural capital. Often, individuals who
choose the ‘simple life’ belong to privileged social groups (white, educated,
non-disabled) and, as such, do not face the stigma and discrimination for
choosing an alternative lifestyle that those in other groups might face for being
forced to live such a life. Second, whilst these examples show that choice
matters, they are insufficient to demonstrate that only capabilities matter. It
might be better to live off frozen ready-meals as a statement of a subcultural

¹⁴ See thetinylife.com. ¹⁵ Westhale 2015.

    75

identity rather than because that is all you can afford, but if these fail to
provide you with a decently balanced diet it might also matter that you lack the
actual functioning of being adequately nourished, regardless of whether this
results from your choice.
However, it is important to clarify what we mean by saying something
‘matters’. Functioning achievement matters in the sense that an individual will
likely be better off if they have, for example, a more nutritious diet. But it does
not follow that the state should aim to ensure people live well in such ways,
rather than protecting their capacity to do so. We frequently allow individuals
to have, or risk having, very low functioning achievements when we are
assured that they are choosing these outcomes for themselves: for example,
by engaging in dangerous sports, consuming drugs like alcohol and tobacco,
using convenient yet potentially dangerous modes of transportation, or under-
going cosmetic surgeries or procedures.¹⁶ I will return to these points later but,
for now, merely wish to demonstrate the prima facie plausibility of restricting
the domain of justice to providing individuals with the opportunity to direct
their lives and make important decisions for themselves, rather than aiming to
ensure their lives go a certain way.

3.4 Capabilities: The Good, the Bad, and the Trivial

Even if we can agree that a capability theory of justice should promote

capabilities, and not functionings, a further source of disagreement concerns
whether our focus should be restricted to capabilities to perform good or
valuable functionings, rather than any functioning. Whilst most of the
examples of functionings discussed so far are of desirable things to be or do,
functionings can, or course, also be bad: for example, being malnourished
or unhealthy, being able to abuse our partner or drive a gas-guzzling vehicle.
Or, indeed, they can be trivial: for example, being mildly irritated or bored,
being able to wiggle our toes or decide when to scratch our nose. (We might
also disagree on how the various functionings should be sorted into these
categories.) A complete account of how well someone’s life is going would
need to take all these abilities into consideration, but surely these are not all

¹⁶ Of course, the idea of an anti-paternalist state is not without its critics (e.g. Conly 2013; Hanna
2018a; LeGrand and New 2015). However, we need not accept at this stage that all paternalistic
interference is unjustified, only that the state sometimes has reason to prioritise choice. I defend
anti-paternalism in Chapter 7.
76      

entitlements. Does the state really owe us the ability to abuse our partner or,
indeed, to wiggle our toes?
Perhaps the question seems rhetorical, but the answer is not self-evident,
and some theorists contend that all functionings are relevant to an account of
justice. Most notably, Peter Vallentyne argues:

it is a mistake to hold that there are some functionings . . . the opportunities

for which are irrelevant to justice . . . for any functioning, there are some
circumstances in which it would affect the quality of one’s life, and justice is
concerned with the opportunities for a good life . . . no capability that con-
tributes to the value of a person’s capabilities should be excluded.¹⁷

Many functionings may be inessential to a good human life, but could poten-
tially enhance the quality of an individual’s life. Vallentyne gives the example
of ‘being able to sing beautifully’.¹⁸ If something could make a positive
contribution to someone’s life, he says, it should not be excluded from the
scope of justice. Vallentyne’s goal is to argue that a plausible version of the
capability approach is identical to the most plausible ‘equal opportunity for
well-being’ view.¹⁹ However, given that Vallentyne starts by assuming that
‘justice is concerned with the opportunities for a good life’,²⁰ his conclusion
that the two approaches collapse into each other is hardly surprising.
However, it is far from obvious that this is what justice is concerned with.
Justice, of course, is a nebulous concept, and it may be that different theories of
justice will serve in different domains. However, focussing on justice as a
feature of social institutions, and taking the fairly uncontroversial view that
justice demands showing all people equal concern and respect, it is not
unreasonable to think that its appropriate goal is not to promote well-being
by any means. Some capabilities—such as singing beautifully, wiggling our
toes, or abusing others—are beyond the scope of justice, in the sense that we
are not entitled to perform these functionings (at least for their own sakes) and
are not subject to injustice if we cannot perform them (assuming we are not
prevented from functioning by unjust means).²¹ It is plausible that individuals
are entitled to access functionings that have little or no impact on their quality
of life (votes for the politically apathetic, say), and are not entitled to access to
some that might radically improve their well-being (for example, singing
lessons for an aspiring performer). The fact that a functioning might positively

¹⁷ Vallentyne 2006: 82–3. ¹⁸ Vallentyne 2006: 82. ¹⁹ Vallentyne 2005, 2006.

²⁰ Vallentyne 2006: 82. ²¹ These parenthetical caveats will be unpacked below.
    77

influence our well-being is insufficient to demonstrate that we are entitled to

perform it, unless we simply assume that our entitlements are merely a means
to ensure our well-being surpasses some threshold. Certainly, it does not seem
to follow from a commitment to equal concern and respect.²²
Another route to expanding the scope of a capability approach to justice is
to explicitly incorporate individuals’ lack of access to negative functionings
(which I call negative capabilities) alongside their capabilities for positive
functionings (positive capabilities). For example, it might seem that consider-
ing whether individuals are free from pain, not merely whether they have
opportunities for pleasure, might identify important differences in peoples’
lives. In a first case, two individuals might both lack a positive capability—for
example, the capability to be nourished—but differ in their access to a negative
one—for example, the capability not to starve. The situation in which an
individual has this negative, but not positive, capability could occur if the
only diet they have access to is unvaried and unbalanced (they lack the
capability for proper nourishment), but they are not at risk of involuntary
starvation (they have access to sufficient calories). In a second case, we might
distinguish two individuals who both have a positive capability—for example,
to be part of a social network—but, again, differ in whether they also have an
associated negative capability—for example, the capacity to be in social con-
texts without experiencing painful anxiety.
Having the freedom from negative functionings, as well as the freedom to
perform positive ones, is undoubtedly important, yet the capability approach
already has the resources to accommodate these cases: negative capabilities are
already implicit within the approach. Even without embarking on a detailed
exploration of what the ‘substantive freedom’ entailed by having a capability
involves, it is clear that any plausible view requires that we have the ability to
perform functionings in the absence of avoidable pain, distress, or other
negative functionings.²³ Thus, we simply do not have a positive capability if
we do not also have the capability to be free of associated negative capabilities.
As such, in the second of the above cases the person with anxiety does not have
the capability to form social networks: they may be able, in some sense, to
achieve this outcome but the costs are too high to say they have the real
freedom to do so. Similarly, someone does not have the capability to exit a
space station if all that awaits them is a quick death.²⁴

²² See Dworkin (2000: 11–64; 2011: 352–63) for extensive argument on this point.
²³ The conditions of having the real freedom entailed by a capability are discussed in §4.2.
²⁴ See e.g. Olsaretti (1998); Goodin and Jackon (2007); Taylor (1979); MacCallum (1967); Pettit
(1997); Carter (1999).
78      

In the former of the above cases, neither individual has the capability to be
well-nourished, though one lacks the capability to be even minimally
adequately nourished. Little hangs on whether these cases are phrased in
terms of access to positive functionings (minimally adequate nourishment)
or freedom from negative ones (starvation), though in my view the discussion
is less convoluted when we focus on whether individuals have positive cap-
abilities, and simply ensure the threshold of positive capabilities are set at an
adequate level. The deeper worry here might be that any approach that focuses
on thresholds will fail to take account of the degree to which individuals fall
below it. This worry is distinct from the question of including negative
capabilities, and will be discussed below (§4.6).
It might be argued that negative functionings should not be included as
items we should lack access to (negative capabilities), but as items individuals
have a claim to be enabled to perform, despite their being negative in the sense
that they harm others or themselves. These are not cases that Vallentyne has in
mind, but are surely a consequence of including every capability. It would
certainly be unfair to interpret Vallentyne as suggesting that an abusive
partner might have the right to continue their abuse if it has a sufficiently
positive impact on their well-being. Nonetheless, it seems that the contribu-
tion to the abuser’s quality of life provides some reason for this to be provided
as a matter of justice, even if this is decisively outweighed by the negative
impact on their victim, and even this concession seems deeply implausible.²⁵
One response would be to simply bite the bullet and allow that negative
capabilities have value when they increase well-being, yet insist that this
would never in fact result in any all-things-considered entitlements to harm
others. Alternatively, if a more objective account of well-being is used, then
some behaviours and actions will be deemed to be bad for people regardless of
the value they attach to them. This raises the more general concern regarding
how we determine which functionings are ‘trivial’ or ‘bad’, and how certain
definitions may be used to ignore or diminish abilities that play a hugely
important role in many lives—including disabled individuals’—and place
these beyond the scope of justice. I defer discussion of this to §4.6.
I will finish this section by returning to the two caveats made above to the
claim that granting the ability to perform every trivial or negative functioning
is beyond the scope of justice. These were that (a) we are not entitled to them
for their own sakes, and (b) we are not denied these capabilities by unjust

²⁵ Also see Claassen and Düwell (2013) who criticise Nussbaum’s approach on the basis that she
does not have a clear explanation for why such negative capabilities should be excluded.
    79

means. First, then, it is important to emphasise that, although there are some
functionings we are not entitled to perform for their own sake (being able to
sing beautifully or wiggle our toes, perhaps), few functionings are universally,
context-independently trivial. This is because, as Chapter 2 argued, giving
individuals a decent life (or a dignified life, or the ability to participate as a
democratic equal) does not necessitate the performance of specific function-
ings. For example, walking, specifically, is inessential for a decent life, dignity,
and democratic equality, even if some form of mobility is required. However, if
an individual with extensive paralysis can be mobile only with a motorised
wheelchair that they control by wiggling their toes, then this specific function-
ing might be one they are entitled to. Thus, an ordinarily trivial functioning
(toe wiggling, in this example) may become an entitlement if it essential to a
general capability (mobility).
To be clear, the reason for its inclusion is not for the contribution it might
make to an individual’s well-being, but that it is the only way they can have
control in a domain that does fall within the scope of justice. Contrast this case
with an individual who nurses the hope of becoming a professional ballet
dancer, yet lacks the capability to achieve this goal as the result of their
inflexible toes. Assuming good alternative careers are available to them, their
failure to achieve their specific goal may not ground a claim of redistributive
justice. Similarly, the capability to sing beautifully would not be a concern of
justice if it were merely a means to pursue the private goal of being more
popular or famous, but it might be if singing were (hypothetically) the
dominant form of communication and we could not make ourselves under-
stood in our atonal singing voice. We might believe that both individuals’ well-
being (their quality of life) and their (dis)advantage (their access to entitle-
ments) should be measured in terms of capabilities, but it does not follow from
this that those capabilities we are entitled to are identical to those that will
increase our well-being.²⁶
The second caveat—that individuals should not be denied opportunities by
unjust means—emphasises that, even if we are not entitled to be able to
perform bad or trivial functionings, this does not imply that no injustice is
ever done when we are prevented from performing them. Thus, we can say
both that we are not owed compensation for our inflexible toes, and that we
would be wronged if someone physically prevented us from wiggling them by
injecting us with a paralytic. This is especially important to emphasise in the
context of disability given that, as we have seen, many of the barriers that

²⁶ Contra Vallentyne 2005, 2006; and arguably Cohen 1993 1994 (see Begon 2016a for discussion).
80      

restrict the capabilities of individuals with impairments are not neither

unavoidable nor natural, as is so often assumed. As well as the obvious
injustice of direct discrimination and exclusion, failing to take reasonable
measures to mitigate an exclusionary status quo may also be unjust (see §2.4).
This conclusion does not require the assumption that individuals are
universally entitled to all opportunities they are excluded from. This is
important since, as §2.6 mentioned, and §4.4 will argue, I contend that
individuals should be entitled to general and multiply realisable opportunities
rather than particular functionings. Thus, there will be many functionings
individuals are not entitled to as a matter of justice on my view—and
not merely trivial functionings, but many that are widely considered deeply
significant, such as walking, talking, and hearing. Nonetheless, if we accept
that reasonable accommodations should be made to include those whom a
system is biased against, then the means by which someone is prevented from,
or enabled to, function may still be unjust. For example, as Anderson famously
noted, it would far more disrespectful to deny individuals with impairments
access to public schools than to amusement parks.²⁷ Yet many of the reasons
individuals with physical and cognitive impairments are unable to access
amusement parks arise from features of them that are designed to accommo-
date only the able-bodied.
As an example, consider the issue of whether queues for rides should be
actual (people turn up and stand in line) or virtual (people turn up, take a
ticket, and return at a designated time). Whilst the former approach is more
common it can prove exclusionary to individuals with various impairments:
individuals with fatigue or pain conditions may be unable to queue for a long
time, especially if this requires standing or being in the sun, and individuals
with autistic spectrum conditions might not understand why they are being
made to wait, or find the process of queuing unbearably frustrating.²⁸ In some
cases, universal access may be impossible or prohibitively expensive, in which
case the failure to go to any lengths to provide it may not be unjust. If it is
simply unfeasible to make exhilarating rollercoasters safe for individuals with
heart conditions, or to secure individuals with severe paralysis into the seats of
some rides, then these excluded individuals are not wronged, since this does
not indicate unfair bias and they are not entitled to the particular functioning.
But failing to mitigate the effects of arbitrary and exclusionary norms and
infrastructure can be unjust, and this is so whether we are being denied access

²⁷ Anderson 1999: 332. ²⁸ Thanks to Kevin Todd Mintz for the latter example.
    81

to public schools or to Disneyworld—though, of course, we may have add-

itional reasons of justice to secure access in the former case.²⁹

3.5 Against Resourcism: Means, Ends,

and Conversion Factors

It is all very well to say that if we are to take a capability theory of justice it
should aim to provide individuals with opportunities (rather than outcomes)
in particular parts of their life (rather than expanding the scope of entitlement
to include any functioning that might contribute to an individual’s well-
being). Yet more remains to be said regarding why we should adopt the
capability approach over alternative accounts of distributive justice, and over
resourcist approaches, in particular. The core difference between capability
and resourcist approaches hangs on the degree to which ‘their principles of
justice are sensitive to internal individual differences, and environmental
features and social norms that interact with these differences’.³⁰ Resourcists
focus on providing individuals with means to pursuing ends: that is, with a
bundle of resources, broadly or narrowly construed.³¹ If all individuals were
equally capable of making effective use of these goods, or converting them
into valuable opportunities and outcomes, then resourcism would not differ
from the capability approach: equality of resources would be identical to
equality of capabilities. However, individuals obviously differ in their needs,
abilities, and talents and, further, occupy different social, political, and
environmental contexts. As such, focusing on the resources individuals
have access to will be a poor proxy for determining the kinds of lives they
are actually able to lead.
Sen has called the ‘personal characteristics that govern the conversion of
primary goods into the person’s ability to promote her ends’, their ‘conversion
capacities’.³² However, this focus on personal characteristics or internal fea-
tures of individuals is too narrow, and ignores the influence of external social
and environmental features. As such, the language of ‘conversion factors’ may

²⁹ I am not, therefore, suggesting that social justice only requires that we cease discriminating against
disabled individuals. Barclay (2010) has powerfully criticised Anderson and Pogge for taking this
approach. I am making the much less controversial point that part of what justice requires is the
mitigation of avoidable discriminatory exclusion.
³⁰ Anderson 2010a: 87.
³¹ See e.g. Brighouse and Robeyns 2010; Daniels 1990; Dworkin 2011: 356–63; 2000: 65–119; Pogge
2002; Rawls 1999.
³² Sen 1999a: 74. Also see Sen 1992: 31–8; 1990; 1999a: 35–53.
82      

be more appropriate.³³ As Chapter 1 discussed in the context of identifying

causes of disablement, these might include: (a) services, systems, policies, and
institutions; (b) norms and attitudes; (c) the natural and material environ-
ment; (d) products, technology, and resources; and (e) support networks,
relationships, and community.
Individuals with socially salient impairments are the go-to example of those
who need additional resources to achieve equal opportunities and outcomes.
Yet they are not the only ones whose opportunities are influenced by their
personal features (like age, education, and social background), or by the
policies and norms in place, or features of the physical environment, their
access to resources, the level of medical and technological development, or the
support networks they can draw on. For example, someone with a high
metabolism will require more food than average to be nourished. Or, in a
social context in which norms and biases undermine women’s ability to
pursue a career in engineering, perhaps due to discriminatory hiring practices
or by undermining women’s confidence in their ability to pursue this profes-
sion, they may require additional resources to overcome such barriers. Or, in
an environment in which pollution is rife, people may need additional
resources to be healthy, such as masks or air filters, or treatments for respira-
tory issues that are caused or exacerbated by low-quality air. For capability
theorists, what matters is what individuals are actually able to do given the
conversion factors they face. The capability approach is, therefore, well-placed
to accommodate the multiple factors that interact to cause both disability and
many other forms of disadvantage.
The disagreement between capability theorists and resourcists concerns
whether we should focus on providing individuals with means (resources) or
ends (the capabilities these resources enable), and not on what the relevant
means or ends are. Capability theorists agree that resources are amongst the
means to ensuring individuals have what they are entitled to, though not
the sole means. Resourcists, meanwhile, could hardly attempt to identify the
value of resources independently of their usefulness in enabling individuals to
function, and, ultimately, the opportunity to form and pursue their own
conception of the good. As arch-resourcist Thomas Pogge says, ‘what ultim-
ately matters is that opportunities to pursue human goals are distributed
equitably in some sense’.³⁴
Yet it is sometimes objected that capability theorists also supply individuals
with means, and so may be thought to face the same problem they have posed

³³ Crocker and Robeyns 2010: 68; Robeyns 2017: 45–7. ³⁴ Pogge 2002: 191.
    83

for resourcists: they ensure individuals have access to certain standard

opportunities, but they do not ensure all individuals are equally able to achieve
their specific goals, which is surely what each of us ultimately cares about.³⁵
The opportunity for mobility, for example, is valuable only insofar as it allows
individuals to, more specifically, achieve the goals of taking their children to
school, visiting their friends, going out to the countryside, going shopping for
groceries, and so on.
There are, however, good reasons for capability theorists to focus on general
rather than specific capabilities. First, it may often be either self-defeating or
infeasible to promote individuals’ ultimate ends as a direct policy goal. The
value of many of our goals derives from the fact that we set and pursue them
for ourselves, free from the interference and encouragement of the state.
Pursuing leisure, forming relationships, or exercising imagination would not
have the same value if we were externally directed or nudged into doing these
things. Second, accepting that resources and capabilities have value insofar as
they enable individuals to pursue their particular goals and projects does not
commit us to the view that justice demands that we maximise people’s ability
to do so. We may wish to accommodate other concerns, such as the value of
individual autonomy and the limits of legitimate government power. Finally,
some capabilities may have value as ends in themselves, even if they are not
used by individuals and so do not improve their individual quality of life: the
opportunity to vote for the politically apathetic, for example.³⁶
For these reasons, capability theories of justice understand the goal of just
public policy to be to provide individuals with opportunities, which they can
then utilise as they wish. As Robeyns says: ‘the end of policy making and
institutional design is to provide people with general capabilities, whereas the
ends of persons are more specific capabilities’.³⁷ Focusing on providing indi-
viduals with general opportunities and allowing them to control their pursuit
of their ultimate ends is, therefore, not problematic in the same way as
providing resources without taking account of the (general or specific) ends
this actually allows individuals to pursue. Capability theorists might not
enable individuals to have identical opportunities or to equally satisfy their
personal ambitions (depending on their goals, ambitions, and how they
choose to utilise their general capabilities), but resourcists fail to ensure that
some individuals have the opportunity to pursue any goals if their personal,

³⁵ E.g. see Pogge 2002: 223 nn.78, 83; Carter 2014.

³⁶ These points will be discussed further in the next chapter, and in Chapter 7.
³⁷ Robeyns 2017: 49.
84      

social, and environmental conversion factors mean they cannot convert their
resource entitlement into any actual functionings.

3.6 Resourcism: The Less Demanding Alternative?

Given that Sen (and many subsequent authors) have brought into clear relief
the disconnect that will exist between resource shares and opportunities for as
long as individuals differ in their capacities and environment, we might
wonder what reason there is to remain committed to resourcist accounts.³⁸
There are two broad forms of response that resourcists might make. The first
pulls in exactly the opposite direction to the preceding worry about the
capability approach’s failure to equally enable individuals to pursue their
specific ends. Instead, some resourcists argue that even ensuring individuals
have equal access to more general capabilities will simply prove to be too
demanding. Pogge, for example, contends that, whilst there may be some
intuitive appeal to ‘the feeling it is not fair that some human beings are less
healthy, bright, tall, handsome, athletic, or cheerful than others or just need
more food, water, or clothing’,³⁹ it simply is not the role of the state to mitigate
this kind of unfairness. A resourcist would accept that their approach may
not provide individuals with identical opportunities, but argue that an
approach that did so would be far too demanding. In other words, they simply

³⁸ A further worry that might be raised against resourcist approaches is that not only is the
compensation provided inadequate, it also will not be provided to all those who need it. This is
because, as Pogge notes, ‘on a resourcist view, the causal origins of special needs and disabilities are
morally significant’ (Pogge 2002: 22). Pogge contends that all individuals should be entitled to a
standard bundle of resources, and will be compensated for any unjust deprivation they have suffered.
Thus, if an individual has additional needs as the result of past injustice or a currently unjust
institutional order then they will be entitled to compensation to mitigate this, but they will not be
entitled to assistance for non-standard needs that are not so caused. The Dworkinian version of
resourcism is slightly more permissive insofar as it would also compensate individuals for impairments
(or any additional needs) produced by bad brute luck, such as a childhood accident or a genetic
predisposition. However, neither approach would provide assistance if additional needs were a
consequence of an individual’s ambitions or tastes: paraplegia caused by a bungee jumping injury,
say. The consequences of providing only for standard needs or rigidly holding individuals responsible
for the consequences of their choices might seem unpalatable (see Anderson 1999, for a powerful
rejection of luck egalitarian approaches). Nonetheless, if we accept either of these resourcist views then
we would have to conclude that this is simply all that justice requires. However, such commitments are
not essential to resourcism, and could conceivably be incorporated into a version of the capability
approach, though I know of no approach that has done so. Capability theorists necessarily hold
individuals responsible for inequalities in non-central capabilities and functioning outcomes, but a
more responsibility-sensitive capability theory could be devised—perhaps providing opportunities only
to those who don’t bear responsibility for their inability. Given this debate is somewhat orthogonal to
the choice of a metric, I will not discuss it further here.
³⁹ Pogge 2002: 208.
    85

bite the bullet, but insist that the problems faced by alternative accounts are
more serious.
Whether this criticism has any purchase will depend on the specific iter-
ation of the capability approach and, in particular, on which capabilities or
functionings are selected as valuable and how they are specified. If a capability
theory of justice argued that individuals should be equal in their capability to
achieve every human functioning—including their ability to be bright, tall,
handsome, athletic, and cheerful—this may indeed be implausibly overde-
manding. However, the only theorists who take this line are those who argue
for the collapse of the capability approach into equal opportunity for welfare
(notably, Vallentyne and Arneson). I have already considered some problems
with this approach (§3.4; also §4.6 and §4.7) and, indeed, it is not the view
of any capability theorist who maintains the distinctiveness of the approach
(as neither resourcist nor welfarist).
There are various influential accounts of how we should identify the
capabilities all individuals are entitled to. Anderson argues we should have
the capabilities necessary to function as equal citizens in a democratic society;
Rutger Claassen suggests we should have the capabilities necessary to express
autonomous agency; and Nussbaum focuses on the capabilities necessary to
live a decent (or, perhaps, dignified) human life.⁴⁰ Any of these approaches
would exclude a number of capabilities as being beyond their scope. For
example, Anderson discusses an example of a woman with lung cancer, who
would be entitled to medical treatment, but

would not be entitled to compensation for the loss of enjoyment of life

brought about by her confinement in the hospital and reduced lung capacity,
for the dread she feels upon contemplating her mortality, or for the reproach
of her relatives who disapprove of her lifestyle.⁴¹

Capability equality, then, need not (and usually does not) mean equality of
every capability.
However, it might seem that all of these accounts could nevertheless be
interpreted to include some version of the opportunities Pogge mentions:
perhaps individuals will be entitled to some degree of intelligence, enough
height to access various social spaces, a lack of any disfigurement that would
lead to ostracism, enough ‘athleticism’ to have independent control of mobility

⁴⁰ Anderson 1999, 2010a; Claassen 2018, 2017; Nussbaum 2011a, 2006a, 2000a.
⁴¹ Anderson 1999: 327.
86      

and so on. However, as these qualifications show, this does not suggest
an entitlement to opportunities to be equal in intelligence, height, beauty,
athleticism, and temperament. Capabilities should be specified by appealing
to the principle that is used to identify them, and individuals’ capability
for democratic equality, expressing autonomous agency, or leading a decent or
dignified life certainly does not require the equal provision of these capabilities.⁴²
A much lower threshold is clearly sufficient. This problem—of advocating
unachievable and perhaps undesirable equality of specific functionings—is also
avoided on my own account. According to this, capabilities are understood as the
ability to exert control in certain domains of our lives rather than the capacity to
perform specific functionings. Never mind attempting to ensure equal athleti-
cism, then, people need not even be equal in their ability to walk.

3.7 Beyond Simple Resourcism

A second resourcist response to their failure to take account of differences in

individuals’ ability to convert resources into actual opportunities involves
reconceptualising what resources constitute. At the most simplistic level,
‘resources’ can be taken to mean income or wealth. A slightly more developed
version might include material goods of any sort. This would permit the
inclusion of public goods, such as access to public spaces or public transport,
which could not be readily converted into a monetary value, or seen as solely
in the possession of a single individual. A yet more nuanced account might
also include rights and freedoms. Rawls’s list of primary goods is, of course, the
archetypical such account. Primary goods are ‘things citizens need as free and
equal persons living a complete life’, and include: basic rights and liberties;
freedom of movement, and free choice amongst a diverse variety of occupa-
tions; powers of offices and positions of authority and responsibility; income
and wealth, understood as all-purpose means to achieve a variety of ends; and
the social bases of self-respect, understood as the recognition by social insti-
tutions that gives citizens a sense of self-worth and the confidence to advance
their ends.⁴³ However, all such accounts focus on ‘things’ that people ought to
have access to, and hence fail to take account of specific features of persons or

⁴² Sen argues that the capabilities should be determined through a process of public reasoning and
deliberation (2009, 1999a). The content is thus indeterminate, but it seems reasonable to suggest that it
would also exclude equality of such capabilities, given the infeasibility of providing them if nothing else.
⁴³ Rawls 2001: 58–9.
    87

their circumstances, and the influence this may have on their ability to convert
these things into actual functioning achievements.⁴⁴
The internal resourcist critique of these accounts diagnoses the problem as
being with the kinds of resources focused on. Specifically, identifying the
worst-off solely on the basis of their possession of social primary goods (like
income and wealth, opportunities and powers, or rights and liberties, which
are distributed directly by social institutions) at the expense of calculating their
share of natural primary goods (like health, intelligence, imagination, and
talents, which are affected by social institutions but not directly distributed by
them).⁴⁵ Dworkin’s is undoubtedly the most well-developed resourcist
account that takes seriously the importance of both social and natural primary
goods—or, in Dworkin’s terminology, personal and impersonal resources.⁴⁶
Centrally Dworkin argues that ‘a political community should aim to erase or
mitigate difference between people in their personal resources’, that is, their
‘health, strength, and talent’.⁴⁷ This is achieved through providing those who
are deficient in personal resources, such as ‘people who are physically handi-
capped or otherwise unable to earn a satisfactory income’, with additional
impersonal resources.⁴⁸
Although Dworkin concludes that the capability approach (or ‘Sen’s theory’)
is merely ‘equality of resources in a different vocabulary’,⁴⁹ this is not the case.
Dworkin’s view aims to neutralise if possible, and mitigate if not, the effects of
the ‘deficits’ in some individuals’ internal resource share. As such, this
approach is too focused on individuals themselves, at the expense of their
social and material environment, and takes this context as given.⁵⁰ This gives
rise to three interrelated problems.
First, it fails to take account of the effect of context on individuals’ ability to
function and the value of their resource share. Thus, it fails to identify an
important vector of disadvantage. Second, it identifies the ‘problem’ of indi-
viduals who are less healthy, strong, or talented, as rooted in their internal

⁴⁴ Some of the primary goods—notably the social bases of self-respect and free choice amongst
occupations—do not seem much like ‘things’ that we either have or lack, but capabilities that we can be
enabled to possess by diverse means (depending on our conversion factors). Indeed, many iterations of
the capability approach take themselves to be developments of Rawls’s theory rather than diverging
from it. Nonetheless, Rawls and many Rawlsians wish to maintain that primary goods are distinct from
capabilities, so I will grant that this is so.
⁴⁵ Kymlicka 2002: 64–5, 70–2. ⁴⁶ See Dworkin 2000: 65–119; 2011: 356–63.
⁴⁷ Dworkin 2000: 286.
⁴⁸ Dworkin 2000: 286. Also see Clayton 2000: 69. The other central element of Dworkin’s view, not
relevant to the current discussion, is that individuals should not be compensated for differences in their
personality: ‘for differences traceable to the fact that some people’s tastes and ambitions are expensive
and other people’s cheap, for instance’ (Dworkin 2000: 286). (See n.38.)
⁴⁹ Dworkin 2000: 303. ⁵⁰ See e.g. Terzi 2008: 132–3; Wasserman 2009: 219–20.
88      

deficiencies as opposed to unjust structures. Thus, it reinforces a conception of

individuals with impairments, or who lack marketable talents as inferior,
flawed, or defective, and in need of compensation for their imperfections. As
well as being inaccurate, this reinforces a damaging, yet dominant, conception
of individuals with impairments as ‘substandard’ compared to non-impaired
individuals.
Third, this suggests that the site of change and intervention should be
individuals rather than the environment. To pick a simplistic example: it
suggests that the state should eliminate mobility impairments or distribute
individual mobility equipment rather than making the built environment
more accessible to those who are mobile through non-standard means (for
example, using wheelchairs, crutches, prostheses, electric scooters, and so on).
There may be cases when such individual interventions are the most appro-
priate response—if, for example, it is impossible or prohibitively expensive to
make the environment sufficiently inclusive. However, the automatic focus on
individual impersonal resource compensation misses a significant portion of
the effective ways to respond to impairment and, indeed, to broader instances
of diversity in individual conversion factors. All these problems can be avoided
if we focus on what people can actually do given the combination of their
impersonal resource shares, personal features, and social, political, and mater-
ial environment.

3.8 Conclusion

The goal of this chapter has been to begin to motivate the claim that our
account of distributive justice should be a version of the capability approach,
and to show that the advantages of this approach are most clearly evident in
the context of impairment and disability. In summary, capability theories have
three central advantages over resourcist approaches. First, they can take
account of both internal and external conversion factors, and thus of the
multiple and interrelated factors that cause disability, and other forms of
disadvantage. Second, they can avoid the assumption that having an impair-
ment leaves an individual worse off. We need not figure out whether an
individual has her fair share of personal resources; we need only determine
whether she has the capabilities she ought. This will inevitably be the result of a
wide combination of factors, and lacking the opportunities you are entitled to
does not imply inferiority. Further, there is no reason to think that individuals
with impairments will necessarily or uniquely be amongst those who lack the
    89

relevant capabilities. Third, the approach emphasises how unjust background

conditions can psychologically, as well as physically, prevent individuals from
having access to opportunities. Even if someone is not physically prevented
from achieving some functioning, they may still lack the capability to perform
it due to lowered expectations as a result of adapting to mistreatment or
oppression, the influence of unjust power relations, or discriminatory social
norms.⁵¹ The capability approach can acknowledge that individuals in such
circumstances lack the real freedom that they should have, as I will now discuss.

⁵¹ This is discussed in much greater length in Chapter 6.

 4
Capabilities for Control

4.1 Introduction

I have argued that individuals are disabled when they lack opportunities they
are entitled to as a result of an impairment. The previous chapter has provided
some reasons to think that what all individuals are entitled to is certain central
capabilities. But what does it mean to have a capability? And will the focus on
providing people with valuable opportunities be exclusionary for individuals
with impairments, whose opportunities are often limited? I argue that this is
not so if we reject the standard understanding of capabilities as the freedom to
choose whether or not to perform certain valuable functionings or combin-
ations of functionings. This standard construal of capabilities is unable to
accommodate individuals who do not value, or cannot perform, the function-
ings deemed central: for example, individuals with mobility impairments may
be unable to walk or run, individuals with sensory impairments may be unable
to hear or see, individuals with social impairments may be unable to engage
in some forms of social interactions. They can, however, have more general
opportunities (to be mobile, to have sensory experiences, to form social
relationships), and they can value having control and choice in these parts
of their life.
The lesson we should draw from this is that we should place less emphasis
on identifying valuable functionings and then guaranteeing that everyone has
access to an identical set of such functionings. Instead, we should focus on
ensuring everyone is able to exercise control over various parts of their life—
such as their mobility, communication, or social interactions—whilst avoiding
the paternalistic imposition of a perfectionist ideal in these domains. Thus,
I contend that capabilities should be conceptualised as opportunities to exer-
cise freedom, as meaningful control, in certain domains of our life. By not
specifying centrally important functionings but only domains of choice, this
approach allows individuals to exert genuine control over central parts of their
lives, regardless of the content of their choices.
This approach avoids endorsing a perfectionist account of what constitutes
the ‘right’ way to function or a hierarchy of functioning, and instead

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0005
    91

acknowledges that different ways of functioning can be interchangeable and

equally valuable: that it is not necessarily worse to achieve mobility in a
wheelchair than it is by walking.¹ This is particularly important in the context
of disability, since it shifts the goal from normalising bodies and minds
(promoting identical functionings) to allowing individuals to control their
functioning outcome (their mobility, their health, their social interactions)
without specifying the form this should take (walking, eliminating physical
impairments, engaging in neurotypical social interactions, and so on). Thus,
this approach demonstrates appropriate respect for the lives and choices of
individuals with impairments, and allows us to develop a theory of justice that
accommodates individuals with diverse needs, preferences, and abilities.
I will begin by clarifying what it means to say that someone has the ‘real
freedom’ to pursue functionings (§4.2), before considering exactly what it is
that we should have the real freedom to do. I will outline the problems with the
standard formulation of capabilities as the ability to perform valuable func-
tionings (§4.3), and argue instead that they should be conceptualised as
opportunities to exercise control in certain central domains of our life (§4.4).
I will then defend this account against the objections that it either over- or
under-emphasises freedom (§4.5). I finish by considering some potential
objections. I respond, first, to those who disagree that individuals should
have capabilities to control certain domains of their lives, rather than ensuring
access to actual functionings (§4.6); and second, to those who dispute the
claim that individuals should be entitled to an irreducibly plural list of these
capabilities, even when some of these are not exercised (§4.7). By defending a
more fleshed-out account of what justice demands and what individuals are
entitled to, the content of my claim that individuals are disabled when they
lack these entitlements will become more concrete.

4.2 Having the Capability: From Formal Freedom

to Freedom as Control

What, exactly, does it mean to possess a capability? Does a woman have the
capability to be nourished if her husband takes her share of the food and keeps

¹ This is not intended to imply that we should be concerned only with what individuals can do, and
not with how difficult it is to do these things, as we have seen. It may, then, be worse to achieve mobility
using a wheelchair if its use entails serious fatigue or pain, and our understanding of capabilities should
reflect this. This is another lesson for our theory of distributive justice that can be learned from a
consideration of disability (as §4.2 will discuss).
92      

it for himself, and she does not wish (or feel able) to complain? What if she
must choose between providing adequate nourishment for her children or
herself? What if her access to nourishment depends on the continued existence
of a food bank? What if she does not consume an adequate number of calories
in an effort to conform to a socially prescribed body shape? Whilst there is
much that is powerful about Sen’s distinction between the fasting and the
starving individual, determining which is which—or whose undernourish-
ment is truly by choice—is often more complicated. It is common to insist
that having a capability requires internal and external preconditions so that an
individual can genuinely voluntarily choose how and whether to function, as
I will shortly outline. Whilst I do not dispute this, I contend that this is
insufficient to have the genuine control necessary for a capability.
First, to clarify the importance of the internal and external preconditions of
a capability, it is helpful to utilise Nussbaum’s distinction between three forms
of capability: basic, internal, and combined.² Basic capabilities are individuals’
potential, latent abilities, or capacities: the ‘innate equipment . . . necessary . . .
for developing the more advanced capabilities’.³ Most infants possess the basic
capabilities for speech, practical reason, and fine motor skills, but the cultiva-
tion of these nascent capacities requires education, development, and practice,
as well as the confidence and self-respect to actually exercise these abilities. For
example, for an individual with Down’s syndrome to have the capacity for
practical reasoning (evaluating, choosing, and pursuing goals or ways of life)
they do not only require an appropriate education. In addition, they need to
develop a sense of themselves as the kind of being who can direct their own
lives in this way. If the dominant understanding of individuals with Down’s is
as incapable, incompetent, or helpless, this may affect their ability to perform
functionings they would otherwise be capable of. When individuals possess the
developed psychological traits actually sufficient to perform a functioning,
they have the internal capability to perform it.
Yet even this is insufficient to ensure an individual is actually able to
perform a functioning, since there may be external factors that prevent them
from doing so. The focus is usually on legal prohibitions or engrained social

² See Nussbaum 2000a: 83–6; Nussbaum 2002b: 132–3; Nussbaum 2011a: 20–4.
³ Nussbaum 2000a: 84. Somewhat confusingly, Sen uses the term ‘basic capabilities’ to mean
something different. In his early work it is used interchangeably with ‘capability’ as the metric of
distributive justice. Later, he uses it to refer to a threshold level of a capability: ‘the ability to satisfy
certain elementary and crucially important functionings up to certain levels’ (Sen 1992: 45 n. 19). For
Sen, then, basic capabilities are ‘the freedom to do some basic things considered necessary for survival
and to avoid or escape poverty or other serious deprivations’ (Robeyns and Byskov 2020). For further
clarification of this terminological confusion see Robeyns (2017: 94–6).
    93

norms that restrict opportunities, but we should also include the various other
features of our social, political, and material environment, which can cause
impairments to be disabling (see §1.4). Only when an individual has an
internal capability combined with the external conditions and resources
necessary to exercise their ability do they have a combined capability. This is
what capability theorists are concerned to promote, and is what I will usually
mean when I refer to ‘capabilities’.
Capabilities, then, are substantive or real freedoms, not merely formal
opportunities.⁴ Individuals with impairments do not have the capability to
obtain meaningful employment simply because there is no law preventing
them from entering the job market. To have this capability also requires, inter
alia, education in an appropriate format and at a sufficient level to be qualified
for a range of jobs, an absence of explicit discrimination and implicit bias that
might prevent them from being shortlisted or hired despite being suitably
qualified, and not being subject to social norms and stigmas that might inhibit
individuals from seeking work in the first place.
Yet even this is not enough to have the substantive opportunity to pursue
some chosen option. Ideally, capabilities should also be possessed (i) inde-
pendently of the arbitrary will of another and regardless of their content,
(ii) securely, (iii) with a robust probability that the capability is actually
realisable, and (iv) as part of a consistent capability set that does not require
unreasonable sacrifices or tragic choices. All these four criteria are scalar: our
capabilities can be possessed more or less independently, securely, and
robustly, and our capability sets may be more or less consistent. If an
individual falls below some threshold level of any of these criteria, it may
be that they cannot be meaningfully said to possess a capability at all.
However, more often these are dimensions along which the quality of
individuals’ capabilities will differ.
We have reasons of justice to move individuals up the scale on all these
criteria, with the ultimate goal of ensuring they are able to exercise their
capabilities autonomously and free from undue external pressure. However,
in many cases achieving complete or maximal levels of independence, security,
robustness, and set compatibility will simply be impossible or implausible.
Individual cases will need to be considered on their merits, and in light of what
is possible given technological development, resource scarcity, and competing
demands. I will not, therefore, attempt to precisely fix the thresholds for each
of these criteria, and will instead discuss some illustrative examples.

⁴ E.g. Sen 2009: 227; Nussbaum 2000a: 86.

94      

4.2.1 Independence

First, consider an example Sen gives of an individual who has the capability ‘to
go out of her house whenever she wants and to move around freely’ because
she ‘is always helped by volunteers with goodwill’, and one who is enabled to
do so by servants who ‘obey—and have to obey—her command’.⁵ (To avoid
the ethical implications of employing servants, we can substitute machines
that fulfil a similar role.) The salient distinction is between a person who is
dependent on others’ goodwill for control over her mobility, and someone
who is not. Sen wishes to resist Pettit’s proposal that capabilities must incorp-
orate the requirement that individuals’ freedom is permit-independent.⁶
Hence, Sen acknowledges that there is certainly something valuable about
not having to rely on others, but, he claims, all the capability approach should
aim to capture is whether someone has the ability leave their house, however
this is achieved.⁷
As ever, whether this is satisfactory will depend on the application of our
particular capability theory. If we were devising a theory of well-being, for
example, we might sometimes need only to ascertain whether individuals
can perform certain functionings. However, part of the distinctive appeal of
capability theories of justice is that they aim to provide individuals with the
ability to direct their lives (regardless of the size of the resource bundle
this requires, and independent of how much welfare this generates). If
individuals can only pursue options that coincide with the goodwill of
the more powerful, and only for as long as this goodwill lasts, we may
reasonably question whether they have the substantive freedom that cap-
abilities are supposed to encapsulate. As Cohen puts it, freedom ‘worthy of
the name’ must entail ‘the circumstance of genuine control over one’s life’.⁸
Thus, in the context of justice at least, we should follow Pettit and require
capabilities to entail permit-independence, such that the decisiveness of our
preferences does not depend on the ‘gratuitous favour’, or arbitrary will, of a
third party.⁹

⁵ Sen 2001: 54. See Begon (2016a: 58) for further discussion.
⁶ Pettit 2007, 2010. ⁷ Sen 2001: 52–6.
⁸ Cohen 1995: 102. It is worth making explicit that I do not mean to imply that all forms of
interdependence or vulnerability are bad, or that complete independence should always be prized.
Merely, that in some domains of our life we should be able to make choices and pursue a conception of
the good without being forced to rely on others’ willingness to provide assistance.
⁹ Pettit 2007, 2010; Begon 2016a. (Also see §1.4.) This has also been called context- or favour-
independence, but Pettit (2010: 98–9) prefers the label of permit-independence, since this better
captures the significant feature of our context (the permission of a third party).
    95

An additional sense in which capabilities should be independent, also

emphasised by Pettit, requires individuals’ choices to be decisive regardless
of what they choose. Having this content-independent freedom means indi-
viduals are not only enabled to perform those functionings deemed valuable.
Thus, the person in Sen’s example should not only be free to leave the house
but also free to stay inside all day long if they wish (and not be dragged to the
park by well-meaning volunteers). If the core of the capability approach
concerns what people are ‘really able to do and what kind of person . . . they
[are] able to be’¹⁰ then this is best captured if capabilities—at least as the metric
of distributive justice—are interpreted as providing people with the content-
and permit-independent freedom to choose how they want their lives to go.¹¹
We may be both better-off and freer if we can leave our house than if we
cannot, whatever the conditions under which we can do so, but we do not have
what we are entitled to unless we have the real freedom to control our
mobility.

4.2.2 Security

To truly possess a capability, our ability to perform a functioning should not

be precarious. Living at risk is both harmful and inimical to the possession of
the real freedom to control how our life goes. However well someone’s life
actually goes, they are seriously harmed if they live at risk of losing their access
to central functionings. Even if these risks never materialise, individuals will
live with the fear and anxiety about how their circumstances might change,
and with the difficulty of planning their life under such uncertain conditions,
and are forced to take actions to mitigate the risks they face.¹² Individuals who
are reliant on foodbanks, work on zero-hour contracts, or depend on bureau-
cratic benefit systems under which they may be subject to unpredictable
sanctions or delays in payment all experience this precariousness. Even if
they never actually experience hunger, are never evicted from their home, or
are never unable to pay their electricity bill they cannot be said to fully have the

¹⁰ Robeyns 2017: 9.
¹¹ Note that a commitment to content-independence need not imply that individuals must be
entitled to every possible capability. It is compatible with restricting the scope of justice to certain
capabilities, though meaningfully possessing these capabilities will involve the ability to make choices in
these domains (as §4.4 will discuss).
¹² The significance of risk and its relevance to the capability approach is most clearly articulated in
Wolff and de-Shalit (2007).
96      

capability to achieve these functionings since they lack control over the
achievement of all these outcomes.
Part of the problem with lacking permit-independence is such precarity: we
never know when or if our benefactor’s favour will be withdrawn. However,
we may experience permit-dependence without precarity (for example, a
spouse who is reliant on their partner for their economic security yet (justi-
fiably) does not fear that this support will be withdrawn, though nothing
would prevent it), and precarity without permit-dependence (for example,
farmers whose livelihood is dependent on the climate). Moreover, precarious
access to one functioning will often spread to other functionings: risks to
employment lead to risks to nutrition if we rely on our income to buy food, to
risks to our health if we are stressed and poorly nourished, to risks to our ability
to form and maintain a social network if we cannot afford to socialise or are too
anxious and distracted to engage with our friends, and so on.¹³ Correspondingly,
providing secure access to ‘fertile functionings’ (in Wolff and de-Shalit’s ter-
minology)—functionings that are likely to secure further functionings—will
ensure individuals have other capabilities.¹⁴

4.2.3 Robustness

The focus on capabilities rather than functionings is often justified with

reference to individual responsibility: individuals are entitled to opportunities
and it is up to them what use they make of them. However, this connection
only holds if possessing a capability entails a reasonable chance that an
individual’s attempt to exercise a functioning will succeed.¹⁵ Determining
what probability of success is reasonable will depend on features of particular
cases. In general, though, we cannot claim to have achieved capability equality
if individuals’ probabilities of success differ radically, and cannot claim to have
achieved capability sufficiency if some people’s chance of success is remote.
Further, if the reasons for these discrepancies are arbitrary and avoidable, this
is likely to be an instance of injustice.
For example, all students with the relevant qualification may, in some sense,
have the capability to gain a place at an elite university. However, if, as the

¹³ Wolff and de-Shalit (2007: 70–2) call this ‘cross-category risk’. Disadvantages that carry this risk
of having negative effects on other functionings they call ‘corrosive disadvantage’ (see Wolff and de-
Shalit 2007: 121–2, 133–49).
¹⁴ Wolff and de-Shalit 2007: 133–54.
¹⁵ The label of ‘robustness’ for this criterion is Robeyns’s (2017: 97–8).
    97

result of explicit and implicit biases, the chances of admittance are much
higher for white, straight, non-impaired men, with wealthy parents and a
long family history of attending the institution, we can hardly claim that
justice is done because every candidate has some chance of success. Further,
it would seem disingenuous to say to another candidate—non-white, queer,
impaired, working-class, and without a family history of university attendance—
‘you too have the opportunity to attend, what more can you expect?’. They
certainly have the right to demand more. It would be similarly misleading to
claim that someone with MS has the capability to travel to work on public
transport if every one attempt in five (say) is successful, yet on other occasions
they are overcome by pain or fatigue, a lack of proper assistance, or inappro-
priate accessibility modifications. These individuals can hardly be described as
having a ‘genuine opportunity’ or ‘real freedom’.

4.2.4 Set-consistency

Lastly, having a capability not only requires that our attempt has a robust
chance of success but that the cost of the attempt is not unreasonably high and,
in particular, does not require individuals to sacrifice some capabilities to
achieve other functionings. The goal of a capability approach to justice should
not be to provide access to individual capabilities but to capability sets that can
be consistently exercised without requiring unreasonable sacrifices or tragic
choices. Wolff and de-Shalit consider cases in which the pursuit of some
functionings renders access to other functionings precarious, which they
label ‘inverse cross-category risk’.¹⁶ They consider, in particular, an individual
who undertakes dangerous work to secure access to food and thereby puts
their ability to be healthy (or, indeed, alive) at risk.
This is an important subset of cases, but ‘unacceptable costs’ are not
identical to inverse cross-category risks. In some cases, the cost of per-
forming a functioning may not be the risk of losing a capability but instead
either the actual loss of another ability; or the sacrifice, or risk of sacri-
ficing, another’s capability; or the sacrifice, or risk of sacrificing, the same
capability in the long term; or, though the cost is high, it may not require
the loss of a central capability. For example, imagine a paraplegic individual
can either afford to purchase a wheelchair that will enable her to be
mobile or pay the rent on her apartment, but lacks the capability to achieve

¹⁶ Wolff and de-Shalit 2007: 70.

98      

both mobility and shelter. Alternatively, imagine a parent who must choose
between working so that they have the resources to feed and clothe their
children, and being able to spend time with them, raise them, and form
valuable parent–child relationships. Either choice will jeopardise both their
own capabilities and their children’s. Or, consider the individuals discussed
in §2.7, who are able to perform all relevant functionings but only with
associated pain or fatigue. They cannot be said to have the capability to be
mobile if walking causes excruciating pain or their wheelchair produces great
discomfort.
We might describe individuals in such cases as simply lacking the capability
for these incompatible functionings, or as lacking an important capability set.
I believe there are reasons to adopt the former approach. For example, imagine
that an impaired student applying to university would be admitted only if they
were willing to undergo a procedure to eliminate this impairment. (Or, for
those who think this would be a very good thing for the student and not really
a cost at all, imagine the situation is reversed and admission depends on being
impaired: a student has the capability to go to university, but they will be
required to undergo an amputation (say) first.) We can specify either case such
that the student does not lose access to any other functionings as the result of
losing or gaining an impairment. The cost nonetheless seems far too high to
say that they have a genuine opportunity to enter university. Similarly, we can
say that a person lacks the capability to fly to Australia both if they simply
cannot afford the airfare and if buying the ticket will require clearing out their
life-savings and selling everything of value that they own. A functioning must
be a realistic option for an individual if we are to say they have the capability to
achieve it.
It might be objected that exercising our capabilities often restricts our other
options: pursing one career might foreclose opportunities to pursue others,
deciding to have children might restrict one’s control over one’s leisure time,
deciding not to have children by undergoing sterilisation will restrict one’s
future ability to conceive. Such opportunity loss surely is not unjust. The
difference is that in the troubling cases above the associated costs were
contingent, avoidable, and not consented to. Having control and pursuing a
conception of the good will require that we are able to sacrifice some oppor-
tunities, but these sacrifices should not be imposed on individuals (see §4.6).
Thus, when determining what individuals are entitled to, and whether they
have what they are entitled to, we should take account not only of whether
they can exercise control in some particular domain (independently, securely,
and robustly) but also of whether they can do this without foregoing, against
    99

their wishes, other important areas of control in either their own lives or
others’, or paying some other unreasonably high cost.

4.3 The Problem with Capabilities to Function

Although capability theories differ in many dimensions, all share the termin-
ology of functionings (as doings and beings) and capabilities (as the real
freedom to achieve these functionings). And, whilst they may disagree about
which functionings are valuable and how they identified, they agree that an
account of valuable functionings is needed. Thus: ‘What matters . . . is that
people enjoy certain specific capabilities, not that they enjoy capability as such,
and in order to identify such specific capabilities we shall still need to refer to a
list of independently specified functionings.’¹⁷ It is true that the distinctive
contribution of the capability approach would be lost if it were simply about
providing unspecified freedom (see §4.7), or indeed bad or trivial freedoms
(§3.4). However, the ability to perform independently specified functionings is
not the only way in which our freedom can be specified. Nor, I think, is it the
right one.
My objection to understanding capabilities as abilities to function is not that
by doing so we inevitably promote actual achievements, rather than mere
opportunities, as a requirement of justice in the political realm. As will become
clear, even without directly promoting particular functioning, a commitment
to their value can be exclusionary to those who do not value and/or cannot
perform some of them. I will grant, then, that identifying central, valuable
functionings can be compatible with insisting, as Nussbaum does, that
‘capability to function, not actual functioning, should be the goal of legislation
and public planning’.¹⁸ I will leave aside those places in which she appears to
violate her own liberal commitments and argue that the state may, in fact, have
reason to override individual choice in order to promote actual functioning or
the capacity for future functioning.¹⁹ Instead, I will take her at her word as a
political liberal, who deems it ‘inappropriate for any particular comprehensive
conception of ethical value to be endorsed by politics’.²⁰ Thus, whatever

¹⁷ Carter 2014: 91.

¹⁸ Nussbaum 1992: 221. I focus on Nussbaum’s account here as the most developed liberal capability
theory of justice.
¹⁹ See Begon 2017 for discussion. Also, Claassen 2014a; Claassen and Düwell 2013; Jaggar 2006;
Nelson 2008; Terlazzo 2014; §4.6. For examples of this in Nussbaum’s work, see Nussbaum 2000a 55,
74, 87–93; 2004: 199; 2011a: 25.
²⁰ Nussbaum 2000b: 124.
100      

Nussbaum’s personal views on the superiority of life in which capabilities are

exercised, the plausibility of her view does not depend on everyone sharing a
particular conception of the good. Crucially, Nussbaum seeks an overlapping
consensus on the value of the central capabilities ‘for political purposes only,
not as a comprehensive guide to life’.²¹ Thus, individuals are not required to
endorse the value of the underlying functionings, or pursue them within their
own lives. Indeed, people need not even ‘endorse freedom as a general good’,²²
but merely a short list of freedoms for political purposes.
Given that Nussbaum is so clear that in the political realm we should not
endorse or enforce a comprehensive conception of the good, it might seem
unimportant that she specifies the functionings everyone should be entitled to
perform. After all, individuals still retain the freedom to set and pursue their
own goals. Yet I will argue that, despite Nussbaum’s attempts to make her
approach neutral, the conceptualisation of capabilities as the ability to perform
specific functionings will always threaten this neutrality. This is because,
although individuals need not include every functioning in their particular
conception of the good, Nussbaum does insist that the list could be the object
of an overlapping consensus. Thus, for her, we must all be able to agree that
the capability to perform these functionings is necessary for a decent human
life.²³ However, this will never be true of at least some individuals who lack the
capability to perform some central functionings.
Her claim that the list could be the object of an overlapping consensus does
not require that there is, at present, no disagreement on the content of the list.
Thus, the problem cases are not those in which people choose not to perform a
central functioning, either occasionally (for example, fasting), or more per-
manently (for example, celibacy). Individuals might do this for the sake of
another central functioning: for a celibate monk, a failure to function in one
way (sexual satisfaction) may be necessary for another functioning he values
more highly (spiritual fulfilment). Or, someone might sacrifice central for
non-central functionings: for example, starving oneself to meet some ideal of
beauty, or choosing the pleasures of drug-use at the expense of one’s long-term
health. These individuals do not threaten the possibility of an overlapping

²¹ Nussbaum 2011a: 90. ²² Nussbaum 2006b: 60.

²³ Nussbaum has moved towards talking in terms of the constituents of a dignified human life
(Nussbaum 2006a, 2008), though it is not always clear how this concept should be cashed out (see
Formosa and Mackenzie 2014; Claassen 2014b). My focus is on how capabilities are conceptualised
rather than how they are selected, so I will not attempt to determine whether we ought to talk in terms
of decency or dignity, or exactly how these might differ. The examples I consider have force on any
plausible reading of any of the main candidate criteria used to identify the list of central capabilities
(including Anderson’s or Claassen’s, mentioned above).
    101

consensus on the central capabilities to function because they can agree on the
value of these opportunities for political purposes, even if they have different
priorities in their own lives. In the first set of cases, the very meaning of the
sacrifice requires an acceptance of the value of the functioning that is being
forgone: fasting and celibacy are only meaningful sacrifices because nourish-
ment and sexual satisfaction have value. In the second set of cases, individuals
do not consciously sacrifice a valuable functioning, but merely find that their
other priorities conflict with it. Again, this does not imply that they dispute its
value or importance, and they would likely prefer circumstances in which both
goals could be achieved (nourishment and the requisite level of slimness;
health and drug use).
There are other cases, however, where individuals who choose not to
perform some functioning do not consider themselves to be giving up some-
thing valuable. Possible examples would be misanthropes who believe there is
no value in human association, or people who entirely lack the capacity for
humour.²⁴ However, it can be hard to be sure that these individuals really see
the opportunity to perform these functionings as inessential to a decent life: for
some, it may be part of their self-conception to have the opportunity to form
human relationships, say, and yet choose not to. It would be hard to under-
stand oneself as a misanthrope having simply lacked the chance to interact
with other people, so we may value having the opportunity so we can reject it.
More clear-cut cases, therefore, are those in which individuals are incapable of
performing a functioning, and consider their lives to be fully flourishing (and
decent and dignified) nonetheless. These individuals lack the capability to
function—we cannot have the real freedom to do something impossible—and
so might naturally dispute the inclusion of the opportunity to perform this
supposedly necessary functioning on the list of central capabilities.
For example, consider asexuality, wherein individuals do not experience
sexual attraction to others. Asexuality is a diverse category, but at least some
asexual individuals will not experience any sexual desire. Such individuals,
whilst typically capable of performing sexual acts, cannot be said to derive
sexual satisfaction from them.²⁵ They might choose to engage in sexual activity
to achieve some other end (such as the satisfaction of another’s sexual or
emotional needs), but they cannot choose to perform the functioning of sexual

²⁴ Nelson 2008: 99; Claassen and Düwell 2013: 496–7.

²⁵ Brunning and McKeever (2021) argue that it is possible to experience sexual desire without sexual
attraction. I am happy to accept this point and merely claim (as Brunning and McKeever allow) that
sexual desire will not be experienced by some asexual persons. (Also see AVEN 2020; Decker 2015;
Bogaert 2015.)
102      

satisfaction, since they have failed to achieve what is uniquely valuable about
it—a satisfying sexual experience. Nussbaum includes ‘having opportunities
for sexual satisfaction’ as part of the capability for bodily integrity,²⁶ but
insofar as some individuals cannot achieve this functioning, they cannot be
provided with the real freedom to perform it.
Similarly, some individuals with Autistic Spectrum Conditions (ASC) may
be unable to perform some elements of the supposedly central functionings of
affiliation and forming emotional attachments. Consider Nussbaum’s descrip-
tion of the former capability, which requires being able ‘to recognise and show
concern for other human beings, to engage in various forms of social inter-
action; to be able to imagine the situation of another’.²⁷ This seems incom-
patible with the experiences of those individuals with ASC who have difficulty
with social interaction, communication, and expressions of empathy.²⁸ Yet, as
the thriving neurodiversity movement makes clear, many individuals with
ASC do not consider themselves worse off as a result. Those with ASC, they
argue, are not ‘damaged or deficient, but rather . . . exhibit . . . a very natural
form of human variation’.²⁹
Asexual individuals, too, are rightly offended by those who consider their
sexual orientation a pathology that requires therapy or hormone treatment. As
one asexual person comments, ‘there’s nothing to fix because we’re not
broken’.³⁰ In both these cases, then, individuals cannot perform a central
functioning, and yet can plausibly claim to have a flourishing life in its
absence. These are not merely cases in which opportunities that matter a
great deal to some people are considered relatively trivial by others.³¹ Rather,
these are cases in which, for as long as these individuals continue to be
asexual or autistic, it will be impossible for at least some of them to have the
capability to perform all of the central functionings, and so impossible to
achieve consensus on their being essential.

²⁶ Nussbaum 2000a: 78. ²⁷ Nussbaum 2011b: 34.

²⁸ To be diagnosed with autism, one needs to have ‘persistent deficits in social communication and
social interaction across multiple contexts . . . [and] restricted, repetitive patterns of behavior, interests,
or activities’ (American Psychiatric Association 2013).
²⁹ Begon and Billington 2019: 188. Indeed, as Robeyns points out, whilst some individuals with ASC
will lack ‘genuine access’ because ‘they do not have the innate characteristics that are needed to truly
enjoy some of these capabilities’, the same may also be true of non-autistics (Robeyns 2016: 387).
Robeyns does not discuss examples, but if she is right this only strengthens the concern I raise about the
exclusionary nature of capabilities to perform specific functionings. In the next chapter I defend the
claim that impairments, including ASC, should be understood as atypicalities rather than defects.
³⁰ Asexuality Archive (2012).
³¹ An objection can be made that the capability approach goes wrong in providing individuals with
opportunities they do not personally value, though I will question the force of this worry in §4.6.
    103

A simple solution would be to therefore excise the ability to perform these

functionings from the list of capabilities that all individuals are entitled to, but
given the importance of these functionings in many other lives, we may be
reluctant to do so. This reluctance may be bolstered by the thought that there
is no harm in including the items on the list. Individuals will only be provided
with capabilities, after all, and so will not be compelled to function in ways
they do not value or are incapable of. Further, for Nussbaum in particular, the
possibility of an overlapping consensus provides only an ancillary justification
for items’ inclusion: ultimately, what matters is their substantive value.³²
Perhaps she could simply accede, then, that an overlapping consensus on
the central capabilities will not be possible after all.
However, we should be reluctant to give up on the possibility of an
overlapping consensus. Ensuring that the content of our account of distribu-
tive justice could be the object of an overlapping consensus demonstrates
appropriate respect for those to whom the theory will apply, and avoids
imposing an alien conception of value that they can never recognise.
Insisting that capabilities are freedoms to function in specific ways means
that some individuals can never be part of a consensus on the necessity of the
ability to perform these functionings—their own lives are clear evidence that
the functioning is not essential—but I will argue that consensus is possible on
the value of capabilities to control. A further benefit of this approach is that it
avoids another troubling implication of focusing on enabling specific func-
tionings: that those who wish to engage in alternative, comparable but unusual
forms of functionings may not be entitled to state assistance. For example, an
individual with ASC may be entitled to ‘normalising’ therapies but not assist-
ance to engage in atypical forms of communication and social interaction.
Even if we set these two worries aside, we still have reason to reject
capability-to-function approaches due to the exclusionary consequences of
justifying the value of capabilities on the basis of the value of the underlying
functioning, for those who cannot perform, and do not value, the functioning.
An advocate of such justification might respond that everyone can have a
capability to function, since even those currently unable to perform it can be
offered a ‘cure’ for whatever condition constitutes a barrier to functioning,
though they would not, of course, be compelled to undergo such cures. After
all, individuals are only provided with capabilities and not forced to function.
They might also point out, as §2.4 considered, that the demands of justice need
not be fully realisable at present. Thus, the fact that we cannot currently cure

³² See Jaggar 2006: 309–20; Terlazzo 2019. Also e.g. Nussbaum 2001: 74.
104      

cancer does not mean that a cancer patient is a counterexample to the claim
that individuals are entitled to the capability for health: justice demands that
this opportunity be provided if feasible, and that we should work towards
making it feasible.³³ It might, then, be argued that we should understand
asexual individuals and those with ASC or other impairments in the same
way. Even if it is currently impossible for asexual individuals to achieve sexual
satisfaction, or for individuals with ASC to function neurotypically, we could
imagine circumstances in which this were possible—where all individuals were
enabled to have the full range of functionings—and justice might demand that
we work to achieve this.³⁴ As such, they are no more ignored or excluded by
the inclusion of capabilities to perform inaccessible functionings than the
cancer patient.
The most obvious difference between these cases and something like cancer,
however, is that people who have cancer do not tend to defend their life and
lack of functioning as valuable, or to understand this feature of themselves as
part of their identity. This difference means that the justification of the
capability will be justifiably offensive to the asexual and autistic individuals
where it would not be to the cancer patient. There is nothing objectionable in
telling someone with cancer that in a more just world we would have a reliable
way to eliminate their illness, and that justice requires that we try to realise this
possibility. This is not true of asexuality and autism.
Or is it? These simple labels cover a heterogeneous set of experiences,
opinions, and preferences. Thus, whilst many autistic and asexual individuals
would find it deeply offensive to suggest that their life lacks something ‘good
and important’ and that their conditions should therefore by cured, others
might welcome such a cure. It might seem, therefore, that making a cure
available, but not forcing anyone to undergo it, is the right approach: everyone
has the capability to function, and they can choose whether they make use of

³³ The concept of ‘health’ is controversial, and having the capability for health should not be
understood to mean the ability to eliminate all of one’s impairments—indeed, given the breadth of
impairment this would be difficult to imagine (as Chapter 5 will discuss). It is relatively uncontrover-
sial, however, to suggest that being healthy requires being free from cancer, and this is all I am
claiming here.
³⁴ Nussbaum’s discussion of disability in Frontiers of Justice suggests she may endorse such a
response. She is rightly critical of the role that social institutions and norms play in causing various
impairments to be disabling, and insists that nearly all individuals could perform the central function-
ings ‘if only public spaces could be adequately designed to support them’ (Nussbaum 2006a: 189).
However, when institutional change is insufficient to enable capabilities she argues ‘we could cure [a]
condition . . . because it is good, indeed important, for a human being to be able to function in these
ways’ (Nussbaum 2006a: 193). Nussbaum is here discussing Eva Kittay’s daughter Sesha, who is
‘ “severely-profoundly” mentally retarded’ (Kittay 2005: 127). She remains undecided on whether the
view would also entail ‘engineering away’ Down’s syndrome, or Asperger’s, or blindness and deafness,
and refuses to ‘speak clearly against’ this possibility (Nussbaum 2006a: 193).
    105

it. For example, even those in the Deaf community who are deeply opposed to
cochlear implants, going so far as to describe the practice of giving them
to deaf infants as a form of ‘genocide’, would not want to prohibit their use by
deaf adults.³⁵
So, am I suggesting that the deep offence that some individuals feel at the
idea of curing their impairment should give them a veto, preventing others
from accessing treatments and ways of functioning they might very reasonably
wish to pursue? I am not. My objection is not that treatments are offered,
nor that someone feels offended, but that the basis of their justification is
offensive. Specifically, that they are taken to be necessary in all cases, implying
an irredeemable deficiency in the lives of those who choose to retain their
impairment. If capabilities are justified on the basis of the value of specific
functionings, then individuals who cannot perform them must be considered
to need the opportunity to be rid of their condition in order to lead a decent
life.³⁶ The fact that people are permitted not to exercise this opportunity, and
so live an inferior life, hardly avoids the insult.
The importance of avoiding the state endorsement of the value of particular
ways of life is a theme in much political liberal writing. Nussbaum describes
this as ‘expressive subordination’, which is ‘subordination that consists in
being publically ranked beneath others’.³⁷ To see why this is problematic
consider two alternative justifications that could be offered to a reasonable
atheist for providing the capability for religious expression. One states that
engaging in religious expression is a good and important human functioning,
the opportunity to engage in which is essential to a decent life. The second
acknowledges that this functioning forms part of some reasonable conceptions
of the good, and that it has value in these lives. The latter justification does not
rank the life of the atheist below that of the religious practitioner.

³⁵ For some discussion of the controversy surrounding cochlear implants, see e.g. Napoli et al. 2015;
Pray and Jordan 2010; Sparrow 2005, 2010. Note that Deaf (capital d) denotes members of the Deaf
community, as compared to those with impaired hearing, who may or may not consider themselves
part of this cultural group.
³⁶ The idea that such functionings are essential not merely to a decent life, but to any kind of human
life, and that those who cannot perform them are consequently less than human, is, sadly, not
uncommon. For example, a survey of attitudes towards asexuality concluded that ‘sexual desire is
considered a key component of human nature and those lacking it are viewed as relatively deficient, less
human and disliked’ (Hodson and McInnis in Mosbergen 2013). Or, as Julie Decker says, ‘[w]e are
perceived as not being fully human because sexual attraction and sexual relationships are seen as
something alive, healthy people do’ (Decker in Mosbergen 2013). The ‘overarching master narrative of
society dictates that “being fully human” or “being a complete person” is dependent on experiencing
sexual attraction’ (Decker 2015).
³⁷ Nussbaum 2011b: 35.
106      

It is perhaps even more important to avoid such ranking when we are doing
this on the basis of central features of individuals’ identities, rather than their
choices. We should not subscribe to a theory of distributive justice that tells
those individuals who see their impairment as a source of identity that justice
requires that we are able to ‘fix’ you, that we should give you this opportunity if
we could, and that it is a great shame that we are unable to do this just yet. But
such expressive subordination is avoided if capabilities are conceptualised
instead as the ability to control certain domains of our life. Treatments
would then be offered as one way to control our functioning outcome in
some domain, as opposed to the only way to achieve a specific valuable
functioning. In addition, this approach could be the subject of an overlapping
consensus amongst those who cannot perform and do not value supposedly
central functionings. Further, as we will see, it provides a wider range of
opportunities, such that all individuals are given opportunities they can
reasonably possess and value.

4.4 Capabilities to Control

To reiterate, the problems with capabilities to function should not lead us to

abandon the capability approach or cull the list of capabilities, but instead
change how capabilities are conceptualised: as opportunities to exercise con-
trol in certain central domains of our life.³⁸ We also should not abandon the
methodology of identifying the central features of a decent life through
internal consideration and interpersonal discussion, resulting in an overlap-
ping consensus for political purposes, amongst individuals with otherwise
diverse conceptions of the good. However, we should change what we come
to a consensus on. Rather than asking ‘which functionings are essential to a
decent life?’, we should ask ‘which domains ought people have control over if
they are to lead a decent life?’ Our goal, then, should be to enable individuals to
form and pursue their own conception of the good, ensuring adequate control
over a variety of options in specific parts of their lives.
Thus, capabilities should no longer be the chance to perform (or not) a
particular functioning, but the substantive opportunity to function in what-
ever way we choose in a domain we have agreed to be important. The
capability for bodily health, for example, will not be the opportunity to achieve
a predefined conception of bodily health, or to choose not to achieve it.

³⁸ A version of the following argument appears in Begon (2017).

    107

It would be the opportunity to control our bodily health, and have the
substantive power to determine and achieve the functionings we prefer in
that domain, even if they are not widely considered ‘healthy’. It is the domain
(our health), not the functioning (being conventionally healthy), that is iden-
tified as being valuable. This means that we can derive a specific list of
capabilities without a specific conception of a flourishing life.³⁹
There are three related benefits to this understanding of capabilities. First, it
allows individuals to be included in the consensus on the central capabilities
without having to acknowledge the value of functioning they do not value or
cannot perform. Individuals with different preferences can agree on the value
of having choice and control without needing to agree on what it is best to
choose. For example, by focusing on ‘control over matters of sexual expres-
sion’, rather than ‘opportunities for sexual satisfaction’, a religious conserva-
tive, a free-love hippy, and a committed celibate could all agree that this is
important. They would, of course, have radically different views about the
proper use that should be made of such control but they can, nonetheless,
accept the central value of setting our own ends in these domains of our life.⁴⁰
Additionally, and more importantly, individuals who cannot perform a func-
tioning, and so cannot possess the corresponding capability to function, can
still possess, and value, a capability to control. As discussed, asexual individ-
uals cannot possess the capability for sexual satisfaction, and may resent the
offer of therapies designed to enable this experience. They can, however,
exercise and value the capability to control their sexuality and sexual life:
this is exactly what asexuality advocacy groups, such as AVEN, campaign for.
This may, for example, involve education programmes to allow for the wider
recognition of asexuality as a sexual identity.
The second benefit is that individuals with non-standard preferences and
functioning capabilities are not simply provided with the opportunity to be
‘normal’—the capability to perform some functioning deemed valuable.
Instead, they are provided with the less specific freedom to determine, and

³⁹ It is worth emphasising that in advocating control I am not suggesting that we should aim to give
people as many options as possible. As has been well-established, providing a great number of similar
and/or valueless options will not increase someone’s control over their life, and may even decrease it.
For example, imagine a single mother who must spend a great deal of time choosing between similar-
quality private nurseries and health insurance plans for her children. She seems to have less ability to
exercise control over her life than a mother who has good-quality state-provided childcare and
healthcare automatically provided, and thus more time for alternative pursuits or to consider more
significant decisions.
⁴⁰ Inclusion of such individuals would not have a repressive effect: individuals cannot enforce their
view of what is ‘right’, since they would be asked only what individuals should control, and not the
proper use that should be made of this control.
108      

pursue, their own goals within a specific domain. This may include the
availability of treatments, but this is not all that would be offered. Providing
individuals with control, without predetermining the proper use they should
make of this control, better captures the capability approach’s central concern
to ensure that everyone has space to form and pursue their own conception of
the good, independently of the content of their decisions. If we are to respect
autonomy, and treat individuals as the origin of ends, we must allow them to
control their bodily health, their sexual satisfaction, their emotional attach-
ments, and so on, and to choose the ends that they wish to pursue in these
domains.
This approach is also consistent with understanding impairment as differ-
ence, and disability as the disadvantages that may be associated with it,
depending on personal features and wider context. Focusing on control
means that justice does not demand the elimination of impairment or the
different ways of functioning it may produce. It merely demands the elimin-
ation of disadvantage, understood as loss of control in a central domain. Since
such control does not necessitate the ability to function in specific ways, this
does not require that impairments are removed. Given that, as the next
chapter will argue, impairments are neutral, this is an important implication.
The third benefit, then, is that when treatments are offered, they are not
justified on the basis that they are necessary because a life lacks something
‘good and important’ unless we can perform a functioning. Rather, they are
offered because the functioning is one valuable option amongst many that may
be part of some reasonable conceptions of the good. For example, consider
deaf and/or blind individuals, who arguably lack the capability to ‘use the
senses’ since they cannot achieve the functioning of hearing and/or sight.⁴¹
These individuals can, nonetheless, possess and value control over their
sensory experience, and the ability to achieve other sensory functionings,
some of which may be shared by hearing and sighted individuals, and some
which may not. Allowing for a range of goals in this domain would mean that
we would not just focus on, for example, the provision of cochlear implants,
thereby making available the ability to perform the functioning identified as
valuable (hearing). Further, insofar as cochlear implants are made available,
they are provided as one route to functioning amongst others, and not as the
only means to achieve an essential opportunity. Instead, individuals would be
entitled to the resources they need to control their sensory experience—as
well, indeed, as control over their ability to function in other domains, which

⁴¹ For further discussion of this case see Barnes (2009a); Begon (2015).
    109

may be restricted if there is limited accommodation for their impairment.

Hence they may be provided, for example, with education in an appropriate
medium, translators and guides, Braille books and keyboards, audible and
visible traffic signals, and so on. There need be no suggestion that the ability
and opportunity to perform specific functionings, and only these functionings,
are important or essential to a decent life. Indeed, even if individuals would
prefer to hear or see, they will not necessarily be entitled to these opportunities
if they already have sufficient options in the sensory domain, and thus already
possess the control they are entitled to—no more than someone who would
prefer to spend their leisure time driving a sports car should be entitled to one
if they already had other leisure activities available to them.
Two possible objections to this approach may suggest themselves. First,
even if a range of options is provided, the definition of domains of control, and
the justification of their inclusion in the list of central capabilities, may still
seem to involve some reference to valuable functionings. Control over our
sensory experience, for example, is included because functionings in this
domain are deemed valuable; control over sexual expression matters because
the various functionings achievements that this expression could entail matter.
Thus, the approach might seem vulnerable to the criticisms raised against
capabilities to function.
It is true that we could only pick out the important domains of control
by reference to the valuable functionings that individuals may pursue in
this domain. However, the domains need not simply be defined as a list of
functionings that have been predetermined to be important. Just as Nussbaum
emphasises that the list of capabilities is ‘open-ended and humble’ and open to
revision,⁴² so the same should be said of the contents of the domains of control.
What these domains entail should be determined in consultation with
those with a variety of experiences, preferences, needs, and abilities, and
should be revised as necessary to include further functioning experiences
within the domain that can be shown to be valuable. Indeed, as the next
chapter will discuss, a far greater role should be given to individuals with
impairments in determining the various forms of functionings that should
make up this domain.
Further, the very inclusion of a multiplicity of functioning achievements
within the domain avoids the central and problematic assumption underlying
capabilities to function: that specific functionings are taken to be valuable
or even necessary to any decent life. Defending the value of a life with

⁴² Nussbaum 2000a: 77.

110      

impairment is not to defend a life lived without functioning achievements, but

with atypical functioning achievements. I agree with Nussbaum’s contention
that we can come to agreement on the shape of a decent human life. I only
suggest that this shape should be sketched in far broader brushstrokes.
Certainly, we could hardly describe a life as decent if it included no capacity
for sensory experiences, no opportunities to be mobile, no ability to commu-
nicate with our fellow persons, and so on. It does not follow from this, though,
that all individuals must be able to hear and see, to walk and run, and to talk
and listen.
The second possible objection is that, since control requires that individuals
have a range of options to choose from, and since individuals’ impairments
can sometimes restrict their option sets, a capabilities-to-control approach
may also seem to necessarily consider at least some impairments to be
disadvantageous. The status of impairment and disability as disadvantageous
will be explored fully in the next chapter. Nonetheless, a response can be
sketched by noting, first, that few impairments leave individuals with only one
functioning option in a domain. Although asexual individuals cannot achieve
sexual satisfaction, this does not mean that there is only one form of sexual
expression open to them. The asexual community is a diverse one, in which
some choose to engage in various forms of sexual relationships (although they
themselves cannot experience sexual attraction), some engage in romantic
but non-sexual relationships (and may identify as hetero-, homo-, or bi-
romantic), and some engage in neither sexual nor romantic relationships.
The diversity of potential functioning achievements is even more apparent
amongst ASC individuals, as their very emphasis on neurodiversity makes
clear. Individuals with sensory impairments, too, have diverse sensory experi-
ences. Consider Helen Keller’s powerful account:

My world is built of touch-sensations, devoid of physical colour and sound;

but without colour and sound, it breathes and throbs with life. Every object is
associated in my mind with tactual qualities which, combined in countless
ways give me a sense of power, or of beauty, or of incongruity.⁴³

Thus, impairments do not ordinarily remove individuals’ ability to control

their functionings or foreclose all their options, and indeed often open up new
possibilities. Further, it is also true that all individuals will be unable to
perform some functionings, and so lack certain experiences. For example, as

⁴³ Keller in Barnes 2009a: 15.

    111

Keller continues: ‘[r]emember that you, dependent on your sight, do not

realise how many things are tangible’.⁴⁴
Thus, having an impairment need not make an individual unusually
restricted. Or, more precisely, impairments assumed to be disabling need
not remove individuals’ capabilities for control, and impairments not usually
assumed to be disabling can seriously restrict individuals’ option sets. It is
important to be careful about language here. As has been discussed, all
individuals have impairments of some form or other, though some are widely
seen as disabilities, some are not, and for some the classification will vary
depending on social context and the preconceptions of whoever is making the
judgement. For example, many would contend that blindness is ‘obviously’ a
disability, whilst short-sightedness is ‘obviously’ not, despite the sometimes
fuzzy border between cases. We cannot simply accede to such intuitions, but
need a principled way to identify the impairments that are disabling.
Specifically, I have argued, individuals are disabled when they lack opportun-
ities they are entitled to.
My goal here is not to offer a classification of conditions that are and are not
disabling (because disadvantageous). Indeed, a central claim of my approach
is that there is no single answer as to whether some impairment is disabling
(as the next chapter explores). What matters is the impact of an impairment
(on access to capabilities to control), and this impact depends not just on
the impairment itself, but all the many factors outlined in §1.4. I have
argued that whilst some individuals with presumptively disabling impairments
cannot perform functionings considered central on capability-to-function
approaches, they can have a range of options in central domains, and can
(and should) be enabled to exercise control over their pursuit of these options.
Contrary to the objection presented here, then, being unable to perform
certain functionings does not rule out the possibility of control. And being
entitled to control in a domain does not require having every opportunity.
Indeed, as Wasserman and Asch point out, ‘[h]aving more ways to realise a
good does not mean that you can realise it more fully’.⁴⁵ This is especially
evident given that all individuals face limitations. Such limitations are not
disabling unless they prevent individuals from having the control they are
entitled to, and our intuitions about what ‘counts’ as a disability are not a good
guide to when this is so. We should not, in other words, be attempting to
generate the ‘correct’ judgement about paradigm cases.

⁴⁴ Keller in Barnes 2009a: 15. ⁴⁵ Wasserman and Asch 2014: 157.

112      

However, I also wish to make clear that I am not making the opposite claim:
that paradigm impairments are never disabling. Not everyone with Helen
Keller’s or Harriet McBryde Johnson’s impairments will have the circum-
stances, support, and internal and external resources to have the same oppor-
tunities and control they experienced. Their testimony demonstrates not that
such control is inevitable, only that it is possible—and, with more support,
could be experienced more widely. There will, then, certainly be cases in which
individuals’ options are sufficiently restricted that they cannot meaningfully be
said to be able to control their functioning outcome in some important
domain of their life, given the current availability of resources and the devel-
opment of technology. Indeed, this may be true very often in the highly
imperfect context in which most impairments are experienced. In such
cases, these impairments are, for now, disadvantageous, and hence disabling
(§5.4). My point, here, is only that this need not be true of all presumptively
disabling impairments.

4.5 Capabilities to Control: Too Little Freedom

or Too Much?

It may further be objected either that too many assumptions are being made
here about what is valuable in human life, unjustifiably ruling out some
conceptions of the good, or too few assumptions about what is valuable,
allowing individuals to make deeply harmful decisions, and, indeed, promot-
ing and supporting such ‘bad’ functionings. First, then, although this approach
makes fewer value assumptions than capability-to-function approaches, it may
be objected that it ‘writes-in’ a commitment to autonomy, and forces choice on
those who do not value it. For example, a religious conservative might agree
that human sexuality is important, but vehemently disagree that our choices in
this domain should be open, limited by nothing but a concern for the consent
of, and potential harm to, others.⁴⁶ They may object even more to being
expected to fund others’ choices in this domain.
My response would be that a commitment to autonomy is a necessary
feature of most liberal positions, and I will not attempt to defend liberalism
to non-liberals here. That this is incompatible with the values of illiberal and
intolerant individuals should not necessarily concern us. First, though indi-
viduals are provided with the capacity to set and pursue their own ends in

⁴⁶ For an argument that this should be our approach, see Begon (2019).
    113

certain domains of their life, they are not forced to choose autonomy-centred
goals. It is compatible with the approach to allow individuals to use the control
they are provided with to choose a way of life that makes little active use of this
control. As §4.6 will discuss, a commitment to the value of capabilities need
not entail an assumption that individuals should choose a life involving
autonomy or maintain their autonomy over time.⁴⁷
If the concern is, instead, that people will resent allowing, and indeed
enabling, others to have such control, then I would echo Nussbaum, who
frequently notes that it is self-subverting to tolerate the intolerant, or to defend
‘strongly nonrelativistic’ local norms out of a commitment to relativism.⁴⁸ If
we accept political liberalism, we need only enable individuals to pursue their
own conception of the good, not enable them to force this conception on
others (however much they would value the chance to do so). Hence, as
Nelson notes, ‘the exclusion of liberty-violating accounts of the good life is
the only departure from neutrality explicitly allowed in . . . [Rawls’s] theory’.⁴⁹
Individuals have the opportunity to pursue their conception of the good in
domains of their own life but are denied the opportunity to limit others’
freedoms, through act or omission.
Second, it may be objected that, rather than assuming too much, the
approach assumes too little: by providing individuals with relatively abstract
opportunities, and allowing them to function as they wish, the approach has
no recourse against individuals making ‘bad’ choices. Further, it may support
or even promote such choices. Claassen and Düwell, for example, have raised
the concern that Nussbaum’s capability approach may justify supporting
‘dark’ capabilities, such as cruelty and aggression. They argue that we must
distinguish the abilities that exert a moral claim to be politically protected and
developed from those that do not.⁵⁰
In response, it should be pointed out, first, that the capability approach will
not allow individuals to develop and exercise capabilities such that they inhibit
the capabilities of others. A central concern for both Nussbaum and Sen is the
separateness of persons, and they vehemently oppose utilitarian approaches
that aggregate individual interests. We can assume, then, that any version of
the capability approach will not allow someone to exercise their capabilities in

⁴⁷ Perfectionist liberals may object to this approach for just this reason (e.g. Chambers 2008:
159–202). (Also see Chapter 7.)
⁴⁸ Nussbaum 2000a: 49. ⁴⁹ Nelson 2008: 102.
⁵⁰ Claassen and Düwell 2013: 495–8. Note that Claassen now favours a view that is neutral on all but
‘the value of autonomy’ (Claassen 2018: 6). The state, he insists, should not ‘create a set of valuable
options or discourage disvaluable options’ (Claassen 2018: 21).
114      

such a way that it avoidably causes someone else to fall below the specified
threshold of some other capability. Whilst there may be some hard cases,
requiring difficult choices about the allocation of a limited bundle of resources,
an individual certainly would not have their capacity for aggression enabled if
this required the provision of a stream of victims on which to express it.
Leaving aside cases of harm to others, then, should we be concerned that a
capability to control approach may permit, or even promote, ‘unvirtuous’
functionings? Should individuals be provided with outlets for their aggression?
Should individuals be allowed to use their control over their health, say, to
undergo non-essential and seemingly harmful medical procedures, such as
FGC or the voluntary removal of a healthy limb? Simply put, the answer is yes:
individuals certainly should not be prevented from pursuing their conception
of the good even when the content might seem distasteful to some.⁵¹
The more difficult question concerns whether such functionings would be
supported, enabled, and politically protected. This answer, I think, will depend
on the role of the functioning for the individual concerned. For example,
removing a healthy limb might be part of an individual’s capacity to control
their bodily integrity if they have bodily integrity identity disorder (BIID) and
can only feel ‘complete’ if the limb is removed. Or, for some, such an
amputation might be an important part of sexual expression.⁵² As a compo-
nent of domains that are agreed to be important this functioning might,
therefore, be amongst the options that are positively enabled: for example,
by making surgeries available on the NHS.⁵³
This distinction—between what should be permitted and what should be
enabled as a matter of justice—is crucial, though often elided. For example,
Carter contends that insofar as the capability approach aims to provide
opportunities to perform specific valuable functionings (or to choose not to),
then no value will be placed on ‘dysfunctionings’: being a couch potato, eating

⁵¹ I am certainly not suggesting that FGC as it is currently practised—forcibly, against young girls, in
unsterile conditions—should be permitted. Rather, I contend that if the procedure was chosen
voluntarily, by consenting adults, then political liberals should not interfere with this self-regarding
decision, just as they would permit cosmetic procedures including (the sometimes very similar)
labioplasty. Nussbaum’s (1999: 118–29) attempt to maintain that political liberalism is consistent
with a ban on FGC is unsuccessful, as Chambers (2008: 159–202) convincingly demonstrates.
⁵² Both motivations are discussed in Noll and Kasten (2014); the former in Bayne and Levy (2005);
the latter in Blom et al. (2017).
⁵³ There might anyway be good reasons to do this given the dangers inherent when individuals take
matters into their own hands, and in light of the positive reports of well-being of those who are
successful (Reynolds 2018; Noll and Kasten 2014; Bayne and Levy 2005; First 2005). This approach is
further supported in the next chapter, where I defend understanding impairment as atypicality: this
then implies that the surgical removal of a limb need not entail harming the patient, undermining a
central objection to its provision (also see Reynolds 2018).
    115

unhealthily, or engaging in non-social and anti-social behaviours.⁵⁴ Whilst

‘liberals would normally tolerate’ such behaviours, Carter insists that capability
theorists can ‘prescribe . . . [them] only to the extent that each of them can be
shown to have value’, either as a valuable functioning or an alternative to a
valuable functioning.⁵⁵ For this reason, he suggests that to make the capability
approach more permissive, we should value ‘capability as such’, or content-
independent freedom, as well as specific capabilities.⁵⁶
However, we need not choose between capabilities to perform valuable
functionings (or not), on the one hand, and ‘capability as such’, on the
other. I agree that we should not be overly prescriptive about the kinds of
functionings individuals should choose and should enable them to pursue
their own conception of the good. Thus, I do not and need not dispute Carter’s
claims that, given persons’ status as setters of their own ends, it would be
inappropriate and disrespectful to set their ends for them, and that a situation
is better when an individual has more freedom than less.⁵⁷ However, it does
not follow from accepting this that we must also believe that the state should
enable people to achieve all of their ends, maximise their freedom, or ensure
they have access to ‘capability as such’. Instead, we can focus on enabling
individuals to be mobile, say, however they choose: supporting individuals’
access to mobility aids, prostheses, modified vehicles, and so on (and even if
their mobility impairment results from their own decision to remove a limb).
Yet control over mobility does not require that every desired option is
provided. Control over some parts of our life may be beyond the scope of
justice and the concern of the state. Plausibly, buying a sports car is neither
necessary for control over our mobility, nor part of any other central domain.
Thus, the state is not required to enable this functioning. This need not imply
the paternalist judgement that sports cars are disvaluable, or that driving them
is a dysfunctioning (at least assuming their emissions do not constitute a harm
to others). Neither seems true. The state need not proscribe this functioning,
then—it simply is not required to ensure that all individuals can access it.⁵⁸
Moreover, there is a benefit in such cases to adopting my approach over
Carter’s. Whilst Carter advocates providing individuals with non-specific
‘capability as such’, he argues that we can identify appropriate beneficiaries
on the basis of their lack of access to specific capabilities. (He advocates a

⁵⁴ Carter 2014: 91–2. ⁵⁵ Carter 2014: 92.

⁵⁶ Carter 2014: 95–6. ⁵⁷ Carter 2014: 93.
⁵⁸ It is true that the state is endorsing a judgement that such functionings are inessential to a decent
life, but this does not seem to generate reasonable grounds for an insult, or an accusation of
paternalism, in such cases.
116      

mixed account in which both capability as such and specific capabilities

have value.) Thus, he gives an example of an individual who lacks the
capability to walk, and is deemed worse off for this reason. He can choose to
spend his (cash) compensation as he chooses, but it is this inability that
warrants the provision of compensation.⁵⁹ Given the individual will not be
interfered with, this is not a case of paternalism. Nonetheless, this demon-
strates the unwarranted and disrespectful assumption discussed above that
individuals ought to be able to function in very specific ways, and if they
cannot we can automatically assume they are worse-off and entitled to com-
pensation. It can certainly be argued that a life goes better when an individual
can control their mobility, but we should not be so wedded to the idea that
specific functionings (like walking) matter.
My contention, then, is not that individuals should be able to do ‘whatever
they want’, but that they should be able to control central domains of their life,
where this will require access to a range of acceptable options. Some domains,
and some functionings, might sometimes be beyond the scope of justice.
Nonetheless, unlike capability to function approaches, I am not committed
to the value of a specific and narrow set of functionings and the conceptions of
the good that incorporate them. Thus, to point out that my approach will
support and protect various non-standard conceptions of the good is not to
critique my approach but to state it—and one of its primary attractions. The
aim of my approach is to ensure that we have substantive support not just to
perform a functioning—for example, to be healthy—and the chance not
to perform it. We should also have the substantive opportunity to be what is
‘unhealthy’ on many views: to live with ASC or deafness or blindness; to
amputate a limb; to engage in sexual activities that leave bruises or scars, or
engage in no sexual activity at all.

4.6 The Value of Capabilities

A final line of objection questions whether providing capabilities of any form

should be the goal of a theory of justice: if such theories ‘should be concerned
only to promote the well-being of persons’ rather than their agency, then
indeed capabilities would not be our central concern.⁶⁰ This gives rise to two

⁵⁹ Carter 2014: 96.

⁶⁰ Arneson 2006: 35 (also see 2000a, 2010). Others who share this concern include Cohen (1993,
1994), Fleurbaey (2002, 2006), Richardson (2000), and Vallentyne (2005, 2006 (see §3.4)).
    117

interrelated worries. First, it might be argued that a capability theory of justice

overvalues capabilities at the expense of functionings. A weaker version of this
worry holds that, though freedom is valuable in itself, achievements matter
too. The stronger version suggests capabilities have no value unless they are
actually used. Both views contend that functionings should be promoted as
well as, or instead of, capabilities. Second, given that at least some items on any
capability list will not be universally utilised or valued, it may be argued that
the focus on providing all individuals with access to specific capabilities should
be dropped, allowing instead for trade-offs between commensurable function-
ings and the provision of a wider array of opportunities. In response, I defend
both the commitment to capabilities not functionings (in this section), and to
a universal, plural, non-fungible list of capabilities (in the next). In brief,
I claim, first, that it is not just the ability to achieve outcomes, but the ability
to achieve them for ourselves, that is of central and ineliminable value in the
context of justice; and, second, that the scope of distributive justice should not
be all-encompassing.
First, then, do capability theorists place an excessive degree of emphasis on
leading an autonomous, self-directed life? Cohen criticised Sen along these
lines, arguing that the ‘athleticism’ of his approach leaves no space for the
value of benefits that are achieved passively.⁶¹ For example, the health we
achieve from being in a malaria-free environment or from the rays of the sun,
or nourishment from a nutrient drip. As Cohen notes: ‘[t]here are two values
associated with the successful exercise of freedom. One is that the world
conforms to my will and the other is that it is I who achieve that result.
Sometimes the second value does not matter much.’⁶² Similarly, Fleurbaey
insists that freedom is not the only thing that matters: ‘even those who like
freedom a lot do also care in general about their achievements’.⁶³ This is the
more moderate version of the worry that capabilities are over-valued, insofar
as it does not deny the intrinsic value of freedom, but simply emphasises that
achievements matter too.

⁶¹ Cohen 1993, 1994. This is a rather unfair criticism of Sen in particular, given that Sen distin-
guishes the value of agency and well-being and sees both as important: it matters that I control my life
and it matters that my life goes well, and these concerns are not reducible to each other (see Sen 1992,
1993; Crocker and Robeyns 2010; Cudd 2014). Nonetheless, Cohen’s critique does have bite against
those iterations of the capability approach that focus on agency, including my own. Further, Cohen and
Sen are in genuine disagreement over whether individuals are free if the world conforms to our will
even if we did not bring this conformity about. For Sen (1993: 44), ‘freedom has many aspects . . . and it
would be a mistake to think of achievements only in terms of active choice by oneself’. I follow Cohen in
understanding freedom to require control (also see Begon 2016a). (For a defence of Sen, see Olsaretti
(2005).)
⁶² Cohen 1994: 124. ⁶³ Fleurbaey 2006: 306.
118      

Indeed, this is the view that is taken by many capability theorists, including
both Sen and Nussbaum: ‘freedom has intrinsic value . . . [but t]he capabilities
would be pointless and idle if they were never used’.⁶⁴ Or, even more starkly:
‘if there were no functioning of any kind in a life, we could hardly applaud
it, no matter what opportunities it contained’.⁶⁵ As such, much of the
criticism of the capability approach for valorising autonomy is misplaced:
many capability theories, including the most influential, allow that sometimes
functionings should be promoted rather than capabilities. Nonetheless, this
criticism is worth engaging with since I unambiguously endorse a commitment
to capabilities (as the ability to control certain domains of our life), rather than
the achievement of valuable functionings.
The stronger version of this criticism, most clearly articulated by Arneson,
goes further and insists that capabilities do not have any value in themselves.
Rather, they are merely a means to ensuring individuals perform valuable
functionings. Indeed, Arneson claims it would be ‘fetishistic’ to postulate that
well-being freedom has ‘intrinsic moral importance’ independent of the value
of actual well-being.⁶⁶ On Arneson’s view, then, freedom only has value if it
actually improves lives: ‘capability provision that in no way enhances anyone’s
life is pointless’.⁶⁷ He finds it yet more inexplicable that a theory of justice
would deem people entitled to a capability that ‘will be ignored or wasted or
mishandled by the individual to whom it is provided’.⁶⁸ Simply put, Arneson’s
complaint is that ‘justice according to the capability approach obligates society
to “help” [people] in ways that do [them] no good by [their] own lights’.⁶⁹
Much of this dispute arises from a deeper disagreement regarding whether
theories of justice should aim to promote well-being on a case-by-case basis. It
is undoubtedly true that we only consider individuals to be entitled to some-
thing as a matter of justice if we think it is, in some sense, good for people.

⁶⁴ Nussbaum 2011b: 25. As n.59 notes, Sen values both agency and well-being.
⁶⁵ Nussbaum 2000a: 87. Nussbaum’s primary justifications for promoting actual functioning are,
first, that some of the capabilities are themselves so important ‘that we are sometimes justified in
promoting functioning rather than simply capability’ (Nussbaum 2000a: 91); and second, that we can
promote functionings to ensure that individuals do not lose access to capabilities, or ‘crucial areas of
empowerment’ (Nussbaum 2004: 199) in future. Yet it should also be emphasised that Nussbaum’s
writing on this topic is ambiguous, and she often argues that the goal of just public policy should be to
provide access to capabilities, not functionings. (For further discussion of Nussbaum’s view see Begon
2017.).
⁶⁶ Arneson 2006: 34. Olsaretti (2005) also takes an instrumentalist approach to the value of
capabilities. On her view, it is not enough simply to perform valuable functionings, we should also
endorse these functionings, and this is best achieved if individuals have the capability to function or not.
Thus, agency should be promoted merely as a means. And, indeed, only if the causal connection
between having capabilities and endorsing valuable functionings actually holds (see Begon 2016a:
58–60).
⁶⁷ Arneson 2000a: 60. ⁶⁸ Arneson 2006: 32. ⁶⁹ Arneson 2000a: 61.
    119

However, welfarists like Arneson and Vallentyne assume that the goal of each
instance of (re)distribution is to improve someone’s quality of life. Hence,
they claim justice should never demand the provision of opportunities that
are not good for some particular individual in their particular context. Yet
many theorists of justice contend that individuals are entitled to opportun-
ities (or resources, or rights) regardless of the benefits they bring, or fail to
bring, in specific cases. This distinction can be seen most clearly when we
consider opportunities that are indeed likely be ignored, wasted, or mis-
handled by some, yet plausibly remain a distributive entitlement. The
opportunity to vote is perhaps the most obvious and uncontroversial
example: few would suggest that we should abandon universal suffrage
because some individuals ignore, waste, or mishandle their right to vote.
Similar points apply to the opportunity to found a family for the resolutely
childless, the opportunity to engage in sexual relationships for the deter-
mined celibate, or the opportunity for leisure for the committed workaholic:
it is plausible to suggest that these individuals are entitled to these oppor-
tunities nonetheless.
One line of response to the welfarist worry would be to point out the ways in
which the provision of capabilities can be beneficial even when they are not
exercised: that a life with opportunities to vote, have children, engage in sexual
relationships, and leisure activities is better than one without, even if the
functioning achievement in the two lives is identical. Indeed, I believe this is
so, for at least three reasons. First, because providing opportunities to exercise
control in central domains of our lives demonstrates an appropriate attitude of
respect toward people as equal members of the community, or as fellow
persons: everyone should have the chance to have their voice heard in political
decision-making, everyone should have access to employment that leaves
them with time to pursue leisure activities if they wish, and so on. Being
treated with the respect to which we are entitled may reasonably be
deemed to improve our quality of life compared to a scenario in which we
are disrespected.
Second, having an opportunity to do something significantly changes the
nature of our decision not to do it. Most obviously, it is a decision. Choosing to
be a workaholic is very different from never having time for hobbies (see §4.3).
The value of being in control of our functioning achievements does not lie
merely in the outcomes that we achieve, but in the ability to take charge of our
own lives. Our political goal, then, should be empowerment and not mere
functioning (see Chapter 7). Third, providing opportunities, even if unused,
leaves options open if individuals change their goals or conception of the good
120      

in future.⁷⁰ Individuals can choose to change their work–life balance or retrain

for a different career, they can become politically engaged, or pursue non-
biological routes to parenthood. There is value in feeling that we can change
our mind and pursue a different way of life, even if we are not one of those who
in fact does so.
Thus, even by Arneson’s own lights there might be some value to ‘wasted’
opportunities.⁷¹ Nonetheless, the core dispute remains. The capability
approach is motivated by the plausibility of the claim that there is something
valuable about having freedom as control over how our life goes. Welfarists
like Arneson and Vallentyne contend that capabilities should only be provided
in those cases when they will actually improve quality of life. The fact that
freedom is sometimes conducive to well-being is, as far as they are concerned,
insufficient to ground a general policy of capability provision, since in actual,
specific cases well-being could be better promoted by other means, such as
directly promoting or compelling functioning.
It is worth noting that, contra Arneson’s caricature of their position,
capability theorists do not advocate the provision of ever more freedom,
even when it is not desired and might lead the recipient astray. Arneson
imagines someone who enjoys a life of seeing opera and playing football,
but begins using cocaine at the expense of other pursuits as a result of the
additional freedom that more money provides them.⁷² Yet capability theories
of justice do not argue for maximal freedom, but the provision of specific
freedoms, usually to some threshold level.⁷³ It is much less obvious how
freedom of this constrained sort will be positively bad for people.
Nonetheless, I will grant Arneson that central capabilities might sometimes
be mishandled, and undoubtedly there will be many cases where capabilities
are at least wasted or ignored.
Given this, are individuals still entitled to opportunities in these cases?
Unequivocally, yes: we have a right to dispose of our private property, for
example, even if some individuals use this right in ways that do them no good
by their own lights—selling it on a whim, gambling it away, gifting it to
some temporary object of their affection, and so on. Imagine someone who

⁷⁰ Of course, individuals may exercise their capabilities in ways that foreclose them changing their
mind (see §4.2). Allowing individuals freedom as control encompasses the opportunity to remove their
own options in this way.
⁷¹ Indeed, Arneson (2006: 32) agrees that freedoms have value as constituents of some items on his
objective list of the components of well-being.
⁷² Arneson 2006: 33.
⁷³ Arneson (2006) has a distinct set of objections to sufficientarianism, which I will not engage
with here.
    121

does not have the vote but never wanted it, or someone who was subject to a
programme of sterilisation because individuals in their social group were
deemed incapable of looking after their offspring but never desired children,
or an individual who is excluded from educational institutions due to ableist
bias and a lack of accommodation but does not see the value in education. Or,
if we follow Arneson and understand well-being objectively (so these individ-
uals’ well-being is not high despite their opinion to the contrary), then instead
imagine that these individuals are, respectively, compelled to vote, prevented
from being sterilised (and so subject to the constant fear of unwanted preg-
nancy for themselves or their partner), or made to stay in formal education
(instead of pursuing a career in a domain they would prefer). The treatment of
the individuals in both sets of cases seems clearly unjust, even if we assume
their (subjective or objective) well-being is appropriately high. Yet Arneson
and Vallentyne see the opposite answer as similarly self-evident.
Perhaps much of the disagreement arises from a deep clash of intuitions of a
form that parallels disagreements around paternalism.⁷⁴ I do not doubt that
individual well-being might improve in some cases if policy were designed to
nudge, manipulate, or even compel actual functioning. After all, if paternalist
policies were always simply self-defeating there would be no interesting debate
regarding their permissibility. However, the fact the paternalist policies can, at
least sometimes, achieve what they are designed to—that is, make individuals
better off by bypassing individuals’ choices or ignoring their autonomous
capacities—certainly does not settle the question of whether we should pursue
them. I can thus only appeal to the contention that competent adults
should be provided with the freedom to control the course of their own
lives, to pursue their goals, and to function as they wish, even if they some-
times make mistakes.⁷⁵
A further objection Arneson might push against the provision of capabil-
ities that will be unused or misused is that it is wasteful: why spend our limited
resources on this instead of something that will actually make people better
off? Again, one partial response can be provided in terms Arneson would
accept. This is to point out that the provision of unused capabilities need not
be expensive. Providing capabilities for ongoing education and retraining, for
example, does not require that we run classes in which no one has registered
on the off-chance someone shows up. Similarly, the provision of accessible

⁷⁴ A debate in which Arneson has made influential arguments on both sides (e.g. Arneson 1980,
1989, compared with Arneson 2000a, 2013). (See Begon 2016b.)
⁷⁵ As Chapter 7 explores in much greater detail.
122      

course materials, for example, can take account of those registered on a course
(insofar as policies of universal design are infeasible). We need not, then,
produce material in Braille, as audio files, in large type, and so on, ‘just in case’.
Thus, in many cases, the costs of capability provision only manifest on uptake.
However, cases may remain when this is not so: we may be required to run
courses with inefficiently low numbers of attendees, for example, or run local
elections in wards like Liverpool Central with a 12.7 per cent turnout rate and
little uncertainty about the outcome.⁷⁶ We might question whether this is a
good use of limited resources.
There are, of course, many questions that can be asked about how our
limited resources should be used. For example, should the state fund art
galleries when people die of avoidable diseases? Should we provide computers
for dyslexic children when children all over the world (including in developed
countries) do not have enough to eat? Or, in effective altruists’ favourite
example, should we fund the expensive process of training guide dogs when
many lives could be saved if this money were instead spent on mosquito nets
to prevent malaria? There are no simple answers to these questions, and few
would find it easy to answer ‘no’ in all these cases. The problem of weighing
questions of efficiency against fairness and the achievement of other valuable
ends in the context of resource allocation is hardly unique to the capability
approach, then.
Moreover, we should be cautious of simply accepting such dichotomies.
Scarcity can be exaggerated and is frequently manufactured. Often the prob-
lem is not that there is not enough to go around but that it is concentrated in
the wrong hands.⁷⁷ We must refuse to allow social norms, political intransi-
gence, or systems of government to determine what justice requires.
Otherwise, unjust norms, systems, and structures could avoid this label simply
by becoming sufficiently entrenched (see §2.4). Thus, in a world in which we
can afford to both train guide dogs and buy malaria nets, justice might require
that we do both. Similarly, justice might require us to run adult education
classes with low enrolment and run elections with low turnouts, as well as
providing opportunities that are more appreciated. When practicalities mean
the requirements of justice cannot be fully realised, it might be that less utilised
opportunities receive lower priority.⁷⁸ Yet this does not undermine the general

⁷⁶ https://www.bbc.co.uk/news/uk-england-35999707.
⁷⁷ E.g. consider Sen’s (1980, 1981) influential argument that famines take place in situations of
moderate to good food availability, or Singer’s (1972, 2009) arguments that the affluent should do more
to help the world’s poorest.
⁷⁸ The weighting of different capabilities is discussed below (§4.7).
    123

claim (which Arneson disputes) that individuals have a claim of justice that
they be provided.
However, a capability approach to justice would not only provide oppor-
tunities that are under-utilised or wasted. It would also not provide oppor-
tunities that would be used, appreciated, and significantly impact quality of
life: for example, cash to spend as we wish, rather than free healthcare,
education, and voting systems. In the end, then, the dispute revolves around
the question of what we think justice is really about. Welfarists like Arneson
and Vallentyne assert that it must involve improving individuals’ quality of life
in every case, whilst I (alongside many others) dispute the claim that justice
is reducible to well-being promotion. I contend instead that a just society
is one that respects individuals as free and equal, and enables them to
exercise control in certain central domains of their lives. One in which they
have the chance to lead a life with high well-being, but not one which aims to
maximise well-being with every state action. Thus, even if individuals would
rather have sports cars, rollercoaster rides, or musical instruments than access
to the voting booth or public schools, they are entitled to the latter and not the
former. Individuals are entitled to capabilities even if they use them in ways
we disapprove of, or which make them worse off, and even if we could instead
provide them with some functioning outcome that would make their life better
by nudging, manipulating, or forcing them to make better choices.

4.7 Why Capability Lists

A further consequence if we were to promote individuals’ well-being rather

than their freedom in specific domains is that this would generate a single scale
of well-being on which individuals can be placed and ranked. And if our goal is
to ensure that individuals reach a particular well-being ‘score’, and ‘well-being
scores can be assigned to all functionings to which one has access’,⁷⁹ then
(leaving aside the difficulty of assigning these scores) all individuals can be
judged, simply, to have achieved or not achieved that score. Further, it does
not matter how the target well-being score is reached as long as it is—even if it
is achieved with trivial functionings, by overriding individuals’ autonomous
choices about how they function, or by substituting deficiencies in one domain
for greater opportunity in others. No special value is attached to achieving

⁷⁹ Arneson 2006: 25.

124      

specific functionings or having access to a range of functionings, except insofar

as some variation might be necessary for well-being.
Thus, when the focus is firmly on well-being, there is no place for the
thought, central to many iterations of the capability approach, that certain
freedoms are of particular importance as entitlements of justice, and a special
kind of wrong is done when these freedoms are infringed: ‘people feel that the
abridgment is not just a cost to be borne; it is a cost of a distinctive kind,
involving a violation of basic justice’.⁸⁰ Nor, indeed, is there space for the
contention that the list of valuable capabilities is irreducibly plural,⁸¹ that
individuals are entitled to each of these capabilities, and that an absence of
one capability cannot be substituted with another.
I suggest that these implications of welfarist views should be resisted: we
should not adopt such an overall ranking, or allow capabilities to be traded.
Compare, for example, a person who lacks the time and resources to engage in
leisure activities but has control over their bodily integrity, with someone in
the reverse situation (a wealthy individual who is raped by their spouse,
perhaps). We might be able to come to a view about which of these individuals
is better-off. However, simply ranking these individuals according to a single
metric—well-being, say—and thereby describing the person who must work
three jobs to make ends meet as ‘better-off ’ than the wealthy victim of rape
and abuse is surely not the right way to characterise this case, or others like it.
This is because such general judgements may blind us, first, to the fact that
both individuals lack what they are entitled to and, second, to the independent
and incommensurable value of these different domains. We cannot, for
example, compensate someone for a lack of bodily integrity with additional
opportunities for leisure, or vice versa. Even if we do genuinely manage to
improve their well-being by this method, this should not be what a theory of
justice advocates.
Of course, in some contexts we may need to make judgements about the
comparative urgency of individuals’ unmet needs, as well as the difficulty of
meeting them, in order to determine how to allocate scarce resources.
However, accepting a pluralist approach, according to which we cannot negate
the injustice of an absence of one capability with more of another, need not
commit us to the view that capabilities are so deeply incommensurable that we
can never make such judgements about relative weighting or prioritisation of
needs. This is not the place to defend a specific list of valuable capabilities or to

⁸⁰ Nussbaum 2006b: 56. ⁸¹ E.g. see Nussbaum 2000a: 81.

    125

determine how the items on the list should be identified and weighted. This is
a project that has been taken up and analysed by many others.⁸²
I am, however, in broad agreement with the approach employed by
Wolff and de-Shalit, who adopt a method of dynamic public reflective equi-
librium to identify valuable capabilities and provide guidance about their
relative weightings.⁸³ Ultimately, they advocate a system of complex evaluation,
utilising multiple approaches to weighting, generated through democratic
consultation.⁸⁴ However, given that disadvantages in different domains tend
to cluster, they note that the same groups will be identified as the least advan-
taged across multiple evaluation methods, making it easier to identify whose
needs should be prioritised. I will not attempt to recapitulate the details of their
approach, but merely raise it as evidence that accepting an irreducible plurality
of capabilities does not preclude weighting competing interests and identifying
the least advantaged in a way that can provide practical policy guidance.⁸⁵
Pragmatic considerations do not, therefore, necessitate a move towards
adopting a single scale. More important, though, are the theoretical reasons
to avoid linearising our understanding of disadvantage and to instead
acknowledge that lives can go better and worse on all sorts of dimensions.
Some of these dimensions may be more significant than others, and may
have knock-on effects (either good or bad) in other domains of our life.
Nonetheless, simply asserting, say, that a rich person with MS is ‘worse off ’
than a person without money (as Vallentyne does) unhelpfully removes

⁸² E.g. Byskov 2017; Claassen 2011; Jaggar 2006; Nussbaum 1992, 2000a, 2011a; Robeyns 2003; 2005;
Sen 2004; Wolff and de-Shalit 2007. I broadly accept the content of Wolff and de-Shalit’s modified
version of Nussbaum’s list, though of course I would understand these as capabilities for control, not
functioning: life; bodily health; bodily integrity; sense, imagination, and thought; emotions; practical
reason; affiliation; other species; play; control over one’s environment; plus doing good to others; living
in a law-abiding fashion; understanding the law; communication (Wolff and de-Shalit 2007: 45–62;
Nussbaum 2011b: 33–4). I would also accept Nussabaum’s subdivisions: for example, bodily integrity
incorporates control over one’s mobility and one’s sexual life.
⁸³ Dynamic public reflective equilibrium is a process whereby initial theoretical assumptions are
tested and, where appropriate, revised in light of the theories and intuitions of participants who have
experience of disadvantage or of working with those who do (Wolff and de-Shalit 2007: 41–3). In
addition to providing support for, and additions to, Nussbaum’s list of central capabilities, Wolff and
de-Shalit also identify a subset of these that are of special importance: life, bodily health, bodily
integrity, affiliation or belonging, control over one’s environment, and sense, imagination, and thought
(Wolff and de-Shalit 2007: 106). These are particularly significant in themselves, constitute fertile
functionings when possessed (‘functionings, which spread their good effects over several categories’),
and corrosive disadvantages when absent (‘disadvantage that has negative effects on other functionings’)
(Wolff and de-Shalit 2007: 121–2). Thus, it may be that rectifying a lack of control in these particular
domains should be a priority.
⁸⁴ Wolff and de-Shalit 2007: 101.
⁸⁵ This is not dissimilar from Sen’s claim that incommensurable values and incomplete orderings
can be resolved through a process of social choice, wherein a consensus is reached on the relative weight
of the various dimensions (see e.g. Sen 1999a: 75–9; 2009: 235–6). Wolff and de-Shalit provide far more
concrete guidance, however.
126      

nuance from our understanding of their experience.⁸⁶ It might be that neither

individual has what they are entitled to (though we should, of course, be
cautious of assuming that merely having MS warrants this conclusion, see
n. 86). It might even be that the needs of the individual with MS should take
priority if we can only provide assistance in one of these cases. However, none
of this implies that the able-bodied person ‘beats’ them in some overall
ranking—in some ways each of their lives goes worse than the other, and in
some ways better. This is so for almost any pairwise comparison we might
make. It is important, then, to acknowledge the plurality of value in individ-
uals’ lives and the multiplicity of domains over which they should have
control, without which they do not have what they are entitled to, no matter
how well their life is going. As such, the irreducible multidimensionality of the
capability approach should not be lost.

4.8 Conclusion

As a partial theory of justice, the capability approach is centrally motivated by

the goal of ensuring that all individuals have the substantive opportunity to
lead a decent human life, and to decide for themselves the form this life should
take. The preceding chapter outlined some of the benefits of adopting cap-
abilities as a metric of justice: by focusing on outcomes, and the multiple
factors that can influence them, the capability approach avoids ranking indi-
viduals on their internal capacities, recognises the influence of contextual
factors, and takes account of the ways in which unjust background conditions
can undermine both opportunities and the use individuals make of them.

⁸⁶ Vallentyne makes this claim on the basis that MS is so bad that even if a person with MS had all
the available money (in an imagined, two-person world) they would be worse off than the person who
had no money: ‘the absence of multiple sclerosis is infinitely more valuable than any (finite) amount of
money’ (Vallentyne 2006: 87). There are a number of things to object to about this case. I will mention
two of them. First, it provides a paradigm example of the unwarranted yet frequent assumption that
having an impairment is a tragedy, and can be employed as a case of extreme disadvantage without
much thought. This is not the place to discuss the status of impairment: that will come in the next
chapter. Nonetheless, it is worth flagging the deep implausibility of this case. Whilst it may be
reasonable to classify MS as something that tends to make individuals’ lives go worse (hence, not
value-neutral), it is not reasonable to assume that it is, in every case, (infinitely) worse than abject
poverty. Second, the focus on money is misleading. We might, for example, reasonably say that
someone is better off if they become disease free, despite losing £10,000 of income. However, if their
past higher income had been used to achieve functionings that they can now achieve much more
cheaply—for example, because they no longer need to purchase pain relief, hire home help, or undergo
expensive specialist treatments—then they have hardly lost one capability at the expense of another.
Rather, they have the same capabilities, but now require fewer resources to achieve them as a result of a
change in their conversion factors (see §3.5).
    127

This chapter has argued that the focus on capabilities to control avoids the
insulting justifications for capabilities to function, is better able to include
individuals who do not value central functionings in the overlapping consen-
sus, and enables people to function in non-standard ways.
The aim of these chapters has been threefold. First, to defend an account of
distributive justice that avoids a commitment to a restrictive conception of the
good, and so avoids privileging certain ways of functioning. This means that
individuals with impairments often considered to be paradigm disabilities
need not count as being disabled, despite being unable to perform specific,
valuable functionings. Second, to flesh out a conception of individual entitle-
ments, in order to clarify what it means to claim, as I have, that individuals are
disabled only when they lack the opportunities they are entitled to (as a result
of having an impairment in a particular context). Finally, to demonstrate that
taking human diversity and the fact of impairment seriously can, and should,
influence our decision about the kind of theory of distributive justice to adopt.
 5
Neutral Impairment, Disadvantageous
Disability

5.1 Introduction

Is it bad to be disabled? Are disabled individuals, ceteris paribus, worse off

than non-disabled individuals? Is disability compatible with health? The
answers to these questions might seem obvious, yet there is little agreement
on what the ‘obvious’ answers are. Many disabled individuals do not consider
their disability a bad feature of their life, they do not think they are, in general,
worse off than non-disabled individuals, and in many cases consider them-
selves to be perfectly healthy. Experiencing the world as an autistic person, for
example, need hardly be a bad thing, and deafness need not constitute a state
of ill-health. Yet for many others, the opposite conclusions seem self-evident:
it is bad to be unable to walk, or hear, or see, or to lack an intuitive sense of
what other people are thinking and feeling; a life with these features is, all else
being equal, worse than a life without them; and it is necessarily impossible to
be both healthy and disabled if health entails ‘optimal or normal functioning’
and disability ‘sub-optimal functioning’.¹
Whilst this debate is often framed in terms of disability, the core dispute in
fact concerns the relationship between well-being and impairment. Recall
that disability is widely understood as a restriction in our ability to perform
important tasks. These tasks might be those an individual is ‘likely to be
required to do or would benefit from doing’,² those that would increase her
welfare,³ or, in my view, those they are entitled to be able to perform as a
matter of justice. In any case, on a wide range of views disability is definition-
ally disadvantageous. More controversial is the status of impairment.
Impairments are physical or cognitive features that cause deviation from a
biological or statistical norm of functioning for our species, usually taken to

¹ Bickenbach 2013: 826. Also see: Boorse 1975, 1977; Daniels 1985; Kahane and Savulescu 2016,
2009; McMahan 2005; Savulescu and Kahane 2009; Singer 2005.
² Buchanan et al. 2000: 287. ³ Kahane and Savulescu 2009.

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0006
  ,   129

mean suboptimal functioning. Indeed, even those who defend a neutral

account of disability, or argue that disability is compatible with health, none-
theless accept that impairment implies deficiency, shortfall, or limitation.⁴
Restricted vision, for example, is an impairment, whilst the inability to read
written language might be part of the associated disability.
Impairment is often what non-disabled individuals have in mind when
considering the ‘badness’ of disability: they wonder what it would be like
if their functioning capacity changed, and their conclusions are often
pessimistic.⁵ Disabled individuals, too, tend to focus on functional difference
when defending their condition. There is nothing inherently worse, they
contend, about having different forms of sensory, motor, or cognitive func-
tionings, though in certain contexts these can generate serious restrictions.
Yet the negative understanding of impairment—indeed, the very word
itself—seems to lend support to those who take a negative view of disabled
life, bolstering the widespread assumption that impairments constitute bar-
riers to a normal life, and so ought to be eliminated if possible. Further, the
negative conception of impairment is also in danger of undermining both the
coherence of disability pride (why be proud of a deficiency?) and the possi-
bility of solidarity (dividing the ‘genuinely’ deficient from those merely per-
ceived as such, as we will see). Finally, beyond an intuitive sense of where the
boundaries of this category fall, it is notoriously difficult to consistently
delineate which forms of functioning are the suboptimal ones.
These four worries have led some theorists to simply abandon the concept
of impairment, and so abandon, too, the central distinction between disability
and impairment (§5.2). Instead, I defend both the value of having a concept of
impairment, and its distinction from disability (§5.3). However, I argue for a
neutral account of impairment, as atypicality rather than deficiency. On this
view, impairment is a broad church, and may include conditions that are
relatively inert or even beneficial in our current context. Disabilities, therefore,
are limitations resulting from different, not deficient, ways of functioning.
I argue, first, that the conception of impairment plays an important role in
our theorising about disability and, second, that many objections to it can be
side-stepped by moving from deficiency to difference (§5.4).
The move to a neutral conception of impairment may seem to be a
straightforwardly positive change from the perspective of the disability rights

⁴ E.g. Aas 2016: 1266–7; Hull 1998: 209–10; Silvers 2003; §1.4.
⁵ Although there is, of course, ample evidence that contrary to the expectations and presumptions of
non-disabled people, ‘the disabled are often about as happy as healthy people are’ (Moller 2011: 189).
130      

movement and the cause of disability pride. Disability is no longer associated

with deficient physical or cognitive forms or processes, but merely different
ones. Thus, there is no question about the potential irrationality of being
proud of something that is, by definition, worse. Yet, on the other hand, the
broadening of the underlying notion of impairment might seem to render the
category of disability itself overly broad, diluting the sense of identity and
solidarity shared by a group with a specific set of experiences, and undercut-
ting its social usefulness as a banner under which to fight against injustice,
exclusion, and rights violations. In response to these worries about over-
inclusiveness, in §5.5 I argue, first, that the fact that impairment will indeed
be an expansive grouping need not imply that the same must be true of
disability, and second, that solidarity does not require a fixed and determinate
line between ‘insiders’ and ‘outsiders’. Thus, the benefits of my justice account
of disability as a means of identifying and responding to injustice are by no
means lost by pairing it with a neutral account of impairment.
I end by considering what kind of difference the impairments usually
deemed disabling really are: bad difference, mere difference, or, as I shall
contend, complex difference (§5.6 and §5.7). To explain that contention,
I should clarify from the start that I distinguish disadvantage, understood as
a lack of what we are entitled to, from harm, understood as what negatively
impacts individuals’ well-being. By saying impairments are complex differ-
ences, then, I mean that they are not necessarily harmful or disadvantageous,
though they may be. In addition, whilst disabilities are, of course, disadvanta-
geous by definition on my view, they need not be harms. Thus, from the outset
it should be emphasised (again) that in the vast majority of cases individuals
with impairments have, or can be enabled to have, a wide variety of oppor-
tunities. This is especially so if our focus is on enabling control over our
functioning at a particular level, rather than in a particular form: individuals
with many mobility impairments can often achieve high levels of mobility,
even if not the ability to walk; individuals with ASC might form meaningful
intimate relationships, even if unable to engage in neurotypical forms of social
interaction; deaf individuals can communicate even if they cannot speak, and
so on. There is, then, nothing desirable about normalcy in itself, nor any
reason to think that atypical functioning is necessarily incompatible with
leading a good life.
Returning to the questions that opened this chapter, it is clear that none of
these should be given universally positive or negative answers. As previous
chapters have argued, in the context of justice our focus should be on the
restrictions in the central opportunities that people face in particular contexts.
  ,   131

Whether an individuals’ impairment disadvantages them, and whether it

lowers their well-being, will depend on the person and their circumstances.
Given the complex effects of individual difference, we should aim to eliminate
the subset of these effects that are disadvantageous, not attempt to eliminate
difference—that is, impairment—itself. Rather, we should accommodate it.
A neutral account of impairment best enables this approach.

5.2 The Debate about Impairment

As Chapter 1 considered, it is standard in much work on disability to distin-

guish impairment, seen as some physical or cognitive deficiency, from disabil-
ity, as a restriction in the ability to perform tasks. Impairment is seen either as
directly causing such restriction (on the medical model) or to result in such
restriction due to external features, such as exclusionary or discriminatory
infrastructure and social norms (on the social model), or, more plausibly, a
combination of the two. We should clearly reject the suggestion that impair-
ment, unmediated by context, is the exclusive cause of disability. Yet given the
essentially embodied nature of disability, we also cannot trace all the disad-
vantages associated with disability to social causes, nor should we expect that
disability would necessarily be eliminated in a fully just society.⁶ For example,
deficiency in the ability to hear is an impairment that is associated both with
various direct restrictions, such as being unable to listen to music or hear
voices, and with various socially mediated forms of exclusion, such as being
unable to go to cinema because of a lack of subtitles or hearing aid loops, or
not being shortlisted for jobs due to prejudice or lack of workplace accom-
modations. Thus, disability should allow for the inclusion of the experiences of
both atypical functioning (i.e. of being impaired) and of being marked for
exclusion or living in an environment poorly suited to one’s form of function-
ing (i.e. of being impaired in a given context), when these restrictions are
disadvantageous.
So far, so familiar: impairment is an individual’s physical or cognitive state,
and disability comprises certain direct and indirect consequences of impair-
ment. Yet it remains to be asked why this language and this distinction is so
widespread, and why it might be helpful. Most obviously, this was a wise
tactical move by activists fighting against the medicalisation of disability,

⁶ Indeed, proponents of the social model (of which I am not one) need not be committed to either
claim, despite common caricatures of this view (see Chapter 1 for extensive discussion and references).
132      

and against the focus on ‘cures’ over accommodation. The approach was
‘politically liberating’ because it ‘provided a basis to question the dominance
of medicine in explaining disability’.⁷ The rhetoric of claiming that ‘disable-
ment is nothing to do with the body’⁸ also helped to draw attention to the ways
in which individuals’ opportunities can be restricted by the poor fit between
norms and infrastructure, on the one hand, and some forms of functioning, on
the other. This helped to shift the focus away from ‘fixing’ individuals, and
towards the problems that certain social and environmental structures caused
and the ways they could be modified to avoid those problems: focusing on
accommodation and barrier removal, and not simply attempting to find cures.
This conceptual framework constitutes a significant theoretical contribution
from within the disability rights movement, reflecting the self-conception of
many disabled individuals. It has had, and continues to have, a substantial
impact on how disability is understood. In particular, it has helped to under-
mine the assumption that disabled bodies are, in some sense, defective. It is an
important tool for drawing people’s attention to the various external causes of
disability, to the importance of accommodation, and to the core problem of
‘fit’ between our bodies and minds and prevailing social norms and
infrastructure—something rarely reflected on by those who do not experience
dissonance between their physical and cognitive functionings and their con-
text. All of which should give us strong reasons to think that this distinction
should not be lightly discarded. Nonetheless, we should still ask whether
distinguishing impairment and disability actually provides conceptual clarity
or improves the precision of discussion in this area, especially given that some
scholars, notably Elizabeth Barnes, have forcefully argued that it does not.
Barnes’s first objection to the concept of impairment is that it merely
functions as a way of capturing the intuitive distinction between forms of
socially mediated disadvantage—such as sexual orientation, or our gender or
racial identity—and those which disabled individuals experience, but without
actually providing a justification or explanation of this intuition.⁹ There is
clearly an intuitive distinction between the stigmatisation of a body for its
perceived feminine characteristics or for exhibiting a phenotype associated
with a particular racial category, and the stigmatisation of a body for its
perceived physical or cognitive deficiencies. There are clear reasons, too, why
we would want to distinguish these phenomena, and avoid conflating the

⁷ Vehmas and Riddle 2020: 349. Also see: Shakespeare and Watson 2001; Shakespeare 2014, and
§1.8.
⁸ Oliver 1996: 35. ⁹ Barnes 2018: 1156–8.
  ,   133

experiences of all such stigmatised groups into a single category—describing all

as ‘disabled’, perhaps. However, Barnes insists that the promise the concept of
impairment gives of grounding the distinctiveness of the disabled experience is a
false one: it does no more than appeal directly to our intuitions about which
individuals are disabled. In her view, no account of impairment can be provided
that will satisfactorily distinguish amongst the various causes of disadvantage, so
instead we simply appeal to our intuitions to decide difficult cases by fiat:
homosexuality is not an impairment, for example, and ASCs are.
Further, the intuitions to which we appeal concern our judgements about
disability. Thus, Barnes maintains that there is no difference between the
judgement that ‘x is an impairment’ and that ‘someone with x is disabled in
context C’.¹⁰ What matters are the rules used to judge who is disabled, and not
the physical (or cognitive) conditions experienced by those to whom these
rules apply.¹¹ Thus, judgements about disability do not depend on an account
of impairment. Rather, impaired individuals are identified on the basis of a
prior judgement about who is disabled. For example, paraplegia will constitute
an impairment only if someone is disabled by it in a given context. Recall that,
for Barnes, paraplegia is a disability if it is amongst the conditions the
disability rights movement are seeking to promote justice for, given their
rules about making solidarity judgements.¹²
Barnes’s first objection, then, is that there is no explanatorily and exten-
sionally adequate account of impairment that can ground our judgement that
someone is disabled. On the contrary, our judgements about disability ground
our claims about impairment. As such, she suggests that we abandon talk of
impairment and focus instead on determining what actually grounds our
judgements about disability. In other words, she believes that all important
philosophical work can be done by talking in terms of disability alone.
Barnes’s second objection is that drawing a distinction between disability
and impairment will lead to an overly disembodied view of disability. As noted
above, a plausible account of disability must incorporate both the exclusion,
stigma, and disadvantage associated with one’s physical and cognitive atypi-
calities, and the more personal and direct experience of living with such
difference. If impairment is used to capture the embodied element of the

¹⁰ Barnes 2018: 1156.

¹¹ Barnes only discusses physical, not cognitive, disabilities, hence the parentheses. I am concerned
with both (and, indeed, am not convinced they can be straightforwardly separated, e.g. see Shepherd
2020: 133).
¹² Barnes 2016a: 43–38; Barnes 2018: 1155. See §1.7.
134      

disabled experience, disability might seem to simply refer to exclusion, ignoring

the ‘messy reality’ of having a disabled body (or mind).¹³

5.3 In Defence of Impairment

I agree with Barnes about both the importance and difficulty of devising an
explanatorily adequate account of impairment. It is this that motivates my
adoption of a neutral account, as the subsequent two sections will explore.
However, I disagree that there is no meaningful distinction between saying
that something is an impairment and saying that this leads to disability in
some context, and I disagree, too, that introducing a concept of impairment
necessarily divorces the physical and cognitive reality of the disabled experi-
ence from an account of disability.
Beginning with the former point: Barnes might be right that some of the
conditions we most frequently think of as being impairments are those that
result in disabilities in contexts we are familiar with. Thus, the paradigm cases
of impairment—such as restrictions in sensory functionings or mobility,
learning difficulties, or limited social functionings—are those that frequently
lead to disabilities in our society. However, the fact that someone is disabled as
a result of impairment x in one context does not demonstrate that x is
universally disabling for all individuals in all contexts. Indeed, it may not (or
may no longer) be disabling for individuals in any context. Yet this does not
seem to alter our judgement that the form of functioning is an impairment.
Myopia, for example, is not disabling in our society, given easy access to
glasses, contact lenses, corrective surgery, and environmental modifications,
but it might be in other places, or at other times, where no such solutions are
or were available. We would not, therefore, describe this ‘x’ as a disability in
our current context (or perhaps any context), though we would reasonably
describe it as an impairment.
This suggests that our judgements about whether something is an impair-
ment are not reducible to whether it is currently disabling. Thus, Barnes seems
descriptively wrong to insist that our intuitions about disability are prior and
do all the work of identification, given there are many cases in which this is not
so. Rather, we appear to be able to make independent judgements about what
constitute impairments without determining their association with disability.
This is so even when we focus on Barnes’s account of disability and restrict our

¹³ Barnes 2018: 1158.

  ,   135

attention to rule-based solidarity judgements: it is hardly a stretch that no

disability rights movement employs rules that imply they should be fighting
for justice for short-sighted people (on the plausible assumption that this is not
a group suffering from injustice), yet this does not seem to undermine the
strength of our intuition that these individuals are impaired.
Of course, our intuitions might be wrong. Indeed, insofar as impairment
implies deficiency I would contend that our intuitions are often mistaken. The
point here is only to note that our intuitions and judgements about these two
concepts can be independent. I also believe there are good reasons to think
that they should be maintained as independent concepts. Two of these reasons
have been mentioned already. First, it can be a useful means to avoid over-
medicalisation and a narrow focus on ‘cures’ or elimination of difference.
Second, it can be a way of distinguishing disability from other forms of socially
mediated disadvantage. The above discussion now suggests a third: it is useful
to be able to describe the shared experience of atypical functioning of those
who have some condition—or impairment—whether they are disabled by it
or not.
Thus, all individuals with myopia, or paraplegia, or dyslexia have an
impairment, and have some shared experiences of living with their impair-
ment, though not all have the experience of being disabled by it. For example, a
well-off dyslexic individual in our society, with access to high-quality and
appropriately modified educational institutions, may not be disabled by their
impairment any more than a myopic contact lens wearer is disabled by theirs.
Nonetheless, they may share significant experiences with less advantaged
dyslexic individuals whose condition was not diagnosed or appropriately
accommodated. For example, both may experience difficulties in using
‘letter-sound relationships (phonics) in combination with context to identify
unknown words’,¹⁴ or embarrassment or discomfort at disclosure.¹⁵ Of course,
one of these individuals has much better methods in place for dealing with
such experiences, but the experiences of such forms of functioning are none-
theless shared in a way to which a non-dyslexic individual can have no direct
access. Thus, it seems to be worth identifying instances and experiences of
atypical functioning, even if the connection to social disadvantage is severed in
some contexts. In other words, it seems worth identifying impairment as a
category.
However, this might seem to raise the spectre of disembodied disability—
Barnes’s second worry. She might object that the experience of functional

¹⁴ Schatschneider and Torgeson 2004: 758. ¹⁵ Alexander-Passe 2015.

136      

difference is now solely encapsulated by the concept of impairment, and

disability refers only to social aspects, such as stigma and exclusion. But this
is not so. Impairment captures the idea of having a body or mind that functions
atypically (on standard views, deficiently), and disability captures what it
means to have such a body or mind in some particular context. In other
words, disability should be understood as some possible consequences of
having an impairment—specifically, on my view, those consequences that
restrict our access to opportunities we are entitled to—and this will include
elements of our bodily and cognitive experiences. Thus, individuals are not
disabled merely because they are unable to function in a normal mode (cannot
walk, for example), but because they function at a significantly different level
and, specifically, at a level below that to which they are entitled (for example,
lacking control over their mobility).¹⁶
Given that our bodily and cognitive features are always mediated by our
environment, what it means to have these features will always depend to some
degree on the context in which they are experienced. As such, there is no non-
contextual answer to the question of what impact any impairment will have, or
to whether some way of functioning is good or bad—though some may be
better suited to more circumstances or to circumstances we have a higher
probability of facing. Arthritis, for example, is a painful condition that fre-
quently makes life difficult for those who have it. Nonetheless, there is no
answer to exactly what this impairment involves—neither what its impact will
be nor the normative valence of these outcomes—that is independent of
external factors, such as climate, terrain, standard methods of travel, available
medicine, and technology. Some impairments may make life more difficult or
complicated in most imaginable contexts (though we should be cautious of
limiting the scope of our imagination), but its consequences are nonetheless
dependent on that environment. Disability is necessarily a contextualised
concept on my view (and ought to be, if it is taken to have normative valence).
However, the impact of context should not be narrowly construed as merely
stigma and exclusion, as subsequent sections will consider further. For now,
my goals are only to defend the meaningfulness and usefulness of distinguish-
ing impairment and disability—that is, between certain ways of functioning
and the context-determined impact of doing so—and to demonstrate that
drawing this distinction does not imply that disability is disembodied.

¹⁶ My focus is the specific context of justice, however I am happy to allow that there might be
contexts in which it is expedient to draw attention to a different set of the externally mediated effects of
impairment, and so employ a difference conception of disability.
  ,   137

5.4 Impairment as Atypicality

In the preceding discussion, impairment was generally described as constitut-

ing atypical forms of functioning. However, it is more ordinarily understood
as a ‘physical, mental or sensory functional limitation’,¹⁷ a ‘sub-optimal
functioning’,¹⁸ a ‘significant shortfall from biologically normal functioning’,¹⁹
or ‘a dysfunctional bodily state that limits a major life activity’.²⁰ In other
words, impairment is usually synonymous with deficiency or dysfunction. It is
assumed that if we do not restrict the relevant atypicalities to dysfunctions—
indeed, often those dysfunctions that significantly limit major activities—then
the category of impairment will be too broad, and thus so too will be the
associated category of disability. In particular, it might seem that a broad
account of impairment will imply that people can qualify as being disabled as a
result of functioning that is ‘better’ than average, or due to latent or hidden
pathologies. Further, that it might prevent us from distinguishing disabilities
from non-pathological conditions with similar effects, or from other features
of identity, for example, minority sexual preferences, sex, racial identity, or
some forms of gender dysphoria.
I will respond to these worries about over-inclusiveness in the next section.
Here, I will consider the benefits of an inclusive, neutral approach, which
avoids having to identify which of the various atypical ways of functioning are
the defective, dysfunctional, or ‘worse’ ways—worse, that is, independent of a
specific context. There are three key benefits to the neutral approach. First,
suggesting there are independent criteria that can pick out the abnormalities
that are simply ‘worse’ or dysfunctional implies that normalisation is the only
appropriate solution to such dysfunction. The neutral approach does not have
this assumption in favour of normalisation. Second, whilst it is uncontrover-
sial to note that some forms of physical and cognitive functioning do not fit
well in certain environmental contexts, attempting to determine which ways of
functioning are non-contextually worse is far more difficult and problematic.
This is not a task that the neutral approach need attempt. Finally, a conception
of impairment as dysfunction encourages a distinction in the disabled com-
munity between those who are ‘really’ impaired (hence, defective), and those
who are wrongly perceived as such. The neutral approach will not draw this
distinction.

¹⁷ Hull 1998: 209. ¹⁸ Bickenbach 2013: 826.

¹⁹ Aas 2016: 1266. ²⁰ Howard and Aas 2018: 1128.
138      

All of these points require elaboration but, before providing it, I will make a
quick note on terminology. Thus far I have retained the language of impair-
ment, given that it is widely used and I wish to avoid the unnecessary
proliferation of terminology, and to avoid weakening the link to the existing
literature on impairment (see §1.2). This has also allowed me to remain
inclusive of both impairment-as-deficiency and impairment-as-difference
views. However, now I am explicitly defending a neutral account of impair-
ment there are reasons to use a different vocabulary to avoid having to
repeatedly specify which conception of impairment I have in mind. This
rationale is strengthened by the fact that the negative etymological root of
‘impairment’ means the negative connotations are difficult to shake off. Thus,
I will henceforth treat ‘anomaly’ as synonymous with ‘neutral impairment’.
First, then, the medicalisation of disability and the language of normalcy
have long been used to stigmatise disabled individuals and to locate the
‘problem’ of disability as internal to them. This does not follow from the
mere use of a concept of impairment, but from its negative construal. For
example, Jerome Bickenbach has defended an account of impairment as
dysfunction according to which having an impairment means being in ill
health. He insists that there is no reason to resist this move, and that it can
seem a ‘mystery’ why someone with an impairment ‘would ever deny, or feel
uncomfortable being told, that their impairment is a health problem’.²¹
Indeed, he claims that understanding impairment as mere human variation
will ill-serve the interests of disabled individuals for whom ‘access to
impairment-related health care resources is an important human rights
issue’.²² These rights claims, he argues, cannot be grounded unless we allow
that disability is associated with ‘health problems’.
Bickenbach does acknowledge that states of ill health—and, we might add,
states of neediness and vulnerability more generally—can be stigmatised. His
proposed solution is that we try to ‘disabuse people’ of this view.²³ Leaving
aside the difficulty of eliminating such stigma (which Bickenbach fails to
acknowledge), the problem with the medicalisation of impairment is not
merely attitudinal. It is also the associated implication that the appropriate
response to disability must involve eliminating or curing the underlying
suboptimal functioning and seeking normalcy: supplying cochlear implants
to remove the dysfunction of deafness, for example, or encouraging individ-
uals with ASC to suppress repetitive and self-soothing behaviours, and other
forms of atypical functioning. As already discussed, our goal should not be to

²¹ Bickenbach 2013: 821. ²² Bickenbach 2001: 50. ²³ Bickenbach 2013: 827.

  ,   139

restore a normal form of functioning, but to ensure an individual can function

at an appropriate level, without associated pain, discomfort, or fatigue.
This is not to say that medical intervention—even medical intervention that
seeks to eliminate an anomaly—is always an inappropriate response. Indeed,
in the current context, our best solution to a number of anomalies may involve
removing them, and in some cases this may seem true for the foreseeable
future. For example, it is not unreasonable to focus on the wider provision of
drugs like Orkambi to treat cystic fibrosis, given the lowered life expectancy
associated with the condition and the fact that many of the symptoms cannot
easily be accommodated by structural changes.²⁴ The objectionable implica-
tion of dysfunctional accounts of impairment is the assumption that finding
and providing such cures should be the default solution in every case—the
only way to rectify the ‘decrement of health’²⁵ impairment entails. The neutral
account does not assume impairment is a problem, thus need not assume we
should always seek to remove it. However, it also need not deny that impair-
ments can be a problem.
The second problem with the negative construal of impairment is that it
may not be possible to identify criteria that will satisfactorily determine the
boundaries of this category: the ways of functioning that are non-contextually
worse. The most obvious approach is to appeal to ‘normal species functioning’,
classically defined as the statistically typical contribution for an organism’s
reference class of their parts or processes to their survival and reproduction,
with a reference class being an age group of a sex of a race of a species.²⁶
However, despite the claims of its proponents, this does not generate an
objective, uncontroversial account of impairment as dysfunction.²⁷
First, as Elselijn Kingma has argued, much hangs on the selection of an
appropriate reference class.²⁸ For example, not experiencing menstruation
would constitute a deviation from normal functioning for a young adult
female, but would be normal for a pre-pubescent or post-menopausal female,
or for a male. Yet though the use of reference classes seems plausible in many
cases, Kingma demonstrates that Boorse’s appeal to ‘a natural class of organ-
isms of uniform functional design’²⁹ provides no objective justification for his
selection of reference classes since diseases (and impairments) can occur in

²⁴ For some of the controversy surrounding the licensing of Orkambi see e.g. Cystic Fibrosis Trust
2019; Wise 2019; Boseley 2019.
²⁵ Bickenbach 2013: 826. ²⁶ Boorse 1977: 555.
²⁷ For example, Kahane and Savulescu (2012), in arguing for the derivative moral importance of
statistical normality take it as obvious that we can distinguish deviations from the statistical norm in the
negative and positive direction. For a critique of their argument, see Reynolds (2018).
²⁸ Kingma 2007. ²⁹ Boorse 1977: 562.
140      

nature, be common, be of uniform design, be written in our genes, and

be maintained by natural selection. Why, then, should race, age, and sex be
considered characteristics of natural reference classes, whilst Down’s syn-
drome, blindness, or sickle-cell anaemia are not?
Ultimately Kingma’s point is similar to Barnes’s, though in the context of
defining health rather than disability: there is no empirical, value-free justifi-
cation for the claim that a post-menopausal woman is normal for their
reference class, as opposed to being an impaired member of a broader class
of post-pubescent women, whereas an autistic person is an impaired (or non-
healthy) member of the neurotypical reference class, as opposed to a normal
member of an ASC reference class. However we cash out the idea of deviation
from normal species functioning, it will not cover all the cases we would wish
unless we import some additional assumptions or considerations, which then
do the actual work of determining what counts. Thus, the mere concept of
unhealthiness (for Kingma), or impairment (for Barnes), as negative devi-
ations from normal species functioning offers little independent guidance.
Barnes has also outlined some of the implausible implications were any of
the candidate theories of deviation from normal functioning employed. She
argues that appeals to functionings that promote ‘survival and reproduction’,
to abilities that most people have, or to inabilities that will have a substantial
impact on daily life, will all generate counterintuitive cases. After all, Michael
Phelps’s marfanoid habitus, which gives him the lanky physique that makes
him an excellent swimmer, also puts him at greater risk of cardiac problems,
and so does not promote survival. Further, being able to roll our tongue is an
ability most people have, and being a petite woman might have a substantial
impact on our daily life, but neither the inability to tongue-roll nor petiteness
would usually be considered an impairment.³⁰ Barnes raises these examples as
cases that cannot plausibly be considered disabilities, though she is similarly
sceptical about the possibility of identifying the ‘feature(s) bodies [or minds]
with impairments have in common’,³¹ and these cases therefore function
equally well as counter-examples to deficiency accounts of impairment.
It may be responded that this line of criticism simply shows that existing
accounts are flawed, and not that the development of an account of relevant
dysfunction is impossible. However, we can see that the problem is deeper
than this when we recall that impairment is a decontextualised concept,
concerning the nature of the body and mind itself and not its interaction
with a particular set of circumstances. We cannot determine whether

³⁰ Barnes 2016a: 13–20. ³¹ Barnes 2016a: 25 (my emphasis).

  ,   141

particular ways of functioning are necessarily worse without knowing more

about the context in which they will be exercised. The environment is not
something to be layered on, which might help or hinder an already defective
body. Rather, it determines what constitutes a defect. This is not undermined
by the observation that some forms of functioning are likely to be disadvan-
tageous in many prospective contexts—perhaps any conceivable context. Even
seemingly easy cases, such as depression or arthritis, are not necessarily,
universally, and context-independently worse. For example, for a young man
in Britain in 1916, having arthritis that exempted him from military service
may have been significantly advantageous. This does not imply a commitment
to the implausible claim that being in pain, experiencing depression, or having
lowered life expectancy are not bad things—merely to the much weaker claim
that our assessment of the overall status of any anomaly must always depend
on the context in which it is experienced.
Given the impossibility of such abstract determinations, the content of the
category of impairment-as-dysfunction is in fact determined by judgements
about the kind of functionings we believe it is worse to have, focusing on what
seems disadvantageous in familiar contexts, or on features that tend to be
stigmatised. Thus, far from being objective, the category is constructed on the
basis of prejudice and unexamined assumptions.³² The judgement that being
unusually short is an impairment (that is, defective) whilst being unusually tall
is not, for example, is not grounded in some feature that short stature shares
with blindness, and Down’s, and paraplegia, but not with being tall. Such
assumptions are made all the more dangerous by masquerading as a value-
neutral, scientific category. If, however, we no longer see impairments as
deviations from the norm in a certain (negative) direction, but any form of
variation or atypical functioning—as anomalies—then these problems are
avoided. We need not claim to have found the unifying feature identifying
all and only negative impairments, and since we are not concerned with the
negative there is no room for prejudicial assumptions about which forms of
functioning are worse nor any need to restrict our focus to paradigm impair-
ments (usually those that are currently disabling in familiar contexts).
The third and final reason to adopt a neutral account is that it provides a
means of responding to a worry that might be raised about under-inclusive-
ness: namely, that if impairment is deficiency, then those whose functioning
cannot plausibly be described as deficient will not be considered disabled.

³² See e.g. Amundson 2000; Reynolds 2017, 2018; Shakespeare 2014; Tremain 2001, 2002;
Wasserman et al. 2016a.
142      

This is a problem because some apparently paradigm cases of disability do

not seem to be associated with a state that can plausibly be described as
an impairment in the negative sense. Perhaps the clearest example of this is
individuals with some (though certainly not all) forms of ASC, many of whom
strongly reject the moniker of deficiency in favour of neurodiversity.³³ To
accommodate such cases, thinkers as diverse as Bickenbach and Sean Aas have
suggested that individuals might be disabled even if they do not have an
impairment, as long as their condition is perceived as, or treated as if it were,
an impairment.³⁴ Aas, for example, contends that ‘to be disabled is to not be
medically needy; but rather, to occupy a social position defined in part by
false/misleading/threatening representations of medical need’.³⁵ The sugges-
tion is that these individuals share at least the socialised element of the
disabled experience—being marked for exclusion, being stigmatised, and
the false perceptions of their condition—even if not the embodied, impair-
ment element.
The problem with this approach is that it leads to a distinction between
those who are ‘really’ dysfunctional, and those who are merely perceived as
such. First, it is not clear on what basis a deep distinction can be drawn
between, say, blindness as dysfunction and autism as difference. Both are
ways of functioning differently that might be correlated with disadvantage.
Both can be valued by their possessors and are compatible with living a good
life. Thus, it is unclear why some individuals’ refusal to conceptualise their
form of functioning as a deficiency is taken more seriously than that of others.
Yet the more significant problem is not the difficulty of identifying who is
deficient and who merely atypical, but rather the fact that attempting to do so
is unnecessarily divisive and misleadingly obscures an element of shared
experience. Contrary to the claims of Aas and Bickenbach, it is not merely
the social element of the disabled experience that those who are socialised as
disabled share with those who are, ostensibly, truly impaired. They also share
the experience of functioning differently: of being in a functioning minority.³⁶
This experience is shared with individuals with paradigm ‘defective’ impair-
ments, such as sensory loss or mobility restriction. Like neurodiverse individ-
uals, many do not experience their functioning as anything other than atypical.

³³ See e.g. Baron-Cohen 2017; Begon and Billington 2019; Davidson and Orsini 2013; Kapp et al.
2012; Robeyns 2016; Sinclair 2012.
³⁴ Aas 2016; Howard and Aas 2018; Bickenbach 2013: 826. Indeed, the ADA has taken this approach
(see Francis 2018 for discussion).
³⁵ Aas 2016: 1273.
³⁶ This is a common theme in much of the writing by individuals with ASC (e.g. Grandin and Panek
2013; Higashida 2013; Prince-Hughes 2002; Williams 1999).
  ,   143

Yet the sense that others do not experience or move around in the world as we
do is a significant part of the embodied experience of disability, distinct from
how other people respond to, misunderstand, or are prejudiced against,
unusual forms of functioning.
Further, narrowly restricting the socialised element of the disability of those
who are apparently not genuinely impaired to false, misleading, or threatening
representations of their condition and their neediness fails to capture the
difficulties and complexities that comes with being in a functional minority.
This may include awkward or uncomfortable reactions to one’s unexpected or
unusual behaviours, which may be borne not out of exclusion, stigma, or false
perceptions but merely lack of familiarity. Finally, then, in addition to the
difficulty of making the distinction between the truly dysfunctional and those
merely perceived to be so, and the failure to account for important elements of
shared life experience across the divide, reducing the disability of the neuro-
diverse, and perhaps others, to merely the false and threatening social reac-
tions to their condition, undermines and ignores the embodied part of their
experience. In a more just world, individuals with ASC may not be represented
misleadingly and, as a result, some of them may no longer be disabled. Yet they
would still be impaired in the sense of functioning atypically, and would
continue to face the challenges associated with this. This can be captured
with a broader, neutral conception of impairment—or anomaly.

5.5 The Over-Inclusiveness Objection

My contention, then, is that impairments should be understood as atypical

functionings—anomalies—which may be good or bad (depending on their
impact on quality of life or individual welfare) or disabling or not (depending
on whether they restrict access to opportunities we are entitled to) in particular
contexts. Impairment itself, therefore, becomes a broad church, which may
include conditions that are relatively inert or even beneficial in our current
context. On the one hand, it may seem that this approach would be welcomed
by disability activists. If disability is no longer associated with deficient
physical or cognitive forms or processes, but merely different ones, then
there can be no suggestion of irrationality for being proud of a feature that
is, by definition, worse, nor an assumption that our goal must be the elimin-
ation of these conditions. Yet, on the other hand, the underlying notion of
impairment might now seem to have been rendered too broad, undermining
disability pride by diluting the sense of identity and solidarity shared by a
144      

group with a specific set of experiences. In particular, this might seem to

leave us with no way of excluding from the category of disability those
who: (i) function better, not worse; (ii) have latent or hidden pathologies;
(iii) function in ways with similar effects to genuine disorders, but without
having those disorders; or (iv) experience exclusion and stigma as a result of
their race or gender identity, sexuality, or some other socially mediated vector
of disadvantage.
Before turning to analyse these four cases in detail, I would dispute the
general claim that an inclusive account of impairment must undermine
disabled pride and solidarity. First, since disability involves anomalous func-
tioning that, in a particular context, results in the loss or restriction of
important opportunities, then even if many people will be deemed impaired
they will not also be disabled unless these consequences follow. Thus, the
inclusiveness of the category of impairment is a separate question to the
inclusiveness of the category of disability. The scope of the latter category
will depend on the restrictions that are deemed disabling (as Chapter 2
discussed). According to my own account, many anomalies will not generate
disability (as the restriction in access to the capabilities individuals are
entitled to).
Moreover, there are positive reasons to allow the disability rights movement
to be a relatively broad church, and reasons, too, not to believe that impair-
ment will become meaningless unless its boundaries are carefully policed. An
inclusive approach captures the shared experiences of those who once were,
but are no longer, disabled though are still impaired (perhaps because their
impairment is now effectively accommodated), or those who anticipate
becoming disabled in future (perhaps because they have a degenerative illness
that will likely result in a disabling impairment, or because they fear their
impairment will cease being accommodated in the future). A further case here
might be individuals who plan to undergo a form of body modification that
will likely restrict their opportunities such that they are disabled: body integ-
rity identity disorder (BIID), wherein an individual seeks to amputate a
healthy limb, is the obvious example.³⁷ One interpretation of such cases is
that these individuals are not, but plan to become impaired (and, perhaps,
disabled). A more plausible reading, at least for those who experience body
dysmorphia and a strong sense of discomfort and unease at the possession of a
limb they dissociate from, is that they are currently impaired (and may be
disabled), and plan to eliminate one anomaly (body dysmorphia) by acquiring

³⁷ See §4.5 for discussion and references.

  ,   145

another (the loss of a limb), which may render them more or less disabled,
depending on context.
To emphasise, if individuals currently lack an anomaly that relevantly
restricts their abilities, then they will not be disabled. Nonetheless, there
may be considerable elements of shared experience amongst this wider
group, and there is little reason to think that acknowledging this will pose
a threat to the cogency of disabled pride and disabled identity. For example,
there is no suggestion that provisions for disabled individuals should be
opened up to the wider community of the impaired. On the contrary,
individuals only have a claim to additional resources on the basis of their
different functioning if this actually results in a restriction to their entitle-
ments. A neutral account of impairment simply allows for the distinct
recognition of functional difference.
These worries seem especially misplaced given that the disability rights
movement is a way for individuals who are excluded and stigmatised as a
result of their atypical functioning to fight for their rights and inclusion.
Hence, it seems very unlikely that people will attempt to claim solidarity on
the basis of atypicalities with few significant consequences, and which do not,
therefore, disable them. On the contrary, individuals are inclined to shrug off
both the mantle of impairment (as in the ASC case) and disability. Students
with dyslexia, for example, may report feeling uncomfortable about being
offered assistance from disability support services, and many successful dis-
abled individuals insist that they ‘do not really consider themselves to be
disabled’. In part, no doubt, this results from the inappropriately and often
unjustifiably negative connotations of both terms.
Simple categorisation in this domain is so difficult exactly because disability
constitutes such a diverse group, yet with certain core, shared experiences (of
atypical functioning and resulting exclusion). We should allow everyone who
has these experiences to have a voice in determining the impact their atypical
functioning has on their life, and should be cautious of excluding individuals
simply because their experiences are not like our own or do not fit into the
paradigm model of the disabled experience. This is not to say that any spurious
claim should be accepted—there are criteria for inclusion as both impaired
and disabled, which give guidance on when these terms are being inappropri-
ately applied. Nonetheless, an openness to shared experiences in perhaps
surprising places need not be antithetical to solidarity. These points can be
illustrated by examining the four kinds of cases that seem in danger of being
wrongly categorised as instances of disability because they will be categorised
as impairment on the neutral view.
146      

5.5.1 Functioning Better

If we no longer require impairment to involve functioning in a way that is

in some sense substandard, then does this mean that individuals with better
than normal functioning might count as being impaired? It will. And though
counterintuitive given the etymological roots of the term, this implication
should not be resisted. It was the very difficulty of discerning a meaningful
way of distinguishing (context-independently) better or worse functionings
that in part motivated the move towards a neutral account of impairment.
Indeed, when we turn to any particular case it becomes clear that few atypi-
calities can simply be classed as beneficial. Especially so if we attempt to make
this judgement without reference to context. The same atypicality that con-
tributes to Phelps’s swimming ability also increases his risk of cardiac prob-
lems; the ASC that might make someone better at processing patterns and
numbers might also make them worse at reading social cues, and so on. These
need not be cases of simple trade-offs—losing sight and gaining better hearing,
say. The consequences of any form of functioning are simply complex, as §5.6
will discuss.
However, assume that it is possible to imagine a person whose atypicalities
are positive in a decontextualised way: they have some combination of cogni-
tive and physical abilities that would be of benefit in any context—high
intelligence, athletic ability, physical beauty, and so on (assuming these qual-
ities can be defined without reference to context). It is true that on a neutral
account they would be deemed impaired, but now that this simply means
atypical or anomalous, this is hardly implausible in itself. More importantly,
they would never be disabled since the consequences of their atypicality are, by
assumption, solely positive. (Note that if we did not assert this by fiat, it is
improbable that any way of functioning would always, on balance, have a
positive impact: the highly intelligent might alienate, or feel alienated from,
those around them, the physically attractive (especially if female) may feel
objectified, reduced to their appearance, and not taken seriously, and so on.
The best we can say about desirable features or ways of functioning is that
they have a high probability of being beneficial, but the effects of all are
complex and cannot be assessed without knowledge of an individual and
their environment.)
What, though, about more realistic cases, such as Phelps or the high-
functioning autistic person? They too are impaired, and whether they are
disabled will depend on the restrictions this causes (and the restrictions we
take to be constitutive of disability). On the one hand, Phelps’s risk of cardiac
  ,   147

problems does not seem to render him disabled, as will be discussed in the
context of latent pathologies below. On the other hand, though, such an
autistic person might be disabled if the reactions to their behaviours, such as
a lack of eye-contact or stimming, mean that they are passed over for jobs for
which they are well qualified, or unable to form emotional connections or
social networks (and even if they consider their condition beneficial overall).
Yet in another, quite similar context, with different social norms and etiquette,
their anomaly might have a wholly positive impact on their life and oppor-
tunities. In neither case, though, does their impairment or anomaly necessarily
lead to their being categorised as disabled.
I suggest enhancement should be treated in the same way: enhancements
that entail anomalous functioning are impairments, and when they cause their
possessor to be disadvantaged, they are disabling. This does not imply all
enhancements are disabling, or even that all are impairments. As Kahane and
Savulescu have emphasised most enhancements move individuals within
‘the existing range of human capacities and dispositions’—normal range
human enhancement—and not beyond it—supranormal enhancement.³⁸ It
is likely that I define the range of normal more narrowly than they do, since
impairments are certainly within the existing range of human capacities
and dispositions—though, in Kahane and Savulescu’s view, at the ‘lower’
end of that range—yet constitute anomalies on my view.³⁹ The boundaries
of what constitutes normal are, as discussed, fuzzy and determined as much by
social expectations and norms as by human physiology. Nonetheless,
making a person with slightly below average intelligence, stamina, or concen-
tration slightly above average would probably not involve creating anomalous
functioning.⁴⁰ Interventions that ‘radically transform (even transcend) normal

³⁸ Kahane and Savulescu 2015: 133.

³⁹ Despite acknowledging that all changes in capacities have complex positive and negative effects,
Kahane and Savulescu (2015: 137–8) nonetheless seem confident that we can ask—and answer—the
question: ‘should we increase or enhance [property] X?’ The assumption that we can pretty clearly rank
any change in capacity as positive or negative is necessary for their commitment to the value of
enhancement and to procreative beneficence (Savulescu and Kahane 2009; Savulescu 2001). We need to
be committed to the view that paradigm impairments are overall bad for people, and qualities like
higher intelligence or stamina are overall good to make such judgements about how humans should, in
general, be modified (or selected). I dispute that such features can be judged non-contextually good or
bad (§5.6, §5.7).
⁴⁰ I say ‘probably’ because we can imagine a situation in which such a process is carried out over
generations until the average moves quite far from what it currently is, and expectations for success in
the job market and social interactions shift too. The non-enhanced, whose intelligence is at the previous
average, may now have anomalous functioning in this context, and so be impaired. Enhancement can
thus create impairment not merely among the enhanced, but in the rest of the population by shifting
our expectations (see §2.7).
148      

human capacities’⁴¹ clearly would involve the creation of individuals who

function anomalously, and hence have impairments on my view.
Is it unacceptably implausible to suggest that such individuals—apparently
‘perfected’—should be described as impaired and even disabled? Again, inso-
far as impaired means (and should ideally be replaced with) anomalous,
this does not seem in the least implausible. The difference between such
(imagined) individuals and the rest of humanity is what leads to the use of
labels like ‘posthuman’ and ‘transhuman’ in this literature.⁴² They are certainly
anomalous, then, but are they disabled? If proponents of enhancement are
right and these processes can be a means to ‘achieving an ideal of what
constitutes a good and flourishing person’,⁴³ then there will be no association
with disadvantage and so no disability. If, as seems plausible to me, we are
unable to identify qualities that are simply good or bad for us, and/or changes
have unanticipated consequences, and disadvantage does result, then it is
indeed possible that the enhanced will be disabled. For example, if genetic
enhancements to someone’s intellectual capacities render their abilities suffi-
ciently above the norm that they feel alienated from, and unable to relate to,
the rest of the non-enhanced population, and as such lack the ability to form
social relationships, then they would be disabled on my definition. This also
means that we cannot maintain the simple dichotomy between impaired and
enhanced individuals, and so the simple assumption that entirely distinct rules
should govern the permissibility of intervening to prevent or remove enhance-
ments on the one hand, and impairments on the other.
Far from being hard cases that my account struggles to accommodate,
I think these implications help to highlight the benefits of my account.
A central goal of this chapter is to challenge the assumption that we can
identify context-independently negative (or positive) deviations from a species
norm. There do certainly seem to be some easy cases—prodigious memory
seems positive, dementia negative, say—and it may initially seem odd to
suggest that both conditions are alike insofar as both are impairments and
may result in disability. Yet it is important to be clear about what this means.
I need not deny that paradigm enhancements are less likely to be disabling
than paradigm impairments (at least in the current social context). Nor need
I deny the ‘insurmountable realities attached to some impairments’.⁴⁴ As the
subsequent sections argue, insisting that atypical forms of functioning cannot
be categorised as good or bad without considering their impact on particular

⁴¹ Kahane and Savulescu 2015: 133. ⁴² See Giubilini and Sanyal 2015.
⁴³ Fowler 2015: 68. ⁴⁴ Vehmas and Watson 2014: 642.
  ,   149

individuals, in particular contexts, does not preclude acknowledging that

some of these consequences may be bad (perhaps for most). However, my
view also guards against the assumption that any atypicalities are good or bad
simpliciter: the loss of abilities many able-bodied individuals fear need not
prevent us having all the opportunities we are entitled to, and features many
would welcome may, in fact, lead to the loss of central capabilities.
In my view, then, it would be both conceptually incoherent and unnecessary
to attempt to distinguish disabilities that result from ‘dysfunction’ and those
that result from ‘enhanced’ functioning. First, I deny that differences can be
meaningfully classified in this way, prior to a consideration of their impact
(disabling or otherwise). Second, if someone is unable to access meaningful
work, form social relationships, participate in a political community, or move
freely within their environment, surely what matters is that they lack these
opportunities, and not whether they lack them due to some form of function-
ing others tend to welcome or repudiate (except insofar as this may inform our
approach to rectifying the capability loss). It is the very assumption that the
category of atypical functionings that are dysfunctional is stable and signifi-
cant that I aim to refute in this chapter.

5.5.2 Latent or Hidden Pathologies

Most standard accounts of impairment can straightforwardly exclude latent or

hidden pathologies on the basis that asymptomatic pathologies do not seem
dysfunctional or to limit major life activities. Whether they constitute func-
tional atypicalities will, in some cases, depend on the kind of functionings we
have in mind. The pancreatic functioning of a person with asymptomatic
diabetes might be atypical, for example, but does not generate atypical func-
tionings at a level that would be experienced or observed by the individual.
I would contend that impairment should be restricted to observable functional
atypicalities to preserve the connection with embodied experience. This is lost
if we include all unusual underlying physiological structures and processes,
since these are ‘found in “normal” long-lived individuals’.⁴⁵ For example, Ron
Amundson discusses an example of a UK university student with previously
undiagnosed hydrocephaly, who had normal intellectual and social function-
ing despite having ‘no more than 10% of the average person’s brain tissue’.⁴⁶

⁴⁵ Bergman et al. 1995–2020. ⁴⁶ Amundson 2000: 40.

150      

At the level of bodily process his functioning was certainly atypical, though his
day-to-day functioning was perfectly normal.
We should focus on the observable, not merely the observed, since some
cases may not be recognised by the affected individual. For example, a person
who is unaware of their low pancreatic function may still experience low
energy or mood swings as a result while failing to identify the cause. Or they
may fail to identify that their energy levels are unusually low if they have never
experienced any other way of being. Since these unacknowledged symptoms
nonetheless form part of the individual’s experience they should be considered
an impairment.
The question of whether latent or hidden pathologies will result in disability
might seem to have a more obvious answer: being latent or hidden these
pathologies will not restrict individuals’ opportunities, and so will not be
disabling. However, there are some complex cases. First, if treated prejudicially
as a result of misconceptions about the pathology, individuals’ opportunities
might nonetheless be restricted. For example, an HIV-positive person may be
asymptomatic but nonetheless subject to stigma, exclusion, and bias. Whilst
this is clearly a case of disadvantage, this does not seem to be a disability. Given
that the individual does not experience atypical functioning they are not, on
my view, impaired. Thus, insofar as disability is restricted to those disadvan-
tages associated with impairment, these disadvantages are not disabilities.
This is perhaps especially clear in this particular case since the prejudice
and ill-treatment is likely the result of a combination of factors, including
homophobia and negative judgements about drug-users (however misplaced).
Undoubtedly, though, such individuals will share significant experiences with
disabled individuals, as do those subject to other forms of discrimination, as
I discuss below.
Second, there may be cases in which the symptoms of an impairment-
causing pathology are hidden from an individual because their opportunities
are anyway limited. For example, imagine an individual who is photosensitive
but has been prevented from going outside, and so is unaffected by their
condition. Although this condition is not observed by its sufferer it is observ-
able (under more normal conditions), and hence constitutes an impairment.
However, whether the restrictions they face count as a disability depends on
the reason they have been subject to them. If, on receiving this diagnosis,
overprotective parents have thus limited their child’s activity, this (over)
reaction to the impairment disables their child—likely much more so than if
they were to take a more measured approach to the condition. If, on the other
hand, the individual’s restrictions are in no way related to their impairment,
  ,   151

but are, for example, the result of deeply sexist norms that do not permit
unmarried women to leave the family home, then they are not disabled—
though they are, of course, disadvantaged. Further, they might become dis-
abled if they were no longer subject to sexist norms, but were then restricted by
their photosensitivity.⁴⁷

5.5.3 Distinguishing ‘Genuine’ Disorders

Many theorists consider it essential that an account of disability is able to

distinguish ‘genuine’ disorders from normal conditions that have functionally
similar affects—ensuring, for example, that we differentiate between those
with disorders like sleep apnoea and those who simply do not like getting
out of bed in the morning;⁴⁸ or distinguishing depression from ‘normal’
sadness following a major distressing life event; or ordinary premenstrual
emotional swings from Premenstrual Dysphoric Disorder.⁴⁹ Some thinkers,
for example Leslie Francis, have claimed that we need the concept of (negative)
impairment to draw such distinctions, whilst others, such as Barnes, worry
that the concept will itself lead to pathologising, and thus stigmatising, normal
behaviours and responses. My inclusive account might seem vulnerable to
both types of concern. First, it will include, and so might seem to conflate,
ordinary grief, reluctance to get out of bed, and premenstrual mood swings
with the ‘genuine’ dysfunctions. Thus, it may seem that it fails to draw the
distinction impairment is meant to. Second, this inclusiveness may seem to
lead to even more widespread pathologising than an ordinary, negative
impairment account: all anomalies are deemed impairments.
But such concerns would be unfounded. It may be that some cases—not
liking to get up very early, perhaps—will simply be excluded on the basis
that this kind of functioning cannot plausibly be described as atypical.
Nonetheless, it is true that many cases not normally considered disorders are
indeed anomalous, and so will count as impairments on my account, and, if this
leads to a restriction in relevant opportunities, will be considered disabilities too.
Yet this is not a conclusion we should be at pains to avoid, since it does not have
the negative implications Francis and Barnes suppose. Certainly, it is important
to understand the different causes behind someone’s insomnia—depression,
transient grief, sleep apnoea, or bad choices in sleeping arrangements—in order

⁴⁷ Compare this example to that of the dyslexic girl migrating to the UK, discussed in §2.7.
⁴⁸ Francis 2018: 1146–7. ⁴⁹ Barnes 2018: 1161–2.
152      

to know how best to respond. It is important, too, to identify and prioritise those
people whose opportunities are most restricted, or are most subject to stigma, or
whose issues are least likely to resolve without assistance.
However, it is far from clear that it is either possible or desirable to clearly
distinguish the point at which grief becomes depression, or even struggling to get
out of bed every day becomes a sleep disorder.⁵⁰ It is right to take an inclusive
approach that aims to provide all individuals with the help that they need, and it is
a benefit of such inclusion that it acknowledges the impact that apparently minor
forms of atypical functioning can have on people’s lives. Insomnia, for example, is
often not taken seriously by GPs and other medical professionals, but the impact
on individuals’ lives and health may be profound (and can be effectively mitigated
with specialised cognitive behavioural therapy).⁵¹ A similar line of argument
would support the (controversial) suggestion that we should cease to diagnose
and distinguish individuals with dyslexia, and instead simply aim to provide early
intervention for all children who experience reading difficulties, who currently
may not get the help they need if their difficulties are not diagnosed.⁵²
Further, what we consider to be a disorder is a transient and shifting
category: the learning difficulties experienced by individuals with dyslexia
were once dismissed as the result of mere lack of intelligence, for example,
and ASCs blamed on a lack of maternal affection. The stability and usefulness
of these, and many other characterisations, continue to be the subject of
debate. We should, therefore, be less eager to exclude apparently borderline
cases. Instead, I contend, we should allow the category of impairment to be
expansive but focus on the nature of its consequences in a particular context—
that is, whether it is disabling.

5.5.4 Distinguishing Disability from

Other Forms of Disadvantage

Impairment is frequently used to mark out disability as a category, distinct

from other forms of disadvantage, such as those based on race or gender

⁵⁰ It might be objected that there is a clear means by which grief and depression, say, can be
distinguished: one is an appropriate emotional response, the other is not. I do not, and need not, deny
that drawing distinctions along these lines is possible, since my claim is not that there is no sense in
which these different sorts of cases can be distinguished. Rather, I only claim that in the context of
determining whether a form of functioning is atypical, and whether it leads to a restriction in relevant
opportunities, such distinctions are neither relevant nor useful.
⁵¹ Parkin 2018; Taylor and Pruiksma 2014.
⁵² See e.g. Elliott and Grigorenko 2014; Elliott et al. 2016; Gibbs and Elliott 2020.
  ,   153

identity, or sexual orientation. Will a neutral account of impairment be able to

reliably draw this distinction? Some cases seem to be excluded from the outset,
since whatever differences they entail can hardly be described as functioning
atypicalities—‘atypical functioning’ hardly seems constitutive of race, for
example, however race is defined.⁵³ It also seems a stretch to describe minority
sexual orientations in these terms. Same-sex romantic and sexual preferences,
for example, are arguably atypical and might restrict our opportunities in a
number of contexts: most obviously as a result of social stigma though, in
addition, this may (for now) prevent individuals having a biological child with
their romantic partner. However, these restrictions cannot plausibly be
described as stemming from atypical physical or cognitive functionings, but
from preferences and the unjust and unjustified treatment of them.
Sex-based differences may seem to present a harder case, since there do
seem to be differences in the functional capacities of average male and female
bodied persons. Most obvious are variations relating to differences in their
reproductive capacities, though these are not the only differences: for example,
females may typically experience different symptoms of a heart attack than
males. We do not need to accept the existence of a strict binary, or a
metaphysically deep reason to appeal to reference classes based on sex cat-
egories, to allow that our assessment of typical functionings can be indexed to
the capacities of biological females and males. For example, a male unable to
bear young is not anomalous, so not impaired in my sense.
It might seem that trans persons are a difficult case to accommodate here.
Whilst I cannot attempt a full account of this complex and controversial topic,
it is not immediately obvious that this must be so. Roughly I would suggest,
first, that either conforming to, or subverting, socially constructed gendered
behaviours should not be considered a form of physical or cognitive function-
ing, but a preference about how to live our lives. Thus, individuals whose
gender presentation is non-binary or does not conform to their assigned sex
should not be understood as having an atypical functioning capacity.
However, individuals who change their sex identity, or change their body to
corresponded to their true sex, might be impaired if they are unable to perform
functionings that are typical for this sex. For example, trans women who are

⁵³ Clearly this is not the place to defend a particular account of race. However, it is worth noting that
most plausible accounts are broadly constructivist, rejecting the idea of ‘racial essences’, but nonetheless
allowing that race possesses a contingent ‘social objectivity’ (Mills 1998: 48). As Haslanger (2000: 43)
puts it: ‘Race is socially real, even if a biological fiction’. If race is, indeed, a matter of racial identity and
identification not biology, then it is hard to see how or why this would be correlated with any kind of
atypical functioning (e.g. Appiah 1996; Shelby 2002).
154      

unable to bear young or menstruate could be said to function atypically, and

could potentially be disabled too, depending on the consequences of such
restrictions, just as natal women who cannot so function might be. This
certainly does not imply that there is anything pathological about mere
possession of a trans identity. I follow the DSM-5, then, in accepting that
‘gender nonconformity is not itself a mental disorder’.⁵⁴ This would not, of
course, hold true if we do not believe individuals can choose or change their
sex, in which case these trans women would be normal functioning male-
bodied persons with female gender identity and presentation—and thus, not
impaired. I take no position on this debate here, but merely mean to sketch
how various approaches would be accommodated by my account and to show
that they can be accommodated.⁵⁵
A neutral account of impairment can, then, provide some guidance on
distinguishing disability from other dimensions of disadvantage, without
falling prey to the manifold difficulties in determining which of the ways of
functioning are dysfunctional. The core role of impairment in this context is to
enable us to distinguish restrictions in options that result from atypical forms
of functioning, and those that result from (prejudice against) certain prefer-
ences, presentation, or appearance. Undoubtedly, then, there are structural
similarities between all these kinds of cases and disability—being subject to
stigma, prejudice, misconceptions, and lowered expectations as a result of
one’s presumed capacities based on a superficial assessment of one’s physical
appearance and behaviours—but this should not, and need not, lead to their
conflation.

5.6 Mere Difference or Complex Difference?

Impairments are atypical functionings or anomalies, whilst disability is a

subset of the consequences of having an impairment in a specific context.
I now turn to the normative questions with which the chapter opened.
Specifically, under what circumstances are such anomalous functionings
and their consequences disadvantageous (causing an absence of something

⁵⁴ American Psychiatric Association 2013. I leave aside here the question of whether there may
nonetheless be reasons for trans persons to choose to adopt a disabled status if this is the only way to
ensure their rights are protected (as Wahlert and Gill 2017 argue).
⁵⁵ For further discussion of these issues see e.g. Bettcher (2014a, 2014b), Byrne (2020); Dembroff
(2018), Jenkins (2016).
  ,   155

individuals are entitled to)?⁵⁶ And under what circumstances are they harmful
(negatively impacting individuals’ well-being)? The central questions, here,
concern the kinds of opportunities that might be restricted and the impact this
tends to have on individuals’ lives. I argue that impairments are not necessarily
harmful or disadvantageous, and whilst disabilities, on my view, are disadvan-
tageous by definition, they need not cause harm. Centrally, I argue that we
should be cautious of assuming that there is something especially bad about
having the kinds of impairments usually considered disabling, and should not
assume that it is appropriate to seek to eliminate them (see §7.9). As such,
I strongly reject taking a ‘bad-difference’ view of disability and, especially,
impairment: functioning differently need not be ‘bad’, understood as either
harm or disadvantage. However, I do not adhere either to a ‘mere-difference’,
or neutral view. I will consider and reject two versions of the mere-difference
view, and argue instead for the complex-difference view. Then, in the next
section, I will defend this view against arguments that the loss of opportunities
entailed by disability and impairment mean it is a bad difference after all.
First, Barnes has mounted an influential recent defence of the mere-
difference view. According to her version of the view, disability is simply
something ‘that’s bad for you with respect to some aspects of your life . . .
[and] good for you with respect to other aspects of your life’.⁵⁷ Her view does
not, therefore, imply that there is no difference in average well-being between
disabled and non-disabled people (in an ableist world); it does not imply that
being disabled cannot involve the loss of some intrinsic goods (only that it is
not merely this loss); and it does not imply that all the bad-making features of
disability are the result of social norms and infrastructure (even if many are).⁵⁸
Insofar as Barnes’s contention is that there is not an inherent, intrinsic, or
automatic connection between having a disability (or, in my terminology,
having an impairment that is usually disabling) and being overall worse-off,
I am in complete agreement with her. However, by including, as she does, a
vast diversity of conditions within a unified category of disability, this con-
clusion becomes obviously true, yet somewhat empty.⁵⁹
Disability, in Barnes’s broad sense, is not necessarily a bad difference, but
still encompasses conditions (whether few or many) that are bad differences.

⁵⁶ Recall that a situation in which individuals are disadvantaged is not fully just since they lack what
they are entitled to, though this may not be the result of injustice (see §2.5).
⁵⁷ Barnes 2016a: 79. ⁵⁸ See §1.7 for further discussion of Barnes’s approach.
⁵⁹ On Barnes’s weak interpretation of neutrality, also see Wasserman 2018: 254–6; Hawkins 2018:
463–6; Dougherty 2014. Note that I need not be committed to the strong claim that Barnes’s account
cannot be action-guiding in any context, only that it is unhelpfully vague in the context of theorising
about justice (which is not, in any case, Barnes’s concern).
156      

Her approach, therefore, allows us to conclude that we should not assume that
all disabilities warrant removal, but it does not commit us to accepting that all
disabilities are neutral features that it would be wrong to remove.⁶⁰ Thus, in
the context of attempting to devise just policy proposals, at least, Barnes’s
emphasis on the terms mere difference and neutrality is importantly misleading.
She is right to be irritated at those who insist that her view implies that ‘the
positive effects [of disability] have to balance the negative ones’, or claim that
she ignores the fact that in our current, ableist context disability can often
significantly reduce our good options.⁶¹ She very clearly rejects the former
claim and accepts the latter.
Nonetheless, it is perhaps unsurprising that her mere-difference view is
assumed to mean that differences are not bad, as opposed to not necessarily
bad, and Barnes does sometimes trade on this rhetoric. For example, the key
thesis of her book is that ‘to be physically disabled is not to have a defective
body, but simply to have a minority body’,⁶² and whilst this is true in a sense, it
can gloss over the fact that some impairments, in some contexts, do lead to
worse forms of functioning—indeed, in Barnes’s own words that ‘disability
can, in and of itself and even in ideal social conditions, be a harm’.⁶³ As such, it
simply would not be as objectionable to work to eliminate the genes that cause
cystic fibrosis or multiple sclerosis as it would to eliminate genes that give rise
to the phenotypes associated with racial categories. The terminology of com-
plex difference better captures the multi-faceted impact impairments can have
on different lives.⁶⁴
Alternatively, we might adopt a more substantive version of the mere-
difference view, which, perhaps more reflective of the term, implies disability
does not encompass any negative differences. This involves restricting the
scope of disability to explicitly exclude those elements most likely to be negative,
such as pain or lowered life expectancy. In other words, we ‘need not deny that
impairments have undesirable aspects, such as pain or discomfort—merely that
those aspects are not within the scope, or part of the meaning, of disability’.⁶⁵
This view perhaps underlies Silvers’s suggestion that individuals deserve health-
care resources ‘for their pain, not for their disability’.⁶⁶ But how are we to
determine which elements of some condition are inessential or separable from

⁶⁰ See §1.7 and §7.9. ⁶¹ Kahane and Savulescu 2016: 777, 776. See §1.7 for further discussion.
⁶² Barnes 2016a: 1. ⁶³ Barnes 2009b: 338.
⁶⁴ Indeed, despite Barnes’s explicit and vocal commitment to a mere-difference view, it might
be argued that her approach could be more plausibly seen as a complex-difference or ‘mixed’ view.
(E.g. see Stoner 2016: 939.)
⁶⁵ Wasserman et al. 2016a. Also see Wasserman et al. 2016b. ⁶⁶ Silvers 2003: 478.
  ,   157

it? How, indeed, are we to determine which the ‘undesirable’ aspects are given
that different individuals will resent, or dissociate from, different features, and
that they will have different impacts on different lives?
Certainly, disability should not be reduced to these negative elements,
especially given that the correlation between these aspects and the underlying
impairment or anomaly depends on the context in which they are experienced.
Nor should we assume that an anomaly necessitates a general neediness or
vulnerability, or implies that we must experience all potentially undesirable
aspects of any form of disablement—for example, that a wheelchair user must
be in pain, or that a deaf person is cognitively impaired.⁶⁷ Indeed, we should
not assume that even paradigm impairments must result in any disability at
all, as Chapter 2 considered at length. Nonetheless, an important part of being
disabled involves living in a body, or with a mind, that functions in ways that
are atypical—and sometimes uncomfortable, inconvenient, or painful. Such
aspects are not all that disability is, but they are an intrinsic part of some
experiences of impairment and it does a disservice to those who have these
experiences to simply define them away from what it means to be disabled. For
individuals who have conditions such as rheumatoid arthritis, fibromyalgia, or
Duchenne muscular dystrophy, it is not clear what their disability would
consist in if the elements that cause pain and discomfort were removed.
Campbell and Stramondo, for example, allow that if pain were ‘an essential
feature of some disability, we grant that it is plausible that that disability is
intrinsically bad in that respect’.⁶⁸ Yet they also question whether pain really is
an essential feature even of disabilities usually ‘defined in terms of . . . pain’,
such as fibromyalgia.⁶⁹ They point out that, if a drug were developed
that eliminates pain without eliminating other symptoms of the condition,
we would not say the drug eliminates fibromyalgia, just as effective contact
lenses do not eliminate myopia but merely manage its negative effects. From
this they conclude ‘pain is not part of what fibromyalgia is, but is instead a
contingently correlated symptom’.⁷⁰ However, this is a rather misleading use
of ‘contingent’. It does not merely capture the externally, artificially imposed
effects of impairments. Rather, it seems that all effects of all impairments are
contingent in this sense: the inability to process auditory information might
only be a contingent part of deafness, insofar as cochlear implants or hearing
aids might mitigate these (negative?) effects. As such, it is not clear on what

⁶⁷ This point might seem so obvious as to hardly need saying, but many individuals with visible
impairments report being treated as though they are needy, suffering, or incapable in all domains.
⁶⁸ Campbell and Stramondo 2017: 157. ⁶⁹ Campbell and Stramondo 2017: 178.
⁷⁰ Campbell and Stramondo 2017: 179.
158      

principled basis pain, depression, and other seemingly negative elements

should be seen as contingent or inessential to a condition. It is much more
illuminating, then, to understand such symptoms as necessary features of
various impairments, yet to focus on what impact these effects have on an
individual’s life and opportunities. In other words, to allow that impairments
encompass multi-faceted, complex sets of difference but to focus on deter-
mining whether these impairments are disabling.

5.7 Bad Difference or Complex Difference?

We should not, therefore, pursue an account that ignores or sets aside the bad
elements of impairments. Yet nor should we overestimate their negative
effects.⁷¹ Demonstrating that the two most plausible iterations of the mere-
difference account seem mistaken, and that impairments may, in some ways,
be bad for us, may be sufficient to show that we should take a complex, rather
than a mere-difference view. This section will argue that a complex-difference
view should also be preferred to a bad-difference view. Other theorists have
also defended the view that impairments may have a mix of positive and
negative impacts on our well-being. One plausible approach is Ian Stoner’s. He
argues that paradigm impairments may make individuals ‘little-w’ worse-off,
in that they are ‘hassles’ that cause episodes of frustration and annoyance that
life would be easier without. However, they do not make individuals ‘big-W’
Worse-off, since they do not prevent individuals from discovering and pur-
suing rewarding projects, and so do not impair their ‘prospects for securing
the substantive correlates of a good life’.⁷² Similarly, Campbell and Stramondo
argue that ‘having a disability does not prevent one achieving any of the major
goods of life’, though they argue disabilities are not instrumentally, compara-
tively, or overridingly neutral.⁷³
I need not defend a specific view here. For my purposes, the important point
is that the differences impairments make to individuals’ lives are complex, and

⁷¹ On the dangers of relying on our imagination about what disabled life is like, see e.g. Amundson
2000; Mackenzie and Scully 2008; Shakespeare 1994, 2014.
⁷² Stoner 2016. Note that Stoner’s understanding of being ‘big-W’ Worse-off is closely related to my
view about individual entitlements. Whilst individuals are not entitled to be free from hassles,
irritations, and annoyances that may decrease their well-being (little-w), individuals are entitled to
form and pursue a conception of the good (perhaps, Stoner’s Well-being). Whether this distinction is
described as being between different sub-concepts of well-being (as Stoner contends), or between well-
being and justice, does not seem significant in this context.
⁷³ Campbell and Stramondo 2017: 157. Also see Bognar 2016; Kahane and Savulescu 2016;
Schramme 2014; Shepherd 2020; Stramondo 2017; Wasserman and Asch 2014.
  ,   159

even when they do restrict, or make more difficult, our pursuit of certain
opportunities, they are not necessarily disadvantageous. Part of the reason for
this is that, as the previous chapter discussed, we are entitled to control in
broadly construed domains. Having such control is achievable by a variety of
means, and so will require an ability to function at a certain level, but not
necessarily the ability to perform specific functionings: hence, mobility does
not require the ability to walk, sensory experience does not require using all
five senses, communication does not require the ability to talk and listen,
artistic enjoyment does not require the ability to experience all modes of
creative expression (sculpture and music and paintings and poetry). Each of
the individual functionings that achieve the more general end might be
valuable, but this does not imply that all of the means to the end are necessary
prerequisites to meet the demands of justice (or even for individual welfare).⁷⁴
On the one hand, the claim that certain very specific forms of functioning
are neither essential components of a good human life nor necessarily part
of our distributive entitlements might seem obvious. Opportunities for leisure
and play might be important, but why specify that these must be for football,
or swimming, or chess—or indeed, all three? Opportunities to pursue mean-
ingful work might be important, but why specify that we must have the
chance to be nuclear physicists, or fighter pilots, or politicians? Yet, on the
other hand, some specific options may seem to have a particular value that
cannot simply be substituted by some other hobby or career path, especially
when we have committed to and deeply value that option. Thus, some would
argue that anomalous functioning is a bad difference when it involves
functioning loss.
It is certainly true that losing the ability to perform a valued functioning can
reduce our quality of life. Hence, even the most ardent supporter of a mere-
difference view would not deny that the process of transition to the loss of a
functioning will often be difficult and painful. However, this does not tell us
that the end state is bad, merely that adjusting to major change—especially a
change that necessitates the sacrifice of a treasured pursuit—will be bad for us,
at least temporarily. Individuals, after all, have a considerable capacity to shift
their goals, and frequently do so in response to much more mundane setbacks.
Further, we should not assume because we are attached to some activity or
pursuit that it must be an essential component of every life: my life might be
meaningless without access to music, but this hardly means that every decent

⁷⁴ Also see Wasserman and Asch 2014; Stoner 2016.

160      

human life must include such access, and even I may adjust my expectations
over time.⁷⁵
More importantly, even where a functioning is necessary for an individual’s
well-being, this does not mean they must be entitled to the ability to perform
this functioning, and so disadvantaged when they lack this opportunity.
Again, my primary concern here is with questions of justice. If I am right
that individuals are entitled to control in certain broad domains of their life,
where this involves having sufficient ability to meaningfully direct the
course of their life, then this does not require that individuals have access to
every option, nor every option they prefer—hence, mobility but not necessar-
ily walking, or leisure but not necessarily a sports car. As such, it is certainly
not necessary that we seek cosmetic normalcy, or superficial functioning
equality. In the context of justice, our focus should be on level of functioning
rather than a specific mode, and so the inability to engage in a specific form of
functioning should not automatically imply that an individual lacks the
opportunities they are entitled to.
However, objections might be raised to this rather quick dismissal of
the significance of losing valuable functionings, especially by those who
advocate for bad-difference views exactly on the basis that there is something
bad about losing access to valuable functionings. I will focus here on two
arguments made by Jeff McMahan. First, McMahan argues that losses of
ability only appear neutral because they ‘can be compensated for by other
abilities that develop to fulfil its functions’,⁷⁶ such as heightened hearing to
compensate for loss of sight. He insists these losses cannot really be ‘entirely
neutral’ since if they were ‘they ought also to be neutral in combination; but
they are not’.⁷⁷
However, McMahan is wrong to suggest that losing an ability is only non-
negative due to compensation, and wrong, too, to suggest that the fact that
losing many abilities ‘would in combination make a life worse’⁷⁸ demonstrates
they are individually negative. We would not say that our inability to be a
surgeon is ‘compensated’ by our ability to be a teacher: these are different ways
of functioning, with different benefits and challenges, and each can contribute
to a valuable life. Similarly, sight loss is not compensated for by superior
hearing in the sense that, bat-like, individuals can achieve something like the

⁷⁵ I consider adaptation in the context of both welfare and justice much more extensively in the next
chapter.
⁷⁶ McMahan 2005: 96. McMahan’s views have frequently been subjected to criticism (e.g. Barnes
2016a: 76–7; Stoner 2016: 938–9; Wasserman and Asch 2014: 156–8).
⁷⁷ McMahan 2005: 96. ⁷⁸ McMahan 2005: 96.
  ,   161

experience that sight gives us, but through a different sensory function. Rather,
blind individuals have their own forms of functioning that may be superior in
some respects, and worse in others, but overall are perfectly compatible with
leading a good life and having what we are entitled to.
Furthermore, the ‘combination problem’ McMahan outlines does not
demonstrate that the loss of abilities associated with impairment is not
really neutral after all. Again, the analogy with a lack of talent is useful: if
we lacked all talents, this would make our life worse, but this does not mean
that the ordinarily talented person, lacking several talents, is badly off, or
even worse off than the polymath. On the contrary, we might think that a
degree of specialism in our abilities could improve our life, insofar as we
need not be plagued by the feeling that we are letting our talents in one
area go uncultivated when we choose to focus our energies in another
direction.⁷⁹ This is not to deny that there are impairments that restrict so
many opportunities, or opportunities with such a wide-ranging impact on
our lives, that they will make individuals worse off. Yet this does nothing to
justify the blanket judgement that any opportunity loss associated with
impairment is similarly bad—just as our limited talents need not be harmful
or disadvantageous, though complete incompetence in every area probably
would be.
McMahan’s second argument is that lacking an ability ‘that is instrumen-
tally valuable to those who have it is, in general, an obstacle to the achieve-
ment of the full range of goods characteristic of human life’.⁸⁰ Again, this
does not seem to be the case. As noted above, a good life does not require
access to everything that might make a life good. Further, each addition of
individually valuable abilities or options will not always be good for us: at
the very least, at some point they are unlikely to make a meaningful differ-
ence to our lives since we simply will not be able to make use of every form of
potentially valuable flourishing. Indeed, it might even make life worse if it
causes us to be paralysed with indecision or filled with a deep sense of
regret about our wasted potential. Yet it may be objected that, whilst this
may be true of some things (abilities to pursue different careers, say), it is not
universally true since there are some abilities that are not interchangeable in
this way. Thus, the lack of these will simply make a life go worse. The ability

⁷⁹ A much more plausible way of understanding such cases is in terms of thresholds: below a certain
point, a lack of options will be bad for us. However, the fact such a threshold exists does not imply that
losses above this threshold are not neutral—as noted, losses might even be beneficial if it is optimal to
have options in a range below the maximum.
⁸⁰ McMahan 2005: 96.
162      

to appreciate music is one of the most oft-cited examples here. It is not

enough, it might be claimed, to have access to some forms of artistic or
aesthetic enjoyment: music matters.
In response, first, it should be noted that this loss is not uniquely the
consequence of impairments usually considered disabling. Individuals
who are tone deaf, or merely musically apathetic, also do not partake of
this functioning, and we do not generally doubt that these individuals can
have a good quality of life, or conclude that they need ‘fixing’. We might
argue that it is different to choose not to listen to music because we prefer
other art forms than to never have had the chance. However, many indi-
viduals who have the ability to hear nonetheless lack the capacity to appre-
ciate music—and it is surely appreciating as opposed to merely hearing
music that has value. Thus, we should be cautious of exaggerating the
tragedy of the opportunity loss associated with paradigm impairments,
and of ignoring the restrictions in experiences and options all people face.
Clearly, following sports and supporting teams brings huge value and
meaning to people’s lives, for example, yet this is not a form of flourishing
I will ever have access to. Eliminating the kind of impairments that concern
McMahan would, therefore, do nothing to remove the problem (if it is a
problem) that some valuable opportunities and functionings are foreclosed
to every person.
It may, though, be objected that the problem with paradigm impairments is
not merely that they entail the loss of particular options, but that they close a
whole domain of valuable experiences (for example, individuals with very
severe cognitive impairments may struggle to engage in any meaningful social
interactions), or that they are accompanied by negative functionings (most
obviously, pain). This is not an objection I can respond to, because these are
not claims I deny. I am not, then, attempting to make the strong claim that
impairments never make our life worse, or never have bad consequences
(beyond those that arise from social and environmental causes). Indeed,
I know of no one who defends this strong claim, though it is against this
straw man that many of those who argue for the badness of disability pitch
their response. Rather, I am defending the much weaker claim that impair-
ments, even when they involve the loss of the ability to perform a functioning
that has value in the lives of those who possess it, is not automatically or
necessarily bad for us and is compatible with a good life. Further—and far
more significantly given that my concern is justice—I contend that having an
impairment is, in a great many cases, compatible with having the capabilities
we are entitled to, and hence need not be disadvantageous.
  ,   163

5.8 Conclusion

There may be reasons to worry about having some anomalous functionings,

then, but this is insufficient to give us reason to believe that the paradigm cases
of ‘negative’ or disabling impairment are especially or uniquely bad. First,
having access to fewer forms of functioning does not mean that an individual’s
functioning achievement is worse, or makes their life go worse overall. It may
place individuals at greater risk of lacking what they are entitled to (and of low
quality of life) if they have fewer modes by which a level of functioning can be
reached. However, this is true of many people: an individual whose happiness
and livelihood rest on their capacity to write novels may be at great risk of
lacking both well-being and entitlements when faced with writers’ block.
Second, being in a functioning minority might sometimes make life more
difficult and, in some respects, worse (and not just due to prejudice and bias).
For as long as most people communicate verbally, for example, being deaf may
remain challenging. In most conceivable circumstances living in chronic pain
will, all else being equal, make a life worse. This does not imply, though, that
the modes of functioning employed by individuals with impairments—or
anomalies—are always worse. In part, it may merely be difficult to be at one
end of the spectrum of human variation, whether at the end perceived as
positive or negative. The effects of paradigm impairments, then, are often no
worse than other forms of difference taken far less seriously.
Both mere-difference and bad-difference views can lead to the imposition of
blanket judgements about whether these conditions can or must be cured,
whether these lives are worse, and whether it is wrong to create individuals
who will lead them. Such universal judgements occlude diversity and can be
deeply misleading as a basis for policy proposals (as the final chapter con-
siders). Whilst generalisations might sometimes prove to be a useful heuristic,
particularly in the context of public policy, we should as far as possible
remain sensitive to the diverse impact of contextual and personal factors on
individuals’ experience of impairment. It is these effects we should worry
about, especially when they restrict people’s access to the opportunities they
are entitled to, and not the mere fact that individuals function in a wide
variety of ways.
Thus, just as we should not assume deafness, Down’s, or paraplegia are to be
avoided, nor should we assume additional sensory capacities, heightened
cognitive functioning, or improved athleticism should be welcomed. Our
unexamined intuitions regarding what constitutes a positive change, enhanced
capacities, or superior abilities (and the reverse) should be set aside. Instead,
164      

we should analyse the degree to which a form of functioning actually inhibits

individuals’ access to acceptable opportunities in central domains of their life.
Our focus, then, should be shifted to mitigating and removing the disadvan-
tageous effects of atypical functioning, and not eliminating the difference itself.
In short: this approach allows us to campaign against disability (as restriction)
without suggesting that we should seek to normalise human bodies and minds
(remove impairment), or even identify what is normal and what is ‘deficient’.
 6
Disambiguating Adaptive Preferences
When, and Why, Should Testimony be Trusted?

6.1 The Dilemma of Adaptive Preferences

The idea that impairments are differences rather than deficiencies, and do not,
or need not, lead to the loss of opportunities we are entitled to, accords with
the views of many people who have impairments widely considered disabling.
Many such individuals rate their subjective quality of life highly,¹ resist the
suggestion that their conditions should be cured, and refuse cures when
offered. They argue for the right to have disabled children, and against policies
of pre-natal screening for various disabilities. They march in Disability Pride
parades, and insist that disabled lives are not inherently worse than non-
disabled lives. Further, as the preceding chapter has maintained, the arguments
that supposedly demonstrate that paradigm impairments are necessarily bad
differences are unconvincing. Nonetheless, the idea that the functioning differ-
ence or restriction associated with disability is disadvantageous—indeed, a tragic
way of being—remains a widely held, and apparently common-sense, view
amongst most able-bodied (and some disabled) people.
Faced with such a divergence of views, whom should we trust? Thus far
I have made the implicit assumption that one benefit of my approach is that it
is in line with the self-conception of individuals widely considered to be
disabled.² After all, disabled people may seem best placed to understand the
realities of living with their conditions. As such, we may be tempted to
conclude that disabled individuals’ testimonies should simply be relied upon.

¹ This is so surprising to individuals without disabilities, many of whom consider the prospect of
becoming disabled as one of the greatest misfortunes that can befall us, that it has been called the
‘disability paradox’ (Albrecht and Devlieger 1999; also see Moller 2011; Mitchell 2018; Ubel et al. 2005).
² In what follows I will sometimes refer to ‘disabled people’ (or equivalent) as shorthand for the
more convoluted ‘people who have impairments widely considered disabling’. The meaning should be
clear from context—what is relevant here is that individuals are assumed to be, and treated as, disabled
in the current context. This should not be taken to conflict with my view that individuals are disabled
only when they lack opportunities they are entitled to and that, as such, many of those with
impairments widely considered disabling are not, in fact, disabled.

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0007
166      

However, if this were right, then the arguments in the previous chapter would
be unnecessary, and we could, instead, simply engage in the empirical task of
delineating disabled individuals’ views about their conditions. Yet things are
not so simple. In fact, we may think that disabled individuals’ experiences are
exactly what give us reason to doubt their testimony. It is widely acknowledged
that in conditions of great hardship or deprivation, individuals may cope with
their circumstances by claiming to prefer and, indeed, coming to prefer, their
situation to any alternative. However, such ‘adaptive preferences’ need not
constitute decisive evidence that these conditions are not, in fact, deplorable.
We do not, for example, consider women’s support of sexist and patriarchal
norms and institutions grounds for justifying them, nor an abused partner’s
preference to remain in the relationship a mitigation of this abuse.³
In a similar way, one might suggest that disabled individuals’ apparent
satisfaction with their lives is not reliable evidence against the claim that
many paradigm impairments must be disadvantageous. Indeed, this sugges-
tion might seem like a neat solution: if disabled individuals’ preferences are
simply deformed by the terrible circumstances in which they were forged, then
we can treat their positive claims about their impairments as suspect, and
maintain our intuition that their lives are tragic after all.⁴ It is certainly a
solution that allows many people to maintain their strongly held intuitions.
However, we should be cautious about simply ignoring the preferences and
beliefs of a group who are already subject to considerable injustice. The
silencing or mistrust of already underrepresented groups may constitute a
serious epistemic injustice, quite apart from the possibility of being used to
justify interference in their lives and choices. There has, on this basis, been a
considerable backlash, for example, against the tendency amongst Western
feminists to see women in the Third World as the ‘dupes of patriarchy’, whose
preferences can be ignored.⁵ Elizabeth Barnes, too, has argued against adaptive
preference models of disabled individuals’ preferences on the basis that when
‘misapplied . . . [they] can simply entrench pre-existing biases’.⁶
However, such worries about the misapplication of the concept of adaptive
preferences should not lead us to abandon it entirely. Misuse of the term

³ See e.g. Khader 2011; Nussbaum 2001; Sen 1999b; Terlazzo 2016.
⁴ My focus is cases where merely preferring a life with an impairment is thought to cast doubt on the
reliability of an individual’s testimony. Thus, I am concerned with physical and cognitive impairments
that are not considered, in themselves, to radically undermine an individual’s capacity for autonomous
agency. I do not consider the issues raised by severe cognitive impairment (nor take a view on which
should be considered ‘severe’).
⁵ Narayan 2002: 418. Also see Khader 2011, 2012, 2013, 2019; Jaggar 2005.
⁶ Barnes 2016a: 137.
    167

should indeed be guarded against, but if we were to simply take individuals’

claims to be satisfied with their lot at face value, this may mean that mistreat-
ment, injustice, and oppression are neither recognised nor rectified. If those
subject to oppression do not identify their own mistreatment, they may not be
entitled to redress. Indeed, it may not be considered mistreatment at all, since
‘no one complains’. Rather than opting for one or other horn of this dilemma,
I suggest that a balance must be struck between taking individuals’ preferences
for their current way of being as definitive, and ignoring them entirely. In
other words, a theory of disability or disability justice that took no account of
disabled individuals’ views would be unacceptable; yet justifiable worries about
the epistemic injustice of unwarranted diagnoses of adaptive preferences
should not lead us to simply take individuals’ testimony at face value.
A further problem with diagnosing adaptive preferences, less frequently
acknowledged, is the vagueness of the term, and the breadth of phenomena it
can be used to cover. A wide variety of experiences and attitudes can be
roughly grouped, and rejected, as adaptive preferences, though many of
these may not seem much like preferences. Further, we might reasonably
think that if the defining characteristic is adaptation, then all preferences
formed in response to our circumstances should be included. It should be
clarified from the outset, then, that ‘adaptive preferences’ as the term is used in
the literature is both broader and narrower than an intuitive understanding of
the concept. Broader because it includes not just preferences, but something
like ‘evaluative states’,⁷ and narrower because it includes only those adapta-
tions that are, in some sense, unreliable. However, even accepting these
restrictions, ambiguity remains: in particular, concerning what constitutes
relevant unreliability.
My goal in this chapter, therefore, is to respond to the problems of both
misapplication and vagueness by clarifying the various phenomena loosely
categorised as adaptive preferences. In particular, I distinguish between adap-
tive preferences that are an unreliable guide to individuals’ interests (‘well-
being adaptive preferences’), and those that are a poor guide to just policy and,
in particular, to distributive entitlements (‘justice adaptive preferences’).
Drawing this distinction allows us to see the different ways in which individ-
uals’ preferences may be unreliably adaptive, and the different responses that
may be appropriate. It also makes clear that we should not simply distinguish
individuals with adaptive preferences from those without, nor should we

⁷ As Terlazzo (2016: 206) notes, the term incorporates both ‘comparative preferences proper’ and
‘non-comparative states like desires, judgements of values, etc.’
168      

distinguish preferences that are adaptive (for all purposes, in all contexts) from
those that are not. Many, if not all, individuals possess some form of adaptive
preferences at some time, yet it is only a subset of these (justice adaptive
preferences) that warrant exclusion from determining the content of an
account of just policy. Further, whilst all adaptive preferences are non-
autonomous, only well-being adaptive preferences are irrational. As such,
preferences may be deemed unreliable from the perspective of justice, without
impugning individuals’ rational capacities. Thus, correctly diagnosing an
individual’s preference as adaptive need not constitute epistemic injustice,
nor insultingly malign their capacities as a rational agent.
I will begin by outlining how traditional, procedural accounts of adaptive
preferences differ significantly from the kinds of adaptive preferences that
raise concerns in the context of social justice (§6.2). I then outline how these
proceduralist accounts of adaptive preferences—which I call well-being adap-
tive preferences—should be understood (§6.3). Next, I argue that it is import-
ant to maintain an account of adaptation of this form, even if our concern is
primarily those forms of adaptation relevant to justice (§6.4). Identifying those
adaptive preferences that are relevant to justice requires a more substantive
approach than pure proceduralism can provide (§6.5). However, such sub-
stantive accounts should not come at the expense of identifying procedural
flaws in preference formation: both approaches are needed to understand the
different ways in which individuals’ preferences may be unreliably adaptive
(§6.6), and the different responses that may be appropriate (§6.7).

6.2 Adaptation and the Political Project

Well-being adaptive preferences are adaptive preferences as classically under-

stood. Roughly, individuals adapt to constrained options in ways that render
their resulting preferences procedurally irrational and non-autonomous, and
so undermine their possessor’s authority about their own well-being. These
classic accounts of adaptive preferences, developed by Jon Elster and his
critics, are modelled on the ‘Fox and Grapes’ parable.⁸ On realising he cannot
reach the grapes he desires, the fox insists ‘grapes are too sour for foxes’, and
he did not want them anyway. It is assumed this response is irrational, and not
a reflection of his best interests.

⁸ E.g. Elster 1987; Bovens 2002; Bruckner 2009; Colburn 2011; Zimmerman 2003.
    169

These accounts root this irrationality in procedural flaws in the process

of preference formation. For example, that the preferences were formed
(necessarily) unconsciously in light of the diminished options, or do not
cohere with our other beliefs and preferences (in the case of the fox, regarding
the ideal sweetness of fruit).⁹ They focus on capturing the idea—central to our
intuitions about the fable—that the fox’s preference is unreliable because he is
unconsciously ‘fooling himself ’: he has failed to acknowledge his limitations,
or to recognise that the real reason he no longer prefers grapes does not
concern their sourness. After all, he would still happily consume more easily
accessible grapes. But if, instead, the fox responded to being unable to reach
the grapes by consciously cultivating a taste for a sweeter, lower-hanging fruit,
and so would no longer consume grapes even when offered, then this would
not be an adaptive preference, but conscious, rational, and autonomous
character planning. It is suggested, on such accounts, that we should not
distrust such preferences in the same way, since even though they are formed
in reaction to our circumstances (and hence are adaptive in a sense) they can
more plausibly be considered a new but genuine preference that reflects our
new goals and character.
In addition to the literature focused on procedural flaws in the formation of
adaptive preferences, and their resulting lack of autonomy, there has also been
considerable discussion of how the problem of adaptation operates in real-
world contexts, with particular focus on situations of economic deprivation
and gender-based oppression.¹⁰ It has been widely observed that individuals
express happiness and satisfaction in the face of deprivation, oppression, and
suffering when they become used to such circumstances, and that apparent
improvements in their material circumstances do not always translate into
reports of higher levels of happiness or satisfaction. This can be observed at the
societal level: for example, whilst richer individuals within a country tend to
report greater levels of happiness, improvements in the wealth and income in a
country as a whole do not improve average reported happiness.¹¹ It can also be
illustrated with more specific, individual cases. Sen, for example, famously
discusses ‘the hopeless beggar . . . the dominated housewife, the hardened
unemployed . . . [who] take pleasures in the small mercies and manage to
suppress intense suffering for the necessity of continued survival’.¹²

⁹ These are the views of Colburn (2011), Elster (1987), and Bovens (2002) respectively.
¹⁰ The most prominent voices here are, perhaps, Nussbaum (e.g. 2000a, 2001), Sen (e.g. 1988,
1999b), and Khader (e.g. 2011, 2013).
¹¹ This is the Easterlin paradox (see e.g. Moller 2011 for further discussion).
¹² Sen 1988: 45–6.
170      

Given that these cases are also referred to as adaptive preferences, it might
be natural to assume that these are also of the procedurally flawed, ‘sour
grapes’ form—especially given that a clear alternative account is often not
provided. Yet examples such as Sen’s belie this.¹³ These individuals are not
obviously ‘fooling themselves’. Indeed, many seem well aware of their suffer-
ing, and make considerable conscious effort to suppress it. Further, unlike the
fox, these do not seem to be cases in which forming these preferences is
irrational. We might say the same of individuals who revise their expectations
up in light of improved prospects in their country: there is nothing obviously
procedurally irrational about this process. Such cases are part of the wide-
spread phenomenon of hedonic adaptation, wherein, after a change in
circumstances—whether winning the lottery or losing the use of one’s legs—
‘people tend to adapt fairly quickly to the state they are in . . . and adjust their
reported utility accordingly’.¹⁴ Sometimes, such realigning of our standards in
light of our particular circumstances seems perfectly appropriate and benign:
it is rational for individuals to expect more when the standard of living in their
country increases. At other times, the realignment seems to conceal injustice:
unemployment does not cease to be a concern if individuals become ‘hard-
ened’ to the suffering it entails.
The standard, procedural accounts seem ill-placed to distinguish such
problematic adaptive preferences from benign adaptation (or, ‘adapted pref-
erences’ in Polly Mitchell’s terminology¹⁵). As such, they will be inadequate in
various social justice contexts. First, standard accounts are over-inclusive,
since they may include as adaptive preferences those that are unconsidered
yet do not seem unreliable from the perspective of justice, and would not
warrant public mistrust or political interference. For example, we may mis-
trust a woman’s unconsidered preference to undergo female genital cutting
(FGC), but not a similarly unconsidered preference not to undergo FGC.
Second, these accounts are under-inclusive, since they fail to include many
paradigm cases, such as those listed by Sen. As noted, these frequently lack the
relevant procedural flaws in their formation. For example, we may mistrust a
choice to undergo FGC that is procedurally autonomous and rational, if it is
made against unjust background conditions—if, say, it is the only means to
marriage, which in turn is necessary for economic security or social status—

¹³ See Khader (2011: 8–10, 42–6) for discussion of the discrepancy between Sen’s examples and
Elster’s account.
¹⁴ Mitchell 2018: 1007. Also see Kahneman 2000; Kahneman and Sugden 2005; Dolan and
Kahneman 2008.
¹⁵ Mitchell 2018.
    171

just as we mistrust the preferences of Sen’s beggar, housewife, or unemployed

person. Note that such mistrust does not involve questioning individuals’
rational capacities (as §6.6 will consider at length). On the contrary, my
point is that observing that a preference or choice like this does meet the
requirements of procedural rationality does not end discussion on whether
justice has been done and whether there may be a case for state intervention.
We must also take account of the social norms and wider context that
determine what preferences and decisions may become rational amongst a
constrained option set—one in which FGC is the best alternative, for example.
Given these problems, we may be tempted to conclude that, in the context
of social justice, these procedural accounts of adaptive preferences are simply
useless. Indeed, this is the conclusion many of those writing in this context
have come to. Martha Nussbaum, for example, rejects proceduralist accounts,
arguing that ‘[Elster] needs something he does not give us, a substantive theory
of justice and central goods’,¹⁶ if we are to be able to distinguish the adapta-
tions relevant to justice from those that are not. Serene Khader has also argued
against such accounts on the basis that they will lead us to mistrust preferences
that are not ‘worthy of public interrogation’,¹⁷ thus generating conclusions
about when public mistrust and interference are appropriate that are incon-
sistent with our intuitions.¹⁸
Such concerns are well summarised in Rosa Terlazzo’s insistence that all
accounts of adaptive preferences must meet the following criterion of ‘political
efficacy’:

An account must provide us with a valuable tool for combatting marginal-

ization and oppression, by explaining which preferences deserve social
suspicion and why they do so. It must include paradigm cases of adaptive
preference and exclude preferences that have nothing to do with the political
project.¹⁹

Traditional, proceduralist accounts will undoubtedly fail this test. As dis-

cussed, they will include preferences that have nothing to do with justice and
the political project—for example, the unconscious decision to reject FGC, and
the fox’s rejection of the inaccessible grapes (and other more realistic but
similarly trivial examples). In addition, and as the next section will discuss
further, they will exclude preferences that are relevant to justice and do deserve

¹⁶ Nussbaum 2001: 79. ¹⁷ Khader 2011: 75. ¹⁸ Khader 2011: 74–106; Khader 2009.
¹⁹ Terlazzo 2016: 210. Also, see Nussbaum 2000a, 2001; Barnes 2016a, 2009a.
172      

social suspicion. Yet this failure is hardly surprising given that they were never
designed to pass such a test, concerned as they are with procedural rationality
rather than social justice.
I agree, therefore, that we need an account of adaptive preferences that is
politically efficacious, and it is for this reason that I offer an account of justice
adaptive preferences: preferences that are a poor guide to individuals’ entitle-
ments. However, it does not follow that there is no value in maintaining an
account of adaptive preferences as procedurally non-autonomous and
irrational, and so a poor guide to individuals’ interests. That is, an account
of what I call well-being adaptive preferences. Some justice adaptive prefer-
ences are rational (and so not also well-being adaptive), whilst others are both
justice and well-being adaptive. Justice and well-being adaptive preferences,
then, are overlapping but distinct categories, and both should play a role in our
understanding and diagnosis of adaptive preferences since different responses
will be appropriate in different sorts of case (§6.7).

6.3 Well-Being Adaptive Preferences

If, then, an account of procedurally non-autonomous preferences is needed,

how is such an account to be cashed out? Elster saw the distinctiveness of
adaptive preferences in the fact that they are formed unconsciously, or ‘behind
our backs’, in light of diminished options.²⁰ Nussbaum, amongst others,
has argued convincingly that Elster’s account is likely to be over-inclusive
even by his own lights, demanding conditions of procedural rationality few
of our preferences meet. As she points out, we are not—and should not be—
‘suspicious of any desire that is formed through [unconscious] adjustment to
reality’.²¹ This is not simply the criticism that this account cannot pick out
adaptive preferences relevant to social justice, but the broader claim that it
cannot distinguish unreliable preferences in any context. Too many of our
preferences are unconsciously formed for this account to be useful: for
example,‘[t]he desire for food is not normally induced through conscious
hunger creation’.²² This is true, too, of more significant preferences: for
example, abandoning our childhood career aspirations for reasons we are
unconscious of, is not usually thought to cast doubt on the rationality or
reliability of our commitment to our adult occupation.

²⁰ Elster 1987: 117–19. ²¹ Nussbaum 2001: 78. ²² Colburn 2011: 57.

    173

Other authors have offered alternative accounts of why fox-and-grapes

cases of adaptation tend to be irrational, whilst character planning is not.
For example, Luc Bovens argues that adaptive preferences are irrational
because they do not cohere with our other beliefs and preferences.²³ As
such, these preferences cannot be reflectively endorsed by an agent. I focus
here, though, on Ben Colburn’s autonomy-based account, since this seems to
best capture the problem with these kinds of cases. In brief, Colburn argues
that adaptive preferences are the result of ‘covert influences’. By this he means
that the reasons for our adaptive preferences are necessarily hidden from us,
such that they could not explain our commitments if we were conscious of
them—we would either adduce independent reasons for our preference or
repudiate it. Thus, the fox’s claim that grapes are sour could no longer explain
his preference not to eat them if he were aware that the real (and covert) cause
of this preference was their inaccessibility. He would need another reason to
reject the grapes, or must give up the repudiation of them.
Colburn’s account does not imply that all unconsciously formed preferences
are adaptive (as Elster’s account seems to), nor that any influence on our
preferences is suspect. Rather, we should be suspicious of adaptive preferences
because they result from influences their possessor does not, and would not,
reflectively endorse, given their current disposition and convictions. The fox
does not believe being difficult to access is a good basis on which to form
preferences about fruit. More realistically, imagine that we realise that the
reason we have revised our career aspirations is that we lack the talent to
pursue our original dream, and possess the aptitude for our current path. This
need not cause us to doubt our goals. Although we may have been unconscious
of the reason for our career choice prior to reflection, this is a reason most would
accept on becoming aware of it (that is, they would reflectively endorse this
influence on their preferences). Compare this with a case in which we only
abandoned our dream of being a scientist due to being subtly undermined by a
sexist science teacher. We do not endorse his view (that ‘women are not cut out
for science’), so this could not justify our preference for non-scientific pursuits.
Possessing an adaptive preference of this sort—one arising from covert
influences—involves a lack of procedural rationality, since we are acting for
reasons we could not accept. More significantly, these preferences lack auton-
omy, understood as ‘deciding for oneself what is valuable, and living one’s life
in accordance with that decision’.²⁴ Deciding for oneself, Colburn argues,

²³ Bovens 2002. Bruckner (2009) offers a similar account. Also see Zimmerman (2003).
²⁴ Colburn 2011: 61; 2010: 19.
174      

requires both that we would endorse our values if we were to reflect on them,
and that our decisions and values are formed free from factors that undermine
our independence.²⁵ It is the independence condition that cannot be met when
an individual is subject to a covert influence: we cannot be said to be deciding
for ourselves when our reasons for actions are necessarily hidden from us.
Thus, individuals’ autonomy is undermined. This may be most obvious in
cases such as subliminal advertising and hypnotic suggestion, but also captures
cases such as the woman who has rejected science due to the more subtle
machinations of her teacher.
It also seems that our well-being would improve if we were no longer subject
to covert influences. In other words, we would be better off if we were to revise
the resulting preferences (or our reasons for holding them) such that we no
longer possessed preferences or desires for reasons we could not accept. On a
simplistic, hedonic conception of welfare this may not be true. After all,
individuals might genuinely experience a state of happiness or satisfaction
even when they have been subject to covert influences: the above individual
may feel real pleasure pursuing her new career even if she is necessarily
unconscious of her reason for choosing it. However, this may simply seem
to constitute evidence that well-being should not be reduced to positive mental
states given the malleability of what can produce these states.²⁶
On the contrary, it is plausible to believe that individuals’ well-being
consists in their life going well according to standards they could consciously
accept. This is a subjective view insofar as our well-being depends on what we
individually value, and not on our ability to access a particular set of goods in
which we might see no value.²⁷ However, it does not imply that well-being is
reduced to the achievement of some positive mental state, irrespective of the
means by which this is reached (be it brainwashing, hypnosis, or acclimation
to long-term deprivation). Thus, even if someone is able to reliably report
whether they are experiencing a state of happiness or satisfaction, this does not
imply that they are also a reliable guide to their own well-being in this wider
sense. An autistic individual excluded from most face-to-face socialising by
exclusionary social norms that stigmatise their unusual behaviours (stimming,
limited eye-contact, direct manner, say) might experience genuine happiness
in the online communities and relationships they engage with instead.
Nonetheless, it seems plausible to suggest they would be better off if they

²⁵ Colburn 2010: 25; 2011: 63.

²⁶ As Sen (1999a: 63) e.g. has convincingly argued: ‘the mental metric of pleasure or desire is just too
malleable to be a firm guide to deprivation and disadvantage’.
²⁷ As per Arneson’s (2000a, 2006) view; see §4.6 for further discussion.
    175

were not pushed into expressing their sociability in this manner for reasons
they cannot endorse. In other words, their well-being would increase if they
could form (and fulfil) their preferences regarding social relationships without
covert influences.
Since my interest is in questions of justice, I will not get too side-tracked by
the debates over the various conceptions of well-being.²⁸ However, I will say a
little more in support of the claim that we should not assume that fulfilling
preferences formed as a result of covert influences will improve their
possessor’s well-being. First, these preferences are unstable insofar as they
are based on reasons the possessor could not endorse, and so will be subject to
change whenever the true root of these preferences is made known to them.
Relatedly, we can expect that individuals would prefer a situation in which
they had preferences for reasons they could endorse, perhaps even if this is one
in which they were less satisfied. Our well-being surely depends on achieving
goods or goals that we believe are valuable. Well-being adaptive preferences,
arising from covert influences, do not reflect our values. Instead, they depend
on norms, actions, or circumstances that we could not agree to be good
reasons for our current preferences.
This is not to say, however, that those who are subject to covert influences
must disavow the resulting preferences. Even if we originally embarked on a
particular path or life plan for reasons we could not have endorsed, it may be
that after pursuing it for a sustained time we can now give independent
reasons for our current preference that we could endorse. Returning to our
dissuaded scientist example: we might not endorse the (sexist) reasons that led
us to begin pursuing philosophy instead, but if we now see more value in
philosophy than science, then we can endorse our current preference for
it. The same goes for the autistic individual now immersed in online commu-
nities. We can escape the impact of the covert influence and decide for
ourselves that our goals are valuable. If we do so, this would cease to be a
well-being adaptive preference, and achieving this goal would contribute to
our well-being.
There will, of course, be borderline cases in which it is difficult to determine
whether someone really is necessarily unconscious of the reason for their
preference. Imagine, for example, a disabled individual who learns, perhaps
in therapy, that they have some preference (not going to university, say) as a
result of being subject to stigma. Yet despite ‘knowing’ the primary cause of
their preference, they do not attend to it, push it to the back of their mind, and

²⁸ For a useful discussion of well-being and adaptation see van der Deijl (2017, 2020).
176      

deceive themselves into thinking they have this preference for other reasons.
In a sense they are conscious of the real reason for the preference (the stigma).
However, insofar as they continue to deceive themselves about the reason for
their preference, fail to take on board or internalise the additional evidence
they now have, and are unable to adduce reasons to continue to hold this
preference that they could accept, then they are both not truly conscious of this
influence and—more importantly—could not continue to hold the preference
if they were to become properly conscious of it (absent further reasons). Thus,
insofar as the covert influence continues to operate, it seems that this remains
a well-being adaptive preference.²⁹
It may be objected at this point that my view of well-being could open the
way for unjustified paternalism. I have suggested that we should not assume
that preferences and desires are a reliable guide to an individual’s well-being if
they are necessarily unconscious of their reason for holding them, regardless of
their mental state. Instead, well-being depends on autonomously identifying
and pursuing a conception of the good, and the covert influence underlying
adaptative preferences prevents this. But, some may object, if people are happy
or satisfied, what business is it of ours how they came to be that way, and why
should the state intervene to disabuse them of the real causes of their prefer-
ences? Fleurbaey and Blanchet, for example, argue that ‘the correct attitude
about this phenomenon [of adaptation] is to worry about the ills to which
people do not adapt, such as pain and noise, rather than insisting that the ills
to which they do adapt are still priorities’.³⁰ However, such worries misidentify
the goal of this account. As §6.2 discussed, diagnosing a preference as well-
being adaptive need not imply that state intervention is justified. As I will
argue at length, it is justice adaptive preferences that warrant such interference,
and well-being adaptive preferences may, but need not, be justice adaptive.
Thus, there certainly will be cases in which the state has no business either
questioning the autonomous basis of individuals’ happiness or attempting to
improve their well-being by improving the autonomy of their preference
formation. However, this, I contend, is because it is not the state’s business
to attempt to maximise individuals’ well-being (as the next chapter will
discuss). It is, however, the business of the state to ensure that individuals
are not deprived of opportunities they are entitled to as the result of covert
influences, especially when this influence is itself exerted through unjust

²⁹ Thanks to Adam Cureton and David Wasserman for raising, and pushing me to address, such
cases.
³⁰ Fleurbaey and Blanchet 2013: 169.
    177

means. That not all well-being adaptive preferences warrant public interven-
tion, then, does not mean none of them do or that they should never be the
priority of the state, as §6.6 will show. Whilst Colburn’s account does a good
job of capturing the phenomenon that motivated Elster, it is not, as we have
seen, a good candidate for achieving the goals of a political project. Not all
well-being adaptive preferences should be a political priority, but in addition,
some paradigm adaptive preferences, meriting intervention, are not well-being
adaptive. As we will see, being well-being adaptive is only one way in which a
preference may be relevantly unreliable.
Before concluding this section, it is worth briefly considering how this
discussion of well-being adaptive preferences relates to Barnes’s ‘discordance
view’, since hers is perhaps the only other account of adaptive preferences
developed directly in relation to disability. Barnes argues that preferences are
unreliably adaptive if there is ‘a clear disconnect’ with ‘the rest of the person’s
life’, including ‘what produces things like anxiety, fear, and antipathy in their
lives’.³¹ This may, therefore, seem to capture a similar phenomenon to
Colburn’s focus on covert influences. If these preferences are not compatible
with, or connected to, an individual’s other commitments—and, indeed, cause
negative feelings in other contexts—then we may not be able to consciously
endorse our reason for holding them.
Like Colburn, Barnes is not focusing on the context of justice or attempting
to provide an account that fulfils the criterion of political efficacy. Like
Colburn’s account, then, Barnes’s does not seem able to capture all the cases
we would want to capture in the context of a political project. For example, an
abused woman might, as Barnes points out, experience anxiety, fear, and
depression. However, women who have internalised more mundane forms
of sexist oppression may not experience such cognitive dissonance. This surely
would not give us reason to overlook such adaptation from the perspective of
justice, yet the fact that these women experience no anxiety is sufficient to
conclude, on Barnes’s view, that this is not an adaptive preference. These
preferences would also not be well-being adaptive, on the Colburnian account
I have adopted, insofar as the individuals could endorse the reason for their
preferences—they really believe a woman’s place is in the home, say. As such,
these preferences may be a reliable reflection of the women’s interests and
values, given the norms they currently ascribe to. Thus, satisfying this prefer-
ence could improve their well-being. Nonetheless, as I will argue, such pref-
erences may still warrant public concern (they may be justice adaptive).

³¹ Barnes 2016a: 140.

178      

A further problem with Barnes’s account is over-inclusivity, first, in a way

mirroring the objection to Colburn’s and other pure proceduralist accounts,
whereby trivial cases of adaptation (the fox, or tastes for particular food or
music) may be well-being adaptive, but do not warrant public scrutiny,
mistrust, or interference. In addition, though, Barnes’s account faces a further
over-inclusivity objection. Situations and conditions we on the whole prefer
can cause stress, anxiety, and fear, yet such preferences are not obviously
unreliable (from the perspective of either justice or well-being). We may hate
public speaking, avoid it whenever we can, and feel great anxiety at the
prospect of it, yet choose to engage in it in pursuit of a goal about which we
feel deeply passionate. There seems no reason to distrust the preference to
engage in public speaking in this case.
It is worth emphasising that Barnes’s primary concern is the testimonial
injustice entailed by unwarranted attributions of adaptive preferences—in
particular, that our negative identity-prejudicial stereotypes of disabled indi-
viduals cause hearers to give an unduly deflated level of credibility to their
testimony.³² Barnes characterises the inclination to set aside disability-positive
testimony as ‘merely adaptive’, and hence untrustworthy, as an instance of the
general phenomenon of the failure to respect disabled individuals as trust-
worthy, credible givers of knowledge. Avoiding misdiagnosis and over-
application of an adaptive preference model is certainly important. The
harm of not being respected as a giver of knowledge has been amply demon-
strated,³³ and disabled individuals are often subject to this harm. Nonetheless,
as it stands, her positive proposal provides insufficient guidance as to when
adaptive preference diagnoses are warranted, and failing to diagnose genuine
adaptive preferences may, from the perspective of justice, be as much of a
problem as misdiagnosis.

6.4 Why Identify Well-Being Adaptive Preferences?

To summarise, well-being adaptive preferences are procedurally irrational

(since we act for reasons we could not endorse) and non-autonomous (since
they are the result of covert influences, and so lack independence). For this
reason, they seem to be a poor guide to individuals’ interests (if independent
reasons cannot be adduced for them). As such, fulfilling these preferences may
not improve individuals’ well-being as I understand it (achieving ends they

³² See Barnes 2016a: 135–9; Fricker 2007. ³³ E.g. Fricker 2007; Dotson 2012; Matthes 2016.
    179

endorse as having value) even if doing so makes them happy. In the context of
social justice, however, this concept is both under- and over-inclusive. First, it
excludes preferences that are rational, autonomous, and so not well-being
adaptive, but which nonetheless are a poor guide to our distributive entitle-
ments and so should be overlooked for the purposes of justice.³⁴ For example,
the preferences of disabled individuals not to enter exclusionary institutions of
higher education may be rational, and a good guide to their interests in these
unjust circumstances. They may be consistent with their other views (about
wishing to avoid oppressive relationships or demeaning situations) and they
would endorse the reason for these preferences. Yet their satisfaction should
not imply justice has been done.
Second, an account of well-being adaptive preferences includes preferences
that are irrelevant to justice. The fox, for example, has a well-being adaptive
preference, but this does not seem to be an injustice the state should rectify. As
a more concrete case, consider preferences for particular aesthetic experiences.
Deaf individuals, for example, may come to prefer a life without music as a
result of influences they would not reflectively endorse. To be clear, this is
likely true of many individuals’ tastes, and there is no reason to think that
individuals with impairments are particularly vulnerable to such covert influ-
ences: hearing individuals’ preferences for some of the dominant styles of
music in their culture may be of a similar form. As Chapter 4 discussed, whilst
justice entitles individuals to certain opportunities—maybe even opportunities
for aesthetic experiences—it does not entitle individuals to every possible
opportunity in a domain, or even the opportunities they would most prefer
(assuming scarcity of resources). Thus, if deaf individuals have access to other
forms of aesthetic experience, their inability to hear music is not a concern of
justice, so nor is it a concern of justice if their preference not to listen to music
is well-being adaptive—no more than people’s preferences for certain musical
styles is.
Given that an account of well-being adaptive preferences fails to pick out
cases of adaptation that are relevant to justice, it may seem that we should
follow in the footsteps of others working on adaptive preferences in the
context of social justice and simply abandon any attempt to identify such
preferences. I disagree. First, the arguments in favour of doing so tend to be
based on the assumption that diagnosing a well-being adaptive preference will

³⁴ I suggest that there are three categories of such preferences: non-autonomous character planning,
justice adaptive choices, and the coherent internalisation of oppressive norms, in addition to prefer-
ences that are both well-being adaptive (hence irrational) and also relevant to justice. All four categories
are discussed in §6.6.
180      

wrongly imply that its possessor is unusually irrational or unreflective, or

that their preferences warrant public concern or interference.³⁵ However,
since well-being adaptive preferences are not intended to meet the criteria of
political efficacy, they need not (and cannot) capture only paradigm cases where
interference seems justified. We can, therefore, acknowledge that many—
perhaps all—people have well-being adaptive preferences.³⁶ Adopting an
account of well-being adaptive preferences, then, does not require reaching
the dubious conclusions that only oppressed and mistreated individuals have
them, nor that they are less reflective or worse at reasoning than others, nor that
possessing well-being adaptive preferences justifies state interference.
Furthermore, there are positive reasons to adopt such an account. Most
obviously, it is useful as a means to achieving conceptual clarity. If, as
I contend, the term ‘adaptive preferences’ is used to capture two distinct
phenomena—wherein those discussing adaptive preferences in the context
of social justice have in mind something quite different from those discussing
adaptive preferences in the context of autonomy and rational choice—it is
worth making this explicit. There are also important practical benefits to
distinguishing well-being and justice adaptive preferences. Justice adaptive
preferences need not also be well-being adaptive, but the two categories do
overlap and, as §6.7 considers, we should treat rational justice adaptive
preferences differently from justice adaptive preferences that are also well-
being adaptive.³⁷ Drawing this distinction also makes clear that having a
justice adaptive preference need not imply we are irrational, and that not
having any justice adaptive preferences does not necessarily mean our prefer-
ence set is fully rational and autonomous.
Moreover, given that domains of justice might change, it is worth identify-
ing well-being adaptive preferences that are now irrelevant from the point of
view of justice, since they may cease to be so. In the past, ensuring access to
aesthetic experiences and leisure may have been considered to be beyond the
scope of government activity—in future, other things might be included.
Finally, whilst well-being adaptive preferences may not warrant direct state
action, acknowledging the irrationality of many of our preferences may guide

³⁵ Khader (2011: 80), for example, objects to autonomy-based accounts on the basis that they entail
the ‘dangerous and unwarranted empirical assumption . . . that people with adaptive preferences reflect
on their behaviour less than everyone else’.
³⁶ However, whilst well-being adaptive preferences are likely to be common, individuals who are
subject to oppression are more likely to have justice adaptive preferences as well, as we will see.
³⁷ Specifically, the former case may simply require an alteration of circumstances so that an
individual’s choices can reflect their preferences, whilst the latter may require deliberation, discussion,
and consciousness-raising, alongside the provision of options.
    181

individual behaviour: encouraging us to critically reflect on our preferences

and, if we cannot find reasons for them we would reflectively endorse, change
them. Further, though it may be unjustifiably paternalist for a state to interfere
in an individual’s well-being adaptive preferences, if they are widely prevalent
the state may legitimately attempt to reduce this prevalence—for example,
with measures that promote individuals’ capacity for reflection or which create
opportunities for reflection. This might include public information campaigns
outlining the benefits of unconsidered options, such as careers, exercise, or
leisure pursuits unfamiliar to, or overlooked, by at least some groups.³⁸

6.5 Perfectionism, Justice, and the Need for

a Substantive Approach

If our goal is to combat injustice, and ensure that preferences that endorse and
perpetuate oppression are disregarded, then a more substantive account of
adaptive preferences is needed: one that takes account of the content of
preferences, and not merely the process by which they were formed. At the
far end of this spectrum is Nussbaum’s account, according to which adaptive
preferences are simply those with the wrong content: preferences for what we
ought not to prefer, formed in light of diminished options.³⁹ Hence, for
Nussbaum, identifying adaptive preferences requires ‘a substantive theory of
justice and central goods’.⁴⁰ For her, whether a preference is rational, ‘con-
sidered’, or even formed in the presence of substantive conditions, such as ‘an
absence of traditional hierarchy, absence of fear, and a sense of one’s worth
and dignity’, is not decisive in determining whether it is reliable.⁴¹ As Barnes
explains Nussbaum’s view: a ‘change in preferences is only problematic insofar
as it leads to a preference for something which one should not, ceteris paribus,
prefer’.⁴² Thus, if an unconsidered preference is for a good way of life
(economic empowerment, say) and a considered preference is not (for physical
abuse, for example), it is the former that is reliable.⁴³

³⁸ This seems to be required by Colburn’s (2010: 94–8) ‘autonomy-minded liberalism’ (also see
§7.8).
³⁹ Nussbaum 2001; 2000a: 122–42. On some interpretations, this is Sen’s view too (e.g. Barnes
2016a; 2009a). However, Sen’s approach is complex and not always clearly defined, and I will not take a
stand on how it should best be understood here (see Qizilbash 2007).
⁴⁰ Nussbaum 2001: 79. ⁴¹ Nussbaum 2001: 74. ⁴² Barnes 2009a: 5.
⁴³ E.g. ‘[a] habituated preference not to have an item on the list [of central functionings] . . . will not
count in the social choice function, and the equally habituated preference to have these things will
count’ (Nussbaum 2001: 84). Although Nussbaum insists that, since ‘desire is not brutish, but an
intelligent reaching-out-for-the-good, we ought to show it some respect . . . in the process of
182      

Nussbaum’s perfectionist approach will certainly allow us to identify

paradigm cases, and provide a tool for combating oppression. We may
worry, however, that this approach will be overly narrow, including as reliable
only preferences for what has been predetermined to be valuable. Thus, as
Terlazzo has argued, ‘persons whose conceptions of the good reject the items
on the perfectionist list will not be shown the respect granted to those with
“proper” preferences’.⁴⁴ Particularly worrying in the current context, it may
lead to the exclusion of disabled individuals’ preferences if disability or
impairment is taken to be suboptimal, since a preference for these forms of
functioning or ways of life will not be deemed ‘proper’. Indeed, Barnes has
argued that this is exactly the implication of adopting Nussbaum’s substantive
perfectionist approach. She contends that since ‘[p]hysical disability repre-
sents, according to the capabilities approach, an absence of one or more basic
goods (bodily integrity, physical health etc . . . )’,⁴⁵ then it ‘cannot be as optimal
as a relevantly similar non-disabled life’.⁴⁶ Thus, disabled individuals’ prefer-
ences for their life cannot be considered reliable on this view. As I have argued
both in earlier chapters and elsewhere, the capability approach need not
evaluate disability so negatively.⁴⁷ Nonetheless, even if disability-positive tes-
timony is not automatically deemed unreliably adaptive, the general worry
remains that Nussbaum’s approach excludes those who dispute her concep-
tion of value, and ‘counts a number of widely and sincerely-held conceptions
of the good as necessarily non-autonomous’.⁴⁸
In light of these problems, both Khader and Terlazzo offer substantive
proceduralist accounts of adaptive preferences, both of them devised for the
development context and focusing on the preferences of women in the devel-
oping world. Both set substantive conditions on preference formation, but aim
to avoid ruling out any preferences on the basis of their content alone. For
Khader, ‘APs are preferences incompatible with an agent’s basic wellbeing . . .
formed under unjust conditions’.⁴⁹ Khader, then, is more willing to focus on
content, seeing adaptive preferences as any ‘behavior or belief whereby an
individual perpetuates her deprivation’, though this must be ‘causally related

justification’, this merely involves giving individuals’ desires ‘a heuristic role . . . [and] a modest ancillary
role in political justification’ (Nussbaum 2004: 200). Whilst Nussbaum is optimistic that proceduralist,
informed-desire accounts will coincide with her substantive approach, then, such proceduralism does
no serious justificatory work. Given this, it may seem that Nussbaum is not ‘relying’ on preferences at
all. However, it has been argued that proceduralism should be given a greater justificatory role in her
account (e.g. Jaggar (2006: 309–20)).
⁴⁴ Terlazzo 2014: 186. ⁴⁵ Barnes 2009a: 2. ⁴⁶ Barnes 2009a: 6.
⁴⁷ Begon 2015. It is possible, though, that this is an implication of Nussbaum’s account (e.g.
Nussbaum 2006a: 155–223).
⁴⁸ Terlazzo 2016: 214. ⁴⁹ Khader 2012: 302.
    183

to her deprivation’, in the sense that it ‘would disappear upon exposure to

superior conditions and/or information’.⁵⁰ As such, we have special reason to
worry about preferences with specific content—those that perpetuate depriv-
ation, independently and substantively defined—but they are only adaptive
preferences if they arose in particular circumstances. On this view, preferences
formed in the same circumstances but with different content are not
adaptive preferences (an oppressed woman’s preference for pineapple over
mango); nor are preferences ‘nonconducive to basic flourishing’⁵¹ formed in
conditions conducive to flourishing (endangering one’s life by engaging in
extreme sports).
Terlazzo argues that individuals should engage in critical reflection on their
preferences, and do so in the presence of recognised alternatives that should be
both valuable and ‘live’, in the sense that individuals can reasonably see
themselves exercising them given their current values and ambitions.⁵² The
substantive element of this procedural approach concerns the specification of
these valuable live options. Further, to meet the criteria of political efficacy she
argues that we should restrict the category of adaptive preferences to ‘core
preferences: that is, preferences that are centrally important to a person’s
broader plan of life or sense of self ’.⁵³ Terlazzo is particularly concerned to
remain neutral about the content of adaptive preferences, but whilst she does
not necessarily rule out any conceptions of the good, she does contend that a
‘unifying characteristic’ of all adaptive preferences will be that they are ones
that people would be ‘highly unlikely to want to have were they to reflect on
their own preferences in a considered way and in better circumstances’.⁵⁴ This
may suggest an implicit judgement about their likely content.
Both Khader and Terlazzo emphasise that having adaptive preferences need
not be irrational, or imply an agent lacks the capacity for autonomy. Khader
delineates different forms of adaptive preferences.⁵⁵ She allows for the exist-
ence of ‘paradigmatic adaptive preferences’, wherein someone ‘perpetuates
injustice against herself because of a near-completely distorted worldview’.⁵⁶
More often, though, individuals are mistaken in some domains, but do not
entirely lack the capacity for autonomous choice and critical reflection. For
example: individuals may internalise some aspects of their oppression but
question others; they may be mistaken about facts, but not values; or they may
engage in forced trade-offs amongst well-being in different domains of their

⁵⁰ Khader 2013: 313. ⁵¹ Khader 2011: 17. ⁵² Terlazzo 2016: 215.

⁵³ Terlazzo 2016: 216. ⁵⁴ Terlazzo 2014: 195. ⁵⁵ Khader 2013: 317–20.
⁵⁶ Khader 2013: 311.
184      

lives. Along similar lines Terlazzo emphasises the distinction between global
and local autonomy: individuals can have the global capacity for autonomy
despite occasional failures to exercise it locally.⁵⁷
Yet though they delineate various ways in which preferences may be
unreliable, these approaches still aim to provide a unified account, identifying
a single central feature that typifies all adaptive preferences.⁵⁸ They make a
binary judgement between preferences that are adaptive and unreliable, and
those that are not adaptive and are reliable. Such proceduralist accounts
therefore obscure the different ways in which preferences may be adaptive.
Nonetheless, as I will argue, a substantive proceduralist account will play an
important role in identifying when preferences are justice adaptive.⁵⁹

6.6 Justice Adaptive Preferences

On at least some welfarist accounts of distributive justice, no tenable distinc-

tion can be drawn between well-being and justice adaptive preferences: if
individuals’ interests directly determine their entitlements, then if their pref-
erences are a reliable guide to their best interests, they are also a reliable guide
to their distributive entitlements (and vice versa). Further, this distinction—
and, indeed, any discussion of adaptive preferences at all—will also have no
role in, or relevance for, objectivist theories that understand individuals’
entitlements (and perhaps also their well-being) to be determined entirely
independently of their preferences. However, many theories of distributive
justice do not fall into either of these camps: whilst individuals’ entitlements
are not reducible to what would make them happy or satisfied, neither are they
determined wholly independently of the views and values of those to whom
they are provided. In this group we might include Dworkin’s equality of
resources, Nussbaum’s capability approach, Arneson’s luck egalitarianism,
Cohen’s equal access to advantage, Rawls’s theory of justice—and, of course,
the version of the capability approach outlined in Chapter 4, on which I will
focus here.⁶⁰
Preferences play two primary roles in such theories of distributive justice.
First, preferences play some part in determining the content of our distributive

⁵⁷ Terlazzo 2016: 217–20. Also on this distinction, see Colburn (2010: 4, 21).
⁵⁸ This is also true of Barnes’s (2016a; 2009a) account, discussed above.
⁵⁹ Also see Begon (2015: 248–9).
⁶⁰ On the former approaches, see e.g. Dworkin 2000; Nussbaum 2000a; Arneson 2000b; Cohen 2011;
Rawls 1999.
    185

entitlements. Broadly, I accept that preferences and other evaluative states

have epistemic value, providing evidence of what is good for people, and so
which capabilities ought to be central. Nussbaum, for example, hopes that if
individuals undergo a process of reflective equilibrium, testing their con-
sidered moral judgements against their conception of what a flourishing life
entails, we can reach an overlapping consensus on the functionings that are
central to human life amongst a large, cross-cultural group.
However, whilst I accept this general approach, it is worth flagging some key
differences with my position. First, since I have argued that capabilities should
be conceptualised as opportunities to control domains of our life (rather than
abilities to function), I suggest the consensus should be reached on these
central domains, rather than central functionings: mobility not walking, com-
munication not speech, to return to our previous examples. Next, I agree with
Wolff and de-Shalit’s emphasis on ensuring the process of reflective equilib-
rium is public: that is, taking the theories and intuitions of the general public as
‘a fundamental point of input’.⁶¹ In this context, particular care should be
taken to incorporate the views of those who have experience of living with an
impairment or caring for those who do. Finally, as noted above, Nussbaum
uses this process as a heuristic device to confirm, rather than independently
identify, the content of the list of central capabilities: ‘the independent moral
argument must take priority’.⁶² For Nussbaum, then, ‘when the moral
approach we favour seems to clash with people’s desires’, we can permissibly
conclude that ‘ “what the majority desires here is wrong” ’, and should not
lead to the elimination of a central capability.⁶³ I favour giving this process a
more independent role, assuming an appropriately wide range of views is
incorporated.⁶⁴
Adaptive preferences in this context might limit the content of our entitle-
ments, and the realm of state assistance. Insofar as our goal is to reach an
overlapping consensus amongst a large group, anomalous individual cases
may not pose too great a challenge: if a few individuals come to believe that
control over their mobility, education, health, or bodily integrity is unimport-
ant this may not seem to undermine the wider consensus on the value of these
opportunities. However, insofar as oppressive practices and structural injust-
ices often work to covertly influence groups of individuals with shared char-
acteristics, by pressuring them or limiting their opportunities, this raises the

⁶¹ Wolff and de-Shalit 2007: 43. ⁶² Nussbaum 2004: 200. ⁶³ Nussbaum 2004: 201.
⁶⁴ See Wolff and de-Shalit 2007; Jaggar 2006. Since what is advocated is a process of reflective
equilibrium, the suggestion is not that the unexamined views or intuitions of the majority should
simply be decisive in determining the content of our entitlements.
186      

possibility that there will be larger groups who reject what might otherwise be
deemed central capabilities.
Such challenges to the content of the list of central capabilities might seem
to occur only if we simply accept people’s unexamined judgements about what
is valuable. The inclusion of a process of reflective equilibrium is, of course,
designed to avoid this: putting pressure, for example, on the consistency of the
claim that control over bodily integrity is important in general, but not when it
comes to women’s (especially disabled women’s) control over their reproduct-
ive rights. Indeed, the reason to adopt substantive proceduralist accounts, such
as those of Khader and Terlazzo outlined above, is to weed out adaptive
preferences and their influence on limiting the scope of our entitlements.
Thus, the point here is not that adaptive preferences will inevitably restrict the
realm of distributive entitlements. Nor does this give us a reason to abandon the
use of views and testimony in identifying the central capabilities.⁶⁵ Rather,
adaptive preferences present a problem that can, and should, be avoided, and
so are a factor we should be mindful of when designing our deliberative
processes. This need not lead us to adopt a substantive, perfectionist view of
the preferences that are bad for us and so should be ignored (qua Nussbaum) or
treated with suspicion (qua Khader). However, as I will consider, it will require
the adoption of a substantive proceduralist account that requires, inter alia, that
individuals have some experience of the capability to control that they are
rejecting before their rejection is taken to be reliable.⁶⁶
The second role for our preferences is to determine whether we make use of
the opportunities or resources to which we are entitled in a particular case.
Liberal approaches are unlikely to compel individuals to function, both
because the intervention required would be illegitimately paternalist, and
because an individual’s autonomous decision not to exercise an opportunity
legitimates its absence. Recall that one of the central insights of the capability
approach is that a starving, but not a fasting, individual is necessarily a
concern of justice. The problem with adaptive preferences in this context is
that they may lead us to wrongly consider some instance of self-harm or self-
sacrifice to be voluntary, and hence not unjust. For example, we may fail to
recognise that an individual who is apparently malnourished by choice does
not really have the capability to be nourished, perhaps because they have

⁶⁵ As e.g. Nussbaum (2000a, 2004) and Arneson (2000b) do.

⁶⁶ Note that this does not mean individuals must have experienced some functioning, predeter-
mined as valuable, since capabilities to function are not what individuals are entitled to. Thus, we can
have experience of the capability for mobility without walking, communication without talking,
forming social relationships without neurotypical social engagement, and so on. (See §6.7.)
    187

adapted to and internalised norms that demand that women give up their food
to male family members. Their claim that they are choosing to make this
sacrifice may be insufficient to demonstrate they are really exercising
control here.
The well-being adaptive preferences on which I have primarily focused can
potentially cause problems in either of these two roles: whenever we adapt in
response to covert influences, our views on the content of our entitlements,
and on whether we take up these entitlements, will be procedurally non-
autonomous and irrational, and not a reflection of our considered judgements.
Any resulting limitation on the scope of our entitlements or instance of self-
sacrifice is thus suspect. However, as I have noted repeatedly, certain adaptive
preferences may be a political priority, even if they are not well-being adaptive,
in that they should neither determine our general entitlements nor justify self-
sacrifice as voluntary. Preferences that are thus problematic from the perspec-
tive of justice, I call (naturally enough) justice adaptive preferences. In add-
ition to well-being adaptive preferences that are also justice adaptive (category
iv), I identify three further categories of justice adaptive preferences: (i) non-
autonomous character planning (where individuals consciously downgrade
opportunities they have been wrongly denied), (ii) justice adaptive choices
(where individuals’ counterfactual preferences are autonomous and reliable,
but their choices are rendered non-autonomous by the limitations of their
circumstances), and (iii) the coherent internalisation of oppressive norms
(where individuals internalise a self-conception according to which they are
not entitled to certain opportunities). I shall consider all four in turn.

6.6.1 Non-Autonomous Character Planning

When discussing traditional accounts of adaptive preferences, I noted that

they are usually defined in direct contrast to conscious, rational, and autono-
mous character planning. We cannot trust the fox who could not accept that
the reason he believes grapes are sour is their inaccessibility, and has no
further autonomous reason for this preference; but this is not true of the fox
who consciously and autonomously cultivates a preference for sweeter, lower-
hanging fruit. However, the stark contrast in this example may be misleading.
Consciously revising our preferences in light of diminished options can be
autonomous, but it need not be, depending on the degree to which our options
are constrained. For example, someone who has adapted to ableist biases in
hiring practices, and concluded that they do not really need opportunities for
188      

meaningful work, may prioritise personal relationships and hobbies instead.

Further, they may be aware of the reasons for this change of priorities, or
would not repudiate them if they were. Such preferences are rational, and do
not demonstrate a failure of the individual’s autonomous capacities. Indeed,
these preferences may function as a reliable guide to their interests (in these
unjust circumstances)—they are not well-being adaptive.
Nonetheless, we surely should not use them to conclude either that the
opportunity for meaningful employment is, in general, not a concern of
justice, or that this individual has voluntarily chosen to forgo this opportun-
ity. The reason to thus distrust this preference is indeed that it is not
autonomous, but not because the individual who has formed it is a defective
agent, who lacks, or has misused, the capacity for autonomous choice.
Rather, their circumstances have prevented them from being autonomous:
from deciding for themselves what is valuable, and being able to pursue it.
Specifically, these circumstances, and other circumstances in which individ-
uals are unjustly constrained, undercut individuals’ independence: they are
subject to factors that undermine their ability to ‘decide for themselves’.⁶⁷
Yet in this case, their independence is not undermined by covert influences
(as §6.3 considered) but by a lack of acceptable alternatives, which the agent
may be well aware of.
This example makes clear the need for a form of substantive proceduralism
to identify preferences that are non-autonomous, and thus unreliable, in this
way. There are two ways in which we could understand individuals’ options in
such cases as being unjustly constrained. First, they have been prevented by
unjust processes from having options they would otherwise have—in other
words, the circumstances determining the availability of options were unjust.
Second, they lack options they are entitled to—in other words, it is the content
of their options that is objectionable.
Whilst both conditions are met in the above case, I believe it is the latter that
is decisive. Imagine an individual who adapts to the lack of opportunities for
meaningful employment (understood in the broadest possible sense, and not
necessarily connected to monetary remuneration) because (implausibly) we
lack the technology or resources to enable them to participate in any possible
occupation in our society. Although the process of exclusion is not unjust, the
adaptation here is still relevantly unreliable. We have, I contend, no reason to
conclude that this is not a capability we should seek to provide to all people

⁶⁷ For a discussion of independence as a necessary component of autonomy, see Colburn (2011:

26–31).
    189

and to which they are entitled, nor that the individual who gives up on this
realm of their life has made the sacrifice voluntarily.⁶⁸ Thus, this example
begins to demonstrate the importance of first allowing individuals to have
opportunities to exert control in central domains of their life (having access to
all the central capabilities) before their rejection of the capacity for such
control can be fully trusted.

6.6.2 Justice Adaptive Choices

The second category can be called ‘justice adaptive choices’. When individuals
engage in character planning, their underlying preferences change. However,
there are also cases in which we can distinguish what people choose (given
limited options) from what they counterfactually prefer. For example, a
disabled individual may choose not to pursue higher education on the basis
that the institutions in which it is offered tend to be exclusive and unaccom-
modating. This may be consistent with their further preferences (not wishing
to associate with people or institutions that mistreat them), and being aware of
their reasons need not lead them to change their mind. However, they may
prefer a situation in which accessible higher education was available. In such a
case, neither the choice nor the preference should be diagnosed as a well-being
adaptive preference: both are rational, and may be a good guide to their
interests given the available options. As above, such cases do not impugn the
individual’s capacity for autonomous choice, but, also as above, their circum-
stances prevent them from being autonomous, and thus scrutiny, mistrust, or
even intervention may be warranted.
It may be objected that such individuals do not really have adaptive
preferences at all, since their counterfactual preferences are autonomous, and
reliable from the point of view of justice. This is certainly true, and it is
important to acknowledge that, when options are constrained, individuals’
choices may only reflect their preferences in a very local sense—I choose not to
go to university because I prefer not to attend these (exclusionary) universities.
As such, their counterfactual preferences can inform a theory of justice: we
have no reason to doubt their testimony in determining an account of our

⁶⁸ Recall that a capability can be amongst our distributive entitlements even if it is not currently
feasible to provide it to all individuals (see §2.4). This individual still retains a conditional entitlement
to the capability: they should be provided it if it were possible, and the state should work to make it
possible.
190      

general entitlements, and no reason to think these individuals would threaten

the consensus on the value of certain domains of control. Circumstances
undermine the independence, and so autonomy of their choices (regarding
whether to go to university, in this case), but not the autonomy of their wider
judgements or preferences (control over life-long education, say).
Nonetheless, sometimes individuals’ choices are the only guidance
we have regarding their preferences, and when they are rendered non-
autonomous in this way they should be deemed justice adaptive. That is,
their choices should not be used to suggest that what they choose to forgo is
generally unimportant, or that their choice to forgo it is voluntary. For
example, if an individual with a mobility impairment is forced to choose
between modifying their home so they can move around it more easily, and
pursuing leisure activities, the fact that they choose the former is not
evidence that they dispute the value of leisure in human life, nor that they
have voluntarily chosen not to pursue it.

6.6.3 The Coherent Internalisation of Oppressive Norms

In other cases, individuals’ reasons to acquiesce in their mistreatment are more

deep-rooted, and are not accompanied by counterfactual preferences for a
more just alternative. For example, if the above individual has internalised a
conception of disabled people as ‘not worthy’ of higher education, they may
not prefer it to be available in an accessible form since they simply think ‘it’s
not for people like me’. Further, the internalisation may be sufficiently com-
plete that this view coheres with their other beliefs and preferences, and they
would not change under conscious examination, or repudiate the cause of
their preference if they were aware of it. After all, they may now really believe
higher education, say, is not for the likes of them. Or, imagine that the sexist
science teacher in §6.3 was so successful that we, on being made aware of the
reasons we gave up science, would say that the teacher did the right thing,
since science, after all, is ‘not for women’.
Indeed, in such cases, acquiescent individuals may be right that they would
not be able to flourish in such environments, though this does not confirm the
truth of their general conviction that they lack the capacity or ability to learn
the skills or complete the tasks such institutions aim to teach. It may be
objected that this is psychologically implausible—experience of oneself and
others as competent and intelligent will be incongruent with such a conception
of disabled people (or women). Nonetheless, it is possible for sustained
    191

oppression to lead to a consistent conception of oneself as worthless, at least

with regards to certain opportunities.⁶⁹
These preferences would not fall foul of Colburn’s covert influence criteria,
since these individuals can and would consciously accept the reason for their
preference and may even think those who influenced them acted correctly.
Thus, as in the case of justice adaptive choices, we may have no reason to
doubt their possessors’ authority regarding their well-being given their current
preference set.⁷⁰ In the case of justice adaptive choices, we make a choice that does
not reflect what we would ideally prefer because it is the best option available:
our preferences are a reliable guide to our interests but we are unable to satisfy
them. In these cases, we have shifted our complete set of preferences and values
in light of external oppressive influences. Thus, if we accept that individuals’ well-
being depends on their current priorities and values (as §6.3 suggested), then
their lives will go better if these preferences are satisfied. Yet accepting this does
not imply that these preferences are reliable from the perspective of justice.
What matters here is whether these preferences should determine (and
indeed limit) individuals’ distributive entitlements at either the general or
individual level, and I contend they should not. Although this influence
on our preferences or choices is not covert (we are not necessarily unconscious
of it) it surely undermines their independence, and so their autonomy. When
individuals exist in a context in which they are railroaded into one particular
(and very restricted) view of their abilities and their options, they have hardly
been enabled to identify a plan of life for themselves. Again, this does not
impugn their capacities, but merely acknowledges how circumstances can
undermine the autonomy of our choices and preferences. Thus, allowing
that individuals may be rational and a reliable guide to their interests even
when their self-conception is seriously undermined does not mean that their
acquiescence will be deemed autonomous, nor their oppression legitimised.
Only when they have had access to the capability to control their life in central
domains should their rejection of options be treated as reliable.

6.6.4 Justice and Well-Being Adaptive Preferences

As the last three cases have demonstrated, justice adaptive preferences need
not also be well-being adaptive, but this does not mean these two categories

⁶⁹ See Khader (2011: 13–17; 2012: 307) on partial and complete losses of self-worth.
⁷⁰ For discussion of similar cases, see Baber (2007: 199–200).
192      

cannot overlap. The inaccessibility of grapes may not be a concern of justice,

but many preferences that arise as the result of covert influences are relevant to
justice, and so are the kinds of thing we should discount when determining the
contents of our entitlements or whether a sacrifice is genuinely voluntary.
Well-being adaptive preferences are justice adaptive, then, when the dimin-
ished options to which individuals irrationally adapt occur in domains of
their life over which they ought, as a matter of justice, to be able to exercise
autonomy and pursue their own conception of the good. For example, a deaf
individual’s preferences not to listen to music and not to engage politically
might both be well-being adaptive preferences, but only the latter is also a
justice adaptive preference.⁷¹ It is, therefore, important to emphasise that
whilst it is possible—indeed, likely—that many individuals will have well-
being adaptive preferences, for those in unjust circumstances these are more
likely to be justice adaptive preferences too, given the kinds of covert influ-
ences to which they are subject.

6.7 Responding to Justice Adaptive Preferences

Part of the reason that it is so important to disambiguate the various ways in

which, and the reasons why, preferences may be adaptive, is that different
forms of intervention will be called for in different cases. Broadly speaking,
when the problem is rooted in individuals’ circumstances rather than their
mistakes or deficient capacities, then our interventions should focus on alter-
ing these circumstances. Of course, in many cases these cannot be simply
pulled apart since circumstances can problematically influence individuals’
judgements. The distinction can be most clearly seen in cases where only
individuals’ choices are adaptive, and their counterfactual preferences are
reliable. Here, our focus should be on providing the acceptable alternatives
that they lack. Moreover, we should engage in a process of deliberation with
the affected individuals to determine what the content of these alternatives
should be, since we have no reason to discount their counterfactual prefer-
ences. For example, reforms to higher education institutions ought to
be informed (inter alia) by those who currently choose not to enter these
institutions due to their exclusionary character. Thus, the state should not

⁷¹ Assuming our entitlements are conceptualised as domains of control, and assuming (plausibly)
that the capacity for the specific functioning of listening to music is not required to have control in any
domain, whilst completely and non-autonomously opting out of engagement in the political sphere
does imply a lack of control in a domain in which we ought to be able to decide for ourselves.
    193

intervene in individuals’ adaptive choices (on the grounds that they are
adaptive choices), but should provide the conditions that allow them to choose
in a way that reflects their underlying preferences.
However, in at least some cases of non-autonomous character planning, the
coherent internalisation of oppressive norms, and well-being and justice
adaptive preferences, it is likely that the provision of further options will be
insufficient. In these cases, individuals’ adaptation to diminished options may
have led to some degree of value distortion. Those who have undergone non-
autonomous character planning have consciously devalued the opportunities
they have been deprived of. Those who have undergone the coherent intern-
alisation of oppressive norms have internalised a conception of themselves as
not entitled to, or worthy of, or fitted for various opportunities. And those who
have irrational justice adaptive preferences reject certain opportunities for
reasons they could not, themselves, accept.
The provision of acceptable alternatives may nonetheless be of benefit in
these cases too. The core of all cases of justice adaptive preferences is that they
are non-autonomous, and thus unreliable, as a result of constrained options,
and that they concern content relevant to justice. Providing access to the
ability to exercise genuine control in central domains is not just what individ-
uals are entitled to then, but also provides the preconditions for reasoning
about the content of these entitlements and determining whether and how to
use them.⁷² When we have an acceptable range of options and the real freedom
to choose between them then we will not have to engage in character planning
to prefer those few options we do have (category (i)); will be less likely to
internalise the oppressive norms that may be causing this restriction (category
(iii)); and will be less likely to be subject to covert influences on preference
formation (category (iv)). (As well as individuals’ choices being able to reflect
their counterfactual preferences—category (ii).)
Nonetheless, when value distortion has occurred, mere provision of alter-
natives may be insufficient to undo the damage, and prevent these preferences
from being justice adaptive. To demonstrate this point I will consider how
such value distortions can undermine the reliability of preferences in the two
roles for them I have outlined in the context of justice. These are, first, their
effect on the content of our general entitlements, and second, their use in
determining whether an instance of self-deprivation is really voluntary.

⁷² Whilst there is a degree of circularity here, it need not be vicious. Rather, it is the simple claim that
the preconditions of coming to a reflective equilibrium on the content of our theory of justice should
take place in ‘conditions favorable for deliberation and judgment in general’ (Rawls 1999: 42). (Also see
Begon 2015.)
194      

First, then, an individual may repudiate the value of an opportunity or

capability entirely. Such a preference might take the form: ‘I live well without
x, so x cannot be essential to a decent life’. For example, autistic individuals
may question whether certain forms of social interaction are as essential as
neurotypical individuals assume; deaf individuals may question whether hear-
ing is essential; and paraplegic individuals may doubt the importance of being
able to walk. I have suggested that we should adopt a form of substantive
proceduralism, at least requiring that individuals experience control of the
relevant sort before their rejection of the value of a distributive entitlement
should be taken as a reason to consider vetoing its inclusion from a list of such
entitlements. It is important to emphasise from the outset, though, that if we
accept the conclusions of Chapter 4, then our entitlements should be concep-
tualised at a relatively high level of abstraction: as capabilities for control, not
capabilities to function. If this is the case, then individuals who repudiate, for
example, the importance of being able to walk or hear, will not threaten
the possibility of overlapping consensus. The relevant opportunities here are
for mobility and aesthetic experience, and these individuals have not been
deprived of these capabilities, nor do they repudiate their value.
Nonetheless, when individuals do reject the capability to control a domain
of their life in the absence of experience of it, we may have warrant to set this
preference aside as unreliably adaptive: we should not discard widely shared
intuitions about the control central to a decent human life solely on the basis
of the intuitions of individuals who have never had such control. Disabled
individuals who reject the value of work, education, or relationships simply
because they have been deprived of them as the result of ableist biases, for
example, should not lead us to conclude that these opportunities are not a
concern of justice.⁷³
As we have seen, the loss of opportunities and control that is frequently
experienced by disabled individuals most often occurs as the result of external
environment and social norms. Whilst paradigm impairments (like more
mundane inabilities) generally involve the loss of specific functionings, most
need not lead to the loss of more general control, and thus need not be
disadvantageous.⁷⁴ In many cases, then, we have no reason to think that
disabled individuals will possess justice adaptive preferences that pose a threat
to the breadth of our distributive entitlements. Consider, again, Helen Keller’s
description of the depth and value of her sensory experience:

⁷³ I argue for this approach in greater detail elsewhere (Begon 2015).

⁷⁴ As Chapter 5 has argued.
    195

My world is built of touch-sensations, devoid of physical colour and sound;

but without colour and sound, it breathes and throbs with life. Every object is
associated in my mind with tactual qualities which, combined in countless
ways give me a sense of power, or of beauty, or of incongruity.⁷⁵

Far from implying that individuals with sensory impairments cannot possess
or value control of their sensory experience, it surely shows just how valuable
and important control over this domain can be, regardless of whether some
particular functionings are closed to us.⁷⁶
Thus, the preferences of individuals with non-disabling impairments, who
do not lack opportunities they are entitled to, have been formed in the right
circumstances: with experience of the central capabilities. As such, we have no
reason to question the reliability of their preferences in the context of justice.
Yet we also have no reason to want to find some rationale for distrusting their
preferences, since we have no basis for thinking that they have any reason to
repudiate the central capabilities. Indeed, even individuals who do lack all
control over some part of their life seem unlikely to insist that such control
lacks value. Yet, if they did, it does not seem insulting to suggest they need to
have this experience before their repudiation should be decisive in determin-
ing the value of a domain. Especially, perhaps, given that the experience
that is required is of general control (mobility, say), and not of specific
putatively valuable functionings (such as walking). As such, we will only
diagnose individuals’ preferences as justice adaptive in relation to determining
general entitlements in a very narrow range of cases.
It may seem more likely that adaptive preferences grounded in value
distortions will lead to problematic adaptive preferences in the individual
domain. If individuals are taught that ‘people like them’ are not worthy of
some valuable opportunity or the ability to control some part of their life,
this may not cause them to doubt the value or importance of such control
in general, but could lead them to repudiate its value in their life. Such
preferences might take the form: ‘x might be good for others, but it’s not for
people like me—and I’m happy without it’. For example, an individual with
Down’s syndrome might take the view that education is valuable, but conclude
that ‘it’s not for them’ if their educative environment is not an inclusive one.

⁷⁵ Keller in Barnes 2009a: 15. See §4.4.

⁷⁶ This is contrary to Barnes’s (2009a) claim that individuals with impairments (or, in Barnes’s
terminology, disabilities) would be unable to possess certain central capabilities.
196      

These individuals’ preferences may be used to determine the opportunities

that are relevant to justice, but should not be used as evidence that an
individual has what they are entitled to: the individual with Down’s does not
dispute the value of education, but their ‘choice’ or preference not to engage in
it should not lead us to conclude that justice has been done. The capability to
control part of our life does not merely require formal freedom from interfer-
ence, but far more substantive conditions (as §4.2 discussed). Importantly in
this context, making a choice or forming a preference in some domain will
require both that acceptable alternatives are available to an individual, and that
these choices are considered live options, such that individuals have a genuine
‘internal capability’.⁷⁷ Exactly how these meaningful opportunities can be
provided is a complex question, but it will likely require some form of
deliberation with affected individuals, which involves consciousness-raising
(such that they understand the significance and potential value of certain
options) as well as the provision of opportunities.
In summary, justice adaptive choices can be avoided by providing individ-
uals with options that reflect their preferences, where they are entitled to these
as a matter of justice. However, insofar as other justice adaptive preferences
involve value distortions, avoiding these requires both that individuals have
acceptable options open to them (to ensure their values are not distorted by a
lack of exposure to reasonable alternatives in the first place) and being taught
to see why these options might be valuable and, more importantly, to see
themselves as the kind of people who can and should be able to exercise them.

6.8 Conclusion: Distrust without Insult?

Having a (justice) adaptive preference, then, need not involve a failure of

procedural rationality or the suggestion that an individual is fooling them-
selves. However, although not all adaptive preferences are irrational, they are
all non-autonomous. Thus, we may worry that the account remains vulnerable
to the objection that diagnosing adaptive preferences has the insulting impli-
cation that individuals with adaptive preferences are defective agents or
‘dupes’. This is of particular concern given that I have suggested that individ-
uals who are subject to injustice—including individuals with many paradigm
impairments—will be more likely to have justice adaptive preferences. It is
important to emphasise, therefore, that acknowledging adaptive preferences

⁷⁷ E.g. Nussbaum 2011b: 20–3, and see §4.2.

    197

are non-autonomous does not imply that diagnosing someone as having one
must be insulting.⁷⁸
First, in many cases adaptive preferences are a rational response to unjust
circumstances. In these instances, autonomy is undermined by limitations of
circumstance rather than defects in individuals’ capacities. Second, having a
value distortion on some particular issue does not imply an individual is
wholly unreliable, or lacks the capacity to make autonomous choices or
form reliable preferences in any area of their life. The fact that individuals
have faced restrictions in their options in one domain, making the autonomy
of their goals and preferences in this domain suspect, gives us no reason to
think that this will spill over into areas in which they have not faced these
restrictions: again, because the problem lies in circumstances rather than
defective agential capacities.
Finally, many (all?) individuals will possess some well-being adaptive pref-
erences: preferences we hold for reasons that are necessarily opaque to us.
Insofar as these preferences are irrational, they do indicate flawed reasoning in
a sense. However, this does not imply that individuals with irrational justice
adaptive preferences are unusually irrational or incapable of formulating
consistent preferences. We all adapt to restrictions in our options, and many
of us do so in irrational ways (we have well-being adaptive preferences).
Indeed, there is ample evidence that our decision-making is often flawed
and overly reliant on heuristics, and the real reasons behind our choices are
frequently obscure to us: for example, post-hoc rationalisation, present bias,
and the phenomenon of anchoring.⁷⁹ In the context of the political project—
that is, our decisions about what individuals are entitled to and whether their
decisions to reject some offered opportunity for control is truly voluntary—
most of these mistakes are as irrelevant as the reason why the fox has given up
his desire for grapes. Yet for those individuals unfortunate enough to face
restrictions as the result of oppression and mistreatment, the resulting adap-
tations are more likely to be a concern of justice (hence, justice adaptive
preferences).
The reason individuals in unjust circumstances are more prone to justice
adaptive preferences is not, then, that they are more defective agents, and
whilst it might seem (literally) to add insult to injury to doubt the preferences
and testimony of individuals already subject to injustice, the alternative is

⁷⁸ I do not mean to downplay individuals’ experiences, and do not deny that (especially given
currently unjust social circumstances) this may be experienced as insulting.
⁷⁹ E.g. see Kahneman (2011); Kahneman et al. (1982); Le Grand and New (2015).
198      

worse. If we fail to acknowledge the ways in which individuals’ values may be

distorted as a result of mistreatment, indoctrination, and restriction of options,
and that this can prevent them from effectively pursuing those options they
prefer, then we are in danger of perpetuating and upholding the unjust status
quo that restricts these options in the first place. I began this chapter with a
dilemma: on the one hand, we do not want to simply disregard members of
oppressed groups as untrustworthy agents, yet on the other, we do not want to
allow that cases in which individuals have adapted to deprivation no longer
constitute injustices. By clarifying what is entailed by a diagnosis of adaptive
preferences, I have demonstrated that such a diagnosis does not imply dis-
missing someone as a ‘dupe’, and whilst misdiagnosis should be avoided this
does not mean all views and preferences should be beyond question.
 7
Don’t do it for my Sake
Providing Control, Avoiding Paternalism, and
Applying the Justice Account of Disability

7.1 Introduction

Disability, as I have defined it, involves lacking the capabilities we are entitled
to. More specifically, lacking the ability to control central domains of our life
as a result of an impairment (anomalous, not deficient, physical or cognitive
functioning) in combination with various features of our circumstances
(social, political, and environmental), our individual resource shares, and
other personal features. From this it follows that when individuals are disabled
by their impairments they are owed state assistance to ensure that they do have
the opportunities they are entitled to, and hence will cease to be disabled, as far
as possible, though they may remain impaired. Yet the relationship between
disabled individuals and the state institutions that ought to ensure their just
treatment has a deeply troubled history. All too often disabled individuals are
portrayed as passive recipients, on whom a beneficent state chooses to bestow
assistance to compensate them for the misfortunes that they face—if the state
takes the needs of disabled individuals seriously at all.
There is much to object to in this portrayal. First, as I have already argued,
much state assistance is not supererogatory—which it can choose to bestow or
not—but morally required, not least because disability is often the result of
unjust institutional structures, social norms, and environmental circumstances
the state designs, upholds, or fails to rectify. Even when this is not the case, a
commitment to the capability approach I have outlined implies individuals are
entitled to acceptable options in a domain, even if this requires additional
resources in light of their atypical functioning capacities. Second, the state’s
goal should not be to counterbalance or offset the effects of some perceived
misfortune, by attempting to ‘cure’ impairments if possible and providing
individuals with some equivalently valuable good if not. As I have also argued,
whilst impairments can cause limitations we are currently unable to rectify,

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0008
200      

in many cases they are not the only or most significant cause of individuals’
disadvantage. Rather, we should focus on ensuring all individuals can reach an
acceptable level of functioning, whatever atypical mode this takes, rather than
enabling identical opportunity sets. We should not, therefore, assume impair-
ments are the problem nor should our goal be to neutralise their effect as much
as we are able. Justice demands the elimination of neither the diverse ways of
being that impairment enables nor their negative effects on well-being (though
these tend to be much less extensive than the able-bodied assume).
My concern in this chapter is with the third element of the above charac-
terisation: of disabled individuals as passive recipients on whom help is
bestowed. It is an unavoidable feature of any redistributive scheme that it
will involve some degree of intervention with its members. This is the basis of
right-libertarian objections to such systems. However, there need be nothing
objectionable about state policies that in some way restrict our options when
the goal is to enable both us and the rest of society to have greater capacity to
direct the course of our own lives. To pick the most obvious example: taxation
reduces our freedom to spend the totality of our income as we wish, but state
spending on infrastructure, education, healthcare, and law enforcement
(amongst much else) provides the context in which we are able to earn and
securely possess that income in the first place.
Controversy might arise when individuals dispute the goals of state policy
or the means by which they pursue them. Yet my concern here is not with
ensuring that all individuals are entirely satisfied with, or wholly approve
of, every policy pursued by the state, but that individuals are appropriately
respected as autonomous agents. This, I argue, requires (at least) a commit-
ment to anti-paternalism. The importance of the commitment to anti-
paternalism is particularly relevant in the context of disability. As the previous
chapter has made clear, disabled individuals are particularly vulnerable to
being considered too unreliable to have their preferences and agency
respected. However, as that chapter has also shown, there is no reason to
think disabled individuals are more prone to failures of rationality or auton-
omy than other individuals (except insofar as their circumstances reduce their
options and may thus deform their choices or preference formation).
Continuing this book’s central theme, then, a focus on disability allows us to
recognise a more general point about theories of justice: that our justification
for redistributive policies should include a commitment to anti-paternalism.
The goal of this chapter is to defend a particular account of anti-
paternalism, to show why paternalism can be objectionable, and to demon-
strate that my own version of the capability approach is not objectionably
  ’       201

paternalist. First, then, whilst the language of paternalism is common—and

commonly used to criticise disability policy—providing a definition has been a
matter of considerable and ongoing debate. I will reject the popular ‘insult
account’ of paternalism in favour of the authority view, according to which
individuals have authority over their own good which paternalist policies fail to
respect. Thus, a commitment to anti-paternalism means believing that individ-
uals’ well-being cannot be used as a reason for action, or as justification for a
policy, if those individuals repudiate the putative benefit, since doing so will fail
to show appropriate recognition respect to autonomous agents in two ways (§7.2).
I next consider why individuals may not permit their good to be used as a
reason (§7.3), and what respecting this authority requires in the context of
distributive justice (§7.4). It might seem that the capability approach can never
advocate paternalist intervention, since it does not advocate intervention at
all—it simply offers individuals opportunities. I show that these offers can
constitute intervention in the relevant sense (§7.5, §7.6), but that they can be
justified without unauthorised appeals to individuals’ good (§7.7). Hence,
paternalism can be avoided. I then argue that when individuals are acting
non-autonomously they may lack absolute authority over their own good, and
thus, that our anti-paternalism should be soft paternalism: there is not the
same strict prohibition on interfering with non-autonomous choices (§7.8).
Finally, §7.9 briefly considers the implications of my account for parental
decisions regarding their current and future children.
When policies are pursued that do not track the preferences of the individ-
uals they are supposed to serve, this is undoubtedly objectionable. However,
I will argue that is especially objectionable when these policies are justified—
and perhaps can only be justified—by appeal to the interests of supposed
beneficiaries who repudiate these benefits. Yet paternalist redistribution is not
merely objectionable because it misuses individuals’ good, but because it
mistakes the very aim of redistribution. As I have argued repeatedly, the
goal of the state should not be to benefit individuals as much as they are
able, but to ensure they can control the course of their lives. Thus, if the state
were to promote well-being it must have the authority of putative beneficiaries
to do so, but this should not be the state’s goal. Redistributive paternalists go
wrong, then, not merely by misusing well-being, but by having well-being
promotion as their core objective. Our goal should, instead, be providing
people with the capabilities to control central domains of their lives.
Providing this control and allowing people to be self-directing will leave plenty
of scope for individuals to make mistakes, but this is what respecting people as
autonomous agents demands.
202      

7.2 Defending Anti-Paternalism

Before considering how paternalism can be avoided, it is essential to under-

stand how it should be understood. In general terms, paternalism involves an
action (interference, very broadly construed), performed for some particular
reason, (the good of the person interfered with), and without their consent.
Or, as Kalle Grill puts it, to be paternalist, ‘an action must be unwelcome,
interfering, and benevolent’.¹ Anti-paternalism involves accepting that a ‘con-
sideration of a person’s good should not ever count in favor of limiting
her liberty’—again, without her consent.² Grill helpfully describes the anti-
paternalist commitment as a filter preventing certain reasons (the promotion
of a person’s good) functioning as a reason to limit their liberty.³
Yet we may wonder why the use of such a reason is objectionable. After all,
acting to promote others’ good ordinarily seems laudable. To foreshadow my
later argument, I defend an authority view and argue that this suggests a dual
account of the wrong of paternalism. First, insofar as individuals are authori-
tative regarding their own good, using their good as a reason contrary to their
wishes will mean that any interference made on this basis lacks the very
justification it claims. As we will see, how serious this is will depend on the
nature of the interference, and what other reasons justify the action if motives
are mixed. Second, ignoring an individual’s authority over their good and
impermissibly using it as a reason demonstrates a lack of recognition respect
for them. That is, it involves a failure to show someone the basic level of
respect to which we are all entitled as moral equals.⁴ Arguably neither feature
is unique to paternalism, but I suggest that the combination of the two
captures the core of what is objectionable about paternalism.⁵

¹ Grill 2018: 54. On paternalism as an action-reason, see Grill (2007). This definition is not
uncontroversial, but I believe Grill is right that it captures the normative core of paternalism. For
alternative accounts see e.g. Dworkin (2013, 2014), Archard (1980), Begon (2016b), Coons and Weber
(2013), Cornell (2015).
² Grill 2015: 49. Some might consider this final caveat unnecessary: if the interfered with individual
consents then we will not need to limit her liberty. However, assuming we allow individuals to make
binding decisions about their future selves, then this will not necessarily be true. For example, the
sailors limited Odysseus’s liberty, arguably for his own good, but given he consented this does not seem
to be a case of paternalism. (For more on paternalism and consent, see Husak 2015.) It may also be
objected that individuals need not actively consent that their good be used as a reason: we can act on the
presumption that others’ goods are available as a reason unless an individual exercises their right to
refuse (see Parry 2017: 357–8). As I argue below (§7.7), whilst active consent might not be required,
mere absence of refusal may also be insufficient, at least in cases of state paternalism.
³ Grill 2015: 52.
⁴ Darwall 1977. Also see Fox (2019) for an application to paternalism, discussed further below.
⁵ Some contend the latter element is unique to paternalism. However, I believe we should restrict the
category of paternalism to cases that plausibly involve interference with the paternalised individual, and
  ’       203

Before defending this view, however, I will briefly outline some of the debate
on this topic.⁶ One common justification of anti-paternalism is that paternal-
ism is, in some way, self-defeating. First, we may believe individuals’ well-
being is partly constituted by autonomous choice: the act of setting and
pursing our own goals is valuable in itself. Or, as Mill famously puts it: a
person’s ‘own mode of laying out his existence is best, not because it is best in
itself but because it is his own mode’.⁷ Therefore, paternalist interference with
someone’s decisions will not, in fact, promote their well-being. Second, we
may hold that individuals have unique epistemic access to information about
what is in their best interests and how their well-being can best be promoted.
Since outsiders will tend to be mistaken about these facts, paternalist interfer-
ence will often be misdirected, and so again will not promote well-being. To
return to Mill: when the public interferes with personal conduct ‘the odds are
that it interferes wrongly and in the wrong place’.⁸ Third, it might seem that,
even if the paternaliser does manage to accurately identify and promote
aspects of their object’s well-being, the harms of intervention will outweigh
any benefits bestowed—for example, the harm of resentment might offset any
gains received. Thus, paternalist interference will still not promote overall
well-being.
These arguments draw attention to important considerations for any puta-
tive paternalist: autonomously forming and pursing a conception of the good
is an important part of a good human life, individuals are often in an
epistemically privileged position regarding what is in their interests and
what will best promote them, and paternalist interventions are often misdir-
ected or ultimately self-defeating. Concerns of this sort may provide a good
basis on which to object to many putatively paternalist policies. However, they
do not provide principled reasons to reject to paternalism. The latter two
arguments are contingent, and do not rule out paternalism if it were better
informed and more sensitively imposed. The first argument relies on the
controversial assumption that persons’ choices seldom set back their own
well-being. But this seems implausible given all we know about people’s
tendency to irrationality, short-sightedness, and weakness of will, as well as
our propensity to prioritise things other than our own well-being. Further, it
is clear that sometimes very minor interferences can generate very big

an individual’s good might be used as a reason for actions other than interfering with them. Most
obviously, and most relevantly in the context of devising redistributive policies, as a reason to interfere
with others—for example, by taxing them. I discuss these third-party cases further below.
⁶ I have done this in much greater detail elsewhere (Begon 2016b).
⁷ Mill 1974 (1859): 133. ⁸ Mill 1974 (1859): 151.
204      

welfare gains.⁹ Thus, whilst we must balance the benefits of any paternalist
intervention against its complete costs, it is implausible to insist that paternal-
ism will never work. Indeed, there would be little reason to debate the merits of
pursuing paternalist policies if there was simply no such thing as successful
paternalism. So, if a commitment to paternalism merely requires that ‘inter-
vention is permissible when, all things considered, it serves the target’s best
interests’,¹⁰ then the question that remains for the anti-paternalist is: why not
make people better off?
The most common answer is that there is something uniquely insulting
about such paternalism, even if successful. It involves treating an adult as if
they were a dependent child, or ‘as if he or she (at least temporarily) lacks the
ability to rationally pursue his or her own good’.¹¹ Paternalism, it might seem,
involves the substitution of the paternaliser’s judgement for the paternalisee’s
on the assumption that the former’s judgement is superior, and hence the
latter’s in some way inferior.¹² It is certainly natural to find it objectionable
when someone forms such a ‘negative judgement’ about our practical reason-
ing, willpower, or emotion management.¹³ No one wants to be treated as if
they are ‘too stupid to run their own lives’.¹⁴ However, it is not clear that this
apparent ‘insult’ is either unique to paternalism or unjustified.
One way of understanding the insult of paternalism is that it demonstrates a
lack of appraisal respect: that is, respect grounded in a positive appraisal of a
person’s competence or aptitude.¹⁵ Judgements about individuals’ inferior
reasoning ability or willpower certainly seem to indicate a negative appraisal,
but are we wronged by others when they form such judgements? Not obvi-
ously. After all, ‘we are not morally entitled to demand that others think well
of us’.¹⁶ Further, given that it has been amply demonstrated that mistakes in
reasoning and weakness of will are common, such judgements might only
indicate that someone has determined our ‘rationality is imperfect, and so is
open to the kinds of error we all are’.¹⁷ Moreover, as David Enoch has argued,
whether or not someone has the ability to deliberate and act rationally is
‘a factual question, and whether or not we should believe this is, arguably, fully

⁹ The various ‘libertarian paternalist’ or nudge policies are perhaps the most obvious example here:
altering the placement of healthy or unhealthy options in a cafeteria, or changing a pension plan from
being opt-in to opt-out may produce significant benefits, yet will arguably involve no loss of liberty at all
(see e.g. Sunstein and Thaler 2003; Sunstein 2014; Thaler and Sunstein 2008; and for discussion see e.g.
Bovens 2009; Goodwin 2012; Grüne-Yanoff 2012; Kelly 2013; Le Grand and New 2015; Mills 2013; 2015;
Mitchell 2005; Wilkinson 2013).
¹⁰ Hanna 2018a: 228 (my emphasis).
¹¹ Quong 2011: 101. See Begon (2016a: 361–7) for discussion of insult accounts.
¹² Shiffrin 2000: 218. ¹³ Quong 2011: 81–2. ¹⁴ Anderson 1999: 330.
¹⁵ Darwall 1977. ¹⁶ Fox 2019: 325. ¹⁷ De Marneffe 2006: 80.
  ’       205

determined by the evidence’.¹⁸ Thus, our judgement of someone’s capacities

and whether they have ‘manifested characteristics that make [them] deserving
of . . . positive appraisal’¹⁹ is an epistemic question. We should not be insulted
by an appropriately informed and well-supported negative appraisal of us,
even if we would prefer its content were otherwise.
It is worth briefly flagging that, even if we allow that the process of reaching
a negative judgement is an epistemic question, this does not imply that
questions of (in)justice are entirely irrelevant. We may not always form a
judgement that reflects the relevant evidence, and the failure to do so may not
indicate a mere lack of epistemic virtues. If our negative judgement results
from a prejudicial stereotype, we may be guilty of testimonial injustice.²⁰ This
is particularly relevant in the context of paternalism directed towards disabled
individuals since, as the previous chapter discussed, prejudicial stereotypes
regarding disabled individuals’ incompetence, irrationality, and untrust-
worthiness have frequently led to their testimony being ascribed an unwar-
ranted degree of unreliability, and hence being easily and regularly dismissed.
This is an important reminder that much of what is objectionable about
paternalism concerns its (mis)application in practice, which can happen all
too easily. However, I am assuming that successful paternalism is possible, and
can grant that this will be paternalism free of testimonial injustice—in part,
because testimonial injustice generates harms,²¹ so any resulting paternalism is
unlikely to advance an individual’s interests all things considered. Yet even if
paternalism grounded in a negative prejudicial stereotype did promote well-
being, we can grant that the most plausible defence of paternalism would
eschew apparently successful paternalist interventions that instantiate injust-
ice or violate some other normative constraint.
Leaving aside these cases, then, we can conclude that our low appraisal of
someone’s ability to successfully promote their own good might be justified.
Perhaps a proponent of the insult view could respond that the problem with
paternalism is not judging someone incompetent, but acting on this judge-
ment: not believing someone is child-like, but treating them as if they are.
However, this approach also faces problems. First, it does not seem to capture

¹⁸ Enoch 2016: 29. ¹⁹ Darwall 1977: 38–9.

²⁰ Fricker describes testimonial injustice as occurring ‘when prejudice causes a hearer to give a
deflated level of credibility to a speaker's word’ (Fricker 2007: 1). For further discussion, see e.g. Dotson
2012; Piovarchy 2021; Wanderer 2012. Note, I am not concerned here with epistemic paternalism,
according to which ‘we are sometimes justified in interfering with the inquiry of another for her own
epistemic good without consulting her on the issue’ (Ahlstrom-Vij 2013: 4), but testimonial injustice
committed in pursuit of ‘standard’ paternalism.
²¹ Fricker 2007: 43–59.
206      

the spirit of many defences of the view, which rely on the intuition that being
judged stupid, incompetent, or irrational is insulting. Indeed, one prominent
proponent of the insult view—Jonathan Quong—denies that interference is a
necessary component of paternalism.²² Second, it does not seem to identify
something that is uniquely wrong about paternalism.²³ Exactly the same
analysis would apply to cases in which we interfere with someone not for
their sake, but for someone else’s. For example, imagine we take it upon
ourselves to check and alter a colleague’s lecture slides out of a (justified)
low assessment of their competence, and consequent fear they will impart
incorrect information to students about essay deadlines or seminars times. We
tend to think there is something specifically wrong about paternalism, but on
the insult account such third-party cases would be wrong (when they are) for
exactly the same reason.²⁴
More fundamentally, it is not clear that the insult view identifies a feature of
actions that is wrong-making. This may be how we assess the above lecture
example: the interfering lecturer did nothing wrong. Or, for a clearer case,
consider judges withholding information from jurors for fear it will bias their
decision-making process. Assuming this is done for the sake of giving the
defendant a fair trial—and not for the epistemic good of the jurors—this is not
a case of paternalism.²⁵ Nonetheless, it instantiates the supposedly core wrong
of paternalism on the insult view—interference with an individual justified by
a negative judgement about their capacities—yet does not seem wrong. Such
examples illustrate that there need be nothing insulting—or wrong—about
either forming or acting on the belief that people may make mistakes in
reasoning. This is especially evident when the judgement is grounded in
general beliefs about flawed human reasoning, and not the incompetence of
specific individuals, as is true of most state paternalism.
I believe that the problem with paternalism is, indeed, acting on a negative
judgement and not merely forming it. In other words, we cannot expect others
not to make (justifiably) low assessments of our capacities, but we can expect
them not to act on this basis—we might not always direct our lives well, but
they are still our lives to direct. The insult view cannot explain why this is so,

²² Quong: 2011. ²³ Fox 2019.

²⁴ This is a point that Shiffrin (2000) acknowledges, and which leads her to conclude that such third-
party cases are normatively equivalent to paternalism. Yet no other authors are willing to bite this
bullet, likely because this moves us away from what seems paternalism’s distinctive feature: that the
paternaliser acts benevolently to the paternalisee.
²⁵ Kristoffer Ahlstrom-Vij (2018) uses this as a core example of epistemic paternalism. However, on
the action-reason view of paternalism, the reason for the judge’s interference matters, and without the
benevolent motive is not paternalist.
  ’       207

but the authority account of anti-paternalism can. The problem is not the lack
of appraisal respect shown to our judgements as the insult view suggests, but
the lack of recognition respect demonstrated by acting on them, as I will
explain. The insult view perhaps remains so plausible because it can indeed be
insulting when we are not shown appropriate recognition respect, but this
does not mean the core wrong can be grounded in this insult.
According to the authority account we are authoritative over what is best for
us regardless of whether we are always right. As such, we have the right to act
as we will in matters concerning our own good, and a right to control whether
anyone is permitted to use our good as a reason. The former point may be
more familiar. As Daniel Groll argues, if individuals are competent their will
should be structurally decisive: ‘it is meant to supplant the reason-giving force
of other considerations not because it outweighs those other considerations
but because it is meant to silence or exclude those other considerations’.²⁶
Treating someone’s will as authoritative can be contrasted with treating it as
part of an all-things-considered assessment of their good—i.e. as substantially
decisive. This latter approach should be reserved for those whose will is not
‘intact’—for example, some individuals with dementia—and so lack the rele-
vant authority over their own good.²⁷
Having our will treated as merely substantially (and not structurally)
decisive need not involve interfering with someone’s decisions given that, for
the reasons noted earlier, this may not actually promote their well-being. For
example, interventions may be physically and psychologically distressing, and
even for individuals who lack competence, part of what makes their life to go
well may be that their will plays a central role in determining its shape.²⁸ Yet
even if we act in accordance with someone’s will because it is their will there
seems to be something objectionable, indeed paternalist, about treating com-
petent individuals as if it is just one reason amongst others²⁹—for example, a
doctor weighing someone’s refusal to undergo a medical treatment amongst its
pros and cons, rather than the refusal simply ending the discussion.
The idea, then, is that we do not need to believe that individuals choose well,
but we must nonetheless treat their will as authoritative in matters concerning
their own interests. Groll focuses on how this should silence other consider-
ations: if I have made a decision in some sphere over which I have legitimate
control, any further considerations are simply irrelevant. This same authority

²⁶ Groll 2012: 701. ²⁷ Groll 2012: 718. ²⁸ Groll 2012: 704.

²⁹ For Groll, having one’s will treated as only substantially decisive is always paternalist, but only
impermissible if an individual is competent. I will consider issues of competence in §7.8.
208      

over our interests also implies our good cannot be utilised as a reason without
our consent—the idea captured by Grill’s anti-paternalist filter.³⁰ We are the
sole authority over the use of our own good, and, without permission (however
understood), others simply cannot use this good when deciding how to treat
us. Jonathan Parry has helpfully described this right as the Power of Prudential
Exclusion (PPE). As he puts it:

individuals have a normative power to control whether others may act to

promote their good. Just as others may not use my body and property
without my authorization, they may not justify their actions by appeal to
reasons grounded in my good if I validly refuse to be benefited.³¹

If we grant individuals this authority, it will be impermissible to use the

benefits to an individual as a reason to act if they repudiate these benefits.
As mentioned at the outset of this section, this raises two problems for the
paternalist: first, paternalist intervention will lack its putative justification and
so fail to show appropriate recognition respect for its object qua unjustified
intervention; and second, using an individual’s good as a reason without their
authority violates their PPE, constituting a further failure to show appropriate
recognition respect. First, then, the paternalist interference will lack the
justification the paternalist takes it to have if their appeal to individuals’
benefit is illegitimate. As such, it will be wrong in the way any unjustified
interference would be in the absence of a legitimate reason. This helps to
provide a response to the paternalist objection that anti-paternalists fetishise
individuals’ autonomy at the expense of their good.³² After all, paternalists
insist, our ‘status as an independent adult’ is not undermined by every
violation of liberty: being required to wear our seatbelt, for example.³³
Whilst it might seem that some interferences are too minor to worry about,
this thought usually comes with the qualification when it does enough good.
Even small interventions are objectionable when arbitrary and lacking justifi-
cation. Our status as an independent adult does seem to be undermined by a
state or individual engaging in a series of petty intrusions in our life—wear a
seatbelt, consume less salt and sugar, save for retirement, don’t smoke inside

³⁰ Grill 2015. Note that Grill does not endorse anti-paternalism, but merely argues that the filter is
the best way to understand it.
³¹ Parry 2017: 362. Groll (2018) also outlines a similar view.
³² E.g. Archard 1994; Arneson 2005: 264, 278; Sunstein and Thaler 2003: 1165.
³³ De Marneffe 2006: 84.
  ’       209

public places, drink less alcohol—if there is no legitimate reason for these
interferences.
Of course, our intuitions about such cases are complicated by the fact that
there are often many good reasons for such intrusions. Even if we accept that
individuals have authority over the use of their own good, and that some of the
targets of these interventions repudiate these benefits, we may still consider
these policies justified on the basis of their benefits to targets or third parties
who welcome them, in at least some cases (as §7.3 considers). Note that my
goal here is not to engage in a ‘project of reconciliation’, as Peter De Marneffe
describes the attempt to find anti-paternalist justifications for seemingly
paternalist policies.³⁴ Instead, my point is simply that certain actions (inter-
ferences) should not be performed for a certain reason (the good of the person
interfered with). That does not necessarily rule out the same action being
paired with some other reason and being permissible. This thought is not
uncommon: it would be wrong to shove a person aside in order to secure a
better seat in the cinema, but permissible to do so to prevent them stepping on
a child or knocking into a treasured vase. The disagreement amongst anti-
paternalists and paternalists concerns whether providing a repudiated benefit
constitutes the right sort of reason.
For the authority approach anti-paternalist who denies that it is, the
intervention is objectionable not because it demonstrates an insulting failure
to positively appraise the target’s capacities (lack of appraisal respect), but
because we fail to appropriately respect her status as an autonomous moral
agent (lack of recognition respect). We cannot demand that others believe we
are never mistaken, but we can expect to be treated as the kind of being who is
entitled to control our life and, perhaps, make our own mistakes. We are not
properly respected when subject to paternalist, and hence unjustified, inter-
ference: ‘[t]o interfere in a context where I ordinarily have the right to be the
one in charge is to call my status into question’.³⁵ To emphasise, this interfer-
ence need not be restricted to undermining particularly valuable liberties or
central capacities.³⁶ A more serious unjustified intervention might be more
objectionable, but any such interference will entail a lack of recognition respect
to an autonomous agent.
This lack of respect is not unique to paternalism: it is displayed in any
intervention in an agent’s legitimate sphere of agency that lacks proper
justification. What is ‘special’ about paternalism is that we are not interfered

³⁴ De Marneffe 2006. ³⁵ Fox 2019: 323.

³⁶ Contra e.g. Fox (2019), De Marneffe (2006, 2013); Shafer-Landau (2005).
210      

with callously or negligently, we are not ignored by someone pursuing their

own project, or used as a means to some wider goal. It is our interests that the
paternaliser is trying to promote. We may believe, as Carl Fox has argued, that
this benevolent motive mitigates the wrong of the interference: ‘[i]f disres-
pecting the agency of a human being delivers a deep insult to her, then . . . it is
partially mitigated if it comes from a place of fellow-feeling and concern’.³⁷
I disagree. Whilst the care shown for our target may demonstrate that we
mean well, the use of their good as a reason represents, I believe, an additional
lack of recognition respect in many contexts. I consider when we might have
authority to use someone’s good as a reason in §7.7, but it is worth flagging
now that the degree of explicit consent necessary will vary by context. For
example, in the context of close personal relationships, or when serious danger
is imminent, we may have presumptive authority in the absence of explicit
refusal. I suggest this presumption does not apply to state justifications of
public policy.
This is the second prong of my dual account of the wrong of paternalist
interference: ignoring an individual’s authority over their own good—
violating their PPE—also fails to respect their status as an autonomous
agent. The object of paternalism is wronged both by the use of their good as
a reason without their consent, and because this does not constitute the right
sort of reason, meaning the intervention lacks justification. It is this further
lack of recognition respect that sets paternalism normatively apart from other
instances of unjustified intervention. This means that, even in cases in which
some interference is all-things-considered justified (by some other reason),
there is still reason to object to the action performed for the paternalist reason
(or, the paternalist action-reason): the paternalisee has not been considered
appropriately authoritative over the use of their good as a reason.
However, arguably, using someone’s good without their authority need not
only occur in cases of paternalism. This reason may also be employed to justify
interference with others—for example, using a putative (and repudiated)
benefit to some as a reason to tax others. As noted, insofar as paternalism
involves interference with someone for their own good, interfering with third
parties for their sake will not count as paternalism. Intuitions differ on these
cases, as §7.6 will consider, though my own view is that such cases share one of
the distinctive wrongs of paternalism but not the other: they fail to respect
individuals’ authority over the use of their own good, but do not involve
interfering in an individual’s legitimate sphere of agency without (appropriate)

³⁷ Fox 2019: 329.

  ’       211

justification. Thus, paternalism is not normatively unique because it wrongs

individuals in a way that is never replicated in non-paternalist acts. The
combination of wrongs does seem unique, however. Both are grounded in
what individuals are entitled to as autonomous, self-directing agents, but there
are two different ways in which paternalist action-reasons fail to respect this.
The committed paternalist may, at this point, remain unconvinced.
Nonetheless, I hope to have demonstrated the possibility of a coherent and
consistent account of anti-paternalism—the authority view—and to have
provided some reasons why paternalist action-reasons might be objectionable.
On this view, the commitment to anti-paternalism is rooted in the broad
conviction that autonomous individuals should be shown appropriate recog-
nition respect as self-directing agents who have authority over their own good.
This tends to be grounded in the further belief that individuals should be
positively enabled to form and pursue their own conception of the good, as
well as being permitted to make their own mistakes. Paternalism is just one
barrier to this goal, but an important one, especially for individuals considered
unreliable, irresponsible, or incompetent, as disabled individuals often are.

7.3 State Assistance and the Anti-Paternalist Filter

What are the implications of a commitment to anti-paternalism in the context

of distributive justice? The two failures of respect I have identified might both
be exhibited in redistributive policies. To see how, let us begin with an
example. Imagine that a state is trying to determine how it should fulfil its
obligations to deaf individuals. We can assume that, in this context, individ-
uals with this impairment lack the capabilities they are entitled to. Hence, they
are disabled, and are owed assistance to rectify this injustice. Imagine, further,
that the government concludes that the problem here clearly results from the
fact that deaf individuals cannot hear, so resolves to fulfil their obligation by
funding research and development into cochlear implants that can be suc-
cessfully used at any age and for any form of hearing loss. This is an expensive
programme, but they decide that it can be justified on the basis of the benefits
it will bring to the entire deaf community. Now imagine that a significant
proportion of deaf individuals repudiate these benefits. If we were only to take
account of the benefits to those who welcomed such a ‘cure’, the gains would
be insufficient to justify the considerable costs. The state is aware of this
but decides to pursue the policy anyway, reasoning that ‘they’ll thank us in
the end’.
212      

There is much to unpack in this case. Perhaps the most natural response to
such an example is to object that the supposed beneficiaries will not be
ultimately grateful: they are best placed to know what is in their interests,
and if they believe implants are not good for them then they are not. Further,
as a purely practical matter, we might think that if people do not want to be
cured then they will not utilise the treatment when it is available. Thus, it will
necessarily not benefit them, and the time and money spent on research and
development will be wasted. Note that since the major expense is assumed to
involve the invention of the implants and not their individual production, low
take-up will not significantly lower the costs. As such, the costs will remain
high and the pool of beneficiaries will shrink as individuals opt out, meaning
the benefits no longer justify the expense. In other words, we might simply
object that this attempt at paternalism will be unsuccessful.
This is an important consideration, and is certainly one good reason not to
ignore the views of supposed beneficiaries. It is likely to be especially relevant
in such cases given the much-discussed tendency of able-bodied individuals to
overestimate the ‘badness’ of impairment, and so assume the best solution is
always to eliminate impairment if we can. However, as discussed, whilst we
must guard against failed paternalism we should also allow that successful
paternalism is possible. Thus, we should grant that in at least some cases
policy-makers will get it right or ‘know better’, and those they aim to help
would actually benefit.³⁸ Perhaps if the implants are widely available, individ-
uals will be persuaded to follow the crowd, accept the implants, and find that
they do actually increase their well-being. Perhaps the mere availability of an
alternative is beneficial in some cases.
If this point is granted, then we might next wonder whether this example
proves exactly the opposite of what I intend: instead of showing how objec-
tionable redistributive paternalism might be, it merely demonstrates the
implausibility of employing the anti-paternalist filter in this context.
Employing the filter means that if an expenditure could only be justified on
the basis of the benefit to some disadvantaged group but a sufficient number
do not authorise the use of their good as a reason, then the pursuit of this goal
cannot be justified after all. But why, we might ask, should some of the
prospective beneficiaries get to ‘veto’ the provision of this benefit, and prevent
its being provided to those who would welcome it? Indeed, we might think

³⁸ Recall that on my view this means they are benefited according to standards they can accept (see
§6.3). This does not rule out successful paternalism: even if we assess individual’s best interests against
their own values, they may lack the knowledge or ability to promote these interests for themselves.
  ’       213

that the entire group ought to allow their good to be counted in the cost-
benefit calculation to ensure that those who welcome the benefit will be
provided with it.
Parry has considered a similar worry in a rather different context. He argues
that individuals’ authority over the use of their good—their PPE—implies that
defending others is morally justified only ‘if those to be defended do not validly
refuse defensive intervention’.³⁹ As such, otherwise permissible cases of other-
defence will be rendered impermissible if a sufficient number of victims refuse
to be saved, given that we should ‘treat victims’ refusal as morally equivalent to
their nonexistence’.⁴⁰ For example, if we can only redirect the standard
runaway trolley away from the five people and towards the one at this ratio,
and one of the five rejects this benefit, then it will no longer be permissible
to save the other four. The analogous objection here is that given that our
consent (seriously) affects others we may be morally required to consent, and
further, if it would be wrong to refuse to consent then any refusal would be
morally invalid.
However, refusing to consent does not just needlessly withhold a benefit
from those who desire it—or merely cause the deaths of four individuals on a
trolley-track. The reason that the good of those who repudiate the benefits is
needed to justify the action is that it is also associated with significant costs:
directing the trolley away from the five will kill one person on the side-track.
In our example, sinking significant resources into developing cochlear
implants will mean less can be spent on the provision of other benefits, on
any plausible assumptions regarding the scarcity of resources (see §7.6). If
these costs cannot be justified with reference only to the good of those who
authorise its use, it is not clear why those who do not welcome the benefit
should be compelled to create a situation in which the cost-benefit analysis
falls out in favour of the willing beneficiaries.⁴¹ No more than there is any
reason for people to become or remain deaf to increase the beneficiary pool. If
we grant individuals’ authority over their own good and an action cannot be
justified without it, they are under no obligation to allow their benefit to
count.⁴² I have the right to decide what is done for my sake.

³⁹ Parry 2017: 357. ⁴⁰ Parry 2017: 387.

⁴¹ See Parry (2017: 388–94). As he puts it: ‘if the preexisting reasons in favor of killing the innocent
are insufficiently weighty to justify doing so, then how can this generate any reason to add additional
considerations in favor of harming to the scales in order to tip the balance?’ (Parry 2017: 393).
⁴² Plausibly, however, they are permitted to have their benefit count. If an individual’s authority over
their good allows them to control how it is used, we may be allowed to consent altruistically. If this is
214      

Next, we may wonder why individuals would want to repudiate an offered

benefit. On the one hand, they may simply be mistaken about what would be
best for them. As the previous section noted, a core commitment of anti-
paternalism involves respecting autonomous individuals’ capacity to be self-
directing, which includes accepting that sometimes they will head in the wrong
direction. Thus, the possibility of mistakes will not deter an anti-paternalist.⁴³
We might, however, raise the deeper worry that the mistaken rejection of the
benefit is involuntary or non-autonomous: perhaps because an individual is
ill-informed, perhaps because they have been pressured by some covert influ-
ence (§6.3), perhaps because they have had to adapt to a constrained set of
options (§6.6), or perhaps because they are experiencing an emotional or
psychological crisis. As Joel Feinberg points out, such choices may be as
‘alien to [us] as the choices of someone else’.⁴⁴
Certainly, the anti-paternalist commitment to respect individuals as
autonomous agents only applies when they are acting autonomously. Thus,
any plausible version of anti-paternalism will be soft paternalist, meaning that
whilst we should not interfere with voluntary choices (regardless of their
consequences), involuntary choices may sometimes be subject to intervention.
The most famous example in support of this view is Mill’s individual about to
cross a rotten bridge unawares: surely no anti-paternalist would prohibit
intervening to ensure they were aware of the bridge’s state, so their decision
to endanger their life was really voluntary.⁴⁵ For now I will assume that
individuals’ repudiation of the benefits is genuinely voluntary or autonomous,
and briefly return to cases where it is not in §7.8.
On the other hand, though, this repudiation need not be indicative of a
mistake—autonomous or otherwise. Individuals may simply be committed to

true, this would provide an explanation for Grill’s suggestion that in cases of group consent ‘altruistic
consenters that consent or not in conflict with their own interests will count for less than those who
consent or not consistently with their own interest’ (Grill 2009: 156). In the case under discussion, this
means deaf individuals voting for the cochlear implants for the sake of those deaf individuals who
welcome them would ‘count for more’ than hearing individuals with similarly altruistic motives. The
reason this would be so is that deaf individuals can choose to have their good included in the calculus,
whilst hearing individuals have no benefit they can add to the balance of reasons in the same way. It
might be responded that, if deaf individuals repudiate the benefit, they also lack a benefit to add to the
calculus. However, if we are assuming the paternalist aim will be successful, this is not so.
⁴³ Note that this may mean that those who choose poorly are not only worse-off than they could
have been but worse-off than those capable of choosing better. Thus, anti-paternalism may seem to
contribute to anti-egalitarian outcomes, and Arneson (2005) objects to anti-paternalism on this basis
(see Voigt 2015). Given individuals’ have an ongoing entitlement to control, on my view, this limits how
bad the consequences for bad-choosers will be. However, it is true that respecting autonomy, and
providing access to capabilities rather than trying to achieve functionings, will entail greater inequality
of outcome than some more interventionist approaches.
⁴⁴ Feinberg 1986: 12. ⁴⁵ Mill 1974 (1859): 166.
  ’       215

values other than doing what maximises their welfare. Most obviously, in the
case of impairments, individuals may value it as constitutive of their identity,
or value the associated membership of a particular community or group. It is
true of many features of our identity that our life might be better without
them—things are perhaps easier for straight, white, able-bodied men, but this
does not mean we would choose to forgo features of ourselves that do not
conform to this model. This analogy perhaps also helps to make clear what is
so objectionable about the opening example for those who do not feel its
immediate intuitive pull. Imagine, instead, that conversion therapy was the
policy under consideration for non-heterosexual individuals (and try to
imagine, too, that it would actually make its targets all-things-considered
better off ). Many people would find it repellent that the state would fund
such a policy (though could perhaps reconcile themselves to it insofar as some
genuinely and autonomously welcomed it). Yet the idea that this policy is
justified—indeed, could only be justified—by reference to its benefits to recipi-
ents who repudiate it, surely represents a grievous insult and lack of respect.
My right to determine my conception of the good demands not just the
provision of opportunities and an absence of unjustified interference, but
also that this good is not used, without my consent, to justify benefits
I reject. Redistributive policies justified by unauthorised appeals to individuals’
good thus encapsulate both core elements of the wrong of paternalism: failure
to respect individuals’ authority over their good, and interference without
justification.⁴⁶

7.4 Anti-Paternalism in a Theory of Justice

The main danger point for paternalism in theories of distributive justice is not
ordinarily taken to occur during the practical business of determining how
individuals’ entitlements are to be fulfilled, as in the earlier example. But,
instead, at the prior level of determining what these entitlements are. If this
involves a perfectionist commitment to items that are essential to any good
human life, then even if freedom is amongst these, it will always be something
to be weighed in the balance and potentially outweighed. This is part of
the reason I have repeatedly emphasised the importance of distinguishing
well-being and justice. As §3.3 and §3.4 discussed, we might believe that both

⁴⁶ As §7.5 will demonstrate, even the provision of options, as in the examples discussed here, can
constitute intervention of the relevant sort.
216      

well-being and justice should be conceptualised in terms of capabilities

without grounding our entitlement to capabilities in the fact that they are
components of well-being (contrary to many approaches⁴⁷). My claim is not
that the selection of domains of control will be unrelated to what makes a life
go well. Indeed, part of what it is to say that controlling our health, our
mobility, our social relationships, or our sexual life is central to human life is
to say that decisions in these areas will have significant impacts on well-being.
Rather, my point is that a theory of justice should not be aiming to ensure
some defined level of well-being, nor need every principle of justice directly
promote well-being.
If well-being were our goal, then given all we know about how poorly people
choose it would surely be unwise to remain committed to capabilities rather
than functionings, as so many critics have noted.⁴⁸ After all, if we have agreed
that a good life involves performing particular functionings, and justice
requires that we ensure people are living good lives, surely there will be
times when functioning ought to be encouraged, or even compelled.
Welfarist positions, including such iterations of the capability approach, will,
therefore, always be vulnerable to sliding into paternalism: if the state is in the
business of welfare promotion, there are likely to be times when it will be
appropriate to directly promote welfare. If our goal is to equalise or maximise
welfare, we simply will not be able to leave the bad choosers alone. Even if we
are committed only to equal opportunity or capabilities to function well, it
may be difficult to remain committed to individuals’ absolute authority over
their own good when we believe we know that is really in their interests, and
especially when they might sign away their future capacity for choice. Thus, as
we have seen, many theorists do permit the promotion of functionings in some
circumstances—perhaps as a mere short-term means to encourage future
functioning or perhaps more long-term—and those who do not are criticised
for leaving people worse off than they might be.⁴⁹
My own, capability-to-control account (defended in Chapters 3 and 4)
avoids a commitment to the value of specific functionings, and so avoids the

⁴⁷ For example, the Stanford Encyclopaedia entry on the capability approach opens by asserting:
‘The capability approach is a theoretical framework that entails two normative claims: first, the claim
that the freedom to achieve well-being is of primary moral importance and, second, that well-being
should be understood in terms of people’s capabilities and functionings’ (Robeyns and Byskov 2020).
A capability approach to justice need not accept either.
⁴⁸ Except perhaps contingently: because actually endorsing functionings is important, or because the
badness of an intervention will always outweigh any benefits it brings (see Olsaretti 2005; Arneson
2006; §4.6).
⁴⁹ See Claassen 2014a for an overview, also Arneson 1999, 2000a; Carter 2014; Deneulin 2002;
Nussbaum 2000a, 2011a; and §4.3.
  ’       217

conflict that plagues capability-to-function approaches (if we know what’s

good for people, why not give it to them?). It is grounded in the claim that
individuals have the authority to direct their own lives and that a theory of
justice should enable them to do so. As such, it is not mysterious why
capabilities are our goal. Indeed, by identifying domains of control rather
than endorsing the value of particular functionings it avoids ‘the perfectionist
view that some functionings are objectively good’, which some have claimed
the capability approach ‘cannot avoid’.⁵⁰ The remainder of this section will
outline the benefits of this approach at the level of selecting a capability list:
since it requires no commitment to a particular view of what is good for
people, there is no danger that individuals’ authority over their good will be
overridden to paternalistically enforce this. Nonetheless, states cannot provide
every possible opportunity, so as §7.5 and §7.6 demonstrate, the possibility of
paternalism may still threaten my approach at the level of policy specification.
In §7.7 I argue that this, too, can be avoided.
To briefly reiterate what earlier chapters have argued, there are two core
problems raised by identifying a list of capabilities by appealing to a set of
valuable functionings all individuals should have the chance to perform. First,
arriving at a single list of valuable functionings in the face of persistent
disagreement amongst conceptions of the good will exclude individuals who
do not value or cannot perform these functionings, as well as making an
expressive judgement that their way of life is inferior. Further, insofar as the
inclusion of these items depends on their (near) universal value, then this
might seem to entail using these individuals’ good as a justification without
their consent. For example, many deaf and blind individuals repudiate the
benefit of using all five senses, many autistic persons repudiate the benefits of
neurotypical social interaction, and many asexual individuals repudiate the
benefits of sexual satisfaction. Thus, the process of selecting a capabilities list
may be in danger of violating individuals’ authority over their own good.
Second, focusing on the opportunity to achieve specific functionings rather
than providing more content-neutral freedom will mean individuals are not
truly enabled to form and pursue their own conception of the good when this
does not involve performing functionings identified as valuable. Most obvi-
ously this will mean that individuals will not be positively enabled to achieve
their goals when these are not amongst the functionings deemed valuable.
More worryingly, this may lead to individuals being encouraged or forced
into doing what is ‘best’ (see §4.6). As noted above, while a commitment

⁵⁰ Khader 2018: 210. Also see Carter 2014; and §4.5 for discussion.
218      

to substantive perfectionism does not necessarily entail its paternalistic

imposition, it often does so. Further, as the next section will argue, merely
endorsing and enabling specific options without compelling their uptake may
itself constitute (potentially paternalist) intervention.
One response to the first worry would be to argue that we need not claim
that performing each and every one of the central functionings would be
necessary to all good human lives. Thus, we need not utilise the good of
dissenters as a means of including items on the list.⁵¹ However, this would
not be a plausible response for versions of the approach that assume the
central functionings are those that people would, in general, benefit from
performing (even if they should not be forced to do so).⁵² This indicates that
the entitlement to specific capabilities is grounded in the (near) universal value
of the functioning, such that if it became clear that the functioning was
valueless to a sizeable minority then its status on the list would be reasonably
questioned. This implies an appeal is being made to the good of individuals
who repudiate the supposedly beneficial functioning: the capability for sexual
satisfaction should stay on the list, for example, because this functioning is an
important part of human life, even for asexual individuals (regardless of their
views). Yet even if the appeal to dissenters’ good can be avoided, it remains the
case that justifications for capabilities to function tend to be based on the
potentially insulting assumption that some ways of being are simply superior,
so those who are necessarily incapable of engaging in them are inferior and
those who choose to reject them are mistaken (§4.3).
By focusing on capabilities to control, not function, my own approach is less
vulnerable to these worries. Avoiding a commitment to the value of specific
functionings means the justification of the approach need not involve an
appeal to the general value of a restrictive list of functionings, and so is
more inclusive of those who choose to, or must, function in non-standard
ways: even those who repudiate central functionings can be part of an overlap-
ping consensus on the more general value of control. Further, providing control
in central domains—an acceptable range of options rather than the ability to
choose whether to perform a specific functioning—grants more content-neutral
freedom. Individuals are enabled to be mobile, form relationships, control their

⁵¹ Note that merely using an individual’s good despite their refusal is not paternalist until it is
employed as a reason for action. However, capability lists are not intended as mere philosophical
exercises, but as guides to state action.
⁵² E.g. Arneson (2010, 2006) and Olsaretti (2005) explicitly, but arguably Nussbaum too (e.g.
Nussbaum 2000a: 55, 74, 87–93; 2004: 199; 2011a: 25; and see §4.3).
  ’       219

sexual life, use their senses, and engage in leisure, without any claim about what
exercising control in these areas should involve.
Yet it might nonetheless be objected that, since I am only committed to
providing control in specific central domains and not simply ‘capability as
such’,⁵³ then I will also have to appeal to individuals’ good. After all, why are
mobility, sociality, sexuality, sensory experience, and leisure amongst the
concerns of justice except insofar as these areas are claimed to be important
for good human lives? I need not deny that my account allows for some small
degree of perfectionism. What matters is that it is not exclusionary in the way
that thicker versions of perfectionism can be and, more importantly, that it
will not justify an unacceptable degree of paternalism.
First, then, we need only be committed to a very thin conception of a good
human life: one according to which it is important to be able to exert autonomous
control, by choosing amongst acceptable options, in certain sorts of domains.
Few functionings are necessarily ruled out or in, since most functionings could be
subsumed under some broad domain.⁵⁴ Further, I claim only that individuals
should be able to live autonomously, not that they must do so. Believing individ-
uals are entitled to direct their life need not entail a commitment to the view that
they must choose an autonomous life. Thus, individuals may exercise their
control by choosing to lead a non-autonomous life—by ceding their control.
The scope of justice is restricted, then, not by specifying the shape of a good
life, but because individuals need only have an acceptable range of options in a
domain, and not every functioning or their most preferred functioning. Hence,
we can have all we are entitled to if we have control over our mobility,
relationships, and so on, even if there are certain modes of functioning closed
to us. Given that so few substantive commitments are required, it is difficult to
imagine the coherent rejection of the value of control, even if it will simply be
exercised to give up that control. Thus, it is plausible that formulating our
list of capabilities-to-control will not require an appeal to individuals’ good
without their assent.
More significantly, merely identifying central domains in which individuals
should be able to exercise control is not, itself, paternalist. Whilst a list of
capabilities to function offers reasons for particular actions (providing cap-
abilities to function), a list of domains of control remains neutral about the
offers that will be made and the policies pursued. Yet this might seem like a

⁵³ Carter 2014: 91; §4.3.

⁵⁴ I am more interested in how capabilities should be conceptualised than in devising a capability
list, but I am happy to broadly endorse Wolff and de-Shalit’s modification of Nussbaum’s list, as §4.7
considered (Wolff and de-Shalit 2007: 45–62; Nussbaum 2011b: 33–4).
220      

distinction without a difference, and it may be objected that my view simply

shifts the problem rather than solving it: for as long as resources are scarce we
are going to have to make decisions about which options individuals can
choose amongst when they are exercising their control. Thus, the state may
end up in the position described earlier: deciding how to provide individuals
with what they are entitled to where this will involve offering some opportun-
ities within a domain and not others. At the level of specification, then,
paternalism may threaten in two ways. First, the state selection of options
within a domain might involve a justificatory process that makes an unauthor-
ised appeal to the good of putative beneficiaries. Second, this justification may
then be paired with an action that involves interference with said beneficiaries.
In other words, we may fail to respect the anti-paternalist filter on the use of
reasons, and this may generate a paternalist act.

7.5 Intervention beyond Interference

However, what is the intervention here? We are merely identifying opportun-

ities individuals will be offered, not constraining their freedom or forcing them
to live ‘well’. Can the offer of a capability—especially the mere capability to
exercise control in some domain of our life—really constitute an intervention?
If not, the possibility of paternalism seems to be entirely off the table, whatever
reasons are used to select offered opportunities. Failure to respect individuals’
authority over their own good may always be objectionable, but, on the
account I have defended, it is not paternalist unless we use this good as a
reason to interfere. Paternalism involves both a violation of individuals’ PPE
and an intervention that thus potentially lacks justification.
Whilst the most serious interventions will involve direct coercion, the impos-
ition of force, or the complete removal of options, this is not, as Kristin Voigt
points out, ‘the only—and perhaps not even the most common—way to interfere
paternalistically with someone’s decisions’.⁵⁵ Drawing on this broader account of
intervention, I will argue that there are three ways in which the state offering
opportunities may constitute interference. First, providing some opportunities
generally involves not providing others, so some individuals’ preferred options
are not enabled and may be more expensive. Second, offering opportunities may
constitute an endorsement of their value, and even persuade individuals to
pursue them. Third, and most directly, many people will be taxed to fund

⁵⁵ Voigt 2015: 94.

  ’       221

them. (Additionally, option provision will involve interfering with others, most
obviously through taxation. Whilst I have suggested such third-party cases are
not paternalist, they do share a failure to respect individuals’ authority over their
good.) The expansive account of intervention motivates the concern that despite
my effort to focus on content-neutral freedom as control, the need to specify
capabilities leaves open the possibility of objectionable paternalism. §7.7 argues
that this can be avoided. I will begin by outlining some common examples of
non-coercive paternalism before applying them to the case of capability provi-
sion in §7.6.

7.5.1 Raising and Reducing Costs

Paternalism may involve making options more expensive: for example, impos-
ing ‘sin taxes’ on high-sugar or high-fat foods and drinks, or on cigarettes.
Indeed, arguably all government policies are of this form, where at the far end
of the scale the ‘price’ of an option is imprisonment.⁵⁶ The state can rarely
directly force individuals into making some choice or completely remove some
option. They can make actions and substances illegal, and impose various
penalties for breaking this law, but this cannot rule out the possibility of
illegally pursing this option. This is true of most interpersonal paternalism
too: we can rarely truly force someone, and even more rarely successfully
promote their good all-things-considered if we did. Thus, for example, we
might try to stop a friend breaking their vow of quitting smoking by buying
up their favourite brand of cigarettes in nearby shops for a few days. This
might make them better off. Spending a few days forcibly pulling cigarettes out
of their mouth is a much less plausible candidate for successful paternalism
(or successful friendship).
The central lesson here is that intervention should be understood to involve
attempts to change an individual’s behaviour or improve their character by
altering their option set. We might achieve this by making options not only
more expensive but also ‘cheaper’ or more desirable. For example, subsidising
healthy foods or gym memberships, or offering tax-free bikes. Relatedly, we
might change an individual’s menu of options by adding items. For example,
Seana Shiffrin argues that, if there are options an individual would prefer not
to have (perhaps because they find too much choice overwhelming, or worry
about yielding to temptation), then to provide these options anyway (perhaps

⁵⁶ See Voigt 2012, 2015.

222      

because we think they would have a better character if tested, or that the
option they are trying to avoid is a good one) can be paternalist.⁵⁷ For example,
offering drinks to a friend who is trying to stay sober because we think they
would enjoy themselves more if they accepted a pint. This, too, involves an
effort to ensure an individual acts in the way that we deem to be best for them
by altering their option set, so should also be counted as intervention in this
context.

7.5.2 Pre-emption: Information and Persuasion

Decisions about whether and how to provide information may also constitute
interference. If we accept the soft paternalist aim of allowing individuals to set
and pursue their own goals when they do so voluntarily, it may seem that
providing information can hardly interfere with this. We might interfere by
acting against their choice, ensuring they act in line with their choice, or acting
on their behalf before they have had a chance to make a choice, but can
providing information they will use to make a better-informed choice really
constitute intervention? It seems that it can. Whilst there may be nothing
paternalist about ensuring someone does not accidently expose themselves to
harm or the risk of harm, many people will choose to engage in risky activities,
and this may involve not wanting to know the details of the risks they face. As
Arneson says: ‘a person’s settled values, attitudes, and desires may include a
disposition to impetuous decision-making’.⁵⁸
Contrary to the standard assumption that education simply cannot be
paternalist,⁵⁹ foisting information on people or even pushing them to carefully
consider their decision can constitute an intervention, which might be pater-
nalist. For example, requiring individuals to attend road safety courses, having
mandatory cooling-off periods for contracts and purchases, or placing health
warnings on cigarette packets might all be paternalist if engaged in for the sake
of the paternalisee’s good. ‘The choice not to gather information is, after all, a
choice’⁶⁰—and interference with that choice is interference. Thus, providing
unwanted information in a way that is hard to avoid constitutes intervention.

⁵⁷ Shiffrin 2000: 213–14.

⁵⁸ Arneson 2005: 267. He continues: ‘and an aversion to the careful consideration that renders choices
voluntary’. My focus is on autonomous choice, which requires that decisions are made independently and
that the individual can endorse the end were she to reflect on it. This does not rule out spontaneity,
and does not imply ill-considered choices are non-autonomous (see §6.3 and §7.8).
⁵⁹ E.g. Abel et al. 2021: 51–2. ⁶⁰ Hanna 2018a: 158.
  ’       223

Further, a commitment to soft paternalism does not rule out allowing individuals
to autonomously make ill-considered, ill-informed, spontaneous choices.
Yet more controversially, some rational persuasion has been argued to
constitute a potentially paternalist intervention. George Tsai has claimed
that the provision of reasons may pre-empt someone’s deliberative activities,
and undermine their ability to be self-directing.⁶¹ For example, if we are
playing a board game and as soon as it is your turn I lay out all the reasons
why some particular move is the best one, I prevent you from making a
decision yourself. I do not force you to act, or even change the profile of
your option set, but by reasoning for you before you have had a chance to
consider the matter yourself, I thwart the exercise of your agency. In a similar
vein, withholding information might also be paternalist. For example, a doctor
choosing not to disclose information about an unsubsidised medication to a
patient because they believe being aware of such an unaffordable possibility
would be detrimental to their well-being.⁶² Again, this action prevents some-
one deciding for themselves.
Thus, we should add that relevant interventions involve attempts to change
behaviour or improve decision-making by not only altering an option set but
also undermining our ability to choose amongst those options for ourselves. It
should be emphasised that I am not claiming such interventions can never be
justified, only that they should not be performed for a paternalist reason.
There may be many legitimate reasons why information should be withheld,
for example: not least, that it is infeasible to provide all information, and that
judgements about relevance and appropriateness must be made. To reiterate, a
distinctive feature of my dual account of the wrong of paternalism is that it
does not require that the intervention itself be presumptively impermissible.
Interventions that are presumptively permissible may still be wrong when
performed for a paternalist reason, since they do not demonstrate appropriate
recognition respect, both by falling to respect individual’s authority over their
own good, and by intervening—even if in a very minor way—without appro-
priate justification.

7.5.3 Refusals to Assist

The same could be said about our final example: refusals to assist. These
too may be performed for a variety of reasons, and though I claim they

⁶¹ Tsai 2014: 92–5. ⁶² Voigt 2015: 94–5.

224      

can constitute an intervention, this does not imply that they constitute a
presumptively impermissible intervention. Consider, for example, refusing to
help someone build some shelves because we think it would be good for them
to learn carpentry, and do not trust them to learn without this push.⁶³ Or
refusing to lend someone £50 out of fear they will use it to fund their heroin
habit.⁶⁴ These cases will be classed as paternalist only if the reason for refusal
concerns the good of the individual interfered with, and not, say, a self-
regarding reason: the inconvenience of providing help or being unable to
comfortably do without £50.
Yet it may be objected that merely refusing to assist does not constitute
intervention since these acts are within the paternaliser’s sphere of agency—it
is their time or their money. This is Quong’s view. Thus, in order to include
these cases as paternalist he drops the requirement that paternalism must
involve interference. Instead, according to his ‘judgemental definition’, pater-
nalism simply involves an attempt to ‘improve the welfare, good, happiness,
needs, interests or values of agent B with regard to a particular decision or
situation that B faces . . . motivated by a negative judgement about B’s ability’.⁶⁵
This opens the possibility that attempting to improve someone’s welfare by
interfering with someone else constitutes paternalism: for example, your
paying off my competitors in a race because you do not trust me to win
otherwise. As noted, my own view is that such cases share a failure to respect
individuals’ authority over their own good, but are not themselves paternalist.
However, there is surely a scale here, rather than a clear binary distinction
between interference and non-interference. Insofar as refusals to assist involve
changing the profile of an individual’s option set and, if successful, altering
their behaviour (causing an individual to practise their carpentry or refrain
from using heroin), this does seem like a case of interference.⁶⁶ Such interfer-
ence might not stand in need of much justification, but this does not mean it is
unobjectionable in the absence of a legitimate one, which I contend paternalist
intervention lacks. Insofar as interfering with the race does not change my

⁶³ Shiffrin 2000: 213. ⁶⁴ Quong 2011: 79–80.

⁶⁵ Quong 2011: 80. I have also rejected the view that paternalism is distinguished by an insulting
negative judgement in favour of the authority account (paternalism distinctively involves violating
individuals’ authority over their good).
⁶⁶ Remember that if our reason for refusal is different—we do not want them to buy heroin with our
money, but do not believe our refusal will decrease the likelihood of the purchase being made, say—
then the refusal would not be paternalist. This is closely analogous to Shiffrin’s discussion of the
unconscionability doctrine: whilst the anti-paternalist filter should rule out an appeal to the good of the
exploited as a reason for this policy, the state may uphold the doctrine because it does not want to be
involved in enforcing contracts whose terms are one-sided, exploitative, or unfair (Shiffrin 2000: 221–4,
and see Grill 2015: 58–9 for discussion).
  ’       225

options or ability to choose between them, but only changes the significance of
my actions (my race-time is now the winning one), this does not seem to cross
the threshold of interference.
Yet it may be argued that if sufficiently thoroughgoing it does change my
options: now I cannot but win. Indeed, it may be objected that, in many third-
party cases, if benefits are successfully bestowed, then the benefited individual
will have their options altered, or their capacity to choose amongst them
restricted. Even when this it true it seems worth separating the two elements
of these cases—the paternalist interference (with an individual for their own
sake) and the third-party interference (for another’s sake)—even if both are
instantiated in a single action. This seems to me to allow greater clarity,
meaning, for example, that we can judge one action-reason pair permissible
and the other not. Yet intuitions on these cases vary widely, and ultimately
little hangs on whether third-party cases are labelled paternalist. What matters
is that the wrong is rooted in the violation of an individual’s PPE and the lack
of justification for the interference.

7.6 The Paternalism of Unwanted Offers

The category of interventionist actions that may be performed for a paternalist

reason is, therefore, broad, including at least: increasing options’ costs, adding
options, or making them cheaper; pre-empting judgment via information
provision, or via rational persuasion; and refusing to assist. This can be
broadly summarised as actions that aim to change an individual’s behaviour
or improve their character by altering their option set, or undermine their
ability to choose amongst those options for themselves. I will now outline
three ways in which the provision of capabilities can be included as interven-
tionist actions.
First, when resources are scarce the decision to fund some means of
assistance will inevitably come with opportunity costs. Thus, the decision to
offer cochlear implants is not just a decision to offer cochlear implants, but
also a decision not to pursue various other avenues by which the disadvantage
faced by deaf individuals might be mitigated—improving and providing
speech-to-text software, for example. Recall that, in the example I describe,
much of the cost arises from the process of developing the technology. This is
not merely a convenient thought experiment: many assistive technologies
involve such one-off costs. This means costs will not depend on level of uptake
and cannot simply be reduced depending on the number of individuals who
226      

wish to utilise them. Thus, we may not be able to fund the development of
multiple alternatives. Further, though the opportunity costs need not neces-
sarily be borne by deaf individuals, if the costs of developing implants can only
be justified on the basis of the benefits to all or most of the deaf population,
and insofar as the government understands this as the way it will provide the
deaf community with the opportunities they are entitled to, then it is plausible
to think that this will reduce their inclination to fund alternatives.
The choice to spend resources developing cochlear implants, then, will
make the preferred, alternative options of many deaf individuals more expen-
sive than they might otherwise have been. Of course, whether these options are
‘more expensive’ will depend on the relevant counterfactual. The software, say,
will be more expensive than it would be if it were government subsidised, but if
individuals only have an entitlement to control in a domain and not to some
specific way of exercising that control, then it is not clear that this is the
relevant comparison. To this, we could respond that, if some individuals do
not consider cochlear implants an acceptable option, then we have not actually
succeeded in enabling them to have control. There may be controversy, in
some cases, in determining what constitutes an acceptable option. Certainly,
we cannot allow individuals to demand that they will not settle for less than
some outrageously expensive option (‘I can only be mobile in a Lamborghini’).
Indeed, part of the motivation behind the development of the capability
approach was to avoid welfarist commitments to simply taking ‘account of
wants only as they are given’,⁶⁷ and to provide certain opportunities regardless
of the welfare they generate and regardless of the resources they require (§3.3).
However, an opportunity to have control only at the expense of removing
what many consider a central feature of their identity can hardly be considered
within the realm of the acceptable. A judgement call will be needed in many
cases, and there may be instances in which individuals are deemed to have the
control they are entitled to, though they are only provided with offers they
would reject: perhaps a ‘cure’ is all that can be offered in cases where we lack
the resources or technology to mitigate the effects of an impairment by any
other means. However, this is clearly not true of the example of deafness under
consideration. Thus, insofar as some individuals lack what they are entitled to
when only cochlear implants are offered, it does not seem unreasonable to
compare this with the equally achievable scenario in which they are provided
with control when assessing the comparative expense of their options.

⁶⁷ Elster 1982: 237.

  ’       227

By making other options more expensive, this falls into the first of the above
categories of intervention. Of course, it does not follow this must be paternal-
ist, never mind all-things-considered impermissible. Pursuing an option that
leaves some impaired individuals without the control they are entitled to (i.e.
disabled), or able to achieve this control only at greater expense, may be the
best or only option (hence permissible), and need not be justified by an appeal
to the benefits to those who repudiate it (hence not paternalist). As §1.3
considered at length, we may simply be unable to prevent all impairments
from being disabling, even when the cause of disablement is external, social,
and contingent. When the needs of individuals with different impairments
conflict, we might need to make difficult decisions based on the number
affected, the level of disadvantage, and the benefits of the proposed policy.
The outcome may leave some disabled, when a different decision would have
avoided this, and that may be permissible. What is not permissible is deter-
mining this outcome on the basis of benefits to individuals who would
repudiate it. That is, funding an option individuals reject because of the
benefits it would bring to them. Further, even if we conclude this does not
make the non-state-funded options more expensive, it may still constitute a
failure to assist. Understood in this way, too, not providing an option may
affect individuals’ options sets and the ability to choose between them.
A second way in which offers may constitute intervention is by encouraging
people to make particular choices. Most obviously, and mirroring the
above discussion, this will make some options cheaper (as others were made
more expensive), or available when they otherwise would not have been.
Additionally, by offering particular forms of functioning and not others the
state endorses, explicitly or implicitly, the value of these options. For example,
if cochlear implants are the state-funded response to deafness, then this is the
option medical professionals will tend to discuss with deaf individuals (or their
parents or carers). They are likely to emphasise the benefits of this option and
may persuade individuals of the advantages of this course of action. They may
withhold information about other, unsubsidised alternatives. By having a
trusted professional simply present cochlear implants as ‘what we do in such
cases’, such policies may pre-empt individuals’ ability to decide for themselves.
It is possible that in some cases the impact will be even greater, such that the
default option is not merely promoted but becomes the only feasible alterna-
tive. For example, if individuals with hearing-loss tend to opt for the new
implants, it may cease to be cost effective to provide and research other forms
of accommodation. Further, Sign might be less widely used and so may
become a less tenable form of communication. It is such scenarios many
228      

Deaf individuals fear, and which lead them to object to the current promotion
of cochlear implants as having the potential for ‘cultural genocide’.⁶⁸
The third, and perhaps most direct, way in which the state provision of
an option will involve interference arises from the collection of taxation to
fund it.⁶⁹ Most obviously, if individuals pay tax, then they will be directly
funding the benefit they repudiate. Even if they are not a taxpayer, if we
understand contribution to, and membership of, a scheme of social cooper-
ation more broadly, then we might think that all participating members of
society are contributing to the provision of state-funded benefits even if they
do not make a direct monetary contribution. Note that my point here is not
that people should not have to fund benefits they disagree with: for example,
that those with particular religious convictions should not have to fund
contraception, or that libertarians should not have to fund any redistributive
policies (see §4.5). I have argued that all individuals are entitled to exercise
control in central domains of their life, and we can neither opt out of this
scheme of cooperation nor decide for others how they should exercise this
control. What is objectionable about the case of paternalist taxation is not that
I am being asked to help someone else in a manner I disagree with, but
that I am being compelled to help myself in a way I reject. In other words,
and to reiterate, my authority over my own good is not being respected in such
cases, and, if the resulting intervention can only be justified on this basis, I am
being interfered with without justification.
In summary, then, despite being mere offers of opportunities, the state
provision of capabilities does involve various forms of (interrelated) interven-
tionist actions, which may be paired with a paternalist reason.

7.7 Specifying Capabilities without the Good

Thus far I have argued that paternalism involves intervening with an individ-
ual to provide them with a benefit they repudiate. This fails to respect their

⁶⁸ See Sparrow (2005) for an overview.

⁶⁹ A further objection to paternalist policies of this sort might be that they undermine relational
equality. Whilst I have rejected the view that paternalism necessarily involves an objectionable lack of
appraisal respect, it may nonetheless be the case that systematically ignoring some individuals’ views
about what would best promote their interests might demonstrate a negative judgement about their
capacities that could ultimately undermine their ability to relate to others as equals. As Voigt notes:
‘whenever individuals find that particular choices are blocked for paternalism-for-equality type
reasons, this will indicate to them that the options they would have chosen were considered to reflect
poor judgement’ (Voigt 2015: 98).
  ’       229

authority over their own good, and any interference performed for this reason
may lack justification. Intervening with an autonomous agent without justifi-
cation demonstrates a further lack of respect. Merely providing opportunities
to exercise control in central domains initially seems not to endorse a thick
perfectionist account of the functionings that are good for people, nor prone to
interfering to impose its values. However, given the practical requirement to
select opportunities to make available now, and to choose where to engage in
research and development to determine what will be made available, then it
seems some choices must be made about which functionings to offer. And
merely providing capabilities to function can constitute interference, especially
when these comes at the expense of other alternatives.
How, then, can paternalist intervention be avoided? The most obvious
answer is to refrain from unauthorised appeals to individuals’ good. I have,
so far, been fairly non-committal about when individuals’ good can be used as
a reason. This is because there is no simple answer. One approach would be to
assume, as a baseline, that we may use it unless an individual withdraws their
consent. This is plausible in cases in which the benefit is obvious, few would
not so consent, and consultation is impractical: for example, shoving an
individual out of the path of a car. It is also plausible in the context of close
personal relationships. Indeed, part of what characterises such relationships is
the presumption that we can (and perhaps should) use each others’ good as a
reason, though of course this permission can be withdrawn. This is less
plausible in the state paternalism cases that concern us here: this does not
involve a close relationship, there is rarely univocal support for policy
responses to impairment, so we can expect that at least some of the intended
beneficiaries might withhold their consent for their good to be counted, and it
is unlikely that consultation is impossible.
However, whilst beneficiaries should be canvassed, requiring explicit con-
sent from every individual may be overly demanding. (Indeed, impossible if
some of the beneficiaries are yet to be born, as is often true when embarking on
a long-term research programme.) Thus, a less individualised approach may
be used. For example, if consulting with a sample of the deaf community
regarding the desirability of cochlear implants demonstrates that views are
heterogeneous—some would welcome them, some consider them a threat to
their culture and identity—and the benefit can only be justified by appeal to all
or most deaf individuals, then clearly the policy should not be pursued. If the
calculations are more finely balanced, further consultation might be needed. If
we fear our consultation process is not representative of all intended
beneficiaries—perhaps it is dominated by long-established Deaf activists, at
230      

the expense of those who are newly experiencing hearing-loss; or dominated

by hearing parents of deaf children at the expense of the hearing impaired—
then the process should be improved.
That this would be an implication of my account is surely a benefit rather
than a cause for concern. It is not unreasonable to hold that respecting all
citizens as democratic equals should involve taking their views about policies
designed to benefit them seriously. Indeed, such consultation is valuable not
merely as a means of respecting individuals’ authority over their good.
Without it, any attempt to make people better off is likely to be ineffective:
those living with an impairment are well-placed to know the kinds of changes
that would improve their quality of life, and when benefits are provided in
opposition to beneficiaries’ values and commitments then often they simply
will not be used. ‘Nothing about us without us’ is a central slogan of the
Disability Rights Movement for a good reason. Thus, even pro-paternalists
should accept such consultation if they hope to be successful.
More fundamentally, though, I contend that the goal of distributive justice
is not to promote a predetermined conception of individuals’ good, but to
enable them to form and pursue their own conception of the good. As has
already been considered, if equalising or maximising welfare was our goal,
capabilities—and especially the broadly construed capabilities for control—
would be a poor choice as a means to achieve it (see §3.4 and §7.4). Few are a
perfectly reliable guide to what would be in their best interests in every
scenario, and fewer still will always reliably stick to the path that would
leave them best off once identified. If our goal is not to ensure that people
live well but to enable them to live according to their own plans and by their
own volition, the reason to focus on capabilities is clear. And this focus should
be maintained when deliberating about how the rather abstract capabilities for
control should be operationalised as policy: again, the goal should not be
bestowing all the benefits we can, but enabling control via the provision of
acceptable options.
As discussed, we cannot determine what is acceptable without some con-
sideration of individuals’ interests and the options they consider viable.
However, our aim should not be ensuring individuals have what they most
prefer, but to allow them to set and pursue ends in central areas of their life. As
such, the justification for including a specific option is that it is valued rather
than valuable. The implications of this are, first, that this implies no judgement
regarding how people should choose and so no suggestion that we are mis-
taken if we fail to opt for some particular functioning, or if our preferred
option is not amongst those offered. Second, the process of selecting these
  ’       231

options will not involve an unauthorised appeal to individuals’ good: both

because the process will involve consultation, and because it will not seek to
identify what is generally good for people in any case. Thus, we should not be
seeking individuals’ views on whether they have any thwarted preferences or if
there is anything the state could do to improve their quality of life. Rather, we
should seek the views of potentially disadvantaged individuals on whether they
have what they are entitled to (control), and what they would require to rectify
this lack.⁷⁰
For example, if we were to consult with deaf individuals, our goal should not
be to ascertain whether they would prefer to hear, but whether being deaf
deprives them of control they are entitled to. Thus, we should not be attempt-
ing to determine how they feel about their deafness—for example, ‘are you
satisfied?’ or ‘do you wish your hearing were better?’—but what they are able
to do—for example, ‘do you have access to adequate education in an appro-
priate form?’, ‘can you access leisure activities?’, or ‘do you have chances to
form social relationships?’—and what they might need to have those oppor-
tunities they lack. Recall that our goal should be to provide control to a
threshold level rather than enabling specific functionings, so there is nothing
inherently or necessarily disadvantageous about being unable to hear—or
being unable to perform various other specific functionings (§2.7). We should
resist the ‘ableist preferences’ that ‘uncritically assert that it is better for a child
[or anyone] to walk than roll, speak than sign, read print than read Braille,
spell independently than use a spell-check, and hang out with nondisabled
kids as opposed to other disabled kids’.⁷¹ In other words, we should resist the
push to normalise, and instead aim to ensure that individuals can be fully
participating members of society regardless of their atypical modes of func-
tioning (§5.4).⁷²
Thus, there is no reason to assume that the only, or the best, means of
enabling control will involve the elimination of impairment. However, this
also does not imply ‘cures’ are ruled out. The point of this chapter’s central
example is decidedly not to suggest that cochlear implants can never be offered
in the absence of universal support. In a majority-hearing society, depending
on other available technology, it may be amongst the best ways to offer some

⁷⁰ As I have argued elsewhere (Begon 2021b). ⁷¹ Hehir 2002: 3.

⁷² As Ashley Taylor points out: ‘if people with disabilities are to be fully included in political decision
making, their differential means of expression and communication would need to be recognized both
within schools and in broader society’ (Taylor 2012: 119). However, the scope of our concern should be
broader: it is not only capabilities for inclusion in political decision-making that individuals are
entitled to.
232      

people control. In general, my contention that curing impairments should not

be the default option does not imply funding this can never be justified.
Crucially, what must be avoided is the assumption that individuals would be
better-off if they could function normally (hear, in this example), that making
people better-off should be goal of justice, and that the benefits of this
functioning can justify the policy regardless of the views of those on whom
it is bestowed.
We should not fund implants if they are not welcomed by enough people to
justify the expenditure, but we also should not fund them simply because
enough people would welcome them. Ultimately justice does not demand that
we offer implants simply because enough people would prefer to hear, any
more than the state is not obliged to offer cosmetic surgery just because enough
people would prefer to be beautiful. Considerations of individuals’ good
should be subsidiary rather than central to specifying the requirements of
distributive justice: determining whether individuals have the acceptable
options they need for control in a domain, rather than being an end in itself.
Returning to our example, given the deep objections many deaf individuals
have to cures (echoed by those with many other impairments), it is highly
unlikely they will have what they are entitled to simply because they are
offered an undesired opportunity. So even if justified as one response to
deafness, this is unlikely to be sufficient for the state to fulfil its obligations
to deaf individuals.⁷³ Reaching such determinations about what is required to
provide individuals with capabilities to control will not require violating the
anti-paternalist filter by making unauthorised appeals to individuals’ good. As
such, the resulting offers will not constitute paternalist interventions. Thus,
individuals will be shown appropriate respect as autonomous agents, as well as
being enabled to exercise control over their lives.

7.8 Soft Paternalism, Autonomy,

and Background Conditions

However, it may be objected that the problem with allowing such control is
that those who have it will misuse it. As frequently noted, individuals can
make mistakes. Indeed, as Chapter 6 argued, adaptation to restricted options

⁷³ Recall that this might fulfil state obligations if no alternatives were available. Although, as §2.4
discussed, this does not preclude an obligation to develop new options: the parameters of the feasible
are far from fixed.
  ’       233

may influence individuals’ decisions, causing some to fail to utilise, or even

value, the opportunities they are entitled to. It is unsurprising that people
reject various functionings when they have been forced to adapt to their
absence—for example, deciding that education or meaningful work or social
relationships are ‘not for them’.⁷⁴ Should recognition respect really be priori-
tised over living a good life? Should being able to direct our life be prioritised
over having our life go in the right direction?
One immediate line of response here would be to reiterate that our version
of anti-paternalism should be soft paternalist. In other words, to insist that
only autonomous choices should be immune from interference. As noted
earlier, it would be implausible indeed to insist that we cannot interfere even
if an individual is being coerced or lacks knowledge of relevant empirical
facts—regarding the rottenness of a bridge, for example. Determining exactly
at what point on the scale of autonomy intervention is warranted is not
straightforward.⁷⁵ Crucially, though we also should not set overly demanding
conditions for a decision to count as autonomous—as Feinberg arguably does
by excluding choices made in a disturbed emotional state, based on mistaken
reasoning, or not carefully considered.⁷⁶ We should not attempt to rule out
spontaneous, ill-considered, and ill-informed choices when embarked on
autonomously, if our goal is to allow people to be self-directing (see §7.5.2).
The thinly perfectionist liberalism I advocate implies that we should aim to
ensure people act with second-order autonomy: actively and willingly choos-
ing an overall way of life without compulsion or influence. However, the state
should not demand that people express autonomy in every decision they
make and every rule or norm they follow. Thus, individuals can be allowed
to act without first-order autonomy, which may include making risky and
spontaneous decisions, or becoming a member of a group that requires
adherence to rules they do not critically endorse.⁷⁷ Hence I would rule out

⁷⁴ Adaptive preferences may also have an influence on, and potentially render unreliable, individ-
uals’ views regarding the more general scope of our distributive entitlements (§6.6). However, my
concern here is with the specification and exercise of our entitlements.
⁷⁵ Depending on the decision, this may seem to either allow interference in an overly broad range of
cases (Fateh-Moghadam and Gutmann 2014; Hanna 2018a; Mullin 2014) or else be too permissive to
choices that are just autonomous enough (Arneson 2005; De Marneffe 2006, 2013; Shafer-Landau
2005). For discussion see Begon (2016b).
⁷⁶ Feinberg 1986: 104–6. To mitigate this problem, Feinberg qualifies that it is only when such
choices are harmful that they can be subject to intervention. However, this raises the further thorny
issue of determining what constitutes a harm, which may allow for the further imposition of a thick
perfectionist conception of the good (see Begon 2013). For example, if we assume losing our hearing is a
harm and gaining it is not, there may be a much higher bar for a decision to refuse a cochlear implant
than the decision to get one.
⁷⁷ For a useful discussion of this distinction, and a defence of more thickly perfectionist liberalism
see Chambers (2008).
234      

allowing someone to ignorantly cross a rotten bridge unmolested if their

ignorance is chosen.⁷⁸ Employing Colburn’s account of autonomy (outlined
in §6.3) means choices should meet the endorsement and independence
conditions, such that individuals decide for themselves what is valuable and
are able to pursue it. This may include them deciding that what is valuable is a
life that does not involve the expression of first-order autonomy. Thus, the soft
paternalism that I advocate allows interference only with choices that lack
second-order autonomy.
Soft paternalism can be misused to exclude individuals with impairments by
setting conditions for autonomous choice they cannot meet: perhaps mere
diagnosis of a cognitive impairment is taken to warrant a judgement of
autonomy deficit; perhaps a preference for an impairment is taken as of
evidence of mistaken reasoning. Such assumptions must be avoided. We
should focus on an individual’s capacity in a particular context, and not
assume this is identical, across all contexts, for all people with the same
impairment. And we should not assume that the content of a decision
means it cannot reflect an autonomous choice. Individuals with impairments
are as capable of choosing and acting autonomously as everyone, but also as
capable of failing to. Thus, the possibility of abuse should not dissuade us from
a commitment to soft paternalism. As with other normative powers, an
individual’s PPE is only binding when autonomously expressed.
Does this mean that nothing can be done to correct apparent mistakes if
individuals meet the autonomy threshold? Not quite. Individuals are owed
capabilities, understood as substantive opportunities or ‘real freedoms’ (§4.2).
If the reason someone rejects these opportunities is systematic bias and
oppression, ensuring access to these capabilities would speak in favour of
eliminating this. This implies that in the long run, a system designed according
to the principles of the capability approach would work to eradicate the
systematic injustice that distorts individuals’ expectations and views of their
capacities. Only by doing so can capabilities for control be provided.
Specifically, this would involve working to ensure all individuals have experi-
ence of accessing the central capabilities (§6.7).

⁷⁸ There are many ways in which this case can be complicated, with which I cannot engage fully here
(see Hanna 2018a, 2018b, 2012). For example, what if the bridge-crosser does not want to cross a rotten
bridge but did not want to expend the effort required to discover which bridges are rotten? Should this
risk be considered autonomously chosen, and thus immune to intervention? In brief, my view is that
insofar as they wish not to expend the effort of educating themselves, but do not wish to cross the
bridge in ignorance, intervention would be permissible by a soft paternalist (contrary to Jason Hanna’s
interpretation of the case).
  ’       235

Yet it is important to emphasise that our goal is not to ensure that people
never make mistakes—or even to determine what constitute mistakes—but to
ensure that they choose autonomously and authentically. In other words, to
allow people to make judgements that have arisen in sufficiently favourable
circumstances that they can be said to be authentic to the individual’s person-
ality, that they would endorse on further reflection, and would not reject if
more informed of relevant empirical facts.⁷⁹ Whilst I have argued that we
should not allow the views of individuals who have not experienced the central
capabilities to determine what these capabilities are (§6.6), many individuals
with such justice adaptive preferences will be neither irrational nor non-
autonomous. As such, they retain authority over their good. It is only when
individuals have non-autonomous, well-being adaptive preferences (see §6.3)
that a commitment to soft paternalism may mean interference is warranted.
Those who have merely comprehensively adapted to diminished options, and
so reject seemingly valuable options as a part of an autonomously formed and
coherent set of preferences and values, retain their PPE. Thus, for example, we
would not excise the capability for forming social relationships from the list of
entitlements on the basis of its rejection by those who have not experienced it,
but would not use their lack of experience as a reason to persuade or force
them to engage in it.
In brief: when they are choosing autonomously, we should allow people to
decide how to make the best of a bad situation, whilst also working to improve
that situation. For example, if an individual chooses to accept cochlear
implants as the only way to enter the job market or form relationships,
we should not interfere with this decision. The decision may lack global or
second-order autonomy—they are influenced into choosing an overall
way of life as a hearing, rather than a deaf, person due to the lack of
accommodation—but can still instantiate local autonomy—it authentically
reflects their view about how to balance the options that are available to
them. However, insofar as our goal is to enable people to have second-order
autonomy, the fact that the state should not paternalistically interfere with the
decision does not mean the situation creates no obligation to act. The state
should enable individuals to actively and autonomously choose their overall
way of life, and not require them to choose between their impairment and the
capabilities they are entitled to. The same reasoning applies to many other
cases, including an individual contemplating conversion therapy: the decision

⁷⁹ See Dworkin 2000: 160.

236      

should not be interfered with when locally autonomous, but the wider
circumstances that push people into making it should nonetheless be altered.
In the short run, then, interference would not be permissible even if
individuals’ autonomous and authentic preferences involve the repudiation
of apparent benefits. If we value autonomy and avoiding paternalism then we
should allow people to make decisions for themselves even when we think
their choices are mistaken, and even when their decisions are grounded in
distorting influences that justice demands should ultimately be eliminated.⁸⁰
Allowing individuals to truly decide for themselves what is valuable, and
giving them the ability to pursue it might require allowing individuals to
autonomously choose both non-autonomous ways of life, and forms of func-
tioning that leave them worse off than they might otherwise have been.

7.9 Cures and Parental Paternalism

The central case on which I have focused concerns a state decision to offer a
cure for an impairment to adults, who will be free to use this or not. However,
curing impairments is more commonly considered in the context of parental
decisions on behalf of existing children, or regarding whether to bring children
with impairments into existence, via pre-implantation genetic diagnosis
(PGD) or selective abortion. Insofar as these decisions concern children, it
may seem that these are not cases in which worries about paternalism apply.
Indeed, these may seem to be cases in which paternalism is exactly appropri-
ate. There is no autonomous agent to whom recognition respect is owed in the
case of infants, foetuses, and (especially) yet-to-be-implanted embryos, and
parents must make decisions on their children’s behalf, and should aim to
promote their welfare when doing so. Nonetheless, many of the same consid-
erations will be significant, and can provide guidance in these cases too. They
may lack autonomous capacities now, but they can still be respected as future
autonomous agents.
Centrally, this requires that the state ensures all people have what they are
entitled to. The guiding consideration behind state policy, therefore, should
be guaranteeing not only children’s welfare but also their future ability to have
central capabilities for control. Further, whilst we might allow parents some

⁸⁰ See Begon (2021b). A benefit of an approach, like my own, where people collectively determine
the content of entitlements is that all continue to be entitled to opportunities even if a minority do not
utilise them. This may encourage their future uptake and further undermine distorting influences.
  ’       237

discretion in how to best promote welfare, access to future capabilities should,

as far as possible, be assured. Possessing these capabilities will give people the
chance to have a decent quality of life but, as we have seen, may not give them
the best possible life, especially if they make bad decisions. More generally,
being justly treated is no guarantee of high quality of life, and unjust treatment
need not always be bad for our welfare. For example, experiencing ‘solidarity
in indignation’ may improve the well-being of individuals subject to legal
discrimination, whilst those used to privileged treatment might experience
reduced well-being if treated as a ‘mere’ equal.⁸¹ The relationship between
disability and well-being is not, therefore, predetermined by my account:
disabled individuals definitionally lack what they are entitled to, but need
not have low quality of life. The relationship between impairment and well-
being is even more open to variety, as Chapter 5 explored.⁸² More importantly,
merely having an impairment does not necessarily prevent individuals from
having what they are entitled to, and as such, there is no general requirement
to eliminate impairments—in adults or children.
However, whilst we should not assume impairments usually considered
disabling warrant cures or removal, we should also avoid generalising in the
other direction, and insisting, like Barnes, that there is no discrepancy between
causing and removing any impairment.⁸³ As I have repeatedly emphasised, the
effects impairments have on individual lives in different circumstances are
diverse, and we must be attentive to specific features of cases to determine
their probable effects (see §1.7, §5.6). I will make two broad claims regarding
parental decisions. First, concerning selection, I suggest that we should not
select for disabling impairments (those that prevent us having what we are
entitled to), but can select for or against non-disabling impairments. This
accords with Joseph Stramondo’s conclusion that parental decisions regarding
PGD should be limited by the requirement that the child has ‘an acceptable
opportunity range’ if reasonable accommodations are made.⁸⁴ This leaves
considerable scope for parental discretion if we accept the complex-difference
account of disability, defended in Chapter 5, since ‘many modes of functioning

⁸¹ Schemmel 2011: 140–1.

⁸² As McBryde Johnson (2003) succinctly summarises: ‘the presence or absence of disability doesn’t
predict quality of life’.
⁸³ Barnes 2014.
⁸⁴ Stramondo 2017: 476. The ‘reasonable accommodation’ clause included by Stramondo (2017:
485–90) is important to give us some guidance on the large grey area, covering impairments that are
disabling only due to a failure of state institutions or those that are not but in danger of becoming so
(under a new administration, say).
238      

can still allow for a plenty broad complement of life opportunities’, though of
course not ‘any mode of functioning will do this’.⁸⁵
Second, causing or removing impairments raises a number of additional
complexities which I will not pursue in detail here. I merely contend that
impairments should be treated analogously with other forms of human vari-
ation, some of which will have no significant impact on our access to our
entitlements, some will mean our access to them will be restricted in current
circumstances, and some will make it extremely difficult to ever have the
control we are entitled to. If we think parents should not alter innocuous
features like hair colour, or contingently disadvantageous features like sex,
race, or sexuality in discriminatory contexts, then the same should go for
impairments in these classes. Some will believe that parents should be permit-
ted to select or engineer their children to ensure they have some desired suite
of ‘features’, others that doing so conflicts with the unconditional welcome
children are owed, or that a child’s right to an open future requires that we do
not make significant inessential alterations to their body or mind until they
have the capacity to understand the consequences of these decisions.⁸⁶ On the
one hand, respecting individuals’ authority over their good does seem to
require refraining from making inessential decisions grounded in assumptions
about what will be best for them. Yet, on the other, much of parenting involves
making just this sort of decision. I will not, therefore, take a view on the correct
position here, only that the relevant aspect of the feature we are selecting for or
against should be its effect on the child’s ability to have the control they are
entitled to. The mere fact that it is an impairment, and so will render their
functioning atypical, provides no justification for treating it differently to any
other form of variation.

7.10 Conclusion

Disabled individuals are all too often treated like children, and like children
this can entail the assumption that it is appropriate to act in what the state
believes to be their best interests, rather than allowing them to direct their own
life. Yet in the vast majority of cases, individuals with impairments are, and
should be treated as, autonomous agents. This means that policies designed to

⁸⁵ Stramondo 2017: 484; §5.6.

⁸⁶ For some competing views on this point, see: Asch and Wasserman 2005; Bou-Habib and
Olsaretti 2014; McMahan 2005; Savulescu and Kahane 2009; Stramondo and Campbell 2020;
Wasserman 2009.
  ’       239

benefit them should not be justified by unauthorised appeals to their good. Yet
they also should not be justified by authorised appeals to their good, since the
goal of just public policy should not be welfare promotion. The state must
ensure all individuals can direct the course of their own lives, not guarantee
them a particular level of well-being, contrary to their autonomous decisions.
The problem with the state acting as if it knows what is in the best interests
of those it aims to assist without consulting them might seem clear: often it
does not, and its policies will therefore be ineffective. Whilst frequently this is
indeed the case, I have argued that even when the state gets it right—engages
in successful paternalism—it would still not be justified in doing so. First,
policies that appeal to the good of their beneficiaries and yet are repudiated by
these beneficiaries are paternalist. As such, they fail to show appropriate
recognition respect, both by using individuals’ good as a reason without
authority, and then potentially intervening in their lives without adequate
justification. Second, they mistake the proper goal of just policy, which is not
to achieve some level of well-being, but to enable people to have what they are
entitled to. On my view of entitlements, this means allowing people to choose
and pursue their own goals in central domains of their life. These might not
always make their life as good as possible. However, sometimes well-being is
not what we care about, and even when it is, we ought to have the freedom to
risk it, and to make our own mistakes. Consultation with those affected by a
policy prior to implementation is essential, then, but not to make sure we are
effectively increasing welfare, but so policies meet the demands of justice and
secure central capabilities for those they are designed for.
 Conclusion

When thinking about disability in the context of justice, our focus should not
be on the ways in which people’s bodies and minds function differently.
Instead, I contend that we should shift our focus to the limitations that this
leads to—for particular people, in particular contexts—and, I claim, only a
subset of these limitations matter. Specifically, those that prevent individuals
from having control in certain central domains of their life, by restricting the
availability of acceptable options or the ability to freely choose between them.
The problem that disability raises is not the mere fact of difference, but the
ways in which that difference is accommodated (or not) and the limitations it
may cause. Our focus, therefore, should not be on whether someone has
received a particular diagnosis, or on the most visible deviations from what
we take to be the species norm, but on whether people have access to central
capabilities.
Yet it may be objected that by focusing so narrowly on whether individuals
lack what they are entitled to as a matter of justice, I leave out much that is
central to the experience of disability. There are various ways in which this
concern could be cashed out. On one iteration, an opponent might agree that
disability should be understood as lacking what we are entitled to, but dispute
limiting the scope to what we are entitled to as a matter of justice, and not what
morality more broadly construed might demand. Alternatively, it may be
objected that focusing on an absence of entitlements is too narrow, however
broadly this is understood, because doing so ignores features of the disabled
experience that cannot plausibly be explicated in the language of entitlements.
I begin with the latter worry, which can itself be split into two concerns.
First, individuals will not be disabled by losses that they are not, in any sense,
entitled to, but which may nonetheless be valuable or contribute to their well-
being. Second, it may seem objectionable that disability is defined entirely in
terms of the lack of valuable capabilities, and so does not definitionally include
more positive elements of the disabled experience: access to community,
sources of communal solidarity and identity, diverse ways of experiencing,
understanding, and interpreting the world, acquiring new skills and abilities,
and so on.

Disability Through the Lens of Justice. Jessica Begon, Oxford University Press. © Jessica Begon 2023.
DOI: 10.1093/oso/9780198875611.003.0009
  241

However, though neither well-being losses nor these more positive elements
are constitutive of disability on my view, this need not imply that these
elements cannot be accounted for, in the sense of being acknowledged as
significant elements of many disabled people’s experiences. Disability is
defined as lacking opportunities we are entitled to as a result of impairment
(as atypicality not deficiency). The connection to impairment is essential to
avoid disability being synonymous with disadvantage, whatever the cause.
However, this also means that disability is defined both as lacking what we
are entitled to, and as functioning atypically. This latter component may lead
to losses of well-being that are not matters of entitlement (for example, a lack
of valuable sensory experiences), and may lead, too, to many positive experi-
ences both as a direct result of the atypical functioning (for example, new
experiences, understanding, and skills) and those more indirectly related to it
(for example, community membership, solidarity, and pride). Thus, disability
on my account is acknowledged to involve experiences both difficult and
desirable; both unpleasant and life-affirming.
An account of disability that excluded such elements would certainly be
objectionable. However, there are also problems with making these elements
of disability central to our definition. Whilst disability can be associated with
well-being losses, it need not be, and atypical functionings that inhibit well-
being are not always appropriately characterised as disabilities. Indeed, as §1.6
argued, defining disability by its contribution to well-being loss will render the
category entirely ubiquitous, and so of little use in identifying a significant
dimension of disadvantage and the individuals who experience it. Attempting
to identify some positive experiences that can provide a plausible grounding
for the category of disability will be similarly unsuccessful, given the diversity
of such experiences—even all those entitled to membership in the disabled
community (Barnes’s criteria, §1.7) will not be part of this community, nor
have identical experiences within it if they are. Further, insofar as we want to
identify disability as useful in the context of recognising and rectifying injust-
ice, it is essential to include those who do not share any of the many positive
experiences of living with impairment.
Of course, such arguments assume that our account of disability should be
useful in the context of justice and, again, I leave open whether in different
contexts and for different purposes, a different account of disability might be
more appropriate (see §1.8). My core point here is simply that insisting that
some part of the disabled experience should not be written into the definition,
as a constitutive component of disability, does not mean the significance and
prevalence of these experiences will be occluded. Indeed, my neutral account
242      

of impairment makes clear the range of physical and cognitive experiences and
ways of being—both positive and negative—that may be part of disability.
I turn now to the second objection: that there is no reason to restrict
disability to diminished abilities to do what we are entitled to as a matter of
justice, as opposed to what we are entitled to be able to do as a matter of
morality more broadly. Insofar as we believe the domain of morality is broader
than the domain of justice, it may seem to render my account of disability
unjustifiably narrow to restrict it only to the loss of justice-relevant entitle-
ments. This concern was raised by a reviewer of the manuscript, and it is worth
considering a version of the counterexample they raise to appreciate the full
force of this worry. Imagine, then, an individual with Down’s syndrome. She
lives in a society in which everyone studiously respects her rights, treats her
with basic civility, complies with relevant laws, helps her when she needs it,
and so on. However, everyone also has no interest in forming bonds of
friendship or love with her, snubs her, fails to get to know her, does not
appreciate her or her interests, and does not have special trust in her. The
criticism goes that this individual lacks the genuine control over her close
personal relationships that she is morally entitled to. However, on the assump-
tion that justice is not concerned with the intricacies of special relationships
(beyond a basic opportunity to form them) then she will have all she is entitled
to as a matter of justice, and so will not be disabled on my account.
I agree that it would be implausible if such cases were excluded from my
account, yet I believe they can be accommodated. Ultimately, the disagreement
arises from a different understanding of the scope of justice employed in the
example. First, adopting a capability metric of justice allows for the inclusion
of apparently private matters as entitlements of justice. Second, and more
fundamentally, we must determine what is meant by justice and how this
differs from the demands of morality. Whilst I accept the standard view
that justice is what we are owed from the state, this is open to a variety of
interpretations. I suggest that the realm of state responsibility should
be broadly construed. Not only does this seem the best interpretation of
justice, it will also avoid an overly narrow account of disability. I consider
each point in turn.
First, then, it is worth reiterating the significance of our selection of a
particular metric of justice. As Chapters 3 and 4 argued, I believe we should
adopt a version of the capability approach, according to which all individuals
are entitled to exercise genuine control in central domains of their lives.
Simply accepting this metric immediately calls into question the claim that
justice does not apply to the intricacies of special relationships. Whilst it is true
  243

that the state should not be concerned with the ‘intricacies’, in the sense of
micro-managing personal interactions (more on this below), our relationships
are nonetheless incorporated into a capability approach to justice in two ways.
First, relationships are included directly, as items on the list of central
capabilities: for example, ‘affiliation’ and ‘emotions’ on Nussbaum’s list, or
‘doing good to other’s’ on Wolff and De-Shalit’s modification of it (which
I broadly adopt, see §4.7). Second, indirectly, as the preconditions for having
the genuine control necessary for other capabilities, since this requires both
the external and internal preconditions of choice. As Chapter 6 argues, being
in oppressive relationships or subject to unjust norms can distort the process
of preference formation: unattainable opportunities may be consciously or
unconsciously downgraded, individuals’ choices may not reflect their prefer-
ences, or individuals may internalise a self-conception according to which they
are not entitled to certain opportunities (see §6.6). All of this can undermine
the degree to which individuals can exercise genuine control. If, as in the above
example, an individual with Down’s syndrome is raised in a context where
they are excluded from caring or loving interactions, then this is likely to
affect their sense of self-worth such that they lack not only capabilities
such as affiliation but also less directly affected capabilities, such as partici-
pating effectively in political choices, having the real opportunity to hold
property or to seek employment. They may, for example, consider them-
selves ‘unworthy’ of engaging in such pursuits or (consciously or uncon-
sciously) lower their judgement of their value.
Thus, on a capability view, the demands of justice are not exhausted by mere
compliance with the demands of basic civility, absence of rights violations, and
obeying the laws. Without access to some others who have at least some
interest in forming the bonds of special relationships, individuals may directly
lack some capabilities—to live with and toward others, to love those who love
and care for them. And due to the distorting influence on their capacity to
form and act on their preferences that living in a condition of constant
ostracisation can be expected to have, they are likely to lack other capabilities
too. As Kimberley Brownlee has convincingly argued, having ‘minimally
adequate opportunities for decent or supportive human contact including
interpersonal interaction, associative inclusion, and interdependent care’ is
an essential part of a minimally decent human life.¹ Further, such supportive
interpersonal contact is not merely intrinsically valuable, but also instrumen-
tally valuable as a means to the ‘development and the maintenance of . . .

¹ Brownlee 2013: 199.

244      

cognitive, physical, emotional, linguistic, and social abilities’,² which in turn

are necessary to exercise many capabilities. Brownlee focuses primarily on
cases of extreme isolation or brutal treatment, but even more mundane social
exclusion, such as persistent coldness or standoffishness, can undermine such
capacities, especially of those who are otherwise vulnerable.
Detailed arguments in favour of a capability approach to justice have
already been given, but it is worth noting that this highlights another of its
benefits: its ability to accommodate the importance of relationships, and
acknowledge the ‘potential problem of justice’ when differences in power
and status lead to unequal relationships.³ It places central emphasis on the
harms of ‘social subordination’, and of denying some members of society ‘the
status of full partners in interaction, capable of participating on a par with
the rest’.⁴ I will not attempt to reiterate the arguments of relational egalitarians
and recognition theorists in favour of including a concern for how we relate to
each other within the realm of justice here. Rather, I merely note that the
capability approach accommodates the concerns of relational theories in the
two ways outlined: as components of particular capabilities, and the precon-
ditions of having them. Further, the capability approach demonstrates that,
contrary to the claims made by some of these theorists, we need not choose
between a distributive or relational theory of justice. Indeed, though often
framed as a distributive theory (distributing capabilities), it is also employed
by relational egalitarians (most obviously Anderson).⁵ Thus, insofar as it is
concerned both with how individuals stand in relation to each other, inde-
pendently of the knock-on effects on wider quality of life, and with the
distribution of valuable opportunities and the ability to make use of them, it
seems best characterised as a ‘pluralist egalitarian’ account of justice.⁶
However, even if capabilities are accepted as the metric of justice, and so
relationships are potentially within its scope, it may be objected that cases like
the ostracised individual with Down’s syndrome still cannot be accommo-
dated, because all the state can do to secure relationships is to allow the
freedom to form them and basic opportunities to do so. Once this is done,
the individual has all they are entitled to as a matter of justice, and so will not
be considered disabled by their remaining exclusion on my view. The second

² Brownlee 2013: 200.

³ Schemmel 2011: 125. Note, too, that the capability approach need not endorse the perfectionism
Schemmel (2011: 141) attributes to it. As Chapter 4 argued, our commitment to enabling all individuals
to exercise control in their life need not be grounded in the assumption that exercising particular
functionings, or even having the capability to do so, will be good for people.
⁴ Fraser 2000: 114. ⁵ For example: Anderson 1999; 2010b.
⁶ Following Kasper Lippert-Rasmussen’s (2015, 2016) terminology.
  245

point of disagreement, then, concerns the various possible interpretations of

the claim that justice is ‘what we are owed from the state’.
The strongest interpretation is that a claim of justice must be correlated with
a duty that could be enforced against a specific state, which is currently able to
fulfil it. This leads to the narrowest understanding of the scope of justice and,
in particular, means it will primarily be restricted to action in the public, not
the private, sphere—and hence exclude cases such as the individual with
Down’s. This is because if demands of justice must be both realisable (i.e.
meet the feasibility constraint) and achievable via direct state action, this leads
to at least three limits on what they can encompass.
First, the enforcement tools of the state tend to be rather blunt: universally
applicable legal sanctions, education programmes, or changes in the design of
public spaces, for example. These are slow-acting and general in application,
and thus unlikely to be effective in regulating problematic personal inter-
actions. Thus, even if some morally acceptable methods of intervening in such
interactions did exist, these may not be available within the apparatus of the
state. Second, a possible intervention may be rendered morally unacceptable if
performed by the state. That is, even if the state could engage in fine-grained
interventions in our personal lives and relationships we may not want the state
to reach into such private matters, especially by means we would consider
morally unacceptable, such as constant, invasive surveillance, or insulting
assessments of the quality of our choices. Third, there will be cases in which
no means of meeting a putative claim of justice are available to any agent. For
example, as I discuss in §2.4, we may lack the resources, technology, or
knowledge to enable some individuals with serious impairments to exercise
control in some central domains of their lives.
On a narrow reading of justice, then, cases in which the state cannot fulfil a
duty, the state should not fulfil a duty, or no one can fulfil a duty would all fall
outside its scope. However, this is not the view that I endorse. In §2.4 I reject
the feasibility constraint on justice, arguing that entitlements should be under-
stood as conditional entitlements. My focus there was to argue that the third
set of cases above should be included: it may be unjust that individuals are
deprived of access to central capabilities, even if the lack cannot be rectified.
However, these arguments apply more broadly, since cases of infeasibility
also include those in which claims can be fulfilled, but not (permissibly) by
the state. For similar reasons, I contend that these too may be instances of
injustices.
It is important to emphasise that this is an interpretation of the claim that
justice is the realm of state responsibility, rather than a rejection of this claim.
246      

My contention is that ascertaining the boundaries of this realm should

not depend on whether a specific state can currently fulfil some claim, but
rather a counterfactual or conditional test: if the state could deploy some
permissible method to rectify a putative injustice, then should it? Answering
in the affirmative is, I believe, sufficient for judging something to fall within
the scope of justice, even if there is little chance of the former clause being
met—either because we cannot imagine it ever being possible for the injustice
to be rectified, or because we cannot imagine the state could find a morally
permissible way to rectify it. For example, the gendered, unequal division of
domestic labour is unjust, even though the state should not directly intervene
in the distribution of household chores.
This does not lead the realm of justice to collapse into morality more
generally, since a meaningful distinction is retained from cases of wrongdoing
that are simply not the business of the state. Individual instances of promise
breaking, or failures of courtesy, for example, are not things the state is obliged
to rectify (regardless of the availability of morally acceptable means).⁷ This can
be contrasted with cases of structural oppression (such as the previous) or
consistently disrespectful treatment of some class of people. For example, an
individual instance of having one’s credibility wrongly deflated is not unjust,
whilst having it consistently deflated, across time and context, on the basis of a
negative prejudice attached to an identity feature, would constitute a (testi-
monial) injustice. The same can be said of individual instances of being treated
disrespectfully by some unpleasant individual (a moral failing on their part),
compared to being constantly and consistently overlooked, ignored, or other-
wise disrespected. The latter cases, unlike the former, are injustices, and are apt
for state intervention—on the condition that morally permissible means are
available.
It is also important to emphasise that more means are available to the state
than have traditionally been assumed. The state either cannot or should not
engage in constant policing of individual interactions to ensure, inter alia,
courtesy, fair and respectful treatment, appropriate attributions of credibility,
and an openness to marginalised hermeneutic resources. Yet this only means
that the state cannot prevent specific manifestations of such structural injust-
ices, and not that they cannot work to undermine the embedded norms and
practices that generate them. For example, returning to the individual with

⁷ It would also exclude those inequalities that do not seem to generate any entitlements, and thus
corresponding duties of rectification, such as Salieri’s complaint regarding Mozart’s superior musical
talents (Anderson 2010b: 10).
  247

Down’s syndrome, it is likely true that the state cannot act to ensure either a
desirable functioning outcome (they have social relationships), or, more
importantly, that this particular individual has the capability to form such
relationships (the genuine, not merely basic, opportunity, see §4.2). If not
actually impossible, any such attempt would likely be impermissibly interven-
tionist. Nonetheless, there are actions the state should take—for example,
education programmes to battle social norms that lead to the reluctance to
engage with individuals with Down’s; or screening applicants for state-
provided social work, healthcare, and education to ensure that they are not
prejudiced, guaranteeing some access to individuals with whom social bonds
may form. Such actions would, at least, reduce the possibility of future
occurrences and may even enhance this individual’s future capabilities.
In other words, and returning to the first point, if our metric is capabilities,
then much more of the private realm is implicated within the scope of justice,
since justice is interpreted as requiring that we have the substantive ability to
exercise control in many parts of our life—including the personal—and
genuine control usually requires freedom from interpersonal oppression.
Indeed, it seems likely that when someone is disadvantaged—that is, lacks
the capabilities they are entitled to—as a result of an impairment, then this will
be due to structural injustice. Outside of the realm of philosophical thought
experiments—someone with Down’s only happens to meet individuals who
just so happen not to want to associate with them, as the result of their
idiosyncratic preferences, not traceable to any systematic bias or hierarchical
systems—disadvantage as the result of an identity feature, such as impairment,
does not simply happen due to individual moral failings but due to systematic
marginalisation and exclusion. Thus, in such cases, we should assume, until
given evidence to the contrary, that there is a structural injustice the state can
and should work to rectify.
The adoption of the capabilities metric therefore reduces the significance of
the disagreement over whether duties of justice must be realisable by a
particular state, insofar as it expands the realm of state duties. Nonetheless,
it does not resolve the dispute. Even if there are more cases in which the state
can feasibly do something it certainly will not be able to fully rectify every
disadvantage, and sometimes there will simply be nothing at all the state can
(permissibly) do. Yet I would still contend that these constitute injustices
because, if the state could provide the absent capabilities, then they should
do so. Thus, even if the state is unable to prevent the ostracisation of the
individual with Down’s, for example, it remains unjust—if the state could
rectify it, it should. (Of course, if a misanthropically minded individual,
248      

impaired or otherwise, chooses to reject such opportunities, then there would

be no injustice in the subsequent absence of functioning achievement.)
No doubt some will find this interpretation of justice counterintuitive, but
I believe rejecting it has equally counterintuitive consequences. If Dworkin is
right that justice is, at its core, about showing equal concern and respect to all
people, then the narrow account of justice would exclude many instances of
apparent injustice. It is surely unjust, for example, when individuals live in
societies in which people like them are considered and treated as less capable
of contributing usefully, less likely to be trusted or more difficult to understand
when attempting to communicate, or simply not a viable prospect as a friend
or romantic partner. There may be no simple way for the state to dissolve the
social hierarchies that lead to such treatment, but whatever it can (permissibly)
do, it should.
To some extent, though, the disagreement may merely be semantic insofar
as it would be possible to agree regarding the content of disability, but insist
that this includes a lack of what we are entitled to as a matter of justice and
(some of) morality. What is substantively important, and what leads me to talk
in terms of justice, is the core distinction between disability as the systematic
deprivation of what we are entitled to, and the loss of opportunities that would
simply improve our quality of life. I believe that the opportunity to form
meaningful personal relationships should be placed squarely in the former
category. The ability to interact with, and be recognised as equals by, our
fellow human beings is an essential part of a decent life, and one which all
individuals are entitled to as a matter of justice. As reflection on capability
deprivation in disabled lives illustrates, lacking control in central domains of
our life seems best characterised as injustice, even when we can identify no
state able to eliminate it. Nonetheless, if others prefer to talk in terms of ‘justice
plus’ some other moral entitlements, they may do so. Crucially, though, the
combination of my broad account of justice—as capabilities for control, and
jettisoning the feasibility constraint—and my definition of disability—as lack-
ing what we are entitled to as a matter of justice as a result of an impairment—
does not exclude the kinds of disadvantages the state cannot immediately
provide.
Thus, though in some ways the arguments advanced in the various chapters
are independent—a reader may agree that disability involves lacking what we
are entitled to as a matter of justice, but disagree about what we are entitled to;
they may agree that we are entitled to capabilities for control, but disagree that
this should ground our understanding of disability; they might agree impair-
ment should be understood as atypicality, but dispute its relationship to
  249

disability, and so on—I believe (unsurprisingly!) that they work best together.
Combined with some accounts of justice my definition of disability would be
implausibly narrow (though so too would be such accounts of justice).
Combined with an understanding of impairment as deficiency, my account
of disability may obscure the positive elements of disabled experience, or
suggest impairments should be eliminated. This is not, of course, my view.
On the contrary, my central claim has been that our goal should not be to
normalise, should not be to remove atypical forms of functioning, nor to
ensure that all have the opportunity to perform the functionings we deem
valuable or best. We should not view disability with an eye to what is
regrettable, or what makes life harder—and not just because we get this so
wrong when it comes to disability. Everyone has difficulties to struggle with
and barriers to overcome, and it is not the state’s business to try and flatten
them all on our behalf. What the state should do, is to ensure that we all have
the substantive opportunity to choose amongst options, and the ability to
pursue them. We should, in other words, provide individuals with the chance
to control their functioning in central areas of their life, without specifying the
use they should make of this control or ranking various ways of functionings
as superior and inferior. It is possible that forcing or persuading people to
function in certain ways may sometimes make them better-off, but giving
everyone the best possible life should not be the goal of the state. Rather, it
should be to meet the demands of justice. Reflecting on disability and impair-
ment has helped us to see that this requires the provision of choice, and not
attempting to secure uniformity in people’s mode of functioning. Yet these
lessons are applicable beyond the context of achieving disability justice. The
goal of a theory of justice should be to support a diversity of abilities, of needs,
and of preferences, and to allow all people to choose for themselves, make
mistakes for themselves, and live their lives as autonomous agents.
 References

Aas, Sean (2016) ‘Disabled—Therefore, Unhealthy?’, Ethical Theory and Moral Practice 9
(5): 1259–74
Abel, Will, Elizabeth Kahn, Tom Parr, and Andrew Walton (2021) Introducing Political
Philosophy: A Policy-Drive Approach (Oxford: Oxford University Press)
Ahlstrom-Vij, Kristoffer (2013) Epistemic Paternalism: A Defence (Basingstoke: Palgrave
Macmillan)
Ahlstrom-Vij, Kristoffer (2018) ‘Epistemic Paternalism’, in Kalle Grill and Jason Hanna
(eds), The Routledge Handbook of the Philosophy of Paternalism (Oxford: Routledge),
261–73
Albrecht, Gary, and Patrick Devlieger (1999) ‘The Disability Paradox: High Quality of Life
Against All the Odds’, Social Science and Medicine 48(8): 977–88
Alexander-Passe, Neil (2015) ‘The Dyslexia Experience: Difference, Disclosure, Labelling,
Discrimination and Stigma’, Asia Pacific Journal of Developmental Differences 2(2):
202–33
American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental
Disorders, 5th edition (DSM-5) (Arlington, VA: American Psychiatric Association)
Amundson, Ron (2000) ‘Against Normal Function’, Studies in the History and Philosophy of
Science, Part C 31: 33–53
Anderson, Elizabeth (1999) ‘What is the Point of Equality?’, Ethics 109: 287–337
Anderson, Elizabeth (2010a) ‘Justifying the Capabilities Approach to Justice’, in Harry
Brighouse and Ingrid Robeyns (eds), Measuring Justice: Primary Goods and Capabilities
(Cambridge: Cambridge University Press), 81–100
Anderson, Elizabeth (2010b) ‘The Fundamental Disagreement between Luck Egalitarians
and Relational Egalitarians’, Canadian Journal of Philosophy 36: 1–23
Appiah, Kwame Anthony (1996) ‘Race, Culture, Identity: Misunderstood Connection’, in
K. A. Appiah and A. Guttmann (eds), Color Conscious: The Political Morality of Race
(Princeton: Princeton University Press), 30–105
Archard, David (1980) ‘Paternalism Defined’, Analysis 50: 36–42
Archard, David (1994) ‘For our own Good’, Australasian Journal of Philosophy 72(3):
283–93
Armitage, Richard, and Laura Nellums (2020) ‘The COVID-19 Response Must be
Disability Inclusive’, The Lancet Public Health 5: 5. DOI: https://doi.org/10.1016/
S2468-2667(20)30076-1
Arneson, Richard (1980) ‘Mill versus Paternalism’, Ethics 90(4): 470–89
Arneson, Richard (1989) ‘Equality and Equal Opportunity for Welfare’, Philosophical
Studies 56: 77–93
Arneson, Richard (1999) ‘Against Rawlsian Equality of Opportunity’, Philosophical Studies
93: 77–112
Arneson, Richard (2000a) ‘Perfectionism and Politics’, Ethics 111(1): 37–63
Arneson, Richard (2000b) ‘Luck Egalitarianism and Prioritarianism’, Ethics 110(2): 339–49
252 

Arneson, Richard (2005) ‘Joel Feinberg and the Justification of Hard Paternalism’, Legal
Theory 11: 259–84
Arneson, Richard (2006) ‘Distributive Justice and Basic Capability Equality: “Good
Enough” is Not Good Enough’, in Alexander Kaufman (ed.), Capabilities Equality:
Basic Issues and Problems (New York: Routledge), 17–43
Arneson, Richard (2010) ‘Two cheers for capabilities’, in Harry Brighouse and Ingrid
Robeyns (eds), Measuring Justice: Primary Goods and Capabilities (Cambridge:
Cambridge University Press), 101–27
Arneson, Richard (2013) ‘Paternalism and the principle of fairness’, in C. Coons and
M. Weber (eds), Paternalism: Theory and Practice (Cambridge: Cambridge University
Press), 134–56
Asch, Adrienne, and David Wasserman (2005) ‘Where is the Sin in Synecdoche?’, in David
Wasserman, Jerome Bickenbach, and Robert Wachbroit (eds), Quality of Life and
Human Difference: Genetic Testing, Health Care (New York: Cambridge University
Press), 172–216
Asexuality Archive (2012) ‘Things that are Not Asexuality’, Asexuality Archive, 27 May.
Available at: https://www.asexualityarchive.com/things-that-are-not-asexuality/. Last
accessed 1 December 2020
Austen, Jane (2004) Pride and Prejudice, ed. James Kinsley (Oxford: Oxford University
Press)
AVEN (Asexuality Visibility and Education Network), www.asexuality.org. Last accessed
1 December 2020
Baber, H. E. (2007) ‘Adaptive Preferences’, Social Theory and Practice 33(1): 105–26
Barclay, Linda (2010) ‘Disability, Respect and Justice’, Journal of Applied Philosophy 27(2):
154–71
Barclay, Linda (2018) Disability with Dignity: Justice, Human Rights and Equal Status
(New York: Routledge)
Barnes, Elizabeth (2009a) ‘Disability and Adaptive Preference’, Philosophical Perspectives
23: 1–22
Barnes, Elizabeth (2009b) ‘Disability, Minority and Difference’, Journal of Applied
Philosophy 26(4): 337–55
Barnes, Elizabeth (2014) ‘Valuing Disability, Causing Disability’, Ethics 125: 88–113
Barnes, Elizabeth (2016a) The Minority Body: A Theory of Disability (Oxford: Oxford
University Press)
Barnes, Elizabeth (2016b) ‘Reply to Kahane and Savulescu’, Res Philosophica 93(1): 295–309
Barnes, Elizabeth (2018) ‘Against Impairment: Replies to Aas, Howard, and Francis’,
Philosophical Studies 175: 1151–62
Baron-Cohen, Simon (2017) ‘Editorial Perspective: Neurodiversity: A Revolutionary
Concept for Autism and Psychiatry’, Journal of Child Psychology and Psychiatry 58(6):
744–7
Bayne, Tim, and Neil Levy (2005) ‘Amputees by Choice: Body Integrity Identity Disorder
and the Ethics of Amputation’, Journal of Applied Philosophy 22(1): 75–86
Begon, Jessica (2013) ‘Policy without Paternalism: A Capability Approach to Legitimate
State Action.’ Thesis submitted for Ph.D., University of Sheffield. Available at:
http://etheses.whiterose.ac.uk/id/eprint/5412
Begon, Jessica (2015) ‘What are Adaptive Preferences? Exclusion and Disability in the
Capability Approach’, Journal of Applied Philosophy 32: 241–57
Begon, Jessica (2016a) ‘Athletic Policy, Passive Well-Being: Defending Freedom in the
Capability Approach’, Economics and Philosophy 32: 51–73
  253

Begon, Jessica (2016b) ‘Paternalism’, Analysis 76(3): 355–73

Begon, Jessica (2017) ‘Capabilities for All? From Capabilities to Function, to Capabilities to
Control’, Social Theory and Practice 43: 154–79
Begon, Jessica (2018) ‘Elizabeth Barnes, The Minority Body: A Theory of Disability’, Journal
of Moral Philosophy 15(1): 100–3
Begon, Jessica (2019) ‘Sexual Perversion: A Liberal Account’, Journal of Social Philosophy
50(3): 341–62
Begon, Jessica (2020) ‘Disability, Rationality, and Justice: Disambiguating Adaptive
Preferences’, in David Wasserman and Adam Cureton (eds), The Oxford Handbook of
Philosophy and Disability (Oxford: Oxford University Press), 343–59
Begon, Jessica (2021a) ‘Disability: A Justice-Based Account’, Philosophical Studies 178(3):
935–62
Begon, Jessica (2021b) ‘Disadvantage, Disagreement, and Disability: Re-evaluating the
Continuity Test’, Critical Review of International Social and Political Philosophy (online first)
Begon, Rob, and Tom Billington (2019) ‘Between Category and Experience: Constructing
Autism, Constructing Critical Practice’, Educational Psychology in Practice 35(2): 184–96
Bergman, Ronald A., Adel K. Afifi, and Ryosuke Miyauchi (1995–2020) Illustrated
Encyclopedia of Human Anatomic Variation. Available at: https://www.anatomyatlases.
org/AnatomicVariants/AnatomyHP.shtml>
Bérubé, Michael (1996) Life as we Know it: A Father, a Family, and an Exceptional Child
(New York: Vintage)
Bettcher, Talia Mae (2014a) ‘Feminist Perspectives on Trans Issues’, in Edward Zalta (ed.),
The Stanford Encyclopaedia of Philosophy (Palo Alto, CA: Stanford University Press)
Bettcher, Talia Mae (2014b) ‘Trapped in the Wrong Theory: Rethinking Trans Oppression
and Resistance’, Signs 39(2): 383–406
Bickenbach, Jerome (2001) ‘Disability Studies and Bioethics: A Comment on Kuczewski’,
American Journal of Bioethics 1(3): 49–50
Bickenbach, Jerome (2013) ‘Disability, “Being Unhealthy,” and Rights to Health’, Journal of
Law and Medical Ethics 41: 821–8
Blom, Rianne, Raoul Hennekam, and Damiaan Denys (2012) ‘Body Identity Integrity
Disorder’, PLoS One 7(4)
Blom Rianne M., Sija J. van der Wal, Nienke C. Vulink, Damiaan Denys (2017) ‘Role of
Sexuality in Body Integrity Identity Disorder (BIID): A Cross-Sectional Internet-Based
Survey Study’, Journal of Sexual Medicine 14(8): 1028–35
Bogaert, Anthony (2015) Understanding Asexuality (Plymouth: Rowman & Littlefield)
Bognar, Greg (2016) ‘Is Disability Mere Difference?’, Journal of Medical Ethics 42(1): 46–9
Boorse, Christopher (1975) ‘On the Distinction between Disease and Illness’, Philosophy
and Public Affairs 5: 49–68
Boorse, Christopher (1977) ‘Health as a Theoretical Concept’, Philosophy of Science 44(4):
542–73
Boseley, Sarah (2019) ‘600 Years’ Supply of Cystic Fibrosis Drug Destroyed in Price Row’,
The Guardian, 27 March. Available at: https://www.theguardian.com/society/2019/mar/
27/600-years-supply-of-cystic-fibrosis-drug-destroyed-in-price-row (accessed 16 July
2019)
Bou-Habib, Paul, and Serena Olsaretti (2014) ‘Autonomy and Children’s Well-Being’, in
Alex Bagattini and Colin Macleod (eds), The Nature of Children’s Well-Being. Indicators
and Research 9 (Dordrecht: Springer)
Bovens, Luc (2002) ‘Sour Grapes and Character Planning’, Journal of Philosophy 86(2):
57–78
254 

Bovens, Luc (2009) ‘The Ethics of Nudge’, in Till Grüne-Yanoff and Sven Ove Hansson
(eds), Preference Change: Approaches from Philosophy, Economics and Psychology
(Berlin: Springer), 207–19
Brighouse, Harry, and Ingrid Robeyns (eds) (2010) Measuring Justice: Primary Goods and
Capabilities (Cambridge: Cambridge University Press)
Brownlee, Kimberley (2013) ‘The Human Right Against Social Deprivation’, Philosophical
Quarterly 63(251): 199–222
Bruckner, Donald (2009) ‘In Defense of Adaptive Preferences’, Philosophical Studies 142(3):
307–24
Brunning, Luke, and Natasha McKeever (2021) ‘Asexuality’, Journal of Applied Philosophy
38(3): 497–517
Buchanan, Allen, D. Dan W. Brock, Norman Daniels, and Daniel Wikler (2000) From
Chance to Choice: Genetics and Justice (Cambridge: Cambridge University Press)
Byrne, Alex (2020) ‘Are Women Adult Human Females?’, Philosophical Studies 177:
3783–3803
Byskov, Morten Fibieger (2017) ‘Democracy, Philosophy, and the Selection of Capabilities’,
Journal of Human Development and Capabilities 18(1): 1–16
Campbell, Stephen M., and Joseph A. Stramondo (2016) ‘Review: The Minority Body by
Elizabeth Barnes’, Notre Dame Philosophical Reviews. Available at: https://ndpr.nd.edu/
news/the-minority-body-a-theory-of-disability/
Campbell, Stephen M., and Joseph A. Stramondo (2017) ‘The Complicated Relationship of
Disability and Well-Being’, Kennedy Institute of Ethics Journal 27(2): 151–84
Carter, Ian (1999) A Measure of Freedom (Oxford: Oxford University Press)
Carter, Ian (2014) ‘Is the Capability Approach Paternalist?’, Economics and Philosophy 30:
75–98
Chambers, Clare (2008) Sex, Culture, and Justice: The Limits of Choice (State College, PA:
Pennsylvania State University Press)
Claassen, Rutger (2011) ‘Making Capability Lists: Philosophy versus Democracy’, Political
Studies 59(3): 491–508
Claassen, Rutger (2014a) ‘Capability Paternalism’, Economics and Philosophy 30: 57–73
Claassen, Rutger (2014b) ‘Human Dignity in the Capability Approach’, in Marcus Düwell,
Jens Braarvig, Roger Brownsword, and Dietmar Mieth (eds), The Cambridge Handbook
of Human Dignity (Cambridge: Cambridge University Press), 240–9
Claassen, Rutger (2017) ‘An Agency-Based Capability Theory of Justice’, European Journal
of Philosophy 25(4): 1279–1304
Claassen, Rutger (2018) Capabilities in a Just Society: A Theory of Navigational Agency
(Cambridge: Cambridge University Press)
Claassen, Rutger, and Marcus Düwell (2013) ‘The Foundations of Capability Theory:
Comparing Nussbaum and Gewirth’, Ethical Theory and Moral Practice 16: 493–510
Clayton, Matthew (2000) ‘The Resources of Liberal Equality’, Imprints 5(1): 63–84
Cohen, G. A. (1993) ‘Equality of What? On Welfare, Goods, and Capabilities’, in Martha
Nussbaum and Amartya Sen (eds), Quality of Life (Oxford: Oxford University Press), 9–29
Cohen, G. A. (1994) ‘Amartya Sen’s Unequal World’, New Left Review 203: 117–29
Cohen, G. A. (1995) Self-Ownership, Freedom, and Equality (Cambridge: Cambridge
University Press)
Cohen, G. A. (2008) Rescuing Justice and Equality (Cambridge, MA: Harvard University
Press)
Cohen, G. A. (2011) ‘On the Currency of Egalitarian Justice’, in M. Otsuka (ed.), On the
Currency of Egalitarian Justice and Other Essays in Political Philosophy (Princeton:
Princeton University Press), 3–43
  255

Colburn, Ben (2010) Autonomy and Liberalism (New York: Routledge)

Colburn, Ben (2011) ‘Autonomy and Adaptive Preferences’, Utilitas 23(1): 52–71
Colburn, Ben (2014) ‘Disadvantage, Autonomy, and the Continuity Test’, Journal of
Applied Philosophy 31(3): 254–70
Conly, Sarah (2013) Against Autonomy: Justifying Coercive Paternalism (Cambridge:
Cambridge University Press)
Coons, Christian, and Michael Weber (2013) ‘Introduction: Paternalism: Issues and
Trends’, in Christian Coons and Michael Weber (eds), Paternalism: Theory and
Practice (Cambridge: Cambridge University Press), 1–24
Cornell, Nicholas (2015) ‘A Third Theory of Paternalism’, Michigan Law Review 113:
1295–1336
Crisp, Roger (2006) Reasons and the Good (Oxford: Oxford University Press)
Crocker, David, and Ingrid Robeyns (2010) ‘Capability and Agency’, in Christopher
W. Morris (ed.), Amartya Sen (Cambridge: Cambridge University Press), 60–90
Cudd, Ann (2014) ‘Commitment as Motivation: Amartya Sen’s Theory of Agency and the
Explanation of Behaviour’, Economics and Philosophy 30: 35–56
Cureton, Adam (2016) ‘Offensive Beneficence’, Journal of the American Philosophical
Association 2(1): 74–90
Cystic Fibrosis Trust (2019) ‘Orkambi’. Available at: www.cysticfibrosis.org.uk/orkambi
(last accessed 15 July 2019)
Daniels, Norman (1985) Just Health Care (Cambridge: Cambridge University Press)
Daniels, Norman (1990) ‘Equality of What: Welfare, Resources, or Capabilities?’,
Philosophy and Phenomenological Research 50: 273–96
Daniels, Norman, Susannah Rose, and Ellen Daniels Zide (2009) ‘Disability, Adaptation,
and Inclusion’, in Kimberley Brownlee and Adam Cureton (eds) Disability and
Disadvantage (Oxford: Oxford University Press), 54–85
Darwall, Stephen (1977) ‘Two Kinds of Respect’, Ethics 88(1): 36–49
Davidson, Joyce, and Michael Orsini (2013) Worlds of Autism: Across the Spectrum of
Neurological Difference (Minneapolis, MN: University of Minnesota Press)
De Marneffe, Peter (2006) ‘Avoiding Paternalism’, Philosophy and Public Affairs 34: 68–94
De Marneffe, Peter (2013) ‘Self-Sovereignty and Paternalism’, in Christian Coons and
Michael Weber (eds), Paternalism: Theory and Practice (Cambridge: Cambridge
University Press), 56–73
Decker, Julie (2015) The Invisible Orientation: An Introduction to Asexuality (New York:
Skyhorse Publishing)
Dembroff, Robin (2018) ‘Real Talk on the Metaphysics of Gender’, Philosophical Topics
46(2): 21–50
Deneulin, Séverine (2002) ‘Perfectionism, Paternalism and Liberalism in Sen and
Nussbaum’s Capability Approach’, Review of Political Economy 14: 497–518
Dolan, Paul, and Daniel Kahneman (2008) ‘Interpretations of Utility and their Implications
for the Valuation of Health’, Economic Journal 118(525): 215–34
Dotson, Kristie (2012) ‘A Cautionary Tale: On Limiting Epistemic Oppression’, Frontiers:
A Journal of Women Studies 33(1): 24–47
Dougherty, Tom (2014) ‘Ethics Discussion at PEA Soup: Elizabeth Barnes’ “Valuing
Disability, Causing Disability” ’, PEA Soup. Available at: https://peasoup.typepad.com/
peasoup/2014/11/ethics-discussion-at-pea-soup-elizabeth-barnes-valuing-disability-
causing-disability-with-critical-p.html
Duffy, Simon (2013) A Fair Society? How the Cuts Target Disabled People (Sheffield: Centre
for Welfare Reform). Available at: https://www.centreforwelfarereform.org/uploads/
attachment/354/a-fair-society.pdf (last accessed 17 June 2021).
256 

Dworkin, Gerald (2013) ‘Defining Paternalism’, in Christian Coons and Michael Weber (eds.),
Paternalism: Theory and Practice (Cambridge: Cambridge University Press), pp. 25–38
Dworkin, Gerald (2014) ‘Paternalism’, in Edward N. Zalta (ed.), The Stanford Encyclopedia of
Philosophy. Available at: http://plato.stanford.edu/archives/sum2014/entries/paternalism/
Dworkin, Ronald (2000) Sovereign Virtue (Cambridge, MA: Harvard University Press)
Dworkin, Ronald (2002) ‘Sovereign Virtue Revisited’, Ethics 113(1): 106–43
Dworkin, Ronald (2011) Justice for Hedgehogs (Cambridge, MA: Harvard University Press)
Elliott, Julian G., and Elena L. Grigorenko (2014) The Dyslexia Debate (Cambridge:
Cambridge University Press)
Elliott, Julian G., Rod Nicholson, and Andrew Davis (2016) Dyslexia: Developing the Debate
(London: Bloomsbury Academic)
Elster, Jon (1982) ‘Sour Grapes’, in Amartya Sen and Bernard Williams (eds),
Utilitarianism and Beyond (Cambridge: Cambridge University Press), 219–38
Elster, Jon (1987) Sour Grapes: Studies in the Subversion of Rationality (Cambridge:
Cambridge University Press)
Engel, George L. (1977) ‘The Need for a New Medical Model: A Challenge for Biomedicine’,
Science 196: 129–36
Enoch, David (2016) ‘What’s Wrong with Paternalism: Autonomy, Belief, and Action’,
Proceedings of the Aristotelian Society 116(1): 21–48
Fateh-Moghadam, Bijan, and Thomas Gutmann (2014) ‘Governing [through] Autonomy:
The Moral and Legal Limits of “Soft Paternalism” ’, Ethical Theory and Moral Practice
17(3): 383–97
Feinberg, Joel (1986) The Moral Limits of the Criminal Law, iii. Harm to Self (New York:
Oxford University Press)
First, Michael (2005) ‘Desire for Amputation of a Limb: Paraphilia, Psychosis, or a New
Type of Identity Disorder’, Psychological Medicine 35(6): 919–28
Fleurbaey, Marc (2002) ‘Development, Capabilities, and Freedom’, Studies in Comparative
International Development 37(2): 71–7
Fleurbaey, Marc (2006) ‘Capabilities, Functionings and Refined Functionings’, Journal of
Human Development 7: 299–310
Fleurbaey, Marc, and Didier Blanchet (2013) Beyond GDP: Measuring Welfare and
Assessing Sustainability (Oxford: Oxford University Press)
Formosa, Paul, and Catriona Mackenzie (2014) ‘Nussbaum, Kant, and the Capabilities
Approach to Dignity’, Ethical Theory and Moral Practice 17(5): 875–92
Fowler, Tim (2015) ‘In Defence of State Directed Enhancement’, Journal of Applied
Philosophy 32(1): 67–81
Fox, Carl (2019) ‘What’s Special about the Insult of Paternalism?’, Law and Philosophy 38:
313–34
Francis, Leslie (2018) ‘Understanding Disability Civil Rights Non-Categorically: The Minority
Body and the Americans with Disabilities Act’, Philosophical Studies 175: 1135–49
Fraser, Nancy (2000) ‘Rethinking Recognition’, New Left Review 3: 107–20
Fricker, Miranda (2007) Epistemic Injustice: Power and the Ethics of Knowing (Oxford:
Oxford University Press)
Gallagher, Paul (2019) ‘NHS Apologises to Autistic Girl for Keeping her Locked in Solitary
Confinement’, The I (17 October). Available at: https://inews.co.uk/news/health/nhs-
apology-bethany-girl-autism-solitary-confinement-351894 (accessed 5 November 2020)
Gay, Roxane (2017) Hunger: A Memoir of (my) Body (London: Corsair)
Gheaus, Anca (2013) ‘The Feasibility Constraint on the Concept of Justice’, Philosophical
Quarterly 63(252): 445–64
  257

Gibbs, Simon J., and Julian G. Elliott (2020) ‘The Dyslexia Debate: Life without the Label’,
Oxford Review of Education 46(4): 487–500
Gilabert, Pablo (2011) ‘Debate: Feasibility and Socialism’, Journal of Political Philosophy
19(1): 52–63
Gilabert, Pablo, and Holly Lawford-Smith (2012) ‘Political Feasibility: A Conceptual
Exploration’, Political Studies 60: 809–25
Giubilini, Alberto, and Sagar Sanyal (2015) ‘The Ethics of Human Enhancement’,
Philosophy Compass 104: 233–43
Glasby, Jon, Yanan Zhang, Matthew R. Bennet, and Patrick Hall (2021) ‘A Lost Decade?
A Renewed Case for Adult Social Care Reform in England’, Journal of Social Policy 50(2):
406–37
Goodin, Robert, and Jackson, Frank (2007) ‘Freedom from Fear’, Philosophy and Public
Affairs 35(3): 249–65
Goodwin, Tom (2012) ‘Why We Should Reject ‘Nudge’’, Politics 32(2): 85–92
Grandin, Temple, and Richard Panek (2013) The Autistic Brain: Thinking across the
Spectrum (New York: Houghton Mifflin Harcourt)
Gregory, Alex (2020) ‘Disability as Inability’, Journal of Ethics and Social Philosophy 18(1):
23–48
Griffin, James (1986) Well-Being: Its Meaning, Measurement, and Moral Importance
(Oxford: Oxford University Press)
Grill, Kalle (2007) ‘The Normative Core of Paternalism’, Res Publica 13: 441–58
Grill, Kalle (2009) ‘Liberalism, Altruism and Group Consent’, Public Health Ethics 2(2):
146–57
Grill, Kalle (2015) ‘Antipaternalism as a Filter on Reasons’, in Thomas Schramme (ed.),
New Perspectives on Paternalism and Health Care (Dordrecht: Springer), 47–63
Grill, Kalle (2018) ‘Paternalism by and towards Groups’, in Kalle Grill and Jason Hanna
(eds), The Routledge Handbook of the Philosophy of Paternalism (Abingdon: Routledge),
46–58
Groll, Daniel (2012) ‘Paternalism, Respect, and the Will’, Ethics 122: 692–720
Groll, Daniel (2018) ‘Paternalism and Rights’, in Kalle Grill and Jason Hanna (eds), The
Routledge Handbook of the Philosophy of Paternalism (Abingdon: Routledge), 119–38
Grüne-Yanoff, Till (2012) ‘Old Wine in New Casks: Libertarian Paternalism Still Violates
Liberal Principles’, Social Choice and Welfare 38: 635–45
Gutmann, Amy (1987) Democratic Education (Princeton: Princeton University Press)
Hanna, Jason (2012) ‘Paternalism and the Ill-Informed Agent’, Journal of Ethics 16: 421–39
Hanna, Jason (2018a) In our Best Interest: A Defense of Paternalism (Oxford: Oxford
University Press)
Hanna, Jason (2018b) ‘Hard and Soft Paternalism’, in Kalle Grill and Jason Hanna (eds),
The Routledge Handbook of the Philosophy of Paternalism (Oxford: Routledge), 24–34
Harris, John (2000) ‘Is there a Coherent Social Conception of Disability?’, Journal of
Medical Ethics 26: 95–100
Harris, John (2001) ‘One Principle and Three Fallacies of Disability Studies’, Journal of
Medical Ethics 27: 383–7
Haslanger, Sally (2000) ‘Gender and Race: (What) are they? (What) do we Want them to
Be?’, Nous 34(1): 31–55
Haslanger, Sally (2012) Resisting Reality: Social Construction and Social Critique (Oxford:
Oxford University Press)
Hawkins, Jennifer (2018) ‘Book Review: Elizabeth Barnes, The Minority Body: A Theory of
Disability’, Ethics 128(2): 462–7
258 

Hehir, Thomas (2002) ‘Eliminating Ableism in Education’, Harvard Educational Review

72(1): 1–32
Higashida, Naoki (2013) The Reason I Jump: One Boy’s Voice from the Silence of Autism
(London: Hodder & Stoughton)
Hoque, Kim, Nick Bacon, and Dave Parr (2014) ‘Employer Disability Practice in Britain:
Assessing the Impact of the Positive about Disabled People “Two Ticks” Symbol’, Work,
Employment, and Society 28(3): 430–51
Howard, Dana, and Sean Aas (2018) ‘On Valuing Impairment’, Philosophical Studies 175:
1113–33
Hull, Richard (1998) ‘Defining Disability: A Philosophical Approach’, Res Publica 4(2): 199–210
Husak, Douglas (2015) ‘Paternalism and Consent’, in Thomas Schramme (ed.), New
Perspectives on Paternalism and Health Care (Dordrecht: Springer), 67–86
Institute for Government (2019) ‘Performance Tracker: Adult Social Care’. Available at:
<https://www.instituteforgovernment.org.uk/publication/performance-tracker-2019/adult-
social-care> (last accessed 17 June 2021)
Institute for Government (2020) ‘Performance Tracker 2020’. Available at: https://www.
instituteforgovernment.org.uk/sites/default/files/publications/performance-tracker-2020.
pdf (last accessed 17 June 2021)
Jaggar, Alison (2005) ‘Saving Amina: Global Justice for Women and Intercultural
Dialogue’, Ethics and International Affairs 19: 55–75
Jaggar, Alison (2006) ‘Reasoning about Well-Being: Nussbaum’s Methods of Justifying the
Capabilities’, Journal of Political Philosophy 14: 301–22
Jenkins, Katherine (2016) ‘Amelioration and Inclusion: Gender Identity and the Concept of
Woman’, Ethics 126: 394–421
Kahane, Guy, and Julian Savulescu (2009) ‘The Welfarist Account of Disability’, in
Kimberley Brownlee and Adam Cureton (eds), Disability and Disadvantage (Oxford:
Oxford University Press), 14–53
Kahane, Guy, and Julian Savulescu (2012) ‘The Concept of Harm and the Significance of
Normality’, Journal of Applied Philosophy 29(4): 318–32
Kahane, Guy, and Julian Savulescu (2015) ‘Normal Human Variation: Refocussing the
Enhancement Debate’, Bioethics 29(2): 133–43
Kahane, Guy, and Julian Savulescu (2016) ‘Disability and Mere Difference’, Ethics 126:
774–88
Kahneman, Daniel (2000) ‘Experienced Utility and Objective Happiness: A Moment-Based
Approach’ in Daniel Kahneman and Amos Tversky (eds), Choices, Values, and Frames
(New York: Cambridge University Press), 187–208
Kahneman, Daniel (2011) Thinking Fast and Slow (London: Allen Lane)
Kahneman, Daniel, Paul Slovic, and Amos Tversky (eds) (1982) Judgement Under
Uncertainty: Heuristics and Biases (Cambridge: Cambridge University Press)
Kahneman, Daniel, and Robert Sugden (2005) ‘Experienced Utility as a Standard of Policy
Evaluation’, Environmental and Resource Economics 32(1): 161–81
Kapp, Steven K., Kristen Gillespie-Lynch, Lauren E. Sherman, and Ted Hutman (2012)
‘Deficit, Difference, or Both? Autism and Neurodiversity’, Developmental Psychology
49(1): 59–71
Kelly, Jamie (2013) ‘Libertarian Paternalism, Utilitarianism, and Justice’, in Christian
Coons and Michael Weber (eds), Paternalism: Theory and Practice (Cambridge:
Cambridge University Press), 216–30
Khader, Serene (2009) ‘Adaptive Preferences and Procedural Autonomy’, Journal of
Human Development and Capabilities 10(2): 169–87
  259

Khader, Serene (2011) Adaptive Preferences and Women’s Empowerment (Oxford: Oxford
University Press)
Khader, Serene (2012) ‘Must Theorising about Adaptive Preferences Deny Women’s
Agency?’, Journal of Applied Philosophy 29(4): 302–17
Khader, Serene (2013) ‘Identifying Adaptive Preferences in Practice: Lessons from
Postcolonial Feminisms’, Journal of Global Ethics 9(3): 311–27
Khader, Serene (2018) ‘Should the Capability Approach be Paternalistic?’, in Kalle Grill and
Jason Hanna (eds), The Routledge Handbook of the Philosophy of Paternalism
(Abingdon: Routledge), 206–19
Khader, Serene (2019) Decolonizing Universalism: A Transnational Feminist Ethic (Oxford:
Oxford University Press)
Kingma, Elselijn (2007) ‘What is it to be Healthy?’, Analysis 67(2): 128–33
Kittay, Eva Feder (2005) ‘At the Margins of Moral Personhood’, Ethics 116: 100–31
Kittay, Eva Feder (2011) ‘Forever Small: The Strange Case of Ashley X’, Hypatia 26: 610–31
Kuper, Hannah, Lena Morgon Banks, Tess Bright, Calum Davey, and Tom Shakespeare
(2020) ‘Disability-Inclusive COVID-19 Response: What it is, Why it is Important and
What we can Learn from the United Kingdom’s Response’ Wellcome Open Research 5:
79. DOI: 10.12688/wellcomeopenres.15833.1
Kymlicka, Will (2002) Contemporary Political Philosophy: An Introduction, 2nd edition
(Oxford: Oxford University Press)
Lawford-Smith, Holly (2013) ‘Understanding Political Feasibility’, Journal of Political
Philosophy 21(3): 243–59
Lazenby, Hugh (2016) ‘Mistakes and the Continuity Test’, Politics, Philosophy, and
Economics 15(2): 190–205
Le Grand, Julian, and Bill New (2015) Government Paternalism: Nanny State or Helpful
Friend? (Princeton: Princeton University Press)
Lippert-Rasmussen, Kasper (2015) ‘Luck Egalitarianism versus Relational Egalitarianism:
On the Prospects of a Pluralist Account of Egalitarian Justice’, Canadian Journal of
Philosophy 45(2): 220–41
Lippert-Rasmussen, Kasper (2016) Luck Egalitarianism (London: Bloomsbury)
MacCallum, Gerald (1967) ‘Negative and Positive Freedom’, Philosophical Review 76(3):
312–34
Mackenzie, Catriona, and Jackie Leach Scully (2008) ‘Moral Imagination, Disability, and
Embodiment’, Journal of Applied Philosophy 24(4): 335–31
Matthes, Erich Hatala (2016) ‘Cultural Appropriation without Cultural Essentialism’, Social
Theory and Practice 42(2): 343–66
McBryde Johnson, Harriet (2003) ‘Unspeakable Conversations’, New York Times Magazine,
16 February. Available at: <http://www.nytimes.com/2003/02/16/magazine/unspeakable-
conversations.html>
McMahan, Jeff (2005) ‘Causing Disabled People to Exist and Causing People to Be
Disabled’, Ethics 116: 77–99
Mill, John Stuart (1974 (1859)) On Liberty, ed. Gertrude Himmelfarb (London: Penguin)
Mills, Charles (1998) ‘ “But What are you Really?”: The Metaphysics of Race’, in Blackness
Visible: Essays on Philosophy and Race (Ithaca, NY: Cornell University Press), 41–66
Mills, Chris (2013) ‘Why Nudges Matter: A Reply to Goodwin’, Politics 33: 28–36.
Mills, Chris (2015) ‘The Heteronomy of Choice Architecture’, Review of Philosophy and
Psychology 6: 495–509
Mitchell, Gregory (2005) ‘Libertarian Paternalism is an Oxymoron’, Northwestern
University Law Review 99: 1245–77
260 

Mitchell, Polly (2018) ‘Adaptive Preferences, Adapted Preferences’, Mind 127(508):

1003–25
Mitra, Sophie (2017) Disability, Health, and Human Development (New York: Palgrave
Macmillan)
Mitra, Sophie, Aleksandra Posarac, and Brandon C. Vick (2011) ‘Disability and
Poverty in Developing Countries: A Snapshot from the World Health Survey’, World
Bank Social Protection Working Paper 1109. Available at: https://ssrn.com/abstract=
1908128
Moller, Dan (2011) ‘Wealth, Disability, and Happiness’, Philosophy and Public Affairs
39(2): 177–206
Morris, Jenny (2011) Rethinking Disability Policy (York: Joseph Rowntree Foundation).
Available at: <https://www.jrf.org.uk/sites/default/files/jrf/migrated/files/disability-policy-
equality-summary.pdf> (last accessed 17 June 2021)
Mosbergen, Dominique (2013) ‘Battling Asexual Discrimination, Sexual Violence and
“Corrective” Rape’, The Huffington Post. Available at: http://www.huffingtonpost.com/
2013/06/20/asexual-discrimination_n_3380551.html?1371733068
Müller, Sabine (2009) ‘Body Integrity Identity Disorder (BIID): Is the Amputation of
Healthy Limbs Ethically Justified?’, American Journal of Bioethics 9(1): 36–43
Mullin, Amy (2014) ‘Children, Paternalism and the Development of Autonomy’, Ethical
Theory and Moral Practice 17(3): 413–26
Napoli, Donna Jo, et al. (2015) ‘Should All Deaf Children Learn Sign Language?’, Pediatrics
136(1): 170–6
Narayan, Uma (2002) ‘Minds of their own: Choices, Autonomy, Cultural Practices, and
Other Women’, in Louise Antony and Charlotte Witt (eds), A Mind of one’s own
(Boulder, CO: Westview Press), 418–33
Nelson, Eric (2008) ‘From Primary Goods to Capabilities: Distributive Justice and the
Problem of Neutrality’, Political Theory 36: 93–122
Newell, Christopher (1999) ‘The Social Nature of Disability, Disease and Genetics:
A Response to Gillam, Persson, Holtug, Draper and Chadwick’, Journal of Medical
Ethics 25: 172–5
Noll, Sarah, and Erich Kasten (2014) ‘Body Integrity Identity Disorder (BIID): How
Satisfied are Successful Wannabes’, Psychology and Behavioural Sciences 3(6): 222–32
Nussbaum, Martha C. (1992) ‘Human Functioning and Social Justice: In Defence of
Aristotelian Essentialism’, Political Theory 20(2): 202–46
Nussbaum, Martha C. (1993) ‘Non-Relative Virtues: An Aristotelian Approach’, in Martha
C. Nussbaum and Amartya Sen (eds), Quality of Life (Oxford: Oxford University Press),
242–69
Nussbaum, Martha C. (1999) Sex and Social Justice (Oxford: Oxford University Press)
Nussbaum, Martha C. (2000a) Women and Human Development: The Capabilities
Approach (Cambridge: Cambridge University Press)
Nussbaum, Martha C. (2000b) ‘Aristotle, Politics and Human Capabilities: A Response to
Antony, Arneson, Charlesworth, and Mulgan’, Ethics 111 (2000): 102–40
Nussbaum, Martha C. (2001) ‘Symposium on Amartya Sen’s Philosophy: 5 Adaptive
Preferences and Women’s Options’, Economics and Philosophy 17(1): 67–88
Nussbaum, Martha C. (2002a) ‘Capabilities and Disabilities: Justice for Mentally Disabled
Citizens’, Philosophical Topics 30(2): 133–65
Nussbaum, Martha C. (2002b) ‘Capabilities and Human Rights’, in Pablo De Greiff and
Ciaran Cronin (eds), Global Justice and Transnational Politics (Cambridge, MA: MIT
Press), 117–50
  261

Nussbaum, Martha C. (2004) ‘On Hearing Women’s Voices: A Reply to Susan Okin’,
Philosophy and Public Affairs 32(2): 193–205
Nussbaum, Martha C. (2006a) Frontiers of Justice: Disability, Nationality, and Species
Membership (Cambridge, MA: Harvard University Press)
Nussbaum, Martha C. (2006b) ‘Capabilities as Fundamental Entitlements: Sen and Social
Justice’, in Alexander Kaufman (ed.), Capabilities Equality: Basic Issues and Problems
(New York: Routledge), 44–70
Nussbaum, Martha C. (2008) ‘Human Dignity and Political Entitlements’, in Edmund
D. Pellegrino, Adam Schulman, and Thomas W. Merrill (eds), Human Dignity and
Bioethics (Notre Dame, IN: Notre Dame University Press), 351–80
Nussbaum, Martha C. (2011a) Creating Capabilities: The Human Development Approach
(Cambridge, MA: Harvard University Press)
Nussbaum, Martha C. (2011b) ‘Perfectionist Liberalism and Political Liberalism’,
Philosophy and Public Affairs 39: 3–45
Oliver, Michael (1996) Understanding Disability: From Theory to Practice (Basingstoke:
Palgrave)
Olsaretti, Serena (1998) ‘Freedom, Force and Choice: Against the Rights-Based Definition
of Voluntariness’, Journal of Political Philosophy 6(1): 53–78
Olsaretti, Serena (2005) ‘Endorsement and Freedom in Amartya Sen’s Capability
Approach’, Economics and Philosophy 21(1): 89–108
Parfit, Derek (1984) Reasons and Persons (Oxford: Oxford University Press)
Parkin, Simon (2018) ‘Finally, a Cure for Insomnia?’, The Guardian, 14 September 2018.
Available at: <https://www.theguardian.com/news/2018/sep/14/finally-a-cure-for-insom
nia> (last accessed 10 December 2020).
Parry, Jonathan (2017) ‘Defensive Harm, Consent, and Intervention’, Philosophy and Public
Affairs 45(4): 356–96
Perry, Stephen (2009) ‘Risk, Harm, Interests, and Rights’, in Tim Lewens (ed.), Risk:
Philosophical Perspectives (Abingdon: Routledge), 190–209
Pettit, Philip (1997) Republicanism: A Theory of Freedom and Government (Oxford: Oxford
University Press)
Pettit, Philip (2007) ‘Capability and Freedom: A Defence of Sen’, Economics and Philosophy
17: 1–20
Pettit, Philip (2010) ‘Freedom in the Spirit of Sen’, in C. W. Morris (ed.), Amartya Sen
(Cambridge: Cambridge University Press), 91–114
Piovarchy, Adam (2021) ‘Responsibility for Testimonial Injustice’, Philosophical Studies
178: 597–615
Pogge, Thomas (2002) ‘Can the Capability Approach be Justified?’, Philosophical Topics 30:
167–228
Pray, Janet L., and I. King Jordan (2010) ‘The Deaf Community and Culture at a
Crossroads: Issues and Challenges’, Journal of Social Work, Disability and
Rehabilitation 9(2–3): 168–93
Prince-Hughes, Dawn (ed.) (2002) Aquamarine Blue 5: Personal Stories of College Students
with Autism (Athens, OH: Swallow Press/Ohio University Press)
Qizilbash, Mozaffar (2007) ‘Well-Being, Adaptation and Human Limitations’, in S. Olsaretti
(ed.), Preferences and Well-Being (Cambridge: Cambridge University Press), 83–109
Quong, Jonathan (2011) Liberalism without Perfection (Oxford: Oxford University Press)
Rawls, John (1993) Political Liberalism, expanded edition (New York: Columbia University
Press)
Rawls, John (1999) A Theory of Justice (Cambridge, MA: Harvard University Press)
262 

Rawls, John (2001) Justice as Fairness: A Restatement (Cambridge, MA: Harvard University
Press)
Reynolds, Joel Michael (2017) ‘ “I’d Rather be Dead than Disabled”: The Ableist Conflation
and the Meanings of Disability’, Review of Communication 17(3): 149–63
Reynolds, Joel Michael (2018) ‘Toward a Critical Theory of Harm: Ableism, Normativity,
and Transability (BIID)’, APA Newsletter on Philosophy and Medicine 16(1): 37–46
Richardson, Henry (2000) ‘Some Limitations of Nussbaum’s Capabilities’, Quinnipiac Law
Review 19: 309–32
Riddle, Christopher A. (2013) ‘Defining Disability: Metaphysical Not Political’, Medicine,
Health Care, and Philosophy 16: 377–84
Riddle, Christopher A. (2020) ‘Why we do Not Need a “Stronger” Social Model of
Disability’, Disability and Society 35(9): 1509–13
Robeyns, Ingrid (2003) ‘Sen’s Capability Approach and Gender Inequality: Selecting
Relevant Capabilities’, Feminist Economics 9(2–3): 61–92
Robeyns, Ingrid (2006) ‘The Capability Approach in Practice’, Journal of Political
Philosophy 14(3): 351–76
Robeyns, Ingrid (2016) ‘Conceptualising Well-Being for Autistic Persons’, Journal of
Medical Ethics 42(6): 383–90
Robeyns, Ingrid (2017) Wellbeing, Freedom and Social Justice: The Capability Approach
Re-Examined (Cambridge: Open Book Publishers)
Robeyns, Ingrid (2005) ‘Selecting Capabilities for Quality of Life Measurement’, Social
Indicators Research 74(1): 191–215
Robeyns, Ingrid, and Morten Fibieger Byskov (2020) ‘The Capability Approach’, in Edward
N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Available at: <https://plato.
stanford.edu/entries/capability-approach>
Savulescu, Julian (2001) ‘Procreative Beneficence: Why we Should Select the Best Children’,
Bioethics 16(5/6): 413–26
Savulescu, Julian, and Guy Kahane (2009) ‘The Moral Obligation to Create Children with
the Best Chance of the Best Life’, Bioethics 23: 274–90
Scanlon, T. M. (1975) ‘Preference and Urgency’, ournal of Philosophy 72(19): 655–69
Scanlon, T. M. (1998) What We Owe to Each Other (Cambridge, MA: Harvard, Belknap
Press)
Schatschneider, Christopher, and Joseph K. Torgesen (2004) ‘Using our Current
Understanding of Dyslexia to Support Early Identification and Intervention’, Journal
of Child Neurology 19(10): 759–65
Schemmel, Christian (2011) ‘Distributive and Relational Equality’, Politics, Philosophy and
Economics, 11(2): 123–48
Schramme, Thomas (2014) ‘Disability (Not) as a Harmful Condition: The Received View
Challenged’, in Jerome E. Bickenbach, Franziska Felder, and Barbara Schmitz (eds),
Disability and the Good Human Life (Cambridge: Cambridge University Press), 72–92
Sen, Amartya (1980) ‘Famines’, World Development 8(9): 613–21
Sen, Amartya (1981) ‘Ingredients of Famine Analysis: Availability and Entitlements’,
Quarterly Journal of Economics 96(3): 433–64
Sen, Amartya (1988) On Ethics and Economics (Malden, MA: Blackwell)
Sen, Amartya (1990) ‘Justice: Means versus Freedoms’, Philosophy and Public Affairs 19:
111–21
Sen, Amartya (1992) Inequality Reexamined (Oxford: Oxford University Press)
Sen, Amartya (1993) ‘Capability and Well-Being’, in Martha C. Nussbaum and Amartya
Sen (eds), Quality of Life (Oxford: Oxford University Press), 30–53
  263

Sen, Amartya (1999a) Development as Freedom (Oxford: Oxford University Press)

Sen, Amartya (1999b) Commodities and Capabilities (Oxford: Oxford University Press)
Sen, Amartya (2001) ‘Reply’, Economics and Philosophy ?: 51–66
Sen, Amartya (2004) ‘Capabilities, Lists, and Public Reason: Continuing the Conversation’,
Feminist Economics 10(3): 77–80
Sen, Amartya (2009) The Idea of Justice (London: Penguin)
Shafer-Landau, Russ (2005) ‘Liberalism and Paternalism’, Legal Theory 11: 169–91
Shakespeare, Tom (1994) ‘Cultural Representations of Disabled People: Dustbins for
Disavowal?’, Disability and Society 9(3): 283–99
Shakespeare, Tom (2014a) Disability Rights and Wrongs Revisited (New York: Routledge)
Shakespeare, Tom (2014b) ‘Nasty, Brutish, and Short? On the Predicament of Disability
and Embodiment’, in Jerome E. Bickenbach, Franziska Felder, and Barbara Schmitz
(eds), Disability and the Good Human Life (Cambridge: Cambridge University Press),
93–112
Shakespeare, Tom, and Nicholas Watson (2001) ‘The Social Model of Disability: An
Outdated Ideology?’, Research in Social Science and Disability 2: 9–28
Shakespeare, Tom, Nicholas Watson, Richard Brunner, Jane Cullingworth, Shaffa Hameed,
Nathaniel Scherer, Charlotte Pearson, and Veronika Reichenberger (2021) ‘Disabled
People in Britain and the Impact of the COVID-19 Pandemic’, Preprints. Available at:
<https://www.preprints.org/manuscript/202101.0563/v1>
Shelby, Tommie (2002) ‘Foundations of Black Solidarity: Collective Identity or Common
Oppression?’, Ethics 112: 231–66
Shepherd, Joshua (2020) ‘Disabilities and Well-Being: The Bad and the Neutral’, in David
Wasserman and Adam Cureton (eds), The Oxford Handbook of Philosophy and
Disability (Oxford: Oxford University Press), 123–37
Shew, Ashley (2017) ‘The Long Way around’, The Chronicle of Higher Education, 17
September. Available at: <http://www.chronicle.com/article/It-Shouldn-t-Be-This-Hard-
to/241166>
Shiffrin, Seana (2000) ‘Paternalism, Unconscionability Doctrine, and Accommodation’,
Philosophy and Public Affairs 29: 205–50
Silvers, Anita (2003) ‘On the Possibility and Desirability of Constructing a Neutral
Conception of Disability’, Theoretical Medicine 24: 471–87
Silvers, Anita (2009) ‘No Talent? Beyond the Worst Off! A Diverse Theory of Justice for
Disability’, in Kimberley Brownlee and Adam Cureton (eds), Disability and
Disadvantage (Oxford: Oxford University Press), 163–99
Sinclair, Jim (2012) ‘Don’t Mourn for us’, Autonomy, the Critical Journal of
Interdisciplinary Autism Studies 1(1): 1–4
Singer, Peter (1972) ‘Famine, Affluence, and Morality’, Philosophy and Public Affairs 1(3):
229–43
Singer, Peter (2005) ‘Ethics and Disability: A Response to Koch’, Journal of Disability Policy
Studies 16(2): 130–3
Singer, Peter (2009) The Life you Can Save: How to Play your Part in Ending World Poverty
(Oxford: Pan Macmillan)
Smith, Steven R. (2001) ‘The Social Construction of Talent: A Defence of Justice as
Reciprocity’, Journal of Political Philosophy 9(1): 19–37
Sparrow, Robert (2005) ‘Defending Deaf Culture: The Case of Cochlear Implants’, Journal
of Political Philosophy 13(2): 135–52
Sparrow, Robert (2010) ‘Implants and Ethnocide: Learning from the Cochlear Implant
Controversy’, Disability and Society 25(4): 455–66
264 

Stanton, Mike (2006) ‘What is Neurodiversity?’, Action for Autism. Available at: <http://
mikestanton.wordpress.com/my-autism-pages/what-is-neurodiversity>
Stemplowska, Zofia, and Adam Swift (2012) ‘Ideal and Non-Ideal Theory’, in David
Estlund (ed.), The Oxford Handbook of Political Philosophy (Oxford: Oxford
University Press), 373–89
Stoner, Ian (2016) ‘Ways to be Worse Off ’, Res Philosophica 93(4): 921–49
Stramondo, Joseph, (2017) ‘Disabled by Design: Justifying and Limiting Parental Authority
to Choose Future Children with Pre-Implantation Genetic Diagnosis’, Kennedy Institute
of Ethics Journal 27(4): 475–500
Stramondo, Joseph A., and Stephen M. Campbell (2020) ‘Causing Disability, Causing Non-
Disability: What’s the Moral Difference?’, in Adam Cureton and David Wasserman
(eds), The Oxford Handbook of Philosophy and Disability (Oxford: Oxford University
Press), 138–57
Sumner, Wayne (1996) Welfare, Happiness, and Ethics (Oxford: Oxford University Press)
Sunstein, Cass R. (2014) Why Nudge? The Politics of Libertarian Paternalism (New Haven:
Yale University Press)
Sunstein, Cass R., and Richard H. Thaler (2003) ‘Libertarian Paternalism is Not an
Oxymoron’, University of Chicago Law Review 70: 1159–1202
Taylor, Ashley (2012) ‘Addressing Ableism in Schooling and Society? The Capabilities
Approach and Students with Disabilities’, Philosophy of Education, 113–21
Taylor, Charles (1979) ‘What’s Wrong with Negative Liberty?’ in Alan Ryan (ed.), The Idea
of Freedom: Essays in Honour of Isiah Berlin (Oxford: Oxford University Press)
Taylor, Daniel J., and Kristi E. Pruiksma (2014) ‘Cognitive and Behavioural Therapy for
Insomnia (CBT-I) in Psychiatric Populations: A Systematic Review’, International
Review of Psychiatry 26(2): 205–13
Terlazzo, Rosa (2014) ‘The Perfectionism of Nussbaum’s Adaptive Preferences’, Journal of
Global Ethics 10(2): 183–98
Terlazzo, Rosa (2016) ‘Conceptualizing Adaptive Preferences Respectfully: An Indirectly
Substantive Account’, Journal of Political Philosophy 24(2): 206–26
Terlazzo, Rosa (2019) ‘How Politically Liberal Should the Capabilities Approach Want to
Be?’, Politics, Philosophy and Economics 18(3): 282–304
Terzi, Lorella (2004) ‘The Social Model of Disability: A Philosophical Critique’, Journal of
Applied Philosophy 21(2): 141–57
Terzi, Lorella (2008) Justice and Equality in Education: A Capability Perspective on
Disability and Special Educational Needs (London: Continuum)
Terzi, Lorella (2009) ‘Vagaries of the Natural Lottery? Human Diversity, Disability, and
Justice: A Capability Perspective’, in Kimberley Brownlee and Adam Cureton (eds),
Disability and Disadvantage (Oxford: Oxford University Press), 86–111
Thaler, Richard H., and Cass R. Sunstein (2008) Nudge: Improving Decisions about Health,
Wealth and Happiness (London: Penguin)
Tremain, Shelley (2001) ‘On the Government of Disability’, Social Theory and Practice
27(4): 617–36
Tremain, Shelley (2002) ‘On the Subject of Impairment’, in Marian Corker and Tom
Shakespeare (eds), Disability/Postmodernity: Embodying Disability Theory (London:
Continuum), 32–47
Tsai, George (2014) ‘Rational Persuasion as Paternalism’, Philosophy and Public Affairs 42:
78–112
Ubel, Peter A., George Loewenstein, Norbert Schwarz, and Dylan Smith (2005)
‘Misimagining the Unimaginable: The Disability Paradox and Health Care Decision
Making’ Health Psychology 24(4, suppl.): S57–S62
  265

Vallentyne, Peter (2005) ‘Debate: Capabilities versus Opportunities for Well-Being’, Journal
of Political Philosophy 13(3): 359–71
Vallentyne, Peter (2006) ‘Capability versus Opportunity for Well-Being’, in Alexander
Kaufman (ed.), Capabilities Equality: Basic Issues and Problems (New York: Routledge),
79–92
van der Deijl, Willem (2017) ‘Which Problem of Adaptation?’, Utilitas 29(4): 474–92
van der Deijl, Willem (2020) ‘A Challenge for Capability Measures of Wellbeing’, Social
Theory and Practice 46(3): 605–31
Vehmas, Simo (2004) ‘Ethical Analysis of the Concept of Disability’, Mental Retardation
42(3): 209–22.
Vehmas, Simo, and Christopher A. Riddle (2020) ‘What Can Philosophy Tell us about
Disability?’, in Nick Watson and Simo Vehmas (eds), Routledge Handbook of Disability
Studies (Oxford: Routledge), 348–61
Vehmas, Simo, and Nick Watson (2014) ‘Moral Wrongs, Disadvantages, and Disability:
A Critique of Critical Disability Studies’, Disability and Society 29(4): 638–50
Voigt, Kristin (2012) ‘Incentives, Health Promotion and Equality’, Health Economics,
Policy, and Law 7: 263–83
Voigt, Kristin (2015) ‘Paternalism and Equality’, in Thomas Schramme (ed.), New
Perspectives on Paternalism and Health Care (Dordrecht: Springer), 87–100
Wahlert, Lance, and Sabrina Gill (2017) ‘Pathological, Disabled, Transgender: The Ethics,
History, Laws, and Contradictions in Models that Best Serve Transgender Rights’,
Kennedy Institute of Ethics Journal 27(2): 249–66
Wanderer, Jeremy (2012) ‘Addressing Testimonial Injustice: Being Ignored and Being
Rejected’, Philosophical Quarterly 62(246): 148–69
Wasserman, David (2006) ‘Disability, Capability, and Thresholds for Distributive Justice’,
in Alexander Kaufman (ed.), Capabilities Equality: Basic Issues and Problems (New York:
Routledge), 214–34
Wasserman, David (2009) ‘Ethical Constraints on Allowing or Causing the Existence of
People with Disabilities’, in Kimberley Brownlee and Adam Cureton (eds), Disability and
Disadvantage (Oxford: Oxford University Press), 319–51
Wasserman, David (2018) ‘Book Review: The Minority Body: A Theory of Disability’,
Philosophical Review 127(2): 251–6
Wasserman, David, and Adrienne Asch (2014) ‘Understanding the Relationship between
Disability and Well-Being’, in Jerome E. Bickenbach, Franziska Felder, and Barbara Schmitz
(eds), Disability and the Good Human Life (Cambridge: Cambridge University Press), 139–67
Wasserman, David, Adrienne Asch, Jeffrey Blustein, and Daniel Putnam (2016a)
‘Disability: Definitions, Models, Experience’, in E.N. Zalta (ed.), Stanford
Encyclopaedia of Philosophy. Available at: <https://plato.stanford.edu/entries/disability>
Wasserman, David, Adrienne Asch, Jeffrey Blustein, and Daniel Putnam (2016b)
‘Disability: Health, Well-Being, and Personal Relationships’, in E.N. Zalta (ed.),
Stanford Encyclopaedia of Philosophy. Available at: <https://plato.stanford.edu/
entries/disability-health>
Watson, Nick (2002) ‘Well, I Know this is Going to Sound Very Strange to you, But I don’t
See myself as a Disabled Person: Identity and Disability’, Disability and Society 17(5):
509–27
West, Lindy (2016) Shrill: Notes from a Loud Woman (London: Quercus)
Westhale, July (2015) ‘The Troubling Trendiness of Poverty Appropriation’, The
Establishment. Available at: <https://theestablishment.co/the-troubling-trendiness-of-
poverty-appropriation-4d3681406320> (accessed 23 April 2018)
Wilkinson, T. M. (2013) ‘Nudging and Manipulation’, Political Studies 61: 341–55
266 

Williams, Andrew (2002) ‘Dworkin on Capability’, Ethics 113(1): 23–39

Williams, Donna (1999) Nobody Nowhere: The Remarkable Autobiography of an Autistic
Girl (London: Jessica Kingsley Publishers)
Wise, Jacqui (2019) ‘NHS and Vertex Remain Deadlocked over Price of Cystic Fibrosis
Drug’, The BMJ 364.
Wolff, Jonathan (2009a) ‘Disability among Equals’, in Kimberley Brownlee and Adam
Cureton (eds), Disability and Disadvantage, (Oxford: Oxford University Press), 112–37
Wolff, Jonathan (2009b) ‘Disability, Status Enhancement, Personal Enhancement, and
Resource Allocation’, Economics and Philosophy 25: 49–68
Wolff, Jonathan (2009c) ‘Disability, Risk, and the Social Determinants of Health’, Public
Health Ethics 2: 214–23
Wolff, Jonathan, and Avner de-Shalit (2007) Disadvantage (Oxford: Oxford University
Press)
World Health Organization (2002) Towards a Common Language for Functioning,
Disability and Health: ICF, The International Classification of Functioning, Disability
and Health (Geneva: World Health Organisation). Available at: <http://www.who.int/
classifications/icf/icfbeginnersguide.pdf?ua=1> (accessed 7 November 2017)
World Health Organization (2017) ‘Health Topics: Disability’ (Geneva: World Health
Organisation). Available at: <http://www.who.int/topics/disabilities/en/> (accessed 7
November 2017)
Youngs, Ian (2019) ‘Meet the Artist Who Designed a Hotel Room that’s Difficult to Stay in’,
BBC News, 17 December. Available at: <https://www.bbc.co.uk/news/entertainment-
arts-50670333> (accessed 5 November 2020)
Zimmerman, David (2003) ‘Sour Grapes, Self-Abnegation and Character Building’, The
Monist 86: 220–41
 Index

For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on
occasion, appear on only one of those pages.

Aas, Sean 38, 141–2 Mitchell 170

ableism 15, 31, 120–1, 155–6, 187–8, 194, 231 Nussbaum 171–2, 181–2, 181n43, 184–5
abuse Rawls 184
adaptive preferences 165–6, 177 reflective equilibrium 184–6
bad functionings 75–6 Sen 169–71, 174n26
and entitlement to perform substantive approach 168, 181–4
functions 76–7 substantive proceduralist
and obese individuals 63n44, 64 approach 154–5, 182–4, 194
partner abuse examples 75–6, 78, 124, Terlazzo 171, 182–4, 186
165–6, 177, 234 as unreliable 176–8, 182, 184
and wheelchair users 16 vagueness 167–8
WHO example 22 value distortion 193, 195–7
achondroplasia 62 well-being adaptive preferences 168,
adaptive preferences. see also justice 172–81, 187–8, 214
adaptive preferences Wolff 185
Arneson 184 aesthetic enjoyment
Barnes 166, 177–8, 181–2 adaptive preferences 179–81
benign adaptation 170 blindness 46–7
Blanchet 176 justice adaptive preferences 191–2
borderline cases 175–6 justice-based account of disability 44–5
Bovens 173 music appreciation 161–2
Cohen 184 aesthetic experiences
Colburn 173–4, 176–8, 191 deaf individuals 46–7, 59, 179, 191–2
core preferences 183 altruism 122, 213n42
covert influences 174–5, 187–8, 191, 214 amputation of a healthy limb. see body
critical reflection 183–4 integrity identity disorder (BIID)
disability paradox 165n2 Amundson, Ron 149–50
divergent views 165–8 Anderson, Elizabeth 56–7, 80–1, 81n29, 85
Dworkin 184 antidiscrimination legislation 22
Elster 168–9, 171–3, 176–7 anti-paternalism 202–11
Fleurbacy 176 Arneson’s objection 214n43
formation 168–72, 176 authority approach 206–7, 209, 211
‘Fox and Grapes’ parable 168–73, 179, cochlear implants 211–13, 213n42, 232
187–8, 192–3 and consent 202, 213
insulting implications 196–8 covert influence 214
justice proceduralist accounts 168, cures 211–12
184–92, 214 Grill 202, 207–8, 213n42
Keller 194–5 and justice theory 215–20
Khader 171, 182–4, 186 and liberty 202
misapplication of the concept 165–7 Mill’s rotten bridge 214
268 
anti-paternalism (cont.)
mistakes 211, 214–15
state assistance 75n16, 200–1, 211–15
will as decisive 207
Arneson, Richard 118–22, 184, 214n43,
222, 222n58
arthritis. see pain conditions
Asch, Adrienne 111
asexuality. see also sexuality
as deficiency 105n36
exclusion example 32–3
functioning of sexual satisfaction 101–4,
107, 110, 218
rejection of the benefit of sexual
satisfaction 217
atheism 105
authority approach 8–9, 200–2,
206–7, 209
Autistic Spectrum Conditions (ASC)
adaptive preferences 194
amusement park example 80–1
capabilities for control 102–5
covert influences 174–5
deficiency vs diversity 141–2
defining disability 11–12
dysfunction vs difference 142–3
exclusion from education 66
functioning better trade-off 146–7
and impairment 110, 130, 132–3, 138–42
as not disabling 62, 128, 145
social interaction 50–1, 146–7,
174–5, 217
solitary confinement example 51
as transient disorder 152
autonomy
ability to express agency 85–6
adaptive preferences 165–70, 173–4,
176–7, 183–4, 189–90
capabilities for control 93, 107–9,
112–13, 117–18, 121, 123–4
Colburn’s account 173, 233–4
disabled as autonomous agents 8–9
forming social relationships 235–6
maximizing one’s welfare 214–15
non-autonomous character
planning 168–9, 173, 187, 193
and paternalism 201
state assistance 219, 222–3, 222n58,
232–6
Barclay, Linda 18, 81n29
Barnes, Elizabeth
on adaptive preferences 166, 181–2
causing and removing
impairment 237–8
concept of negative impairment 151–2
deviation from normal functioning 140
disability and impairment 12
discordance view 177–8
mere-difference view 155–6
objections to the concept of
impairment 132–4
social constructionist approach 32–5, 241
value of distinguishing difference to
disability 134–5
Bickenbach, Jerome 138–9, 141–2
BIID (body integrity identity disorder). see
body integrity identity disorder (BIID)
Blanchet, Didier 176
blindness
aesthetic experiences 46–7
Braille 5, 108–9, 121–2, 231
classification as disability 111
context 19
control 108–9
dysfunction 141–2
‘engineering away,’ 104n34, 238
entitlements 47–8
impairments 12, 18
and natural reference class 139–40
as not disabling 62
opportunities 46–7, 116
rejection of the benefit of all five
senses 217
sensory compensation 160–1
sensory experiences 5, 46–7
treatments 108–9
board games example 223
bodily integrity
capabilities for control 50, 101–2, 114,
124, 125nn82, 84, 185–6
rape example 124
in substantive perfectionist
approach 182
body dysmorphia 144–5
body integrity identity disorder
(BIID) 33–4, 62–3, 114–16, 144–5
Boorse, Christopher 139–40
Bovens, Luc 173
  269

Browne, Kimberly 243–4 security 95–6

Brunning, Luke 101n25 Sen 91–2, 94–5, 117–18, 122n77
Buchanan, Allen 24–7 set-consistency 97–9
on social institutions and
Campbell, Stephen M. 35n85, 157–8 impairments 104n34
cancer examples 85, 103–4 Vallentyne 120–1
capabilities for control value of capabilities 91, 97–9, 118
Arneson 118–21 ‘wasted opportunities,’ 120–1
autonomy 93, 107–9, 112–13, 117–18, Wolff 96n13, 97, 124–5
121, 123–4 capability approach 69–70. see also
basic capabilities 92 mobility
bodily integrity 50, 101–2, 114, 124, accommodate the importance of
125nn82, 84, 185–6 relationships 244
Brunning 101n25 Anderson 56–7, 80–1, 85
capability for bodily health 106–7 anti-paternalist state 75n16
capability sets 97–8 Browne 243–4
Carter 114–16 capabilities compared to
central domains 91, 95–6 functionings 70, 72–5, 215–16, 226
Cohen 94, 117 capability equality 85–6
content-independent freedom 95 capability theory of justice 70–2
dealing with limited resources 122 Claasen 85
de-Shalit 96n13, 97, 124–5 Cohen 43–4, 49–50
in domains of choice 90, 218–19 compared to resourcist approach 70,
elimination of impairment 108 81–6, 88–9
fertile functionings 96, 125n83 discrimination 74–5, 79–80, 81n29
Fleurbacy 117 and distributive justice 69–70
formal freedom 196 Dworkin’s resourcist approach 84n38,
impairment treatment as 87–8
offensive 105–6, 108–9, 218 educational provisions 57–8
issues with capabilities to function 91, functioning 77
99–106 functioning achievement 75
Johnson 112 functionings 76–8
judging capabilities 91, 123–6 functionings (doings and beings) 70–1
Keller 110–12 Gay 63–4
Kittay 104n34 Gilabert 49–50
McKeever 101n25 Johnson 56
Nussbaum 92–3, 99–103, 104n34, 105, negative capabilities 78
109–10, 113–14, 118 negative capabilities (lack of access) 77
overlapping consensus 99–101, 103, 106 negative functionings 78–9
over/under-emphasis on 91, 96–7 Nussbaum 71, 85, 184
Pettit 94–5 opportunities vs achievements 73
problem with capabilities to as overly demanding 84–6
function 94–5 and paternalism 216
real freedoms to pursue Pogge 81n29, 82, 84–6
functionings 91–9 primary goods as resources 86–7
risk and real freedom 95–6 Rawls’s primary goods 86–7, 87n44
Robeyns 96–7, 102n29 relevant functionings 71–2
robustness 96–7 resourcism 86–8
scarcity 122–5 Robeyns 71–2
270 
capability approach (cont.)
Sen 76–7, 84–5
Silvers 57–8
Stanford Encyclopaedia entry 216n47
and the state 72
trivial functionings 75–6, 78–80
Vallentyne 76, 78
valuable capabilities 70, 75–81, 215–16
Wolff 55–6
capability to function approach 216–17
careers. see also employment
adaptive preferences 172–3, 180–1
ballet dancer example 79
capabilities 79, 98–9, 120–1
dissuaded science teacher example 173,
175, 190
engineering example 82
mobility 21
opportunities 119–20, 159
park ranger example 21
singer example 79
carpentry example 223–5
Carter, Ian 114–16
celibacy examples 100–1, 107–8, 118–19
chronic pain. see pain conditions
cigarettes, paternalist examples 221–3
Claasen, Rutger 85
cochlear implants
consultation process 229–30
as contingent impairment 157–8
curing the dysfunction 138–9, 232
d/Deaf individuals 8–9, 59, 108–9,
225–30
entitlements 59, 232
as genocide 104–5, 227–8
opportunity costs 225–6
state assistance and anti-paternalism
211–13, 213n42, 232
as a state endorsed value 227
Cohen, G. A. 43–4, 49–50, 94, 117, 184
Colburn, Ben 173–4, 176–8, 191,
233–4
conversion therapy 214–15, 235–6
cures. see also cochlear implants
Parry 213n42
paternalism 226, 236–8
policy development 230–1
state assistance 211–12, 236–8
cystic fibrosis example 139, 156
Davis, Alison 15–16
Deaf community 104–5, 167, 228–30.
see also cochlear implants; hearing
impairment
deaf individuals. see also cochlear implants;
deafness
adaptive preferences 194
aesthetic experiences 46–7, 59, 191–2
cochlear implants 8–9, 59, 108–9,
225–30
communication 46–7, 58–9, 130
consultation 231
justice adaptive preferences 191–2
rejection of the benefit of all five
senses 217
state assistance 211, 213n42, 231–2
deafness. see also deaf individuals
capabilities 5
communication 163–4
as disabling 19
‘engineering away,’ 104n34
Harris 28n49
as impairment 11–12
as not a disability 41, 66–7
as not disabling 62
opportunity to be unhealthy 116
and well-being 33–4
De Marneffe, Peter 222
dementia examples 148–9, 207
depression 26–7, 62–3, 140–1, 151–2,
157–8, 177
de-Shalit, Avner 96n13, 97, 124–5,
184–5, 243
diabetes 63, 149–50
disability, defining
Aas 38
ameliorative approach 36–9, 241–2
the Ashley treatment 29
Barclay 18
Barnes 12, 17, 32–5
Buchanan 24–7
contextual factors 17–19, 21–4, 82
Davis 15–16
disability rights movement 13–14
due to just and unjust social
causes 19–20
elimination of impairments 13
external factors 21–4
Gregory 25n44, 26n46
  271

Haslanger 36, 38 restricts abilities 64

Howard 38 solidarity in indignation example 236–7
the hybrid model 20–4, 131 distributive justice theories 43–4, 230
and impairment 11–15, 34 domestic labour example 245–6
justice 29–30 Down’s syndrome examples
Kahane 28–30, 31n60, 33–4 capabilities 50
Kittay 29 defining disability 37–8
as neutral 35n85 education 66, 195–6, 246–7
Oliver 17n20 ‘engineering away,’ 104n34
performing tasks 10, 24–5, 39 forming relationships 242–8
public policy 30 and natural reference class 139–40
Rawls 30 as not disabling 62, 163–4
Riddle 37 personal factors 21
Savulescu 28–30, 31n60, 33–4 Duchenne muscular dystrophy
Shakespeare 18, 37 example 157
Smith 16 Dworkin, Ronald 84n38, 87–8, 184, 248
the social model 13–15, 17–20, 38–9 dyslexia
species norms and impairment 11, 24–7 distinguishing impairment from
terminology 165n2 disability 135, 145
Terzi 27, 28n49 the immigrant girl example 67, 151n47
Watson 37 institutional practices 22
welfarist approach 12, 28–31, 33–4, 36, reading on paper 18
39, 55–6 state funding 122
what is valuable or essential to life 26 as transient disorder 152
WHO 11n7, 12–13, 21–4
Wolff 12 economic deprivation 169
disability pride 6, 129–30, 143–4, 165 education
disability rights movement. see also accessible course materials 121–2
neurodiversity movement Anderson’s public schools vs
Barnes on defining disability 32–3 Amusement parks 80–1
conceptions of impairments 129–30 assumed to not be paternalistic 222–3
on consultation 230 Braille 5, 108–9, 121–2
defining disability 13–14 Down’s syndrome 66, 195–6, 246–7
distinguishing impairment and enrollment numbers 122–3
disability 131–2, 134–5 entitlements 120–3
indeterminacy 33 justice adaptive choices 189–90
over-inclusiveness of impairment 144–5 justice adaptive preferences 192–3
the social model 14 justice-based approach 57–9
discrimination. see also antidiscrimination paternalism 222–3
legislation taxation 200
Anderson 81n29 Elizabeth Bennett example 52–3
Barclay 81n29 Elster, Jon 168–9, 171–3, 176–7
capabilities for control 93 employment. see also careers; unemployed
capability approach 74–5, 79–80, person example; workaholic examples
81n29 ableist hiring practices 187–8
impairment 150–64, 238 autism 65–6
informal institutional practices 22, 82 capabilities 96, 110
justice-based account 50–1 Down’s syndrome 243
Pogge 81n29 functionings 70–1, 96
272 

employment (cont.) Gregory, Alex 25n44, 26n46

leprosy example 17, 40 Grill, Kalle 202, 207–8, 213n42
ME 67 Groll, Daniel 207–8
meaningful 93, 188–9 guide dogs example 122–3
opportunities 50–1, 119
empowerment 118n65, 119–20, 181 Haslanger, Sally 36, 38, 153n53
Enoch, David 204–5 hearing impairment 19–20, 57–9, 131.
epistemic paternalism 205n20, 206n25 see also Deaf community
hearing loss. see cochlear implants
fasting individual examples 73–4, 91–2, hedonism 43–4, 170, 174
100–1, 111, 186–7. see also heroin example 223–5
nourishment HIV-positive person example 150
fatigue examples holding property example 243
amusement parks 80–1 homosexuality examples. see sexuality
and disability status 67–8 Howard, Dana 38
multiple sclerosis 96–7 hydrocephaly example 149–50
performing functionings 26–7, 61, 97–8,
138–9 impaired vision. see blindness; vision
welfare deficits 43–4 impairments
feasibility impairment. see also blindness; capabilities
and the justice-based account 45–8 for control; disability, defining; hearing
the feasibility constraint on justice 48–50, impairment; vision impairments
53–4, 245, 248 Aas 141–2
Feinberg, Joel 214, 233 Amundson 149–50
female genital cutting (FGC) examples 114, as anomaly 129, 137–43, 231
170–2 ‘atypical functioning’ and race 152–3
feminism 32–3, 166 bad difference 130, 158–62
Fleurbacy, Marc 117, 176 Barnes 12, 132–5, 140, 151–2, 155–6
forming relationships Bickenbach 138–9, 141–2
and adaptive preferences 235–6 ‘big-W’ Worse-off 158
capabilities 83 Boorse’s uniform functional
Down’s syndrome examples 242–8 design 139–40
entitlements 41–2 Campbell 157–8
Fox, Carl 202n4, 209–10 complex difference 130, 151–2, 158–62
Francis, Leslie 151–2 definition 1–2, 10–13, 128–9
freedom of movement 50–1, 65, 86–7 discrimination 66, 150, 238
disembodied view of disability 133–6
Gay, Roxanne 63–4 distinguished from disability 129, 131–6
gender 21, 38 distinguishing disability from other
gender-based oppression 169 disadvantages 152–4
gender dysphoria 137 enhancements 147–9
gendered domestic labour example 245–6 entitlement to control 158–61
gender identity 6, 27, 31, 132–3, 143–4, Francis 151–2
152–3, 153–4 functioning better than normal 146–9
genetic diagnosis. see pre-implantation Haslanger 153n53
genetic diagnosis (PGD) hidden pathologies 137, 143–4, 149–51
genetic enhancement example 148 Kahane on enhancements 147–8
Gheaus, Anca 49n12 Kingma’s appropriate reference
Gilabert, Pablo 49–50 class 139–40

losing ability to perform a
function 159–62
McMahan 160–2
medicalisation of disability 131–2,
135, 138
negative impairments 128–9, 141, 151
over-inclusiveness 129–30, 137, 143–54
physical and cognitive reality 133–4
Savulescu on enhancements 147–8
sex-based differences 153
Silvers, Anita 156–7
social norms and entitlements 60
Stoner 158
Stramondo 157–8
value of distinguishing difference to
disability 129, 134–6
insomnia 151–2
insult
adaptive preferences 7–8, 167–8, 196–8
capabilities to function 126–7
curing impairment 105
functionings 115n58
and impairment 22, 27
paternalism as 204–5
insult account 200–1, 205–7
Johnson, Harriet McBryde 56, 112
judges withholding information
example 206
justice adaptive preferences 184–92. see also
deaf individuals
adaptive preferences 51n20, 165–8, 171,
177, 180
choices 187, 189–90, 193
internalisation of oppressive norms 169,
187, 193, 243
mobility 190
non-autonomous character
planning 168–9, 173, 187, 193
oppression 186
responding to 168, 180, 192–6
and well-being adaptive preferences 7–8,
167–8, 172, 176, 184, 187, 191–3
justice-based account
access to education 57–9
access to entitlements 42–5
access to opportunities 44–5
bias 65–6
and capability approach 55–6
 273
changing disability status over time 66–8
cognitive impairments 65
difficulty performing functionings 61
disabled/impaired compared to
disadvantaged 246
disadvantaged compared to disabled/
impairment 42
discrimination 50–1, 66
entitlement identification 48–54, 245
entitlements without hierarchy 54–7
feasibility 45–8
the feasibility constraint 48–50, 53–4,
245, 248
impairments and restriction of
opportunities 65–6
internal capabilities 64
levelling-down entitlements 46–7, 51
minimal functioning 57–61, 80
mistreatment 50–1
norms 65–6
Nussbaum 64n50
paradigm cases 62
reasonable expectations about decent
human life 52–3
relevant opportunity loss 56
social norm approach 61
species norm approach 52–3, 61
state assistance assumptions 62, 241
sufficientarian approaches 47–8
thinning-out entitlements 47–8
treatment of impairments 60
unjust behaviour 49–51
welfarist approaches 55–6
justice proceduralist accounts 168, 184–92,
214. see also proceduralist accounts;
substantive proceduralist approach
Kahane, Guy 28–30, 31n60, 33–4, 147–8
Keller, Helen 110–12, 194–5
Khader, Serene 171, 182–4, 186
Kingma, Elselijn 139–40
Kittay, Eva Feder 29, 104n34
lecture slides example 205–6
leisure activities
adaptive preferences 180–1, 190
capabilities for control 98–9, 108–9,
118–20, 124
consultation 231
274 
leisure activities (cont.)
control in central domains 218–19
entitlement 44–5, 52–3, 58–9
functionings 71, 83
opportunities 63–4, 159–60
leprosy examples 11–12
liberalism 112–13. see also perfectionist
liberalism; political liberalism
libertarian paternalism (nudge
policies) 204n9
life expectancy, lowered 51, 139–41, 156–7
list of capabilities
and functionings 217, 219–20
Nussbaum 109–10, 124–5, 243
paternalism 217–19, 218n51
relationships 243
revision 109–10
locked-in syndrome examples 47–9, 54–5
marfanoid habitus 140. see also Phelps,
Michael
marginalization 171
McKeever, Natasha 101n25
McMahan, Jeff 160–2
the medical model 3, 11, 15–17, 21, 27,
36–7, 131
menstruation examples 139–40, 151,
153–4
migraine example 67
Mill, John Stuart 203, 214
misanthropes 30, 101, 247–8
Mitchell, Polly 170
mobility
capabilities for control 90–1, 115
entitlements and species normal 52–3
individual who cannot walk
example 115–16
justice adaptive preferences 190
locked-in syndrome examples 47–9,
54–5
paraplegic individuals 44, 47, 55–6, 97–8
selection of domains of control 215–16
toe wiggling examples 75–80
value of opportunities 78
walking as inessential 78–9
wheelchair use 5, 18, 55–9, 61, 65, 78–9
mobility impairment
entitlement to state assistance 59
as not a disability 66–7
morality 240, 242, 245–6, 248
Mr. Darcy example 52–3
multiple sclerosis (MS) examples 96–7,
125–6, 156
myopia. see vision impairments
neurodiversity movement 102. see also
disability rights movement
nourishment
achieved passively example 117
bad functionings 75–6
capabilities 78, 91–2, 111
capability approach 69–71
capability sets 97–8
entitlements 49–50
fasting individual examples 73–4, 91–2,
100–1, 111, 186–7
metabolism example 82
pregnant women example 72–3
and real freedom 91–2
risk and real freedom 95–6
starvation examples 73–4, 77–8, 91–2,
100–1, 186–7
Nussbaum, Martha C.
adaptive preferences 184–5
capabilities to live a decent life 85
capability approach 85, 184
‘dark’ capabilities 113–14
on Elster 171–2
expressive subordination 105
functions essential to decent human
life 71
goal of legislation and public
planning 99–100
internal capabilities 64n50
list of capabilities 109–10, 124–5, 243
opportunities for sexual
satisfaction 101–2
overlapping consensus 103
overlapping consensus of value of
capabilities 99–101
perfectionist approach 181–2, 186
on social institutions and
impairments 104n34
three forms of capability 92–3
on tolerating the intolerant 113
value of capabilities 118
value of opportunities and
outcomes 73

obesity example 38, 62–4
Oliver, Mike 17n20
oppression. see also gender-based
oppression; state oppression; structural
oppression
adaptive preferences 171, 177, 180
internalisation of oppressive norms 187,
190–1, 193, 243
norms 7
perfectionist approach 182
self identification of 7, 166–7
option sets 110–11, 170–1, 221–5
pain conditions
arthritis examples 37–8, 136, 140–1
capability sets 97–8
chronic pain examples 17–19, 26–7,
61, 163
contingent symptom vs disability 156–7
fibromyalgia example 157–8
rheumatoid arthritis example 157
welfare deficits 43–4
paralysis examples 17, 30–1, 78–81,
161–2
paraplegia
adaptive preferences 194
avoiding assumptions 163–4
Barnes’s conception as disability 133
capability sets 97–8
entitlements 44, 47
as impairment 133, 135, 141
as not disabling 62
relevant opportunity loss 55–6
resourcism 84n38
parental paternalism 201, 236–8
parenthood examples 97–8, 119–21,
150–1, 229–30, 236–8. see also
sterilisation
park ranger example 21
Parry, Jonathan 207–9, 213, 213n42
paternalism. see also epistemic paternalism;
libertarian paternalism (nudge
policies); parental paternalism;
putative paternalism; redistributive
paternalism; soft paternalism;
successful paternalism
adaptive preferences 176
and anti-paternalism 202–11
and autonomy 201, 232–6
 275
Barnes 237–8
Colburn’s account of autonomy 173,
233–4
consent 209–10, 229–30
consultation 229–30
content-neutral freedom 217–21
costs 221–2
cures 226, 236–8
defined 202
Feinberg 233
Grill 202
insult account 200–1, 205–7
as insulting 204–5
interference 201–3, 205–6, 219–25
intervention 201, 203–4, 208–9, 217–18,
220–5, 229
list of capabilities 217–19, 218n51
mistakes 201, 204–5, 234
negative judgments 204–7, 224, 228n69
option sets 214, 221–5
persuasion 222–3, 233
and Power of Prudential Exclusion
(PPE) 220, 225, 234–5
providing information 222–3
Quong’s judgmental definition 224
refusal of assistance 207, 213
refusing to assist 223–5
Shiffrin 221–2, 224n66
specifying capabilities without the
good 201, 209–10, 220–1, 228–32
Stramondo 237–8
Tsai 223
unwanted offers 201, 210–11, 213, 225–8
value of capabilities 121
Voight 228n69
and welfarist approach 216
paternalist intervention. see paternalism
patriarchal norms 7, 165–6
perfectionist approach 181–2, 186, 216–17,
219, 228–9. see also Nussbaum,
Martha C.
perfectionist liberalism 233–4. see also
liberalism; political liberalism
Pettit, Philip 94–5
PGD (pre-implantation genetic
diagnosis) 236–8
Phelps, Michael 140, 146–7
photosensitive individuals example 150–1
pluralism 124–5
276 
pluralist egalitarian 244–5
pluralist welfarism 43–4
Pogge, Thomas 81n29, 84–6
policy. see cures; disability, defining; justice
adaptive preferences; state assistance
political liberalism 113, 114n51. see also
liberalism; perfectionist liberalism
pollution example 82
poverty tourism (poverty appropriation)
functionings and capabilities 74–5
Power of Prudential Exclusion (PPE)
and paternalism 220, 225, 234–5
state assistance 207–10, 213, 220
power relations 83, 88–9, 244
pregnant woman examples 72–3, 120–1
pre-implantation genetic diagnosis
(PGD) 236–8
Premenstrual Dysphoric Disorder
example 151
proceduralist accounts 171–2, 178. see also
justice proceduralist accounts;
substantive proceduralist approach
putative paternalism 203–4
Quong, Jonathan 205–6, 224
race (running) example 224–5
rape example
bodily integrity 124
Rawls, John 30, 86–7, 87n44, 184
recognition theorists 244
redistributive paternalism 201
reflective equilibrium 1, 38–9, 69, 125,
184–6
relational egalitarians 244
religious conservative example 107, 112
religious convictions example 228
religious expression example 105
resourcist approach
compared to capability approach 70,
81–4, 88–9
Dworkin 84n38, 87–8
expanding the definition 86–8
as less demanding than capability
approach 84–6
paraplegia 84n38
Pogge 82, 84–6
Riddle, Christopher 37
Robeyns, Ingrid 71–2, 83–4, 96–7, 102n29
runaway trolley example 213
Savulescu, Julian 28–30, 31n60, 33–4, 147–8
schizophrenic individuals example 65
seatbelt example 208–9
Sen, Amartya
capabilities and the independent
individual 94–5, 117–18
famines 122n77
fasting and starving individual 91–2
hopeless beggar example 169–71
personal characteristics 76–7
on positive mental states 174n26
value of opportunities and outcomes 73
sexuality. see also asexuality
celibacy examples 100–1, 107–8, 118–19
conversion therapy examples 214–15,
235–6
homosexuality examples 34, 38–9, 132–3
selection of domains of control 215–16
sexual satisfaction
asexuality 101–4, 107, 110, 218
capabilities to control 112
celibacy 100–1
Shakespeare, Tom 18, 37
Shiffrin, Seana 221–2, 224n66
shoving a person examples 209, 229
sickle-cell anaemia example 139–40
Silvers, Anita 57–8, 156–7
singing example 76–7
the single mother example 107n39
Smith, Steven 16
smoking. see cigarettes, paternalist
examples
sobriety, paternalist example 221–2
the social model
Barnes’s rejection of 32
criticisms of 36–9
defining disability 10–11, 13–15, 17–20,
36, 38–9
impairment 131
social status 23–4, 74–5, 170–1, 244
soft paternalism 201, 214, 222–3, 233–4
spina bifida example 15–16
sports car example 115
spousal abuse examples 75–6, 78, 124,
165–6, 177, 234
starving individuals examples, see
nourishment
state assistance. see also taxation
anti-paternalism 201, 211–15
and the anti-paternalist filter 209
  277

Arneson 222n58 difficulty eliminating 138–9

assumption that impairment must be disability as internal 138
disadvantageous 62 as disabling 17
and autonomy 232–6 employment 93
costs 221–2 functionings 145, 153–4
cures 212, 231–2, 236–8 Haslangerian account 38
De Marneffe 222 HIV-positive example 150
disability, defining 29 impairment distinguished from
disabled people as passive recipients who disability 135–6
are bestowed help 200 obese individuals 62–4
domains of control 218–20 opportunities 151–2
enforcement tools 245 privileged social groups 74–5
Enoch 204–5 from social position not medical
entitlements 59 need 141–3
Feinberg 214 Stoner, Ian 158
Fox 209–10 Stramondo, Joseph 35n85, 157–8, 237–8
a good life 219, 232–3 structural injustice 186, 246–7
Grill 202, 213n42 structural oppression 246
Groll 207–8 substantive approach 168, 181–4. see also
intervention and interference 220–5 substantive perfectionist approach;
interventions as morally substantive proceduralist approach
unacceptable 245 substantive opportunities (capabilities). see
limiting of Odysseus’s liberty 202n2 capability approach
as morally required 199–200 substantive perfectionist approach 182,
Parry 207–9, 213 217–18
paternalism as threat 216–17 substantive proceduralist approach 154–5,
persuasion 222–3 182–4, 194. see also justice proceduralist
Power of Prudential Exclusion accounts; proceduralist accounts
(PPE) 207–10, 213, 220 successful paternalism 8–9, 203–5, 212,
Quong 205–6 221, 239
redistributive paternalism 201 sufficientarian approaches 47–8
refusing to assist 223–5
specifying capabilities without the taxation 195, 210–11, 220–1, 228
good 228–32 Terlazzo, Rosa 171, 182–4, 186
testimonial injustice 205 Terzi, Lorella 27, 28n49
unwanted offers 225–8 Tiny House Movement example 74–5
value of control 148 trans women examples 32–3, 153–4
Voight 220 Tsai, George 223
well-being 215–16
state oppression 43–4. see also oppression unemployed person example 169–71.
state policy, goals see also employment
Arneson’s objection to anti-
paternalism 214n43 Vallentyne, Peter 76, 78, 120–1, 125–6
sterilisation 98–9, 120–1. see also vision impairments. see also blindness
parenthood examples entitlements 3–4, 19–20, 59–60, 67
stigma myopia 134–5, 157–8
adaptive preferences 175–6 short-sightedness 3–4, 59–60, 111,
Barnes 151 134–5
body characteristics vs Voight, Kristin 220, 228n69
deficiencies 132–3 voting examples 118–23
278 

Wasserman, David 111 capabilities for control 90–1

Watson, Nicholas 37 environmental factors 22, 64–5
welfarist approach 43–4, 241. see also as impaired 13–14
Arneson, Richard; Vallentyne, Peter mobility 5, 18, 55–9, 61, 65, 78–9
accounts of disability 12, 28–31, 33–4, relevant opportunity loss 56–7
36, 39, 55–6 restriction of abilities 65
accounts of justice 43–4 social norms 15–16
capabilities 119, 123–4 WHO
opportunities 118–19, 226 defining disability and
and paternalism 216 impairment 11n7, 12–13
well-being 184 external factors 21–4
what is valuable or essential to human personal factors 21
life 26 Wolff, Jonathan 12, 55–6, 96n13, 97,
wheelchair use 124–5, 184–5, 243
assumptions 157 women pursing careers 82, 173, 175, 190
capabilities 47, 61 workaholic examples 118–20