Vice Epistemology

Vice Epistemology
Some of the most problematic human behaviors involve vices of the mind such as
arrogance, closed-mindedness, dogmatism, gullibility, and intellectual cowardice,
as well as wishful or conspiratorial thinking. What sorts of things are epistemic
vices? How do we detect and mitigate them? How and why do these vices prevent
us from acquiring knowledge, and what is their role in sustaining patterns of igno-
rance? What is their relation to implicit or unconscious bias? How do epistemic
vices and systems of social oppression relate to one another? Do we unwittingly
absorb such traits from the process of socialization and communities around us?
Are epistemic vices traits for which we can be blamed? Can there be institutional
and collective epistemic vices?
This book seeks to answer these important questions about the vices of the
mind and their roles in our social and epistemic lives, and is the first collection of
its kind. Organized into four parts, chapters by outstanding scholars explore the
nature of epistemic vices, specific examples of these vices, and case studies in
applied vice epistemology, including education and politics.
Alongside these foundational questions, the volume offers sophisticated
accounts of vices both new and familiar. These include epistemic arrogance and
servility, epistemic injustice, epistemic snobbishness, conspiratorial thinking,
procrastination, and forms of closed-mindedness.
Vice Epistemology is essential reading for students of ethics, epistemology,
and virtue theory, and various areas of applied, feminist, and social philosophy.
It will also be of interest to practitioners, scholars, and activists in politics, law,
and education.
Ian James Kidd is an Assistant Professor of Philosophy at the University of

Nottingham, UK.
Heather Battaly is a Professor of Philosophy at the University of Connecticut,

USA.
Quassim Cassam is a Professor of Philosophy at the University of Warwick, UK.

Vice Epistemology
Edited by Ian James Kidd,

Heather Battaly, and
Quassim Cassam
First published 2021
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2021 selection and editorial matter, Ian James Kidd, Heather Battaly and
Quassim Cassam; individual chapters, the contributors
The right of Ian James Kidd, Heather Battaly and Quassim Cassam to be
identified as the authors of the editorial material, and of the authors for their
individual chapters, has been asserted in accordance with sections 77 and
78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Kidd, Ian James, 1983- editor. | Battaly, Heather D., 1969- editor. |
Cassam, Quassim, editor.
Title: Vice epistemology / edited by Ian James Kidd, Heather Battaly, and
Quassim Cassam.
Description: Abingdon, Oxon; New York, NY: Routledge, 2020. |
Includes bibliographical references and index. |
Identifiers: LCCN 2020016484 (print) | LCCN 2020016485 (ebook) |
ISBN 9781138504431 (hardback) | ISBN 9781315146058 (ebook)
Subjects: LCSH: Vices. | Knowledge, Theory of.
Classification: LCC BD176 .V53 2020 (print) | LCC BD176 (ebook) |
DDC 179/.8–dc23
LC record available at https://lccn.loc.gov/2020016484
LC ebook record available at https://lccn.loc.gov/2020016485
ISBN: 978-1-138-50443-1 (hbk)
ISBN: 978-1-315-14605-8 (ebk)
Typeset in Times New Roman
by Deanta Global Publishing Services, Chennai, India
Man is only a reed, the weakest in nature, but he is a thinking reed … Let us then
strive to think well; that is the basic principle of morality.
Blaise Pascal, Pensées, §200 (Lafuma)
For all those of us struggling against the vices of the mind
Contents
Notes on contributors xi
Acknowledgments xiv
Introduction: from epistemic vices to vice epistemology 1

IAN JAMES KIDD, HEATHER BATTALY, AND QUASSIM CASSAM
PART I
Foundational issues 19
1 The structure of intellectual vices 21

JASON BAEHR
2 The metaphysical foundations of vice epistemology 37

QUASSIM CASSAM
3 Ignorance, arrogance, and privilege: vice epistemology and the

epistemology of ignorance 53
ALESSANDRA TANESINI
4 Epistemic corruption and social oppression 69

IAN JAMES KIDD
PART II
Collectives, institutions, and networks 87
5 Institutional epistemic vices: the case of inferential inertia 89

MIRANDA FRICKER
x Contents
6 Capital vices, institutional failures, and epistemic neglect in a
county jail 108
JOSÉ MEDINA
7 Implicit bias and epistemic vice 126

JULES HOLROYD
8 Vectors of epistemic insecurity 148

EMILY SULLIVAN AND MARK ALFANO
PART III
Analyses of specific vices 165
9 Quitting, procrastinating, and slacking off 167

HEATHER BATTALY
10 Epistemic insensitivity: an insidious and consequential vice 189

MAURA PRIEST
11 Intellectual snobs 208

CHARLIE CRERAR
PART IV
Applied vice epistemology 223
12 Teaching to the test: how schools discourage phronesis 225

CASEY JOHNSON
13 Vices of questioning in public discourse 239

LANI WATSON
Index 259
Contributors
Ian James Kidd is an Assistant Professor of Philosophy at the University of

Nottingham, UK. His research interests include character epistemology, femi-
nist ethics and social philosophy, and Buddhist and classical Chinese philoso-
phies. Earlier edited collections include The Routledge Handbook to Epistemic
Injustice (with José Medina and Gaile Pohlhaus, Jr.). His website is www.ian-
jameskidd.weebly.com.
Heather Battaly is a Professor of Philosophy at the University of Connecticut.
She is author of Virtue (Polity 2015) and editor of Virtue and Vice, Moral
and Epistemic (Blackwell 2010) and The Routledge Handbook of Virtue
Epistemology (2019). She works on virtues and vices in epistemology and
ethics.
Quassim Cassam is a Professor of Philosophy at the University of Warwick,
UK. He is the author of six books, including Vices of the Mind: From the
Intellectual to the Political (Oxford 2019), Conspiracy Theories (Polity 2019),
and Self-Knowledge for Humans (Oxford 2014). As well as vice epistemology,
his research interests include the philosophy of terrorism and extremism.
Mark Alfano is an Associate Professor of Philosophy at Macquarie University,
Australia. He uses tools and methods from philosophy, psychology, and com-
puter science to explore topics in social epistemology, moral psychology, and
digital humanities. His books include Character as Moral Fiction (Cambridge
2013) and Nietzsche’s Moral Psychology (Cambridge 2019).
Jason Baehr is a Professor of Philosophy at Loyola Marymount University in
Los Angeles, California. He specializes in theoretical and applied virtue epis-
temology. He is author of The Inquiring Mind (Oxford 2011) and editor of
Intellectual Virtues and Education: Essays in Applied Virtue Epistemology
(Routledge 2016). From 2012 to 2015, he directed the Intellectual Virtues and
Education Project at LMU, which involved founding the Intellectual Virtues
Academy, a public charter middle school in Long Beach, CA. He is currently
working on two books: a second monograph in theoretical virtue epistemology
and a book for teachers, provisionally titled Deep in Thought: A Philosophical
xii Contributors
and Practical Guide to Teaching for Intellectual Virtues (under contract with
Harvard Education Press).
Charlie Crerar is an Assistant Research Professor at the University of
Connecticut. He works primarily on issues in virtue and social epistemology.
Recent publications include a collection on Harms and Wrongs in Epistemic
Practice (Cambridge 2018, co-edited with Simon Barker and Trystan Goetze),
and papers on epistemic injustice, intellectual charity, and intellectual vice.
Miranda Fricker is Presidential Professor of Philosophy at The Graduate
Center, City University of New York. Her research is primarily in moral phi-
losophy, feminist philosophy, and social epistemology. She is the author of
Epistemic Injustice: Power and the Ethics of Knowing (Oxford 2007); co-
author of Reading Ethics: Selected Texts with Interactive Commentary (Wiley-
Blackwell 2009); and co-editor of a number of edited collections, most recently
The Routledge Handbook of Social Epistemology (2019). She is an Associate
Editor of the Journal of the American Philosophical Association, and a Fellow
of the British Academy.
Jules Holroyd is a Vice-Chancellor’s Fellow at the University of Sheffield
and co-director of its Centre for Engaged Philosophy, UK. Her teaching and
research focus on questions concerning the ways in which we are implicated in
injustices and span the topics of political philosophy, moral psychology, and
social philosophy (in particular, feminist philosophy and philosophy of race).
She has published extensively on discrimination, implicit bias, moral respon-
sibility, and feminist ethics.
Casey Johnson is an Assistant Professor in the Department of Politics and
Philosophy at the University of Idaho. Her main areas of interest include
social epistemology and feminist philosophy of language. Casey was the edi-
tor of the volume Voicing Dissent: The Ethics and Epistemology of Making
Disagreement Public (Routledge 2018). Her current work concerns our obliga-
tions to meet one another’s epistemic needs.
José Medina is Walter Dill Scott Professor at Northwestern University, Illinois.
He works primarily in critical race theory, feminist and queer theory, politi-
cal philosophy, communication theory, and social epistemology. His books
include The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic Injustice, and Resistant Imaginations (Oxford 2012). Current book
projects include Racial Violence and Epistemic Activism and Theories of the
Flesh: Latin-American and US Latina Feminist Theories (with Andrea Pitts
and Mariana Ortega).
Maura Priest is an Assistant Professor and bioethicist in the Department of
Philosophy in the School of Historical, Philosophical, and Religious Studies at
Arizona State University. Her research interests span biomedical ethics (espe-
cially disability, disability and mental health, and paediatric issues), normative
Contributors xiii
and applied ethics, and virtue ethics and epistemology. Her forthcoming book
is Can We Trust Elites? (Routledge 2021).
Emily Sullivan is an Assistant Professor of Philosophy at Eindhoven University
of Technology, the Netherlands. Her research explores the intersection between
philosophy and data and computer science and investigates how explanations
and data-driven models can promote and be designed for epistemic values.
Alessandra Tanesini is a Professor of Philosophy at Cardiff University, Wales.
Her current work lies at the intersection of ethics, the philosophy of lan-
guage, and epistemology with a focus on epistemic vice, silencing, prejudice,
and ignorance. Her new book The Mismeasure of the Self: A Study in Vice
Epistemology is forthcoming with Oxford University Press.
Lani Watson is a Leverhulme Early Career Fellow at the University of Edinburgh,
Scotland. Her research focuses on questions and questioning, drawing pri-
marily on virtue, vice, and social epistemology. Her forthcoming book with
Routledge is titled The Right to Know: Epistemic Rights and Why We Need
Them. Her website is www.philosophyofquestions.com.
Acknowledgments
Ian is grateful to the University of Nottingham for a period of research leave. We

are grateful to the University of Connecticut for sponsoring the Vice Epistemology
Conference in April 2019. We thank all conference participants, especially
our commentators: Teresa Allen, Josh Dolin, Yuhan Liang, Dan Licitra, Alice
Monypenny, Heather Muraviov, Nate Sheff, Heather Spradley, Taylor Tate, and
Cody Turner. We are also grateful to Adam Johnson and the team at Routledge
for their excellent work.
Introduction
From epistemic vices to vice epistemology
Ian James Kidd, Heather Battaly, and Quassim Cassam
This is the first volume dedicated to the emerging discipline of vice epistemology:
the study of the nature, identity, and epistemological significance of epistemic
vices. The most commonly discussed are perhaps arrogance, closedmindedness,
and dogmatism, but there are many more. Some epistemic vices are esoteric, cur-
rently known only to those with specialist training – think of vices like testimonial
injustice, epistemic self-indulgence, epistemic insensibility. To this list, one can
add more familiar epistemic vices, such as conspiratorial thinking, wishful think-
ing, gullibility, or cynicism. Such vices of the mind contrast with epistemic vir-
tues, the dispositions, attitudes, and ways of thinking that constitute excellences
of epistemic character, like open-mindedness and intellectual humility. Taken
together, epistemic virtues and vices are entrenched parts of our everyday ways of
describing and assessing epistemic character.
A defining feature of the vices of the mind is that they make us bad think-
ers, insofar as they prevent us from acquiring and sharing knowledge, express
bad motives and desires, or interfere with our individual and collective epis-
temic functioning, in all sorts of ways. Some are irksome, others are dangerous.
Assessing the nature and scope of the badness of epistemic vices is a main task
for vice epistemology. Closely related is the more ameliorative project of trying
to identify effective ways to prevent them from developing or to weaken those
already in place. These diagnostic and ameliorative efforts are integrated for the
following reason: we worry about epistemic vices because we, and the people
with whom we share the world, have them, and their effects alarm, anger, disturb,
and distress us in all sorts of ways.
The discipline of vice epistemology can be thought of as starting from these very
general thoughts. Our epistemic behavior and character are integral to our lives as
human beings at practically every level. Activities such as arguing, criticizing, and
evaluating are essential to social intercourse, from the trivial to the momentous, and
from the idiosyncratic to the universal. Our social and political existences depend
on our being capable of effectively assessing, explaining, and understanding aspects
of our world – from its economic structures to its legislative processes. Our moral
agency requires us to be able to form reliable judgments of other people, to be able
to engage in self-reflexive critical scrutiny of our own motives and actions, and to
acquire and properly deploy knowledge and understanding of people and situations.
2 Kidd, Battaly, and Cassam
Much of our moral wrongdoing involves epistemic failures, such as thoughtless-
ness, for instance. Moreover, the effective development and performance of some
of our major shared institutions and projects also require us, individually and col-
lectively, to have certain epistemic virtues and to avoid certain epistemic vices.
The obvious examples, here, are such culturally prestigious epistemic projects as
the modern scientific enterprise, but any large-scale cultural projects will require
us to be knowers of certain sorts. The quality of our epistemic characters therefore
affects, in complex and substantive ways, the wider course and quality of our lives,
as individuals and as members of communities and cultures.
How bad are epistemic vices? Such vices of the mind as arrogance and dogma-
tism are sometimes irritating but bearable, at least when they are confined to the
office boor who knows it all, or the diehard music fan who won’t budge from their
conviction that their favored band is best of all. Typically, though, epistemic vices
will tend to have more severe effects. Arrogance and dogmatism can contribute
to patterns of social exclusion and oppression – ramifying with wider patterns of
racist and sexist bias, supercharged by cultural tendencies to political polariza-
tion, amplified by the suboptimal social and epistemic structures of modern life.
In these cases, the vices of the mind contribute to serious, systematic patterns of
injurious, often violent mistreatment, injustice, and oppression. If our thinking is
not right, then much else is at risk. At this point, vice epistemology starts to merge
into the new discipline of political epistemology (Hannon and de Ridder 2010).
Stepping back from these large-scale social and political contexts, there are
still many other reasons to worry about epistemic vices. Maybe we want to think
well for its own sake – if, like Pascal, one believes that ‘to think well is the basic
principle of morality’ (Pensées §200, Lafuma). Maybe we want to do our best to
honor certain epistemic ideals, such as objectivity, integrity, or epistemic respon-
sibility, the latter concept being one of the original motivators for early virtue
epistemology (Code 1987). Maybe we are committed to aspirational normative
ideals that require of us virtues of the mind as well as the heart, reflecting a vision
of ourselves as creatures for whom a good life necessarily consists of epistemic
as well as ethical excellences. Maybe we are interested in the vices of the mind
because, when lined up alongside the ethical vices, they help to confirm a misan-
thropic vision of humankind as infused with failings and corruptions. All of these
reasons, and others, can motivate the study of epistemic vices.
We think there are many reasons to want to take epistemic vices seriously.
Below, we take a quick tour through the historical study of epistemic vice, and
the contemporary development of vice epistemology. We highlight key themes in
contemporary work on epistemic vice and vice epistemology, and offer an over-
view of the volume’s chapters. We close with some open questions that warrant
further exploration.
The historical study of epistemic vices: a brief tour

The systematic study of epistemic failings and vices is nothing new since there
have been implicit and explicit studies of epistemic failings in several of the
From epistemic vices to vice epistemology 3
world’s philosophical traditions. Such investigations are not always articulated
in the terms of contemporary vice epistemology, of course, since not all cul-
tures recognize the distinction between ethical and epistemic vices. Kǒngzǐ, for
instance, notes a set of vices that systematically interfere with individual epis-
temic agency, such as foolishness, rigidity, and lack of a love of learning, a
deficient appetite for epistemic goods (Confucius §17.8).1 Similarly, the early
Daoist, Zhuāngzǐ, describes an array of epistemic character traits opposed to
the clarity of vision and lucidity proper to the sage, which can be articulated as
vices like arrogance and dogmatism; on his analysis, these typically arise from
failures to grasp the perspectival character of our perceptions, convictions, and
evaluations.
Within the early Buddhist texts of the authoritative Pali Canon, one finds very
complex taxonomies of character failings, with the detailed listings of ‘cankers’,
‘taints’, ‘defilements’, and ‘fetters’ presented in the Majjhima Nikāya and Digha
Nikāya (see Nyanatiloka 2004), accompanied by similar lists and condemna-
tions in other texts, too.2 Such failings are ontologically diverse, since they
include many kinds of things, including character traits, sensibilities, affective
dispositions, and corrupted styles of thinking and attention. Keown (1991: 63f)
argues that early Buddhism, at least, recognizes the twofold Aristotelian dis-
tinction: there are epistemic vices rooted in moha, ‘delusion’, and ethical vices
rooted in lobha, ‘greed’, and dosa, ‘hatred’. All of our vices and failings are
ultimately grounded in these three akusala-mūla or ‘unwholesome roots’ (see,
e.g., Visuddhiamagga XXII, 49, 65). Moreover, other early Buddhist texts add
other epistemic failings, which include various samyojana, ‘fetters’ – false and
recalcitrant beliefs which entrap us within the cycle of rebirth, the most famous
of which is the delusory belief in a substantial, persisting self (sakkāya-diṭṭhi).
There are also incapacitating forms of doubt or uncertainty, especially about the
Buddha’s awakeness and the nine supermundane consciousnesses (vicikicchā),
as described in the Saṅgīti Sutta (e.g., Digha Nikaya 33). The early Buddhist
texts of the Pali Canon therefore offer rich prospects for studies in cross-cultural
vice epistemology, perhaps made all the more interesting by the specific sote-
riological context of that tradition and its concern with human salvation. After
all, our ‘cankers’, ‘defilements’, and other vices and failings are fundamentally
of concern because they perpetuate our entrapment within samsara, the cycle of
rebirth and dukkha.
In ancient Greece, Aristotle is known to have introduced the distinct cate-
gory of epistemic virtues (Nicomachean Ethics VI), though it is worth exploring
whether he had a corresponding notion of epistemic vices. Whether or not they
had an explicit notion of epistemic vice, it is clear that Greek thinkers were sensi-
tive to epistemic failings. Socrates, for one, castigates various forms of epistemi-
cally vicious behavior on the part of his interlocutors. A main target for him was
‘hubris’, a tendency towards inflated confidence in one’s beliefs, which, in the
case of ethical knowledge, is manifested by Polus in Gorgias and Euthyphro in
his eponymous dialogue. (Specific denunciations of dogmatism and hubris can
also be found in Charmides 166c7–d2 and Alcibiades 1 118a4–5.) We could also,
more speculatively, consider other ancient Greeks as having a pronounced hostil-
ity toward certain epistemic vices: think of Heraclitus’ criticisms of the foolish-
ness of the masses and the dogmatism and conceit of ‘polymaths’ (McKirahan
§§10.1–18), or the condemnations of the corrupting effects of learning and study
that run through the apothegms of the Cynics. Bion, for instance, targets intel-
lectual conceit (oiēsis), while Tiresias warns of the self-satisfaction that, for him,
runs throughout too much ‘philosophico-jibber-jabber’ (quoted and referenced in
Desmond 2014: 128, 65).
Use of a vocabulary of vices to describe and criticize forms of objection-
able epistemic behavior is one thing, of course, while the systematic philo-
sophical study that uses an articulate framework of epistemic vice is quite
another. Systematic vice theory really began, in the West at least, during the
early Christian tradition, with a focus on ethical and spiritual vice (see DeYoung
2009). Arguably, the first fully fledged vice epistemologies were developed dur-
ing the seventeenth century, likely with Sir Francis Bacon’s analysis of the ‘Idols
of the Mind’ (idola mentis), in Sections 34 to 55 of his Novum Organum of 1620.
Although few scholars of his work interpret the Idols in character-epistemic
terms, many of them are clearly individual and collective epistemic vices and
failings. Indeed, the Idols are defined by two scholars as ‘impediments of vari-
ous kinds which interfere with the processes of clear human reasoning’ (Jardine
and Silverthorne in Bacon 2000 [1620]: xix). Bacon’s sensitivity to our innate
and acquired epistemic vices and failings was taken up in early modern English
natural philosophy. The exploration of epistemic vices and failings found its way
into John Locke’s influential Essay Concerning Human Understanding, which
addresses the ‘Vices [that] oppose or menace our Endeavours’ (Essay § 3.4.18),
and into Locke’s surveys of various epistemic vices in his educational essay, Of
the Conduct of the Understanding (Locke 1996, Locke 2008). An important tra-
dition also emerged in early modern feminist vice epistemology, mostly focused
on education, represented by Mary Astell’s 1694 study, A Serious Proposal to
the Ladies and, just over a century later, Mary Wollstonecraft’s 1792 classic, A
Vindication of the Rights of Woman (Astell 2002, Wollstonecraft 1995). Each
is notable for its sensitivity to the gendered character of certain epistemic vices
and the ways that, under the influence of misogynistic social norms, the forms of
education afforded to women tended to be deeply epistemically corrupting (see
Kidd 2019).
These two early modern English traditions in vice epistemology are com-
plex and integrated with wider philosophical debates about natural science,
education, and the social and political rights of women. They invite more
extensive study both by vice epistemologists and specialist historians of phi-
losophy. The subsequent tale of vice epistemology is less clear, from the late
eighteenth century onwards, despite a clear enduring interest in arrogance,
dogmatism, and related vices. And even if vice epistemology went cold during
that period, that in itself demands explanation. In the course of those investi-
gations, we are confident that other precursor ventures into vice epistemology
will be revealed.
The contemporary development of vice
epistemology: a brief tour
What is clear, though, is that the modern emergence of vice epistemology traces
its roots to the development of its sister discipline, virtue epistemology. In the
early 1980s, various internal developments within analytic epistemology led to a
desire for new ways to think about seemingly intractable problems.
A crucial moment was Ernest Sosa’s (1980) paper, ‘The Raft and the Pyramid’,
soon to be followed by early formative work aimed at thickening epistemology
by Lorraine Code (1987) and James Montmarquet (1987). Highlighting the then-
neglected theme of epistemic responsibility, roughly modeled on similar themes
in ethics, Code and Montmarquet set the stage for Linda Zagzebski’s now-classic
book, Virtues of the Mind (1996). More than anything else, this established virtue
epistemology as a new, distinctive area for epistemologists, inspiring systematic
work on the nature of epistemic (or intellectual) virtue and, later, sustained stud-
ies of specific character virtues. Some crucial later studies were Robert Roberts
and W. Jay Wood’s (2007) Intellectual Virtues, and Jason Baehr’s (2011) The
Inquiring Mind.
The epistemic, or intellectual, vices were generally rather recessive in these
foundational studies other than an article by Casey Swank (2000). (We use ‘epis-
temic’ and ‘intellectual’ interchangeably, as applied to virtues and vices.) An hon-
orable exception is Miranda Fricker’s landmark Epistemic Injustice: Power and
the Ethics of Knowing (2007), which inspired new interest in the sources, nature,
and correction of the phenomenon of epistemic injustice (see Kidd, Medina, and
Pohlhaus 2017). Although Fricker’s theoretical framework was feminist virtue
epistemology, few subsequent scholars took up her invitation to conceive of
testimonial injustice (the main form she focuses on) as an epistemic vice (see
Battaly 2017 for a corrective study). It was therefore welcome when, in 2013,
José Medina’s The Epistemology of Resistance drew upon a variety of sources –
from Fricker to critical race theory and social epistemology – to offer perhaps the
first substantive study of epistemic vices. Medina’s book applies a sophisticated
vice-theoretical framework to analyze gendered and racial oppression and social
activism, giving specific attention to the ways that epistemic character and agency
are shaped, for better or worse, by the subject’s location within social practices
and systems of power.
The interest in epistemic vices represented by Fricker and Medina tended to
locate vice epistemology within the vicinity of feminist and race epistemologies.
But another crucial theoretical advance in the development of vice epistemology
was an important series of papers by Heather Battaly that drew upon the resources
of Aristotelian virtue theory. With this work, vice epistemology really got going,
although the field was not named until the publication of Quassim Cassam’s
paper ‘Vice Epistemology’ (2016). Battaly charted the varieties of epistemic vice,
described the major models for normative appraisal of epistemic character traits,
and developed original accounts of several neglected vices, such as epistemic
self-indulgence and epistemic insensibility (see Battaly 2010, 2013, 2014). Since
this brings us up to the present day, we now turn to survey the main themes of
contemporary vice epistemology.
Key themes in contemporary vice epistemology

The agenda of vice epistemology has thus far been influenced by two main fac-
tors. First, it has inherited a set of questions from the disciplines of virtue eth-
ics, virtue epistemology, and social epistemology. These include questions about
the nature of a vice, the ways that social structures affect ethical and epistemic
agency, and the ways that epistemic character relates to debates about knowledge
or justification. Jason Baehr once proposed thinking of virtue epistemology as
having ‘conservative’ and ‘autonomous’ forms: the former construes epistemic
virtues as significant insofar as they illuminate the agenda of analytic epistemol-
ogy, the latter recognizes that the virtues of the mind are of interest independently
of their significance to those debates (Baehr 2011: §1.2.2). The same could be
said of vice epistemology. Some of the current work engages with issues like the
justification of our beliefs, but there is also ‘autonomous’ work, for instance, on
epistemic vice and social oppression.
The second factor – or set of factors – that have clearly informed the agenda
and concerns of vice epistemology are wider cultural and political developments.
This is clearest in such modern ills as ‘post-truth politics’ and ‘epistemic bubbles’,
exemplifications of epistemic vice by major figures in (for instance) the British
and American political establishment, and decisive political events whose expla-
nation requires appeals to the arrogance, dogmatism, and other vices of political
actors.3 This is not to say that vice epistemology owes its contemporary fortunes
solely to those cultural and political developments, but they do underscore an
important point about the epistemically vicious character of the contemporary
social world. Cultural concern can also motivate an interest in virtues, of course,
as Bernard Williams demonstrated in Truth and Truthfulness – but the fact that
concern has shifted from virtues of truth to vices of truth says something about
how things have changed since the publication of that book (the cultural context
of concerns about truthfulness as a virtue are mainly discussed in Williams 2002:
ch.1). We could therefore say that if virtue epistemology developed alongside
social epistemology, then perhaps vice epistemology should develop alongside an
anti-social epistemology (see Alfano, Klein, and de Ridder Forthcoming).
Thus far, contemporary work in vice epistemology has centered around three
key themes.
1 Foundational work on the structure and features of epistemic vices, and their
impact on knowledge.
2 Analyses of specific epistemic vices.
3 Case studies in applied vice epistemology.
1 Much of the extant work in vice epistemology has explored questions related
to the first theme – the structure and features of epistemic vice. For starters,
theorists have offered competing views of what makes epistemic vices bad.
Some have argued that bad motivations are constituent components of epis-
temic vices (Battaly 2014; Crerar 2018; Tanesini 2018b), while others have
argued that epistemic vices are bad because they obstruct justified belief and
knowledge (Cassam 2016, 2019a). Vice epistemologists have likewise begun
to explore whether agents are blameworthy for their epistemic vices, paying
special attention to different notions of blameworthiness – some that involve
control and accountability, and others that do not (Battaly 2016, 2019; Cassam
2019a; Holroyd 2017; Fricker 2007, 2016). Though there is disagreement
over several of the main features of epistemic vice, much of the field is united
in thinking that epistemic vices can be stealthy or self-concealing; it is com-
monly held that an agent’s being (e.g.) closed-minded makes it difficult for
her to detect this vice in herself (Fricker 2007; Cassam 2019a; Holroyd 2017;
Medina 2013). Something similar may be true for the vice of willful herme-
neutical ignorance, the refusal of dominantly situated knowers to acknowl-
edge epistemic tools developed from the experienced world of those situated
marginally (Pohlhaus Jr. 2012). More generally, though extant work in the
field has tended to focus on the vices of individual agents, this focus is neither
exclusive nor necessary. Scholars working at the intersection of vice episte-
mology and social epistemology have argued that the possession of epistemic
vices isn’t limited to individual agents – groups can have epistemic vices, too
(Fricker 2010; Lahroodi 2019). Finally, theorists have begun to offer some
strategies for ameliorating epistemic vices (Cassam 2019a; Holroyd 2016;
Sherman and Goguen 2019), recognizing that these strategies must be sensi-
tive to whatever features the vices end up having, and to whatever conditions
end up facilitating the vices (Kidd 2018a).
2 The nascent literature on vice epistemology has likewise begun to provide
analyses of specific epistemic vices. Much of this work has focused on two
families of vices – the vices of epistemic injustice, and the vices associated
with the virtue of intellectual humility. As indicated above, Fricker’s (2007)
and Medina’s (2013) accounts of testimonial and hermeneutical injustice
have generated a substantial literature of their own. With respect to the
vices associated with intellectual humility, Alessandra Tanesini (2016) has
analyzed intellectual arrogance and haughtiness, and intellectual timidity
and servility (2018a), while Roberts and Wood (2019) have explored what
they call ‘vices of pride’, such as domination and hyper-autonomy. Also on
offer are analyses of the vices of closed-mindedness (Battaly 2018; Cassam
2019a), gullibility (Cassam 2019a), epistemic malevolence (Baehr 2010),
epistemic intemperance (Bloomfield 2019), and epistemic self-indulgence
(Battaly 2010).
3 Vice epistemologists have also begun to apply their analyses of specific epis-
temic vices to important domains of our lives. One of the main reasons to be
interested in epistemic vices is because we are often worried about epistemi-
cally vicious people – dogmatic politicians, arrogant doctors, lazy media con-
sumers, and so on. Vicious people create problems in the world, and much
of the modern interest in vice epistemology reflects these sorts of practical
concerns. Accordingly, several theorists have examined epistemic vices in
the domain of political discourse, including Cassam (2019b) who focuses on
conspiratorial thinking, and Michael Lynch (2019) who explores arrogance.
With respect to online discourse, Thi Nguyen (2018) has explored epistemic
vices connected to echo chambers and epistemic bubbles. In the domain of
education, Battaly (2013) has tackled epistemic insensibility in higher educa-
tion policy in Britain and the US (see further Baehr 2015 and Kotzee 2013).
Biddle, Kidd, and Leuschner (2017) argue that attacks on climate scientists
by a variety of sceptics, deniers, and ‘doubt-mongers’ are intended to cause
those scientists to develop the vice of epistemic timidity, while Kidd and
Carel (2014) explore the vices of epistemic injustice within clinical and
healthcare practice. Kidd (2016, 2017, 2018c) develops a methodology for
charging others with epistemic vice and uses it to assess charges of arrogance
and dogmatism against advocates of scientism.
Summary of chapters
Part I concerns foundational issues in vice epistemology, starting with ontological
questions about the sorts of things that epistemic vices are. In his chapter, Jason
Baehr argues that the structure of intellectual vices does not mirror the struc-
ture of intellectual virtues. For Baehr, possessing an intellectual virtue requires
excellence along ‘four-dimensions’ – virtue requires the motivation for epistemic
goods, good judgment, competence, and proper affect. In contrast, possessing an
intellectual vice does not require being defective along all four of these dimen-
sions. Baehr suggests that defective epistemic motivation can itself be sufficient
for intellectual vice. He likewise explores whether defective epistemic motivation
is necessary for possessing intellectual vice, tackling counter-examples proposed
in the recent literature.
A starting point is the set of ontological questions concerning the nature of
epistemic vices. Quassim Cassam initiated ‘vice ontology’ in an earlier paper
(Cassam 2017) and in his chapter for this volume offers a systematic set of reflec-
tions on the three main questions in that project: what kinds of things are epis-
temic vices, how are they individuated, to what are those distinctions answerable?
Consistent with his earlier work, Cassam defends vice-pluralism – the conviction
that epistemic vices can be many kinds of things, including character traits, atti-
tudes, and ways of thinking (Aristotle, by contrast, was a vice-monist, regard-
ing all vices as character traits). The individuation of vices can rely on direct
or indirect approaches, where the former focus on the vice itself and the latter
on our vice-concepts, like ‘closedmindedness’. This depends on the question of
what our vice-concepts are answerable to – a vital matter when we are confronted
with rival accounts of some vice. Cassam endorses the midway position that our
vice-concepts depend on the combination of our conceptual resources (especially
the vice-concepts we have contingently inherited) and empirical realities about
people’s psychology and conduct. Cassam ends by emphasizing that, however
complex epistemic vices are, they are only ever one of many styles of explana-
tion for people’s objectionable epistemic conduct. The others include situational,
ideological, moral, and political-rational explanations – a pluralism that a vice
epistemologist does well to honor.
Alessandra Tanesini’s chapter, ‘Ignorance, arrogance, and privilege’, traces
out some connections between vice epistemology and the epistemology of igno-
rance. She starts by arguing that certain epistemic vices are best construed as
sensibilities, and that these play a role in many forms of ‘active’ or ‘motivated
ignorance’. Epistemically vicious sensibilities make a person insensitive to what
is salient, given their epistemic goals, and they can ramify with epistemically
vicious attitudes. Tanesini argues that a set of psychological and social relations
between ignorance, arrogance, and privilege help to generate and sustain two spe-
cific vices – ‘racial insensitivity’ and ‘intellectual arrogance’. Her account then
concludes by noting important investigative directions for further studies of rela-
tionships between the epistemologies of vice and ignorance.
Continuing the theme of the social dimensions of epistemic vice, Ian James
Kidd’s ‘Epistemic corruption and social oppression’ explores the role vice epis-
temology can play in understanding the impact of social oppression on epistemic
character. Kidd maps several ways in which oppressive conditions can be ‘epis-
temically corrupting’ – i.e., can facilitate the development and exercise of epis-
temic vices. He argues that combating this corruption requires a specific kind
of vice epistemology, a ‘critical character epistemology’. One that is simultane-
ously sensitive to the different ways in which vices develop in differently situated
agents, informed by the epistemic and non-epistemic harms of vices, and alive to
the possibility that whether a trait counts as a vice might depend on context.
Part II explores epistemic vices at the collective, institutional, and network
levels. Everyone who accepts the existence of epistemic vices agrees that their
bearers include individual epistemic agents. But what about vice-bearers at wider
levels? Most accounts of epistemic vices are individualist in their orientation: they
focus on the epistemic vices of individuals. One might wonder, however, whether
institutions or collectives can also be epistemically vicious and, if so, what their
epistemic vices might be. These issues are taken up by Miranda Fricker, José
Medina, Jules Holroyd, and Emily Sullivan, and Mark Alfano.
In her contribution, Fricker asks why one might care about the question
whether institutions can be said to have vices of any kind. As she notes, the idea
of an institutional vice has at least a tentative foothold in public discourse about
institutions. In developing her account, she draws heavily on the notion of an
institutional ethos, understood as the institutional analogue of an individual’s
character. An institution’s ethos consists of the collective dispositions, values, and
evaluative attitudes that orientate its activities. Fricker argues that institutional
epistemic vices are displayed ‘whenever there are culpable lapses in the institu-
tion’s epistemic ethos and/or in the implementation of ends’. One such vice is
inferential inertia, which she argues is rapidly becoming part of our normal insti-
tutional environment. In one notorious case involving the British Broadcasting
Corporation (the BBC), the informational compartmentalization of the institution
ensured that scattered items of information about a sexual predator never amassed
into a body of evidence, but instead remained dispersed and inferentially inert
within the organization. As this example illustrates, inferential inertia is an insti-
tutional vice that needs to be distinguished and properly understood, especially
when public trust in institutions is in decline.
The institutional vice that Medina highlights is incredulity, which is a specific
form of the epistemic vice of testimonial insensitivity. This can be both a personal
and an institutional failing. Medina gives the example a detention facility inmate
whose medical emergency was ignored by the guards on duty despite repeated
pleas for help from other inmates. In this case, the institution itself, and not just
the particular detention officers who happened to be on duty, exhibited the epis-
temic vice of incredulity. The incredulity of individual guards who ignored pleas
for help for a distressed inmate vitiated their proper epistemic functioning as well
as their epistemic relationship with inmates. The sense in which the institution
exhibited the epistemic vice of incredulity is that it lacked adequate protocols
for handling emergency calls, trained its guards not to trust or empathize with
inmates, and operated deficient accountability procedures. Since the institution
did not treat inmates as trustworthy it became ill-equipped to ‘carry out prop-
erly basic epistemic operations required for providing care and protection for the
inmates’. This is a capital institutional vice where, as in the present case, it para-
lyzes or vitiates the institution’s overall epistemic functioning. Such vices can be
corrected by various forms of what Medina calls epistemic activism. This is much
more than consciousness raising. It is ‘an attempt to meliorate epistemic dynamics
and institutional frameworks so that capital epistemic vices are uprooted and the
work towards epistemic justice can begin’.
In her chapter, Jules Holroyd focuses on collective rather than institutional
vices. Her question is whether implicit biases are epistemic vices. This is not an
easy question since implicit biases are heterogeneous, and different vice episte-
mologists offer different accounts of epistemic vice. Holroyd puts pressure on the
idea that the implicit biases of individuals are epistemic vices by noting the low
predictive validity of implicit biases and the low test–retest reliability of meas-
ures of implicit bias. However, this leaves it open that implicit biases are collec-
tive epistemic vices. A collective or group C can be said to have vice V to the
extent that C is disposed to behave in ways characteristic of V. Collective vices
as Holroyd understands them do not require joint commitment to a bad end or
motive. Instead, they can emerge through negligence, via ‘hidden hand’ mecha-
nisms. Holroyd concludes by noting that being able to call out collectives as epis-
temically vicious may serve an important ameliorative function in addressing the
problematic patterns of bias in which we are implicated.
Rounding off Part II, Emily Sullivan and Mark Alfano point out that we often
rely on networks of other agents in arriving at our beliefs, both on-line and off-
line. Their chapter develops the concept of an insecure epistemic network – in
which one relies on very few independent sources, as in an echo chamber or
star-network. Sullivan and Alfano argue that an insecure network can negatively
impact one’s epistemic well-being, by making one less intellectually autonomous
and less likely to get true beliefs. They explore virtues and vices that are con-
nected to monitoring, adjusting, and restructuring one’s own epistemic networks
and those of others. Along the way, they distinguish epistemic security from epis-
temic safety.
Part III offers studies of specific epistemic vices. Heather Battaly opens with an
analysis of three closely related epistemic character traits – quitting, procrastinat-
ing, and slacking off. Each reflects different ways to lack the trait of perseverance.
Battaly’s argument is that these three traits are not always epistemically vicious
and might under certain conditions be virtuous. She draws on the virtue-theoretic
concept of akrasia and some concerns about the ‘fit’ between traits and hostile
epistemic environments to pose complicated questions about our appraisals of the
normative status of epistemic traits, then ends with a genealogy of some related
traits, such as apathy, folly, complacency, and resignation. All of this suggests,
against Aristotle, that there are several vices of deficiency for the trait of persever-
ance, if not for others, too.
In her chapter, Maura Priest argues that epistemic insensitivity is an interper-
sonal, or other-regarding, epistemic vice – one that tends to negatively impact the
epistemic endeavors of other agents. She conceives of epistemic insensitivity as
a tendency to fail to recognize, or fail to respond appropriately to, other agents’
epistemic concerns. Priest explores several manifestations of this vice, includ-
ing expertise insensitivity, which occurs when (e.g.) an expert is oblivious to or
doesn’t care that she is using terms of art that her interlocutor fails to understand.
Priest likewise distinguishes the vice of epistemic insensitivity from the vices of
oversensitivity, incivility, and arrogance.
Is intellectual snobbery a specifically intellectual vice? Charlie Crerar argues
that it is. It is an intellectual vice that affects the practice of making intellectual
evaluations. For Crerar, what makes someone an intellectual snob is not the con-
tent or even the strength of their intellectual evaluations, but rather the consid-
erations upon which these evaluations are based. A ‘snobbery of motives’ is one
form of intellectual snobbery. A person who is an intellectual snob in this sense
is not only disposed to appraise things on the basis of their intellectual status, but
does out of a desire to feel or appear superior to some other individual or group.
A different type of intellectual snobbery consists in a ‘snobbery of sensibilities’.
This consists in what Crerar describes as ‘an excessive sensitivity to intellectual
status that leads to unwarranted conclusions of intellectual superiority and inferi-
ority’. Crucially, however, being a snob is not just a matter of having an unwar-
ranted preoccupation with status. The vice consists in having a preoccupation
with considerations of intellectual status in a way that leads one to look down
on others whom one views as inferior. Crerar concludes his discussion with the
observation that the latter type of intellectual snobbery should be of special inter-
est to epistemologists who are looking to move beyond a motivational account of
intellectual vice.
The volume closes with Part IV – applied vice epistemology. Applied vice
epistemology. That term refers both to attempts to apply vice-epistemic insights
to practical domains (like healthcare and education) and the reciprocal appeal
to actual epistemic practice to inform thinking about epistemic vices. We think
that looking for epistemic vice in practice is crucial: among other things, it can
reveal new vices, articulate the different forms that vicious conduct can take, and
provide effective ameliorative strategies. After all, we often tend to spot epis-
temic vices when they manifest within our epistemic practices, especially those
practices that are integral to the organization and operation of our societies – the
media, law, education, healthcare, and so on.
Casey Johnson focuses on education, arguing that contemporary educational
practices discourage the development of epistemic phronesis. This refers to
a cultivated capacity to intelligently govern the application of one’s epistemic
skills, abilities, and other attainments – something which Johnson argues is often
thwarted by developments within education, such as the imperative to ‘teach to
the test’ which was encouraged in the United States by the No Child Left Behind
Act. Her analysis of epistemic corruption shows what vice epistemology can con-
tribute to contemporary critical discourses of education, as well as deepening our
understanding of a positive exemplar, the epistemic phronimos.
Lani Watson explores the connections between vice epistemology and the
practices of questioning that are central to our everyday, social, and political lives.
She starts with an account of bad questioning, divided into failures to identify a
proper content for one’s questions and failures to properly perform questions – for
instance, by asking in ways that are aggressive, intrusive, or misleading. Watson
organizes these into a sophisticated taxonomy of forms of bad questioning prac-
tice, which she then applies to a careful analysis of several recent, high-profile
cases of public political discourses, including the Scottish independence and
Brexit referenda in the United Kingdom. As those examples suggest, epistemic
vicious conduct can often be of deep importance to the integrity and future of our
social and political systems.
Open questions
Unsurprisingly, there are many open questions in vice epistemology that warrant
further exploration. Here, we mention just six sets of questions.
First, there are issues in vice ontology. What are epistemic vices? Some can-
didates are character traits, attitudes, sensibilities, ways of thinking, emotions,
and other aspects of our minds or characters. Some favor monistic answers,
while others go pluralistic. As well as their intrinsic interest, these ontological
questions also inform how we go about studying the epistemic vices – through
conceptual analysis, say, or appeal to empirical psychology. They can also
inform our ameliorative decisions, since what the vices are can inform our deci-
sions about how best to transform them, if that is even a possibility. Vice ontol-
ogy can also include vice taxonomy, the project of cogently ordering the vices
of the mind. We can start by providing lists of those vices, for sure, but a list is
only so useful.
Second, there are issues concerning the classification of epistemic vices, which
follows closely on the heels of vice ontology. How should we cogently organize
the vices of the mind? Are our current lists of those vices historically contingent?
Are we ‘missing’ some epistemic vices from our lists? Have certain kinds or clus-
ters of epistemic vices enjoyed attention, to the occlusion of others, as some his-
torically minded philosophers argue (Manson 2012, Smith 2016)? If so, how can
we spot those ‘missing’ vices? Can historical study of epistemic vices and failings
of the kind in Kidd (2018b) help us identify them and describe the wider contets
of thought which lent them salience? In pursuing these questions, we can turn
to work by social and intellectual historians on virtues, vices, and character (see
DeYoung 2009, Kivisto 2014, Paul and van Dongen 2017, Petkov 2012).
Third, which structural conditions facilitate and amplify epistemic vices?
Which structural conditions might ameliorate epistemic vices? What sorts of con-
cepts do we need to theorize the dynamic relationships between character, struc-
tures, and virtue and vice? Will we need to give different answers, depending on
the specific vice in question? Will some conditions facilitate some vices while
ameliorating others? Will there be any common features of ameliorative structural
conditions that could be employed across the varied domains of public discourse,
online media consumption, education, and healthcare practice? Can epistemic
vices be legitimately attributed to abstract objects, too, such as doctrines, stances,
or policies?
Relatedly, fourth, what are the features of group epistemic vice? Are there
epistemic vices that can only be had by groups or collectives, and not by individu-
als? Are the strategies for ameliorating vices had by a group different from the
strategies for ameliorating vices had by individuals? Does collective epistemic
vice extend into institutional vices? How does thinking about epistemic vices in
collective ways affect the methodology and agenda of vice epistemology – for
instance, does it require us to embrace feminist frameworks attentive to power
relations, social situation, and systems of oppression (Kidd 2020)? For these
issues, vice epistemologists should take guidance from earlier work in feminist
critical character theory and epistemology (see Daukas 2019, Dillon 2012).
Fifth, which epistemic vices contribute to the pollution of our online environ-
ment, via the creation and dissemination of ‘fake news’? And, which epistemic
vices do those polluted online environments, in turn, help to facilitate? Hot on
the heels of contemporary ‘social virtue epistemology’, there’s surely scope for
studies of the interrelations of vice epistemology and social epistemology – per-
haps in the direction of an anti-social epistemology. Some topics of concern will
be socially scaffolded processes of epistemic corruption, epistemic vices and the
media, ameliorative projects aimed at the social-epistemic structures, and the
development of effective methods for studying the social, political, and structural
dimensions of epistemically vicious conduct.
Sixth, how is vice epistemology connected to epistemic resistance and libera-
tion? Do some traditional epistemic virtues, such as intellectual humility, inhibit
epistemic resistance and liberation? If so, might intellectual arrogance count as
a liberatory epistemic virtue? Can vice epistemology be a source of social and
political activism?4 What are the connections between oppression, marginaliza-
tion, and epistemic character?
We hope this volume helps to inspire further work on epistemic vices and vice
epistemology along the lines indicated by these sorts of questions. We also hope
it inspires applications of the field to other areas, both in and outside of academic
philosophy.
Notes
1 There is a small literature on Confucian character epistemology, although most of it
focuses on the epistemic virtues, rather than the epistemic vices: see Kidd (2018a), Li
(2016), and Tsai (2014).
2 Three exemplary Buddhist catalogues of our failings from the Pali Canon are the
Sabbāsava Sutta in The Middle Length Discourses of the Buddha: A Translation of the
Majjhima Nikāya, 3rd ed., translated by Bikkhu Ñānamoli and Bhikkhu Bodhi (Boston,
MA: Wisdom Publications, 2009); the Udumbarika-Sihanāda Sutta in The Long
Discourses of the Buddha: A Translation of the Digha Nikāya, trans. by Maurice Walsh
(Boston, MA: Wisdom Publications, 1995); and Buddhaghosa’s Visuddmimagga (5th
century bce), trans. by Bhikkhu Nyanamoli (Kandy, Sri Lanka: Buddhist Publication
Society, 2011).
3 At a panel on epistemic vice at the 2016 meeting of the American Philosophical
Association Pacific Division, a delegate who missed the panel asked what it was about.
Upon being told the topics included arrogance and dogmatism, they replied, ‘Oh, polit-
ical epistemology, then’.
4 Within contemporary philosophy, these sorts of applied social and political issues tend
to dominate, but, in other cultures and historical periods, the vices of the mind were
worrisome for other reasons – as obstacles to our salvation, for instance.
References
Alfano, Mark, Colin Klein, and Jeroen de Ridder (eds.). Forthcoming. Social Virtue
Epistemology. London: Routledge.
Aristotle. 1998. Nicomachean Ethics (c. 350 BCE), trans. D. Ross. New York: Oxford
University Press.
Astell, Mary. 2002. A Serious Proposal to the Ladies. Parts I and II [1694], ed. Patricia
Springborg. Ontario: Broadview Literary Text.
Bacon, Francis. 2000. Novum Organum [1620], eds. Lisa Jardine and Michael Silverthorne.
Cambridge: Cambridge University Press.
Baehr, Jason. 2010. “Epistemic Malevolence,” Metaphilosophy 41(1–2): 189–213.
Baehr, Jason. 2011. The Inquiring Mind: On Intellectual Virtues and Virtue Epistemology.
Oxford: Oxford University Press.
Baehr, Jason (ed.). 2015. Intellectual Virtues and Education: Essays in Applied Virtue
Epistemology. London: Routledge.
Battaly, Heather (ed.). 2010. “Virtue and Vice: Moral and Epistemic,”. Metaphilosophy
41(1–2).
Battaly, Heather. 2010. “Epistemic Self-Indulgence,” Metaphilosophy 41(1–2): 214–34.
Battaly, Heather. 2013. “Detecting Epistemic Vice in Higher Education Policy: Epistemic
Insensibility in the Seven Solutions and the REF,” Journal of Philosophy of Education
47(2): 263–80.
Battaly, Heather. 2014. “Varieties of Epistemic Vice,” in Jon Matheson and Rico Vitz
(eds.), The Ethics of Belief. Oxford: Oxford University Press, 51–76.
Battaly, Heather. 2016. “Epistemic Virtue and Vice: Reliabilism, Responsibilism, and
Personalism,” in Chienkuo Mi, Michael Slote, and Ernest Sosa (eds.), Moral and
Intellectual Virtues in Western and Chinese Philosophy. New York: Routledge,
99–120.
Battaly, Heather. 2017. “Testimonial Injustice, Epistemic Vice, and Vice Epistemology,”
in Ian James Kidd, José Medina, and Gaile Pohlhaus (eds.), The Routledge Handbook
to Epistemic Injustice. New York: Routledge, 223–231.
Battaly, Heather. 2018. “Closed-Mindedness and Dogmatism,” Episteme 15(3): 261–282.
Battaly, Heather. 2019. “Vice Epistemology has a Responsibility Problem,” Philosophical
Issues 29(1): 24–36.
Biddle, Justin, Ian James Kidd, and Anna Leuschner. 2017. “Epistemic Corruption and
Manufactured Doubt: The Case of Climate Science,” Public Affairs Quarterly 31(3):
165–187.
Bloomfield, Paul. 2019. “Epistemic Intemperance,” American Philosophical Quarterly
56(2): 109–124.
Carel, Havi and Ian James Kidd. 2014. “Epistemic Injustice in Healthcare: A Philosophical
Analysis,” Medicine, Healthcare, and Philosophy 17(4): 529–40.
Cassam, Quassim. 2016. “Vice Epistemology,” The Monist 99(3): 159–180.
Cassam, Quassim. 2017. “Vice Ontology,” Social Epistemology Review and Reply
Collective 6(11): 20–27.
Cassam, Quassim. 2019a. Vices of the Mind: From the Intellectual to the Political. Oxford:
Oxford University Press.
Cassam, Quassim. 2019b. Conspiracy Theories. Cambridge: Polity.
Code, Lorraine. 1987. Epistemic Responsibility. Hanover, NE: University Press of New
England.
Confucius (Kǒngzı̌ ). 2003. Analects: With Selections from Traditional Commentaries,
trans. Edward Slingerland. Indianapolis, IN: Hackett.
Crerar, Charlie. 2018. “Motivational Approaches to Intellectual Vice,” Australasian
Journal of Philosophy 96(4): 753–766.
Daukas, Nancy. 2019. “Feminist Virtue Epistemology,” in Heather Battaly (ed.), The
Routledge Handbook of Virtue Epistemology. New York: Routledge, 379–391.
Desmond, William. 2014. Cynics. London: Routledge.
DeYoung, Rebecca Konyndyk. 2009. Glittering Vices: A New Look at the Seven Deadly
Sins and Their Remedies. Grand Rapids, MI: Brazos Press.
Dillon, Robin. 2012. “Critical Character Theory: Toward a Feminist Perspective on ‘Vice’
(and ‘Virtue’),” in Sharon L. Crasnow and Anita M. Superson (eds.), Out from the
Shadows: Analytical Feminist Contributions to Traditional Philosophy. New York:
Oxford University Press, 83–114.
Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford:
Fricker, Miranda. 2010. “Can There Be Institutional Virtues?,” in Tamar Gendler and John
Hawthorne (eds.), Oxford Studies in Epistemology, vol. 3. Oxford: Oxford University
Press, 235–52.
Fricker, Miranda. 2016. “Fault and No-Fault Responsibility for Implicit Prejudice,” in
Michael Brady and Miranda Fricker (eds.), The Epistemic Life of Groups. Oxford:
Hannon, Michael and Jeroen de Ridder (eds.). 2020. The Routledge Handbook to Political
Epistemology. New York: Routledge.
Holroyd, Jules and Daniel Kelly. 2016. “Implicit Bias, Character, and Control,” in Alberto
Masala and Jonathan Webber (eds.), From Personality to Virtue. Oxford: Oxford
University Press, 106–133.
Holroyd, Jules. 2017. “Responsibility for Implicit Bias,” Philosophy Compass 12(3).
doi:10.1111/phc3.12410
Keown, Damien. 1991. The Nature of Buddhist Ethics. London: Macmillan.
Kidd, Ian James. 2016. “Charging Others with Epistemic Vice,” The Monist 99(3):
181–197.
Kidd, Ian James. 2017. “Epistemic Vices in Public Debate: The Case of ‘New Atheism’,”
in Christopher Cotter and Philip Quadrio (eds.), New Atheism’s Legacy: Critical
Perspectives from Philosophy and the Social Sciences. Dordrecht, the Netherlands:
Springer, 51–68.
Kidd, Ian James. 2018a. “Confucianism, Curiosity, and Moral Self-Cultivation,” in Ilhan
Inan, Lani Watson, Dennis Whitcomb, and Safiye Yigit (eds.), The Moral Psychology
of Curiosity. New York: Rowman & Littlefield, 97–116.
Kidd, Ian James. 2018b. “Deep Epistemic Vices,” Journal of Philosophical Research 43:
43–67.
Kidd, Ian James. 2018c. “Is Scientism Epistemically Vicious?” in Jeroen de Ridder, Rik
Peels, and Rene van Woudenberg (eds.), Scientism: Prospects and Problems. Oxford:
Kidd, Ian James. 2019. “Epistemic Corruption and Education,” Episteme 16(2): 220–235.
Kidd, Ian James. 2020. “Epistemic Vices and Feminist Philosophies of Science,” in Kristen
Intemann and Sharon Crasnow (eds.), The Routledge Handbook to Feminist Philosophy
of Science. New York: Routledge.
Kidd, Ian James, José Medina, and Gaile Pohlhaus (eds.). 2017. The Routledge Handbook
to Epistemic Injustice. New York: Routledge.
Kivisto, Sari. 2014. The Vices of Learning: Morality and Knowledge in Early Modern
Universities. Leiden, the Netherlands: Brill.
Kotzee, Ben (ed.). 2013. Education and the Growth of Knowledge: Perspectives from
Social and Virtue Epistemology. Oxford: Wiley-Blackwell.
Lahroodi, Reza. 2019. “Virtue Epistemology and Collective Epistemology,” in Heather
Battaly (ed.), The Routledge Handbook of Virtue Epistemology. New York: Routledge,
407–419.
Li, Jin. 2016. “Humility in Learning: A Confucian Perspective,” Journal of Moral
Education 45(2): 147–165.
Locke, John. 1996. Some Thoughts Concerning Education and Of the Conduct of the
Understanding [1706], eds., Ruth W. Grant and Nathan Tarcov. Indianapolis, IN: Hackett.
Locke, John. 2008. An Essay Concerning Human Understanding [1690], ed. Pauline
Phemister. Oxford: Oxford University Press.
Lynch, Michael P. 2019. Know-It-All Society. New York: Liveright Publishing Corporation.
Manson, Neil C. 2012. “Epistemic Restraint and the Vice of Curiosity,” Philosophy 87:
239–259.
Medina, José. 2013. The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic Injustice, and Resistant Imaginations. Oxford: Oxford University Press.
Montmarquet, James. 1987. “Epistemic Virtue,” Mind 96: 482–497.
Nguyen, C. Thi. 2018. “Echo Chambers and Epistemic Bubbles,” Episteme. doi:10.1017/
epi.2018.32
Nyanatiloka. 2004. Buddhist Dictionary: Manual of Buddhist Terms and Doctrines, ed.
Nyanaponika. Kandy, Sri Lanka: Buddhist Publication Society.
Pascal, Blaise. 1980. Pensées [1670], trans. A. J. Krailsheimer. London: Penguin.
Paul, Herman and Jeroen van Dongen (eds.). 2017. Epistemic Virtues in the Sciences and
the Humanities. Dordrecht, the Netherlands: Springer.
Petkov, Kiril. 2012. “The Cultural Career of a ‘Minor’ Vice: Arrogance in the Medieval
Treatise on Sin,” in Richard G. Newhauser and Susan J. Ridyard (eds.), Sin in Medieval
and Early Modern Culture: The Tradition of the Seven Deadly Sins. York: York
Medieval Press, 43–84.
Pohlhaus Jr., Gaile. 2012. “Relational Knowing and Epistemic Injustice: Toward a Theory
of Willful Hermeneutical Ignorance,” Hypatia 27(4): 715–735.
Roberts, Robert C. and W. Jay Wood. 2007. Intellectual Virtues: An Essay in Regulative
Epistemology. Oxford: Oxford University Press.
Roberts, Robert C. and W. Jay Wood. 2019. “Understanding, Humility, and the Vices of
Pride,” in Heather Battaly (ed.), The Routledge Handbook of Virtue Epistemology. New
York: Routledge, 363–375.
Sherman, Benjamin R. and Stacey Goguen (eds.). 2019. Overcoming Epistemic Injustice.
Lanham, MD: Rowman & Littlefield.
Smith, Richard. 2016. “The Virtues of Unknowing,” Journal of Philosophy of Education
50(2): 272–284.
Sosa, Ernest. 1980. “The Raft and the Pyramid,” Midwest Studies in Philosophy 5(1): 3–26.
Swank, Casey. 2000. “Epistemic Vice,” in Guy Axtell (ed.), Knowledge, Belief, and
Character: Readings in Virtue Epistemology, Lanham, MD: Rowman and Littlefield
Publishers, 195–204.
Tanesini, Alessandra. 2016. ““Calm Down Dear”: Intellectual Arrogance, Silencing and
Ignorance,” Aristotelian Society: Supplementary Volume 90(1): 71–92.
Tanesini, Alessandra. 2018a. “Intellectual Servility and Timidity,” Journal of Philosophical
Research 43: 21–41.
Tanesini, Alessandra. 2018b. “Epistemic Vice and Motivation,” Metaphilosophy 49(3):
350–367.
Tsai, Cheng-Hung. 2014. “Xunzi and Virtue Epistemology,” Universitas 41(3): 121–142.
Wollstonecraft, Mary. 1995. A Vindication of the Rights of Men with A Vindication of the
Rights of Woman and Hints [1792], ed. Sylvana Tomaselli. Cambridge: Cambridge
University Press.
Zagzebski, Linda. 1996. Virtues of the Mind: An Inquiry into the Nature of Virtue and the
Ethical Foundations of Knowledge. Cambridge: Cambridge University Press.
Part I
Foundational issues
1 The structure of intellectual vices
Jason Baehr
Much of virtue epistemology turns on an insight concerning the relationship

between intellectual character and epistemic achievement. The insight is that
many prized epistemic accomplishments are underwritten by the intellectual
character strengths of the persons who make them. Ground-breaking scientific
discoveries, brilliant philosophical treatises, and illuminating histories frequently
manifest the curiosity, intellectual autonomy, intellectual humility, and related
attributes of their authors. These qualities are intellectual virtues; they are excel-
lences of intellectual character.
Intellectual vices, on the other hand, are defects of intellectual character, such
as intellectual laziness, intellectual cowardice, narrow-mindedness, closed-mind-
edness, and dogmatism. My focus in this chapter is on the structure of intellectual
vices. In particular, I am interested in whether the structure of intellectual vices
mirrors that of intellectual virtues. As such, this chapter is intended as a contribu-
tion to what might be called “foundational vice epistemology.”1
In the first section, I provide an overview of the structure of intellectual virtues.
Next, I consider whether intellectual vices exhibit a symmetrical structure. This
consideration leads, in the final section of the chapter, to an assessment of “moti-
vationalism” concerning intellectual vice.
1.1 The structure of intellectual virtues

I begin with some general features of intellectual virtues as I understand them.
First, intellectual virtues are strengths of intellectual character. In Aristotelian
terms, they are dispositions to act, think, and feel in particular (rational or excel-
lent) ways. Like Aristotle’s moral virtues, they also occupy a midpoint between
two (or more) extremes. Intellectual humility, for instance, can be understood as a
mean between intellectual arrogance (deficiency) and intellectual self-deprecation
(excess). Open-mindedness is a mean between closed-mindedness (deficiency)
and gullibility (excess). And intellectual courage is a mean between intellectual
cowardice (deficiency) and intellectual recklessness (excess).
Second, intellectual virtues can be differentiated from (familiar or straight-
forwardly) moral virtues on the basis of their ultimate aim or object, which is
distinctively epistemic. Intellectual virtues flow from something like a “love” or
22 Jason Baehr
intrinsic concern for epistemic goods, such as truth, knowledge, understanding,
and wisdom. They are, as James Montmarquet notes, the character traits that a
“truth-desiring person would want to have” (1993: 30).
Third, intellectual virtues contribute to their possessor’s “personal worth.”
This does not mean that intellectually virtuous persons are “worth more” or pos-
sess a greater inherent dignity than persons who lack intellectual virtues. Rather,
the idea is that intellectual virtues make us excellent or admirable as persons.
There is no contradiction in thinking, as many do, that while all persons share a
common worth or value qua persons, some are better persons or better qua per-
sons than others. My notion of personal worth corresponds to the latter but not the
former type of value. The fact that intellectual virtues contribute to their posses-
sor’s personal worth is explainable (at least partly) in terms of their motivational
basis. That is, a “love” of or intrinsic concern with epistemic goods is personally
admirable or excellent; it reflects well on its possessor qua person.2
In previous work (2015), I have defended a “four-dimensional” account of the
structure of intellectual virtues, according to which each intellectual virtue has
a competence dimension, a motivational dimension, a judgment dimension, and
an affective dimension. These dimensions are not necessarily discrete parts or
components of an intellectual virtue. The model is consistent with the possibility
that, say, two dimensions of an intellectual virtue might have their basis in a single
constitutive part. Nevertheless, as I hope to demonstrate, it remains theoretically
useful to distinguish between all four dimensions.
1.1.1 Competence dimension

I turn now to a brief overview of this model. First, for every intellectual virtue V,
there is a skill, activity, or competence characteristic of V on the basis of which
V can be distinguished from other intellectual virtues. To possess an intellectual
virtue—open-mindedness, say—is to be skilled or competent at a certain sort of
intellectual activity, for example, at perspective-switching. Similarly, curiosity
involves competence at asking thoughtful and insightful questions. And intel-
lectual humility involves the skill of attending to and owning one’s intellectual
limitations. Moreover, it is on the basis of these characteristic competences or
skills that we are able to distinguish one intellectual virtue from another. While
all intellectual virtues have a common motivational basis (a “love” of epistemic
goods), they differ one from another in respect of their characteristic competences
or skills.
1.1.2 Motivational dimension

However, being skilled at perspective-switching is not sufficient for possessing
the virtue of open-mindedness. Nor is being competent at formulating thoughtful
and insightful questions sufficient for the virtue of curiosity. For one can pos-
sess such skills but be unmotivated to use them, and thereby fail to possess the
virtues in question. Accordingly, to possess an intellectual virtue V, one must be
The structure of intellectual vices 23
motivated to engage in the activity, deploy the skill, or manifest the competence
proper to V—and be motivated to do so (at least partly) out of an intrinsic concern
for or “love” of epistemic goods. This requirement is consistent, of course, with
the possibility—indeed the reality—that intellectually virtuous persons often care
about or pursue knowledge partly for instrumental reasons.
1.1.3 Judgment dimension

This might appear to be a more or less complete picture. If a person is capable
of perspective-switching, and motivated to perspective-switch, what could pre-
vent him from possessing the virtue of open-mindedness? A possible reply is
that he might lack good judgment about when or for how long or in what way to
perspective-switch. As such, he might regularly fail to hit the “mean” with respect
to perspective-switching. Therefore, to possess an intellectual virtue V, one must
have good judgment with respect to when, how often, in what amount, toward
whom, and so on, to manifest the competence proper to V. Put another way, every
intellectual virtue contains an element of phronesis or practical wisdom.
1.1.4 Affective dimension
A further dimension of intellectual virtues also merits attention. We can come
at it by considering Aristotle’s notion of enkrateia or continence (NE.VII). For
Aristotle, the continent agent is one who reliably acts virtuously, and does so out
of sound ethical judgment, but who fails to enjoy or take appropriate pleasure in
so acting. On his view, continent persons fall short of moral virtue.
My view of intellectual virtues stands in partial contrast to Aristotle’s view of
moral virtues. I am inclined to think that a robustly enkratic person can be mini-
mally intellectually virtuous, particularly if the person satisfies the motivational
requirement on intellectual virtue. An inquirer with an unstinting commitment to
truth, who reliably and intelligently manifests virtuous intellectual competences,
but takes no pleasure in doing so, seems to me to exhibit personal excellence suf-
ficient for minimal virtue. That said, I also find plausible Aristotle’s claim that
pleasure “completes [virtuous] activity … as a sort of supervenient end” (NE,
1175a). Put another way, I think pleasure and other appropriate affections are
necessary for the possession of full virtue.
More precisely, I maintain that to possess an intellectual virtue V in its fullness,
one must be disposed to manifest the affective or feeling states proper to V. This
entails neither that a fully virtuous person always enjoys manifesting her virtue-
relevant competence, nor that the relevant affective or feeling states are limited
to pleasure, delight, etc. The intellectually courageous person who puts herself
in harm’s way in order to discover or communicate the truth may rarely (if ever)
enjoy doing so. Instead, her virtue is more likely to be manifested in feelings of
confidence or self-control.3
A final observation about these dimensions is that they can be instanti-
ated to a greater or lesser degree. A person can be more or less competent at
24 Jason Baehr
perspective-switching, have better or worse judgment about when to perspective-
switch, or enjoy perspective-switching to a greater or lesser extent. Accordingly,
intellectual virtue possession is itself a matter of degree: once a certain threshold
is met, minimal virtue is attained; however, for any intellectual virtue, there may
be a significant developmental or normative distance between a minimal posses-
sion of the virtue and its full or maximal possession.
To summarize the discussion up to this point: intellectual virtues (1) are
strengths of intellectual character that (2) contribute to personal worth, (3) are
rooted in a concern with or “love” of epistemic goods, (4) have at least four
dimensions, and (5) are possessed in degrees.
1.2 The structure of intellectual vices

Suppose the structure of intellectual virtues is more or less as described. And
suppose we are also interested in understanding the structure of intellectual vices.
Several questions come to mind, including: do intellectual vices exhibit an analo-
gous four-part structure? More precisely, to possess an intellectual vice, must one
be defective along all four dimensions of an intellectual virtue? Or might it be suf-
ficient that one is defective along just one (or two or three) of these dimensions?4
I begin with three brief points about intellectual vices as I understand them.
First, like intellectual virtues, they are attributes of intellectual character; more
specifically, and by contrast with intellectual virtues, they are defects of intellec-
tual character. Second, like intellectual virtues, they can be possessed to a greater
or lesser extent. Third, intellectual vices make a negative contribution to personal
worth. They reflect negatively on their possessor qua person.
Let us begin with the following conjecture: one possesses an intellectual vice
only if one is defective with respect to all four dimensions of one or more intel-
lectual virtues. On this view, the structure of intellectual vices mirrors that of
intellectual virtues. To possess an intellectual vice, one must exhibit defective
competence, motivation, judgment, and affection.5
In contrast with this claim, recall Aristotle’s remark that while “it is possible to
fail in many ways … to succeed is possible only in one way … For men are good
in but one way, but bad in many” (NE, II.6). This suggests that our conjecture may
be too strong—that while several things must go right for the development of an
intellectual virtue, relatively few things need to go wrong for the possession of an
intellectual vice. Indeed, I think something like Aristotle’s view is applicable to
intellectual virtues and vices.
1.2.1 Defective motivation

To make good on this claim, I begin by noting that defective intellectual motiva-
tion can be sufficient for the possession of an intellectual vice. Imagine a person
who possesses a wide range of virtue-relevant competences and a reasonably good
sense of when (etc.) these competences should be manifested, but who is indiffer-
ent to truth or knowledge, and therefore systematically refrains from thinking or
inquiring in intellectually virtuous ways.6 Such a person might be intellectually
lazy, and his laziness might be an intellectual vice. A similar point holds for intel-
lectual carelessness. I might be aware of what would be required for giving an
accurate characterization of a complex philosophical position, and have the ability
to make such a characterization, but out of carelessness fail to do so. If I do this
often enough or across a wide enough range of contexts, I might possess the vice
of intellectual carelessness.7 Here a deficiency of virtuous epistemic motivation
appears to be sufficient for the possession of an intellectual vice.8
Consider, in addition, the vice of epistemic malevolence, which I elsewhere
describe as opposition to the epistemic good (or to epistemic goods) as such
(2010: 204). Here as well someone might be capable of listening and responding
to his interlocutors in open-minded and intellectually generous ways, and have a
good sense of when (etc.) to do so, yet systematically refrain from manifesting
these abilities on account of his malevolence. In this case, a positively bad epis-
temic motivation would be sufficient for the possession of an intellectual vice.9
A variation on this case is worth considering. Imagine an epistemically malev-
olent agent who is skilled at open-minded and empathic thinking and who conniv-
ingly deploys these skills in an effort to sow cognitive dissonance and ignorance
among his interlocutors. On my view, if open-minded and empathic thinking are
sufficiently ingrained in the person’s way of thinking and reasoning, it might be
accurate to say that he has the traits or attributes of open-mindedness and intellec-
tual empathy. Intuitively, however, these traits would not be intellectual virtues in
the present sense. Indeed, they would appear to be vices, making their possessor
a worse person or worse qua person. The explanation for this, again, consists in
the agent’s epistemically malevolent motivation. This leads to the surprising (but
I think ultimately correct) view that the sorts of traits we often think of as intellec-
tual virtues (e.g. open-mindedness and cognitive empathy) can, on motivational
grounds, turn out to be intellectual vices.10
I note two findings from the discussion thus far:
Finding #1: defective epistemic motivation (whether positive or negative)

can be sufficient for the possession of an intellectual vice.11
Finding #2: if ill-motivated, traits we tend to think of as intellectual virtues
can be intellectual vices.
1.2.2 Defective judgment

Can defective practical judgment also be sufficient for the possession of an intellec-
tual vice? There is at least some reason to think it can. Recall the possibility, noted
in the previous section, that a person might be skilled at perspective-switching,
and be motivated to use this skill, but have poor judgment about when, how often,
toward whom, etc., she should do so. Such a person might, as a result of her lack
of judgment, end up being either closed-minded (if her judgment is too restrictive)
or gullible (if her judgment is too permissive). Similarly, a person might be skilled
26 Jason Baehr
at and motivated to resist her fearful impulses in an epistemic context while being
indiscriminate about when or how often she should manifest this skill. As a con-
sequence, she might become intellectually reckless or foolhardy.
However, it is important to ask whether the kind of closedmindedness, gul-
libility, and intellectual recklessness just alluded to necessarily would be genuine
intellectual vices. Whether they would be vices depends, it seems, on whether
the person in question can (in some sense) be considered responsible for or in
relation to her lack of virtuous judgment. Suppose, for instance, that her lack of
judgment is a function of her (entirely inculpable) unfamiliarity with features of
her epistemic environment, such that she lacks an accurate sense of when per-
spective-switching is (or isn’t) likely to be epistemically appropriate. Or suppose
her lack of judgment is due to an innate and intractable (and therefore inculpable)
cognitive disability. In cases like this, the attributes in question would not reflect
negatively on their possessor qua person; as such, they would not count as intel-
lectual vices, on the present view.12
Later in the chapter we will consider some possible implications of this point
for our understanding of the relationship between the judgment and motivational
aspects of intellectual vice. In the meantime, we may draw the following, fairly
modest conclusion concerning defective judgment and intellectual vice:
Finding #3: defective judgment can be sufficient for the possession of an

intellectual vice only if the person in question is (in some sense) responsible
for this defect.
Though relatively modest, this finding raises the thorny question of exactly what
sense or type of responsibility is required in order for a defect to count as an
intellectual vice. Some argue that for a feature of an agent to contribute to his
possession of an intellectual virtue or vice, the agent himself must be responsible
for this feature in the sense of having cultivated or brought it about (Zagzebski
1996: 104–105, 116–125). I cannot address this or similar claims within the lim-
its of this chapter. Instead, I will simply register that such a view strikes me as
objectionably narrow. I heartily endorse Heather Battaly’s recent recommenda-
tion that vice epistemologists pay closer attention to “non-voluntarist” accounts
of moral responsibility (2019: 9–10).13 In keeping with this, I suggest that when
it comes to thinking about ways in which defective judgment might contribute to
intellectual vice, we would do well to opt for a wide rather than a narrow concept
of responsibility.
1.2.3 Defective competence

What, then, about the competence dimension of an intellectual vice? Can a person
possess an intellectual vice strictly on account of lacking the skill or competence
proper to one or more intellectual virtues? To answer this question, we need to try
to imagine a case in which a person lacks the competence proper to a given virtue
despite being motivated to manifest this competence and having good judgment
about when (etc.) to do so. There is, it seems, a psychological implausibility about
such a scenario. If a person is genuinely motivated, say, to pay careful attention to
important details, or to avoid careless mistakes, and if this person has good judg-
ment about when (etc.) to do these things, why wouldn’t he also be competent to
do them? What might explain his incompetence, if not defective motivation or
judgment?
This underscores a notable feature of the competences proper to intellectual
virtues: namely, that they are not, at least in some of their most basic forms, par-
ticularly technically demanding.14 As a general rule, if a person really wants to
pay attention to important details, avoid careless mistakes, or acknowledge her
intellectual limitations, she is able to do so. In this respect, virtue-specific compe-
tences differ from many other cognitive competences or skills, the cultivation of
which can take a great deal of practice and training.
That said, I think we can successfully identify cases in which a person is genu-
inely motivated to manifest a virtue-relevant competence, has a good sense of
when (etc.) to manifest this competence, but is incapable of doing so. Here too
a natural cognitive disability might prevent someone from, say, attending to or
focusing on important details, even if the person is robustly motivated to do other-
wise. Or, consider the sort of competence characteristic of intellectual autonomy,
which involves thinking for oneself, drawing one’s own conclusions, and so on.
A person who has been brainwashed, or who has been raised in an extremely
epistemically deferential environment, might see the value of thinking in intel-
lectually autonomous ways, desire very much to think in these ways, have a good
sense of when (etc.) to do so, but still find himself without the skill or ability to
think for himself.
Note, however, that in the sorts of scenarios just described, the persons in ques-
tion are unlikely to merit a charge of intellectual vice. Their epistemic defects are
unlikely to reflect negatively on them qua persons.15 And the reason, it seems,
is that they would not appear to be responsible (even in a wide sense) for their
lack of virtuous competence. Might there, then, be other cases in which a person
satisfies the motivational and judgment conditions for a particular virtue but still
fails (culpably) to satisfy the competence condition and consequently possesses
an intellectual vice? This is a tricky issue, and one to which we will return below.
Presently, we may draw a conclusion regarding competence and intellectual
vice that parallels the conclusion regarding good judgment:
Finding #4: a lack of virtue-relevant competence can be sufficient for the

possession of an intellectual vice only if the person in question is (in a broad
sense) responsible for this defect.16
Before turning to the affective dimension, I want to consider a further point related
to competence and intellectual vice. We have been considering whether a failure
to possess a virtue-relevant competence might be sufficient for the possession
of an intellectual vice. We have yet to consider whether intellectual vices might
involve competences of their own.
28 Jason Baehr
We have already found that it is not a necessary feature of intellectual vices
that each vice have a unique or characteristic competence. For, again, a mere lack
of virtue-relevant skill can be sufficient for the possession of a vice, provided
that the person is (broadly) responsible for this deficiency. Nevertheless, it would
appear that at least some vices are such that, to possess them (or to possess them
fully), one must possess certain skills characteristic of these vices. This includes
skills associated with self-deception, keeping counter-evidence at bay (intellec-
tual dishonesty), misrepresenting opposing viewpoints (closed-mindedness), and
the like. Plausibly, with the right sort of practice or training, one can become
“better”—more competent at—these mental operations. Moreover, such compe-
tences are expressive of the vices in question. A closed-minded person who is
especially skilled at misrepresenting to herself the content or plausibility of other
people’s views is more closed-minded than a closed-minded person who is still in
the process of acquiring this skill. This leads to a further conclusion concerning
competence and intellectual vice:
Finding #5: there are intellectual vices such that, to possess them fully or
maximally, one must possess certain skills or competences proper to these
vices.
1.2.4 Defective affection
On, then, to the affective dimension of intellectual virtues. Can one’s failure to
manifest the affections proper to an intellectual virtue be sufficient for the pos-
session of an intellectual vice? Here there would appear to be an asymmetry with
the dimensions considered above. Consider again Aristotle’s enkratic or “conti-
nent” agent. According to Aristotle, while this person is not virtuous, neither is
he vicious.
A similar point holds, I would suggest, with respect to epistemic continence.
A person who, say, reliably probes for deeper understanding (thoroughness), but
who never enjoys the activity of thorough inquiry, while less than fully virtuous in
respect of his thoroughness, surely is not intellectually vicious. Similarly, some-
one who reliably asks thoughtful and insightful questions, but takes no delight in
such wonderment, while less than fully virtuously inquisitive, does not, as such,
manifest an intellectual vice.
It remains, however, that certain characteristically vicious affections can com-
pound or amplify the badness of one’s vices. For instance, a person who enjoys
his chronic intellectual laziness or sloppiness, delights in cognitive distortion or
obfuscation, or takes pleasure in concealing his intellectual mistakes or misrep-
resenting others’ views, would be more vicious than one who performs the same
intellectual actions but derives no pleasure from them. We may, then, draw the
following conclusions concerning affections and intellectual vices:
Finding #6: a deficiency in virtuous affections is not, by itself, sufficient for

the possession of an intellectual vice.
Finding #7: vicious affections are necessary for the full or maximal posses-
sion of at least some intellectual vices.
1.3 Motivationalism
We have found that the structure of intellectual vices is not symmetrical with that
of intellectual virtues in the sense that, to possess an intellectual vice, a person
need not be defective across all four dimensions of an intellectual virtue. In the
present section, I consider the implications of our findings for “motivationalism”
about intellectual virtue, which I here take to be the thesis that defective intel-
lectual motivation is a necessary condition for the possession of an intellectual
vice.17 On the one hand, we have found that the most obvious or straightforward
way a person can be intellectually vicious is motivational in nature: viz. by failing
to care sufficiently about epistemic goods (e.g. intellectual laziness or careless-
ness) or by being outright opposed to them (e.g. epistemic malevolence). While
this might appear to tell in favor of motivationalism, we have also encountered
some evidence for thinking that a lack of virtuous competence or judgment can
be sufficient for the possession of an intellectual vice (provided that the person is
broadly responsible for this defect).
1.3.1 Motivation and responsibility

To better understand whether the proposed model of intellectual vice is compat-
ible with motivationalism, we must return to the relevant notion of responsibility.
Specifically, we must consider: on account of what might a person be (relevantly,
broadly) responsible for his lack or virtuous judgment or competence?
One salient possibility underscores a potential connection between respon-
sibility and motivation. Imagine someone whose gullibility is a function of her
culpable lack of good judgment about when (etc.) it is appropriate to perspec-
tive-switch. Again, what might explain this lack of good judgment? One obvi-
ous possibility is that the person has not cared much or has been insufficiently
concerned with developing an accurate sense of when her tendency to perspec-
tive-switch might lead her astray. Here, the person’s lack of good judgment
clearly has a motivational source (it is rooted in her lack of care or concern). If
this is how we should think about vice-constituting defective judgment across
the board, then the account of intellectual vice sketched above may be consist-
ent with motivationalism.18 For again, the suggested picture is that the kind
of defective judgment that can contribute to intellectual vice necessarily has a
motivational basis.
Consider as well a scenario involving a (culpable) lack of virtuous competence.
Imagine a person who is motivated to think for himself, has a reasonably good
sense of when he should do so, but has yet to develop this ability. Suppose, fur-
ther, that he really ought to have developed the ability to think for himself at this
point—that his failure to have done so is not a function of cognitive disability,
insufficient opportunities to practice thinking for himself, or the like. What might
30 Jason Baehr
explain this failure? Here as well it is tempting to think in motivational terms, for
example, in terms of an insufficient interest in or concern with learning to think
for himself or with the epistemic goods such thinking can deliver.
These cases suggest a potential problem with our earlier attempts (in connec-
tion with findings #3 and #4) to imagine a person who is motivated to manifest
a virtue-relevant competence out of an intrinsic concern for epistemic goods, yet
fails to possess either the competence itself or good judgment about when (etc.)
the competence should be manifested. While these initially seemed like genuine
possibilities, we now have some reason to doubt that they are. Specifically, the
cases just considered underscore the possibility that any (culpable, vice-constitut-
ing) failure to possess the competence or judgment proper to a virtue will entail
a prior failure of virtuous motivation.19 If this were right, then motivationalism
about intellectual vice apparently would be in the clear, for it would turn out that
a (culpable) deficiency of virtuous competence or judgment is sufficient for intel-
lectual vice only given a corresponding motivational failure.20
But is it right to think that a (culpable) failure of virtuous competence or judg-
ment entails a failure of virtuous epistemic motivation? I am uncertain about
this. It depends, at a minimum, on how exactly we understand the motivational
dimension of intellectual virtues. This dimension was characterized above as fol-
lows: to possess an intellectual virtue V, one must be motivated to engage in the
activity, deploy the skill, or manifest the competence proper to V—and to do so
(at least partly) out of an intrinsic concern for or “love” of worthy epistemic
ends. Understood in this way, it seems doubtful that instantiating the motivational
dimension of an intellectual virtue precludes a prior culpable failure to instantiate
at least the judgment dimension of that virtue.21
To see why, consider again someone who satisfies the motivational condition
with respect to the competence of perspective-switching. Ex hypothesi this person
is motivated to perspective-switch and her motivation is rooted at least partly
in an intrinsic concern with epistemic goods. Why should this particular state
rule out the possibility that, as a result of a kind of inattentiveness or intellectual
laziness, the person might (culpably) have failed to develop an accurate sense of
when her disposition to perspective-switch is likely to lead her astray? Why must
a sincere and robust motivation to take up alternative perspectives, even when
rooted in an intrinsic concern with epistemic goods, rule out this other sort of
motivational deficiency (viz. a deficiency that could explain the person’s lack of
good judgment about her disposition)?22
I see no reason to think that it must. However, to my mind, this does not settle
the question of whether motivationalism is correct about intellectual vice. Rather,
I think the more salient question is whether the foregoing characterization of the
motivational dimension is too narrow. Again, if a person fails, as a result of inat-
tentiveness or intellectual laziness, to develop good judgment concerning when
a virtue-relevant competence might lead her away from the truth, there is a clear
sense in which the person is insufficiently concerned with or motivated by truth.
In Robert Adams’s (2006) terms, it suggests that the person is not sufficiently
“for” epistemic goods.
This underscores the need for a broader characterization of the motivational
dimension of virtue. It also underscores the possibility that motivationalism might
still be correct about intellectual vice. For, again, it suggests that a motivational
deficiency underlies the sorts of failures in judgment and competence which in
turn contribute to the possession of an intellectual vice.
A full defense of this claim would require further reflection on the question of
whether it is possible to possess a culpable or vice-constituting defect of judgment
or competence that is not rooted in a defect of epistemic motivation. That is, what
alternative explanation might there be of why a person possesses a genuinely
vicious defect of judgment or competence? It is not immediately clear what such
an alternative explanation might be. However, the question merits more attention
than I am able to give it here.
1.2.3 Purported counterexamples to motivationalism

In the remainder of the chapter, I want to look briefly at five cases that have been
raised in the literature against motivationalist accounts of intellectual vice and to
sketch some initial reasons for thinking that none presents a very clear or intrac-
table problem for these accounts.23
The first is the character of Oblomov, from the Ivan Goncharov novel of the
same name, as discussed by Charlie Crerar (2018: 757f). As Crerar describes him,
Oblomov is “a parody of a lazy young nobleman, who is almost totally incapable
of making any decisions or undertaking any actions.” He is, Crerar says, “a limit
case for laziness and incuriosity … that reflect badly upon him, and for which
he seems blameworthy.” Yet, his vices “are motivated, not by an active desire to
avoid knowledge or to remain ignorant, but by an utter indifference to the two”
(757).
I have no objection to this assessment. While it may present a problem for ver-
sions of motivationalism according to which the possession of an intellectual vice
requires the possession of positively bad epistemic motives, I agree with Crerar
that such views are too restrictive.24 Again, on the view put forth here, a person
can possess an epistemic vice on account of being positively against epistemic
goods (epistemic malevolence) or by being insufficiently for them (intellectual
laziness and carelessness).
The second case is that of Galileo. Drawing on a characterization from Bob
Roberts and Jay Wood (2007: 254), Crerar describes Galileo as being “ultimately
motivated by epistemic goods” but also an “arrogant genius.” Galileo’s sense of
“his own intellectual superiority” leads to his being “closed-minded in his interac-
tions with others” (2018: 750). Crerar seems not to take seriously enough the pos-
sibility that Galileo might be virtuous in certain respects but vicious in others, that
is, that he might be virtuous in respect of his desire for truth, but vicious in respect
of his intellectual arrogance. The critical question is how to understand Galileo’s
arrogance and closed-mindedness, which are manifested in his judgment that he
has nothing to learn from his interlocutors. Ex hypothesi Galileo’s reasoning is
irrational and unreasonable.25 On account of what, then, might his sense of his
32 Jason Baehr
own intellectual superiority lead him, mistakenly and irrationally, to ignore or
minimize the potential epistemic contributions of others? One natural possibility
is that his abilities and prior intellectual achievements have gone to his head, such
that he has become less thoughtful about or attentive to his intellectual limitations
or to how he might learn from others. As this possibility illustrates, it is not at all
difficult to imagine that Galileo’s poor judgment might have a motivational basis.
And if it does have such a basis, then the defender of motivationalism can agree
that Galileo is vicious in respect of his arrogance and closedmindedness (while
also acknowledging that he is, perhaps, virtuous to the extent that he continues to
desire the truth).
A similar response applies to a third case discussed by Crerar (2018: 759–760).
Dave is a politician whose extremely privileged upbringing has “bestowed upon
him a flawed understanding of what constitutes an intelligent and reliable person”
(759). As a result, he believes that the “only people worth listening to” are people
with backgrounds like his own. Once elected to office, Dave surrounds himself
with and listens exclusively to the perspectives of people who satisfy his elitist
criteria. He is “closed-minded … prejudiced, partial, and a snob” (760). Yet, says
Crerar, he is also well-motivated on account of his desire to ascertain the truth
about the likely outcomes of his political decisions (760).
This good motivation notwithstanding, Dave’s thinking continues to manifest
closed-mindedness, prejudice, partiality, and snobbery.26 If so, I think we can be
confident that Dave ought to have known better than to categorically dismiss the
potential epistemic contributions of anyone other than himself and his elite associ-
ates.27 And here again it seems plausible to think of this shortcoming in motiva-
tional terms. Had Dave been more conscientious in his reasoning, had he cared
more about making accurate judgments concerning whose perspectives might
have genuine merit, he would not have exemplified intellectual vice.
A fourth case is discussed by Alessandra Tanesini (2018: 356).28 It features
Olivia, a conspiracy theorist who “seems to be motivated by the truth and yet is
intellectually vicious.” On the one hand, Olivia “does not want to believe conspir-
acies come what may; she only wants to believe them if they are true.” However,
she “also wants the conspiracies to be true, and it is this motivation that guides
her inquiries.” As a result, “she has a blind spot; she is closed-minded about
the falsity of her pet theories” (356). As with Galileo and Dave, Olivia appar-
ently has mixed epistemic motives. While caring about truth in certain respects,
or to a certain extent, the fact that she “wants the conspiracies to be true” and
that this desire “guides her inquiries” makes clear that Olivia is motivationally
deficient. Accordingly, this case also appears to be consistent with the core of
motivationalism.29
A fifth and final case comes from Tanesini (2018) via James Montmarquet
(1993: 25). Gail is concerned, with good reason apparently, that she is too gul-
lible. Out of “a desire for what is intellectually good,” she attempts to correct
this defect by being less open to the views of others. Over time, she develops “a
tendency not to listen to contrary views” and “thus turns herself into a dogmatic
person” (356). This case can be handled in either of a couple of ways. If, for the
duration of the time in question, Gail does her level best to compensate for her
natural gullibility, then while she likely will not exhibit a defective motivation,
neither will her “dogmatism” (if in fact it can be described as such) be a genuine
intellectual vice. Tanesini makes a similar assessment, commenting: “If we were
to ask [Gail] why she behaves as she does, she would make reference to her gul-
lible temperament, which requires her to be particularly active in counterbalanc-
ing this tendency” (261). On the other hand, if Gail’s counterbalancing begins
in this sensible and well-motivated way, but over time morphs into unthinking
dismissiveness and dogmatism, then while Gail might eventually be intellectually
vicious, it will, at this later point, be questionable whether her counterbalancing is
well-motivated, that is, whether it is really driven any longer by an earnest attempt
to improve the accuracy of her beliefs. Either way, the case appears not to pose a
serious problem for motivationalism.
1.4 Conclusion
We began by asking whether the structure of intellectual vices is symmetrical
with that of intellectual virtues. In response, we quickly encountered an asym-
metry. Minimally, a person need not be defective along the judgment or compe-
tence dimensions of an intellectual virtue in order to possess an intellectual vice.
Defective epistemic motivation can, by itself, form the basis of an intellectual
vice. We also considered some initial evidence for thinking that both defective
judgment and defective competence can form the basis of an intellectual vice.
On closer inspection, however, we were led to the possibility that any vice-mak-
ing failures of judgment or competence might presuppose a motivational failure.
Given certain connections between vice, motivation, and responsibility, we con-
cluded that before trying to settle this issue, vice epistemologists would do well
to consult the literature in moral psychology on “non-voluntarist” accounts of
responsibility. These points also led to a consideration of “motivationalism” about
intellectual vice. Here we found that while several recent objections to motiva-
tionalism appear to come up short, an adequate defense of this view requires
closer attention to the exact nature and scope of the motivational dimension of
intellectual virtues.30
Notes
1 This label is drawn from Ian James Kidd’s division of the field (2017: 1). For a recent
substantial overview and contribution to this vice epistemology, see Cassam (2019).
2 It does not necessarily follow that intellectual virtues are moral virtues, at least in any
narrow sense of “moral.” For it may be that personal worth or excellence has both
moral and intellectual dimensions. For more on these points, see my (2011: Ch. 6) and
Adams (2006).
3 Aristotle makes a similar point about courage (simpliciter) in NE.III.6.
4 By “sufficient” here my concern is with the possibility that a person might possess a
vice strictly on account of being defective along (say) just a single dimension of a given
virtue, that is, while possessing the features proper to the virtue’s other dimensions.
34 Jason Baehr
5 Per note 4, by “exhibit defective competence, motivation, judgment, and affection,” I
mean fail to possess virtuous competence, motivation, judgment, and affection.
6 As I do with this imagined (extremely sketchy) scenario, I will mostly be bracketing a
concern with the affective dimension in my discussion of the remaining three dimen-
sions. I do this merely to minimize the complexity of the examples.
7 I say “might” here and in the preceding sentence in light of the fact that the defective
motivation might need to satisfy other conditions—e.g. it might need to be something
that the person in question is (broadly) responsible for. More on this point below.
8 Though I will not pause to explore this point here, epistemic motivation can also be
defective on account of being excessive. For example, an excessive concern for accu-
racy or for certainty (both epistemic goods) might lead to a kind of hyper-carefulness
or epistemic scrupulosity. As this point suggests, I am not here purporting to identify
the full range of ways in which defective motivation can contribute to intellectual vice.
Thanks to Josh Dolin for raising this issue with me.
9 Per note 4 above, it would be sufficient in the sense that he need not (and indeed does
not) lack the corresponding virtuous competence or judgment. Nevertheless, might it
be necessary that he possess certain malevolence-specific competences? I will have
more to say about vice-specific competences momentarily. However, while I cannot do
justice to this issue here, my own sense is that such competences are not required for
the minimal possession of epistemic malevolence (though they likely are required for
its full possession).
10 For a helpful development of this point, see (Battaly 2018).
11 Note that I am not claiming that defective epistemic motivation is categorically or
always sufficient for the possession of an intellectual vice—only that it can be. And,
again, by “sufficient” I mean that a person with defective epistemic motivation can be
vicious even if he doesn’t lack—that is, even if he possesses—the corresponding virtu-
ous competence, judgment, etc. Again, see note 4 above. This leaves open the possibil-
ity that a person might, say, have defective epistemic motivation without meeting the
threshold for the possession of an intellectual vice. Related to this, it also leaves open
the possibility that virtues and vices should be understood as contraries vs. contradicto-
ries. Thanks to Heather Battaly for pressing me on this point.
12 I think something similar should be said about a person who, say, is born and raised
in a community that prizes closed-mindedness and as a consequence lacks an accu-
rate sense about when (etc.) to perspective-switch provided that the person’s evidence
supports thinking of closed-minded activity as epistemically appropriate or beneficial
(which seems possible given the community into which she’s been born). For it is
difficult to see why engaging in intellectual activity that one judges (inaccurately but
non-culpably) to be truth-conducive (say) should reflect negatively on who one is as
a person (vs. on the results of said activity). Thanks to Heather Spradley for a helpful
exchange on this issue.
13 This includes, for instance, work by Robert Adams (1985), Gary Watson (1996),
George Sher (2006), and Angela Smith (2008).
14 This is consistent with the possibility that in their more sophisticated (i.e. less basic)
forms, the skills proper to intellectual virtues can be quite demanding.
15 I say “unlikely” because there may be iterations of the case that would merit a differ-
ent assessment. In particular, if the agency of the persons in question is manifested in
their lack of the relevant ability or if they fail (culpably) to at least attempt to develop
the skill when presented with opportunities to do so, then their lack of virtuous ability
might reflect negatively on them qua persons and form the basis of an intellectual vice.
Whether this would still be consistent with their satisfaction of the motivational condi-
tion on the corresponding virtue remains an open question, which I will revisit shortly.
I’m grateful to Maura Priest and Heather Spradley for helpful input on this and related
points.
16 In fact, a similar point holds with respect to vice-constituting failures of epistemic
motivation (and thus to finding #1). That is, a person’s defective epistemic motivation
can suffice for the possession of an intellectual vice only if the person is (broadly)
responsible for this motivation. I have waited to introduce the complication of inculpa-
ble failures of virtue only because I think inculpable failures of judgment and skill are
more common or likely than inculpable failures of motivation.
17 There does not appear to be much agreement on what precisely such an account might
involve or require. See e.g. Crerar 2018, Cassam 2016 and 2019, and Battaly 2015.
However, I take it that my characterization is at least in the ballpark of these others.
18 Battaly (2015: 74) suggests something like this point. See also Zagzebski (1996: 207).
19 In opposition to this view, it is not enough to show that a person who in certain respects
is epistemically well-motivated might nevertheless manifest a (culpable, vice-consti-
tuting) shortcoming of virtuous judgment or competence. For the same person might
be poorly motivated in other respects, and this defective motivation might explain his
defective judgment or competence. More on this point below. Were it stipulated that
the person’s epistemic motives are entirely above board, the suggestion, again, is that
his defective judgment or competence would not be culpable—or at least not culpable
in a way that reflects negatively on his personal worth. I am grateful to Maura Priest for
some helpful critical feedback on this point.
20 Whether this prior motivational failure is itself a vice-making feature vs. a mere psy-
chological precondition for such is not an issue I will attempt to settle here. I do think
it is an interesting question, however, and am grateful to Charlie Crerar for raising it.
21 I am much less doubtful about this in connection with the competence dimension.
That is, it seems much less plausible that a person might, say, (1) culpably lack the
ability to perspective-switch while being (2) genuinely motivated to deploy this abil-
ity. Accordingly, my focus here will be on the relationship between the motivational
dimension and the judgment (vs. the competence) dimension.
22 See Crerar (2018: 759–760) for a related discussion.
23 My remarks on these cases, brief as they must be, are not intended as the final word.
Several of the cases are complex and lacking in important details, and therefore capable
of being interpreted in different ways and calling for different replies. Still, I hope my
treatment of them makes clear that motivationalism is not obviously or immediately
susceptible to them.
24 Tanesini (2018) defends a view like this. For her response to this case, see pp. 354–355.
25 There may, of course, be a version of this case in which Galileo is right to ignore the
views and input of others (e.g. if he is transparently intellectually superior and they are
transparently inferior on all relevant matters). But then his behavior would not, I take
it, be intellectually arrogant (even if it might be perceived as such by those ignorant of
the cognitive discrepancy).
26 If not, then there is no problem for motivationalism, since we no longer have a case of
epistemic vice that (purportedly) can’t be explained in motivational terms.
27 Were this not the case—for instance, if Dave’s dismissal were entirely a function of
how he was raised and he cannot in any sense be held responsible it—then the claim
that Dave is viciously closed-minded, etc., would be much less plausible, as would any
objection to motivationalism based on this case.
28 This is a helpful variation on Cassam’s (2016, 2019) Oliver case, which Crerar (2018)
rightly notes is ambiguous.
29 It is less easily accommodated by Tanesini, who subscribes to a more demanding view
of intellectual vice. For her discussion of this case, see her (2018: 356).
30 I am especially grateful to two students, Blake Colquitt and Josh Dolin, for several
lengthy and rigorous conversations about the ideas developed in this chapter. I’m also
grateful to my colleague (and friend) Dan Speak for the same. Heather Battaly, Charlie
Crerar, Rie Iizuka, Maura Priest, and Heather Spradley provided very helpful and gen-
36 Jason Baehr
erous written feedback or responses to an earlier draft. Finally, the chapter benefited
from feedback received from audiences at the University of Edinburgh, the University
of Connecticut, and California State University San Bernardino.
References
Adams, R. (1985) “Involuntary Sins,” The Philosophical Review 94(1): 3–31.
Adams, R. (2006) A Theory of Virtue (Oxford: Oxford University Press).
Aristotle. (1984) “Nicomachean Ethics,” in J. Barnes (ed.) The Complete Works of Aristotle
(Princeton, NJ: Princeton University Press).
Baehr, J. (2010) “Epistemic Malevolence,” Metaphilosophy 41: 189–213.
Baehr, J. (2011) The Inquiring Mind: On Intellectual Virtues and Virtue Epistemology
(Oxford: Oxford University Press).
Baehr, J. (2015) “The Four Dimensions of an Intellectual Virtue,” in C. Mi, M. Slote, and
E. Sosa (eds.), Moral and Intellectual Virtues in Western and Chinese Philosophy (New
York: Routledge), 86–98.
Battaly, H. (2015) Virtue (Cambridge: Polity).
Battaly, H. (2018) “Can Closed-Mindedness be an Intellectual Virtue?” Royal Institute of
Philosophy Supplement 84: 23–45.
Battaly, H. (2019) “Vice Epistemology Has a Responsibility Problem,” Philosophical
Issues 29(1): 24–36.
Cassam, Q. (2016) “Vice Epistemology,” The Monist 99: 159–180.
Cassam, Q. (2019) Vices of the Mind: From the Intellectual to the Political (Oxford:
Oxford University Press).
Crerar, C. (2018) “Motivational Approaches to Intellectual Vice,” Australasian Journal of
Philosophy 96(4): 753–766.
Kidd, I. J. (2017) “Capital Epistemic Vices,” Social Epistemology Review and Reply
Montmarquet, J. (1993) Epistemic Virtue and Doxastic Responsibility (Lanham, MD:
Rowman and Littlefield).
Roberts, R. C. and W. J. Wood. (2007) Intellectual Virtues: An Essay in Regulative
Epistemology (Oxford: Oxford University Press).
Sher, G. (2006) “Out of Control,” Ethics 116: 285–301.
Smith, A. (2008). “Control, Responsibility, and Moral Assessment,” Philosophical Studies
138: 367–392.
Tanesini, A. (2018) “Intellectual Vice and Motivation,” Metaphilosophy 49(3): 350–367.
Watson, G. (1996) “Two Faces of Responsibility,” Philosophical Topics 24(2): 227–248.
Zagzebski, L. (1996) Virtues of the Mind: An Inquiry into the Ethical Foundations of
Knowledge (Cambridge: Cambridge University Press).
2 The metaphysical foundations
of vice epistemology
Quassim Cassam
2.1
Vice epistemology is the philosophical study of the nature, identity and signifi-
cance of intellectual or epistemic vices. Among the questions that vice episte-
mologists might reasonably be expected to answer are the following:
1 What kind of thing is an epistemic vice?

2 How are epistemic vices individuated?
3 To what are competing accounts of epistemic vices answerable?
The first of these will be referred to here as the Kind Question, the second as the
Individuation Question and the third as the Answerability Question. These three
questions are ontological or metaphysical, and they bring into focus the meta-
physical foundations of vice epistemology. The third question is especially press-
ing. Where there are disagreements about what to count as an epistemic vice, or
about how specific vices are individuated, it is natural to wonder how such disa-
greements are to be resolved. Choosing between competing accounts of epistemic
vice, or of particular epistemic vices, requires clarity about the facts to which rival
theories are answerable.
One approach to the Kind Question is vice monism. This says that epistemic
vices are one kind of thing. One version of vice monism insists that epistemic
vices are character traits. Other versions of vice monism can be imagined. Vice
pluralism allows that there are different varieties of epistemic vice, between
which it isn’t necessary or even advisable to choose. As well as character traits
and attitudes, epistemic vices might include thinking styles, emotions, cognitive
faculties and cognitive biases. Different vice pluralists have different views about
what to include and what to exclude but they all agree that epistemic vices come
in different shapes and sizes. They aren’t one kind of thing.1
For present purposes, to individuate an object or kind is to draw a boundary
around it in thought or perception, to distinguish it from other objects or other kinds.
Individuating in this sense is the same thing as differentiating, and differentiating
is something that thinkers do. An indirect approach to the Individuation Question
holds that specific epistemic vices such as dogmatism and closed-mindedness are
38 Quassim Cassam
individuated by analysing our concepts. Conceptual analysis seeks to identify con-
ceptually necessary and sufficient conditions for the application of a concept. It
proceeds by the method of cases.2 In the present context, this means constructing
and examining examples of epistemic conduct with a view to determining whether
and how particular vice concepts apply in those cases. If there are cases in which,
intuitively, the concept of closed-mindedness applies but the concept of dogma-
tism does not then this would support the conclusion that these are distinct vices.
A direct approach holds that if our interest is in the individuation of epistemic
vices then we should focus on those vices themselves rather than on our concepts
of them. A similar point has been made by Hilary Kornblith in response to the
suggestion that knowledge is best studied by analysing the concept of knowl-
edge: ‘the subject matter of epistemology is knowledge itself, not our concept of
knowledge’ (2002: 1).3 Something like this thought is the basis of the suggestion
that the subject matter of vice epistemology should be our epistemic vices them-
selves rather than our concepts of them. But what would it even be to individuate
epistemic vices themselves without engaging in conceptual analysis or employing
the method of cases? The answer to this question is far from obvious. It conceiv-
able that in reality the direct approach employs many of the same techniques for
individuating epistemic vices as the indirect approach. In that case, one might
conclude that the differences between the direct and indirect approaches are more
ideological than practical.4
The Individuation Question leads directly to the Answerability Question.
Where there are different and incompatible accounts of, say, the relationship
between one epistemic vice and another, there is the challenge of deciding which
one is better. It is hard to choose between rival accounts without having an answer
to the Answerability Question. What is the reality to which the rival accounts are
trying to do justice? Realism is the view that there can only be one thing to which
accounts of epistemic vices are answerable: the nature of the vices themselves.5
The realist models the individuation and classification of vices on the individu-
ation and classification of diseases. For example, disease taxonomies classify
Alzheimer’s and vascular dementia as distinct forms of dementia because there
is a real distinction between one condition and another. The physical basis of
Alzheimer’s is different from the physical basis of vascular dementia. The bound-
ary between one epistemic vice and another might not be physical but the realist
nevertheless regards it as a real distinction.
At the opposite extreme is conceptualism. In its most general form, conceptu-
alism says that there are no lines in nature between one kind of thing and another.6
All distinctions between kinds or types are the product of conceptualisation. For
present purposes conceptualism needn’t take a view about diseases but it does
insist that the distinction between one epistemic vice and another is simply a
reflection of a difference in concepts. There are no ‘real’ divisions to which the
individuation of epistemic vices is accountable. All we have to go on is the ordi-
nary usage of vice terms and our intuitions about where the line between one vice
and another falls.7 This is very much in keeping with the indirect approach to the
Individuation Question.
The metaphysical foundations 39
The three sections that follow will address, in turn, each of the three basic
metaphysical questions for vice epistemology. In Section 2.2 a case will be made
for vice pluralism. Section 2.3 will argue that the direct/indirect dichotomy is a
false one and recommend a different way of proceeding, one that focuses on the
function or purpose of vice attributions. It will be argued in Section 2.4 that nei-
ther realism nor conceptualism offers a satisfactory answer to the Answerability
Question. The perspective to be developed here is more pragmatic. We cannot
hope to understand what epistemic vices are, how they are individuated, or what
makes one account better than another without a serious examination of whether
and why it is useful (if it is) to describe people as epistemically vicious or virtu-
ous. Section 2.5 will raise a question about the usefulness of vice attributions.
2.2
Vice monists tend to assume rather than argue explicitly for vice monism. For
example, it is taken for granted by ‘responsibilists’ that epistemic virtues and
vices are character traits. This is in line with Aristotle’s account of virtues as
‘states of character’ but is there an argument for the view that epistemic vices are
limited to such states? One argument that can be extracted from the work of Linda
Zagzebski is what might be called the argument from depth: the premise of this
argument is that virtues and vices are deep traits of a person. A virtue, Zagzebski
argues, is a ‘deep quality of a person, closely identified with her selfhood’ (1996:
104). Once a virtue or a vice develops, ‘it becomes entrenched in a person’s char-
acter and becomes a kind of second nature’ (1996: 116). These and other such
formulations can be read as restricting virtues and vices to character traits on the
grounds that only character traits can have the requisite depth.
One question about the argument from depth is whether specifically intellec-
tual virtues and vices need to be as deep as Zagzebski supposed. A vice is ‘a
quality we would ascribe to a person if asked to describe her after her death’
(1996: 135) but there are minor intellectual vices such as obscurity or pretentious-
ness that would hardly merit a mention in a person’s obituary.8 It might also be
objected that personal qualities other than character traits can be deep enough
to count as vices. Many of a person’s attitudes are closely identified with her
selfhood, but attitudes are stances or postures rather than character traits. In her
list of intellectual vices Zagzebski includes prejudice and wishful thinking. Yet
prejudice is a judgement or attitude rather than a character trait. Having strong
prejudices might be a character trait but prejudices themselves aren’t character
traits. Wishful thinking is a mode of thought rather than a character trait. If wish-
ful thinking, prejudice and character traits like closed-mindedness are genuine
epistemic vices, as they appear to be, then it follows immediately that epistemic
vices aren’t one kind of thing. It is a further question whether the three types of
epistemic vice all spring from a common root or whether two of the three types
are somehow reducible to the third. These are not questions that will be addressed
here, other than to say that there is no obvious way of reducing the three types to
one. It certainly looks like there are epistemic vices of genuinely different kinds.9
40 Quassim Cassam
An explicit theoretical defence of vice pluralism has been provided by Heather
Battaly.10 Her argument is this: virtues are qualities that make one an excellent
person whereas vices are qualities that make us worse people. These very thin
general conceptions of virtue and vice can be fleshed out in different ways. Having
a hard-wired cognitive faculty like excellent vision is one way to be an excellent
person. Another way is to be open-minded. By the same token, bad vision is one
way to be a defective person, closed-mindedness is another. Bad vision and other
such defects are ‘reliabilist’ vices whereas closed-mindedness is a ‘responsibilist’
vice. In this way, Battaly’s pluralism accommodates both varieties of epistemic
vice. It also accommodates two different intuitions about virtues and vices. One is
that luck plays an important role in whether we are virtuous or vicious. The other
is that bad effects over which one has no control should not render one vicious.
Reliabilism accommodates the first intuition, responsibilism the second, but vice
pluralism accommodates both.
One concern one might have about this argument for vice pluralism centres on
the idea of a better or worse person. It is far from obvious that cognitive disabili-
ties, such as bad vision or forgetfulness, make one a ‘worse person’.11 One does
not become a worse person if one loses one’s vision as a result of an illness or the
ageing process. Care also needs to be taken here to avoid suggestions of ableism.
Furthermore, as Liezl van Zyl has argued, the reliabilist and responsibilist intui-
tions about virtue and vice that Battaly seeks to accommodate are not just differ-
ent but incompatible. If luck plays an important role in whether we are virtuous
or vicious then why can’t bad effects over which one has no control render one
vicious? It is not a virtue of vice pluralism that it endorses contradictory verdicts
about what does and doesn’t constitute a vice. This is not an objection to vice plu-
ralism per se but to one specific version of this doctrine. There are, however, other
less contentious ways of being a vice pluralist. One way is to be an obstructivist.12
Obstructivism refrains from claiming that epistemic vices make one a worse
person. It argues that the intuitive notion of an epistemic vice is that of a per-
sonal quality that, as Medina puts it, ‘gets in the way of knowledge’ (2013: 30).
Epistemic vices systematically obstruct the gaining, keeping or sharing of knowl-
edge. The personal attributes that do that are a mixed bag. They include character
traits like closed-mindedness, attitudes like prejudice and arrogance and modes of
thinking such as wishful thinking. In that case, why not also include conditions
such as insomnia that make it harder to gain and retain knowledge? One reason
for excluding insomnia is that epistemic or intellectual vices must be conditions
of the intellect. In addition, being epistemically vicious is something for which a
person can fairly be blamed or criticised. This rules out insomnia as well as for-
getfulness, even though the latter might be regarded as a condition of the intellect.
This version of vice pluralism is relatively relaxed about admitting additional
varieties of epistemic vice. It doesn’t insist that only character traits, attitudes and
modes of thinking can be epistemic vices. It allows that it can sometimes be hard
to say whether something is or is not an epistemic vice. Cognitive biases might
not qualify if they aren’t personal qualities or defects for which a person can be
blamed or criticised. It isn’t clear, however, that they aren’t personal attributes
or that people can’t be criticised for their biases. The key point is that in the first
instance epistemic vices make one a worse gainer, keeper or sharer of knowledge
rather than a worse person.13 On one issue, however, obstructivism is in complete
agreement with Battaly: epistemic vices are not one kind of thing.
2.3
How are epistemic vices individuated? How do we draw the line between one
epistemic vice and another? Even if the subject matter of vice epistemology is
our epistemic vices themselves, it is arguable that our concepts are our only guide
to their structure. This would explain why conceptual analysis is the appropri-
ate method for vice-individuation. The labelling of this approach as ‘indirect’
might be questioned on the basis that there is no alternative ‘direct’ method. For
example, if we are interested in understanding the relationship between closed-
mindedness and dogmatism the only effective way of doing that is to analyse
the corresponding concepts. If these concepts are such that it is possible for one
of them to apply without the other applying then the only reasonable conclusion
would be that these are in fact distinct vices.14
Battaly’s work is again relevant here since she argues that there are indeed
requirements for dogmatism that aren’t requirements for closed-mindedness.15
The latter is an unwillingness or inability to engage with relevant intellectual
options. It does not require one to have already made up one’s mind. Dogmatism
does. It is an unwillingness to engage seriously with relevant alternatives to the
beliefs one already holds. On this account, dogmatism is a form of closed-mind-
edness but it is possible to be closed-minded without being dogmatic. The basis
of such claims is conceptual analysis. As Battaly recognises, other analyses of the
relationship between closed-mindedness and dogmatism are possible. For exam-
ple, she considers and rejects the proposal that closed-mindedness is an unwilling-
ness or inability to transcend a default cognitive option.16 In effect, she employs
the method of cases to undermine this proposal: she devises cases in which it
seems that a person is genuinely closed-minded without having, non-trivially, a
default cognitive option or position.
The necessary conditions for closed-mindedness that interest Battaly are con-
ceptually necessary, and disputes about such conditions are notoriously difficult
to resolve. The method of cases might be employed to determine what is concep-
tually necessary for closed-mindedness but will not provide a resolution if people
have conflicting intuitions about whether a given case is or is not a genuine case
of closed-mindedness. A given individual’s intuitions about hypothetical cases
might be a guide to their concept but it is a further question to what extent they
reveal conceptually necessary conditions for the application of ‘the’ concept of
closed-mindedness.17 One might take the view that there isn’t a single, univocal
concept, and that there is bound to be an element of stipulation in an individual
theorist’s conception of what it is to be closed-minded.
There is, however, another way of proceeding. Instead of individuating epis-
temic vices by conceptual analysis one might instead ask how, when and why it is
42 Quassim Cassam
useful and appropriate to employ vice concepts – concepts of epistemic vices – to
attribute epistemic vices to another person. With luck, an answer to this question
will shed some light on how epistemic vices are individuated. The inspiration for
this approach is Edwards Craig’s ‘state of nature’ account of knowledge.18 Craig
describes himself as ‘creeping up on the concept of knowledge’ (1990: 3) by ask-
ing what its role in our life might be. Its purpose ‘should be at least as interesting as
its analysis’ (1990: 2), and this leads to the suggestion that ‘the concept of knowl-
edge is used to flag approved sources of information’ (1990: 11). The fact that this
is what the concept is used for is a clue to what its characteristics might be.
When a person is judged to be closed-minded or dogmatic or to display some
other epistemic vice, such vice attributions appear to serve a number of different
purposes. They can be explanatory, evaluative or cautionary.19 We suppose that a
person’s epistemic conduct can sometimes be explained by their epistemic vices.
In attributing an epistemic vice to someone we are also implicitly evaluating them
on account of their being epistemically vicious in that respect. The implicit eval-
uation is negative rather than positive. Finally, the judgement that someone is
epistemically vicious can serve as a warning to proceed with caution in relying
on them for knowledge or understanding. To say that vice attributions are used
to explain, evaluate and warn is to make a point about what might be called their
functional role and, by implication, the functional role of vice concepts. If V1 and
V2 are both epistemic vices but attributions of V1 have a different functional role
from attributions of V2 then that is an indication that V1 and V2 are distinct vices.
An illustration of the explanatory role of vice attributions is the well-docu-
mented case of the Yom Kippur surprise.20 On Yom Kippur 1973, Egyptian and
Syrian forces launched a surprise attack on Israel on two fronts. Israel’s mili-
tary was taken by surprise despite the prior availability of excellent intelligence
indicating an impending attack. A subsequent study by Uri Bar-Joseph and Arie
Kruglanski blamed the intelligence failure on the closed-mindedness of Israel’s
Director of Military Intelligence and his senior Egyptian Affairs specialist.21 The
study concluded that these individuals ignored evidence of an impending attack
because they had a particularly high need for cognitive closure and had already
made up their minds that Egypt and Syria would not attack. The ‘because’ in this
formulation is both causal and explanatory. Bar-Joseph and Kruglanski’s hypoth-
esis is that the attribution of the epistemic vice of closed-mindedness to two senior
intelligence officers enables us to explain their lapses. If these individuals had
been more open-minded, they might have been more inclined to pay attention to
clear indications that Egypt and Syria were preparing to attack.
In explaining an event like the Yom Kippur surprise by reference to the epis-
temic vices of specific individuals, it is important to mark certain distinctions.
There is, for example, the distinction between an intelligence analyst who is com-
mitted to a particular doctrine – say the doctrine that Egypt and Syria wouldn’t
attack – and an analyst who is disposed to ignore evidence against whatever he
happens to believe but who is not committed in advance to a specific doctrine.
Bar-Joseph and Kruglanski mark this distinction by distinguishing between the
non-specific and specific need for closure. The former is ‘the desire for a confident
judgment on an issue, any confident judgment as compared to confusion and
ambiguity’ (2003: 80). The latter is ‘the desire for a judgment of a particular
content’ (2003: 80). A person with a specific need for closure has a specific bias,
not just a bias against any view that is opposed to his.
A different way to mark this distinction would be distinguish the dogmatism of
the intelligence analyst whose need for closure is specific from the mere closed-
mindedness of the analyst whose need for closure is non-specific. Even if both
analysts are dismissive of evidence pointing to an impending attack the explana-
tion is different in the two cases. The dogmatic analyst is dismissive of such evi-
dence not simply because of a general disposition to stick to his guns but because
the evidence is at odds with a specific doctrine – that the Arabs wouldn’t attack –
to which he is dogmatically committed. The merely closed-minded analyst would
have been receptive to evidence of an attack if he already believed that an attack
was likely. He is dismissive of such evidence only because he happens already to
have concluded that the Arabs wouldn’t attack and has a general disposition to be
dismissive of information that is at odds with his prior conception, whatever that
conception happens to be.
In this example, closed-mindedness and dogmatism are distinguished not by
analysing the corresponding concepts but by reflecting on the role of vice attri-
butions in enabling us to flag salient differences between different explanations
of a person’s epistemic conduct. It might even be hypothesised that concepts
like closed-mindedness and dogmatism came into being as subtly distinct con-
cepts precisely in order to make it possible for us to draw correspondingly subtle
explanatory distinctions. These and other epistemic vice concepts earn their keep
by helping us to make sense of the many different ways in which a person’s epis-
temic conduct can be flawed. A satisfactory account of how specific epistemic
vices are individuated must track differences in their explanatory roles.
With regard to the evaluative function of vice attributions, one view is that, as
Gabriele Taylor puts it, ‘merely to use the labels “virtue” and “vice” is to indicate
candidates for praise and blame’ (2006: 6). It should be noted, however, that our
evaluative repertoire is much more extensive than this suggests.22 There are vices
for which people are criticised but not blamed. Stupidity, understood as foolish-
ness rather than lack of intelligence, is one such vice. In George Sher’s terminol-
ogy, epistemic vices cast a ‘negative shadow’ over the person whose vices they
are but different vices do this in different ways.23 Virtue and vice attributions are
evaluative in the way that references or letters of recommendation are evalua-
tive, and this has a bearing on how virtues and vices are individuated. What we
expect from a person who is described in a letter as arrogant is different from
what we expect from a person who is described as overconfident. This is a sign
that arrogance and overconfidence are not the same vice. In the same way, closed-
mindedness and dogmatism aren’t the same epistemic vice if they generate differ-
ent expectations. Reflection on such differences is a tool for delineating epistemic
vices. V1 and V2 are not the same epistemic vice if the epistemic conduct one
would expect from someone who has V1 is different from the epistemic conduct
one would expect from a person who has V2.
44 Quassim Cassam
Letters of recommendation can be cautionary as well as evaluative. A letter
that goes into great detail about a person’s epistemic vices but is silent about
their virtues is sending a message to the reader: for example, do not hire this
person. Different epistemic vices provide different reasons for rejecting a candi-
date. Again, this tells us something about the relationship between one vice and
another. The fact that a person is arrogant might not be seen as a reason for reject-
ing their application to be a foreign exchange trader. In contrast, overconfidence
might be regarded as a fatal defect. V1 and V2 are not the same epistemic vice
if it is coherent to regard V1 but not V2 as relevant to a person’s suitability for a
particular professional role.24 How can they be the same vice if it matters whether
they have V1 but not whether they have V2?
Is the functional role approach to the Individuation Question direct or indi-
rect? The case for treating this approach as indirect is that it seeks to individuate
epistemic vices by reflecting on various aspects of vice attributions. On the other
hand, the functional role approach doesn’t offer an analysis of the vice concepts
that are employed in vice attributions and does not see itself as creeping up on
the concept of an epistemic vice. Just as Craig’s state of nature theory can be read
as an account of knowledge rather than just the concept of knowledge, so there is
nothing wrong with regarding the study of epistemic vice attributions as reveal-
ing the nature of the epistemic vices themselves.25 There is little to be gained by
describing the functional role approach either as ‘direct’ or as ‘indirect’. What
matters is whether, one way or another, it tells us something about the nature of
the vices to which it is applied.
2.4
What are the implications of the functional role approach to the Individuation
Question for the Answerability Question? To put this issue into context, con-
sider again the contrast between realism and conceptualism. Realism insists that
philosophical accounts of epistemic vices are answerable to the nature of the
vices themselves and that the distinction between one vice and another is a real
distinction. Epistemic vices are like diseases or other natural kinds and are no
less real. Like natural kinds, they have real essences, though not physiological
real essences. The realist’s motto is that there are lines in nature and these lines
include the line between one vice and another as well as one disease and another.26
It is not clear, however, in what sense the distinction between one epistemic
vice and another is a ‘distinction in nature’. One might suppose that epistemic
vices have psychological rather than physiological essences. For example, the
need for closure might be posited as the psychological essence of closed-minded-
ness. Yet it is conceivable that what grounds a person’s closed-mindedness is not
a need for closure but arrogance. By the same token, people aren’t motivated to be
stupid by an independently identifiable need or desire. This makes it hard to see in
what sense epistemic vices have psychological real essences. Realism helps itself
to the idea that epistemic vices are ‘natural’ but does little to address the suspicion
that its conception of naturalness lacks substance.
What is the conceptualist alternative to realism? The commitments of concep-
tualism are highlighted by Locke’s theory of ‘mixed modes’. Modes are depend-
ent existences that can only exist as the qualities of a substance. Simple modes
are combinations of the same simple idea whereas mixed modes combine ideas
of several different kinds.27 For example, theft is a mixed mode since the idea of
theft is the idea of the concealed change of possession of something without the
consent of the proprietor. Locke maintains that ideas of modes are ‘voluntary
Collections of simple Ideas, which the Mind puts together, without any reference
to any real Archetypes’ (II.xxxi.3).28 It follows that these ideas can’t fail to be
adequate since, as Ayers puts it on Locke’s behalf, we form these ideas ‘without
the need to refer to reality’ (1991: 57).29 Consider the idea of courage:
He that at first put together the Idea of Danger perceived, absence of disorder
from Fear, sedate consideration of what was justly to be done, and executing
it without that disturbance, or being deterred by the danger of it, had certainly
in his Mind that complex Idea made up of that Combination: and intending
it to be nothing else, but what it is; nor to have any other simple Ideas, but
what it hath, it could not also be but an adequate idea: and laying this up in
his Memory, with the name Courage annexed to it, to signifie it to others, and
denominate from thence any Action he should observe to agree with it, had
thereby a Standard to measure and dominate Actions by, as they agreed to it.
(II.xxxi.3)
It is reality that sets the standard for our ideas of substances. With mixed modes,
our ideas set the standard for reality, so that an action is courageous just if it has
the features that our idea of courage brings together. Locke doesn’t deny that
ideas of mixed modes can be formed by observation. For the most part, however,
ideas of modes are the products of invention, of the ‘voluntary putting together of
several simple Ideas in our own minds’ (II.xxii.9), without prior observation. It
is, in an important sense, arbitrary how we choose to put together simple ideas to
form complex ideas of mixed modes.
A striking consequence of Locke’s conceptualism is that there is no external
standard by reference to which disputes about what is and is not part of the idea
of mixed modes can be settled. Again Locke uses the example of courage to make
his point. Suppose that one person X’s idea of a courageous act includes the idea
of ‘sedate consideration’ of ‘what is fittest to be done’ (II.xxx.4). This is the idea
of ‘an Action which may exist’, but another person Y has a different idea accord-
ing to which a courageous action is one that is performed ‘without using one’s
Reason or Industry’. Such actions are also possible, and Y’s idea is as ‘real’ as
X’s. An action that displays courage by X’s lights might fail to do so by Y’s lights
and vice versa, but it seems that the only respect in which Y’s idea might count as
‘wrong, imperfect, or inadequate’ (II.xxxi.5) is if Y intends his idea of courage to
be the same as X’s. Apart from that, both ideas are equally good and equally valid.
Suppose, then, that one vice epistemologist insists, while another denies, that
closed-mindedness involves an unwillingness to engage seriously with relevant
46 Quassim Cassam
alternatives to the beliefs one already holds. Can it not be argued in this case that
both ideas of closed-mindedness are equally good and valid? Not if one of the two
ideas does a significantly better job of flagging a distinction that is important for
the purposes of explanation or evaluation. If it is important for such purposes to
distinguish epistemic vice V1 from another vice V2 then the distinction between
these vices is not arbitrary and there is something beyond our ideas to which it is
answerable: the fact that some explanations are better than others.30 If a distinction
is useful or even indispensable in practice then it is, at least to this extent, a real
distinction in a non-quixotic sense of ‘real’.
Much depends, therefore, on whether it is true, as argued above, that certain
ways of delineating epistemic vices such as closed-mindedness and dogmatism
are better than others, better, that is, relative to an interest in explaining and evalu-
ating people’s epistemic conduct. If so, then this is the reality to which competing
accounts of our epistemic vices are answerable. It is also the reality that concep-
tualism is in danger of disregarding if it insists that the choice between competing
accounts of epistemic vices is arbitrary or that there is no need to refer to reality in
deciding where to locate the line between one epistemic vice and another. There
may be forms of conceptualism that play down the arbitrariness of vice individu-
ation, but the extreme conceptualism about mixed modes to which Locke is com-
mitted is not one of them.31
Realism is also problematic since the functional role approach does not sup-
port the idea that epistemic vices are natural kinds or that the line between one
epistemic vice and another is analogous to the distinction between one disease
and another. This is not the sense in which the distinction between two epis-
temic vices is a ‘real’ distinction. An analogy might help to make the point
clearer: a taxonomy of belief that is interested in explaining human action, and
not just categorising it, will almost certainly have to recognise religious beliefs
as a distinct type of belief. One reason, perhaps, is that religious beliefs have a
distinctive content but it is also true that a person’s religious beliefs can explain
many aspects of their conduct – the fact that they pray regularly, for example
– that their non-religious beliefs cannot. Religious beliefs have a distinctive
functional role.32 However, few would be tempted by the notion that there is
a real distinction ‘in nature’ between the religious and non-religious beliefs,
or that religious beliefs constitute a natural kind. An unqualified realism about
types of belief is a non-starter, and the same goes for an unqualified realism
about epistemic vices.
In both cases, it is necessary to find a middle way between conceptualism and
realism. A label for this middle way is conceptualist realism.33 A less cumber-
some label that captures the link between the reality of a distinction and its useful-
ness is pragmatism. The Answerability Question invites vice epistemologists to
decide whether competing accounts of epistemic vice are answerable to our con-
cepts or to extra-conceptual reality. Conceptualist realism regards this as a false
choice. Since our concepts and classificatory choices are shaped by reality it fol-
lows that the two options aren’t mutually exclusive. The way that epistemic vices
are individuated is both a reflection of our conceptual scheme and the explanatory
and other realities by which our thinking is influenced. Epistemic vices lack the
ontological stability of genuine natural kinds but they aren’t arbitrary constructs
if they enable us to make sense of epistemic reality, including numerous varieties
of flawed epistemic conduct that are part of that reality.
2.5
For the conceptualist realist the crux of the matter is whether vice attributions are
actually as useful as has been claimed here. If one epistemic vice is distinguished
from another on the basis that they explain different things it had better be the
case that epistemic vices are explanatory. In offering an explanation of the flawed
epistemic conduct of people who should know better, vice attributions offer a
partial but significant answer to certain ‘how-possible’ questions.34 For example:
how was it possible for Israeli military intelligence to overlook or ignore com-
pelling evidence of an impending attack? Because of the epistemic vices of key
individuals in the story of the Yom Kippur surprise. What might otherwise be
hard to understand becomes intelligible in the light of the vice attribution or, as
one might call it, the vice explanation. A vice explanation is not an explanation
of an epistemic vice but an explanation of a person’s conduct by reference to an
epistemic vice.
It might be objected that vice explanations lack the explanatory power that has
been claimed for them and that they only serve to obscure more pertinent fac-
tors in many cases. For example, from the standpoint of what Mark Alfano calls
‘epistemic situationism’, a person’s epistemic conduct usually has much more to
do with situational factors than with intellectual or other character traits.35 Some
situationists even see this as part of an argument against the very existence of such
traits and, by implication, the existence of intellectual character vices. Yet there
are many examples of defective epistemic conduct that are not amenable to expla-
nation in situational terms. For example, much has been written about the conduct
of senior members of the Bush administration in the run-up to the 2003 American
invasion of Iraq. The flaws in this conduct have been attributed to a wide vari-
ety of epistemic vices, including arrogance, imperviousness to evidence and an
inability to deal with mistakes.36 It is questionable whether in this case there are
plausible situational alternatives to such vice explanations.
Other alternatives to vice explanations focus on sub-personal, socio-struc-
tural or ideological factors. For example, belief in conspiracy theories has been
explained by reference to the intellectual vices of conspiracy theorists but some
psychologists argue that conspiracy theories are sustained by sub-personal biases
such as proportionality and intentionality bias.37 A socio-structural approach
might be encouraged by studies that indicate that a tendency to see conspiracies
everywhere is associated with adverse life circumstances.38 Finally, there is evi-
dence that conspiracy theories are associated with extremist political ideologies,
including anti-Semitic ideologies.39 A person who subscribes to such an ideology
might be more inclined to endorse ideologically motivated conspiracy theories. In
such cases, it is the person’s ideology rather than their epistemic vices that is the
48 Quassim Cassam
key to their thinking. The importance of ideology tends to be underestimated by
vice epistemologists. There are apparently epistemic failings that have more to do
with politics than anything else.40
One question that arises is whether ideological explanations are an alternative
to vice explanations. It might be suggested, for example, that a commitment to
extremist ideologies is itself something that calls for an explanation in terms of
the extremist’s epistemic vices. On the other hand, it is important not to underes-
timate the extent to which ideologies are expressive of a person’s values. At least
on the face of it, the fact that a person is an anti-Semitic conspiracy theorist says
more about their moral failings than their epistemic vices. Hitler did what he did
and believed what he believed because he was morally depraved and not because
he was closed-minded. No doubt he was closed-minded, but when it comes to
explaining his conduct his epistemic vices pale into insignificance when com-
pared to his moral defects.
The lesson of such examples is that one and the same phenomenon can be
explained at multiple different levels and by reference to multiple different factors:
situational, ideological, moral, psychological and epistemological. To describe a
person as closed-minded is, in effect, to pass comment on their psychology and
their epistemological outlook. Vice explanations can be trumped by explanations
of other types, and it’s not clear in such cases whether there is even such a thing as
the ‘right’ explanation. Vice epistemologists are perhaps too inclined to promote
vice explanations of what they see as flawed epistemic conduct even in cases
where plausible alternative explanations are available.41 This is an epistemic vice
of vice epistemologists. The contrary intellectual virtue that they need to cultivate
is the virtue of only resorting to vice explanations where such explanations are
called for.
As the case of the Yom Kippur surprise shows, there are cases where such
explanations are more compelling than any alternative. If this were not so the
appeal of vice epistemology would be considerably diminished. Given that vice
explanations are sometimes successful, the remaining question is how a person’s
epistemic vices are to be explained. José Medina proposes that vices like arro-
gance and closed-mindedness are among the structural and systematic vices of the
privileged.42 If this proposal is correct then there is an opening here for vice expla-
nations to take account of socio-structural factors. However, the fact that some of
the epistemic vices that figure in vice explanations are amenable to explanation
in socio-structural terms is not a reason for doubting their existence. It remains
true that it is sometimes plausible and necessary to explain why people think and
reason as they do by reference to their epistemic vices. That is the reality to which
philosophical thinking about epistemic vices is answerable.43
Notes
1 For a defence of vice pluralism see Battaly 2014 and 2015a. For a critique, see Van Zyl
2015.
2 There is a vast philosophical literature on the method of cases. See, for example,
Goldman 2007, Kornblith 2007 and Sosa 2007.
3 By focusing on the concept of knowledge, Kornblith argues, ‘we only succeed in
changing the subject: instead of talking about knowledge, we end up talking about our
concept of knowledge’ (2002: 9–10).
4 For further discussion of these issues see Kornblith 2002, especially chapter 1.
5 As Ian James Kidd notes, epistemic vices have historically been held to be answerable
to metaphysically deep features of the world. What Kidd (2018) calls ‘deep epistemic
vices’ are ones whose identity and intelligibility are ultimately determined by a deep
conception of human nature or the nature of reality.
6 As David Wiggins puts it, ‘there are no “lines” in nature (even though, after the imposi-
tion of lines, there are edges for us to find there). It is we, sharing the benign illusion
that there is just one way to do this, who impose lines on nature, not arbitrarily or in
just any way, but in ways that are determined for us by our constitution and ecology,
by the scale appropriate to our physical size in relation to the rest of the world, and by
our intellectual and practical concerns’ (1986: 170). The view that there are concept-
independent lines in nature is defended in Ayers 1991.
7 Except in the case of comparatively esoteric epistemic vices, in relation to which we
may have no strong intuitions.
8 It should be noted, however, that what counts as a major or minor epistemic vice is
highly context-relative.
9 See Cassam 2019a for a defence of this approach.
10 See, for example, Battaly 2015a and 2015b. Pluralism, as Battaly understands it, is the
view that there are different kinds of virtues and that ‘different qualities can make one
a better person in different ways’ (2015a: 9).
11 The claim that it does is made by Heather Battaly. She writes: ‘Vices are qualities
that make us defective people … a person can be defective in a variety of ways: for
instance, she can be defective insofar as she has bad vision; or insofar as she lacks logi-
cal skills; or insofar as she is dogmatic, unjust, or cruel’ (2014: 52).
12 See Cassam 2019a for an exposition and defence of obstructivism.
13 I leave open the possibility that epistemic vices make one a worse person by making
one a worse gainer, keeper, or sharer of knowledge.
14 Distinct concepts like water and H2O can be concepts of the same property but in this
case it isn’t possible for one of these concepts to apply without the other applying. If
there are cases in which the concept of closed-mindedness applies but the concept of
dogmatism does not then these are distinct concepts and distinct properties.
15 See Battaly 2018.
16 She attributes this view to Jason Baehr.
17 For present purposes a concept can be understood as ‘a psychological structure or state
that underpins a cognizer’s deployment of a natural-language predicate’ (Goldman & Pust
2002: 83). If people have markedly different contents associated with one and the same
predicate, then, as Goldman and Pust note, ‘philosophical analysis must be satisfied with
using intuitions to get at each person’s distinct concept; it must be prepared, if necessary,
to abandon the assumption that the content of each person’s concept can be generalized to
others’ (2002: 86). However, Goldman and Pust are not convinced that this is necessary.
18 In Craig 1990.
19 A fuller account of the functional role of vice attributions would also need to take
account of what Ian James Kidd calls ‘epistemic vice-charging’, the critical practice
of charging other people with epistemic vice. According to Kidd, ‘the practice of vice-
charging should ultimately be ameliorative’ (2016: 192).
20 See Bar-Joseph & Kruglanski 2003, Bar-Joseph 2005, and Cassam 2019a: 28–52.
21 Bar-Joseph & Kruglanski 2003.
22 As Julia Driver points out, ‘we sometimes, and indeed often do, make critical com-
ments about someone’s intellect without blaming them’ (2000: 132).
23 Sher 2006: 58.
24 On the role relativity of virtues and vices, see Pigden 2017.
50 Quassim Cassam
25 It is an odd and unnecessary feature of Craig’s account that he insists on representing it
as an account of the concept of knowledge.
26 It is worth noting that the distinction between one disease and another is less ‘natural’
and less straightforward than realism assumes. As Peter Toon notes, debates about the
boundaries of a disease are often ‘really evaluative debates about the boundary between
illness and wellness’ (2014: 57). Furthermore, when a disease is defined on the basis
of a continuous variable ‘the normal and the abnormal merge imperceptibly into each
other, so the boundary is to some extent arbitrary’ (2014: 58). The realist who insists
that there are sharp distinctions in reality between epistemic vices is thinking specifi-
cally of diseases with sharp boundaries and that are defined on a categorical variable.
For example, it isn’t possible to have a touch of bubonic plague. See Cassam 2017 for
further discussion of the analogy between epistemic vices and diseases.
27 Locke’s examples of mixed modes include beauty, theft, obligation, drunkenness, a lie,
hypocrisy, sacrilege, murder, appeal, triumph, wrestling, fencing, boldness, habit, testi-
ness, running, speaking, revenge, gratitude, polygamy, justice, liberality and courage.
This list is from Perry 1967.
28 All references in this form are to a book, chapter and section of Locke 1975, originally
published in 1689.
29 Locke illustrates the arbitrariness of mixed modes by noting that we have the complex
idea of patricide but no special idea for the killing of a son or a sheep.
30 An explanation as I understand it is part of the natural world, and might remain undis-
covered. For an account of the contrast between this ‘objectivist’ conception of expla-
nation and more subjectivist approaches, see Bird 2005.
31 Locke’s conceptualism is a form of what David Wiggins calls ‘anti-realist conceptu-
alism’, as distinct from ‘sober conceptualism’. Anti-realist conceptualism holds that
nothing prevents us from dissecting reality completely differently from what we are
used to. See Wiggins 1980: 138–140.
32 Their functional role is not unrelated to their content.
33 This label is borrowed from Wiggins 1980.
34 See Cassam 2007 for an account of the nature of how-possible questions.
35 Alfano 2013.
36 See Ricks 2007 and Cassam 2019a: 1–27.
37 An intellectual vice account of conspiracy theories is given in Cassam 2015. For a
response see Pigden 2017. A cognitive bias approach is defended in Brotherton 2015.
38 Freeman & Bentall 2017.
39 See Byford 2011 and Cassam 2019b.
40 This is the central point of Cassam 2019b.
41 See Cassam 2015 for an example of this.
42 Medina 2013: 30.
43 For helpful comments and suggestions, I thank Heather Battaly, Josh Dolin, Ian James
Kidd and other participants at the April 2019 UConn Vice Epistemology Conference.
References
Alfano, M. (2013), Character as Moral Fiction (Cambridge: Cambridge University Press).
Ayers, M. R. (1991), Locke, volume 2: Ontology (London: Routledge).
Bar-Joseph, U. (2005), The Watchman Fell Asleep: The Surprise of Yom Kippur and Its
Sources (Albany: State University of New York Press).
Bar-Joseph, U. and Kruglanski, A. (2003), ‘Intelligence Failure and Need for Cognitive
Closure: On the Psychology of the Yom Kippur Surprise’, Political Psychology, 24: 75–99.
Battaly, H. (2014), ‘Varieties of Epistemic Vice’, in J. Matheson and R. Vitz (eds.), The
Ethics of Belief: Individual and Social (Oxford: Oxford University Press): 51–76.
Battaly, H. (2015a), ‘A Pluralist Theory of Virtue’, in M. Alfano (ed.), Current
Controversies in Virtue Theory (New York/London: Routledge): 7–22.
Battaly, H. (2015b), Virtue (Cambridge: Polity Press).
Battaly, H. (2018), ‘Closed-Mindedness and Dogmatism’, Episteme, 15: 261–82.
Bird, A. (2005), ‘Explanation and Metaphysics’, Synthese, 143: 89–107.
Brotherton, R. (2015), Suspicious Minds: Why We Believe Conspiracy Theories (London:
Bloomsbury Sigma).
Byford, J. (2011), Conspiracy Theories: A Critical Introduction (Basingstoke: Palgrave
Macmillan).
Cassam, Q. (2007), The Possibility of Knowledge (Oxford: Oxford University Press).
Cassam, Q. (2015), ‘Bad Thinkers’, Aeon (https://aeon.co/essays/the-intellectual-chara
cter-of-conspiracy-theorists).
Cassam, Q. (2017), ‘Vice Ontology’, Social Epistemology Review and Reply Collective,
6: 20–27.
Cassam, Q. (2019a), Vices of the Mind: From the Intellectual to the Political (Oxford:
Cassam, Q. (2019b), Conspiracy Theories (Cambridge: Polity Press).
Craig, E. (1990), Knowledge and the State of Nature: An Essay in Conceptual Synthesis
Driver, J. (2000), ‘Moral and Epistemic Virtue’, in G. Axtell (ed.) Knowledge, Belief,
and Character: Readings in Virtue Epistemology (Lanham: Rowman & Littlefield
Publishers Inc.): 123–34.
Freeman, D. and Bentall, R. (2017), ‘The Concomitants of Conspiracy Concerns’, Social
Psychiatry and Psychiatric Epidemiology, 52: 595–604.
Goldman, A. (2007), ‘Philosophical Intuitions: Their Target, Their Source, and Their
Epistemic Status’, Grazer Philosophische Studien, 74: 1–26.
Goldman, A. and Pust, J. (2002), ‘Philosophical Theory and Intuitional Evidence’, in
A. Goldman (ed.), Pathways to Knowledge: Private and Public (Oxford: Oxford
University Press): 73–94.
Kidd, I. J. (2016), ‘Charging Others with Epistemic Vice’, The Monist, 99: 181–97.
Kidd, I. J. (2018), ‘Deep Epistemic Vices’, Journal of Philosophical Research, 43: 43–67.
Kornblith, H. (2002), Knowledge and Its Place in Nature (Oxford: Oxford University
Press).
Kornblith, H. (2007), ‘Naturalism and Intuitions’, Grazer Philosophische Studien, 74:
27–49.
Locke, J. (1975), An Essay Concerning Human Understanding, ed. P. H. Nidditch (Oxford:
Medina, J. (2013), The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic Injustice, and Resistant Imaginations (Oxford: Oxford University Press).
Perry, D. L. (1967), ‘Locke on Mixed Modes, Relations, and Knowledge’, Journal of the
History of Philosophy, 5: 219–35.
Pigden, C. R. (2017), ‘Are Conspiracy Theorists Epistemically Vicious?’, in K. Lippert-
Rasmussen, K. Brownlee, and D. Coady (eds.), A Companion to Applied Philosophy
(Chichester: Wiley Blackwell): 120–132.
Ricks, T. E. (2007), Fiasco: The American Military Adventure in Iraq (London: Penguin
Books).
Sher, G. (2006), In Praise of Blame (Oxford: Oxford University Press).
Sosa, E. (2007), ‘Intuitions: Their Nature and Epistemic Efficacy’, Grazer Philosophische
Studien, 74: 51–67.
52 Quassim Cassam
Taylor, G. (2006), Deadly Vices (Oxford: Clarendon Press).
Toon, P. (2014), A Flourishing Practice? (London: Royal College of General Practitioners).
Van Zyl, L. (2015), ‘Against Radical Pluralism’, in M. Alfano (ed.), Current Controversies
in Virtue Theory (New York/London: Routledge): 22–34.
Wiggins, D. (1980), Sameness and Substance (Oxford: Blackwell).
Wiggins, D. (1986), ‘On Singling out an Object Determinately’, in P. Pettit and J.
McDowell (eds.), Subject, Thought, and Context (Oxford: Clarendon Press): 169080.
Zagzebski, L. (1996), Virtues of the Mind: An Inquiry into the Nature of Virtue and the
Ethical Foundations of Knowledge (Cambridge: Cambridge University Press).
3 Ignorance, arrogance, and privilege
Vice epistemology and the epistemology of
ignorance
Alessandra Tanesini
At the start of the #metoo protests, many men professed genuine surprise about
the prevalence of sexual harassment, whilst many women could not figure out
how men could have been so ignorant. Black people have long observed that
a similar apparent commitment to ignorance about race is widespread among
whites. In a blog post originally written in 2004, the British journalist, Reni Eddo-
Lodge, reported that she had given up talking about race to white people because
the majority simply refuse to accept the reality of structural racism. She notes
that when she speaks, she can see that “their eyes shut down and harden. It’s like
treacle is poured into their ears, blocking up their ear canals” (Eddo-Lodge, 2017,
p. ix).
This chapter offers a vice-epistemological account of the ignorance of those
who are invested in not knowing, with a special focus on white ignorance, under-
stood as a kind of racial insensitivity. I have two main aims. The first is to show
that some forms of ignorance are the product of epistemic vices, conceived as
sensibilities. The second is to explain the mutually re-enforcing connection that
exists between arrogance, ignorance, and privilege. The chapter consists of three
sections: the first offers a definition of active ignorance. The second provides an
account of one epistemically vicious sensibility – racial insensitivity. The third
section illustrates how arrogance feeds on privilege, produces ignorance, and wid-
ens inequalities that entrench privilege.
3.1 Varieties of ignorance

Ignorance is a heterogeneous phenomenon. It comprises at least the presence of
false beliefs, and the absence of true ones. One may thus define ignorance in the
following terms: a person is ignorant about a fact that p, whenever that person
either disbelieves p (e.g., believes that it is false), or suspends belief about it (e.g.,
has not settled on either p or not p), or has no doxastic attitude toward p (e.g., the
thought as to whether p has never entered her mind) (Peels, 2017, p. 169).1
These states of ignorance can be brought into being in different ways. Basic
or plain ignorance occurs when disbelief, belief suspension, or absence of dox-
astic attitude are the result of mere accident, bad luck, or of cognitive shortcom-
ings. Thus, for instance, a person might be ignorant of many mathematical truths,
54 Alessandra Tanesini
because he has never studied mathematics. Whilst his lack of interest might find
deeper and troublesome explanations in some cases, it might also on occasion be
the result of an autonomous choice to focus one’s intellectual efforts elsewhere.
Active ignorance, on the other hand, has a different and less savoury aetiology.
I borrow the term from theorists of the epistemology of ignorance (Medina, 2013;
Mills, 2007; Sullivan & Tuana, 2007; Tuana, 2006). At first pass, one can define
it as “the kind of ignorance that is deeply invested in not knowing” (Medina,
2016, p. 182). There are broadly speaking at least three kinds of psychological and
social mechanisms whose involvement in the production of the state of ignorance
makes that output an instance of active ignorance. First, ignorance can be the
result of cognition that is motivated by goals other than accuracy.2 One might list
self-deception and wishful thinking among its paradigmatic examples. Second,
ignorance might be the result of deception, misinformation, and doubt-mongering.
Central cases include the ignorance of people who have been lied to, who have been
targeted by propaganda and misinformation campaigns.3 Third, ignorance might
be the product of finding oneself in a cognitive niche that promotes it, and whose
success at fostering ignorance makes the niche self-sustaining. Supermarkets are
a good example of this phenomenon. They are designed to maximise the chances
of impulse buying by fostering the right amount of shoppers’ ignorance of the
location of goods on the shelves. Supermarket staff achieve this goal of the
company by moving things around often enough to cause some confusion, but not
so often as to alienate their customers.4
What these three kinds of case have in common is that in these instances igno-
rance is the product of, in Medina’s words, “epistemic labour” (Medina, 2016).
That is, active ignorance is the output of mechanisms – psychological and social
– functioning well, rather than failing to operate or malfunctioning. These are
mechanisms that are either designed or selected for their ability to obfuscate
or misinform. This is why active ignorance is ignorance as a perverse kind of
achievement (Spelman, 2007; Tanesini, 2018b).
It is useful to compare this tripartite characterisation of active ignorance with
the taxonomy provided by Tuana (2006). Tuana singles out six forms of igno-
rance, not all of which are active in the sense described above. Wilful ignorance,
understood as the will not to know, and ignorance that flows from a lack of inter-
est in things that should be of concern to one, fit the motivated cognition para-
digm. In these cases, individuals are ignorant because they adopt a strategy of
looking away, ignoring, or simply avoiding thinking too hard. Ignorance that is a
result of being kept ignorant by the actions of others fits the deception paradigm.
As Tuana points out there are at least two forms of such ignorance. The first is the
dissemination of propaganda and the active blockage of the transmission of useful
information.5 The second is the constant undermining of people’s confidence in
their own epistemic authority. This process can result in a tendency not to trust
one’s own convictions that increases ignorance by undermining the ability to form
full beliefs. Ignorance can also be the result of systemic environmental mecha-
nisms that block the development of knowledge. This fifth case fits the paradigm
of cognitive niches that scaffold ignorance.
Ignorance, arrogance, and privilege 55
Active ignorance is thus heterogeneous. It is also often connected to moral and
epistemic vices. In some cases ignorance is the outcome of epistemically vicious
motivations to ignore something to which one should attend. In other cases, igno-
rance is the effect of other people’s malevolent intentions toward one. In addi-
tion, it would seem that ignorance is a characteristic outcome of several epistemic
character vices such as laziness and, as I argue in the final section of this chapter,
intellectual arrogance.
That said, ignorance is not always bad. For example, sometimes there are
moral requirements not to know some things in order to safeguard others’ privacy
(cf. Manson, 2012). Ignorance can also be of instrumental epistemic value. For
instance, good experimental design often requires that scientists are ignorant of
whom is receiving a placebo when testing the efficacy of a drug. Therefore, there
are cases when being wilfully ignorant is not epistemically vicious but might actu-
ally be virtuous.6 A difference between good and bad ignorance lies in the con-
scious or subconscious ultimate motives for keeping oneself ignorant. The kind of
active ignorance that I identify as intellectually vicious is often driven by morally
bad motivations, such as the desire to feel superior to other people or to avoid
thinking about one’s complicity in injustice. Another difference between good
and bad ignorance lies in its purpose. Some ignorance is ultimately in the service
of knowledge and of good moral outcomes. Vicious ignorance, instead, generally
obstructs effective and responsible inquiry.7
All of these forms of ignorance differ from Tuana’s sixth kind of ignorance
which she calls “loving ignorance”. This ignorance is not really ignorance per se
since it is an acknowledgment of the epistemic limitations of one’s own perspec-
tive. It is, thus, akin to the adoption of an attitude of humility. Whilst active igno-
rance, when driven by bad motivations, is often invisible to the ignorant person,
loving ignorance consists precisely in making one’s first-order ignorance visible.
Thus understood, loving ignorance is actually knowledge, rather than ignorance,
of one’s own ignorance.
The notion of active ignorance that I have sketched here is related to, but not
the same as, Charles Mills’ notion of white ignorance (2007, 2015). Mills’ notion
is more restricted since it focuses exclusively on ignorance that is caused by social
facts about whiteness conceived as a structure of domination, rather than as a
biological construct. Hence, Mills’ account does not include those forms of igno-
rance that are the result of different social structures that rank some groups as
superior and others as inferior such as heteronormativity or patriarchy. It is easy,
however, to see how one could provide accounts of straight or male ignorance
along Millsian lines.
Mill’s account of ignorance, however, might elide rather than help to clarify
the differences between types of active ignorance. In his view, white ignorance
is ignorance that is caused by, in the sense of being counterfactually dependent
on, some social facts about whiteness (Mills, 2015, p. 218).8 The account entails
that, for example, a black person’s ignorance of higher mathematics if it is coun-
terfactually dependent on white privilege as a structural injustice is an instance
of white ignorance. I agree with Mills that this would be an instance of active
ignorance. However, it is also a kind of ignorance that is very different in nature
from a white person’s ignorance about the facts of her own white privilege. In my
view in order to develop ways of minimising the occurrence of active ignorance
we need theories that detail the differences among types of ignorance more than
Mills’ account does.
Relatedly, because he does not focus on the differences among kinds of active
ignorance, Mills’ account would need supplementation if we are to make princi-
pled distinctions between instances of active ignorance where the ignorant person
should have known better from cases where ignorance is less culpable. That said,
as I mention below, our moral focus when addressing this pernicious ignorance
should not be directed toward apportioning blame. Instead, we should strive to
take responsibility for our ignorance and help others to see the importance of
doing the same.
I have argued that active ignorance is heterogeneous. In this chapter, I focus on
one of its species. My interest is with ignorance that is the result of a morally dubi-
ous desire not to know. This is active ignorance as motivated ignorance where
the motivations driving cognition are morally questionable and at variance with
accuracy. The motives that I discuss here are primarily related to the need to self-
enhance and feel good about oneself.9 Broadly speaking, this kind of ignorance
has two flavours. The first involves motivated cognition that leads to the forma-
tion of false beliefs.10 It can be thought as a will not to know. It is exemplified by
wishful thinking and by other forms of motivated believing (cf., Scott-Kakures,
2000). The second involves motivated avoidance of cognition that leads to the
absence of belief. It can be thought as comprising strategies of cultivated disen-
gagement with issues or studied avoidance of investigation often out of fear of
what one might discover (Spelman, 2007).
Motivated ignorance, unless it is the outcome of a conscious decision and in
the service of moral and epistemic goods, is hard to dislodge because it tends to
be invisible to those who suffer from it.11 The sturdiness of this kind of ignorance
is partly explained by its irrationality. It is impossible to sustain a wishful belief
whilst being fully aware of its wishful nature. Suppose, for example, that one
believes that one’s intellectual achievements are wholly attributable to one’s hard
work and talents rather than, to some extent, to one’s white privilege. If one came
to realise that one holds this belief because one is in fact complicit in an unjust
system that unfairly benefits one, rather than because of the evidence suggesting
it is true, one could not hold on to the belief and to the positive view of oneself.
More broadly, because of the constitutive links of belief to truth or to justification,
the discovery that one has no good epistemic reasons for a belief that one holds for
motives other than accuracy undermines one’s ability to sustain that belief. Since,
however, one’s non-epistemic motives for holding onto the belief are strong, the
pedigree of the belief must be of necessity invisible to one.
The invisibility of the kind of motivated ignorance that results in the absence
of true belief rather than the presence of false ones is caused by a different
dynamic.12 These are cases where subjects are motivated to avoid cognition and
the resultant belief formation. These cognitive strategies are also invisible to
the agent when the motives for thought avoidance are morally objectionable
by one’s own lights. Since omissions based on such motives are violations of
one’s personal standards, they are inconsistent with one’s conception of oneself.
They have therefore the potential to generate experiences of extreme discomfort.
Individuals tend to reduce this cognitive dissonance by justifying their behav-
iour.13 This leads to the formation of self-serving explanations of the reasons
motivating ignorance. This is the dynamic at work, for instance, when white
people refer to their ignorance to justify their lack of engagement with discus-
sion of race. They might explain that they do not know about such things and
do not wish to offend by misspeaking. Whilst such claims might on a few occa-
sions be evidence of “loving ignorance”, they often are ways of denying one’s
complacency about, and perhaps complicity in, the mechanisms that keep white
privilege in place (DiAngelo, 2012).
This second form of motivated ignorance might be more widespread among
those who occupy dominant positions. If one is sufficiently privileged so that he
can insulate himself from some realities, it is easier to preserve one’s absence of
belief than to work to sustain one’s false believing. The person who thinks that
his achievement is wholly due to his efforts and talents is often in the presence
of potential counterevidence.14 He is therefore frequently forced to discount or
explain away these considerations. This approach is cognitively costly because it
requires a lot of effortful motivated cognition. It is also possibly quite uncomfort-
able since it involves numerous instances of dissonance-inducing events that one
must work hard to reduce. Simply ignoring the issue, if one can do this, is instead
less cognitively effortful. One does not have to constantly find new ways to dis-
miss counterevidence. Instead, one might simply fail even to entertain whether
one’s achievements might be at least in part due to one’s own racial privilege.
Ignorance in these cases is, as they say, bliss.
3.2 Ignorance as a vicious sensibility

In this section I provide a vice-epistemological account of some forms of moti-
vated ignorance. I argue that individuals who have a high dispositional tendency
to motivated ignorance manifest a vicious sensibility that is a kind of epistemic
vice. I first briefly outline the view that epistemic vices include sensibilities as
well as character traits. I explain what a vicious sensibility consists in by way of
a contrast with the epistemically virtuous sensibility of being observant. Finally,
I compare my account of this type of vice with Medina’s view that racial insensi-
tivity is a kind of numbness. Within this framework I illustrate the affective and
motivational aspects of motivated ignorance. I thus show that ignorance is not a
purely doxastic phenomenon.
Epistemic vices are not all of one kind.15 Some are character traits such as,
arguably, intellectual arrogance and dogmatism. Others are bad cognitive habits
such as always jumping prematurely to conclusions. But there are also epistemic
vices that are best described as sensibilities. They involve deploying one’s facul-
ties in ways that systematically hinder effective and responsible inquiry.16 Vices
as sensibilities include the sensibilities that are responsible for motivated igno-
rance but also for testimonial injustices (Fricker, 2007).
The idea that virtues and vices are, or include, sensibilities has a long tradition
in virtue ethics. It is often endorsed by defenders of the view that moral facts can
be perceived by those who have developed the right kind of sensitivity (Clarke,
2018; McDowell, 1998). In epistemology, Hookway’s (2003) account of the
epistemic virtue of being observant is an analogue of this approach. Twenty-
twenty vision is not sufficient to be excellent at acquiring knowledge by means
of sight. While it is true that without reliable vision, one is not able to observe
well, in order to excel one must also be a skilled observer.17 What is characteris-
tic of the observant person is that she is motivated and able to direct her attention
to those portions of the visual field that are salient given her overall epistemic
goals. Since foveal vision is more accurate than peripheral vision, in order to see
well we need to know where to look. Being observant thus presupposes being
skilled at looking. That is, being observant requires knowing-how to look. In
short, the epistemic virtue of being observant includes the skill of looking in the
right places given one’s epistemic goal combined with the motivation to deploy
that skill.
While vision is a native cognitive faculty, good observation is learnt. A range
of activities require good observational skills. These include working as a radiog-
rapher or as an ornithologist. Those who are good at these professions have been
taught how to observe. Slowly, they have acquired the ability to look in the right
places, and to interpret what they see. They have learnt to notice what is salient,
and to make sense of visual patterns that are possibly ambiguous or would be just
a mess to the untrained eye. Examples such as these demonstrate that the virtue
of being observant is learnt through practice. They also indicate that this virtue
is highly domain-specific. The excellent radiologist might not make an excellent
ornithologist and vice-versa. Partly this is explained by the fact that observation
relies on the possession of a large amount of knowledge. The person who knows
nothing about cancer cannot tell what to look for in an X-ray. These considera-
tions suggest that there is not one virtue of being observant in general. Rather,
insofar as being observant is an acquired skill, this virtue is relative to a field of
knowledge. Thus, there is one virtue of being observant for a radiographer and
another for a bird watcher.
Being observant in a given domain, I have argued, involves a skill that impor-
tantly includes directing one’s attention to the locations in the visual field that are
salient given one’s epistemic goals. These are the locations that can supply the
information required to arrive at accurate answers to the questions that implic-
itly or explicitly guide one’s perceptual activity. For example, the observant fish
watcher directs her attention to the gills in order to ascertain whether, given the
general appearance of the fish that she is looking at, she is in the presence of a sur-
geon or a doctor fish. This direction of attention is on occasion based on conscious
decisions as in the example just supplied. At least as often, however, it is initiated
by emotional responses to the situation. The radiographer for instance might have
her attention captured by a pattern on the X-ray that causes her to feel anxious or
worried. Her anxiety makes her look more closely at that location in the visual
field rather than focus her attention elsewhere.18 These considerations point to the
crucial motivational contribution of affective states, such as episodic emotions,
to the virtue of being observant. Emotions supply the motivational impetus to
search for relevant visual information, but they also direct that search by capturing
attention to some locations that are identified as salient and marked as positive or
negative.19
We are now able to see why being observant is better thought of as a sensibil-
ity, that includes an affective component, rather than as a pure cognitive skill.
Knowing how and where to look involves being emotionally attuned to one’s
current situation. The radiographer becomes good at her job partly by training her
emotional responses to the X-ray (cf. Lance & Tanesini, 2004). Through trial and
error, practice and mentoring, she acquires the ability to look at an X-ray quickly
and only to pay close attention to those areas that at first sight worry her or cause
some anxiety. This emotional training is the development of a sensibility that
makes her visually sensitive to the facts that are salient given her epistemic goals.
One way to understand vicious sensibilities is to contrast them with virtuous
ones. More specifically, one might think that the vices opposed to the virtue of
being observant include all the multifarious ways in which one may lack this vir-
tue and thus be rightly described as inobservant. Here, I adopt a different take. I
reserve the label of epistemic vice for the presence of sensibilities that make one
insensitive to what is salient given one’s epistemic goals. In short, if virtue con-
sists in the presence of a sensibility that is an emotional response that is sensitive
to what matters, vice is not the mere absence of virtue, but the presence of a kind
of sensibility that comprises the emotional response of being insensitive to what
is salient given one’s epistemic goals.20
A person who lacks the virtue of being observant might be someone who has
her head in the clouds. She is often lost in her own thoughts and does not pay
much attention to her immediate surroundings. If employed as a bartender, this
person may fail to notice whether customers are present, or the order in which
they need to be served. This person’s behaviour can be usefully contrasted with
that of a person who, because of her motivated ignorance, tends to overlook older
white women when serving customers. This behaviour is very common. Some
people, including white women who have entered middle age, often experience
the feeling that they are invisible because they are not addressed or made eye-con-
tact with: they are, literally, overlooked. The people who do not pay attention to
them are not distracted by their own thoughts. They are paying attention to other
people. Their attention is captured by the white men, or, on occasion, the younger
white women in the room. These people have trained their emotional reactions to
focus on some people as salient and to discount the presence of people of differ-
ent kinds as distractors to ignore. They have tuned out some people and attuned
themselves to detect others. In short, they have cultivated insensitivity to the pres-
ence, needs, or features of some people. It is this sort of acquired insensitivity that
can be vicious morally and epistemically if it deflects one’s attention away from
at least some of one’s moral and epistemic goals.
I have argued that epistemic virtues can be sensibilities that involve skilled
emotional responses that are sensitive to what is epistemically salient (given some
of the agent’s goals) in the surrounding environment. Conversely, some epis-
temic vices are sensibilities that involve emotional responses that are trained to
be insensitive to, by failing to direct attention to or guiding attention away from,
what is epistemically salient (given some of the agent’s goals) in the surrounding
environment.21 In what follows, I first explain the role of insensitive sensibility in
motivated ignorance, before comparing my account to Medina’s view that racial
insensitivity is a kind of numbness.
I have described motivated ignorance as ignorance (that is, absence of true
belief or presence of false belief) that results from a desire not to know. I have
also claimed that in its least cognitively effortful incarnation, such ignorance is
the product of avoidance of cognition. This avoidance is cultivated, and it pre-
supposes that one is sufficiently privileged to be able to afford not having an
opinion on a given issue. That is, one must to some extent enjoy the luxury of
not needing to know in order to satisfy the need not to know.22 One way in which
people habituate themselves to remaining ignorant is by becoming insensitive.
Thus, insensitivity as a sensibility is a frequent component of motivated igno-
rance. There is more to motivated ignorance than the development of this kind of
sensibility, but its acquisition is often a pre-requisite of motivated ignorance at
least in those cases in which the motivations driving such ignorance are morally
or epistemically dubious.
The account offered above is largely consistent with Medina’s view that active
ignorance about race is a form of insensitivity that he describes as a kind of numb-
ness (Medina, 2016). What I have presented here fleshes out, and generalises,
some aspects of his insightful proposals. I wish, however, to highlight a worry
about his description of insensitivity in general and racial insensitivity in particu-
lar as numbness.
Medina justifies this terminological choice on at least two grounds. First, he
wants to highlight the affective dimension of insensitivity. Second, he wishes to
jettison the term, often used in the relevant literature, “blindness” because of its
discriminatory connotations (Medina, 2013, pp. xi–xiii). I share these motiva-
tions and agree that “numbness” is an improvement on “blindness”. However,
numbness is often used to indicate an overall loss of emotional tone. The person
who is numb is the person who, for whatever reasons, feels nothing. Those who
would rather not know the truth about racial privilege and how they benefit from
it are likely to lack empathy for the plight of ethnic minorities. Hence, they are
numb in that sense. Numbness, however, might not be an important component
of their emotional reactions to black people. Insensitivity as a sensibility might,
and often does, involve heightened emotional reactions (Ashton-James & Tracy,
2012; Haddock & Gebauer, 2011). The person who is motivated not to know
about white privilege might be extremely sensitive to the bodily presence of her
black work colleagues. She might experience physical closeness as discomfort-
ing; she might avoid eye and bodily contact and increase the physical distance
between her and them (Amodio & Devine, 2006).
These behaviours are the result of negative emotional reactions to black people.
These emotional reactions often draw attention away from what is epistemically
salient in these circumstances because they prevent one from paying attention to
what is being said or to how one is treating one’s colleagues. Thus, racial insen-
sitivity as a sensibility is often characterised by a negative emotional stance that
leads one to avert one’s eyes and body away from the other person. In turn, this
aversion results in not noticing, because one is ignoring, what is epistemically
salient for one in the exchange.
The description of some of the behaviour characteristic of racial insensitivity
as a sensibility bears close connections to what is described in the psychologi-
cal literature as aversive racism (Pearson et al., 2009). Aversive racism is often
defined as implicit prejudice and avoidant behaviour in the context of sincerely
held, non-prejudicial beliefs accompanied by good intentions. Thus, aversive rac-
ists are those who are implicitly biased, even though they explicitly hold egalitar-
ian views. I sidestep the issues of implicit bias here (cf. Holroyd, this volume).
Instead, I focus on the combination of avoidant behaviour with egalitarian beliefs
about racial groups to highlight three points that serve to clarify the nature of
racial insensitivity and to support the claim that it is an epistemically vicious
sensibility.
First, insensitive sensibilities are epistemically deficient because they consist
of trained emotional responses that draw attention away from what is epistemi-
cally salient in context. The empirical literature on aversive racism also indi-
cates that these sensibilities lead to limited searches for considerations that are
relevant to belief formation. Aversive racists are closed-minded because they
are motivated not to look extensively for relevant evidence (Roets et al., 2015).
The account of aversive behaviour as an expression of an insensitive sensibility
explains why and how such narrow search occurs. Emotions play a crucial role by
capturing and directing attention to the features of the environment that are rel-
evant to find information required to form accurate answers to the questions guid-
ing one’s epistemic inquiry. Insensitive sensibilities involve emotional responses
that inhibit wide searches for information whilst directing attention away from
what matters epistemically and toward other features of the situation that one’s
fear or anger highlights as important (Maio & Haddock, 2015, ch. 6; Tiendens &
Linton, 2001).
Second, the negative emotional reactions that capture attention in cases of
insensitive sensibility are often not fitting. These reactions misrepresent the situ-
ation since black people are not usually a threat. Further, they lead one astray
because they direct attention in ways that inhibit the formation of accurate judge-
ments about the situation. In this way, insensitive sensibilities are at the service
of the agents’ desire to be ignorant and avoid noticing things. However, they also
undermine agents’ own epistemic goals since these ostensibly include acquiring
an adequate understanding of the situation they find themselves in. Those white
students who keep silent because they claim to know nothing about race, would,
if asked, justify their ignorance by citing the difficulty of gaining information.
They would not claim disinterest in the issue. I presume here that, even though
they may be somewhat deluded, these students are not lying. They genuinely want
to know about race. However, their non-conscious desire not to know is stronger.
This is also why it is possible for some of these students to have egalitarian beliefs
about racial groups.
Third, the suggestion that insensitive racial sensibilities might be associated
with aversive racism raises questions about people’s moral responsibility for
their racially insensitive sensibility. There are several reasons against blaming
individuals for their sensibilities and for the active ignorance that they promote.
First, blame attributions might prove counterproductive since there is evidence
that people, when they are labelled as vicious, change their behaviour to fit the
label provided that they find the description somewhat plausible (Alfano, 2013,
pp. 96–97). Thus, blaming people for their racist sensibilities might make them
more racist. Second, it is often the case that others lack the required standing to
cast aspersions on those whose sensibilities are insensitive to race. For example,
it would be hypocritical of many who suffer from the same vices to blame others
for theirs.23
That said, whilst backward-looking blame attribution might be sometimes
unwarranted and frequently inadvisable, there is scope for promoting the taking
of responsibility, irrespective of culpability.24 That is, there are sound moral rea-
sons why we should all take shared responsibility for addressing the extensive
moral and epistemic harms caused by motivated ignorance and other forms of
active ignorance. With regard to racial insensitivity, the beneficiaries of white
privilege have special responsibilities to acknowledge and address their epistemic
limitations.25
3.3 Ignorance and arrogance

Arrogance, including intellectual arrogance, always tends to bring ignorance in
its trail.26 In this concluding section, I present an explanation of why this might
be the case. Arrogance, I argue, promotes ignorance because the self-important
self-conception that is the trademark of the arrogant can only be sustained by
purposeful cultivated self-ignorance.
I have shown elsewhere that there is a kind of arrogance that I have labelled
superbia that consists in a disposition to do others down in order to excel (Tanesini,
2016, 2018a). Those who have this character trait gain their sense of self-worth
by feeling superior to other people. This form of arrogance manifests itself in
behaviours designed to “big oneself up”, such as bragging, boasting, or arrogating
special entitlements, and in activities that diminish other people by humiliating
or intimidating them and discounting or dismissing their views. Ultimately, these
behaviours are in the service of defending a sense of self-esteem whose positivity
is fragile. For this reason, I have argued, arrogance is best thought of as a defen-
sive form of high self-esteem (Tanesini, 2018a, 2019).
If, as I hope it is granted, despite differences in ability and achievement among
human beings, no person is likely to be superior to all his acquaintances in all
respects, the individual who, in order to think well of himself, needs to feel
superior to others can only sustain a sense of superiority at the price of self-delu-
sion. He is likely to be ignorant about at least some of his own weaknesses and of
other people’s strengths. His ignorance is motivated by the need to self-enhance.
This motivation skews cognitive processes such as reasoning, thinking, evalu-
ating, remembering, and perceiving. The desire to enhance one’s sense of self-
worth biases cognition in at least two ways. First, it restricts the range of potential
evidence that is considered. The person whose thinking is driven by the need to
boost the ego focuses attention only on those factors that promote or inhibit the
satisfaction of this need. Second, it skews outcomes by setting asymmetry of error
costs (Scott-Kakures, 2000). For a person who feels that his self-esteem is under
attack, mistaking a threat for something non-threatening is costlier than the oppo-
site error. This person needs less evidence to conclude that something is a threat
to his self-esteem than he would need to reach the opposite conclusion. Hence,
arrogant individuals are extremely defensive and always ready to pounce in anger.
Arrogance therefore can only be sustained by actively remaining ignorant.
But an individual’s motivations are not sufficient to keep themselves ignorant.
Ignorance is often a luxury that only the privileged can afford. I have claimed
that the arrogant person can sustain self-belief only by feeling that he is better
than others. I have also argued that he can maintain this feeling only by avoiding
inconvenient truths and being very selective with the evidence. However, in non-
pathological cases, even individuals who are arrogant might be forced to modify
their opinions of their own abilities if they are faced with inconvertible evidence
of failure or with evidence that other people do not hold them in high regard.
It is virtually impossible for people who belong to less powerful groups to
avoid encountering such evidence that chastises any arrogance they might har-
bour. Their self-esteem is likely to be deflated by their awareness that many
members of dominant groups think of them in negative terms. This low self-
esteem might be experienced as shame for whom one is (Fanon, 1986). In addi-
tion, proper pride often spurs stigmatised individuals to try hard in order to prove
society wrong. Unfortunately, such attempts often result in underperformance,
because of the influence of fear of failure on cognitive activity (Steele, 2010). In
short, the combined effect of society’s judgement and of anxiety when faced with
stereotype-threat inducing situations provides those who are subordinated with
plenty of genuine and misleading evidence of their shortcomings.
Members of privileged groups instead can more easily avoid facing such poten-
tial evidence. Their privilege affords them a good education and other opportu-
nities to improve their abilities. In addition, other members of society generally
presume that they are capable, intelligent, and morally upright. They are frequently
praised for their successes whilst their shortcomings are often explained away as
due to bad luck or other circumstances beyond their control. Further, they might
often even have the luxury of not-trying to achieve. They are thus able to explain
their successes as the product of natural talent and rationalise failures as the result
of lack of interest and application. Either way, no possible outcome could be
treated as evidence of lack of ability. This strategy, known as self-handicapping,
is practiced by those whose high self-esteem is defensive (Lupien et al., 2010).
At Oxford and Cambridge third-class degrees – the lowest of the honours degrees
– are widely regarded as better than upper second (the degree classification just
below a first) because they show that one had other interests and, unlike slightly
better degrees, do not provide evidence that one is not the smartest. Of course, it
is only those whose third-class degree is from Oxbridge who can afford this strat-
egy. They are likely to have connections that will secure good jobs in any case.
What these considerations show is that arrogance is hard to preserve in the
absence of social privilege.27 This is because arrogance is predicated on being able
to sustain a large dose of motivated self-ignorance and is thus often accompanied
by an insensitive sensibility to the facts of one’s privilege. The privileged are best
placed to insulate themselves sufficiently from reality to avoid confronting their
own limitations. So, arrogance requires ignorance and frequently presupposes
privilege, but arrogance also bolster the inequalities that constitute privilege. I have
claimed that the arrogant person diminishes others in order to feel superior to them.
These behaviours often succeed in undermining other people’s efforts. Those who
are arrogant put obstacles in the way of others’ success; they humiliate and intimi-
date them. As a result, the recipients of these behaviours are prevented from devel-
oping their abilities to their full potential. Hence, arrogant behaviour contributes to
creating those social inequalities it requires to sustain itself less effortfully.28
This chapter has primarily focused on the relations between two epistemic vices
– racial insensitivity and intellectual arrogance – and active ignorance because of
their tight connections. I have argued that privilege often enables the kind of igno-
rance that arrogant people cultivate in order to persevere in their arrogant ways. I
have also indicated that the development of insensitive sensibilities is often cru-
cial when keeping oneself ignorant of the things one does not want to know. It is
likely that other epistemic character vices such as laziness, for example, might be
closely related to sensibilities that are especially suited to the production of igno-
rance (Medina, 2013, pp. 145–148). The exploration of these connections would
be required to provide a fuller account than I was able to provide here of the rela-
tions between the epistemologies of vice and of ignorance.
Notes
1 I set aside here the issue whether true belief which is short of knowledge is a kind of
ignorance which is at the centre of the dispute between supporters of the standard and
of the new view of ignorance (Le Morvan & Peels, 2016). I merely wish to record that
in my view there are cases where a person is not ignorant of a fact because they have
a true belief about it, even though their belief is not knowledge because it is not suf-
ficiently justified.
2 On motivated cognition see Kunda (1990).
3 These are the topic of several contributions to Proctor and Schiebinger (2008).
4 I have discussed these three kinds of case in my (2018b). I should add that they are not
mutually exclusive because one piece of ignorance can be the product of both decep-
tion and self-deception, for instance.
5 But note that one can be complacent or even complicit about the process; when this
happens one is deceived or misinformed but also engages in processes of self-obfusca-
tion.
6 Thanks to Quassim Cassam and Ian Kidd for independently pressing this point.
7 I have argued elsewhere for a motivational account of what makes epistemic vices
vicious (2018c).
8 His gloss on causation as counterfactual dependence might also open his account to
well-known counterexamples that beset these theories of causation.
9 There is good social psychological evidence supporting the view that this is an impor-
tant motivation in attitude formation and maintenance (Watt et al., 2008).
10 Or to the withholding of belief about whether p or not p.
11 The reason for such invisibility is cognitive dissonance (see note 13 below). This is not
the same mechanism discussed by Cassam (2015, 2019) when arguing that some vices
are stealthy.
12 That said, there is bound to be some overlap between these strategies for keeping one-
self ignorant. In both kinds of case agents will to some extent avoid considering incon-
venient evidence and engage in biased information gathering activities.
13 This is what is predicted by the Self-Standards model of cognitive dissonance. For an
overview, see Cooper (2008, pp. 104–116).
14 Unless, that is, he can afford to insulate himself from such evidence. In the final section
I show how social privilege creates the opportunities for such insulation.
15 The view that not all epistemic vices are character traits is ably defended by Cassam
(2019, this volume).
16 I borrow the notion of effective and responsible inquiry from Cassam (2016). This is
inquiry that reliably leads to knowledge and that leads to blameless belief because it
is belief that does not violate any intellectual obligations. As I explain below, I do not
think of sensibilities as being vicious exclusively because their bad effects but also
because of the motivations that animate them.
17 One can subscribe to the view that sensibilities are only virtuous if they are reliable
without endorsing a virtue-reliabilist account for them. My suggestion that they have
an emotional component indicates that sensibilities are virtuous only when they have
both good effects and good motivations.
18 On the view that emotions capture attention and thus motivate us to look in the right
places for information that leads to accurate judgement see Brady (2013). For a defence
of the role of anxiety in good judgment see Kurth (2018).
19 Emotions do not need to be themselves epistemic reasons for belief to fulfil this role, so
they need not be akin to perceptions of values. Instead, they may serve as mechanisms
for accessing those reasons by being something like conscious alarm bells that are reli-
ably connected to unconscious processing. When the bell rings we are alerted to where
to go to find answers.
20 Insensitivity might involve lack of awareness of what is actually salient or an excessive
preoccupation with some aspects that are not salient or are not deserving of the level of
attention that is directed at them. Hence, oversensitivity is in my vocabulary a form of
insensitivity since it entails that one’s sensibility is not well-calibrated or sensitive to
what is salient.
21 In my account vicious sensibilities are self-stultifying because they are driven by goals
–such as to feel good about oneself – rather than the desire for accuracy that the agent
also has but which is not efficacious in directing attention. It is also possible that an agent
lacks epistemic goals that they should have. Such an individual might be rightly criticis-
able for her disinterest which causes her ignorance. However, in cases such as this one,
it is at best unclear whether what is at work is some of sort of insensitive sensibility or
a shortcoming of a different kind such as the presence of an epistemic character vice.
Thanks to Alice Monypenny and to Quassim Cassam for pressing this point.
22 Hence, active ignorance is oftentimes the obverse of the epistemically privileged stand-
point of the subordinated (Collins, 2000, p. 11; Mills, 2007, p. 15).
23 Tognazzini and Coates (2018) offer an overview of some of these issues concerning the
moral standing of would-be blamers.
24 On the distinction between backward-looking attributions of responsibility and a for-
ward-looking assumption of responsibility see Card (1996). See Medina (2013, ch.
4) for an extensive discussion of the conditions under which one might be said to be
responsible for one’s ignorance.
25 See Medina (2013) for an extensive discussion of how to resist insensitivity.
26 The converse does not appear to hold about motivated ignorance in general. However,
a case can perhaps be made that racial insensitivity, as a particular type of motivated
ignorance, is often a symptom of complacency. The latter, even though not necessarily
arrogant, is something that can only be afforded by the privileged.
27 This is one of the reasons why arrogance is a vice associated with privilege and entitle-
ment. I do not intend to suggest either that all privileged individuals are arrogant or that
they are the only people beset by this character trait. Rather, my point is that privilege
facilitates the formation of an arrogant character, whilst underprivilege is a kind of real-
ity check that makes arrogance much less likely.
28 I would like to thank Alice Monypenny, Quassim Cassam, and Ian James Kidd for their
extremely helpful comments on an earlier draft. I am also grateful to Heather Battaly
for organising the Vice Epistemology Conference where this chapter was first aired.
Thanks also to audiences at the Universities of Connecticut, Glasgow, and Kent for
their comments on some of this material.
References
Alfano, M. (2013). Character as Moral Fiction. Cambridge: Cambridge University Press.
Amodio, D. M., & Devine, P. G. (2006). Stereotyping and evaluation in implicit race
bias: evidence for independent constructs and unique effects on behavior. Journal of
Personality and Social Psychology, 91(4), 652–661.
Ashton-James, C. E., & Tracy, J. L. (2012). Pride and prejudice: how feelings about the
self influence judgments of others. Personality and Social Psychology Bulletin, 38(4),
466–476.
Brady, M. S. (2013). Emotional Insight: The Epistemic Role of Emotional Experience.
New York: Oxford University Press.
Card, C. (1996). The Unnatural Lottery: Character and Moral Luck. Philadelphia: Temple
University Press.
Cassam, Q. (2015). Stealthy vices. Social Epistemology Review and Reply Collective,
4(10), 19–25.
Cassam, Q. (2016). Vice epistemology. The Monist, 99(2), 159–180.
Cassam, Q. (2019). Vices of the Mind: From the Intellectual to the Political. Oxford:
Cassam, Q. (2020). The metaphysics of vice epistemology. In Ian James Kidd, Heather
Battaly, and Quassim Cassam (eds.), Vice Epistemology, 37–52. London: Routledge.
Clarke, B. (2018). Virtue as a sensitivity. In N. E. Snow (Ed.), The Oxford Handbook of
Virtue (pp. 35–56). New York: Oxford University Press.
Collins, P. H. (2000). Black Feminist Thought: Knowledge, Consciousness, and the Politics
of Empowerment (10th Anniversary ed.). New York/London: Routledge.
Cooper, J. (2008). Cognitive Dissonance: Fifty Years of a Classic Theory. Los Angeles,
CA/London: SAGE.
DiAngelo, R. (2012). Nothing to add: a challenge to white silence in racial discussions.
Understanding & Dismantling Privilege, 2(1), 1–17.
Eddo-Lodge, R. (2017). Why I’m no Longer Talking to White People about Race. London:
Bloomsbury.
Fanon, F. (1986). Black Skin, White Masks (C. L. Markmann, Trans.). London: Pluto Press.
Fricker, M. (2007). Epistemic Injustice: Power & the Ethics of Knowing. Oxford:
Clarendon.
Haddock, G., & Gebauer, J. E. (2011). Defensive self-esteem impacts attention, attitude
strength, and self-affirmation processes. Journal of Experimental Social Psychology,
47(6), 1276–1284.
Holroyd, J. (2020). Implicit bias and epistemic vice. In Ian James Kidd, Heather Battaly,
and Quassim Cassam (eds.), Vice Epistemology (pp. 126–147). London: Routledge.
Hookway, C. (2003). Affective states and epistemic immediacy. Metaphilosophy, 34(1–2),
78–96.
Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3),
480–498.
Kurth, C. (2018). The Anxious Mind: An Investigation into the Varieties and Virtues of
Anxiety. Cambridge, MA/London: The MIT Press.
Lance, M., & Tanesini, A. (2004). Emotion and rationality. Canadian Journal of
Philosophy, 34(sup1), 275–295.
Le Morvan, P., & Peels, R. (2016). The nature of ignorance: two views. In R. Peels &
M. Blaauw (Eds.), The Epistemic Dimensions of Ignorance (pp. 12–32). Cambridge:
Cambridge University Press.
Lupien, S. P., Seery, M. D., & Almonte, J. L. (2010). Discrepant and congruent high
selfesteem: behavioral self-handicapping as a preemptive defensive strategy. Journal
of Experimental Social Psychology, 46, 1105–1108.
Maio, G. R., & Haddock, G. (2015). The Psychology of Attitudes and Attitude Change (2
ed.). London: SAGE.
Manson, N. C. (2012). Epistemic restraint and the vice of curiosity. Philosophy, 87(02),
239–259.
McDowell, J. H. (1998). Mind, Value, and Reality. Cambridge, MA/London: Harvard
University Press.
Medina, J. (2013). The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic Injustice, and Resistant Imaginations. Oxford/New York: Oxford University
Press.
Medina, J. (2016). Ignorance and racial insensitivity. In R. Peels & M. Blaauw (eds.), The
Epistemic Dimensions of Ignorance (pp. 178–201). Cambridge: Cambridge University
Press.
Mills, C. W. (2007). White ignorance. In S. Sullivan & N. Tuana (eds.), Race and
Epistemologies of Ignorance (pp. 13–38). Albany: State University of New York Press.
Mills, C. W. (2015). Global white ignorance. In M. Gross & L. McGoey (eds.), Routledge
International Handbook of Ignorance Studies (pp. 217–227). London/New York:
Routledge.
Pearson, A. R., Dovidio, J. F., & Gaertner, S. L. (2009). The nature of contemporary
prejudice: insights from aversive racism. Social and Personality Psychology Compass,
3(3), 314–338.
Peels, R. (2017). Responsible Belief: A Theory in Ethics and Epistemology. New York:
Proctor, R. N., & Schiebinger, L. (2008). Agnotology: The Making and Unmaking of
Ignorance. Stanford: Stanford University Press.
Roets, A., Kruglanski, A. W., Kossowska, M., Pierro, A., Hong, Y.-y., James, M. O., &
Mark, P. Z. (2015). The motivated gatekeeper of our minds: new directions in need for
closure theory and research. Advances in Experimental Social Psychology, 52, 221–283.
Scott-Kakures, D. (2000). Motivated believing: wishful and unwelcome. Noûs, 34(3),
348–375.
Spelman, E. V. (2007). Managing ignorance. In S. Sullivan & N. Tuana (eds.), Race and
Epistemologies of Ignorance (pp. 119–131). Albany: State University of New York
Press.
Steele, C. (2010). Whistling Vivaldi and Other Clues to How Stereotypes Affect Us. New
York/London: W. W. Norton.
Sullivan, S., & Tuana, N. (eds.). (2007). Race and Epistemologies of Ignorance. Albany:
State University of New York Press.
Tanesini, A. (2016). ‘Calm down, dear’: intellectual arrogance, silencing and ignorance.
Aristotelian Society Supplementary Volume, 90(1), 71–92.
Tanesini, A. (2018a). Arrogance, anger and debate. Symposion: Theoretical and Applied
Inquiries in Philosophy and Social Sciences, 5(2 (Special issue on Skeptical Problems
in Political Epistemology, edited by Scott Aikin and Tempest Henning)), 213–227.
doi:10.5840/symposion20185217
Tanesini, A. (2018b). Collective amnesia and epistemic injustice. In J. A. Carter, A. Clark,
J. Kallestrup, S. O. Palermos, & D. Pritchard (eds.), Socially Extended Epistemology
(pp. 195–219). Oxford: Oxford University Press.
Tanesini, A. (2018c). Epistemic vice and motivation. Metaphilosophy, 49(3), 350–367.
Tanesini, A. (2019). Reducing arrogance in public debate. In J. Arthur (ed.), Virtues in the
Public Sphere (pp. 28–38). London: Routledge.
Tiendens, L. Z., & Linton, S. (2001). Judgment under emotional certainty and uncertainty:
the effects of specific emotions on information processing. Journal of Personality and
Social Psychology, 81(6), 973–988.
Tognazzini, N., & Coates, D. J. (2018). Blame. In E. N. Zalta (ed.), The Stanford
Encyclopedia of Philosophy (Fall 2018 ed., pp. 1–42). Retrieved from https://plato.s
tanford.edu/archives/fall2018/entries/blame/.
Tuana, N. (2006). The speculum of ignorance: The women’s health movement and
epistemologies of ignorance. Hypatia, 21(3), 1–19.
Watt, S. E., Maio, G. R., Haddock, G., & Johnson, B. T. (2008). Attitude functions in
persuasion: matching, involvement, self-affirmation, and hierarchy. In W. D. Crano &
R. Prislin (eds.), Attitudes and Attitude Change (pp. 189–211). New York: Psychology
Press.
4 Epistemic corruption
and social oppression
Ian James Kidd
The dominant understanding of vice, virtue, and character needs to be revised in

light of the recognition by feminist theorists, among others, that character is not
simply a matter of what is inside the individual, for which the individual is wholly
and solely responsible, but is also a matter of interpersonal, social, cultural, and
political contexts.
(Robin Dillon, 2012: 90)
4.1 Character and oppression

A key insight of liberatory philosophies is a double connection between character
and social oppression. The operation of oppressive social systems requires
oppressors characterised by what Lisa Tessman calls ‘the ordinary vices of
domination’, such as cruelty, indifference, contempt, and arrogance, which make
for ‘degraded’, ‘twisted’ forms of moral character (2005: 54). Correspondingly,
subjection to behaviours and systems characterised by such vices itself will tend
to damage the characters of the oppressed, causing what Tessman, inspired by
the work of Claudia Card (1996), dubs ‘moral damage’, in the sense that ‘the
self under oppression can be morally damaged, prevented from developing or
exercising some of the virtues’ (2005: 4).
Such characterological damage has been explored through feminist and
critical race-theoretic philosophies within the Western tradition, at least back
to early work by Mary Astell and Mary Wollstonecraft at the start and end
of the seventeenth century (see Kidd 2019a: §2). Although much of this work
has focused on forms of damage to moral character, there has been a latent
sensitivity to damaged epistemic characters, too. Astell and Wollstonecraft’s
critiques of misogynistic social cultures exposed the damage to women’s
epistemic capacities and characters, as later critical race epistemologies did for
the damaged epistemic characters of members of subordinated racial groups. If
subjection to conditions of oppression damages the epistemic character of the
oppressed, then it’s reasonable to expect engagement with that fact from vice
epistemologists. Moreover, it’s also reasonable to expect them to have useful
resources to offer to those who want to understand the character-epistemic
dimensions of social oppression.
70 Ian James Kidd
A sensitivity to the connections between forms of social oppression and dam-
aged or malformed epistemic character is certainly a feature of contemporary
vice epistemology, for at least three reasons. To start with, a general interest in
epistemic vices should naturally pay at least some attention to the question of
how epistemic subjects come to acquire or develop those vices. Granted, such
aetiological interest need not necessarily attend to the features of the social world
focused upon by feminist and critical race theorists, such as intersectionally struc-
tured social identities, systems of social privilege, power hierarchies, and unjust
social institutions and traditions. Some prefer to focus on the psychological bases
of epistemic vices, or situational factors, or wider sociological, cultural, and ide-
ological conditions that structure our development – and maldevelopment – as
epistemic subjects (cf. Cassam 2019, Medina 2013).
A second reason to expect vice epistemologists to attend to the connections
between oppression and character is that many of them are working within femi-
nist philosophical frameworks. Standout examples include the work of Miranda
Fricker (2007) and José Medina (2012) on epistemic injustice, and the work
of Alessandra Tanesini (2016, 2018) on the vices of humility: arrogance and
superbia in the case of the privileged, and timidity and servility in the case of
the oppressed. A final reason why vice epistemologists ought to attend to char-
acter and oppression is that most of them have at least latent ameliorative aspi-
rations: one of the main reasons for systematically studying the epistemic vices
is that epistemically vicious people and patterns of behaviour are sufficiently
objectionable that it becomes imperative to find effective ways to reduce the
incidence of vices. This is what Quassim Cassam recently called ‘the project of
vice reduction’, whose main motivation is the sense that a world of unchecked
epistemically vicious behaviour is just ‘too ghastly to contemplate’ (2019: 186,
187).
A first step towards a vice epistemological analysis of the characterological
damage to the members of groups subjected to forms of social oppression will
be the development of a suitable concept to describe that phenomenon. A good
clue comes from some remarks made by Fricker and Medina when describing
the effects on people of sustained subjection to forms of epistemic injustice:
Fricker says that subjection to gendered and racialised epistemic injustices tends
to ‘inhibit’ or ‘thwart’ the development of subjectivity and character (2007: 49,
58). She also argues that internalisation of sexist and racist norms, values, and
assumptions ‘corrupts’ the epistemic sensibilities and dispositions of both oppres-
sors and the oppressed (2007: 93, 131, 138). Medina uses a similar language when
characterising an epistemic vice as ‘a set of corrupted attitudes and dispositions’,
and when arguing that as the period and intensity of a person’s subjection to con-
ditions of oppression increase, so ‘their epistemic character [will] tend to become
more corrupted’ (2012: 29, 72).
Although neither Fricker nor Medina think of the members of oppressed
groups as the passive victims of their environments, both find it natural to describe
the damage done to epistemic character using the term ‘corruption’ – a term that
gathers together other terms they use to describe such damage, such as ‘inhibit’,
Epistemic corruption and social oppression 71
‘thwart’, and ‘erode’. Taking my cue from Fricker and Medina’s use of the term,
my aim in this chapter is to offer a working analysis of the concept of epistemic
corruption. I think this concept could play a useful role in our efforts to under-
stand the connections between social oppression and epistemic character dam-
age and therefore enhance the ameliorative potential of vice epistemology. But
the concept also requires us to critically reconsider some of the conceptual and
methodological commitments of vice epistemology. I therefore end the chapter by
arguing that the concept of epistemic corruption can inspire the development of a
distinctive sort of vice epistemology that – inspired by the work of Robin Dillon
(2012) – I will label critical character epistemology.
4.2 The concept of epistemic corruption

The term ‘corruption’ has enjoyed a long career as a means of articulating moral
damage and other forms of degeneration experienced by people, whether as a
result of external or social conditions, or internal factors such as failures of the
will. Such rhetorics of corruption have often brought with them a vocabulary of
vice. Gabriele Taylor remarks that, where the virtues heal and benefit oneself and
others, ‘the vices corrupt and destroy’ (2006: 126), while Judith Shklar remarks
that vices ‘dominate and corrupt’ our character (1984: 200), while Alasdair
MacIntyre remarks that ‘the corruption of institutions is always in part at least
an effect of the vices’ (2013: 227). Within the Western tradition, the moral uses
of the term ‘corruption’ have generally referred to the loss or deterioration of the
essential, typically positive qualities of a thing, and typically in ways that funda-
mentally compromise its integrity or constitution. Although it need not be coupled
to characterological concepts, that is how I will use the term, albeit confined only
to epistemic character, rather than moral and political corruption (on the latter, see
Rothstein and Varraich 2017).
I propose that epistemic corruption occurs when one’s epistemic character
comes to be damaged due to one’s interaction with persons, conditions, processes,
doctrines, or structures that facilitate the development and exercise of epistemic
vices. Three comments on this initial definition. First, ‘damage’ can be under-
stood in two ways: the deterioration of any pre-existing virtues and integrity
already present in the subject’s character, or the failure of the subject to develop
an epistemic character characterised by virtues and integrity. I label these active
corruption and passive corruption, respectively, where the distinction is between
damage done to qualities they currently possess and to their development.
Second, we need to add some terminology: the corruptee is the person or thing
being corrupted and the corruptors are the persons or things doing the corrupting.
Sometimes, of course, they may be same thing, since some subjects are complicit
in their epistemic self-corruption. And third, the term ‘facilitate’ is purposefully
broad: depending on the context, it can mean ‘encourage’, ‘justify’, ‘legitimate’,
‘motivate’, ‘promote’, ‘provide conditions for’, and so on. We can roughly divide
these into two main modes of facilitation: material conditions and motivational
conditions.
72 Ian James Kidd
Since there are many ways to materially and motivationally facilitate the
development and exercise of epistemic vices, our job as epistemologists is easier
if we distinguish different modes of epistemic corruption. Without pretending to
be comprehensive, consider the following five modes – with the provisos that oth-
ers exist, and, in practice, that modes of corruption tend to merge into one another:
1 Acquisition: a corruptor can enable the acquisition of novel epistemic vices,

ones not previously a feature of the subject’s epistemic character. Imagine
a student with no natural dispositions or prior tendencies to arrogance who,
under the influence of very charismatic but arrogant teachers within a school
environment that rewards arrogant actions, comes to acquire the vice of
arrogance.
2 Activation: a corruptor can activate dormant epistemic vices, ones typically
latent or inactive in the subject’s epistemic character. Imagine a student with
some underlying arrogant tendencies that are, as a matter of luck, always just
‘below the surface’, never showing themselves. Unfortunately, the student
then has the bad luck to move to a school whose culture encourages inflated
forms of self-confidence, bordering on arrogance, which activate their dor-
mant arrogant dispositions.
The next three ‘modes of corruption’ are different, insofar as they involve changes
to vices already present and active in a subject’s character.
3 Propagation: a corruptor can increase the scope of a vice, the extent to which
it affects the range of the subject’s character. Imagine a person whose vices
only affect a certain set of activities or topics, such as someone dogmatic only
about music, but not about politics, science, or anything else. Unfortunately,
their initially localised dogmatism is amplified under the influence of their
peer group, spreading to encompass more and more topics. Soon, they are
dogmatic about all sorts of subjects, if not about all things, until that vice
comes to ‘infect their whole character’, in Annette Baier’s (1995: 274) useful
phrase.
4 Stabilisation: a corruptor can also increase the stability of a vice, reducing the
chances of the vice’s susceptibility to disruption. Imagine an arrogant person
prone to saying and doing arrogant things, but whose saving grace is also that
they don’t always act arrogantly and typically cease when challenged. Their
arrogance is unstable, since it fluxes on and off and is relatively easily acted
upon by others. Unfortunately, the social conventions under which they oper-
ate come to be transformed in ways that tend to stabilise their vices – certain
norms of censure break down, for instance, as the society slowly starts to
become more tolerant of public displays of arrogance. The consequence is
that the vice is gradually stabilised. Where it was once fluctuating, blinking
‘on’ and ‘off’, it has become stable and highly resistant to destabilisation.
5 Intensification: a corruptor can also increase the strength of a vice. Imagine a
person with only a weak tendency to dogmatism: confronted with challenges
to their beliefs, they are irksomely stubborn, but not aggressively resist-
ant, and if asked to defend their views are prone to shrug and ‘let it go’.
Unfortunately, they are internalising certain attitudes and assumptions as a
result of their increasingly privileged social and professional identities. The
consequence is that their once weak, incipient dogmatism slowly mutates
into a raging dogmatic hubris of a strongly adversarial, agonistic form.
These five modes of epistemic corruption should suffice to indicate some of the
main ways a subject could become more epistemically vicious as a result of inter-
actions with corruptors. The relevance and significance of the different modes
will depend on specifics of specific epistemic subjects, the structure and psychol-
ogy of the epistemic vices, and different socio-epistemic environments. To spell
out some of these, let me note some presuppositions of this analysis of epistemic
corruption.
4.3 Presuppositions and predicaments

To start with the most obvious presupposition, subjects have epistemic characters,
and these consist of a variety of epistemic dispositions or traits of fairly malleable
scope, stability, and strength. These include dispositions to argue, evaluate, judge,
perceive, reason, reflect, and understand in certain ways and, for most people,
most of these dispositions will lack the stability and strength that are constitu-
tive of virtues or vices. Our epistemic characters aren’t therefore exhausted by
our virtues and vices. Most of us have perhaps a handful of virtues or vices and
an array of more-or-less virtuous and vicious dispositions, which could be devel-
oped into the stable forms of virtue and vice – an ontology of character recently
defended with due reference to empirical psychology by Christian Miller (2017).
When an epistemic disposition or trait gains in strength, scope, and stability, then
it becomes a virtue (if positively valenced) and a vice (if negatively valenced).
A second presupposition – or set of presuppositions – is that a subject’s epis-
temic character is an active, ongoing product of an array of psychological, devel-
opmental, interpersonal, and structural-contextual factors. Much goes into the
shaping of one’s epistemic character, and a main purpose of the concept of epis-
temic corruption is to help us identify the relevant sorts of factors: our psychologi-
cal profiles, our patterns of social interactions, the diversity and the quality of our
peers, the forms and dynamics of our social institutions, the variety of epistemic
exemplars available as models of emulation, and the wider ramifying structures
of power and privilege which intersect and interconnect all of these and more. But
since this makes for a very complex picture, it’s helpful to draw on José Medina’s
concept of an ‘epistemic predicament’.
An epistemic predicament is the particular, contingent, and changing array of
epistemic challenges, deficits, needs, obstacles, and threats that a subject experi-
ences as a result of their positionality in the social world (2012: 28f, 34f). Writing
of differently socially situated subjects, Medina emphasises that ‘their epistemic
deficits are different, and their resources to overcome these deficits and to resist
74 Ian James Kidd
dominant ideologies are also different [as are the ways they] accrue epistemic
gains and losses’ (2012: 28). Given all of this, one’s epistemic predicament pro-
foundly influences the ways one can develop and deploy one’s epistemic agency.
I think this includes one’s vulnerability to, and capacities to resist, epistemically
corrupting conditions and processes. Medina separates the ‘predicaments of the
privileged’ from the ‘predicaments of the oppressed’, although he also empha-
sises their internal variety and fluidity relative to one another. Those who occupy
social positions of privilege are especially vulnerable to what he calls the ‘vices
of the privileged’, such as arrogance, closedmindedness, and epistemic laziness,
at least in certain of their forms (2012: §1.1). Although those vices are not con-
fined to the privileged, nor are they an inevitable feature of the epistemic char-
acter of the privileged, they do lie in their path as obstacles which they ought to
take special care to avoid. By contrast, members of socially oppressed groups are
especially susceptible to developing such vices as servility and timidity, although
these are neither confined to them, nor unanimous among the epistemic characters
of the oppressed.
A key feature of an epistemic predicament is that it shapes the specific range
of epistemic vice to which one is especially susceptible and the range of resources
one has for detecting and resisting epistemically corrupting influences and condi-
tions. The epistemic predicaments of the oppressed may include the challenge
of trying to devise effective strategies to protect and, if possible, to restore, their
fragile testimonial credibility and epistemic confidence. The epistemic predica-
ments of the privileged may include the challenge of resisting the acute tempta-
tions to epistemically arrogant patterns of behaviour that come with a privileged
social identity.
My account of epistemic corruption also includes several presuppositions
about the nature and activity of epistemically corrupting conditions. Such condi-
tions vary along at least four axes. Their scope is the breadth of their influence
within a given community, institution, or socio-epistemic environment and can
be broader or narrower. Their strength is the power of their capacity to corrupt,
where more strongly corrupting conditions are those more reliably capable of
corrupting subjects. Their stability is the capacity of those conditions to maintain
their corrupting power in the face of efforts to disrupt them or contingent changes
in the wider environment. Their specificity is the range of vices the conditions can
‘corrupt for’, where we should distinguish specifically corrupting conditions that
facilitate a specific range of vices – such as appetitive, rather than alethic, vices –
and generically corrupting conditions which will tend to facilitate a whole range
of vices. I’d speculate that epistemically monocultural environments can facilitate
a whole range of epistemic vices, from arrogance and closed-mindedness to dog-
matism and laziness to myopia and obliviousness.
I also presuppose that epistemic corruption is a diachronic and dynamic pro-
cess, one that unfolds over time and is dynamic in the double sense that it is active
and consists of both corrupting and counter-corrupting tendencies – a ‘push and
pull’. Epistemic corruption should not be thought of as a single, one-off instance
of radical character damage, although perhaps certain dramatic or traumatic events
could have this sort of destructive effect. In most cases, epistemic corruption takes
the form of a complex, constant, cumulative series of interactions, experiences,
and influences that is temporally extended and socially scaffolded, which is why
studying it must use the resources of feminist and social epistemologies (Dillon
2012, Daukas 2019).
The dynamic character of epistemic corruption also has a happy upshot. All but
the most hostile epistemic environments contain at least some edifying aspects.
We can imagine a radically epistemically corrupting environment, populated by
forthright exemplars of epistemic vice and whose practices and structures are
designed to systematically, materially, and motivationally facilitate epistemically
vicious conduct – a nightmarish scenario. But such environments are extremely
rare, and most actual epistemic environments are partially rather than radically
corrupting. Even within small communities, such as a workplace, one can almost
always find a variety of virtuous and vicious epistemic exemplars, for instance.
4.4 Epistemically corrupting conditions

A fundamental aspiration of vice epistemology ought to be the improvement of
the epistemic characters of epistemic subjects and, as a corollary, the provision of
practicable methods for identifying and correcting epistemically corrupting con-
ditions. Such work must be at once philosophically and empirically sophisticated,
bringing together vice epistemology, sociology, and psychology as well as other
allied disciplines, such as educational studies (cf. Kidd 2019b).
I content myself with sketching out some examples of generically epistemically
corrupting conditions. Their purpose is both to help fill out the idea of epistemic
corruption and, more importantly, to help make the case for the development of
the distinctive form of critical character epistemology to which the subsequent
sections are devoted.
1 The absence of exemplars of virtue: a community may lack any positive

exemplars who model epistemic virtue. Those within the environment there-
fore lack admirable and emulable exemplars of epistemic virtue, who can
practically exemplify the virtues and/or provide theoretical elaboration of
virtues, to speak in the terms of the exemplarist character theory developed
by Linda Zagzebski (2017). Such absence does not, of course, automatically
put people on a path to vice, but it does close off at least one important path
to virtue (Croce and Vaccarezza 2017).
2 The derogation of exemplars of virtue: a community has exemplars of virtue,
but they could be subject to derogation by other members of that community –
sneered at, or mocked, or derided either openly or in private. Imagine a
philosophy department where the ‘tender-hearted’ professors who are fair-
minded and temperate in debates are mocked for being ‘soft’ or for lacking
the aggressively adversarial spirit allegedly constitutive of properly impres-
sive philosophising – a pattern of derogation criticised by feminist argumen-
tation theorists (Rooney 2010).
76 Ian James Kidd
3 The valorisation of vicious conduct and exemplars: some or all of the mem-
bers of a community may celebrate exemplars of epistemic vice and also
promote, celebrate, and reward epistemically vicious behaviour. Seeing that
social and professional goods such as respect, esteem, praise, and recogni-
tion reliably accrue to vicious exemplars can provide incentive to epistemic
viciousness. Moreover, the public celebration and advancement of the epis-
temically vicious sends clear messages about the possibilities available, if
one cultivates those vices in one’s own life.
4 The ‘rebranding’ of vices as virtues: some people will become aware of the
ways that corrupting conditions are starting to affect their epistemic character
for the worse and they might try to take countermeasures. An obvious way to
respond to those countermeasures is to try to disguise or conceal the fact of
their being corrupted, and one way is to ‘rebrand’ vices as virtues (cf. Dillon
2012: 99). Think of how arrogance can become confidence, or dogmatism can
become tenacity. If successful, this conceals the fact of corruption and so ena-
bles it to continue unhindered, at least until the concealment can be revealed.
An outstanding analyst of such rebranding is the Duc de La Rochefoucauld,
whose Reflections: Or, Sentences and Moral Maxims of 1664 opens with
the maxim, ‘Our virtues are most frequently but vices disguised’ – a reflec-
tion of an early modern European culture of cynicism about virtue, explored
in Michael Moriarty’s interesting book, Disguised Vices (2011), especially
chapter 12.
5 The establishment of conditions that increase the exercise costs of virtue:
virtues will typically incur exercise costs, whether practical, psychological,
social, or epistemic. An effective way to discourage the cultivation or exer-
cise of virtue is to increase the costs incurred by its exercise. Think of typical
costs for exercising the virtue of honesty: upsetting friends, embarrassing the
powerful, angering those who wanted things kept hidden, sacrificing oppor-
tunities whose availability was contingent on one’s failure to honestly reveal
their existence to others, and so on. By increasing exercise costs for epistemic
virtues, the path to vice becomes easier and more attractive. Again, increas-
ing exercise costs does not automatically make a person vicious, but does
alter their capacity to explore the space of character-epistemic developmental
possibilities in the direction of the virtues of the mind.
6 The establishment of structures that encourage the exercise of vice: these
refer to ways of materially and motivationally facilitating epistemic vicious-
ness. Imagine a doctor working for a tobacco company, which incentivises
acts of dishonesty by financially rewarding the publication of journal and
newspaper articles that insincerely question the connections between smok-
ing and various diseases – this being one of many forms of epistemically
corrupting conditions created by the tobacco industry, as documented by his-
torians and sociologists of science, including Robert Proctor, who refers to it
as ‘an unparalleled corruption of science’ (2011: 561).
7 The establishment of policies whose enactment requires the exercise of vice:
our social and institutional environments are often organised and directed by
policies, which create and empower certain norms, practices, and forms of
organisation. If enacting a policy in these ways requires the exercise of vices,
then the policy will be corrupting in the material and motivational senses.
Several critics have argued that certain higher educational policies in the
United States and United Kingdom do this in the case of the appetitive and
alethic vices, like epistemic insensibility (Battaly 2013, Cooper 2008).
No doubt there are many other forms of corrupting conditions, as well as vari-
ations in the forms taken by the ones sketched above. There is also interesting
work to do in tracing out in detail the correlations between modes of corruption,
corrupting conditions, and specific vices. There is also more to say, of course,
about the psychology of epistemic corruption, the ways that the structure of dif-
ferent vices relates to different modes of corruption, and so on. Such details are
best provided in conjunction with specific concrete case studies, for which at the
moment the best sources are analyses of epistemic corruption in education (see
Kidd 2015 and 2019b).
For now, I hope these examples suffice to give a fuller picture of the nature
and sources of epistemic corruption. I want to ask how this awareness of the phe-
nomenon of epistemic corruption may require changes to the ways we conceive
of epistemic vices and also how we practice vice epistemology.
4.5 Critical character epistemology

The phenomenon of epistemic corruption clearly relates to a variety of issues and
themes of vice epistemology – most obviously, to aetiological questions about the
ways that subjects come to have epistemic vices, ameliorative questions about the
possibility of and practices for vice reduction, and explanatory questions about
the relationship between vice-based and other ways of explaining problematic
epistemic behaviour (structural and situational, say). It seems clear that theorising
epistemic corruption requires a constant, direct sensitivity to agent–environment
interactions and the ways that epistemic character and agency are deeply shaped
by the norms, practices, structures, and power systems of social environments.
I want to argue that taking seriously epistemic corruption, as a pervasive fea-
ture of our socio-epistemic environments and a major factor in the development
of epistemic character, needs a style of vice epistemology with certain distinc-
tive conceptual and methodological commitments. I call this a critical character
epistemology, taking as its model and inspiration the critical character theory
developed by Robin Dillon. It aims ‘to understand moral character as affected by
domination and subordination and by the struggles both to maintain and to resist
and overthrow them’ (2012: 84). It challenges traditional forms of philosophi-
cal character theory for their double neglect of the socio-political dimensions of
character and the negative side of character, primarily vice and oppression (and,
I would add, corruption).
Dillon emphasises several other features, three of which stand out for my pur-
poses. First, critical character theory conceives of individuals as situated within
78 Ian James Kidd
social practices and institutions that are organised and animated by systems of
power and shape, what Medina calls their ‘epistemic predicaments’. Second, its
conceptual pluralism: an active sensitivity to the complex interconnections among
the core organising concepts of ethical theory, rather than taking one – such as
character, action, or intention – as fundamental. Third, critical character theory
shares with critical theory the deep aspiration ‘to liberate human beings from the
circumstances that enslave them’ (Horkheimer 1982: 244). But where critical the-
ory tends to focus on material, social, and political dimensions of oppression, for
Dillon, critical character theory ‘springs from the recognition that enslavement is
not only social and material but also operates on and through character’ (2012: 85).
A critical character epistemology aims to deploy Dillon’s insights with an
eye on those specifically epistemic characterological dimensions of oppression,
as articulated using the concept of epistemic corruption. Analyses of the dam-
age done to the epistemic characters of oppressed subjects can only be fully
achieved with a vice epistemology characterised by the sort of socialised, plural-
istic, and liberatory stance built into critical character theory by Dillon. I therefore
want to sketch out some options for a suitably revised form of critical character
epistemology.
I focus on three features, which I label aetiological sensitivity, axiological plu-
ralism, and normative contextualism. Of these, the last will likely prove the most
contentious.
4.5.1 Aetiological sensitivity

A concern with epistemically vicious characters should naturally invite an inter-
est in the processes or conditions that contribute to the formation and retention
of epistemic vices. Much of this interest is currently reflected in debates about
agential responsibility for epistemic vices, such as the distinctions between
acquisition-responsibility and retention-responsibility (Cassam 2019: ch.6) and
between responsibilist and personalist forms of vice (Battaly 2016). But these are
specific uses of what I have elsewhere called aetiological sensitivity (Kidd 2016).
It names a commitment to actively attend to the complex, contingent conditions
under which the epistemic characters of subjects develop (the Greek, aitía, means
‘cause’, and in medicine, ‘aetiology’ refers to both the causation or origination of
a disease and its study).
Aetiological sensitivity is a methodological commitment that urges the theorist
to combine description of the current state of epistemic character with attentive-
ness to its developmental history. To describe someone as vicious is to describe
the state of their character. To describe them as corrupted is to say something
about how they got into that state. It’s for this reason that a critical character epis-
temologist will usually prefer to refer to corrupted characters, rather than vicious
characters. (Compare describing someone as a radical with describing them as
having been radicalised – the latter term, of course, is used in terrorism studies,
a discipline whose use of aetiological explanations has been critically discussed
by Cassam 2018.)
Since epistemic character can be shaped by psychological, developmental,
interpersonal, structural, and historical factors, there are many sorts of factors to
which one may need to be aetiologically sensitive. Our characters are shaped by
our attitudes, cognitive styles, and temperaments; by the particular sequencing
and patterns of events and experiences we undergo during our lives; by the diver-
sity and characteristics of the subjects with whom we can or must interact, includ-
ing the particular ways they respond to our developing epistemic characters; and
by our emplacement within systems of social practices and institutions that organ-
ise the epistemic possibilities that are available and made salient for the members
of different groups; by the wider cultural and ideological contexts within which
play out emerging and entrenched structures of power, such as the inherited, self-
sustaining systems of ‘white ignorance’ described by Charles Mills (2007).
Luckily, aetiological sensitivity does not require appeal to these multifactorial
explanations of epistemic corruption in every case. Sometimes narrower forms of
explanation will do the job (cf. Cassam 2019: 27). Sometimes, though, our analy-
ses of epistemic corruption must go ‘all the way back’, into the contingencies of
history, and ‘all the way down’, to the underlying social structures or theoretical
visions shaping conceptions of our epistemic predicament. Explanatory adequacy
depends on one’s explanatory aims. Medina, for instance, typically prefers to tell
socially and historically complex stories about the ‘sociogenesis’ of epistemic
character traits – the idea that ‘epistemic virtues and vices [have] distinctive lines
of social development’ – because he wants to explore the connections between
epistemic corruption and sexist and racist structures and cultures (2012: 30).
There are two sorts of vice epistemology that typically show sensitivity to the
aetiology of epistemic vices and to the phenomenon of epistemic corruption. The
first is work in character epistemologies of education, much of which is concerned
with the possibility that educational practices and systems can be epistemically
corrupting (Battaly 2013, Johnson 2020, Kidd 2019). If an educator worries about
the vices of their students, then they should identify the sources of those vices,
especially if their own teaching practices maybe among them. A better under-
standing of the aetiology of vice can help us better educate for virtuous epistemic
character. We also find aetiological sensitivity in liberatory vice epistemologies,
such as Medina’s and Tanesini’s accounts of the epistemically corrupting effects
of systems of subordination and domination, for the privileged and the marginal-
ised, alike. Aetiology here plays a double role. It serves the ameliorative aspira-
tion of liberating people from oppression, since dismantling corrupting conditions
is an aim of that project. It also helps to encourage a properly pluralistic stance on
our evaluations of, and responses to, epistemically vicious subjects.
Consider evaluative responses to vice. Cassam argues that epistemic vices are
always proper objects of criticism, since vices are failings independent of their
aetiology. The other typical evaluative response to viciousness is blame, on which
virtue and vice theorists have tended disproportionately to focus. But this is dou-
bly problematic. First, certain forms of blaming are part of a standard strategy of
oppression, which aims to obscure the unjust structures that corruptively generate
those vices and failings – a sensitivity informing what Dillon calls ‘the politics of
80 Ian James Kidd
character appraisal’ (2012: 100; cf. Medina 2012: ch.4). Second, a myopic focus
on blaming occludes other responses to corrupted subjects, like anger, disappoint-
ment, regret, and sadness. Their appropriateness is often determined by aetiologi-
cal considerations: if your vices are the product of sustained subjection to acutely
corrupting conditions from which escape was impossible, then the appropriate
responses may be anger at those conditions and sadness at the damage done to
your character (cf. Cassam 2019: 21–22).
Aetiological sensitivity also encourages a more pluralistic sense of our practi-
cal responses to epistemically vicious characters. By tracing the conditions that
corrupt our characters, one can better identify what one can do to reduce the
incidence and intensity of vices. Corrective practical strategies can be individual
(self-discipline, edifying instruction) or structural (social activism, educational
reform). Sometimes, one should act at both levels, bearing in mind the fuzzi-
ness of the agents/structures distinction and Dillon’s guiding vision of character
as ‘fluid, dynamic, and contextualised’, and ‘processive rather than substantive’
(2012: 105). Crucially, our decisions about the appropriateness of different strat-
egies should be informed by the aetiology of the vicious character of different
subjects. Employing individual strategies for vices that are products of epistemi-
cally corrupting structures means we are tricked into playing a febrile form of
ameliorative whack-a-mole, rather than really transforming the surrounding struc-
tures. Developing those practical possibilities into the future will, of course, be
closely tied to questions about the nature of aetiological explanations: are they
more akin to mechanical explanations, aimed at identifying relevant underlying
processes, or more like an autobiographical description of the particular course of
one’s development into the person one becomes?
Aetiologically sensitive analyses of epistemic vices motivate the conviction
that responses to corrupted agents should therefore be evaluatively and practically
pluralistic. When confronted with corrupted subjects with vicious characters,
our evaluative responses can include anger, blame, disappointment, frustration,
regret, and sadness; moreover, these are often messily bound up with one another
in affectively and interpersonally complex ways. A consequence of this for criti-
cal character epistemology is that a default suspicion of blaming, which a critical
character epistemologist will regard as one response among others. Blaming is
often counterproductive, unjust, and liable to perpetuate patterns of oppression.
Dillon puts the point well in her advice that when confronted with corrupted sub-
jects, the point is not to blame but ‘to determine what to do, with or for whom,
and how’ (2012: 92).
4.5.2 Axiological pluralism

Axiological pluralism is the conviction that a proper appraisal of the badness of
vices cannot always be achieved if it appeals only to epistemic values. Most analy-
ses of epistemic vices are axiologically monistic insofar as they appeal to one type
of value, namely, epistemic value. Consider the main conceptions of epistemic
vice. Consequentialist analyses construe vices as tending systematically to create
a preponderance of bad epistemic effects, or failing to cause a preponderance of
good epistemic effects for the bearer, other agents, or the environment (according
to Cassam’s obstructivism, for instance, epistemic vices are traits that obstruct
the gaining, keeping, and sharing of knowledge). Motivational analyses construe
vices as either expressing bad epistemic motives, such as indifference to truth or
a desire to thwart the epistemic agency of others, or as marking the absence of
good epistemic motives and values, such as what Linda Zagzebski argued is the
fundamental motivation for all epistemic virtues, the desire for ‘cognitive contact
with reality’ (1996: 167). On both conceptions, the badness of epistemic vices is
defined in relation to epistemic values, whether a lack of love of truth, or of desire
to acquire knowledge, or a lack of aspiration to ‘cognitive contact with reality’.
Granted, none of these deny that epistemically vicious conduct is also bad non-
epistemically, too. But consideration of the non-epistemic tends to be a secondary
aspect of their analyses.
I think analyses of epistemic corruption motivate a rejection of axiological
monism. Appraisals of epistemic corruption require axiological pluralism. In a
weaker form, pluralism says that appraisal of the badness of at least some epis-
temic vices requires appeal to epistemic and non-epistemic, ethical, socio-political
values. Explaining the wrongs of vicious testimonial injustice requires inclusion
of the moral wrongs, social harms, and political injustices caused by unfair defla-
tions of agential testimonial credibility (see Congdon 2017; Pohlhaus, Jr. 2014). It
would, at the very least, be odd to attempt any fully satisfying explanation of the
wrongs of testimonial injustice that didn’t reference its integrated moral, social,
and political dimensions (Fricker 2007: 44, 54).
In its stronger forms, axiological pluralism says that appraisal of the badness of
at least some epistemic vices requires a rejection of the epistemic/non-epistemic
distinction. Nancy Daukas, for instance, speaks of ‘ethico-epistemic’ traits and
dispositions (2019: 381) while Fricker speaks of ‘hybrid’ ethico-epistemic virtues
(2007: §5.2). This more radical pluralism maintains that the epistemic, ethical,
social, and political dimensions of life are too indissolubly bound up with one
another to be even notionally separated. Indeed, the very attempt to separate them
may invite suspicion, given Kristie Dotson’s (2012) lucid warning that doing so
can serve to conceal and so perpetuate certain aspects of social oppression. If the
epistemic is political, then studies of corruption need a richer axiology.
4.5.3 Normative contextualism

The third feature of a critical character epistemology, perhaps also the most con-
tentious, is normative contextualism: the conviction that the normative status of
some or all epistemic character traits is dependent on the epistemic predicament
of a given epistemic subject. This plays on Dillon’s critical challenges to conven-
tional classifications of traits as virtues or vices: certain traits traditionally classi-
fied as virtues may be vices for the oppressed, and likewise in the case of ‘vices’
which could be ‘reclaimed’ as ‘liberatory virtues’ (2012: 98). The conclusion of
the critical character theorist is that the valence, appropriateness, and possibility
82 Ian James Kidd
of certain epistemic character traits is contextually sensitive. We should not pre-
judge the normative status of traits – as vicious or virtuous – since much depends
on the socially situated epistemic predicament of the subject in question.
Normative contextualism can take a variety of forms, with the weaker thesis
being that at least some epistemic character traits have a default normative status,
that can be distorted under certain epistemically hostile conditions. Consider Heather
Battaly’s account of the trait of closedmindedness, characterised as an unwilling-
ness or inability to engage seriously with relevant epistemic options. She argues
that, under epistemically hostile conditions, this trait is an effects-virtue, since its
exercise would tend systematically to help one retain true beliefs, avert epistemic
opportunity costs, and pursue one’s own epistemic projects (Battaly 2018a, 2018b:
§4). It may at least be a ‘burdened’ epistemic virtue, in Lisa Tessman’s sense of
‘traits that make a contribution to human flourishing—if they succeed in doing so at
all—only because they enable survival of or resistance to oppression, while in other
ways they detract from their bearer’s well-being’ (2005: 95).
The stronger form of normative contextualism denies that epistemic character
traits have even a default normative status. Independently of a specific epistemic
predicament, there are no grounds for normative evaluation of a trait as virtuous
or as vicious. A critic might protest that at least some traits must have a default
normative status, perhaps high-fidelity vices such as cruelty or epistemic malevo-
lence. But the critical character epistemologist regards this as an empirical claim,
vulnerable to the provision of concrete cases where those traits serve the epis-
temic and other interests of oppressed subjects. Perhaps the most overt advocate
of a strong form of normative contextualism is Dillon, who argues that ‘character
and character assessment may be deeply context dependent’ (2012: 100):
[According to a critical character theory] character dispositions would be

understood to be inculcated, nurtured, directed, shaped, and given significance
and moral valence as vice or virtue in certain ways in certain kinds of people
by social interactions and social institutions and traditions that situate peo-
ple differentially in power hierarchies; and we would understand vices in and
among individuals as, among other things, dispositions that support, direct,
shape, and give significance and value to social interactions and institutions.
(2012: 104)
The trick, of course, will be to develop accounts of epistemic character traits and
dispositions that are properly neutral, in the sense of not prejudging their nor-
mative status. Daukas gives the example of Roberts and Wood’s account of the
vice of vanity as an ‘excessive concern to be well-regarded by other people and
thus a hypersensitivity to the view that others take of oneself’ (2007: 259). She
argues that the trait of vanity functions as a virtue of the members of oppressed
groups, for whom intense concern about others’ perception of them ‘expresses a
realistic caution, a pragmatically necessary vigilance in self-monitoring’ (2019:
381). A critic will object that since vanity was characterised in terms of ‘exces-
sive’ concern and sensitivity, it is vicious by definition. But the critical character
epistemologist responds that Roberts and Wood’s original definition was imper-
fect, since it failed to honour trait-neutrality. Perhaps it is better to define the
trait of vanity as the disposition to actively and deliberately configure one’s self-
presentation to try to ensure one is perceived and evaluated positively by others.
Characterising the trait in this way is hopefully neutral, since that disposition
could take on virtuous or vicious forms, depending, for instance, on the sorts of
motives or effects with which it can become associated.
A critical character epistemologist proposes that we start by conceiving of
epistemic character traits as normatively neutral, only assigning them the status
of virtue or vice once we have considered carefully the range of values judged
to be proper to epistemic character evaluation and the epistemic predicaments of
different socially situated subjects. What we might end up with is a more complex
picture of a variety of epistemic character traits whose normative status is much
more contingent than is tolerable for traditional character theory.
4.6 Conclusions
I have offered a working analysis of the phenomenon of epistemic corruption
and argued it should play a more central role within vice epistemology. An epis-
temically corrupted subject has experienced certain forms of characterological
damage due to interactions with features and members of the social world that
facilitate the development and exercise of vices. There are several modes of epis-
temic corruption and complex stories to tell about the ways that a subject experi-
ences and resists the epistemically corrupting effects of their environment. We
also find strong precedent for analyses of epistemic corruption in contemporary
and historic vice epistemological projects, including precursor work in early mod-
ern English feminist vice epistemology and more recent critical race-theoretic
character epistemologies, such as those of W.E.B. Du Bois (see Kidd 2018: §2A).
I also proposed that studying epistemic corruption as a mode of oppression
requires a critical character epistemology. It ought to be aetiologically sensitive to
the complexity and contingency of the conditions that shape epistemic character,
axiologically pluralistic in the range of values used in appraisals of epistemic
character traits, and normatively contextualist about the status of character traits
for differently socially situated predicaments of different agents. Finally, critical
character epistemology is ultimately liberatory in its aspirations. It aims to iden-
tify and dismantle epistemically corrupting conditions and to find ways to repair
damaged epistemic characters. It is unclear how different a critical character epis-
temology would be from the current forms of vice epistemology. I suspect the real
points of difference will be axiological pluralism and normative contextualism,
especially in their stronger forms, but, hopefully, time will tell.
Acknowledgements
I am grateful for many constructive discussions with the Vice Squad and for the
comments and encouragement of Alice Monypenny, Taylor Tate, and audiences
84 Ian James Kidd
at the Universities of Connecticut, Durham, Nottingham, and Sheffield. The ideas
in this paper are also obviously deeply indebted to Robin Dillon’s work.
References
Baier, A. (1995) Moral Prejudices: Essays on Ethics (Cambridge, MA: Harvard University
Press).
Battaly, H. (2013) “Detecting Epistemic Vice in Higher Education Policy: Epistemic
Insensibility in the Seven Solutions and the REF,” The Journal of Philosophy of
Education 47 (2): 263–280.
Battaly, H. (2016) “Epistemic Virtue and Vice: Reliabilism, Responsibilism, and
Personalism,” in C. Mi, M. Slote, and E. Sosa (eds.), Moral and Intellectual Virtues in
Chinese and Western Philosophy: The Turn Towards Virtue (New York: Routledge),
99–120.
Battaly, H. (2018a) “Closed-Mindedness and Dogmatism,” Episteme 15 (3): 261–282.
Battaly, H. (2018b) “Can Closed-Mindedness be an Intellectual Virtue?” in S. Barker, C.
Crerar, and T. Goetze (eds.), Harms and Wrongs in Epistemic Practice, Royal Institute
of Philosophy Supplement 84: 23–45.
Card, C. (1996) The Unnatural Lottery: Character and Moral Luck (Philadelphia: Temple
University Press).
Cassam, Q. (2012) “Vice Epistemology,” The Monist 99 (3): 159–180.
Cassam, Q. (2018) “The Epistemology of Terrorism and Radicalisation,” in S. Barker, C.
Crerar, and T. Goetze (eds.), Harms and Wrongs in Epistemic Practice, Royal Institute
of Philosophy Supplement 84: 187–209.
Cassam, Q. (2019) Vices of the Mind: From the Intellectual to the Political (Oxford:
Congdon, M. (2017) “What’s Wrong with Epistemic Injustice? Harm, Vice, Objectification,
Misrecognition,” in I.J. Kidd, G. Pohlhaus, Jr. and J. Medina (eds.) The Routledge
Handbook of Epistemic Injustice (New York: Routledge), 243–253.
Cooper, D.E. (2008) “Teaching and Truthfulness,” Studies in Philosophy and Education
27: 79–87.
Croce, M. and M.S. Vaccarezza. (2017) “Educating Through Exemplars: Alternative Paths
to Virtue,” Theory and Research in Education 15 (1): 5–19.
Daukas, N. (2019) “Feminist Virtue Epistemology,” in H. Battaly (ed.), The Routledge
Handbook of Virtue Epistemology (New York: Routledge), 379–391.
Dillon, R. (2012) “Critical Character Theory: Toward a Feminist Perspective on ‘Vice’
(and ‘Virtue’),” in S.L. Crasnow and A.M. Superson (eds.), Out from the Shadows:
Analytical Feminist Contributions to Traditional Philosophy (New York: Oxford
University Press), 83–114.
Dotson, K. (2012) “A Cautionary Tale: On Limiting Epistemic Oppression,” Frontiers: A
Journal of Women Studies 33 (1): 24–47.
Fricker, M. (2007) Epistemic Injustice: Power and the Ethics of Knowing (Oxford: Oxford
University Press).
Horkheimer, M. (1982) Critical Theory (New York: Seabury Press).
Johnson, C. (2020) “Teaching to the Test: How Schools Discourage Phronesis,” in I.J.
Kidd, Q. Cassam, and H. Battaly (eds.), Vice Epistemology (New York: Routledge).
Kidd, I.J. (2015) “Educating for Intellectual Humility,” in Jason Baehr (ed.), Intellectual
Virtues and Education: Essays in Applied Virtue Epistemology (London: Routledge,
2015), 54–70.
Kidd, I.J. (2016) “Charging Others with Epistemic Vice,” The Monist, 99 (3): 181–197.
Kidd, I.J. (2019a) “Deep Epistemic Vices,” Journal of Philosophical Research 43: 43–67.
Kidd, I.J. (2019b) “Epistemic Corruption and Education,” Episteme 16 (2): 220–235.
Macintyre, A. (2013) After Virtue: A Study in Moral Theory, 3rd ed. (London: Bloomsbury).
Medina, J. (2012) The Epistemology of Resistance: Gender and Racial Oppression,
Miller, C. (2017) The Character Gap: How Good Are We? (Oxford: Oxford University
Press).
Mills, C. (2007) “White Ignorance,” in S. Sullivan and N. Tuana, (eds.), Race and
Epistemologies of Ignorance (New York: SUNY Press), 11–38.
Moriarty, M. (2011) Disguised Vices: Theories of Virtue in Early Modern French Thought
Pohlhaus, Jr., G. (2014) “Discerning the Primary Epistemic Harm in Cases of Testimonial
Injustice,” Social Epistemology 28 (2): 99–114.
Proctor, R.N. (2011) Golden Holocaust: Origins of the Cigarette Catastrophe and the
Case for Abolition (Berkeley: University of California Press).
Roberts, R.C. and W.J. Wood. (2007) Intellectual Virtues: An Essay in Regulative
Epistemology (Oxford: Oxford University Press).
Rochefoucauld, F, de La. (2008) Collected Maxims and Other Reflections [1664], trans
E.H. Blackmore, A.M. Blackmore, and F. Giguère (Oxford: Oxford University Press).
Rooney, P. (2010) “Philosophy, Adversarial Argumentation, and Embattled Reason,”
Informal Logic 30 (3): 203–234.
Rothstein, B. and A. Varraich. (2017) Making Sense of Corruption (Cambridge: Cambridge
University Press).
Shklar, J. (1984) Ordinary Vices (Cambridge, MA: Harvard University Press).
Tanesini, A. (2016) “Calm Down, Dear: Intellectual Arrogance, Silencing and Ignorance,”
Aristotelian Society Supplementary Volume 90 (1): 71–92.
Tanesini, A. (2018) “Intellectual Servility and Timidity,” Journal of Philosophical
Research 43: 21–41.
Taylor, G. (2006) Deadly Vices (Oxford: Clarendon).
Tessman, L. (2005) Burdened Virtues: Virtue Ethics for Liberatory Struggles (New York:
Zagzebski, L. (1996) Virtues of the Mind: An Inquiry into the Nature of Virtue and the
Ethical Foundations of Knowledge (Cambridge: Cambridge University Press).
Zagzebski, L. (2017) Exemplarist Moral Theory (Oxford: Oxford University Press).
Part II
Collectives, institutions, and

networks
5 Institutional epistemic vices
The case of inferential inertia
Miranda Fricker
One of the achievements of virtue epistemology has been the identification of an

array of epistemic virtues that had not previously been distinguished or focused
on in traditional epistemology.1 Similarly, it is an ongoing achievement of what
Quassim Cassam has called ‘vice epistemology’ to explore the underbelly of the
same domain—the epistemic vices that either mirror or in less direct ways reflect
the constellation of virtues. There is also room for a certain hybridity across the
two main domains in which the notions of virtue and vice find application—the
epistemic and the ethical. In earlier work I tried to bring attention to two virtues
that are hybrid in that they are both epistemic and ethical in kind, since they aim
equally at the ultimate ends of truth and justice. These were ‘testimonial justice’
and ‘hermeneutical justice’ (Fricker 2007), both virtues of epistemic justice, and I
subsequently tried to show how these virtues might manifest themselves not only
as individual virtues but alternatively as institutional virtues, gatekeeping certain
kinds of political power (Fricker 2010 & 2013).2 José Medina has explored further
virtues and vices in the domain of epistemic injustice—virtues such as humil-
ity, curiosity/diligence, and open-mindedness; vices such as epistemic laziness,
arrogance, and closed-mindedness (Medina 2013)—and, I take it, all such virtues
and vices will be hybrids inasmuch as the wrongs that they pre-empt or that flow
from them respectively are at once ethical and epistemic in kind. Just as there can
be hybrid ethical-epistemic virtues, then, so there can be hybrid ethical-epistemic
vices.
Why might we care about the question of whether institutions can be said to
have vices of any kind? Why not content ourselves with using a vocabulary of,
say, efficiency and inefficiency, functionality and dysfunctionality, relative to the
institution’s goals or purposes? Surely we are able to critically assess these things,
and the relative merits of the institutional goals and purposes too, without hav-
ing to talk specifically in terms of virtue and vice? These are terms which after
all strike many as having alienating overtones of high church or, alternatively,
high classicism, depending on whether they ring moralistic or simply archaic.
This is a fair question, but on the other hand, let’s not forget that (to fleetingly
sloganize) meaning is use—or at any rate these concepts and their overtones are
not static, unless and until we stop using them. There might be good reasons to
rehabilitate and normalize the notions of virtue and vice, even for institutions, so
90 Miranda Fricker
that unwanted overtones are silenced and ideas of virtues and vices come, more
completely, to seem like a proper part of our contemporary normative equipment
for ethical evaluation. Indeed this is one way of picturing what much of our recent
theorizing about virtue and vice is gradually working to achieve.
More specifically, an important reason to engage in the ongoing modernization
of these concepts is that something approximating the idea of institutional vice
has for some time had a tentative foothold in public discourse, but one we perhaps
do not yet conceive very clearly either in the public domain or in philosophy. The
idea of institutional vice, or virtue for that matter, has even now received rela-
tively little philosophical attention compared with individual virtue and vice, so
perhaps there is some useful work to be done on this score.3 A central role for the
idea of institutional vice came sharply to the fore in British culture in 1999 with
the publication of the Macpherson Report on the handling of the racially moti-
vated murder six years earlier of a teenager, Stephen Lawrence, in South East
London (Macpherson 1999). That report described the London Metropolitan
Police as ‘institutionally racist’, and this marked a watershed moment of public
acknowledgement of the deep permeation of racism in a central and powerful
institution such as the capital’s police force.4 If anything is a vice in an institution
then racism is, and I therefore consider the idea of a police force being found to
be institutionally racist as furnishing us with a central and prominent example
of something that can be properly theorized as institutional vice. While I would
not pretend that the idiom of institutional vice is the only one in which we could
make good sense of the various phenomena of institutional racism (we could,
for instance, restrict ourselves to talk of institutional prejudice, inequality, bias,
dysfunctionality, failure of protocol …) still I would argue that there is a robust
purpose for which the vocabulary of institutional virtue and vice is distinctively
well-placed to serve. That purpose is basically one of picking out aspects of insti-
tutions that are the collective analogue of an individual agent’s character, but
where the actual individuals whose combined epistemic agency comprises the
institution’s epistemic agency need not, as individuals, have any of the traits or
attitudes of the institution. While some institutional vices will depend upon some
significant number of the individual officers having the vice themselves, at least
when in role as an officer of that institution, other vices will be more structural
in kind, and the notion of institutional vice is well-designed to be applied to both
sorts of case. Or so I aim to show.
5.1 Ethos matters

What is distinctive about the idea of an analogue to an individual’s character?
Why put oneself to all the philosophical trouble of substantiating the idea of
institutional character? Even if it is philosophically do-able, do we really need
it? The answer lies in how far we value the possibility that (at least some of) our
institutions have an ethos from which their procedures and judgements flow.5
Only through sustaining an ethos that guides and explains their conduct might
an institution—the NHS, parliament, the police force, the BBC, the care system,
Institutional epistemic vices 91
the judiciary—genuinely stand for something, and constitute part of the fabric
of what we believe or hope is good about the culture of which they are a part.
Most saliently perhaps, in a democracy it may be important to us as citizens that
the judicial system operate not only in a way that is well-designed to deliver
right results—fair sentencing in the criminal courts, for instance—but moreover
that the institutional mechanisms and procedures that furnish these right results
are fuelled by appropriate values. Were the right results (the fair sentencing)
produced by a miracle of clever incentivization and efficiency mechanisms, this
would not be enough. While there is a place for incentivization and efficiency
mechanisms in any institution—performance reviews, prospects of promotion,
disbarring and dismissal for anyone found to be corrupt, and so on—the point
remains that no institution can produce justice proper except by way of reasons
of justice (Fricker 2013). In the institutional setting this can only be a matter of
the value commitments being embodied in the processes, including the epistemic
processes involved in fact-finding and adjudicating, by way of an appropriate
ethos. Ethos, at least as I am using the notion, looks to be the only way that insti-
tutional bodies can incorporate intrinsic values in their agency as institutions.
It is comprised of collective motivational dispositions and evaluative attitudes
within the institutional body, of which the various good or bad ends orientate the
institution’s activities.
A different aspect of the explanation why we regard an appropriate ethos, and
not merely appropriate outcomes, as important for at least some of our institutions
is that a chief way of convincing people that a given outcome is appropriate—fair,
or just, or a correct application of the rules—is to show them the appropriate value
commitments that were in fact brought to bear in the process that produced it. A
good deal of our confidence in institutional judgements and actions flows from
our degree of confidence in its value commitments—epistemic values such as
actually caring about the truth and the gathering of proper evidence, as opposed to
just securing a conviction, for instance. That is why transparency in many of these
processes is a good thing—we get to see the ethos of the institution laid more or
less bare in the record (video of interviews, Hansard, minutes of meetings, etc.).
If, for example, a local government body entrusted with certain town planning
decisions in an area of redevelopment fails to consult long-term local residents,
or consults only private residents and ignores council housing residents, then we
might well describe that institutional body as ‘high-handed’, ‘arrogant’, perhaps
‘cynical’, not only ethically but intellectually—from the point of view of data
gathering and achieving a proper perspective for judgement. In itself the outcome
judgement (a new supermarket and fitness centre in place of the beloved but little-
used Victorian municipal swimming pool) might be found by all to be the right
result all things considered, yet still if residents are signally unimpressed with the
values implicit in the way the authorities came to the planning decision, then this
will tend to cast doubt on the quality of the outcome judgement itself. Institutional
ethos matters, then, partly because the presence or absence of appropriate value
priorities behind any given item or process of institutional epistemic conduct is a
factor in determining confidence and satisfaction levels in the outcome judgement
92 Miranda Fricker
itself. Just as value-dispositions matter in our evaluations of individual agents,
then, in many of our evaluations of institutional agency, ethos matters.
Insofar as this is a key rationale for modernizing the ideas of virtue and vice
with a view to rendering them fully applicable to contemporary institutions, it
is worth noticing that it entails a commitment to a broadly (though not exclu-
sively) motivational or disposition-based conception of virtue/vice. Were one
to adopt instead a purely skill-based conception that makes no mention of the
agent’s motivational states and dispositions, then there would be no distinctive
purpose for virtue and vice talk in relation to institutions. We might just as
well stick with the familiar, thinly performance-oriented terms of assessment,
evaluating institutional epistemic agency in relation to informational accuracy,
size of data sets, evidential thoroughness, breadth of fact-finding, soundness of
predictions, and other aspects of epistemic performance construed in a way that
does not draw on ethos. So the general question, ‘What is distinctive about the
idiom of virtue and vice when it comes to evaluating institutional conduct?’,
has furnished an answer that makes a purely skill-based conception otiose. I
shall therefore pursue my line of thought about institutional vice on the basis
that epistemic virtues involve good epistemic dispositions and attitudes as well
as reliability in achieving good epistemic ends. This means I will be using a
broadly responsibilist conception as opposed to a reliabilist conception (Axtell
2000; Battaly 2010).6
In saying we care, and should care, about an institution’s ethos, I have so far
put the point only positively: we want justice proper, and so we want the judiciary
to produce just outcomes from an appropriately stable commitment to the value of
justice—an ethos of justice. However, as soon as one puts the point positively in
this way, the negative counterpart quickly comes to mind: the idea of institutional
ethos is equally important because we need to be able to think about our institu-
tions critically, in terms of their faults, whether stable and systematic, or fleeting
and one-off. Like individual agents, institutional bodies can obviously have fleet-
ing lapses of judgement that might even be described as ‘out of character’. That is
why it is always such an important question to ask of an apparently one-off lapse
whether it is indeed fleeting and out of character or whether in fact, beneath the
surface, there has been a deterioration of ethos more widely in a branch of the
organization. This question was in the air, for example, when on the 9th February
2018 it was revealed by The Times that ‘Top Oxfam staff paid Haiti survivors for
sex’, after which a crucial question was whether similar abuses had been commit-
ted by other Oxfam aid workers, and whether Oxfam had in any way covered up
allegations or important details of the case. Assessing how far a given lapse is a
one-off event or an expression of a more systemic decline in institutional ethos is
an important question, partly because it determines whether or not a general loss
of faith in the organization is a warranted response, and because of the implica-
tions for what it will take to fix the organization.
We need, then, to be able to make reference to institutional ethos when engaging
in evaluative judgements about institutions’ practical and epistemic functioning.
And this means we have an interest in understanding what social-metaphysical
commitments are entailed by such talk. What, then, is the structure of an institu-
tional ethos?
5.2 Modelling ethos

I am proposing the idea of an ethos as the institutional analogue of an individual’s
character in the virtue-theoretic sense—a set of interrelated dispositions and atti-
tudes, where (in the case of a virtuous person) these are conceived as temporally
and counter-factually stable motives towards good ultimate and mediate ends.
This will be the same whether we are considering ethical or intellectual character,
though the good end(s) are fewer in the case of the intellectual. (Value monists
would say truth is the only ultimate end of epistemic virtue, so that different epis-
temic virtues are exclusively individuated by reference to their differing mediate
ends; pluralists, by contrast, allow that an intellectual virtue might aim variously
at truth or knowledge or understanding. Zagzebski (1996) helpfully glosses these
by talking in terms of ‘cognitive contact with reality’.) In modelling the phe-
nomenon of institutional ethos, there are two broad theoretical approaches avail-
able to us that are applied to various sorts of group phenomena, such as group
intentionality, group belief, and so on: summative and collective. In general, it
should not be thought of as a competition, for both models represent perfectly
real possibilities, and indeed each kind of group is frequently realized in institu-
tional life. For example, if an exam board were to move by way of majority vote
to produce the judgement that a given candidate has satisfied the requirements
for conferral of degree, the group and its decision is structured summatively. If
a similar exam board in a neighbouring department moves by way of consensus
towards what Margaret Gilbert calls a ‘joint commitment’ to the judgement that
a given candidate has satisfied the requirements for conferral of degree, then the
group is structured as a collective, and in the committed sense she calls a ‘plural
subject’. In this latter case the group judges ‘as a body’, to use Gilbert’s evocative
phrase (Gilbert 2000). This will involve the individual board members expressing
willingness, under conditions of common knowledge (each knows that the others
know …), to endorse or go along with the judgement.7
By contrast, the summative model of institutional ethos would cast institutions
or institutional bodies within broader institutions (faculties within a university,
or squads within a police force) as capable of having an ethos only as a matter
of the aggregate of individual officers’ value commitments. These may or may
not be adequately consistent to add up to a coherent ethos, and so for any such
aggregation of individual values it will remain an open question whether they
amass into a coherent ethos or not. But it is worth noting that a further possibility
for institutional ethos is a mixed economy of collective and summative structures.
If, for instance, members of the executive branch of an institution were to jointly
commit to a certain set of values in their deliberations and judgements, while
the implementation branches of the same organization were charged with sim-
ply implementing the policies (without any joint commitment, let us imagine, to
the values from which they flowed), then there could easily be some significant
94 Miranda Fricker
mismatch between the jointly committed values of the executive and the values
of the officers implementing them on the ground. They may or may not like it, but
they do it regardless as part of their job because that’s the new policy they have
been told to implement. Notably, in a case where the executive body is trying to
bring about institutional change, we must positively expect such mismatch. This
mixed economy picture presents us with a fairly typical top-down form of institu-
tional change (or failure to change, depending on how it turns out), and one of the
reasons why top-down change can ultimately fail is precisely because successful
institutional change so often depends on the new ethos being genuinely taken up
at the implementation level of the organization—the officers in the field, whether
policing, or healthcare, or development work. While there is room for a mismatch
between newly committed values in the executive and their mere implementation
at the ground-level, then, in many situations (recall the case of Oxfam) if those
values are not in fact stably held by the officers in the field, whether by dint of
joint commitment or personal commitment, then there may be trouble ahead.
Anyone looking to make good sense of institutional ethos will need to make
room for non-summative possibilities, including the mixed economy model above.
I have started with Gilbert’s view, and will argue that it offers the appropriate col-
lectivist model for present purposes, but we should first acknowledge that there
is a range of differing possibilities of a genuinely collectivist kind. The various
options can all be helpfully conceived in relation to the degree of commitment to
the shared intention/value/activity, as it may be. Michael Bratman’s view of col-
lective intentionality requires no commitment as such, in the sense that the interde-
pendent intentions of members of the group who each intend to play their part in
a planned group activity, such as painting a house together, are strictly pro tem,
so that someone who breaks away is not burdened by any residual commitment
they are failing to honour (Bratman 1999, 2014). Christian List and Philip Pettit
(2011) do not employ any stronger idea of commitment to group plan than we find
in Bratman; indeed, they are explicit in using more or less his conception. Raimo
Tuomela’s conception of ‘we-thinking’ does involve a certain commitment to
the group endeavour, but he employs the notion of ‘ethos’ to capture the flavour
of that commitment, and it does not seem to entail that a break-away member of
a we-intending group would thereby be the proper object of rebuke (Tuomela
2013). If this broad contrast between Gilbert’s view and the others in the literature
is correct, then Gilbert’s account emerges as distinctive in embedding a strong
notion of commitment in the very mechanism of group agency—the formation of
the plural subject. The joint commitment involved in the making of any collective
judgement—for instance, the examining board’s judgement that the examinee has
satisfied the requirements for conferral of degree—survives even the most com-
plete lapse of intention, participation, or interest on the part of a given individual
member of the group. For Gilbert, if a member reneges on a joint commitment,
they are therein a proper object of rebuke or a demand for an explanation.
One can become fully party to a joint commitment of this kind even if one per-
sonally disagrees with the content of the commitment, for one can go along with a
given judgement that the candidate has satisfied the requirements for conferral of
degree by merely ‘letting it stand’, even if, as a private individual, you would not
make that judgement. Perhaps you take a dim view of one of their exam results,
considering conflicting examiners’ marks to have been resolved in the wrong
direction, but because your colleagues on the board take a different view, you
have decided to acquiesce and allow a consensus. Going along with a judgement
by ‘letting it stand’ in this way is sufficient for being fully party to the jointly com-
mitted judgement. And it is perhaps worth noting that this feature of the view is
highly desirable, because it is what enables boards and committees and other insti-
tutional bodies to achieve unity in group judgements even while there may remain
candid disagreement at the individual level. This is how Gilbert’s model works to
allow potentially radical differences between a judgement of the collective body
and the judgements of individuals that compose it (Gilbert 1987, 2002). Still, once
an individual group member is party to such a commitment, they are on the hook.
This binding aspect of Gilbert’s plural subjectivity naturally renders it some-
what susceptible to objection when the model is applied to the breezier forms of
group activity that seem more easy-come-easy-go. If people dance together for
a while in a salsa club, and then one of them has had enough and wanders off to
get a drink, isn’t that perfectly fine? No need for explanation, surely, let alone
rebuke—that would be weird. That kind of dancing together is non-committal,
but it is still dancing together. It lasts as long as it lasts, and that is part of the free-
dom of it. Bratman and Tuomela’s models can easily accommodate this kind of
case, though obviously not Gilbert’s which is too demanding on the commitment
front. But here I believe the correct conclusion to draw is that a certain pluralism
is in order to accommodate the full range of ways we can do things together. It
is precisely the bindingness of Gilbert’s model that makes it the right one for
present purposes, for in order to make sense of the bindingness of ethos we need
the commitment involved in the joint commitment that creates the plural subject
of the ethos. An ethos is precisely not something pro tem, but something commit-
ted and intersubjectively binding by way of potential apt rebuke. A set of values
that one can ditch when it no longer suits is no ethos at all, but mere lip service.
The ethos of a group or institutional body is something that binds its members
because it consists of value commitments in the robust sense of commitments that
are temporally and counter-factually stable, or at least meant to be temporally and
counter-factually stable. If a member proves their commitment less than appropri-
ately stable by reneging on it, then they are properly subject to rebuke. I conclude
that the joint commitment model is the distinctively appropriate model to employ
in elaborating the collective value-dispositions involved in forming a given insti-
tutional ethos, and therefore the appropriate model in particular for the idea of an
institutional epistemic ethos.
To use the Gilbertian apparatus for these purposes is basically to engage in a
piece of analogical thinking. Indeed, all talk of plural subjectivity is based on the
elaboration of an analogy between the individual level and the group level, so
that we may earn the right to speak literally of groups doing things like making
a careful judgement that a given candidate has satisfied all the requirements for
conferral of degree. Gilbert explains the very idea of a plural subject by way of
96 Miranda Fricker
an analogy between the relation of an individual’s will and the action it produces
with the relation of a group’s pooled wills and the relation with the group action
produced:
these wills will be directed at that end, as if they belonged to a single person.
That is, the coherence of the behaviour which is their output will approximate
in coherence to the output of the will of a single person acting in pursuit of a
goal of his own.
(Gilbert 1989: 211)
For our distinctly epistemic purposes at the level of institutions, my suggestion is

that we invoke a similar analogy. We say that when an institutional body, like an
examinations board, comes to a careful judgement that the candidate has satisfied
the requirements, the board members’ individual wills (in the aspect relevant to
epistemic agency) are directed at the end of coming to a careful judgement as if
they belonged to a single adjudicator. On this basis, we may speak simply and
literally of the exam board making the careful judgement in question.
As we have seen, this model requires individual members to express willing-
ness, under conditions of common knowledge, to at least go along with the group
judgement made. Transposing this now to our proposed case of institutional epis-
temic ethos—an ethos, for example, of truthfulness and fact-checking in public
office—we might say of an institutional body such as a branch of government:
the individual officers all endorse, or at least go along with, the set of values that
comprise an ethos of truthfulness and fact-checking in public office. Thus, parties
to this jointly committed ethos would in theory rebuke anyone who was found to
have lied or bullshitted in public office. It is worth emphasizing, as Gilbert does,
that the expression of willingness can be very minimal. Indeed, in some contexts
one can surely count as expressing willingness by default, by simply failing to
object. Imagine, for example, a case in which there is general suspicion of a fig-
ure in political office that he is using his influence for personal political gain in
an upcoming election—and yet no one is yet talking about it openly, and there
remains a kind of group pretence that everything is as it should be, or at least
within the bandwidth of normal political dealings. Those officers of the political
institution in question are in the unfortunate position that their silence is a way of
going along with the fiction that everything is okay. But it isn’t, and in their con-
tinued silence they passively become party to a joint commitment whose content
is incompatible with the good institutional epistemic ethos proper to a democratic
government. It will take a whistle-blower to break the silence, and that is a seri-
ously costly thing to be for all sorts of reasons, but one feature of the pressures on
a potential whistle-blower is that they are currently party to a joint commitment
of silence. Whistle-blowers are admired by many, but they are also rebuked and
often abused by many who wish to discredit them. Much of that is more politics
and sheer threat in the context of power-mongering, but one aspect of it is proper
to the basic normativity of joint commitment. It is built into Gilbert’s model that
if you do at least go along with the silence for a while, you thereby become party
to the joint commitment to conserve the status quo, so that if you then shift your
stance and blow the whistle you are pro tanto a proper object of rebuke. In a real
political case that may well be the least of an actual whistle-blower’s worries, but
it is a real normative feature of such situations nonetheless, and one that can be
exploited by those with an interest in conserving the silence.
As Gilbert herself has emphasized in relation to group belief, this brings out
the insidious potential of joint commitment: once made, it brings genuine norma-
tive pressure to bear, and there is no guarantee that this is to a good end of any
kind. As in the example just given, it might work in the interests of a corrupt poli-
tician by creating genuine normative pressure not to break ranks. We can see bet-
ter just how easily—how passively—this kind of situation can be manufactured
if we reflect, further, on the phenomenon of ‘accommodation’ as Rae Langton
has recently explored it. She takes up David Lewis’s notion of accommodation
as a feature of scorekeeping in a language game (Lewis 1979). Lewis coins the
idea to capture how a presupposition can enter circulation in a conversation and
be accommodated if it is not actively challenged, and Langton elaborates how
powerful this mechanism can be as a means of introducing prejudiced, stereo-
typical, or hateful ideas into conversational circulation. She calls assertions that
enter in by way of presupposition ‘backdoor testimony’—assertions, indeed tell-
ings of an inexplicit kind, that surreptitiously become accommodated and thereby
perpetuated without challenge (Langton 2018). Similarly, I would suggest that
the phenomenon of accommodation is useful for seeing one way that a default
‘expression of willingness’ to go along with the content of a given joint commit-
ment can be a powerful force. An official can passively become party to a joint
commitment to keep quiet about a political leader’s corrupt lies just by failing to
dare to be a whistle-blower. One thereby accommodates the presupposition that
the leader had engaged merely in acceptable levels of political hyperbole or bom-
bast, even if one knows perfectly well they were lying. In such a case, a person
with decent epistemic values becomes party to a committed toleration of corrupt
mendacity, thereby passively betraying whatever may be left of the decent epis-
temic ethos of truthfulness in public office.
Becoming party to a joint commitment, then, is easy—frighteningly so in some
contexts—and this can be manipulated by those who wish to keep certain atti-
tudes in circulation. The net result at group level is that the institutional body
in question behaves in a way that departs from a pre-existing ethos of truthful-
ness and fact-checking. In this respect Gilbert’s model emerges once again as
an excellent fit for modelling institutional ethos; only now we are concentrating
on its credentials for modelling bad or vicious ethos. Every time officers in the
government of our imagined truthless political leader go along with something
he has said or presupposed, passively letting it stand and thereby accommodating
it, they raise the level of pressure—genuine normative pressure—to conserve the
status quo. The mechanism of joint commitment helps us understand a norma-
tive aspect of conspiracies of silence, but more importantly for present purposes
it lays bare the mechanism of how institutional bodies can behave in ways that
depart from, and help deteriorate, a pre-existing epistemic ethos of truthfulness
98 Miranda Fricker
and fact-checking. The insidious joint commitment to letting stand and thereby
increasingly accommodating truthless content threatens to end in institutional
epistemic failure—shifts of enduring commitment that risk entailing substantial
erosion of good epistemic ethos in the medium to long term, bringing about a cor-
ruption of institutional character.
5.3 Modelling institutional epistemic vice

Epistemic vices, like vices in general, can pertain to acts or behaviours, and they
can pertain to agents. When a given agent displays a stable pattern of vicious
actions and behaviours, then we attribute the vice not only to that which is done
but also to the doer themselves. I committed myself earlier to a broadly respon-
sibilist conception of virtuous action whose distinctive feature is its motivational
richness—an inner or motivational element. No other model would be able to cap-
ture the importance we implicitly attach to the ethos at the heart of certain of our
institutions. But it also includes a reliabilist aspect—an outer, or performative ele-
ment. On the responsibilist conception, an epistemically virtuous agent is some-
one who acts from a temporally and counter-factually stable good motive and
where the good end of that motive is reliably achieved (Battaly 2010; Zagzebski
1996). Thus, we are presented with both an inner and an outer element of virtue,
and therefore two distinct areas for potential failure and lapse into vice.
But not all failures will indicate vice. There must surely be some more neutral
ground, not least because virtue itself comes in degrees, as an agent is increas-
ingly habituated and spontaneous in her responses. If an agent—individual or
institutional—acts in a manner that is less than ideal, but not culpably bad, then it
would not be natural to use the word ‘vice’ to describe either them or their action.
Imagine, for instance, an institution that has a good ethos, and so the inner ele-
ment is fine, yet is inefficient in its performative aspect, so the outer aspect is less
than it might be. A school, perhaps, with teachers who care about doing a good
job, but an IT system for homework submission and marking that is not well-
managed or well-used, and makes for occasionally serious communicative fail-
ures both with students and with parents. This is not an institution we would hold
up as an exemplar of epistemic virtue as regards its information-sharing practices;
indeed we would criticize it, but we would not go so far as to describe it as guilty
of any epistemic vice.8
This said, we can return to the previous point that the responsibilist conception
makes for two distinct areas of potential culpable failure—the inner ethos (stable
motives), and the outer performance (achievement of the ends of those motives).
If vices are culpable failures of virtue, then epistemic vices are culpable failures of
epistemic virtue either in respect of ethos and/or in respect of (what we might call)
implementation. Imagine our school does seriously mess things up one year, so
that teachers have practically given up using the online homework system but no
proper alternative has been put in place. And imagine the mess up is bad enough—
disruptive enough to cause a real loss of confidence on the part of the students and
parents—so that the school is culpable. Now we are looking at a behaviour on the
part of the school that would count as vicious as regards its practices of informa-
tion-sharing. But this ‘thinking vice’9 might yet be out of character for the school,
so it does not yet imply that the school itself—the institution—has the vice. That
attribution would require a temporally and counter-factually extended pattern of
such culpable lapses of virtue as regards information-sharing practices. So let us
now imagine our school ten years on, after a decade of becoming increasingly
inefficient and disorganized, even while it had opportunities to do better. The
teachers have become disenchanted and fed up so that despondency and laziness
has infected the whole-school ethos, and/or, despite continued underlying value
commitments, the school has simply fallen into repeated performative failures in
the implementation of its policies on information-sharing. One way or the other
this would now be a school that displayed an epistemic vice of bad information-
sharing. The various culpable failures of virtue have congealed into a systematic
failure, so that the very character of the institution has been changed for the worse.
Institutional epistemic vice is a matter of culpable epistemic bad habits, where the
culpable lapses might be in ethos or in implementation, or in both.
Charlie Crerar has critiqued what he calls the ‘mirror view’ of epistemic vices,
one form of which depicts vices as always positively aiming at an epistemically
bad end, mirroring the way virtues always aim at a good one (Crerar 2018). The
critique is persuasive—in fact it is not easy to dream up even a single psychologi-
cally coherent epistemic vice of that kind because of the fundamental investment
in truth/knowledge that all epistemic subjects as such have. Jason Baehr imagines
a far-fetched case of epistemic malevolence that would fit the bill (Baehr 2010),
and perhaps another promising prospect in this regard might be persistent kinds of
self-deception—imagine someone with a long-term investment in lying to them-
selves about how talented they are, for instance. Still, such cases will surely be
unusual at best, so that on the whole any motivational disorder constituting an
epistemic vice will instead take the negative form of an inadequate commitment
to good epistemic ends. These ends might be the ultimate end of cognitive contact
with reality—as in the case where the politician shows a flagrant disregard for the
truth—or, alternatively, any of the mediate ends whose epistemic value consists
in their functioning as a means to achieving that ultimate end. Such mediate ends
might be, for instance, looking carefully at the evidence, fact-checking, being
open to counter-arguments, realizing when one’s evidence base is too narrow, and
so on. It may not be possible to be epistemically virtuous without an appropriate
commitment to the ultimate end of cognitive contact with reality, but it does not
follow that only a lapse in relation to that ultimate end can indicate epistemic vice.
On the contrary, I contend that someone might be epistemically vicious precisely
because of persistent lapses in relation to a mediate end—such as fact-checking—
even if their ultimate epistemic commitment remained intact. This possibility is
entirely compatible with the background theoretical idea that only the value of the
ultimate end confers value on the mediate end—so, for instance, the only reason
fact-checking matters epistemically is because fact-checking promotes cognitive
contact with reality. Indeed, my contention positively relies on the instrumental
connection between mediate and ultimate ends, for what makes it epistemically
100 Miranda Fricker
bad, and potentially culpable, to fail to fact-check is precisely that failing to fact-
check is bad from the point of view of achieving cognitive contact with reality. It
is because of this instrumental connection between fact-checking and cognitive
contact with reality that a lapse in the former can, if culpable, constitute an epis-
temic vice. As regards this kind of epistemic vice, the distinguishing feature is
that the subject is to blame for how at least one of their mediate motives is failing
to align with the ultimate end of cognitive contact with reality.
Crerar imagines two figures, Galileo and Dave, each of whom seems to present
a clear case of epistemic vice and yet each of whom is equally clearly committed
to epistemically good ultimate ends.10 Galileo is individually brilliant and cares
about the truth, but he is also epistemically arrogant in his scornful neglect of the
views of his colleagues; Dave displays a lamentable narrow-mindedness of the
privileged, even though he, too, cares about the truth. I agree these characters
surely display epistemic vices, and I agree that they would represent a puzzle to
anyone arguing for a conception of epistemic vice that required a motivational
lapse in respect of ultimate ends. But for the reasons offered above, I would inter-
pret Galileo and Dave as exemplifying epistemic vices in virtue of the fact that
each is culpably unmotivated towards a relevant mediate end—respectively, that
of listening to one’s fellow researchers’ informed opinions, and that of awareness
of how privilege is affecting one’s social perceptions. Such culpable lapses in
relation to mediate epistemic ends is perfectly sufficient for epistemic vice.11 Like
Crerar, I do not hold to any exclusively motivational conception of vice, but this
is not because I do not see Galileo and Dave’s epistemic failings as motivational
failings. I interpret both Galileo and Dave as displaying vices by exhibiting medi-
ate motivational failings that undermine their epistemic orientation to cognitive
contact with reality, even while they both may remain psychologically motivated
to achieve it. For me, the reason to reject an exclusively motivational account of
vice is simply that motivational failure is not the only route to vice, since a culpa-
ble lapse in the outer, performative aspect of virtue remains an independent possi-
bility. Remember our informationally challenged school at the moment where its
flawed information-sharing practices have become seriously entrenched, despite
a continuing good ethos. This school is displaying an epistemic vice of sloppy
information-sharing even though there is nothing wrong with it at the motiva-
tional level of ethos; the problem and the culpability is all at the performative
level. A person or institution can display epistemic vice simply through persistent
performative failure, even if the motivational commitments, mediate and ulti-
mate, are all that they should be.
On the view I am putting forward, then, epistemic vice consists in some culpa-
ble lapse of epistemic virtue either (i) in its inner aspect of mediate and/or ultimate
motivations to good epistemic ends, and/or (ii) in its outer aspect of performance—
the achievement of those ends. A motivational and/or performative lapse that is
bad enough to warrant blame is bad enough to warrant the label ‘vice’. Where it
is persistent it will constitute a vice of epistemic character and not merely a more
fleeting vice of thinking. Putting together the earlier picture of institutional ethos
with this conception of epistemic vice, we can say that institutional epistemic
vices are displayed—either in thinking or, where persistent, also at the level of
institutional character—whenever there are culpable lapses in the institution’s
epistemic ethos and/or in the implementation of its ends.
5.4 The institutional vice of inferential inertia

A salient rationale for a philosophical vindication of the idea of institutional epis-
temic vice is that there may be some epistemic vices that are especially worth
identifying in their institutional form, either because they are especially perni-
cious in that form, or because they are especially fixable, or both. I want to draw
attention to an epistemic vice I will call the vice of inferential inertia. I think we
easily recognize a certain scenario in an individual hearer, who is not guilty of
perpetrating any testimonial injustice exactly, for their credibility judgement of
the speaker is not depressed by prejudice of any kind, and indeed (let us imag-
ine) they do believe her; and yet … nothing else happens by way of epistemic
follow-through. Imagine a case of someone telling a colleague or co-worker of a
crime committed in the workplace. Imagine the colleague believes what she tells
them—they assent, they express genuine sympathy or shock, or whatever is in
order—and yet … somehow the evidential bearing of what they have been told
does not impact anywhere (else) in their belief system. Perhaps they are resistant
to the implications of this particular piece of news, and hope it is a one-off, or
perhaps they are epistemically lazy or scared or unimaginative when it comes to
shaking their sense of the status quo. Such a person, let us imagine, fails to draw
any inferences at all, does not alter her other beliefs one iota, even though they
are likely seriously undermined by what she has been told and now believes. For
instance, their other beliefs about the perpetrator remain unaltered, or at least are
certainly not altered in a manner appropriate to the evidence.12 They somehow
hold the contradictory beliefs in suspension without making the cognitive effort
to draw the inferences that are there to be drawn, even tentatively. This person
believes what they are told, but the new information never gets to have its evi-
dential impact. They are guilty of a distinctive epistemic vice, that of inferential
inertia. In such a case, though the speaker is not misjudged epistemically, still she
is just as frustrated in her aim to bring the hearer to appreciate the implications of
what she’s saying as she would have been in an ordinary case of testimonial injus-
tice. From the point of view of inferential uptake, she might just as well have not
been believed. We might slot this phenomenon of inferential inertia into relation
with testimonial injustice by saying that insofar as any case of inferential inertia
is the product of prejudice, then it is a close relation of testimonial injustice, and
instantiates a hybrid ethical-epistemic vice, or, alternatively, insofar as a given
case is the result of some other kind of epistemically culpable error, it displays a
plain epistemic vice—in this case the kind of stupidity inherent in failing to grasp
clear implications of new information.13
I think we can easily identify various institutional forms of exactly this epis-
temic vice. How many feedback forms does one fill out, whether online or by
hand, after a doctor’s appointment, a retail experience, a meal at a restaurant, an
102 Miranda Fricker
online purchase, or even a trip to the dentist, where the much-vaunted feedback,
accompanied by apparently sincere declarations of just how much they really
want to know how they can do better, in fact passes into an institutional void. Too
often the fact of having such a mechanism notionally in operation is all that the
institution really cares about (they can tick that box), so that the evidential import
of any of the content actually fed back is entirely lost. This is the institutional
epistemic vice of inferential inertia, and it is rapidly becoming a normal part of
our institutional environment.
Looking to a gravely serious example from UK institutional life we can see
a quite different way in which it can be important to diagnose an institutional
epistemic vice: in the BBC-commissioned independent review led by Dame Janet
Smith into the BBC’s culture and practices in the years when Jimmy Savile was
committing multiple predatory sexual crimes, we see that this epistemic vice of
inferential inertia was effectively a key part of the diagnosis of what went wrong
institutionally speaking so that he was allowed, even enabled, to commit these
crimes undetected for so long.
Dame Janet’s review emphasizes certain cultural aspects of the BBC at the
time, one of which is a climate of not complaining and in particular not complain-
ing about the Talent:
5.4.1 The culture of not complaining about the talent

54. As I have said, there was a culture of not complaining about anything. The
culture of not complaining about a member of the Talent was even stronger.
Members of the Talent, such as Savile, were to a real degree, protected from
complaint. The first reason for this is because of a deference or even adula-
tion which was, and still can be, accorded to celebrity in our society. The
second reason was because of the attitude within the BBC towards the Talent.
The evidence I heard suggested that the Talent was treated with kid gloves
and rarely challenged. An example of this is the attitude of C51’s supervisor
when he was told that Savile had sexually assaulted C51 (see paragraphs
5.254–5.255 of my Report). His immediate reply was ‘Keep your mouth shut,
he is a VIP’.
(Smith 2016)
The review also emphasizes a ‘culture of separation’ and the ‘silo mentality’ that
entailed there was very little information-sharing between different parts of the
BBC. Competitiveness between departments exacerbated the situation, since it
incentivized secretiveness about anything that might prove a liability to one’s
home department. The net result was that even when suspicions were raised or
a complaint made in one place these would not go any further in the epistemic
economy of the organization:
61. This sense of separation could mean that a concern which arose in one
part of the BBC would not be transmitted to or discussed with another part.
For example, in 1973, Douglas Muggeridge does not appear to have shared
his concern about Savile with anyone in Television. I accept that, if an issue
was considered by the Board of Management, it would be known of by senior
management across the BBC. For example, when concerns arose about pos-
sible misconduct at Top of the Pops, there was some discussion at a meeting
of the Board of Management. Soon afterwards, there was discussion about
this kind of issue at the Management Director Radio’s weekly meetings. But
if an issue was not raised at such a meeting, its chances of going across the
BBC were slight.
62. At a lower level, there could be a reluctance to discuss a problem
which arose in one department with personnel in another. This seems to have
been attributable to the sense of competitiveness which prevailed between
programme making departments.
(Smith 2016)
Now if we consider the BBC as a collective epistemic subject, what we are pre-
sented with here is a characterization of an organization whose informational
states were radically unintegrated both because of an ethos failing and because
of a structural performative failing. The ethos failings consist in the climate of
not making complaints about the Talent; combined with the competitiveness
between departments. The report also goes into ‘the macho culture’ in the organi-
zation especially around issues of sexual harassment. The structural performative
failing is organizational—the fact that different departments were unintegrated
and lacked channels of communication between them that would have enabled
information-sharing:
63 Even within the same programme, there could be difficulties of commu-

nication in relation to complaints. Staff working on a programme would not
necessarily have the same line manager. Staff working on the production
team would be part of a line of management which ran through the producer,
maybe to an executive producer and from there to the head of department.
Other staff would have a different management line – for example, the floor
manager would not report to the programme producer but to his or her own
line manager in the Studio Management Department. That was because the
provision of floor management was a central service provided to a programme.
Cameramen, sound engineers and audience supervisors had similar separate
management structures. This separation seems to me to have the potential for
preventing anyone in management from seeing the bigger picture.
5.4.2 The macho culture

64. Another reason why complaints or concerns of a sexual nature might not
have been passed up the BBC as they should have been related to the ‘macho
culture’ which some witnesses said was present in some (but not all) depart-
ments of the BBC. Particular complaint was made about the behaviour and
104 Miranda Fricker
attitudes of technical staff (who were almost entirely male) and of manage-
ment in Radio 1 and Television’s Light Entertainment Department, where
there [were] very few women in senior positions. I have the impression that
sexual harassment was more common in the Light Entertainment Department
and BBC Radio 1 (the areas where Savile worked) than in many other parts of
the BBC. Women found it difficult to report sexual harassment. Generally,
the attitude of the male managers was thought to be unsympathetic and, of
course, there were very few female managers.
(Smith 2016)
A figure such as Savile could consequently operate relatively freely in his sex-
ual predation in the knowledge that suspicions raised in one department were
unlikely to spread, and therefore unlikely to be treated by anybody as eviden-
tially significant. The portrait of the BBC is as an organization which was at
that time seriously epistemically unintegrated when it came to the kind of infor-
mation that was needed to properly pick up on what Savile was doing. People
would make complaints, and even if they were believed, the informational con-
tent would go nowhere, receiving little to no inferential follow-through. The
informational compartmentalization of the organization effectively ensured that
the scattered items of information would never amass into a body of evidence,
and be treated as such, but would remain inferentially inert epistemic particles
dispersed in the organization. The upshot is a portrait of an organization that
had serious ethos problems of shielding the Talent, inter-departmental competi-
tiveness and protectionism, and a ‘macho culture’ especially as regards sexual
harassment. In addition to these culpable defects in ethos, there was the signifi-
cant structural performative failure relating to extreme compartmentalization
and consequent inadequate information-sharing. All of this adds up, epistemi-
cally speaking, to a paradigm example of the institutional vice of inferential
inertia. The only upside is that its diagnosis instructively lays bare exactly the
kinds of innovations required to improve the situation and to help ensure against
recurrences.
I have proposed a conception of epistemic vice such that any culpable lapse
in motivational and/or performative elements of epistemic virtue is sufficient
for it. And I have applied Gilbert’s joint commitment conception of collective
agency in order to elaborate what is involved in having an institutional epis-
temic ethos. An institutional body whose actions systematically betray a jointly
committed good epistemic ethos, and/or whose performance systematically fails
to achieve the good ends of that ethos, is an institution that displays an epistemic
vice. Lastly, I have offered a sketch of one institutional epistemic vice in par-
ticular, which I have called the vice of inferential inertia. Though it can occur
in individuals, its institutional form seems particularly relevant to social life
these days. It can occur in both trivial and deadly serious forms, and it has been
publicly revealed to have found disastrous expression in a valued institution. I
hope to have shown it is an institutional epistemic vice worth distinguishing and
understanding.14
Notes
1 See Baehr (2011), Battaly (2010 & 2015), Cassam (2019), Medina (2013). The term
‘vice epistemology’ was, I believe, coined by Cassam in his 2016 paper of that name.
2 For a defence of the need, indeed ‘inevitability’, of such virtues see Madva (2019).
3 See Lahroodi (2007), Sandin (2007), Fricker (2010 & 2013), and Konsellmann Ziv
(2012). See also, however, the treatment of group epistemic polarization by Broncano-
Berrocal and Carter (2020), which treats the tendency for a given group to incline
towards a more extreme belief than the beliefs of any of its constituent individual mem-
bers as a collective epistemic vice. I thank Charlie Crerar for directing me to their work
on this.
4 I discuss this case in more detail in Fricker (2013) and my present purpose is to develop
more fully some initial ideas I put forward there concerning institutional vice.
5 For an account of collective agency, specifically those cast as ‘we-mode groups’, that
makes use of a notion of an ethos as part of their characteristic ‘we-thinking’, see
Tuomela (2007, 2013 and 2017).
6 The germinal opus for the motivationally rich conception of intellectual virtue, which
later attracted the label ‘responsibilist’, is Linda Zagzebski’s Virtues of the Mind (1996).
7 Gilbert (1989) is the locus classicus; more recently see, for instance, Gilbert (2013);
and for a focus on the epistemic see, for example, Gilbert (2004).
8 In this I may differ slightly from Cassam, who explicitly leaves room for the possibility
of vices that are criticisable but non-blameworthy because, for instance, the agents lack
the power to correct the intellectual defect in question, such as the cultural prejudice
that infected their judgement (Cassam 2019; ix, and ch. 6, especially p. 97). Instead
I would tend to categorize such cases as in principle blameworthy—some people in
the same context after all were able to resist the prejudicial pressures of the day—but
where the cultural context of prejudice might make excuses applicable to reduce the
appropriate level of blame.
9 ‘Closed-mindedness as a quality of a particular piece of thinking is a thinking vice, an
epistemically vicious way of thinking or ‘thinking style’. It is one thing to be closed-
minded and another to think closed-mindedly’ (Cassam 2019: 56).
10 The example of Galileo’s arrogance is from Robert Roberts and Jay Wood (2007: 254),
as quoted in Crerar (2018).
11 See also Tanesini (2019).
12 Eliana Peck has suggested in conversation that the case of Christine Blasey Ford’s testi-
mony to the Senate Judiciary Committee (September 2018) concerning the nomination
of Brett Kavanaugh to the Supreme Court is a prime example of this. Blasey Ford was
widely considered a manifestly credible witness, and it also seems was in fact believed,
and yet her testimony may as well have not been believed for all the relevant inferential
activity it provoked regarding the question in hand.
13 Inferential inertia might also be involved in what Patrick Bondy (2010) calls ‘argu-
mentative injustice’, which is presented as an adaptation of testimonial injustice, since
it involves a hearer giving a prejudicially depressed level of credibility to a speaker’s
argument. (Bondy, however, focuses exclusively on the effects of negative identity
prejudice, rather than prejudice more generally.)
14 I thank Quassim Cassam and Charlie Crerar for helpful comments on an earlier draft.
References
Axtell, Guy (2000) ed. Knowledge, Belief, and Character (Lanham, MD: Rowman and
Littlefield).
Baehr, Jason (2010) “Epistemic Malevolence,” in H. Battaly (ed.), Virtue and Vice: Moral
and Epistemic (Chichester: Wiley-Blackwell): 189–213.
106 Miranda Fricker
Baehr, Jason (2011) The Inquiring Mind: On Intellectual Virtues and Virtue Epistemology
Battaly, Heather (ed.) (2010) Virtue and Vice: Moral and Epistemic (Chichester:
Wiley-Blackwell).
Battaly, Heather (2015) “A Pluralist Theory of Virtue,” in Mark Alfano (ed.), Current
Controversies in Virtue Theory (New York: Routledge).
Bondy, Patrick (2010) “Argumentative Injustice,” Informal Logic 30 (3): 263–278.
Bratman, Michael E. (1999) Faces of Intention: Selected Essays on Intention and Agency
(Cambridge: Cambridge University Press).
Bratman, Michael E. (2014) Shared Agency: A Planning Theory of Acting Together
(Oxford: Oxford Univeristy Press).
Broncano-Berrocal, Fernando, and J. Adam Carter. (2020). The Philosophy of Group
Polarization (London: Routledge).
Cassam, Quassim (2016) “Vice Epistemology,” Monist 99 (2): 159–180.
Cassam, Quassim (2019) Vices of the Mind: From the Intellectual to the Political (Oxford:
Crerar, Charlie (2018) “Motivational Approaches to Intellectual Vice,” Australasian
Journal of Philosophy 96 (4): 753–766.
Fricker, Miranda (2007) Epistemic Injustice: Power and the Ethics of Knowing (Oxford:
Fricker, Miranda (2010) “Can There Be Institutional Virtues?” in Tamar S. Gendler
and John Hawthorne (eds.), Oxford Studies in Epistemology (Special Theme: Social
Epistemology) (Oxford: Oxford University Press), 235–252.
Fricker, Miranda (2013) ‘Epistemic Justice as a Condition of Political Freedom,” Synthese
190 (7): 1317–1332.
Gilbert, Margaret (1987) “Modelling Collective Belief,” Synthese 73 (1): 185–204.
Gilbert, Margaret (1989) On Social Facts (Princeton, NJ: Princeton University Press).
Gilbert, Margaret (2000) Sociality and Responsibility: New Essays in Plural Subject
Theory (Lanham, MD: Rowman and Littlefield).
Gilbert, Margaret (2002) “Belief and Acceptance as Features of Groups,” Protosociology
16: 35–69.
Gilbert, Margaret (2004) “Collective Epistemology,” Episteme 1 (2): 95–107.
Gilbert, Margaret (2013) Joint Commitment: How We Make the Social World (Oxford:
Konsellman Ziv, Anita (2012) “Institutional Virtue: How Consensus Matters,”
Philosophical Studies 161 (1): 87–96.
Lahroodi, Reza (2007) “Collective Epistemic Virtues,” Social Epistemology, 21: 281–97.
Langton, Rae (2018) “Blocking as Counter-Speech,” in Daniel Harris, Daniel Fogal, and
Matt Moss (eds.), New Work on Speech Acts (New York: Oxford University Press):
144–164.
Lewis, David (1979) “Scorekeeping in a Language Game,” Journal of Philosophical Logic
8: 339–359.
List, Christian and Pettit, Philip (2011) Group Agency: The Possibility, Design, and Status
of Corporate Agents (Oxford: Oxford University Press).
Macpherson, Sir William (1999) The Stephen Lawrence Inquiry: Report of an Inquiry by
Sir William Macpherson of Cluny (London: The Stationery Office) https://assets.publi
shing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/2771
11/4262.pdf
Madva, Alex (2019) “The Inevitability of Aiming for Virtue,” in Stacey Goguen and
Benjamin R. Sherman (eds.), Overcoming Epistemic Injustice (London: Rowman &
Littlefield): 85–99.
Medina, José (2013) The Epistemology of Resistance: Gender and Racial Oppression,
Roberts, Robert and Wood, Jay (2007) Intellectual Virtues: An Essay in Regulative
Epistemology (Oxford: Clarendon Press).
Sandin, Per (2007) “Collective Military Virtues,” Journal of Military Ethics 6(4): 303–314.
Smith, Dame Janet (2016) The Dame Janet Smith Review Report: An Independent Review
into the BBC’s Culture and Practices During the Jimmy Savile and Stuart Hall Years
(London: BBC).
Tanesini, Alessandra (2019) “Epistemic Vice and Motivation,” Metaphilosophy 49 (3):
350–367.
Tuomela, Raimo (2007) The Philosophy of Sociality (Oxford: Oxford University Press).
Tuomela, Raimo (2013) Social Ontology (New York: Oxford University Press).
Tuomela, Raimo (2017) “The Limits of Groups, An Author Responds,” Social Epistemology
Review and Reply Collective 6 (11): 28–33.
Zagzebski, Linda Trinkaus (1996) Virtues of the Mind: An Inquiry into the Nature of Virtue
and the Ethical Foundations of Knowledge (Cambridge: Cambridge University Press).
6 Capital vices, institutional
failures, and epistemic
neglect in a county jail
José Medina
This chapter examines the phenomenon of communicative and epistemic neglect,

that is, the phenomenon of being disregarded in communicative and epistemic
practices. The chapter elucidates this phenomenon through the case study of the
neglect of inmates in a county jail in Durham, North Carolina. In my analysis
I will highlight the vices that we can ascribe both to individuals and to institu-
tions within the criminal justice system. I will also address the kinds of activism
needed, inside and outside an institution such as a county jail, in order to produce
effective transformations. I will focus on what I call epistemic activism and the
kinds of interpersonal and institutional transformations it can help achieve.
Using the theoretical framework of vice epistemology, this chapter studies the
phenomenon of communicative and epistemic neglect, that is, the phenomenon of
being abandoned as a subject of communication and knowledge. Victims of com-
municative and epistemic neglect still have communicative and epistemic agency,
but their speech acts are ignored or receive defective uptake, and their poten-
tial contributions to knowledge and understanding are blocked or diminished.
Communicative and epistemic neglect are closely related to moral neglect, that is,
the phenomenon of being excluded from the moral community or being neglected
as a moral subject worthy of respect and dignity. Phenomena of communica-
tive and epistemic neglect interact with phenomena of moral neglect in complex
ways. To begin with, moral neglect will typically lead to, and will be typically
manifested in and through, forms of communicative and epistemic neglect: being
excluded as a moral subject will often result in not being talked to or in not having
one’s words taken seriously for the production and dissemination of knowledge
and understanding. On the other hand, it is also important to note that communica-
tive and epistemic neglect is itself a form of moral neglect, and it involves a moral
harm since it consists in not being treated with the dignity and respect of a subject
capable of communication and knowledge. Moreover, communicative and epis-
temic neglect will deepen and entrench previously existing forms of moral neglect
and can facilitate new ones. Although the communicative, epistemic, and moral
aspects of neglect are difficult to separate from one another and I will address all
of them in this chapter, I will focus on the distinctively epistemic side of being
neglected and the epistemic harms it produces, while paying attention also to the
communicative and moral elements of epistemic neglect and the resulting harms.
Capital vices and epistemic neglect 109
This chapter aims at drawing a distinction between capital and non-capital or
venial epistemic vices of subjects and institutions as they appear in cases of epis-
temic neglect. More specifically, I will examine how a capital epistemic vice of
an institution can lead to corrupted sensibilities and epistemic dysfunctions in
testimonial dynamics in which people’s expressive behavior is systematically dis-
regarded. I will argue that once an epistemic pathology is triggered by a capital
epistemic vice of an institution (such as the police or the criminal justice system),
two things need to happen to repair the pathology and to produce fair epistemic
dynamics: (1) redesigning the institutions with new protocols and procedures that
guarantee fair epistemic treatment; and (2) new training and the creation of a new
climate and culture, so that a new sensibility can flourish and epistemic virtues
can be cultivated.
6.1 Introduction
In the months leading to his death in 2016, Matthew McCain, a jail inmate at the
Durham County Detention Facility (hereafter DCDF), had reported months of
insufficient and inconsistent medical treatment for his diabetes and epilepsy while
incarcerated.
Early on January 19th 2016, McCain suffered a seizure while detainees in his
pod, repeatedly pressed emergency call buttons and yelled for help. According
to detainee witnesses, these calls were not answered by the detention officer
on duty, and McCain died without assistance. The Durham Sheriff’s Office
does not publically announce in-custody deaths, except to oversight officials
and the deceased’s registered next of kin, so previous deaths in the jail have
garnered no press coverage and little public attention. This time, however,
detainee witnesses immediately reached out through intermediaries to the
community organizers of Inside-Outside Alliance. Convinced that McCain’s
death was a direct result of medical neglect and unresponsiveness by detention
officers, detainee witnesses wanted to alert McCain’s loved ones and draw
attention to conditions in the jail. Together with McCain’s family, Inside-
Outside Alliance broke the story to local news media and published reports
on its own website, Amplify Voices Inside. In turn, inquiries from journalists
prompted the Sheriff’s Office and the Durham County Department of Public
Health to release public statements on McCain’s death, and to acknowledge
two other recent deaths in the jail.
The guards on duty during McCain’s death displayed a complete lack of concern
for his well-being, which is primarily a moral failing, but they also displayed a
systematic epistemic failing: the guards repeatedly contended that they do not
take seriously the inmates’ emergency calls because they do not believe them.
The excerpt quoted above is from a forthcoming co-authored essay that I wrote
with activist-scholar Matt Whitt entitled “Epistemic Activism and the Politics of
Credibility: Testimonial Injustice Inside/Outside a North Carolina Jail.” In this
110 José Medina
essay we explore how the epistemic failing of the guards on duty led to McCain’s
death by medical neglect, but we also address how this cannot be properly under-
stood as an isolated instance to be blamed solely on the failings of the guards
on duty. The recent deaths of two other jail inmates were acknowledged by the
authorities, and other such deaths in recent years were investigated.
Clearly we are dealing here with a pattern and a systemic problem in which we
need to attend not only to the epistemic failings of the individuals involved—it is
not enough to say that we are dealing with “a few bad apples” or a few bad deten-
tion officers—but also to the epistemic failings of the institution—the orchard
itself, so to speak, is corrupted. We are dealing with a vitiated institutional space
which, far from guaranteeing that the voices of the jail inmates are properly heard,
stacks the decks against their voices receiving proper uptake, that is, an institu-
tional space that encourages epistemic dysfunctions and in which epistemic vices
flourish. The focus of my analysis will be the epistemic vice of testimonial insen-
sitivity both as a personal and as an institutional epistemic failing.
What I call testimonial insensitivity can be understood not as a single, mon-
olithic epistemic vice but, rather, as an umbrella term that includes a complex
constellation of epistemic vices that can take many shapes—just as epistemic
injustice is a generic, umbrella term that contains sub-types and different kinds of
epistemic vices. As Quassim Cassam (2019) has explained, in cases of epistemi-
cally vicious behavior it is often the case that different kinds of vices converge
and work together, and, as he shows with great analytical depth, there is a wide
variety of epistemic vices that can be classified into three categories: vicious atti-
tudes, vicious personality traits, and vicious ways of thinking. In my analysis of
the particular shape that testimonial insensitivity took in the case of McCain’s
death by medical neglect at DCDF, the epistemic vice that will take center-stage
is incredulity1—which, interestingly, is a vicious doxastic attitude that has not
received as much attention as its opposite, the epistemic vice of excessive cre-
dulity or gullibility. But there are many other vicious attitudes and personality
traits that support the testimonial insensitivity of detention officers at DCDF: in
particular, closed-mindedness, epistemic arrogance, incuriosity, and epistemic
laziness. And there is also a vicious way of thinking at work here which we can
describe as prejudicial or stigmatizing thinking: namely, thinking according to a
carceral logic that treats detained and incarcerated subjects as intrinsically suspect
and untrustworthy, and treats inmates as cry-babies in need of discipline rather
than care and protection (Cacho 2012, Medina and Whitt forthcoming). But more
importantly, cases of death by medical neglect at DCDF such as McCain’s could
not have happened without the active complicity of the carceral institution itself,
and the epistemic dysfunction behind the unattended calls for help reveals epis-
temically vicious structural conditions: namely, structural conditions in which the
incredulity and testimonial insensitivity of detention officers could be acted on
with impunity.
I will argue that the institution itself, and not just the particular detention
officers who happen to be on duty, exhibits the epistemic vice of incredulity, the
vicious attitude of not trusting the inmate’s voices and not taking their words at
face value systematically, as a matter of epistemic policy. Such institutional epis-
temic vice could be discontinued, and the testimonial insensitivity of detention
officers short-circuited, by protocols that force detention officers to take seriously
and respond to emergency calls and, more generally, by institutional epistemic
policies according to which an inmate’s words and expressive behavior have to be
taken at face value. This is what activist organizations such as the Inside/Outside
Alliance demand when they suggest that carceral institutions should implement a
policy of Believing Inmates. But critics reply that carceral institutions would fall
into the epistemic vice of gullibility if they were to follow this epistemic policy. I
will come back to this issue, but it suffices to note for now that, since it is the duty
of a carceral institution and of its officers to protect the safety and well-being of
inmates, it is difficult to see that there could be such thing as taking their emer-
gency calls too seriously or believing them too much when it comes to matters of
life and death.2 In addition to this, the specific carceral institution I am focusing
on is a county jail in which the overwhelming majority of inmates are awaiting
trial and have not been convicted of any crime, and DCDF has a legal and an
ethical obligation to presume their innocence. However, the evidence strongly
suggests that stigmas of criminality vitiate DCDF’s institutional attitudes towards
the inmates and, unfortunately, this seems to be typical of all carceral institutions
in the US, including jails.3
Following Cassam (2019), I will not focus on the specific motivation or etiol-
ogy of the epistemic vices at work in the testimonial insensitivity of the detention
officers or of the jail as an institution, but rather on their consequences. Cassam
has defended a compelling consequentialist view of epistemic vice that he calls
obstructivism. According to obstructivism, epistemic vices are to be understood
as attitudes, personality traits, or ways of thinking that “get in the way of knowl-
edge,” that is, that obstruct epistemic activities. I will follow Cassam’s obstructiv-
ist view in my analysis and in the exploration of the question of which epistemic
vices can be deemed capital vices in terms of their disastrous consequences. My
aim will be to explore what, within the testimonial insensitivity of DCDF and its
guards, counts as a capital epistemic vice that derails the epistemic interactions
between inmates and guards and obstructs the very possibility of sharing knowl-
edge. What could have possibly gone so wrong that the calls for help on behalf of
a dying individual could go unanswered, blatantly and outrageously disregarded
precisely by those in charge of protecting his life? What could possibly have gone
so calamitously wrong in the sensibility of those guards so that they were able to
leave those desperate cries for help unattended? And how is an institution set up
so that it allows its officers to exercise an epistemic insensitivity that incapacitates
them to listen properly to the voices of subjects in need? This chapter offers some
preliminary answers to these questions.
In the first section I will elucidate how certain vices at the core of a testi-
monial sensibility can disable a subject’s proper epistemic functioning, and in
the second section I will elucidate how epistemic vices at the core of an institu-
tion’s design can vitiate its whole modus operandi. In these diagnostic sections
I will try to distinguish between venial epistemic vices that do not compromise
112 José Medina
the epistemic character or functioning of a subject or system in toto and can be
treated through sporadic measures and interventions, on the one hand, and capital
epistemic vices that paralyze or vitiate the overall epistemic functioning of an
individual or system, on the other.4 What I call capital (or cardinal) epistemic
vices are those with maximal obstructionist power, that is, those vices that put
spikes in the wheels of a cognitive mechanism or epistemic system, so to speak.
Note that as I draw this distinction, it is not a categorical but a gradual distinction,
so that we can talk about epistemic vices being more or less capital. On my view,
capital epistemic vices need highlighting just because of their particularly harm-
ful consequences and obstructionist potential, and not necessarily because they
have genetic or explanatory priority over other (venial) epistemic vices. Capital
epistemic vices are those that derail the overall epistemic functioning of individu-
als and institutions.
In the final, brief, concluding section, I will explore the ways in which we
can resist the formation or continuation of capital epistemic vices. I will argue
for combining a character-based approach and an institutional approach, so that
we use a hybrid view of epistemic resistance that addresses the interpersonal/
interactive and the structural/institutional aspects of deep epistemic corruptions
or dysfunctions simultaneously.
6.2 Falling into the abyss of epistemic corruption

(I): corrupted testimonial sensibility
On my view, there is no catalog of capital epistemic vices we can provide. The
distinction between capital and non-capital or venial epistemic vices, first sug-
gested by Ian James Kidd (2017), cannot be drawn as an absolute distinction
by looking at each epistemic vice in isolation according to some pre-established
ethics or politics of knowing that postulates a fixed hierarchy of epistemic goods
and valuable features. Rather, the distinction is a practical and functional one
that has to be drawn case by case by looking at how the epistemic vice in ques-
tion functions holistically within the subject or institution that has it and within
particular contexts, and by looking at what the epistemic vice does epistemically
to the subject or institution in question as well as to others whom it affects. There
are two different criteria that we could use to draw a consequentialist distinction
between capital and non-capital when we examine a particular epistemic vice in
a particular context (or set of contexts): (1) according to the scope and depth of
the epistemic disablement that the vice produces in the subject or institution that
exhibits it (epistemic self-harm); and (2) according to the scope and depth of the
epistemic harms that it produces for others in (or through) the relevant epistemic
interactions. We can say that an epistemic vice has become capital for a particu-
lar subject or institution in a particular context if (1) the vice leads to forms of
epistemic disablement that obstruct epistemic well-functioning and derail epis-
temic cooperation; or (2) the vice leads to epistemic interactions that endanger
the epistemic dignity and agency of others in important ways. I will call crite-
rion (1) the lack-of-proper-epistemic-functioning criterion, and criterion (2) the
lack-of-proper-epistemic-relationality criterion. If an epistemic flaw triggers the
satisfaction of one of these criteria, that alone is sufficient to establish that the flaw
has become a capital epistemic vice. On my view, each of these criteria is inde-
pendently sufficient to establish that an epistemic vice has become capital, and
although very often both criteria will be satisfied simultaneously, as we shall see,
there can be cases in which only one of them is satisfied and that alone will suffice.
Note also that the application of these criteria has to be carefully contextualized.
The key question is whether an epistemic flaw or failing satisfies one of these
criteria in the relevant contexts, not in every possible context. Thus, for example,
in the case of the epistemic neglect of inmates by detention officers in DCDF,
the key issue is whether the officers’ incredulity with respect to inmates’ utter-
ances and expressive behavior disables their proper epistemic functioning in the
county jail, not in every possible context; for, indeed, these officers may function
properly in epistemic activities and interactions outside the county jail, but if,
because of the vicious attitude of incredulity with respect to inmates, they do not
process information adequately, they become incapable of giving proper uptake,
they become epistemically lazy, etc., we can say that the vice of incredulity has
indeed vitiated their entire epistemic functioning inside the jail and, within that
context, they have become corrupted epistemic subjects.
Note that there are good reasons to think that the lack-of-proper-epistemic-
functioning criterion and the lack-of-proper-epistemic-relationality criterion will
often be satisfied simultaneously since it is to be expected that there will be traffic
and interaction between proper epistemic functioning and proper epistemic rela-
tionality. On the one hand, if the inability to function properly epistemically deep-
ens (with epistemic failings such as being closed-minded, thinking arrogantly,
being epistemically lazy and careless, etc., piling up), the growing epistemic
disablement is likely to vitiate one’s epistemic relations to others (one’s ability
to listen properly, to trust, to give uptake, to make good use of others’ epistemic
contributions, etc.). In other words, sustained epistemic disablement undermines
epistemic cooperation and cuts one off from healthy epistemic relations. On the
other hand, if one’s epistemic relations to others become vitiated, this will surely
affect one’s own epistemic development by diminishing the opportunities to
properly exercise one’s epistemic capacities and cultivate epistemic virtues while
increasing the opportunities to develop epistemic limitations and flaws. In other
words, vitiated epistemic relations with others are likely to undermine one’s epis-
temic flourishing and to contribute to the development of epistemic vices.
Now that I have offered a preliminary description of the criteria by which we
can deem an epistemic flaw a capital epistemic vice, let’s look more closely into
how exactly not only the detention officers in DCDF who neglected inmates’
calls for help, but also the institution itself, can be said to satisfy these criteria
and, therefore, be guilty of (blameworthy or at least criticizable for)5 a capital
epistemic vice. In particular, let’s examine the different kinds of harms that can
be identified according to the two criteria laid out: epistemic self-harms as identi-
fied in the lack-of-proper-epistemic-functioning criterion, and epistemic harms
to others as identified in the lack-of-proper-epistemic-relationality criterion. For,
114 José Medina
indeed, what motivates drawing a distinction between capital and venial epistemic
vices is the need to be more attentive to (and to repair) the harms that result from
such vices, starting with the deepest and most menacing harms.
Far from being a mere academic exercise, drawing this distinction has a clear
payoff. Usually epistemic vices come in complex and broad clusters within a cor-
rupted mind or an epistemic pathology, and it is important to identify which vices
lie at the very core of the epistemic dysfunction, corruption, or pathology—which
ones have the most harmful consequences—so that they receive special attention
and they are given priority in our corrective and reparative practices of epistemic
resistance. For example, in some cases, misplaced incredulity may become a capi-
tal epistemic vice and should then receive our most urgent attention because it
corrupts the subject’s or institution’s epistemic functioning, resulting in serious
epistemic self-harm, but also because it forecloses or thwarts epistemic interac-
tions in which people are not heard, or their voices are distorted, or their testi-
mony is unfairly discredited, etc., therefore resulting in serious epistemic harms
to others. In other cases, however, misplaced incredulity may function as a venial
epistemic vice that does not have serious harmful consequences, either in terms of
epistemic self-harm for its possessor or in terms of harms to others. Let’s see how
incredulity with respect to inmates’ utterances and expressive behavior seems to
have functioned in those cases of epistemic neglect at DCDF that resulted in the
death of inmates by medical neglect.
Consider the incredulity of detention officers at DCDF as it operates when
they tend to disregard inmates’ emergency calls because they do not believe that
they are really in need of urgent help. From a consequentialist and obstruction-
ist perspective, this incredulous attitude is vicious in the sense that it vitiates the
proper epistemic functioning of the officers as well as their epistemic relationality
with inmates. What are the epistemic harms that result from the epistemic vice of
incredulity in this case? First, the officers’ incredulity satisfies the lack-of-proper-
epistemic-functioning criterion since it epistemically disables the subjects in ques-
tion in carrying out epistemic tasks they are supposed to do, such as gathering
information about the inmates’ safety and well-being. Moreover, the officers’
incredulity is embedded in a pattern of epistemic self-harm, that is, in the corrup-
tion of the officers’ testimonial sensibility, which includes epistemic flaws that
accrue as a result of their vicious incredulous attitude, such as the following: their
epistemic arrogance and incuriosity, their lack of motivation to find out whether
or not the emergency calls are warranted, their epistemic laziness in making no
efforts to find out more about the situation, etc.6
In the second place, the officers’ incredulity satisfies the lack-of-proper-epis-
temic-relationality criterion as well: because of their incredulous attitude with
respect to the inmates, they disregarded their emergency calls and this prevented
inmates from giving testimony while witnessing Mr. McCain die in front of their
very eyes, resulting in epistemic as well as moral and physical harms for the
inmates. Following Fricker’s analysis of the different harms of epistemic injus-
tice (2007), let’s distinguish between, on the one hand, the primary (epistemic)
harms that occur at the moment of the epistemic interaction, or missed encounter,
between the inmates and the detention officers before Mr. McCain’s death, and,
on the other hand, the secondary (epistemic and non-epistemic) harms that occur
as a result of the epistemic neglect. In the first place, the primary and immedi-
ate harms that inmates receive at the moment of the epistemic neglect are two-
fold: first, the epistemic harm of being cut off from effective communication,
having their expressive behavior rendered ineffective, so that their knowledge
is prevented from being shared and the vital information they possess about Mr.
McCain’s critical condition is not circulated; and second, the hybrid, ethico-epis-
temic harm of receiving no respect as a subject of knowledge, the violation of
the dignity of inmates as speaking subjects capable of giving testimony (a deep
epistemic and moral harm). In the second place, besides the primary harms that
are constitutive of the testimonial injustice suffered by inmates at DCDF, there
are also secondary harms that result from the act of epistemic neglect: clear addi-
tional harms that are both epistemic and non-epistemic, negative ways in which
the epistemic neglect reverberates in other epistemic and non-epistemic transac-
tions in the jail. The secondary epistemic harms that result from the inmates being
distrusted and epistemically neglected include the epistemic repercussions that
such distrust and neglect have for the inmates’ epistemic agency and character in
future interactions, for example, by undermining their epistemic confidence, their
capacity to trust, and also by increasing the likelihood of testimonial smother-
ing7 or self-silencing in future interactions with the guards and other jail officials.
The secondary non-epistemic harms that result from the inmates’ being distrusted
and epistemically neglected include all the non-epistemic consequences that such
distrust and neglect have for the inmates’ life in jail: negative psychological con-
sequences such as anxiety, depression, etc.; and negative practical consequences
such as an increased vulnerability to being unfairly disciplined and subject to
physical harm and injury, including death, as in Mr. McCain’s case.8
Of course, there can be cases of inmates issuing false emergency calls at
DCDF, as the detention officers are prompt to point out. But note that even in
those cases the policy of Believing Inmates is good epistemic policy and does
not result in a problematic credulous attitude or the epistemic vice of gullibility
(except perhaps in exceptional contexts in which the frequency of false alarms
has become extremely high).9 Note that even in false-emergency-calls cases not
following the policy of Believing Inmates and exercising distrust without any
grounds will result in epistemic harms. Even in such cases baseless incredulity
functions as a capital epistemic vice in my view, because even in such cases the
incredulity satisfies both criteria, although the second one only in a very lim-
ited way. The satisfaction of the lack-of-proper-epistemic-functioning criterion
is clear because, even in these cases, the incredulity epistemically disables the
guards in their interaction with the inmates (even if there is no pay-off resulting
from those interactions) and is likely to contribute to the corruption of the offic-
ers’ testimonial sensibility, incentivizing other epistemic vices such as epistemic
arrogance and epistemic laziness.
On the other hand, the satisfaction of the lack-of-proper-epistemic-relation-
ality criterion may not be so clear in cases of false emergency calls, but I would
116 José Medina
argue that even in these cases there is a partial satisfaction of the criterion: what
we do not have is the primary epistemic harm of obstructing knowledge from
being circulated, but what we still do have is the indignity of being treated as a
second-class citizen in the epistemic community, the ethico-epistemic harm of
being disrespected as a subject of knowledge. Also, in false-emergency-call cases,
secondary epistemic and practical harms can still accrue as a result of the ethico-
epistemic harm of epistemic neglect.10 So the policy of Believing Inmates is good
epistemic policy at least for contexts of communication that concerns matters of
life and death such as in responding to emergency calls, for, in such contexts, this
policy guarantees proper epistemic functioning and proper epistemic relationality.
6.3 Falling into the abyss of epistemic corruption

(II): epistemically corrupted institutions
As I pointed out from the beginning, after Mr. McCain’s death by medical and
epistemic neglect came to light, under the pressure of the activist organization
Inside/Outside Alliance, it was uncovered that there have been multiple similar
cases at DCDF in the preceding months and many more in the preceding years.
Once we notice that these are not isolated events, we cannot simply attribute the
epistemic failings only to the individual epistemic vices of the particular guards
on duty in each case. There is in fact a pattern of epistemic neglect at DCDF. In
the light of this pattern, we can (and should) move the analysis of epistemic vices
from the personal to the institutional level. But how do we establish the presence
of a capital epistemic vice in an institution? What counts as institutional epistemic
behavior? One way of proceeding is by paying attention to what the institution
should have been doing to guarantee the fair epistemic treatment of inmates and
to guarantee, as much as possible, that tragic cases like the epistemic neglect of
emergency calls do not happen. In this way we can detect the epistemic negli-
gence of the institution itself as setting the stage for specific instances of epistemic
neglect in the interaction between inmates and detention officers.
To begin with, a carceral institution can and should have protocols that make
it impossible for guards to arbitrarily disregard emergency calls, so that as soon
as an emergency button is pushed and/or a cry for help is heard, it is required that
a guard be immediately sent to check on the alleged emergency and all activi-
ties stop until this requirement is fulfilled. And in order to prevent systematic
epistemic neglect in different areas of carceral life, institutional protocols that
implement a communicative-epistemic policy of taking the expressive behavior
of inmates seriously are also needed in non-emergency situations. In particular,
protocols are needed for guaranteeing that inmates’ complaints and grievances
are properly handled. At DCDF, inmates do have opportunities to file grievances,
which is crucial in order to ensure that they have epistemic subjectivity and agency
to give testimony about their problems and concerns in carceral life. The problem
is that such grievances are rarely properly processed, responded to, and acted on.
Protocols and proper procedures are needed here to ensure that inmates’ voices
are respected, and their utterances and pronouncements are taken seriously and as
deserving of trust. Unfortunately, that is not the case at DCDF. As Medina and
Whitt’s (forthcoming) analysis of testimonial dysfunctions within DCDF under-
scores, hundreds of detainee letters collected by Inside-Outside Alliance offer
hundreds of examples of jail staff verbally disregarding, minimizing, or dismiss-
ing detainee complaints and requests out of hand. They write:
A computerized system now enables detainees to file grievances and monitor

their status, and this ostensibly makes the grievance process more efficient
and transparent. However, the grievances are routed to over one hundred
potentially responsible parties, and detainees report that grievances are not
resolved, progress is not updated, or that complaints and requests fall into
gaps in the system.11 In this way, the computerized system reinforces a social-
epistemic context in which detainee voices are systematically disadvantaged
within the jail, even as it provides the appearance of technological impartial-
ity and efficiency.
(Medina and Whitt forthcoming, my emphasis)
Besides instituting and implementing protocols for properly handling grievances,

requests, and emergency calls, there are at least two other things a carceral institu-
tion should be doing to ensure that the epistemic dignity of inmates is respected
and sufficient levels of trust are maintained: the proper training of officers and
jail staff to guarantee proper epistemic functioning and proper relationality vis-
à-vis the inmates; and the use of procedures of accountability so that officers and
jail staff are held accountable for epistemic mistreatment, that is, so that they
face consequences if they unfairly distrust inmates or treat them in epistemically
undignified and disrespectful ways. As Medina and Whitt’s analysis of testimonial
dysfunctions at DCDF shows, “correctional officers are explicitly and implicitly
trained to not trust, fraternize with, or empathize with incarcerated persons.” The
institution lacks training processes that foster empathy, unprejudiced attitudes,
and good listening habits. Carceral institutions should work with psychologists
and educational experts who can help them to design training aimed at mitigating
or bypassing the influence of prejudices.
Carceral institutions should also consult with formerly incarcerated subjects
in designing training practices so that their perspectives and vulnerabilities as
subjects of knowledge under the care and protection of the institution are taken
into account and properly addressed. It would serve carceral institutions well to
pay particular attention in the design of their training practices to especially vul-
nerable populations within the criminal justice system of the US such as African-
Americans and Latinos who are disproportionately incarcerated and mistreated
within the prison system (see esp. Rios 2011). In this essay I have not focused on
race/racism as a key factor in the epistemic mistreatment of incarcerated subjects,
but, although the prejudices and stigmatizations concerning criminality affect all
incarcerated subjects, they do not affect them all equally, and inmates of color are
more vulnerable to epistemic disrespect and unfair distrust based on prejudicial
and stigmatizing ways of thinking—it is indeed not accidental that Mr. McCain as
118 José Medina
well as many of those who have died at DCDF by epistemic and medical neglect
were African-American.
Finally, the best protocols for proper epistemic functioning and proper epis-
temic relationality will be ineffective if officers and jail staff are not held account-
able when they violate these protocols, and the best training for instilling healthy
epistemic attitudes and habits will be ineffective if officers and jail staff do not
act in consequence with their training and their contravening the training has no
consequences. A carceral institutional commitment to a communicative-epistemic
policy of taking inmates’ words and expressive behavior seriously requires clear
accountability procedures that are made known to officers and jail staff and are
conscientiously followed to guarantee that infringing the institution’s expecta-
tions about epistemic respect for inmates and proper epistemic treatment cannot
happen with impunity.
The three institutional deficiencies I have highlighted—the lack of adequate
protocols, of proper training, and of accountability procedures—clearly dem-
onstrate that the institution has failed to take proactive steps to guarantee that
the inmates’ expressive behavior is taken seriously and trusted as a matter of
communicative-epistemic policy.12 The institution exhibits the epistemic vice
of incredulity with respect to the inmates since it has not instituted the default
attitude of trusting inmates unless evidence to the contrary surfaces. And the
epistemic vice can be ascribed to the institution on the basis of these failures
alone and independently of actual patterns of distrust exhibited by officers on
the grounds. Note that the epistemic vice of unfairly distrusting inmates can be
attributed to the institution even when there is no pattern of epistemic mistreat-
ment of inmates that can be detected over time by the accumulation of cases.
For example, think of the following counterfactual scenario, which is different
from DCDF and other crowded county jails: a jail with so few inmates and with
such a quick turn-around that inmates are processed in and out of the facilities
without there being much of a chance for a pattern of epistemic neglect to be
formed and clearly exhibited. Still, the epistemic vice can be counterfactually
attributed to the jail as a carceral institution given its lack of protocols, training,
and accountability procedures to prevent epistemic neglect from occurring. A
single instance of epistemic neglect in this scenario is likely to be attributed to
the flawed detention officers on the grounds, but, in fact, it is already a systemic
problem because of the absence of adequate protocols, training, and account-
ability procedures.
By lacking a communicative-epistemic policy of taking inmates’ voices seri-
ously and systematically failing to instill and demand healthy trusting attitudes
and habits, the county jail can be said to exhibit an institutional attitude of unwar-
ranted distrust and incredulity with respect to the inmates, and this vicious epis-
temic attitude does satisfy the two criteria I laid out for counting as a capital
epistemic vice. On the one hand, the lack of epistemic proper functioning clearly
happens at the institutional level: DCDF, as an institution, does not treat inmates
as trustworthy in their grievances, requests, and emergency calls, and there-
fore becomes epistemically ill-equipped to carry out properly basic epistemic
operations required for providing care and protection for the inmates. It is not
only an individual failure of the particular jail officers and administrators whom
these grievances, requests, and emergency calls happen to reach, but also a failure
of the institutional set-up that makes proper epistemic uptake unlikely or a matter
of luck. As a result, DCDF itself, as an institution, and not only its officers and
staff, fails to respect inmates as subjects of knowledge and to establish a proper
epistemic relationality with them. Therefore, the institutional vice of incredulity
can also be deemed capital according to the second criterion.
But where is this institutional, capital vice of incredulity grounded? A car-
ceral institution that does not have adequate communicative-epistemic policies
of listening to and trusting its inmates is likely to harbor not only misplaced
disinterest and distrust, but also a prejudicial mindset, that is, to harbor not only
vicious attitudes but also a vicious way of thinking. The prejudicial and stig-
matizing way of thinking that depicts inmates as intrinsically untrustworthy can
also operate as a capital epistemic vice of both the officers and the institution.
But although a lot has been written on the prejudices and stigmas of criminal-
ity as they operate in the American criminal justice system and how they make
detained and incarcerated people—and especially people of color—vulnerable
to mistreatments of all sorts (epistemic and non-epistemic), the epistemic vice
of employing a prejudicial and stigmatizing way of thinking is more elusive and
more difficult to prove or confirm with clear evidence than the epistemic vice
of incredulity. For this reason, I have bracketed the trickier issue of the pos-
sibly prejudicial and stigmatizing way of thinking exhibited in the institutional
behavior of DCDF, and I have focused on the issue of the institutional incredu-
lous attitude towards inmates and the capital institutional epistemic vice this
attitude amounts to. My elucidations here offer only the preliminary steps for a
full analysis of the institutional epistemic vices involved in cases of neglect in
the criminal justice system.
6.4 Fighting capital epistemic vices: resisting and

preventing deep epistemic corruption
As Fricker (2010) has suggested, there are “basic institutional virtues” that we
should expect all public institutions to exhibit because “they are of fundamental
importance to the legitimacy of the polity in general,” and, Fricker goes on to say,
“institutional virtues of epistemic justice, most notably the virtue of testimonial
justice, are basic in this sense” (2010: 250). By the same token, we can also say
that the institutional vices of epistemic injustice—and of testimonial injustice in
particular—should be thought of as “basic” vices with a particular kind of central-
ity for public life and for the epistemic life of institutions. Building on this insight
and going beyond it, in this essay I have examined some of the different vices that
may be contained within the broader vice of testimonial injustice and I have laid
out epistemic criteria that can allow us to draw a distinction between capital and
venial epistemic vices, so that we identify those capital vices that are at the core
of processes of epistemic corruption both for subjects and for institutions. With
120 José Medina
an obstructionist view of epistemic vice in the background, the motivation for
providing a finer-grained diagnostics of epistemic vices—and of the epistemic
vices of testimonial injustice in particular—was to be able to offer some guidance
for corrective and reparative practices, that is, to be able to identify the epistemic
vices that were most damaging, with the most harmful consequences, and there-
fore capital, so that they could be addressed first; and by resisting them first, we
could stop the epistemic bleeding—so to speak—where it was most needed, giv-
ing priority to the prevention of the most nefarious harms and trying to halt the
process of epistemic corruption at its foundations.
In my previous work I have tried to shed light on struggles against epistemic
oppression and marginalization and efforts to fight against epistemic corruption
by elucidating what I call epistemic resistance (Medina 2013). In my most recent
work, I have elucidated our concerted efforts at epistemic resistance on the ground
as practices of epistemic activism (Medina 2018, Medina and Whitt forthcoming).
What are the forms of epistemic resistance and the forms of epistemic activism
that need to be mobilized to correct and repair capital epistemic vices that are
creating fatal epistemic dysfunctions at DCDF? Here we have to identify practices
of epistemic activism that will lead to interventions and transformations that will
improve both interpersonal testimonial sensibilities and institutional epistemic
policies and designs.
As explained by Medina and Whitt (forthcoming), we can think of epistemic
activism as concerted efforts and interventions in epistemic practices that aim
to “augment the epistemic agency of unfairly disadvantaged subjects, amplify-
ing their voices and facilitating the development and exercise of their epistemic
capacities.” Epistemic activism can take many different shapes and forms. Its
strategies and tactics will be dictated by who engages in it, in what contexts,
and against what patterns of interaction and institutional frameworks. Differently
situated subjects, both oppressed and non-oppressed subjects, can become epis-
temic activists. “Oppressed subjects can become epistemic activists—sometimes
by necessity if not by choice—when they actively fight against their epistemic
marginalization and work towards forms of self-empowerment that can achieve
the epistemic agency they are unjustly denied” (ibid.). Within carceral contexts,
inmates themselves can (and often do) become epistemic activists by denouncing
and trying to resist unfair patterns of epistemic neglect, and by expressing epis-
temic solidarity by backing up one another’s testimonies, so that they mitigate the
harmful consequences that individual acts of protest typically encounter. A good
example of epistemic activism cultivated by jail inmates at DCDF is provided by
Medina and Whitt:
At DCDF an unknown number of detainees recently organized the “First Five

Grieving Committee,” a “non-violent” and “non-gang affiliated” cooperative
that anonymizes and amplifies the grievances of individual detainees. By
working together, the members of the Committee have successfully directed
their concerns to the Durham County Sheriff, whereas individual grievances
are typically heard—if they are heard at all—by subordinate staff members.
This is an instance of epistemic activism, within the context of the jail, start-
ing to ameliorate the testimonial disadvantage that detainees face.
(Medina and Whitt forthcoming)
Not only oppressed subjects but also their allies can engage in epistemic activ-
ism. In cases of epistemic neglect within carceral contexts, “activists, scholars
and journalists, family members, political leaders, social media participants, and
in short the general public can join forces with jail detainees to help ensure that
their voices are heard, and their concerns are addressed” (Medina and Whitt forth-
coming). I want to pay special attention here to the work of epistemic resistance
of the Inside-Outside Alliance (hereafter IOA), a local activist organization that
describes itself as “a group of people trying to support the struggles of those inside
(or formerly inside) Durham County jail, and their families and friends.”13 IOA
members—friends and family of incarcerated subjects, formerly incarcerated sub-
jects, and activists—engage in epistemic interventions, programs, and initiatives,
which they subsume under the heading Amplify Voices Inside. There are two kinds
of interventions in the epistemic activism of IOA that I want to highlight: epis-
temic resistance inside the jail and epistemic resistance outside the jail.
In the first place, IOA members use their voices and epistemic agency within
the jail itself to echo, support, and empower the neglected voices of inmates and
to put pressure on DCDF and its workers to meliorate their dynamics and policies.
Think of cases of epistemic neglect of inmates’ grievances. Detainees often com-
plain that their unaddressed grievances disappear in the system, and when these
grievances have been especially urgent, they have worked with IOA members to
put external pressure on jail administrators.
In November 2014, the activist group organized call-ins to overwhelm

administrators’ phone lines when evening meals were reduced to two cold
sandwiches. They adopted similar tactics in September 2015, when jail staff
would not grant emergency medical transfer to a detainee in severe pain. In
these actions, phone calls from diverse community members—many of whom
do not consider themselves to be “activists”— echoed detainee grievances in
ways that made them more difficult to disregard or disbelieve. Additionally,
the phone calls reminded jail staff of their accountability to the local com-
munity. In these ways, the actions temporarily disrupted typical patterns of
interaction inside and outside of the jail, and indicated the possibility for
alternative, less dysfunctional patterns. For alternative patterns to take hold,
however, it may take repeated activist interventions.
(Medina and Whitt forthcoming, my emphasis)
What this epistemic activism can achieve is the (at least temporary) interruption
of dysfunctional dynamics of epistemic neglect, making it difficult (at least tem-
porarily) for detention officers and jail staff to act on epistemic vices (such as
incredulity) that rise to the level of capital vices. The epistemic resistance of IOA
members by itself will not create fair epistemic dynamics, but it can help trigger
122 José Medina
a process of amelioration of testimonial dynamics and it puts pressure on indi-
viduals and institutions to discontinue epistemic vices and improve testimonial
sensibilities.
In the second place, IOA’s epistemic activism also takes place outside the
county jail in their attempts to procure epistemic standing for inmates’ perspec-
tives and some degrees of epistemic agency for their voices in the outside world.
As Medina and Whitt put it,
detainees’ voices rarely reach places of political authority without being dis-
torted, translated into other idioms or discourses, or ventriloquized by others.
For this reason, it is important to have forms of epistemic activism in which
outside allies lend their voices as instruments or extensions of the detainees’
own, without interpreting or translating them.
IOA members do this “by reading detainees’ letters in City Council meetings
and County Commissioner meetings, disrupting ‘business as usual’ with the tes-
timonies of individuals who have been excluded from the sites of official power.”
Other ways in which IOA members seek to amplify detainee voices include:
“publishing their letters verbatim, usually without context or commentary, on the
website Amplify Voices Inside”; and publishing “detainee letters and artwork in a
print magazine called Feedback” (ibid.). In these different ways IOA members try
to ensure that the voices of inmates are heard in the outside world and their stories,
problems, and concerns neglected inside the jail can reach other institutions and
authorities as well as the general public.
Epistemic activism against epistemic corruption and capital epistemic vices
is much more than consciousness-raising; it is an attempt to meliorate epistemic
dynamics and institutional frameworks so that capital epistemic vices are uprooted
and the work towards epistemic justice can begin. In carceral contexts, this work
of epistemic resistance needs to happen both at the level of interpersonal dynam-
ics, targeting the testimonial insensitivity of carceral workers, and at the level of
the institution itself, targeting the epistemic vices of the institutional structure in
its protocols and procedures or lack thereof. The activist interventions of IOA I
have highlighted show clear ways in which the epistemic attitudes and habits of
guards and jail staff are put under pressure to change and improve by becom-
ing accountable to outside publics and authorities. But they also have to become
accountable within the institutional framework of the jail itself, and this calls not
only for a melioration of the testimonial sensibility of the officers on the grounds,
but also for institutional transformations that result in in-house accountability as
well as in new protocols and training procedures.
Individual and institutional capital epistemic vices work in tandem and feed
each other, and corrective and reparative practices of epistemic pathologies
need to take both interpersonal and institutional measures because both interper-
sonal dynamics and institutional frameworks need to be meliorated simultane-
ously. Solutions or corrective measures that are purely interpersonal or purely
institutional are doomed to fail. Purely interpersonal correctives disregard the
institutional aspects of the dysfunction and assume that more epistemically just
institutions will emerge as a by-product of interpersonal epistemic meliorations.
On the other hand, purely institutional correctives by themselves are also insuf-
ficient, for they disregard the interpersonal aspects of the dysfunction and assume
that more epistemically just interpersonal dynamics will emerge as a by-product
of institutional epistemic meliorations. Both of these approaches are misguided.
We need a hybrid approach that is both interpersonal and institutional. The epis-
temic activism I have elucidated in this section gives a good illustration of how
to target both institutional epistemic vices and vicious interpersonal dynamics
simultaneously.
It is important to emphasize that the work of epistemic resistance and the eradi-
cation of capital epistemic vices require not simply activist interventions, but con-
stant, sustained efforts in the perfectionist struggle toward epistemic virtue and
fair epistemic dynamics within institutional frameworks. In a carceral institution
such as DCDF, this epistemic perfectionism has to be constantly cultivated in the
three areas of institutional design and practice I have highlighted. First, it is not
enough to institute adequate protocols, but it is also necessary to use assessment
processes for the revision of these protocols and to engage in sustained efforts
toward changing the institutional culture and institutional values so that inmates’
voices and perspectives are respected and treated fairly. Second, a carceral institu-
tion such as DCDF needs to provide not only adequate training, but also constant
re-training, which is crucial not only to address the epistemic flaws and failings
of workers who are already in the institution, as opposed to newcomers, but also
for revisiting and continuing the epistemic melioration of attitudes, habits, and
dynamics initiated in the training process, since no training practice will be able
to guarantee once and for all that epistemic vices will not set in. Third, the institu-
tion needs not only to have adequate accountability procedures, but also to follow
them conscientiously so as to ensure that protocols are followed, that the received
training and re-training are incorporated in daily epistemic interactions, and that
the institutional culture and values operate properly in the embodied sensibility of
people on the grounds.
In this chapter, I have offered a preliminary analysis of capital epistemic vices
that can afflict both subjects and institutions by laying out two criteria to identify
such vices and by applying the analysis to a specific case of capital testimonial
injustice in a carceral institution. This provides only the first step towards future
work on the diagnostics of capital epistemic vices and, more importantly, on the
reparative and corrective practices needed to eradicate them.
Notes
1 Needless to say, incredulity is not always an epistemic vice: in the fable of the boy who
cried wolf, the farmers who ignored the boy the third time are not epistemically vicious
for doing so. Here context is everything, and there are indeed contexts in which incre-
dulity can be epistemically defensible, depending on what has happened in the past.
2 Matters of life and death raise the bar extremely high for judgments about the admis-
sibility or advisability of incredulity. But even here there could be contexts in which
124 José Medina
incredulity may not be an epistemic failing. As Quassim Cassam has pointed out to me
in personal correspondence, it is not epistemically vicious to ignore a fire alarm when it
tends to malfunction and systematically goes off without there being a fire, even though
a building fire is potentially a matter of life and death. The number of false alarms and
the evidence available about the malfunction are indeed crucial considerations in this
case for assessing the admissibility and advisability of incredulity.
3 See Cacho (2012) and Rios (2011). See also Medina and Whitt (forthcoming) for how
this can be detected at DCDF in particular.
4 Ian James Kidd (2017) has introduced the concept of capital epistemic vices. I draw
inspiration from his remarks in personal correspondence, but I develop the distinction
in this essay in my own way.
5 I am using “guilt” and “blameworthiness” here in a loose way to allude to the respon-
sibility of a subject or an institution for the vices they harbor. But it may advisable to
focus on criticizability (rather than culpability) as the most apt normative category, as
Cassam (2019) has argued.
6 One may wonder why I want to highlight the vicious epistemic attitude of incredulity
and not one of the other flaws, such as incuriosity or epistemic laziness. Acknowledging
that these other flaws are also capital epistemic vices in this case would not be a prob-
lem for my analysis, but I want to give center-stage to incredulity in this case because
the other flaws work in tandem with incredulity but not independently, whereas incre-
dulity does not seem to require that the other flaws be already there prior to the incredu-
lous attitude (e.g. it does seem necessary that the officers be already epistemically lazy
prior to succumbing to incredulity). But in the case under examination the attitude of
incredulity may not work fully independently of all other epistemic vices; in particular,
the incredulity of DCDF officers seems to be the product of a vicious way of thinking: a
prejudicial and stigmatizing mindset that depicts inmates as intrinsically untrustworthy.
I will comment on the connection between the vicious attitude of incredulity and the
prejudicial way of thinking of a stigmatizing carceral logic later in the chapter.
7 See Dotson (2011).
8 Needless to say, in this context, deeming a harm secondary does not mean that it is
less important, but only that it does not happen immediately at the moment of the
epistemic interaction in question, but rather as a result of it. The epistemic neglect of
the emergency calls by detention officers at DCDF contributed to the eventual death
of Mr. McCain in January 2016, but Mr. McCain’s death is not a direct and immediate
consequence of such neglect.
9 See endnotes 1–2 for a brief discussion of this important qualification.
10 Note that there could be, of course, different kinds of cases that depart from the cases
of epistemic neglect by incredulity that I am describing. In particular, there can be sce-
narios of ethical neglect in which the emergency calls of inmates could be disregarded
by guards not out of incredulity but out of lack of care (for example, cases in which
the guards and/or the carceral institution itself do not see inmates as truly deserving of
basic protections and medical care). These cases will not include the indignity of being
considered intrinsically untrustworthy, that is, the primary harm of being disrespected
as a subject of knowledge capable of giving reliable testimony.
11 Durham Jail Investigation Team, “Initial Report of Grievances, Rules, Backgrounder,”
internal report prepared May 2016, and “Durham Jail Investigation Team Infopack
for Human Relations Commission,” report prepared September 2016. Both reports are
based on Durham County public records, detainee letters, news coverage, and informa-
tion provided by DCDF Public Information Officers. Available upon request.
12 This is what activist organizations such as IOA seem to be demanding with the slogan
Believing Prisoners. Such a slogan highlights the need for a communicative-epistemic
policy of giving proper uptake to prisoners and taking their words (or communicative
attempts) seriously and (in principle) as worthy of belief.
13 See www.amplifyvoices.com.
References
Cacho, Lisa M. 2012. Social Death: Racialized Rightlessness and the Criminalization of
the Unprotected. New York: New York University Press.
Cassam, Quassim. 2019. Vices of the Mind: From the Intellectual to the Political. Oxford:
Dotson, Kristie. 2011. “Tracking Epistemic Violence, Tracking Practices of Silencing.”
Hypatia 26 (2): 236–257.
Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford/
New York: Oxford University Press.
Fricker, Miranda. 2010. “Can There Be Institutional Virtues?” In T. Szabo Gendler and
J. Hawthorne (eds.), Oxford Studies in Epistemology Volume 3: Social Epistemology.
Oxford/New York: Oxford University Press, pp. 235–252.
Kidd, Ian James. 2017. “Capital Epistemic Vices.” Social Epistemology Review and Reply
Collective 6 (8): 11–16.
Medina, José. 2013. The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic Injustice, and Resistant Imaginations. Oxford/New York: Oxford University
Press.
Medina, José. 2018. “Resisting Racist Propaganda: Distorted Visual Communication and
Epistemic Activism.” The Southern Journal of Philosophy 56: 50–75.
Medina, José, and Whitt, Matt. Forthcoming. “Epistemic Activism and the Politics of
Credibility: Testimonial Injustice Inside/Outside a North Carolina Jail.” In Nancy
McHugh and Heidi Grasswick (eds.), Making the Case. Albany: State University of
New York Press.
Rios, Victor. 2011. Punished: Policing the Lives of Black and Latino Boys. New York:
New York University Press.
7 Implicit bias and epistemic vice
Jules Holroyd
Can implicit biases be properly thought of as epistemic vices? I start by sketching

the contours of implicit biases (Section 7.1), and then turn to the recent claim,
from Cassam, that implicit biases are epistemic vices (Section 7.2). However, I
argue that concerns about the stability of implicit biases and their role in produc-
ing behavior make for difficulties in establishing that implicit biases of individu-
als are epistemic vices (Section 7.3). I then consider a recently developed model
which prompts us to consider implicit biases as properties of groups (Section 7.4).
This raises the question of whether implicit biases might constitute collective
epistemic vice. I suggest that there is a way to make sense of this claim, but it
requires rethinking how we conceptualize collective epistemic vice (Section 7.5).
These re-conceptualizations can be independently motivated. I close by marshal-
ing some considerations in favor of using the terminology of vice to capture these
defects of collective epistemic practice (Section 7.6).
7.1 What are implicit biases?

Implicit biases are heterogeneous phenomena. Authors tend to point to various
features that implicit biases share: they operate quickly and automatically, they
may be difficult for the agent to control or be aware of, they may be arational
or limited in the extent to which they are guided by the norms that govern other
mental states. Biases may differ in the extent to which they manifest each of these
characteristics. There are different kinds of biases, that appear to operate in differ-
ent ways, and may be differently related to individuals’ other attitudes, motives,
and beliefs. Different biases are related to different kinds of behavior. For exam-
ple, some biases might affect judgments (how competent another is); others might
affect one’s manner (e.g. how friendly one is). Exactly how we should character-
ize implicit biases is contentious, and not a matter that needs to be settled here
(see Holroyd 2016 for an overview and critique of various ways of conceiving
implicit biases).
The manifestation of implicit biases in behavior is also a complex phenome-
non. Certain patterns of implicit biases have been pervasively found in large-scale
studies that use implicit measures to access biases on which people are unable
or unwilling to report: biases against women, black people, minority ethnicity
Implicit bias and epistemic vice 127
individuals, and other socially stigmatized groups. This is unsurprising, given the
interaction between our cognitions and background patterns of social inequality
and injustice (see Madva 2016 for discussion of how to understand this inter-
action). Moreover, that implicit biases are found to be pervasive resonates with
reports of discrimination as persistent and pervasive from those who experience it
(see e.g. Williams, 2014; Valian, 2005, Sue et al. 2008). The pervasiveness with
which biases are found has led some to posit implicit biases as important explana-
tory factors in understanding persisting patterns of exclusion and discrimination
(Greenwald et al. 2015). If very many people, even only occasionally, behave in
ways that express implicit bias, a pattern of discriminatory judgment and behavior
could emerge, with exclusionary consequences.1 Next, I introduce some particular
kinds of associations, and the sorts of behaviors in which we might find these
biases expressed.
Implicit gender bias and judgments of competence or leadership: various stud-
ies indicate gender bias in the evaluation of CVs, whereby women’s CVs are
judged to demonstrate less competence, or merit lower pay grades, than commen-
surate CVs of male counterparts (Bertrand et al. 2005; Moss-Racusin et al. 2012).
Studies have also shown that women are more strongly associated with notions to
do with the family than with career-oriented notions (with which men are associ-
ated) (Rudman & Kilianski 2000), and that women are less strongly associated
with leadership roles than men (Valian 2005).2
At issue here, then, are the associations themselves (between women and fam-
ily or supporting roles, men and career or leadership roles) and behaviors that
appear to be underpinned by implicit biases: judgments of lesser competence,
and lesser recognition, or undervaluing, of women’s achievements. When women
are viewed through the lens of stereotypes, or judged to be less professionally
competent, or to have less intellectual acumen or leadership, due to implicit bias,
should we think that those who make these judgments display an epistemic vice?
It can be difficult to reach any evaluations or judgments of agents for their per-
formance on implicit measures in laboratory contexts – not least because biases
may be visible in these contexts because all else is held fixed (in a way that is
rarely the case in ‘real-world’ scenarios). So it will be helpful to have in mind a
real-world example.
Implicit bias outside the lab: the following scenario is anonymously reported
on the ‘What is it like to be a woman in philosophy?’ blog:
I was at a bar with three colleagues, each of whom are a) male, b) my friends,
and c) self-identified feminists. So there were four philosophers in a bar, at
a 3:1 male-to-female ratio. The table was discussing a book that only half in
attendance had actually read. Now, I was one of the two folks who had read
the book. It should surprise you, then, to learn that for the life of me, I could
not get a word in edgewise! 3/4 people were talking, and only 1/3 of those
speaking had read the book under discussion, but every freakin’ time I tried
to speak, I was summarily shut down, talked over, and/or ignored … I was
disheartened and sad to be treated this way by my friends. I picked up my
128 Jules Holroyd
phone, only to find that it was out of batteries, and tossed it back down on
the table, frustratedly. One colleague took notice of my frustration and asked
what was the matter, to which I responded rather directly, “Well there is
nothing else for me to do at this table, and now my phone is out of batteries.”
His response? “That sucks. So anyway, how was your weekend with [my
partner]?” Shocked and appalled by this totally unnatural segue, I retorted,
“We don’t have to stop talking about philosophy!” [implying of course: just
because you’re going to include me, now.] Totally unawares, he sincerely
replied, “No! I really wanted to know how your weekend was!” He didn’t
even realize what he had done …
All three of these guys are my friends, they are self-identified feminists,
and they take themselves to be good allies. I’ll bet if I told this story back to
them in another context, all three of those guys would be appalled. But from
the inside, they had no idea what they were doing.3
Of course, since implicit measures (or measures of any kind) were not used in this
case, we cannot know that implicit biases were at work here. But let us interpret
the case on the plausible assumption that this is an instance in which implicit
bias is driving the behavior. This seems credible if we accept that the author is
accurate in her judgment that her colleagues are friends, feminists, and would-be
allies. They do not subscribe to the belief that ‘women have nothing of value to
contribute to philosophical discussions’, say; when they are dismissive of her con-
tributions, they seem not to intend to devalue her, nor to realize that this is what
they are doing. It is not a stretch to explain such behavior in terms of implicit asso-
ciations between women and family (rather than career) – especially given their
willingness to include her in a discussion about her relationship! – or tendencies
to evaluate women as less competent, or to undervalue women’s contributions.
Such behavior may be influenced by implicit associations such as those mani-
fested in the studies described above. Insofar as such behavior expresses implicit
gender biases, are the philosophers who so behave displaying an epistemic vice?
7.2 The prima facie case for implicit biases as intellectual vices
In this section, I outline the contours of Cassam’s recent claim that implicit biases
are epistemic vices.4
7.2.1 Bias impedes knowledge

On Cassam’s ‘obstructivist’ account of epistemic vice (2016 and developed in
his 2019):
OBS: an epistemic vice is a blameworthy or otherwise reprehensible charac-

ter trait, attitude, or way of thinking that systematically obstructs the gaining,
keeping, or sharing of knowledge.
(2019, 23)
For example, gullibility is an intellectual vice because it is a trait that hinders
responsible knowledge acquisition, leading the inquirer to rely on unreliable
sources and leap to unsupported conclusions. This analysis of epistemic vice
is helpfully expansive, taking in not only traits, but also attitudes or ways of
thinking. Cassam is also explicit that cognitive biases, including implicit biases,
may also be candidates for epistemic vice. Drawing on the idea (from Banaji &
Greenwald, 2016) that implicit biases are ‘habits of thought’ Cassam argues that
implicit biases can be understood as epistemically harmful attitudes (2019, 168–
173). The cases on which he focuses are those of weapons biases: the tendency to
misperceive items as weapons when primed with black faces (Payne 2006). Such
biases are implicated in potentially lethal harms (such as racist patterns of police
shootings), as well as epistemic harms (impeding perceptual knowledge). Implicit
biases of this sort, and the kind we considered above, seem clearly reprehensible –
that is, criticizable – and Cassam makes the case that such biases are blameworthy
insofar as agents are what he calls ‘revision responsible’ for them – that is, if we
have the control to weaken or rid ourselves of such biases.5 Since there is reason to
suppose that biases are to some degree malleable, and can be weakened by various
forms of self-manipulation (see Holroyd & Kelly 2016), this, together with their
obstructive role in inquiry, would suffice on Cassam’s account, to make implicit
biases epistemic vices. Our examples (from Section 7.1) might be thought to add
plausibility to this claim: in experimental studies, gender biases hinder knowledge
about the competences and value of women, and knowledge sharing is certainly
impeded by the gender biases of the colleagues involved in the ‘What is it like…?’
example.
However, I want to flag the following issue now, to which I will return (in
Section 7.3, below): namely, the requirement, in OBS, that vices systematically
obstruct knowledge. This claim might pose difficulties for the obstructivist’s con-
tention that implicit biases in individuals are epistemic vices.
7.2.2 Other conceptions of epistemic vice

Cassam’s conception of epistemic vice is not the only one. An alternative view
has it that ‘vices will be qualities that reliably produce the bad’ (Battaly, 2014,
56). For epistemic vices, the bads at issue will, paradigmatically, be false beliefs.
On this conception (the reliabilist conception of vice (see also Battaly 2015), a
trait or mode of thinking – be it hard-wired or acquired – is vicious if it reliably
produces bad effects or outcomes. This relation to bad effects is necessary and
sufficient for a quality to be a vice (2014, 57).
Battaly also sketches a ‘responsibilist’ conception of vice. This view focuses
on the blameworthy psychology of the agent: ‘bad motives, false conceptions
of the good, dispositions to perform bad actions … are required for vice’ (2014,
62). This view – the responsibilist view – is driven by the thought that vices are
those aspects of our character that are within our control. Another intuition sup-
porting responsibilism is that what matters is that our character expresses what
we care about and value (62). I introduce these views to note that Cassam’s is
130 Jules Holroyd
not the only conception on which a case might be made for implicit biases as
vices. Implicit biases may produce bad epistemic effects (false beliefs), or may
be rooted in blameworthy psychologies (see e.g. Holroyd 2012, Holroyd & Kelly
2016, Brownstein 2016).6 One might think that these alternative conceptions, like
Cassam’s, could also accommodate the claims that implicit biases are epistemic
vices.
However, these accounts commit to the view that implicit biases should reli-
ably produce bad effects, or are stable traits of the agent. In the next section, I
argue these requirements (like that of systematic obstruction of knowledge) pose
difficulties for establishing that implicit biases, in individuals, are epistemic vices.
7.3 The challenges

7.3.1 Predictive validity
Does an individual’s implicit bias systematically obstruct inquiry or reliably pro-
duce false beliefs (or other epistemic bads)? To systematically impede inquiry,
bias need not invariably do so. Cassam rather follows Driver (2001) in requiring
rather that the connection between the vice and the bad epistemic outcome be
‘non-accidental’ (2019, 11). This is intended to rule out cases where luck plays
a role: ‘to make room for the possibility that epistemic vice can have unexpected
effects in particular cases’ (12). For example, if an implicit bias occasionally and
unexpectedly promoted, rather than obstructed, knowledge that need not under-
mine its putative status as a vice. But what is meant by ‘non-accidental’, pre-
cisely? In a useful footnote, Cassam asks what we might think of counterfactual
scenarios in which something we presently consider a virtue (open-mindedness)
‘normally gets in the way of knowledge’ (fn 25 at p.12, my italics). So our ques-
tion is whether an individual’s implicit biases systematically, that is, normally, or
in the usual run of things (without luck or causal deviance), obstruct knowledge.
Recent meta-analyses examining the relationship between individuals’ implicit
biases and behavioral outcomes are highly pertinent to this issue. Greenwald et al.
(2015), in their defense of implicit measures, examine the ‘predictive validity’
of individuals’ implicit biases: namely, the extent to which how an individual
performs on an implicit measure enables us to predict how they will behave (the
correlation between biases and certain behaviors). They point out that the pre-
dictive validity of the IAT is what psychologists would call ‘low’7 – that is, an
individual’s score on an Implicit Association Test (e.g. whether they have asso-
ciations between women/family and men/career-related notions) does not allow
us to predict with confidence how that individual will behave towards women.
They take this to be perfectly consistent with their defense of implicit measures –
and I return to this shortly – but the point for now is that it problematizes the claim
that an individual’s implicit bias will be systematically obstructive of inquiry,
or a reliable producer of bad effects. That implicit biases are poor predictors of
behavior suggests that they don’t normally, in the usual run of things, produce
such bad effects. Occasionally they do, but often they do not (producing neutral
or non-discriminatory behavior). Note that the bad effects at issue include both
behaviors, e.g. how far away an individual might sit from the target of the bias,
and judgments, e.g. of competence or value. The latter concerns whether knowl-
edge (accurate judgment) is promoted or obstructed: bad epistemic effects.
The issue is simply that, whilst implicit biases might be pervasive, they aren’t
particularly good predictors of whether individuals will behave in discriminatory
or knowledge obstructing ways. That is to say, in many instances in which we
find an individual harbors an implicit bias, we don’t find a strong relationship to
such behaviors.8
There are various ways in which this issue might be addressed: the first would
be to appeal to Cassam’s distinction (2019: 58–68) between a character vice –
possessed by a person – and a thinking vice – a vicious way of thinking that can
be displayed on occasion even by those who do not have the related character
vice. One could on occasion be, e.g. closed-minded, thereby displaying a thinking
vice, without being a closed-minded person in general. This could enable us to say
that implicit biases are thinking vices: when that mode of thinking is displayed, it
obstructs inquiry or produces false beliefs.9
However, it is not clear that this deals with the problem. Consider the distinc-
tion between the presence of an implicit association in an agent’s mental economy
(e.g. between women and family-oriented notions), and the activation and use of
that association in a particular deliberative episode (biased thinking). The meta-
analyses don’t directly address this issue, but since they concern how individuals
perform on an implicit measure (which activates a bias) and their subsequent per-
formance on some behavioral measure, there is reason to believe that they concern
episodes of biased thinking. So the meta-analyses should also lead us to conclude
that biased thinking weakly correlates with behavioral outcomes. That is, whilst
an individual may engage in episodes of biased thinking (they might make asso-
ciations between men and career, and women and family, or might automatically
undervalue the qualifications of women), these thoughts may be overridden by
other, non-biased aspects of their deliberative processes. The systematic – normal,
in the usual run of things – relationship between episodes of biased thinking and
biased behavior also faces the challenge from predictive validity.
A second option might be to consider implicit bias as a low-fidelity, rather than
high-fidelity vice (Alfano, 2013, 31–32, discussed in Cassam 2019, 32–34). High-
fidelity traits, Cassam suggests, require near perfect consistency: one is not gener-
ous unless one behaves generously quite consistently. But many ordinary vices,
he suggests, are low-fidelity: occasional expression suffices for the vice. One
doesn’t have to be consistently cruel to be cruel: one episode suffices. Would bias
best be construed as a low- or high-fidelity vice? Are only those who behave in
biased ways on a regular basis displaying the vice of bias, or is a one-off instance
of bias sufficient for someone to qualify as vicious (as is plausibly the case for,
e.g. vicious cruelty)?
Whilst Cassam argues that many ordinary vices are low-fidelity, I am inclined
to think that bias is akin to closed-mindedness, which Cassam characterizes as a
high-fidelity vice. An individual who is generally open-minded, but has a domain
132 Jules Holroyd
in which they display closed-mindedness, seems not to have the vice of closed-
mindedness – that domain is one in which they behave in out-of-character ways.
Likewise with bias: an individual who on occasion displays biased thinking need
not have the vice of bias – they in this instance behave in a biased way, which is
out of character.
Consider this issue of predictive validity in light of the example from the ‘What
is it like…?’ blog. We might reasonably infer that the behavior of the author’s
colleagues is not routine for them: were that the case, it is perhaps unlikely that
the author would describe them as friends, much less feminist allies. Rather, the
incident is notable, we can infer, because even friends and card-carrying femi-
nists might on occasion manifest implicit biases. As emphasized by researchers
on implicit bias, such biases are pervasive and all of us are at risk of, on occasion,
manifesting bias in behavior.10
Implicit biases in an individual’s mental economy don’t appear to produce the
relevant (bad) consequences with the required systematicity to establish them as
epistemic vices. So it is not clear we can establish the conclusion that implicit bias
systematically or reliably impedes inquiry or produces false belief; it is not clear
that implicit bias is an epistemic vice.11
How do those who defend the explanatory importance of implicit biases in
understanding discrimination deal with the issue of predictive validity? Greenwald
et al. (2015) maintain that implicit biases are significant, despite their low predic-
tive validity, by pointing to the cumulative effects of implicit biases when they
are manifested, even just occasionally, by very many people. Using statistical
modeling they show that across a large number of people, implicit biases that cor-
relate weakly with individual behaviors could nonetheless manifest in significant
behavioral outcomes across the group as a whole. This suggests that we might do
better to consider the phenomenon of implicit biases at the level of groups. Before
considering this option, let us turn to the other consideration: whether implicit
biases might be thought of as stable traits.
7.3.2 Stability
One might endorse a conception of vice where what matters is that the vices are
stable character traits. The issue of the stability of implicit biases has been hotly
contested in recent writings. This contention rests on the fact that implicit meas-
ures – such as those mentioned in Section 7.1 above – have been found to have
low test-retest reliability. That is to say, as Gawronski puts it ‘a person’s score
on an implicit measure today provides limited information about this person’s
score on the same measure at a later time’ (2019, 583). This is not what would
be expected if the measures tracked an individuals’ stably expressed traits.12 So,
some have concluded that the measures instead access rather more transient states
of the agent: what happens to be in mind at a particular time: ‘the momentary
activation of associations in memory’ (Gawronski, 583). If that is the case, then
it puts pressure on the idea that implicit biases – as measured by the sorts of tests
described in Section 7.1 – are vices. A ‘momentary activation’ is certainly not a
stable trait, which would pose a challenge for accounts according to which epis-
temic vices should be stable traits.13
A competing interpretation of test-retest reliability findings is to acknowledge
that what individuals have in mind on any one occasion is of course dependent on
contextual factors, such that it is no surprise to find that across a range of contexts,
the extent to which an individual expresses bias on an implicit measure varies. It
is after all well-known that implicit biases are malleable: they are highly sensitive
to features of the context. This has to do both with the person and their situation.
Regarding the person, how tired or distracted they are, on any particular occasion,
affects how susceptible individuals are to implicitly biased modes of thinking.
And context takes in features of the situation: with whom one is interacting, what
exemplars from different social groups are encountered (stereotypical or counter-
stereotypical) (see Dasgupta & Asgari, 2004), the environment in which a person
is encountered, what pressures from social norms are exerted, and so on. We store
a rather complex set of information, which can include problematic stereotypes
and evaluations; which subset of that stored information is activated depends on
the context (see Gawronski 2019).
This way of interpreting the findings about test-retest reliability somewhat
vindicates the implicit measures: it is not surprising that there is relatively low
test-retest reliability. But it still poses a challenge to the idea that implicit biases
are stable features of individuals that qualify as character traits in the way the
responsibilist requires.14 Consider again the ‘What is it like…?’ example. For all
we know, the colleagues in this scenario have varying results on implicit measures
(this is likely, if they are in keeping with much of the population). And, as noted,
to the extent that they display implicit bias here, this seems noteworthy because
it is not in keeping with the rest of their characters. They may display implicit
biases, but they do not appear to evince a stable character trait in doing so. This
poses difficulties for any account of vice according to which it is a stable trait.
I have suggested that the recent analyses showing the low predictive valid-
ity of implicit biases, and the low test-retest reliability of measures of implicit
biases, puts pressure on the idea that implicit biases could constitute epistemic
vices in individuals. However, these concerns should not lead us to reduce the
extent to which we are concerned about implicit biases.15 The challenges confront
the specific idea that implicit biases in individuals are epistemic vices.16 But these
challenges have also motivated a new way of conceiving of implicit biases, which
prompts us to consider the issue of collective epistemic vice. Next, I introduce the
new model of implicit biases, and then turn to consider collective epistemic vice.
7.4 The bias of crowds

Despite the fact that individuals’ scores on implicit measures are unstable, and
vary from one occasion to the next, there is remarkable stability in aggregate
levels of implicit biases across groups (Payne et al. 2017). This suggests that,
whilst individuals’ biases are unstable, and individuals’ biases do shift, the nature
of that individual shift is limited in a way that does not undermine the mean
134 Jules Holroyd
level of bias of a group. Moreover, whilst implicit measures are weak predictors
of individual behavioral outcomes, the aggregate implicit biases of a group are
more strongly associated with differential outcomes. Payne et al. draw on analy-
ses that show that in countries in which the aggregate level of implicit gender
bias is higher (in particular, the association with men and STEM subjects), there
are greater gender-based achievement gaps in science and math subjects (Nosek
et al. 2009); in regions with higher implicit racial biases (associating black people
with negative notions such as danger or crime) there are greater racial disparities
in police shootings (more black people are shot) (Hehman et al. 2018). What can
explain the stable aggregate levels of implicit bias, and the stronger relationship
with disparate outcomes, despite instability and weak predictions generated at the
individual level?
Payne et al. propose that we should see implicit biases as an attribute of situ-
ations or contexts, rather than individuals (2017, 236). By this, I take it that they
want to emphasize the contribution of contextual factors to the ways individuals
behave, such that patterns of biased behavior emerge across samples operating
within a particular context. Indeed, their spelling out of this claim is that situa-
tions, or contexts, encode or contain social stereotypes (we might also appeal to
other aspects of a social context, such as scripts, narratives, and aspects of social
meaning (cf. Haslanger 2015)). Features of a particular situation affect what is
situationally accessible.17 For example, if a stereotype of women as nurturing car-
ers is prominent, that will affect the extent to which that stereotype is accessible
to individuals. Likewise, if the majority of caring roles are in fact occupied by
women, or if prominent representations portray women in such roles, this will also
affect the extent to which a stereotype is situationally accessible. Since implicit
measures record the stereotypes and associations that are accessible, individuals
in that situation will display biases (on implicit measures). Indeed, the situation-
ally accessible biases are fairly constant, so if the relevant features of the situation
and all else were held completely fixed, we could expect that the individual levels
of bias expressed would remain fairly constant (there would be good test-retest
reliability). But we aren’t mere sponges or mirrors of our situations. The extent
to which stereotypes are accessible changes for individuals across time and con-
text, depending on who we interact with, what thoughts we have, what our latest
interactions or engagements were, how present in mind stereotypes are, and other
aspects of our mental lives, etc. However, across the sample as a whole, the rela-
tive constancy of the background situation, and the stereotypes in that context,
contribute to a pattern of implicit bias emerging, which is (a) more stable, and (b)
strongly associated with disparate outcomes. I suggest that one way to interpret
these claims is that implicit bias is something manifested stably, in a way that
affects behavioral outcomes, in collectives or groups.
To speculatively flesh out an example: take the group of academic philoso-
phers in Anglophone institutions. Any individual philosopher, we would expect,
would demonstrate varying levels of gender bias on implicit measures. But the sit-
uationally accessible associations and stereotypes are fairly constant: in addition
to the background conditions of gender inequality that prevail in wider society,18
philosophy is stereotyped as male, much of the canonical literature taught and
taken as giving rise to central research questions is by male philosophers, only
recently have efforts been made to include more women and scholars of color in
curricula and in research events, and to uncover the contributions of marginalized
philosophers to the canon. Whilst individual measures of implicit bias would vary
from day to day (depending on what literature had just been read, with which col-
leagues one had engaged, what blogs one had read or contributed to), we would
expect a fairly stable mean level of bias across a large sample of academic phi-
losophers in Anglophone institutions. And, if in keeping with findings in other
contexts, we would expect this to better predict discriminatory outcomes across
the profession than individual bias predicts individual behaviors. This example is
speculative, since it is modeled on Payne et al.’s Bias of Crowds way of under-
standing bias, rather than underpinned by systematically gathered data looking at
implicit measures and behavioral outcomes in this context. But of course, it fits
with what limited data we do have about gender and under-representation in phi-
losophy19 and with the fact that plenty of anecdotal evidence points to patterns of
(e.g.) gender bias. Many women in philosophy experience some form of gender
bias, some of the time;20 few individuals who (presumably) have implicit gender
biases express them all or even much of the time. All that is needed is that many
express gender bias some of the time, even just occasionally – as in the ‘What is
it like…?’ example – for deleterious and discriminatory outcomes to take effect.
This is explained by the Bias of Crowds model.
On this model, whereby implicit bias is a stable property of groups, and mani-
fests stably in group behavior, should we think of it as a collective epistemic
vice? On the assumption that, at least in the context under discussion, implicit
gender biases obstruct inquiry (in the sorts of ways described in our ‘What is it
like…?’ case) and produce the sorts of negative epistemic outcomes associated
with exclusion of philosophers who otherwise have much to contribute, I focus on
the question of whether it is a collective epistemic vice. Much will depend on the
conception of collective vice at issue, to which I now turn.
7.5 Collective vice

The contours of the case – the Bias of Crowds – I have described are as follows:
the collective or group at issue is a relatively loosely formed group of individuals:
members of a nation, or region, or profession – without any particular institutional
structure unifying those individuals. Nonetheless, across those individuals, we
find certain patterns of behavior which produce certain outcomes. These patterns
of behavior are not intentionally coordinated. The outcomes are not aimed for.
Is it idiosyncratic to think of such cases as instances of collective vice? Loose
collectives have been considered candidates for collective virtue or vice before:
Slote’s (2001) account of group agency extends to societies, broadly construed;
Beggs (2003) considers his account of institutional virtue as applicable to the
polis. It is not uncommon to attribute vices to loosely constituted groups: Medina
writes of the epistemic arrogance of the ‘powerful and privileged’, for example
136 Jules Holroyd
(2013, 31). That the group is loosely constituted need not be an obstacle to seeing
the Bias of Crowds as a vice.21 What matter is whether they meet other conditions
for collective vice.
7.5.1 Joint commitment22

On one prominent account of collective virtue and vice, what is crucial is that
there is a group or collective constituted by individuals operating under a par-
ticular practical identity (team member, or participant in some endeavor). Each
individual takes on a joint commitment to some virtuous (or vicious) motive, or
to some virtuous (or vicious) end that will be pursued by some good (or poor)
method (Fricker, 2010, 241, 243).23 The virtuous members of the night watch each
take on a commitment to vigilance, say. Joint commitment, on Fricker’s account,
involves a practical and cognitive component. Cognitively, the participants each
‘take on’ a responsibility to do something, and will involve an awareness that one
is committing (2010, 245).24 Practically, this means that reneging on the commit-
ment will be accompanied by, at least, a demand for an explanation.
The Bias of Crowds model obviously won’t count as collective vice on this
model. It is entirely implausible to suppose that there is a joint commitment
to some bad epistemic motive, or bad epistemic end, involved in cases where
groups stably manifest implicit biases – that each participant of the loosely con-
nected group has committed to make discriminatory judgments about the value
of women philosophers, say, or to ignore contributions, or dismiss lines of argu-
ment. Of course, there may be pockets of bad epistemic motives, and there will
most likely be bad epistemic outcomes (loss of important knowledge, failures of
understanding, fruitful lines of enquiry not pursued). But it is hard to make the
case that these are outcomes that members comprising a group commit to pursing,
in any meaningful way of understanding that.
Is this the only option, though? Perhaps we need not establish that vicious
joint commitments are taken up. Indeed, at some points in Fricker’s discussion
there is the suggestion that at least in the case of collective epistemic vice (if not
virtue), participants need not actively take on a commitment to a bad motive or
end; rather, it suffices that they fail to commit to a good motive or end.25 In this
respect there is an asymmetry between vice and virtue.26 In her example of the
collectively vicious night watchmen – a bunch of slackers who nod off, entertain
themselves, and ‘in one or another manner signally failing to jointly commit to the
end of vigilance’ (243), Fricker writes that ‘given that vigilance and negligence
are exclusive opposites for a night watch, the watch thereby displays the collec-
tive vice of negligence’ (243). Merely failing to commit to some good end can, in
some cases, suffice to constitute epistemic vice.
On one reading of Fricker’s night watch case is that the failure to commit to a
virtue itself signals a vice.27 This seems to be Fricker’s own understanding of the
case, and one which applies here, since vigilance and negligence are exclusive
opposites, as she puts it. But this is a limitation of her account: insofar as there
are virtues in relation to which a failure to commit need not, thereby, signal vice,
these cases will not be captured by the joint commitment model. And indeed,
there do seem to be such cases. A failure to jointly commit to courage need not
signal cowardice; a group that does not jointly commit to generosity need not
signal miserly thriftiness. In the context of biases of crowds: we might hope that
a group would jointly commit to fair-mindedness; but a failure to do so does not,
in itself, signify closed-minded prejudice. Such failures might simply signify that
a group has other priorities: a commitment to cautious research rather than cour-
age; a commitment to prudential budgeting rather than generosity. Or – particu-
larly in the case of implicit bias – a failure to commit to fair-mindedness might
simply signal a failure to realize that any specific commitment on the matter is
needed.
There will also be some vices that collectives may manifest without any joint
commitment to bad ends or motives, and which are not signaled by failing to com-
mit to the opposite virtue. A collective or group may display the vice of disorgani-
zation without having jointly committed to being disorganized. Nor does a failure
to commit to good principles of organization signal a commitment to this vice.
The UK government’s approach to Brexit negotiations is a good example of this.
A group might display the vice of closed-mindedness without having jointly com-
mitted to this stance. Nor does a failure to commit to open-mindedness signal a
commitment to this vice. The trans-exclusionary organization ‘A Woman’s Place’
is a good example of this. An institution may display the vice of petty bureaucracy
without its members having jointly committed to opacity and obstructive modes
of operating. Nor does a failure to commit to well-justified efficiency signal a
commitment to this vice. Various helplines for utilities services exemplify this
vice. And, we might contend, a group may display bias without jointly commit-
ting to biased ways of thinking. Nor need a failure to commit to fair-mindedness
signal a commitment to bias.
The joint commitment route to understanding collective virtue and vice, then,
does not seem a promising one for capturing biases of crowds as collective epis-
temic vice. But there seem to be independent reasons for departing from the joint
commitment model of collective vice. Some vices that collectives may display
– disorganization, closed-mindedness, petty bureaucracy – are ill-captured by the
joint commitment model. What other options might there be?
7.5.2 Invisible hand mechanisms and dispositions to behave

A suggestive but under-explored alternative is also present in Fricker’s paper:
that virtues or vices might emerge by ‘invisible hand’ mechanisms, whereby the
group feature is not reflected at the individual level, but might emerge through
and ‘be explained by the way in which the individual level features synthesize
to create a quite different feature at group level’ (239). Fricker’s virtue-based
example: a jury might be constituted by prejudiced members whose prejudices all
cancel each other out, such that the overall judgment reached shows no prejudice
or imbalance.28 However, Fricker doubts such invisible hand accounts are well
placed to capture virtue, suggesting that the relationship between the supposed
138 Jules Holroyd
virtue and good conduct should not be accidental or a fluke. Rather, ‘the good
conduct should be performed because of the good motive or skill’ (239). This
non-accidental relationship seems not to be present in cases in which the trait in
question (fair-mindedness) emerges because the prejudices happen to cancel each
other out. The jury doesn’t seem creditworthy for their fair-minded verdict, she
suggests; in her view ‘the same point applies to vice’ (240).
But we have already seen that there could be reason for treating virtue and vice
asymmetrically. This may be another instance in which the conditions for virtue
and vice are not symmetrical. We could accept Fricker’s claim that, in the case
of collective virtue, the feature should not emerge accidentally. But in the case
of vice, we could maintain that, because negligence is one of the ways that vices
can emerge, mere accident of how the individual traits synthesize can produce a
collective vice.29
The general observation that groups could be vicious through negligence
seems to open the door to the invisible hand mechanism being one through which
genuine collective vices can arise. Suppose a jury is comprised of 12 fair-minded
individuals, but they fail to consider the way that, in their group dynamics, these
qualities may not be reflected; good norms of group discussion are not estab-
lished, some members dominate the discussion, assuming that others will speak
up if they disagree. Through negligence, the individual features of the group syn-
thesize to produce a poorly functioning collective, that lacks the fair-mindedness
that each of the constituent members individually possesses; the verdict instead
is ill-informed by evidence and manifests closed-minded prejudice. It is ‘mere
accident’ that this feature has emerged, in the sense that the jury did not commit to
it, and its emergence is not intentional. That does not undermine the case for such
a feature of the group being vicious. The group dynamic will reliably produce
epistemically poor decisions. One might hold, then, that vices can emerge from
invisible hand mechanisms, even if virtues cannot.
We can draw on Byerly and Byerly’s (2016) account of collective virtue to
develop an account of collective vice that can accommodate invisible hand vices.
According to their basic account of collective virtue:
Collective virtue: ‘a collective C has virtue V to the extent that C is disposed

to behave in ways characteristic of V under appropriate circumstances.’
(43)30
Thus, a jury has the virtue of fair-mindedness if it is disposed to behave in ways

characteristic of fair-mindedness under appropriate circumstances. If the consti-
tution of the jury makes it such that it is so disposed, then it has the virtue of
fair-mindedness – whether or not this constitution is ‘mere accident’, and whether
or not the members have taken on any commitments to that end. We can readily
apply this analysis to collective vice:
Collective vice: a collective C has vice V to the extent that C is disposed to

behave in ways characteristic of V under appropriate circumstances.
The case for collective vice thus construed may be stronger than for collec-
tive virtue, if one is swayed by the idea that the negligent production of vice
should be accommodated, even if the accidental production of virtue should
not.31 A jury is closed-minded or prejudiced to the extent that it is disposed to
reach prejudiced verdicts. A group is disorganized to the extent that it is dis-
posed to behave in poorly administered ways (failing to have a representative
at important meetings, having incoherent policies, uninformed representatives,
etc.). A collective is closed-minded to the extent that it is disposed to behave
in closed-minded ways (ignoring important evidence and arguments, select-
ing only evidence that supports the group’s stated aims, question-begging in
debates). An institution displays the vice of pettiness if it is disposed to behave
in petty bureaucratic ways (opaque and obstructive procedures, unbending and
inflexible adherence to protocols, insistence on procedural norms even when
irrational). A collective displays the vice of bias to the extent that it is dis-
posed to behave in discriminatory ways (patterns of behavior that disadvan-
tage some demographic, exclusion or devaluing of the contributions of some).32
Insofar as these examples appear to be cases of vice that collectives or groups
manifest, and insofar as they are manifested without joint commitment to some
bad motive or end, they can nonetheless be accommodated by the dispositional
account of collective vice.33
This account has some advantages over the joint commitment account: as
Byerly and Byerly point out, a group may commit to virtue without, in fact, being
disposed to behave in virtuous ways. Commitments count for little if they are
empty. Invisible hand cases also point to the importance of dispositions to behave,
rather than commitment. If vices can emerge in collectives, and these are mani-
fested in the dispositions to behave of the collective, then again, the importance
of commitment – at least for some vices – is undermined. If one thinks an account
should capture invisible hand cases (of vice, if not of virtue), this will also be an
advantage of the dispositional, over the joint commitment, analysis. There is inde-
pendent motivation, then, for moving away from the joint commitment account as
providing necessary conditions for collective vice. Some collective vices might
be instantiated through joint commitments to bad ends. But others may emerge
through invisible hand mechanisms and manifest in the collective’s dispositions
to behave in ways characteristic of that vice.
Where does this leave us in thinking about implicit bias? This way of mak-
ing sense of collective vice is particularly helpful for thinking about the Bias of
Crowds model. On this analysis a group has the vice of bias when it is disposed to
behave in ways characteristic of bias under appropriate circumstances. For exam-
ple, the collective of academic philosophers in Anglophone institutions would
have the disposition to gender bias to the extent that the collective is disposed to
behave in ways characteristic of gender bias (undervaluing women’s contribu-
tions, practices that exclude women from participation in research events, failures
to represent women’s contributions to the discipline on curricula, and so on).34 We
could also appeal to further evidence (where it is available) of such dispositions: a
stable mean level of bias found across a group would be strong evidence that the
140 Jules Holroyd
collective has the relevant disposition. And such dispositions could be evidenced
where there is a strong relationship with disparate outcomes for different demo-
graphics within that group.
As noted, the collectives at issue in the Bias of Crowds model are large and
loosely connected ones: nations, populations across certain regions. The extent
to which we consider these samples as collectives will rest on questions in social
metaphysics that cannot be settled here. But there does not seem to be any obvious
reason for which we should not treat such large samples of individuals as collec-
tives, if we find stable propensities to behave across such populations.
This view of collective vice will face an objection recently advanced by Cordell
(2017): that what I have identified is a feature of a collective, but does not amount
to a substantive vice.35 This is for two reasons: first, he argues that if a feature is
to be diagnosed as a substantive virtue or vice, then it must be something that the
agent (the collective) can reflect on as something to be cultivated or eliminated
from their functioning. But collectives of this sort (he argues) lack the requisite
processes of reflection. Second, Cordell suggests that one could avoid this first
concern by being purely instrumental about virtues or vices: whatever feature pro-
duces good or bad effects (irrespective of any mechanism for reflection on these
features) is a virtue or vice of the collective. But this instrumentalist picture is not
well suited to capture the extent to which the collective is an agent: the group has
a feature, but it is not a feature produced by the agent.
These objections may have some promise when directed towards an account of
collective virtue. Perhaps for a trait to be genuinely credit-worthy it does have to
be intentionally produced – perhaps via mechanisms of reflection – by the collec-
tive agent. However, I see no reason to accept these claims with respect to collec-
tive vices. As we have seen, vices of collectives could result from negligence, and
so by their nature will not be the result of intentional production, or the fruits of
a reflective mechanism that has decided to cultivate a particular feature. My view
is that it would be an excessively restrictive view of collective vice to insist that
they cannot be produced by negligence.
In sum: I have argued that we have good reasons to reject the claim that col-
lective vice requires joint commitment to some bad end or motive, and that a case
can be made for vices that emerge – through negligence – via ‘invisible-hand’
mechanisms. This seems true in at least some cases for vices such as closed-mind-
edness, prejudice, disorganization, or pettiness – and, in particular, bias. This can
be captured by an account of collective vice, drawing on Byerly and Byerly, that
focuses on the disposition to behave in ways characteristic of vice. Where these
dispositions or propensities affect knowledge-seeking activities, then, they can be
properly described as collective epistemic vice. Implicit biases, when understood
on the Bias of Crowds model, are contenders for collective epistemic vice.
7.6 Vice charging, individual and collective

I have suggested that there are obstacles to determining that implicit biases are
epistemic vices in the individual case. But I argued that we should think collective
vices can be captured on the ‘disposition to behave’ analysis, and can emerge
without joint commitments. On that analysis, we can claim that implicit biases
manifested by groups – the Biases of Crowds – are collective vices. Where pat-
terns of implicit bias across groups serve to obstruct knowledge-seeking or pro-
duce bad epistemic effects they will be collective epistemic vices. But why should
we want to be able to make such a claim? What is gained by being able to diag-
nose biases as vicious? What is the advantage of being able to call out collectives
as vicious?
As Kidd argues, charging an agent with vice should serve an ameliorative
function (2016, 192); the aim should be to do so in a constructive spirit, with a
view to improving the character or conduct of others.36 There are good reasons to
suppose that characterizing patterns of behavior of groups and collectives as vices
can serve an ameliorative function: first, doing so identifies a systematic defect in
the conduct of the collective – a defect which many of the individuals comprising
the collective would find reprehensible. This is particularly so in instances where
the defect has emerged from invisible hand mechanisms, and where no individu-
als have committed to bringing about the conduct or outcomes that have emerged.
Second, vice charging in the case of collectives can prompt members of the collec-
tive to reflect on how they sustain certain patterns of behavior, albeit unintention-
ally. It can draw the attention of individuals to ways in which they, with others,
are complicit in problematic patterns of behavior and outcome, despite their indi-
vidual subscription to good values, or despite good individual intentions. Third,
drawing attention to individuals’ roles in perpetrating collective vices, in this way,
might be a particularly good way of motivating change. Fourth, this is particularly
so because it prompts members of the collective to focus not just on what they do,
qua individual, but also on the structures, norms, and practices that enable these
vices to be enacted at the level of the collective. Finally, it focuses attention on
what collective measures are needed to avoid these problematic dispositions, and
highlights the importance of collective, rather than individual, virtue in addressing
these issues.37 Seeing the Bias of Crowds as a collective epistemic vice, then, may
serve an important ameliorative function in addressing the problematic patterns of
bias in which we are implicated.
Notes
1 This pattern may be part of a ‘perfect storm’ of factors all pointing towards exclusion
(see Antony 2012).
2 Goff and Kahn (2013) show that in such studies, the paradigm ‘woman’ that partici-
pants have in mind is a white woman. As such, we should be cautious about general-
izing these findings to women of color, who likely face biases that encode the ways
in which gender is racialized. Similarly, they urge caution about generalizing studies
about associations with black people, which may really hone in on stereotypes about
black men. As such there is a lacuna in the research on implicit bias that is only recently
starting to be addressed (see Theim et al. 2019 on the biases that might target black
women in particular).
3 https://beingawomaninphilosophy.wordpress.com/2014/01/12/ins idious-norms/
[posted 2014, accessed April 2019].
142 Jules Holroyd
4 It is also plausible that being influenced by bias is related to other epistemic vices:
closed-mindedness, dogmatism, epistemic negligence, perhaps. I set aside the interest-
ing task of teasing out the relationship between biases and other vices for another time.
5 But on this account, biases need not be blameworthy – merely criticizable – in order to
constitute vices.
6 Interestingly, in the debate about the blameworthiness of agents for implicit biases,
some who have held back from arguing that bias is blameworthy have tried to establish
that nonetheless aretaic evaluations of the agent – evaluations that appeal to virtue or
vice terms – are nonetheless apt (see e.g. Zheng 2016, Brownstein 2016). These authors
(appealingly, I think) detach blameworthiness from the kind of virtue and vice attribu-
tions in a way that is starkly at odds with the characterization of vice on the responsibil-
ist view.
7 But crucially, not as low as argued by the Oswald et al. (2013) meta-analysis, whose
inclusion criteria Greenwald et al. critique. Note that their meta-analyses focused on
implicit measures of racial attitudes.
8 Note that this is unequivocally not to say that implicit biases might have epistemic
benefits (cf. Gendler 2011 for this claim, which I find problematic for the reasons elu-
cidated in Puddifoot 2017 and Saul 2018).
9 Compare Levy’s argument for the conclusion that those who express implicit racial
bias are, in some important respect, racist; even if there are other aspects of their char-
acter that are not racist, or anti-racist (2017).
10 Compare the oft-quoted remark from Jesse Jackson: ‘There is nothing more painful
to me at this stage in my life than to walk down the street and hear footsteps and start
thinking about robbery. Then look around and see somebody White and feel relieved’
(Remarks at a meeting of Operation PUSH in Chicago (27 November 1993). Quoted
in “Crime: New Frontier – Jesse Jackson Calls It Top Civil-Rights Issue” by Mary A.
Johnson, 29 November 1993, Chicago Sun-Times). The quote is used to illustrate that
even those dedicated to anti-racism, and themselves stigmatized by the stereotypes
at issue, can on occasion manifest implicit bias. As such, the behaviour of the col-
leagues in our example is consistent with them being card-carrying feminists (though
of course, we rely on the author’s description which provides scant information about
their commitments, compared to the abundant evidence of Jackson’s anti-racist activ-
ism).
11 Note that my claim is not that implicit biases could never be part of an epistemic vice
that an individual possesses. In cases where implicit bias props up and is supported by
explicit bias, for example, we may well find epistemic vice (and other vices). My claim
is simply that implicit bias itself may not meet conditions for epistemic vice.
12 See also Brownstein et al. (2019) for discussion of whether implicit measures access
traits (variously construed) or states.
13 In fact, nor do such transient states seem to qualify as modes of thinking, even. The
term ‘modes of thinking’ suggests default assumptions or inference patterns that indi-
viduals tend on balance to rely on – not a mere momentary activation captured in labo-
ratory conditions.
14 The idea that individuals’ characters are constituted by how individuals react in particu-
lar contexts – rather than as context-free fixed points – is a familiar and much discussed
one (see Brownstein et al (2020) for discussion of this issue).
15 Also for reasons rehearsed in Holroyd and Saul (2019): namely that low predictive
validity still gives cause for some concern that biases might, on occasion, manifest;
and that the reliability is not markedly worse than other well-established measures; and
that the degree of variation on implicit measures is in keeping with a general pattern of
expressed biases. One’s bias might vary in strength, but less likely in valence.
16 Denying they are vices is perfectly consistent with thinking they are blameworthy in a
range of ways (see Holroyd et al. 2017 for an overview of claims about responsibility,
blameworthiness, and implicit bias).
17 Situational accessibility is contrasted with chronic accessibility (what is available to
the agent given their psychological make-up), but as the authors note, these two kinds
of accessibility will interact (Payne et al 2017, 236).
18 As Saul 2013 notes, these wider societal background conditions cannot be the whole of
the story, because philosophy is much worse, in terms of gender inclusion, than other
subjects in the humanities and most others across academia.
19 See Holroyd and Saul 2019 for an overview of some of the relevant data on inclusion in
philosophy. This draws on data from Beebee and Saul 2011, Norlock 2011, Botts et al.
2014 inter alia.
20 Though as reports on the ‘What is it like…?’ blog indicate, some of these experiences
look to be the result of blatant and explicit sexism.
21 Note, though, that the sort of groups I have in mind above are unlikely to meet Beggs’
(2003) conditions for constituting a collective (solidarity and decision procedures).
22 An assumption in what follows is that the discussion is premised on an anti-summativ-
ist conception of vice – that is, a conception whereby the collective vice is not reducible
to vices of the individual. This is precisely what is at stake in discussions of group-level
implicit biases – the property of the group (stable biases that correlate with disparate
outcomes) is precisely what is harder to establish at the individual level. I do not mean
to suggest that there is nothing defensible about summativist conceptions, but simply
that such accounts will not be the right model for the case in hand. For discussion
of summativist and anti-summativist approaches, see Fricker 2010, Lahroodi 2007,
Cordell 2017, Byerly & Byerly 2016.
23 Fricker also notes some reliability condition will also be needed, to ensure the relation-
ship between the motive or method and good outcome.
24 Though as Fricker emphasizes, it need not involve awareness that one is committing
to something qua virtue, nor the reliable relationship between that motive or way of
proceeding and good outcomes.
25 Compare Battaly’s concept 2* that requires not that individuals commit to a bad motive,
but that they fail to commit to a good motive. On a strong reading, Battaly argues, this
is an implausible view (2014, 64).
26 I have learnt much about these putative asymmetries from discussions with Charlie
Crerar. See also Crerar (2018).
27 An alternative reading would have it that there is tacit joint commitment between the
watch members. They are aware that they are each taking on a certain – bad, negligent
– way of proceeding, and expect each other to follow suit. This rendering of Fricker’s
watch case is consistent with her analysis of virtue and vice in terms of joint commit-
ment, but will not capture the Bias of Crowds. Consider the speculative example from
the last section: it stretches credulity to suppose that academic philosophers have tacitly
committed to ignoring certain contributions, or dismissing and undervaluing lines of
argument coming from women.
28 Another example Fricker uses is that of a debating society, the members of which are
prejudiced but whose prejudices cancel each other out such that the debate overall
shows no prejudice. One might find this example stretches credulity, since a non-preju-
diced debate concerns not just the balance of views expressed, but also the contents of
what is expressed. For this reason I focus on the jury example (since jury deliberations
are not revealed, any prejudiced contents expressed will not be known).
29 Compare Battaly’s remark (2014, 64) that in individuals vices can negligently emerge.
30 They offer a more complex formulation of this basic account (at p. 43), which makes
clear that the virtue can be construed in terms of group-dependent properties that indi-
vidual members have. However, because I find their argument from multiple realizabil-
ity convincing (an argument which purports to show that groups can have properties that
are not reducible to the individual realizers of those properties), I stick with this more
basic formulation. Nothing in the argument turns on this though, so readers are free to
substitute the more complex formulation from Byerly and Byerly should they see fit.
144 Jules Holroyd
31 I want to remain agnostic on what we should say about invisible hand mechanisms pro-
ducing virtue. My main point is that whatever we say about virtue, a case can be made
that collective vices can emerge through these invisible-hand mechanisms, negligence
being one of the key ways in which they can do so.
32 Note that whilst the emergence of the group-level property is not intentional, in the case
of bias it is not ‘mere accident’; social structures of racism and sexism are effective
engineers of these group-level properties.
33 We might ultimately come to quite different views regarding the vices of institutions
and groups or collectives on a number of matters, such as their collective responsi-
bility and blameworthiness, as well as forward-looking responsibilities for correcting
vice. These issues, which would have to address the hierarchical structures and power
dynamics involved in each, are beyond the scope of this chapter.
34 Beggs suggests that ‘practice’ might be considered the group analogue to individual
disposition (2003, 51). Practice on his account is understood as ‘the social grammars
(the types) that an individual agent’s actions manifest (the tokens)’ (466).
35 He advances another line of objection, targeted at Fricker’s joint commitment account:
that she has not provided an account of an irreducibly collective virtue – rather, he
argues, the virtues can be reduced to the commitments of individuals in their group-
oriented roles. Since I think there are other reasons to depart from Fricker’s joint com-
mitment account, I set aside this concern here. It is clear that the feature of the group
with which I am concerned – bias – is irreducible to members of the collective, given
the considerations raised in Section 7.3.
36 Kidd also notes that vice charges should ‘build in a suitably complex account of agen-
tial epistemic responsibility’ (2016, 194) and in particular one that is sensitive to the
aetiology of the vice. As noted in endnote 6, I find attractive a view according to which
vice attribution does not depend on establishing blameworthiness. Of course, there will
be many interesting and complex questions to address regarding collective responsibil-
ity or blame for implicit bias. In particular, it will be important in this context to be
sensitive to the power dynamics within the group, especially when it comes to forward-
looking responsibility: whose responsibility it is to undertake, or lead the way in taking,
corrective steps.
37 See Anderson (2012) for concerns that a focus on individual virtue is an insufficient
corrective for addressing implicit biases.
References
Alfano, M. (2013). Character as Moral Fiction. Cambridge: Cambridge University Press.
Anderson, E. (2012). “Epistemic justice as a virtue of social institutions.” Social
Epistemology, 26(2), 163–173.
Antony, L. (2012). “Different voices or perfect storm? Why are there so few women in
philosophy?” Journal of Social Philosophy, 43(3), 227–255.
Banaji, M. R., and Greenwald, A. G. (2016). Blindspot: Hidden Biases of Good People.
New York: Delacorte Press.
Battaly, H. (2014). “Varieties of epistemic vice.” In: R. Vitz and J. Matheson (eds.), The
Ethics of Belief. Oxford: Oxfod University Press, 51–76.
Battaly, H. (2015). “Epistemic virtue and vice: Reliabilism, responsibilism, and
personalism.” In: C. Mi, M. Slote, and E. Sosa (eds.), Moral and Intellectual Virtues in
Western and Chinese Philosophy. New York: Routledge, 109–130.
Beebee, H., and Saul, J. (2011). “Women in philosophy in the UK.” A Report by the British
Philosophical Association and the Society for Women in Philosophy in the UK.
Beggs, D. (2003). “The idea of group moral virtue.” Journal of Social Philosophy, 34(3),
457–474.
Bertrand, M., Chugh, D., and Mullainathan, S. (2005). “Implicit discrimination.” American
Economic Review, 95(2), 94–98.
Botts, T., Bright, L. K., Cherry, M., Mallarangeng, G., and Spencer, Q. (2014). “What is
the state of blacks in philosophy?” Critical Philosophy of Race, 2(2), 224–242.
Brownstein, M. (2016). “Attributionism and moral responsibility for implicit bias.” Review
of Philosophy and Psychology, 7(4), 765–786.
Brownstein, M., Madva, A., and Gawronski, B. (2019). “What do implicit measures
measure?” Wiley Interdisciplinary Reviews: Cognitive Science, 10: e1501. https://doi
.org/10.1002/wcs.1501.
Brownstein, M., Madva, A., and Gawronski, B. (2020). “Understanding implicit bias:
Putting the criticism into perspective.” Pacific Philosophical Quarterly, https://doi.org
/10.1111/papq.12302.
Byerly, T. R., and Byerly, M. (2016). “Collective virtue.” Journal of Value Inquiry, 50(1),
33–50.
Cassam, Q. (2016). “Vice epistemology.” The Monist, 99(2), 159–180.
Cassam, Q. (2019). Vices of the Mind: From the Intellectual to the Political. Oxford:
Cordell, S. (2017). “Group virtues: No great leap forward with collectivism.” Res Publica,
23, 43–59.
Crerar, C. (2018). “Motivational approaches to intellectual vice.” Australasian Journal of
Philosophy, 96(4), 753–766.
Dasgupta, N., and Asgari, S. (2004). “Seeing is believing: Exposure to counterstereotypic
women leaders and its effect on the malleability of automatic gender stereotyping.”
Journal of Experimental Social Psychology, 40(5), 642–658.
Driver, J. (2001). Uneasy Virtue. Cambridge: Cambridge University Press.
Fricker, M. (2010). “Can there be institutional virtues?” Oxford Studies in Epistemology,
3, 235–253.
Gawronski, B. (2019). “Six lessons for a cogent science of implicit bias and its criticism.”
Perspectives on Psychological Science, 14(4), 1745691619826015.
Gendler, T. S. (2011). “On the epistemic costs of implicit bias.” Philosophical Studies,
156(1), 33.
Goff, P. A., and K. B. Kahn. (2013). “How psychological science impedes intersectional
thinking.” Du Bois Review: Social Science Research on Race, 10(2), 365–384.
Greenwald, A. G., Banaji, M. R., and Nosek, B. A. (2015). “Statistically small effects of
the Implicit Association Test can have societally large effects.” Journal of Personality
and Social Psychology, 108(4), 553–561.
Haslanger, S. (2015). “Distinguished lecture: Social structure, narrative and explanation.”
Canadian Journal of Philosophy, 45(1), 1–15.
Hehman, E., Flake, J. K., and Calanchini, J. (2018). “Disproportionate use of lethal force
in policing is associated with regional racial biases of residents.” Social Psychological
and Personality Science, 9(4), 393–401.
Holroyd, J. (2016). “What do we want from a model of implicit cognition?” Proceedings
of the Aristotelian Society, 116(2), 153–179.
Holroyd, Jules. (2012). “Responsibility for Implicit Bias.” Journal of Social Philosophy
43(3), 274–306.
Holroyd, J. D., and Kelly, D. (2016). “Implicit bias, character and control.” In: A. Masala
and J. Webber (eds.), From Personality to Virtue: Essays on the Philosophy of
Character. Oxford: Oxford University Press, 106–133.
146 Jules Holroyd
Holroyd, J., Scaife, R., and Stafford, T. (2017). “Responsibility for implicit bias.”
Philosophy Compass, 12(3), e12410.
Holroyd, J., and Saul, J. (2019). “Implicit bias research and reform efforts in philosophy:
A defence.” Philosophical Topics, 48(2).
Insidious Norms, post at “What is it like to be a woman in philosophy?’ https://be
ingawomaninphilosophy.wordpress.com/2014/01/12/insidious-norms/ [posted 2014,
accessed April 2019].
Kidd, I. J. (2016). “Charging others with epistemic vice.” The Monist, 99(2), 181–197.
Lahroodi, R. (2007). “Collective epistemic virtues.” Social Epistemology, 21(3), 281–297.
Levy, N. (2017). “Am I a racist? Implicit bias and the ascription of racism.” The
Philosophical Quarterly, 67(268), 534–551.
Madva, A. (2016). “A plea for anti-anti-individualism: How oversimple psychology
misleads social policy.” Ergo, an Open Access Journal of Philosophy, 3.
Medina, J. (2013). The Epistemology of Resistance: Gender and Racial Oppression,
Epistemic Injustice, and the Social Imagination. New York: Oxford University Press.
Moss-Racusin, C., Dovidio, J., Brescoll, V., Graham, M., and Handelsman, J. (2012).
“Science faculty’s subtle gender biases favor male students.” PNAS, 109(41),
16474–16479.
Norlock, K. (2011). “Women in the profession: A report to the CSW.” American
Philosophical Association (February).
Nosek, B. A., Smyth, F. L., Sriram, N., Lindner, N. M., Devos, T., Ayala, A., and Kesebir,
S. (2009). “National differences in gender–science stereotypes predict national sex
differences in science and math achievement.” Proceedings of the National Academy of
Sciences, 106(26), 10593–10597.
Oswald, F. L., Mitchell, G., Blanton, H., Jaccard, J., and Tetlock, P. E. (2013). “Predicting
ethnic and racial discrimination: A meta-analysis of IAT criterion studies.” Journal of
Personality and Social Psychology, 105(2), 171–192.
Payne, B. K. (2006). “Weapon Bias: Split-Second Decisions and Unintended Stereotyping.”
Current Directions in Psychological Science, 15(6), 287–291. https://doi.org/10.1111/j.
1467-8721.2006.00454.x
Payne, B. Keith, Vuletich, Heidi A., and Lundberg, Kristjen B. (2017). “The bias of
crowds: How implicit bias bridges personal and systemic prejudice.” Psychological
Inquiry, 28(4), 233–248.
Puddifoot, K. (2017). “Stereotyping: The multifactorial view.” Philosophical Topics,
45(1), 137–156.
Rudman, L. A., and Kilianski, S. E. (2000). “Implicit and explicit attitudes toward female
authority.” Personality and Social Psychology Bulletin, 26(11), 1315–1328.
Saul, J. (2013). “Implicit bias, stereotype threat, and women in philosophy.” In: K.
Hutchinson and F. Jenkins (eds.), Women in Philosophy: What Needs to Change.
Oxford: Oxford University Press, 39–60.
Saul, J. (2018). “(How) should we tell implicit bias stories?” Disputatio, 10(50), 217–244.
Sue, Derald Wing, Capodilupo, Christina M., and Holder, Aisha. (2008). “Racial
microaggressions in the life experience of Black Americans.” Professional Psychology:
Research and Practice, 39(3), 329.
Slote, M. (2001). Morals from Motives. Oxford: Oxford University Press.
Thiem, K. C., Neel, R., Simpson, A. J., & Todd, A. R. (2019). “Are black women and girls
associated with danger? Implicit racial bias at the intersection of target age and gender.”
Personality and social psychology bulletin, 45(10), 1427–1439.
Valian, V. (2005). “Beyond gender schemas: Improving the advancement of women in
academia.” Hypatia, 20(3), 198–213.
Williams, Joan C. (2014). “Double jeopardy? An empirical study with implications for the
debates over implicit bias and intersectionality.” Harvard Journal of Law & Gender,
37, 185.
Zheng, R. (2016). “Attributability, accountability and implicit attitudes.” In: M. Brownstein
and J. Saul (Eds.), Implicit Bias and Philosophy, Vol. 2. Oxford: Oxford University
Press, 62–89.
8 Vectors of epistemic insecurity
Emily Sullivan and Mark Alfano
8.1 Introduction
Virtue epistemologists have largely neglected the ways in which epistemic virtue
functions in social epistemic environments of inter-connected information shar-
ers. Testimonial exchanges between two people receive the bulk of the attention.1
Moreover, only recently have epistemologists recognized that epistemic virtues,
like moral virtues, can be other-regarding (Kawall 2002; Fricker 2007). The slow
arrival of the social within virtue epistemology is likely due to the fact that vir-
tue is mainly understood as localized within an individual (Battaly 2008; Greco
2010; Montmarquet 1992; Sosa 2007; Zagzebski 1996).2 In this paper, we high-
light the ways in which epistemic virtue and vice depend on the larger structure of
one’s epistemic community (Alfano and Skorburg 2017a, 2017b). In particular,
we address the way that modal epistemic standings and the virtues and vices that
accompany these standings are networked.
We first consider the familiar modal epistemic standing of safety, which
obtains when the epistemic agent could not easily have believed falsely (Pritchard
2007). We argue that safety in a social network context is best understood as
vector-relativized. One’s belief is safe only if, holding constant the structure of
one’s network of informants (and one’s informants, and one’s informants’ inform-
ants … ), there is no close possible world in which one’s belief turns out false.
We then introduce a complementary modal epistemic standings, belief-security
and network-security, to cover a related phenomenon. One’s belief is secure only
if, given small perturbations in the structure of one’s network of informants (and
one’s informants, and one’s informants’ informants … ), there is no close pos-
sible world in which one’s belief turns out false. One’s network is secure only if,
given one’s network of informants, and given small perturbations in the structure
of one’s network of informants, one does not depend only on a small number of
non-independent sources for information. Given this framework, we discuss the
virtues and vices that are operative within the social epistemic context and how
these dispositions navigate the trade-offs between security and safety narrowly
construed. Understanding epistemic virtue and vice as depending on the larger
structure of one’s epistemic community enables us to capture the unique quali-
ties of social epistemic communities and the modal epistemic standings operative
within them.
Vectors of epistemic insecurity 149
8.2 Toward vector-relativized modal epistemic standings
Nozick (1981) argued that a modal requirement on knowledge is necessary. Beliefs
are epistemically better to the extent that they are sensitive. A belief is sensitive
iff S knows p via a method M only if, were p false, S would not believe p via M.
Nozick’s notion of sensitivity has since been abandoned for an alternative modal
standing: safety. Safety seeks to capture Nozick’s intuition that knowledge has a
modal requirement without the purported drawbacks of skepticism, abandoning
closure, and failing to allow for the possibility of knowledge about necessary
truths (Pritchard 2009; Sosa 1999a, 1999b; Williamson 2002). Safety continues to
play a central role in anti-luck epistemology. Recently, Pritchard (2016) proposed
that a related concept, epistemic risk, is more fundamental than safety.
We aren’t interested here in providing a set of necessary and sufficient condi-
tions for knowledge. Instead, we are interested in a range of modal standings
that—regardless of whether they are constitutive of knowledge—are epistemically
(dis)valuable in social epistemic networks. The central discussions surrounding
these modal epistemic concepts focus almost exclusively on cases of testimo-
nial knowledge between two people or the knowledge one person gains from her
immediate environment. Moreover, discussions of testimony almost exclusively
ignore the complexities that emerge when individuals are embedded within a
structured community, and how this complexity might impact modal epistemic
standings. In this section we argue that understanding safety as vector-relativized
brings modal epistemic standings in line with a truly social epistemology.
Generally speaking, a belief is safe when it could not have easily been false.
While there are many different formulations of safety (Pritchard 2009; Sosa
1999a, 1999b; Williamson 2002), the basic idea is that knowledge is not compat-
ible with a certain type of luck. In cases where luck is one of the main reasons
why a person’s belief is true, it seems that the person does not know. Much ink
has been spilled explicating the conditions under which a belief is safe (Kelp
2009; Rabinowitz 2011). A central part of the nuance comes in explicating what it
means for a possible world to be close and what degree of closeness undermines
knowledge. Leaving the discussion of close possible worlds aside, we turn to a
related issue that many agree on: safety is basis-relative.
In judging whether one’s belief is safe, we need to hold fixed the basis on
which, or method by which, the belief was formed (Nozick 1981: 179). For exam-
ple, if Alberto formed a true belief based on Jane’s testimony, then we should
consider the close possible worlds where that belief, based on Jane’s testimony, is
false. Even if it is (closely) possible that Alberto could believe the same proposi-
tion by reading it in the newspaper, or by consulting a Magic 8-Ball, these facts
do not impinge on the question whether Alberto’s belief—based on Jane’s testi-
mony—is safe. This results in the following definition of safety:
Basis-relative safety:
S believes that p safely on basis B iff there is no close possible world in which
S falsely believes p on B.
150 Emily Sullivan and Mark Alfano
Cases of testimonial belief raise interesting questions concerning whether the tes-
tifier’s belief must be safe in order for the receiver’s belief to be safe. Goldberg
(2005, 2007) argues that one can in fact have a safe belief from unsafe testimony.
He asks us to consider the following case:
Milk Carton: Mary observes a small carton of milk in Frank’s fridge at 7:40am.
She tells Sonny (who always has cereal with milk for breakfast) that there is
milk in the fridge. Sonny forms the true belief based on Mary’s testimony that
there is milk in the fridge. However, unbeknownst to both of them, this is just a
matter of luck. On every other morning, except this one, Frank at 7:30 empties
the milk carton and places the empty carton back in the fridge. However, since
Frank is in the kitchen with both Mary and Sonny, in all possible worlds where
the carton was empty, Frank would have interjected and corrected Mary’s testi-
mony. Thus, there is no possible world in which Sonny’s belief is false.
(Goldberg 2005: 302)
The idea is that even though Mary’s testimony is unsafe, Sonny still forms a safe
belief based on her testimony because there is no close possible world in which his
belief is false. Debate has since followed about whether Goldberg’s intuition about
unsafe testimony is right, and whether Sonny’s belief is actually a belief based on
testimony (Lackey 2008; Pelling 2013). We are not here interested in joining that
debate. Instead, we want to highlight that this case serves to broaden the scope of
how we should think of modal epistemic standings such as safety. In particular,
milk carton shows that safety is vector-relative.3 It is our contention that Sonny
seems to have a safe belief because he is not in an epistemic dyad, where informa-
tion is shared from exactly one person to exactly one other person. Instead, Sonny
is in an epistemic network with two sources, one actual (Mary) and the other merely
potential (Frank). The presence of both these sources entails that the belief Sonny
has about the milk in the fridge will be true in all close possible worlds. Thus, once
we move beyond epistemic dyads to a truly social epistemology it is evident that the
structure of one’s epistemic network greatly impacts whether one has knowledge.
We therefore propose the following account of safety in a social epistemic
network:
Vector-relative safety:
S believes that p safely within epistemic network N iff there is no close pos-
sible world in which S falsely believes that p in N.
Vector-relative safety can also explain why someone’s belief is unsafe. Consider
another case introduced by Pritchard (2010: 77–79).
House Fire: Imagine that Campbell comes home to find his house on fire. The
fire department is already on the scene. Campbell sees a number of people
dressed in fire protective gear. He approaches one of these people and inquires
about the cause of the house fire. He receives testimony from the fire official that
the fire was caused by faulty wiring. Campbell then forms the true belief that the
fire was caused by faulty wiring. However, unbeknownst to Campbell, the other
people he saw dressed in fire gear were not fire fighters but people on their way
to a firefighter themed costume party. Campbell’s belief that the fire was caused
by faulty writing could easily have been false, because in close possible worlds
Campbell did not ask a real fire official but an unreliable party goer.
House fire is a classic of case of environmental luck. Campbell finds himself

in a bad epistemic environment, and as such does not have a safe belief. Notice
that this case is also vector-relative. Campbell’s belief is unsafe because he finds
himself in an epistemic network of mostly party-goers disposed to give false or
unreliable testimony.
In the classic fake barn county case (Goldman, 1976), the bad epistemic envi-
ronment is not specifically a bad social epistemic environment, so the vector-
relative aspect of safety does not arise. However, in testimonial cases it is useful
to think of modal epistemic standings as vector-relative to better capture and
diagnose epistemic success and failure. In this regard, we are not suggesting a
replacement for basis-relative safety. We are suggesting that in social epistemic
cases the basis of one’s belief is essentially dependent on the structure of one’s
epistemic network. Furthermore, as we argue in Section 8.4, understanding safety
as vector-relative uncovers steps agents can take to place themselves in a better
epistemic position and cultivate the virtues needed (and avoid related vices) in a
truly interconnected social epistemic network.
8.3 Epistemic security

Vector-relativized safety is a modal standing that captures someone’s epistemic
position given the current structure of her epistemic network. We hold the net-
work structure fixed when considering counterfactual possibilities of false belief.
We argued above that thinking of safety in this way better captures epistemic suc-
cess and failure in a truly social testimonial network.
Once we accept the usefulness of a modal epistemic standing that keeps the
network structure fixed, it’s natural to think about another modal epistemic stand-
ing that considers close possible worlds of (slightly) different network structures.
Call this epistemic security. Unlike safety, which tracks whether one is currently
in a good social epistemic environment, epistemic security tracks how robust this
environment is to changes. This is especially important in cases in which there
are members of the social epistemic community intentionally working to deceive,
mislead, and manipulate. There are two notions of epistemic security: the security
of one’s network, and the security of one’s belief.
Belief-security:
S believes that p securely within epistemic network N iff there is no close

possible world in which S falsely believes that p in N′, where N′ is a network
that can be generated by adding or deleting a small number of testimonial
links in N.
Network-security:
S is in a secure epistemic network N iff S’s epistemic well-being in N does

not depend on a small number of independent sources and there is no close
possible world in which S’s epistemic well-being depends on a small number
of sources in N′, where N′ is a network that can be generated by adding or
deleting a small number of testimonial links in N.
First consider belief-security. While all epistemically secure beliefs are safe, not
all safe beliefs are secure. Mixed cases involving safety but not security occur
when just a few small changes to the geometry of the testimonial network would
result in the agent no longer believing the true proposition in question. For exam-
ple, consider a case in which S receives testimony about p from three other agents:
A, B, and C. In the actual world, A and B truly testify that p, whereas C falsely tes-
tifies that ~p. Now consider a counterfactual scenario in which C testifies to both
A and B that ~p before they have a chance to speak to S. In light of C’s testimony,
both A and B abandon their belief in p and so are not inclined to testify that p to
S. In this nearby possible world, the only testimony that S receives indicates that
~p. Thus, while her belief is safe (i.e., true in close possible worlds holding fixed
the network structure), it is not secure.
On the other hand, network-security considers network structure only. We
abstract away from the exact testimony and beliefs of the agents in question. This
means that occupying a more secure network may increase the likelihood of true
beliefs in the future, but it may not. If agents happen to be clued in to reliable and
trustworthy sources—even if the network is insecure—they will still fare well.
However, if an agent is in a network with several independent sources, but none
of them are trustworthy, then she will not fare well, despite the network-security.
Just as safety is a modal epistemic standing that captures how luck is incom-
patible with knowledge, epistemic security captures the way that the structure of
one’s network, and the beliefs one forms in that network, should be immune to
luck and resilient to bad actors. It shouldn’t just be a matter of happenstance that
someone gains true beliefs given the network structure that she occupies. The
epistemic value of belief-security should be intuitive. Belief-security concerns the
likelihood of someone maintaining a true belief in the face of network changes.
However, the value of an epistemically secure network may strike some as unin-
tuitive. Thus, in this paper we mostly develop a case for network-security.
An epistemically secure network is one where the epistemic well-being of
an agent isn’t dependent on the epistemic good-will of one or just a few other
agents. It is our contention that epistemic network-security is worth aiming for
not only if it increases the likelihood of true beliefs, but also because secure net-
works promote epistemic growth and epistemic autonomy. To see why, consider
a non-epistemic case. Kant (1797) in the Metaphysics of Morals (MM) argues
that political rights are necessary for a person’s freedom and autonomy to be
respected. Without these rights, the well-being of someone is dependent on the
good-will of others. Korsgaard (2012: 2) puts the point nicely:
[Kant] argued that without the institution of enforce-able legal rights, our
relationships with each other must be characterized by the unilateral domina-
tion of some individuals over others. The problem is not, or not merely, that
the strong are likely to tyrannize over the weak. Even if the strong were scru-
pulous about not interfering with the actions or the possessions of the weak,
still, without rights, the weak would be able to act on their own judgment and
retain their own possessions only on the sufferance of the strong (MM 6:312).
Since her innate right to freedom is violated when one person is dependent
on some other person’s good-will, Kant thinks it is a duty, and not just a con-
venience, for human beings to live in a political state in which every person’s
rights are enforced and upheld (MM 6:307–8).
In the case where the leader of the society has good-will, rights will not make a
practical difference to those in the society. However, their well-being is still less
secure. In the close possible worlds where the leader is replaced, or the strong
have a change of heart (for the worse), then the weak have no protections. So even
if political security does not make a practical difference in day-to-day lives, it is
still valuable. It is a matter of luck, or happenstance, that each person’s well-being
is respected.
We find analogous results in the epistemic case. If someone is in an insecure
testimonial network but the source he relies on is reliable and accurate, the lack of
security is epistemically problematic. The agent lacks a clear sense of epistemic
autonomy, with little opportunity to grow epistemically. Moreover, in close pos-
sible worlds where the source is not reliable or has a change of heart, the hearer of
testimony loses whatever good epistemic position he had. Figure 8.1a shows such
an insecure testimonial network (from the perspective of any of the outer nodes).
In this star-network the center node (the source node) is the only source the outer
nodes (receivers of testimony) are drawing on. Each outer node is dependent on
the center to supply accurate and truthful information (Alfano 2016; Freeman
1978). Indeed, star-networks are associated with a number of problematic and
harmful practices. Sexual predators and their targets often form a star-network,
with the predator at the center and the victims on the points of the star. This keeps
the victims from effectively communicating with one another, and coordinating
or cooperating against the predator (Fire, Katz, Elovici 2012). Star-networks are
also associated with financial fraud (Šubelj, Furlan, Bajec 2011), academic fraud
(Callaway 2011), and terrorist activities (Reid et al. 2005; Krebs 2002).
Making small adjustments to the network where the outer nodes themselves
become directly connected (Figure 8.1b), makes the center node’s network more
secure. The outer nodes can confer with each other on the merits of the center
node’s testimony. Epistemic well-being is no longer dependent solely on one
other person.
(a) (b)
Figure 8.1 (a) Star-network. (b) More secure network
In more connected networks, the structures become more complicated, and

it may take some effort for an agent to be aware of the level of network-security
she has. Figure 8.2 shows an example of an insecure network that at first may
seem secure (Alfano et al. 2018). Imagine that Sana (represented by the white
node) finds herself in an epistemic network consisting of four different people to
whom she is directly connected. When she seeks information about some topic,
she enters a conversation with each person separately. She finds that every time
the same three people (represented by black nodes) always give the same answer.
Sometimes their answer is the same as that provided by the fourth person (rep-
resented by the grey node), sometimes it isn’t. According to Sana’s immediate
perspective, it seems as though she is in a secure network with four independ-
ent sources that help her to converge on the truth. However, when we zoom out
we see that the three people who always agree are actually just passing on the
information from one single source. So in reality Sana is drawing from only two
sources, not four. Her position is less secure than she thought. Three of the people
she talks with are not genuine sources of information, but mere conduits. They are
simply passing along information. Only when Sana monitors the structure of her
network with the goal of increasing her network-security does it become transpar-
ent that she should seek out more independent sources to guard against misleading
and false information (or just overly amplified information).4
Cases like the one modeled by Figure 8.2 seem to be increasingly common in
today’s social media environment. Large platforms such as Facebook and Twitter
specialize in amplifying viral content, which makes it seem more prevalent (and
thus more likely to be true) than it really is. Moreover, organized groups of trolls
(e.g., on 4chan and 8chan), advertising firms, and political consultancies make
efforts to hijack the amplification process and spread their preferred messages. To
do so, they manipulate the structure of online testimonial networks. This process
is essentially one of undermining network-security for the sake of some other
Figure 8.2 Insecure network
motive (the lulz, profit, or desired political outcomes, respectively). Only when
we have a clear grasp of what network-security is and why it is valuable does this
phenomenon make sense.
Taking a network perspective on how we access information from a commu-
nity of information sharers, we see that there are more than two possible epis-
temic roles in testimonial exchanges. There are receivers (those who listen to
information), sources (those who are the primary sources of information), and
conduits (those who pass on information from sources) (Sullivan et al. 2020).5
The line between a conduit and a source is not always clear cut. Someone who
passes along information he heard from another source can be more or less reflec-
tive before passing the information along. Someone who brings their own back-
ground knowledge to bear on the information, and engages in an independent
check before sharing, more closely resembles a source than a conduit. For the
sake of this paper, we treat conduits as simply passing along information in a
minimally reflective way, such that the conduit could reasonably “pass the buck”
to the original source.6 We expect that someone who brings their own relevant
background knowledge to bear on an issue, or who carries out an independent
check, could not reasonably pass the buck. Importantly, tracking network-security
requires monitoring the structure of one’s network as well as the epistemic roles
(receiver, conduit, source) that individuals in the network play. This allows agents
to restructure their networks, making them more resilient. It also enables them to
maintain intellectual autonomy, and not to depend overmuch on the epistemic
good-will of others.
That said, it’s worth addressing an important disanalogy between Kant’s politi-
cal case and the epistemic case. There is no authority upholding “epistemic rights”
in the epistemic case.7 So in the end we are all dependent on the larger epistemic
community. This is made salient in Pritchard’s house fire case. Campbell is in a
relatively secure network given the number of sources at his disposal, but it is still
the case that many of these sources, if they are disposed to mislead, could prohibit
Campbell from gaining knowledge. So while the structure of the network qua struc-
ture is secure, there is still a lack of safety and belief-security. This highlights the
point that network-security is just one of several features that need attending to in
social epistemic networks. It also suggests that there are trade-offs between security
and safety. Thus, part of acting virtuously in a social network is navigating these
trade-offs in a way that does not lead to vicious epistemic behavior. In the next sec-
tion we consider the virtues and vices involved in negotiating the trade-offs within
insecure epistemic environments and the trade-offs between security and safety.
8.4 Virtue and vice in social epistemic networks

Several epistemic virtues and vices are relevant in social epistemic environments.
Many of these virtues are already widely discussed, such as open-mindedness
(Riggs 2010). However, if what we have been saying about social epistemic
communities is on the right track, maintaining knowledge in a social epistemic
environment requires distinct considerations.8 We want to draw attention to three
classes of virtues that only arise in social epistemic networks: monitoring, adjust-
ing, and restructuring. To our knowledge, these dispositions have been almost
entirely neglected by virtue epistemologists, who tend to favor a very individual-
istic approach. As we will see below, all three types of dispositions have similar
structures to more traditional epistemic virtues. They involve sub-dispositions
related to attention, motivation, cognition, and so on. In addition, as we shall
argue, they are scaffolded on one another. One can only embody an effective
adjusting or restructuring virtue if one is sufficiently adept at monitoring. In short,
monitoring makes one alert to imperfections in the structure of one’s epistemic
network, and those imperfections can be addressed either by leaving the structure
intact while modulating one’s trust or credence in various sources, or by altering
the structure itself. In addition, given that these virtues are operative in a social
epistemic environment, they may be both self-regarding and other-regarding.
Moreover, vices, such as dogmatism, in the social epistemic context are mani-
fested through vicious monitoring, adjusting, and restructuring.
8.4.1 Monitoring
In order to benefit from the knowledge embodied in one’s social network, one
should monitor and understand the structure of that network. This applies to both
safe and secure belief. In milk carton, Sonny’s belief is safe because of the struc-
ture of his network. Sonny would do well to monitor this structure so he can
be attuned to any actions he could take to improve his epistemic position, and
whether he should believe the testimony of Mary or seek further testimony. It is
only through monitoring this structure that these considerations come to the fore.
While monitoring the structure of the network is important, one must also
monitor the epistemic roles and track-records of those in the network. Do my
sources have a reliable track-record, or do they often provide false or misleading
information? Are my sources independent, or are they conduits simply amplifying
the messages of others? In the former case, I may be able to benefit from the wis-
dom of crowds, as the Condorcet Jury Theorem and related proofs indicate (List
2001; Masterton et al. 2016; Sullivan et al. 2020). In the latter, I may not. Even
though monitoring the epistemic roles and track-records of others goes beyond the
structure of the network, it is only through contextualizing these sources within a
network structure that I am able to shape current and future epistemic behavior.
That said, the way modal standings like safety help shape future behavior is not
always obvious. Gardiner (2017), for example, criticizes current conceptions of
safety because they do not allow agents to actively select for safe belief. According
to Gardiner, since safety is an external condition on knowledge and only con-
cerned with counterfactual properties of a specific belief, it does not affect future
beliefs. Furthermore, since safety concerns only nearby possible worlds, select-
ing for safe belief in the actual world is not possible. Instead, Gardiner argues,
safety simply tracks whether someone currently is in good epistemic environment,
without guiding future behavior. However, by understanding safety (and security)
as vector-relative these worries fall away. Monitoring, when done virtuously, is
explicitly attuned to locating epistemic opportunities and threats in one’s network.
Monitoring requires seeing the structure of one’s network as signaling possible
network improvements and signaling how to weigh differing testimony both in the
actual world and nearby counterfactuals. Considering epistemic sources apart from
the network context, and considering whether one’s belief is safe apart from net-
work context, leaves one vulnerable to epistemically vicious dispositions and hab-
its of behavior, such as dogmatism and closed-mindedness. For example, without
monitoring the structure of my network, I may be in an epistemic echo-chamber
that, while appearing epistemically diverse, actually cuts me off from potential
knowledge today and tomorrow. Thus, monitoring virtues concern the way that
agents actively keep track of their epistemic position in various domains and con-
texts, and why keeping track of this position is epistemically beneficial.
Monitoring virtues can also be other-regarding. I can benefit others by recom-
mending sources to them, or telling them to stop listening to certain sources to
increase their safety and security. But I can only do this if I monitor the structure
of their social networks, the track-record of these sources, and the epistemic roles
of those in their network. Moreover, monitoring can be done with epistemically
malevolent or benevolent motivations; one can monitor with the aim of finding
ways to improve another person’s epistemic position or to undermine it.
Consider a case of malevolent monitoring. One of Baehr’s (2010) examples
of personal epistemic malevolence seems especially relevant. Baehr asks us to
consider how Frederick Douglass was treated with epistemic malevolence by his
owners. The patriarch of the house, Tom, upon discovering his wife Sofia teach-
ing Douglass how to read, actively worked to sever their ties. Tom was monitor-
ing the structure of Douglass’s epistemic network with epistemically malevolent
motivations. He wanted to undermine Douglass’s epistemic well-being. This was
only possible through vicious monitoring.
Vicious monitoring can also be epistemically negligent, instead of malevolent.
Online social media platforms, such as Facebook and Twitter, currently embody
this type of vicious monitoring. Both platforms actively monitor the epistemic
network of each user. However, their monitoring is not done to increase the
epistemic well-being of the users; instead monitoring is guided solely by the profit
motive. Concern for epistemic values is absent. Moreover, the monitoring of these
platforms is arguably done in a vicious way, as seen by the privacy violations
associated with the Cambridge Analytical scandal (Timberg et al. 2018).
Monitoring in a virtuous way with virtuous aims takes effort. It involves
weighing different values, not just maximizing true and minimizing false beliefs.
The values of privacy and epistemic autonomy, in addition to truth, are especially
salient in this context.
8.4.2 Adjusting
One reason for monitoring one’s network is to be able to know how to calibrate
and adjust the weight one should give to sources and pieces of information spread
throughout the network. Every real social epistemic network is imperfect, at least
to some extent. If I manage to monitor the structure of my own network suffi-
ciently well, I may be able to adjust my credences to account for its imperfections.
The monitoring virtue is thus conceptually prior to the adjusting virtue. And the
two are distinct. In principle, I could monitor the structure of my testimonial net-
work adequately without being disposed to take into account the imperfections I
identify when updating my beliefs. Likewise, I could monitor the epistemic track-
records of my sources adequately without being disposed to distrust those who
have proven themselves unreliable. Thus, adjusting virtues govern how someone
should utilize the knowledge she gained from monitoring her network’s structure,
sources’ track-records, and the epistemic roles of the agents in her network. For
example, if someone is in a network structured in a way similar to Figure 8.2, she
should not ascribe more weight to the information coming from the conduits just
because the information is repeated by more people. Instead, she should consider
the reliability of their source node (the black node on the far left). Information
distributed in a secure network should be weighted differently, as compared to
information distributed in an insecure network. Thus, as with monitoring virtues,
it is the structure of the network that gives shape to adjusting virtues. It is not
possible to virtuously weigh the testimony of someone within a social network
without considering the structure of the network. Failure to do so will preclude
knowing whether someone is a conduit or a source, which in turn leads to failure
in correctly assessing whether one’s belief is safe or secure.
Adjusting virtues can also be other-regarding. I may be able to benefit others
by suggesting that they put more or less trust in various sources located in their
social epistemic network. Contrariwise, I may be able to harm them epistemically
by making opposite suggestions. The ability to do so depends on other-regarding
monitoring dispositions, but exercising that ability (ir)responsibly is its own epis-
temic virtue or vice. Failing to adjust one’s beliefs (and not suggesting adjust-
ments to others) based on the structure of the network not only risks developing
adjusting vices, but risks developing more deep-seated vices like dogmatism and
close-mindedness. In a social epistemic environment, networked vices are inti-
mately connected to and give rise to other vices.
8.4.3 Restructuring
While all real social epistemic networks are imperfect, sometimes they are so
flawed that they need to be modified. Networks can (to some extent) be rewired.
This could involve seeking out new sources, no longer listening to sources one
had previously trusted, or effecting more distal changes in the structure of the net-
work. Doing this well depends on sufficiently successful monitoring, recognition
that attempts to adjust credences are not up to the task, and the motivation and
capacity to identify efficient and effective changes that one has the power to enact.
The latter dispositions are components of restructuring as self-regarding social
epistemic virtues. And as with the other dispositions in this taxonomy, one could
embody correlative vices instead of virtues. One could, for instance, be disposed
to cut oneself off from reliable testifiers, plug oneself into networks that amplify
fake news and conspiracy theories, and so on.
How exactly do and should we navigate the process of restructuring our own
and others’ testimonial networks? Levy (2017) argues that you should cut yourself
off completely from sources of fake, misleading, and unreliable news. He cites a
wealth of psychological studies that suggest that humans are easily persuaded by
false information, even if they know full well that the information is false. Humans
have cognitive biases such that we tend to misremember sources of information
and believe fictions (e.g., Marsh et al. 2016; Prentice et al. 1997; Wheeler et al.
1999). Levy argues that even being exposed to false information leaves us vulner-
able to acquiring false beliefs. Through the lens of our framework, Levy values
safety and belief-security over network-security: we should limit the number of
sources to only those that reliably provide good information. While this might
seem like a restructuring virtue, we want to suggest that, for several reasons, such
behavior can actually manifest as a restructuring vice.
First, by cutting myself off from an untrustworthy source, I end up more depend-
ent on the remaining sources I do trust. Thus, by making myself less vulnerable
against that untrustworthy source, my network becomes less secure, and I become
more vulnerable to my remaining sources. This can become a problem. Sources can
change slowly over time with respect to how reliable, independent, and epistemi-
cally well-intentioned they are. As a real-world example, small local media compa-
nies in the United States are undergoing a takeover by a single company, Sinclair
Media, that has a clear ideological agenda (Stelter 2018). Safeguarding network-
security by keeping many different types of sources in one’s network can better
guard against this type of epistemic takeover. Furthermore, it provides agents with
the opportunity for epistemic growth, despite the vulnerability to false informa-
tion. Engaging even with propaganda can serve to develop better epistemic skills,
if the propaganda serves as a negative epistemic exemplar. It is possible to learn
what to avoid and how to spot similar but different bad epistemic behavior in future
instances from bad epistemic examples (Alfano 2013; Sullivan and Alfano 2019).
Second, limiting any engagement with unreliable sources lessens my poten-
tial to develop other-regarding restructuring virtues. Other-regarding restructur-
ing virtues involve being disposed to help others rewire their trust (and distrust)
networks so that they are epistemically better off and less vulnerable. However,
if I am overly concerned with limiting my own exposure to false and unreliable
information, then I will be unable to advise others how to better their networks.
In order to help others, I need to monitor their networks. Part of this monitoring
will include exposing myself to false and misleading information. This is not to
say there is never a source that you should sever ties with. Instead, we contend
that the solution to false and intentionally misleading news is not a divide and
conquer strategy. One should take a more encompassing view of what it means to
do well epistemically. A single-minded concern for the truth of one’s own beliefs,
neglecting network-security, can cut one off from other epistemic desiderata, and
from developing important other-regarding epistemic virtues. What Levy and
others who advocate cutting oneself off from fake news and related phenomena
neglect is that there is a sort of collective action problem here: single-mindedly
focusing on the verisimilitude of my own beliefs may lead me to neglect the epis-
temic well-being of my community.
Lastly, it is imperative to discuss other-regarding restructuring vices. Getting
other people to stop trusting reliable sources and to plug themselves into ampli-
fiers of fake news and conspiracy theories is a restructuring epistemic vice. As
discussed alongside Figure 8.1a, this practice is often employed by sexual harass-
ers and abusers, and perpetrators of financial and academic fraud. Such actors
seek to make others epistemically dependent on them. This can be done with
malevolent intentions, or it may not. Cases with malevolent intentions (including
those already mentioned) are easy to see. In the case of Fredrick Douglass, Tom
actively restructured Douglass’s epistemic network for the worse. Tom limited
the potential for Douglass’s epistemic growth and intellectual autonomy by sever-
ing ties between Douglass and his epistemic informants.
Even someone who has well-intentioned motivations and has access to the
truth, but reduces others’ network-security, making others epistemically depend-
ent on them (even for their own good), displays vicious behavior. Plato’s philoso-
pher king is a perfect example. The philosopher king keeps the public cut off from
art and fiction, tells the public untruths, but all for the sake of their own epistemic
well-being. However, despite Plato’s enticing epistemic arguments, we contend
that this too is epistemically vicious. It reduces network-security. It makes people
less intellectually autonomous and less able to enjoy epistemic growth.
8.5 Conclusion
In this paper we argued that social epistemology needs to expand its toolkit to
include modal epistemic standings and epistemic virtues and vices as networked
concepts. The structure of one’s epistemic network gives shape to the related vir-
tues and the nature of modal epistemic standings. Conceptualizing modal epis-
temic standings and virtues as inherently networked allows us to see that safe
belief is relative to one’s surrounding network, and that network-security (and
belief-security) is just as important. Not only should our beliefs be safe in the cur-
rent network, but also secure in nearby possible networks.
We also introduced three classes of virtues (and correlative vices) that allow us
to navigate the social epistemic realm: monitoring, adjusting, and restructuring.
These virtues can be both self-regarding and other-regarding. They govern the
trade-offs between safety and belief-security and network-security, and are mani-
fested in different ways depending on the structure of one’s epistemic network.
Much more needs to be done to develop and find the correct balance between
safety, belief-security, and network-security and between self-regarding and
other-regarding considerations. What we have done here is take the first step in
articulating a framework within which to expand virtue epistemology and to make
social epistemology truly social.
Notes
1 Some exceptions include discussions about what it means for groups to testify
(Tollefsen 2007; Lackey 2015, 2018) and discussions about which network struc-
tures are conducive to sharing knowledge within idealized communities of scientific
researchers (Holman and Bruner 2015; Rosenstock et al. 2017; Zollman 2007).
2 For a socialized alternative, see Alfano and Skorburg (2017b) and Skorburg (2018).
3 Goldberg (2005) argues that milk carton suggests that there are local invariances in
one’s environment that should be held fixed in determining whether someone has a safe
belief. Our notion of vector-relative safety is in the spirit of Goldberg’s account, but it
is more specifically tied to one’s social epistemic network.
4 See Kelly (2010) for a related discussion of independence. Kelly only considers cases
in which someone is prompted to revise a belief based on disagreement with multiple
(in)dependent testifiers. The more fundamental question that we address here is what a
desirable testimonial network looks like in the first place.
5 A further interesting question is whether a group or network of individuals could itself
count as a source. For example, when one reads The New York Times, one could treat
the individual journalist whose byline appears with the article as the source of one’s
knowledge. Alternatively, one could treat the Times itself as one’s source. We suspect
that most groups are not sufficiently organized and structured to count as epistemic
agents with the power to testify in their own right. However, this does not preclude
certain groups from counting as epistemic agents and hence as sources. For more on
these ideas, see Tollefsen (2007) and Fricker (2012).
6 For a discussion about buck passing and testimony, see Baker and Clark (2018).
7 There are laws against fraud, false advertising, and other types of crimes that have an
epistemic dimension. However, since free speech is a cornerstone of democracy, the
scope of epistemic rights recognized by the state might be more limited than the scope
of political rights. For an account of epistemic rights, see Watson (2018).
8 One might wonder whether these virtues are necessary conditions for one’s true tes-
timonial beliefs counting as knowledge. We are agnostic on that question here. At the
very least, these dispositions are epistemically valuable, even if they are not necessary
for knowledge.
References
Alfano, M. 2013. Character as Moral Fiction. Cambridge: Cambridge University Press.
Alfano, M. 2016. “The Topology of Communities of Trust.” Russian Sociological Review,
15(4), 30–56.
Alfano, M., Cunningham, S., Meulemans, W., Rutter, I., Sondag, M., Speckmann, B., &
Sullivan, E. 2018. “Social Network-Epistemology.” In 2018 IEEE 14th International
Conference on e-Science (e-Science), 29 Oct–1 Nov 2018, 320–321.
Alfano, M., & Skorburg, A. 2017a. “Extended Knowledge, the Recognition Heuristic, and
Epistemic Injustice.” In D. Pritchard, J. Kallestrup, O. Palermos, & A. Carter (eds.),
Extended Knowledge. Oxford: Oxford University Press, pp. 239–65.
Alfano, M., & Skorburg, A. 2017b. “The Embedded and Extended Character Hypotheses.”
In J. Kiverstein (ed.), Handbook of Philosophy of the Social Mind. London: Routledge,
pp. 465–78.
Baehr, J. 2010. “Epistemic Malevolence.” Metaphilosophy, 41(1–2), 189–213.
Baker, J., & Clark, P. 2018. “Epistemic Buck-Passing and the Interpersonal View of
Testimony.” Canadian Journal of Philosophy, 48(2), 178–99.
Battaly, H. 2008. “Virtue Epistemology.” Philosophy Compass, 3(4), 639–63.
Callaway, E. 2011. “Report Finds Massive Fraud at Dutch Universities.” Nature, 479, 15.
Fire, M., Katz, G., & Elovici, Y. 2012. “Strangers Intrusion Detection: Detecting Spammers
and Fake Profiles in Social Networks Based on Topology Anomalies.” Human Journal,
1(1), 26–39.
Freeman, L. 1978. “Centrality in Social Networks Conceptual Clarification.” Social
Networks, 1(3), 215–39.
Fricker, M. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford
University Press.
Fricker, M. 2012. “Group Testimony? The Making of a Collective Good Informant.”
Philosophy and Phenomenological Research, 84(2), 249–76.
Gardiner, G. 2017. “Safety’s Swamp: Against the Value of Modal Stability.” American
Philosophical Quarterly, 54(2), 119–29.
Goldberg, S. 2005. “Testimonial Knowledge Through Unsafe Testimony.” Analysis, 65,
302–11.
Goldberg, S. 2007. “How Lucky Can You Get?” Synthese, 158, 315–27.
Goldman, A. I. 1976. “Discrimination and Perceptual Knowledge.” The Journal of
Philosophy, 73(20), 771–91.
Greco, J. 2010. Achieving Knowledge: A Virtue-Theoretic Account of Epistemic
Normativity. Cambridge: Cambridge University Press.
Holman, B., & Bruner, J. P. 2015. “The Problem of Intransigently Biased Agents.”
Philosophy of Science, 82(5), 956–68.
Kant, I. 1797. The Metaphysics of Morals. Trans. and ed. Gregor Mary. Cambridge
Texts in the History of Philosophy series. Cambridge: Cambridge University Press,
1996.
Kawall, J. 2002. “Other–Regarding Epistemic Virtues.” Ratio, 15(3), 257–75.
Kelly, T. 2010. “Peer Disagreement and Higher-Order Evidence.” In R. Feldman & T.
Warfield (eds.), Disagreement. Oxford: Oxford University Press, pp. 111–74.
Kelp, C. 2009. “Knowledge and Safety.” Journal of Philosophical Research, 34, 21–31.
Korsgaard, C. M. 2012. “A Kantian Case for Animal Rights.” In M. Michael, D. Kühne, &
J. Hänni (eds.), Animal Law-Tier and Rect: Developments and Perspectives in the 21st
Century. Zurich: Dike Verlag, 3–27.
Krebs, V. 2002. “Mapping Networks of Terrorist Cells.” Connections, 24(3), 43–52.
Lackey, J. 2018. “Group Assertion.” Erkenntnis, 83(1), 21–42.
Lackey, J. A. 2015. “A Deflationary Account of Group Testimony.” In J. Lackey (ed),
Essays in Collective Epistemology. Oxford: Oxford University Press, pp. 64–95.
Lackey, Jennifer. 2008. Learning from Words: Testimony as a Source of Knowledge.
Levy, N. 2017. “The Bad News About Fake News.” Social Epistemology Review and
Reply Collective, 6(8), 20–36.
List, C., & Goodin, R. E. 2001. “Epistemic Democracy: Generalizing the Condorcet Jury
Theorem.” Journal of Political Philosophy, 9(3), 277–306.
Marsh, E. J., Cantor, A. D., & Brashier, N. M. 2016. “Believing that Humans Swallow
Spiders in Their Sleep: False Beliefs as Side Effects of the Processes that Support
Accurate Knowledge.” Psychology of Learning and Motivation, 64, 93–132.
Masterton, G., Olsson, E. J., & Angere, S. 2016. “Linking as Voting : How the Condorcet
Jury Theorem in Political Science is Relevant to Webometrics.” Scientometrics, 106(3),
945–66.
Montmarquet, J. A. 1992. Epistemic Virtue and Doxastic Responsibility. Lanham, MD:
Rowman & Littlefield.
Nozick, R. 1981. Philosophical Explanations. Oxford: Oxford University Press.
Pelling, C. 2013. “Testimony, Testimonial Belief, and Safety.” Philosophical Studies,
164(1), 205–17.
Prentice, D. A., Gerrig, R. J., & Bailis, D. S. 1997. “What Readers Bring to the Processing
of Fictional Texts.” Psychonomic Bulletin & Review, 4(3), 416–20.
Pritchard, D. 2007. “Anti-Luck Epistemology.” Synthese, 158, 277–98.
Pritchard, D. 2009. “Safety-Based Epistemology: Whither Now?” Journal of Philosophical
Research, 34, 33–45.
Pritchard, D. 2010. “Knowledge and Understanding.” In Pritchard, Haddock, and Miller,
(eds.), The Nature and Value of Knowledge: Three Investigations. Oxford: Oxford
University Press, pp. 3–90.
Pritchard, D. 2016. “Epistemic Risk.” Journal of Philosophy, 113(11), 550–71.
Rabinowitz, D. 2011. “The Safety Condition for Knowledge.” Internet Encyclopedia of
Philosophy.
Reid, E., Qin, J., Zhou, Y., Lai, G., Sageman, M., Weimann, G., & Chen, H. 2005.
“Collecting and Analyzing the Presence of Terrorists on the Web: A Case Study of
Jihad Websites.” In P. Kantor, G. Muresan, F. Roberts, D. D. Zeng, F. Y. Wang, H. C.
Chen, & R. C. Merkl (eds.), Intelligence and Security Informatics. Dorcrecht: Springer,
402–11.
Riggs, W. 2010. “Open-Mindedness.” Metaphilosophy, 41(1–2), 172–88.
Rosenstock, S., Bruner, J., & O’Connor, C. 2017. “In Epistemic Networks, Is Less Really
More?” Philosophy of Science, 84(2), 234–52.
Skorburg, J. A. 2018. “Where Are Virtues?” Philosophical Studies, 176(9), 2331–49.
Sosa, E. 1999a. “How to Defeat Opposition to Moore.” Philosophical Perspectives, 13,
141–54.
Sosa, E. 1999b. “How Must Knowledge be Modally Related to What is Known?”
Philosophical Topics, 26(1–2), 373–84.
Sosa, E. 2007. A Virtue Epistemology: Apt Belief and Reflective Knowledge (Vol. 1).
Stelter, B. 2018. “Sinclair Made Dozens of Local News Anchors Recite the Same Script.”
The New York Times, 2 April.
Šubelj, L., Furlan, U., & Bajec, M. 2011. “An Expert System for Detecting Automobiles
Insurance Fraud Using Social Network Analysis.” Expert Systems Applications, 38(1),
1039–52.
Sullivan, E., & Alfano, M. 2019. “Negative Epistemic Exemplars.” In B. Sherman & S.
Goguen (eds.), Overcoming Epistemic Injustice: Social and Psychological Perspectives.
Lanham, MD: Rowman & Littlefield, pp. 17–32.
Sullivan, E., Sondag, M., Rutter, I., Meulemans, W., Cunningham, S., Speckmann, B.,
& Alfano, M. 2020. “Can Real Social Epistemic Networks Deliver the Wisdom of
Crowds?” In T. Lombrozo, J. Knobe, & S. Nichols (eds.), Oxford Studies in Experimental
Philosophy (Vol. 3). Oxford: Oxford University Press, pp. 29–63.
Timberg, C., Dwoskin, E., Zapotosky, M., & Barrett, D. 2018. “Facebook’s Disclosures
Under Scrutiny as Federal Agencies Join Probe of Tech Giant’s Role in Sharing Data
with Cambridge Analytica.” The Washington Post, 2 July.
Tollefsen, D. 2007. “Group Testimony.” Social Epistemology, 21(3), 299–311.
Watson, L. 2018. “Systematic Epistemic Rights Violations in the Media: A Brexit Case
Study.” Social Epistemology, 32(2), 88–102.
Wheeler, C., Green, M. C., & Brock, T. C. 1999. “Fictional Narratives Change Beliefs:
Replications of Prentice, Gerrig, and Bailis (1997) with Mixed Corroboration.”
Psychonomic Bulletin & Review, 6(1), 136–41.
Williamson, T. 2002. Knowledge and its Limits. Oxford: Oxford University Press.
Zagzebski, L. T., & Zagaebski, L. T. 1996. Virtues of the Mind: An Inquiry into the
Nature of Virtue and the Ethical Foundations of Knowledge. Cambridge: Cambridge
University Press.
Zollman, K. J. 2007. “The Communication Structure of Epistemic Communities.”
Philosophy of Science, 74(5), 574–87.
Part III
Analyses of specific vices

9 Quitting, procrastinating,
and slacking off
Heather Battaly
Compare four sets of people, all of whom have intellectual goals and all of whom
encounter obstacles to achieving their goals. Members of the first set tend to aban-
don their intellectual goals when they hit an obstacle. They drop a class after
failing the first homework assignment. They scrap an idea for a project as soon
as they hear objections. They forsake a line of inquiry when it becomes boring.
They abandon their goals because they don’t want to put in the effort required to
achieve them. When the going gets tough, they quit.
Members of the second set don’t abandon their intellectual goals when they
hit an obstacle. Instead, they delay, even though they think they should act. They
delay starting the final paper for the course until the night before the deadline.
They postpone working on the overdue book project. They put off grading papers
until they receive email warnings from the Dean’s office. They typically feel
awful about delaying, but, nevertheless, delay because they don’t want to put in
the effort. When the going gets tough, these folks procrastinate.
Members of the third set are neither quitters nor procrastinators. When they hit
an obstacle, they don’t abandon their intellectual goals. Nor do they delay against
their better judgment. Instead, they avoid putting in effort in a variety of other
ways. Some don’t delay at all; they move forward but avoid obstacles by skipping
steps, submitting shoddy reports. Others deliberately delay and waste time, think-
ing it is fine to drag out straightforward assignments. Yet others simply do noth-
ing: consider free-riders in team projects who surf on-line while everyone else
works. Though this set is diverse, all of its members have something important in
common: they intentionally take the easy route, knowing (or at least truly believ-
ing) their work will suffer. When the going gets tough, they slack off.
In contrast with all three of the above groups, members of the fourth set put in
effort to overcome the obstacles they encounter and stick with their intellectual
goals. They stay in class and work hard to try to pass the next assignment. They
write a rough draft of their paper well before the deadline. They work on the book
project, and the grading. They try to overcome obstacles, neither avoiding them
by skipping steps nor succumbing to them by doing nothing. When the going gets
tough, they persevere.
This chapter focuses on the traits of quitting, procrastinating, and slacking off,
which are different ways to lack the trait of intellectual perseverance. It explores
168 Heather Battaly
what makes the traits of quitting, procrastinating, and slacking off intellectual
vices when they are. But it does not assume that these traits are always intellec-
tual vices, and even suggests some conditions in which they might be intellectual
virtues.
The first section draws on recent work in virtue epistemology to provide an
account of intellectual perseverance (Battaly 2017; King 2014; King 2019). It
distinguishes between the trait of intellectual perseverance and the virtue that
goes by the same name, arguing that the trait of intellectual perseverance is not
a virtue when one has it to excess—roughly, when one doesn’t know when to
quit. Sections 9.2 through 9.4 propose working definitions of the traits of quitting,
procrastinating, and slacking off, as deficiencies of, or ways of lacking, the trait of
intellectual perseverance. I examine why these traits are intellectual vices when
they are, while leaving open the possibility that they sometimes fail to be intel-
lectual vices and might even be intellectual virtues. The final section canvasses
a number of related traits including laziness, apathy, complacency, resignation,
and folly.
9.1 Intellectual perseverance

Let’s begin with an analysis of the trait of intellectual perseverance, without
presupposing that this trait is automatically an intellectual virtue. There are
advantages to keeping these questions separate—to separating our analysis of
intellectual perseverance as a character trait from our investigation into its status
as an intellectual virtue. This approach can help us home in on what makes intel-
lectual perseverance a virtue when it is one, while leaving open the possibility
that it can fail to be an intellectual virtue. It is an example of what Ian James Kidd
(2020) calls ‘normative contextualism.’ Normative contextualists initially con-
ceive of intellectual character traits as normatively neutral, and then investigate
what turns those neutral character traits into intellectual virtues or vices.
9.1.1 The trait

So, what is the trait of intellectual perseverance (IP)? By way of introduction, it is
a disposition to continue to perform intellectual actions, so as to overcome obsta-
cles, in pursuit of one’s intellectual goals (Battaly 2017). Intellectual persever-
ance is a subset of general perseverance (GP), the latter of which does not restrict
its goals and actions to the intellectual.
Which goals are intellectual? Roughly, intellectual goals are those that aim
at intellectual objects. They aim at pursuing or avoiding, among other things,
knowledge, beliefs, ideas, learning, understanding, and inquiry. Agents have been
known to adopt a wide variety of intellectual goals. Examples include figuring out
the answer to a question, learning a language, understanding another person’s per-
spective, mastering a particular field of knowledge, gaining just enough knowl-
edge to get promoted, avoiding knowledge of a subject matter at all costs, and
avoiding uncomfortable ideas. Contrast these with goals that are not intellectual
Quitting, procrastinating, and slacking off 169
per se and fall under the purview of general perseverance. Some examples include
exercising more, eating better, ending one’s marriage, and spending less time with
one’s family.
The trait of IP allows an agent’s intellectual goals to be sub-standard. This is
as it should be, since agents sometimes persevere with respect to sub-standard
intellectual goals, e.g., the goals of amassing trivia about celebrities, of avoid-
ing knowledge of anthropogenic climate change, etc. But, the trait of IP does not
allow intellectual goals to be too easily achieved. It excludes goals that are so eas-
ily achieved that they don’t permit any obstacles (the goal of solving ‘1 + 1 = x’)
and don’t require any effort on the part of the agent.
Obstacles make it difficult for an agent to achieve her intellectual goals and
require effort to overcome. Some agents will encounter obstacles, and need per-
severance, where others don’t.1 To use a simple case, we can expect logic stu-
dents, but not logic professors, to encounter obstacles when doing basic proofs.
Accordingly, obstacles, and perseverance itself, must be indexed to the agent in
question (King 2014: 3795). Obstacles can be internal or external. Internal obsta-
cles, generated by one’s own psyche, include frustration (with a research project),
boredom and drudgery (in compiling data for a monthly report), and the desire to
do something easier, or more pleasurable, or less tedious. External obstacles can
be generated by the environment or by the goals themselves. Some environments
are rife with distraction, others with discouragement, yet others with repetitive
tasks. Environments that categorically preclude an agent’s pursuit of intellectual
goals or deny her the opportunity to perform intellectual actions are a limiting
case. Moreover, some goals will be difficult to achieve even in hospitable envi-
ronments—it is challenging to solve Russell’s paradox.
Agents with the trait of IP are disposed to successfully overcome routine obsta-
cles, such as the temptation to play video games instead of doing one’s home-
work. When obstacles are extremely difficult—e.g., as they are in finding a cure
for AIDS—agents with IP will try to overcome them, even if they don’t succeed
and never accomplish their intellectual goals. Persevering with respect to a goal
is distinct from accomplishing the goal—one can persevere without completing a
project, and (as we will see below) one can complete a project without persever-
ing. Since agents and their intellectual goals and obstacles vary, the actions that
agents perform (in trying) to overcome obstacles will also vary. Some standard
examples include generating replies to objections, resisting the temptation to for-
sake a frustrating line of inquiry for something easier, and trying an alternative
solution to a problem after an initial failure. Though these specific actions vary,
they are all intellectual actions—roughly, voluntary actions of the mind, which
may or may not manifest in observable behavior.
There are two points to drive home about intellectual actions. First, the cat-
egory of ‘intellectual actions’ is much broader than the category of ‘actions that
agents with IP perform (in trying) to overcome obstacles.’ For starters, intellectual
actions aren’t restricted to IP in any way, and are involved in all intellectual char-
acter traits. Further, guessing and jumping to conclusions are intellectual actions,
but these aren’t involved in overcoming obstacles so much as in avoiding them.
170 Heather Battaly
Finally, obstacles, setbacks, difficulties, and so forth need not even be on the
scene for an agent to perform intellectual actions. For many adults, the intellectual
actions of adding 405 + 68, and reading and understanding a few sentences in a
news story (in one’s native language) pose no obstacles whatsoever.
Second, the trait of IP allows an agent’s intellectual actions to be sub-standard.
This is because it allows an agent’s intellectual goals to be sub-standard. Consider
Hunter, an agent with the trait of IP whose goal is to avoid knowing anything
about white privilege. Hunter overcomes the obstacles that he regularly encoun-
ters—evidence of white privilege in his workplace and community, speeches by
world leaders that address white privilege, students in his classes who mention
white privilege, talks about white privilege at his university—by ignoring them.2
He thus performs intellectual actions that an intellectually virtuous agent would
not perform, in the service of a goal that an intellectually virtuous agent would not
have. This brings us to the distinction between the trait of IP and the virtue of IP.
9.1.2 The virtue

We have seen that the trait of IP involves staying the course—continuing to per-
form intellectual actions in an effort to overcome obstacles in pursuit of one’s
intellectual goals. Marie Curie had the trait of IP. She ran hundreds of experi-
ments in order to isolate a decigram of radium from several tons of pitchblende.
Bertrand Russell had the trait of IP. He stared at a blank sheet of paper every day
in the summers of 1903 and 1904 trying to solve, what has come to be known as,
Russell’s paradox.
But Albert Michelson and Edward Morley also had the trait of IP. In the late
1880s, Michelson and Morley conducted one failed experiment after another in
an effort to measure the alleged phenomenon of ether drift. Rather than give up
the assumption that ether existed and abandon their line of inquiry, they stayed
the course. Similarly, English professor Grady Tripp—the protagonist of Michael
Chabon’s novel Wonder Boys—has the trait of IP. Tripp won’t give up on the
novel he is writing, even though he has written 2,600 pages and gotten no closer to
the end. Chabon himself devoted 5 years and 1,500 pages to a manuscript before
abandoning it (Gorney 2010). These agents have the trait of IP, but they don’t
have the intellectual virtue of IP. At least, they don’t have the intellectual virtue
of IP on an Aristotelian analysis, which is one of the two dominant analyses of
intellectual virtue in the literature.3 Why not?
The short answer is that these agents act inappropriately because they lack
good judgment. They have latched onto intellectual goals that are in fact doomed
to fail, and that they should have learned were doomed to fail (given the lack of
progress they had made), and have refused to give them up! They persevered even
when they should have abandoned their goals. In other words, they persevered
to excess, whereas Aristotelian virtue requires hitting the mean in one’s actions.
Gary Watson makes a similar point about benevolence: ‘the word “benevolence”
names both a general concern for others (which may be excessive, and lead to bad
action) and the qualified and informed concern that constitutes the virtue’ (1984:
68). Just as an excess of the trait of benevolence can cause one to donate to an
organization without considering its merit, an excess of the trait of intellectual
perseverance can cause one to stick with projects that are ill-fated. It can cause
one to perform actions that are characteristic of the intellectual vice of intransi-
gence, which Nathan King describes as a disposition to ‘give up too late or not at
all’ and to ‘persist in … projects long after [they] should be abandoned’ (2014:
3786).
The Aristotelian virtue of IP requires good judgment about when to persevere
with respect to a goal and when to abandon the goal. Good judgment reins in
excess—it prevents agents from staying the course when they should quit.4 It like-
wise reins in deficiency—it prevents agents from quitting when they should stay
the course (see below). It enables agents to hit the mean in their actions—to stay
the course when it is appropriate. Unsurprisingly, it can be difficult to determine
exactly when to persevere with respect to a goal and when to quit. Though it is
clear that Michelson and Morley, and Tripp should have quit, most cases aren’t
as easy to judge. Still, two criteria can be helpful.5 First, good judgment involves
balancing the importance of a goal against the difficulty of achieving it. Some
intellectual goals, like figuring out a cure for AIDS, rank high on the scale of dif-
ficulty, but also high on the scale of importance, rendering the judgment to stay
the course. Other goals, such as assessing academic programs, can suffer the fate
of ranking high on the scale of difficultly and low enough on the scale of impor-
tance to render the judgment to quit. The same fate awaits trivial goals—figuring
out Nolan Ryan’s ERA in Tuesday-games—which rank even lower on the scale
of importance. Second, good judgment factors in the agent’s skill-set and epis-
temic opportunity costs. Arguably, the late Toni Morrison would have amassed
epistemic opportunity costs had she continued to edit the work of other black
authors instead of quitting editing to write her novel Beloved. In contrast, an edi-
tor without the skill-set to write fiction, let alone an important novel, would amass
epistemic opportunity costs by abandoning his editorial goals.6
There is more to the Aristotelian virtue of IP than good judgment and appropri-
ate action. At a minimum, agents also need good motivation. Jason Baehr (2020)
argues that for a character trait to be an intellectual virtue, it must be grounded
in a motivation for epistemic goods, such as truth, knowledge, and understand-
ing. In many of the agents named above—including Curie, Russell, and even
Michelson and Morley—the trait of IP is grounded in a motivation for epistemic
goods. Russell, for instance, overcame obstacles and stayed the course because
he wanted to understand. But the same can’t be said of Hunter, who is motivated
to avoid knowing anything about white privilege. When an agent’s trait of IP is
driven by the motivation to protect his own views, or to win arguments, or to
compete with peers, it is not an intellectual virtue.7
9.2 Quitting
There are several traits that are deficiencies of, or ways of lacking, the trait of
intellectual perseverance. When an agent hits obstacles to her intellectual goals,
172 Heather Battaly
she might tend to quit, or procrastinate, or slack off, rather than persevere. Let’s
begin with an analysis of the trait of quitting. As above, we will first propose a
normatively neutral analysis of the trait, and then investigate its status as an intel-
lectual vice. This can help us home in on what makes the trait of quitting a vice
when it is one.
The trait of quitting is arguably a disposition to abandon one’s intellectual
goals when one hits obstacles. Quitters do not perform intellectual actions in
an effort to overcome obstacles—they do not persevere. But nor do they delay
action, or intentionally take the easy route, in pursuit of their goals. When they hit
obstacles, they give up their intellectual goals altogether.
When is quitting an intellectual vice? The answer depends on what we think
intellectual vices are. Mirroring the Aristotelian analysis of virtue in Section 9.1,
we might think of intellectual vices as bad intellectual character traits (Battaly
2014). So, just as intellectual virtues involve dispositions to perform appropriate
intellectual actions, intellectual vices will involve dispositions to perform inap-
propriate intellectual actions. As we saw above, intellectual virtues also involve
good judgment and good intellectual motivations. Accordingly, we can consider
whether intellectual vices likewise involve bad judgment and bad intellectual
motivations, like those of Hunter above, or at least the absence of good intellec-
tual motivations, like the motivation for truth.8
With this basic structure in mind, we can investigate whether the trait of quit-
ting is an intellectual vice. Importantly, our target is the trait, or disposition, of
quitting. Dispositions are tendencies. They tell us what an agent would do, were
she to encounter a particular type of situation. In this case, the disposition tells us
that the agent would abandon her intellectual goals were she to encounter obsta-
cles. To drive this point home: she would consistently abandon her intellectual
goals in the face of obstacles—she wouldn’t abandon some goals but persevere
with respect to most others, she would (with few exceptions) abandon them all.9
As we can glean from the discussion of obstacles above, agents often encounter
obstacles to their intellectual goals, though some agents will encounter different
(or more) obstacles than others. Accordingly, we can identify quitters by look-
ing for agents who consistently abandon their intellectual goals when they hit
obstacles.
These agents give up their intellectual projects whenever the going gets tough.
They drop their classes as soon as the material gets challenging. They scrap ideas
for projects upon sensing the first whiff of criticism. They forsake lines of inquiry
the moment they become boring or tedious. And they behave in these ways over
and over again. Granted, these agents also quit intellectual projects that should
be abandoned. They quit the projects of measuring ether drift, of writing (ill-con-
ceived) novels, and of computing Nolan Ryan’s ERA in Tuesday-games. Some of
their acts of quitting end up being appropriate (albeit unwittingly). Nevertheless,
we can expect the vast majority of their acts of quitting to be inappropriate.
Provided that they generally adopt goals that are appropriate for them—provided
that their classes, ideas for projects, etc., aren’t ill-fated, and are sufficiently impor-
tant and suited to their skill-sets—these agents are quitting when they shouldn’t.
Their actions do not match those of people with the virtue of IP, who would stick
with the aforementioned goals, striving to overcome the challenging material, the
whiffs of criticism, and the tedium. Instead, their actions match those of a person
with the vice of irresolution, which in King’s words, involves ‘a disposition to
give up too early on one’s intellectual projects in the face of obstacles’ (2014:
3786). In short, provided that there is nothing wrong with an agent’s intellectual
goals, the disposition to quit will be a disposition to perform inappropriate intel-
lectual actions.10
Suppose that agents generally adopt intellectual goals that are appropriate for
them (our goals at, e.g., school and work tend to be appropriate). Does this mean
that the disposition to quit is an intellectual vice? It, at least, means that quitting is
not an intellectual virtue, since intellectual virtues require dispositions of appro-
priate intellectual action. By itself, is the disposition to perform inappropriate
actions enough to make quitting an intellectual vice? That is an open question for
vice epistemologists. But factoring in the motivations of quitters can help tip the
scales. Suppose that the quitters above abandon their intellectual goals because
they don’t want to put in the effort to overcome obstacles. As Kidd remarks, ‘a
person may not care enough … to put in the epistemic work’ (2017: 15). Quitters
may not care enough about epistemic goods to put in the effort it takes to acquire
them—their motivation for epistemic goods may be deficient (Crerar 2018: 758).
Accordingly, even if their disposition to perform inappropriate actions isn’t suf-
ficient for vice on its own, that disposition combined with a lack of appropriate
motivation makes a stronger case for vice-possession.
Is the disposition to quit ever an intellectual virtue? For starters, we can reiter-
ate that an individual act of quitting can be appropriate even when the general dis-
position to quit would be a vice. Case in point: Michelson and Morley should have
abandoned their goal of measuring ether drift; here, it was appropriate to quit.
Similarly, it can be appropriate for an agent to abandon a difficult intellectual goal
whose importance is low or whose continued pursuit would produce epistemic
opportunity costs. In short, it can be appropriate to quit a particular project, even
when a general disposition to quit would be a vice. So, are there any conditions in
which this general disposition would be an intellectual virtue?
Here is a suggestion. Imagine an inhospitable epistemic environment, in which
agents are systematically prevented from pursuing appropriate intellectual goals.
Agents usually only have opportunities to pursue intellectual goals that are ill-
fated or trivial, and don’t get to pursue goals that are important.11 We could think
of such cases along the lines of demon-worlds, in which agents only ever get
to pursue goals to square the circle or to amass knowledge of celebrities. But it
might be more useful to recall the ways in which women have been, and in some
places continue to be, systematically excluded from formal education and directed
toward more trivial topics. When women are prevented from studying subjects
such as mathematics, literature, history, and science, and directed to trivial sub-
jects (e.g., the truths of conventional etiquette),12 the general disposition to quit
starts to look less like an intellectual vice and more like a virtue. It might be appro-
priate to quit the trivial intellectual projects into which one has been channeled.
174 Heather Battaly
On the assumption that those are (with few exceptions) the only projects to be
had, the general disposition to quit would then be a disposition of appropriate
intellectual action. By itself, that isn’t enough to make it an intellectual character
virtue. The agents in question would at least need to be motivated to quit because
they cared about and wanted to pursue more important truths. Crucially, they
would also need good judgment—which would prevent them from quitting on the
rare occasion when goals were appropriate. Though we still need further details, I
hope to have taken us some way toward conditions in which quitting might count
as an intellectual virtue. In Orwellian scenarios, quitting might even count as a
virtue of resistance.13
9.3 Procrastinating
When procrastinators hit obstacles, they don’t abandon their intellectual goals.
They aren’t quitters. But nor do they continue to perform intellectual actions in
pursuit of their goals—they don’t persevere either. Instead, they delay acting in
pursuit of their goals. Following the procedure above, we will first propose a
normatively neutral analysis of the trait of procrastinating, which distinguishes it
from the traits of perseverance and quitting. We will then investigate its evalua-
tive status as an intellectual vice.
9.3.1 The trait

The trait of procrastination is arguably a disposition to intentionally delay acting
in pursuit of intellectual goals that one believes one should be actively pursuing.
Specifically, it is a disposition to do this when one hits obstacles. Let’s explore
four key features of this working definition.
First, unlike quitters, people with the trait of procrastination hold onto their
intellectual goals when they encounter obstacles. They put off acting in pursuit
of their intellectual goals, but they do not abandon them.14 Sarah Stroud (2010:
65) insightfully argues that these goals can be vague and unspecific—one need
not have a well-developed goal in order to count as procrastinating. Indeed, one
can have a vague goal, like writing a grant proposal ‘sometime this year,’ and
procrastinate with respect to making more specific plans to achieve it. The point
is that procrastination requires having an intellectual goal of some sort: one must
have a goal with respect to x in order to count as procrastinating with respect to x.
Second, when procrastinators hit obstacles, they intentionally delay acting in
pursuit of their intellectual goals. This means that though they have the time and
opportunity to act, they intentionally avoid acting now, and typically, intend to
act later. This also means that people who fall behind need not be procrastina-
tors. An analogy from professional cycling may be helpful. Cyclists in an ardu-
ous multi-stage race like the Tour de France can fall behind even when they are
cycling as fast as they can. Falling behind in, e.g., writing a book, is consistent
with persevering—with overcoming obstacles and continuing to act in pursuit of
that goal—and does not entail any intentional delay in acting.
Third, according to our working definition, intentionally delaying action in
pursuit of a goal is not sufficient for procrastination. Suppose Daphne has simulta-
neous looming deadlines: one for a publication on which she knows she has much
left to do, and the other for an update of an extant letter of recommendation, which
she knows will take her less than an hour. She does not procrastinate when she
intentionally delays working on the letter in order to make progress on the pub-
lication. Some cases of intentionally delayed action are simply exercises of good
time-management. In Stroud’s words, ‘procrastination cannot be equated with
putting something off’ (2010: 53). Procrastination requires a further condition.
Which condition? This brings us to our fourth key feature. According to our
working definition, to procrastinate, an agent must intentionally delay acting in
pursuit of goals that she believes she should be actively pursuing. Thus, Daphne
didn’t procrastinate in working on the letter because she didn’t believe that she
should have been working on it. Note that on our definition, an agent who should
in fact be actively pursuing a particular goal, but doesn’t herself believe this, is not
procrastinating. But why not claim that agents procrastinate whenever they should
in fact be actively pursuing their goals, whether or not they believe this? On that
analysis, Daphne didn’t procrastinate because she should not, in fact, have been
working on the letter. Since there is disagreement in the literature over which of
these conditions to adopt, our working definition requires further defense.
Let’s begin defending it by laying out three different proposals: the first—our
working definition—has a subjective bent, the second has an objective bent, the
third combines these.
Subjective proposal: procrastination is a disposition to intentionally delay act-

ing in pursuit of intellectual goals that one believes one should be actively
pursuing.
Objective proposal: procrastination is a disposition to intentionally delay acting
in pursuit of intellectual goals that one should in fact be actively pursuing.
Combined proposal: procrastination is a disposition to intentionally delay acting
in pursuit of intellectual goals that one believes one should be actively pursu-
ing and that one should in fact be actively pursuing.
Situating these in the literature, Stroud (2010: 57) endorses a version of the
subjective proposal. Christine Tappolet (2010: 121) ascribes the objective pro-
posal—what she calls ‘blind procrastination’—to Chrisoula Andreou’s (2007:
183) analysis of procrastination as delaying ‘what one should—relative to one’s
ends and information—have done sooner.’ Duncan MacIntosh (2010) arguably
endorses the combined proposal.15
Why isn’t the objective proposal correct? For starters, the condition in the
objective proposal isn’t sufficient for procrastination. An agent can satisfy that
condition—she can intentionally delay acting in pursuit of goals that she should
in fact be actively pursuing—without procrastinating. To see why, consider the
following case. Optimistic Opal has been assigned a new course to teach next
semester. Suppose it will take her a considerable amount of time to design a
176 Heather Battaly
syllabus for the course—she should, in fact, start working on it soon. But Opal
doesn’t realize this—she is overly optimistic about her ability to design the syl-
labus and she under-estimates the obstacles. Opal thus believes, albeit falsely, that
she can put off work on the syllabus until the weekend before the semester begins.
Accordingly, she delays working on it. Though Opal intentionally delays acting
in pursuit of one of her intellectual goals, and though this is a goal that she should
in fact be actively pursuing, she isn’t procrastinating. She isn’t procrastinating
because she doesn’t think she needs to work on the syllabus. To put the point
differently, her problem isn’t procrastination, it is mistaken belief, specifically
the mistaken belief that she shouldn’t be working on the syllabus. We can even
assume that Opal would have started working on the syllabus had she realized
that she should have. Had her beliefs been correct, Opal wouldn’t have delayed.
Opal has a problem alright, but the objective proposal misdiagnoses it. For
Opal’s problem to be procrastination, she would need to believe that she should
have been working on the syllabus. Procrastination requires a conflict between
what one believes one should do, and what one actually does or fails to do. We
need this subjective condition in order to avoid misdiagnosing the problem.
Taking it on board thus gives us a conceptual advantage over the objective pro-
posal. Taking the subjective condition on board likewise gives us an explanatory
advantage: it explains the awful feelings of regret that typically accompany pro-
crastination, whereas the objective proposal does not.
Now, you might agree that the objective condition isn’t sufficient for procrasti-
nation and that some sort of subjective condition is necessary, but you might worry
that the details of the subjective condition above aren’t quite right. Specifically,
you might worry that procrastinators don’t need to believe that they should be
actively pursuing a goal. (1) Can’t someone procrastinate in coming to have such
a belief in the first place? (2) Don’t some procrastinators deceive themselves into
believing that they shouldn’t be actively pursuing their goals?
Regarding the first worry, consider a teenager who has the goal of applying to
college, but puts off thinking about this goal, and never forms the belief that she
should be working on college applications. Our teenager should have formed that
belief (and would have, had she given the matter sufficient thought), but didn’t.
So, the worry is that our teenager is procrastinating in delaying work on her appli-
cations, even though she doesn’t believe that she should be working on them. My
response is to hold the line. Our teenager does have a problem, but it isn’t yet the
problem of procrastination. It looks like procrastination because her actions match
those of a procrastinator, but her psychology is different—she doesn’t think she
is doing anything wrong and doesn’t feel regret. Her problem is instead one of
negligence, which might ultimately be rooted in laziness or apathy. We can put
things off for lots of ‘reasons,’ many of which are not as psychologically complex
as procrastination.
Let’s move to the second worry: can the subjective condition allow for cases
of procrastination that involve self-deception? That could depend on what self-
deception entails. Suppose self-deception entails holding contradictory beliefs.
Suppose I intentionally delay working on a publication with an imminent deadline
because I have deluded myself, at some level, into thinking that other projects are
more important; nevertheless I retain the implicit belief that I really should be
working on the publication—when it comes time to request an extension, I curse
myself, admitting that I knew all along that I should have been working on it. The
subjective condition above does count this as a case of procrastination, since I still
(implicitly) believe that I should be working on the publication. So far, so good
for our subjective condition.
But what if self-deception does not entail an implicit belief that I should be
working on the publication? What if I initially delay action while believing that
I should be working on the publication, but then utterly and completely delude
myself into believing that it can wait? Since my belief that I should be working on
it has been entirely eradicated, our subjective condition won’t count this as a case
of procrastination. But is that the wrong result? Perhaps not. Perhaps, in changing
beliefs, my problem also changes—it morphs from being a problem of procras-
tination to being one of self-deception—in which case, our subjective condition,
once again, gets the right result.
Still, it might be hard to shake the intuition that our teenager and our self-
deceiver are procrastinating, in which case, we might need to tinker with our
subjective condition. I encourage future tinkering. In the meantime, I will be
defending the subjective condition, keeping in mind that amendments might be
forthcoming.
Should we combine the subjective condition with the objective condition,
plumping for the combined proposal? Or should we endorse the subjective pro-
posal on its own? For an agent to procrastinate, must she correctly believe that
she should be acting? Or would any belief that she should be acting—even a false
one—suffice for procrastination?
I will be plumping for the subjective proposal—our working definition—and
arguing against the combined proposal. We should reject the combined proposal
because the objective condition isn’t necessary for procrastination either. Agents
can procrastinate with respect to a project even when they shouldn’t in fact be
working on it. In other words, agents can falsely believe that they should be work-
ing and still be procrastinating. To see why, consider the case of Mona, who
has been assigned the task of writing her organization’s next quarterly report.
Mona has never done a task like this before, and needs to find out what the report
requires in order to complete it. She assumes that the report will require a consid-
erable amount of work on her part, and thus believes that she should start working
on it soon. She puts a reminder in her calendar to investigate the requirements
of the report. Still, she dreads getting started because of the onerous work she
thinks the report will demand. When the reminder pops up, she delays it, feeling
regret and anxiety about doing so. Mona intentionally delays acting in pursuit of
an intellectual goal that she believes she should be actively pursuing. Thus far,
she meets the condition in the subjective proposal, which is also a condition in the
combined proposal. Now suppose that she doesn’t meet the objective condition
in the combined proposal. It turns out that the report isn’t onerous, it will be easy
for Mona to complete, and she has plenty of time. Given the other intellectual
178 Heather Battaly
projects currently in her inbox, Mona’s belief that she should be actively working
on the report is false. Does Mona still count as procrastinating? I submit that she
does. She is intentionally putting off working on an intellectual project that she
believes she should be working on. As a ‘bonus,’ she feels badly about it. Her
belief that she should be working on the report doesn’t need to be true for her
to be procrastinating. It doesn’t need to be a fact that she should be working on
it. Relatedly, consider agents who (despite empirical evidence to the contrary)16
really do produce their best work under pressure. Suppose one such agent, Milton,
delays working on a paper, while believing he should be working on it, and feeling
terribly about failing to work on it. (Assume Milton doesn’t realize that he excels
under pressure.)17 He nevertheless produces a marvelous paper at the last minute.
Given Milton’s ability to excel under pressure, working on the paper wasn’t press-
ing for him—he had plenty of time. We can thus assume that Milton’s belief that
he should have been working on the paper was false: as a matter of fact, given his
other intellectual commitments, he should not have been working on the paper.
Nevertheless, he was procrastinating.18
By way of comparison, Mona and Milton do have something in common with
Opal. All of them have mistaken beliefs. But Mona and Milton mistakenly believe
that they should be working on their projects, whereas Opal mistakenly believes
that she shouldn’t be working on hers. As a result, Mona and Milton have a prob-
lem that Opal doesn’t have—they are procrastinating, and she is not.19 We can
point out a further difference between them. I have been assuming that Opal, and
Mona and Milton, are all disposed to intentionally delay action. Opal is disposed
to do so because she is disposed to over-estimate her abilities and under-estimate
obstacles. Mona and Milton are disposed to do so because they tend to delay
action when they think a project will be onerous. The difference is this: Opal’s
delays of action can be ‘fixed’ wholesale by correcting her beliefs, but Mona’s
and Milton’s can’t. Granted, correcting their beliefs can fix their delays of action
when they falsely believe that they should act. But Mona and Milton will also
delay action when they truly believe that they should act. To put the point differ-
ently, we can’t fix procrastination simply by correcting an agent’s beliefs because
there will be plenty of cases in which agents procrastinate while knowing (and
truly believing) that they should be acting! When students (or professors) procras-
tinate in writing papers until the night before the deadline, they typically know
(and truly believe) that they should have started earlier. The point of the previous
paragraph is that procrastination doesn’t require such beliefs to be true: one can
procrastinate even when one should not in fact be working on a project. But, of
course, one can also procrastinate when one should in fact be working on a project
and knows this full well.
9.3.2 Vice, akrasia, and virtue

In the last section, I argued that the trait of procrastination is a disposition to
intentionally delay acting in pursuit of an intellectual goal that one believes one
should be actively pursuing (whether that belief is true or false). We have seen
that procrastinators postpone working on book projects, quarterly reports, and
grading, and put off writing papers until the last minute. They delay as soon as the
going gets tough. And they delay while believing they should be actively pursuing
their projects.
Is the trait of procrastination, so described, an intellectual vice? Here, too,
the answer depends on what intellectual vices are. Suppose that intellectual vices
are bad character traits that at least involve dispositions to perform inappropriate
intellectual actions. Are procrastinators disposed to act inappropriately? I submit
that they are, provided that they tend to procrastinate with respect to intellectual
goals that they should, in fact, be actively pursuing. In other words, they will
be disposed to act inappropriately, provided that they tend to procrastinate with
respect to goals that they correctly believe they should be actively pursuing. Thus,
agents whose trait of procrastination ranges widely over, e.g., pressing book pro-
jects, papers, grading, and the like are acting inappropriately. Their actions do
not match those of people with the virtue of IP, who would be working on these
projects.
This means that procrastination with respect to goals one should in fact be
actively pursuing isn’t an intellectual virtue. But, by itself, the disposition to per-
form inappropriate intellectual actions may not be enough to make such procras-
tination an intellectual vice. As we suggested above, intellectual vices might also
involve bad motivation and bad judgment. So, do the motivations of procrastina-
tors tip the scales toward vice? Quintessential procrastinators retain some motiva-
tion for epistemic goods—they care about truth more than quitters do and enough
to stick with their projects. Nevertheless, their epistemic motivations are deficient.
They ultimately care more about avoiding the effort, or perceived effort, it would
take to work on a project than they do about pursuing truth. Their deficient moti-
vations are weak enough to disqualify them from virtue, but are they weak enough
to qualify them for vice? And what about their judgment? Procrastinators who
correctly believe that they should be acting arguably have good judgment rather
than bad.
Concerns like these led Aristotle to distinguish between vice and akrasia
(weakness of will). On his view, ‘akrasia is not vice … for akrasia is contrary to
choice while vice is in accordance with choice’ (NE.1151a6–7). For starters, this
means that akratic people act contrary to their beliefs about what they should do.
More specifically, Aristotle argues that akratic people know what they should
do (their beliefs are true), and want to act appropriately, but also have compet-
ing appetites which outweigh their desires to act appropriately (NE.1151a21–24).
Consequently, they act inappropriately and regret it. Aristotle thinks that unlike
akratic people, vicious people act in accordance with what they believe they
should do. But, since vicious people have false beliefs about what that is, they too
act inappropriately, though with conviction and without regret (NE.1150a18–21).
We need not endorse every detail of Aristotle’s analysis of vice. For present pur-
poses, it is enough that a distinction between akrasia and vice can be drawn.
This distinction can help us diagnose procrastinators who correctly believe that
they should be actively pursuing their goals. These procrastinators have good
180 Heather Battaly
judgment, or at least true beliefs about what they should do. Their desires for truth
are likewise outweighed by their desires to avoid effort, resulting in inappropri-
ate delays, which they regret. Accordingly, I submit that these procrastinators are
akratic, even if they aren’t vicious.20
Given the analysis above, could the trait of procrastination ever be an intellec-
tual virtue? Arguably, some individual acts of procrastinating will be condition-
ally appropriate. Recall Mona and Milton who falsely believed that they should
have been acting in pursuit of their goals. Mona and Milton ended up doing the
right thing: given their other intellectual projects, it was appropriate for them to
delay action. On the condition that their beliefs were false—the quarterly report
and the paper were not in fact pressing—their individual acts of procrastinating
were appropriate. Suppose we are willing to go that far. What happens when
we then try to generalize from individual acts to a stable disposition? The short
answer is that we end up with cases of ‘inverse akrasia’ rather than virtue.21 We
can generate cases where procrastination involves a general disposition to act
appropriately, but these cases come at a cost. In them, agents will be disposed to
have false beliefs about what they should do. They will have terrible judgment,
which ensures that their delays will be appropriate, but disqualifies them from
virtue.
Given its ties to akrasia, the trait of procrastination isn’t likely to be an intel-
lectual character virtue. But there is a different way of thinking about intellectual
virtues, and here procrastination might stand a better chance. Instead of conceiv-
ing of intellectual virtues as character traits that require dispositions of appro-
priate intellectual action, good judgment, and the motivation for truth, we can
conceive of them as qualities that reliably produce good epistemic effects. On this
analysis, intellectual virtues are qualities that reliably produce true beliefs and
knowledge.22 We can add further good effects to this traditional list, such as capi-
talizing on epistemic opportunities, and completing worthwhile intellectual pro-
jects. John Perry (2012) seems to be thinking of ‘structured procrastination’ along
these lines. He argues that structured procrastinators are those who complete a
range of ‘difficult, timely, and important tasks’ that they wouldn’t have otherwise
completed, as a way of avoiding work on projects that are even more important
(2012: 3). Structured procrastinators, unlike quitters, get a lot of important work
done—they reliably produce good epistemic effects. Still, Perry admits that much
procrastination is not structured and leads to bad results. The psychological lit-
erature (Kim and Seo 2015) confirms that procrastination tends to produce bad
epistemic effects. It produces work that contains omissions and mistakes—or at
the very least, work that is less careful, less thorough, and less insightful than it
would have been, had the agent persevered. So the trait of procrastination won’t
be an easy fit for effects-based virtue either, and is instead a likely candidate for
effects-based vice.
Nevertheless, there will be inhospitable epistemic environments where the
general disposition to procrastinate will produce better epistemic effects than the
trait of IP. Consider the Oceania of George Orwell’s 1984, where most intellec-
tual projects aim at generating and disseminating falsehoods. Here, the general
disposition to procrastinate will at least slow the production of bad epistemic
effects. In this environment, we would recommend the general disposition to pro-
crastinate over the general disposition of IP, even if we weren’t willing to count
procrastination as an outright effects-virtue. We would likewise recommend the
general disposition to quit, which will produce even better epistemic effects than
procrastination. In Oceania, the disposition to quit intellectual projects might be
our least bad option.
9.4 Slacking off

Finally, let’s address a third way of lacking the trait of intellectual perseverance:
slacking off. The trait of slacking off is arguably a disposition to intentionally
take an easy path in pursuit of one’s intellectual goals, knowing (or at least truly
believing) that one’s work will suffer. Whereas people with the trait of IP are
disposed to overcome obstacles, people with the trait of slacking off are disposed
to avoid them, or succumb to them.
Importantly, agents can slack off without quitting or procrastinating. Agents
who slack off need not abandon their intellectual goals, when they hit an obstacle.
Students can intentionally take an easy path in completing assignments without
quitting—perhaps because they still want to pass and graduate, even if they don’t
care much about the assignments themselves. Likewise, one can intentionally take
an easy path without delaying action at all, or without doing so against one’s bet-
ter judgment. One can slack off because one thinks a goal isn’t important enough
to be actively pursuing. So, slacking off does not imply procrastinating either.23
Perhaps what goes ‘slack’ in cases of slacking off is the agent’s degree of com-
mitment to her goals: in slacking off, her motivation to pursue her goals, or her
valuing of them, is reduced.
There are four key features of this working definition. First, slacking off
requires intentionally taking an easy path, knowing (or at least truly believing)
that one’s work will suffer. Intentionally taking an easy path isn’t, by itself, suf-
ficient for slacking off. Sometimes taking the easy path is a mark of efficiency!
To slack off, one must intentionally take an easy path, knowing (or at least truly
believing) that this will reduce the quality or rate of one’s work.
Second, slacking off requires the intention to take an easy path, and the belief
that one’s work will suffer. This means that agents who take an easy path without
intending to do so, or without believing their work will suffer, are not slacking
off. Consider the freshman student who doesn’t include a thesis statement in her
first university-level paper because she didn’t know one was needed. Suppose
she was never asked to write a thesis statement in high school, and her current
professor never mentioned that the paper required a thesis statement. In writing a
‘stream of consciousness’ paper, our freshman skipped an important step, avoided
an obstacle, took (what was for her) an easy route, and produced a flawed paper.
But she didn’t intend to do any of this and didn’t believe her paper would be
flawed—she simply didn’t know that a thesis statement was required. Indeed, we
can assume that had she known this, she would have worked hard to generate a
182 Heather Battaly
thesis statement. She wouldn’t have avoided the obstacle of generating one; she
would have persevered in trying to overcome that obstacle.
Our freshman has a problem, but it isn’t slacking off, it is ignorance about the
required components of a university-level paper. Like Opal above, whose delays
of action can be fixed wholesale by correcting her beliefs, our freshman’s step-
skipping can be fixed wholesale by supplying her with knowledge.24 But slacking
off isn’t the kind of problem that can be fixed by a shot of knowledge. Agents who
slack off already know (or truly believe) that they are taking an easy path and that
their work will suffer.
One might object that these conditions make our definition of slacking off too
narrow. They only get the right answer about our freshman because she couldn’t
have been expected to know that she was taking an easy path or that her work
would suffer. Our conditions exclude agents who should have known that they
were taking an easy path, or that their work would suffer, but didn’t. Here, too,
my response is to hold the line. These agents do have a problem, but it isn’t slack-
ing off. Their problem is negligent ignorance, which may ultimately be rooted in
apathy or laziness.
This brings us to the third key feature. Slacking off requires succeeding in
taking an easy path, and actually reducing the quality or rate of one’s work. This
means that agents who intend to take an easy path and believe their work will suf-
fer, but do not actually succeed in taking an easy path or in reducing the quality of
their work, are not slacking off. Suppose it is your job to write the department’s
assessment report. Given the looming deadlines of your other intellectual pro-
jects—preparing a presentation, finishing an article—you intend to take an easy
path in analyzing the assessment data, fully aware that this will reduce the quality
of your report. But, because of your perfectionist tendencies, you don’t actually
take an easy path—you end up running multiple analyses of the data, and even
consult the stats department. You don’t skip steps, as is your intention. I submit
that you aren’t slacking off. You are trying to slack off, but failing: some agents
can’t produce C-level work even when they intend to.25
Let’s turn to the fourth key feature of slacking off. There are different ways
to take an easy path, some of which involve avoiding obstacles and others of
which involve succumbing to them. For instance, agents can take an easy path
by step-skipping, phoning it in, wasting time, or free riding. As rough sketches,
consider Stan the step-skipper and Philippa who phones it in. Both intention-
ally circumvent obstacles. Suppose Stan and Philippa are each preparing to teach
modules on ethics in their intro courses; Stan for the first time, Philippa for the
tenth. Stan intentionally takes an easy path by not including detailed analyses
of any ethical theories. He knows that good modules include such analyses and
that his module will suffer, but he wants to avoid the work it will take to pro-
duce them. Instead, he relies on easy platitudes (‘The Golden Rule’), intention-
ally skips detailed analysis, and delivers a shoddy module. Philippa intentionally
takes an easy path by using the same class lectures she used last year. These may
already contain appropriately detailed and updated analyses of the major ethical
theories—Philippa needn’t have skipped any steps. Nevertheless, using last year’s
lectures poses no obstacles for her—it is easy for her to do, given her abilities.
Now add that Philippa knows she could make the module better with further revi-
sions. But, since she wants to avoid the effort it will take to improve the module,
she intentionally phones it in. (On whether this can be virtuous, see below.)
Next, consider Tim the time-waster and Frida the free rider, both of whom
intentionally take paths that allow them to succumb to obstacles. Tim intention-
ally takes ages to complete straightforward projects, knowing that this will reduce
his rate of work (and prevent him from being assigned further or more difficult
tasks).26 Though he has the ability and opportunity to complete projects quickly, he
intentionally drags them out so as to make minimal effort (now and in the future),
succumbing to boredom and other internal obstacles along the way. Tim spends
most of his working hours wasting time on social media. Note that time-wasters
need not be procrastinators—time-wasters may believe that it is fine to drag tasks
out. Frida the free rider also succumbs to obstacles. She intentionally relies on
the other members of her team to overcome obstacles, and doesn’t contribute any
effort to do so herself. Frida knows that this burdens the team and will reduce the
quality of their work. In sum, all four of the above agents intentionally take an easy
path, and all four slack off according to our working definition. None perseveres.
Is the trait of slacking off, so described, an intellectual vice? Like quitting,
slacking off will involve a disposition of inappropriate action, provided that
agents generally adopt goals that are appropriate for them. In short, if an agent’s
intellectual goals (e.g., at school and work) tend to be appropriate for her, then her
disposition to slack off will issue inappropriate actions. Her actions won’t match
those of people with the virtue of IP, who would confront and (try to) overcome
obstacles rather than circumvent or succumb to them. Nor will her motivations
match those of people with the virtue of IP, if she slacks off because she doesn’t
care enough about truth. Indeed, agents who slack off might only care about
truth for its instrumental value—they might only maintain their intellectual goals
because quitting would get them kicked out of school or fired.
The previous paragraph makes a case for slacking off as an intellectual char-
acter vice. Slacking off with respect to goals that are appropriate might also be
an effects-vice (producing a preponderance of bad epistemic effects), particularly
when it comes to skipping steps. Step-skippers, who are (e.g.) running medical
tests on a patient, can easily end up with more false beliefs than true ones. While
wasting time, phoning it in, and free riding can also produce false beliefs, they
may be more likely to inhibit the overall quality and number of truths produced.
Time-wasters who succeed in producing truths might have produced even more,
had they persevered. Similarly, free riding and phoning it in can result in less
insightful truths than perseverance would have produced.
Could the trait of slacking off be an intellectual virtue? Importantly, some indi-
vidual acts of slacking off will be appropriate—slacking off can be the right thing
to do when one has other intellectual goals with higher priority. If you have hard
deadlines for an important publication, grading, and the assessment report, then
it is appropriate to phone in the assessment report. If you didn’t phone it in, you
would be amassing epistemic opportunity costs. As John Perry (2012: 20) puts the
184 Heather Battaly
point: ‘You must ask yourself some questions: How useful would a perfect job be
here? How much more useful would it be than a merely adequate job? Or even a
half-assed job?’ Indeed, if Philippa’s ethics module is already in good shape, it
can be appropriate for her to phone it in, given that she has other more important
projects with pressing deadlines. Philippa may even have hit the point of dimin-
ishing epistemic returns on the ethics module, and may phone it in because she
wants to devote energy to other more pressing projects.
Are there any conditions in which a general disposition to slack off would be
an intellectual virtue? Like its brethren, the disposition to slack off will come clos-
est to being an intellectual virtue in inhospitable epistemic environments, where
trivial or deceitful intellectual projects are the only ones on offer. In such environ-
ments, we will recommend slacking off over IP, even if we aren’t willing to count
slacking off as an outright intellectual virtue. To explicate, slacking off—wasting
time—will produce better epistemic effects in Orwell’s Oceania than IP, even if it
falls short of producing a preponderance of good epistemic effects and, thus, isn’t
an outright effects-virtue. We may likewise advise agents who are stuck doing
trivial projects to slack off—to phone them in—rather than persevere, even if
slacking off falls short of an outright intellectual character virtue. I note that quit-
ting is arguably a better candidate for intellectual virtue in inhospitable environ-
ments than either procrastinating or slacking off. For those of us who are stuck in
inhospitable environments, quitting may be our best bet.
9.5 Some related traits

I have argued that the traits of quitting, procrastinating, and slacking off are defi-
ciencies of intellectual perseverance. The trait of IP is a disposition to continue to
perform intellectual actions, so as to overcome obstacles, in pursuit of one’s intel-
lectual goals. By contrast, quitting is a disposition to abandon one’s intellectual
goals, slacking off is a disposition to intentionally take an easy path in pursuing
them, knowing (truly believing) that one’s work will suffer, and procrastinating is
a disposition to intentionally delay acting in pursuit of goals that one believes one
should be actively pursuing. I have identified some conditions that make quitting,
procrastinating, and slacking off intellectual vices (or in the case of procrastina-
tion, akratic), while also suggesting that they might approximate intellectual vir-
tues in inhospitable epistemic environments. This means, contra Aristotelianism,
that there are several vices of deficiency associated with the virtue of IP.
In closing, we can begin to map a genealogy of related traits: apathy, folly,
complacency, and resignation.27 Quitting, procrastinating, and slacking off don’t
entail any of these. Agents can quit, procrastinate, or slack off when they care about
their intellectual goals, exercise excellent judgment about them, and know their
intellectual limitations and strengths. For instance, I may care about my current
book project, know that the topic is worthwhile and that I should be working on it,
know my strengths and limitations, and still slack off because I am lazy. Laziness
arguably involves not making an effort to overcome obstacles to an intellectual
goal because one doesn’t care enough about the goal to make the effort.28 I slack
off because it is hard to write a book, and (right now at least) I care more about
watching TV than I do about putting in the work required to start the next chapter.
In contrast with laziness, apathy arguably involves not caring enough to adopt
intellectual goals in the first place. As Jason Kawall puts the point, ‘the paradigm
apathetic person … fails to value or concern herself with the … project at all’
(2006: 349).29 Arguably, folly is a way of exercising bad judgment with respect to
intellectual goals, e.g., of falsely believing that appropriate goals are inappropriate
(or vice-versa).30 Complacency and resignation arguably involve over-estimating
and under-estimating (respectively) one’s intellectual strengths. For Kawall, com-
placency involves an ‘overestimate of one’s accomplishments’ which produces
‘excessive self-satisfaction’ and ‘an insufficiently strong desire … to maintain …
an appropriate level of accomplishment’ which, in turn, yields a lack of appropriate
action (2006: 345).31 These traits do prevent agents from having the virtue of IP. To
explicate, folly and resignation can cause agents to quit projects that are appropri-
ate for them. Apathy can cause agents to fail to adopt and pursue such projects in
the first place. Finally, complacency can cause agents to inappropriately delay act-
ing, or fail to perform further actions, in pursuit of such projects.32 But cultivating
the virtue of IP isn’t the best way to correct these traits. Arguably, apathy is best
corrected by curiosity, folly by good judgment, complacency by intellectual humil-
ity, and resignation by intellectual pride. Cultivating the virtue of IP is, arguably,
the best way to correct laziness, and the best way to correct vicious quitting, pro-
crastinating, and slacking off that result from laziness. My hope is that this essay
will help to generate further analysis of all of these traits, and of their connections
to one another and to the virtue of intellectual perseverance.33
Notes
1 Privileged agents who encounter relatively few obstacles may fail to realize that
oppressed persons encounter significantly more. Privileged agents should thus be cau-
tious in ascribing vicious forms of quitting, procrastinating, and slacking off to others.
2 Lest readers think that these aren’t obstacles and ignoring evidence is always easy,
Medina argues that some perspectives ‘require an enormous amount of effort to be hid-
den and ignored’ (2013: 34–35).
3 Advocates include Baehr (2011); Montmarquet (1993); and Zagzebski (1996). The
Aristotelian analysis is also called ‘virtue-responsibilism.’ The other dominant analysis
is ‘virtue-reliabilism.’
4 The good judgment described above is a manifestation of phronesis (practical wisdom),
which enables agents to hit the mean in their actions.
5 For further analysis of good judgment, see King (2014: 3787–3789); King (2019: 261–
264).
6 Editorial goals are intellectual ones. They are other-directed, aiming at authors and the
dissemination of their work.
7 For further analysis of the intellectual virtue of IP, see Battaly (2017); King (2014);
King (2019).
8 See Crerar (2018); Tanesini (2018); Baehr (2020). On my view, agents need not be
blameworthy for possessing intellectual vices.
9 Trait possession can be a matter of degree—a person can be a quitter to a greater or
lesser degree. A person who usually quits, even if she doesn’t invariably quit, can still
186 Heather Battaly
count as possessing the trait of quitting, provided that possessing the trait involves
meeting or exceeding a threshold. See Swanton (2003: 24).
10 Of course, an agent might have moral reasons to quit an intellectual project that out-
weigh her intellectual reasons to continue.
11 Our baseline shifts from the default assumption that agents generally adopt appropriate
intellectual goals to the default assumption that agents generally adopt inappropriate
intellectual goals.
12 Thanks to Ian James Kidd for noticing a connection to Astell (1697).
13 For instance, in a repressive, government-sponsored curriculum that re-writes history,
quitting one’s courses might be a virtue of resistance for students.
14 Tenenbaum (2010: 144) likewise distinguishes between ‘irresolution’ and procrastina-
tion.
15 MacIntosh (2010: 69) argues that procrastination is ‘imprudent delay, where one puts
off until tomorrow what one admits would … be better done today.’
16 Kim and Seo (2015).
17 Milton’s belief that he should be working on the paper may even be unjustified.
18 Compare Milton*, who has learned over decades of work that he excels under pressure.
Milton* does intentionally delay working on the article, but because he knows himself
well, he doesn’t believe that he should be working on the article (nor does he feel badly
about delaying). Like Daphne, he doesn’t procrastinate.
19 Does the subjective proposal cast the net too widely? Suppose you are overcommitted
with many goals and falsely (and unjustifiedly) believe that you should be actively pur-
suing them all. You also intentionally delay acting in pursuit of all but one of them. Are
you still procrastinating with respect to the rest? I think so: such agents have several
problems, one of which is procrastination.
20 See also Baker (2010).
21 Aristotle considers such cases at NE.1146a26–30. On inverse akrasia, see Bennett
(1974) and Arpaly (2003: 75).
22 This is the view known as ‘virtue-reliabilism.’ Advocates include Sosa (2007) and
Greco (2010).
23 We can leave open the question of whether quitting implies slacking off. Procrastinating
will not imply slacking off because slacking off requires that one succeed in taking an
easy path whereas procrastinating does not. Mona (above) is procrastinating, but on the
assumption that she is working on her other projects, she is not slacking off.
24 Relatedly, intentionally taking an easy path that in fact reduces the quality of one’s
work isn’t sufficient for slacking off. Agents can meet those conditions while (falsely)
believing that they are being efficient. Such agents have mistaken beliefs, but they are
not slacking off.
25 Alternatively, consider an agent who succeeds in taking an easy path, fulfilling that
intention. But her belief that her work will suffer is false. This agent is (unwittingly)
being efficient, not slacking off (though she has a mistaken belief).
26 See also King (2014: 3789) on lollygaggers.
27 We might also consider the traits of haste and patience. See Dolin and Baehr (ms);
Kawall (forthcoming); Locke (1706, sect. 25).
28 On laziness, see Kidd (2017), Medina (2013: 33, 39), Roberts and West (2015: 2559,
2570). On sloth see DeYoung (2009: ch. 4).
29 See also O’Connor (2018) on idleness, and King (2014: 3786) on indifference. In
Battaly (2015), I call people with apathy ‘slackers.’
30 See King (2014: 3788).
31 On complacency and resignation, see Kawall (2006). Resignation might be common
among students who have fixed mindsets. On fixed and growth mindsets, see Dweck
et al. (2014).
32 Complacency does not entail procrastination. The complacent person over-estimates
her strengths and under-estimates her limitations and thus falsely believes that she
should not act. Whereas procrastinators believe that they should act. Surprisingly, nor
does complacency entail slacking off. Complacent agents are overconfident about their
abilities, self-satisfied with their performance, and believe they have no need to act
further. They think they are doing fine and don’t need to improve. Accordingly, they
won’t realize that they are, e.g., managing their time badly or failing to put in appropri-
ate effort. Unlike slackers, who intentionally waste time and phone it in, complacent
agents won’t intentionally do these things.
33 Thanks to Ian James Kidd, Jason Kawall, Josh Dolin, and Paul Bloomfield for extremely
helpful comments. Thanks also to Don Baxter, Mona Kulkarni Caron, Andrew Cortens,
Charlie Crerar, Kate Elgin, Sam Elgin, Ed Ferrier, Jane Gordon, Seisuke Hayakawa,
Chris Innes, Masashi Kasaki, Brian Kierland, Yuhan Liang, Tracy Llanera, Koichiro
Misawa, Clifford Roth, Kana Sato, Kunimasa Sato, Lynne Tirrell, and audiences at
Keiai University, Boise State University, the Pacific APA, and the University of
Connecticut.
References
Andreou, Chrisoula. 2007. “Understanding Procrastination.” Journal for the Theory of
Social Behavior 37(2): 183–193.
Aristotle. 1984. Nicomachean Ethics. In J. Barnes (ed.), The Complete Works of Aristotle.
Princeton, NJ: Princeton University Press, 1729–1867
Arpaly, Nomy. 2003. Unprincipled Virtue. Oxford: Oxford University Press.
Astell, Mary. 1697. A Serious Proposal to the Ladies, for the Advancement of Their True
and Greatest Interest. http://www.gutenberg.org/files/54984/54984-h/54984-h.htm
Accessed June 2, 2020.
Baehr, Jason. 2011. The Inquiring Mind: On Intellectual Virtues and Virtue Epistemology.
Baehr, Jason. 2020. “The Structure of Intellectual Vices.” In I. J. Kidd, H. Battaly, and Q.
Cassam (eds.), Vice Epistemology. London: Routledge.
Baker, Jennifer A. 2010. “Procrastination as Vice.” In C. Andreou and M. S. White (eds.),
The Thief of Time. Oxford: Oxford University Press, 165–181.
Battaly, Heather. 2014. “Varieties of Epistemic Vice.” In J. Matheson and R. Vitz (eds.),
The Ethics of Belief. Oxford: Oxford University Press, 51–76.
Battaly, Heather. 2015. Virtue. Cambridge: Polity Press.
Battaly, Heather. 2017. “Intellectual Perseverance.” Journal of Moral Philosophy 14(6):
669–697.
Bennett, Jonathan. 1974. “The Conscience of Huckleberry Finn.” Philosophy 49: 123–134.
Crerar, Charlie. 2018. “Motivational Approaches to Intellectual Vice.” Australasian
Journal of Philosophy 96(4): 753–766.
DeYoung, Rebecca Konyndyk. 2009. Glittering Vices. Grand Rapids, MI: Brazos Press.
Dolin, Josh and Jason Baehr. Manuscript. “Intellectual Patience and Perseverance.”
Dweck, C., G. M. Walton, and G. L. Cohen. 2014. “Academic Tenacity: Mindsets and
Skills that Promote Long-Term Learning.” Bill & Melinda Gates Foundation. http://
k12education.gatesfoundation.org/resource/academic-tenacity-mindsets-and-skills-tha
t-promote-long-term-learning/ Accessed June 2, 2020.
Gorney, D. 2010. “Michael Chabon: How to Salvage a Wrecked Novel.” The Atlantic.
http://www.theatlantic.com/entertainment/archive/2010/12/michael-chabon-how-to-s
alvage-a-wrecked-novel/68665/ Accessed June 2, 2020.
Greco, John. 2010. Achieving Knowledge. Cambridge: Cambridge University Press.
Kawall, Jason. 2006. “On Complacency.” American Philosophical Quarterly 43(4): 343–355.
188 Heather Battaly
Kawall, Jason. Forthcoming. “Patience and Sustainability.” In J. Kawall (ed.), The Virtues
of Sustainability. Oxford University Press.
Kidd, Ian James. 2017. “Capital Epistemic Vices.” Social Epistemology Review and Reply
Kidd, Ian James. 2020. “Epistemic Corruption and Social Oppression.” In I. J. Kidd, H.
Battaly, and Q. Cassam (eds.), Vice Epistemology. London: Routledge.
Kim, Kyung Ryung and Eun Hee Seo. 2015. “The Relationship Between Procrastination
and Academic Performance: A Meta-Analysis.” Personality and Individual Differences
82: 26–33.
King, Nathan. 2019. “Intellectual Perseverance.” In H. Battaly (ed.), The Routledge
Handbook of Virtue Epistemology. New York: Routledge, 256–269.
King, Nathan L. 2014. “Erratum to: Perseverance as an Intellectual Virtue.” Synthese 191:
3779–3801.
Locke, John. 1996/1706. Some Thoughts Concerning Education and Of the Conduct of the
Understanding. Indianapolis, IN: Hackett Publishing..
MacIntosh, Duncan. 2010. “Intransitive Preferences, Vagueness, and the Structure of
Procrastination.” In C. Andreou and M. S. White (eds.), The Thief of Time. Oxford:
Medina, José. 2013. The Epistemology of Resistance. Oxford: Oxford University Press.
Montmarquet, James A. 1993. Epistemic Virtue and Doxastic Responsibility. Lanham,
MD: Rowman & Littlefield.
O’Connor, Brian. 2018. Idleness. Princeton, NJ: Princeton University Press.
Perry, John. 2012. The Art of Procrastination. New York: Workman Publishing.
Roberts, Robert C. and Ryan West. 2015. “Natural Epistemic Defects and Corrective
Virtues.” Synthese 192: 2557–2576.
Sosa, Ernest. 2007. A Virtue Epistemology. Oxford: Oxford University Press.
Stroud, Sarah. 2010. “Is Procrastination Weakness of Will?” In C. Andreou and M. S.
White (eds.), The Thief of Time. Oxford: Oxford University Press, 51–67.
Swanton, Christine. 2003. Virtue Ethics. Oxford: Oxford University Press.
Tanesini, Alessandra. 2018. “Epistemic Vice and Motivation.” Metaphilosophy 49(3):
350–367.
Tappolet, Christine. 2010. “Procrastination and Personal Identity.” In C. Andreou and M.
S. White (eds.), The Thief of Time. Oxford: Oxford University Press, 115–129.
Tenenbaum, Sergio. 2010. “The Vice of Procrastination.” In C. Andreou and M. S. White
(eds.), The Thief of Time. Oxford: Oxford University Press, 130–150.
Watson, Gary. 1984. “Virtues in Excess.” Philosophical Studies 46(1): 57–74.
Zagzebski, Linda. 1996. Virtues of the Mind: An Essay on the Ethical Foundations of
Knowledge. Cambridge: Cambridge University Press.
10 Epistemic insensitivity
An insidious and consequential vice
Maura Priest
10.1 Introduction
Imagine a carefully designed symposium session at an American Philosophical
Association (APA) meeting. The session consists of three speakers – two well-
established experts and an up-and-coming youngster. All talks touch on novel
issues, and the commentators were given papers well in advance, allowing time
to write careful and insightful commentary. Unlike most symposia, audience
members were also given papers in advance. Hence, not only are the speakers
well-versed in the subject, but so are the questioners. From this description, it
might seem like things are not only in epistemic order, but that the epistemic envi-
ronment is of especially high quality. Hence, the odds of meaningful epistemic
exchange might seem higher than the average philosophy talk. However, filling
in a few more details, details that might seem prima facie “non-epistemic”, can
turn things around completely. At least, that is the gist of what this paper argues.
Consider these details:
1 The meeting is scheduled from 11am to 2pm.

2 Like many symposium sessions, the first 2.5 hours are without audience
questions.
3 At least half of the scholars in the room are unaware of the cognitive science
that the other half understands in-depth.
4 Two of the three speakers are very impatient persons.
Perhaps, the above list shows some potential for epistemic trouble. Still it might
be surprising to argue that the problematic effects of the above factors can over-
come all the epistemic virtues laid out before. This paper will suggest that the
tendency to overlook contingent factors that influence “epistemic uptake” is not
only a vice, but an especially consequential vice. I call this vice epistemic insensi-
tivity.1 When I say effects are consequential, I mean that there are significant and
negative epistemic effects. If effects are “epistemic”, they influence the acquisi-
tion and dissemination of knowledge, understanding, and other epistemic goods.
Negative epistemic effects influence the aforementioned in negative ways, i.e.,
make the dissemination of knowledge, and so on, less likely.
190 Maura Priest
Before delving into the specifics of epistemic insensitivity, Section 10.2 makes
a case for understanding virtue and vice themselves in a new light. More spe-
cifically, Section 10.2 argues that the harms of epistemic insensitivity are of a
different kind than other traditional epistemic vices. Section 10.3 starts by explor-
ing epistemic insensitivity in general terms. This broad-brush discussion sets the
stage for the next discussion, i.e., exploring the common ways in which epistemic
insensitivity manifests. While there are probably thousands of possible displays of
epistemic insensitivity, I narrow in on four especially common categories: value
insensitivity, expertise insensitivity, physiological insensitivity, and interest insen-
sitivity. Section 10.4 explains that possessing the vice of epistemic insensitivity
does not require the exemplifying of insensitivity in any particular category, and
certainly not the exemplifying of them all, but rather exemplifying any particular
manifestation of the vice consistently enough that it can be understood as a char-
acter trait. Section 10.5 explores the opposing vice and middle virtue that are on
the same spectrum as epistemic insensitivity, i.e., the virtue of sensitivity and the
vice of oversensitivity. Section 10.6 covers potential objections.
10.2 Two types of virtue

Virtue is a type of excellence. But there are many excellences. One debate in
the virtue epistemology literature has been between the two types of excellences
described as “reliabilism” and “responsibilism”.2 These virtues differ, for instance,
insofar as the former are closer to a skill and the latter closer to a character trait.
But one factor they have in common is that both consider traits virtuous insofar
as they aid an agent in acquiring true beliefs, wisdom, knowledge, etc. In other
words, they help the virtue-holder acquire valuable epistemic goods. Such virtues
are admirable and latch onto critical facets of our epistemic lives. One criticism,
however, is this is a rather narrow vision for all epistemic virtue.
Let us think about athletes and athletic excellence. Athletes show virtue in their
craft (i.e., the craft of athletics) insofar as they are successful at setting records,
scoring points, winning games, and otherwise demonstrating their own ability to
excel athletically – in other words, insofar as they, personally, demonstrate supe-
rior athletic skill. This makes an athlete great. Perhaps some athletes also show
related talents, for example, maybe they are at especially good at bringing out
great athletic performance in children learning the craft. We might consider this
a good thing, but it says little about the athlete’s own excellence. In judging great
athleticism, we must know about the accomplishments of the athlete themselves.
Consider a different field of expertise. Let us look at physicians who are
experts in their matter of expertise, i.e., in health. Unlike with athletes, the physi-
cian’s excellence is not explained in how well he or she personally demonstrates
superior health. Indeed, we can imagine a physician who is not only unhealthy,
but one who gets progressively unhealthier every year. This is compatible with the
same physician demonstrating excellence in his field, indeed progressive excel-
lence where each prior success builds on the last. This is because a physician’s
excellence is demonstrated not in improving their own health, but in improving
Epistemic insensitivity 191
the health of others. Physicians manifest excellence insofar as their skills allow
them to cure others, or maintain their health, or suggest lifestyle changes that bet-
ter the patient’s well-being. While some might praise a physician for personally
being in excellent health, perhaps this way the physician “sets a good example”,
this would nonetheless be tangential to the sense of excellence traditionally attrib-
uted to physicians.
I contend that epistemology should care about both senses of excellence just
mentioned. However, for most of its history it seems to have been excessively
focused on the athletic sense of excellence at the expense of the physician sense.
Epistemic virtues (both reliabilism and responsibilism) are typically traits that aid
the virtue-holder’s own acquisition of knowledge, and vices are those that inhibit
such. There has been some talk of physician-like virtues, but they are few and
far between.3 It also seems there has not been a clear enough distinction made
between these two kinds.
With the above distinction in mind, let me make the case for why physician-
like virtues matter. It is a simple case that comes with just a few premises. To
begin: epistemology is a subject concerned with knowledge and understanding,
and thereby the acquisition of knowledge and understanding, i.e., with inquiry.
Sometimes knowledge and understanding are acquired by an agent in complete
independence. Imagine Descartes, for example, sitting alone in his pajamas, in
deep thought about geometric equations. Descartes might acquire some theoreti-
cal knowledge in this fashion. Or imagine Darwin, alone in the wilderness, who
stumbles across a porcupine and quietly observes. In so doing Darwin acquires
knowledge about porcupine behavior. But these cases are not only not exhaustive
of the way we acquire knowledge, they are also not typical. In many if not most
instances, what we know is influenced by our social surroundings, i.e., surround-
ings impacted by other epistemic agents. Testimony is an obvious case, but other
people issue influence in other ways. Friends might serve as a study partner, or
encourage us to join a book club, or entice us into considering a particular ideo-
logical viewpoint (even one which they personally reject). The point is: social
interaction frequently influences knowledge acquisition. Hence, if we are to care
about inquiry and how people acquire various sorts of epistemic goods, then we
ought to care about how some epistemic agents affect the epistemic lives of other
epistemic agents. We should care not only that individual epistemic agents have
habits that help themselves personally, but also that epistemic agents are success-
ful in aiding others to acquire the same. The second is every bit as important as
the first, for as social creatures a meaningful proportion of all our beliefs were
acquired with the help of other agents. In other words, these beliefs were acquired
interpersonally. Hence, I call the physician-like epistemic virtues, interpersonal,
while the athlete-like ones I call personal virtues. Interpersonal virtues can be
thought of as a type of “other-regarding” virtue, i.e., these types of virtues, when
exercised, have the tendency to aid knowledge acquisition in persons other than
the virtue holder.
Responsibilist epistemologists have typically understood epistemic virtues in
an Aristotelian sense, i.e., as lying on a spectrum of extremes. The middle state is
192 Maura Priest
virtue, while each opposing is a vice. Hence if there can be interpersonal virtues,
there can also be interpersonal vices. I am understanding an interpersonal epis-
temic vice as a character trait that negatively influences the epistemic endeavors
of others. This negative impact is manifest by making it less likely that other epis-
temic agents will acquire knowledge, wisdom, understanding, and other epistemic
goods. What I call epistemic insensitivity is primarily a vice of this kind, i.e.,
an interpersonal vice. While at times insensitivity negatively impacts one’s own
epistemic life, the most telling consequences are ones that impact not the agent
himself, but others. In the next section, some of the most common forms of insen-
sitivity are discussed. We will see that the harmful epistemic impact stretches
widely, perhaps more widely than many would have expected.
10.3 Four breeds of insensitivity

10.3.1 Expertise insensitivity
Imagine a conversation between “Fred”, a professor of climate science, and
“John”, his next-door neighbor. Suppose that John has a bachelor’s degree in com-
munication from the local state university (he graduated 20 years ago). John now
has a middle-class job in human resources. He is not sure what to make of climate
change. He expresses this to Fred, noting that he heard in the 1970s that there were
concerns about the ice age that never panned out.4 John also notes that he knows
the earth is billions of years old. Since records of climate were not kept that long
ago, he does not see how anyone can know that today’s climate is a pattern that
is really different from the past. His neighbor Fred begins with a genuine desire
to educate his neighbor, and hopefully turn him away from doubt into belief. The
problem, however, is Fred uses a number of scientific concepts that John does not
understand.5 The concepts are second nature to him, and seem obvious. But they
are not obvious to his neighbor. Hence very early in the conversation John starts
to get confused, and with confusion he brings up objections. Fred is annoyed and
points out that earlier comments already explain those objections away. At this
point, it should be obvious to Fred that his conversation partner is lost. But to
whatever extent Fred might be aware of such, he does not take the opportunity to
correct his earlier mistake, i.e., he does not go back and explain the concepts in
simpler terms. Instead, he moves on to other concepts that are equally confusing.
Fred first shows insensitivity by not assessing John’s background knowledge.
When experts know that they have more background knowledge than their inter-
locutor, the sensitive response (before getting into deep conversation) is to make
sure they are starting on the same page. It is only by knowing where the other
starts from that an expert can effectively teach the non-expert. But skipping this
initial assessment seems a common mistake. Besides, whether common or not, it
is a mistake that there are epistemic reasons to avoid. If this assessment is skipped,
the odds of importing knowledge drop precipitously.
The second place where Fred shows insensitivity is in missing that John got lost
mid-conversation. And Fred ought to have realized this; the line of questioning
revealed as much. A properly sensitive agent would immediately perceive ques-
tions as key pieces of epistemic information along the lines of: “The knowledge I
want to convey is getting lost in translation. I need to stop, back-up, and make sure
that these translation mishaps are corrected”. As the conversation goes on, there
might be further cues that John has missed a point or gotten confused. Sensitive
and insensitive agents will have opposite reactions. The former will take these
cues as a reason to pause, rephrase, and reassess. In other words, they take action
to make sure that their testimony is not merely heard but understood. The insensi-
tive agent is oblivious to these cues. As such the insensitive agent will carry on
as if his conversation partner were latching on to every word. An opportunity to
disseminate knowledge is wasted. This is expertise insensitivity.
Expertise insensitivity is obviously a detrimental vice of a teacher. Educators
unaware of lost students not only fail to teach in that instant, but this vice has con-
sequences for the next semester as slides, assignments, and class activities remain
unchanged. It is also unfortunate to be an audience member of an academic talk
featuring an insensitive speaker. It is quite common, after all, for academic talks
to be attended by non-specialists. The epistemically sensitive will speak to reach
all, while the insensitive speaks to a room full of glazed-over eyeballs. And it is
not only in traditional education settings that problems arise. As the example with
John and Fred shows, everyday conversations can also be plagued by expertise
insensitivity. If an agent manifests this type of vice regularly and frequently, i.e.,
if an agent has the vice of epistemic insensitivity, the potential for lost knowledge
can be substantial.
What about the non-expert? Can they show epistemic insensitivity, or more par-
ticularly, something called “expertise insensitivity”? Yes, on both accounts. Let me
first make clear that in any two-way conversation (i.e., most conversations, maybe
all depending on how you define conversation) it is ideal that both parties are epis-
temically sensitive, and detrimental when either party is epistemically insensitive.
However, epistemic insensitivity is perhaps most problematic when exhibited by
an expert talking to a non-expert (because of the critical missed opportunity for
teaching and knowledge transfer.) Even so, epistemic insensitivity is always an
epistemic negative in any conversational setting. And expertise insensitivity can
stretch beyond the insensitivity of the expert. Expertise insensitivity is simply
being insensitive to the epistemically important role that expertise plays in a con-
versation. The non-expert can show epistemic insensitivity to an expert, with unfa-
vorable epistemic results (and regardless of the results, it is disrespectful.) If, for
instance, I (someone without a medical degree) am talking to a physician, I might
show expertise insensitivity by talking about the pharmacology class I took as an
undergraduate and why that class justifies rejecting the medication that my physi-
cian recommends. This shows insensitivity to the fact that my physician might be
insulted by my presumption that my own “expertise” matters more than his, in
spite of his superior qualifications. An epistemically sensitive agent could share the
same information with the physician, but in a more modest tone that would respect
the physician’s intellectual authority, for example, “I’m curious what you would
say about something I remembered from an undergraduate class…”
194 Maura Priest
Insensitivity can be due to either ignorance or lack of concern. You can notice
that interlocutors are getting lost while just not giving a damn, or you might be
oblivious to them getting lost in the first place. Both are insensitive. What is com-
mon to both is the lack of response to epistemic feedback.
I will not argue for this in depth, but I think it is a common feature of virtues
and vices generally that they can manifest via distinct components. In this regard,
epistemic insensitivity can be thought to have the following components:
1 Failing to recognize other agents’ epistemic concerns (which can be owing to

either ignorance or lack of concern).
2 The failure to respond to, or properly take into account, the epistemic con-
cerns of other agents. (If you are ignorant, you almost meet this by default.
However, agents can be aware of concerns but do nothing about them.)
When such virtues and vices have component manifestations, posses-
sion of the virtue or vice does not depend on the agent manifesting all parts.
Epistemically insensitive agents can have both component features, but they
might only have one. Agents that possess component 1 will typically, almost
by default, also have 2, i.e., those unaware of an agent’s concerns will either
not respond to them, or only respond incidentally.
10.3.2 Value insensitivity

When considering the best way to spread information, it is helpful to remember
epistemic agents are not purely rational robots. As such, the following types of
considerations which we would not need to worry about with robots, we will need
to worry about with persons:
1 Whether persons believe their own viewpoint is being given a fair shake.
2 Whether persons believe their deepest held values are being attacked.
3 Whether the testimony persons hear fits with their cultural upbringing.
Epistemically insensitive agents tend to ignore the importance of 1–3. Or said

more exactly, one way of manifesting epistemic insensitivity is to ignore 1–3.
Moreover, those who behave in such careless ways are also prone to blame the
victim, i.e., to claim that if the relevant conversation partner or audience member
learned nothing from the epistemic exchange, the blame for ignorance must fall
on the ignorant. Consider the falling conversation:
Sandy: I’m a big fan of presidential candidate Smith.

Jim: How could you say that? Someone with such outlandish views on animal
rights can never get elected.
Sandy: I agree with him on animal rights.
Jim: But don’t you get it, most people don’t because your views are extreme. If we
want our party to win in November we can’t vote for Smith.
Sandy: My views aren’t extreme and I’m going to vote for him regardless.
Jim: It is because of people like you that we lose the election. You are putting your
emotions ahead of rationality.
There are many things we might say about the above conversation. But one of
the first, it seems, is that the following two statements might both be true: (1) Jim
is correct that candidate Smith has no chance of winning in that district, but also
(2) Jim is epistemically culpable for behaving insensitively. If Jim had refrained
from attacking Sandy’s values, Sandy would have been more inclined to consider
Jim’s point seriously and perhaps think twice about whether candidate Smith is
the best strategic choice. But the insensitivity put Sandy on the defensive, and
she was no longer in a position to calmly reflect on Jim’s claims. (Although Jim’s
claims are about strategy, they are still epistemically loaded as they are also truths
in the empirical world.) Jim might have been correct that Sandy was putting her
emotions ahead of rationality. But he doesn’t consider that this is partly his fault,
and that even if Sandy is failing epistemically, he might be failing epistemically
as well.
Different types of value questioning can overlap: one person’s deeply held per-
sonal values might conflict with another’s cultural upbringing. Consider someone
who was raised hunting and eating meat, call her “Jenny”. Perhaps Jenny is not
deeply committed to eating meat, but this is just what she is familiar with and
considers typical. An insensitive person might start a conversation with Jenny by
condemning all meat eaters as grossly immoral. Jenny presumably does not con-
sider herself or her family grossly immoral. When the conversation turns in that
direction, the plusses and minuses of vegetarianism are put to the sidelines. Jenny
is likely to become focused solely on the defensive. An epistemically sensitive
person would refrain from this type of characterization in order to focus on the
ethical arguments about meat eating, the pain animals suffer, and so on. The sensi-
tive interlocutor is going to have higher odds of widening Jenny’s understanding
about animal ethics.
Epistemic failures need not be directly tied to bad personal epistemic habits.
They can be tied to bad interpersonal habits. And bad interpersonal habits include
unawareness of the way presentation styles affect epistemic uptake. Epistemically
sensitive agents are known to do right even when their conversation partner does
wrong.
At this point, one might wonder whether epistemic insensitivity requires the
help of other virtues like civility and respect for others. Or perhaps, whether by
showing epistemic sensitivity, (especially value sensitivity) you thereby also
manifest civility and respect. In this sense, I see a disanalogy between the virtue of
epistemic sensitivity and the vice of epistemic insensitivity. Epistemic insensitivity
is a form of disrespect. It is a failure to give another epistemic agent the attention
and concern they deserve in virtue of their membership in a community of rational
equals. “Rational equals” does not imply that everyone is “equally rational”, nor
that we are all equally skilled epistemically. There is nothing insensitive in recog-
nizing expertise and epistemic differences. “Rational equals” suggests that adult
196 Maura Priest
humans (barring some particular cognitive disabilities) are rational creatures, and
this rationality explains our equal political and moral standing. The right to be an
intellectually active community member should not depend on being an expert,
but only passing the bar of rationality. When an epistemically insensitive agent
fails to acknowledge another’s intellectual input, they disrespect the other’s status
as a member of the rational community.
Epistemic sensitivity, to some extent, does manifest intellectual respect. But
there are so many ways to be disrespectful that epistemic insensitivity is not a
foolproof shield. After all, an agent might carefully listen to another, but shortly
thereafter question their motives aggressively. This agent failed to properly
respect their interlocutor, despite first showing epistemic sensitivity. This, I sus-
pect would also be a form of uncivil behavior. Hence, epistemic sensitivity (the
virtue), while itself likely enhances civility, other things being equal, need not
entail civility.
What about persons who have entire systems of false belief? Consider some-
one who believes the first moon landing was staged. This false belief is likely
accompanied by many other false beliefs, such as a government conspiracy which
explains why there was incentive to falsify the landing.6 Or consider someone
who believes vaccines cause autism. This can be accompanied by other question-
able beliefs, such as the belief that almost all physicians and medical researchers
are nothing but pawns of “big pharma”. As false as these beliefs are, there are
sometimes good reasons to “hear the other side out”. This is because people who
do not believe they are being heard are unlikely to hear you. Hence, in private
conversation with a conspiracist theorist, one can do epistemic good by hearing
out false ideas.7
Now some might object as follows. First, “those taken in by conspiracy have
little chance of changing their mind”. At times this might be true, and it is up to
each epistemic agent to consider whether the person with whom you engage is
worth the conversation. But still, there are those who once believed conspiracy
theories who then changed their minds. Change is possible. Even if you are not
the mind-changer in a single conversation, you might start the process of recon-
sidering ideas. But you are unlikely to do this by dismissing the conspiracy theo-
rist’s views. Dismissiveness is likely to cause anger and resentment. Moreover,
it shows you do not see them as a fellow epistemic agent capable of judging and
responding to evidence. If they are capable of judging and responding to evi-
dence, it might do good to hear why they believe as they do. Understanding this
might make it easier to respond in effective ways. I once had a conversation with
someone convinced that socialism would quickly lead to a civil war. I listened
to their reasons, and it soon became clear that they had a completely inaccurate
perception of what socialism actually consisted in. Once they understood, I could
immediately see them reconsider their views.
A second objection is that: “even considering outlandish views gives false cre-
dence to nonsense, thus leading more people to take them seriously”. This might
be true in some circumstances where the audience is vast and impersonal, i.e.,
on radio, in newspapers, on television, and other forms of mass media. Often it
is best to not give conspiracy theories serious consideration, for doing this leads
others to think the ideas are themselves serious. But there are more caveats to
this objection than I usually hear others admit. First, the following two epistemic
stances are possible: (1) not discussing outlandish ideas in a serious way, while
(2) not being dismissive of those who hold these ideas. If the ideas are really
outlandish, and therefore not taken seriously, there is no need to mention them at
all. If they are outlandish but a large portion of the populace does take them seri-
ously, then it may be worth giving the view some air time to get that large swath
of the population to engage in conversation. There is not always an easy answer,
but dismissiveness is rarely justified epistemically. Lastly and most importantly,
everyday conversations are not always aired on mass media. When talking to a
single person or small group, there is little reason to worry that considering their
ideas will lead to any type of society-wide influence.
The claim is not that there is never cause, or never epistemic cause, to question
values. Rather the claim is that the epistemically sensitive person is aware of the
costs of doing so, and the insensitive person is not. Moreover, the epistemically
sensitive person considers the costs of questioning commitments, while the insen-
sitive person does not. All of this is compatible with questioning the aptness of
commitments in certain circumstances.
10.3.3 Interest insensitivity

Consider a situation similar to our opening example. Imagine a philosophy book
session, with three speakers and three commentators. The book topic is philoso-
phy of medicine. The audience and commentators consist of both medical profes-
sionals and philosophers. Suppose, also, that the format session involves nearly
three hours of speaking before any engagement from the audience. Further sup-
pose that the philosophers speak using highly technical philosophical language.
Likewise, the medical professionals use highly technical medical language. The
session goes on.
In the described situation, various physiological and interest insensitivities are
at play. Let us start with interest insensitivity. This is the type of insensitivity that
fails to notice that interest in the relevant epistemic topic is waning. It is uncon-
troversial that epistemic uptake improves with interest in a subject, and there is
plenty of scientific literature to back this up.8 Being bored comes with an epis-
temic cost. Epistemic sharing with someone who isn’t listening is not sharing at
all. Hence, the first problem with the philosophical book session is its length in the
absence of breaks or discussions. Even when persons are interested in a subject,
three hours of listening to complex material will strain the focus of most. Even
more, a room is never full of “the most attentive”. Some are more attentive than
others, and in any large audience there will be some members who have particular
trouble with attention span, and others who are in-between. Hence, neglecting to
consider the length of speaking times is neglecting to consider an important epis-
temic consideration, i.e., the epistemic purview of all audience members. It is not
only epistemically misguided, it is straightforwardly exclusive.
198 Maura Priest
10.3.4 Physiological insensitivity
Like interest insensitivity, physiological insensitivity impedes epistemic concen-
tration and thereby the acquisition of epistemic goods. However, physiological
insensitivity is completely divorced from epistemic and intellectual issues under
discussion. It is possible that a talk is well-prepared, gripping, easy to understand,
and yet audience members take in little information. Physiological insensitivity
might explain the problem. Imagine, for instance, that an expected conference
breakfast never arrived. Moreover, the last talk before lunch is scheduled to con-
clude at 1:00pm. Most in the room are very hungry, and most are aware that most
others are very hungry. Nonetheless, suppose the speaker continues to talk past
the allotted time, ignoring cues from the chair. The speaker, by showing insen-
sitivity to the audience members’ physical needs, is thereby guilty of epistemic
insensitivity. And this epistemic insensitivity has consequences: hungry scholars
focusing on their upcoming lunch are distracted, not playing their “A” game, and
so on. There are, of course, other examples. An especially hot or cold room makes
learning more difficult and uncomfortable, as do bright lights, loud noises, and a
microphone that is not well-programmed. All of these things might seem minor,
and hardly vicious. But I would argue otherwise. While in any given instance
the consequences are rarely extreme, a general intellectual culture of ignoring
physiological needs adds up: bits of knowledge sharing lost here and there can be
substantial over time.
10.4 On different manifestations

At first it might seem odd to categorize different manifestations of a single vice.
If they manifest in these different ways, one might wonder, are they really the
same vice at all? Upon reflection, however, it seems we can say something similar
about all vices, i.e., that there are some common ways the vice might manifest,
even though these common ways are not exhaustive. For instance, consider greed.
One common display is refusing to donate money to charitable causes. Another is
never giving even a little of your time to those who need it. A third common form
of greed is an obsession with accumulating wealth. Agents can be greedy and
manifest only one of these displays of greediness, or they can manifest them all.
What matters is that however greed might be manifested, the greedy person main-
tains a disposition to show greed itself, no matter what form. Likewise, someone
can be epistemically insensitive by only manifesting one of the aforementioned
types of insensitivity, or by manifesting them all. What matters for possessing the
vice is that there is a consistent disposition to manifest insensitivity, in whatever
form.
But if all of the types of insensitivity mentioned are manifestations of the same
vice, what is the use in distinguishing between them? Well, if we know the com-
mon forms of insensitivity, then it seems more likely we will recognize insensi-
tivity when we see it. One problem with certain types of vices like insensitivity
is that they are under-discussed and under-recognized. We might just shrug our
shoulders at insensitivity when we ought to criticize it. By describing particular
instances of insensitivity, we make it easier to call out these vices (in ourselves
and others) and thereby correct for them. In addition, by using different examples
of insensitivity we can see what all types have in common. If we were to only
focus on one kind of insensitivity, we might mistake non-essential features as
part of insensitivity itself. But when we see common features across different
types of attitudes and behaviors, we can identify those features as essential to
the vice itself. With insensitivity, we see the way that a refusal to take account of
contingent epistemic features can have negative epistemic consequences and be
disrespectful. Even without the bad consequences, epistemic insensitivity does
wrong to other epistemic agents in not giving them the respect they deserve as
fellow members of the epistemic community.
Speaking of recognizing vice across cases, some might argue that what I am
calling epistemic insensitivity is just the manifestation of other vices. Perhaps
poor listening skills are just incivility, arrogance, or being uncharitable. I will
first say that this criticism can be launched at almost any vice or virtue. Maybe
we do not need a vice called “selfishness”, when we already have, say, greed and
arrogance. However, what objections like this are missing is that vices can be
distinguished from one another insofar as their components relate to each other in
unique ways, insofar as the components are more or less central to the vice, and
insofar as the components of the vice might or might not require a certain degree
of intensity. The components of the vice are only part of the vice: how they fit
together is of equal conceptual importance.
Many would say that selfishness and greed are distinct vices. However, there
is significant overlap, i.e., component features that contribute to the possession of
the vice of selfishness are sometimes the same component features of greed. What
makes the vices distinct is the specific role played by each component feature.
For instance, “putting self above others” might be a component feature of both
greed and selfishness. Yet this component is more central to selfishness. Likewise,
“obsession with wealth accumulation” is more central to greed. Likewise, the vice
of passive aggressiveness arguably shares component features with the vices of
manipulativeness, dishonesty, and cowardice.
Despite overlaps of component features of many virtues and vices, distinguish-
ing traits seems worthwhile. Using the description “he is passive-aggressive” has a
communitive power that effectively portrays a certain descriptive content. Maybe
you could offer the same descriptive content by saying, “he is manipulative in a
way that shows both dishonesty and cowardness”. Sometimes these alternative
descriptions work well. But often they fail to capture the same content. Even
when vices are conceptually close, differentiating them helps us communicate.
The above said, I will briefly layout key differences between epistemic insen-
sitivity and other vices. Let me start with incivility. Incivility seems both broader
and narrower than epistemic insensitivity. It is broader because there are ways
to manifest incivility that do not bear any relation to epistemic insensitivity.
Responding to another agent rudely and aggressively is usually uncivil. But rude-
ness and aggressiveness bear less relation to insensitivity. Suppose that I know
200 Maura Priest
my friend Alex is unusually competitive. Alex wants to be engaged in a task with
a goal, not passively contemplating. Knowing this, I engage in an aggressive con-
versation that awakens Alex’s competitive side. In caring about bringing out the
intellectual best in Alex I manifest sensitivity. In this way, incivility requires more
than epistemic insensitivity.
I know many whose intellectual style is always civil. They are always polite,
they always pause to listen to the response of their interlocutors, they never make
rash accusations about motives, etc. Yet they are still epistemically insensitive.
While they listen to their interlocutors, they also treat them all the same. When
attending conferences, they have the same style of engagement, disregarding the
differences in the communities and cultures. This is civil but not sensitive. In this
way, epistemic sensitivity is more demanding than civility.
I hope it is becoming clear how virtues and vices that might initially seem to
have much in common can also be different in critical respects. I will finish with
the difference between epistemic arrogance and epistemic insensitivity. While the
epistemically arrogant are often epistemically insensitive, they need not be. An
epistemically insensitive agent need not be arrogant, for their insensitivity might
be explained primarily in terms of ignorance and carelessness. But epistemic
arrogance requires a type of mindset that cannot be explained by carelessness
or negligence alone. Lastly, imagine a vice called something like “uncharitable-
ness”, consisting in the disposition to interpret others uncharitably. This is a vice,
indeed, but not the vice of epistemic insensitivity. The epistemically insensitive
do not actively attempt to misrepresent others, they just fail to pay proper atten-
tion to contingent features that might sometimes result in unintentional misrep-
resentation. Misrepresentation is an incidental, not essential, feature of epistemic
insensitivity.
10.5 A spectrum of sensitivity

Aristotelian vices, like insensitivity, lie on a spectrum. One consequence of this
spectrum view of virtue and vice is that there can be borderline cases, i.e., cases
in which it is unclear whether someone has the vice or not. To possess epistemic
insensitivity requires not merely showing insensitivity on occasion, but that one is
disposed to show insensitivity in a wide range of circumstances. This, of course, is
consistent with not showing it for years at a time, or potentially ever. Disposition
is not action. I am arguing that epistemic insensitivity is an interpersonal vice,
hence manifesting this vice requires the presence of other agents. We can theo-
retically imagine an isolated epistemic agent who is disposed to insensitivity, but
never manifests it, for there are simply no interpersonal opportunities.
We’ve already mentioned the virtue on insensitivity’s spectrum, namely sensi-
tivity. While the insensitive person is disposed to ignore contingent circumstances
that influence the acquisition of epistemic goods, the sensitive person is disposed
to the opposite, i.e., to be aware of the contingent circumstances and to respond in
a way that mitigates the obstacles to knowledge (or understanding, wisdom, etc.).
What then, is the opposite vice? It might seem that it is hard to be “too aware” of
contingent circumstances. However, there are ways in which agents can indeed
be oversensitive to contingent features. Consider, for instance, value sensitivity.
An insensitive agent ignores his interlocutor’s values and makes no effort to par-
ticipate in the discussion in a way that respects these values. The sensitive agent,
on the other hand, is acutely aware of the relevant values and manages to steer the
conversation so that not only the manifestation of blatant disrespect is avoided,
but also (insofar as showing respect might forward epistemic ends), respect is
displayed.
Oversensitive agents take sensitivity to an unproductive extreme. They are so
concerned with contingent factors that they fail to focus on the epistemic. Imagine,
for instance, someone who is so concerned with not offending that they dance
around topics or hide the truth. Suppose Tina suspects that Jane was upset because
three months ago, her car was broken into. Tina and Jane are talking local politics.
Tina, however, refuses to bring up anything even remotely related to crime out of
fear of hurting Jane. Not only does this fear prevent productive discussion, but it
is also overwrought insofar as Jane would be fine talking about these things.
In this sense oversensitivity is patronizing: oversensitive agents assume their
interlocutor is incapable of handling a wide variety of honest epistemic inquiry.
Suppose a professor cancels his class because the university football team lost a
game. This professor is not giving his students enough credit for their ability to
focus in non-ideal situations. Another way of showing oversensitivity is to agree
with everything someone says out of fear of hurting or offending. Suppose, for
instance, that Betty is a devout Catholic and her friend Jim is an atheist. When
Betty talks about her faith, Jim often pretends to agree so as to avoid offending
Betty’s sensitivities. But this lacks due respect to Betty as an epistemic agent. Jim
not only withholds truths from Betty, but he also refuses to engage in genuine
intellectual discussion. Presumably, if Jim did trust Betty as an epistemic agent,
then such engagement wouldn’t be a problem. The key to hitting the middle ground
between insensitivity and oversensitivity is to simultaneously respect one’s intel-
lectual partners as autonomous capable thinkers while also realizing they are
autonomous, capable, human thinkers who have their frailties. The insensitive
agent forgets about the human part. The oversensitive agent forgets about the
autonomous thinker part. The sensitive agent gets things just right, while also
being cognizant of differences between persons.
10.6 Objections
Some might object to my description of epistemic insensitivity as a vice. I imag-
ine one objection might be this: by insisting that epistemic insensitivity is a vice,
I encourage indulgent epistemic behaviors, i.e., listening to people who shouldn’t
be listened to. After all, these persons are unreasonable and there is no hope of
changing their minds. Likewise, discouraging what I call insensitivity encourages
people to listen to those who aren’t only wrong, but perhaps purposely spreading
false information. Not everyone, the critic might contend, is worthy of epistemic
respect.
202 Maura Priest
Let me repeat what epistemic sensitivity is not: epistemic sensitivity does not
mean listening to persons in all circumstances, no matter what. It does not demand
engaging with trolls or anyone else out to cause epistemic (or just ordinary) trou-
ble. There is a wide range of situations in which an epistemically virtuous person
would choose not to have a conversation. One reason might simply be that he is
too busy with other tasks. The virtuous person knows his time is limited, and uses
it in morally virtuous ways. Likewise, the epistemically virtuous person knows his
epistemic time is limited and does not waste it on fruitless endeavors. Engaging
trolls is not epistemically sensitive, but oversensitive. Oversensitivity can be man-
ifest by going too far in trying to listen to someone who clearly has subversion
as his intellectual goal. The insensitive person, on the other hand, dismisses intel-
lectual engagement for weak or non-compelling reasons. The sensitive person
knows when to engage.
The key point is that the epistemically sensitive person need not listen to every
person in every circumstance. What matters is that when the epistemically sensi-
tive person does choose to engage he does so in a way that is free of the vices of
insensitivity and oversensitivity. Thus: if there is going to be conversation at all,
it should be done with sensitivity. If there is no point in trying to be sensitive,
then there is probably no point in trying to have the conversation in the first place.
Epistemic resources would be better spent on other things.
Notwithstanding the just mentioned clarifications, some might argue there are
still cases in which it makes sense to engage with a speaker in straightforwardly
insensitive ways. Perhaps someone has views that are so harmful, it is to the ben-
efit of others that those views are dismissed out of hand, and even dismissed
arrogantly. My first response is to point out that we must distinguish between
epistemic and moral benefit. The type of insensitivity that concerns me is epis-
temic sensitivity. Hence, it is possible that epistemic goals and moral goals con-
flict, and that there may be cases in which there is some moral benefit in acting
with epistemic insensitivity. It does not follow, however, that there would be epis-
temic benefit.
The critic might say: but sometimes there is epistemic benefit to acting in ways
that are generically insensitive. My response: yes, perhaps. It is compatible with
my thesis to admit to unique cases in which epistemically insensitive behavior
leads to epistemically beneficial results. Because we are talking about an epis-
temic vice and not an epistemic rule, this possibility does not conflict with my
claims. Consider an analogy. Perhaps being cowardly sometimes results in moral
good. It does not follow that being cowardly is virtuous, much less that moral
persons are cowardly. Likewise, even if there are rare circumstances in which it
makes epistemic sense to be insensitive, it does not follow that insensitivity is an
epistemic virtue. The general disposition toward epistemic insensitivity makes
one a worse epistemic agent overall. It also happens to be a vice that tends to have
especially harmful epistemic consequences. This is compatible with there being
positive consequences in special circumstances.
A different objection might contend that this vice I describe is not “essentially”
epistemic. The thought might be that if a character trait is not directly related to
the acquisition of knowledge then it is not “really” epistemic. Along similar lines,
one might take issue with an epistemic virtue primarily affecting other epistemic
agents. Let me address each issue in turn.
First, I am unconcerned about what is “essentially” epistemic. It is unclear
to me what that means. I am concerned with what is generally epistemic. The
tendency to question virtues and vices that do not directly relate to knowledge
acquisition has a long history. For decades epistemology was nearly exclusively
focused on justification, knowledge, and belief. This narrow focus was at the
expense of other epistemic goods and activities. Yet if knowledge is epistemic, it
only makes sense that so too are the activities which lead to knowledge acquisi-
tion. In other words, inquiry is epistemic. Moreover, joint inquiry is no less epis-
temic simply because it involves multiple agents. If knowledge is epistemic, so
is inquiry, and so then is joint inquiry. This paper is focused on joint inquiry and
how it can be done well or badly, or in ways likely to promote epistemic ends or
likely to inhibit them.
As far as my focus on agents other than the virtue holder, return to my earlier
division of physician and athlete virtue. Some want to deny that there is any-
thing equivalent to physician-like virtues in epistemology. But again, let us go
back to the beginning, i.e., what is epistemology? At minimum, epistemology
is not only about knowledge but also justification and belief. If justification and
belief are part of epistemology, it seems epistemology is not only about knowl-
edge but related concepts. Inquiry seems a related concept, and it is hard to deny
that inquiry is often done with others. Even more, it is at least plausible that other
agents have as great an influence on our own knowledge acquisition as we do
ourselves. If this is true, then it seems those who care about epistemology should
care about habits which increase the odds that person-to-person interactions will
result in favorable epistemic outcomes. And caring about this is just caring about
a special class of epistemic virtue (interpersonal) and vice, i.e., traits that influ-
ence other persons and their pursuit of epistemic goods.
In one sense, it is of non-critical importance whether we call epistemic sensi-
tivity and insensitivity “epistemic” or give it some other name. What matters is
the phenomenon, what we call it is secondary. Yet in another sense, there is some
value in putting these traits in the same class as traditional, individually centered,
epistemic traits. Because doing so might encourage epistemologists to focus on
these traits, and give them needed attention. This seems especially important
since these types of other-directed traits have been overlooked. Lastly, there is
no principled reason to prefer athlete-type traits over physician-like traits. Both
are of central importance when it comes to the acquisition and dissemination of
epistemic goods. If the overall epistemic state of society matters (and not just a
few epistemic super stars), then other-directed virtue and vice should be front and
center in epistemic discussions.
Should we (epistemically) blame those who are epistemically insensitive in a
broad sense of “blame”? I would say “yes” in the same way that I would regard-
ing their other epistemic vices. We should blame the vice in the abstract, but not
blame the particular persons who possess the vice, even though they are, indeed,
204 Maura Priest
blameworthy in a broad sense. Individually, this means we hold both the abstract
belief that “epistemic insensitivity is bad”, and also the belief that individuals with
these traits are blameworthy, i.e., we recognize that having the vice makes them
a worse epistemic agent. Blaming in the abstract, on a social level, means that
conversations between friends, family, and coworkers recognize and condemn
epistemic insensitivity. Consider the following:
“I admire the bravery of the victims who came forward”.

“The greed of the administration in using that donation for their own salary is
outrageous”.
“Laughing at Jacob for crying in the office is awful, just heartless. Everyone
knows he just lost his brother”.
The above are instances of publicly praising virtues and blaming vice. We can
imagine further examples where such blame and praise are expressed in public
fashion, i.e., through a media source. Such blame and praise shape our social per-
ceptions of what is and is not acceptable. Ideally, epistemic insensitivity should
be blamed in this fashion. A caveat is that the direct (face-to-face) expression of
blame is usually unhelpful. Persons do not respond well to this type of criticism.
Notwithstanding, persons do seem to respond to social approval and disproval,
understood abstractly, i.e., we are less inclined to do what is socially unacceptable.
Public condemnation plays a role in shaping the socially acceptable. The public
condemnation of epistemic insensitivity makes it less socially acceptable. Making
it less socially acceptable makes it less common. This is epistemically helpful.
10.7 Conclusion
When we enter into epistemic engagement, we might think that what matters most
are true beliefs. And indeed, true beliefs are important. If we lack true beliefs, we
cannot share them. But this is only part of the story. There is a host of contin-
gent, and not directly epistemic, features of discussion, conversation, and mutual
epistemic inquiry that can have great impact on the extent to which knowledge
and other epistemic goods are exchanged. These contingent features can be just
as impactful as other features more “directly” epistemic. Excellent analytic rea-
soning ability might directly influence knowledge acquisition. But suppose this
excellent reasoner is epistemically insensitive. Someone more sensitive, but less
analytically skilled, might improve the epistemic world far more than our insensi-
tive genius. Humans are not epistemic angels; contingent features which ought
not impact knowledge acquisition do, in fact, influence it. Hence, if we care about
imparting epistemic goods in the real world and not just the ideal world, we should
care about epistemic insensitivity.
In arguing that epistemic insensitivity is a vice, I have argued that not all epis-
temic vices affect an agent’s own epistemic state. Or at least, they need not. A
different class of vices concerns how an agent’s own epistemic habits influence
epistemic goods acquired by others.
Notes
1 Woomer (2017) writes about “agential insensitivity”. While the language we use is
similar, the concepts are very different. In particular, my concern is about how an agent
shows insensitivity to other agents, not to evidence. In addition, the consequences of
insensitivity, on my account, fall mostly outside of the insensitive agent.
2 A lot has been written on reliabilism vs. responsibilism. Some examples include Greco
2000, 2011; Baehr 2006; Battaly 2008; and Fleisher 2017.
3 Perhaps one of the articles that gets closest to what I’m getting at here and in my
2017 piece (“Intellectual Humility: An Interpersonal Theory”) is Kawall 2002. Anther
article along these lines is Driver 2003. Driver says the best way to distinguish between
moral and epistemic virtue is in their consequences, i.e., epistemic virtues have posi-
tive epistemic consequences. This is similar to my justification for physician-like epis-
temic virtues, i.e., that we should care about good epistemic consequences overall, and
many overall epistemic consequences concern how our personal epistemic behaviors
have interpersonal consequences. Other articles which touch on the altruistic side of
epistemic virtue include Lackey 2018; Fallis and Whitcomb 2009; Fallis 2004; and
Paternotte and Ivanova 2017.
4 While there were some scientists who predicted an ice age for the 1970s, this was far
from a consensus view and many other scientists predicted warming. A good summary
of the myth and facts can be found in Kessler 2015.
5 Carel and Kidd (2014) discuss a similar problem in regards to the healthcare field.
Carel and Kidd argue that medical professionals do their patients an epistemic injustice
by using medical terminology which they cannot possibly follow. I would say this is an
example of a physician showing insensitivity to his patients and both their epistemic
and healthcare needs.
6 Philosophical work on conspiracy theories includes Sunstein 2009; Coady 2003,
2007a, 2007b, and 2019; Graumann and Moscovici 1987; Keeley 1999; Dentith 2014;
Raikka 2009; Pigden 2007; Clarke 2007; and Levy 2007.
7 Burroughs and Tollesfsen (2016) argue that children are victims of Miranda Fricker’s
type of epistemic injustice. We tend to distrust children without good reason, based
merely on the fact they are children. While being a child might give us reason to
think someone has less overall knowledge and experience, this isn’t necessarily rea-
son to dismiss them out of hand. We owe them the epistemic service of listening. I
would say something similar about the uneducated conspiracy theorists, and those
who hold unreasonable beliefs. We might dismiss all testimony from these types out
of hand, which is unfair for several reasons. First, people can be wrong about many
things and yet right about a few things (without listening we don’t know what they
are right about.) Moreover, even if people are wrong, they might have good reasons
for their beliefs. And lastly, it simply seems a type of epistemic injustice to dismiss
out of hand, and not even offer the basic epistemic courtesy of listening. Aberdein
(2014) discusses similar ideas, i.e., he stresses the importance of listening in argu-
ment.
8 A sampling of this literature includes Tobias 1994; Renninger 2015; Schiefele 1999;
Kpolovie 2014; and Silvia 2008.
References
Aberdein, A. (2014). In defence of virtue: The legitimacy of agent-based argument
appraisal. Informal Logic 34(1), 77–93.
Baehr, J. (2006). Character, reliability and virtue epistemology. The Philosophical
Quarterly 56(223), 193–212.
Battaly, H. (2008). Virtue epistemology. Philosophy Compass 3(4), 639–663.
206 Maura Priest
Burroughs, M. D., and D. Tollefsen. (2016). Learning to listen: Epistemic injustice and the
child. Episteme 13(3), 359–377.
Carel, H., and I. J. Kidd. (2014). Epistemic injustice in healthcare: A philosophical analysis.
Medicine, Health Care and Philosophy 17(4), 529–540.
Clarke, S. (2007). Conspiracy theories and the internet: Controlled demolition and arrested
development. Episteme 4(2), 167–180.
Coady, D. (2003). Conspiracy theories and official stories. International Journal of Applied
Philosophy 17(2), 197–209.
Coady, D. (2007a). Are conspiracy theorists irrational? Episteme 4(2), 193–204.
Coady, D. (2007b). Introduction: Conspiracy theories. Episteme 4(2), 131–134.
Coady, D. (2019). Conspiracy Theories: The Philosophical Debate. London: Routledge.
Dentith, M. R. (2014). The Philosophy of Conspiracy Theories. Basingstoke: Palgrave
Macmillan.
Driver, J. (2003). The conflation of moral and epistemic virtue. Metaphilosophy 34(3),
367–383.
Fallis, D. (2004). Epistemic value theory and information ethics. Minds and Machines
14(1), 101–117.
Fallis, D., and D. Whitcomb. (2009). Epistemic values and information management. The
Information Society 25(3), 175–189.
Greco, J. (2000). Two kinds of intellectual virtue. Philosophy and Phenomenological
Research 60(1), 179–184.
Greco, J. (2011). Virtues in epistemology. In P. K. Moser (ed.), The Oxford Handbook of
Epistemology (pp. 287–315). Oxford: Oxford University Press.
Fleisher, W. (2017). Virtuous distinctions. Synthese 194(8), 2973–3003.
Kawall, J. (2002). Other–regarding epistemic virtues. Ratio 15(3), 257–275.
Keeley, B. L. (1999). Of conspiracy theories. The Journal of Philosophy 96(3), 109–126.
Kessler, G. (2015). Huckabee’s claim that ‘global freezing’ theories from the 1970s shows
the science is “not as settled” on climate change. Washington Post. https://www.was
hingtonpost.com/news/fact-checker/wp/2015/06/25/huckabees-claim-that-global-free
zing-theories-from-the-1970s-shows-the-science-is-not-as-settled-on-climate-change/
?utm_term=.be8d8b1d5685 (accessed March 18, 2019).
Kpolovie, P. J., A. I. Joe, and T. Okoto. (2014). Academic achievement prediction: Role
of interest in learning and attitude towards school. International Journal of Humanities
Social Sciences and Education 1(11), 73–100.
Lackey, J. (2018, online first). The duty to object. Philosophy and Phenomenological
Research. doi:10.1111/phpr.12563
Levy, N. (2007). Radically socialized knowledge and conspiracy theories. Episteme 4(2),
181–192.
Moscovici, S., and C. F. Graumann. (1987). Changing Conceptions of Conspiracy. New
York: Springer-Verlag.
Paternotte, C., and M. Ivanova. (2017). Virtues and vices in scientific practice. Synthese
194(5), 1787–1807.
Pigden, C. (2007). Conspiracy theories and the conventional wisdom. Episteme 4(2),
219–232.
Räikkä, J. (2009). On political conspiracy theories. Journal of Political Philosophy 17(2),
185–201.
Renninger, K. A., S. Hidi, and A. Krapp (eds.). (2015). The Role of Interest in Learning
and Development (Vol. 1). New York: Psychology Press.
Schiefele, U. (1999). Interest and learning from text. Scientific Studies of Reading 3(3),
257–279.
Silvia, P. J. (2008). Interest—The curious emotion. Current Directions in Psychological
Science 17(1), 57–60.
Thaler, R. H., and C. R. Sunstein (2009). Nudge: Improving Decisions about Health,
Wealth and Happiness. London: Penguin Books.
Tobias, S. (1994). Interest, prior knowledge, and learning. Review of Educational Research
64(1), 37–54.
Woomer, L. (2017). Agential insensitivity and socially supported ignorance. Episteme
16(1), 73–91.
11 Intellectual snobs1
Charlie Crerar
11.1 Introduction
I’ll start by introducing the two antiheroes of this chapter. Both are drawn from
Muriel Spark’s 1988 A Far Cry from Kensington, a novel whose setting – the
publishing world in 1950s London – provides rich pickings for those interested
in intellectual snobbery. Its antagonist is a thin-skinned writer named Hector
Bartlett, who attempts a campaign of revenge against the novel’s narrator, Mrs
Hawkins, after she puts voice to the popular sentiment that he is a pretentious
hack. Mrs Hawkins first introduces Hector Bartlett to us as follows:
He used to waylay me in Green Park on my way to work or on my way home.

Occasionally this amused me, for I might egg him on to show off his social
superiority, and, not less, the superior learning that he claimed. For he knew
the titles of all the right books, and the names of the authors, but it amounted
to nothing; he had read very little … His writings writhed and ached with
twists and turns and tergiversations, inept words, fanciful repetitions, far-
fetched verbosity, and long, Latin-based words.
(Spark 1988: 39–40)
Hector Bartlett’s intellectual shortcomings, both characterological and authorial,

are detailed at some length in the novel. Less prominent but equally interesting
for present purposes are those of our second antihero: Martin York. Martin York
is Mrs Hawkins’ employer at Ullswater Press, a publishing firm that he eventu-
ally runs into the ground through both financial and creative mismanagement. A
significant contributor to this, we discover, was his tendency to offer commissions
to authors of questionable talent:
Sometimes, I think, his desire to sign up these [frightful] books for his pub-
lishing house was not due to a lack of discrimination so much as to the com-
mon fallacy which assumes that if a person is a good, vivacious talker he is
bound to be a good writer. This is by no means the case. But Martin York
had another, special illusion: he felt that men or women of upper-class back-
ground and education were bound to have advantages of talent over writers
Intellectual snobs 209
of more modest origins … Publishers, for obvious reasons, attempt to make
friends with their authors; Martin York tried to make authors of his friends.
(ibid.: 38)
I suspect many philosophers will have come across a few Hector Bartletts or
Martin Yorks in their time: people who know (and who want you to know that
they know) all the right books and authors and who load their writing with con-
voluted jargon, or who judge intellectual contributions on the prestige and back-
ground of those who present them, rather than the merit of the contribution itself.
Of course, such characters are not confined to academia and publishing. They are
recognisable in the sneering online commenter who would sooner correct some-
one’s grammar than engage with the substance of their arguments, in the friend
who insists on mocking Harry Potter as ‘a children’s book’, and in the man who
treats his long-distant Oxford education as a license to assume intellectual superi-
ority over any interlocutor.
Each of these examples are manifesting a shared epistemic vice: they are all
intellectual snobs.2 My aim in this chapter is to develop an account of intellectual
snobbery, and to draw some lessons from this specific vice for vice epistemolo-
gists more generally. In Section 11.2 I briefly motivate the idea that there is a
form of snobbery that is a specifically intellectual vice. In Section 11.3 I draw a
distinction between intellectual status and intellectual merit, and highlight the role
it plays in snobbish judgements. Sections 11.4 and 11.5 are then devoted to the
discussion of two distinct forms of intellectual snobbery: snobbery of motives and
snobbery of sensibilities. Section 11.6 concludes.
11.2 Snobbery as an intellectual vice

The first puzzle when thinking about intellectual snobbery concerns the kind of
vice we’re dealing with. Is it actually an intellectual vice, something that picks
out a specifically intellectual failing? Or is it really just snobbery about intellect,
that is, a moral vice as it operates in the intellectual domain? Compare, for exam-
ple, another trait that appears to straddle the moral–epistemic divide in this way:
intellectual courage. Rather than trying to theorise a form of courage that is, in
some sense, distinctively intellectual, virtue epistemologists have tended to think
of intellectual courage as just a more general trait – courage – as manifested in the
intellectual domain. Jason Baehr, for example, builds his account of intellectual
courage on an account of courage simpliciter, which involves ‘responding in a
certain way to a conflict between the achievement of a particular good and one’s
own safety or well-being’. What makes intellectual courage specifically ‘intel-
lectual’, he argues, is just that ‘the good in question is necessarily an intellectual
one’ (Baehr 2011: 164).
Perhaps an account of intellectual snobbery could follow the same strategy:
start with an account of snobbery as a moral vice, before extrapolating to the
intellectual domain. Although he does not talk about intellectual snobs specifi-
cally, Emrys Westacott (2012) can be read as taking this approach. Snobbery as a
210 Charlie Crerar
moral vice, for Westacott, involves ‘believing without sufficient justification that
you are superior to another person in certain respects because you belong to or are
associated with some group’ (2012: 115). The ‘intellectual’ element then comes
in through some of the groups Westacott identifies as potentially providing the
basis for such a belief: groups based on education, erudition, occupation, reading
habits, and so on. The intellectual snob, on this account, would just be someone
who thinks him- or herself a better person, and does so because of one of these
intellectual facts about themselves.
Note, though, just how minimal the sense is in which this counts as an ‘intel-
lectual’ vice. Intellectual courage, in Baehr’s sense, may be derivative of a more
general trait, but it nonetheless picks out a quality that is an important part of
good epistemic agency. Snobbery about intellect in Westcott’s sense, however,
will have at most an indirect effect on one’s approach to epistemic practice. The
failing it picks out – the tendency to think of oneself as a better person than others
– is a moral one, and it is largely by-the-by that this judgement is made on an intel-
lectual basis. The intellectual snobs I am interested in are those whose character
corrupts their epistemic conduct in a much more direct way.
My suggestion is that intellectual snobbery proper affects how one conducts
one of the most fundamental epistemic practices: the practice of making intel-
lectual evaluations. Intellectual evaluations are assessments of some object
as good or bad according to some value that is distinctively intellectual. I use
this term broadly along a variety of axes. First, we make intellectual evalua-
tions about all manner of different objects: propositions, arguments, intellectual
agents (oneself included), intellectual objects (such as books or institutions),
and intellectual practices. Second, these evaluations invoke all sorts of different
values. Some of these are narrowly epistemic, as when we assess the truth of
a proposition, the credibility of an interlocutor, or the validity of an argument.
Many, however, are intellectual in a broader sense, in that they affect how one
conducts and directs oneself as an intellectual agent: the projects one engages
in, the ways one conducts one’s inquiries, the habits and sensibilities one looks
to cultivate, and so on. Thus, evaluations of a paper as interesting, a research
area as worthwhile, or a potential university as appealing are all, at least in part,
intellectual evaluations.3 Third, some intellectual evaluations are categorical (a
proposition is either true or not), but many are comparative (as when we judge
this argument to be more convincing than that one). And fourth, intellectual
evaluations can be either occurrent or dispositional. They might take the form
of either a specific act of appraisal, or an abiding and perhaps subconscious taste
or preference.
That there is some connection between intellectual snobbery and the practice
of making intellectual evaluations is fairly clear. In fact, sometimes we describe
peoples as snobs simply because of their tendency make very robust evaluations of
superiority and inferiority: the friend who reflexively tuts when they see someone
reading a tabloid newspaper, for example. This colloquial use of snob, though, is
clearly too broad to capture the idea of snobbery as a vice. Intellectual evaluations
are a central part of our epistemic lives, and virtue cannot be a matter of avoiding
or shying away from them. Rather, it is about making them in a responsible fash-
ion. It is here where we should look for our account of intellectual snobbery.
11.3 Intellectual status and intellectual merit

The question at hand, then, is how to differentiate the properly discerning intel-
lectual agent – one who makes intellectual evaluations, and potentially even quite
robust ones, but who does so on an appropriate basis – from the intellectual snob.
At this point, it will be helpful to return to Hector Bartlett, the first of our para-
digm snobs. Recall that the reason why Mrs Hawkins initially found his knowl-
edge of ‘all the right books, and the names of all the authors’ so amusing was
the fact that, in spite of all his name-dropping, he had clearly ‘read very little’
(Spark 1988: 39). A reviewer reaches a similar conclusion when attempting to
make headway with the manuscript for Bartlett’s intended masterpiece, grandly
titled The Eternal Quest: A Study of the Romantic-Humanist Position. Bartlett, the
reviewer quickly concludes, is ‘completely phoney’: ‘On every page Nietzsche,
Aristotle, Goethe, Ibsen, Freud, Jung, Huxley, Kierkegaard, and no grasp whatso-
ever of any of them’ (ibid.: 94–95).
The problem with Bartlett is not simply that he is interested in Nietzsche,
Aristotle, and the rest;4 these are, after all, major intellectual figures, and engag-
ing with their work can be a fruitful and fulfilling exercise.5 The problem is not
even that he does not fully understand them; this is not ideal for an author work-
ing on these figures, but nor is it necessarily indicative of intellectual vice. The
problem, instead, is that he affects understanding that he evidently lacks, with
the clear insinuation that he does so precisely because of the intellectual heft that
these authors carry. In short, what makes him a snob is not that he is interested
in weighty books and fashionable authors, but that he is only interested in them
because they are weighty and fashionable.
My initial claim, then, is that what makes someone an intellectual snob is not
the content or even the strength of their intellectual evaluations, but rather the
considerations upon which these evaluations are based.6 I have already gestured
at the relevant distinction, but we can clarify it further by introducing some rough
terminology. First, there are what I refer to as an object’s intellectual merits.7
In essence, intellectual merits are an object’s positive intellectual qualities: its
strengths, proficiencies, benefits, and so on. Clearly this is a loose and heteroge-
neous category, and there is no one quality that is the quality of being intellectu-
ally meritorious. The intellectual merits of a person might be their intellectual
virtues, skills, or intelligence, the merits of an argument might include its validity
or explanatory power, the merits of a university might be the quality of education
it provides or its research output, and so on. What constitutes an intellectual merit
in the case of a specific evaluation will be determined both by the nature of the
object in question and the purposes and context of the evaluation.
Second, there is an object’s intellectual status. This is the way it is perceived
and the reputation it enjoys, qua intellectual object, within a society or social
group. Sometimes intellectual status is awarded as a fairly direct response to
212 Charlie Crerar
perceived intellectual merit, with the status enjoyed by revered public intellectu-
als and challenging best-sellers good examples of this. Typically, though, the rela-
tion between status and merit is more complicated. For a start, certain properties
can come to enjoy intellectual status not because they themselves constitute a way
of being intellectually meritorious, but because they are associated with proper-
ties that do. An affiliation with a prestigious university is not itself an intellectual
merit, for example, but it confers intellectual status on account of the connection
believed to exist between that affiliation and intellectually meritorious qualities.
Of course, a range of more problematic attributes can come to be associated with
intellectual merit in this way, as when accents come to be associated with levels of
education or age with a refinement of interest. The connection between status and
merit is further complicated when broader societal values and prejudices come
to affect the distribution of intellectual status. In the UK, for example, qualities
like sobriety, tradition, and eloquence all enjoy intellectual status to an extent that
cannot be explained by appeal to even a (supposed) indirect connection to intel-
lectual merit.
Intellectual evaluations, we have seen, are evaluations of an object’s intellec-
tual qualities. They therefore just are, at least in part, evaluations of intellectual
merit. This does not mean, however, that they will always actually be based upon
or responsive to intellectual merits. Other factors might serve to mediate – or cor-
rupt – these evaluations. This is what happened with Hector Bartlett, and what I
would suggest happens with intellectual snobs generally. They make their intel-
lectual evaluations not (or not primarily) on the basis of a direct assessment of an
object’s intellectual merits, but on the basis of its status or reputation within their
social group.
This insight – that the intellectual snob appeals to considerations of status
when making intellectual evaluations – only brings us part of the way to develop-
ing an account of intellectual snobbery. This is because considerations of intel-
lectual status actually play a role in many of our judgements, often in ways that
seem entirely unobjectionable. This point has recently been noted by Alessandra
Tanesini, in a discussion of the role played by esteem – which is relevantly similar
to intellectual status8 – in facilitating judgements about the testimony of experts:
Individuals are often faced with the task of adjudicating between contradic-
tory testimonies, or of deciding whether to change their pre-existing opinions
in the light of the views expressed by their critics. It is not always feasible or
possible to proceed by assessing independently the likely truth of the views
themselves. One may lack either the resources or the knowledge required
rationally to evaluate the positions at hand. Further, one may also be unable
to evaluate the competence of the disagreeing would-be experts. In some of
these cases esteem supplies evidence that assists one’s evaluation.
(Tanesini 2018a: 53)
As for expert testimony specifically, so too for intellectual evaluations generally.

We often find ourselves unable to assess the merits of intellectual objects directly.
Sometimes, as Tanesini notes, this is because we lack the time, skills, or resources
to do so. Alternatively, it may be because doing so would defeat the point of
making that evaluation in the first place. When moving into a new research area,
for example, you might want to know the most important papers published on
that topic so you can work out where to start your reading. The best place to start
reading, however, is not something you can determine by directly assessing the
contributions to that field, since doing so would require doing just the thing you’re
trying to strategise: namely, starting reading.
As Tanesini points out, intellectual status can provide a valuable workaround
in situations like these. The correlation between status and merit is by no means
perfect, as we have just seen, but sometimes it is all we have to go on. As such,
it cannot be the case that making evaluations on this basis is problematic across
the board. What, then, is vicious about how the intellectual snob does this? This
question will be my focus for the remainder of this chapter.
11.4 Snobbish motives

A helpful starting point will be to think in more general terms about what it is that
makes traits intellectually vicious. One common approach here is to offer a moti-
vational analysis of intellectual vice. According to this motivational approach,
intellectual vices are traits that involve bad epistemic motivations (Baehr 2010;
Battaly 2016; Tanesini 2018b).
These motivations come in two parts. First, there are the proximate motivations
of a given vice: a set of dispositions to act, think, and feel in certain ways. Thus,
the proximate motivations of closed-mindedness centrally involve something
like the motivation to ignore relevant epistemic options, stubbornness involves
the motivation to persist in some approach in defiance of feedback from others,
and so on. Sets of proximate motivations are unique to specific vices, even if, as
these examples just offered illustrate, there will often be some overlap between
vices. Theories of the nature of specific vices are therefore, to a significant extent,
accounts of their distinctive proximate motivations. What makes any of these
sets of proximate motivations intellectual vices in the first place, however, is
the fact that they are underpinned by a shared set of ultimate motivations. Thus,
according to the motivational approach, all intellectually vicious character traits
are ultimately motivated by some desire for epistemic bads: they are patterns of
thought, feeling, and action that one exercises in an attempt to remain ignorant,
or to acquire false or comforting beliefs, or beliefs that fit with your pre-existing
worldview.9 So, the closed-minded will ignore relevant epistemic options, and
will do so because they wish to remain ignorant about some uncomfortable truth.
We can identify a form of intellectual snobbery that fits the framework of the
motivational approach. The proximate motivations are clear enough: intellectual
snobbery involves a motivation to evaluate things, at least in part, on the basis of
their intellectual status, rather than according to their actual intellectual merits.
It is the ultimate motivations that are more interesting. This is because not just
any set of ultimate ends, even when conjoined with these proximate motivations,
214 Charlie Crerar
count as snobbish. Imagine someone who, like Hector Bartlett, bases all his
judgements about which authors to engage with purely on their ranking within
the Western literary canon. The reason he does so is simply because he is scared
of going out on a limb. He works in a deeply traditional and hierarchical field, and
thinks that if he goes beyond the canon he will open himself up to ridicule and,
potentially, a loss of livelihood. He knows he shouldn’t be relying upon status,
but does so anyway because his main interest is preserving his place in the pro-
fession. Certainly his profession is a snobbish one, and these snobbish norms are
something he is helping to perpetuate. However, snobbery doesn’t quite pick out
his individual character failing. Rather, he seems more like an intellectual coward,
or a conformist.
Compare this with the actual Hector Bartlett. Far from just trying to fit in and
go unnoticed, his main concern is with setting himself apart: specifically, with
establishing himself as intellectually superior to other people, as smarter, more
cultured, and as moving in better circles. The only reason he wants to be a writer,
we learn, is because he fancies himself as ‘a great critic, a sort of thinker’, and
he ‘wants to see his name in print’ (Spark 1988: 142). Similarly, it is his desire
to ‘show off his social superiority, and, not less, the superior learning that he
claimed’ that led him to name-drop incessantly in front of Mrs Hawkins, someone
whom he assumes to be his inferior and who will thus, he thinks, be impressed
by his behaviour (ibid.: 39). It is this underlying desire, to feel or appear superior
to other people, that explains how sensitive he is to intellectual status. He hopes
that by associating himself with markers of intellectual refinement – by knowing
all the right books, fraternising with the right people, and peppering his writing
with the right jargon – people will come to think of him as some superior intellect.
The temptation at this point might be to dismiss Hector Bartlett as a slightly
tragic figure, bumbling and pretentious perhaps, but nothing worse. This, though,
is only part of the story. As Judith Shklar reminds us, when thinking about snob-
bery it is important that we don’t see ‘only the upward striving’ and ‘[forget] the
kick aimed downward’ (Shklar 1984: 89). In other words, if we are too busy roll-
ing our eyes at Bartlett’s tendency to shamelessly court the great and the good at
parties, then we might miss his more obviously pernicious impact upon the less
status-worthy, the ‘literary agents and authors of little fame’ who are left hovering
‘warily round the fringe’ (Spark 1988: 92). Superiority is an inherently relative
notion: it requires not just being better than some other person or group, but that
person or group being worse than you. Accordingly, the snob who is attempting
to establish their superiority will be equal parts ingratiating and sneery, distanc-
ing themselves from the low-brow or marginalised just as avidly as they pursue a
connection with the high-brow and established.
In its manifestations as both an upward striving and a kick aimed downwards,
this desire to feel or appear superior to others is recognisable in a host of other
intellectual snobs, from the person who turns down participation in an interesting
research project only because it is associated with a comparatively minor insti-
tution, to the person whose go-to move in online debates is to correct the other
person’s grammar. This, then, is our first form of intellectual snobbery, which we
can think of as a ‘snobbery of motives’. It picks out the person who is disposed
to appraise things on the basis of their intellectual status, and who does out of a
desire to feel or appear superior to some other individual or group.
11.5 Snobbish sensibilities

My discussion of snobbery of motives was predicated upon a motivational anal-
ysis of intellectual vice. Although highly influential, this analysis has recently
been called into question by theorists who have queried whether the presence of
bad epistemic motivations is always necessary for intellectual vice (Crerar 2018,
Cassam 2019). It is in this connection that intellectual snobbery proves a particu-
larly interesting case for vice epistemologists. This is because, whilst there is a
form of intellectual snob who, like Hector Bartlett, is characterised by bad moti-
vations, this is not the only form of intellectual snob.
It is here that our second paradigm snob comes into play. This, recall, is
Martin York, a director who sunk the publishing firm Mrs Hawkins worked at
because he only published books by ‘his fellow-officers of war-time, or for-
mer school friends’ (Spark 1988: 17). As we have already seen, Mrs Hawkins
speculates that York’s bad judgement was not due to a ‘lack of discrimina-
tion’, by which she presumably means he did have some grounds for making
these decisions. The problem, instead, was that the grounds were shoddy ones.
Specifically, he tended to think that ‘if a person is a good, vivacious talker he
is bound to be a good writer’, and that those ‘of upper-class background and
education were bound to have advantages of talent over writers of more modest
origins’ (ibid.: 38).
Martin York’s problem, we can now recognise, was that he would base his
evaluation of manuscripts on facts about the intellectual status of its author – facts
about their background, education, and mannerisms, and how these resonate with
society’s image of a good writer – and not on an actual appreciation of its intel-
lectual merits. In this sensitivity to status, he is therefore very like Hector Bartlett.
However, the two are also crucially different. Bartlett’s disposition is problematic
because it is rooted in a desire to feel or appear superior to others. At no point,
however, is there any indication that York’s judgements are similarly motivated.
We have just seen, for a start, that Mrs Hawkins does not attribute his reliance on
status to the self-aggrandising intent that she recognises in Bartlett; indeed, she
relates that York is keenly aware of the need to sign good books (rather than just
books that make him seem good), if only for the good of the firm (Spark 1988: 17).
Instead, she attributes his failings to the sincere, albeit woefully mistaken, view
that aspects of an author’s background really are predictive of intellectual merit.
Furthermore, unlike the deeply insecure Bartlett, York’s intellectual superiority
relative to the general population is something of which he seems totally assured.
Even at his nadir, imprisoned for fraud and castigated by the press, he remains
unfailing in his conviction that he deserves a place amongst the intelligentsia of
high society. As he earnestly proclaims to Mrs Hawkins from his cell, ‘I have a
first-rate brain, some say brilliant’ (ibid.: 47).
216 Charlie Crerar
The puzzle the motivational approach was invoked to solve, recall, was to
explain what is vicious about the snob’s appeals to intellectual status, given that
many such appeals are entirely unobjectionable. The motivational analysis can
explain this in cases like Bartlett’s precisely because he is using status quite dif-
ferently from the ways suggested by Tanesini. Specifically, he is not treating it
as a work-around in judgements of intellectual merit, but as a shortcut to help
cultivate a certain image of himself. The same, however, cannot be said of Martin
York. The reason he appeals to intellectual status is because he thinks it provides
a reliable shortcut when making evaluations of intellectual merit.10 The problem,
of course, is that he is wrong about this; or, to put the point more perspicuously,
he is too sensitive to considerations of intellectual status, such that he makes use
of them even in cases where to do so is clearly irresponsible. As a consequence, he
habitually dismisses authors and their work when he shouldn’t, grants dispropor-
tionate credence to those who clearly don’t warrant it, and maintains an absurdly
inflated view of his own intellectual prowess despite an abundance of evidence
to the contrary.
The kind of snobbery on display here is thus crucially different from the snob-
bery of motives discussed previously. My suggestion is that what makes Martin
York an intellectual snob is not anything to do with the quality of his epistemic
ends, but rather is to do with the calibration of his epistemic sensibilities. Our sen-
sibilities are particular patterns of attention that jointly comprise our characteristic
way of looking at the world. In the epistemic case, they determine the considera-
tions to which we are particularly sensitive in the conduct of our epistemic activi-
ties, and those that we tend to ignore. To borrow a phrase from Miranda Fricker,
these sensibilities enable us to see the world in ‘epistemic colour’ (2007: 71),
such that certain aspects of a situation strike one as more relevant to epistemic
judgements than others. The problem for Martin York is that considerations of
intellectual status strike him as more vivid than they should. Thanks, presumably,
to his upbringing, education, and social circles, he has developed a view of the
world according to which intellectual status is seen as a particularly powerful pre-
dictor, if not determinant, of intellectual merit. So close is this connection in his
mind that even when his epistemic ends are entirely appropriate – he genuinely
wants to appraise the true quality of a manuscript, for example – considerations of
intellectual status will nonetheless strike him as amongst the most salient factors
to consider.
Although I suspect that vice epistemologists would benefit from devoting more
attention to the role of epistemic sensibilities within one’s intellectual character,
fleshing out the details of this account goes far beyond the scope of this chapter.
My aim here, instead, is more focused: to establish that, in addition to the snob-
bery of motivations already discussed, there is also a snobbery of sensibilities.
The former is the snobbery of the epistemologist who dismisses feminist philoso-
phy as ‘not real philosophy’ in order to shore up their image of themselves as a
more rarefied intellect; the latter is the snobbery of the epistemologist who just
accepts that they are a superior intellect, because of the status enjoyed by their
research area. The former is the snobbery of someone who only reads broadsheet
newspapers in an attempt to cultivate an air of intellectual refinement; the latter
is the snobbery of someone who genuinely believes they are more intellectually
refined than others on the basis of the newspaper they read.
This, then, is the general idea behind the notion of snobbish sensibilities: that it
is an excessive sensitivity to intellectual status that leads to unwarranted conclu-
sions of intellectual superiority and inferiority. Much remains to be said, however,
about where exactly this snob goes wrong in their intellectual evaluations, and
what it is that makes their doing so snobbish. I shall discuss each of these issues
in turn.
11.5.1 Status as proxy

One helpful way to think about the role that intellectual status plays in responsible
intellectual evaluations is as a proxy for intellectual merit. In general, a proxy vari-
able is a variable that correlates more or less reliably with some further variable, the
‘target’ variable. We employ a proxy variable when we have some interest in meas-
uring the target variable, but where doing so directly is either impossible or else
excessively time- or resource-consuming. This difference in ease of measurement
can motivate the use of a proxy even when its correlation with the target variable is
only quite loose. So, for example, social scientists use average income as a proxy
for well-being not because it provides a perfect indication of the latter, but because
it gives a rough impression at only a fraction of the difficulty of measurement.
As I noted in Section 11.3, there is often at least an indirect connection between
intellectual status (the intellectual reputation some object enjoys) and intellectual
merit (its actual intellectual qualities). This is why employing the former as a
proxy for the latter is not necessarily an epistemically irresponsible thing to do.
The motivational snob, of course, doesn’t use status as a proxy in this way; all
they care about is how status might reflect back upon themselves. This explains
where they go wrong in their evaluations. Things are more complicated for the
person with snobbish sensibilities, since they do use status as a proxy. Martin
York, we have seen, genuinely did think that class and verbosity were correlated
with writing ability. The question we need to ask, therefore, is when it is respon-
sible to use status as a proxy, and when it is not.
The basic point to bear in mind here is that proxies rely for their explanatory
power on the presence of a correlation between the proxy and target variables.
Generally speaking, the stronger the correlation, the better the proxy. Perhaps the
most straightforward manifestation of a snobbish sensibility, therefore, will sim-
ply be someone’s taking intellectual status to be relevant when it is in fact not, or
where the correlation between status and merit is only very weak. Alternatively,
they might use a marker of intellectual status that is in fact correlated with some
form of intellectual merit in an evaluation of a quite different form of merit. Bear
in mind that I am using ‘intellectual merit’ not to pick out a specific property, but
as a catch-all to accommodate the full diversity of positive intellectual attributes.
In most cases, even justifiably earned intellectual status will correlate with only
a few dimensions of intellectual merit. Thus, even if Martin York was right that
218 Charlie Crerar
there is a connection between facts about someone’s educational history and, say,
their knowledge-base – someone with a literature degree, for example, is likely to
know more about aspects of literature than the general population – he would still
not be warranted in making the sweeping judgements that he does.
Intellectual status’ role as a proxy for merit can also explain the snob’s attitudes
towards other forms of evidence. We have seen that proxies are typically invoked
when there is a paucity of easily accessible direct evidence about the target vari-
able. The existence of a very strong proxy correlation, however, might justify
deferring to that proxy even in the presence of direct evidence. In other words, it
is sometimes reasonable to trust a reliable and easily accessible proxy over one’s
own capacity to interpret evidence directly. Take, for example, someone who has
never read Wittgenstein, but who ranks him as a profound and important philoso-
pher on the basis of his reputation. There is nothing obviously snobbish about this
judgement, since within the community of academic philosophers there probably
is quite a strong correlation between someone’s enjoying a positive intellectual
status qua philosopher and their being intellectually meritorious in some of the
ways that are constitutive of excellence as a philosopher.11 In fact, the strength of
this correlation is such that it is not obviously snobbish for this person to continue
to rank Wittgenstein as a profound and important philosopher even after engaging
with his work directly, and finding it somewhat underwhelming.
The problem with the snob, of course, is that they take status to be far more
reliable as a proxy then it in fact is. As a consequence, they will take this sceptical
attitude towards direct evidence when there is not the strength of proxy relation
to justify it. Martin York, for example, did have direct evidence of the things he
was trying to measure indirectly – he was reading the manuscripts, after all – but
he continually discounted its significance relative to the evidence provided by the
proxy.
11.5.2 Snobbish judgements

Thus far, I have identified snobbish sensibilities simply with an undue sensitivity
to considerations of intellectual status. This, though, is in need of further refine-
ment, since it is clear that not just any such sensitivity counts as snobbish. To see
why, consider the informal role that markers of intellectual status play in exclud-
ing students from less privileged backgrounds from elite universities. It is often
suggested that one mechanism of exclusion at institutions like Oxbridge is that
students from lower-income backgrounds or without a family history of higher
education feel out of place when confronted by these universities’ international
reputation, illustrious history, and imposing architecture. You might have all the
academic attributes required to flourish there, and yet still be thrown – as one
student put it – by the ‘simple intimidation of “oh God this room is in a literal
castle”’ (Curtis 2017).
At least sometimes, what is going wrong here is that these prospective stu-
dents are being too sensitive to considerations of intellectual status: they are judg-
ing their suitability for a particular education on the status their own background
conveys vis-à-vis the status of that institution. However, although there is clearly
something amiss with these evaluations, it seems wrong to describe such students
as snobs. This, surely, is an epithet that we reserve for the privileged student, the
one who concludes on the basis of their family history and prior educational back-
ground that they ‘belong’ in such environs.
This contrast serves to illustrate an analogous point to the example of the
conformist version of Hector Bartlett, introduced in our discussion of snobbish
motives. In both cases, what is highlighted is the connection between intellectual
snobbery and judgements of one’s own superiority relative to some other indi-
vidual or group. Being a snob is not just to have an unwarranted preoccupation
with status, but to have a preoccupation with status that, amongst other things,
leads one to look down on others whom one views as inferior. The refinement
needed for snobbery of sensibilities is therefore as follows: it is the vice of being
excessively attuned to considerations of intellectual status, in a way that leads one
to conclude that one is superior to some other individual or group.
Insofar as this refinement suggests that a snobbish sensibility only manifests
itself in relative evaluations of the intellectual merits of different people, this
might seem an unwelcome restriction on the scope of intellectual snobbery. After
all, many of the examples of paradigm snobbish judgements that I have given
throughout this chapter are judgements about the intellectual merits of ‘things’:
books, institutions, methodologies, and so on.12 Of course, sometimes evaluations
of things are premised quite directly on evaluations of people. Martin York’s
snobbery, for example, might manifest itself in a dismissive attitude towards a
manuscript written by a working-class person, but what he is really dismissive of
here is the aptitudes of the working class. Other examples, though, are trickier to
accommodate. Consider another kind of literature snob: not someone (like Hector
Bartlett) who is determined to cultivate an air of refinement and intellect, or some-
one (like Martin York) who bases their evaluations on the social background of
authors, but someone who simply thinks that reading anything other than ‘the
classics’ is a waste of time. Can my account accommodate the intuition that there
is something snobbish going on here?
I think it can, but only because I suspect that the whiff of snobbery that per-
vades this example depends on the assumption that his wrinkling his nose at
certain genres of literature is, in fact, accompanied by a wrinkling of the nose
at certain kinds of people. One possibility is that his taste in literature is itself
guided, in part, by judgements about which genres are appropriate for different
kinds of people. He might condescendingly dismiss certain types of writing as
‘chick lit’, say, or as suitable only for people without the education to appreciate
the classics. These dismissals thus invoke judgements about the merits of differ-
ent groups of people, judgements that are based upon facts about their intellectual
status and that, at least implicitly, class him as one of the intellectually superior.
Such judgements are, of course, paradigmatically snobbish.
Let’s try to close off this reading. Suppose that our literature snob simply
thinks of light fiction as a totally vacuous form of literature, and does so without
presupposing a condescending view of any particular type of person. As he might
220 Charlie Crerar
put it, he simply thinks that we are all ‘better than that’. In this case, I would have
to concede that there is nothing snobbish about his tastes in literature, since in no
way do they involve a judgement of his own superiority. Nonetheless, perhaps
there is a different snobbery at play here that can explain our intuitions: namely,
the snobbery of evaluating yourself as intellectually superior on the basis of your
intellectual evaluations. Tastes, preferences, and judgement are all viable candi-
dates for designations of intellectual status: people can be and are intellectually
revered for their taste in literature, film, radio stations, and so on. This creates
space for a form of higher-order snobbery, according to which the light fiction
cynic thinks of himself as superior because he, unlike everyone else, has seen the
light when it comes to literature.
There are thus two ways in which snobbery of sensibilities – a sensitivity to
intellectual status that leads to unwarranted conclusions of one’s own superiority
– can affect evaluations of the intellectual merits of things: it might be a sense of
one’s own superiority that underlies these evaluations, and one can judge oneself
to be superior on the basis of these evaluations. If we strip this example of any hint
of either of these attitudes, then we would be left with someone who concludes,
on the basis of intellectual status, that light fiction is not something of any intel-
lectual merit, but who also maintains that there is nothing intellectually inferior
about those who do read it. This person would not count as an intellectual snob
in either of the senses I have identified. This, I think, is exactly the right result.
Harbouring a blanket and stubborn aversion to a whole family of literature might
still be intellectually vicious: it might be closed-minded, or conformist, or a form
of epistemic insensibility (Battaly 2013). However, absent some sneering judge-
ment about the intellectual merits of different types of people, snobbery does not
seem quite right.
11.6 Conclusion
In this chapter, I have identified two distinct kinds of intellectual snob. One is the
person who guides their intellectual evaluations by way of an appeal to intellec-
tual status, and who does so in an attempt to make themselves feel or appear supe-
rior to others. The other is the person who is simply too sensitive to intellectual
status, and who thus habitually draws conclusions of their own superiority relative
to others. These traits are clearly related: both are dispositions to make use of
intellectual status, and both have a close connection to a view (desired or actual)
of one’s intellectual superiority. As important, though, are their differences, with
one a vice of motivation, the other a vice of sensibility. It is the latter that will
prove of most interest to epistemologists looking to move beyond a motivational
account of intellectual vice.
Notes
1 I am grateful to Quassim Cassam, Miranda Fricker, Jules Holroyd, and Cody Turner
for very helpful comments on drafts of this chapter. Thanks also to audiences in
Manchester, Vienna, Sheffield, and Connecticut, and to Pauline and Graham Crerar,
whose book recommendations are always appropriately sensitive to intellectual merits.
2 As I have implied, there are really two forms of snobbery here: snobbery as akin to a
sneery pretentiousness, and snobbery as akin to elitism. I will argue that, when fully
spelt out, both traits constitute a form of intellectual snobbery, and that whilst there
are interesting differences between the two there are also important commonalities. I
have noticed, though, that people sometimes have different intuitive reactions to these
vignettes; specifically, that people with British or other European backgrounds some-
times claim that the real snobbery is Martin York’s comfortable elitism, whilst North
Americans are more prone to identify it with the affected disdain of Hector Bartlett.
I suspect that this discrepancy has something to do with the functioning of the class
systems in these different societies, though I cannot pursue this point here. In any case,
my aim in this chapter is to vindicate the viciousness of both forms of snobbery.
3 Some objects are apt for more than one form of evaluation: books and films, for exam-
ple, might be assessed both intellectually and aesthetically. The boundaries between
these domains will often not be sharp.
4 This is not what Spark is trying to convey, in any case, though we might think that there
is something amiss with someone whose intellectual horizons consist exclusively of
white European men.
5 Or so I presume, based on their reputations. I don’t think this makes me a snob, for
reasons I discuss Section 11.5.1.
6 My discussion here is heavily influenced by Kieran’s (2010) account of aesthetic snob-
bery.
7 Henceforth, I use ‘object’ as a generic term for anything that can serve as the object of
an intellectual evaluation.
8 Esteem is a positive (or negative) evaluative attitude directed at a person or group for
their good or bad qualities (Tanesini 2018a: 49). Tanesini is mainly interested in esteem
as conferred by individuals, whilst I am thinking of intellectual status as something
social. This is closer to what Tanesini calls ‘reputation’.
9 Alternatively, we might think of vicious motivations as involving the absence of a
concern for epistemic goods.
10 Of course, bad faith or self-deception about his ultimate ends is one explanation here. I
see no reason why this should be the only plausible explanation.
11 That the presence of a good reputation is correlated with intellectual merit does not
entail that the absence of a good reputation is strongly correlated with a lack of intel-
lectual merit. Proxies can be asymmetrically reliable.
12 This distinction between snobbery about people and snobbery about things is also an
issue for Westacott (2012).
References
Baehr, J. (2010) ‘Epistemic Malevolence’, in Battaly, H. (ed.) Virtue and Vice, Moral and
Epistemic (Chichester: Wiley-Blackwell), 189–213.
Baehr, J. (2011) The Inquiring Mind: On Intellectual Virtue and Virtue Epistemology
Battaly, H. (2013) ‘Detecting Epistemic Vice in Higher Education Policy: Epistemic
Insensibility in the Seven Solutions and the REF’, Journal of Philosophy of Education
47 (2), 263–280.
Battaly, H. (2016) ‘Epistemic Virtue and Vice: Responsibilism, Reliabilism, and
Personalism’, in Mi, C., M. Slote, and E. Sosa. (eds.) Moral and Intellectual Virtues in
Chinese and Western Philosophy: The Turn Towards Virtue (New York: Routledge),
99–120.
222 Charlie Crerar
Cassam, Q. (2019) Vices of the Mind (New York: Oxford University Press).
Crerar, C. (2018) ‘Motivational Approaches to Intellectual Vice’, Australasian Journal of
Philosophy 96 (4), 753–766.
Curtis, D. (2017) ‘Oxford is Accepting More State School Pupils – In the South East at
Least’, New Statesman. Available at: https://www.newstatesman.com/politics/educati
on/2017/10/oxford-accepting-more-state-school-pupils-south-east-least (accessed on
March 20th 2019).
Fricker, M. (2007) Epistemic Injustice: Power and the Ethics of Knowing (Oxford: Oxford
University Press).
Kieran, M. (2010) ‘The Vice of Snobbery: Aesthetic Knowledge, Justification, and Virtue
in Art Appreciation’, The Philosophical Quarterly 60 (239), 243–263.
Shklar, J. (1984) Ordinary Vices (Cambridge, MA: Belknap Press).
Spark, M. (1988) [2017] A Far Cry from Kensington (Edinburgh: Polygon Books).
Tanesini, A. (2018a) ‘Caring for Esteem and Intellectual Reputation: Some Epistemic
Benefits and Harms’, in Barker, S., C. Crerar, and T. Goetze (eds.) Royal Institute
of Philosophy Supplement: Harms and Wrongs in Epistemic Practice (Cambridge:
Cambridge University Press), 47–67.
Tanesini, A. (2018b) ‘Epistemic Vice and Motivation’, Metaphilosophy 49 (3), 350–367.
Westacott, E. (2012) The Virtues of Our Vices (Princeton, NJ: Princeton University Press).
Part IV
Applied vice epistemology

12 Teaching to the test
How schools discourage phronesis
Casey Johnson
In this chapter I defend the claim that dominant approaches to contemporary

education discourage epistemic virtue not by encouraging the development of
vices directly, but by discouraging students from developing epistemic phrone-
sis. Students subjected to education by rote memorization or test-focused teach-
ing are discouraged from knowing when and how to use their epistemic virtues
in school. Nonetheless, they still display epistemic virtues in many parts of their
lives – they simply don’t seem to know that school is a place where these virtues
apply. Contemporary schooling, then, is epistemically corrupting in a way that
is related, but not identical, to Kidd’s sense of the word: it corrupts by prevent-
ing the development of phronesis (Kidd, 2015). And it ought not to be – edu-
cational contexts like formal schooling should not only help students develop
their epistemic virtues, but should also help them use those virtues in the right
way at the right time. In other words, schools should be one place where epis-
temic phronesis is encouraged and developed. This chapter will build on work
on educating for the virtues from Battaly (Battaly, 2006), Watson (Watson,
2018), Kidd (Kidd, 2017), and others, as well as from data on educational out-
comes from students in the U.S. after “No Child Left Behind” (Duffy, Giordano,
Farrell, Paneque, & Crump, 2009; Hursh, 2007; McCarthey, 2008; Rushton &
Juola-Rushton, 2008).
12.1 Sarah and Wallace

In the first season of the HBO show, The Wire, there’s a scene involving a math
problem. Wallace, a teenager who cares for younger children, is awoken by one
of his young charges, Sarah, who needs his help with her math homework. Sarah
attends a Baltimore, MD, public school. Being a Baltimore public school means,
as the show’s viewers know, that Sarah’s school is understaffed, underfunded,
and (almost certainly) failing to meet state educational standards. Sarah and
Wallace are also involved in illegal drug sales. Sarah and Wallace have the fol-
lowing exchange:
“Yo, Wallace.”
“What?”
226 Casey Johnson
“What’s this about here?”
“This one here? ‘A bus traveling on Central Avenue begins its route by picking
up eight passengers.
Next, it picks up four more, and then an additional two, while discharging one.
At the next-to-last stop, three passengers get off the bus and another two get on.
How many passengers are on the bus at the last stop?’ Just do it in your head.”
“Seven, right? … Eight?”
“Damn, Sarah, look. Close your eyes. You working a ground stash. 20 tall pinks.
Two fiends come up and ask for two each, another one cops three. Then
Bodie hands you 10 more, but a white guy rolls up in a car waves you down,
and pays for eight. How many vials you got left?”
“Fifteen.”
“How the fuck can you keep the count right, but not do the book problem?”
“Count be wrong, they’ll fuck you up.”
In this exchange, Sarah has the skills necessary to do both math problems. Indeed,
the problem that Wallace gives her is more complicated than the one she needs
to do for homework. However, she does not, or is not able to bring those skills to
bear on the book problem. What is going on here?
It may be that the explanation is simple: Sarah herself suggests that the degree
of motivation is different. She is highly motivated by the threat of violence to
keep accurate count of how many vials of drugs she has. There is no such threat in
the case of the book problem. So she is, perhaps, simply insufficiently motivated
to apply her math skills to the book problem. However, I don’t think this is a
convincing explanation for the whole of the difference between these cases. She
is motivated to solve the book problem – she asks her de facto guardian for help
with it before school. Something other than motivation must be playing a role in
this case.
It may be that Sarah’s case seems implausible. It is, after all, from a work
of fiction. We might think the show’s writers just invented a conversation that
might happen in some fictional contexts, but deny that this conversation is real-
istic or even provides much insight into real-world experiences. However, we do
see cases like Sarah’s happen in real-life educational contexts. Phillip Stevens, an
Apache sociologist, gives examples of this in his work with Vanessa Anthony-
Stevens and Sheila Nicholas. Stevens recounts the ways in which native students
can struggle in Western White educational systems when the values, norms, and
vocabularies differ from those in their homes and communities:
Educational scenarios where students are told new names to familiar actions
can be absurd. However, if students can come to understand that concepts
like tessellation have foundation in events such as stacking mesquite wood,
rather than existing in isolation upon clean, white papers covered in abstract
lined shapes, these same students can take this knowledge and find purpose
in it so that it makes sense in their chosen community.
(Anthony-Stevens, Stevens, & Nicholas, 2017)
Teaching to the test 227
The students who Stevens describes are familiar with a pertinent skill or concept
or at least have the building blocks for such a skill. However, the schools they’re
required to attend fail to draw out and to develop that familiarity.
So, we cannot discount Sarah’s case as simply fantastic and implausible. And
we cannot explain her situation away by pointing to some failure of motivation
on her part. We have to explain why it is that a student who has certain skills
or familiarity with certain concepts fails to use them in formal educational con-
texts. This will be the project of this chapter. I propose that part of the cause of
Sarah’s situation is that schools discourage a particular kind of virtuous thinking
– they discourage what I’ll call epistemic phronesis. I’ll defend this claim by first
explaining that concept before going on to argue that contemporary education
plays a role in discouraging epistemic phronesis. I’ll close by considering future
work.
12.2 Epistemic phronesis

In his Nicomachean Ethics, Aristotle defines phronesis as “a true and reasoned
state of capacity to act with regard to the things that are good or bad for a man”
(1104b5). Phronesis is a kind of cultivated practical know-how. The person who
has phronesis – the phronimos – knows when x is a suitable action, given the sort
of context she is in. She has insight into how to act in a particular situation, mani-
fested by a certain exemplary style of agency.
The relationship between phronesis and other virtues is contentious. Some
argue that phronesis is the “intellectual excellence that is operative in behavior
that manifests good character” (McDowell, 1998, pp. 27–28). This would mean
that phronesis is one intellectual virtue that a person can have. It may also be that
the phronimos just has all the other virtues – that being phronetic is necessary
and sufficient for being virtuous. Or, we might hold that phronesis is a virtue, but
one of several higher-order virtues. Adjudicating this disagreement is beyond the
scope of the current chapter. Indeed, my thesis should follow, albeit with some
changes in wording, however the relationship is best understood. For the purposes
of this chapter, I will follow Paul Bloomfield who, in his writing about phrone-
sis, argues that “phronesis is necessary for all the virtues but is not sufficient”
(Bloomfield, 2013, p. 288). We must be phronetic in order to manifest the other
virtues, though we may need to do other things as well.
The phronimos is able to accurately perceive what is to be done. This abil-
ity builds on a number of others. As Rosalind Hursthouse and Glen Pettigrove
put it, “moral sensitivity, perception, imagination, and judgement informed by
experience—phronesis in short—is needed to apply rules or principles correctly”
(Hursthouse & Pettigrove, 2018). This means that knowing some set of rules
or norms is insufficient. A person must also apply those rules in a way that is
informed by the relevant features of the situation together with her experience.
Only then will she have phronesis.
A person has phronesis, as I understand it, when she is able to do two things:
she must be able to sufficiently accurately assess the situation in which she finds
228 Casey Johnson
herself. This is the perceptual/sensitivity part. She must also use that assessment
to decide what to do. This might be by activating the right mental habits, or it
might be by engaging in a degree of reflection on and understanding of her rel-
evant experiences. She must properly assess the situation, and she must properly
use that assessment, and then she will be in a position to know that this is an occa-
sion to do that.
A case may help illuminate phronesis further. Imagine that Sally is trying to
be virtuous. She attempts to manifest the correct degree of courage, and to hit the
golden mean of temperance. However, she makes mistakes about when to use the
skills associated with each virtue. She attempts to be temperate when courage is
called for. She has the skills to be virtuous, but she enacts them in the wrong con-
texts. Or, perhaps worse, she never identifies the contexts in which to use them.
She’s practiced being temperate, or learned to be humble, but she fails to recognize
those occasions on which she should bring those skills into play. We can imagine
that Sally would be, counterfactually, virtuous if she only knew when to do what.
Because of her lack of phronesis, Sally’s attempts to be virtuous are foiled.
With this basic understanding of phronesis in place, we can return to Sarah
and her math problem. Sarah has the skills necessary to do math. She knows how
to follow the rules of arithmetic, even if she doesn’t know what those rules are
called. She may not even know how to explain the rules – this would be a kind of
meta-mathematical knowledge. However, it is clear from her conversation with
Wallace that Sarah’s problem is not that she cannot do the math problem. Her
problem, instead, is that she doesn’t know that this is a problem in which she can
bring those particular arithmetic skills to bear. Sarah does not know that home-
work is an occasion to put her math-relevant experiences to use. This is either
because she fails to assess the homework context properly or because she fails to
use that assessment properly in deciding what to do.
Of course, arithmetic skills aren’t always considered to be a virtue. Linda
Zagzebski distinguishes between skills and virtues saying that skills have con-
tingent value – they are good insofar as a situation makes them useful. Virtues,
according to Zagzebski, have intrinsic value (Zagzebski, 1996). Similarly, Jason
Baehr claims that skills are “abilities to perform certain reasonably specific or
technical intellectual tasks” (Baehr, 2011, p. 29). We usually think of intellec-
tual virtues as including such traits as curiosity, humility, tenacity, diligence,
etc. While Sarah’s math problem-solving skills are intellectual skills, they aren’t
clearly character traits and they don’t clearly have intrinsic value. This does not
mean, however, that virtues and skills aren’t closely related. Many virtues involve
enacting a skill. Verbal skills can help one to be a virtuous communicator, for
example. It is important to keep in mind, though, that virtues and not skills are
involved in phronesis. To better understand epistemic phronesis, we’ll make a
brief detour into some of the details of virtue epistemology.
There are two main schools of thought regarding epistemic virtues and vices.
One, defended Zagzebski (among others), has it that virtues and vices are acquired
character traits. They are, “deep quality of a person, closely identified with her
selfhood” (Zagzebski, 1996, p. 104). Call this virtue-responsibilism. The other
main position, virtue-reliabilism, contends that virtues and vices are cognitive
faculties, virtuous when they’re stable and reliable, and vicious otherwise. Virtue-
reliabilism has been defended most notably by Ernie Sosa (Sosa, 1980). To get
clear on epistemic phronesis, I’ll discuss epistemic phronesis as it fits with each
of these approaches to epistemic virtues and vices.
First, we’ll turn to virtue-reliabilists. Sosa developed virtue-reliabilism in order
to answer certain problems for his preferred account of justification, arguing that a
person can be justified in forming the belief that x just in case that belief was formed
by a reliable process (Sosa, 1980). A process is reliable just in case it produces more
(maybe quite a few more) true beliefs than false ones. This focus on reliable pro-
cesses leads virtue-realiabilists to diagnose as virtues traits like attentiveness, good
memory, acute reasoning, etc. Persons who are disposed to attain more true beliefs
than false ones – that is who use reliable belief-forming processes, are virtuous on
this picture. Virtue-reliabilists are comparatively less interested in accounting for
the various moving parts of Aristotelean virtue ethics and so don’t say much about
phronesis. We can, however, fill in some details. A process may be reliable in some
contexts and not in others, according to virtue-reliabilism. So, a person might be
said to be a phronimos on such a view when she knows which belief-forming pro-
cess to use in which context. She’ll use whichever is or are reliable.
Next, let’s take the virtue-responsibilists. For a person to have an epistemic
virtue, on this view, that person must have the characteristic in question in a
particularly deep way. In describing this, Zagzebski writes that the characteris-
tic “becomes entrenched in a person’s character and becomes a kind of second
nature” (Zagzebski, 1996, p. 116). In addition to this depth, the virtuous person
is also motivated use the character trait in question in order to form true beliefs.
So, if a person has the virtue of curiosity, her curiosity comes naturally, and she is
motivated to use her curiosity-related skills to improve her doxastic position. The
virtue-responsibilist conceives of epistemic virtues as character-based traits such
as inquisitiveness, humility, intellectual courage, etc. (Watson, 2015).
Zagzebski’s view of phronesis is enlightening, not least for emphasizing that
learning phronesis is a social matter. People learn to be virtuous by following
the example of a phronimos. People learn what is to be done by seeing a practi-
cally wise person acting virtuously. Phronesis must be acquired by social experi-
ence. Epistemic phronesis, then, is a matter of being well-placed in a sufficiently
healthy epistemic community. Indeed, Zagzebski says:
When I am in a position of trying to find out whether or not to believe some-

thing, it is important for me to connect to the social network of beliefs in the
proper way. I must know where to look – what books to read, which people
to consult.
(Zagzebski, 1996, p. 228)
And this social network must be sufficiently receptive, supportive, and sensitive
to my epistemic needs and abilities. If I am in such a position, I will better know
which intellectual skills and traits to employ.1
230 Casey Johnson
These are competing notions of intellectual virtues that produce competing
concepts of phronesis. However, by using either one we can understand why
phronesis is important and why formal education discourages it. To show this,
and with these possible understandings of epistemic phronesis in mind I’ll close
this section by considering two additional cases. Together with Sarah’s case dis-
cussed above, these cases demonstrate that a person might have the skills involved
in an epistemic virtue (or have the trait to some degree, or use the process in some
context), yet lack phronesis and so fail to deploy these virtues in formal educa-
tional contexts.
Consider Jamie, a college student from a conservative Christian background.
His parents taught him to value formal education as a means to an end and to treat
authority figures with respect. Jamie finished high school with good grades and
decent SAT scores and enrolled in his state’s land grant university. Jamie partici-
pates in a Christian fellowship group in a church just off campus. The group meets
weekly to discuss matters of theology and faith over pizza. Jamie is frank and
vocal in these conversations. He asks relevant, careful, and challenging questions.
He takes argumentative and epistemic risks and demonstrates some flexibility
about his beliefs at a granular level.
Jamie is also enrolled in a political philosophy class to fulfil a general educa-
tion requirement. Jamie disagrees with several of the values that he thinks are
taken for granted in the class, but he does not speak up. When his professor gives
arguments for liberal viewpoints, Jamie fails to listen in an open-minded way. He
explains away her points as mere “identity politics” and “political correctness”.
The views and theories are, to Jamie, like a kind of fiction he must learn about,
but one that has little to do with his life. Jamie does not contribute to classroom
discussions, though he is careful to do the reading and performs well on exams.
He behaves respectfully toward his professor but does not deeply engage with the
course content.
On my analysis, Jamie lacks phronesis, and so doesn’t know how to follow the
appropriate epistemic principles in this formal educational context. He has many
of the epistemic skills necessary to make him a virtuous interlocutor. He is curious
and inquisitive. He is open-minded to other people’s arguments regarding some
subject-specific challenges to his beliefs. Yet when he is in a formal educational
context, all of those traits fall by the wayside. Jamie understands his job in the
classroom to be to read and regurgitate the information his professor tells him. He
does not take any of the content onboard, nor does he engage with it critically. He
is intransigent, not because he is engaging with the course content and digging
into his antecedent views, but rather because he does not see the theories as hav-
ing anything to do with him. As Jamie sees it, he is doing his part in the classroom.
It doesn’t occur to him that he should or even could challenge the professor and
engage in a dialog with her. If he did think about it, he wouldn’t see the point.
This, he would think, is not the place where we do that. He has the skills, but he
does not see the classroom as the place to deploy them.
Next, consider Alex, an 18-year-old who wants to enter a technical college
once she graduates high school. She plans to be a dental hygienist and knows that
she needs to do well in her biology class to get into her preferred program. One
of the class requirements is a research paper on cell mitosis. Alex missed several
days of class last semester due to a bad flu and does not have the notes on mitosis.
She is entirely paralyzed by this situation. She has access to the library and the
Internet, but she does not know how to use them in this novel way. This is despite
the fact that Alex is a social media master. Alex can find out who is going to
whose party with a few clicks and taps. She negotiates tricky social situations with
grace and flexibility. She is able to seamlessly navigate different apps, to control
what information goes where, and to solve complicated social and technological
problems. When she doesn’t know how to solve a problem she is facing, Alex
tries novel variations of solutions that have worked before on similar problems.
She asks friends for information in careful, subtle, and effective ways. She is
confident that there is a solution without being cocky that she already knows what
the solution is.
On my analysis, Alex, like Jamie and Sarah, lacks phronesis. She has epis-
temic traits relevant to solving her biology class problem: she is creative, and
innovative and determined. She is humble and also tenacious. But she manifests
these traits only outside of the classroom. She does not have experience being sen-
sitive to epistemic possibilities in formal educational contexts the way she does
in her social life. She fails to perceive that this is a context in which enacting the
virtues would be right and proper.
If we take on virtue-reliabilist’s understanding of phronesis, we can see that
Jamie and Alex do not know what belief-forming processes to use in which con-
texts. Jamie does not know that questioning and challenging are reliable belief-
forming processes in his classroom. While his religious upbringing has encouraged
the development of those sorts of processes in the context of fellowship, his for-
mal education has not done so. He doesn’t know that this is a place to use those
skills. Alex does not know that she can use the same skills in the classroom that
make her so successful online.2
If we take virtue-responsibilism, on the other hand, we can see that Jamie does
not see his professor as the sort of person with whom he can converse. His previ-
ous formal education discouraged him from doing this, and did not provide him
with any role models. He does not call on these traits as second nature, because he
doesn’t know that this is a context in which to use them. And something similar
is happening in Alex’s case. She is motivated to seek the truth about mitosis, but
she doesn’t know – and no one has shown her – that she can be just as creative in
her academic problem solving. School has not provided Alex or Jamie with any
role models with phronesis. It has not given them the opportunity to develop their
sense of what to do when.
For Alex, Jamie, and Sarah, a lack of phronesis renders them unable to assess
their epistemic environment properly. They have the relevant skills, they are
developing the necessary virtues, but they fail to assess their situation properly
and so fail to use that assessment in deciding what to do. This means that they
don’t know what is to be done. They are not yet vicious, but they are nonetheless
kept from being virtuous.
232 Casey Johnson
To be clear, unlike Sarah’s case, or the case of the Apache student, these
cases entirely are contrived: I made them up. So, I’m not claiming that these
are real-life examples for which any good theory must account. I’m merely
suggesting that these cases are plausible, and that the idea of a failure of epis-
temic phronesis allows us to get traction on what is going on in them. When we
encounter students or other epistemic agents who appear not to have epistemic
virtues in formal educational contexts, I encourage us to consider whether they
are vicious, or whether they lack phronesis. I’m deliberately emphasizing for-
mal educational contexts. This is because I think, as I’ll detail in the next sec-
tion, that contemporary public education is a large part of the cause of failures
of epistemic phronesis.
12.3 Educating against phronesis

Contemporary public education in the United States is dominated by the legacy
of the 2001 No Child Left Behind Act (NCLB). The act was adopted, in part,
to increase federal and state-wide accountability for teachers and schools and to
close the achievement gap between high- and low-performing schools – espe-
cially in those cases in which performance is correlated with school funding and/
or race. NCLB established requirements for teacher qualifications and for student
achievement (Duffy et al., 2009). Schools that fail to meet state standards receive,
in accordance with this act, various sanctions often including increased control of
schools by state and federal authorities. As David Hursh puts it,
These include providing students with: “supplemental services in the com-

munity such as tutoring, afterschool programs, remedial classes or summer
school,” replacing the school staff, implementing a new curriculum, “decreas-
ing management authority, appointing an outside expert to advise the school,
extending the school day or year, or reorganizing the school internally.”
Schools failing for five consecutive years must either reopen as a charter
school, replace all or most of the school staff who are relevant to the failure to
make adequate yearly progress, or turn over the operations either to the state
or to “a private company with a demonstrated record of effectiveness” (US
Department of Education, 2003c, pp. 6–9). Many of the “remedies,” such as
tutoring, remedial classes and replacing the administration, provide opportu-
nities for private corporations to profit from public funding.
(Hursh, 2007, p. 297)
These sanctions imposed an incentive structure on public schools to meet the

standards laid out by NCLB. For student achievement, the act mandated regular
periodic testing to establish whether students are meeting standards. These tests
have high stakes for teachers, school administrators, and schools themselves. This
has encouraged increased testing in math and reading, the areas emphasized by the
act, increased “teaching to the test”, and decreased focus on subjects not empha-
sized by the act (Hursh, 2007). These subjects include social sciences, so-called
enrichment subjects like music and art, and, until very recently, natural sciences
(Jennings & Rentner, 2006).
This notion of “teaching to the test” is in our vernacular, but what does it
involve? Education seems to require assessment, and certainly, we want to know
how well schools are educating their students. However, “teaching to the test” is
negatively connoted – it is often taken to be emblematic of the ills of contempo-
rary education and education policy. So why is teaching to the test bad?
Teaching to the test has involved two distinct aspects. First, there are the tests
themselves – often rote-based and repetitive, these tests incentivize memorization
and “parroting” rather than deep understanding. Second, there is the testing sched-
ule – frequent high-stakes testing interrupts the flow of the school year and dis-
tracts teachers from other educational goals that would better meet their students’
needs. We can imagine having either of these aspects without the other. Indeed,
some states have made efforts under the new Every Student Succeeds Act (ESSA)
to change the testing schedule. ESSA, which replaced NCLB in 2015, allows
states to determine when and how often to test their students. We’ll briefly return
to the ESSA below, noting here that while the new act may improve on NLCB,
standardized testing remains central to the legislation. This means, in part, that
we can anticipate that schools will continue to be incentivized to teach to the test.
According to Louis Volante, teaching to the test involves certain kinds of
test preparation practices. These might include repetitive exposure of students
to questions from past tests, or test questions “cloned” from past test (questions
with some variables and the answer changed). It might include rote memoriza-
tion of likely test answers. In some cases, teachers faced with high-stakes testing
have even given students test answers, changed student answer forms, or other-
wise improperly influenced test results (Volante, 2004). Standardized testing is
supposed to measure performance on various kinds of activities, testing types of
skills, and using these as a measure of background ability. This means that teach-
ing to the test, as described by Volante, renders standardized tests impractical and
useless.
So, teaching to the test is bad if we want to use standardized tests to assess
student achievement. Teaching to the test is also bad for students’ learning out-
comes.3 As Joan Herman puts it, “the time focused on test content has narrowed
the curriculum by overemphasizing basic-skill subjects and neglecting higher-
order thinking skills” (Herman, 1992, p. 74). Examples of higher-order thinking
skills include thinking creatively, solving novel problems, constructing research
projects, and identifying implicit assumptions (Bloom, Englehart, Furst, Hill, &
Krathwohl, 1956; Zohar & Dori, 2003). It is worth noting that many of these
higher-order thinking skills are involved or play a role in epistemic virtues.
Consider curiosity, creativity, and humility, as examples.
The effects of the NCLB continue to echo through public education even
though ESSA has replaced that legislation. ESSA makes two notable changes to
NCLB. The first is to allow states, rather than the federal government, to deter-
mine the details of the schedule of consequences for schools that fail to meet state
standards. The second is to give states more control over the number and length
234 Casey Johnson
of tests that students take. One year into the transition from NCLB to ESSA,
25 states reported interest in decreasing the amount of testing to which their stu-
dents are subjected (Rentner, Kober, & Frizzell, 2017). This, perhaps, reflects a
realization that this testing has deleterious educational effects.
Students who receive their education by rote memorization or test-focused
teaching receive a particular kind of education. This kind of education is, at least
in some ways, quite narrow and fails to encourage higher-order thinking. Even as
state standards expand to include natural sciences, students’ success is measured
by performance on a test. These tests must, to do the job they’re designed to do,
be standardized. This means that the students must be able to perform a task in
a particular way. As Martha Nussbaum puts it, teaching to the test “produces an
atmosphere of student passivity and teacher routinization” (Nussbaum, 2016, p.
134). The test cannot be flexible enough to accommodate students who under-
stand two-dimensional drawings of tessellations or tessellations as they appear in
stacked wood. Teaching to the test emphasizes first-order thinking over higher-
order skill building, excluding things like curiosity, creativity, and other virtues.
Further, teaching to the test is not serving the U.S. particularly well. In addi-
tion to the harms faced by individual students, particularly those in historically
failing schools, the United States is not performing well on international com-
parisons. According to the Programme for International Student Assessment data
from 2015, the U.S. is ranked 38th out of 71 countries for math performance,
and 24th in science (Desilver, 2017). This, despite the focus of NCLB and ESSA
on math. Further, if Martha Nussbaum is correct, by teaching to the test, we are
producing students who lack the creativity and higher-order thinking necessary to
be productive members of society or well-rounded and participating members of
democracy (Nussbaum, 2016).
Beyond this, I contend that students who are taught to the test are discour-
aged from knowing when and how to manifest their epistemic virtues in school.
Nonetheless, they may (and I believe likely do) still display epistemic virtues in
many parts of their lives – they simply don’t seem to know that school is a place
where these virtues apply. This is why the claim of this chapter is not merely
that young people lack phronesis – this would not be surprising given that phro-
nesis is learned through experience and young people necessarily lack experi-
ence. Instead, the claim of this chapter is that contemporary schooling in the U.S.
actively discourages students from developing phronesis. Teaching in a narrow,
rote, test-based way gives students the impression that epistemic work is matter
of regurgitating the teacher-/test-sanctioned answer. Many students don’t realize
that educational contexts are appropriate for curiosity, humility, and other epis-
temic virtues because in their test-based experience, that development is neither
recognized nor rewarded.
For students who are particularly savvy or reflexive, the lack of reward might
be interrupting their development of phronesis at a slightly different stage. They
might have some idea that creativity or conviction could be helpful in an educa-
tional context, but nonetheless take what might be called the “safe” or “conserva-
tive” route, instead. Consider Jennifer, who is a friend and classmate of Jamie’s.
Jennifer and Jamie attend the same church, Jennifer also participates in the lively
biblical debates, and Jennifer is also enrolled in Jamie’s political science class.
She begins to grapple with the content presented in class, attempting to assimilate
it with her views where she can and come up with objections otherwise. However,
like Jamie, Jennifer never raises her hand to voice her objections. She expresses
no dissent in class, and so has no practice doing so. She knows from her educa-
tional background that regurgitation of content is what it takes to get a good grade,
so that’s what she does. Rather than take social, educational, and epistemic risks,
she plays it safe and dutifully repeats what she’s memorized. She falls short of
phronesis because she does not recognize the value of engaging her nascent vir-
tues in this educational context. If I’m right, then we might usefully understand
contemporary schooling in the U.S. as epistemically corrupting. Ian James Kidd
introduces a notion of epistemic corruption as it relates to epistemic vices and vir-
tues especially in the educational context (Kidd, 2019). I will spend the remainder
of this section extending Kidd’s notion of epistemic corruption to cover cases
wherein phronesis is discouraged.
According to Kidd, a set of practices (i.e. an educational context or set of
testimonial norms) is epistemically corrupting insofar as it fails to promote
epistemic attainments. These practices can involve either active corruption, or
passive corruption, or both. Active corruption encourages agents to develop
epistemic vices. This might involve a set of practices that encourages dogma-
tism, or arrogance. Passive corruption, on the other hand, fails to encourage or
enable epistemic virtues (Kidd, n.d.). And Kidd is clear that educational con-
texts often offer sets of epistemically corrupting practices. I think that Kidd is
correct that contemporary schooling can be corrupting in these ways. A student
might be encouraged to be dogmatic by attending a school that rewards hard-
headed conviction in one’s beliefs despite the evidence. A student might also
be discouraged from developing curiosity if, for example, her teacher scolds
her for asking too many questions. Kidd’s work helpfully highlights the ways
in which formal education can keep people from developing the virtues, or can
encourage epistemic viciousness
I want to suggest, however, that a set of practices might be neutral with regard
to virtues and vices, but nonetheless discourage the kind of accurate perception
and experiential reflection required for someone to develop phronesis. Certainly,
active corruption is not necessary for a set of practices to discourage phronesis.
Some schools might encourage vicious behavior, but they need not in order to
discourage students from developing phronesis. On the view I’m defending, an
agent might be able to enact the virtues in some contexts, but fail to do so in
their formal education. And I think this is largely friendly to Kidd’s view; as
Kidd himself notes, epistemic corruption “does not, of course, imply that students
are thereby debarred from cultivating their virtues, since education is only one
place where they can do that” (Kidd, 2015). However, if educational contexts
themselves are neither encouraging vice, nor discouraging virtue (as virtues are
being developed), yet they are epistemically corrupting, they must be doing so in
a different way. I suggest that this is by way of discouraging epistemic phronesis.
236 Casey Johnson
As the cases of Sarah and Alex and Jamie and Jennifer show, even if students
are able to develop their virtues in other places, they are not always able to bring
those virtues to bear in educational contexts. Contemporary education discour-
ages them from doing so. The mistake that formal education is making, in all
of these cases, is cordoning off educational contexts from the rest of students’
epistemic lives. The message is that while perhaps in the streets, or in the fellow-
ship hall, or online, certain virtues are called for, this is not the place for that. In
here – that is, in school – we take tests, we memorize answer sets, and we solve
the problems in these and only these ways. Contemporary schooling, then, may
be actively corrupting, it may be passively corrupting (in Kidd’s senses), and it
might also be corrupting in an additional way: it might prevent the development
and exercise of phronesis.
12.4 Future work

In this section I will sketch some avenues for future work. I will restrict myself to
work at the intersection of philosophy and educational practice, though this work
clearly suggests political, curricular, and policy projects as well. Philosophers
who are convinced by at least some of what I’ve said above might consider these
subsequent projects.
First, there is important work already in progress on educating for the virtues.
Heather Battaly is using virtue epistemology to make practical suggestions to
educators (Battaly, 2006). Lani Watson has laid out the value of educating for
virtues like curiosity and good questioning (Watson, 2018). These are projects
in virtue epistemology as it applies to education in a more or less concrete way.
I suggest adding to that work some careful thinking about what educating for
phronesis would look like. Some work has begun on this already (Noel, 1999).
Because phronesis is developed by experience, this work could be especially rich
and important in fixing educational practice. If I’m right in the above then our
current educational system, at least in the U.S., actively works against students
developing phronesis. This future project would, in part, be an attempt to sort out
how to rectify that.
Second, identifying the relevant epistemic phronimos. What does an epistemic
phronimos look like? And how could we identify one? Are there sensitivities
to practices and principles that are specific to formal educational contexts? Or
trackable ways that epistemic principles are manifested in formal education? The
phronimos, for Aristotle, is an important measure of virtuousness. The golden
virtuous mean between vices is not the sort of thing that can be discovered a
priori. However, we can discover it if we can know what the phronimos would
do. It would be useful, then, to be able to identify people with epistemic phro-
nesis. Maria Silvia Vaccarezza and Michel Croce have relevant work on moral
exemplars, building from the work of Zagzebski (Croce & Silvia Vaccarezza,
2017). It would be of interest to investigate the ways their account would need
to be adjusted to discover epistemic exemplars. Perhaps these would be good
teachers. Perhaps students could learn from their experience. Perhaps, however,
these would be exemplars from outside of education. After all, if we retain our
educational commitment to teaching to the test, students will have to continue to
look outside of formal education to develop their virtues and their phronesis. The
question, then, would be translating those epistemic practices into legibility and
usefulness in formal schooling contexts.
Finally and relatedly, there is work to be done picking up and developing
Zagzebski’s idea that well-functioning social groups are instrumental for good
epistemic agency. Recall that according to Zagzebski, part of the process of form-
ing a belief involves participating in a social network of beliefs. If this is right,
and if, as I contend, this participating is also instrumental for phronesis, then there
is an interesting intersection between questions in social epistemology and ques-
tions in virtue epistemology. What kinds of social networks help us to develop
phronesis? What kinds of social networks do epistemic exemplars enjoy? These,
and the above, are the sorts of theoretical questions we might ask, given that epis-
temic phronesis is necessary for deploying the virtues.
Notes
1 I am grateful to Ian James Kidd for pointing this out. Kidd’s work on epistemi-
cally corrupting environments is relevant here and throughout. I elaborate on points
related to the sensitivity requirement in my as yet unpublished work on epistemic
vulnerability.
2 This is, in some relevant ways, the dual of what Kidd, in this volume, calls propaga-
tion. For Kidd, a person’s character can become increasingly more vicious as previ-
ously localized vices spread to “affect the range of a subject’s character” (Kidd 2020)
A phronimos’ virtues have affected the range of that subject’s character so the virtues
can be called upon when it is proper to do so.
3 At least insofar as we count higher-order thinking and problem-solving skills as among
our learning goals.
References
Anthony-Stevens, V., Stevens, P., & Nicholas, S. (2017). Raiding and alliances: Indigenous
educational sovereignty as social justice. Journal of Critical Thought and Praxis, 6(1), 3.
Baehr, J. (2011). The Inquiring Mind: On Intellectual Virtues and Virtue Epistemology.
Battaly, H. (2006). Teaching intellectual virtues: Applying virtue epistemology in the
classroom. Teaching Philosophy, 29(3), 191–222.
Bloom, B. S., Englehart, M. D., Furst, E. J., Hill, W. H., & Krathwohl, D. R. (1956).
Taxonomy of educational objectives: Handbook I. Cognitive Domain. New York:
David McKay.
Bloomfield, P. (2013). Some intellectual aspects of the cardinal virtues. In M. Timmons (Ed.),
Oxford Studies in Normative Ethics (pp. 287–313). Oxford: Oxford University Press.
Croce, M., & Silvia Vaccarezza, M. (2017). Educating through exemplars: Alternative
paths to virtue. Theory and Research in Education, 15(1), 5–19.
Desilver, D. (2017). U.S. students’ academic achievement still lags that of their peers in
many other countries. PEW Research Centre, https://www.pewresearch.org/fact-tank
/2017/02/15/u-s-students-internationally-math-science/ Accessed 2 June, 2020.
238 Casey Johnson
Duffy, M., Giordano, V. A., Farrell, J. B., Paneque, O. M., & Crump, G. B. (2009). No
Child Left Behind: Values and research issues in high‐stakes assessments. Counseling
and Values, 53(1), 53–66.
Herman, J. L. (1992). What research tells us about good assessment. Educational
Leadership, 49(8), 74–78.
Hursh, D. (2007). Exacerbating inequality: The failed promise of the No Child Left Behind
Act. Race Ethnicity and Education, 10(3), 295–308.
Hursthouse, R., & Pettigrove, G. (2018). Virtue ethics. In E. N. Zalta (Ed.), Stanford
Encyclopedia of Philosophy, https://plato.stanford.edu/archives/win2018/entries/ethics
-virtue/ Accessed 18 September, 2020
Jennings, J., & Rentner, D. S. (2006). Ten big effects of the no child left behind act on
public schools. Phi Delta Kappan, 88(2), 110–113.
Kidd, I. J. (2015). Educating for intellectual humility. In J. Baehr (Ed.), Intellectual Virtues
and Education (pp. 54–70). New York: Routledge.
Kidd, I. J. (2017). Capital epistemic vices, Social Epistemology Reply and Review Collectie
8(8), 11–16.
Kidd, I. J. (2019). Epistemic corruption and education. Episteme, 16(2), 220–235.
Kidd, I. J. (n.d.). Epistemic corruption and social oppression. In I. J. Kidd, Q. Cassam, &
H. Battaly (Eds.), Vice Epistemology. London: Routledge.
McCarthey, S. J. (2008). The impact of No Child Left Behind on teachers’ writing
instruction. Written Communication, 25(4), 462–505.
McDowell, J. (1998). Mind, Value, and Reality. Cambridge, MA: Harvard University
Press.
Noel, J. (1999). On the varieties of phronesis. Educational Philosophy and Theory, 31(3),
273–289.
Nussbaum, M. C. (2016). Not for Profit: Why Democracy Needs the Humanities-Updated
Edition. Princeton, NJ: Princeton University Press.
Rentner, D. S., Kober, N., & Frizzell, M. (2017). Planning for Progress: States Reflect on
Year One Implementation of ESSA. Center on Education Policy. https://files.eric.ed.gov
/fulltext/ED583013.pdf
Rushton, S., & Juola-Rushton, A. (2008). Classroom learning environment, brain research
and the no child left behind initiative: 6 years later. Early Childhood Education Journal,
36(1), 87.
Sosa, E. (1980). The raft and the pyramid: Coherence versus foundations in the theory of
knowledge. Midwest Studies in Philosophy, 5(1), 3–26.
Volante, L. (2004). Teaching to the test: What every educator and policy-maker should
know. Canadian Journal of Educational Administration and Policy 35. https://cjc-rcc
.ucalgary.ca/index.php/cjeap/article/view/42715
Watson, L. (2015). What is inquisitiveness. American Philosophical Quarterly, 52(3),
273–287.
Watson, L. (2018). Educating for good questioning: A tool for intellectual virtues education.
Acta Analytica, 33(3), 353–370.
Zagzebski, L. T. (1996). Virtues of the Mind: An Inquiry into the Nature of Virtue and the
Zohar, A., & Dori, Y. J. (2003). Higher order thinking skills and low-achieving students:
Are they mutually exclusive? The Journal of the Learning Sciences, 12(2), 145–181.
13 Vices of questioning in
public discourse
Lani Watson
Questioning is ubiquitous and habitual in our daily lives. We ask questions all
the time, often without reflecting consciously on the practice. Sometimes it goes
well, sometimes it doesn’t, and often we don’t notice the difference. Questions
are also a familiar feature of public discourse, and here the difference between
good and bad questioning can have important and sometimes damaging effects:
a leading question that influences the results of a referendum, a loaded question
that forces a prejudicial response in public debate, the aggressive or insensitive
questioning of journalists hungry for a story. In this chapter I investigate what
makes questioning bad (Section 13.1) then offer a taxonomy of bad questioning
practices (Section 13.2). Drawing on examples of questioning in contemporary
politics, I go on to discuss the nature and impact of bad questioning in the public
sphere (Section 13.3). I argue that bad questioning is an intellectual failing often
expressed in intellectual vices such as negligence, closed-mindedness and arro-
gance (Section 13.4). As such, bad questioning in the public sphere degrades the
professional character of, for example, journalists and politicians and undermines
the wider role that they play in our epistemic communities. I conclude that greater
attention should be paid to questioning practices in public and political forums in
order to check and maintain the epistemic and characterological integrity of key
social institutions (Section 13.5).
13.1 What is bad questioning?

Most of us have some intuitive grasp of what bad questioning looks like. We
know that asking a toddler for directions to the train station, or typing ‘where
are my keys’ into Google, is probably not the best way to get the information we
are after. But what exactly is it that makes questioning like this bad? In order to
answer this, it will be instructive first of all, to take a look at what makes question-
ing good. This will shed light on the many ways in which one can fail to be a good
questioner. Many (although not all) of these lead to bad questioning. I will draw
on an analysis of good questioning that I have provided in more detail elsewhere
(Watson 2018). As a starting point, I take questioning to be essentially a form of
information-elicitation. When one engages in questioning, whether good or bad,
one is typically in the business of trying to find things out. This is not, of course,
240 Lani Watson
the only reason we ask questions, but I take it to be the defining function of the
practice. A question is a question in virtue of its information-eliciting function.
Put another way, a question is an information-eliciting tool. This is so, even if we
sometimes, perhaps often, use the tool for some other reason. We may ask a ques-
tion in order to be polite, to show compassion or to humiliate someone. In each
case, the question is an information-eliciting tool used to achieve a further goal.
Any tool can be used to greater or lesser effect. In order to be a good questioner,
one must use the tool, that is, engage in the activity of eliciting information, skil-
fully. This elevates good questioning above mere questioning in two ways. In
cases of good questioning we do not simply want to elicit information, rather we
want to (1) competently elicit information that is (2) worth having. What exactly
does this mean?
Starting with competence, our efforts to elicit information must not overly rely
on acquiring information by luck or chance, even if those do, in certain cases,
enable one to successfully acquire information. Conversely, while successfully
eliciting information requires that one acquire that information, competently elic-
iting information does not; one can act competently in order to elicit information
even if one is unsuccessful in actually acquiring it. The good questioner will do
just this; she will act competently in order to elicit information. Good question-
ing also requires that the information one elicits is worthwhile. When we engage
in good questioning we do not merely want to elicit any information, however
competently, but to elicit information that is worthwhile, relevant or significant in
some sense; information that is worth having. This requires one to exercise judge-
ment about what information one elicits. The good questioner will avoid eliciting
trivial or disvaluable information, as well as the large amount of irrelevant or
insignificant information that is available to her. As such, the good questioner acts
competently in order to elicit worthwhile information.
These distinct aspects of good questioning can be approximately aligned with
the content and the performance of a question. The content of a question refers
to the information being sought. In other words, the content is what the question
asks, its subject matter. If one asks ‘what is the time?’, the content of the question
is information about the time. The good questioner must decide what to ask and
the content of her question must be worthwhile information. The performance of a
question refers to the manner in which it is asked. In other words, the performance
is how, when and where a question is asked, as well as whom it is asked of. One
may ask abruptly, forcefully, politely or in Spanish. One may ask in the morning
or just before the meeting. One may ask at the bus stop or on live television. One
may ask an adult or a child or a search engine. These are all aspects of the per-
formance of a question. The good questioner must decide how, when, where and
whom to ask and, in doing so, put herself in a position where she is most likely
to get the information she is after. That is what it means to act competently in the
case of good questioning. A good questioner acts competently in order to elicit
worthwhile information by determining what, how, when, where and whom to
ask. In other words, she asks the right thing, of the right source, at the right time
and place, in the right way. Clearly, there is more that can be said about what
Vices of questioning in public discourse 241
makes questioning good. My aim has been to lay the groundwork for our present
focus; what makes questioning bad?
The analysis of good questioning provides an insight into bad questioning.
Simply put, the bad questioner falls short on one or more of the aspects of good
questioning; she asks the wrong thing, or the wrong source, or she asks at the
wrong time, or in the wrong place, or in the wrong way. Of course, she may
do more than one of these things simultaneously. It is easier to be a bad ques-
tioner than a good one. Good questioning is relatively hard and can go wrong
in a multitude of ways. When it does, it often amounts to bad questioning. As
such, there are two broadly distinct ways in which bad questioning can arise.
Either the content of the question is not worthwhile, and so the questioner fails
to identify what she should ask, or the performance (asking) of the question is
not competent, and so the questioner fails to identify how, when, where and/or
whom she should ask. Bad questioning is information-elicitation gone awry in
one of these ways.
That is not to say that the bad questioner always fails to get the information
she is after. Nonetheless, being a bad questioner will, more often than not, impede
epistemic progress. This explains why asking a toddler for directions to the train
station amounts to bad questioning. Assuming that one actually wants to get to the
train station, asking a toddler is unlikely to elicit the relevant information: one is
asking the right thing of the wrong person. Likewise, one is almost certainly not
going to find one’s keys by typing ‘where are my keys’ into Google (unless they
happen to be next to the keyboard). Again, one is asking the right thing but con-
sulting the wrong source. If, instead of finding one’s keys, one actually wants to
know what the capital of Eritrea is, then typing ‘where are my keys’ into Google is
bad questioning of a different sort; one is asking the right source, the wrong thing.
Such examples multiply easily.
Significantly, note that what makes this questioning bad is the impact that
it has on the questioner’s epistemic progress. Bad questioning tends to make
it less likely that one will get the worthwhile information one needs or wants.
Questioning is bad insofar as it prevents or impedes the elicitation of worthwhile
information, hence my earlier claim that a question is a question in virtue of its
information-eliciting function, that is, the defining function of the practice. Good
questioning requires the questioner to elicit worthwhile information competently.
Bad questioning is questioning that prevents or impedes this, for one or more
of the reasons above. This allows for questions that fail to elicit worthwhile
information for some other reason. One can ask a good question which fails to
elicit any information at all. Imagine, for example, that the person one asks for
directions to the train station has taken a vow of silence, of which one was not
and could not reasonably have been, aware. One may still be engaging in good
questioning whilst failing to find out where the train station is. Precisely put, a
question is not bad because it prevents or impedes the elicitation of worthwhile
information. Rather, a question is bad if it prevents or impedes the elicitation of
worthwhile information because the questioner has asked the wrong thing, or the
wrong source, or at the wrong time, or in the wrong place, or in the wrong way.
242 Lani Watson
Questioning that prevents or impedes the elicitation of worthwhile information,
for one or more of these reasons, is bad questioning qua questioning.
Contrast this with the many ways in which questioning can be bad according to
some other standard. There is, for example, an important sense in which a ques-
tion can be morally bad. Indeed, the cases I discuss below have a distinctly moral
and/or political dimension. Without diminishing the significance of this dimen-
sion, I want to show how the moral and/or political dimensions of questioning can
themselves prevent or impede a questioner from eliciting worthwhile informa-
tion. Questioning that is morally bad can (and often will) be bad questioning qua
questioning. What makes a question bad qua question is the manner in which it
prevents or impedes the elicitation of worthwhile information and so epistemic
progress more generally.
This helps to illuminate the relationship between bad questioning and intellec-
tual vice. Bad questioning is not an intellectual vice itself, just as good questioning
is not an intellectual virtue. Rather, good questioning is an intellectual skill found
in the exercise of many of the intellectual virtues including, for example, inquisi-
tiveness, open-mindedness, intellectual humility and intellectual courage (Watson
2018). In much the same way, bad questioning is an intellectual failing found in
the exercise of many of the intellectual vices including, for example, carelessness,
dogmatism, prejudice, arrogance, closed-mindedness and negligence. Insofar as
the intellectual vices ‘impede effective and responsible inquiry’ (Cassam 2016),
bad questioning is a mechanism of intellectual vice. I will explore this relationship
further in Section 13.4.
Naturally, the above analysis of bad questioning is oversimplified. Questioning
is a complex practice and determining what makes one good or bad at it is cor-
respondingly complex. Simplified accounts of both good and bad questioning are
required in order to allow for analysis at a useful level of generality. That being
said, much of our intuitive grasp of bad questioning is captured by these accounts.
The accounts, for example, are sensitive to variations in context: the good ques-
tioner must be able to make context-sensitive judgments about when, where and
whom to ask. The bad questioner may fail to do so in any number of different
ways. There is, likewise, nothing in the account of good questioning that sug-
gests there is only one way to be a good questioner. There may be many equally
viable sources of information in any given instance, or many equally worthwhile
things to find out. Thus, while good questioning is relatively hard compared to
bad questioning, given the number of ways in which one can fall short, it is not
necessarily hard in and of itself, given the number of ways in which one can be a
good questioner and the many opportunities one has for practising the skill. Good
questioning can be as simple as asking a person with a watch, ‘what is the time?’
Furthermore, my intention is not to draw a hard line between good and bad
questioning; both must surely come in degrees. If one types ‘where are my keys’
into Google when one actually wants to know what the capital of Eritrea is, one is
clearly going wrong. But this is an unlikely and outlandish example. Subtly and
more plausibly one may want to know which South American countries border
Eritrea and so type this into Google. Strictly speaking, one has asked the wrong
question – Eritrea is a country in northeast Africa. Does this amount to bad question-
ing? In fact, when one types ‘which South American countries border Eritrea’ into
Google, the search engine provides the following information, without judgment:
Eritrea is located in the Horn of Africa and is bordered on the northeast and
east by the Red Sea, on the west and northwest by Sudan, on the south by
Ethiopia, and on the southeast by Djibouti.
One finds out both that Eritrea is in Africa and which countries border it. There is
a sense in which one has asked the wrong question in order to find this out. But
it’s not clear that one’s epistemic progress was prevented or impeded by doing so,
especially given the extraordinary power of the search engine. Should we there-
fore say that this amounts to good questioning? That doesn’t seem clear-cut either.
Technically speaking, one has found out information in response to a question one
didn’t even ask. Ultimately, this question is probably somewhere in the middle; nei-
ther particularly bad nor particularly good. That kind of result should be expected.
Again, since questioning is a complex practice, the evaluation of questions on a
case-by-case basis will be correspondingly complex. In fact, I think we can expect
many cases to be ambiguous in this way. My aim is not to carve questioning up into
two unambiguous evaluative categories – the good and the bad – but to provide a
simplified tool for analysing questions and questioning in complex real-world set-
tings. With that in mind, let’s move on to the taxonomy of bad questioning.
13.2 A taxonomy of bad questioning

Bad questioning is questioning that prevents or impedes the elicitation of worth-
while information because the questioner has asked either the wrong thing, or in
the wrong way, or at the wrong time, or place, or the wrong source. This tells us
what bad questioning is. In order to understand how bad questioning operates in
practice, it will be useful to examine different types of bad questions as they arise
in everyday life. The taxonomy presented in this section provides an overview
– albeit an incomplete one for there are surely bad question types that are not
accounted for and can be added. Nonetheless, the taxonomy covers a wide range
of question types, many of which will be familiar to the reader and, given that the
study of bad questioning is in its infancy, this will serve as a good starting point.
I provide summaries in Figure 13.1 and a more detailed explication for the more
technical question types follows. I also discuss the anomalous place of rhetorical
questions in the taxonomy.
Excepting this anomaly, the question types are broadly aligned with either the
content or the performance dimension of questioning. If a question type is aligned
with the content dimension, then asking a question like this may prevent or impede
the questioner from eliciting worthwhile information. If a question type is aligned
with the performance dimension, then asking a question like this may prevent or
impede the questioner from eliciting that information competently. This is not an
exact science. Often bad questioning will involve falling short in both dimensions
244 Lani Watson
and they will regularly act together. A person’s judgments about whom to ask, for
example, will often be influenced by what they are trying to find out, and so on.
Aligning the different question types with these dimensions is nonetheless useful
for the purposes of identifying and diagnosing bad questioning and, in turn, iden-
tifying the role that bad questioning plays in intellectual vice.
Note also that not all of these practices will always amount to bad questioning.
In fact, arguably none of them will always amount to bad questioning. Whether
they do will depend on whether they prevent or impede the elicitation of worth-
while information for one of the five reasons discussed in Section 13.1. Indeed,
open and closed questions, in particular, will often amount to good question-
ing and both have been advocated by education theorists and practitioners (e.g.
Dillon 1988; Gershon 2013; Worley 2019). On the other hand, one might think
that aggressive or insensitive questions are always bad but, again, this will depend
on whether or not they always prevent or impede the elicitation of worthwhile
information, which seems unlikely. Imagine an aggressive question asked in order
to reveal the location of a hidden bomb.
Similarly, not all of these bad questioning practices amount to intellectually
vicious behaviour and, as we will see in Section 13.4, bad questioning is not nec-
essary for the expression of intellectual vice. Nonetheless, understanding different
types of bad questioning will allow for a more detailed analysis of its relationship
to intellectually vicious behaviour in the public sphere. The purpose of the tax-
onomy is to shed light on familiar question types that can, and often do, amount
to bad questioning qua questioning. This will allow us to better evaluate ques-
tions, determine where they can and do go wrong and, ultimately, identify what is
required in order to ask better questions, in our personal interactions and in public
discourse, and so to diminish intellectual vice.
Many of the question types in this taxonomy need little explication.
Understanding what it means to misdirect a question, for example – to ask the
wrong person or source – is relatively easy, even if identifying when or why a
question is misdirected will not always be straightforward. Others require fur-
ther explication. Absolute questions are questions that contain absolutes such as
‘always’, and ‘ever’, for example, ‘Do you ever use public transport?’, ‘Do you
always vote Labour?’ Absolute questions can be especially problematic when
employed in surveys and polls because they often dramatically reduce the list of
plausible answers and consequently make responses harder to interpret or less
useful. Likewise, compound questions can be problematic in polls. Compound
questions are questions that ask more than one thing but limit the respondent’s
options so that they cannot, or are less likely to, provide all of the relevant infor-
mation. Consider the question, ‘How satisfied are you with the public transport
and services in your area?’ It is not clear how to answer this question if one is very
satisfied with the public transport but less satisfied with the services. Compound
questions can be used to deliberately confuse the respondent or manipulate her
answer. In this way, they are related to complex questions.
Complex questions are questions that contain a presupposition that is not explic-
itly stated such as, ‘Do you vote Labour or Conservative?’ (The presupposition
BAD QUESTIONING
CONTENT PERFORMANCE
Quesõner fails to iden˜fy Quesõner fails to iden˜fy HOW,
WHAT to ask WHEN, WHERE, or WHO to ask
ABSOLUTE Questions that contain absolutes. AGGRESSIVE Questions asked in an aggressive
manner.
CLOSED Questions that require a yes or no CONVOLUTED Questions that are difficult to
answer. follow.
COMPLEX Questions that obscure a DISTRACTING Questions that contain distracting
information.
(AKA TRICK) presupposition.
COMPOUND Questions that ask more than one INEPT Questions asked through an
thing. unsuitable medium.
INAPPROPRIATE Questions that seek out INSENSITIVE Questions asked in an insensitive
inappropriate information. manner.
LEADING Questions biased in favour of an MISDIRECTED Questions asked of the wrong
answer. source.
LOADED Questions that contain a contentious MISPLACED Questions asked in the wrong
presupposition. place.
MISGUIDED Questions that seek out irrelevant MISTIMED Questions asked at the wrong
information. time.
OPEN Questions that require more than a
yes or no.
‘Rhetorical quesõns’: Interrogaves
SLIPPERY Questions that change the
that do not aim at elicing informaon.
information being sought.
Figure 13.1 An incomplete taxonomy of bad questioning

246 Lani Watson
being that the respondent votes for one or the other.) Complex questions are often
referred to as ‘trick questions’ because they can be used to trick a respondent
into accepting a presupposition that they wouldn’t accept if it were stated explic-
itly. Often the presupposition is not immediately obvious, hence the ‘trick’. Even
when the presupposition is obvious, complex questions can be used to force the
respondent into a difficult dialogical position as when, for example, the prosecu-
tor asks the defendant, ‘When did you stop beating your wife?’ The respondent
is forced either to accept the presupposition (that she used to beat her wife) or
refuse to answer the question (which may be incriminating in itself). Complex and
compound questions can both give rise to objections in court. If such an objection
is sustained, the question must be withdrawn and asked in a series of separate
questions so as to avoid confusing the respondent or tricking them into an unin-
tentional admission or commitment hidden in an obscured presupposition. When
used in this way, complex questions are closely related to loaded questions.
Loaded questions are questions that contain a contentious presupposition. The
question ‘When did you stop beating your wife?’ is both complex and loaded
because it contains a presupposition that is not explicitly stated (that the defend-
ant used to beat her wife) and that presupposition is contentious (whether the
defendant did indeed beat her wife is, let us assume, one of the accusations at
issue in the trial). Loaded questions are in turn closely related to leading ques-
tions. Leading questions are questions that encourage the respondent to answer in
a particular way. In a trial or deposition loaded or leading questions can also give
rise to objections from the opposing party, as when one hears in legal dramas that
‘the defence is leading the witness’. If such an objection is sustained, the question
must be withdrawn. As with the other question types, however, not all leading
questions are bad. Indeed, they are a staple of good parenting as when a mother
encourages her toddler to consume more fruit by asking ‘Do you want a delicious
apple as a special treat?’ Nonetheless, all of these forms of questioning can be
dangerous and harmful when employed in certain contexts or with intellectually
vicious motivations.
It is also worth commenting on inept, inappropriate and misguided questions.
Inept questions are questions asked through an unsuitable medium. This could
mean asking a question in the wrong language, using the wrong technology or
employing the wrong means of communication. Inappropriate questions are
questions that seek out inappropriate information. By inappropriate I mean to
capture something quite broad: in general, information that it is in some sense not
appropriate for the questioner to seek to possess. This could mean asking for pri-
vate or sensitive information, asking for legally redacted information or asking for
information that is harmful in some way. What exactly makes information inap-
propriate is a more complex issue than I can address here, involving underlying
and unresolved questions concerning the nature of epistemic value (Schmitt and
Lahroodi 2008; Brady 2009; Haddock, Millar and Pritchard 2009) and the virtue
of epistemic temperance (Inman 2015). Nonetheless, if we can make sense of the
idea that there is some information that it is not appropriate for a person to seek to
possess, then we can make sense of inappropriate questions.
Misguided questions are questions that seek out irrelevant (as opposed to inap-
propriate) information. By irrelevant I, again, mean to capture something quite
broad: in general, information that is not relevant to the questioner’s epistemic
goals. This could mean, for example, asking what colour shirt the leader of a
political party is wearing in order to decide whether or not to vote for them or
asking what age a woman is in order to decide whether or not she is a suitable
candidate for a job. What exactly makes information irrelevant again involves
deeper questions concerning, for example, the nature of relevance (Schaffer 2001)
and the virtues and/or vices of open- and closed-mindedness (Riggs 2010; Battaly
2018). Nonetheless, if we can make sense of the idea that there is some informa-
tion that is not relevant to a person’s epistemic goals, then we can make sense of
misguided questions.
We now have some sense of the nature and range of the bad questioning prac-
tices represented by the taxonomy. Before proceeding it is also worth noting the
anomalous position of rhetorical questions. Rhetorical questions are anomalous
because they are not properly questions in the sense that they are not defined by an
information-eliciting function. Indeed, they are defined precisely by the absence
of this function. Rhetorical questions are ‘questions’ – or more accurately, inter-
rogative sentences – that do not seek to elicit information. They can nonetheless
be deployed in public discourse for a variety of unscrupulous reasons. Their ori-
gins in the ancient practice of political rhetoric speak to their significance as a
political speech act. Despite this and their interrogative form, however, rhetorical
questions cannot qualify as either good or bad questions qua questions, given that
they are not defined by the function of information-elicitation.1
13.3 Bad questioning in public discourse

What does bad questioning look like in public discourse and what impact does
it have? In this section, I present real-world examples of bad questioning in the
public sphere to illustrate the taxonomy and highlight the negative impact that
bad questioning has in contemporary political life. There is not space to cover all
of the question types listed in the taxonomy, nor would it be possible to provide a
truly representative sample from the many examples available in recent politics.
The examples have been selected to demonstrate the significance of questions
and questioning for the preservation or degradation of public discourse and of the
epistemic environments in which that discourse is conducted. I begin by examin-
ing referendum questions to demonstrate the impact of good and bad questions in
democratic politics and then move on to consider a recent case of bad questioning.
The latter will bring us to a more detailed discussion of the relationship between
bad questioning and intellectual vice in public discourse.
13.3.1 Scottish Independence referendum and Brexit

One of the most obvious ways to demonstrate the significance of good and
bad questions in political life is by examining the formation of referendum
248 Lani Watson
questions. Two recent examples in the UK bear witness to this: firstly, the Scottish
Independence referendum, held on 18 September 2014, and secondly, the UK
referendum on EU membership, held on 23 June 2016, better known as Brexit.
In both cases the referendum question that was initially proposed was reformu-
lated after consultation and question-testing by the UK Electoral Commission, an
independent body set up to regulate election finance and standards.2 The Electoral
Commission guidelines state:
A referendum question should present the options clearly, simply and neu-
trally. So it should: be easy to understand, be to the point, be unambiguous,
avoid encouraging voters to consider one response more favourably than
another, avoid misleading voters.3
Both the Scottish Independence and Brexit referendum questions were assessed
according to these guidelines. In the first case, the initial question proposed by the
Scottish National Party (SNP) read: ‘Do you agree that Scotland should be an inde-
pendent country?’ The Electoral Commission report cites public opinion research
and consultations with political party members in the Scottish Parliament, aca-
demics and lawyers, as well as groups such as Age Scotland, Dyslexia Scotland
and Outside the Box (a charity for people with learning disabilities). The report
concluded that the initial question should be reworded. Issues with the ques-
tion focused primarily on the use of the phrase ‘Do you agree…’. The Electoral
Commission stated:
In our view, while there is no evidence to suggest that ‘Do you agree…?’
is intended deliberately to encourage voters to consider one answer more
favourably than another, the responses we have received demonstrate that
‘Do you agree…?’ can be seen by people as encouraging such a response. 4
As a result, the Scottish referendum question was changed to ‘Should Scotland

be an independent country?’ This question was judged to be more neutral and so
less likely to ‘encourage voters to consider one response more favourably than
another’, in line with the question guidelines.
The same process was carried out by the Electoral Commission with respect to
the Brexit referendum question. During this process similar concerns were raised
about the neutrality of the initial question, ‘Should the United Kingdom remain a
member of the European Union?’ Specifically, some participants in the consulta-
tion process suggested that ‘only specifying the “remain” option in the question
could influence voters’.5 While it was acknowledged in the report that the effects
of this were probably negligible, the wording of the question was nonetheless
changed to reflect this concern. The final referendum question read, ‘Should the
United Kingdom remain a member of the European Union or leave the European
Union?’
One might argue that these small changes in question-wording are insignifi-
cant. This argument was made by political science professor Matt Qvortrop in
the report on the Scottish Independence question. Qvortrop provided statistical
analyses of 74 referendums on independence or self-government from 1980 to
2011. On this basis he concluded that there was no firm evidence that the initial
Scottish referendum question containing ‘Do you agree…’ would create bias. In
a 2013 article for the BBC, Qvortrop is quoted as saying, ‘The overall conclusion
one can draw if one looks around the world, is that the question itself extremely
rarely has an impact on the outcome of the referendum’.6 This conclusion is based
on the idea that, at least when it comes to major referendums, people have typi-
cally made up their minds in advance of entering the polling booth and minor
differences in the wording of the question are unlikely to influence them at that
stage. This seems right, but it is worth noting that the wording of a referendum
question has an influence beyond voting decisions in the polling booth. In both the
cases discussed, the major campaigns on each side of the debate were based on the
questions as they were stated, thus, the ‘Yes’ and ‘No’ campaigns in the Scottish
Independence referendum and the ‘Vote Leave’ and ‘Vote Remain’ campaigns in
the case of Brexit. It seems plausible that ensuring the neutrality of the questions
themselves avoids giving a rhetorical edge to one side or the other in the public
debate.
Perhaps more importantly, it is worth reiterating that these relatively minor
changes were made to questions that already met most of the Electoral Commission
standards unequivocally. The question assessments conducted by the commission
did not turn bad referendum questions into good ones, they turned good referen-
dum questions into better ones. This is a positive outcome in itself. The signifi-
cance of good referendum questions is, however, highlighted clearly when one
considers the impact of a more noticeably bad one.
13.3.2 New Zealand corporal punishment referendum

The New Zealand corporal punishment referendum, held from 31 July to
21 August 2009, was a citizens-initiated referendum concerning the legal status of
parental corporal punishment. In it the people of New Zealand were asked to vote
on whether parental corporal punishment should be a criminal offence. The ref-
erendum was initiated in response to the passing of the so-called ‘anti-smacking’
law in 2007, which removed parental correction as a defence for assault against
children. Campaign groups seeking to re-establish this defence gained the req-
uisite public support for a referendum on the issue in which voters were asked,
‘Should a smack as part of good parental correction be a criminal offence in New
Zealand?’ Voter turnout was 56.1% and 87.4% of voters answered ‘No’. This
result suggests that a majority favoured reinstating the parental correction defence
and, in effect, overturning the anti-smacking law.
The question is clearly both loaded and leading. It contains a contentious pre-
supposition (that a smack can be part of good parental correction) and is biased
in favour of a particular answer (voting ‘Yes’ suggests that one believes good
parental correction should be a criminal offence). As such, this is a bad question.
Notice that it is not merely bad on moral or political grounds. It bad because it
250 Lani Watson
prevents the elicitation of worthwhile information. Indeed, this is largely what
makes it bad on moral and political grounds. In the case of a referendum, the
worthwhile information being sought is information about the views of the pub-
lic on a particular issue. If the collection of that information is corrupted by bias
in the process – in this case by a loaded and leading question – then the informa-
tion itself will be corrupted. Access to a true representation of public opinion is
denied and, as such, information about public opinion is prevented from coming
to light.
The question was heavily criticised. The then-Prime Minister, John Key, was
quoted as having commented that ‘the question is a bit ambiguous and could be
read a number of different ways’.7 The Labour Party Leader, Phil Goff, likewise
argued, ‘the question implies that if you vote “yes” that you’re in favour of crimi-
nal sanctions being taken against reasonable parents – actually nobody believes
that’.8 Crucially, no change to the law was made following the referendum despite
what would appear to be public support in favour of overturning the anti-smack-
ing law. Whatever one thinks about the issue of parental corporal punishment, this
fact is worth pausing to consider; 87.4% of voters ostensibly voted in favour of
a change in the law. Yet the result was almost entirely dismissed by the govern-
ment, prompting anger and public protest in the weeks following the referendum.9
Given the government’s clear and warranted criticism of the question, it is, I think,
plausible that the question wording played an important role in the government’s
willingness and ability to dismiss the referendum result on the basis that it did not
provide a true representation of public opinion. Had the question been less biased,
and the result the same, this dismissal would have been less easily justified and a
more legitimate cause for public outcry. At the very least, it seems clear that poor
question formation led to the perception of a dubious and unreliable referendum
result in this case and plausibly to a false representation of public opinion. Bad
questions like this can have significant political ramifications, particularly in the
context of a referendum where the question plays an important role in represent-
ing the issue at stake and is often at the heart of the public debate. This provides a
clear example of the impact of bad questions in the public sphere.
13.3.3 Jeremy Paxman 2017 pre-election interviews

Referendum questions provide a useful illustration of the significance of good and
bad questions in democratic politics. Moving beyond the role of individual ques-
tions, however, a different public setting in which questioning plays an important
role is in the media. Journalists, news anchors and broadcasters, for example, are
regularly tasked with asking important, timely or probing questions in order to
uncover truths about issues of societal import for the general public. They are, to
some extent, professional questioners. As such, by examining journalistic ques-
tioning we can observe the significance, not only of bad individual questions but
of bad questioning strategies. The following examples are taken from televised
interviews conducted prior to the 2017 UK general election by the well-known
British broadcaster, Jeremy Paxman.
Paxman is renowned for his ‘tough-talking’ reputation, particularly deployed
in interviews with politicians, and is associated with a distinctive questioning
strategy involving sustained repetition of a single question. This strategy is
intended to expose evasive interlocutors if they fail to answer and is, in itself,
neither good nor bad. Paxman’s deployment of the strategy won him acclaim in
his early career, perhaps most notably in a 1997 interview with the then-Home
Secretary, Michael Howard, who was asked the same question 12 times by
Paxman in a televised interview for Newsnight, which he declined to answer
directly. Howard’s failure to answer the question directly was taken by many
as an admission of guilt or wrongdoing on his part. Paxman received general
public approval for the strategy and continued to build his journalistic reputa-
tion on this basis. The same approach 20 years later, however, attracted criti-
cism from those watching Paxman’s live, televised interviews with the leaders
of the two major British political parties, Jeremy Corbyn (Labour) and Theresa
May (Conservative), 11 days prior to the 2017 UK general election. An excerpt
from the first few minutes of the Corbyn interview demonstrates the strategy
well:
Paxman: ‘You promise in this [Labour Party Manifesto] to renew Trident [UK
Nuclear Programme]’
Corbyn: ‘It was a conference decision by the Labour Party and as the leader of the
party I accept the democracy of our party and in answer –’
Paxman: [speaking over] ‘Is that morally right?’
Corbyn: ‘– in answer to the questions put earlier I made the point that as Prime
Minister I will do all I can to bring about a nuclear free world –’
Paxman: ‘Sure’
Corbyn: ‘– because I’m horrified at the very idea … at the very … horrified at the
very idea of a nuclear attack anywhere’
Paxman: [speaking over] ‘Here you promise to renew … you promise to renew …
you promise to renew a nuclear weapon’
Corbyn: ‘It’s there in the –’
Paxman: [speaking over] ‘Is that morally right?’
Corbyn: ‘Listen, it’s there in our manifesto because our conference voted for it. I
have to accept that decision –’
Paxman: [speaking over] ‘Do you think that it’s morally right?’
Corbyn: Wait a minute, wait a minute, can I finish? Can I finish? What I want to
see … what I want to see –’
Paxman: [speaking over] ‘I’m asking you perfectly simply, do you think it’s mor-
ally right?’
Corbyn: ‘– what I want to see is a nuclear free world. That means –
Paxman: [speaking over] ‘Of course, everybody wants to see that.’
Corbyn: ‘Well, I’m not so sure about that … I’m not so sure … I’m not so sure
about that –’
Paxman: [speaking over] ‘But is it morally right if that’s what you … if that’s
what you believe, is it morally right to renew a nuclear deterrent?’
252 Lani Watson
Corbyn: ‘That is the decision that’s been taken. We will work for a nuclear free
world. We will work through the nuclear non-proliferation treaty to achieve
that. That surely is something well worth doing –’
Paxman: [speaking over] ‘I note you don’t answer.’
Paxman repeats the question ‘is that morally right?’ – or a version of it – six times
during this short exchange. It is a closed question so the only direct answers pos-
sible are yes and no. At the end of the exchange Paxman accuses Corbyn of not
answering the question despite the fact that Corbyn says, or at least attempts to
say, a number of things in response. It is tempting to conclude that Paxman judges
Corbyn to have not answered the question because he does not answer it with a
yes or no. Technically, perhaps. But is that indicative of an evasive approach to
the question on Corbyn’s part or of a faulty questioning strategy on Paxman’s
part?
I think in this instance it is the latter. We can, for example, ask what would
satisfy Paxman that the question had been answered. The most complete ver-
sion of the question (which is given at the end of the exchange) is this: ‘if that’s
what you believe, is it morally right to renew a nuclear deterrent?’ Imagine that,
instead of trying to explain his position, Corbyn had answered this question with
a simple yes or a no. Would Paxman be satisfied? The question is closed but it is
nonetheless a complicated question. It is a question about the moral permissibility
and indeed ‘rightness’ of the decision to include a statement in a political party
manifesto that has been democratically agreed upon by the party members but
with which the leader of the party – to whom the question is addressed – person-
ally disagrees. It is deserving of a more detailed answer than a simple yes or no,
something which Paxman surely recognises. His decision to ask a closed rather
than open question here looks to be a trap; the question warrants a more detailed
response but in trying to offer it, Corbyn is accused of ‘not answering the ques-
tion’. This is a hallmark of Paxman’s aggressive, closed questioning strategy.
We see this questioning strategy in operation throughout the interviews with
Corbyn and May. A second excerpt is worth examining, taken from the first few
minutes of the May interview:
Paxman: ‘Hang on a second … you said in March last year that we would be more
secure, more prosperous and more influential, virtually in those words, if we
stayed in the European Union and now you want to take us out of it’
May: ‘And I also said that the sky wouldn’t fall in if we left the European Union’
Paxman: ‘So you have changed your mind have you?’
May: ‘We gave, we gave … we gave people the choice –’
Paxman: [speaking over] ‘You’ve changed your mind?’
May: ‘– we … I’ll answer that in a minute … we gave people the choice, Jeremy,
and the British people decided to leave the European Union –’
Paxman: ‘Yes’
May: ‘– and I think it’s important for them to see their politicians delivering on
that choice and respecting the will of the people. And what I think’s impor –’
[Applause from audience]
Paxman: ‘So you’ve changed your mind?’
May: ‘What I am now doing is delivering –’
Paxman: [speaking over] ‘Have you changed your mind?’
May: ‘– I think there are huge opportunities –’
Paxman: [speaking over] ‘Have you changed your mind?’
May: ‘Jeremy, I know that you have, err, that you use this tactic and you want me
to –’
Paxman: [speaking over] ‘I’m just trying to get an answer, that’s all. You can
say yes or no, I haven’t changed my mind, yes, I have changed my mind, say
what you like!’
Paxman asks May five times during this short exchange whether she has changed
her mind on the issue of Brexit. Here the question is rather simpler than the
question posed to Corbyn, and one might think that a simple yes or no would
be an adequate and informative answer. May’s refusal to answer with a yes or
no, therefore, arguably indicates a degree of evasiveness on her part. It seems
fair, however, to say that the viewing public is most probably interested in more
than merely knowing whether or not May herself has had a change of heart. Her
answer is plausibly an attempt to respond to a more substantive question which
holds greater interest for the public and, indeed, she is cut off at one point by
spontaneous audience applause.
At any rate, perhaps the most interesting aspect of this exchange for present
purposes concerns the point at which May attempts to challenge Paxman’s ques-
tioning strategy which she describes as a ‘tactic’. Paxman responds by animatedly
declaring that he is ‘just trying to get an answer’ and reveals, at the same time,
that the only answers he is interested in, or would satisfy the question, are yes and
no. Yet he directs May to ‘say what you like’. Given that the only answers he will
accept are yes and no this seems insincere at best and, moreover, is hard to align
with the fact that he speaks over her relentlessly when this is precisely what she is
trying to do. This moment of exposure again suggests that Paxman’s aggressive,
closed questioning strategy is deployed as a dialogical trap.
Importantly, this questioning strategy is a bad one not because it traps the
respondent but because in doing so it prevents or impedes the elicitation of worth-
while information. Remember that these were national television interviews,
conducted live, 11 days before the 2017 UK general election, with the leaders
of the country’s two main political parties. They were the only such interviews
conducted prior to the election and, as such, provided a significant opportunity
for voters to learn about Corbyn and May’s political beliefs, positions and poli-
cies, in an open, accessible manner. Whatever one thinks about the role that this
information plays in the voting decisions of the viewing public, the purpose of the
interviews was, at least in part, to help the electorate to make an informed decision
about who should govern their country.
They are interviews, not spoken political statements, or written manifestos,
or any other form of political communication and so are conducted in the form
254 Lani Watson
of question and answer exchange. The questions are thereby essential to the
process and to the goal of helping the public to make an informed decision on
polling day. The aggressive, closed questioning strategy deployed by Paxman
throughout the interviews proves an impediment to this goal. His pursuit of
yes or no answers to complicated, moral questions leaves more substantive
responses under-examined or, worse, unsaid. His use of dialogical traps, in
place of a genuine pursuit of worthwhile information, renders that informa-
tion difficult or impossible to access. Moreover, his aggressive manner and, at
times, slippery questioning are sources of frustration for the viewing public,
as much as they are for Corbyn and May. Indeed, live-tweeting throughout the
interviews revealed something of this. A small sample of tweets indicates the
flavour of this response:
‘We’ll probably get nuked before Paxman lets someone finish a sentence.’
‘These debates had a lot of potential. Unfortunately Paxman completely
ruined it. I think he tried to be difficult but he just came across as rude.’
‘Any chance Paxman would shut up long enough to let Jeremy answer a
question.’
‘I’d sack him. Can’t let Corbyn answer. He was a terrible interviewer and
he’s biased.’
‘You should be ashamed of Paxman [@Channel4News]. His questioning and
approach is completely unprofessional and unhelpful to the nation when try-
ing to decide.’
Journalist Hannah Jane Parkinson summarised this public response in a Guardian

article after the interviews commenting that Paxman’s approach was ‘disrespect-
ful to audience members and viewers who actually wanted to learn something and
see the interviewees properly challenged’.10 The approach is predominantly one
of bad questioning.
A final question from these interviews is worth noting in order to highlight
Paxman’s particularly provocative questioning style. Paxman opens the interview
with May with the following question: ‘Teresa May, when did you realise that
you’d got the wrong answer to the biggest question of our times in politics?’ This
is not only the first question of the interview but the very first thing that is said.
This question provides a clear example of a complex, loaded question. The logi-
cian Douglas Walton (1999) offers a helpful summary of this question type:
‘When did you stop cheating on your income tax returns?’ The question is
a when-question, so, in order to give a direct answer, the respondent has
to indicate some particular time like, for example, December 2nd, 1976.
However, in this case, if the respondent does give such a specific time as
answer, then it is clear that he has become committed to having cheated on
his income tax returns and, presumably, this is a proposition which generally
he would not be want to concede, or at any rate, would be prejudicial, or not
in his interest to concede.
(Walton, 1999, p. 379)
It is easy to see the similarities between the question Walton uses as a paradig-
matic example of a complex, loaded question, and the one asked by Paxman at
the start of the May interview. It is a ‘when’ question with a set of contentious
presuppositions: (1) that the question to which Paxman is referring is the ‘big-
gest question of our times in politics’ (whatever that means), (2) that May got the
‘wrong answer’ to that question, and (3) that she has since realised it. It is fair to
say that it would not have been in May’s interest to concede any one of these. Just
imagine her answering the question with a date!
There is something almost textbook about Paxman’s use of this complex,
loaded question to open the interview. May’s response is a baffled pause, fol-
lowed by an attempt to clarify which question Paxman is referring to as ‘the big-
gest of our times’, followed by a sort of dialogical stumbling into a defence of her
actions regarding Brexit (which leads into the excerpt quoted above). It is hard
to view this as an informative start to the interview. Perhaps Paxman succeeds in
making May look foolish, but the joke is ultimately on the viewing public who get
little to no epistemic benefit.
The Paxman pre-election interviews are, of course, just one example of bad
questioning in the public sphere – albeit an example taken from a noteworthy
moment in contemporary British politics. The example is intended to illustrate
the broader claim of the chapter, namely, that bad questioning strategies play
a significant and potentially damaging role in public discourse. They determine
what information is made available and what is left unsaid, as well as influencing
how important a topic is taken to be and how deeply it is examined. All of this
ultimately plays a role in determining the extent to which the reading, viewing and
voting public are informed on issues that matter. Moreover, bad questioning, such
as that employed by Paxman, also serves as an expression of intellectual vice in
the public sphere. I have described Paxman’s strategy as one of aggressive, closed
questioning, revealing the vice-language already in play in cases such as this. We
can now examine this relationship in more detail.
13.4 Bad questioning and intellectual vice

Bad questioning is a feature of intellectually vicious behaviour. In order to expli-
cate this it is useful once again to return to good questioning and examine its
relationship to intellectual virtue. I have argued elsewhere that good questioning
is an intellectual skill involved in the exercise of many, if not all, of the intel-
lectual virtues (Watson 2018). What does this mean? Following other authors, I
take it that part of what it is to be intellectually virtuous is to possess and exhibit
certain intellectual skills (e.g. Zagzebski 1996; Roberts and Wood 2007). Thus,
intellectual skills partly constitute intellectual virtues; one cannot be intellectually
virtuous without the exercise of intellectual skill. The skill of good questioning is,
256 Lani Watson
I argue, at least sometimes the intellectual skill which partly constitutes the intel-
lectual virtues of, for example, attentiveness, intellectual autonomy, intellectual
courage and intellectual humility (Watson 2018). In addition, I argue that the
skill of good questioning is necessary for the exercise of virtuous inquisitiveness
(Watson 2015). In other words, good questioning is one way of expressing the
intellectual virtues.
In much the same way, the intellectual failing of bad questioning is involved
in the expression of many, if not all, of the intellectual vices; it at least sometimes
partly constitutes intellectual vices such as negligence, arrogance, prejudice,
closed-mindedness and dogmatism. Bad questioning is one way of expressing
these intellectual vices. That is not to say that bad questioning is necessary for
intellectual vice. A person can be closed-minded or negligent, for example, with-
out asking any questions. Nor is bad questioning typically sufficient for intel-
lectual vice. Typing ‘where are my keys’ into Google is bad questioning but it is
not intellectually vicious. Nonetheless, in cases such as the Paxman pre-election
interviews, bad questioning serves as an expression of intellectual vice. Paxman’s
aggressive, closed questioning reveals, for example, a negligent disregard for the
pursuit of worthwhile information throughout the interviews. Moreover, his reli-
ance on this strategy and unwillingness or inability to change course despite live,
negative feedback is indicative of a degree of arrogance and/or closed-mindedness
on his part resulting, as journalist Hannah Jane Parkinson put it, in ‘disrespect’
for the viewing public. Paxman exhibits intellectual vice through bad questioning.
This is one (among many) ways of expressing intellectual vice.
Notably, Paxman’s reputation as a respected British broadcaster suffered after
the interviews, as seen in the Twitter response and in subsequent discussion of the
interviews by fellow journalists. More important than the effect that this may or
may not have had on Paxman, the presence of intellectual vice in the interviews,
expressed through bad questioning, diminished their value to the viewing and vot-
ing public. Examples such as this degrade the professional character of journalism
more generally and of mainstream media outlets such as the BBC. These outlets
represent a central cog in the wheel of democracy, and their degradation in the
eyes of the public is a cause for concern. As such, the presence of intellectual vice
in the mainstream media and in the wider public sphere raises difficult questions
about, for example, the extent to which journalists and politicians can be trusted,
the reliability and impartiality of experts and news sources and the accountability
of public figures with respect to privileged knowledge and information. To the
extent that bad questions and bad questioning strategies serve as an expression
of intellectual vice within these social institutions they should be identified and
alleviated in order to help check and maintain the epistemic and characterological
integrity of key social institutions.
13.5 Concluding thoughts

I have offered an analysis of bad questioning (Section 13.1), provided a taxon-
omy of bad question types (Section 13.2) and illustrated the harms that can and
do arise from bad questioning as an expression of intellectual vice in the public
sphere (Sections 13.3 and 13.4). The study of bad questioning is nonetheless in
its infancy, and the present chapter is intended to prompt more questions than it
answers. We should ask, for example, why bad questioning prevails in public
discourse and under what conditions it flourishes. Paxman was, to some extent,
exposed for bad questioning in the pre-election interviews, but it is worth remem-
bering that his reputation was largely built on the strategy that he deployed in
these interviews, and it seems unlikely that this is the first and only time that it has
warranted criticism. It is only by paying attention to the questions, as well as the
answers, that bad questioning such as this can be challenged.
Furthermore, we should ask what, if anything, can be done to guard against
bad questioning, particularly when it expresses intellectual vice in public life.
One interesting prospect may lie in examining the regulation of questioning in
legal settings such as a courtroom. Here the dangers of bad questioning are, at
least to some extent, recognised and legislated for, as noted in Section 13.2.
Lawyers are trained and conduct themselves on this basis. No such recognition or
regulation of questioning exists in, for example, the mainstream media or politics
and yet the dangers are equally live. Are there mechanisms by which journalists
conducting interviews or politicians engaging in public debate could be held to
the same professional questioning standards as lawyers in a courtroom? If so,
how could such standards be implemented and enforced? These are questions
that can and should be tackled. By placing a spotlight on questioning practices
in public discourse I hope to have drawn attention to a common but often under-
acknowledged aspect of intellectual conduct and highlighted its relationship to
intellectual vice.
Acknowledgements
Many thanks to Nathan Sheff and Ian James Kidd for detailed comments on drafts
of this chapter and to the audience of the Vice Epistemology Conference at the
University of Connecticut, 2019. This work was supported by a grant from the
Leverhulme Trust [grant number R44476].
Notes
1 The nature, benefits and harms of rhetorical questions will be explored in more detail
in future work.
2 Electoral Commission referendum question-testing for the Scottish Independence and
UK EU membership referendums: www.electoralcommission.org.uk/our-work/our-r
esearch/referendum-question-testing [Accessed: 22 March 2019].
3 Electoral Commission Referendum Question Assessment Guidelines: www.electoralco
mmission.org.uk/__data/assets/pdf_file/0006/82626/Referendum-Question-guidelines
-final.pdf [Accessed: 22 March 2019].
4 Referendum on independence for Scotland: Advice of the Electoral Commission on the
proposed Referendum question (pp. 21–22): www.electoralcommission.org.uk/__data
/assets/pdf_file/0007/153691/Referendum-on-independence-for-Scotland-our-advice
-on-referendum-question.pdf [Accessed: 22 March 2019].
258 Lani Watson
5 Referendum on membership of the European Union: Question testing (p. 9): www.e
lectoralcommission.org.uk/__data/assets/pdf_file/0006/192093/GfK-Report-EU-Refer
endum-Question-Testing-2015-WEB.pdf [Accessed: 22 March 2019].
6 www.bbc.com/news/uk-scotland-20512743 [Accessed: 22 March 2019].
7 ‘Key, Goff won’t vote on smacking referendum’: www.nzherald.co.nz/nz/news/article.
cfm?c_id=1&objectid=10578819 [Accessed: 22 March 2019].
8 ‘Key, Goff won’t vote on smacking referendum’: www.nzherald.co.nz/nz/news/article.
cfm?c_id=1&objectid=10578819 [Accessed: 22 March 2019].
9 ‘One arrest as thousands join “March for Democracy”’: www.nzherald.co.nz/nz/news/
article.cfm?c_id=1&objectid=10610750 [Accessed: 22 March 2019].
10 ‘When did Paxman go from supreme interviewer to shouty interrupter?’: www.thegua
rdian.com/commentisfree/2017/may/30/jeremy-paxman-politician-theresa-may-je
remy-corbyn.
References
Battaly, Heather. 2018. “Closed-Mindedness and Dogmatism”, Episteme 15(3): 261–282.
Brady, Michael. 2009. “Curiosity and the Value of Truth”, in Adrian Haddock, Alan
Millar and Duncan Pritchard (eds.), Epistemic Value, Oxford: Oxford University Press,
265–283.
Cassam, Quassim. 2016. “Vice Epistemology”, The Monist 99(3): 159–180.
Dillon, J. T. 1988. Questioning and Teaching: A Manual of Practice. Oregon: Resource
Publications.
Gershon, Mike. 2013. How to Use Questioning in the Classroom: The Complete Guide.
Published by the author.
Haddock, Adrian, Alan Millar, and Duncan Pritchard (eds.). 2009. Epistemic Value.
Inman, Ross D. 2015. “Epistemic Temperance and the Moral Perils of Intellectual Inquiry”,
Philosophia Christi 17(2): 457–472.
Riggs, Wayne. 2010. “Open-Mindedness”, in Heather Battaly (ed.), Virtue and Vice:
Moral and Epistemic, Oxford: Wiley-Blackwell, 173–188.
Roberts, Robert C. and Wood, W. J. 2007. Intellectual Virtues: An Essay in Regulative
Epistemology. Oxford: Oxford University Press.
Schaffer, Jonathan. 2001. “Knowledge, Relevant Alternatives and Missed Clues”, Analysis
61(271): 202–208.
Schmitt, Frederick F. and Reza Lahroodi. 2008. “The Epistemic Value of Curiosity”,
Educational Theory 58(2): 125–148.
Walton, Douglas. 1999. “The Fallacy of Many Questions: On the Notions of Complexity,
Loadedness and Unfair Entrapment in Interrogative Theory”, Argumentation 13:
379–383.
Watson, Lani. 2015. “What is Inquisitiveness?”, American Philosophical Quarterly 52(3):
273–288.
Watson, Lani. 2018. “Educating for Good Questioning: A Tool for Intellectual Virtues
Education”, Acta Analytica 33(3): 353–370.
Worley, Peter. 2019. Questioning: 100 Ideas for Primary Teachers. London: Bloomsbury.
Zagzebski, Linda. 1996. Virtues of the Mind: An Inquiry into the Nature of Virtue and the
Index
Page numbers in italic indicate figures.
active ignorance 53–56, 60, 62, 64, 65n22 Bias of Crowds model 135–136, 139–140
activism 5, 13, 80, 108; epistemic 10, 108, blame 43, 56, 62, 79–80, 100, 105n8,
120–123, 142n10 144n36, 194, 203–204
aetiological sensitivity 78–80 blameworthy psychologies 129–130
A Far Cry from Kensington 208 Bloomfield, Paul 7, 227
affective dimension 22–23, 27–28, 34n6, 60 Brexit 12, 137, 247–249, 253, 255, 257n2
akrasia 11, 178–180, 186n21 British Broadcasting Corporation (BBC) 9,
Alcibiades 3 90, 102–104, 249, 256
Alfano, Mark 6, 9–10, 47, 62, 131, 148, Buddhism 3, 14n2
153–154, 159 Byerly, Meghan 138–140, 143n30
amelioration 1, 7, 10, 12–13, 49n19, Byerly, Ryan 138–140, 143n30
70–71, 77, 79–80, 121–122, 141
Andreou, Chrisoula 175 capital institutional vices 10
anti-social epistemology 6, 13 Card, Claudia 69
applied vice epistemology 6, 11 carelessness 25, 29, 31, 200, 242
Aristotle 3, 8, 11, 21, 23–24, 28, 33n3, 39, Cassam, Quassim 5, 7–8, 70, 78–81, 89,
179, 186n21, 211, 227, 236 105n1, 105n8, 110–111, 124n2, 126,
Astell, Mary 4, 69 128–131, 215, 242
attention 3, 5, 7, 13, 23, 26–27, 31, 33, 42, character: critical epistemology 9, 13, 71,
58–61, 63, 65n18, 65n20, 65n21, 70, 75, 77–78, 80–83; critical theory 13,
89–90, 101, 108–110, 114, 116–117, 77–78, 81–82; epistemic 1–3, 5–6, 9,
121, 141, 148, 156, 195, 197, 200, 203, 11, 13, 55, 64, 65n21, 69–79, 81–83,
216, 239, 257 100, 112; epistemologies of education
aversive behaviour 61 79; intellectual 21, 24, 47, 93, 168–169,
axiological pluralism 78, 80–81, 83 172, 174, 180, 183–184, 216; moral
Axtell, Guy 92 69, 77; professional 239, 256; trait 3,
5, 8, 11–12, 22, 37, 39–40, 47, 57, 62,
Bacon, Francis 4 65n15, 66n27, 79, 81–83, 128, 132–133,
Baehr, Jason 5–8, 49n16, 99, 157, 171, 168–169, 171–172, 179–180, 190, 192,
209–210, 213, 228 202, 213, 228–229; vices 47, 55, 64,
Baier, Annette 72 65n21, 131, 183; virtue 5, 174, 180, 184
Battaly, Heather 5, 7–8, 11, 26, 40–41, Charmides 3
77–79, 82, 92, 98, 129, 143n25, 148, climate scientists 8
168, 172, 213, 220, 225, 236, 247 Code, Lorraine 2, 5
belief-security 148, 151–152, 155, cognitive contact with reality 81, 93,
159–161 99–100
Believing Inmates (policy) 111, 115–116 collective intentionality 94
260 Index
Condorcet Jury Theorem 156 235; corruptee 71; corruptors 71–73;
Confucius see Kǒngzı intensification 72; passive corruption
consequentialism 80, 111–112, 114, 189 71, 235; propagation 72, 237n2;
conspiracies 32, 47–48, 50n37, 97, ‘rebranding’ 76; stabilisation 72
159–160, 196–197, 205n6, 205n7 epistemic vices: alethic vices 74, 77;
continence 23, 28 apathy 11, 168, 176, 182, 184–185;
Cooper, David E. 77 appetitive vices 74, 77; arrogance
Cordell, Sean 140 1–4, 6–9, 11, 13, 14n3, 21, 31–32,
Craig, Edward 42, 44, 50n24 40, 43–44, 47–48, 53, 55, 57, 62–64,
Crerar, Charlie 7, 11, 31–32, 99–100, 66n27, 69–70, 72, 74, 76, 89, 105n10,
173, 215 110, 114–115, 135, 199–200, 235, 239,
critical race theory 5 242, 256; capital 10, 109, 111–116,
Croce, Michel 75, 236 118–120, 122–123, 124n4, 124n6;
culpability 62, 100, 124n5 classification of 12; collective 1, 4, 10,
curiosity 21–22, 31, 89, 110, 114, 124n6, 13, 103, 105n3, 126, 133, 135–137,
185, 228–229, 233–236 140–141; complacency 11, 57, 64n5,
Cynics 1, 4, 76, 91, 220 66n26, 168, 184–185, 186n31, 186n32,
187n32; dismissiveness 33, 196–197;
Daukas, Nancy 13, 75, 81–82 disorganization 137, 140; dogmatism 1–4,
DeYoung, Rebecca 4, 13 6, 8, 14n3, 21, 33, 37–38, 41, 43, 46,
diligence 89, 228 49n14, 57, 72–74, 76, 142n4, 156–158,
Dillon, Robin 12, 69, 71, 75–82, 84 235, 242, 256; epistemic insensibility 1,
distrust 115, 117–119, 158–159, 205n7 5, 8, 77, 220; epistemic insensitivity 11,
Dotson, Kristie 81, 124 111, 189–190, 192–196, 198–202, 204;
doubt-mongering 54 epistemic intemperance 7; epistemic
Douglass, Fredrick 157, 160 malevolence 7, 25, 29, 31, 34n9, 82, 99,
Driver, Julia 49n22, 130, 205n3 157; epistemic self-indulgence 1, 5, 7;
Du Bois, W.E.B. 83 esoteric epistemic vices 49n7; excessive
credulity 110; folly 11, 168, 184–185;
Eddo-Lodge, Reni 53 ‘four-dimensional’ account of their
education 4, 8, 11–13, 63, 77, 79, 173, structure 22; gullibility 1, 7, 21, 26, 29,
193, 208–212, 215–216, 218–219, 225, 33, 110–111, 115, 129; haughtiness 7;
227, 230–237, 244 hyper-autonomy 7; incredulity 10, 110,
enkrateia 23, 28 113–115, 118–119, 121, 123n1, 123n2,
environmental luck 151 124n2, 124n6, 124n10; incuriosity 31,
epistemic: bubbles 6, 8; confidence 74, 110, 114, 124n6; indifference 31, 69,
115; environment 11, 26, 73–75, 77, 81, 186n29; inferential inertia 9–10,
148, 151, 156–158, 173, 180, 184, 189, 101–102, 104, 105n13; intellectual
231, 247; exemplars 73, 75, 236–237; arrogance 7, 9, 13, 21, 31, 55, 57,
functioning 1, 10, 92, 111–118; labour 62, 64; intellectual laziness 21,
54; neglect 108–109, 113–116, 118, 28–31; intellectual recklessness 21, 26;
120–121, 124n8, 124n10; phronesis 12, intellectual snobbery 11, 208–215, 219,
225, 227–230, 232, 235–237; phronimos 221n2; intellectual timidity 7; petty
12, 236; relationality 113–116, 118– bureaucracy 137, 139; prejudice 32,
119; resistance 13, 112, 114, 120–123; 39–40, 61, 90, 97, 101, 105n8, 105n13,
safety 11; security 11, 151–152; 117, 119, 137–140, 143n28, 212, 242,
self-corruption 71; sensibilities 70, 256; procrastinating 11, 167–168,
216; situationism 47; well-being 10, 174–178, 180–181, 184–185, 185n1,
152–153, 157–158, 160 186n19, 186n23; quitting 11, 167–168,
epistemic corruption 9, 12–13, 71–75, 171–174, 181, 183–185, 185n1, 186n9,
77–79, 81, 83, 112, 116, 119–120, 122, 186n13, 186n23; racial insensitivity 9,
235; acquisition 60, 72, 78, 129, 189, 53, 57, 60–62, 64, 66n26; resignation
191, 198, 200, 203–204; activation 72, 11, 168, 184–185, 186n31; servility
131–132, 142n13; active corruption 71, 7, 70, 74; slacking off 11, 167–168,
Index 261
181–185, 186n23, 186n24, 186n25, Goldberg, Sandy 150, 161n3
187n32; stealthy 7, 65n11; testimonial Gorgias 3
injustice 1, 5, 58, 81, 101, 105n13, Greco, John 148
115, 119–120, 123; thinking-vices 99, group agency 94, 135
105n9, 110, 131; vanity 82–83; venial
epistemic vices 109, 111–112, 114, 119; Haslanger, Sally 134
vice of oversensitivity 190; ‘vices of the hermeneutical justice 89
privileged’ 48, 74; vicious personality ‘hidden hand’ mechanisms 10
traits 110; vicious sensibilities 9, 53, 57, high fidelity vices 82, 131
59, 61, 65n21; wishful thinking 1, 39– Holroyd, Jules 7, 9–10, 61, 126, 129–130,
40, 54, 56; see also capital institutional 142n15
vices; insensitivity Hookway, Christopher 58
epistemic virtues: adjusting 158; Hursthouse, Rosalind 227
competence dimension 22, 26, 33,
35n21; curiosity 21–22, 89, 185, ideological explanations 48
228–229, 233–236; diligence 89, 228; ignorance 7, 9, 25, 53–64, 64n1, 64n4,
intellectual humility 1, 7, 13, 21–22, 65n21, 65n22, 66n24, 66n26, 79, 182,
185, 242, 256; judgment dimension 194, 200; loving 55, 57
22–23, 30; motivational dimension Implicit Association Test 130
22, 30–31, 33, 35n21; other-regarding implicit biases 10, 126–134, 136, 140–141,
191; perseverance 11, 167–169, 171, 142n6, 142n8, 142n11, 143n22, 144n37
174, 181, 183–185; restructuring 156, inquisitiveness 28, 229–230, 242, 256
159; sensitivity 4, 11, 58, 69–70, 77–79, insensitivity: expertise 11, 190, 192–193;
82, 149, 190, 195–196, 200–203, interest 190, 197–198; physiological
215, 217–218, 220, 227–228, 237n1; 190, 198; testimonial 10, 110–111, 122;
tenacity 76, 228; testimonial justice value 190, 194
89, 119 Inside/Outside Alliance (IOA) 109, 111,
ethical vices 2–3 116–117, 121–122, 124n12
‘ethico-epistemic’ traits and dispositions institutional epistemic failure 98, 110
81, 115–116 institutional racism 90; see also vice
ethos 9, 90–101, 103–104, 105n5 intellectual: actions 28, 168–170, 172–174,
Euthyphro 3 179–180, 184; evaluations 11, 210–212,
Every Student Succeeds Act (ESSA) 217, 220, 221n7; goals 167–179, 181,
233–234 183–185, 186n11, 202; merits 211–213,
exemplars 12, 14n2, 73, 75–76, 133, 159, 215, 219–220, 221n1; status 11, 209,
227, 236–237; of epistemic vice 75–76; 211–220, 221n8
of epistemic virtue 75, 98 interviews 91, 250–257
‘invisible hand’ mechanisms 137–141,
Facebook 154, 157 144n31
fact-checking 96–100
‘fake news’ 13, 159–160 jail 108–111, 113, 115, 117–122
feminist epistemology 4–5, 75, 83 Johnson, Casey 12, 79
foundational vice epistemology 21 ‘joint commitment’ 10, 93–98, 104,
free-riders 167 136–137, 139–141, 143n27, 144n35
Fricker, Miranda 5, 7, 9, 58, 70–71, 81, 89,
91, 105n4, 114, 119, 136–138, 143n23, Kant, Immanuel 152–153, 155
143n24, 143n27, 143n28, 144n35, 148, Kawall, Jason 148, 185
205n7, 216 Kidd, Ian James 4–5, 7–9, 13, 33n1, 49n5,
frustration 80, 128, 169, 254 49n19, 69, 75, 77–79, 83, 112, 124n4,
141, 144n36, 168, 173, 205n5, 225,
Galileo 31–32, 35n25, 100 235–236, 237n1, 237n2
gender bias 127–129, 134–135, 139 King, Nathan 168–169, 171
Gilbert, Margaret 93–97, 104, 105n7 Kivisto, Sari 13
Goguen, Stacey 7 Kǒngzı 3, 14n1
262 Index
Kornblith, Hilary 38, 49n3 networks: -security 148, 152, 154–156,
Korsgaard, Christine 153 159–161; social epistemic 149–151,
156, 158–159, 161n3; star-networks 10,
Lackey, Jennifer 150 153, 154
Lahroodi, Reza 7, 246 New Zealand corporal punishment
Langton, Rae 97 referendum 249
Lawrence, Stephen 90 Nguyen, C. Thi 8
Lewis, David 97 Nicomachean Ethics 3, 227
liberation 13 No Child Left Behind Act (NCLB) 12,
liberatory: epistemic virtue 13; vice 232–234
epistemologies 79 normative contextualism 78, 81–83, 168
List, Christian 94, 156 ‘numbness’ 57, 60
Locke, John 4, 45–46, 50n27, 50n29, 50n31 Nussbaum, Martha 234
low-fidelity vices 131
Lynch, Michael 8 Oblomov 31
obstructivism 40–41, 49n12, 81, 111–112,
MacIntosh, Duncan 175, 186n15 114, 116, 120, 128–131, 137, 139
MacIntyre, Alasdair 71 Orwell, George 174, 180, 184; Oceania
Macpherson Report 90 180–181, 184
Manson, Neil C. 13, 55
McCain, Matthew 109–110, 114–117, Pascal, Blaise 2
124n8 Paul, Herman 13
Medina, José 5, 7, 9–10, 40, 48, 54, 57, Paxman, Jeremy 250–257
60, 64, 66n24, 70–71, 73–74, 78–80, 89, personal worth 22, 24, 33n2, 35n19
110, 117, 120–122, 135, 165n2 personalist 78
#metoo 53 Pettigrove, Glen 227
Michelson, Albert 170–171, 173 Pettit, Philip 94
Miller, Christian 73 Plato 160
Mills, Charles 54–56, 79 plural subject 93–95
misinformation 54 Pohlhaus Jr., Gaile 5, 7, 81
modal epistemic standings 148–152, 160 post-truth politics 6
Montmarquet, James 5, 22, 32, 148 pragmatism 39, 46, 82
‘moral damage’ 69, 71 ‘predicaments of the oppressed’ 74
moral neglect 108 ‘predicaments of the privileged’ 74
moral vices see virtue ethics Priest, Maura 11
moral virtues see virtue ethics Pritchard, Duncan 148–150, 155, 246
Moriarty, Michael 76 Proctor, Robert 76
Morley, Edward 170–171, 173 propaganda 54, 159
Morrison, Toni 171
motivation 8, 24–25, 27, 29–33, 34n5, 34n7, questioning/questions: bad 12, 239,
34n8, 34n11, 35n16, 35n19, 58, 63, 65n9, 241–244, 245, 247, 249–250, 254–
70, 81, 111, 114, 120, 139, 156, 159, 257; complex 144n36, 244, 245, 246,
171–173, 179–181, 213, 220, 226–227; 254–255; compound 244, 245, 246;
proximate 213; ultimate 100, 213 good 236, 239–244, 247, 250, 255–
motivationalism 11, 21–23, 25–27, 29–33, 256; inappropriate 245, 246; leading
34n15, 35n20, 35n21, 35n23, 35n26, 239, 245, 246, 250; loaded 239, 245,
35n27, 57, 59, 65n7, 71–72, 75–77, 246, 250, 254–255; misguided 245,
81, 91–92, 98–100, 104, 105n6, 213, 246–247
215–217, 220
racism 90, 117; anti- 142n10; aversive
negative epistemic exemplar see exemplar 61–62; structural 53, 144n32; see also
of epistemic vice institutional
negligence 10, 116, 136, 138, 140, 142n4, reliabilism 40, 190–191, 205n2; see also
144n31, 176, 200, 239, 242, 256 virtue
Index 263
responsibilism 40, 129, 190–191, 205n2; Tuomela, Raimo 94–95
see also virtue Twitter 154, 157, 256
responsibility 2, 5, 26, 29, 33, 56, 62,
66n24, 78, 124n5, 136, 142n16, 144n33, UK referendum on EU membership see
144n36; retention- 78 Brexit
Riggs, Wayne 156, 247
Roberts, Robert C. 5, 7, 31, 82–83, 255 Vaccarezza, Maria Silvia 75, 236
Rochefoucauld, Duc de La 76 Valian, Virginia 127
Rooney, Phyllis 75 vector-relative safety 150, 161n3
Russell, Bertrand 169–171 vice attributions 39, 42–44, 47, 49n19,
142n6, 144n36
Scottish Independence referendum 12, vice ontology 8, 12; conceptualism 38–39,
247–249, 257n2 44–46, 50n31; conceptualist realism 46;
Sherman, Benjamin 7 realism 38–39, 44–46, 50n26; vice-
skill 12, 22–23, 25–28, 30, 34n14, 34n15, monism 8, 37, 39; vice-pluralism 8, 37,
35n10, 49n11, 58–60, 92, 138, 159, 39–40, 48n1
171–172, 190–191, 195, 199, 204, 211, vice reduction see amelioration
213, 226–231, 233–234, 237n3, 242, vice-charging 49n19, 140–141
255–256 vices: aggressiveness 12, 73, 75, 196,
Slote, Michael 135 199–200, 239, 244, 245, 252–256;
Smith, Dame Janet 102–104 contempt 69; cruelty 69, 82, 131;
snobbery 11, 32, 208–217, 219–220, dishonesty 28, 76, 199; dismissiveness
221n2, 221n6, 221n12; of motivations 33, 43, 128, 196–197, 219; of
11, 209, 215–216; of sensibilities 11, domination 69; greed 3, 198–199, 204;
209, 216, 219–220 institutional 9–10, 13, 90, 92, 101, 104,
social: institution 70, 73, 82, 239, 256; 105n4, 119; interpersonal 192, 200;
oppression 6, 69–71, 81; virtue manipulativeness 199; rudeness 199,
epistemology 13 254; selfishness 199; substantive 140;
Sosa, Ernest 5, 148–149, 229 uncharitableness 199–200
Spelman, Elizabeth 54, 56 vicious ways of thinking see epistemic
Stroud, Sarah 174–175 vices/thinking-vices
Sullivan, Emily 9–10, 155, 157, 159 virtue epistemology: autonomous 6, 10,
summative 93–94 27, 54, 160, 201; conservative 6, 230,
234; foundational 5–6, 8, 21
Tanesini, Alessandra 7, 9, 32–33, 35n24, virtues: ethics 6, 58, 229; humility 55,
35n29, 54, 59, 62, 70, 79, 212–213, 216, 70, 89, 228–229, 233–234; intellectual
221n8 autonomy 21, 27, 155, 160, 256;
Tappolet, Christine 175 intellectual courage 21, 209–210,
Taylor, Gabriele 43, 71 229, 242, 256; intellectual humility
‘teaching to the test’ 232–234, 237 1, 7, 13, 21–22, 185, 242, 256;
Tessman, Lisa 69, 82 interpersonal 191–192; monitoring 11,
testimonial: credibility 74, 81; smothering 155–161; -reliabilism 185, 186n22, 229;
115 -responsibilism 185n2, 228, 231; virtues
testimony 97, 105n12, 114–116, 124n10, of epistemic justice 89, 119
149–153, 156–158, 161n6, 191, Virtues of the Mind 5
193–194, 205n7, 212 Volante, Louis 233
time-wasters 183
tobacco industry 76 Watson, Gary 170
trolls 154, 202 Watson, Lani 12, 225, 229, 236, 239, 242,
trust 10, 54, 113, 115, 117, 156, 158–159, 255–256
201, 218 Westacott, Emrys 209–210
truthfulness 6, 96–97 whistle-blowers 96–97
Tuana, Nancy 54–55 white privilege 55–57, 60, 62, 170–171
264 Index
Whitt, Matt 109–110, 117, 120–122 Yom Kippur 42, 47–48
Williams, Bernard 6
Wire, The (TV show) 225 Zagzebski, Linda 5, 26, 39, 75, 81, 93, 98,
Wollstonecraft, Mary 4, 69 105, 148, 228–229, 236–237, 255; see
Wonder Boys 170 also Virtues of the Mind
Wood, W. Jay 5, 7, 31, 82–83 Zhuāngzǐ 3

Vice Epistemology - PDF Version 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Vice Epistemology - PDF Version 1

Uploaded by

Copyright:

Available Formats

Ian James Kidd is an Assistant Professor of Philosophy at the University of

Heather Battaly is a Professor of Philosophy at the University of Connecticut,

Quassim Cassam is a Professor of Philosophy at the University of Warwick, UK.

Edited by Ian James Kidd,

Introduction: from epistemic vices to vice epistemology 1

1 The structure of intellectual vices 21

2 The metaphysical foundations of vice epistemology 37

3 Ignorance, arrogance, and privilege: vice epistemology and the

4 Epistemic corruption and social oppression 69

5 Institutional epistemic vices: the case of inferential inertia 89

7 Implicit bias and epistemic vice 126

8 Vectors of epistemic insecurity 148

9 Quitting, procrastinating, and slacking off 167

10 Epistemic insensitivity: an insidious and consequential vice 189

11 Intellectual snobs 208

12 Teaching to the test: how schools discourage phronesis 225

13 Vices of questioning in public discourse 239

Ian James Kidd is an Assistant Professor of Philosophy at the University of

Ian is grateful to the University of Nottingham for a period of research leave. We

The historical study of epistemic vices: a brief tour

Key themes in contemporary vice epistemology

Much of virtue epistemology turns on an insight concerning the relationship

1.1 The structure of intellectual virtues

1.1.1 Competence dimension

1.1.2 Motivational dimension

1.1.3 Judgment dimension

1.2 The structure of intellectual vices

1.2.1 Defective motivation

Finding #1: defective epistemic motivation (whether positive or negative)

1.2.2 Defective judgment

Finding #3: defective judgment can be sufficient for the possession of an

1.2.3 Defective competence

Finding #4: a lack of virtue-relevant competence can be sufficient for the

Finding #6: a deficiency in virtuous affections is not, by itself, sufficient for

1.3.1 Motivation and responsibility

1.2.3 Purported counterexamples to motivationalism

1 What kind of thing is an epistemic vice?

3.1 Varieties of ignorance

3.2 Ignorance as a vicious sensibility

3.3 Ignorance and arrogance

The dominant understanding of vice, virtue, and character needs to be revised in

4.1 Character and oppression

4.2 The concept of epistemic corruption

1 Acquisition: a corruptor can enable the acquisition of novel epistemic vices,

4.3 Presuppositions and predicaments

4.4 Epistemically corrupting conditions

1 The absence of exemplars of virtue: a community may lack any positive

4.5 Critical character epistemology

4.5.1 Aetiological sensitivity

4.5.2 Axiological pluralism

4.5.3 Normative contextualism

[According to a critical character theory] character dispositions would be

Collectives, institutions, and

One of the achievements of virtue epistemology has been the identification of an

5.1 Ethos matters

5.2 Modelling ethos

For our distinctly epistemic purposes at the level of institutions, my suggestion is

5.3 Modelling institutional epistemic vice

5.4 The institutional vice of inferential inertia

5.4.1 The culture of not complaining about the talent

63 Even within the same programme, there could be difficulties of commu-

5.4.2 The macho culture